Platelet-derived microvesicles modulate the bioenergetic and inflammatory phenotype of human polymorphonuclear leukocytes

Abstract

Platelets release microvesicles (PMVs) into the extracellular milieu upon activation. PMVs retain various platelet components, including functional mitochondria, and actively participate in intercellular communication with immune cells such as polymorphonuclear leukocytes (PMNs). PMVs have been known to modulate the inflammatory response of PMNs under normal physiological condition. Despite growing interest in the transfer of biological material between immune cells, the mitochondrial content shuttling from PMVs to PMNs and the resulting effects have remained unclear. Using freshly isolated PMVs from healthy and consenting donors, we demonstrate that PMVs modulate both the bioenergetic and inflammatory phenotypes of the recipient immune cell. We first confirmed the mitochondrial content transfer and then measured cell viability, mitochondrial respiration, and ATP production. Platelet-derived mitochondria were found associated with PMNs, consequently decreasing caspase-3 activity. PMVs increased mitochondrial activity and ATP levels in the recipient cell. Incubation of PMNs with PMVs containing nonfunctional mitochondria did not affect respiration and caspase-3 activity. This demonstrates that functional and active mitochondria are required for the PMVs to modulate the bioenergenetic phenotype of human PMNs. Finally, we detected the transfer of active 12-lipoxygenase and of cyclooxygenase-1 in the recipient cells, enzymes found specifically in PMVs, and an increase in the production of their respective inflammatory products. These findings suggest that platelet-derived mitochondria play a key role in enhancing the survival and inflammatory function of PMNs in inflammatory conditions.

Platelet-derived microvesicles induces phenotype modulation in neutrophils, marked by increased viability, elevated platelet-derived inflammatory proteins, and lipid mediators, culminating in a proinflammatory neutrophil phenotype.

1. Introduction

Polymorphonuclear leukocytes (PMNs), are abundant circulatory immune cells that play an essential role in the innate immunity. Among the first blood cells recruited to infection sites, they promote the inflammatory response and are directly involved in the elimination of the pathogens by several mechanisms including phagocytosis, neutrophil extracellular traps (NETs), and production of proinflammatory mediators such as eicosanoids and cytokines/chemokines. However, while PMNs are key participants in acute inflammation, they also contribute to chronic inflammatory diseases, like rheumatoid arthritis (RA).^1–3 In fact, PMNs are involved in multiple phases of the RA pathology (reviewed in Wright et al.)¹ and are primarily recruited to the inflamed joints by the lipid mediator leukotriene B₄ (LTB₄)² and chemokine interleukin (IL)-8.⁴ Upon migration into the synovial cavity, PMNs undergo excessive phenotype changes⁵ that results in delayed apoptosis,^5,6 as well as an increase in both biosynthesis of eicosanoids (i.e. leukotrienes, prostaglandins) and cytokines (i.e. IL-8, tumor necrosis factor α, IL-6).^1,2,7

Mitochondria are considered the powerhouse of the cell, generating ATP via the oxidative phosphorylation. For years, the functionality of PMNs’ mitochondria has been debated,⁸ in part because of the very low amount of mitochondrial ATP and cellular respiration detectable in resting cells.^9–11 Consequently, PMNs rely heavily on glycolysis for their energy requirements.^12,13 Circulating PMNs have a relatively short lifespan (8 to 12 h) in comparison with platelets, B cells, T cells, and monocytes/macrophages.¹⁴ The questioned functionality of mitochondria in PMNs has been considered as an explanation for the short lifespan of cells, as mitochondrial dysfunction and decreased ATP production can promote cell apoptosis.^15,16 However, for reasons that are still elusive, the PMN's lifespan increases considerably, up to several days in some cases, under inflammatory conditions.^5,17 This increased lifespan is problematic, as the persistent release of PMN-derived inflammatory mediators such as leukotrienes, prostaglandins, and cytokines is known to significantly contribute to the severity of chronic inflammatory diseases, such as the case in the RA pathology.¹⁸

Platelets are small discoid shaped cells (∼3 to 5 mm in size) better appreciated for their role in promoting wound repair. Recent evidence has shown that they also actively participate in chronic inflammatory diseases, including RA,¹⁹ atherosclerosis,²⁰ and systemic lupus erythematosus.²¹ Physiological agonists such as thrombin, collagen, fibrinogen, or immune complexes induce the activation of the calcium-dependent calpain's pathway in platelets.²² This pathway triggers the remodeling of the cell's cytoskeleton, resulting in the shedding in the extracellular milieu of small extensions of their cytoplasm known as platelet-derived microvesicles (PMVs) (submicron particles of 0.1 to 1 mm in size). This process implicates the transfer of bioactive content of the platelet, including RNAs, cytokines, proteins, organelles, and lipids.^23–25 Interestingly, PMVs, but not whole platelets, are abundantly found in the synovial fluid of RA patients.^19,26,27 Further adding to the complexity of the inflammatory contribution of PMVs to the disease is the discovery that these extracellular vesicles are in fact a highly heterogeneous population.^28,29 Not only are PMVs found to be associated with immune complexes capable of inducing the inflammatory response,³⁰ but also our team discovered a previously unreported subtype of PMVs in the synovial fluid of RA patients that retains functional mitochondria, termed mitochondria-containing microvesicles (mitoMPs).²⁶ Since their initial discovery, mitoMPs have also been detected in systemic lupus erythematosus patients,³¹ in the blood of obese individuals,³² in the plasma of healthy individuals,³³ and in platelet concentrate storage bags.³⁴

Interestingly, both PMVs and mitoMPs have been shown to interact with several cell types and transfer platelet-specific content, such as mitochondria, subsequently reprogramming the bioenergetic state of the recipient cell.^23,35–37 Nonetheless, whether the modulation of the metabolomic state of the recipient cells correlates with an increase in the lifespan of short-lived immune cells, for instance PMNs, remains uninvestigated. Hence, the main objective of this study was to investigate the effects of platelet-derived mitochondrial content in their ability to increase the lifespan of PMNs under inflammatory conditions. Here, we demonstrate that the presence of functional mitochondria in PMVs is required for the increase in both respiration and ATP levels in PMNs. Furthermore, we first report a transfer of inflammatory enzymes, consequently increasing the production of inflammatory lipid mediators in human PMNs. Although PMNs are crucial for the innate immunity response, prolonged survival of these cells can be detrimental, potentially leading to persistent inflammation, tissue damage, and organ dysfunction. A slight increase in the PMN's lifespan can be of significant clinical implications, thus offering new therapeutic opportunities in regulating platelet/PMN interaction in the context of inflammatory disease.

2. Methods

2.1. Cell culture and PLB-985 differentiation

PLB-985 cells, a human monomyelocyte cell line was a generous gift from Dr. Marc Surette (Université de Moncton), which were obtained from DSMZ (ACC139, Leibniz Institute DSMZ). The neutrophil-like phenotype for the PLB-985 cells was obtained with the addition of dimethyl sulfoxide (DMSO) as previously described.^38,39 Briefly, cells were seeded at 2 × 10⁵ cells/mL in RPMI 1640 medium (Corning) supplemented with 10% fetal bovine serum (FBS) (Corning) and 1% penicillin-streptomycin (Pen/Strep) (Thermo Fisher Scientific) at 37 °C under 5% CO₂ atmosphere. Differentiation was initiated with the addition of 1% DMSO (Millipore Sigma). After 72 h, cells were counted and reseeded at 4 × 10⁵ cells/mL in presence of a new culture medium supplemented with 1% DMSO and further incubated for an additional 72 h. The differentiation was confirmed by flow cytometry (Attune NxT; Invitrogen) with anti-human CD11b-Brilliant Violet 421 (BioLegend) labeling. For the entire study, only DMSO-differentiated PLB-985 cells were used.

