Alternative Methods for the Detection of Superoxide Anion Generation in Platelets

Summary

The generation of superoxide anion is essential for the stimulation of platelets and, if dysregulated, critical for thrombotic diseases. Here, we present three protocols for the selective detection of superoxide anions and the study of redox-dependent platelet regulation.

Abstract

Reactive oxygen species (ROS) are highly unstable oxygen-containing molecules. Their chemical instability makes them extremely reactive and gives them the ability to react with important biological molecules such as proteins, nucleic acids, and lipids. Superoxide anions are important ROS generated by the reduction of molecular oxygen reduction (i.e., acquisition of one electron). Despite their initial implication exclusively in aging, degenerative, and pathogenic processes, their participation in important physiological responses has recently become apparent. In the vascular system, superoxide anions have been shown to modulate the differentiation and function of vascular smooth muscle cells, the proliferation and migration of vascular endothelial cells in angiogenesis, the immune response, and the activation of platelets in hemostasis. The role of superoxide anions is particularly important in the dysregulation of platelets and the cardiovascular complications associated with a plethora of conditions, including cancer, infection, inflammation, diabetes, and obesity. It has, therefore, become extremely relevant in cardiovascular research to be able to effectively measure the generation of superoxide anions by human platelets, understand the redox-dependent mechanisms regulating the balance between hemostasis and thrombosis and, eventually, identify novel pharmacological tools for the modulation of platelet responses leading to thrombosis and cardiovascular complications. This study presents three experimental protocols successfully adopted for the detection of superoxide anions in platelets and the study of the redox-dependent mechanisms regulating hemostasis and thrombosis: 1) dihydroethidium (DHE)-based superoxide anion detection by flow cytometry; 2) DHE-based superoxide anion visualization and analysis by single platelet imaging; and 3) spin probe-based quantification of superoxide anion output in platelets by electron paramagnetic resonance (EPR).

Introduction

The superoxide anion (O₂^•-) is the most functionally relevant ROS generated in platelets¹. O₂^•- is the product of the reduction of molecular oxygen and the precursor of many different ROS ². The dismutation of O₂^•- leads to the generation of hydrogen peroxide (H₂O₂) via spontaneous reactions in aqueous solution or reactions catalyzed by superoxide dismutases (SODs³). Although different enzymatic sources have been suggested (e.g., xanthine oxidase⁴, lipoxygenase⁵, cyclooxygenase^6, and nitric oxide synthase⁷), mitochondrial respiration^8,9 and nicotinamide adenine dinucleotide phosphate-oxidases (NOXs)¹⁰ are the most prominent sources of superoxide anion in eukaryotic cells. This also seems to be the case in platelets, where electron leakage from mitochondrial respiration^11,12 and the enzymatic activity of NOXs^13,14 have been described as the main contributors to the superoxide anion output.

Although several studies have focused on the regulation of platelets by O₂^•-, there is no consensus regarding the molecular mechanisms responsible. The modulation of surface receptor activity via direct oxidation and disulfide bond formation has been proposed for different platelet receptors. The positive regulation of integrin αIIbβ3 by ROS via direct oxidation of cysteine residues has been suggested^15,16,17. Similarly, since platelet responses to collagen depend on disulfide-dependent dimerization and consequent dimerization of the glycoprotein VI (GPVI)¹⁸, receptor activity potentiation by ROS-dependent oxidation has been proposed¹⁹, although not fully proven experimentally. Finally, ROS-induced oxidation of the sulfhydryl groups of glycoprotein Ib (GPIb) was shown to promote platelet adhesion and platelet-leukocyte interaction during inflammation²⁰. Conversely, as a possible consequence of decreased sulfhydryl group oxidation and receptor activation, the shedding of the ectodomain of both GPVI and GPIb is diminished by reducing conditions²¹.

Modes of action independently of a direct oxidation of platelet surface receptors have also been proposed. ROS, including O₂^•-, have been shown to positively modulate the collagen receptor GPVI by attenuating the activity of the Src homology region 2-containing protein tyrosine phosphatase 2 (SHP-2), which negatively regulates the signaling cascade of this receptor²². Moreover, O₂^•- can generate ONOO^- (peroxynitrite) by rapid reaction with nitric oxide (NO), which normally inhibits platelets through the NO-sensitive guanylyl cyclase (NO-GC) and the generation of the negative platelet regulator cyclic GMP (cGMP)^23,24. The resulting decrease in NO levels can lead to platelet potentiation. Alternatively, the generation of O₂^•- by NOX2 has been suggested to contribute to lipid peroxidation and isoprostane formation, which is essential for platelet activation and adhesion²⁵. Finally, mitogen-activated protein kinase (MAPK) extracellular signal-regulated kinase 5 (ERK5), a protein kinase proposed as a redox stress sensor in platelets²⁶, is activated by O₂^•- and induces a procoagulant phenotype in platelets (as estimated by flow cytometry-based measurement of phosphatidylserine externalization)²⁷.

