Nội dung

Mitochondria regulate platelet metamorphosis induced by opsonized zymosan A – activation and long-term commitment to cell death Paola Matarrese1, Elisabetta Straface1, Giuseppe Palumbo2, Maurizio Anselmi2, Lucrezia Gambardella1, Barbara Ascione1, Domenico Del Principe2 and Walter Malorni1 1 Department of Therapeutic Research and Medicines Evaluation, Istituto Superiore di Sanita’, Rome, Italy 2 Department of Pediatrics, University of Rome Tor Vergata, Italy Keywords aggregation; apoptosis; mitochondrial membrane potential; platelets; zymosan Correspondence P. Matarrese, Department of Therapeutic Research and Medicines Evaluation, Section of Cell Aging and Degeneration, Istituto Superiore di Sanita’, viale Regina Elena 299, 00161 Rome, Italy Fax: +39 6 49903691 Tel: +39 6 49902010 E-mail: paola.matarrese@iss.it (Received 1 November 2008, revised 25 November 2008, accepted 3 December 2008) doi:10.1111/j.1742-4658.2008.06829.x Changes in the mitochondrial membrane potential play a key role in determining cell fate. Mitochondria membrane hyperpolarization has been found to occur after cell activation, e.g. in lymphocytes, whereas depolarization is associated with apoptosis. The aim of this study was to investigate the effects of an immunological stimulus, i.e. opsonized zymosan A, on human platelet mitochondria by means of flow and static cytometry analyses as well as biochemical methods. We found that opsonized zymosan induced significant changes of platelet morphology at early time points (90 min). This was associated with increased production of reactive oxygen species, and, intriguingly, mitochondrial membrane hyperpolarization. At a later time point (24 h), opsonized zymosan was found to induce increased expression of CD47 adhesion molecule, platelet aggregation, mitochondrial membrane depolarization and phosphatidylserine externalization. Although these late events usually represent signs of apoptosis in nucleated cells, in opsonized zymosan-treated platelets they were not associated with membrane integrity loss, changes in Bcl-2 family protein expression or caspase activation. In addition, pre-treatment with low doses of the ‘mitochondriotropic’ protonophore carbonyl cyanide p-(trifluoro-methoxy)phenylhydrazone counteracted mitochondrial membrane potential alterations, production of reactive oxygen species and phosphatidylserine externalization induced by opsonized zymosan. Our data suggest that mitochondrial hyperpolarization represents a key event in platelet activation and remodeling under opsonized zymosan immunological stimulation, and opsonized zymosan immunological stimulation may represent a useful tool for understanding of the pathogenetic role of platelet alterations associated with vascular complications occurring in metabolic and autoimmune diseases. Several mechanisms are brought into play in order to control the balance between platelet production and destruction. Among these, recent studies have identified a form of apoptosis. Platelets have been shown to be able to undergo apoptosis in response to various stimuli [1,2]. It has been reported that platelet differentiation recapitulates morpho-functional events that are typical of apoptosis, such as trans-bilayer migration of phosphatidylserine (PS) to the outer membrane leaflet [3]. Platelets also express Abbreviations AM, acetoxymethyl ester; DHR123, dihydrorhodamine 123; DIC, differential interference contrast; FCCP, carbonyl cyanide p-(trifluoro-methoxy) phenylhydrazone; IVM, intensified video microscopy; JC-1, 5-5¢,6-6¢-tetrachloro-1,1¢,3,3¢-tetraethylbenzimidazol-carbocyanine iodide; MMP, mitochondrial membrane potential; OPZ, opsonized zymosan; PRP, platelet-rich plasma; PS, phosphatidylserine; ROS, reactive oxygen species. FEBS Journal 276 (2009) 845–856 ª 2009 The Authors Journal compilation ª 2009 FEBS 845 Mitochondria in platelet activation P. Matarrese et al. many components of nucleated-cell apoptosis such as caspases [1,4]. Mitochondria are generally considered to be key players in cell life and death. In addition to energy supply, they have also been demonstrated