2.2. Blood cell isolation and platelet-derived microvesicles production

This study was conducted in accordance with the guidelines of the Declaration of Helsinki of 1975 (revised 2013). The Institutional Review Board statement and approval for studies involving human samples have been granted from the Université de Moncton (project no. 1516–002) research ethics board. PMNs were isolated as previously described.^11,40 Briefly, blood was obtained from healthy, consenting donors in tubes containing the anticoagulant acid citrate dextrose (ACD-A; 75 mM sodium citrate dihydrate, 38 mM citric acid, 136 mM dextrose, pH 4.5) and centrifuged at 275 g for 10 min at room temperature (RT) without brakes. The top layer containing the platelet-rich plasma (PRP) was removed and the bottom layer was subjected to a 3% (w/v) dextran-induced sedimentation (Thermo Fisher Scientific) to remove contaminating erythrocytes. The resulting top layer was centrifuged at 500 g for 5 min at RT without brakes. The pellet was resuspended in 40 mL of Hanks’ balanced salt solution (HBSS) (Corning), then centrifuged on 10 mL of lymphocyte separation medium (1.077 to 1.080 g/mL; Corning) at 800 g for 20 min at RT without brakes. The remaining erythrocytes were removed by a 20 s hypotonic lysis and PMNs were resuspended at 10⁷ cells/mL in RPMI medium (supplemented with 10% FBS and 1% Pen/Strep) in presence of 0.3 U/mL adenosine deaminase (ADA) (Roche Diagnostics).

Platelets were isolated as previously described.^41–43 Briefly, the PRP was isolated and supplemented with 2% (v/v) and 20% (v/v) of initial PRP volume of 0.5 M EDTA and ACD-A, respectively, then centrifuged at 200 g for 4 min at RT without brakes to remove remaining erythrocytes. Platelets in the remaining supernatant were then pelleted following a centrifugation at 1,300 g for 10 min at RT without brakes. The platelets were resuspended at 3 × 10⁸ cells/mL in Tyrode's buffer pH 7.4 (135 mM sodium chloride, 3 mM potassium chloride, 0.3 mM disodium phosphate, 12 mM sodium bicarbonate, 20 mM HEPES, 5 mM glucose, 1 mM magnesium chloride, and 0.5 mg/mL bovine serum albumin).

For the generation of PMVs, platelets were stimulated overnight at RT in presence of 5 mM calcium chloride (CaCl₂) (Millipore Sigma) and 0.1 U/mL thrombin (Millipore Sigma) as previously reported.^24,37 The following day, the solution was centrifuged at 2,500 g for 10 min at RT to pellet remaining platelets, then the supernatant was centrifuged at 17,800 g for 90 min at RT to pellet the PMVs. The PMVs were resuspended in 0.2 mm filtered phosphate buffered saline (PBS) (Corning) and counted by flow cytometry with anti-human CD41 Brilliant Violet 421 (BioLegend) labeling (Fig. S1). All experiments were conducted using freshly isolated PMVs that were never frozen, in order to maintain mitochondrial functional integrity, unless otherwise specified.

To produce dysfunctional PMVs, fresh PMVs were subjected to a freeze-thaw cycle as previously reported.^37,44 PMV samples were frozen at −20 °C for a minimum of 24 h prior to incubation with PMNs to ensure the use of PMVs with functionally inactivated mitochondria.

2.3. PMVs interaction with immune cells

The PLB-985 cells were seeded at 10⁶ cells/mL in RPMI medium (supplemented with 10% FBS and 1% Pen/Strep) and incubated in the presence or absence of PMVs for 1 or 24 h at 37 °C. Following incubations, cells were centrifuged at 350 g for 5 min at RT and washed with PBS before further processing.

For PMN co-incubation with PMVs, we used resting PMNs to mimic circulating cells and activated PMNs to replicate cells under inflammatory conditions in the joint environment as previously described, which we have shown to be essential for optimal internalization of PMVs.²⁴ Activated PMNs were obtained with the addition of 100 U/mL tumor necrosis factor α (Cedarlane), 10 ng/mL granulocyte-macrophage colony-stimulating factor (Cedarlane) and 100 nM 12-HETE (Cayman Chemicals). Both resting and activated PMNs were first washed twice in PBS, and subsequently incubated with concentrations of 1, 10, or 100 PMVs per PMN or in the absence of PMVs for 1, 4, or 24 h at 37 °C. Following incubations, cells were centrifuged at 350 g for 5 min at RT and washed with HBSS before proceeding with further analysis.

To generate PMVs containing fluorescent mitochondria, isolated platelets were labeled with 100 nM MitoTracker Deep Red FM (MTDR) for 15 min. Platelets were then centrifuged at 1,300 g for 10 min at RT without braking and resuspended in Tyrode's buffer pH 7.4 at 3 × 10⁷ cells/mL, then stimulated overnight as previously described.

2.4. Detection of fluorescent mitochondria in immune cells

After incubation with labeled PMVs, 2 × 10⁵ cells were resuspended in 100 µL RPMI medium. Flow cytometry was then used to analyze the samples and determine the percentage of cells interacting with mitochondria of PMV origin.

For confocal microscopy, PMNs prelabeled with CellMask Orange (5 µg/mL) were washed and then incubated for 2 h in poly-L-lysine–coated wells (Millipore Sigma) at 37 °C to promote cell adhesion for the subsequent live-imaging. Images were captured using FV31S-SW software on an Olympus Fluoview FV3000 confocal microscope (Olympus Corporation) and processed using FIJI software. Mitochondria were quantified by counting the number of MTDR fluorescent particles in each cell of 3 independent experiments.

2.5. Lipid mediator production and detection by reversed-phase high-performance liquid chromatography

Lipid mediator production was measured using 5 × 10⁶ PLB-985 resuspended in 500 µL HBSS containing 1.6 mM CaCl₂ and 0.3 U/mL ADA with the stimulation initiated by adding 10 µM arachidonic acid (AA) (Cayman Chemicals) and 10 µM calcium ionophore A23187 (Millipore Sigma) then incubated for 30 min at 37 °C.³⁸ The stimulation was stopped by adding 500 µL 1:1 MeOH:acetonitrile (ACN) (v:v) solution supplemented with 200 ng/mL 19-OH-PGB₂ as the internal standard. Samples were stored at −20 °C for at least 24 h. Prior to high-performance liquid chromatography analysis, samples were centrifuged at 1,000 g for 5 min and supernatants were transferred to standard reversed-phase high-performance liquid chromatography (RP-HPLC) autosampler vials for injection.

