The alteration of circulating exosomal mRNAs in acute myocardial infarction is associated with inammation response mediated by neutrophils

Background Although many cardiovascular disease studies have focused on the characteristics of microRNAs in circulating exosomes, the prole and the potential clinical diagnostic value of plasma exosomal long RNAs (exoLRs) in acute myocardial infarction (AMI) is still unknown. Methods In this study, exosomes isolation and RNA sequencing were applied to achieve the circulating exoLRs prole of 10 AMI patients, 8 stable coronary artery disease (CAD) patients, and 10 healthy individuals. Bioinformatics approaches were used to investigate the features and potential clinical value of exoLRs. Results Each sample from 2 mL of plasma could reliably achieve more than 8000 exosomal messenger RNAs (mRNAs) making up a majority of the total exoLRs. Immune cell types analyzed by CIBERSORT showed that neutrophils and monocytes were signicantly enriched in the AMI group compared to healthy individuals and the CAD group which were consistent with the clinical baseline characteristics. Similarly, the biological processes enrichment analyses of different exosomal mRNAs and co-expression network analysis both indicated that neutrophils activation associated processes were also signicantly enriched in the AMI group. We further identied two exosomal mRNA ALPL and CXCR2 which might be served as potential biomarkers with high diagnostic eciency through co-expression network analysis and receiver operating characteristic curve (ROC). Conclusions In summary, our study explored the alteration of exoLRs in the AMI patients which might associate with the acute inammatory response mediated by neutrophils. We found that exosomal mRNAs ALPL and CXCR2 might be potentially useful for AMI diagnosis. WGCNA: weighted gene co-expression network analysis; MONO: absolute monocyte count; MONO%: the percentage of monocyte; NEUT: absolute neutrophil count; NEUT%: the percentage of neutrophil; CK-MB : creatine kinase, MB Form; HBDH : hydroxybutyrate-dehydrogenase.

associate with the acute in ammatory response mediated by neutrophils. We found that exosomal mRNAs ALPL and CXCR2 might be potentially useful for AMI diagnosis.

Background
Exosomes secreted by most cell types are a class of lipid membrane-enclosed extracellular vesicles ranging in size from 40 to 100 nm [1,2]. These small vesicles not only contain proteins, lipids, RNAs, and metabolites of the source cell but also maintain the compositions stabilized [3]. Therefore, exosomes are now progressively considered as crucial mediators of cell-cell communication and promising potential biomarkers for the disease [4,5].
So far, most studies have focused on examining changes of microRNAs in circulating exosomes and exosomal microRNAs in cardiovascular disease have been well characterized and studied [6]. However, the exosomal mRNAs may not be extensively applied because the production of microRNAs in exosomes is lack of speci c small quantity and lack of speci city [7].
Circular RNA (circRNA), long noncoding RNAs (lncRNA), and messenger RNA (mRNA) known as long RNAs are present and stabilized in exosomes [8]. More and more evidence indicates that the exosomal mRNA may have functional and potential clinical implications [9]. Among various exosomal mRNAs from tubular epithelial cells to macrophages, CCL2 is crucial for albumin-Induced tubulointerstitial in ammation [10]. What's more, two serum exosomal mRNAs, KRTAP5-4 and MAGEA3 could be potential biomarkers to detect colorectal cancer [11]. However, very few studies have concentrated on the characteristics of exosomal long RNAs (exoLRs) in cardiovascular disease, especially for acute myocardial infarction (AMI) known as a common disease with serious consequences in mortality, morbidity, and cost to the society [12]. And the pro le of circulating exoLRs in AMI patients remains unknown.
In this study, we explored the plasma exoLRs pro le of AMI, stable coronary artery disease (CAD), and healthy individuals by RNA sequencing analyses to investigate its features and potential diagnostic value in clinical. The plasma exoLRs pro le could not only re ect the relative fractions of circulating immune cell types but also distinguish patients with AMI from CAD and healthy individuals. Intriguing exosomal mRNAs were identi ed to serve as potential biomarkers for AMI diagnosis. It might contribute to promoting the investigation of the intrinsic mechanisms of AMI.