to be involved in execution of apoptosis via the release of apoptogenic factors such as cytochrome c [5]. Human blood platelet mitochondria play a critical role as they work efficiently as energy factories for both resting and stimulated cells. Mitochondria are also involved in non-ATP-related functions such as oxygen radical generation and apoptotic-like events. In fact, mitochondrial damage has been reported to be a key step in platelet apoptosis, and mitochondrial membrane potential (MMP) is lost during storage and under other conditions [4,6,7]. Changes in the mitochondrial permeability transition pore have been hypothesized to play a role in production of so-called coated platelets, a sub-population of platelets observed upon dualagonist activation, e.g. thrombin plus collagen, that express surface PS and can be induced by activation of the pro-apoptotic Bcl-2 family member Bax [3]. Hence, mitochondria integrity and function appear to act as general regulators of platelet fate in terms of both activation and apoptosis, two processes that, in platelets, appear to share some common features, e.g. PS externalization. Zymosan A, a complex polysaccharide obtained from Saccharomyces cerevisiae, is a complement activator that can be used as a tool to investigate the role of activated platelets in several diseases, including immune complex-mediated inflammation and its vascular complications [8]. Once opsonized, zymosan A has been suggested to activate platelets in a complementand fibrinogen-dependent way. In particular, complement components, such as C5, C6 and C7, are necessary, and IgG binding is also required for zymosan opsonization [9,10]. Hence, opsonized zymosan (OPZ) could represent a suitable model for the study of platelets as ‘inflammatory’ cells. In view of this, we decided to investigate whether this immunological stimulus may induce mitochondria dysfunction and influence platelet fate. We found that mitochondria modifications have a dual role, controlling both platelet activation and death. Results Characterization of morphological modifications induced by OPZ Treatment with OPZ induced two main modifications of platelet morphology: (a) cell remodeling typical of activation at early time points (after 30 and, more markedly, 90 min), and (b) platelet aggregation after 24 h. With regard to cell remodeling, we decided to analyze the actin cytoskeleton organization. This appeared to be significantly modified by OPZ (Fig. 1A): redistribution of actin filaments, forming long actin-positive protrusions (arrows), was detected in platelets exposed for 90 min to zymosan (central panel), and, more especially, to OPZ (right panel). Scanning electron microscopy analysis indicated that exposure to OPZ for 30 (not shown) and 90 min (Fig. 1B) appeared to activate platelets: the typical round morphology of platelets was altered to an activated morphology characterized by emission of thin protrusions. A series of analyses were also performed using CD47, the thrombospondin receptor, a surface molecule involved in cell adhesion [11]. Flow cytometry analyses revealed no changes at early time points, but platelet alterations were observed 24 h after OPZ administration. A platelet sub-population overexpressing CD47 was detected in zymosan-treated, and especially OPZ-treated, platelets (a representative flow cytometry analysis is shown in Fig. 1C). This increased expression was also detected by immunofluorescence analysis, and platelet aggregates were detectable after both zymosan and OPZ treatments (Fig. 1D). The number of aggregates was Fig. 1. Characterization of platelet modifications induced by OPZ. (A, B) Morphological alterations. (A) IVM analysis of actin microfilaments. Actin-positive protrusions (arrows) are visible in treated platelets (90 min). Insets in the middle and right panels show bright-field micrographs that have been electronically inverted to highlight these thin protrusions (arrows). Magnification ·1500. (B) Scanning electron microscopy. Exposure to zymosan A (opsonized and non-opsonized) for 90 min changed round-shaped resting platelets (control, left panel) into activated platelets characterized by the emission of long thin protrusions. Magnification ·4000. (C–F) CD47 expression and cell-aggregation analyses. (C) Histograms representing flow cytometry evaluation of surface expression of the adhesion molecule CD47 are shown in the upper panels. Numbers represent the percentage of highly positive platelets. In the lower panels, dot plots of the physical parameters of the platelet population (one representative experiment) are shown. (D) IVM analysis showing the intracellular distribution of CD47 molecule in zymosan-treated cells (central panel) and OPZ-treated cells (right panel). (E) DIC (Nomarski) micrographs showing cell aggregates in zymosan-treated platelets (central panel) and OPZ-treated platelets (right panel) in comparison with untreated platelets (left panel). In (D) and (E), arrows indicate platelet aggregates. (F) Quantification of cell aggregation by morphometric analysis performed using DIC. Values are means and SD of the results obtained in three independent experiments. *P < 0.01 versus control platelets; P < 0.01 versus zymosan-treated platelets. 846 FEBS Journal 276 (2009) 845–856 ª 2009 The Authors Journal compilation ª 2009 FEBS P. Matarrese et al. Mitochondria in platelet activation A Untreated Zymosan 90 min Opsonized zymosan 90 min Zymosan 24 h Opsonized zymosan 24 h 5 µm B Count 200 200 Untreated 200 5 µm 2.3 100 101 102 103 104 24.7 0 0 0 14.2 100 101 102 103 104 100 101 102 103 104 SSC-Height 1000 SSC-Height 1000 0 FSC-Height 1000 0 0 0 C SSC-Height 1000 Fluorescence 0 FSC-Height 1000 0 FSC-Height 1000 D 10 µm E 5 µm F *°° Aggregates (number/field) 20 15 * 10 5 Untreated Zymosan Opsonized zymosan FEBS Journal 276 (2009) 845–856 ª 2009 The Authors Journal compilation ª 2009 FEBS 847 Mitochondria in platelet activation P. Matarrese et al. evaluated. The results of morphometric analysis by fluorescence microscopy (CD47-stained samples) and differential interference contrast (DIC, i.e. Nomarski microscopy, Fig. 1E,F) clearly indicated that the number of aggregates, which was negligible in control samples, was significantly increased in OPZ-treated samples in comparison with non-opsonized zymosan-treated cells (Fig. 1F). OPZ-induced MMP alterations On the basis of previous studies, which reported increased MMP (hyperpolarization) in conjunction with cell activation, e.g. in lymphocytes [12,13], and MMP loss (depolarization) in conjunction with apoptosis [5], we analyzed this parameter in platelets treated with opsonized and non-opsonized zymosan A at various time points. Quantitative flow cytometry analysis, performed using a JC-1 probe (a representative experiment is shown in Fig. 2A), clearly indicated the presence of a significantly higher percentage of cells with hyperpolarized mitochondria (see boxed areas) after treatment with OPZ (third row) in comparison with either untreated platelets (first row) or platelets treated with non-opsonized zymosan (second row). Importantly, this hyperpolarization of mitochondrial membrane started early after OPZ addition (second column) and peaked after 90 min (third column). Interestingly, 24 h after OPZ addition (third row, fourth column), flow cytometry analysis clearly revealed a significant percentage of cells (approximately 35%) displaying mitochondrial membrane depolarization. These effects were also evident by pooling together data obtained from four independent experiments: Fig. 2B,C shows the percentage of cells with hyperpolarized or depolarized mitochondria, respectively. Altogether, these results indicate that mitochondria of platelets treated with OPZ underwent a marked increase in MMP at early time points (until 90 min), followed by a significant MMP loss at later time points (starting from 24 h). Importantly, in our experimental system, the decrease in MMP was not paralleled by an alteration of Bax (Fig. 2D) or Bak (Fig. 2E) expression levels. OPZ-induced ROS production Mitochondrial hyperpolarization has been related to hyperproduction of reactive oxygen species (ROS) [13,14]. A quantitative time-course analysis of