The dysregulation of O₂^•- and other ROS generation in platelets has been associated with the exaggerated hemostatic response leading to thrombotic cardiovascular complications associated with atherosclerosis, diabetes mellitus, hypertension, obesity, and cancer^28,29. In these pathological settings, the ROS output by platelets is increased, which leads to a potentiation of their adhesive and aggregatory responses. In addition to the effect on platelet responses, the free radical output of platelets may have consequences on other blood cells and vascular structures, which is a poorly understood and underinvestigated area of cardiovascular health³⁰. Despite our limited understanding of the molecular mechanisms linking oxidative stress to thrombotic conditions, the clinical relevance of antioxidants for the protection against cardiovascular disease has received considerable attention. Plasma antioxidant levels have been shown to inversely correlate with the risk of developing cardiovascular conditions, and dietary antioxidant consumption has been shown to protect against coronary artery disease^31,32. Consequently, the use of dietary antioxidants has been advocated as a promising approach for cardiovascular disease prevention^33,34,35. Amongst the effects of ROS generation in platelets, the increase in apoptosis can have important pathophysiological effects^36,37. Overall, reliable protocols to detect and quantify the O₂^•- output by platelets are increasingly relevant in cardiovascular research.

Currently, available techniques for the detection of ROS have important limitations of specificity (i.e., the chemical nature of the oxidant molecules detected is unknown) and reliability (i.e., the unwanted interaction with biological molecules and experimental reagents leads to biased non-physiological results)^38,39. The most commonly used approach for the detection of ROS in platelets is based on the use of dichlorodihydrofluorescein diacetate (DCFDA), which is converted to dichlorodihydrofluorescein (DCFH) by intracellular esterases and consequently to the highly fluorescent dichlorofluorescein (DCF) by cellular oxidants, including hydroxyl radicals and peroxidase-H₂O₂ intermediates^40,41. Despite its wide use, serious questions have been raised regarding the reliability of this approach for the measurement of intracellular ROS³⁸. The oxidation of DCFH to DCF can be, in fact, induced by transition metal ions (e.g., Fe²⁺) or heme-containing enzymes (e.g., cytochromes) instead of ROS⁴². Moreover, DCFDA is converted by cell peroxidases to its semiquinone free radical form (DCF^•-), which is in turn oxidized to DCF by reaction with molecular oxygen (O₂) with the release of O₂^•-, which leads to the artificial amplification of oxidative responses^41,43,44. Therefore, the detection of intracellular ROS by DCFDA is useful for obtaining initial insights but requires cautious consideration and extensive experimental controls^38,39.

This study presents three alternative techniques for the detection and measurement of the key regulator of platelet function O₂^•-1. The first technique is the detection using DHE and flow cytometry, which offers advantages of reliability and specificity over DCFDA. The second technique proposed here also utilizes DHE, but the detection method is live-platelet fluorescence imaging, which allows the study of the generation of O₂^•- upon platelet signaling with fast kinetics and single-cell resolution. Finally, a protocol based on the use of the hydroxylamine spin probe 1-hydroxy-3-methoxycarbonyl-2,2,5,5-tetramethylpyrrolidine (CMH) in EPR resonance experiments offers the possibility of quantifying the rate of O₂^•- generation by platelets and comparing it in different conditions.

Protocol

The collection of peripheral blood from consenting volunteers is approved by the local Ethics Committee and the National Health Service Health Research Authority (REC reference: 21/SC/0215; IRAS ID: 283854).