Patients
This is a case-control study. There were 10 AMI, 8 CAD patients and 10 healthy individuals were recruited at Guangdong Provincial People's Hospital from December 2018 to January 2019. The AMI and CAD patients were diagnosed by laboratory tests and coronary angiography according to the European Society of Cardiology (ESC) guidelines [13,14]. Participants aged between 18-75 years are included, regardless of gender.
The healthy individual's recruitment criteria of included: normal renal and liver function, no history of smoking, malignancy, recent cardiovascular or cerebrovascular event, rheumatological disorders, no diagnosis of chronic heart failure, diabetes, acute or chronic infectious disease, aortic dissection, pulmonary embolism, myocarditis or pericarditis, and congenital heart disease.
Ethical approval of human sample collection was achieved from the Ethics Committee of Guangdong Provincial People's Hospital (No.2018160A), and all patients provided informed consent.
Extraction exosomal RNAs from serum and library construction, sequencing and data analysis 2mL of venous blood from patients were collected into Ethylene Diamine Tetraacetic Acid (EDTA) routine blood tubes. Plasma was separated by centrifugation at 3000 rpm for 10 min at 4℃ and collected into cryogenic vials and stored at −80℃. The exoRNeasy Serum/Plasma kit (Qiagen) was applied to extract the exosomal total RNAs. According to the handbook of the kit, 1 mL plasma was used to isolate exosomal RNAs. SMART technology (Clontech) was applied to construct the RNA-seq libraries. Each RNA sample used for sequencing was performed on the illumine Nova-Seq 6000 System with technical support of Guangzhou Epibiotek Co., Ltd. HISAT2 was applied to align with the sequencing reads data [15]. GENCODE database was used to annotate mRNA and lncRNA. CircRNAs from unmapped reads were identi ed through Acfs [16].

Measurement of exosome characterization
The morphological characteristics of the exosomes were observed by transmission electron microscopy (TEM) and their size and distribution were measured by NanoSight NS500 (NanoSight Ltd, Amesbury, United Kingdom). Exosomal protein markers CD9 and CD63 were detected by western blotting.
Different exosomal long RNA analysis and functional enrichment analysis NetworkAnalyst (http://www.networkanalyst.ca), a visual analytics platform for comprehensive gene expression pro ling [17], was applied to identify different exoLRs between AMI and CAD/normal samples. RNAs with P-value < 0.01 and log 2 fold change ≥ |1| were considered as signi cantly different exoLRs.
Co-expression network of exosomal mRNA Exosomal mRNAs would be rstly ltered if the expression values were less than 1 FPKM in at least 90% samples. The remaining mRNAs with standard deviation greater than 0.2 were fed into an R package for weighted correlation network analysis [19].
Appropriate soft-threshold power was selected to ensure the co-expression network according to the scale-free topology. The weighted adjacencies and correlations were transformed into a topological overlap matrix (TOM), followed by calculating the corresponding dissimilarity (1-TOM). Next, 1-TOM as the distance measure was applied to a hierarchical clustering analysis of genes. A dynamic tree cut algorithm was used to identify modules. According to the correlations of module eigengene and clinical traits, signi cant modules were identi ed with P-value < 0.05. Top 50 exosomal mRNAs connections based on topological overlap in signi cant module were used to construct a network diagram using Cytoscape [20] and the exosomal mRNA with eigengene connectivity > 0.8 in the module related to clinical traits was considered as candidate hub genes [21].

Statistical analysis
Principal component analysis (PCA) was applied to evaluate the variables in the three groups with the R package. Continuous variables with normally distributed were displayed as mean with standard deviation (SD) and analyzed by the t-test. Otherwise, they would be displayed as medians and analyzed by the nonparametric test. Pearson's correlation coe cient was applied to correlations analysis between two variables. RNA expression levels were shown as mean with fragments per kilobase per million (FPKM).
The area under the curves (AUC) of receiver operating characteristic (ROC) was applied to evaluate the speci city and sensitivity of the exosomal mRNA for AMI diagnosis. The PASW Statistics 18.0 software was applied to all statistical analyses. P-value< 0.05 was considered statistically signi cant.

