Detection of Platelet-Derived Microparticles in Blood Samples by Flow Cytometry: The Importance of Choosing the Adequate Control

Detection Platelet-Derived Microparticles in Blood Samples by Flow Cytometry: The Importance of Choosing the Adequate Control Circulating microparticles are cell-derived vesicles that are generally released within the circulation by stressed cells upon tissue injury. Changes in microparticles circulating levels are associated with a wide range of pathologies. Since microparticles retain markers of their cellular origin they are of major interest as novel biomarkers of tissue injury. Platelets-derived microparticles are the focus of many studies as they are associated with several vascular and blood coagulation disorders. Flow cytometry analysis combined with the use of fluorochrome- coupled antibodies directed against specific microparticle markers is the most commonly technique for detecting and quantifying microparticle levels within blood and other biological samples. To be valid, this technique requires verifying the specificity of each of the antibody used by including in the experiments control isotype antibodies. During our studies aiming at characterizing by flow cytometry analysis basal level of PMPs in rat plasma we have consistently observed that commercially obtained-anti CD42 and anti CD61 antibodies and their respective control isotypes lead to positive signal. We have then analyzed and quantified by western blotting experiments total immunoglobulin content of the commercially available either anti CD42 or anti CD61 antibody or their control isotype solutions. We report that, as compared to anti CD42 or anti CD61 antibody solutions, the IgG concentration of the solutions corresponding to their isotype controls is much higher. Furthermore, we demonstrate that if prior to flow cytometry analysis of PMP levels in rat plasma, the isotype control solutions are diluted in such a way that the discrepancy in IgG concentrations between anti CD42 or anti CD61 antibody solutions or their corresponding control isotype solutions are suppressed, we no longer have non-specific signal. Therefore, in the light of our present report, IgG concentrations between specific and isotype control antibody solutions should be normalized for a better analysis of microparticle plasma levels that avoids non-specific signal.


Introduction
Microparticles (MPs) are 0.1 µm and 1 µm diameter vesicles that may be released by injured tissues or stressed cells [1,2]. They are formed by parts of cells plasma membranes and may contain many cell-derived molecules such as receptors, cytoskeleton, signal proteins and effector proteins [3]. Microparticles are characterized by both phosphatidylserines (PS) exposure at their external surface and by the presence of markers reflecting their cellular origin [4,5].
MPs are different from other types of cell-derived vesicles such as exosomes and apoptotic bodies. Exosomes are much smaller (40 to 100 nm diameter) than MPs and do not express PS at their external surface. Unlike exosomes, apoptotic bodies express PS on the outer membrane but their size (1 to 5 µm diameter) is bigger than that of MPs [6]. Under some stress conditions, cytoplasmic calcium level may increase causing the activation of calcium-dependentscramblases and floppases as well as the inhibition of flippase, which leads to the translocation of PS from the inner to the outer leaflet of the plasma membrane [7,8]. Externalized PS contribute to the activation of the prothrombinase complex, which triggers in turn the activation of the blood coagulation cascade and platelet aggregation [9,10]. Externalized PS are also eat-me signals that play a major role in phagocytosis and consequently in tissue lesion and repair [11].
Increases in MP levels have been reported in a variety of prothrombotic and inflammatory states, cardiovascular or autoimmune disorders and infectious diseases. Thus, measuring MP levels in blood or in other biological fluid may provide useful information as a prognostic biomarker for a wide variety of diseases [2,12,13]. In healthy subjects, platelet derived microparticles (PMPs) are the major circulating MPs (70-90 %). Studies have suggested that PMPs promote the expression of adhesion molecules on a variety of cells, stimulate the release of cytokines, alter vascular functions, induce inflammation and angiogenesis, and are also involved in cancer metastasis [13][14][15]. A large variety of markers have been used to characterize PMPs such as CD41, CD41a, CD42a, CD42b, CD42d, CD61, CD62P [2,16].
Flow cytometry using fluorescent probes is the most widely used technique to characterize and quantify MPs in blood as well as other biological liquids. Annexin-V binds strongly and specifically to PS and thereby fluorescent-labelled annexin-V is used for the detection of PS-exposing bodies among which MPs. Therefore, based on the size (50-1000 nm diameter) and on the exposure of PS (annexin-V positive vesicles) flow cytometry analysis enables to detect MPs within a given biological sample. Furthermore, fluorescent-coupled antibodies recognizing a cellular marker may be used to identify the cellular origin of the MPs. However, to ascertain the specificity of the signal, it is necessary to conduct parallel experiments using control isotype for each antibody. Control isotype corresponds to an antibody labelled with the same fluorochrome and produced in the same species as the specific antibody but lacking reactivity with the targeted marker. In the present report, we initially undertook to detect basal levels of platelet-derived microparticles in rat plasma using commercially available anti CD42 and anti CD61 antibodies.
However, the control isotypes of either of these antibodies lead to non-specific signal. We then investigated the raisons for this nonspecific signal and we provide clues on how to avoid such nonspecific signal in MPs detection by flow cytometry experiments.

