Irregular spiking in free calcium concentration in single, human platelets Regulation by modulation of the inositol trisphosphate receptors Roosje M. A. van Gorp 1 , Marion A. H. Feijge 1 , Wim M. J. Vuist 1 , Martin B. Rook 2 and Johan W. M. Heemskerk 1 1 Departments of Biochemistry and Human Biology, University of Maastricht, the Netherlands; 2 Department of Medical Physiology, University Medical Centre Utrecht, the Netherlands Fluorescence ratio imaging indicates that immobilized, aspirin-treated platelets, loaded with Fura-2, respond to inositol 1,4,5-trisphosphate- (InsP 3 )-generating agonists such as thrombin by high-frequency, irregular rises in cytosolic [Ca 2+ ] i with spikes that vary in peak level and peak-to-peak interval. This differs from the regular [ Ca 2+ ] i oscillations observed in other, larger cells. We found that the thiol-reactive compounds thimerosal (10 l M ) and U73122 (10 l M ) evoked similar irregular Ca 2+ responses in platelets, but in this case in the absence of InsP 3 generation. Throm- bin-induced spiking was acutely abolished by inhibiting phospholipase C or elevating intracellular cAMP levels, while spiking with sulfhydryl reagents was only partially blocked by cAMP elevation. Confocal laser scanning microscopy using fluo-3-loaded platelets indicated that, with all agonists or conditions, t he irregular spikes were almost instantaneously raised in various regions within a single platelet. When using saponin-permeabilized platelets, we found that InsP 3 -induced Ca 2+ release from stores was stimulated by modest Ca 2+ concentrations, pointing to a mechanism of InsP 3 -dependent Ca 2+ -induced Ca 2+ release (CICR). This process was completely inhibitable by heparin. The Ca 2+ release b y InsP 3 , but not the C ICR sensor, was negatively regulated by cAMP elevation. Thimerosal treat- ment did n ot release Ca 2+ from intracellular stores, but markedly potentiated the stimulatory effect of InsP 3 .In contrast, U73122 caused a heparin/cAMP-insensitive Ca 2+ leak from stores that differed from those used by InsP 3 . Taken together, these results demonstrate t hat Ins P 3 recep- tor channels play a crucial role in the irregular, spiking Ca 2+ signal of intact platelets, even when induced by agents such as thimerosal or U73122 which do not stimulate InsP 3 for- mation. The irregular Ca 2+ release events appear to be subjected to extensive regulation by: ( a) InsP 3 level, (b) the potentiating effect of elevated Ca 2+ on InsP 3 action via CICR, ( c) Ins P 3 channel sensitization by sulfhydryl (thim- erosal) modification, (d) InsP 3 channel-independent Ca 2+ leak with U73122, and (e) down-regulation via cAMP elevation. The observation that individual Ca 2+ peaks were generated in various parts of a platelet at similar i ntervals and amplitudes points t o effective c ooperation of the various stores in the Ca 2+ -release process. Keywords:Ca 2+ -induced Ca 2+ release; cyclic AMP ; cytosolic Ca 2+ ; inositol trisphosphate; platelets. Most vertebrate cells respond to specific agonists by repetitive spiking or oscillation in cytosolic [Ca 2+ ] i as a consequence of regenerative release of Ca 2+ from stores into the c ytosol through inositol 1,4,5-trisphosphate (InsP 3 ) or ryanodine receptor channels, located in the membrane of the endoplasmic o r sarcoplasmic reticulum, respectively [1]. For large cells such as oocytes and HeLa cells, evidence has been collected that local clusters of InsP 3 receptors in the reticular membrane function as discrete Ca 2+ release s ites. Such local spots, being spaced at intervals of tens of micrometers apart, are taken responsible for so-called elementary Ca 2+ release events [2–4]. At low concentra- tions, InsP 3 may trigger individual release sites, which results i n t he appearance of local Ca 2+ ÔpuffsÕ, i.e. of brief Ca 2+ release events of usually low amplitude. Higher I nsP 3 concentrations cause a summation in amplitude or fre- quency mode of these release events, and lead to recruitment of neighbouring r elease sites. As a consequence, global increases in [Ca 2+ ] i can develop that prop agate through the entire cell as Ca 2+ oscillations or waves. These whole-cell Ca 2+ responses are u sually regular in shape, such in contrast to the local Ca 2+ puffs which are heterogeneous in both amplitude and t ime of appearance. In a variety of cells, the InsP 3 receptor channels play crucial roles in eliciting [Ca 2+ ] i oscillations and puffs [1–4]. Three different InsP 3 receptor isoforms are presently recognized with subtle differences in the regulation of Ca 2+ channel opening. Characteristic for the type 1 InsP 3 receptors is a biphasic effect of cytosolic Ca 2+ on the channel activity, with Ca 2+ stimulating the Ca 2+ release Correspondence to J. W. M. Heemskerk, Departments of Biochemis- try/Human Biology, University of Maastricht, PO Box 616, 6200 mD Maastricht, the Netherlands. Fax: + 31 43 3884160, Tel.: + 31 43 3881671, E-mail: JWM.Heemskerk@bioch.unimaas.nl Abbreviations:CICR,Ca 2+ -induced Ca 2+ release; InsP 3 , inositol 1,4,5-trisphosphate; PGE 1 , prostaglandin E 1 . Note: Part of this paper appears in the PhD Thesis of R. M. A. Beisser-van Gorp (University of Maastricht, the Netherlands). (Received 21 September 2001, revised 21 December 2 001, accepted 22 January 2002) Eur. J. Biochem. 269, 1543–1552 (2002) Ó FEBS 2002 from stores up to 300 n M and inhibiting this activity at higher levels [5–9]. This biphasic effect may control the rising and falling phases of individual Ca 2+ spikes.Thus,at a relatively low [Ca 2+ ] i ,InsP 3 -mediated Ca 2+ release is facilitated by the sensitizing mechanism of Ca 2+ -induced Ca 2+ release ( CICR), whereas at higher Ca 2+ levels the InsP 3 receptors become desensitized. Other factors deter- mining the open probability of the receptor channels are the luminal Ca 2+ concentration in the endoplasmic reticulum [9,10], modulation or oxidation of the receptor sulfhydryl groups [11–14], and phosphorylation b y cAMP-dependen t protein kinase [ 15,16]. Platelets are among the smallest cellular entities in the mammalian body (diameter o f about 2 lm w ith e stimated volume of 6 fL). They acutely respond to Ins P 3 -forming agonists by regenerative Ca 2+ release [ 17–20]. T he [Ca 2+ ] i spiking pattern of platelets is remarkably irregular in shape in comparison to that of larger cells, e.g. of the smoothly oscillating megakaryocytes [21,22]. All thre e InsP 3 receptor isoforms have been identified in platelets, i.e. mostly type 1 and type 2 receptors in addition to some type 3 receptors [23–26]. In p latelet membrane preparations it is