1. Method 1: Superoxide anion detection using DHE by flow cytometry

1. Pre-warm (37 °C) sodium citrate solution (4% w/v) and use it as an anti-coagulant by adding it directly to the blood at the time of venepuncture at a ratio of 1:7 (0.5% w/v final).
2. Draw peripheral human blood from healthy volunteers by median cubital vein venepuncture.
3. Obtain platelet-rich plasma (PRP) from whole blood by an initial centrifugation step at 200 x g for 20 min.
4. Centrifuge PRP at 500 x g for 10 min with the reversible platelet inhibitors indomethacin (10 µM) and PGE1 (40 ng/mL).
5. Resuspend the resulting platelet pellet in modified Tyrode's buffer (145 mM NaCl, 2.9 mM KCl, 10 mM 4-(2-hydroxyethyl)-1-piperazineethane sulfonic acid (HEPES), 1 mM MgCl2, 5 mM glucose, pH 7.3) (37 °C) at a density of 2 x 10⁸ platelets/mL.
6. After isolation, rest platelets for 30 min at 37 °C.
7. Prepare DHE in dimethyl sulfoxide (DMSO) at a stock concentration of 5 mM.
8. Add DHE to the platelet suspension at 5 µM final concentration (dilute the stock solution 1 to 1,000, with final DMSO concentration 0.1% v/v) and incubate for 15 min at 37 °C.
9. Treat platelets with pharmacological agents and/or ROS scavengers for 15 min before stimulation with the desired physiological stimuli and concentrations (e.g., 0.1 unit/mL thrombin or 3 µg/mL collagen).
10. After stimulation, dilute platelet suspensions 1 to 10 in ice-cold modified HEPES buffer.
11. For the flow cytometry, set side scattering (SSC) and forward scattering (FSC) gain at 40 mV and 220 mV, respectively, and use logarithmic scale scatter plots to visualize the particle population in the suspension (example in Figure 1A).
12. (OPTIONAL) Immunostain one aliquot of platelet suspension using an anti-CD41 antibody to identify the population of events corresponding to platelets (example in Figure 1B).
13. Analyse samples by flow cytometry using excitation at 405 nm (violet laser) and emission at 580 nm wavelength, which selectively detects the superoxide anion-specific product 2-hydroxy-ethidium (2OH-Et+) (Figure 2).
    NOTE: Fixation with paraformaldehyde and long-term sample storage are not advisable.

2. Method 2: Superoxide anion detection using DHE by single-platelet fluorescence imaging

1. Coat an 8-well µ-slide on the day of the experiment with 0.1 mg/mL fibrillar collagen I from equine tendons (Horm collagen), 0.2 mg/mL fibrinogen or 1 mg/mL poly-L-lysine (PLL) in modified Tyrode's buffer for 2 h at 37 °C.
2. Wash twice with modified Tyrode's buffer (10 min each).
3. Quench non-specific adhesion by incubation in modified Tyrode's buffer plus 5 mg/mL bovine serum albumin (BSA) for a minimum of 30 min (or until the experiment).
4. Pre-warm (37 °C) sodium citrate solution (4% w/v) and use it as an anti-coagulant by adding it directly to the blood at the time of venepuncture at a ratio of 1:7 (0.5% w/v final).
5. Draw peripheral human blood from healthy volunteers by median cubital vein venepuncture.
6. Obtain platelet-rich plasma (PRP) from whole blood by an initial centrifugation step at 200 x g for 20 min.
7. Centrifuge PRP at 500 x g for 10 min with the reversible platelet inhibitors indomethacin (10 µM) and PGE1 (40 ng/mL).
8. Resuspend the resulting platelet pellet in modified Tyrode's buffer (145 mM NaCl, 2.9 mM KCl, 10 mM HEPES, 1 mM MgCl2, 5 mM glucose, pH 7.3) (37 °C) at a density of 4 x 10⁷ platelets/mL.
9. After isolation, rest platelets for 30 min at 37 °C.
10. Prepare DHE in dimethyl sulfoxide (DMSO) at a stock concentration of 10 mM.
11. Add DHE at 10 µM final concentration to the platelet suspension (dilute the stock solution 1 to 1,000, with final DMSO concentration 0.1% v/v) and incubate for 1 min (37 °C).
12. After the 1 min of incubation with DHE, remove the blocking solution from the chosen wells, position the µ-slide on the microscope stage ready for imaging, and gently dispense platelets.
13. Monitor DHE conversion to 2OH-Et+ by inverted confocal imaging for 10 min (405/580 nm ex/em), with images collected every 10 s with a 40x oil immersion lens.
14. (OPTIONAL) For wells coated with PLL, monitor the superoxide anion generation in response to a soluble agonist (e.g., 0.1 unit/mL thrombin or 3 µg/mL collagen) by adding the agonist 10 min after platelet dispensing and fluorescence image collection for a further 10 min, as described in step 2.13.
15. (OPTIONAL) If required, add superoxide anion scavengers or selective inhibitors (e.g., NOX inhibitor) by dispensing gently to the side of the well during the time course and collect fluorescence images as described in 2.13 for a further 10 min.
16. Quantify single-cell fluorescence analysis by selecting the Region of Interest (ROI) containing representative single platelets and analyzing fluorescence intensity in different frames with the Measure tool of imaging suite ImageJ 1.53t.

