Plasma extracellular vesicle modulate immune cell transcriptional responses following acute myocardial infarction
Lewis Timms, Raphaella Ridley, Adam Braithwaite, Charlotte Hooper, Tanvir Rafe, Carolin Lewe, Sebastian Lemke, Divya Ganesh, Zeyu Chen, Tafadzwa Kufazvinei, Katherine A. Boden, Malene Møller Jørgensen, Rikke Bæk, Rebecca Dragovic, Keith M. Channon, Mohammad Alkhalil
TL;DR
The study shows that plasma extracellular vesicles containing miRNA-320b activate immune cells after heart attacks, potentially offering new ways to improve recovery.
Contribution
The paper identifies miRNA-320b in extracellular vesicles as a novel driver of immune cell activation following acute myocardial infarction.
Findings
Plasma EVs from MI patients are enriched in miRNA-320b, which promotes monocyte adhesion and pro-inflammatory gene expression.
Endothelial cells increase miRNA-320b in EVs during inflammation, linking it to immune cell priming.
EV-miR-320b targets are enriched in neutrophils and monocytes from MI patients, suggesting a role in immune modulation.
Abstract
Plasma extracellular vesicles (EVs) increase during acute myocardial infarction (MI), correlate with myocardial injury, and mobilize immune cells from the spleen to the circulation. These cells are transcriptionally activated even before tissue recruitment, yet the mechanisms driving this priming are unclear. We show that plasma EVs isolated at hospital presentation with MI are enriched in miRNA-320b. Endothelial cells upregulate miRNA-320b in EVs following inflammatory stimulation. Target gene pathway analysis revealed enrichment in adhesion and cytokine signaling. Endothelial EVs promoted monocyte adhesion and induced IL6 and TNF mRNA expression in macrophages while dampening cytokine secretion. RNA-sequencing of MI patient neutrophils and monocytes confirmed significant enrichment of miRNA-320b targets. Peripheral blood mononuclear cells treated with MI plasma EVs showed similar gene…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsExtracellular vesicles in disease · Cardiac Fibrosis and Remodeling · Atherosclerosis and Cardiovascular Diseases
Introduction
Extracellular vesicles (EVs) are lipid-bilayer-encased particles secreted by cells, mirroring selected characteristics of their originating cells by expressing specific surface protein markers and carrying payloads such as lipids, microRNAs, DNA, and metabolites.1^,^2 These EVs can be liberated into the peripheral blood, making plasma EVs highly heterogeneous and serving as biomarkers.3 Additionally, EVs can modulate the function of recipient cells both locally in tissue microenvironments and remotely in other organs,4 including interactions with blood leukocytes.5
EVs derived from a specific organ display distinct heterogeneity, representing the diversity of cells that work together to maintain organ-specific microenvironments to maintain proper function. The heart, which includes cardiomyocytes, endothelial cells, fibroblasts, immune cells, and smooth muscle cells, exemplifies such complexity. Following acute myocardial infarction (MI), there is a significant increase in the concentration of plasma EVs identifiable by EV markers like CD9, CD63, and CD81,6 and show an early enrichment of endothelial cell-associated markers,7^,^8 including vascular cell adhesion molecule-1 (VCAM-1).9^,^10
MI is also associated with a rapid increase in blood leukocytes, including independent elevations in blood neutrophils11 and monocyte populations.12 The concentration of these cells at the time of presentation to the hospital is prognostic of the severity of myocardial injury, cardiac function, and mortality risk.13^,^14^,^15 The concentration of plasma EVs and the magnitude of the peripheral blood immune response show a positive association following MI.9 Endothelial cell-derived EVs are elevated in patients prior to arrival at hospital following MI, and mediate the mobilization of splenic-reserves of neutrophils and monocytes to the peripheral blood, which is dependent on EV-associated VCAM-1.9^,^10
Shortly after MI, peripheral blood neutrophils9 and monocytes12 exhibit differential regulation of gene expression before being recruited to the injured myocardial tissue. This challenges the traditional notion that immune cell responses are solely driven by local signals within the ischemic heart, suggesting instead that transcriptional priming or programming for distinct biological pathways occurs prior to tissue recruitment. However, the mechanisms by which the injured myocardium regulates these immune cell remain poorly understood. This is critical because monocytes, in particular, exhibit distinct functions depending on their recruitment phase16: early-phase proinflammatory monocytes exacerbate injury, while late-phase anti-inflammatory monocytes aid in tissue repair.17 Balancing these functions is crucial, because suppressing proinflammatory responses lowers infarct size acutely following MI, but impairs subsequent healing and repair.18^,^19^,^20^,^21
This study investigates how plasma-derived EV-miRNAs impact the transcriptomes of immune cells during MI. We demonstrate that plasma EV and miRNAs from these plasma EVs are altered following MI and that these miRNAs likely modulate immune cell gene expression, enhancing cell adhesion and inflammatory cytokine pathways. Our findings suggest that plasma EVs play a crucial role in orchestrating the transcriptional immune responses and highlight the potential of using EV-derived miRNAs as therapeutic targets to manage myocardial injury and inflammation.
Results
Plasma EV characterization at presentation with MI and after a 3-month follow-up
Twenty-one patients were recruited who had no prior history of MI with an average age of 64 ± 10 years; 71% of the patients were male. The average troponin measurements were 16.66 ± 20.48 ng/L (inter-quartile range 28.7). One patient was diabetic, 8 patients were hypertensive, 8 patients were active smokers, and 6 patients had a history of ischemic heart disease as previously reported.22
EV markers CD9 (p = 0.0215), CD63 (p < 0.0266), CD81 (p = 0.0215), Annexin V (p = 0.0017), ALIX (p = 0.0061), TSG101 (p = 0.0105), and Flottilin-1 (p = 0.0479), and LAMP-2 (p = 0.0081) were significantly higher at presentation with MI versus follow-up (Figure 1A). Simultaneously, we probed different cell-associated markers, representing endothelial cells, platelets, immune cells, muscle and structural proteins, adipocytes, plasma lipoproteins, and metabolic markers. EpCAM (p = 0.0448), CD142 (p = 0.0448), CD141 (p = 0.0448), p53 (p = 0.0305), CD151 (p = 0.0569), thrombospondin-1 (p = 0.015), CD162 (p = 0.0395), GLUT2 (p = 0.0182), GLUT4 (p = 0.0182), and FATP1 (p = 0.0174) were significantly higher at presentation with MI versus the follow-up timepoint (Figure 1B). The EV-Array does not determine the size and concentration of plasma EV, therefore we applied platelet-poor plasma to size exclusion chromatography (SEC) columns and combined EV enriched fractions for subsequent ultra-centrifugation as previously described.23 Plasma EV number and size distribution profiles were assessed by nanoparticle tracking analysis (NTA) measurements and showed a 1.2-fold difference in their total concentration (p = 0.0556) at time of presentation to hospital with MI versus the 3-month follow-up timepoint, but no overall difference in their size (Figures 1C and 1D). Isolated plasma EV was positive for EV markers Syntenin-1 and CD9. Plasma apolipoprotein B not was not present in isolated plasma EV, and they were negative for cellular GM130, mitochondrial ATP5A, and nuclear histone H3 (Figures 1E and S1). Isolated plasma EV displayed a typical cup-shaped morphology by transmission electron microscopy (TEM) (Figure 1F).Figure 1. Characterization of plasma EV in MI patientsEVs were isolated from the plasma of MI patients at hospital presentation and at 3-month follow-up.(A) EV-Array analysis of canonical EV-associated markers shows significant enrichment at MI presentation (n = 17 patients paired at both timepoints).(B) EV-cell-associated markers were also enriched at presentation with MI. White in the heatmaps denotes the baseline (normalized to 1), corresponding to the 3-month follow-up value.(C) Total EV concentrations measured by NTA of SEC and ultracentrifugation-purified plasma EVs.(D and E) Spaghetti plots showing EV size and particle concentration profiles at the individual patient level across the two timepoints (n = 19 patients paired at both timepoints). (E) Western blotting confirmed the presence of EV markers (CD9 and Syntenin-1), absence of cellular contaminants (GM130, ATP5A, and Histone H3), and depletion of the plasma Apo B.(F) TEM images confirmed typical cup-shaped EV morphology. Scale bars, 500 nm or 100 nm.(G and H) Single-particle interferometric reflectance imaging (NanoView) confirmed tetraspanin-positive EVs (CD9, CD63, and CD81), (H) with associated size and concentration profiles (n = 9–10 at both timepoints). EV marker values at MI presentation were normalized to each patient’s own 3-month follow-up value. In (G), data are presented as group means ± SD. Statistical analysis: (A–C) Wilcoxon matched-pairs signed-rank test; (G) two-way ANOVA with Tukey’s post hoc test. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.
NTA assessment of isolated plasma EV size and concentration is prone to artifacts by protein aggregates, lipoproteins, and NTA can overestimate the size of EVs and does not provide information on tetraspanin-expressing populations. Using single particle interferometric reflectance imaging sensing, we show that these plasma EV isolations were positive for CD9, CD63, and CD81 (Figures 1G and 1H). However, there were no differences between the number of CD9, CD63, and CD81 EV populations between presentation with MI and the follow-up timepoint, or the size and concentration profiles at each timepoint (Figures 1G and 1H). Several EV markers associated with endothelial cells (CD141 and CD151), coagulation and platelet activation (CD142, Thrombospondin-1, and CD162), epithelial activation (EpCAM), cellular stress (p53), and metabolic regulation (GLUT2, GLUT4, and FATP1) were significantly elevated in plasma EVs at the time of MI presentation compared to the 3-month follow-up.