shown that the InsP 3 receptors are susceptible to sulfhydryl modifica- tion and c AMP-dependent phosphorylation [ 24,25]. T here is, however, little evidence that s uch modulation influences InsP 3 receptor f unctioning also in intact platelets [27,28]. In particular, it is controversial whether cAMP-dependen t protein kinase may stimulate InsP 3 -induced Ca 2+ release [29], cause modest inhibition [30,31], or is without effect [32] on the release process. In this report we consider the nature and subcellular organization of the regenerative Ca 2+ release in platelets triggered by I nsP 3 -mobilizing receptor agonists and non- InsP 3 -mobilizing sulfhydryl reagents. We investigated the importance of InsP 3 receptor-dependent CICR in the irregular Ca 2+ signal generation by these agents, and the sensitivity of this signal toward cAMP elevation. We found that th e i rregular s piking Ca 2+ signal of platelets contains several but not all characteristics of local, InsP 3 receptor- dependent Ca 2+ puffs described f or other, larger cells. EXPERIMENTAL PROCEDURES Materials H-Arg-Gly-Asp-Ser-OH (RGDS) was purchased from Bachem (Bubendorf, Switzerland), a nd ultra-pure calcium- free water from Baker (Phillipsburg, NJ, USA). Fura-2, Fluo-3 and Indo-1 acetoxymethyl esters as well as noneste- rified Fluo-3 were bought from Molecular P robes (Leiden, the Netherlands). Manoalide, U73122, U73343 and InsP 3 came from Biomol (Plymouth Meeting, PA, USA), and thimerosal (sodium ethylmercuri-thiosalicylate) was from Janssen (Beerse, Belgium). Other chemicals were obtained from Sigma (St Louis, MO, USA) or Merck (Darmstadt, Germany). Platelet preparation and loading with Ca 2+ probes Blood was collected from healthy volunteers, w ho had n ot taken medication for at least two weeks. Platelet-ric h plasma was p repared by centrifugation [18]. It was incubated with acetoxymethyl ester of Fura-2 (3 l M ) or Fluo-3 (7 l M )in thepresenceoflysineacetylsalicylate(aspirin,100l M )at 37 °C for 45 min. After this loading procedure, the platelets were spun down, washed twice i n the presence of apyrase (0.1 U ADPaseÆmL )1 ), and resuspended in buffer A (pH 7 .45), which was composed of 136 m M NaCl, 10 m M glucose, 5 m M Hepes, 5 m M KCl, 2 m M MgCl 2 ,0.1%(v/v) bovine serum albumin and apyrase (0.2 U ADPaseÆmL )1 ). The suspension was adjusted to 1 · 10 8 plateletsÆmL )1 . Measurement of [Ca 2+ ] i in single, immobilized platelets Aspirin-treated, Fura-2-loaded platelets were immobilized on fibrinogen-coated glass coverslips, as described previ- ously [18]. Briefly, the platelets were allowe d to bind to the surface, and bathed in 0.5 mL buffer A supplemented with 10 l M RGDS, apyrase (0.2 U ADPaseÆmL )1 ) and CaCl 2 (2 m M )at23°C. Agonists and antagonists were given as freshly prepared solutions in bathing medium (0.1 m L). Changes i n Fura-2 fluorescence were recorded in in dividual cells using an inverted N ikon microscope (Tokyo, Japan), equipped with a dichroic mirror, computer-driven excitation and emission filter wheels, and an intensified charge-coupled device camera working at standard video rate (Photonic Sciences, Robertsbridge, UK). A 100-W Xenon lamp was used for illumination. The excitation wavelength was alternated between 340 and 380 nm, and fluorescent light was detected at 505 nm. The light was collected with a 40 · oil objective (Fluor Nikon, numerical aperture 1.3). Final image resolution was 1.0 pixels Ælm )1 , while confocality giving half-maximal intensity in the x–y plane was deter- mined at 2 .3 lm. QUANTICELL 700 software (Visitech, Sun- derland, UK) was used to control the filter wheels and capture the images [33]. Four-times averaged, backgroun d- subtracted fluorescence ratio images were obtained e very second. Calibration of 340/380 nm fluorescence ratio to [Ca 2+ ] i , using lysed platelets, w as as described elsewhere [20]. Fluorescence measurements with suspensions of Fura-2- or Fluo-3-loaded platelets were c arried out as described [20]. High-resolution, confocal images were collected with a Nikon RCM 8000 real-time c onfocal laser scanning system, equipped with an Argon laser. Light was collected with a 60 · oil objective (Apo Nikon, numerical aperture 1.4). Fluo-3-loaded platelets were visualized at a laser power of 87–91 lW, and excitation and emission wavelengths of 488 nm and 500–550 nm, respectively. Using a small pinhole, confocality in the x–y plane was experimentally determined at 0.2 lm ( matching the fi nal image re solution of 6.0 pixelsÆlm )1 ), while confocality along the z axis was 0.5 lm. Because of the limited fluorescence levels in the platelets, image frames were eightfold averaged to give a final temporal resolution of 10 Hz. The Fluo-3 fluorescence level was expressed as a pseudo-ratio value (F/F o )ofthe actual fluorescence intensity ( F) relative to the basal i ntensity of the platelet at r est (F o ), as described elsewhere [3,4]. Calibration was performed as described by Yao et al.[2]. The s ame confocal system was a lso used t o monitor Ind o- 1-labelled platelets, at settings described else where [34]. Measurement of [Ca 2+ ] in suspensions of saponin-permeabilized and intact platelets Aspirin-treated platelets were suspended at a concentration of 1–1.5 · 10 9 mL )1 in buffer B (pH 7.45), composed of 1544 R. M. A. van Gorp et al. (Eur. J. Biochem. 269) Ó FEBS 2002 136 m M NaCl, 20 m M glucose, 5 m M Hepes, 5 m M KCl, 2m M MgCl 2 and 0 .1 m M EGTA prepared in calcium-free water. The platelets were permeabilized with saponin at 23 °C, basically as described elsewhere [31]. Immediately before start of a measurement, a sample of 0.4 mL was added to 1.6 mL of Hepes/KCl buffer pH 7.4 (buffer C), composed of 100 m M KCl, 100 m M sucrose, 20 m M Hepes , 1.4 m M MgCl 2 and 1.25 m M NaN 3 (preparedincalcium- free water). The mixture, in a fluorescence cuvette, was supplemented with 7.5 m M phosphocreatine, 1 m M ATP, 1m M KH 2 PO 4 ,30lgÆmL )1 creatine kinase, 0.6 lgÆmL )1 oligomycin and 1 l M Fluo-3. Permeabilization of the platelets was achieved by addition of 15–20 lgÆmL )1 saponin. After 10 min of stirring, fluorescence was mea- sured and the free Ca 2+ level was titrated to 110 n M by stepwise additions from a 0.05-m M CaCl 2 solution. InsP 3 and other agents were given during the fluorescence recording. Part of the experiments were carrie d out with 0.75 m M phosphocreatine and 0.1 m M ATP. In that case, apyrase (2 UÆmL )1 ) was added after 6 m in of permeabili- zation to degrade A TP. Free Ca 2+ was then adjusted to the desired level, after which InsP 3 was added. Ultra-pure, calcium-free water was used for