3. Method 3: Detection of superoxide anion by platelets using electron paramagnetic resonance (EPR)

1. Pre-warm (37 °C) sodium citrate solution (4% w/v) and use it as an anti-coagulant by adding it directly to the blood at the time of venepuncture at a ratio of 1:7 (0.5% w/v final).
2. Draw peripheral human blood from healthy volunteers by median cubital vein venepuncture.
3. Obtain platelet-rich plasma (PRP) from whole blood by an initial centrifugation step at 200 x g for 20 min.
4. Centrifuge PRP at 500 x g for 10 min with the reversible platelet inhibitors indomethacin (10 µM) and PGE1 (40 ng/mL).
5. Resuspend the resulting platelet pellet in modified Tyrode's buffer (145 mM NaCl, 2.9 mM KCl, 10 mM HEPES, 1 mM MgCl2, 5 mM glucose, pH 7.3) (37 °C) at a density of 2 x 10⁸ platelets/mL.
6. After isolation, rest platelets for 30 min at 37 °C.
7. Add 5 µM diethyldithiocarbamate (DETC) and 25 µM deferoxamine to the platelet suspension.
8. Without incubation, add 200 µM 1-hydroxy-3-methoxycarbonyl-2,2,5,5-tetramethylpyrrolidine (CMH).
9. Load samples onto the aggregometer within aggregation cuvettes (including Teflon-coated stirring magnets).
10. After 1 min, add the stimuli at the desired concentration (e.g., 1 unit/mL thrombin or 3 µg/mL collagen). Incubate for 10 min.
11. (OPTIONAL) Obtain superoxide anion generation at different time points by proceeding to step 3.12 at the desired time.
12. Transfer the platelet suspension from the aggregation cuvette to a microcentrifuge tube and quickly spin down at 6,000 x g for 10 s.
13. Load 50 µL of the supernatant in capillary micropipettes and seal with EPR sealing wax.
14. Transfer samples to the EPR scanner.
15. Set the EPR scanner as follows: center field 3,492.5 G, field sweep 60 G, modulation amplitude 2 G, sweep time 10 s, number of scans 10, microwave frequency 9.39 GHz, microwave power 20 mW, conversion time 327.68 ms, time constant 5242.88 ms.
16. Estimate CMH oxidation to CM^● as the area under the curve (AUC) of the EPR peaks recorded using the above parameters.
17. Build a calibration curve plotting the EPR intensity measured as described in step 3.16 on the y-axis and the concentrations of the commercially available CM^● on the x-axis (e.g., 0, 0.3, 1, 3, 10, and 30 µM).
18. From the CM^● concentration in the samples, obtain the amount of superoxide anion generated by the platelets in the samples using the formulas described in the Representative Results section.

Results

For flow cytometry detection of DHE fluorescence, we show representative results for platelets either resting (Figure 3A) or stimulated with 0.1 unit/mL thrombin (Figure 3B). The O₂^•- output was quantified as platelet mean fluorescence intensity (MFI), as shown for stimulation with 0.1 unit/mL thrombin (Figure 3C) or 3 µg/mL collagen-related peptide (CRP) (Figure 3D). A two/three-fold increase in DHE staining was observed as a result of platelet stimulation. To confirm that O₂^•- is measured, we suggest the use of the cell-permeable O₂^•- scavenger pegylated-superoxide dismutase (PEG-SOD, 100 units/mL), while to identify the source of O₂^•- we used the NOX inhibitor VAS2870 (10 µM). For both collagen and thrombin, PEG-SOD and VAS2870 abolished the increase in DHE fluorescence caused by platelet activation, confirming that the measurement is specific for O₂^•- and that their sources in these conditions are NOX enzymes.