ROS generation during zymosan A treatment was thus performed using flow cytometry. In accordance with the MMP data, increased ROS production was detected in OPZ-treated platelets using dihydrorhodamine 123 (DHR123). A representative experiment is shown in Fig. 3A [compare control platelets (left) and non-opsonized zymosan-treated platelets (middle panel) with OPZ-treated platelets (right)]. The results obtained from four independent experiments are reported in Fig. 3B. In OPZ-treated cells, increased ROS production was detectable at earlier time points (30 and 90 min), but the values detected after 24 h were similar to those found in control samples. OPZ induces PS externalization (but not caspase activation) We analyzed PS externalization in platelets under various experimental conditions. Flow cytometry evaluation of cell-surface expression of PS was performed using annexin V ⁄ trypan blue double staining. Figure 4A shows the results of a representative experiment, and Fig. 4B shows mean values obtained from four independent experiments. These analyses clearly indicated that, in the absence of any stimulus, platelets displayed very low levels of PS at their surface up to 24 h after isolation (first row, bottom right quadrant), and non-opsonized zymosan treatment induced a small increase of the percentage of annexin V-positive cells (second row), whereas a time-dependent increase in PS externalization was observed in OPZ-treated cells (third row, bottom right quadrant). Interestingly, neither non-opsonized nor OPZ-treated platelets were positive for trypan blue dye (see percentages in the upper right quadrant), indicating that the plasma membrane of most cells was undamaged at least up to 24 h (Fig. 4A, third row, fourth column) and 48 h (not shown) after OPZ administration. These results were clearer when data obtained from four independent experiments were Fig. 2. OPZ induces MMP alterations. (A) Biparametric flow cytometry analysis of MMP after staining with JC-1 in untreated platelets (first row), platelets treated with zymosan A (second row) and platelets treated with OPZ (third row) at various time points. The numbers in the boxed areas represent the percentages of cells with hyperpolarized mitochondria. The percentages of cells with depolarized mitochondria are shown below the dashed line. The results obtained in a representative experiment are shown. (B, C) Mean percentage (and SD) of platelets with hyperpolarized (B) or depolarized (C) mitochondria obtained from four cytofluorimetric experiments. Statistical analyses indicate a significant (P < 0.01) increase in cells with hyperpolarized or depolarized mitochondria at early (up to 90 min) and late (24 h) time points, respectively, only in platelets challenged with OPZ. (D, E) Bax (D) and Bak (E) expression levels as evaluated by FACS analysis. The y axis shows the median values of fluorescence as the mean and SD from four independent experiments. 848 FEBS Journal 276 (2009) 845–856 ª 2009 The Authors Journal compilation ª 2009 FEBS P. Matarrese et al. Mitochondria in platelet activation pooled (Fig. 4B). Data on cell viability obtained by the trypan blue exclusion test were confirmed using calceinacetoxymethyl ester (AM) (Fig. 4C). 10.4 9.7 T 24 h 104 T 90 min 104 T 30 min 104 T0 104 A On the basis of these results, analysis of the activation state of the main executioners of apoptosis, i.e. caspases, was required. We found that neither zymosan 11.1 10.6 7.1 100 104 11.5 100 104 104 14.9 12.2 100 104 100 7.1 100 104 104 104 104 100 5.6 100 100 Untreated 16.7 25.1 100 104 12.2 100 100 11.2 100 104 100 64.7 56.1 29.4 15.2 104 104 104 104 104 104 100 34.7 Opsonized zymosan 104 100 100 11.8 104 100 104 30.4 100 100 9.6 100 7.3 100 J-aggregates 100 6.9 100 Zymosan 100 104 J-monomers Untreated C 75 60 45 30 15 0 T0 T 30 min T 90 min Opsonized zymosan Zymosan T 24 h Cells with depolarized mitochondria (%) Cells with hyperpolarized mitochondria (%) B 40 30 20 10 0 Bak 5 4 3 2 1 T 30 min T 90 min T 24 h Bax E Protein expression level (a.u.) Protein expression level (a.u.) D T0 0 5 4 3 2 1 0 T0 T 30 min