The 5-lipoxygenase (5-LO), 12-LO, and cyclooxygenase (COX)-1 products were analyzed with the system described in Hébert et al.⁴³ with modifications. Briefly, using in-line solid phase extraction (SPE) followed by RP-HPLC with UV detection on an Agilent 1260 system equipped with an Automatic liquid sampler, Quaternary pump, Thermostated column compartment, Diode-array detector, Agilent 1290 Flexible cube equipped with G4744A/B Online SPE Direct Injection Kit (including 2 separate 2-position/10-port InfinityLab Quick Change Valves) and built-in isocratic flush pump, and an Agilent 1100 isocratic pump, based on an updated version of the protocol previously detailed.⁸⁶ With the system's valve A in its SPE configuration (position 2), valve B has the advantage of enabling the switch from loading to eluting (positions 1 and 2) (refer to Agilent P/N G4742-90110 Figs. 3 and 4 for schematics). In this configuration, while SPE 1 is connected to the analytical flow path, SPE 2 is washed and reconditioned to enable efficient back-to-back injections without the usual time penalty incurred during single SPE workflows. Successive injections alternate between loading/eluting and reconditioning between SPE 1 or SPE 2, as dictated by valve B's position.

Fig. 3.

PMVs modulate the inflammatory state of PMNs. (A) Lipid mediators released by resting or activated PMNs following a 24-hour co-incubation with or without PMVs followed by a 30-min stimulation with 1 µM thapsigargin. Samples were then processed by LC-MS/MS. (B) Inflammatory protein expression of PMNs after 4 h of incubation with PMVs. Data are shown as the mean ± SD of 3 to 4 biological replicates. Two-tailed ratio paired t test: *P < 0.05, **P < 0.01, ***P < 0.001. Ctl, control cells without PMVs; PMVs, cells incubated with 100 PMVs per cell.

Fig. 4.

PMVs modulate PMN viability. (A) Cellular viability measured by caspase-3 activity of both resting and activated PMNs after incubation with PMVs. (B) Caspase-3 activity of resting and activated PMNs incubated with fresh and dysfunctional PMVs. Data are shown as the mean ± SD of 3 to 4 biological replicates. Two-way analysis of variance followed by Dunnett's multiple comparisons on log transformed raw data with 1:0 group used as control (A). Control refers to control without PMVs. Multiple t tests (B): *P < 0.05, **P < 0.01.

Samples were injected onto the system and loaded on a Bond Elut PLRP-S in-line SPE cartridge (15 to 20 µm pore size, 2.1 internal diameter × 12.5 mm, Agilent) using 0.1% acetic acid as a loading solvent at 0.5 mL/min. In order to ensure analyte retention during in-line SPE, samples were diluted prior to reaching the SPE cartridges to <20% total organic solvent by an Agilent 1100 isocratic pump connected to the sample flow path with a High-Pressure Static Mixing Tee (U-466S; Idex Health and Science LLC) delivering 0.1% aqueous acetic acid. After loading, valve B switched position and the sample was eluted to an InfinityLab Poroshell 120 end-capped C18 column (Agilent; 4.6  internal diameter × 100 mm, 2.7 µm particle size, 120 Å pore size) and eluted based on the elution gradient and mobile phases as described in Hébert et al.⁴³

The UV spectra were recorded at wavelengths of 236 and 270 nm using the diode-array detector, and biosynthetic product identification and quantification were done using a standard mixture containing a known quantity of all the pure analytes (Cayman Chemicals) and by normalizing the peak area relative to the internal standard (19-OH-PGB₂). The 5-LO, 12-LO, and COX-1 products investigated in order of elution were 20-COOH LTB₄ (270 nm); 20-OH LTB₄ (270 nm); LTD₄, LTC₄, and LTE₄ (270 nm); LTB₄ (270 nm); 5,12-diHETE (270 nm); 12-HHTrE (236 nm), 12-HETE (236 nm); and 5-HETE (236 nm).

2.6. Lipid mediator production and detection by liquid chromatography tandem mass spectrometry

Lipid mediator production of PMNs was measured by liquid chromatography tandem mass spectrometry (LC-MS/MS). After 24 h of incubation with PMVs, 2.5 × 10⁶ cells were resuspended in 500 µL HBSS supplemented with 1.6 mM CaCl₂ and 0.3 U/mL ADA. The stimulation was initiated with the addition of 1 µM thapsigargin (Enzo Life Sciences) for 30 min at 37 °C. The stimulation was stopped by adding 250 µL 1:1 MeOH:ACN (v:v) solution supplemented with internal standards of LTB₄-d5, 12-HETE-d8, PGE₂-d4, and TxB2-d4. Samples were stored at −20 °C for at least 24 h. Prior to LC-MS/MS analysis, samples were centrifuged at 1,000 g for 5 min and supernatants were diluted to 10% organic with acidified water (0.1% acetic acid v/v) and then subjected to a solid phase extraction on C-18 columns (HyperSep C18 Cartridges; Thermo Fisher Scientific) for purification and concentration of the samples. Columns containing the samples were washed with 2 mL acidified water, and the lipid mediators were eluted using 3 mL MeOH. The solvent was evaporated using nitrogen gas and the dried sample was resuspended in 200 µL ACN:MeOH (1:1) for injection using LC-MS/MS.

The LC-MS/MS protocol for analysis of eicosanoids was adapted from Hébert et al.⁴³ Chromatographic separation prior to mass spectrometry was accomplished by gradient microscale liquid chromatography (Ultimate 3000; Thermo Fisher Scientific) utilizing a 100 mm × 1.0 mm internal diameter × 5 µm particle size (12 nm pore size) reversed-phase column (YMC-Pack Pro C18; YMC America). Mobile phase composition was as follows: A (0.1 mM HCO₂H in H₂O) and B (0.1 mM HCO₂H in 10:90 H₂O:ACN). Injection volume was 5 µL with elution at 80 µL/min beginning at 10% B mobile phase, ramping to 100% B from 0.2 to 11 min, followed by a hold at 100% B for 5 min before returning to equilibrate at 10% for 9 min (total runtime = 25 min). Postseparation analysis was conducted on a high-resolution/accurate mass quadrupole-Orbitrap mass spectrometer using a pneumatically-assisted electrospray ionization ion source operating in negative ion mode (Q-Exactive; Thermo Fisher Scientific) at maximum resolution of 140,000 over a mass-charge range of 200 to 700 m/z. The mass spectrometer was calibrated using the commercial negative ion calibration mixture from Thermo Fisher Scientific prior to batch analysis. Interface conditions on the Q-Exactive were as follows: sheath, auxiliary, and sweep gas flow rates (5, 3, and 0 AU); ionspray voltage (3.70 kV); heated capillary voltage (275 °C); and S-lens RF (100 V). Data were processed using Xcalibur 3.0.63 software with analyte identity confirmed by retention time, exact mass, and diagnostic ion masses (MRM, N₂ collision gas with normalized collision energy as reported in Table S1) as compared with pure and deuterated standards (Cayman Chemicals, USA). Observed retention times, parent ion mass, and diagnostic fragment ion mass transition (parent m/z > fragment m/z) are reported in Table S1. TXB₂ and TXB₂-d₄ present as 2 peaks in accordance with its anomeric isomerism. Quantification of analytes was conducted by comparing area of biogenic eicosanoids with their deuterated internal standards (12.5 ng/sample) or suitable deuterated surrogate as indicated in Table S1.