For imaging detection of DHE fluorescence, we present data from experiments assessing the generation of O₂^•- at the moment of adhesion to collagen or fibrinogen (and the following 10 min) (Figure 4A) or from experiments aiming to assess the O₂^•- generation from platelets previously adhered to PLL upon stimulation with thrombin (Figure 4B). As expected, adhesion to PLL does not trigger a robust production of O₂^•-. Most likely because this form of adhesion depends on electrostatic interaction without the initiation of any signaling response⁴⁵, which makes this approach suitable to measure the O₂^•- output in response to the addition of soluble agonists. To confirm that this protocol allows the detection of O₂^•-, either scavengers or selective inhibitors can be utilized. We have shown in previous studies that N-acetyl-cysteine (NAC) can scavenge O₂^•- and abolish platelet fluorescence in these experiments⁴⁶. Here, we show that the selective NOX inhibitor VAS2870 effectively and instantly abates fluorescence in platelets adhering to collagen or fibrinogen (Figure 4A) or responding to thrombin (Figure 4B). This suggests that NOXs are the source of O₂^•- in these experimental conditions. Representative movies for DHE fluorescence (405 nm excitation) in response to adhesion to collagen or fibrinogen or the stimulation by thrombin of platelets adhering to PLL are included as Supplementary File 1, Supplementary File 2, and Supplementary File 3, respectively. Of note, PEG-SOD, which has been used successfully in other experimental setups as an O₂^•- scavenger, is not effective in this type of experiment. We have no clear explanation for this at the moment. We can hypothesize that PEG-SOD reaches the cellular compartments in which O₂^•- induces increased fluorescence in the imaging-based assays either with poor efficiency or slow kinetics. Further focused investigations may be required to clarify this point.

We applied the measurement of O₂^•- by EPR in platelets in different previously published studies^{1,13,47,48,49,50}. A crucial step for this technique is the preparation of a calibration curve using commercially available CM^● (i.e., the product of the reaction between the spin probe CMH and O₂^•- and the resonating species measured by EPR). The reaction of CMH and O₂^•- is shown in Figure 5A, and representative examples of the EPR signal induced by different concentrations of CM^● are shown in Figure 5B. The data from different CM^● concentrations were used to build a calibration curve describing the linear correlation between the amount of oxidized CMH and the EPR signal intensity in samples (Figure 5C). The equation below describes the correlation between CM^● concentration and EPR signal intensity and is the starting point for the calculation of the calibration curve for the experiments in this study (Equation 1):

As both CM^● concentrations and their corresponding EPR intensity were obtained experimentally, the Slope and the Y-axis Intercept of the calibration curve can be calculated. In the example shown in Figure 5C, Slope = 1,002,169 and Intercept = 596,383. Therefore, the equation for the calibration curve can be re-written as:

By rearranging the equation above, it is possible to calculate CM^● concentration for different samples based on their EPR intensity values (Equation 2):

Once the concentration of CM^● in the samples is known, it is possible to calculate its generation rate per platelet and unit of time with the equation below (Equation 3):

Using Equation 2, Equation 3 can be rewritten as (Equation 4):

Which in the example shown in Figure 5C can be re-written as:

The CMH oxidation rate will be expressed as attomoles of CM^● generated per platelet per minute. This value can be used to compare the O₂^•- generation rate between experiments and donors, which is not possible with fluorescent probes like DHE (or DCFDA). Fluorescent probes generally only allow an estimate of the effect of different conditions on O₂^•- generation as changes in fluorescence intensity within the same experiment. As for all techniques described in this article, the specificity of detection was confirmed with ROS scavengers (e.g., PEG-SOD, NAC) or selective inhibitors of enzymes generating superoxide anion (e.g., VAS2870 to inhibit NOXs). Here, we present data for platelet stimulation (Figure 6) with thrombin (Figure 6A,C,E) or collagen (Figure 6B,D,F). For both agonists, PEG-SOD was used to confirm that O₂^•- is detected, while the lack of effect for pegylated-catalase (PEG-Cat.) suggests that hydrogen peroxide is not detected by this technique (Figure 6C,D, respectively). VAS2870 abolishes thrombin- (Figure 6E) and collagen-induced (Figure 6F) EPR signals, suggesting that NOXs are the main source of platelet O₂^•- in response to both these agonists.

Figure 1: flow cytometry analysis of platelet suspensions. (A) Typical presentation of forward (FSC) and side scattering (SSC) plots for isolated platelets. (B) The immunostaining of the platelet suspension with an anti-CD41 antibody confirms >98% of the particles in the preparation as platelets. Please click here to view a larger version of this figure.