T 90 min T 24 h FEBS Journal 276 (2009) 845–856 ª 2009 The Authors Journal compilation ª 2009 FEBS T0 T 30 min T 90 min T 24 h 849 Mitochondria in platelet activation P. Matarrese et al. T0 T 30 min T 90 min T 24 h A 200 Opsonized zymosan 200 Zymosan 200 Untreated M = 32.2 M = 44.2 M = 58.6 M = 22.5 0 100 101 102 103 104 0 M = 17.3 M = 21.4 M = 19.1 M = 22.7 0 Events M = 18.4 M = 20.3 M = 17.1 M = 31.6 100 101 102 103 104 100 101 102 103 104 Green fluorescence intensity B 70 Untreated * Zymosan ROS production (a.u.) 60 Opsonized zymosan * 50 40 30 20 10 0 T0 T 30 min T 90 min T 24 h Fig. 3. OPZ induces production of reactive oxygen species. Quantitative cytofluorimetric analysis of ROS production was performed using DHR123. (A) Results obtained in a representative experiment. The values represent the median fluorescence. (B) Mean values (and SD) obtained from four independent experiments. *P < 0.01 versus control and zymosan-treated cells. (Fig. 5A) nor OPZ (Fig. 5B) induced activation of caspases 3 and 9 at any time point considered. In particular, 90 min after zymosan A administration, when we observed the maximum OPZ-induced MMP hyperpolarization (Fig. 2) and ROS production (Fig. 3), activation of caspase 9 (which depends on the release of mitochondrial apoptogenic factors) [5] and caspase 3 (the main enzyme involved in caspase-dependent apoptosis) was negligible. Even if the zymosan A exposure time was prolonged to 24 h, at which time PS externalization and mitochondrial membrane depolarization were observed (see Fig. 4), no significant activation of these caspases was detected. As a positive control, platelets treated with 1 UÆmL)1 of thrombin 850 for 1 h (in the presence or absence of the specific caspase inhibitors) were studied (Fig. 5C). Data obtained in positive controls or in platelets treated for 24 h with opsonized and non-opsonized zymosan A were confirmed by western blot analysis (Fig. 5D). MMP plays a key role in OPZ-mediated effects We examined the effects of an agent that is capable of specifically influencing MMP homeostasis and cell fate [12,14]: the protonophore uncoupler carbonyl cyanide fluorophenyl-hydrazone (FCCP), which is known to hinder the mitochondria hyperpolarization phenomenon at low doses [15]. In particular, the ability of FEBS Journal 276 (2009) 845–856 ª 2009 The Authors Journal compilation ª 2009 FEBS P. Matarrese et al. T 90 min 0.5 2.1 T 24 h 104 0.4 1.9 T 30 min 104 104 T0 104 A Mitochondria in platelet activation 1.1 4.1 5.3 1.4 104 100 100 104 1.3 6.5 0.8 6.1 100 104 104 0.6 5.9 100 100 104 104 104 104 100 100 100 Untreated 2.1 6.7 104 100 104 104 1.3 17.3 100 104 4.4 34.9 Opsonized zymosan 100 0.6 14.3 100 100 100 104 104 100 0.5 7.4 104 104 100 100 100 100 104 104 100 100 Trypan blue 100 Zymosan 100 104 100 104 Annexin V Annexin V positive Trypan blue positive B Untreated 40 Zymosan 40 30 30 30 20 20 20 10 10 10 T 30 T0 T 90 T 24 h 0 200 Events Untreated 24 h T 24 h Zymosan 24 h 0 T0 T 30 T 90 T 24 h Minutes Opsonized zymosan 24 h 5.3 3.4 0 0 2.7 100 101 102 103 104 T 90 Minutes Exposure time 200 Minutes C T 30 T0 200 0 Opsonized zymosan 0 % of cells 40 100 101 102 103 104 100 101 102 103 104 Calcein-AM (green fluorescence) Fig. 4. OPZ induces PS externalization. (A) FACS analysis after double staining with annexin V ⁄ trypan blue. Dot plots from a representative FACS experiment are shown. Numbers represent the percentages of annexin V-positive cells (bottom right quadrant) or annexin V ⁄ trypan blue double positive cells (upper quadrant). Note the very low percentage of cells that are positive for trypan blue. (B) Results obtained from four independent experiments, reported as means and SD. (C) FACS analysis after staining with calcein-AM (which is retained in the cytoplasm of live cells) of platelets treated with zymosan (central panel) or OPZ (right panel) for 24 h. Untreated platelets incubated for 24 h at 37 C (left panel) were used as the control. Numbers represent the percentage of calcein-negative cells. One representative experiment is shown. FCCP to counteract the OPZ-induced effects in terms of