Clinical baseline characteristics of patients
The clinical characteristics of the three groups (10 AMI, 8 CAD, and 10 controls) are presented in Table 1. Acute in ammation levels between CAD group and control group were similar, whereas the AMI group had signi cantly higher acute in ammation levels at study entry. CKMB and HBDH were also higher in AMI patients than the rest subjects.
A brief view of the work ow of human plasma exosomal long RNA-seq and its characteristics in each group Reliable exosomal long RNA-seq data were obtained according to the process of plasma isolation, puri cation of exosomes, exosomes RNA extraction, and RNA-seq library construction ( Figure 1A). Transmission electron microscopy results showed membrane-enclosed structures of exosomes without similar sizes and uniform distribution in a dark background ( Figure 1B). The average diameter of the isolated exosomes was 75.83 nm measured by the NanoSight instrument ( Figure 1C). Membrane markers of exosomes CD9 and CD63 were con rmed by western blots ( Figure 1D). The mRNA constituted 58.46% of total mapped reads. Pseudogenes and circRNA accounted for 12.80% and 11.73%, respectively, whereas lncRNA and antisense RNA were only 7.94% and 7.55%, respectively ( Figure 1E). What's more, we found the numbers of mRNA, circRNA, lncRNA, and pseudogenes in the AMI group were all signi cantly higher than the control group ( Figure 1F).
Exosomal long RNA may re ect relative fractions of immune cell types Twenty-two types of immunocyte from exosomal sequencing data of each sample were investigated through an established computational resource (CIBERSORT) (Figure 2A). The PCA of the immunological pro le showed an un-uniform distribution ( Figure 4C). Memory B cells were the lowest in the CAD group, whereas naïve B cells of the CAD group were the highest. Neutrophils and monocytes were signi cantly enriched in the AMI group which were consistent with the clinical baseline characteristics. The results indicated that circulating exosomal long RNA may re ect the circulating immunological pro le.
Different exosomal long RNA analysis between the AMI and the control Compared to the control group, 296 different exosomal mRNAs consisting of 254 up-and 42 downregulated in the AMI group ( Figure 3A). These 296 mRNAs with signi cant differences were visualized in Figure 3B. However, only a few different lncRNAs or circRNAs were identi ed ( Figure 3C and 3D). To investigate the biological processes enrichment of different exosomal mRNAs, ClusterPro ler was used to analyze and visualize functional pro les (Gene Ontology) of 296 different exosomal mRNAs. Meanwhile, these different exosomal mRNAs were subjected to https://www.networkanalyst.ca for enrichment maps of biological processes. The results of functional pro les analysis were shown as a bar plot ( Figure 3E) and the enrichment map of biological processes was shown as a network ( Figure 3F). Among the top 10 biological processes in Figure 3E and Figure 3F, neutrophil degranulation, neutrophil activation, and neutrophil activation involved in immune response were signi cantly enriched. The enrichment map ( Figure 3F) showed that the in ammatory response might be the core biological processes.
Above all, through the analysis of different exosomal long RNA between the AMI and the control, we found that circulating exosomal mRNAs in the AMI group might play a crucial role in acute in ammation response mediated by neutrophils.
Different exosomal mRNA analysis between AMI and CAD When compared to the CAD group, 230 different exosomal mRNAs consisting of 120 up-and 110 downregulated in the AMI group ( Figure 3A). The results of functional pro les analysis ( Figure 4D) were shown that neutrophil activation played a leading role in biological processes. To further investigate the relationship between AMI and CAD, two intersections of differently up-or down-regulated exosomal mRNAs among AMI, CAD, and control groups were shown in Figure 4B and 4C. There were 35 different exosomal mRNAs (31 up-and 4 down-regulated) overlapped between the comparison of AMI and control and comparison of AMI and CAD. The results of gene ontology analysis of these 35 different exosomal mRNAs showed that myeloid leukocyte activation was mainly enriched ( Figure 4E). Besides, we found that ALPL was special since it was the only one up-regulated exosomal mRNA among the three comparison sets (Figure 4B), which might serve as a potential biomarker for AMI diagnosis.