Figure 2: Chemical reaction of DHE with superoxide anion and other ROS and spectral properties of its products. Structure and spectral properties of DHE and its two oxidation products 2-hydroxy-ethidium (2OH-Et+), generated by reaction with superoxide anions, and ethidium (Et+), generated by reaction with other types of ROS (e.g., hydroxyl radical and hydrogen peroxide). The excitation peak at 405 nm, highlighted by a dashed blue rectangle in the figure, is specific for 2OH-Et+ and can, therefore, be used to detect/measure O₂^•-. Please click here to view a larger version of this figure.

Figure 3: Thrombin- and collagen-dependent superoxide anion generation detected by DHE and flow cytometry. Representative histograms for DHE staining of (A) resting and (B) thrombin-stimulated platelets. The bar indicates DHE staining values below 500, which is 70.7% and 16.1% for resting and thrombin-stimulated platelets, respectively. This technique was utilized to estimate superoxide anion generation in (C) 0.1 unit/mL thrombin and (D) 3 µg/mL collagen-stimulated platelets. PEG-SOD (100 unit/mL) was used to confirm that superoxide anions are measured in this experiment, while 10 µM VAS2870 was used to identify NOX enzymes as the source of superoxide anions. Each data point was calculated as mean fluorescence intensity (MFI) from 50,000 platelets, and each experiment was repeated five independent times (n = 5), with mean ± SEM shown in the graphs. The statistical significance was assessed by one-way ANOVA with Tukey post-test for pairwise comparisons (** = p < 0.01, *** = p < 0.001). Please click here to view a larger version of this figure.

Figure 4: Collagen-, fibrinogen-, and thrombin-dependent superoxide anion generation detected by DHE fluorescence imaging. (A) Superoxide anion generation kinetics was assessed for platelets adhering to either collagen or fibrinogen. Platelets were dispensed on coated µ-slides 1 min after the beginning of the imaging, while 10 µM VAS2870 was added 5 min from the beginning of the image collection. (B) Superoxide anion generation kinetics for platelets upon stimulation with thrombin. Platelets were allowed to adhere to PLL-coated µ-slides for 10 min before the beginning of the image collection. After 2 min, 0.1 unit/mL thrombin was added, and after a further 5 min, 10 µM VAS2870 was added (i.e., 7 min from the beginning of the imaging). The experimental scheme is presented at the top of the panels, while representative images from 0 min, 1 min, 3 min, 6 min, and 9 min are shown in the middle of the panels. Single-platelet DHE fluorescence quantification is shown at the bottom of the panels and was obtained by fluorescence intensity analysis using ImageJ. The data are mean ± SEM from 8-12 platelets per condition in 4 independent experiments. Please click here to view a larger version of this figure.

Figure 5: Calibration curve and CMH oxidation rate calculations. (A) Chemical structure and EPR properties of CMH and CM^●. (B) Representative examples of EPR data for different concentrations of CM^● (100 nM to 10 µM). (C) Calibration curve of EPR intensity versus CM^● concentration. Please click here to view a larger version of this figure.

Figure 6: Detection of superoxide anion production by platelets upon thrombin or collagen stimulation. The superoxide anion output was measured by EPR with the spin probe CMH in platelets stimulated with either (A) 0.1 unit/mL thrombin or (B) 3 µg/mL collagen. (C, D) A concentration of 100 units/mL of the superoxide anion scavenger PEG-SOD or 100 units/mL of the hydrogen peroxide scavenger pegylated catalase (PEG-Cat.) was included in the conditions tested for both agonists. The source of superoxide anion was investigated by co-incubation with the NOX inhibitor VAS2870 (10 µM). Both (E) thrombin and (F) collagen depend on NOX activity for the generation of superoxide anion. Throughout the figure, the CMH/superoxide anion reaction rate was calculated from the EPR traces, as described in the manuscript (Equation 4). Data are mean ± SEM from 5 (C,D) and 4 independent experiments (E,F). The statistical significance was assessed by one-way ANOVA with Tukey post-test for pairwise comparisons (** = p < 0.01, *** = p < 0.001). Please click here to view a larger version of this figure.

Supplementary File 1: Representative movie for DHE fluorescence (405 nm excitation) in response to adhesion to collagen. Please click here to download this File.

Supplementary File 2: Representative movie for DHE fluorescence (405 nm excitation) in response to adhesion fibrinogen. Please click here to download this File.