2.7. Protein expression quantification

PMNs (4 × 10⁶ cells) were resuspended in 400 µL lysis buffer (50 mM Tris-hydrochloride, 150 mM sodium chloride and 2 mM EDTA at pH 7.6) and boiled in a water bath for 10 min. Samples were then homogenized with a 21-gauge needle. For PLB-985, 2 × 10⁶ cells were washed twice with PBS and resuspended in 200 µL NP40 (Alfa Aesar) and incubated 15 min on ice, then centrifuged at 10,000 g for 10 min at 4 °C. Protein immunoblotting analysis was performed by loading the protein content of 2 × 10⁵ cells per well (18 wells 4% to 12% Criterion TGX Stain-Free Precast Gels; Bio-Rad). For PLB-985 cells, stain free of the polyvinylidene fluoride or polyvinylidene difluoride (Millipore Sigma) membrane was used for normalization by total protein content. For PMNs, histone H3 (4499S; Cell Signaling Technology; 1:1,000) expression was used for normalization, as histone H3 expression is absent in anucleate PMVs. Expression of 5-LO (C49G1; Cell Signaling Technology; 1:1,000), 12-LO (PAB965Hu01; Cloud Clone; 1:750), COX-1 (13393-1-AP; Proteintech; 1:1,000), and histone H3 were detected following coupling with horseradish peroxidase–coupled anti-rabbit (70745; Cell Signaling Technology; 1:1,000) secondary antibodies. All the primary antibodies were prepared in Tris-buffered saline with Tween and contained 5% bovine serum albumin and 0.02% sodium azide.

2.8. Fractional rate of protein synthesis

The fractional rate of protein synthesis was measured using the flooding dose assay modified for using stable isotopes.^45,46 Briefly, 4 × 10⁶ PLB-985 were incubated for 1 or 24 h with PMVs in RPMI medium supplemented 1.5 mM phenylalanine (PHE), of which 50% was deuterated (ring-D₅-phenylalanine; Cambridge Isotope Laboratories). Following the incubation period, the cells were centrifuged and the pellets were stored at −80 °C until analysis. The specific enrichment of D₅-PHE was measured as previously described.^45,46 Briefly, samples were homogenized by 3 rounds of sonification in perchloric acid 0.2 M and subjected to overnight acid digestion in HCl 6 M; the supernatant of the first homogenization was kept at −20 °C for further analysis as a protein-free pool of PHE. Solid phase extraction (Bond-Elut-C18, 100 mg, 1 mL; Agilent) was performed on both protein and protein-free pool, and the extracted PHE was eluted with 30% MeOH and evaporated to dryness at 110 °C for 2 h. The extracted PHE was then derivatized with pentafluorobenzyl bromide before being analyzed by gas chromatography coupled to a mass spectrometer. The system consisted of an Agilent gas chromatograph (model 6890N) interfaced with a single quadrupole inert mass selective detector (MSD; model 5973). The rate of protein synthesis was then calculated as indicated in Lamarre et al.⁴⁵ The fractional rate of protein synthesis is presented as % d⁻¹.

2.9. Caspase-3 activity assay

After the incubation with PMVs, 10⁶ PMNs were washed, and the pellets were stored at −80 °C for 24 to 48 h until further analysis. Caspase-3 activity was evaluated with the EnzChek Caspase-3 Activity Assay Kit #1 (Thermo Fisher Scientific) following the manufacturer's protocol. Briefly, cells were resuspended in 50 µL lysis buffer and incubated 30 min on ice. Samples were then centrifuged at 2,300 g for 5 min, and supernatant was transferred to a 96-well plate and 50 µL of the substrate working solution (containing the caspase-3 substrate Z-DEVD-AMC) was added to the wells. After 30 min of incubation in the dark at RT, fluorescence of AMC was read at 342/441 nm (ex/em) on a microplate reader (Synergy H1; BioTek, Thermo Fisher Scientific). Relative fluorescent units (RFU) were used to calculate the fold change compared with the control (1:0) condition.

2.10. ATP quantification

To quantify ATP production, 4 × 10⁵ PMNs were resuspended in 1% Triton X-100 and incubated for 10 min at RT. The cell lysate was then centrifuged at 500 g for 5 min at RT and the supernatant was used to quantify ATP levels with the ATP determination kit (A22066, Invitrogen) using the firefly luciferase-based reaction according to the manufacturer's protocol. Briefly, 10 µL of a standard ATP curve in triplicates and 10 µL samples in technical duplicates were added to a 96-well plate and a microplate reader injected 90 µL of the luciferine-luciferase reagent then read luminescence after 1 s. The concentration of experimental samples was determined using the standard curve samples.

2.11. Mitochondrial membrane potential

Mitochondrial membrane potential was measured using TMRM (Thermo Fisher Scientific) by flow cytometry. Following incubation with PMVs, 3 × 10⁵ PMNs were washed and stained with 100 nM TMRM for 30 min and analyzed on the flow cytometer for fluorescence intensity.

2.12. High-resolution respirometry

To measure mitochondrial respiration, 5 × 10⁶ PMNs were resuspended in MiR05 (0.5 mM ethylene glycol tetraacetic acid, 3 mM magnesium chloride, 60 mM K-Lactobionate, 20 mM Taurine, 10 mM potassium dihydrogen phosphate, 20 mM HEPES, 110 mM D-sucrose, and 1 g/L bovine serum albumin, pH 7.1)⁴⁷ respiration buffer to measure mitochondrial oxygen consumption as previously described.⁴⁸ Briefly, cells were added to high-resolution respirometer (Oxygraph-2k; Oroboros Instruments) chambers and basal respiration was measured. To obtain LEAK respiration, oligomycin (2.5 µM) was added to inhibit the mitochondrial complex V. A titration of CCCP (0.5 µM per step) was performed to determine the maximal capacity of the electron transport system (ETS). To determine the residual oxygen consumption, rotenone (0.5 µM), and antimycin A (2.5 µM) were added to inhibit both complex I and complex III, respectively. All reagents were obtained from Millipore Sigma.

2.13. Mitochondrial metabolic activity

Mitochondrial complex 1 activity was assessed as previously described in Veilleux et al.³⁷ The mitochondrial complex I activity was measured using the CellTiter-Blue kit (CTB) (Promega) according to the manufacturer's protocol. The CTB reagent measures the level of metabolically active cells by reducing redox dye (resazurin) into a measurable fluorescent product (resorufin). Briefly, following incubation with PMVs, 10⁶ PMNs were resuspended in 100 µL complete RPMI medium and 20 µL CTB buffer was added. Cells were incubated for 1 h at 37 °C and the fluorescence was read at 560/590 (ex/em) with a microplate reader. RFU were used to calculate the fold change compared with the control (1:0) condition.

2.14. Transmission electron microscopy

Samples were fixed in 2.5% glutaraldehyde (Sigma-Aldrich) in PBS at 4 °C for 2 h, followed by postfixation in 1% osmium tetroxide (Sigma-Aldrich) at 4 °C for an additional 2 h. The preparation was centrifuged at 17,800 g for 90 min to pellet extracellular vesicles. Pellets were resuspended in PBS and processed at Dalhousie University's EM Facility Core for dehydration, sectioning, and imaging. Briefly, samples were dehydrated through a graded acetone series and embedded in Epon-Araldite resin. Ultrathin sections (100 nm) were cut using an ultramicrotome, mounted on 300-mesh copper grids, stained with 2% uranyl acetate, rinsed, treated with lead citrate, and air-dried. Imaging was performed using a JEOL JEM-1230 transmission electron microscope at 80 kV, with images captured via a Hamamatsu ORCA-HR digital camera.

2.15. Statistical analysis

Two-way analysis of variance (ANOVA) and mixed-effect analysis followed by Tukey's or Dunnett's multiple comparison analysis, ratio t test, multiple t tests, and Student's t test were performed as described in the figure legends. Outliers were determined using Grubb's outlier's test. Normality of data was tested using Shapiro-Wilk normality tests or visualized using QQ plot of residuals. Data that did not meet the criteria for normal distribution were log-transformed, as specified in the figure legends. Data was analyzed using GraphPad Prism software (version 10; GraphPad Software). P values of <0.05 were considered significant.