Co-expression network analysis of exosomal mRNAs in AMI
To further investigate the union of genes related to AMI, we applied WGCNA(weighted gene co-expression network analysis) to clarify the key modules and the highly correlated exosomal mRNAs in AMI. 43 exosomal mRNAs modules were identi ed by the hierarchical clustering dendrogram ( Figure 5A). The association of the 43 co-expression modules was analyzed by the topological overlap matrix plot which consisted of the modules and the corresponding hierarchical clustering dendrogram ( Figure 5B). For module-trait analyses, only the light-yellow module containing 175 exosomal mRNAs was related to AMI ( Figure 5C). Although there was no signi cant enriched KEGG biological pathway in the light-yellow module, the results of GO enrichment analysis indicated that the light-yellow module was involved in in ammation respond mediated by neutrophils such as neutrophil aggregation and chemokine production ( Figure 5D). The association between the light-yellow module and AMI could be also supported by enrichment analyses of different exosomal mRNAs between AMI ( Figure 3E) and control or between AMI and CAD ( Figure 4D). To further analyze the core mRNA of light-yellow modules, the top 50 exosomal mRNAs of the topological overlap matrix in this module were used to construct a network diagram ( Figure 5E). In this network, the exosomal mRNA (with eigengene connectivity > 0.8) in the lightyellow module was considered as core mRNA and was labored with yellow ( Figure 5E). Thus, six mRNAs including ALPL, CXCR2, ELL2, EMC9, FAM129A, and DBF4B were identi ed as core mRNA.

Potential clinical value of exosomal ALPL and CXCR2
To estimate the potential clinical value of the six core exosomal mRNAs, ROC curve analysis was applied ( Figure 6A). The AUC indicated that there were two mRNAs with great predictive accuracy (ALPL (AUC: 0.99); CXCR2 (AUC: 0.98)). The correlation analysis between these two exosomal mRNAs and clinical baseline characteristics ( Figure 6B) indicated that ALPL and CXCR2 were associated with the neutrophil count (R=0.52, R=0.51, respectively). Statistical analysis of expression suggested that exosomal mRNA ALPL and CXCR2 from the AMI plasma were signi cantly higher than the other groups ( Figure 6C, 6D).
Besides, exosomal ALPL in the CAD group was also signi cantly higher than the control group. It might suggest that circulating exosomal ALPL would extremely increase with the progression of coronary plaque to acute myocardial infarction.
Since most results of the analysis of exosomal mRNAs in AMI patients mainly pointed to acute in ammation respond mediated by neutrophils, we tried to apply the PCA with neutrophil count and neutrophil ratio to identify the three groups. The PCA of neutrophil count and neutrophil ratio showed that although the AMI group could separate from the other groups, there was much overlap between CAD and control ( Figure 6E). However, when we applied PCA with the expression value of ALPL and CXCR2, the three groups could separate from each other ( Figure 6F). To sum up, besides the potential of great predictive accuracy, these two exosomal ALPL and CXCR2 might also be good for predictive identi cation.

