Generalities on Flow Cytometry

First developed as a rapid and automatic technique for determining the nuclear DNA content, flow cytometry (FCM) now finds its most common applications in the phenotypic and functional characterization of various cell types because of its ability to detect several specific characteristics of a large number of individual cells in suspension in a time range of seconds.

In the flow cytometer, suspended aligned cells pass through a flow cell where they are hit by the focused beam of a laser. Detectors process the light scattering and, in the case of fluorescent labeling, the emitted fluorescence of each cell. Cell parameters that can be assessed by FCM are divided into two groups:

  • Physical or intrinsic parameters, represented by the forward-scattered light (FS), which is proportional to cell dimensions, and the side-scattered light (SS), which is proportional to the structural complexity of the cell.

  • Fluorescent parameters, obtained by labeling cells, before analysis, with fluorophore-conjugated monoclonal antibodies (targeting specific cell antigens) or cell-permeable fluorescent proteins that are activated at specific excitation wavelengths by a laser beam. The intensity of the fluorophore-emitted light is directly proportional to the antigen density. The wide range of labeled antibodies or cell-permeable fluorescent proteins commercially available permits not only the analysis of several cell parameters in one single shot, but also a detailed study of cell phenotype and function, cell–cell interaction, cell activation, apoptosis, and cell cycle, to name just a few (McCoy 2002).

Advantages of the Use of FCM in Platelet Studies

FCM allows rapid analysis of platelet turnover and count, morphology, structure, activation state, and response to agonists. This means that it is a versatile and reliable diagnostic tool for analysis of platelet disorders, evaluating the effects of antiplatelet agents, and assessment of circulating activated platelets in the context of thrombotic risk.

In recent years, the use of FMC in the field of hemostasis and thrombosis has become widespread thanks to its feasibility compared with time-consuming and expensive conventional platelet assays such as enzyme-linked immunoassay (ELISA), transmission electron microscopy (TEM), fluorescence microscopy, C14 labeling, high performance liquid chromatography (HPLC), particle gel immunoassay (PaGIA), aggregometry, and radioimmunoassay (Carubbi et al. 2014a).

Additionally, the reduced blood volume required for analysis, minimal sample manipulation, and high number of commercially available monoclonal antibodies allowing single-shot multiparametric evaluation make FCM a fundamental technique in platelet studies. However, some aspects need to be carefully considered when using FCM in platelet studies. In particular, the operator should be aware of the risk of platelet activation as a result of inadequate sample collection and/or time delays. The principal advantages and disadvantages of FCM are summarized in Table 1.

Table 1 FCM in platelet studies
Full size table

Methodological Aspects

Peripheral Blood Draw: Suggested Procedure and Sample Preparation

In platelet studies, a careful execution of peripheral blood draw is mandatory to minimize platelet activation. For this purpose, a 20–21G needle is recommended. Additionally, it is helpful to rapidly remove the tourniquet and discard the first 2 mL of blood. The sample should then be conserved at room temperature and processed within 30 min. These precautions are strongly recommended when the activation state of circulating platelets is to be determined.

The choice of anticoagulant relies mainly on specific tests: in the case of ex vivo platelet response to agonists, sodium citrate is recommended because it does not interfere with platelet activation. By contrast, if seeking to minimize platelet activation after sampling, anticoagulants should be chosen that contain platelet inhibitors such as CTAD (citrate, theophylline, adenosine, and dipyridamole). Ethylenediamine tetraacetic acid (EDTA) is suitable for platelet count measurements (Michelson et al. 2000).

Platelet Source

The starting material for platelet studies by FCM can consist of platelet-rich plasma (PRP), washed platelets, or whole blood. The choice depends mainly on the test type. PRP, obtained by blood centrifugation at 125–150 g for 10–20 min, is suggested by some authors for calcium flux analysis.For detection of platelet microparticles, two centrifugations (1500 g for 15 min followed by 13,000 g for 2 min) are recommended in standardized protocols. Washed platelets are utilized in activation studies with strong agonists (i.e., thrombin), which can cause clot formation in the presence of plasma fibrin (Hickerson and Bode 2002; Robert et al. 2009).

Nevertheless, in clinical studies, whole blood is by far the most convenient platelet source (Shattil et al. 1987) because platelets are kept in their physiological milieu of red cells, leukocytes, and plasma components, which clearly affects platelet activation (Santos et al. 1991; LaRosa et al. 1994b). Moreover, the sample is minimally manipulated, preventing artifactual ex vivo activation and potential loss of platelet subpopulations (Michelson et al. 1991; Abrams and Shattil 1991; Shattil et al. 1987). Also, only minuscule volumes of blood are required (2–5 μL), allowing accurate analyses on newborns and profound thrombocythopenic patients (Shattil et al. 1987; Michelson et al. 1991; Rajasekhar et al. 1994). Michelson and colleagues (Michelson et al. 1991; Michelson 1994; Kestin et al. 1993) developed a method for studying platelet activation by thrombin directly in whole blood, based on addition of the synthetic tetrapeptide GPRP (glycyl-l-prolyl-l-arginyl-l-proline) that competitively prevents fibrin polymerization and, via inhibition of fibrinogen binding to its receptor, partially blocks platelet aggregation. The same effect is also mediated by Arg-Gly-Asp (RGD)-containing peptides that, mimicking fibrinogen binding sites, act as glycoprotein IIb/IIIa (GPIIb/IIIa) antagonists. These assays allow simultaneous analysis of the activation state and reactivity of circulating platelets. An alternative is use of the thrombin receptor agonist peptide (TRAP), although it may not fully reflect all aspects of thrombin-induced activation because it is a peptide fragment of the “tethered ligand” receptor for thrombin (Yamamoto et al. 1991).

Expression of Antibody Binding

In FCM, platelet population is conventionally identified by a “morphological gate” according to platelet physical properties and is represented on a dot plot of FS versus logarithmic SS. However, for discrimination of platelets from other blood cells it is more appropraite to use a light scatter parameter versus the fluorescence signal of platelet-specific constitutive antigens (i.e., “immunological gate”). Photomultiplier voltages are adjusted to give light scatter signals in the midrange of the logarithmic scale of the instrument. Amplification of fluorescence signal is selected to give a platelet autofluorescence signal that falls within the first logarithmic decade of the instrument. Because of spectral emission overlap, appropriate electronic color compensation must be set for each combination of antibodies (fluorophores) according to the manufacturer’s instructions and confirmed by each laboratory.

Antibody binding can be expressed as the percentage of cells staining positive for a particular antibody or as mean fluorescence intensity (MFI). The percentage of positive cells identifies a subpopulation of platelets expressing a specific antigen, in comparison with the negative fluorescence of a predominant platelet population. A threshold value for the positive signal intensity is defined on the basis of an appropriate value for false-positive events, conventionally including 1–2 % of positive cells in a one-parameter histogram analysis of a matched negative control sample stained for the determination of nonspecific fluorescence (e.g., sample stained with isotype control). This method is not affected by variations in signal amplification because the negative control signal increases in proportion with the test sample. Moreover, as a result of its high sensitivity, the method allows detection of changes in antigen expression by small subpopulations of cells, which results in a heterogeneous platelet staining pattern. Indeed, the “percentage of positive platelets” method can detect subpopulations of platelets arising from a local in vivo insult and is the most appropriate method for analysis of antigens expressed upon activation. However, this method is inappropriate for measuring variations in the expression of antigens that are homogeneously exposed across platelet subpopulations (i.e., CD41). Moreover, antibody-positive platelets may have very little antigen expressed at their surface.

If the goal is to determine variations in homogeneously expressed antigens or the total amount of platelet surface antigens, MFI is the method of choice. MFI is defined as the mean fluorescence intensity of the analyzed population and represents the mean antigen density on the cell surface. Therefore, an increase or decrease in MFI represents a similar increase or decrease in antigen expression per single platelet. Quantification of the number of antibodies specifically bound to platelets is based on a calibration curve established for each directly conjugated antibody using multiple bead populations with a defined number of specific binding sites. For this purpose, commercial kits equipped with a set of calibrated fluorescent standards and software can be used to determine molecules of equivalent soluble fluorochrome (MESF). An additional advantage of the routine utilization of these standards allows data comparison over time and between different instruments and laboratories (Michelson et al. 2000).

An interesting technique for the calculation of binding index that takes into account both MFI and the percentage of positive platelets has also been described (Hjemdahl et al. 1994; Zeller et al. 1999; Leytin et al. 2000).

FCM Evaluation of Platelet Activation Markers

As previously stated, FCM is a helpful tool for assessing platelet activation state both in vivo (circulating platelets) and ex vivo (platelet reactivity to agonists). When activated, platelets undergo several changes in surface antigen expression and granule release (Fig. 1). FCM can detect these events as changes in the binding of specific monoclonal antibodies that recognize markers of platelet activation. The most common markers are summarized in Fig. 1 and Table 2 and can be divided into surface markers and intracellular markers. Additionally, other parameters of platelet activation are represented by platelet–leukocyte aggregates (PLAs) and platelet microparticles (PMPs).

Fig. 1
figure 1

Surface and intracellular platelet markers in resting and activated platelets studied by flow cytometry

Full size image
Table 2 Flow cytometric markers of platelet activation
Full size table

Activation-Dependent Surface Markers

Compared with other cells, platelet surface antigens constitute a larger proportion of platelet cellular mass and are primarily represented by platelet–membrane receptors. Conformational modifications or changes in the levels of their expression on the cellular surface can be used to trace platelet activation and, for this purpose, FCM is the method of choice for its accuracy and rapidity. Platelet–membrane receptors are grouped into eight types:

  1. 1.

