Abstract
Objective
Based on bioinformatics screening, C-X-C motif chemokine ligand 2 (CXCL2) and fibronectin 1 (FN1) were identified as common targets of acute myocardial infarction (AMI) and neutrophil extracellular traps (NETs). This study aimed to validate their adjunctive differential diagnostic value in emergency department patients with chest pain suspected of AMI, and explore NET associations.
Methods
In this prospective observational study, 349 patients with chest pain suspected of AMI were enrolled between June 2022 and January 2025 and classified as non-AMI (n=226) or AMI (n=123). Serum CXCL2, FN1, and NETs biomarkers (myeloperoxidase-DNA [MPO-DNA] and citrullinated histone H3 [CitH3]) were measured by enzyme-linked immunosorbent assay (ELISA). Multivariable logistic regression, receiver operating characteristic (ROC) curves, and Spearman correlation were performed.
Results
AMI patients had higher proportions of males, smokers, and angina, with elevated heart rate, fasting blood glucose (FBG), glycated hemoglobin (HbA1c), neutrophil-to-lymphocyte ratio (NLR), and platelet-to-lymphocyte ratio (PLR), and reduced high-density lipoprotein cholesterol (HDL-C) versus non-AMI (P<0.05). CXCL2, FN1, MPO-DNA, and CitH3 were elevated in AMI (P<0.05). Multivariable analysis showed heart rate, FBG, HbA1c, NLR, CXCL2, and FN1 as independent AMI factors, and HDL-C was negatively associated (P<0.05). ROC analysis showed AUCs of 0.786 and 0.755 for CXCL2 and FN1 alone, and 0.816 combined. CXCL2 and FN1 correlated positively with MPO-DNA and CitH3 (P<0.001).
Conclusion
Serum CXCL2 and FN1 are elevated in AMI and correlate with NETs biomarkers. Combined detection demonstrated moderate diagnostic performance in patients with chest pain suspected of AMI, suggesting potential as novel auxiliary biomarkers for differential diagnosis in the ED. However, this single-center study had limited sample size and was positioned for differential diagnosis in symptomatic patients rather than asymptomatic screening. Multicenter, large-scale validation is needed to confirm reproducibility and generalizability, and functional studies (in vitro and in vivo) are required to clarify the mechanisms of NETs regulation.
Keywords: acute myocardial infarction, C-X-C motif chemokine ligand 2, fibronectin-1, neutrophil extracellular traps
Graphical Abstract
Introduction
Acute myocardial infarction (AMI), most often triggered by abrupt coronary occlusion with sustained ischemic necrosis of the myocardium, remains one of the most lethal emergent cardiovascular events.1 Although early reperfusion markedly improves prognosis, both electrocardiography and conventional cardiac enzymes lack sufficient sensitivity during the initial hours, underscoring the urgent need for more sensitive and specific serum biomarkers to enable rapid recognition and prompt intervention.2,3
Neutrophil extracellular traps (NETs)—DNA-based lattices decorated with neutrophil granule proteins—exert direct cytotoxicity, enhance platelet adhesion, and ignite the coagulation cascade, thereby accelerating intracoronary thrombosis and precipitating microvascular occlusion that expands infarct size and fuels acute myocardial infarction progression.4–6 These findings identify NETs as a potential new target for early AMI diagnosis and risk stratification.
C-X-C motif chemokine ligand 2 (CXCL2) is a chemokine that primarily recruits neutrophils to sites of inflammation or injury, thereby orchestrating inflammatory responses.7 In murine AMI models, post-ischemic suppression of CXCL2 transcription and secretion curbs neutrophil infiltration into the infarcted myocardium, attenuating inflammation and limiting cardiac damage.8 Fibronectin-1 (FN1), a high-molecular-weight glycoprotein abundant in the extracellular matrix, mediates cell migration, invasion, metastasis, and thrombogenesis.9–11 In mouse myocardial ischemia/reperfusion models, FN1 expression is markedly up-regulated, and its blockade ameliorates oxidative stress and inflammation—via activation of the AMP-LKB1-AMPK axis—thereby reducing myocardial injury.12 CXCL2-mediated neutrophil recruitment and FN1-mediated extracellular matrix destruction may play important roles in the pathophysiology of AMI; however, systematic studies on their dynamic changes in human serum are lacking, and their association with NETs formation remains to be elucidated.
Accordingly, this study aimed to investigate serum CXCL2 and FN1 levels and their differential diagnostic value in patients with chest pain suspected of AMI, and to further analyze their relationship with NETs formation, thereby providing novel theoretical evidence and potential auxiliary biomarkers for the differential diagnosis and mechanistic investigation of AMI.
Materials and Methods
Bioinformatic Analysis
The NCBI Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/) was searched using the keywords “acute myocardial infarction” AND “Homo sapiens”, with the data type restricted to “expression profiling by array”. Among the datasets meeting the search criteria, the final datasets for analysis were selected according to the following criteria: (1) samples were derived from peripheral blood and included both AMI patients and healthy controls, to ensure logical consistency with subsequent clinical serum validation; (2) the Affymetrix microarray platform was employed, ensuring comparability between the two datasets; (3) the sample size was sufficient to meet the statistical power requirements for differential expression analysis. The microarray datasets GSE19339 and GSE66360 were ultimately selected. Analysis was performed using the official online tool GEO2R, which by default calls the limma package for linear modeling. Multiple testing correction was performed using the Benjamini–Hochberg method. The screening thresholds were set at |log2FC| > 1 and FDR < 0.05.