preparation of all buffers, supplements and agonists. Fluo-3 fluorescence intensities (F) were continuously recorded at 488 nm excitation and 526 nm emission wave- lengths (slits of 4 nm), using an SLM-Aminco DMX-1100 spectrofluorometer (Rochester, NY, USA). Calibrations were performed by adding excess amounts of CaCl 2 and EGTA/Tris (1 : 1, mol/mol) to obtain F max and F min values, respectively. Level of [Ca 2+ ] in the suspe nsion was calcu- lated from the binding equation [Ca 2+ ] ¼ K d · b (F–F min )/ (F max –F). Th e s ame fluorometer was also u sed to measure changes in [Ca 2+ ] i in intact platelets loaded with Fura-2 o r Fluo-3 [34]. Measurement of Ins P 3 InsP 3 levels were determined in samples of resting and activated platelets (180 lL, 3.5 · 10 8 cells). Cellular a ctivity was stopped by addition of 75 lL ice-cold 10% (w/v) HClO 4 . After standing on ice for 30 min and centrifuging at 2000 g for 10 min (strictly at 4 °C), supernatants were collected and neutralized to pH 7 with a solution of 1.7 M KOH a nd 75 m M Hepes. After 30 m in on ice, the precipi- tated KClO 4 was removed by another centrifugation step (4 °C). The supernatants were used to measure mass amounts of InsP 3 with a Biotrak radioreceptor assay system (Amersham-Pharmacia, UK). Freshly dissolved InsP 3 was taken as a standard. Statistics Paired data were compared for significance of difference using a Student t-test. Unpaired data were compared by ANOVA . RESULTS Irregular spiking in [Ca 2+ ] i in single platelets independently of Ins P 3 formation Fura-2-loaded p latelets immobilized on fibrinogen often exhibit ÔspontaneousÕ, spiking increases in [Ca 2+ ] i ,which can partially be prevented by treatment of the platelets with aspirin and apyrase (blocking the effects of released thromboxane A 2 and ADP, r espectively) [35]. U sing plate- lets treated with a spirin and apyrase, we compared the effects of v arious G q /phospholipase C-b stimulating recep- tor agonists on Ca 2+ signal generation. Extracellular CaCl 2 was p resent to allow physiological, store-regulated influx of Ca 2+ . Both platelet-activating factor (400 n M )andthe thromboxane A 2 analogue, U46619 (1 l M ), caused repetit- ive increases in [Ca 2+ ] i in single platelets for up to 3 min. In these traces, individual C a 2+ spikesvariedinpeaklevels and occurred after short but variable time intervals (Fig. 1 A,B). The strong agonist thrombin (4 n M )also elicited irregular, spiking rises in [Ca 2+ ] i , but the signal now persisted for more than 5 min (Fig. 1C). These responses differ markedly from the quite regular and symmetric oscillations in [Ca 2+ ] i , which have been reported for larger cells such as rat megakaryocytes [9,10]. To determine the involvement of cytosolic InsP 3 in the irregular spiking process in p latelets, we used t he phospholipase C-inhibiting agents manoalide [36,37] a nd U73122 [38,39]. Addition of manoalide (10 l M ) or a low dose of U73122 Fig. 1. Irregular spiking in [Ca 2+ ] i induced b y phospholipase C-activating agonists. Aspirin- treated, Fura-2-loaded p latelets on a fib rin- ogen surface were stimulated with 0 .4 l M platelet-activating factor ( PAF) (A), 1 l M U46619 (B) or 4 n M thrombin (Thr) (C–F) in thepresenceof1m M CaCl 2 and apyrase (0.1 UADPaseÆmL )1 ). Where i ndicated, 10 l M manoalide (D), 2 l M U73122 (E), or 10 l M PGE 1 (F) was added after stimulation. Fluorescence ratio images were collected from microscopic fields using a came ra-based sys- tem. Traces are Ca 2+ responses of single platelets, representative for 50–100 cells from at least four independent experiments. Ó FEBS 2002 Regulation of calcium spiking in platelets (Eur. J. Biochem. 269) 1545 (2 l M ) shortly after t hrombin completely cancelled t he generation of new [Ca 2+ ] i spikes (Fig. 1 D,E), whereas the U73122 control substance U73343 (2 l M ) was without effect (data not shown). T hrombin-induced [Ca 2+ ] i spiking was also annulled by addition of the cAMP-elevating agent, prostaglandin E 1 (PGE 1 , Fig. 1F). Thus, the irregular spiking process with thrombin apparently depends on continuous generation of InsP 3 and is down-regulated by elevation of the cAMP concentration (see also below). Note that similar, irregular Ca 2+ responses were also obtained when using platelets loaded with Fluo-3 instead of Fura- 2. Membrane-permeable sulfhydryl reagents provide an alternative way of evoking Ca 2+ responses, although occurring in the apparent absence of phospholipase C activation [40,41]. W e used thimerosal, a compound that sensitizes the platelet InsP 3 receptor channels [28,42], and a high dose of U73122 which acts as an N-ethylmaleimide derivative thus affecting o ther en zymes than only phospho - lipase C [38,39,43]. When aspirin-treated platelets on fibri- nogen were treated with thimerosal ( 10 l M ) or U73122 (10 l M ), this resulted in prolonged, irregular spiking in [Ca 2+ ] i after a lag time of one or more minutes (Fig. 2A,B). As U73122 inhibits phospholipase C activity already at 2 l M (see below), the spiking with U73122 is unlikely to result from phospholipase C activation and InsP 3 genera- tion. This conclusion was also drawn for thimerosal, as neither pretreatment with manoalide (Fig. 3A) nor postad- dition of manoalide (Fig. 3B) or a low dose of U73122 (not shown) influenced the spiking induced by thimerosal. In quantitative trms, after 5 min o f stimulation with thimero- sal, peak amplitudes were 667 ± 80 n M [Ca 2+ ] i in the absence of manoalide pretreatment and 586 ± 48 n M after manoalide pretreatment (mean ± SEM, n ¼ 22 cells, P ¼ 0.38). In contrast, p reincubation of the p latelets with 10 l M PGE 1 lowered the amplitude of the thimerosal- induced peaks to 390 ± 70 (n ¼ 24 platelets, P ¼ 0.009) (Fig. 3 C). PGE 1 , when added after thimerosal, gradually inhibited the appearance of new [Ca 2+ ] i spikes, although it did not restore [ Ca 2+ ] i to the basal level (compare Fig. 3A and D). When added after U73122, PGE 1 had a similar effect on the spiking process ( Fig. 2C). In experiments with aspirin-treated platelets in suspen- sion, we verified the effects of these platelet-activating agents on phospholipase C stimulation. Levels of InsP 3 levels were measured at time points where the Ca 2+ signal was still maximal. Thrombin, but not thimerosal, had a potent InsP 3 -elevating effect that was largely abolished by a preincubation with PGE 1 (Table 1). This is in agreement with earlier data [44]. U73122 blocked the thrombin-in duced increase in InsP 3 level at concentrations that also suppressed the thrombin-induced Ca 2+ response. Together these results