    Integrins (αIIbβ3, α2β1)

  2. 2.

    Leucine-rich repeat receptors (GPIb-IX-V complex, Toll-like receptors)

  3. 3.

    Seven-transmembrane receptors (thrombin, prostaglandin, ADP, lipid, and chemokine receptors)

  4. 4.

    Immunoglobulin superfamily (GPVI, FcγRIIa, FcεRI, junction adhesion molecules, intracellular adhesion molecules, PECAM-1, CD47)

  5. 5.

    C-type lectin receptor family (P-selectin, CD72, CD93)

  6. 6.

    Tetraspanins (CD9, CD63, CD82, CD515), glycosyl-phoshatidylinositol-anchored proteins (CD55 and CD59)

  7. 7.

    Tyrosine-kinase receptors (thrombopoietin, insulin and leptin receptor, PDGF receptor)

  8. 8.

    Others (CD36, lysosomal-associated membrane proteins, CD40L) (Clemetson and Clemetson 2013).

The two most widely utilized activation-dependent surface antigens are P-selectin and the αIIbβ3 complex (Shattil et al. 1985; Stenberg et al. 1985) (see Table 2 and Fig. 1).

P-selectin (or CD62P) is a component of the α-granule membrane of resting platelets that mediates adhesion to neutrophils and monocytes. It is expressed on their surface only after platelet degranulation (Fig. 1). Therefore, P-selectin-specific monoclonal antibodies (see Fig. 2 and Table 2) bind exclusively to activated platelets.

Fig. 2
figure 2

Flow cytometric analysis of the most utilized activation markers before and after platelet activation with different agonists (ADP and calcium ionophore A23187). Analysis was performed on whole blood or platelet-rich plasma (PRP), as indicated

Full size image

The αIIbβ3 complex is a receptor for fibrinogen, von Willebrand factor (vWF), vitronectin, and fibronectin. It undergoes a conformational change in response to platelet activation. The change is recognized by the monoclonal antibody PAC-1 (Fig. 1), which binds specifically to the fibrinogen binding site of αIIbβ3 that is exposed during its activation-dependent re-shaping (see Fig. 2 and Table 2).

As an alternative to αIIbβ3-specific monoclonal antibodies, fluorescein isothiocyanate (FITC)-conjugated fibrinogen can also be used to detect the activated form of the surface glycoprotein. Fluorescein–fibrinogen binding is saturable, dependent on agonist activation, but competitively inhibited by unlabeled fibrinogen in the plasma and released from platelet α-granules (Faraday et al. 1994; Heilmann et al. 1994).

Other common markers of platelet activation are represented by CD63 (also known as lysosomal integral membrane protein, LIMP), CD36 (or GPIV), and CD40L. CD63 is a component of lysosomal membranes whose role has yet to be clarified. When granules are released, CD63 is exposed on the platelet surface where its expression increases from 650 to 12,600 molecules per platelet (Fig. 1) (Nieuwenhuis et al. 1987). Although CD63, compared with P-selectin, can be considered a more stable marker of platelet activation because of a reduced tendency to undergo proteolysis, it is not an ideal candidate for platelet activation studies because of a requirement for a greater level of activation for adequate surface antigen expression. Consequently, P-selectin remains the marker of choice for the majority of investigations (Nishibori et al. 1993).

The monoclonal antibody OKM5 is directed against an epitope on GPIV that binds thrombospondin. OKM5 also recognizes resting platelets but its binding is increased by thrombin stimulation (see Fig. 1). The GPIb-IX complex binds to vWF, mediating adhesion to damaged blood vessel walls and is recognized by the monoclonal antibody 6D1. Platelet activation leads to a redistribution of GPIb-IX complexes from the surface into the open canalicular system (Hourdille et al. 1990, 1992; Michelson et al. 1994). The decreased binding of 6D1 to activated platelets is a very sensitive marker of activation both ex vivo and in vivo (Kestin et al. 1993) (Fig. 1).

CD40 ligand (CD40L, also known as CD154) is a trimeric transmembrane protein and member of the tumor necrosis factor family. It is a component of the α-granule membrane of resting platelets and is quickly exposed on the platelet surface after activation. Expression of CD40L on the activated platelet surface is transient because it is rapidly cleaved from the platelet membrane, generating a soluble fragment known as soluble CD40L (Aloui et al. 2014). FCM studies with the monoclonal antibody TRAP-1 demonstrated that CD40L is undetectable on resting platelets and that platelet activation by thrombin results in maximal expression of CD40L within 1 min (Henn et al. 1998). Platelet activation in cardiovascular disease is associated with a significant increase in expression of CD40L (Garlichs et al. 2001; Abu el-Makrem et al. 2009; Pignatelli et al. 2011; Ferroni et al. 2012). However, in the majority of recent papers, detection of soluble CD40L by ELISA is preferred to detection of surface CD40L, probably because of the transient expression of the latter (Ferroni et al. 2012).

Binding of FITC-conjugated annexin V is a widely utilized marker of platelet activation. A crucial step in the activation of the coagulation cascade is the result of a flip–flop in anionic phospholipids (predominantly phosphatidylserine; PS) from the inner to the outer leaflet of the platelet membrane bilayer, which forms a binding substrate for the prothrombinase complex (Fig. 1). Annexin V is a Ca2+-dependent phospholipid-binding protein that allows detection of PS exposure on the outer cell surface of activated platelets (Dachary-Prigent et al. 1993) (see Fig. 2 and Table 2).

Intracellular Markers

FCM can be utilized to evaluate intracellular markers such as specific δ-granule components (serotonin and adenine nucleotides) and specific phosphoproteins. Serotonin release from the δ-granules, where it is stored in steady-state conditions, is a crucial step for platelet aggregation (Fig. 1). However, its measurement has relied for years on expensive and laborious techniques (HPLC, ELISA, C14 labeling). Our group has described an FCM method for assessing intracellular serotonin content in resting and activated platelets after fixation and permeabilization using an anti-serotonin–R-phycoerythrin (RPE) conjugated antibody coupled with anti-CD41 surface marker staining (Gobbi et al. 2003) (see Fig. 2 and Table 2).

Adenine nucleotides can be detected by mepacrine (quinacrine) staining. Mepacrine is a green fluorescent dye capable of selectively binding adenine nucleotides and, consequently, provides information on the uptake and release of δ-granule content (Gordon et al. 1995).

Vasodilator-stimulated phosphoproteins (VASPs) are intracellular signaling molecules that are nonphosphorylated in the basal state and phosphorylated in prostaglandin E1-inhibited platelets. VASPs are targets of the ADP/P2Y12 receptor, a seven-transmembrane domain receptor linked to an inhibitory G-protein. The degree of VASP phosphorylation (reported as platelet reactivity index) reflects P2Y12 activity and can be measured using a commercial assay developed by BioCytex (Schwarz et al. 1999).

Platelet–Leukocyte Aggregates

As previously described in chapter “Platelet-Leukocyte Interactions” (Evangelista et al. 2017), it has been well known since 1989 that activated platelets bind to monocytes or neutrophils via P-selectin interaction with the PSGL-1 counter-receptor on the leukocyte surface (Larsen et al. 1989). It has been demonstrated that the level of circulating platelet–leukocytes aggregates (PLAs) is increased in pathological conditions such as ischemic stroke (Marquardt et al. 2009) and diabetes mellitus (Elalamy et al. 2008) and in patients with coronary artery disease (Mickelson et al. 1996; Furman et al. 1998; Ott et al. 1996; Neumann et al. 1997b; Sarma et al. 2002). PLAs are currently utilized as platelet activation markers (Nagasawa et al. 2013; Pearson et al. 2009) (Fig. 1 and Table 2).

FCM is a sensitive and rapid method for qualitative (mean fluorescence intensity) and quantitative (percentage positive events) measurement of PLAs. To minimize ex vivo platelet activation attributable to centrifugation and washing steps, whole blood analysis is recommended. In addition, sample fixation reduces uncontrolled changes in the platelet surface. Although some authors discourage red cell lysis because it can lead to artifactual activation (Pearson et al. 2009; Li et al. 1997), other investigators have used red cell lysis to provide clearer delineation of platelet and white cell populations, reporting that the lysis procedure does not affect platelet–monocyte aggregation (Furman et al. 1998; Ray et al. 2005; Harding et al. 2007).

Usually, FCM detection of PLAs is based on a combination of a platelet-specific antibody (usually CD41, CD42 and/or CD61) and leukocyte-specific antibody. Specifically, in whole blood analysis, leukocytes can be discriminated from erythrocytes by anti-CD45 binding. Moreover, the specific leukocyte population interacting with platelets can be identified using antibody against CD3 (T-lymphocytes), CD19 (B-lymphocytes), CD56 (natural killer cells), CD14 (monocytes), CD16 (neutrophil), anti-CD4, and anti-CD8 (to discriminate T cell subsets) (Michelson et al. 2001; Yip et al. 2013; Pearson et al. 2009; Nkambule et al. 2015) (Table 2).

Platelet-Derived Microparticles

As previously mentioned in chapter “Platelet-Derived Microparticles” (Cointe et al. 2017), platelet microparticles (PMPs) are fragments with a size ranging from 0.1 to 1 μm. They circulate in the bloodstream at a concentration of 100–1000/μL. They are shed from platelets when activated and during aging and destruction; therefore, they have a phospholipid-based structure with functional receptors belonging to the platelet membrane. Specifically, PMPs express common megakaryocyte-platelet glycoprotein receptors (CD41, CD42b) and platelet activation markers such as PS, P-selectin, and CD63. In recent years, there has been increasing interest in PMPs because of their clinical applications. PMPs are used as platelet activation markers ex vivo, such as after stimulation with specific agonists such as thrombin, collagen, and calcium ionophore A23187 (Michelson et al. 2001; Italiano et al. 2010) (Fig. 1 and Table 2), and also in vivo. In fact, aberrant levels of plasma PMPs have been associated with bleeding or thromboembolic complications and can be considered a biomarker of ongoing thrombosis (Ramacciotti et al. 2009).