Construction of the NETs-related gene set: A NETs-related gene set was constructed through systematic literature search and manual curation. The PubMed database was searched for relevant literature published up to 2022 using the terms “neutrophil extracellular traps”, “NETs”, and “NETosis”. Priority was given to studies that directly identified NETs formation regulatory molecules through proteomics, transcriptomics, or gene editing, as well as core NETs components summarized in systematic reviews. The NETs-related gene lists manually curated from the primary integrated literature13,14 were incorporated. The inclusion criteria were as follows: (a) structural core components of NETs (eg, genes encoding elastase, myeloperoxidase, and histones); (b) key regulatory factors of NETosis (eg, PAD4 and NOX2); (c) chemokines or extracellular matrix molecules associated with NETs release; (d) molecules directly confirmed to be functionally associated with NETs in at least two independent studies. After merging and deduplication, 147 non-redundant genes were ultimately obtained. The aforementioned AMI differentially expressed genes and the NETs-related gene list were uploaded to the Venn online tool (http://bioinformatics.psb.ugent.be/webtools/Venn/) to obtain the intersection, yielding common targets for subsequent analysis.
Study Population
Based on prior data,15 the probability that a patient presenting with chest pain and suspected AMI will receive a final adjudicated diagnosis of AMI is 30.91%. Assuming an absolute precision of 5% (ie, 95% CI width = 10%) and two-sided α = 0.05, PASS 15.0 software indicated that 346 participants would be required to ensure statistical validity.
Inclusion criteria: (1) AMI diagnosed in accordance with the Fourth Universal Definition of Myocardial Infarction (2018 ESC/ACCF/AHA/WHF):16 cardiac troponin (cTn) elevation with at least one value above the 99th percentile upper reference limit (URL) and a serial rise and/or fall, plus at least one of the following: ① symptoms of AMI (eg, chest pain, dyspnea) or new ischemic ECG changes (ST-segment elevation or depression, new pathologic Q waves); ② imaging (eg, cardiac magnetic resonance or echocardiography) demonstrates new loss of myocardial viability or regional wall-motion abnormalities consistent with an ischemic etiology; or ③ coronary angiographic documentation of intracoronary thrombus. (2) Time from symptom onset to admission < 12 h; (3) Age 18–80 years; (4) Written informed consent provided.
Exclusion criteria: (1) Severe hepatic or renal dysfunction; (2) Active malignancy, autoimmune disease, or hematologic disorder; (3) Major infection, surgery, or trauma within the preceding 3 months; (4) Pregnancy or lactation; (5) Pre-hospital thrombolysis or PCI before enrollment; (6) Psychiatric illness or cognitive/language impairment precluding cooperation; (7) Congenital heart disease or severe viral myocarditis; (8) Significant arrhythmia or chronic heart failure deemed to confound interpretation; (9) Use of systemic glucocorticoids, immunosuppressants, biologics, or chemotherapeutic agents within the past month.; (10) Anticipated discharge or transfer to another facility within 24 h.
This prospective, single-center, observational study consecutively enrolled 456 patients presenting with chest pain and suspected AMI who were admitted to our hospital between June 2022 and January 2025. After strict application of inclusion and exclusion criteria, 349 individuals were retained for the final analysis and classified into an AMI group (n = 123) or a non-AMI group (n = 226) based on the adjudicated diagnosis. The participant-flow diagram is shown in Figure 1.
Figure 1.
Participant screening flowchart.
Methods
Data Collection
Baseline information was recorded at admission. General characteristics comprised sex, age, body-mass index (BMI), smoking and alcohol habits, family history of cardiovascular disease, and prior medical history (hypertension, diabetes mellitus, dyslipidemia, angina). Time from chest-pain onset to presentation was also documented.
Laboratory variables obtained on admission included systolic/diastolic blood pressure, heart rate, fasting blood glucose (FBG), glycated hemoglobin (HbA1c), total cholesterol (TC), triglycerides (TG), high-density-lipoprotein cholesterol (HDL-C), low-density-lipoprotein cholesterol (LDL-C), white blood cell (WBC) count, hemoglobin (Hb), neutrophil count (NEUT), lymphocyte count (LYMPH), and platelet count (PLT). The neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR) were calculated from these values.
Measurement of Serum CXCL2, FN1, and NETs-Associated Biomarkers (MPO–DNA, CitH3)
Five milliliters of fasting venous blood was collected from each patient within 24 h of admission, allowed to clot at room temperature for 30 min, and centrifuged at 1000 × g for 15 min; aliquots of resulting serum were stored at −80 °C until analysis and subjected to no more than one freeze–thaw cycle. On the day of assay, sera and all reagents were equilibrated to room temperature. Procedures followed the manufacturers’ instructions exactly: standards and pre-diluted sera were added to microplates pre-coated with capture antibody, incubated, washed, and sequentially treated with biotinylated detection antibody or HRP-conjugated secondary antibody. After further incubation and washing, TMB substrate was added for color development; the reaction was stopped with stop solution and absorbance was read at 450 nm. Concentrations were calculated from a four-parameter logistic standard curve; each sample was assayed in duplicate and repeated if the intra-assay CV exceeded 15%. Every plate included high- and low-level controls, with intra- and inter-assay CVs maintained below 10%. Throughout the process, operators and data analysts were blinded to clinical status; samples were identified only by random codes assigned by personnel not involved in the assay.