Supplementary File 3: Representative movie for DHE fluorescence (405 nm excitation) in response the stimulation by thrombin of platelets adhering to PLL. Please click here to download this File.

Discussion

In this manuscript, we present three different techniques with the potential to advance the capability to investigate the redox-dependent regulation of platelet function via the selective detection of O₂^•-. The first two methods are an improvement on existing techniques because of the redox probe utilized (DHE instead of the more common but less reliable DCFDA). These techniques are, therefore, easily accessible, and most laboratories can adopt them effectively without particular equipment or training costs. The third technique is based on the use of specialized equipment for the analysis of the oxidation of the spin probe CMH by EPR. This technique requires the purchase of specialized equipment and reagents and the dedicated training of the laboratory user. Consequently, it appears suitable for laboratories specifically focusing on the study of redox-dependent events in platelets (or other cells).

For techniques utilizing DHE, the wavelength selection for the excitation is crucial to obtain the specific detection of O₂^•-. 2-hydroxy-ethidium (2OH-Et+) generated by the reaction of DHE with O₂^•- is excited at 405 nm, while ethidium (Et+) generated by the general reaction of DHE with ROS (e.g., hydroxyl radical and hydrogen peroxide) has an excitation peak around 520 nm but no excitation at 405 nm (Figure 1)⁵¹. Using DHE instead of DCFDA protects from the artifacts caused by the direct reaction of this probe with cellular components such as peroxidases, other heme-containing enzymes, and metal ions reducing agents^42,43,44. DHE has been suggested as a safer tool for cellular redox studies^38,39. Here, we propose the use of DHE with two different readout techniques: 1) flow cytometry and 2) fluorescence imaging.

Flow cytometry is generally a technique that allows contemporaneous handling of multiple samples with a low-to-medium throughput⁵². Unfortunately, the approach used in this study does not permit sample fixation. Although the dilution of the samples with ice-cold modified Tyrode's buffer stabilizes the fluorescence in the samples for up to 30 min, the analysis of all samples needs to be performed in this timeframe. This significantly limits the number of samples per experiment. Multiple internally controlled experimental runs within the same day and using the same platelet preparation are possible, although the results should be normalized to controls and not considered as independent experiments. This technique has been used by the authors of this manuscript to highlight the role of redox-dependent platelet modulation in the responses to different agonists and modulators^46,50. This method can be modified by adopting the use of a chemically modified version of DHE that allows targeting of mitochondria (i.e., Mitosox)⁵³, which has been used for platelet studies focusing on NOX2⁵⁴. This is an interesting opportunity that has not yet been pursued by our laboratories.

The fluorescence imaging technique using DHE proposed here has the advantage of determining the time course of O₂^•- generation in platelets upon adhesion (agonists absorbed on the surface of the imaging well) or stimulation (agonists in suspension). This allows the interpretation of the interplay of different phases of the platelet activation process and the evaluation of the contribution of different O₂^•- sources (e.g., different ROS-generating enzymes, e.g., NOXs). No existing technique allows this type of investigation. This comes at a cost in terms of assay throughput, with each tested condition taking 30 min (which includes µ-slide setup, focusing and field selection, imaging, and data saving). Therefore, this technique allows testing of a small number of conditions per working day (i.e., under 10). Although a confocal microscope was used for the data shown here, we have confirmed that equally good results can be obtained with an epifluorescence microscope with 405 nm excitation capability. This reduces the requirement for expensive confocal microscopy setups and makes this technique more affordable.

The EPR protocol has been developed based on previous studies in different cell types⁵⁵. This technique is widely recognized as the most reliable tool for the study of ROS generation and the investigation of redox-dependent events in cells³⁸. In addition to superior reliability compared to other techniques due to the specificity of the reaction of hydroxylamine spin probes with oxygen-free radicals⁵⁶, EPR allows the quantification of the O₂^•- generation rate and its comparison between different individuals and experiments. This is not possible with fluorescence-based techniques, which only allow us to estimate the effect of selected conditions on fluorescence intensity within a single experiment and platelet preparation. Although this EPR protocol was developed as a technique to combine with platelet aggregation for the parallel monitoring of platelet activation and O₂^•- generation¹, these experiments can be performed without the parallel analysis of aggregation. Even without parallel analysis of platelet aggregation, we suggest performing the incubation in an aggregometer to guarantee effective platelet stimulation, which requires the shear stress provided by continuous stirring. If the experiments need to be performed in 1.5 mL microcentrifuge tubes without stirring, the levels of O₂^•- production are significantly lower. Therefore, if this experimental approach is chosen, we suggest performing the CMH incubation on a rotating wheel and with more densely concentrated platelets (i.e., 1 x 10⁹/mL), which leads to a signal sufficiently strong for detection by the EPR spectrometer.