PMP detection techniques have been implemented and new methodologies are currently under investigation, including TEM, ELISA, FCM, atomic force microscopy, nanoparticle tracking analysis, and resistive pulse sensing (Nomura et al. 2009; Dragovic et al. 2011; Yuana et al. 2010). At present, FCM is still the most widely utilized method for PMP identification.

Several methodological issues must be taken into account when analyzing PMPs from peripheral blood. PMP detection can be affected by several sample processing steps, ranging from blood collection, plasma isolation, and storage to staining of phospholipids and surface antigens for determining the cellular origin of the microparticles (Shah et al. 2008; Shet 2008).

Although the standardization of pre-analytical steps remains a challenge, an overall consensus has been reached on the following specific issues (Lacroix et al. 2010, 2012; Yuana et al. 2011): It is recommended that samples are collected carefully to avoid sheer stress and endothelial activation, using a light tourniquet, large needles (20–21G), and discarding the first 2–3 mL of blood. Citrate tubes are preferred over EDTA, because the latter is known to interfere with microparticle measurement (Shah et al. 2008; Trummer et al. 2009). In a hospital setting, it is important to note that patients’ samples are not collected in the laboratory, therefore sample transportation and time delay between blood sampling and microparticle preparation can impact the analysis. It was demonstrated that a time delay of 2 h before the first sample centrifugation is acceptable because any increase in the number of microparticles during this period remains moderate. Moreover, transporting blood tubes in the vertical rather than horizontal position limits the extent of in vitro microparticle generation (Lacroix et al. 2012).

Platelets need to be removed from the plasma to avoid cellular activation, which can lead to inadvertent production of microparticles. For this purpose, the most common protocol for FCM microparticle analysis consists of one centrifugation at 1500 g for 15 min followed by a second centrifugation at 13,000 g × 2 min to obtain platelet-free plasma (Yuana et al. 2011; Robert et al. 2009; Sabatier et al. 2002). However, some investigators perform the analysis directly on platelet-poor plasma obtained after a first blood centrifugation at 1500 g for 15 min and a second re-centrifugation of plasma at 1500 g for10 min (Jy et al. 2004; Enjeti et al. 2008; Dignat-George et al. 2009).

Lacroix and coworkers demonstrated that two successive centrifugations of 2500 g for 15 min at room temperature are more efficient for platelet removal (Lacroix et al. 2012). Dey-Hazra and colleagues also suggested that the filtration of buffer using a 0.2-μm filter is a useful step for reducing the amount of background noise, cell debris, and precipitates. The latter have the same size range as microparticles and could influence or disturb the analysis (Dey-Hazra et al. 2010).

In multicenter studies and prospective trials it is often inevitable that plasma samples are frozen and stored before performing the assay. In these cases, plasma is first snap-frozen in liquid nitrogen before being stored at −80 °C, or, alternatively, plasma is directly frozen at −80 °C (Yuana et al. 2011). It has been demonstrated that directly freezing at −80 °C and quickly thawing in a water bath at 37 °C at the time of the assay do not strongly influence PMP analysis (Lacroix et al. 2012).

Regardless of the biological source (peripheral blood or cell culture), one of the most important steps for PMP detection is the identification of PMP populations according to their light scattering properties. Indeed, PMPs are submicrometer fragments, therefore their discrimination from platelets and debris relies on the setting and resolution of the instruments used.

To promote standardization, the ISTH SSC Working Group on Vascular Biology proposed the use of calibrated latex beads with sizes of 0.5, 0.9, and 3 μm (MegaMix Beads BioCytex, Marseille, France) to adjust the instrument settings and increase the resolution of FCM. The dimensions of this bead population cover both the PMP (0.5 and 0.9 μm) and platelet populations (0.9 and 3 μm) (Lacroix et al. 2010; Robert et al. 2009). A multicenter study, performed with several types of FCM, demonstrated that standardization of PMP enumeration by FCM is feasible but dependent on the intrinsic characteristics of FCM and on the calibration strategy. Although latex beads remain an imperfect model for defining the cutoff of PMP populations, because the refractive index of the plastic beads is different to that of PMPs, they are currently considered useful standards that allow instrument qualification and follow up (Lacroix et al. 2010).

PMP staining with fluorescently labeled antibodies is recommended for more precise identification of PMP populations. PS is a well-established marker for PMPs, regardless of their cellular origin. Fluorophore-labeled annexin V is commonly used to measure the total number of PMPs with FCM. However, more specific platelet antigens should be used to distinguish PMPs from microparticles derived from leukocytes, erythrocytes, and endothelial cells. Von Willebrand factor and fibrinogen receptors are markers for microparticles derived from both platelets and megakaryocytes; however, monoclonal antibodies specific for platelet activation markers such as P-selectin, CD107A, and CD63 are recommended for selective identification of PMPs (Flaumenhaft et al. 2009).

To provide the concentration or absolute count of PMPs in a sample, counting beads of a known concentration are used. When they are used as internal standard, the beads are added to each sample before FCM measurement. For use as an external standard, the counting beads are processed at the same FCM settings/conditions used for the samples. FCM counts the number of beads and/or microparticles in the sample until the acquisition time is reached and, from this number, the concentration of microparticles can be calculated (Yuana et al. 2010; van der Zee et al. 2006; Robert et al. 2009; Shet et al. 2003).

Ex Vivo Platelet Activation

Platelet reactivity after ex vivo stimulation with different agonists can be performed on whole blood, PRP, or washed platelets (Fig. 2). Whole blood analysis allows detection of high degrees of platelet activation, limiting selective cell losses (Shattil et al. 1987).

Several agonists, both fully synthetic or isolated from organisms, are commercially available and allow highly reproducible and standardized ex vivo activation. Such agonists include ADP, stable thromboxane analogs, and PAR1 agonists such as TRAP-6 (SFLLRN) and TRAP-14 (SFLLRNPNDKYEPF) (Giesberts et al. 1995; Michelson et al. 1991). GPVI-specific agonists include collagen-related peptide [Gly-Lys (or Cys)-Hyp-(Gly-Pro-Hyp)10-Gly-Lys (or Cys)-Hyp-Gly] (Morton et al. 1995; Kehrel et al. 1998), the snake toxin convulxin (Polgar et al. 1997), and poly(Pro-Hyp-Gly), which induces platelet aggregation independently of thromboxane A2 and α2β1 (Inoue et al. 2009).

Shear stressed-induced activation can also be used: the friction forces generated by forcing the sample through a capillary tube produces a platelet activation state similar to that registered in vivo at an injury site or stenotic vessel. The activation markers that can be monitored are listed in Table 2.

Main applications of ex vivo platelet activation studies are monitoring of antiplatelet drugs and pathological conditions associated with platelet hypo- or hyperreactivity, as described next.

In Vivo Platelet Activation

An activated platelet phenotype may reflect an ongoing acute thrombotic process such as acute coronary syndrome, as well as chronic pathological conditions such as diabetes, peripheral arteriopathy, allergic asthma, or cancer (Tschoepe et al. 1991; Tomer 2004). Analysis of circulating activated platelets (i.e., in vivo platelet activation) requires extremely careful sample collection and manipulation to avoid activation induced by venipuncture and storage. If the sample cannot be processed within a short time frame, stabilization with activation antagonists and fixatives is strongly encouraged (Shattil et al. 1987; Dovlatova et al. 2014). Platelet subpopulations that are heterogeneous in their activation state can be detected; in fact, the lower limit of detection of platelet activation for FCM assays is about 1 % activated platelets in a sampled population of stimulated and unstimulated platelets (Shattil et al. 1987).

FCM is the most appropriate technique for in vivo platelet activation studies. Basically, all the FCM platelet activation markers previously described can be utilized to assess the activation state of circulating platelets (Table 2), although some concerns have been raised concerning the transient expression of P-selectin and the rapid reversibility of activated αIIbβ3 (Schmitz et al. 1998). The choice of marker relies mainly on the clinical application and the disease of interest. Platelet degranulation markers such as P-selectin have been utilized in hypertension (Stumpf et al. 2005) and activated αIIbβ3 in antiphospholipid antibody syndrome (Joseph et al. 2001). PLAs and PMPs have been extensively studied in solid and hematologic malignancies, acute myocardial infarction, coronary revascularization procedures, diabetes, and asthma (Michelson et al. 2001; Mallat et al. 2000; Pitchford et al. 2003; Falanga et al. 2005).

Platelet Turnover

Platelet Count

Obtaining accurate platelet counts has been a long-standing problem for both pathologists and clinicians, as demonstrated by the fact that platelet counting methods have evolved from manual phase contrast microscopy through the era of automated cell analyzers to FCM, which is currently considered the gold standard technique, as discussed in detail in chapter “Platelet Counting and Measurement of Platelet Dimension” (Noris and Zaninetti 2017).

In FCM, platelets are identified with a fluorescent monoclonal antibody specific to a cluster of differentiation common to all (resting and activated) platelets, with the advantage that platelets are immunologically recognized independently of size. Thus, FCM allows unequivocal identification of platelets from other cellular elements of similar size. This is of great relevance in thrombocytopenic samples, for which platelet counting is known to be problematic because of the presence of giant platelets or particles (i.e., fragmented red cells and immune complexes). In these cases, FCM platelet count (in contrast to impedance analysis or optical counting) can include giant platelets in the count because they are clearly resolved from red cells, and particles with sizes similar to those of platelets are excluded (Ault et al. 1999; Tanaka et al. 1996; Harrison et al. 2001).