indicate that both InsP 3 -generating (thrombin) and non- InsP 3 -generating (sulfhydryl reagents) agents cause irregular [Ca 2+ ] i spiking in platelets. The thrombin-induced spiking and to a lesser extent the thimerosal/U73122-induced spiking appears to be sensitive to cAMP modulation. Regulation of InsP 3 receptor function and store depletion by Ca 2+ , cAMP and sulfhydryl reagents To better understand the effects of these agents on the spiking p rocess, we directly measured the Ca 2+ release through the InsP 3 receptor channels. Therefore, platelets in suspension were permeabilized with saponin under low Ca 2+ -buffering and ATP-regenerating conditions using Fluo-3 as a Ca 2+ probe [31]. In t his experimental system, InsP 3 caused a (nonlinear) dose-dependent increase in [Ca 2+ ] from stores, which was completely suppressed by the InsP 3 receptor antagonist heparin (Fig. 4A). A low con- centration of InsP 3 (50 n M )causedaCa 2+ release of 27 ± 6 pmol per 10 8 platelets (mean ± SEM, n ¼ 7) at a medium free Ca 2+ concentration of 110 n M . The Ca 2+ -dependency of the InsP 3 -evoked Ca 2+ release was evaluated by permeabilization experiments designed as to prevent c hanges in the Ca 2+ store content. Platelets were thus permeabilized at 110 n M [Ca 2+ ], after which apyrase was added (to block Ca 2+ re-uptake), followed by different amounts of Ca 2+ and 50 n M InsP 3 (see Materials and methods). Under these conditions, the Ca 2+ release Fig. 2. Irregular spiking in [Ca 2+ ] i induced by sulfhydry l-reac tive agents. Aspirin-treated, Fura-2-loaded platelets on a surface were sti- mulated with 10 l M thimerosal (TMS) (A) or 10 l M U73122 (B, C). CaCl 2 and apyrase were present (see Fig. 1); PGE 1 (10 l M ) was given as indicated. Calcium r esponses are s hown of single platelets, and are representative for > 50 cells. 1546 R. M. A. van Gorp et al. (Eur. J. Biochem. 269) Ó FEBS 2002 increased about tenfold when the [Ca 2+ ] was raised from 50 to 200 n M , whereas it declined at [Ca 2+ ] above 400 n M (Fig. 4 B). This result thus resembles the biphasic effect of Ca 2+ on InsP 3 -dependent CICR, p reviously observed in preparations from cerebellum, synaptosomes and A7r5 smooth muscle cells [5–7,45], although in the latter sys- tems higher levels of InsP 3 were needed to achieve Ca 2+ release. Preincubation of platelets with 10 l M PGE 1 before permeabilization resulted in a 50% suppression of the Ca 2+ -mobilizing e ffect of InsP 3 but w as without influence on the biphasic effect of Ca 2+ (Fig. 4B and Table 2). Control experiments indicated t hat PGE 1 treatment did not influence the slow Ca 2+ release evoked by the endomem- brane Ca 2+ -ATPase inhibitor thapsigargin (data not shown). Thus, cAMP elevation seems to partially block the InsP 3 receptor channel opening, but not to affect the sensitization mechanism by Ca 2+ . Further experiments with permeabilized platelets were performed under conditions where the Ca 2+ release process was most sensitive to modulation, i.e. at InsP 3 and Ca 2+ concentrations of 50 and 110 n M , respectively. Thrombin activation of the platelets prior to p ermeabilization signifi- cantly increased the amount of Ca 2+ released by InsP 3 (Table 2). This i s possibly due to a decrease in the platelet cAMP level caused by this G i -stimulating agonist [46]. Thimerosal and U73122 had very different effects. Thim- erosal (10 l M ) did not elicit Ca 2+ release, but strongly stimulated InsP 3 -induced Ca 2+ release (Fig. 5A), as repor- ted for hepatocytes a nd other cells [13,40,41]. On the other hand, a h igh dose o f U73122 (10 l M ) caused stro ng release of Ca 2+ by itself (Fig. 5A), which process was insensitive to pretreatment with hep arin or PGE 1 (Table 2). This U73122 reaction was of little effect on subsequent InsP 3 -induced Ca 2+ release. Control experiments showed that the inhib- itory effects of heparin and PGE 1 on InsP 3 -evoked Ca 2+ release were not influenced by U73122 (Table 2). These data thus suggest that InsP 3 and U73122 have additive effects on Ca 2+ release from intracellular stores. To confirm this, InsP 3 was applied at a higher, saturating concentration. With 1 l M InsP 3 ,increasing[Ca 2+ ]from 100 to 200 n M resulted in a 1.7–fold (± 0.2, n ¼ 3) increase in Ca 2+ release; PGE 1 pretreatment reduced the Ca 2+ release by 45%. When given after high InsP 3 (1 l M ), U73122 (10 l M ) s till caused a rapid phase of Ca 2+ release (Fig. 5 B). This suggested that its e ffect was mediated b y Ca 2+ -leak channels different from the InsP 3 receptors. Thapsigargin was used to determine the possible effect of U73122 on (thapsigargin-releasable) Ca 2+ stores [18,27]. In permeabilized platelets, thapsigargin (1 l M )causeda slow but progressive Ca 2+ release, when applied either before or after InsP 3 . However, the release by U73122 was not reduced, but even proceeded faste r, after InsP 3 /thaps- igargin application (Fig. 5B). When applied to suspensions of intact platelets in EGTA-containing medium, thapsigar- gin caused slow and partial Ca 2+ release. In this system, preincubation with U73122 accelerated and potentiated the Ca 2+ release in a similar way as did the InsP 3 -generating agonist thrombin (Fig. 5C). A synergism of thapsigargin- and thrombin-evoked Ca 2+ mobilization in platelets has Fig. 3. Regulation of thimerosal-induced spiking in [Ca 2+ ] i . Immobilized platelets were stimulated with 10 l M thimerosal (TMS) under conditions, as de scribed for Fig. 2. Manoalide (10 l M ) was added at either 5 min before (A) or 1.5 min after (B) thimerosal. In other e xperiments, PGE 1 (10 l M )wasadded at 5 min before (C) or 1.5 min after (D) thimerosal. Traces are typical responses from a single platelet, representative for 50–75 analysed cells. Table 1. Levels of InsP 3 activated platelets. Aspirin -treated platelets (1 · 10 9 mL )1 )in1m M CaCl 2 and apyrase remained unstimulated or were activated with thrombin (10 n M )orthimerosal(10l M ). The platelets were preincubated with U73122 (2 l M )and/orPGE 1 (10 l M ) for 5 min, where indicated. Mass amounts of InsP 3 were determined after 5 s (thrombin) or 60 s (thimerosal) of activation, i.e. when maximal rises in [Ca 2+ ] i were reached, as measured in parallel incu- bations. Data are mean values ± SEM (n ¼ 4–6). ND, not deter- mined. Agonist InsP 3 (pmol/10 8 platelets) No pretreatment PGE 1 pretreatment None 0.48 ± 0.06 ND Thrombin 1.40 ± 0.15 a 0.75 ± 0.04 b U73122 + thrombin 0.51 ± 0.08 c ND Thimerosal 0.58 ± 0.14 0.56 ± 0.10 a P < 0.005; b P < 0.05 compared to the control condition, i.e. no agonist (t-test, two-sided); c n ¼ 3. Ó FEBS 2002 Regulation of calcium spiking in platelets (Eur. J. Biochem. 