3. Results

3.1. PMVs interact with PLB-985 and modulate the inflammatory phenotype by horizontal transfer of inflammatory enzymes

We initially aimed to determine if PMVs can interact with the neutrophil-like PLB-985 cell line. We measured the cells’ interaction with MTDR-labeled mitochondria from PMVs by flow cytometry. PLB-985 cells co-incubated with PMVs prelabeled with MTDR were found to be MTDR positive, with 18.7% and 53.1% association after 1 h and 24 h, respectively (Fig. 1A). Because we confirmed that PMVs interact with our cell line, we then measured several PLB-985 and platelet-specific lipid mediators by RP-HPLC. PLB-985 cells generate lipid mediators via the 5-LO pathway.³⁸ Because leukotrienes are important contributors to inflammatory diseases,² we evaluated the effects of PMVs on the ability of PLB-985 to generate lipid mediators. The PMVs affected the production of LTB₄ in PLB-985 with a slight increase after 1 h from 0.60 to 0.68 ng/10⁶ cells and an increase after 24 h of incubation from 0.64 to 0.87 ng/10⁶ cells (Fig. 1B). While PMVs did not affect LTC₄ and LTD₄ production following a 1 h co-incubation, the production was increased from 20.3 to 23.6 ng/10⁶ cells and from 11.4 to 15.7 ng/10⁶ cells, respectively, after a 24 h incubation with PMVs (Fig. 1B). Other 5-LO products were also detected, such as LTE₄ and 5-HETE; however, their production was not affected by incubation with PMVs (Fig. 1B). A 7.9-fold increase in 12-HETE (0.05 to 0.44 ng/10⁶ cells) was observed following 1 h of incubation with PMVs and a nonsignificant 2.8-fold (from 0.04 to 0.12 ng/10⁶ cells) increase after 24 h. We observed a 1.3-fold increase (from 0.58 to 0.73 ng/10⁶ cells) in the COX product 12-HHTrE following 24 h of incubation with PMVs. The 15-LO product 15-HETE was also detected but not affected by the presence of PMVs. Overall, these results suggest that PMVs increase the production of inflammatory lipid mediators in PLB-985.

Fig. 1.

Mitochondria from PMVs are transferred to and modulate the inflammatory phenotype of PLB-985 cells. (A) Mitochondria from PMVs interact with PLB-985 as determined by flow cytometry following 1 (left) or 24 (right) hours of incubation with MTDR-labeled PMVs. (B) Quantification by RP-HPLC of lipid mediators released by PLB-985 following co-incubation with PMVs. After 1 or 24 h of incubation with PMVs, PLB-985 was stimulated for 30 min with 10 µM AA and 10 µM of A23187. (C) Inflammatory protein expression of PLB-985 after 1 h (left) or 24 h (right) of incubation with PMVs with a representative image of 4 biological replicates. Densitometry analysis was normalized with total protein using total membrane protein (stain-free images). (D) Rate of protein synthesis per day of PLB-985 cells after 1 h (left) or 24 h (right) of incubation with PMVs determined by measuring uptake of D₅-PHE by gas chromatography mass spectrometry analysis. Data are expressed as mean ± SD of 3 to 4 biological replicates. Two-tailed paired t test; for panel B, t tests were performed individually on data from each time point: *P < 0.05, **P < 0.01, ***P < 0.001. Ctl, control cells without PMVs; PMVs, cells incubated with 100 PMVs per cell.

After demonstrating that both the 5-LO and 12-LO pathways increase in PLB-985 following incubation with PMVs, we explored whether alteration in their protein expression was also observed. We first evaluated the 12-LO, a platelet-specific enzyme absent in human PMNs, and is responsible for the conversion of AA into 12-HETE (more specifically 12(S)-HETE). We confirmed that 12-LO was not present in cells incubated in the absence of PMVs; however, protein expression was detected after incubation with PMVs as early as 1 h postincubation (Fig. 1C). The COX-1 protein, which is constitutively expressed in both platelets and PLB-985, was increased by 1.5- and 1.3-fold after 1 and 24 h of incubation with PMVs, respectively (Fig. 1C). The expression of 5-LO remained unchanged following incubation with PMVs (Fig. 1C). We confirmed that the total protein synthesis rate was not affected by PMVs treatment (Fig. 1D). Overall, our result confirms that PMVs can modulate the inflammatory state of PLB-985 by transferring enzymes, such as the platelet-specific 12-LO, involved in the biosynthesis of proinflammatory lipid mediators.

3.2. Mitochondria from platelets associate with human PMNs

Having demonstrated that PMVs can modulate the inflammatory phenotype of a neutrophil-like cell line in vitro, we sought to replicate these observations in an ex vivo approach using freshly isolated PMVs and human PMNs. First, we needed to confirm that PMVs interact with PMNs under our experimental conditions as previously reported.²⁴ Resting and activated PMNs were incubated in presence of 3 different ratios of PMVs prelabeled with MTDR and association was quantified by flow cytometry. Resting PMNs mimic circulating PMNs, while activated PMNs represent cells in an inflammatory environment, such as the joint of RA patients.²⁴ An increased interaction between PMNs and PMVs was observed, primarily with the 1:100 (PMN:PMV) condition, which resulted in a 59% and 65% interaction after 4 and 24 h, respectively, in resting cells and a 79% and 69% interaction after 4 and 24 h, respectively, in activated cells (Fig. 2A). An increase in the interaction was observed in the 1:100 condition compared with the control and 1:1 condition after 4 h in both resting and activated cells (Fig. 2A). Additionally, the interaction in the 1:100 condition was increased compared with all other conditions after 24 h in both resting and activated cells (Fig. 2A).

Fig. 2.

Mitochondria from PMVs interact with PMNs. (A) Mitochondrial interaction with resting and activated PMNs determined by flow cytometry following incubation of PMNs with MTDR-labeled PMVs. (B) Representative image of CellMask Orange–labeled PMNs (red) incubated without PMVs from 4 biological replicates (top). Representative image of MTDR-labeled mitochondria (blue, indicated with white arrows) from PMV origin interacting CellMask Orange–labeled PMNs (red) from 4 biological replicates (bottom). (C) Quantification of PMV mitochondria per cell (n = 85). (D) Transmission electronic microscopy imaging of our preparation of platelet-derived microparticles. Data are expressed as the mean ± SD of 3 biological replicates. (A) Two-way analysis of variance followed by Tukey's multiple comparisons: *P < 0.05, **P < 0.01. Ctl, control cels without PMVs; PMVs, cells incubated with 100 PMVs per cell.

To further validate this interaction, we labeled the PMN cell membrane with CellMask Orange and visualized, using live cell imaging, the platelet-derived mitochondria's interaction with PMNs (Fig. 2B). MTDR-labeled mitochondria were absent in our control condition (1:0) while on average 3 mitochondria from platelet were detected in recipient PMNs (1:100) (Fig. 2C). Taken together, these results indicate that platelet-derived mitochondria associate with human PMNs.