Because most flow cytometers cannot be programmed to process a fixed volume of sample, counting procedures involve indirect derivation of platelet number. For this purpose, various approaches have been proposed (Tanaka et al. 1996; Davis and Bigelow 1999). Specifically, in the late 1990s, an immunoplatelet counting procedure was introduced that was based on a reference standard of fluorescent beads with a predefined concentration. In this method, a known amount of fluorescent calibration beads is added to the sample, and platelet number calculated as bead ratio (Dickerhoff and Von Ruecker 1995; Matzdorff et al. 1998). However, this method lacks standardization and an alternative procedure using FCM platelet count based on red blood cells as internal standard has been demonstrated to satisfy the criteria for a reference method (Harrison et al. 2001).

The International Council for Standardization in Haematology (ICSH) and the International Society of Laboratory Hematology (ISLH) recommend FCM-based counting, based on the ratio of platelets to red blood cells, as a reference method for platelet counting (International Society of Laboratory Hematology Task Force on Platelet Counting, International Council for Standardization in Haematology Expert Panel on Cytometry 2001).

The assay involves dilution of EDTA-anticoagulated blood specimen in a sterile buffered solution, followed by staining of platelets with specific fluorescent antibodies. The stained platelets in solution are diluted to the counting concentration, and the platelets and red blood cells (RBCs) counted on a flow cytometer with thresholds set to discriminate platelets from RBCs on the basis of fluorescence amplitude and scatter amplitude. The RBC/platelet ratio is determined, and the platelet count is calculated from an accurate RBC count of the sample, obtained using a cell counter that meets previous ICSH specifications (International Council for Standardization in Haematology 1994). The main advantage of this technique is that, by using RBCs as internal standard, the count obtained is independent of potential pipetting artifacts (Harrison et al. 2001).

Multicenter studies using different approaches have been undertaken to improve the accuracy of platelet counts (De la Salle et al. 2012; Masters and Harrison 2014; Sehgal et al. 2010; Sandhaus et al. 2002; van der Meer et al. 2012).

Reticulated Platelets

Reticulated platelets (or young platelets) are rich in mRNA. Platelet mRNA derives from megakaryocytes during thrombopoiesis, and, because of its instability, is present in a larger amount in young platelets than in older ones and can be utilized to identify newly released platelets from the bone marrow.

FCM is the method of choice for reticulated platelets analysis. This technique relies on the use of a fluorescent dye, thiazole orange (TO), which has a large fluorescence enhancement and high quantum yield upon binding to nucleic acids, especially RNA. Reticulated platelets can thus be distinguished from mature platelets by FCM according to their dye uptake (Kienast and Schmitz 1990; Bonan et al. 1993).

Protocols have been successively optimized to associate TO labeling with platelet-identifying antibodies for better characterization of platelet populations (Chavda et al. 1996) and a consistent number of papers describing clinical applications of reticulated platelet analysis have been published. However, it also became evident that the FCM assay is prone to methodological variation, which makes it difficult to compare results obtained with different assays; for example, the normal reference range is reported to lie between 1 and 15 % (Romp et al. 1994; Matic et al. 1998). Many factors that contribute to this analytical issue have been identified: type and concentration of fluorescent dye, incubation time and temperature, fixation, RNAse treatment, and FCM data analysis, including gating and threshold settings (Richards and Baglin 1995; Watanabe et al. 1995; Matic et al. 1998; Bonan et al. 1993; Rapi et al. 1998). One of the major problems is that platelets show non-RNA-specific binding of fluorescent dye, resulting in background staining, which is size-dependent (Matic et al. 1998; Robinson et al. 1998; Balduini et al. 1999). New initiatives have been undertaken that are aimed at developing a method with the potential to become a future international reference method (Hedley and Keeney 2013; Machin 2013). According to these studies, reticulated platelet analysis should be performed in whole blood, with stringent control of incubation time and temperatures and with sample fixation after labeling. Concerning gating strategy, platelets should be identified according to CD41 and CD61 expression and then analyzed using a dot-plot of SS versus TO. A threshold of 1 % for TO-positive platelets is defined using a negative control unstained for TO. Platelet enumeration is performed according to the ICSH guidelines using the platelet/RBC ratio as described above (Hedley et al. 2015; Hoffmann 2014).

The main clinical application of reticulated platelet assessment is to determine, in thrombocythopenic patients, whether the low platelet count is a result of abnormally accelerated destruction or of impaired bone marrow output. Because megakaryopoietic activity is low in patients with bone marrow failure, the assumption is that the ratio of reticulated platelets to total platelets is also low. By contrast, in conditions typified by enhanced compensatory megakaryocypoiesis (e.g., immune thrombocytopenia or postchemotherapy recovery) the absolute number of reticulated platelets, and consequently the ratio of reticulated platelets to total platelets, is expected to be high (Kienast and Schmitz 1990; Pons et al. 2010; Thomas-Kaskel et al. 2007; Rinder et al. 1998; Ryningen et al. 2006; Romp et al. 1994; Richards et al. 1996; Catani et al. 1999; Stohlawetz et al. 1999). Interestingly, in the thrombocytopenic phase after chemotherapy and transplantation for hematological malignancies, reticulated platelets have been monitored by FCM and it has been observed that an increase in reticulated platelets precedes the recovery of platelet count by 2–3 days. This creates the opportunity to defer platelet transfusions that would be given if transfusion decisions were based only on platelet count. However, clinical evidence supporting this concept is limited and there is need for randomized, controlled clinical studies in this field (Macchi et al. 2002; Wang et al. 2002; Michur et al. 2008; Chaoui et al. 2005).

Reticulated platelets have also been utilized for risk assessment and drug monitoring in patients with coronary artery disease (Guthikonda et al. 2008; Lakkis et al. 2004; McBane et al. 2014; Perl et al. 2014).

Recently, Hedley and colleagues (Hedley et al. 2015) demonstrated that macrothrombocytopenic patients show an increase in the immature platelet fraction that, in agglutination tests, increases during the formation of platelet clumps. The data indicate that measurement of the immature platelet fraction is influenced by platelet size and could be a useful parameter in the differential diagnosis of macrothrombocytopenia (Miyazaki et al. 2015; Hoffmann 2014).

Clinical Applications

Clinical Disorders Associated with Abnormal Platelet Reactivity

Several thrombotic and nonthrombotic disorders are typified by alterations in the number of circulating activated platelets and an abnormal expression of platelet activation markers. Both these conditions can be evaluated by FCM.

Coronary Syndromes

FCM studies have demonstrated the presence of circulating activated platelets in patients affected by coronary artery disease (CAD), including stable angina, unstable angina, and acute myocardial infarction (AMI), as well as after percutaneous coronary intervention (PCI). In this setting, analysis of platelet activation markers is not only beneficial in terms of diagnosis but also provides information on clinical outcome and can be utilized for the optimization of antiplatelet therapies

Specifically, CD62P-expressing platelets and circulating PLAs at baseline and after stimulation with exogenous agonists are higher in patients with stable CAD, unstable angina, and AMI than in healthy subjects (Furman et al. 1998; Langford et al. 1996; Carubbi et al. 2012).

Interestingly, these two platelet activation markers, both detected by FCM, show a different sensitivity. PLAs are a more sensitive marker of in vivo platelet activation than CD62P in patients that have undergone PCI and with AMI, and are therefore potential candidates as early diagnostic markers of AMI (Michelson et al. 2001; Furman et al. 2001).

Other studies reported an increase in PMPs, CD40L, and platelet collagen receptor (GPVI) in acute coronary syndrome (Katopodis et al. 1997; Bigalke et al. 2007; Ferroni et al. 2012; Skeppholm et al. 2012).

Interestingly, analysis of the expression of platelet activation markers can be of prognostic value for thrombotic risk after PCI. It has been demonstrated that patients developing acute events after PCI showed a higher expression of platelet CD62P, CD63, and the active form of αIIbβ3 before intervention. The authors suggest that FCM analysis of platelet function in PCI candidates can stratify patients into those at high and low risk of cardiac events after PCI (Kabbani et al. 2001, 2003; Tschoepe et al. 1993; Gawaz et al. 1997).

Cerebrovascular Ischemia

It has been extensively demonstrated that almost all platelet activation markers, including CD62P, CD63, activated αIIbβ3, PMPs, and PLAs, are increased in patients with cerebrovascular ischemia (McCabe et al. 2004; Cao et al. 2009; Grau et al. 1998; Zeller et al. 1999; Meiklejohn et al. 2001; Yamazaki et al. 2001; Cherian et al. 2003; Yip et al. 2004; Smout et al. 2009; Tsai et al. 2009; Marquardt et al. 2002; Htun et al. 2006; Minamino et al. 1998; Bigalke et al. 2010; Lee et al. 1993; Geiser et al. 1998; Fateh-Moghadam et al. 2005; Koyama et al. 2003). Interestingly, the expression levels of these markers have been associated with different subtypes of ischemic stroke.

Specifically, patients with large-vessel cerebral infarction elicit higher platelet activation than those with small-vessel infarction, whereas the role of platelet activation is still controversial in cardioembolic stroke (Cao et al. 2009; Grau et al. 1998; Zeller et al. 1999; Meiklejohn et al. 2001; Yamazaki et al. 2001; Cherian et al. 2003; Yip et al. 2004; McCabe et al. 2004; Smout et al. 2009; Tsai et al. 2009; Marquardt et al. 2002; Oberheiden et al. 2012; Turgut et al. 2011). Furthermore, FCM analysis of CD62P, CD63, and PMPs suggests that platelet activation occurs chronically in these diseases, as demonstrated by the fact that, despite the specific effects of antithrombotic therapies, their expression levels are higher at 1, 3, and 6 months before the acute event than in healthy controls (Grau et al. 1998; Cherian et al. 2003; Meiklejohn et al. 2001; Yamazaki et al. 2001; Marquardt et al. 2009).