Plasma EV-miRNAs are enriched at presentation with MI versus a 3-month follow-up
We isolated EV-RNAs and undertook a miRNA-array, which allowed systematic profiling of over 30,000 miRNA including human pre-miRNAs, mature miRNAs, snoRNAs, and scaRNAs. To account for patient-to-patient variation and treatment effects, we used a mixed effects model to determine significantly altered RNAs between presentation with MI and the follow-up timepoint, accounting for age, family history of cardiovascular disease, smoking status, and array quantification batch effects. One hundred forty-five significantly altered RNAs were detected at presentation to hospital with MI versus the 3-month follow-up; p values were adjusted for multiple comparisons (p < 0.05) and had a >1 log_2_ fold change (Figure 2A). miR-320b showed the greatest increase (7.35-fold change; adjusted p = 0.011), whereas miRNA-101-5p showed the greatest decrease at presentation (3.79-fold change; adjusted p = 0.006).Figure 2. Small RNA analysis of plasma EV in MI patients(A) Volcano plot illustrating differentially expressed small RNAs (miRNAs, snoRNAs, etc.) in plasma EVs from MI patients at presentation with MI compared to 3 months later (n = 10 patients paired at both timepoints). Non-significant (NS) RNAs are shown in gray, with log2 fold change (FC) in green, p value < 0.05 in blue, and adjusted p value < 0.05 with log2 FC in red.(B) TissueAtlas expression analysis for miRNA-101-5p.(C) TissueAtlas expression analysis for miRNA-320b, suggesting potential cardiovascular endothelial cell origin.(D) RT-qPCR analysis of miRNA-126-3p, miRNA-320b, and miRNA-101-5p in HUVECs cell pellets normalized to miR-103a-3p, confirming endothelial association. Data are group means ± SD. N = 8 biological replicates per group.
To infer the likely source of plasma EV miRNA-101-5p and miRNA-320b, we used TissueAtlas expression scores.24 miRNA-101-5p showed enrichment within veins (Figure 2B), whereas miRNA-320b showed enrichment within artery and the myocardium (Figure 2C), suggesting cardiovascular endothelial cell origin of these EV-miRNAs and in agreement with our protein analysis of plasma EV.9^,^10 We used human umbilical cord vein endothelial cells (HUVECs) to determine the presence of miRNA-101-5p and miRNA-320b. We used miRNA-126-3p as a known endothelial cell associated miRNA25 to determine the relative abundance of miRNA-101-5p and miRNA-320b in HUVECs. miRNA-320b was detected in HUVECs whereas miRNA-101-5p was not detected; therefore, we focused subsequent analysis on miRNA-320b.
miRNA320b is enriched in inflammatory endothelial cell-derived EV
To determine whether miRNA-320b was expressed in endothelial cell-derived EV and enriched following proinflammatory stimulation, we used a model of endothelial cell activation in vitro. HUVEC total, live, and dead cell numbers between control and TNF-α stimulated conditions were consistent at the end of the stimulation period (Figure 3A), but TNF-α activated cells released significantly more soluble VCAM-1 into cell culture supernatants (p < 0.001) confirming proinflammatory cell stimulation (Figure 3B). Single particle interferometric reflectance imaging sensing showed CD9, CD63, and CD81 positive EV in HUVEC culture supernatants (Figure 3C), with significant enrichment for CD63 and CD9 in TNF-α activated cells (Figure 3D). CD9, CD63, and CD81 HUVEC-derived EV from control and TNF-α culture had similar average particle sizes (Figure 3E). In agreement with previous studies, we show a significant 2.9-fold (p < 0.001) increase in isolated EV from TNF-α stimulated cells (Figure 3F), with a modal increase in small EV (Figure 3G). Small differences in cell number may contribute substantial differences to total numbers of EV. However, the differences in EV count between control and TNF-α conditions remained significant when normalized by cell count (Figure 3H). Isolated HUVEC-derived EV was positive for EV markers syntenin-1, CD9, TSG101 and negative for markers of cellular contamination by GM130. TNF-α-derived HUVEC EV was positive for VCAM-1 (Figure 3I) (Figure S2). Isolated HUVEC-derived EV displays typical cup-shaped morphology by TEM (Figure 3J). TNF-α stimulation did not alter cellular levels of miRNA-320b (Figure 3K); however, isolated EV from TNF-α stimulated HUVECs shows a significant enrichment of miRNA-320b (Figure 3L). miRNA-320b is enriched in endothelial cell-derived EV following inflammatory stimulation.Figure 3. Enrichment of miRNA-320b in endothelial cell-derived EV following inflammatory stimulationHUVECs were cultured under control (black) or TNF-α-stimulated (red) conditions to model endothelial inflammation.(A) Total and viable cell counts at study endpoint confirm similar cell numbers across conditions.(B) Soluble VCAM-1 levels in supernatants were significantly increased following TNF-α stimulation, confirming proinflammatory activation.(C and D) Single-particle interferometric reflectance imaging (NanoView) detected CD9, CD63, and CD81-positive EVs in both conditions. Scale bars, 10 μm. (D) Total captured EV concentrations were significantly higher in TNF-α-stimulated cultures.(E) Average particle diameters of captured EVs did not differ significantly.(F) NTA showed elevated particle concentrations in EV preparations from TNF-α-treated HUVECs.(G) Size distribution profiles of EVs from control and TNF-α conditions.(H) EV counts were normalized to total and live cell numbers, confirming increased release under inflammatory stimulation.(I) Western blot analysis of HUVEC-derived EVs confirmed enrichment of canonical EV markers (TSG101, Syntenin-1, and CD9), absence of cellular contaminant GM130, and presence of VCAM-1 specifically in TNF-α-derived EVs.(J) TEM images showed characteristic EV morphology; scale bars, 1 μm or 500 nm.(K) RT-qPCR showed no change in HUVEC cellular miRNA-320b levels between conditions.(L) Significant enrichment of miRNA-320b was detected in EVs from TNF-α-stimulated HUVECs. miRNA expression was normalized to UniSp6 spike-in control, because endogenous controls (e.g., miR-103a-3p) were undetectable in EV samples. Data are presented as group means ± SD. N = 4 biological replicates per group. Statistical analysis: (B, F, L) unpaired Student’s t test; (D, H) two-way ANOVA with Tukey’s post hoc test. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.
Identification of miR-320b-mRNA targets and gene ontology pathway analysis
miRNA-putative mRNA targets can be extracted from open sources databases. We evaluated the likely consequences of EV-enriched with miRNA-320b on recipient cells by obtaining putative mRNA targets of miRNA-320b from miRWalk, Targetscan, and miRDB. By using three databases, we enhanced the scope of our analysis, by removing duplications from the individual lists for mRNAs that appear more than once we enhance the stringency of our analysis and found an overlap of 375 mRNA targets (Figure 4A). We next determined the likely target biological processes of these putative mRNA targets from each dataset individually and the intersect of the three databases using gene ontology (GO) biological pathway analysis, with a false discovery rate (FDR) < 0.05 and p < 0.05. Five GO biological process pathways were common among all datasets (Figure 4B), including: homophilic cell adhesion via plasma membrane adhesion molecules (GO:0007156) and cell-cell adhesion via plasma-membrane adhesion molecules (GO:0098742). Suggesting that enriched miRNA-320b EV of endothelial cell origin may target cell adhesion in recipient cells.Figure 4. Identification of miRNA-320b mRNA targets and pathway analysis(A) Venn diagram showing the overlap of predicted miRNA-320b targets identified using miRWalk, TargetScan, and miRDB databases.(B) GO biological process analysis of overlapping miRNA-320b targets revealed significant enrichment of pathways related to homophilic cell adhesion via plasma membrane adhesion molecules and cell-cell adhesion.(C) Functional validation of miRNA-320b EV activity in a monocyte adhesion assay. HUVEC-derived EVs (from TNF-α-stimulated cells) enriched in miRNA-320b significantly enhanced THP-1 monocyte adhesion to TNF-α-activated endothelial cell monolayers compared with control EVs. Representative fluorescence microscopy images are shown. Scale bars, 20 μm. Data are presented as group means ± SD. N = 4–5 biological replicates per group. Statistical analysis: (C) one-way ANOVA with Tukey’s post hoc test. ∗∗∗∗p < 0.0001.
Inflammatory endothelial EV enhance monocyte adhesion to activated endothelial cells
Peripheral blood neutrophils and monocytes are among the first cells to arrive at sites of tissue injury by extravasation, which is mediated by cell adhesion molecule binding to activated blood vessels for subsequent diapedesis. Monocytes express very late antigen-4 (VLA-4) on their surface, which enables binding to the glycoprotein VCAM-1 expressed on the surface of activated endothelial cells. We determined the propensity of HUVEC-derived EV, rich in miRNA-320b to modulate THP-1 monocyte binding to activated monolayers of endothelial cells. Overnight incubation of monocytes with EV from control (p < 0.0001) and TNF-α stimulated (p < 0.0001) HUVEC-derived EV showed enhanced affinity to bind to activated endothelial cells versus untreated cells, and this was more pronounced in TNF-α-derived EV than control EV (p < 0.0001) (Figure 4C). We overexpressed miRNA-320b in THP-1 monocytes using lentiviral transduction to determine the effects of miRNA-320b versus a scramble control in isolation of EV. Scramble and miRNA-320b displayed strong transduction using enhanced green fluorescent protein (eGFP) linked to a T2A peptide sequence (Figure S3A). miRNA-320b overexpression was confirmed by RT-qPCR versus scramble controls (Figures S3B and S3C). miRNA-320b overexpressing monocytes adhered to activated HUVEC monolayers less abundantly than scramble control cells (p < 0.05) (Figure S3D). Similarly, transduction of miRNA-320b in HUVECs led to a strong expression of eGFP (Figure S4A) and overexpression of miRNA-320b versus scramble controls (p < 0.01) (Figures S4B and S4C). miRNA-320b HUVEC produced more soluble VCAM-1 in their cell culture media versus scramble cells under basal conditions, but exhibited similar responses when challenged with recombinant human TNF-α (Figure S4D). These data demonstrate that inflammatory endothelial cell-derived EV enhance the propensity of monocytes to transiently bind to activated endothelial cells and that miRNA-320 may contribute to inflammatory activation in endothelial cells.
miRNA-320b mRNAs are enriched in immune cell transcriptomes following MI
MI induces peripheral blood neutrophil and monocyte transcriptional activation and enrichment of genes for distinct biological pathways prior to tissue recruitment.9^,^12 We determined whether miRNA-320b putative mRNA targets from all databases (miRDB, miRWalk, and TargetScan) and the overlap were enriched in monocyte transcriptomes following MI. Peripheral blood monocytes were obtained at presentation to hospital with MI and compared for differentially expressed genes (DEG) versus monocytes obtained from the same patient 48 h following MI and versus stable coronary artery disease patients.12
Monocytes from MI patients arriving at hospital show 509 DEG versus stable control patients (p < 0.05), whereas the same patient’s peripheral blood monocytes 48 h post-MI show 227 DEG (versus stable control patients, p < 0.05). An additional 238 DEG are enriched in MI patients following presentation to hospital versus 48 h post-MI (p < 0.05).12
Monocyte DEG from MI patients following presentation to hospital significantly intersected with miRWalk miRNA-320b mRNAs (358 mRNA targets, p = 4.15 x 10^−11^). Similarly, miRNA-320b mRNA targets intersected with DEG in peripheral blood monocytes 48 h post-MI in the same patients (versus presentation); however, only 154 mRNAs were common to miRWalk (p = 2.02 x 10^−7^). This suggests that miRNA-320b mRNA targets are present in DEGs from peripheral blood monocytes following MI and may regulate more target genes in the immediate hours post-MI.