ROS production, MMP alterations and PS externalization was studied (Fig. 6). In our experiments, a very low dose (20 nm) of FCCP was able to hinder both the early and late events induced by OPZ. In platelets pre-treated with FCCP, the increase in ROS FEBS Journal 276 (2009) 845–856 ª 2009 The Authors Journal compilation ª 2009 FEBS 851 Mitochondria in platelet activation P. Matarrese et al. Caspase 9 200 200 200 0 0 104 100 104 Zymosan Events 200 0 104 100 Caspase 3 104 104 Opsonized zymosan 200 0 0 0 100 T 24 h 104 100 200 100 200 Caspase 9 T 90 min 0 T0 104 0 200 104 100 104 100 Green fluorescence 200 100 T 24 h 0 0 Events 200 0 104 100 Caspase 3 0 100 B T 90 min 200 T0 200 A 100 104 100 104 Green fluorescence 200 200 C Caspase 9 activation control D LEHD-fmk 74.6 9.4 Caspase 3 Positive control 1 2 0 0 104 100 104 200 Caspase 3 activation control 200 Events 32 100 DEVD-fmk 3 13.4 4 104 100 104 20 production induced by OPZ administration for 90 min was significantly reduced (Fig. 6A, compare shaded gray histograms with black histograms). Fittingly, the mitochondrial membrane hyperpolarization state observed 90 min after treatment with OPZ was significantly inhibited by low doses of FCCP (compare boxed areas in Fig. 6B, first and third panels, with Fig. 2B). The same protective effect of FCCP was observed with respect to mitochondrial membrane depolarization induced by 24 h treatment of OPZ (compare areas under the dashed line in Fig. 6B, second and fourth panels, with Fig. 2C). Similarly, the PS 852 5 32 0 0 69.3 100 20 Fig. 5. OPZ does not induce caspase activation. Analysis of the activation state of caspases 3 and 9 in intact living platelets treated with zymosan (A) or opsonized zymosan (B) at various time points. (C) Activation state of caspases 3 and 9 in a positive control represented by platelets treated for 1 h with thrombin (1 UÆmL)1) in medium containing 1 mM Ca2+. Values are the percentage of cells containing these caspases in their active form. Results obtained in a representative experiment are reported. (D) Western blot analysis of caspase 3 in thrombin-treated platelet (lane 2; compare with control in lane 1); untreated platelets (lane 3); platelets treated for 24 h with zymosan (lane 4) or with OPZ (lane 5). externalization observed in platelets treated for 24 h with OPZ (see Fig. 4) was negligible when platelets were pre-exposed to FCCP (Fig. 6C). Discussion Serum opsonized zymosan (from yeast cell walls) is known as a model phagocytic stimulus that interacts with both immunoglobulin and complement receptors, is ingested, and activates oxidative mechanisms. Because OPZ engages at least two types of receptor, the signaling pathways triggered by this stimulus are FEBS Journal 276 (2009) 845–856 ª 2009 The Authors Journal compilation ª 2009 FEBS Opsonized zymosan M = 63 FCCP + Zymosan M = 21 FCCP + Opsonized zymosan M = 32 T 90 min 0 0 Events T 90 min 100 101 102 103 104 100 101 102 103 104 Green fluorescence 15.2 100 104 9.7 15.5 100 104 19.3 100 16.1 100 100 104 T 24 h 104 T 90 min 12.3 10.1 100 104 T 24 h 104 104 T 90 min 9.3 100 J-aggregates FCCP + Opsonized zymosan FCCP + Zymosan B 100 104 100 complex. The stimulus may be considered as an immunological stimulus, which is also able to affect human platelets by inducing an oxidative burst [10]. Here we show that OPZ can trigger platelet metamorphosis [10], consisting of morphological and biochemical changes, that is typical of activation. In fact, platelets rapidly changed from a discoid form to an activated shape characterized by emission of long actin-positive protrusions. At the last time point (24 h), OPZ treatment was found to lead to CD47 overexpression and platelet aggregation. It has been suggested that platelet activation and adhesion are associated with morphological modifications, CD47 overexpression and platelet aggregation [10,11]. However, these changes were accompanied by an early and transient production of ROS, which probably serve, as in other cell systems, as signaling molecules of biological significance. Activation of platelets by OPZ was associated with a transient