Statistical Analysis
All analyses were performed with SPSS 26.0 (IBM, Armonk, NY, USA) and graphs were generated in GraphPad Prism 9.5. Continuous variables were first tested for normality using the Kolmogorov–Smirnov test. Normally distributed data are presented as mean ± SD and compared between groups with the independent-samples t test; non-normally distributed data are expressed as median (interquartile range [P25, P75]) and compared with the Mann–Whitney U-test. Categorical variables are reported as number (percentage) and evaluated with the χ2-test. Multivariable logistic regression (Enter method) was used to identify independent predictors of AMI. The omnibus test was employed to assess the overall significance of the model, the Hosmer–Lemeshow goodness-of-fit test was used to evaluate model calibration, and Nagelkerke R2 was calculated to reflect the explanatory power of the model. ROC curves were constructed to evaluate the diagnostic performance of CXCL2, FN1, and their combination. For the combined model, raw CXCL2 and FN1 values were entered into a binary logistic regression to generate a predicted probability (P), which was then subjected to ROC analysis. The optimal cut-off was defined as the value yielding the maximum Youden index (sensitivity + specificity − 100), and the corresponding sensitivity and specificity are reported. Correlations between serum CXCL2, FN1, and NETs biomarkers (MPO–DNA and CitH3) were assessed with Spearman’s rank correlation. A two-sided P < 0.05 was considered statistically significant.
Results
Bioinformatic Findings
Intersection of AMI-related genes from GSE19339 and GSE66360 with the NETs-related gene set yielded only two overlapping transcripts—CXCL2 and FN1—both significantly up-regulated in the two arrays (Figure 2A–C). Subsequent analyses therefore focused on functional validation and mechanistic interrogation of these two candidates.
Figure 2.
Bioinformatics analysis results. (A) Volcano plot of GSE19339; (B) Volcano plot of GSE66360; (C) Intersection of AMI-related and NET-related genes. The red gene labels in A and B (CXCL2 and FN1) indicate key significantly upregulated genes.
Baseline Characteristics
Compared with the non-AMI cohort, the AMI group had a higher proportion of males, active smokers, and prior angina, together with higher heart rate, fasting blood glucose (FBG), HbA1c, NLR, and PLR, whereas HDL-C was lower (all P < 0.05). No significant between-group differences were observed for age, BMI, alcohol use, family history of cardiovascular disease, hypertension, diabetes, time to chest-pain onset, blood pressure, TC, TG, LDL-C, PLT, Hb, or WBC count (all P > 0.05) (Table 1).
Table 1.
Comparison of Baseline Characteristics Between Non-AMI and AMI Patients
| Indicator | Non-AMI Group (n=226) | AMI Group (n=123) | t/Z/χ2 | P |
|---|---|---|---|---|
| Sex (n/%) | 9.476^ | 0.002 | ||
| Male | 97 (42.92) | 74 (60.16) | ||
| Female | 129 (57.08) | 49 (39.83) | ||
| Age [M (P25, P75), years] | 59 (57, 61) | 59 (57, 62) | −0.561# | 0.575 |
BMI ( , kg/m2) |
22.57±1.82 | 22.48±2.12 | 0.444* | 0.658 |
| Smoking history (n/%) | 92 (40.71) | 77 (62.62) | 15.287^ | 0.000 |
| Alcohol use (n/%) | 52 (23.01) | 34 (27.64) | 0.921^ | 0.337 |
| Family history of cardiovascular disease (n/%) | 51 (22.53) | 36 (29.27) | 1.912^ | 0.167 |
| Hypertension (n/%) | 77 (34.07) | 50 (40.65) | 1.490^ | 0.222 |
| Diabetes mellitus (n/%) | 49 (21.68) | 33 (26.83) | 1.174^ | 0.279 |
| Hyperlipidemia (n/%) | 89 (39.38) | 60 (48.78) | 2.877^ | 0.090 |
| History of angina (n/%) | 51 (22.57) | 75 (60.98) | 25.235^ | 0.000 |
| Time to chest-pain onset [M (P25, P75), h] | 4 (3, 5) | 4 (3, 5) | −0.728# | 0.466 |
| SBP[M (P25, P75), mmHg] | 127 (120, 131) | 126 (121, 133) | −0.398# | 0.691 |
| DBP[M (P25, P75), mmHg] | 76 (69, 84) | 76 (69, 86) | −0.257# | 0.797 |
Heart rate ( , beats/min) |
79.22±12.54 | 88.67±13.22 | −6.594* | 0.000 |
TC ( , mmol/L) |
4.82±1.02 | 4.62±1.05 | 1.747* | 0.081 |
TG ( , mmol/L) |
1.36±0.23 | 1.41±0.35 | −1.650* | 0.100 |
| HDL-C[M (P25, P75), mmol/L] | 1.35 (1.22, 1.50) | 0.97 (0.82, 1.06) | −12.263# | 0.000 |
LDL-C ( , mmol/L) |
3.12±0.52 | 3.06±0.47 | 1.072* | 0.285 |
FBG ( , mmol/L) |
5.58±1.12 | 6.45±1.15 | −6.916* | 0.000 |
| HbA1c [M (P25, P75), %] | 5.6 (5.3, 5.9) | 6.4 (6.0, 6.8) | −11.669# | 0.000 |
| PLT (10^9/L) | 198.26±45.86 | 203.57±42.41 | −1.062* | 0.289 |
Hb ( , g/L) |
146.51±15.21 | 143.91±19.12 | 1.391* | 0.165 |
| WBC [M (P25, P75), 10^9/L] | 5.8 (5.1, 6.6) | 6.0 (5.5, 6.3) | −1.454# | 0.146 |
| NLR [M (P25, P75)] | 2.5 (1.9, 3.0) | 3.6 (2.8, 4.7) | −8.815# | 0.000 |
PLR ( ) |
134.56±25.12 | 143.78±38.69 | −2.691* | 0.007 |
Notes: M, median; P25, 25th percentile; P75, 75th percentile. * t value, #Z value, ^ χ2value.