Animals
Wistar rats were used in our study for preparing platelet poor plasma (PPP). Animal care and manipulations for experimentation were conducted in accordance with the guidelines of the French Agriculture and Forestry Ministry (decree 87849) and of European Communities Council Directive (86/609/EEC) and were approved by the local ethics committee (COMETHEA:CE2012-06).

Preparation of Platelet Poor Plasma (PPP)
Rats were anaesthetized with 4 % isoflurane + oxygen 4 ml/ min for induction and 2 % isoflurane + oxygen 2 ml/min for maintenance. Arterial blood was collected intracardially using 5 ml syringes containing 100 µl of anticoagulant (sodium citrate 0.109 M). PPP is obtained by two successive centrifugations at 2500 g, 20°C for 15 min. Each time, the pellet is removed and the supernatant is collected and stored at -80°C.

Flow Cytometry Analysis
The detection of MPs was performed by flow cytometry using

Statistical Analysis
Western Blot quantification are given as means  SEM.
Comparison between anti CD61-PE, anti CD42d-APC and their control isotype was performed using t-test. Figure 1 represents a typical flow cytometry analysis of rat PPP.

Results
An unstained sample was acquired to set the photomultiplier for considered channels as represented in Figure 1A. In Figure 1B   In A, anti CD42d-APC or its isotype control antibody solutions and in B anti CD61-PE or its isotype control antibody solutions were analyzed at multiple dilutions by western blotting using anti total IgG antibody. In C, D quantification of the intensity of heavy chain band using ImageJ software is represented. **: p<0.01; ***: p<0,005. In all the experiments n was 3 or higher.

Discussion
The initial aim of our study was to set up and standardize a method for detecting and quantifying PMPs in rat PPP and possibly other biological fluids both under basal-and pathological conditions. However, we were confronted in the early steps of our study to a serious problem. Control isotype antibodies, which supposedly should define the background of the assay represented by non-specific antibody binding to MP, resulted in a strong signal comparable to that of the two-specific anti PMPs antibodies used: anti CD42d-APC or in anti CD61-PE. We aimed then at analyzing this experimental problem hoping to provide clues on how to ovoid it in future experiments. Since the same problem occurred with two different antibodies directed against distinct antigens (CD42d and CD61), we can rule out that this problem may not be linked to a given antibody source. In our experimental set-up, we initially used all the antibody solutions that were commercially obtained, at rigorously the same concentration that recommended by the supplier. All the antibodies (specific and isotype control) used in our study are IgG type. IgG heavy chains are much larger proteins than the IgG light chain which may be responsible for non-specific binding to other proteins. Therefore, we sought to estimate in a semi-quantitative manner using western blotting analysis, the IgG concentration of each of the antibody solutions used in our study. Such an analysis revealed that IgG concentrations are very heterogeneous between the 4 samples tested and are much higher in the antibody isotype control solutions than in the specific antibodies (CD42d and CD61) solutions. Furthermore and in confirmation that the higher concentration of IgG in the antibody isotype control solutions is responsible for the non-specific signal observed in flow cytometry analysis described in Figure 1, data depicted in Figure 3 provided evidence that normalizing the concentration of heavy chain IgG by diluting antibody isotype control solutions drastically reducedand ever suppressed the non-specific signal. Andersen et al. [1] have shown that non-specific labelling can be explained by the Fc part of the IgG heavy chain that binds to Fc receptors located on cell surface and therefore potentially on PMPs. These authors suggested to use antibody-based blocking solutions to block these Fc receptors. In the present report, we provide a novel clue to prevent non-specific signal, which is to normalize by dilution, all the antibody solutions used to have a similar heavy chain IgG concentration in all the solutions. In conclusion, the present report revealed a heterogeneity in term of IgG heavy chain concentration between anti CD42d, anti CD61 and their control isotype antibody solutions that distorts the analysis of MPs by flow cytometry. In the light of our finding, we recommend to use western blotting analysis to estimate IgG heavy chain concentration in each solution and to dilute solutions of control isotype antibodies in such a way that the IgG heavy chain concentration is similar in all solutions prior to MPs analysis by flow cytometry.