269) 1547 been described earlier [18], but this can now be extended to thapsigargin- and U73122-evoked responses. From these experiments we concluded that InsP 3 ,Ca 2+ (via CICR) and U73122 c ause additional a mounts of Ca 2+ release both in intact and permeabilized platelets. The sulfhydryl reagent U73122 seems to release Ca 2+ from stores that differ from those used by InsP 3 , in a way insensitive to heparin and cAMP. Puff-like characteristics of [Ca 2+ ] i spiking in single platelets To determine the involvement of different Ca 2+ stores in the [Ca 2+ ] i spiking process in single platelets, we monitored this at higher spatial and temporal resolution. A fast confocal fluorescence laser system was used to produce fluorescent images from immobilized Fluo-3-loaded plate- lets at an image resolution of 6.0 pixels per micrometer and a scanning rate of 10 Hz. Because platelets spread on fibrinogen increase in surface area from about 2–4 lmin diameter (thickness of  0.5 lm), this set-up gave image series of 250–450 pixels per platelet. We fi rst monitored the characteristics of the Ca 2+ release events at low agonist conditions, i.e. the ÔspontaneousÕ [Ca 2+ ] i spikes that are due to autocrine p roduced ADP [ 35]. Quite similar fluctuating patterns in fluorescence were detected in different sub- cellular regions (80–100 pixels) within a single platelet (Fig. 6 A-B). The fluorescence pattern was completely different in the adjacent region of a nearby platelet, proving that the optical resolution was sufficiently high to detect differences between the selected regions. The high temporal resolution allowed precise analysis of the [Ca 2+ ] i spikes. The ÔspontaneousÕ peaks arose after long but variable intervals of 15.1 ± 1.6 s (mean ± SEM, n ¼ 63) (Fig. 6C). The ampli- tudes of the individual peaks were highly variable, but rela- ted to the total peak duration (Fig. 6D). When compared to the usual criteria for low-amplitude Ca 2+ puffs (maximal Fig. 4. InsP 3 -induced CICR in permeabilized platelets. (A) Traces of InsP 3 -induced mobilization of Ca 2+ from stores. Asp irin-treated platelets (3 · 10 8 ÆmL )1 ) permeabilized with saponin in the presence of Fluo-3, as described in Materials and methods. The Ca 2+ level of the medium was a djusted to 110 n M ,InsP 3 was a dded at 50 or 200 n M concentrations, heparin (20 lgÆmL )1 ) was given at 2 min before InsP 3 where indicated. (B) InsP 3 –induced Ca 2+ release as a function of [Ca 2+ ] of the medium. Plate lets were pe rmeabilized with saponin at 110 n M Ca 2+ , after which ATP generation was abolished with apyrase, and Ca 2+ in the medium was changed to the indicated level (x axis). The release o f C a 2+ by 50 n M InsP 3 was measured (y axis). Before permeabilization, the platelets were treated with 10 l M PGE 1 (open circles) or r emained untreated (closed circles). Vertical line is at standard [Ca 2+ ]of110n M . Data are from three or more experiments (mean ± SE M). Table 2. Modulatio n of Ins P 3 -induced Ca 2+ release in permeabilized platelets. Aspirin-treated platelets in KCl/ATP medium containing Fluo-3 were left untreated or were treated with PGE 1 (10 l M ,5min), where indicated. Part of the platelets was activated with thrombin (40 n M , 2 min). The pl atelets were permeabilized with saponin, and [Ca 2+ ] in t he medium wa s adjusted to 110 n M . Thimerosal ( 10 l M ), U73122 (10 l M ) and/or h eparin (2 min, 20 lgÆmL )1 ) w er e added a t 8 min after saponin, as indicated. InsP 3 (50 n M ) was given at 10 min after saponin. Increases in [Ca 2+ ] were measured i n response to the agonist (thrombin, thimerosal or U73122) and I nsP 3 .TheCa 2+ release by 50 n M InsP 3 under control conditions (no pretreatment) was taken as 100% (82 ± 17 n M ,equivalentto27±6pmolper10 8 platelets). Data are mean values ± SEM (n ¼ 3–5). N D, not determined. Agonist Ca 2+ release (% of control) Agonist, no heparin InsP 3 , no heparin InsP 3 with heparin None – 100 (control) 10 ± 3 a + PGE 1 –46±8 a ND Thrombin ND 139 ± 13 a 5±2 a Thimerosal 8 ± 4 217 ± 30 a,b 3±2 a + PGE 1 15 ± 5 92 ± 16 ND U73122 472 ± 24 78 ± 12 c 3±1 a,c + PGE 1 495 ± 35 46 ± 9 c,d ND a P < 0.001 compared to the release by InsP 3 under control con- ditions, i.e. no pretreatment/no other agonist (t-test, two-sided). b Effect of thimerosal was 145 ± 17% of control with 500 instead of 50 n M InsP 3 . c Relative to corresponding control value at 550 n M [Ca 2+ ]. d P < 0.01 compared to control conditions. 1548 R. M. A. van Gorp et al. (Eur. J. Biochem. 269) Ó FEBS 2002 amplitude of < 200 n M and t otal durat ion o f 1–2 s) [2–4], many of the low-amplitude Ca 2+ release events in platelets (< 200 n M ) appear to be of longer dur ation. The high-resolution confocal scanning revealed irregular trains of [Ca 2+ ] i spikes when the Fluo-3-loaded platelets were stimulated with thrombin (Fig. 7A). A gain, no more than minor differences in peak generation were f ound between different subcellular regions. The average peak-to- peak interval was n ow decrea sed to 4 .8 ± 0.3 s (mean ± SEM, n ¼ 67 peaks of 20 cells). This is similar to the highest oscillation frequency reported for ATP-stimulated rat megakaryocytes (peak-to-peak interval per cell varying from 5 to 30 s) [22]. After platelet stimulation with thimerosal, again trains of [Ca 2+ ] i peaks started almost simultaneously in vari ous subcellular parts (Fig. 7B). W ith thimerosal, the average peak-to-peak interval was 8.9 ± 0.8 s (mean ± SEM, n ¼ 50; P < 0.001 compared to thrombin). Thus, regardless of the peak generation frequency, individual Ca 2+ -release events seemed to be generated in various parts of a platelet at quite similar intervals and amplitudes. DISCUSSION Here we describe that InsP 3 -mobilizing agonists (thrombin, U46619 and p latelet-activating factor) as well as agents acting independently of InsP 3 formation (thimerosal and U73122 at 10 l M ) evoke irregular [Ca 2+ ] i spiking in aspirin- treated platelets. T he thrombin-induced spiking app ears to be strictly dependent on InsP 3 formation, because it is abolished by manoalide or low U73122. It is also inhibited by cAMP elevation with PGE 1 , in part due to reduced InsP 3 formation ( probably by phospholipase C inhibition) and in part due to decreased InsP 3 -mediated Ca 2+ release from intracellular stores. On the other hand, the sulfhydryl reagent thimerosal elicits [ Ca 2+ ] i spiking not by increasing the InsP 3 level but by potentiating InsP 3 receptor-mediated Ca 2+ release. This may explain why the Ca 2+ response with thimerosal is only partially