Second, we determined whether the DEG in monocytes following MI and the miRNA-320b-mRNAs collectively favored similar biological pathways by undertaking GO biological pathway analysis for each dataset individually and looked for common pathways among both datasets (FDR < 0.05 and p < 0.05). DEG in post-MI peripheral blood monocytes and the miRNA-320b-mRNA putative targets from miRWalk showed similarity in 29 biological processes (GO biological pathway analysis, FDR < 0.05 and p < 0.05), including: positive regulation of cytokine production (GO:0001819), negative regulation of protein phosphorylation (GO:0001933), pattern recognition receptor (GO:0002221), signaling pathway for myeloid cell differentiation (GO:0030099) and myeloid leukocyte activation (GO:0002274) (Figure 5A).Figure 5. Enrichment of miRNA-320b-mRNA targets in immune cell transcriptomes following MI(A and B) GO biological pathway analysis showing the overlap between miRNA-320b-mRNA targets and DEG in: (A) peripheral blood monocytes following MI, highlighting pathways such as positive regulation of cytokine production and myeloid leukocyte activation; and (B) in peripheral blood neutrophils, showing overlap in biological pathways with miRNA-320b-mRNA targets.(C) Comparison of enriched GO biological pathway terms in monocyte and neutrophil transcriptomes with miRNA-320b-mRNA targets, identifying common pathways such as cytokine production and cell adhesion.
Monocytes are not the first cells to mobilize to the peripheral blood or to arrive at the injured heart following MI. Therefore, we determined whether neutrophils obtained from MI and control (NSTEMI) patients similarly showed interactions with miRNA-320b mRNA targets through DEG intersect analysis following presentation to hospital (versus a 1-month follow-up control sample) from our previously published study.9 MI peripheral blood neutrophils show 924 DEG versus 8 genes from controls. MI DEG significantly intersected with miRWalk-320b mRNA targets (756 mRNAs, p = 1.24 x 10^−5^) and control DEG significantly intersected with miRWalk-320b mRNA targets (4 mRNAs, p = 0.0033).
However, while DEG from neutrophils obtained from MI patients showed an overlap in 118 pathways versus miRWalk miRNA-320b-mRNA targets (GO biological pathway analysis, FDR < 0.05 and p < 0.05) (Figure 5B), control DEG neutrophils showed no overlap with miRNA-320b mRNA targets from any of the databases (miRWalk, Targetscan, and miRDB) or the overlap. MI DEG and miRWalk miRNA-320b-mRNA targets favored pathways for: temperature homeostasis (GO:0001659), positive regulation of cytokine production (GO:0001819), pattern recognition receptor signaling pathway (GO:0002221), myeloid cell homeostasis (GO:0002262), cell activation involved in immune response (GO:0002263), and myeloid leukocyte activation (GO:0002274).
Go biological pathway analysis of MI DEG monocyte and neutrophil genes showed some similarities, and therefore, we hypothesized that miRNA-320b may have similar effects on different cell types. We next compared the significantly (FDR < 0.05 and p < 0.05) enriched GO pathway terms which intersected with miRNA-320b-mRNAs and the DEG from MI peripheral blood monocytes and neutrophils to determine the most abundant pathways. We found myeloid leukocyte activation (GO:0002274) and positive regulation of cytokine production (GO:0001819) as the most frequent terms (Figure 5C). Homophilic cell adhesion via plasma membrane adhesion molecules (GO:0007156) and leukocyte cell-cell adhesion (GO:0007159) were among the second most common pathways. Demonstrating that miRNA-320b-mRNAs are significantly overrepresented in human peripheral blood monocytes and neutrophil transcriptomes following MI and that there is enrichment in pathway activation for cytokine production and cell adhesion in two independent cell types, which contribute to MI proinflammatory injury.
We systematically evaluated the directionality of gene regulation in monocytes and neutrophils using RNA-seq datasets from MI patients.9^,^12 We cross-referenced these criptomes with miRNA-320b-predicted target genes using the three established miRNA-mRNA target datasets.
We observed that only 3%–9% of miRNA-320b-mRNA targets were upregulated in patient cells following MI. miRWalk-320b-mRNA predictions in STEMI neutrophils were only 3% for positively intersecting targets, while 97% were unchanged or downregulated. Similarly, in monocytes, 9% of targets were upregulated, while 91% were not. In TargetScan-predicted targets, just 7% of targets were positively regulated in STEMI neutrophils (Table S1). This directionality is consistent with canonical miRNA-mediated repression of gene expression in recipient immune cells.
Inflammatory endothelial EV induces transcriptional activation but suppresses supernatant cytokines in monocyte-derived macrophages
Since cytokine production emerged as potentially regulated pathway in multiple cell types in GO pathway analysis, we next determined experimentally whether endothelial cell-derived EV from control and TNF-α stimulated conditions impact immune cell cytokine generation. We exposed primary human peripheral blood monocyte-derived macrophages to endothelial cell-derived EV for 24 h prior to subsequent proinflammatory M1-stimulation with lipopolysaccharide and recombinant human IFN-γ. Inflammatory endothelial cell EV significantly enhanced IL-6 mRNA (versus control EV, p < 0.01) (Figure 6A) and elevated TNF mRNA (versus control, p < 0.001) (Figure 6B) 6 h following M1-stimulation. However, protein levels of IL-6 (versus control EV, p < 0.0001) (Figure 6C) and TNF-α (versus control EV, p < 0.0001) (Figure 6D) were significantly blunted by inflammatory-derived EV 18 h after M1-stimulation. THP-1 cells overexpressing miRNA-320 versus scramble control cells show significantly elevated expression of IL-6 (p < 0.01) and TNF-α (p < 0.001) in cell culture supernatants when challenged with M1-stimulation with lipopolysaccharide and recombinant human IFN-γ (Figures S5A and S5B, respectively). Inflammatory activated endothelial cell-derived EV enriched with miRNA-320b augment transcriptional activation of proinflammatory cytokine mRNAs, but suppressed the quantity of soluble proinflammatory cytokine protein in cell culture supernatants.Figure 6. Impact of inflammatory endothelial cell-derived extracellular vesicles (EC-EVs) on monocyte-derived macrophage cytokine productionHuman CD14^+^ monocytes were differentiated into macrophages, pretreated with EC-EVs from control or TNF-α-stimulated HUVECs, and subsequently activated under M1 polarizing conditions (LPS and IFN-γ).(A and B) RT-qPCR analysis showed significantly increased IL6 and TNF mRNA expression in macrophages treated with inflammatory EC-EVs (red) compared to untreated controls (black) (N = 4 biological donors per group).(C and D) In contrast, ELISA of cell culture supernatants revealed reduced secretion of IL-6 and TNF-α protein in TNF-α-EC-EV-treated macrophages under the same M1 conditions (N = 6 biological donors per group). This discrepancy suggests post-transcriptional modulation of cytokine production by inflammatory EC-EVs. Data are presented as group means ± SD. Statistical analysis: two-way ANOVA with Tukey’s post hoc test. ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.
MI peripheral blood monocytes show enrichment for cell adhesion molecules and MAP kinases
To determine the key cellular pathways involved in early monocyte activation following MI in patients and their response to subsequent proinflammatory stimulation, we obtained peripheral blood monocytes from a cohort of patients presenting to hospital with STEMI (N = 3) and used control (NSTEMI) patients (N = 3). Monocytes were obtained prior to percutaneous coronary intervention (PCI) therapy and subject to bulk mRNA sequencing to determine alterations in peripheral blood monocyte transcriptomes. STEMI patient monocytes showed differential enrichment of 82 genes versus NSTEMI control (Figure 7A), which favored biological processes for: response to oxygen levels (GO:0070482) (p = 2.57 × 10^−6^), response to hypoxia (GO:0001666) (p = 2.18 × 10^−6^), positive regulation of cell adhesion (GO:0045785) (p = 1.8210^−6^), cell-cell adhesion (GO:0045785) (p = 1.82 × 10^−6^), positive regulation of cytokine production (GO:0001819) (p = 0.0017), regulation of MAP kinase activity (GO:0043405) (p = 0.0044), and regulation of mononuclear cell proliferation (GO:0032944) (p = 0.0025). We intersected the miRNA-320b mRNA targets from miRWalk, TargetScan, miRDB with these 82 genes DEG in MI patient monocytes and found no significant interactions using a Fishers exact test (p = 0.0511, p = 0.2333, and p = 1, respectively) (Figure 7B).Figure 7. Modulation of monocyte response to EC-EVs(A) MA plot showing DEG in monocytes from MI patients compared to controls, highlighting significantly altered genes in red.(B) Venn diagram showing the intersection of miRNA-320b-mRNA targets with DEG in MI monocytes.(C and D) Comparison of DEG profiles in M1-stimulated monocytes and (D) GO biological pathway from MI and control patients. N = 3 patients per group/condition.
MI peripheral blood monocytes are less responsive to M1-inflammatory activation
We hypothesized that gene expression changes in peripheral blood monocytes from MI patients would display alterations to M1-stimulation with lipopolysaccharide and recombinant human IFN-γ, similar to our in vitro monocyte-derived macrophages exposed to endothelial cell-derived EV. M1-stimulated control patient monocytes showed an induction of 5468 genes versus 3688 genes in M1-stimulated STEMI monocytes (Figure 7C). 2936 genes were common among both STEMI and control M1-stimulated monocytes (p < 0.001); however, 2249 DEG were unique to control and 623 genes unique to STEMI M1-stimulated monocytes (Figure 7C). GO biological pathway analysis (FDR < 0.05 and p < 0.05) showed 971 pathways that were common between both STEMI and control M1-stimulated monocyte groups and included pathways for: response to lipopolysaccharide (GO:0032496) (p = 3.85 × 10^−24^), positive regulation of cytokine production (GO:0001819) (p = 4.17 × 10^−18^), and regulation of cell-cell adhesion (GO:0022407) (p = 4.06 × 10^−15^) (Figure 7D).