3.3. PMVs transfer functional inflammatory proteins to PMNs

PMVs modulate the inflammatory phenotype of the PLB-985 cell line (Fig. 1). Thus, we tested whether PMVs would impact similar modulatory changes in human PMNs by quantifying the lipid mediator production (Fig. 3A). A 2.9-fold increase in 12-HETE, a product primarily generated via the platelet-specific 12-LO enzyme, was observed in resting cells incubated with PMVs (Fig. 3A). Other products generated by platelet-specific enzymes such as COX-1 also displayed an increase in production such as TxB₂ (3.3-fold in resting cells and 3.5-fold in activated cells) and PGE₂ (2.5-fold in resting cells and 1.6-fold in activated cells) when PMNs were incubated with PMVs (Fig. 3A). Of interest, an increase in 5,12-diHETE, a product requiring the combined action of the PMN’s 5-LO and the platelet's 12-LO,⁴⁹ was also observed in both resting and activated PMNs (6.7-fold in resting cells and 3.0-fold in activated cells) (Fig. 3A). The proinflammatory LTB₄ and its by-products were significantly increased by 3.9-fold in the resting cells when incubated with PMVs (Fig. 3A).

Having detected a significant increase in lipid mediators produced by platelet-specific enzymes in both PLB-985 and human PMNs in presence of PMVs, we then investigated if the respective enzymes could be transferred from PMVs to the recipient cell. We observed the transfer of both the platelet-specific 12-LO and COX-1 enzymes from PMVs to PMNs after only 4 h of co-incubations (Fig. 3B). However, 5-LO expression does not seem to be affected by incubation with PMVs (Fig. 3B).

3.4. PMVs modulate the viability of human PMNs

Given the importance of mitochondria in the cell viability,⁸ and that PMNs have a relatively short lifespan (8 to 12 h),¹⁴ we investigated whether freshly isolated PMVs could modulate the lifespan of the recipient cell. We initially examined the effects of PMVs on the PMN viability with the EnzChek caspase-3 activity assay kit. No differences in PMN viability were observed after 1 and 4 h of incubation in both resting and activated conditions (Fig. 4A). However, after 24 h of incubation with PMVs, a tendency toward a decrease in caspase-3 activity was found in the 1:100 ratio resting condition while a significant decrease of 13% and 27% in the 1:10 and 1:100 ratios of the activated conditions, respectively, were observed. Finally, because apoptosis can be regulated by mitochondrial-dependent pathways, the caspase-3 activity assay was also performed on PMNs incubated with dysfunctional PMVs (Fig. 4B). In contrast to our results of fresh PMVs, there was a significant increase in caspase-3 activity in the dysfunctional PMVs condition with activated PMNs, as we observed a 13% and 19% increase in the 1:10 and 1:100 conditions, respectively. Taken together, our ex vivo approach confirms that PMVs modulate the lifespan of the recipient cell, in a mitochondria-dependent mechanism.

3.5. Mitochondrial integrity is necessary for the modulation the bioenergetic phenotype of PMNs

Given that mitochondria are transferred from PMVs to PMNs, we aimed to evaluate whether the mitochondria uptake could alter the bioenergetic phenotype of the recipient cell. We first measured ATP levels and observed an increase in resting PMNs after 1 h, rising from 0.09 µM to 1.03 µM, and after 24 h, from 0.17 to 0.34 µM with the 1:100 condition, compared with cells incubated without PMVs (Fig. 5A). Additionally, we observed that in resting PMNs, the 1:100 condition showed overall higher ATP levels compared with cells incubated without PMVs (Fig. 5A). However, co-incubation of activated PMNs with PMVs did not affect total ATP levels (Fig. 5A). To further evaluate mitochondrial physiology in the recipient cells, we assessed mitochondrial membrane potential using TMRM staining. We observed a dose-dependent increase in TMRM fluorescence of 14%, 24%, and 33% at PMN:PMV ratios of 1:1, 1:10, and 1:100, respectively. Activated PMNs exhibited more modest increases increase of 8% and 16% in TMRM fluorescence at PMN:PMV ratios of 1:10 and 1:100, respectively (Fig. 5B), indicating enhanced mitochondrial polarization. Finally, the cellular respiration was investigated by high-resolution respirometry using the Oroboros instrument. We observed a 2-fold increase in the ETS capacity of activated PMNs incubated with PMVs for 4 h (Fig. 5C).

Fig. 5.

PMVs modulate the bioenergetic state of PMNs. (A) ATP levels of resting and activated PMNs following incubation with fresh PMVs. (B) Increase of mitochondrial membrane potential of PMNs after incubation with fresh PMVs during 4 h measured with TMRM. (C) Mitochondrial respiration of resting and activated PMNs following incubation with or without fresh PMVs for 4 h as measured by high-resolution O₂ respirometry. (D) Mitochondrial respiration of resting and activated PMNs following incubation with or without dysfunctional PMVs after 4 h measured using high-resolution O₂ respirometry. Components of mitochondrial respiration states were evaluated as followed: routine (basal) respiration, LEAK (ATP synthase inhibition), and the maximal capacity of ETS. (E) Mitochondrial complex I redox activity of PMNs after incubation with various PMN:PMV ratios using fresh and dysfunctional PMVs determined by CellTiter Blue assay. Data are presented as RFU fold change with 1:0 ratio used as a control. Data are shown as the mean ± SD of 3 to 4 biological replicates. Two-way analysis of variance followed by Tukey's multiple comparisons were performed on square root transformed data (A resting) or on data (A activated). Multiple t tests on data (C resting, E [except 24 h activated]), on log-transformed data (C activated), on square root data (D), or on data squared (E [24 h activated]): *P < 0.05, **P < 0.01. Values without at least 1 common superscript are significantly different (P < 0.05).

Since the presence of PMVs seemed to impact PMN metabolism, we sought to investigate the importance of the mitochondrial integrity of mitoMPs. We generated dysfunctional mitoMPs by subjecting them to a freeze-thaw cycle³⁷ and subsequently performing co-incubation with PMNs. We initially evaluated the effects of dysfunctional mitoMPs on the cellular respiration. In contrast to fresh preparation of mitoMPs, the frozen-thawed preparation did not modulate respiration of both resting or activated PMNs (Fig. 5C, D). The mitochondrial metabolism was also assessed via the mitochondrial complex I redox activity using a CTB assay. In activated PMNs, we observed a decrease in redox activity after incubation with dysfunctional PMVs compared with fresh PMVs in the 1:10 condition of 47% after 1 h and 43% after 4 h (Fig. 5E). In resting PMNs, there was a 33% decrease in redox activity after 24 h with the 1:100 condition between the fresh and the dysfunctional PMVs (Fig. 5E). Overall, our results indicate that PMVs containing functional mitochondria modulate key physiological bioenergetic process in PMNs.

4. Discussion

PMVs are extracellular vesicles released by platelets following activation known to have a role in inflammation and immunity.⁵⁰ They are active participant in intercellular communication by shuttling bioactive content between cells.⁵¹ In colon cancer, PMVs transfer the platelet-specific 12-LO enzyme to the cancer cells. The recipient cells then acquire the ability to generate the inflammatory lipid mediator 12-HETE, a known mediator implicated in membrane permeabilization.⁵² PMVs are also implicated in the initiation and progression of inflammatory diseases, most notably in RA.¹⁹ More importantly, PMVs’ blood concentration has been associated with the disease severity of RA.⁵³ PMNs are an important cell type that are not only vital for the normal immune response, but also active contributors in autoimmune diseases such as RA.^2,3 It has been previously reported that PMVs interact with PMNs,²⁴ resulting in a transfer of platelet-specific organelles, such as mitochondria, into the recipient cells.²⁶ However, whether this transfer of mitochondria could alter the lifespan of the PMNs and consequently modulate the inflammatory phenotype of the recipient cell remained uninvestigated until now. In this study, we demonstrated important physiological changes to the PMNs following the uptake of PMVs and their functional mitochondria.