Abbreviations: BMI, body-mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; FBG, fasting plasma glucose; HbA1c, glycated hemoglobin A1c; PLT, platelet count; Hb, hemoglobin; WBC, white blood cell; NLR, neutrophil-to-lymphocyte ratio; PLR, platelet-to-lymphocyte ratio.
Comparison of Serum CXCL2 and FN1 Levels Between Non-AMI and AMI Patients
Serum CXCL2 and fibronectin-1 concentrations were markedly elevated in the AMI group relative to the non-AMI group (both P < 0.05) (Table 2, Figure 3).
Table 2.
Comparison of Serum CXCL2 and FN1 Levels Between Non-AMI and AMI Patients
| Group | n | CXCL2 ( , pg/mL) |
FN1 [M (P25, P75), ng/mL] |
|---|---|---|---|
| Non-AMI group | 226 | 48.66±16.44 | 9.8 (8.2, 11.3) |
| AMI group | 123 | 64.52±12.16 | 12.3 (10.2, 14.0) |
| t /Z | −9.390* | −7.875# | |
| P | 0.000 | 0.000 |
Notes: M, median; P25, 25th percentile; P75, 75th percentile. * t value, # Z value.
Abbreviations: AMI, acute myocardial infarction; CXCL2, C-X-C motif chemokine ligand 2; FN1, Fibronectin-1.
Figure 3.
Comparison of serum CXCL2 and FN1 levels between non-AMI and AMI patients. (A) Serum CXCL2 was higher in AMI than non-AMI (t = −9.390, P < 0.001); (B) Serum FN1 was higher in AMI than non-AMI (Z = −7.875, P < 0.001).
Multivariable Logistic Regression Analysis of Factors Influencing AMI Development
Variables that showed statistical significance in the univariable analysis (sex, smoking history, angina history, heart rate, HDL-C, FBG, HbA1c, NLR, PLR, CXCL2, and FN1) were examined for multicollinearity. All variance inflation factors (VIFs) were < 2, indicating absence of collinearity; consequently, these factors were retained for entry into the subsequent multivariable model. Multivariable logistic regression analysis was performed with AMI occurrence (no = 0, yes = 1) as the dependent variable, and sex (female = 0, male = 1), smoking history (no = 0, yes = 1), angina history (no = 0, yes = 1), heart rate, HDLC, FBG, HbA1c, NLR, PLR, CXCL2, and FN1 as independent variables. The results showed that elevated heart rate, FBG, HbA1c, NLR, CXCL2, and FN1 were independent factors associated with AMI, whereas elevated HDLC was an independent negatively associated factor (Table 3). The overall model was statistically significant (Omnibus test: χ2 = 379.173, P < 0.001), demonstrated good calibration (Hosmer–Lemeshow test: χ2 = 9.039, P = 0.339), and showed adequate explanatory power (Nagelkerke R2 = 0.598).
Table 3.