inhibitable by P GE 1 .The N-ethyl maleimide derivative U73122, at a high dose of 10 l M , yet acts in a still different manner. In permeabilized Fig. 6. Confocal monitoring of ÔspontaneousÕ spiking in [Ca 2+ ] i in spread platelets. Fluorescence changes were monitored by confocal laser scanning microscopy in aspirin-treated, Fluo-3-loaded platelets spread on fibrinogen. Apyrase was omitted from the incubation medium (nominally Ca 2+ -free). High-resolution images of 250–450 pixels/ platelet were collected at 10 Hz. (A) Fluorescence recordings from three selected regions of one spread platelet (a-c); and from a region of interest of an adjace nt platelet ( r) (initial value of each trace, F/F o ¼ 1). Insert shows expanded part of curves a–c. (B) Selection of regions of interest of the platelets ( areas  0.8 · 2.5 lm). (C) Histogram of variation in peak-to-peak interval of 15 responsive platelets. (D) Plot of total duration of individual peaks (90% decay) vs. peak amplitude. Data are mean values plus SEM of analysis results from the three regions per platelet. Regression analysis of all data: y ¼ 0.83 + 0.98 x (R 2 ¼ 0.68, P < 0.001). Fig. 5. Calcium mobilization from stores in permeabilized and intact platelets. (A,B) Aspirin-treated platelets were permeabilized with saponin in Fluo-3-containing me dium. A fter [ Ca 2+ ] adjustment to 110 n M ,thimerosal(TMS,10l M ), U73122 (10 l M ) and thapsigargin (TG, 1 l M )were given, as indicated. (A) InsP 3 was added at a low concentration of 50 n M with 3 · 10 8 plateletÆmL )1 (B) InsP 3 was given at a higher concentration (1 l M ), while the platelet concentration was 2 · 10 8 plateletsÆmL )1 . Note that U73122-evoked Ca 2+ release leads to a h igher m edium [Ca 2+ ], which potentiates t he InsP 3 -evoked r elease. (C) Intact, a spirin-treated platelets in suspension (1 · 10 8 plateletsÆmL )1 ), loaded with Fura-2, were stimulated with thrombin (T hr, 4 n M ), thapsigargin (1 l M ) and/or U73122 (10 l M ) in the presence of 1 m M EGTA. Ó FEBS 2002 Regulation of calcium spiking in platelets (Eur. J. Biochem. 269) 1549 platelets, it causes a cAMP/heparin-insensitive Ca 2+ leak that seems to be independent of the InsP 3 receptor-mediated Ca 2+ release. It can thus be envisioned that, in intact platelets, the Ca 2+ release evoked by U73122 s timulates the process of InsP 3 receptor-mediated CICR, and thereby the generation of [Ca 2+ ] i spikes. In a variety of cells, thimerosal is known to react with critical thiol groups controlling InsP 3 -receptor channel opening, which results in repetitive Ca 2+ release at basal levels of InsP 3 [13,14,40,41]. In platelets sulfhydryl groups may s imilarly control I nsP 3 receptor functioning [42]. This agrees with our finding that, i n permeabilized platelets, heparin completely inhibits the t himerosal-enhanced Ca 2+ release by InsP 3 . Taken together, the present work thus indicated t hat the platelet InsP 3 receptors p lay a key role in the regenerative, spiking Ca 2+ release evoked by phospho- lipase C -stimulating and InsP 3 receptor-modulating agents, similarly as established f or other cell types. Using saponin-permeabilized platelets, we foun d that the InsP 3 -evoked Ca 2+ -mobilizing pote ncy changed with the cytosolic Ca 2+ concentration in a biphasic way (Fig. 6 ), similarly as firstly described for neuronal cells [5–7] and later for pancreatic acinar c ells, hepatocytes and smooth muscle cells [45,47,48]. Whereas in many cell t ypes micromolar concentrations of InsP 3 were needed to detect a stimulating effect of Ca 2+ on InsP 3 receptor-mediated Ca 2+ release [5–8,45,47], this could be demonstrated in platelets already low levels of 50–200 n M InsP 3 . It is noted that platelets are relatively rich in type 1 InsP 3 receptors [26], which are quite sensitive to Ca 2+ modulation. For rabbit a nd mouse pancreatic acinar cells, it has been shown that U73122 evokes [Ca 2+ ] i oscillations by potenti- ating the release of Ca 2+ from a InsP 3 -sensitive store compartment [39,43]. This release may lead to increased Ca 2+ influx from the e xternal medium and to subsequen t overloading of InsP 3 -insensitive stores, which in turn can trigger regenerative C a 2+ release [1,43]. A similar mechan- ism, i.e. cooperation of store compartments in [Ca 2+ ] i spiking, may also apply to platelets. Typical for platelets is that the amount of Ca 2+ released by a suboptimal InsP 3 concentration, but not the Ca 2+ sensitivity of the release, is suppressed upon cAMP eleva- tion. There i s little doub t that most o r all cAMP-m ediated effects in platelets are due to cAMP-dependent protein phosphorylation, and that the platelet InsP 3 receptors are targets of cAMP-dependent protein kinase [27]. Earlier, we have reported that thrombin- and thapsigargin-induced Ca 2+ responses in platelets are down-regulated by cAMP analogues and inhibitors of cAMP phosphodiesterase, a nd that cAMP-dependent protein kinase was important in this effect. These cAMP-elevating interven tions also suppressed the InsP 3 -induced Ca 2+ mobilization in saponin-permeabi- lized platelets [ 46]. Together w ith the new evidence it thus becomes clear that cAMP-dependent phosphorylation ren- ders the InsP 3 receptor less active as a Ca 2+ channel [30,31,49], and also that the phosphorylated receptor remains sensitive to change s in [ Ca 2+ ] i (this paper). In this respect, platelets differ from other cells such as hepatocytes, where a ctivation of c AMP-dependent kinase was found to increase the amount of Ca 2+ released by InsP 3 [41]. The confocal laser scanning experiments with Fluo- 3-loaded platelets, permitting a simultaneously high tem- poral and spatial resolution of the Ca 2+ signal, clearly indicated that the [Ca 2+ ] i release events in platelets are highly irregular in s hape, amplitude and frequency, regard- less of whether they are raised by InsP 3 -generating receptor agonists or sulfhydryl-reactive compounds. The experi- ments show that the irregular traces detected in Fura- 2-loaded platelets by camera-based microfluorometry are most probably not artefacts of the ratio imaging procedure. In addition, they detect similar Ca 2+ release events at distant sites within a platelet: this holds not only f or single [Ca 2+ ] i spikes, but also for complex series of consecutive spikes (Figs 6 ,7). Calcium puffs as recorded in large r cells are commonly d efined as single Ca 2+ release events that arise due to the action of multiple InsP 3 receptor c hannels clustered in f unctional units [2–4]. The operating definitions of a Ca 2+ puff vary somewhat, but congregate as a local Ca 2+ release event (diameter about 1 lm) with a maximal amplitude of < 200 n M , a rising time of < 0 .35 s and t otal duration of 1–2 s. The Ca 2+ spikes of platelets resemble the puffs seen in larger cells in local appearance, but differ from these in at least two a spects. First, the platelet spikes appear at a v ariable frequency (0.02–0.3 Hz), regardless o f whether CaCl 2 or EGTA is externally present (see [35]). Second, they are rather broad and do not sum up, i.e. the individual Fig. 7. Uniform [Ca 2+ ] i transients within act i- vated, spread platelets. Fluo-3-loaded platelets were stimulated with (A) thrombin (4 n M , given a t t ¼ 8 s) o r (B) thimerosal (10 l M , given a t t ¼ 0 s) in the presence of 1 m M CaCl 2 and apyrase. High-resolution images were collected by co nfocal laser s canning microscopy, as described for Fig. 6. Fluores- cence recordings are show n from three non- overlapping regions of one platelet (initial value of each trace, F/F o ¼ 1). Inserts give extended parts. Data are representative for 3 or more experiments. 1550 R. M. A. van Gorp et al. (Eur. J. Biochem. 269) Ó FEBS 2002 events do not seem to be subjected t o frequency or amplitude recruitment, such as described for HeLa cells [4]. Because of the small size of platelets with nearby Ca 2+ - ATPases throughout the cell, it is likely that the rate of Ca 2+ pumping rather than the diffusion of released Ca 2+ into the c ytosol (as in bigger cells) d etermines the duration of the platelet spikes. In many cell types, the global release of Ca 2+ is controlled by an intimate interplay between thapsigargin- and InsP 3 -sensitive Ca 2+ store compartments. For instance, in rabbit pancreatic a cinar cells the (thapsigargin-inhibited) compensatory Ca 2+ pumping by endomembrane Ca 2+ - ATPases restricts the Ca 2+ -store depletion by InsP 3 [47]. In mouse lacrimal cells, the thapsigargin-induced Ca 2+ mobilization is dependent on the basal level of InsP 3 and the InsP 3 -receptor func tion [50]. Such a situation m ay also exists in platelets, where both the InsP 3 - a nd thapsigargin- sensitive Ca 2+ store compartments are likely to contribute to the [Ca 2+ ] i spiking [18,27]. In the present paper, we describe that regardless of the type of a gonist, stimulating (thrombin) or sensitizing (thimerosal) InsP 3 receptors or acting primarily independently of InsP 3 receptors (U73122), and regardless o f the type of stores u sed by these agonists, the spiking process w as always irregular in amplitude and frequency and occurred with no more than little subcellular heterogeneity. This situation however, differs from that of pancreatic acinar cells, where even within the voxel of a Ca 2+ Ôhot spotÕ quite different patterns of spike-like events can be observed [51]. This apparently points to a high cooperation of Ca 2+ mobilization from the various stores in platelets to g enerate s maller as well as l arger C a 2+ -release events. In summary, the small platelets forms an attractive model to study the function of InsP 3 receptors, even when induced by agents such as U73122 and thimerosal that do not cause InsP 3 formation. The platelet InsP 3 receptors are subjected to extensive regulation by at least four factors: (a) local InsP 3 levels; (b) the potentiating effect of moderate increases in [Ca 2+ ] i on InsP 3 action via CICR; (c) InsP 3 receptor channel sensitization (thimerosal) and desensitization (mediated by cAMP); and (d) InsP 3 channel-independent Ca 2+ leak with U73122. Given the importance of the Ca 2+ signal for the process of platelet activation, it is likely t hat the highly regulated nature of the Ca 2+ signal plays a n important role in ensuring rapid p latelet deposition a t the right physiological sites during hemostasis and at athero- sclerotic sites during thrombosis. ACKNOWLEDGEMENTS We ack nowledge grants from the Netherlands Heart Foundation and the Netherlands Organization for Scientific Research. REFERENCES 1. Berridge, M.J. (1993) Inositol trisphosphate and calcium signal- ling. Nature 361, 315–325. 2. Yao, Y., Choi, J & Parker, I. (1995) Quantal puffs of intracellular Ca 2+ evoked by inositol trisphosphate in Xenopus oocytes. J. Physiol. 482, 533–553. 3. Parker, I & Yao, Y. (1996) Ca 2+ transients associated with openings of inositol trisphosphate-gated channels in Xenopus oocytes. J. Physiol. 491, 663–668. 4. Bootman, M.D., Berridge, M.J & Lipp, P. (1997) C ooking with calcium: the recipes for composing global signals from elementary events. Cell 91, 367–373. 5. Finch, E.A., Turner, T.J & Goldin, S .M. ( 1991) C alcium as coa- gonist of inositol 1,4,5-trisphosphate-induced calcium release. Science 252, 443–446. 6. Bezprozvanny, I., Watras, J & Ehrlich, B.E. (1991) Bell-shaped calcium-response curves of Ins(1,4,5)P 3 - and calcium-gated channels from endoplasmic reticu lum of cerebellum. Nature 351, 751–754. 7. Iino, M & E ndo, M. (1992) Calcium-dependent immediate feed- back control of inositol 1,4,5-trisphosphate-induced Ca 2+ release. Nature 360, 76–78. 8. Hagar, R.E., Burgstahler, A.D., Nathanson, M.H & Ehrlich, B.E. (1998) Type III InsP 3 receptor channel stays ope n in the presence of increased calcium. Nature 396, 81–84. 9. Missiaen, L., Taylor, C.W & Berridge, M.J. (1991) Spontaneous calcium release from inositol trisphosphate-sensitive calcium stores. Nature 352, 241–244. 10. Combettes, L., Cheek, T.R & Taylor, C.W. (1996) Regulation of inositol trisphosphate recep tors by luminal Ca 2+ contributes to quantal Ca 2+ mobilization. EMBO J. 15, 2086–2093. 11. Renard, D.C., S eitz, M.B & Thomas, A.P. (1992) Oxidized glutathione causes sensitization of calcium release to inositol 1,4,5-trisphosphate in permeabilized hepatocytes. Biochem. J. 284, 507–512. 12. Renard-Rooney, D .C., Joseph, S.K., Seitz, M.B & Thomas, A.P. (1995) Effect o f oxidized glutathione and temperature on inositol 1,4,5-trisphosphate binding in permeabilized hepatocytes. Bio- chem. J. 310, 185–192. 13. Thrower, E.C., Duclohier, H., Lea, E.J., M olle, G & Dawson, A.P. (1996) The inositol 1,4,5-trisphosphate-gated Ca 2+ channel: effect of the protein thiol reagent thimerosal on channel activity. Biochem. J. 318, 61–66. 14. Wu, J., T akeo, T., K amimura, N., Wa da, J., Duga, S., Hoshina, Y & Wakui, M. (1996) Thimerosal modulates the agonist-specific cytosolic Ca 2+ oscillatory patterns in single pancreatic acinar cells of mouse. FEB S Lett. 390, 149–152. 15. Joseph, S.K. (1996) The inositol trisphosphate receptor family. Cell Signalling 8, 1–7. 16. Wojcikiewicz, R.J.H & Luo, S.G. (1998) Phosphorylation of inositol 1,4,5-trispho sphate receptors by c AMP-depe ndent protein kinase. Type I, II and III receptors are differentially susceptible to phosphorylation and ar e p hosphorylated in i ntact cells. J. Biol. Chem. 273, 5670–5677. 