Studying human PMNs ex vivo can be somewhat challenging due in part to the cells’ short lifespan. Therefore, the PLB-985 cell line is a reasonable in vitro alternative to generate a neutrophil-like cells and obtain preliminary data.³⁸ We first confirmed that the inflammatory enzyme 12-LO, a platelet-specific enzyme not found in PMNs, was transferred to PLB-985 cells. Also, increased production of 12-HETE was observed in PLB-985 incubated with PMVs. Platelets are known to transfer inflammatory enzymes, such as 12-LO, to several cell types^52,54; however, this is the first report of a specific transfer of 12-LO in the neutrophil-like PLB-985 cells. In addition to the 12-LO, platelets widely express the COX-1 constitutive enzyme. PMVs also contributed to a modest COX-1 increase in PLB-985. In contrast to 12-LO, we found that COX-1 is expressed at the protein level in PLB-985. This is expected, as COX-1 is a known constitutive protein⁵⁵; therefore, its increase may be attributed to an external source present in PMVs.²⁴ However, unlike 12-LO, the COX-1 activity became more evident after 24 h of incubation with PMVs. This suggests that a secondary mechanism is involved in regulating COX-1 expression, such as a potential increase in transcription or even direct COX-1 RNA transfer from PMVs. While COX-1 results were intriguing, further studies will be needed to elucidate the mechanism involved in the modulation of the enzyme in PLB-985. Of interest, a significant increase of the 5-LO pathway was observed in PLB-985. More specifically, an increase in LTB₄ after 1 h and LTB₄ and LTD₄ after 24 h of incubation was observed. Given the importance of the 5-LO enzyme in chronic inflammatory diseases such as RA,² asthma,⁵⁶ and atherosclerosis,⁵⁷ the contribution of PMVs in promoting the production of neutrophil-like specific lipid mediators demonstrates the important role of extracellular vesicles in regulating the inflammation response.

Although PLB-985 serves as an alternative model of human PMNs for generating preliminary data, we ultimately validated our hypothesis using primary cells. We obtained similar results for lipid mediator production in both our in vitro and ex vivo models. For instance, the incubation with PMVs affected the production of LTB₄ in resting cells and did not affect the production of 5-HETE. These products, detected in all our experimental conditions, are potent chemotactic agents for PMNs with crucial roles in the recruitment of the immune cells to the inflammatory sites.^58,59 Interestingly, as observed in PLB-985, the products from platelet-specific enzymes such as 12-LO and COX-1 were significantly increased after incubations with 100 PMVs per cell. The 12-LO pathway is not active in PMNs, in contrast to PMVs, which suggests an enzyme transfer between these populations, as previously shown in other cell types such as colon cancer cells⁵² and mast cells.⁵⁴ The baseline detection of 12-HETE, TxB₂, and PGE₂ in our experiments could be explained by the presence of platelets in our PMN isolations. It has been previously reported that platelets can bind to PMNs in the blood circulation.⁶⁰ In fact, between 10% and 40% of our PMN isolation was positive for CD41 expression, a glycoprotein IIB/IIa integrin specifically expressed on the platelet's surface, before co-incubation with PMVs (Fig. S2). To confirm that the increase in the production of platelet-specific products was the result of a protein transfer, we investigated the expression of 12-LO and COX-1, 2 enzymes found in PMVs but not in PMNs. We detected both proteins in PMNs following incubations with PMVs but not in the control conditions. The transfer of specific protein content between immune cells could have the potential to modulate the inflammatory response of PMNs. For instance, the lipid mediator 5,12(S)-diHETE is produced by the collaborating activity of the 5-LO and 12-LO enzymes.⁴⁹ While the 5,12(S)-diHETE is abundantly found in inflammatory synovial fluid from RA patients,⁶¹ its physiological function in the inflammatory response remains elusive. Interestingly, we detected a significant increase in the production of this lipid mediator after the incubation of PMNs with PMVs. These results suggest that the transfer of the platelet-specific 12-LO into PMNs could mediate the production of 5,12(S)-diHETE.

While our study utilized thapsigargin stimulation in the absence of exogenous AA, we do not assume that phospholipase activity alone is sufficient to drive robust lipid mediator biosynthesis. Thapsigargin induces intracellular calcium mobilization, which activates cytosolic phospholipase A₂,⁶² leading to the release of endogenous AA from membrane phospholipids.⁶³ This liberated AA then serves as a substrate for downstream enzymatic pathways, including 5-LO, 12-LO, and COX-1, resulting in the production of leukotrienes, HETEs, and prostaglandins. Our findings suggest that PMVs contribute to this cascade not only by enhancing mitochondrial bioenergetics but also by transferring platelet-specific enzymes such as 12-LO and COX-1, thereby amplifying the inflammatory output of PMNs. Future studies should further dissect the interplay between phospholipase activation, AA availability, and vesicle-derived enzymatic contributions to clarify the full scope of this pathway.

Because PMVs contain a subpopulation that retains functional mitochondria, known as mitoMPs,²⁶ we investigated whether organelle transfer would occur under our experimental setting. The intercellular transfer of mitochondria has previously been shown to alter the phenotype of recipient cells. In fact, horizontal mitochondria transfer has been found to enhance resistance to chemotherapy in acute myeloid leukemia⁶⁴ and to restore mitochondrial function in receiving cells.^65,66 In addition, uptake of mitochondrial content improved the viability of cardiomyocytes^67,68 as well as of neurons.^69,70 Therefore, the organelle transfer can serve as a double-edged sword in the inflammatory response. While the proinflammatory phenotype of PMNs in response to pathogens is beneficial to prevent uncontrolled infections, an extended PMN lifespan leading to prolonged release of inflammatory mediators can be harmful to certain tissues. Such a phenomenon is observed in the synovial fluid of RA patients, whereas PMNs have displayed delayed apoptosis compared with their usual lifespan of 6 to 8 h.^71,72 The exact mechanism for this increased viability remains elusive, although many antiapoptosis cytokines are thought to be involved.⁷³ Our results demonstrate that platelet-derived mitochondria have a role in decreasing caspase-3 activity in PMNs, suggesting a potential mechanism in modulating the viability of PMNs at inflammatory sites. Interestingly, when PMNs were co-incubated in presence of PMVs with dysfunctional mitochondria, we observed a significant increase in the caspase-3 activity compared with cells incubated with a fresh isolation of PMVs containing functional mitochondria. The increase caspase-3 activity is in line with base line activity of PMNs incubated in absence of PMVs. These results further support the importance of the mitochondrion integrity in modulating the PMN lifespan. Noteworthy, we encountered a significant technical challenge with our initial experimental approach to assess PMN viability after co-incubation with PMVs. Several standard viability tests could not be used during this study due to the presence of PMVs. The presence of PMVs lacking functional mitochondria expressing surface phosphatidylserine³⁷ on PMNs posed a significant challenge, as these PMVs can adhere to the PMN's surface and interfere with accurate Annexin V-based detection of apoptosis. This confounding factor limited the reliability of these assays in our experimental context, reinforcing our decision to rely on caspase-3 activity as a more specific and robust measure of cellular viability. Additional assays evaluating the mitochondria functions such as MTT or CTB are not accurate in determining the PMN's viability because these tests, measuring mitochondrial activity, also detect mitochondrial content in mitoMPs. Consequently, we investigated the caspase-3 activity in the recipient cells.