Control M1-stimulated monocytes had 522 unique pathways targeting: ATP metabolic process (GO:0046034) (p = 4.25 × 10^−6^), nucleoside triphosphate metabolic process (GO:0009141) (p = 7.48 × 10^−6^), and ribonucleoside triphosphate metabolic process (GO:0009199) (p = 1.67 × 10^−6^). Whereas STEMI M1-stimulated monocytes had 285 unique pathways targeting: cellular extravasation (GO:0045123) (p = 0.0001), regulation of cytosolic calcium ion concentration (GO:0051480) (p = 0.0003), and fatty acid biosynthetic process (GO:0006633) (p = 0.0012) (Figure 7D). Demonstrating diffrential gene activation of STEMI monocytes to MI-stimulation.
MI plasma EV activate PBMC transcriptomes
We therefore hypothesized that plasma EV regulate peripheral blood immune cell gene expression profiles prior to recruitment to the injured myocardium to favor cytokine regulation and cell adhesion. To determine the influence of plasma EV on peripheral blood immune cells in isolation, we obtained peripheral blood mononuclear cells (PBMCs) from healthy male volunteers and exposed them to plasma EVs from STEMI and control (NSTEMI) patients following presentation to hospital and undertook bulk mRNA sequencing.
Control plasma EV induced changes in 67 DEG (Figure 8A) versus 2,498 DEG (Figure 8B**)** by STEMI plasma EV, 35 genes were common among both control and STEMI plasma EV (Figure 8C). MI plasma EV targeted multiple pathways including: positive regulation of cytokine production (GO:0001819) (p = 3.29 x 10^−17^), mononuclear cell differentiation (GO:1903131) (p = 8.43 x 10^−11^), leukocyte migration (GO:0050900) (p = 8.43 x 10^−11^), and positive regulation of cell adhesion (GO:0045785) (p = 1.99 x 10^−9^). We then accounted for differences driven by PBMC donor and the addition of plasma EV to immune cells, which may elicit a generic response by comparing MI versus control plasma EV relative to their control groups and found 684 DEG (Figure 8B). To determine whether these induced transcriptional effects in PBMCs were specific to the time of presentation with MI to hospital, we exposed control and MI plasma EV from the same patients to the same PBMC donors, but from the 3 months post-MI timepoint and found 92 (Figure 8D**)** and 62 DEG (Figure 8E**)**, respectively. Demonstrating that plasma EV from the time of presentation following MI induce distinct transcriptional activation of PBMCs.Figure 8. Transcriptomic activation of PBMCs by plasma EVs from MI patients(A and B) MA plot showing DEG in PBMCs treated with plasma EVs from (A) control (NSTEMI) patients and (B) MI patients, red dots are significantly altered genes.(C and E) Venn diagram showing the overlap of DEG induced by control and MI plasma EVs in PBMCs. MA plot showing DEG in PBMCs treated with plasma EVs from (D) control (NSTEMI) plasma EV and (E) MI patient plasma EV 3 months post-MI.(F–I) GO biological pathway analysis comparing pathways enriched by MI plasma EVs at presentation, 3 months post-MI, and shared pathways. MA plot showing DEG in PBMCs treated with (G) control plasma EVs and (H) MI patient plasma EV at presentation and (I) control (NSTEMI) plasma EV and (J) MI patient plasma EV 3 months post-MI followed by M1-stimulation. N = 5 PBMC donors per group.
We then compared the DEG by Geno Ontology-biological pathway analysis (FDR < 0.05 and p < 0.05) and found 501 pathways, which were unique to STEMI plasma EV at presentation, 64 pathways that were unique to 3 months post-MI and 89 shared pathways (Figure 8F). STEMI plasma EV unique terms favored cytokine-mediated signaling pathway (GO:0019221) (p = 9.69 x 10^−17^), whereas control plasma EV at 3 months post-MI favored pathways for antigen processing and presentation of peptide antigen (GO:0048002) (p = 7.82 x 10^−18^). Thereby demonstrating that STEMI plasma EV obtained at the time of presentation to hospital with MI drive specific immune cell transcriptional activation for cytokine synthesis and cell adhesion.
MI plasma EV modulate proinflammatory stimulation in PBMCs
We further hypothesized that prior exposure of immune cells to plasma EVs from STEMI patients would alter inflammatory activation to subsequent M1-stimulation. Exposing PBMCs to LPS and IFN-ꝩ induced significant alteration in gene expression profiles by upregulating 4336 genes and downregulating 4393 genes (Figure S3) and favored pathways for: cytokine-mediated signaling pathway (GO:0019221) (p = 1.86 x 10^−25^), positive regulation of cytokine production (GO:0001819) (p = 1.71 x 10^−17^), myeloid leukocyte activation (GO:0002274) (p = 1.88 x 10^−22^), positive regulation of leukocyte cell-cell adhesion (GO:1903039) (p = 1.53 x 10^−19^), and positive regulation of cell-cell adhesion (GO:0022409) (p = 9.65 x 10^−19^).
Prior exposure to control plasma EV showed differential regulation of 44 genes (Figure 8G), whereas prior exposure of STEMI plasma EV altered 1830 genes (Figure 8H), showing that STEMI plasma EV drive differential activation of PBMC transcriptomes. Pre-exposure of control (Figure 8I) or STEMI (Figure 8J) plasma EV from 3 months post-MI from the same patients showed 124 and 63 DEG, respectively. Presentation STEMI plasma EV exposed to PBMC prior to M1-stimulation targeted numerous pathways, including positive regulation of cell adhesion (GO:0045785) (p = 0.006) and positive regulation of cytokine production (GO:0001819) (p = 0.0007). Revealing that prior exposure of PBMCs to STEMI plasma EV from the time of presentation impacted transcriptional capacity to M1-stimulation.
miRNA-320b-mRNA targets are overrepresented in immune transcriptome activated by plasma EV at the time of MI
To determine whether miRNA-320b-mRNA targets were overrepresented against the DEG from STEMI plasma EV-treated PBMCs, we undertook a Fishers exact test. miRNA-320b-mRNA targets were significantly overrepresented in STEMI plasma EV-induced DEG in PBMCs when compared against miRDB (p = 3.03 x 10^−8^), TargetScan (p = 3.06 x 10^−7^), and the overlap mRNAs (p = 3.39 x 10^−6^). Control plasma EV only significantly overlapped with miRWalk (p = 0.0035), and there were no significant overlaps between miRNA-320b-mRNA targets and STEMI and control plasma EV DEG 3 months post-MI. Showing that miRNA-320b mRNA targets are more differentially regulated by STEMI plasma EV at the time of presentation with MI.
Discussion
In this study, we investigated the role of plasma EVs in acute MI patients and their potential impact on immune cell transcriptional activation. Our key findings reveal that: there were (I) significant changes in plasma EV characteristics for EV-makers: between presentation with MI and a 3-month follow-up control sample and there were likely cellular contributions from endothelial cells. (II) There were significant differences in the miRNA cargo of these plasma EVs. Specifically, miRNA-320b was significantly enriched in plasma EVs at MI presentation compared to follow-up. (III) miRNA-320b is present in endothelial cells and enriched into endothelial cell-derived EVs following TNF-α stimulation, supporting the role of miRNA-320b in vascular inflammation. (IV) Endothelial cell-derived EVs: induce adhesion of monocytes to activated endothelial cells, and this is augmented by TNF-α-activated endothelial cell-derived EVs; (V) mediate proinflammatory transcriptional changes for IL6 and TNF, but impede the release of these proinflammatory proteins. (VI) miRNA-320b targets are significantly enriched in neutrophil and monocyte transcriptomes in patients following MI. (VII) MI-derived monocytes show an attenuated transcriptional capacity in response to M1-stimulation, but significant enrichment for miRNA-320b-mRNA targets. (VIII) Exposing PBMCs to plasma EV from MI patients induces greater transcriptional activation at the time of presentation versus a 3-month follow-up plasma EV samples from the same patients and enriches miRNA-320b-mRNA targets. (IX) PBMCs exposed to MI plasma EVs show differential regulation of their transcriptomes when subsequently challenged with an M1-stimulus with enrichment for miRNA-320b-mRNAs.
Our analysis demonstrated that the plasma EV profile following presentation to the hospital with MI is enriched for EV consistent with previous studies.6^,^9^,^10^,^26 However, the overall particle concentration of plasma EVs, determined by NTA, was not strikingly different in this cohort of patients. This is likely because patients in previous cohorts were recruited to the Oxford Acute Myocardial Infarction (OxAMI) Study, where blood samples are collected systematically following MI presentation to the hospital, within ∼3–4 h of chest pain9^,^10 and blood samples are obtained prior to PCI; Whereas, in this study, blood for plasma EV analysis was obtained following PCI and at a variable times from chest-pain to presentation. These differences are also likely contributors to the alterations observed in EV-composition. We have previously shown a high abundance of endothelial cell-derived EVs following hospital presentation with MI prior to PCI and a distinct pattern for immune cell-enriched EVs immediately following PCI.9 We also find here a high abundance of endothelial cell associated EVs (CD142, CD141, CD151, and thrombospondin-1), which are likely to originate from activated endothelial cells.