burst in ROS, which was possibly due, at least in part, to the energy metabolism [16]. Indeed, zymosan A, an activator of the alternative complement pathway, has been hypothesized to activate platelets in plasma in a complement- and fibrinogen-dependent way [9,10,17]. Hence, immunological stimulation by OPZ 104 1.6 3.1 4.0 5.4 104 100 100 1.3 104 100 Annexin V 104 2.1 100 104 104 1.5 100 100 104 J-monomers Trypan blue Fig. 6. The mitochondrial membrane potential plays a key role in zymosan-mediated effects. Quantitative flow cytometry evaluation of (A) ROS production, (B) MMP and (C) PS externalization in platelets pre-treated with a low dose of FCCP (20 nM) before addition of opsonized or non-opsonized zymosan A. Pre-treatment with FCCP significantly reduced OPZ-induced ROS production (A, compare shaded gray histograms with black empty histograms), mitochondrial membrane hyperpolarization (compare boxed areas in B, first and third panels, with Fig. 2B), mitochondrial membrane depolarization (compare areas under the dashed line in B, second and fourth panels, with Fig. 2C), and PS externalization (compare numbers in C with those in Fig. 4). The results shown were obtained in one experiment (representative of four). Values in (A) represent median fluorescence; those in (B) and (C) represent percentages of cells. Zymosan M = 15 Events A 200 Mitochondria in platelet activation 200 P. Matarrese et al. 100 7.4 104 could trigger a complex cascade of events leading to platelet remodeling and activation. The main result reported here is that OPZ-stimulated platelets underwent a time-dependent hyperpolarization of mitochondria, which started early (30 min) and lasted until 90 min. Such mitochondrial hyperpolarization was previously observed in T cells, and was considered as an activation-associated event [12]. Although the mechanism underlying this MMP increase remains unclear, it has been suggested that it could be associated with ROS signaling and could represent an early metabolic change ‘preparing’ the cell for the death process [18]. On the basis of our results, we can hypothesize that mitochondrial hyperpolarization could be associated, as in lymphocytes, with platelet activation. In accordance with results obtained in other cell types [14,15], the fact that the ‘stabilizing’ effect of the ‘mitochondriotropic’ drug FCCP prevented OPZ-induced mitochondrial membrane hyperpolarization as well as platelet morphological remodeling seems to suggest that the hyperpolarization state of mitochondria might represent an early transient key event sustaining platelets towards an activated phenotype, probably creating a pseudo-hypoxic redox state characterized by normoxic FEBS Journal 276 (2009) 845–856 ª 2009 The Authors Journal compilation ª 2009 FEBS 853 Mitochondria in platelet activation P. Matarrese et al. decreases of ROS and a shift from oxidative to glycolytic metabolism [19]. As in other systems, low doses of FCCP could inhibit ROS signaling events that lead to the programmed mitochondrial destruction termed mitoptosis [20]. In nucleated cells, mitochondria hyperpolarization occurs early after the apoptotic commitment, and is followed by MMP loss. It is widely accepted that the latter could contribute to apoptosis [5]. Our model system provides some further clues on this matter, and underlines the differences between nucleated and non-nucleated cells. We found that PS externalization, a typical early marker of apoptosis in nucleated cells, also occurs early in OPZ-stimulated platelets, together with ROS production and mitochondria hyperpolarization. Platelets, however, maintain their integrity for a long time (at least 48 h) despite MMP loss and increased PS externalization. As recently reported for other stimuli, including engagement of immunoreceptors [21], OPZ induced non-apoptotic externalization of PS. Furthermore, OPZ-treated platelets dis not show either caspase activation or an increase in Bcl-2 family proteins, nor cell death. On this basis, we can also hypothesize that, in some immunopathological instances, the increased number of platelets could be