Multivariable Logistic Regression Analysis of Factors Influencing AMI Occurrence
| Indicator | β | S. E | Walds | P | OR | 95% CI | VIF |
|---|---|---|---|---|---|---|---|
| Sex | 0.773 | 0.676 | 1.308 | 0.253 | 2.167 | 0.576–8.158 | 1.052 |
| Smoking history | 0.915 | 0.680 | 1.807 | 0.179 | 2.496 | 0.658–9.469 | 1.100 |
| Angina history | 0.091 | 0.689 | 0.018 | 0.895 | 1.096 | 0.284–4.231 | 1.127 |
| Heart rate | 0.091 | 0.029 | 10.029 | 0.002 | 1.095 | 1.035–1.158 | 1.121 |
| HDL-C | −0.951 | 0.288 | 10.923 | 0.001 | 0.386 | 0.220–0.679 | 1.366 |
| FBG | 0.947 | 0.318 | 8.890 | 0.003 | 2.578 | 1.383–4.804 | 1.121 |
| HbA1c | 0.617 | 0.133 | 21.388 | 0.000 | 1.852 | 1.427–2.406 | 1.418 |
| NLR | 1.062 | 0.197 | 29.051 | 0.000 | 2.893 | 1.966–4.256 | 1.305 |
| PLR | 0.020 | 0.011 | 3.707 | 0.054 | 1.020 | 1.000–1.042 | 1.058 |
| CXCL2 | 0.070 | 0.026 | 7.111 | 0.008 | 1.072 | 1.019–1.128 | 1.261 |
| FN1 | 0.630 | 0.168 | 13.975 | 0.000 | 1.877 | 1.349–2.611 | 1.168 |
Abbreviations: HDL-C, high-density lipoprotein cholesterol; FBG, fasting plasma glucose; HbA1c, glycated hemoglobin A1c; NLR, neutrophil-to-lymphocyte ratio; PLR, platelet-to-lymphocyte ratio; CXCL2, C-X-C motif chemokine ligand 2; FN1, fibronectin-1.
Diagnostic Value of Serum CXCL2 and FN1 in the Differential Diagnosis of AMI
ROC analysis yielded AUCs of 0.786 for CXCL2 alone, 0.755 for FN1 alone, and 0.816 for the two markers combined; the composite model also achieved higher sensitivity than either individual biomarker (Table 4, Figure 4).
Table 4.
Diagnostic Value of Serum CXCL2 and FN1 in the Differential Diagnosis of AMI
| Indicator | AUC | 95% CI | Cut-off | Sensitivity (%) | Specificity (%) |
|---|---|---|---|---|---|
| CXCL2 | 0.786 | 0.738–0.834 | 55.25 pg/mL | 81.30 | 69.03 |
| FN1 | 0.755 | 0.701–0.809 | 11.75 ng/mL | 60.16 | 79.20 |
| CXCL2+ FN1 | 0.816 | 0.773–0.860 | - | 90.24 | 65.49 |
Abbreviations: CXCL2, C-X-C motif chemokine ligand 2; FN1, fibronectin-1.
Figure 4.
ROC curves of serum CXCL2 and FN1 for the differential diagnosis of AMI.
Correlation Analysis of Serum CXCL2 and FN1 with NETs Biomarkers in AMI Patients
Serum NETs markers (MPO–DNA and CitH3) were significantly higher in AMI patients. Spearman correlation revealed strong positive associations between both CXCL2 and FN1 and the NETs biomarkers in the AMI group (Table 5, Figure 5).
Table 5.
Comparison of Serum MPO–DNA and CitH3 Levels Between Non-AMI and AMI Patients
| Group | n | MPO-DNA (M (P25, P75), ng/mL) | CitH3 (M (P25, P75), ng/mL) |
|---|---|---|---|
| Non-AMI group | 226 | 1.8 (1.7, 2.0) | 8.25 (6.9, 9.5) |
| AMI group | 123 | 2.3 (2.1, 2.6) | 11.0 (9.0, 13.4) |
| Z | −11.131# | −8.877# | |
| P | 0.000 | 0.000 |
Notes: M, median; P25, 25th percentile; P75, 75th percentile. # denotes Z value.
Abbreviations: AMI, acute myocardial infarction; MPO-DNA, myeloperoxidase-DNA; CitH3, citrullinated histone H3.
Figure 5.
Correlation of serum CXCL2 and FN1 with NET biomarkers. (A and B) Correlation of serum CXCL2 with MPO–DNA and CitH3 levels; (C and D) Correlation of serum FN1 with MPO–DNA and CitH3 levels.
Discussion
This study evaluated the adjunctive differential diagnostic value of serum CXCL2 combined with FN1 in patients with chest pain suspected of AMI, and revealed a close association between these two biomarkers and NETs formation. The results demonstrated that CXCL2 and FN1 were significantly elevated in AMI patients and were independent factors associated with AMI. Combined detection of these two markers showed moderate diagnostic performance in this population, and both were significantly positively correlated with the NETs biomarkers MPO-DNA and CitH3. These findings suggest that CXCL2 and FN1 may serve not only as novel auxiliary biomarkers for the early differential diagnosis of AMI, but also participate in the disease process through promoting NETs formation, thereby providing novel insights for the mechanistic investigation and diagnosis of AMI.
NETs—lattices of DNA, histones, and antimicrobial proteins released by neutrophils—efficiently ensnare and kill pathogens, serving as a cornerstone of antibacterial defense. Excessive NETs formation, however, can occlude vessels, damage tissue, and amplify inflammation.17 In cardiovascular disorders such as AMI and heart failure, NETs abundance and their biomarkers (eg, CitH3 and MPO–DNA) are markedly increased; sustained NETs accumulation fuels a self-perpetuating inflammatory–thrombotic cycle via activation of TLR and NLRP3 inflammasome signaling, thereby enlarging infarct size and worsening cardiac function.18 Conversely, targeted inhibition of NETs generation attenuates myocardial ischemia–reperfusion injury, curbs inflammatory infiltration, and limits necrotic expansion.19 Collectively, these findings indicate that a cascade beginning with excessive NET formation, progressing through amplified inflammation and thrombosis, and culminating in myocardial injury may constitute a central driver in the initiation and progression of acute myocardial infarction.