Aggregation with different leukocyte subtypes was also investigated. Results showed that an increase in monocyte–platelet aggregates is short-lived and could reflect an acute reaction to cerebral ischemia. By contrast, granulocyte–platelet aggregate formation persists into the subacute phase, suggesting that specific PLAs could reflect the prothrombotic state and the inflammatory processes after stroke (Marquardt et al. 2009).

Peripheral Vascular Disease

The use of FCM to assess platelet function in patients with peripheral arterial disease (PAD) has been documented since 1991. Data published in the last 10–15 years show an increase in platelet hyperreactivity and circulating activated platelets in these patients compared with healthy subjects. It has been demonstrated that P-selectin expression, the number of PMPs, and the number of platelet aggregates are significantly higher for both resting and stimulated platelets in the PAD group compared with controls (Robless et al. 2003; Cassar et al. 2003; Koksch et al. 2001; Zeiger et al. 2000). Interestingly, platelet activation parameters are correlated with the severity of vascular disease. P-selectin and PLAs, specifically platelet–monocyte aggregates (PMAs), are significantly increased in patients with severe limb ischemia compared with those with intermittent claudication (Rajagopalan et al. 2007; Tan et al. 2005). Moreover, it has been demonstrated that in the early postoperative period after infrainguinal bypass, the level of PLAs, including PMAs and platelet–neutrophil aggregates, is significantly greater in patients who had experienced graft occlusion compared with patients whose graft remained patent at 6 months. The authors suggest that this analysis allows identification of a patient subset with a high risk of graft occlusion that would benefit from more aggressive antiplatelet therapy (Esposito et al. 2003).

All patients affected by peripheral venous disease, irrespective of the degree of chronic venous insufficiency, are characterized by higher levels of PMAs and platelet–neutrophil aggregates in the bloodstream and an increased propensity to form platelet aggregates in response to platelet agonists than control group subjects. However, this increase in platelet reactivity appears unrelated to the presence of ulceration (Powell et al. 1999). Moreover, higher circulating levels of PMAs persist after complete correction of chronic venous insufficiency, suggesting that the increased number of PMAs identified in these patients is not secondary to the presence of venous reflux, but could be involved in the primary etiology of chronic venous insufficiency (Rohrer et al. 2002).

Cancer

The association between platelet and cancer dates to the mid-1800s, when Armand Trousseau linked venous thrombus formation with an underlying undiagnosed malignancy (Dammacco et al. 2013). Since then, several studies have reported platelets as major players, not only in thrombus generation in neoplastic patients, but also in promoting tumor growth, invasiveness, and angiogenesis. Many authors now point to the “platelet–cancer loop” as a pivotal mechanism in tumorigenesis (Hasselbalch 2014). In this setting, detection of platelet activation markers by FCM is as a useful tool. For solid tumors, most reports concern tumor-derived microparticles. Toth and colleagues showed that PMPs are higher in breast cancer patients than in those with benign lesions. PMP levels correlate with tumor invasiveness, but no association was established with prothrombin levels and thrombin formation (Toth et al. 2008). In hormone-refractory prostate cancer, Helley and coworkers demonstrated a correlation between circulating PMP levels and Gleason score, patient performance status, and overall survival, with a poorer outcome in patients with more than 6867 PMPs/μL (Helley et al. 2009). In patients with gastric cancer, the mean number of CD62P molecules on the platelet surface was significantly higher than in the healthy group, and further increased after stimulation with TRAP (6- to 12-fold in neoplastic patients and threefold in the control group) (Osada et al. 2010).

Platelet activation has been extensively investigated in hematologic malignancies. Myeloproliferative disorders (MPD) are characterized by an increased risk of thrombo-hemorrhagic events, probably related to platelet dysfunction (Finazzi et al. 1996; Wehmeier et al. 1997). Platelet aggregation studies in MPD have revealed a tendency toward spontaneous platelet aggregation and a correlation between agonist-induced platelet hyperactivation and a history of thrombosis (Balduini et al. 1991). Similarly, FCM analyses showed an increased percentage of CD62P-expressing platelets, PLAs, and PMAs in the bloodstream of patients with MPD compared wth healthy controls (Jensen et al. 2000, 2001; Villmow et al. 2002).

Additionally, MPD patients with a previous history of thrombosis or microvascular disturbances had a higher percentage of platelet–granulo/monocytes and PMAs than patients with no history of these events (Jensen et al. 2001). Moreover, in vitro formation of platelet–polymorphonuclear leukocyte aggregates was enhanced in essential thrombocythemia patients without pharmacological antiaggregation but reduced in essential thrombocythemia patients treated with aspirin (Falanga et al. 2005).

Hemorrhage is one of the principal symptoms associated with acute myeloid leukaemia (AML). It may be responsible for a lethal course of the disease and is related to different factors, including thrombocytopenia and defects in platelet function (Cowan et al. 1975; Estey et al. 1982). Leinoe and colleagues demonstrated that in vitro expression of CD62P was reduced in AML patients with bleeding tendency, suggesting that FCM analysis of platelet function is a putative biological marker of hemorrhage in this disease (Leinoe et al. 2004).

AML can arise in a context of genetic predisposition, as in the case of familial platelet disorder with a predisposition to acute myelogenous leukemia (FPD/AML), an autosomal dominant platelet disorder characterized by thrombocytopenia, platelet function defects, and a lifelong risk of the development of hematologic neoplasms. The disorder is associated with germline heterozygous mutations in the transcription factor gene RUNX1. Analysis of defective αIIbβ3-dependent activation pathways has been documented as reduced binding of FITC-conjugated PAC-1 and Alexa Fluor 488-conjugated fibrinogen (Glembotsky et al. 2014).

Monitoring of Antiplatelet Therapies

FCM is a useful tool for monitoring antiplatelet therapies. Medical therapy to reduce platelet activation is the mainstay in prevention of atherothrombotic events in many vascular diseases. Several antiplatelet drug are available, the most popular being aspirin and clopidogrel (a P2Y12 inhibitor), but other P2Y12 and GPIIb/III antagonists have been developed, including prasugrel, ticagrelor, and tirofiban (Ford 2015; Rollini et al. 2016; Savonitto et al. 2015). Clinical trials assessing the efficacy of these drugs have been mainl based on evaluation of platelet reactivity using FCM analysis of both surface antigens and intracellular VASP.

VASP assay, based on FCM, is a reliable index of platelet ability to be activated by ADP. The degree of VASP phosphorylation can be measured in a direct, cost-effective, and automated manner as described above (see “Intracellulars Markers of Platelet Activation”). This method has been used to detect clopidogrel resistance and to compare the efficacy of different P2Y12 inhibitors (Aleil et al. 2005; Bednar et al. 2015).

Detection of surface marker CD62P is used to evaluate the contribution of antithrombotic therapy on platelet activation after acute coronary syndrome and after coronary intervention (Gawaz et al. 1996; Neumann et al. 1997a; Ault et al. 1999). Other activation markers can be utilized, usually in combination with P-selectin, to monitor antiplatelet therapies. Indeed, results from the PRINCIPLE-TIMI 44 study show a correlation between platelet reactivity before and after P2Y12 blockade, as assessed by FCM measurement of CD62P and PMAs (Frelinger et al. 2011). Moreover, we have described an FCM-based method for assessing the effects of clopidogrel and tirofiban on platelet activation using 5 μL of whole blood as starting material. Peripheral blood samples were collected from patients before drug administration and after 2, 6, and 24 h of treatment with tirofiban and clopidogrel alone and in combination. Pharmacological treatment was able to induce modulation of platelet reactivity to ADP-induced activation, as assessed by PAC-1 and P-selectin (Solinas et al. 2009).

Monitoring platelet reactivity is extremely relevant, not only in CAD patients but also in the setting of ischemic stroke, in which it is well established that antiplatelet therapy reduces the frequency of secondary events (Hennekens 2002; Smith et al. 1999; Serebruany et al. 2004, 2005; Yip et al. 2004; Grau et al. 2003; Klinkhardt et al. 2003; Moshfegh et al. 2000). FCM detection of CD62P and CD63 expression shows that platelet activity is significantly more suppressed in patients on clopidogrel than in those taking aspirin in the subacute and convalescent phases of non-cardioembolic ischemic stroke. No time-dependent modulation of CD62P expression could be detected in patients on anticoagulants (warfarin) (Tsai et al. 2010; Yip et al. 2004). An innovative and promising use of FCM is documented by Serebruany and coworkers, combining aggregometry tests with FCM determination of 14 platelet surface receptors. The authors found a marked heterogeneity of platelet characteristics in patients after ischemic stroke, suggesting that bleeding complications and hemorrhagic transformation after aggressive antiplatelet regimens could be related to the decreased or normal baseline platelet characteristics in such patients (Serebruany et al. 2004).

Diagnosis of Platelet Function and/or Number Disorders

Among the clinical applications of FCM platelet analysis, diagnosis of specific platelet function disorders (PFDs) is one of the most relevant. PFDs encompasses a heterogeneous group of both inherited and acquired diseases that affect platelet function and/or number and lead to a defective primary hemostasis. Disorders include Bernard-Soulier syndrome (BSS), Glanzman thromboasthenia (GT), platelet-type von Willebrand disease, storage pool diseases, Scott syndrome, heparin-induced thrombocytopenia (HIT), immune-mediated thrombocytopenias, and rare inherited conditions associated with somatic defects, such as MYH9-related diseases and thrombocytopenias associated with skeletal defects (Carubbi et al. 2014b).