due to a defective death of these cells (although PS-positive) rather than de novo production of these cells. Conversely, other stimuli, such as collagen plus thrombin, have recently been demonstrated to induce PS externalization, a decrease in MMP, increased expression of the Bcl-2 proteins Bax and Bak, caspase activation and cell death [2,3]. In addition, the physiological platelet agonist thrombin also induces Bid, Bax and Bak translocation to the mitochondria and endogenous generation of hydrogen peroxide, which stimulates cytochrome c release and activation of caspases 3 and 9 [22]. Thus, OPZ appears to be a valuable activating agent triggering a long-term commitment to apoptosis (as also suggested by OPZ-induced PS externalization) rather than a typical apoptotic inducer. The previously hypothesized role of PS externalization and its role as an adhesion factor in cell–cell interaction therefore requires reappraisal [4,23]. For instance, platelet binding to dysfunctional endothelium was found to be inhibited by the phosphatidylserine-binding protein annexin V and enhanced by platelet agonists [24]. Give the importance of the loss of functional integrity of platelets in the pathogenesis of cardiovascular complications often associated with diabetes and some autoimmune diseases, e.g. antiphospholipid syndrome or Kawasaki disease [25,26], the results reported here indicate that OPZ could represent a prototypic 854 immunological stimulus for study of the pathogenic mechanisms of these diseases. Experimental procedures Platelet isolation and treatments Blood samples were collected from healthy volunteer blood donors who had taken no drugs for at least 10 days. The platelets were obtained by mixing fresh blood samples with a 1 ⁄ 6th volume of acid ⁄ citrate dextrose (38 mm citric acid, 75 mm Na3 citrate, 135 mm glucose) as anticoagulant. All the experiments were performed in platelet-rich plasma (PRP), which was prepared by centrifugation of blood samples at 150 g for 10 min at room temperature. Opsonized zymosan was obtained as previously reported [27]. Briefly, zymosan A (Sigma Chemical Co., St Louis, MO, USA) was boiled for 20 min and then washed for 5 min three times in NaCl ⁄ Pi. After washing, boiled zymosan A was added to platelet-poor plasma (obtained by centrifugation of PRP at 1000 g) and incubated at 37 C for 30 min. After washing three times, OPZ was ready to use. Control samples were prepared using non-opsonized zymosan A. Stimulation of the platelets was achieved by incubating PRP with 4 mgÆmL)1 OPZ at 37 C for various durations. After incubation with zymosan, PRP was centrifuged at 700 g for 5 min, and washed platelets were prepared for various analyses. For experiments with FCCP (Molecular Probes, Leiden, The Netherlands), PRP was incubated for 10 min with 20 nm FCCP before addition of zymosan A (both opsonized and non-opsonized). Samples treated with FCCP alone were also studied. Platelets were analyzed at various time points (5, 10, 20, 30 and 90 min and 6, 8 and 24 h) after treatment with non-opsonized zymosan A or OPZ. The main changes were detected after 30, 90 min and 24 h: only the results obtained at these time points are shown here. In addition, we also analyzed platelets treated with zymosan A (opsonized and non-opsonized) and immediately washed three times. These were considered as controls to test the responsiveness of platelets immediately after interaction with zymosan A, and this time point is indicated as T0. Scanning electron microscopy Samples were collected and plated on poly-l-lysine-coated slides, and fixed with 2.5% glutaraldehyde in 0.1 m cacodylate buffer (pH 7.4) at room temperature for 20 min. After post-fixation in 1% OsO4 for 30 min, samples were dehydrated through a graded ethanol series, critical point-dried in CO2 and gold-coated by sputtering using a Balzers Union SCD 040 apparatus (Balzers, Weisbaden, Germany). The samples were examined using a Cambridge 360 scanning electron microscope (Leica Microsystem, Wetzlar, Germany). FEBS Journal 276 (2009) 845–856 ª 2009 The Authors Journal compilation ª 2009 FEBS