In this study, serum CXCL2 and FN1 were markedly elevated in AMI patients and correlated positively with the NETs biomarkers MPO–DNA and CitH3, suggesting that both analytes promote AMI pathobiology by modulating NETs formation. CXCL2, a chemokine that rapidly recruits and activates neutrophils, is up-regulated in multiple inflammatory disorders; binding to CXCR2 triggers downstream MAPK and NF-κB signaling, amplifying inflammation and tissue injury.7 In sepsis models, high CXCL2 expression augments hepatic neutrophil infiltration and worsens organ damage, whereas CXCL2 knockdown reduces myeloperoxidase (MPO) activity and injury.20 In the present study, the significant rise in serum CXCL2 implies that CXCR2-mediated recruitment of neutrophils to ischemic myocardium not only intensifies local inflammation and cardiomyocyte damage but also supplies the cellular substrate for NETs formation, further escalating the inflammatory cascade and propelling AMI progression.4,21,22
FN1, a multifunctional glycoprotein, has recently been shown to correlate directly with thrombus volume in cerebral venous thrombosis (CVT); FN1-targeted intervention curbs thrombus growth and lowers re-thrombosis risk.11 Here, the observed up-regulation of serum FN1 in AMI may promote thrombosis via integrin α5β1–PI3K/Akt-mediated activation of endothelial cells and platelets.23 Concurrently, FN1 ligation to integrin β1/α5 perpetuates FAK→Rac-1→NADPH oxidase signaling, rapidly elevating intracellular ROS;24 heightened ROS triggers NOX-independent NETs release. The ensuing NETs amplify coagulation while suppressing fibrinolysis, generating a hypercoagulable milieu that promotes thrombus formation and thereby increases the risk of AMI.25,26 However, these mechanisms were inferred from other disease models and in vitro studies; the specific causal direction and direct regulatory role in AMI remain to be elucidated through functional studies such as in vitro cell experiments and animal models.
Multivariable logistic regression showed that elevated serum CXCL2 and FN1 concentrations were independent factors associated with AMI. ROC curve analysis further demonstrated that both markers alone and in combination possessed certain adjunctive differential diagnostic value in patients with chest pain suspected of AMI, suggesting that CXCL2 and FN1 may serve as auxiliary indicators for risk stratification of acute chest pain in the emergency department, thereby enhancing early identification of AMI. In addition, we found that increased heart rate, FPG, HbA1c, and NLR were independent risk factors for AMI, whereas HDL-C was protective—findings consistent with previous reports.27–31 Collectively, these data suggest that integrated management of glycemia, inflammation, and lipid profile in high-risk individuals may aid AMI prevention, although specific intervention strategies require further clinical investigation.
It should be emphasized that cTn is recognized as the gold-standard biomarker of cardiomyocyte necrosis, and its diagnostic position remains unshakeable in current guidelines. This study was not intended to replace cTn with CXCL2 or FN1. Rather, CXCL2 and FN1 reflect inflammatory–thrombotic cascade activation and NETs formation, biological processes closely related to the pathophysiology of AMI, whereas cTn reflects myocardial necrosis per se; thus, the two represent different dimensions of evaluation. Therefore, we hypothesize that CXCL2/FN1 and cTn may have a complementary rather than competitive relationship: cTn is used to confirm the presence of myocardial injury, whereas CXCL2/FN1 may provide incremental information regarding the intensity of early inflammatory response, neutrophil activation status, and NETs-mediated thrombotic risk. Future studies may further explore the incorporation of CXCL2/FN1 into existing multimarker diagnostic strategies to evaluate whether they can provide additional risk stratification value in early chest pain patients with negative cTn, or to predict inflammation-related complications following AMI. However, this hypothesis remains to be confirmed by prospective, large-scale, and externally validated studies.
This study has several limitations. First, the single-center observational design and limited sample size, together with the bioinformatics analysis incorporating only two GEO datasets, may have affected the comprehensiveness and robustness of target screening. Moreover, the AMI group was not stratified by STEMI/NSTEMI, culprit vessel, or reperfusion strategy, and the control group exhibited high etiological heterogeneity, limiting the generalizability of the conclusions. Second, only a single time-point serum sample obtained within 24 h of admission was measured; although this reflects the acute pathological state, it failed to capture the very early (<3 h) dynamic changes of biomarkers after symptom onset. The lack of longitudinal follow-up after discharge precluded determination of peak time, clearance kinetics, and prognostic associations with myocardial remodeling and MACE. In addition, this study did not incorporate key variables such as Killip class, LVEF, and infarct size, limiting the assessment of dose–response relationships between the biomarkers and cardiac function or infarct area. Third, because cTn is an essential component of the gold-standard diagnostic criteria for AMI, its inclusion in ROC comparison would introduce incorporation bias; therefore, a head-to-head comparison with existing biomarkers was not performed, and the true incremental value of CXCL2 and FN1 awaits further validation. Fourth, based on cross-sectional data, only statistical associations could be established, and reverse causation and residual confounding could not be excluded. The absence of in vitro functional experiments and animal model validation precluded confirmation of the direct regulatory effects of CXCL2 and FN1 on NETs formation and the underlying causal mechanisms. Finally, the conclusions of this study are applicable only to the adjunctive differential diagnosis of patients with chest pain suspected of AMI and should not be extrapolated to screening in asymptomatic populations. Future studies integrating larger-scale transcriptomic cohorts, multicenter prospective dynamic cohorts, and functional investigations are needed to further clarify their clinical translational value.