miRNA-320b is widely reported in cancer progression, where it contributes to cell proliferation and metastasis.27^,^28 Recent studies have expanded the relevance of miRNA-320b in the context of cardiovascular and inflammatory diseases. miRNA-320b modulates cholesterol efflux in macrophages by targeting genes involved in lipid metabolism. miRNA-320b increases atherosclerotic lesions, lesional macrophage numbers, and proinflammatory cytokines release through phosphorylation of NF-kB.29 miRNA-320b plasma concentrations are associated with deep venous thrombosis (DVT), and when combined with D-dimer it improves the diagnostic accuracy of DVT.30 miRNA-320b also regulates cell apoptosis and inflammation in diabetic retinopathy by modulating glucose metabolism and proinflammatory pathways and is a biomarker of metabolic disease in patient serum and EV.31^,^32 Serum miRNA-320b levels are also a biomarker for multiple sclerosis (MS) severity, progression, and brain atrophy,33 and plasma levels of miRNA-320b are elevated in hypertension34 and following severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection in coronavirus disease 2019 (COVID-19) patients.35 A common feature of all of these pathologies is endothelial cell disturbance and endothelial cell activation. Endothelial cells are in very close proximity to the peripheral blood and the parenchyma; therefore, endothelial cells are the first to experience perturbations in tissue microenvironments during pathological processes such as tissue invasion, infection and in response to cellular and tissue stress, such as alterations in oxygen saturation, blood glucose, and elevations in blood pressure. miRNA-320b expression was previously reported in endothelial cells36 and overexpression of miRNA-320b by mimetics can enhance cell proliferation, cell migration and blunts IL-6 and TNF-α release.34 The differential enrichment of miR-320b in plasma EVs at the time of MI in our study also suggests similar vascular endothelial cell origin. Our tissue-specific miRNA expression analysis indicated that miR-320b is particularly enriched in arterial and myocardial tissues, which corroborates its potential cardiovascular origin. Moreover, the observed enrichment of miRNA-320b in endothelial cell-derived EVs, particularly following proinflammatory TNF-α stimulation, points to a significant role for endothelial activation in miRNA-320b regulation post-MI. Interestingly, while EV-mediated delivery of miR-320b was associated with enhanced monocyte binding to activated endothelial cells, cytokine transcription, but reduced secretion of proinflammatory cytokines, forced lentiviral overexpression of miR-320b in monocytes led to, reduced adhesion and amplified cytokine production. These divergent outcomes suggest that the mode of delivery and magnitude of miRNA expression may distinctly influence cell behavior, and highlights the need for dose-sensitive and temporally controlled models when studying EV-miRNA function. While miRNAs are classically understood to repress gene expression in recipient cells, our data show a paradoxical transcriptional upregulation of proinflammatory cytokines (IL-6 and TNF-α) in monocyte-derived macrophages treated with miR-320b-enriched EVs. This may reflect indirect or compensatory regulatory mechanisms such as suppression of negative feedback regulators or post-transcriptional inhibition of protein translation or secretion. These findings highlight the complexity of EV-mediated signaling and the need for time-resolved and cell-type-specific mechanistic studies. We acknowledge that loss-of-function studies such as selective depletion of miRNA-320b from endothelial EVs would provide stronger mechanistic insight, but require careful control of donor cell phenotype, which we are actively addressing through inducible models in ongoing studies. The ability of endothelial-derived EVs to modulate immune cell activation via miRNA-320b also raises the possibility of targeting this axis for therapeutic immunomodulation following MI. Strategies to fine-tune EV miRNA cargo or selectively block EV uptake in specific immune populations may offer new avenues for limiting post-infarct inflammation.
TNF-α is a potent stimulator of endothelial cell activation, it is elevated in MI patients and linked to reperfusion injury.37 However, PBMCs obtained from patients following MI showed blunted responses to cytokine release (IL-6 and TNF-α) when stimulated with agonists in comparison to follow-up samples collected from the patients in the months following the MI.17 Similar reports show that plasma TNF-α concentrations are higher in non-MI patients versus MI patients, when blood is collected directly upon admission to hospital.38 Inflammatory activated endothelial cell-derived EV, rich in miRNA-320b significantly enhances monocyte adhesion to activated endothelial cells suggests a role in promoting inflammatory responses and potentially exacerbating MI in the early phase of inflammation. While inflammatory endothelial cell-derived EVs and MI plasma EV at presentation increased proinflammatory cytokine mRNA levels in macrophages and PBMCs, respectively, they paradoxically suppressed protein levels in macrophages, indicating a complex regulatory mechanism whereby EVs might influence cytokine production at the transcriptional level while modulating protein release. This may also explain the blunted cytokine responses in MI patients in previous clinical investigations.17^,^38 Monocytes obtained from MI patients prior to PCI also showed attenuated transcriptional responses to M1-stimulation, when compared to control patient monocytes; however, the initial induction of DEG following presentation to hospital with MI does not explain the overall lower transcriptional capacity of MI monocytes to M1-stimulation. Following monocyte mobilization to the blood in mice, monocytes undergo a metabolic reprogramming event within the first few hours of inflammatory activation, which enables monocyte migration to the inflamed tissues.39 Monocyte activation places a large energy demand on the cell; therefore, the more energy efficient process of oxidative phosphorylation is utilized over the largely inefficient process of glycolysis. This enables immune cells to undergo inflammatory activation and favor high energy demanding processes such cell migration for tissue recruitment. Inhibition of oxidative phosphorylation drastically reduces the ability of monocytes to migrate39; however, oxidative phosphorylation is associated with anti-inflammatory induction of immune cells and thus may explain the attenuated M1 response in MI patient monocytes.40 This is supported by our data, which shows that MI M1-stimulated monocytes favor fatty acid oxidation pathways. However, further studies are needed to demonstrate altered immune metabolic activity in peripheral blood leukocytes following MI.
The differential gene expression profiles observed in monocytes from MI versus control patients, and the induction of differential transcriptional activation by STEMI plasma EV at presentation alongside the significant overlap between miRNA-320b targets and genes upregulated in MI monocytes, underlines the importance of miRNA-320b in modulating inflammatory pathways. Specifically, the enrichment of pathways related to cell adhesion and cytokine production in MI monocytes, altered metabolic pathways in conjunction with the regulation of MAPK-associated proteins, highlights a potential mechanism through which miRNA-320b affects inflammatory responses and cellular activation of leukocytes following MI.
The influence of plasma EVs on immune cell gene expression and inflammatory activation was further elucidated by our experiments with PBMCs. Exposure to MI plasma EVs led to a broader and more pronounced impact on gene expression compared to control plasma EVs. This suggests that MI-associated plasma EVs might drive a more intense inflammatory response, as indicated by the upregulation of pathways associated with the inflammatory responses and cell adhesion. Plasma EV-transcriptomes are also altered during tumor progressing and capture certain features of the tumor and the patients disease severity.41 Similarly, a more comprehensive understanding of plasma EV-payloads may enable determination of the magnitude and granularity of the peripheral blood immune responses following MI to identify patients for focused therapeutic intervention to enable repair and restoration of the injured myocardial tissues.
Conclusion
Our study provides evidence that plasma EVs structural and cell-associated marker characteristics and their miRNA content, particularly miRNA-320b, change following MI and significantly reflect underlying endothelial activation. The enhanced presence of miRNA-320b in MI plasma EVs and its role in modulating monocyte adhesion and inflammatory pathways offer potential avenues for further research into targeted therapies and biomarkers for immunomodulation in MI. Understanding these mechanisms more comprehensively could lead to strategies for managing inflammation and improving outcomes in patients with MI.
Limitations of the study
Our study has several limitations. First, while we provide evidence that endothelial EV-derived miR-320b modulates immune cell transcriptional responses following MI, definitive loss-of-function experiments were not performed. Selective depletion of miR-320b from endothelial EVs or the use of inducible knockdown approaches would provide stronger mechanistic insight, but such strategies are technically challenging, because miRNA knockout in donor cells induces widespread cellular changes beyond the intended target. Second, although we observed consistent transcriptional upregulation of IL6 and TNF mRNAs with concomitant suppression of protein secretion, the precise mechanisms underlying this post-transcriptional regulation remain to be defined. Additional studies are required to determine whether EVs impair translation efficiency, trafficking, or cytokine release. Collectively, these caveats highlight the need for carefully controlled EV-specific and inducible systems, time-resolved functional studies, and species validation to dissect the relative contribution of miR-320b and other EV-associated signals in cardiovascular immunomodulation.
Resource availability
Lead contact
Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Naveed Akbar ([email protected]).
Materials availability
This study did not generate new unique reagents.
Data and code availability
- •The data underlying this article are available in Gene Expression Omnibus at https://www.ncbi.nlm.nih.gov/geo and can be accessed with [GSE187571](GSE187571) and GSE313527, or upon reasonable request to the corresponding author.
- •Data reported in this paper will be shared by the lead contact upon reasonable request.
- •This paper does not report original code.
- •Any additional information required to reanalyze the data reported in this study is available from the lead contact upon reasonable request.
Acknowledgments
The authors thank the staff at the Oxford Heart Center for the clinical care of patients recruited in the coordination of these studies. N.A. and R.P.C. acknowledge support by research grants from a 10.13039/501100000274British Heart Foundation (BHF) Intermediate Fellowship (N.A.: FS/IBSRF/22/25110); the 10.13039/501100005617British Heart Foundation (BHF) Centre of Research Excellence, Oxford (N.A. and R.P.C.: RE/13/1/30181; RE/18/3/34214; RE/24/130024); 10.13039/501100000274British Heart Foundation Project Grant (N.A. and R.P.C.: PG/18/53/33895); and the Tripartite Immunometabolism Consortium, 10.13039/100032285Novo Nordisk Foundation (R.P.C.: NNF15CC0018486 and NNF20SA0064144) 10.13039/501100013373Oxford Biomedical Research Centre (BRC); Nuffield Benefaction for Medicine and the Wellcome Institutional Strategic Support Fund (ISSF) (N.A.) and a Health Research Bridging Salary Scheme (HRBSS) to N.A. K.A.B. is supported by the Kusuma Trust. The views expressed are those of the author(s) and not necessarily those of the 10.13039/100024063National Health Service. The OxAMI study is supported by the 10.13039/501100000274British Heart Foundation (BHF; grant CH/16/1/32013 to K.M.C.); the 10.13039/501100005617BHF Centre of Research Excellence, Oxford (RG/13/1/30181); and the 10.13039/501100012618National Institute for Health Research Biomedical Research Centre, Oxford. See supplemental acknowledgments for OxAMI details. National Institute for Health and Care Research (NIHR) or the Department of Health.
Author contributions
N.A. led study project administration, funding acquisition, and conceptualization. L.T., R.R., A.B., C.H., T.R., C.L., S.L., D.G., Z.C., T.K., K.A.B., M.M.J., R.B., R.D., K.M.C., M.A., R.P.C., and N.A. undertook data curation, formal analysis, methodology, project administration, supervision, validation, visualization, and writing of the original draft. All authors contributed to writing, review and editing of the final manuscript.
Declaration of interests
The authors declare no competing interests.
Declaration of generative AI and AI-assisted technologies in the writing process
During the preparation of this work, the authors used ChatGPT (OpenAI, GPT-5 model) in order to assist with language editing, grammar refinement, and improvement of sentence clarity. After using this tool, the authors critically reviewed and edited the content to ensure scientific accuracy and integrity, and take full responsibility for the content of this publication.