Discussion
Recent years have witnessed the great advances in exosomal microRNAs in cardiovascular disease study but very few studies revealed the features and potential clinical value of exosomal mRNAs. In this study, we found that the plasma exoLRs largely consisted of mRNAs. The alteration of these mRNAs in AMI patients could indicate the neutrophilic in ammation of the circulatory system. Among these mRNAs, we found ALPL and CXCR2 with good predictive accuracy might be potential biomarkers for AMI diagnosis.
Our exoLRs sequencing data from this study suggested that the plasma exoLRs were mainly mRNAs. It is not surprising to nd the exosomal circRNAs and lncRNAs were expressed in low abundance since they had speci c spatiotemporal expression patterns [22][23][24]. A relatively high number of long RNAs were identi ed in AMI samples. On the one hand, the AMI samples in this study were from relatively aged patients. Previous studies demonstrated that aging was a profound factor in changes of circulating extracellular vesicles in concentration, size, and cargo [25,26]. On the other hand, stress including hypoxia, in ammation or injury could induce cardiomyocytes or other cells to secret exosomes [27].
To provide a better indication of the change exosomal mRNAs in the AMI patients, we applied bioinformatics approaches to analyze different exosomal mRNAs by comparing to CAD patients and healthy individuals. Besides, we also used co-expression network analysis to investigate the exosomal mRNAs in AMI. The results of our analyses all indicated that the acute in ammatory response mediated by neutrophils might be the core biological processes in the AMI patients. These ndings were consistent with the current AMI studies. Since the cardiomyocyte is extremely sensitive to ischemic injury, the reduced blood supply to the myocardium would initially cause the injury and further lead to an intense in ammatory response once the AMI occurred [28,29]. Furthermore, Neutrophil extracellular traps were identi ed as the crucial triggers and structural factors of various forms of thrombosis [30], besides, neutrophilic in ammation was found to in uence the infarct size, healing, and cardiac function after myocardial infarction [31]. Although the exact physiological role of exosome in AMI is still poorly understood, in ammation-related alterations in exosomal RNAs were associated with the biological processes of AMI.
Since miscellaneous immune cells would take parts in cardiac repair during different phases of cardiovascular disease [32], neutrophils were traditionally viewed as biomarkers [33]. Similarly, we found two exosomal mRNA ALPL and CXCR2 might be served as potential diagnostic biomarkers for AMI as they were speci cally enriched in circulating neutrophils which could be con rmed by the Human Protein Atlas (http://www.proteinatlas.org) [34].
The previous study indicated that ALPL could regulate the cardiac brosis during myocardial infarction through TGF-β1/Smads and P53 signaling pathways [35]. Similarly, the inhibitor of ALPL was found to be a potential target to treat cardiovascular disease by attenuating the arterial calci cation in a non-chronic kidney disease context [36]. Consistent with our study, we found the expression level of exosomal mRNA ALPL in the AMI samples was the highest, CAD the next and healthy individuals' samples the lowest. Whether exosomal mRNA ALPL would increase during the progression of coronary plaque to acute myocardial infarction still need to be further studied.
CXCR2, an intriguing biomolecule had dual effects on the myocardial ischemia-reperfusion injury for myocardial damage and cardioprotective effect [37,38]. Nonetheless, increasing evidence showed that CXCR2 might play a crucial cardioprotective role in myocardial infarction through enhancing myeloid progenitor production and upregulating cardiac adhesion molecules [31,39]. Our data showed that the circulating exosomal mRNA CXCR2 was a potential biomarker for AMI with high diagnostic e ciency but the mechanism of secreting the exosomal mRNA CXCR2 need to be further studied.
Until recently, circulating plasma exosomes were shown to interact with a variety of cell types and tissue.
However, the exact physiological role and function of plasma exosomal mRNAs are poorly understood. Whether the plasma exosomal mRNAs would be functional mainly depended on they were intact or fragments. In fact, some exosomal mRNAs were found to be full length and functional [40]. Nevertheless, further investigations of these exosomal mRNAs were still needed. Although few studies focused on the exosomal mRNA of AMI, it should be noted that these ndings still needed to be validated in prospective clinical trials as the sample size was not big.

Conclusions
In summary, our study explored the alteration of exoLRs in the AMI patients which might associate with the acute in ammatory response mediated by neutrophils. Moreover, we found that exosomal mRNAs ALPL and CXCR2 might be potentially useful for AMI diagnosis.        Potential clinical value of ALPL and CXCR2. A: The ROC curve analysis of six core exosomal mRNAs. B: The correlation analysis between these two exosomal mRNAs and clinical baseline characteristics. C-D: The statistical analysis of expression of ALPL and CXCR2 among AMI, CAD, and control group. Results are presented as the mean ± SD (*P < 0.05, **P < 0.01, ***P < 0.001) E: PCA of neutrophil count and neutrophil ratio. F: PCA of the expression of ALPL and CXCR2.