The diagnostic laboratory work-up for PFDs is challenging and involves several methodologies, including bleeding time, light transmission or impedance aggregometry (the leading assay for investigating PFDs), PFA-100, lumiaggregometry, HPLC, fluorescence microscopy, TEM, ELISA, and radioimmunoassay (Paniccia et al. 2015).

The main drawbacks of these tests are a high false-positive rate (~20 %), unreliability in cases of platelet count below 50 × 109/L, lack of standardization, and variation in result interpretation. Additionally, their use is still limited to specialized laboratories (Pai and Hayward 2009; Miller 2009).

In this scenario, FCM is a rapid, reliable, and feasible technique for the diagnosis of PFDs characterized by surface glycoprotein deficiency (e.g., GT and BSS, for which it is the method of choice), storage pool disease, and Scott syndrome (Carubbi et al. 2014b; Nurden and Nurden 2014).

In more detail, diagnosis of BSS is based on the demonstration of GPIb-IX-V deficiency by specific monoclonal antibodies targeting glycoproteins Ib, IX, and V. By this means, abnormalities in the complex structure and levels of expression can be detected, also allowing discrimination between homozygous and heterozygous states (Andrews and Berndt 2013).

In GT, FCM analysis of αIIbβ3 integrin expression by monoclonal antibodies recognizing GPIIb (CD41) and GPIIIa (CD61) is used as a confirmatory diagnostic tool. FCM can determine the levels of expression of αIIbβ3 per platelet, allowing GT patients to be subclassified as having type I, II, or III disease according to the amount of αIIbβ3 present per platelet (respectively <5 %, 10–20 %, or equal to 50 % of the normal amount) (Nurden et al. 2012). Moreover, in heterozygous patients, this methodology can establish αIIbβ3 levels in various platelet populations and, consequently, whether the clinical picture is the result of a global reduction in antigen expression by all platelets or the coexistence of normal platelets and platelets lacking αIIbβ3 (Sharp et al. 1998).

Auto- and allo-antibodies or paraproteins directed against αIIbβ3 can mimic inherited GT and are responsible for so-called acquired GT (aGT). aGT is an extremely rare bleeding disorder whose diagnostic work-up relies on complex and time-consuming laboratory assays that demonstrate the presence of circulating proteins interfering with platelet function. The assays include aggregation tests, platelet adhesion to a collagenated surface, ELISA, and mixing studies. Giannini and coworkers describe an FCM-based method for investigating aGT by assessing PAC-1 and fibrinogen binding to patient’s platelets (absent), expression of the αIIbβ3 complex on patient’s platelets using different monoclonal antibodies against GPIIb (clones SZ22, P2, A2A9/6; decreased) or GPIIIa (clones SZ21, SAP; normal), and PAC-1 and A2A9/6 binding to control platelets in the presence of patient’s serum (reduced). Overall, FCM emerged as the only test able to characterize both the functional effect and the molecular target of the patient’s autoantibody on platelets (Giannini et al. 2008).

FCM has expedited the diagnostic procedure for δ-storage pool disease. FCM assay relies on the same principles as fluorescence microscopy (for decades the gold standard in the diagnosis of these disorders), detecting the fluorescence of mepacrine, a dye that selectively binds to adenine nucleotides stored in δ-granules (Gordon et al. 1995).

FCM can rapidly and effectively detect Scott syndrome platelet abnormality, characterized by defective scrambling of membrane PS, which fails to be exposed on the outer membrane leaflet after platelet activation. This defect can be feasibly identified by the annexin V binding test, which is based on annexin V binding to phospholipid exposed after activation and can reveal a lack of PS exposure (Zwaal et al. 2004).

Experimental applications of FCM in PFDs include assessment of serotonin release in δ-granule storage disease (Gobbi et al. 2003), quantification of von Willebrand binding induced by ristocetin to fresh or formalin-fixed donor platelets for differential diagnosis between platelet-type von Willebrand disease and von Willebrand disease (Giannini et al. 2007, 2010), and analysis of defective platelet activation pathways in the rare FPD/AML disorders (Glembotsky et al. 2014).

Additionally, FCM can offer a simple, easy-to-perform diagnostic pre-test for patients with a bleeding history suggestive of PFD, as proposed by Dovlatovla and colleagues. The authors describe an FCM-based platelet function test to select, among patients with excessive bleeding, those that would benefit from further, extensive platelet phenotyping. The sample is stabilized for up to 9 days using a fixing solution (PAMFix; Platelet Solutions, Nottingham, UK) and then tested for P-selectin, as a marker of α-granule secretion and a general indicator of platelet reactivity, and for CD63 to evaluates δ-granule secretion after stimulation with combinations of ADP, the thromboxane A2 analog U46619, arachidonic acid, epinephrine, and TRAP. The assay shows good agreement with conventional lumiaggregometry and could thus be an appealing screening tool in this complex clinical scenario (Dovlatova et al. 2014).

FCM currently holds a marginal but expanding role in the routine diagnostic work-up of acquired thrombocytopenias such as immune-mediated thrombocytopenias and heparin-induced thrombocytopenia (HIT). In the case of HIT, FCM rapidly detects platelet activation induced by heparin-dependent cell-activating anti-PF4/heparin antibodies by annexin V binding (Tomer 1997), serotonin release (Gobbi et al. 2003), CD62P expression (Vitale et al. 2001), and platelet microparticle formation (Mullier et al. 2010).

Immune-mediated thrombocytopenias include a wide group of disorders typified by a reduction in platelet number as a result of production of antibodies directed against self-platelet antigens (immune thrombocytopenia; ITP) or neonatal/donor platelet antigens, most commonly human platelet antigen (HPA)-1a, as in the case of neonatal alloimmune thrombocytopenia, (NATP) and post-transfusion purpura (PTP). Innovative FCM assays have been developed to identify circulating autoantibodies and platelet-bound autoantibodies in ITP (Tomer et al. 2005; Tomer 2006) and to detect human leukocyte antigen-directed antibodies in the case of alloimmunization after repeated platelet transfusions (Carrick et al. 2011). FCM has proven to be a very sensitive method for detection of alloantibodies, as it is capable of detecting very small amounts of platelet-bound antibodies.

Interestingly, Freliger and colleagues utilized FCM-based platelet activation tests to investigate bleeding tendency in ITP patients. The following markers of activation were considered: CD62P expression, PAC-1-binding, and TRAP-stimulated platelet surface CD42b. The authors conclude that unstimulated platelet surface P-selectin and CD42b expression, together with higher levels of immature platelet fraction and platelet forward light scatter, were associated with a higher bleeding score, independently of platelet count, and could consequently be considered FCM markers of bleeding risk in ITP patients (Frelinger et al. 2015).

In the diagnostic algorithm of NATP, FCM is used to test maternal serum against paternal and maternal platelets and a small panel of platelets from normal group O donors typed for selected common HPA antigens. In the assay, washed platelets are sensitized with maternal or control serum for up to 60 min at room temperature. Platelets are then carefully washed to remove nonspecific immunoglobulins. Platelet-bound antibodies are detected with a fluorescently labeled (usually FITC) polyclonal or monoclonal antibody specific for human immunoglobulin. Results are expressed as the ratio of the fluorescence (mean or peak) emitted by normal platelets sensitized with maternal serum to that emitted by normal platelets incubated in normal serum. To prevent nonspecific binding of the immunoglobulin probe via Fc receptors on the target platelets, the probe antibodies are enzyme-treated to remove the Fc end of the molecule. A second fluorescent label, usually phycoerythrin (PE), can be attached to an anti-human IgM probe to detect IgM anti-platelet antibodies. Tests can be run simultaneously on the same sample of washed sensitized platelets, allowing detection of anti-platelet IgG and IgM in the course of the same acquisition (McFarland 2003; Peterson et al. 2013; Curtis and McFarland 2009). The same assay also applies to PTP, as described in the section “Platelet Crossmatching”.

A common potential drawback of the above-described FCM-based method of detection of serum alloantibodies is the fact that it does not differentiate between platelet-specific (i.e., platelet glycoprotein-directed) and non-platelet-specific (i.e., human luekocyte antigen- and ABO-directed) antibodies, leading to potential non-platelet-specific reactivity.

In addition to the detection of circulating anti-HPA-1a antibodies, FCM has been proposed as a rapid, simple, and reliable tool for platelet immunophenotyping in the setting of large-scale screenings to identify HPA-1a-negative subjects, potentially at risk of NATP, PTP, and refractoriness to platelet transfusion (Killie et al. 2004; Sorel et al. 2004; Tazzari et al. 1998). Of these three alloimmune disorders, NATP requires parental platelet antigen typing when a personal obstetric history suggestive of NATP is present, or in the case of maternal sister(s) with an obstetric history of laboratory-confirmed NATP or suggestive of NATP.

FCM-based platelet immophenotyping relies on the fact that polyporphisms of the HPA-1 system antigen, which account for alloimmune thrombocytopenias, result in different expression of binding sites for monoclonal antibodies of platelet membrane glycoproteins. Weiss and coworkers (Weiss et al. 1995) first characterized a monoclonal antibody, SZ21, directed against GPIIIa and able to distinguish between HPA-1a and HPA-1b genotypes because of its markedly reduced binding to platelets with the HPA-1b genotype. Starting from this finding, Schwippert-Houtermans and coworkers developed and standardized an FCM method for classifying the HPA-1 genotype, utilizing SZ21 monoclonal antibody (Schwippert-Houtermans et al. 2001). This method was subsequently implemented by coupling SZ21 with P2 antibody (targeting the αIIbβ3 complex), which allows discrimination of HPA-1a/1a phenotype from HPA-1a/1b thanks to absence of overlapping of P2/SZ21 mean fluorescence intensity ratios between the two phenotypes (Sorel et al. 2004).