The observed increase in neutrophil viability following mitochondrial transfer from PMVs has important implications for the pathophysiology of RA. PMNs are among the first immune cells recruited to inflamed joints, where they contribute to tissue damage through the release of reactive oxygen species, cytokines, and lipid mediators.^1,2 Under normal conditions, their short lifespan serves as a regulatory mechanism to limit excessive inflammation. However, in RA, PMNs exhibit delayed apoptosis, which exacerbates chronic inflammation.¹⁴ Our findings suggest that the uptake of functional mitochondria from PMVs enhances neutrophil survival by decreasing caspase-3 activity and increasing ATP production. This prolonged lifespan may allow PMNs to sustain their proinflammatory activity within the synovial fluid, thereby amplifying joint damage and disease severity. These results underscore the potential role of platelet-derived mitochondria as modulators of immune cell longevity and highlight a novel mechanism by which platelet-neutrophil interactions may contribute to the persistence of inflammation in RA. Targeting this pathway could offer new therapeutic strategies aimed at restoring neutrophil turnover and mitigating chronic inflammation.

While the precise mechanism by which PMNs internalize PMVs and mitoMPs remains to be elucidated, several conceptual pathways can be considered based on existing literature. Previous studies have implicated receptor-mediated endocytosis, macropinocytosis, and membrane fusion as potential mechanisms for mitochondrial transfer between cells.^74,75 In the context of PMVs, the presence of surface phosphatidylserine and specific integrins may facilitate recognition and uptake by PMNs, possibly involving scavenger receptors or phospholipid-binding proteins.⁷⁶ Additionally, the concerted activity of 12-LO and secreted phospholipase A2-IIA has been shown to mediate PMV internalization in PMNs,²⁴ suggesting a lipid signaling–dependent mechanism. Once internalized, the trafficking of mitochondria within PMNs may involve cytoskeletal elements such as microtubules and motor proteins like Miro1, which have been implicated in mitochondrial transport in other cell types.⁷⁷ Although our study did not directly investigate these pathways, future research should aim to delineate the molecular machinery governing mitochondrial uptake and intracellular trafficking in PMNs, as this could uncover novel therapeutic targets for modulating immune cell function in inflammatory diseases.

A key physiological function of mitochondria is ATP generation. We observed increased ATP levels in resting PMNs, while no significant changes were detected in the activated population. A possible explanation is that when PMNs are incubated in an inflammatory environment, their basal ATP levels increase. Therefore, the effects of platelet-derived mitochondria on ATP levels could be negligible in comparison with the effects of the inflammatory environment. In addition, several physiological functions of activated PMNs require ATP, such as NET formation,⁷⁸ phagocytosis,⁷⁹ and lipid mediator biosynthesis,^80,81 which would consume most of the ATP generated. However, when investigating the cellular respiration in activated PMNs, the maximal ETS capacity was increased. Given that PMNs have very low detectable mitochondrial oxygen consumption,⁹ this observation suggests that platelet-derived mitochondria are functional in the recipient cell. To confirm that these mitochondrial phenotype changes were partially induced by the transfer of functional mitochondria from PMVs, we co-incubated PMNs with dysfunctional platelet-derived mitochondria.^37,44 In the presence of dysfunctional PMVs, we observed no significant differences in mitochondrial respiration between the control cells and those incubated with PMVs. The complex I activity in PMNs was also unaffected by dysfunctional PMVs, in contrast to our fresh preparation of functional PMVs. Our results align with previous studies demonstrating the importance of mitochondrial integrity in mitoMPs for modulating the bioenergetic and malignant characteristics of various cancer cell lines.^35,37,82 In addition to shedding light on the role of platelet-derived mitochondria in modulating the bioenergetic phenotype of PMNs, our study highlights the importance of proper preparation and storage of PMVs for subsequent applications. Further studies are required to investigate whether the internalization mechanism differentiates between fresh and dysfunctional PMVs.

We acknowledge that prolonged activation may influence vesicle composition, particularly in terms of lipid mediators or surface marker expression. However, we have consistently used this activation protocol across multiple studies.^24,35,37,83 Subsequent studies should focus on comparing the cargo and function of PMVs generated under varying activation durations and agonist combinations to further refine the biological relevance of these conditions. Nevertheless, based on our current standardized approach, we have full confidence in the conclusions drawn from our experimental model. We acknowledge the potential impact of individual donor variability on the effects of PMVs on PMNs. In our study, PMVs were isolated from multiple healthy, consenting donors, and all experiments were conducted using freshly prepared PMVs to preserve mitochondrial integrity. Although PMVs were obtained from multiple healthy donors and consistent trends were observed across biological replicates, we did not stratify the data by individual donor or quantify interdonor variability. Given that platelet function and PMV composition can vary due to genetic, environmental, and lifestyle factors, this represents a limitation in our study design. We recommend that future investigations include donor-specific analyses and larger sample sizes to better understand the extent of heterogeneity and its impact on immune cell modulation.

An important consideration in interpreting our findings is the use of the PLB-985 cell line as a model for neutrophil-like behavior. While PLB-985 cells can be differentiated to exhibit several functional characteristics of human PMNs, including phagocytosis and reactive oxygen species production, they remain an immortalized myeloid leukemia line and may not fully recapitulate the phenotype or responsiveness of primary human PMNs. Notably, differences in surface receptor expression, signaling pathways, and vesicle uptake capacity may influence the extent and nature of interactions with PMVs. We selected this model to enable controlled, reproducible studies of PMV internalization, but we recognize its limitations and are currently extending these findings to primary human PMNs to validate the physiological relevance of our observations.

While our study utilized PMVs isolated from healthy donors to establish a baseline model for mitochondrial transfer and neutrophil modulation, we acknowledge that PMVs derived from individuals with inflammatory conditions may differ substantially in composition, bioactivity, and cargo.⁸⁴ Inflammatory environments are known to alter platelet activation profiles, which can influence the molecular content of PMVs, including cytokines, lipid mediators, and surface markers.^83,85 Therefore, the findings presented here may not fully capture the complexity of PMV-neutrophil interactions in pathological settings such as RA. Future studies should incorporate PMVs from RA patients or other disease cohorts to assess disease-specific effects and validate the relevance of our observations in clinical contexts.

Our findings reveal the critical role of platelet-derived mitochondria in modulating the bioenergetic and inflammatory phenotypes of PMNs, highlighting their potential implications in inflammatory diseases such as RA. The transfer of functional mitochondria and inflammatory enzymes like 12-LO enhances PMN survival and lipid mediator production, suggesting a mechanism that could exacerbate chronic inflammation. Future investigations using PMVs derived from individuals with different inflammatory or pathological conditions, such as RA, systemic lupus erythematosus, or cancer, could reveal important disease-specific variations in extravesicle cargo and function. Such studies would further clarify the role of PMVs in modulating neutrophil responses in the context of disease and may uncover novel targets for therapeutic intervention.

Supplementary material

Supplementary material is available at Journal of Leukocyte Biology online.

Funding

This work was supported by the Canadian Institutes of Health Research under grant 511840 and Research New Brunswick under grant RAI_2024_006.

Conflicts of interest. The authors declare that they have no conflicts of interest relevant to this article.

Data availability

Data will be made available upon reasonable request.