STAR★Methods
Key resources table
REAGENT or RESOURCESOURCEIDENTIFIERAntibodiesAlixBiolegend634501; RRID:AB_2268110Annexin VR & D SystemsAF399; RRID:AB_2258445Apo ER & D SystemsMAB41441; RRID:AB_2289763b2BP I/ApolipoproteinHMP Bio0859414, RRID:AB_2334829CD105 (Endoglin)LS BioLS-C149182-200; RRID:AB_11143816CD106R & D SystemsMAB809; RRID:AB_2214234CD11cR&D SystemsMAB1777; RRID:AB_2234107CD138/syndecan-1R&D SystemsAF2780; AF2780CD14R & D SystemsMAB3833; MAB3833CD141/ThrombomodulinR&D Systems; RRID:AB_2201816CD142 (anti-hTF)R & D SystemsMAB2339; RRID:AB_2101338CD146Abcamab24577; RRID:AB_448154CD151R & D SystemsMAB1884; MAB1884CD16BD Biosciences555404, RRID:AB_395804CD162/PSGLR&D SystemsAF3345; RRID:AB_10973837CD206/MMRR & D SystemsAF2534; RRID:AB_2063019CD24BD Biosciences555426; RRID:AB_395820CD3BD Biosciences555337; RRID:AB_395743CD31 (PECAM-1)R & D SystemsAF806; RRID:AB_355617CD36Santa Cruzsc-7309; RRID:AB_627044CD4R & D SystemsMAB379; RRID:AB_2244289CD41Biolegend303702; RRID:AB_314372CD45R & D SystemsMAB1430; RRID:AB_2174120CD56BD Biosciences559043; RRID:AB_397180CD63BioradMCA2142; RRID:AB_324562CD81AncellA302-020A, RRID:AB_1576591CD9Ancell156–020EpCamSanta Cruz BioSc-59782; RRID:AB_10610504Fas LigandR & D SystemsRRID:AB_354703FATP1R&D SystemsMAB3304; RRID:AB_2190617FATP4R&D SystemsMAB3650; RRID:AB_2190634FATP5R&D SystemsMAB3897; RRID:AB_2302056Flotillin-1Abcamab41927; RRID:AB_941621Glut2R&D SystemsMAB1414; RRID:AB_2286240Glut4R&D SystemsMAB86541; RRID:AB_3659231ICAM-1 (CD54)Bender MedSystemsBMS1011; RRID:AB_1709134LAMP-1R & D SystemsMAB4800; RRID:AB_10719137LAMP2R & D SystemsMAB6228; RRID:AB_3658562MyosinAvivaOAAI00680, RRID:AB_3731350N-CadherinAbcamab19348; RRID:AB_444868p53Abcamab26; RRID:AB_303198TGFβ1BD Pharmingen555052; RRID:AB_395673Thrombospondin-1R & D SystemsAF3074; RRID:AB_2201958tPAR & D SystemsAF7449; RRID:AB_3644428TSG101AbnovaH00007251; RRID:AB_10718450VE-CadherinR & D SystemsAF938; RRID:AB_355726VEGFR2Biolegend359902; RRID:AB_2562485Vimentin (VIM)AvivasysbioOAAI00680, RRID:AB_3731350ApoBAbcamab139401, RRID:AB_3731356GM130Abcamab52649; RRID:AB_880266CD9System BiosciencesEXOAB-CD9A-1; RRID:AB_2687469Syntenin-1Abcamab133267; RRID:AB_1116026VCAM-1Abcamab134047; RRID:AB_2721053TSG101Abcamab83; RRID:AB_306450Histone H3Cell signalingD1H2/4499PBiotinylated anti-human CD9AncellClone SN4/C3-3A2Biotinylated anti-human CD63Ancell215-030; RRID:AB_2665375Biotinylated anti-human CD81Ancell302-030; RRID: AB_3731626Biotinylated goat anti-mouse IgGNovus BiologicalsNB7605, RRID:AB_790573Anti-Rabbit IgG (H + L), HRP ConjugatePromegaW4011; RRID:AB_430833Anti-Mouse IgG (H + L), HRP ConjugatePromegaW4021; RRID:AB_430834Bacterial and virus strainsVB UltraStableVector BuilderVB UltraStablepMD2.GAddgenePlasmid #12259pMDLg/pRREAddgenePlasmid #12251pCMV-VSV-GAddgenePlasmid #8454pLV[shRNA]-EGFP:T2A:Puro-U6>{hsa-miR-320b}Vector BuilderEcoli(VB250302-1277jxs)Scramble shRNA lentiviral control vector pLV[shRNA]-EGFP/Puro-U6>Scramble_shRNAVector BuilderEcoli(VB010000-9526zpu)-PBiological samplesBlood ConesNational Health ServiceChemicals, peptides, and recombinant proteinsLife Technologies434315Recombinant Human TNF-alpha ProteinBiotechne210-TA-020/CFLPSSigma-AldrichL2880-10 MGRecombinant Human IFN-gamma ProteinBiotechne285-IF-100/CFCritical commercial assaysELISA IL-6BiotechneDY206ELISA TNF-αBiotechneDY210DuoSet ELISA Ancillary Reagent Kit 2 (5 96 well plates)BiotechneDY008BGeneChip miRNA 4.0 ArraysThermofisher Scientific902411Deposited dataRNA-sequencing of peripheral blood neutrophils from STEMI and NSTEMI patients at presentation and 1 month post-AMIGene Expression OmnibusGSE187571RNA-sequencing of peripheral blood monocytes from STEMI and NSTEMI patients at presentation and 1 month post-AMI and from peripheral blood mononuclear cells treated with plasma extracellular vesicles from patients at presentation and follow up 3 months post-MI.Gene Expression OmnibusGSE313527Experimental models: Cell linesHUVEC – Human Umbilical Vein Endothelial Cells, Pooled, in EGM™LonzaCC-2519THP-1ATCCTIB-202293TATCCCRL-3216OligonucleotidesmiRNA-320bQiagenhsa-miR-320b (YP02119299)miRNA-126-3pQiagenhsa-miR-126-3p (YP00204227)miR-103-3pQiagenendogenous control assay (miR-103-3p)18sThermo Fischer ScientificHs03003631_g1PPIAThermo Fischer ScientificHs04194521_s1TNF-αThermo Fischer ScientificHs00174128_m1IL-6Thermo Fischer ScientificHs00985639_m1Software and algorithmsRStudio is RStudioR Project2025.05.1 + 513Prism versionPrism10.5.0
Experimental model and study participant details
Patients
Written informed consent was obtained from all subjects. All clinical investigations were conducted in accordance with the Declaration of Helsinki, and ethical approval was obtained from the National Research Ethics Service, South Central – Oxford A Research Ethics Committee (REC reference 16/SC/0102) before the commencement of the study. Patients presenting with acute coronary syndromes (ACS), defined according to the third universal definition of myocardial infarction,42 were recruited at Oxford University Hospitals NHS Foundation Trust. Patients were identified upon presentation with ischaemic chest pain, confirmed through electrocardiogram (ECG) changes and/or elevated cardiac troponin, and scheduled for invasive coronary angiography.
Inclusion criteria included patients over 18 years of age, presenting with ACS who had not been exposed to lipid-lowering drugs, including statins, at the time of admission. Patients with contraindications to magnetic resonance imaging (MRI) were excluded from the study.
All patients were managed in accordance with standard care guidelines for ACS. Upon admission, they commenced on intensive statin therapy (daily atorvastatin 80 mg) as part of routine clinical care, in accordance with prevailing guidelines.43
Further patient recruitment occurred via the Oxford Acute Myocardial Infarction Study (OxAMI), with ethical approval for cohort protocols provided by the Oxfordshire Research Ethics Committee (REC 10/H0408/24). In parallel, peripheral venous blood was collected from healthy male volunteers, or leukapheresis cones were obtained from the NHS Blood Bank for comparative analyses (R35181/RE003). The influence of sex on the study outcomes was not specifically evaluated owing to limited sample size; thus, potential sex-related effects cannot be excluded.
PBMCs from healthy volunteers
Peripheral venous blood was collected from healthy male volunteer’s in ethylenediaminetetraacetic acid (EDTA) tubes or leukapheresis cones were obtained from NHS Blood Bank. Written informed consent was provided from all donating participants. PBMCs were resuspended in 5 mmol/L glucose (Sigma-Aldrich) that contained 10% EV-depleted foetal bovine serum (FBS), 2mM L-glutamine and penicillin and streptomycin (Sigma-Aldrich) and cells were cultured in a humidified atmosphere with 5% CO2 at 37°C.
Peripheral blood monocytes from MI patients
Peripheral venous blood was collected into EDTA tubes. Isolated monocytes were incubated in complete 5 mmol/L glucose DMEM (supplemented with 10% FBS, 2% penicillin-streptomycin, and 2 mmol/L glutamine) in a humidified atmosphere with 5% CO2 at 37°C.
Human umbilical vein endothelial cells
Human umbilical vein endothelial cells (HUVECs) (Lonza, UK) were cultured in EGM-2 media supplemented with BulletKit (Lonza, UK) in a humidified atmosphere with 5% CO_2_ at 37°C. HUVECs were maintained with a final concentration of 2% FBS, for HUVEC-EV studies all FBS was depleted of EV by ultracentrifugation at 120,000 x g for 16 hours at 4°C. cell identity was verified by the supplier and cells were routinely tested and confirmed negative for mycoplasma contamination. Cells were discarded after six passages.
THP-1 human monocytes
THP-1 human monocytes (ATCC TIB-202) were cultured in RPMI medium (Sigma-Aldrich) supplemented 10% FBS, 1% antibiotics (Penicillin-Streptomycin) (Sigma-Aldrich), 2 mM L-glutamine (Sigma-Aldrich), and 0.05 nM 2-mercaptoethanol in a humidified atmosphere with 5% CO2 at 37°C. Cells were authenticated and mycoplasma-free.
Details of statistical analyses for all cell-based and patient-derived experiments are provided in the Quantification and Statistical Analysis section.
Method details
Plasma
Peripheral blood was collected in EDTA tubes at the time of presentation to hospital with MI and three months later from the same patients following standard clinical treatment or from healthy volunteers. Peripheral blood was centrifuged for 15 minutes at 1000 x g at 4^o^C. Subsequently plasma was rendered platelet poor by centrifugation at 5000 x g for 10 minutes at 4^o^C and stored in 500 μL aliquots at -80^o^C.