FCM analysis using immunofluorescence labeling with specific alloantisera has been described (Forsberg et al. 1995; Tazzari et al. 1998), but is not able to distinguish HPA-1a/1b hetererozygous from HPA-1a homozygous subjects.

Today, the development of DNA-based methods allowing simultaneous genotyping of as many as 17 HPAs has made serological typing for HPA antigens obsolete (Bertrand and Conti 2015).

Blood Bank Applications

FCM is available in platelet blood banking, especially for (1) enumeration of residual white blood cells in red blood cells and platelet concentrates (PCs); (2) determination of platelet function in PCs by measuring platelet activation markers; (3) sterility testing of PCs to assess bacterial risk; and (4) platelet crossmatching.

Enumeration of Residual White Blood Cells in Red Blood Cells and Platelet Concentrates

FCM methods for counting residual white blood cells (WBC) in PCs are well established and widely used (van der Meer et al. 2012; Fischer et al. 2012). Traditionally, the methods are based on propidium iodide DNA staining for WBC detection and on fluorescent beads, at known concentration, as internal standard for leukocyte enumeration (Barclay et al. 1998; Backteman et al. 2002; Dijkstra-Tiekstra et al. 2004; Santana and Dumont 2006). Other FCM methods have also been proposed; for example, using a fluorescent anti-glycophorin A antibody that allows the simultaneous enumeration of residual WBCs and residual RBCs (Schmidt et al. 2009), and the use of specific WBC fluorescent antibody to discriminate WBCs from other nucleated cells (i.e., nucleated red blood cells) that are responsible for the overestimation of WBC content (Fischer et al. 2012).

Determination of Platelet Function in Platelet Concentrates by Platelet Activation Markers

Platelet transfusions are routinely used as life-saving procedures during surgery, myeloablative therapies, and in patients with particular bleeding disorders. The primary objective of platelet transfusion is to provide a sufficient amount of platelets with preserved hemostatic function. Unfortunately, platelet storage and pathogen reduction technologies (PRTs) cause a decrease in functionality over time, often referred to as “platelet storage lesion” (PSL) (Thon et al. 2008). In vitro platelet function tests, mainly FCM tests, have been used to determine the platelet reactivity of PCs during storage and to compare it with that of PCs prepared with different PRTs.

Although FCM detection of platelet-derived extracellular vesicles and LAMP-1 (lysosme-associated membrane protein-1) have been proposed as new markers of PSL (Black et al. 2015; Sodergren et al. 2015; Pienimaeki-Roemer et al. 2014), the assessment of platelet functionality in PCs primarily relies on the evaluation of activation-dependent changes in platelet surface markers in a resting state and after agonist stimulation (Cardigan et al. 2005). The physiological increase in CD62p, αIIbβ3, and fibrinogen binding in response to platelet agonists is hampered during platelet storage (Curvers et al. 2004; Lozano et al. 1997; Rock et al. 2003; Leytin et al. 2004), whereas GpIII expression is enhanced. This increase, measured both by western blot and FCM, is a result of translation mechanisms that, in turn, may be involved in the initiation or exacerbation of PSL (Thon and Devine 2007).

Several reports now associate GPIbα shedding and increased expression of CD63 and PS to augmented platelet clearance, leading to reduced post-transfusion platelet survival (Ohto and Nollet 2011; Canault et al. 2010; Albanyan et al. 2009a, b; Metzelaar et al. 1993).

A controversial aspect of FCM application in PC storage is the evaluation of P-selectin expression for quality control. Because P-selectin expression is the most commonly applied parameter of platelet activation in PCs stored in blood banks, efforts have made to standardize its measurement (Middelburg et al. 2013; Curvers et al. 2008). Expression of P-selectin is widely used as predictor of platelet survival and function in vivo; furthermore, FCM detection of CD62p is often used to study the effects of pathogen reduction strategies on PSL (Ostrowski et al. 2010; Galan et al. 2011; Johnson et al. 2011; Castrillo et al. 2013; Ignatova et al. 2015).

However, some authors debate the use of P-selectin as a predictor of platelet survival in vivo, providing data in support of the fact it does not mediate platelet clearance. Specifically, in a nonhuman primate model, infused degranulated platelets rapidly lose surface P-selectin to the plasma pool but continue to circulate and function in vivo (Michelson et al. 1996). Moreover, platelets from wild-type and P-selectin knockdown mice had identical life spans, and, in a thrombocytopenic rabbit model, thrombin-activated human platelets that lose surface P-selectin survive and are effective in the rabbit circulation for as long as fresh human platelets (Berger et al. 1998; Krishnamurti et al. 1999).

Future studies are needed to clarify how PC studies translate to platelet function in vivo after transfusion (Cardigan et al. 2005).

Sterility Testing of PCs to Assess Bacterial Risk

One of the most frequent infectious complications in platelet transfusion therapy is related to bacterial contamination of PCs. To preserve platelet function, PCs are stored at room temperature and this facilitates bacterial contamination (Vollmer et al. 2012). Several works have demonstrated that sterility testing of PCs by FCM is a feasible approach both for buffy coat-derived PCs and PRP-derived PCs (Mohr et al. 2006a, b; Schmidt et al. 2006a, b; Lee et al. 2012). Traditionally, TO staining is used for detection of bacteria and, before testing, it is suggested that the PC sample is incubated for 20–24 h at 37 °C, which should be prolonged for up to 2 days for slow-growing bacteria (Mohr et al. 2006a). This inevitably protracts the assay results. In 2009, Dreier and coworkers introduced an FCM-based assay (BactiFlow) that fulfils the requirements for a point-of-issue testing of PCs with a time-to-result of approximately 1 h, combined with a high sensitivity of 150 colony forming units/mL (Dreier et al. 2009). This method was subsequently implemented as a routine method for the identification of contaminated PCs (Vollmer et al. 2011). Multicenter studies have since validated BactiFlow as a very convenient test for PC bacterial screening (Vollmer et al. 2012).

Platelet Crossmatching

Transfusion of crossmatch-compatible platelets is a consolidated strategy for transfusion of alloimmunized patients, who show refractoriness to platelet transfusion (Rebulla 2005; Rebulla et al. 2004).

Serum from alloimmunized patients can be crossmatched with platelets from PCs already available for transfusion or with frozen or refrigerated aliquots of platelets from potential donors. HLA-specific and HPA-specific antibodies in the patient’s plasma react with platelets expressing incompatible antigens. Only platelet components that are compatible are transfused (Stroncek and Rebulla 2007). Several methods have been used to crossmatch patient samples, including ELISA, platelet immunofluorescence (PIFT), solid phase red cell adherence assay (SPRCA), and FCM. In FCM tests, donor platelets are incubated with a patient’s serum or plasma. Binding of the patient’s antibodies on donor platelets is detect using fluorescently labeled anti-human IgG antibodies. A platelet-specific monoclonal antibody (such as anti-CD41) should be added so that the analysis is performed exclusively on the platelet population (Sayed et al. 2011). The test must include the donor serum as negative control. Data are usually presented as the fluorescence ratio, defined as the ratio between the fluorescence intensity of gated platelets after incubation with the patient’s serum and the fluorescence intensity of the negative control (Dohlinger et al. 2005; Sayed et al. 2011). It has been demonstrated that platelet donor selection using FCM crossmatch gives a better clinical outcome for transfusion in many thrombocytopenic alloimmunized patients (Sintnicolaas and Lowenberg 1996; Sayed et al. 2011). Moreover, this method has proved useful for comparing the detection of platelet antibodies in fresh and frozen cells in attempts to standardize the storage of donated platelets (Dohlinger et al. 2005).

Other Applications

Intraplatelet Production of Reactive Oxygen Species

Platelet aggregation is associated with considerable production of reactive oxygen species (ROS), which is not counterbalanced by adequate intracellular content of antioxidants (Krotz et al. 2004). Generation of intraplatelet ROS occurs in both physiological and pathological conditions (Ghoti et al. 2007; Amer et al. 2005; Becatti et al. 2013; Mondal et al. 2015) and has been implicated in the regulation of αIIbβ3 activation, granule secretion, platelet shape change, and, more generally, in platelet reactivity (Begonja et al. 2005).

Detection and quantification of ROS production is relevant for measurment of platelet oxidative stress, and FCM offers a simple and rapid assay for this. The technique essentially relies on the incubation of platelets with dyes that can diffuse across the cell membrane. In basal conditions the dyes are not fluorescent, but, when oxidized by ROS, they emit a fluorescent signal that is detected by the flow cytometer. The most commonly used dyes are 2′,7′-dichlorodihydrofluorescein diacetate (H2-DCFDA) and dihydrorhodamine 123 (DHR). Fluorescence is measured in the FL1 green channel. The test can be performed directly on whole blood.