In summary, this study found that CXCL2 and FN1 levels were significantly associated with AMI and NETs biomarkers, and these two markers may participate in the inflammatory–thrombotic cascade of AMI and drive disease progression through regulating NETs formation. Limitations include the single-center design, restricted sample source, potential selection bias, and reliance on a single admission-time point without serial time-course monitoring or external validation, and the applicable scenario of the conclusions was limited to patients with chest pain suspected of AMI. Multicenter, large-scale cohorts with dynamic follow-up and complementary animal studies are warranted to clarify the clinical translational value and precise mechanisms of CXCL2 and FN1, thereby providing a firmer basis for early recognition and targeted intervention in AMI.
Funding Statement
The authors received no specific funding for this work.
Ethics Statement
This study was approved by the Ethics Committee of Wuhan No.1 Hospital (Approval Number: 2022-15) and adheres to the principles outlined in the Declaration of Helsinki. Written informed consent was obtained from all participants, ensuring their voluntary participation. Furthermore, measures were taken to protect the confidentiality and anonymity of the patients involved in this research.
Disclosure
The authors declare that there is no conflict of interest.
References
- 1.Zhu M, Li Y, Xu Q, Wang W, Liu Y, Liu Y. Acute myocardial infarction: molecular pathogenesis, diagnosis, and clinical management. MedComm. 2025;6(10):e70418. doi: 10.1002/mco2.70418 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Huang X, Bai S, Luo Y. Advances in research on biomarkers associated with acute myocardial infarction: a review. Medicine (Baltimore). 2024;103(15):e37793. doi: 10.1097/md.0000000000037793 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Shi LY, Han YS, Chen J, Li ZB, Li JC, Jiang TT. Screening and identification of potential protein biomarkers for the early diagnosis of acute myocardial infarction. Ann Transl Med. 2021;9(9):743. doi: 10.21037/atm-20-7891 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.He L, Liu R, Yue H, et al. Interaction between neutrophil extracellular traps and cardiomyocytes contributes to atrial fibrillation progression. Signal Transduct Target Ther. 2023;8(1):279. doi: 10.1038/s41392-023-01497-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Mangold A, Hofbauer TM, Ondracek AS, et al. Neutrophil extracellular traps and monocyte subsets at the culprit lesion site of myocardial infarction patients. Sci Rep. 2019;9(1):16304. doi: 10.1038/s41598-019-52671-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Zhang X, Song H, Liu D, et al. S100A12 triggers NETosis to aggravate myocardial infarction injury via the Annexin A5-calcium axis. Nat Commun. 2025;16(1):1746. doi: 10.1038/s41467-025-56978-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Lv Y, Chen C, Han M, et al. CXCL2: a key player in the tumor microenvironment and inflammatory diseases. Cancer Cell Int. 2025;25(1):133. doi: 10.1186/s12935-025-03765-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Mylonas KJ, Turner NA, Bageghni SA, et al. 11β-HSD1 suppresses cardiac fibroblast CXCL2, CXCL5 and neutrophil recruitment to the heart post MI. J Endocrinol. 2017;233(3):315–12. doi: 10.1530/joe-16-0501 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Ji Z, Li X, Wang Y, et al. Fibronectin 1 aggravates colon cancer metastasis by regulating RAP1B protein stability through Akt/CREB signalling pathway. J Cell Mol Med. 2025;29(13):e70702. doi: 10.1111/jcmm.70702 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Zhu X, Li Y, Liu H, Xiao Z. FN1 from cancer-associated fibroblasts orchestrates pancreatic cancer metastasis via integrin-PI3K/AKT signaling. Front Oncol. 2025;15:1595523. doi: 10.3389/fonc.2025.1595523 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Jiang H, Chen J, Zhou Y, et al. The FN1-targeted delivery of low-molecular-weight heparin to venous clots revolutionizes thrombosis treatment in CVT. Biomaterials. 2026;326:123653. doi: 10.1016/j.biomaterials.2025.123653 [DOI] [PubMed] [Google Scholar]