EV-Array
The choice of plasma EV isolation method can introduce method-dependent biases in acquired EV-characteristic, including EV size, concentration and the presence of cell associated EV sub-populations.6 Therefore, plasma EV were initially analysed by EV-Array technology, which utilises immobilised antibodies to bind EV from whole plasma to the array surface, therefore circumventing the need for isolation and quantification of EV-markers and cell associated proteins, which has been described previously.6^,^9^,^44 Antibodies were diluted in PBS with 5% trehalose and printed in triplicates at 200 μg/mL onto slides. Biotinylated goat anti-mouse IgG (Novus Biologicals, CO, USA) was used as a positive control. 75 μL platelet poor plasma sample (per patient / per time point) and Casein Blocking Buffer (10x concentrate, Sigma-Aldrich, MO, USA, catalogue B6429) was applied to each array and incubated for 2 hours at room temperature on an orbital shaker (450 rpm) followed by a static overnight incubation at 4°C. After a wash procedure, each array was incubated with 100 μL detection antibody cocktail (biotinylated anti-human-CD9 (clone SN4/C3-3A2), -CD63 (clone AHN16.1/46-4-5) , and -CD81 (clone 1.3.3.22) (Ancell, MN, USA) diluted 1:1,500 for 2 hours at room temperature with agitation. Following a wash, 100 μL streptavidin-Cy3 (Life Technologies, MA, USA) diluted 1:3,000 was added to each well and incubated for 30 minutes at room temperature on a shaker. Arrays were dried and scanned using a sciREADER FL2 microarray scanner (Scienion AG, DE), at 535 nm and an exposure time of 2,000 milliseconds.
Plasma extracellular vesicle isolations
500 μL of plasma was thawed at room temperature and centrifuged for 10 minutes at 5000 x g. Plasma EV were isolated using size exclusion chromatography (SEC) and ultracentrifugation as previously described.23 Briefly, plasma was applied to qEV columns (iZON) and EV enriched fractions were collected, pooled, combined with phosphate buffered saline (PBS) (ThermoFisher) into quick seal ultra-centrifugation tubes and centrifuged for 120,000 x g for 1 hour at 4^o^C (Beckman Coulter, Optima MAX – XP Ultracentrifuge). Pelleted EV were resuspended in 100 μL of PBS (Thermofisher) or lysed in 100 μL of 1x RIPA buffer (Cell Signalling) supplemented with PhosSTOP (Roche, Basel, Switzerland) and cOmplete (Roche, Basel, Switzerland) and utilised in studies as described in accordance with recommendation from the International Society for Extracellular Vesicles.45
Nanoparticle Tracking Analysis (NTA)
Plasma EV size and concentration was determined by Nanoparticle Tracking Analysis (NTA) using: the Malvern Nanosight 405nm NS500 instrument as previously described.10^,^46 Briefly, samples were diluted in PBS (1/10 to a total volume of 1 mL. Nanosight measurements were obtained at camera level 12 (camera shutter speed: 15 ms, camera gain: 350) using the following script: PRIME, DELAY 5, CAPTURE 60, REPEAT 5. Videos were analysed using NanoSight NTA 2.3 software (Malvern, UK). For cell culture samples, Particle Metrix NTA, Zetaview was used. Samples were diluted (1:1,000) in 1 mL PBS and loaded into the sample chamber. Zetaview settings were: 11 positions, 2 cycles, medium video resolution. Pre-acquisition settings: sensitivity 80, fame 30, shutter speed 100. Post-acquisition settings min brightness 25, max size 1000, min size 5, trace length 15, nm / class 30, classes / decade64.6
Single particle interferometric reflectance imaging sensing (NanoView) analysis
Isolated plasma EV samples were incubated with the ExoView tetraspanin arrays (NanoView Biosciences, EV-TC-TTS-01) overnight at room temperature. The ExoView tetraspanin arrays contain antibodies against CD9, CD63, CD81, and a mouse IgG1 Isotype control in triplicate. After incubation, the ExoView arrays were washed on an orbital shaker with PBS-tween-20 (0.05%). Nanoview chips were then incubated with ExoView tetraspanin labelling antibodies (NanoView Biosciences, EV-TC-AB-01) (anti-CD81 Alexa-555, anti-CD63 Alexa 488, and anti-CD9 Alexa-647) as previously described.47 Nanoview arrays were imaged using the ExoView R100 reader with the ExoScan 2.5.5 acquisition software and analysed using ExoViewer 2.5.0.
Transmission electron microscopy (TEM)
Negative staining of isolated EV was achieved by passive absorption onto grids (300 mesh Cu carbon film) that were glow discharged for 20 seconds at 15 mA (Leica EM ACE 200). Samples were applied to the grid for 2 minutes, blotted, stained with 2% uranyl acetate for 20 seconds and subsequently allowed to air dry. Images were acquired on a 120kV Tecnai 12 TEM (ThermoFisher) equipped with a OneView digital camera (Gatan).
Western blotting
EV associated protein markers were confirmed as previously described.6 Protein lysates were combined with NuPage LDS sample buffer (4x) agent (Invitrogen). Samples were loaded onto a 4-12% bis-tris gradient gel (NuPAGE 4-12% Bis-Tris Protein Gel; 1.5 mm (ThermoFisher Scientific). Separated samples were transferred to nitrocellulose membranes and blocked for non-specific binding in 5 % milk in PBS-tween for 1 hour at room temperature. Membranes were incubated with primary antibodies overnight as detailed in key resource table in 5% milk in PBS-tween at 4^0^C. Membranes were washed three times and incubated with secondary-horse radish peroxidase (HRP) conjugated antibody for one hour at room temperature. Membranes were washed again before incubating with enhance chemiluminescence substrate (Pierce ECL, ThermoFisher Scientific) for imaging (Bio-Rad ChemiDoc MP Imaging system).
Plasma EV-miRNA isolation
Plasma EV-RNAs were isolated using miRNeasy Mini Kits (Qiagen) as per the manufacturer’s instructions.
miRNA arrays
Isolated RNA was checked for quality using Agilent 2100 Bioanalyzer chips (Agilent). GeneChip miRNA 4.0 Arrays (Thermofisher Scientific) were used to detect and quantify plasma EV-miRNAs according to the manufacturer’s instructions by UK Bioinformatics Ltd.
miRNA analysis
GeneChip miRNA 4.0 Arrays were analysed using the oligo48 and LIMMA49 R packages. Briefly, arrays were background corrected and normalised using the RMA algorithm.50^,^51^,^52 Redundant probe expression values were averaged at the gene level. Differential expression was analysed using a mixed effects model comparing groups and timepoints while controlling for array batch, family history of cardiovascular disease, smoking status, and age.
miRNA-mRNA and pathway analysis and comparisons
The most differentially expressed miRNAs were investigated for tissue expression using miRNATissueAtlas2.24 mRNA putative targets were obtained from miRWalk,53 TargetScan54 and miRDB.55 miRNA-mRNA putative targets were analyzed for target pathways using Gene Ontology (GO)56^,^57^,^58 Biological Pathway terms. A Fisher’s exact test was used to compare lists of genes and a reference 20,000 genes was used in calculations using R studio. A background of 20,000 genes and 29,112 total biological processes were used for the calculations, respectively.
Human endothelial cell culture
HUVECs were seeded into T175 cm^2^ tissue culture flasks at a density of 4 x 10^6^ cells per flask in 15 mL standard medium and incubated overnight. The next day, the cell culture supernatants were removed, and the cell monolayers were washed with 15 mL PBS. Subsequently, 15 mL of EV-depleted medium, with or without 10 ng / mL recombinant human tumour necrosis factor-α (TNF-α) (R&D Systems) was exposed to cells at the specified time points.
After incubation, cell cultures were trypsinised and counted using a Countess 3 automated cell counter (ThermoFisher Scientific), and supernatants were collected and centrifuged at 1,000 x g for 10 minutes to pellet cellular debris. The resulting supernatant was transferred to quick-seal ultracentrifugation tubes and centrifuged at 120,000 x g for 2 hours at 4°C (Beckman Coulter, Optima MAX-XP Ultracentrifuge). The resulting EV pellets were resuspended in 100 μL PBS and washed by combining with 13 mL PBS followed by another centrifugation at 120,000 x g for 1 hour. Finally, EV pellets were resuspended in 100 μL PBS or lysed in 100 μL of 1x RIPA buffer for further studies.
RNA extraction and quantification
Total RNA was isolated and purified from cell pellets and EV samples using the miRNeasy Mini Kit (Qiagen) according to the manufacturer’s protocol. The concentration and purity of the RNA were determined with a NanoDrop spectrophotometer (NanoDrop Technologies). For miRNAs cDNA synthesis was performed using the miRCURY LNA RT kit (Qiagen). Quantitative PCR (qPCR) was conducted with the miRCURY LNA SYBR Green PCR Kit (Qiagen), utilizing the miRCURY SYBR Green Master Mix and primers for miR-320b and miR-101-5p (Qiagen). Controls included UniSp6 RNA spike-in, introduced during the cDNA synthesis stage, and a primer for endogenous miRNA, miR-103a-3p. Additionally, a primer for EC-associated miRNA, miR-126-3p, was used (Qiagen). For mRNA cDNA was synthesised using QuantiTect Reverse Transcription RT-PCR kit (Qiagen). mRNA levels were determined by quantitative PCR (qPCR) using TaqMan probes and reagents as per the manufacturer’s instructions.
THP-1 culture and static adhesion assay
THP-1 human monocytes (2 x 10^6^ cells / mL) were treated with 1 x 10^9^ HUVEC-EVs from control or TNF-α stimulated cells or an equal volume of control media that had not been in contact with cells and incubated under standard cell culture conditions for 16 hours. THP-1 monocytes were wash by centrifugation in PBS and stained with PKH67 (Sigma-Aldrich) according to the manufacturer’s instructions. PKH67 stained monocytes were added to confluent monolayers of HUVECs grown on glass cover slides in a 12 well plate, that were activated with recombinant human TNF-α 10 ng / mL for 24 hours prior to the assay. Activated HUVEC monolayers were washed with PBS prior to the addition of 20,000 cells THP-1 cells per well of activated HUVEC monolayers for 1-hour at 37°C. Supernatants were then removed, monolayers and adhered THP-1 cells washed with PBS and 4% paraformaldehyde applied for 10 minutes at room temperature, washed with PBS and mounted onto glass slides using a glycerol mounting medium containing 2-(4-amidinophenyl)-1H-indole-6-carboxamidine (DAPI) mounting media (Abcam, UK) and imaged on a fluorescent microscope.