Calcium Flux

Cytosolic free calcium ions (Ca2+) are important second messengers and markers of platelet activation and reactivity. Therefore, quantitative measurement of intraplatelet calcium provides additional information for evaluation of platelet status. Traditionally, Ca2+ dynamics has been studied by cuvette-based methods, whose main drawback is the necessity of platelet isolation steps that are not only time consuming, but can also interfere with platelet activation status. One of the first FCM-based methods for the assessment of intraplatelet Ca2+ dynamics relied on the use of Indo-1 as fluorescent dye. However, excitation of Indo-1 requires a laser in the UV range (not standard on most flow cytometers) and the staining precludes the combined use of compounds that exhibit fluorescence when excited by UV light (i.e., oxidized lipids/lipoproteins), so this method has very limited application (Dustin 2000). Other dyes have been proposed that have the advantage of being excited by a standard blue argon laser, such as fluo-3-acetoxymethyl ester (Fluo-3) and its derivative, the brighter and more photostable Fluo-4. The fluorescence of Fluo-4 can be detected in the FL1 green channel and is enhanced by binding of Ca2+. Cellular loading and accumulation of these dyes results from esterase-mediated cleavage that convert the lipophilic (pro)dyes into polar membrane-impermeable forms that thereby also acquire the ability to bind Ca2+. These dyes can be coupled with anti-CD61 (Labios et al. 2006) or anti-CD41 (do Ceu Monteiro et al. 1999) antibodies that identify platelet populations. Tests can be run both on PRP and whole blood.

More recently, Assinger and coworkers described a method that, combining Fluo-4 with Fura Red, (another acetoxy-methyl ester derivative whose fluorescence is decreased by Ca2+ binding and is detected in the FL3 red channel) can rule out an enhanced signal resulting from the presence of aggregates (Assinger et al. 2015).

Platelet Cytoskeleton Studies

Actin polymerization in filamentous form (F-actin) is an early event in platelet activation and is associated with shape change, granule centralization, and glycoprotein redistribution. Additionally, in the basal state, F-actin content varies between individuals (Oda et al. 1992) and can be increased in pathological conditions such as type 1 diabetes (Spangenberg et al. 1989).

Variations in platelet F-actin content can be rapidly detected by FCM. This method is uses fluorescent phalloidin and phallacidin derivatives for labeling and quantifying actin filaments. These phallotoxins (isolated from Amanita Phalloides mushroom) are bicyclic peptides that differ by two amino acid residues and can be interchangeably used as probes for F-actin. They stain F-actin at nanomolar concentrations and have similar affinities for both large and small filaments, binding in a stoichiometric ratio of one phallotoxin molecule to one actin subunit.

The most utilized fluorescent phallotoxin probes are 7-nitrobenz-2-oxa-l,3–phallacidin (NBD–phallacidin) (Oda et al. 1992), BIODIPY–phallacidin (Semple et al. 1997) and FITC–phalloidin (LaRosa et al. 1994a).

Prior to incubation with fluorescent phallotoxin probes, platelets can be blocked with unlabeled phallotoxins to reduce nonspecific staining (Semple et al. 1997). Fixation is required.

FCM and Platelet Aggregation

Platelet aggregation can be measured by a whole blood FCM assay (Fox et al. 2004). This method is based on measuring the decrease in the number of single platelets as they form aggregates in stimulated and stirred whole blood. The platelet number can be measured in small subsamples removed from the test tube at different time points to provide kinetic information on platelet aggregation. Platelets are labeled with platelet-specific antibody and the number of red cells is used as a reference for counting individual platelets.

The method is sensitive to microaggregate formation and can provide information on platelet disaggregation, although the approach is quite elaborate. A modification of this protocol has been described, in which platelets are labeled with two platelet-specific markers (one type of antibody conjugated with two different fluorocromes), mixed together, and stimulated with various platelet agonists (phorbol myristate acetate, collagene, ristocetin). The increase in events characterized by both fluorescent labels is representative of aggregate formation.

To date, the performance of this method has been tested in both experimental (mouse models) and clinical (GT patients, full-term neonates at 24-h of life) settings in which limited blood sampling is beneficial. Indeed, FCM proved to be a very promising tool for testing platelet activation and aggregation using a small amount of whole blood (De Cuyper et al. 2013; Baker-Groberg et al. 2016).

RNA Interference

It is well established that platelets contain megakaryocyte-transcribed mRNAs that are translated into proteins in response to physiological stimuli regulating platelet function (Weyrich et al. 2004). Manipulation of platelet RNA expression is a fascinating scenario for both research and clinical purposes, such as for characterization of the specific role of platelet proteins and microRNAs, and for manipulation of platelet function and lifespan to overcome platelet storage lesion in transfusion medicine.

It was previously demonstrated that FCM provides reliable assessment of the efficiency of short interfering RNA (siRNA) in cell populations (Ho et al. 2006). More recently, Hong and coworkers demonstrated, for the first time, that human platelets could be transfected with siRNA and that transfection efficiency could be assessed by FCM. The authors transfected fluorescently labeled siRNAs in human PRP and washed platelets, testing multiple transfection conditions and identifying the optimal method for measuring the fluorescence of transfected platelets and negative controls by FCM. The authors also demonstrated knockdown of the mRNA target in transfected platelets isolated by FCM cell sorting (Hong et al. 2011).

Platelet Count During In Vitro Megakaryocytopoiesis

FCM can be utilized for monitoring platelet production during in vitro megakaryocytopoiesis from purified hematopoietic stem cells. The method allows discrimination of in vitro produced platelets from cellular debris by combining calcein-AM staining (a hydrophilic molecule that confers a green fluorescence to living cells) with anti-CD41 monoclonal antibody. The simultaneous use of calibrated beads provides the absolute platelet count (Fig. 3). This technique is useful for monitoring end-stage in vitro megakaryocytic differentiation and for testing the effect of drugs on in vitro platelet production (Gobbi et al. 2007, 2009, 2013; Nurden et al. 2010; Carubbi et al. 2014a).

Fig. 3
figure 3

Platelet quantification by flow cytometric analysis. (A) Distribution of platelets (Y) and calibrated beads (X) according to their forward scatter (FS) and side scatter (SC, on a logarithmic scale). (B) Calibrated beads appear as a well-defined and clearly distinguishable population (CD41−/calcein-AM−). (C) CD41 and calcein-AM staining allows discrimination between platelets (CD41+/calcein-AM+) and debris (CD41+/calcein-AM−). (D) Ungated plot permits calculation of platelet number according to the number of events in the lower-left quadrant and in the upper-right quadrant

Full size image

Animal Models

Animal models are important tools in laboratory studies. In our setting, platelet function tests in different animal species have a wide range of applications. In particular, ex vivo pharmacological and toxicological studies have been undertaken in experimental animal models, including works of primary veterinary interest and studies in animal models of human platelet-derived disease. Flow cytometry can be applied to platelet studies in animal models, with similar considerations as those described for human platelets.

Given the fact that human and murine hemostasis are overlapping processes, mouse models are by far the most utilized. Indeed, engineered mice mimicking human acquired and inherited bleeding disorders have been developed, helping to clarify the role of specific transcription factors, receptors, and intracellular proteins or the effects of drugs (Aktas et al. 2005; McKenzie and Reilly 2004; Magallon et al. 2011; Graham et al. 2009; Pitchford et al. 2005; Pozgajova et al. 2006; Strassel et al. 2007; Kato et al. 2004; Kassassir et al. 2013; Chow et al. 2010; Musaji et al. 2004).

During FCM platelet analysis in mouse models, we have first to consider that murine platelets are smaller and more concentrated than human platelets (Nieswandt et al. 2004). An FCM method for determining the number and activation state of circulating platelets from a single mouse over an extended period of time has been described, using only 5 μL of blood collected by tail cut. Platelets are identified using a specific fluorescent antibody and a known number of fluorescent beads for counting standardization. The authors demonstrated that tail vein bleeding does not activate platelets and that the method is rapid, accurate, and reproducible (Alugupalli et al. 2001).

In addition, Shipper and colleagues have validated a single-platform protocol for counting human platelets after transfusion and cord blood transplantation in the peripheral blood of NOD/SCID mice, using an anti-human CD41 antibody against human platelets and counting beads (Schipper et al. 2007).

As described for human platelets, FCM analysis of surface exposure of P-selectin and activated αIIbβ3 (recognized by the monoclonal JON/A antibody, equivalent to PAC-1 for murine platelets) (Bergmeier et al. 2002), as well as detection of PLAs and PMPs, are routinely utilized to assess platelet activation in vivo and ex vivo in murine models of different diseases (Pitchford et al. 2005; Ohno et al. 2014; Lamrani et al. 2014; Chen et al. 2003; Henry et al. 2009; Yokoyama et al. 2005).

It has been pointed out that the pig is a very good model for study of atherosclerosis and thrombosis (Vilahur et al. 2011). This model is being increasingly used in cardiovascular and platelet research, and a reliable method for detecting the activation of porcine platelets has been described. The authors identified a set of commercially available antibodies that bind activated platelets, also setting the optimal platelet source (whole blood or PRP) for these purposes (Krajewski et al. 2012).

FCM has also been utilized in veterinary studies investigating platelet function in various species. As an example, dogs harboring inflammatory diseases show an increase in platelet activation markers, such as PLAs and CD62p, and their expression correlates with clinical outcome of the disease (Moritz et al. 2003, 2005; Goddard et al. 2015; Majoy et al. 2015).

More recently, the analysis of platelet surface receptors has been suggested as a useful tool in equine clinical medicine for investigation of new therapeutic strategies for the prevention or treatment of equine recurrent airway obstruction. FCM analysis has demonstrated that horses with this disease show an increase in CD41/61 and a decrease in CD62p platelet expression (Iwaszko-Simonik et al. 2015).

Take Home Messages

  • Platelet morphology and function can be comprehensively evaluated by flow cytometry

  • Flow cytometry has become routinely used not only for research purposes but also in the clinic

  • Relatively high costs and interlaboratory standardization are the main drawbacks that limit the application of flow cytometry