- 12.Zhang Y-L, Li P-B, Han X, Zhang B, Li -H-H, Alvarez S. Blockage of fibronectin 1 ameliorates myocardial ischemia/reperfusion injury in association with activation of AMP-LKB1-AMPK signaling pathway. Oxid Med Cell Longev. 2022;2022(1):6196173. doi: 10.1155/2022/6196173 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Zhang Y, Guo L, Dai Q, et al. A signature for pan-cancer prognosis based on neutrophil extracellular traps. J Immunother Cancer. 2022;10(6):e004210. doi: 10.1136/jitc-2021-004210 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Shi H, Pan Y, Xiang G, et al. A novel NET-related gene signature for predicting DLBCL prognosis. J Transl Med. 2023;21(1):630. doi: 10.1186/s12967-023-04494-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Wei Y, Wang B. The expression levels of plasma dimethylglycine (DMG), human maternally expressed gene 3 (MEG3), and Apelin-12 in patients with acute myocardial infarction and their clinical significance. Ann Palliat Med. 2021;10(2):2175–2183. doi: 10.21037/apm-21-122 [DOI] [PubMed] [Google Scholar]
- 16.Thygesen K, Alpert JS, Jaffe AS, et al. Fourth universal definition of myocardial infarction (2018). J Am Coll Cardiol. 2018;72(18):2231–2264. doi: 10.1016/j.jacc.2018.08.1038 [DOI] [PubMed] [Google Scholar]
- 17.Wang H, Kim SJ, Lei Y, et al. Neutrophil extracellular traps in homeostasis and disease. Signal Transduct Target Ther. 2024;9(1):235. doi: 10.1038/s41392-024-01933-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Natorska J, Ząbczyk M, Undas A. Neutrophil extracellular traps (NETs) in cardiovascular diseases: from molecular mechanisms to therapeutic interventions. Kardiol Pol. 2023;81(12):1205–1216. doi: 10.33963/v.kp.98520 [DOI] [PubMed] [Google Scholar]
- 19.Gu LY, Jia CG, Sheng ZZ, et al. Fibroblast growth factor 21 suppressed neutrophil extracellular traps induced by myocardial ischemia/reperfusion injury via adenosine monophosphate-activated protein kinase. Cardiol Res. 2024;15(5):404–414. doi: 10.14740/cr1705 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Wang G, Huang W, Wang S, et al. Macrophagic extracellular vesicle CXCL2 recruits and activates the neutrophil CXCR2/PKC/NOX4 axis in sepsis. J Immunol. 2021;207(8):2118–2128. doi: 10.4049/jimmunol.2100229 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Hilgendorf I, Frantz S, Frangogiannis NG. Repair of the infarcted heart: cellular effectors, molecular mechanisms and therapeutic opportunities. Circ Res. 2024;134(12):1718–1751. doi: 10.1161/circresaha.124.323658 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Cao Y, Chen M, Jiao X, et al. Neutrophil extracellular traps mediate the crosstalk between plaque microenvironment and unstable carotid plaque formation. Exp Mol Med. 2024;56(8):1717–1735. doi: 10.1038/s12276-024-01281-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Maurer LM, Tomasini-Johansson BR, Mosher DF. Emerging roles of fibronectin in thrombosis. Thromb Res. 2010;125(4):287–291. doi: 10.1016/j.thromres.2009.12.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Chen X, Abair TD, Ibanez MR, et al. Integrin alpha1beta1 controls reactive oxygen species synthesis by negatively regulating epidermal growth factor receptor-mediated Rac activation. Mol Cell Biol. 2007;27(9):3313–3326. doi: 10.1128/mcb.01476-06 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Azzouz D, Palaniyar N. How do ROS induce NETosis? Oxidative DNA damage, DNA repair, and chromatin decondensation. Biomolecules. 2024;14(10). doi: 10.3390/biom14101307 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Wu X, Jing H, Jiang T, et al. Platelet-induced NET in liver cirrhosis: mitochondrial ROS-mediated NETosis and its contribution to the hypercoagulable state. Liver Int. 2025;45(8):e70223. doi: 10.1111/liv.70223 [DOI] [PubMed] [Google Scholar]
- 27.Helleryd E, Rawshani A, Rawshani A, Hjärtstam N, Myredal A, Skoglund K. Association between exercise load, resting heart rate, and maximum heart rate and risk of future ST-segment elevation myocardial infarction (STEMI). Open Heart. 2023;10(2):e002307. doi: 10.1136/openhrt-2023-002307 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Li H, Lin Q, Jiang Z, Zhong G. Analyzing the non-linear relationship between fasting blood glucose levels and Gensini score in patients with STEMI. Front Cardiovasc Med. 2024;11:1427567. doi: 10.3389/fcvm.2024.1427567 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Wang R, Tu S, Tan M, Gao L. Clinical risk factors and characteristics of coronary artery lesions in premature acute myocardial infarction patients. Discov Med. 2024;36(190):2253–2263. doi: 10.24976/Discov.Med.202436190.207 [DOI] [PubMed] [Google Scholar]
- 30.Wang Y, Huang X, Chen F, et al. Association between serum BIN1 level and Killip class in patients with acute myocardial infraction. Nan Fang Yi Ke Da Xue Xue Bao. 2024;44(12):2388–2395. doi: 10.12122/j.issn.1673-4254.2024.12.15 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Li Z, Huang J, Li N. Predictive and prognostic value of high-density lipoprotein cholesterol in young male patients with acute myocardial infarction. Chin Med J (Engl). 2017;130(1):77–82. doi: 10.4103/0366-6999.196581 [DOI] [PMC free article] [PubMed] [Google Scholar]