PBMCs isolations from healthy volunteers
Peripheral venous blood was collected from healthy male volunteer’s in EDTA tubes or leukapheresis cones were obtained from NHS Blood Bank and layered onto Ficoll Paque Plus and centrifuged at 600 x g for 30 min following approval R35181/RE003 and informed written consent. PBMCs were collected from the buffy coat, combined with 50 mL PBS and washed by centrifugation (500 x g for 15 minutes at 4^o^C). PBMCs were resuspended in 5 mmol/L glucose (Sigma-Aldrich) that contained 10% EV-depleted FBS, 2mM L-glutamine and penicillin and streptomycin (Sigma-Aldrich). Cells were counted and plated at a density of 2 x 10^5^ / mL per well of a 48 well plate and rested overnight in humidified incubator. PBMC were exposed to isolated plasma EV exposed for 6 or 18 hours and +/- M1-stimulation with 100 ng / mL LPS (Sigma-Aldrich) and 10 ng / mL of recombinant human interferon-ꝩ (Invitrogen) (R&D Systems) for 6 hours. Following stimulations cells were collected, washed in PBS by centrifugation and lysed using the lysis buffer from the miRvana miRNA kit (Thermofisher Scientific). Samples were snap frozen and stored at -80^o^C.
Peripheral blood monocytes from MI patients
Peripheral venous blood was collected into EDTA tubes was diluted 1:1 with PBS and monocytes were negatively enriched using the EasySep Human monocyte enrichment kit without CD16 depletion (StemCell Technologies, Grenoble, France) as per the manufacturer’s instructions. Isolated monocytes were incubated in complete 5 mmol/L glucose DMEM (supplemented with 10% FBS, 2% penicillin-streptomycin, and 2 mmol/L glutamine) under control conditions or M1-proinflammatory stimulated with 100 ng / mL LPS (Sigma-Aldrich) and 10 ng / mL of recombinant human interferon-ꝩ (Invitrogen) (R&D Systems) for 6 hours at 37^o^C and 5% CO2. Following incubation the cells were centrifuged; supernatants were removed and cell pellets lysed with RLT (Qiagen) and stored at -80^o^C.
Monocyte-derived macrophage isolations and cultures
Monocytes were enriched by preparing a solution containing 1 x 10^7^ PBMCs in PBS, 0.5% bovine serum albumin (BSA) and 2 mM EDTA by diluting MACS BSA Stock Solution (130- 091-376) 1:20 with autoMACS™ Rinsing Solution (130-091- 222). Anti-CD14 microbeads (Miltenyi 130-050-201) were combined with 1 x 10^7^ PBMCs following the manufacturer’s instructions and passed through MS column (Miltenyi 130-042-201). Enriched monocytes were incubated with 50 ng / mL recombinant human recombinant Human M-CSF Protein (R&D systems) at a density of 200,000 / well in a 48 well plate and rested for 3 days prior to experimentation. HUVEC derived EV were exposed for 6 or 18 hours and +/- M1-stimulation with 100 ng / mL LPS (Sigma-Aldrich) and 10 ng /mL of recombinant human interferon-ꝩ (Invitrogen) (R&D Systems) for 6 hours. Cell culture supernatants were collected, and cells were lysed using the lysis buffer from the miRvana miRNA kit (Thermofisher Scientific). Samples were snap frozen and stored at -80^o^C.
Lentiviral vector production and cell transduction
Lentiviral vectors for overexpression of miR-320b (pLV[shRNA]-EGFP:T2A:Puro-U6>{hsa-miR-320b}; VectorBuilder ID: VB250302-1277jxs) and scrambled control (pLV[shRNA]-EGFP:T2A:Puro-U6>Scramble_shRNA; VectorBuilder ID: VB010000-9526zpu) were produced using a third-generation packaging system. HEK-293-T (ATCC) cells were seeded at 80% confluence in high glucose supplemented with 10% FBS and transfected using TransIT®-293 Transfection Reagent in Opti-MEM™ Reduced Serum Medium with the transfer plasmid and the three packaging plasmids: pMDLg/pRRE (gag/pol), pRSV-Rev (rev), and pMD2.G (VSV-G envelope). Viral supernatants were harvested at 48 post-transfection, filtered through 0.45 μm membranes, and stored at -80^o^C. THP-1 monocytes and HUVECs were transduced with lentiviral particles in the presence of TransduceIT® Transduction Reagent (Sigma-Aldrich). After 24 hours, medium was replaced, and cells were maintained for 24–72 hours before further analysis. Transduction efficiency was confirmed by eGFP fluorescence and RT-qPCR for miR-320b expression. THP-1 cells were selected with puromycin for >5 days prior to use in downstream assays.
ELISAs
Endothelial cell activation following stimulation with recombinant TNF-α was confirmed using the human VCAM-1/CD106 DuoSet ELISA kit (R&D Systems). Additionally, human IL-6 and TNF-α were detected in immune cell cultures using the DuoSet ELISA kits from R&D Systems. ELISAs were conducted according to the manufacturer's instructions.
RNA-sequencing
RNA-sequencing was undertaken at the Wellcome Trust Centre for Human Genetics, University of Oxford. RNA was isolated using RNeasy Mini Kit (Qiagen), according to manufacturer’s instructions. The amount of total RNA was quantified using the Qubit Fluorometric Quantitation system (Life Technologies) and the RNA integrity number (RIN) was determined using the Experion Auto-mated Electrophoresis System (Bio-Rad). RNA-seq libraries were prepared with the TruSeq Stranded mRNA Library Prep kit (Illumina) using both Sciclone and Zephyr liquid handling robotics (PerkinElmer). Library concentrations were quantified with the Qubit Fluorometric Quantitation system (Life Technologies) and the size distribution was assessed using the Experion Automated Electrophoresis System (Bio-Rad). For sequencing, samples were diluted and pooled in equimolar amounts and sequenced on Illumina HiSeq 3000/4000 instruments in 50-bp-single-read configuration. Base calls provided by the Illumina Real-Time Analysis (RTA) software were subsequently converted into BAM format (Illumina2bam) before de-multiplexing (BamIndexDecoder) into individual, sample-specific BAM files via Illumina2bam tools (https://github.com/wtsi-npg/illumina2bam).
Transcriptomes analysis
Transcript abundance quantification was carried out from raw RNA-seq reads by Salmon v1.0.059 with GC bias correction and mapping validation options, using the Ensembl human cDNA annotation release 96.60 Analysis of differential gene expression was carried out using DESeq2 v1.26.061 with an experimental design comparing EV groups, stimulations and controlling for PBMC donors. Briefly, pre-filtering was performed to remove cDNAs with fewer than 50 total reads across all samples. Differential expression analysis was then carried out with the default parameters. Shrinkage of log fold changes for visualisation in MA plots was carried out using the ashr R package.62
Quantification and statistical analysis
Statistical analyses were performed using GraphPad Prism (v10.0.3; GraphPad Software, San Diego, CA, USA) and RStudio (v4.2.1; R Foundation for Statistical Computing, Vienna, Austria). Normality of data distributions was assessed using the Shapiro–Wilk test. Depending on distribution, paired or unpaired t-tests, Fisher’s exact tests, or non-parametric Wilcoxon matched-pairs signed-rank tests were used for two-group comparisons. For multiple-group comparisons, one- or two-way ANOVA with Tukey’s post hoc test or mixed-effects models (for repeated measures or non-independent data) were employed where appropriate. A two-tailed p-value of < 0.05 was considered statistically significant. N denotes the number of biological replicates, independent experiments, or individual study participants/patients as specified in each figure legend. Full quantification and statistical details can be found in the figure legends for each dataset. The definition of centre (mean ± SD) and dispersion/precision measures are described in the corresponding figure legends.
Additional resources
Biorender
Biorender.com was used to generate schematic figures.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Jeppesen D.K.Zhang Q.Franklin J.L.Coffey R.J.Extracellular vesicles and nanoparticles: emerging complexities Trends Cell Biol.33202366768110.1016/j.tcb.2023.01.00236737375 PMC 10363204 · doi ↗ · pubmed ↗
- 2van Niel G.D'Angelo G.Raposo G.Shedding light on the cell biology of extracellular vesicles Nat. Rev. Mol. Cell Biol.19201821322810.1038/nrm.2017.12529339798 · doi ↗ · pubmed ↗
- 3Kowal J.Arras G.Colombo M.Jouve M.Morath J.P.Primdal-Bengtson B.Dingli F.Loew D.Tkach M.Théry C.Proteomic comparison defines novel markers to characterize heterogeneous populations of extracellular vesicle subtypes Proc. Natl. Acad. Sci. USA 1132016 E 968E 97710.1073/pnas.152123011326858453 PMC 4776515 · doi ↗ · pubmed ↗
- 4Hoshino A.Costa-Silva B.Shen T.L.Rodrigues G.Hashimoto A.Tesic Mark M.Molina H.Kohsaka S.Di Giannatale A.Ceder S.Tumour exosome integrins determine organotropic metastasis Nature 527201532933510.1038/nature 1575626524530 PMC 4788391 · doi ↗ · pubmed ↗
- 5Driedonks T.Jiang L.Carlson B.Han Z.Liu G.Queen S.E.Shirk E.N.Gololobova O.Liao Z.Nyberg L.H.Pharmacokinetics and biodistribution of extracellular vesicles administered intravenously and intranasally to Macaca nemestrina J. Extracell. Biol.12022 e 5910.1002/jex 2.59PMC 979928336591537 · doi ↗ · pubmed ↗
- 6Paget D.Checa A.Zöhrer B.Heilig R.Shanmuganathan M.Dhaliwal R.Johnson E.Jørgensen M.M.Bæk R.Craig E.W.Comparative and integrated analysis of plasma extracellular vesicle isolation methods in healthy volunteers and patients following myocardial infarction J. Extracell. Biol.12022 e 6610.1002/jex 2.6638939906 PMC 11080728 · doi ↗ · pubmed ↗
- 7Loyer X.Zlatanova I.Devue C.Yin M.Howangyin K.Y.Klaihmon P.Guerin C.L.Kheloufi M.Vilar J.Zannis K.Intra-Cardiac Release of Extracellular Vesicles Shapes Inflammation Following Myocardial Infarction Circ. Res.123201810010610.1161/CIRCRESAHA.117.31132629592957 PMC 6023578 · doi ↗ · pubmed ↗
- 8Khandagale A.Lindahl B.Lind S.B.Shevchenko G.Siegbahn A.Christersson C.Plasma-derived extracellular vesicles from myocardial infarction patients inhibits tumor necrosis factor-alpha induced cardiac cell death Curr. Res. Transl. Med.70202210332310.1016/j.retram.2021.10332334979484 · doi ↗ · pubmed ↗