In summary, here we present three validated techniques for the detection of O₂^•- in platelets. They offer different advantages and can be chosen according to experimental objectives and priorities. The first technique based on the use of DHE by flow cytometry is the least laborious and time-consuming and can, therefore, be chosen as an entry-level analysis. The second technique, based on the use of DHE by fluorescence microscopy, allows an appreciation of the fast kinetics of O₂^•- generation and the study of how different signaling cascades may affect the oxidative response in the seconds time scale (while other techniques only allow endpoint investigation of events occurring within minutes from stimulation). Finally, EPR allows the quantification of the O₂^•- generation rate and its comparison between experiments and individuals. For this reason, the latter technique is particularly useful for clinical studies aiming to characterize the oxidative response in platelets from different individuals (e.g., comparison between patients or patients versus healthy controls). Overall, these three techniques offer significant advantages over existing methods and represent a significant advance in the experimental tools at our disposal for studying redox-dependent events in platelets.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
1-hydroxy-3-methoxycarbonyl-2,2,5,5-tetramethylpyrrolidine (CMH)	Noxygen Science trasfer and Diagnostics GmbH	NOX-02.1-50mg	Reagent for EPR (spin probe)
BD FACSAria III	BD Biosciences	NA	Flow cytometer
Bovine Serum Albumin	Merck/Sigma	A7030	For μ-slide coating
Bruker E-scan M (Noxyscan)	Noxygen Science trasfer and Diagnostics GmbH	NOX-E.11-BES	EPR spectrometer
Catalase–polyethylene glycol (PEG-Cat.)	Merck/Sigma	C4963	Hydrogen peroxide scavenger (specificity control)
ChronoLog Model 490+4	Labmedics/Chronolog	NA	Aggregometer
CM radical	Noxygen Science trasfer and Diagnostics GmbH	NOX-20.1-100mg	Reagent for EPR (calibration control)
deferoxamine	Noxygen Science trasfer and Diagnostics GmbH	NOX-09.1-100mg	Reagent for EPR
diethyldithiocarbamate (DETC)	Noxygen Science trasfer and Diagnostics GmbH	NOX-10.1-1g	Reagent for EPR
Dihydroethidium	Thermo Fisher Scientifics	D11347	Superoxide anion probe
Dimethyl sulfoxide	Merck/Sigma	34869	For stock solution preparation
EPR sealing wax plates	Noxygen Science trasfer and Diagnostics GmbH	NOX-A.3-VPM	Consumable for EPR
EPR-grade water	Noxygen Science trasfer and Diagnostics GmbH	NOX-07.7.1-0.5L	Reagent for EPR
Fibrinogen from human plasma	Merck/Sigma	F4883	For μ-slide coating
FITC anti-human CD41 Antibody	BioLegend	303704	Platelet-specific staining for flow cytometry
Glass cuvettes	Labmedics/Chronolog	P/N 312	Consumable for incubation in aggregometer
Horm Collagen	Labmedics/Chronolog	P/N 385	For platelet stimulation
ImageJ	National Institutes of Health (NIH)	NA	ImageJ 1.53t (Wayne Rasband)
Indomethacin	Merck/Sigma	I7378	For platelet isolation
Micropipettes DURAN 50µl	Noxygen Science trasfer and Diagnostics GmbH	NOX-G.6.1-50µL	Consumable for EPR
Poly-L-lysine hydrochloride	Merck/Sigma	P2658	For μ-slide coating
Prostaglandin E1 (PGE1)	Merck/Sigma	P5515	For platelet isolation
Sodium citrate (4% w/v solution)	Merck/Sigma	S5770	For platelet isolation
Stirring bars (Teflon-coated)	Labmedics/Chronolog	P/N 313	Consumable for incubation in aggregometer
Superoxide dismutase–polyethylene glycol (PEG-SOD)	Merck/Sigma	S9549	Superoxide anion scavenger (specificity control)
Thrombin from human plasma	Merck/Sigma	T6884	For platelet stimulation and μ-slide coating
VAS2870	Enzo Life Science	BML-EI395	NOX inhibitor
Zeiss 510 LSM confocal microscope	Zeiss	NA	Confocal microscope