17. Ozaki, Y., Yatomi, Y., Wakasugi, S., Shirasawa, Y., Saito, H & Kume, S. (1992) Thrombin-induced calcium oscillation in human platelets and Meg-01, a megakaryoblastic leukemia cell line. Biochem. Biophys. Res. Commun. 183, 864–871. 18. Heemskerk, J.W .M., Vis, P., Feijge, M.A.H., Hoyland, J., Mason, W.T & Sage, S .O. (1993) R oles of phospholipase C and Ca 2+ - ATPase in calcium responses of single, fibrinogen-bound platelets. J. Biol. Chem. 268, 356–363. 19. Ariyoshi, H & Salzman, E.W. (1995) Spatial and te mporal change in cytosolic pH a nd [Ca 2+ ] in resting and activated single human platelets. Cell Calcium 17, 317–326. 20. Heemskerk, J.W.M., Feijge, M.A.H., Henneman, L., Rosing, J & Hemker, H.C. (1997) The C a 2+ -mobilizing p otency of a-thrombin and thrombin-receptor-activating peptide on human platelets: concentration and time effects of throm bin-induced Ca 2+ signa- ling. Eur. J. Biochem. 24 9, 547–555. 21. Hussain, J.F & Mahaut-Smith, M.P. (1996) ADP evokes oscilla- tions of an I P 3 -dependent monovalent cation current in rat megakaryocytes. J. Physiol. 497, 121P–122P. 22. Tertyshnikova, S & Fein, A. (1997) [ Ca 2+ ] i oscillations and [Ca 2+ ] i waves in rat megakaryocytes. Cell Calcium 21, 331– 344. Ó FEBS 2002 Regulation of calcium spiking in platelets (Eur. J. Biochem. 269) 1551 23. O’Rourke, F., Matthews, E & Feinstein, M.B. (1995) Purification and cha racteriz ation of the hu man t yp e 1 Ins(1,4,5)P 3 receptor from platelets and com parison with rece ptor subtyp es in other normal and transfo rmed blood cells. Biochem. J. 312, 499–503. 24. El-Daher, S.S., Eigenthaler, M., Walter, U., Furuichi, T., Miyawaki, A., Mikoshiba, K., Kakkar, V.V & Authi, K.S. (1996) Distribution and activation of cAMP- and cGMP-depend ent protein k inases in highly purified hu man platelet plasma and intracellular membranes. Thromb. Haemost. 76, 1063–1071. 25. Quinton, T.M., Brown, K.D & Dean, W.L. (1996) Inositol 1,4,5-trisphosphate-mediated Ca 2+ release from platelet internal membranes is regulated by differential phosphorylation . Bio- chemistry 35, 6865–6871. 26. El-Daher, S.S., Patel, Y., Siddiqua, A., Hassock, S., Edmunds, S., Maddison, B., Patel, G., G oulding, D., Lupu, F., Wojcikiewicz, R.J.H & Authi, K.S. (2000) Distinct localization and function of (1,4,5)IP 3 receptor subtypes and the (1,3,4,5)IP 4 receptor GAP1IP4 BP in highly purified human platelet membranes. Blood 95, 3412–3422. 27. Authi, K.S. (1997) Ca 2+ homeostasis in human platelets. In Handbook o f Experimental Pharmacology, Vol. 126 (von Bruch- hausen, F & Walter, U., eds), pp. 325–370. Springer, Berlin, Germany. 28. Van Gorp, R.M.A., van D am-Mieras, M .C.E., Hornstra, G & Heemskerk, J.W.M. (1997) Effect of membrane-permeable sulf- hydryl reagents and depletion of glutathione on calcium mobili- sation in hu man platelets. Biochem. Pharmacol. 53, 1533–1542. 29. Enouf, J ., Giraud, F., Bre doux, R., Bourdeau, N & Le ´ vy-Tole- dano, S. (1987) Possible role of a cAMP-dependent phos- phorylation in the calcium release mediated by inositol 1,4,5-trisphosphate i n h uman platelet membrane vesicles. Biochim. Biophys. A cta 928, 76–82. 30. Quinton, T.M & Dean, W.T. (1992) Cyclic AMP-dependent phosphorylation of the inositol-1,4,5-trisphosphate receptor in- hibits Ca 2+ release from platelet m embranes. Biochem. Biophys. Res. Commun. 184, 893–899. 31. Cavallini, L., Coassin, M., Borean, A & Alexandre, A. (1996) Prostacyclin and sodium nitroprusside inhibit the activity of the platelet inositol 1,4,5-trisphosphate r eceptor and promote its phosphorylation. J. Biol. Chem. 271, 5545–5551. 32. O’Rourke, F ., Zavoico, G.B & Feinstein, M.B. (1989) Release of Ca 2+ by inositol 1,4,5-trisphosphate in platelet membrane vesicles is n ot dependent on cyclic AMP-dependent protein kinase. Bio- chem. J. 257, 715–721. 33. Heemskerk, J.W.M., Vuist, W .M.J., Feijge, M.A.H., Reuteling- sperger, C.P.M & Lindhout, T. (1997) Collagen but no t fibrinogen surfaces induce bleb formation, expo sure of phosphatid ylserine, and p ro coagu lant activity of adherent platelets: evide nce for reg - ulation by protein tyrosine kinase-dependent Ca 2+ responses. Blood 90, 2615–2625. 34. Van der Velden, H.M., van Kempen, M .J., Wijffels, M.C., van Zijverden, M., Groenewegen, W.A., Alessie, M.A & Jongsma, H.J. (1998) A ltered pattern of connexin 40 distribution in persis- tent atrial fibrillation in the goat . J. Cardiovasc. Electrophysiol. 9, 596–607. 35. Heemskerk, J.W.M., Hoyland, J., Mason, W.T & Sage, S.O. (1992) Spiking in cytosolic calcium concentration in single fibrinogen-bound fura-2-loaded human platelets. Biochem. J. 283, 379–383. 36. Bennett, C.F., Mong, S., W u, H.L.W., Clark, M.A., Wheeler, L.A & Crooke, S.T. (1987) Inhibition of phosphoinositide-specific phospholipase C by mano alide. Mol. Pharmacol. 32, 587–593. 37. Diamond, S.L., Sachs, F & Sigurdson, W.J. (1994) Mechanically induced c alcium mobilization in cultured endothelial cells is dependent on actin and phospholipase. Arterioscler. Thromb. 14, 2000–2006. 38. Heemskerk, J.W.M., Farndale, R.W & Sage, S.O. (1997) Effects of U73122 and U73343 on human platelet calcium signalling and protein tyrosine phosphorylation. Biochim. Biophys. Acta 1355, 81–88. 39. Mogami, H., Lloyd Mills, C & Gallacher, D.V. (1997) Phospho- lipase C inhibitor, U73122, releases intracellular Ca 2+ ,potentiates Ins(1,4,5)-mediated Ca 2+ release and directly activates ion channels in mouse p ancreat ic acinar c ells. Biochem. J. 324,645– 651. 40. Bootman, M.D., Taylor, C.W & Berridge, M.J. (1992) The thiol reagent, th imerosal, evokes Ca 2+ spikes in HeLa cells by sensiti- zing th e inositol 1,4,5-trisphosphate receptor. J. Biol. C hem. 267, 25113–25119. 41. Bird, G.S.J., Burgess, G.M & Putney, J.W. (1993) Sulfhydryl reagents and cAMP-dependent k inase increase t he sensitivity o f the inositol 1 ,4,5-trisphosphate recepto r in h epatocytes. J. Bi ol. Chem. 268, 17917–17923. 42. Adunyah, S & Dean, W.L. (1986) Effe cts of sulfhydryl reagents and other inhibitors on t he Ca 2+ transport and inositol tris- phosphate-induced Ca 2+ release f rom hum a n pl atelets . J. Biol. Chem. 261, 13071–13075. 43. Willems, P.H.G.M., van de Put, F.H.M.M., Engbersen, R., Bosch, R.R., van Hoof, H.J.M & de Pont, J.J.H.H.M. (1994) Induction of C a 2+ oscillations by selective, U73122-mediated depletion of inositol-trisphosphate-sensitive Ca 2+ stores in rabbit pancreatic acinar cells. Pflu ¨ gers Arch. 427 , 23–243. 44. Ryningen, A., Jensen, B.O & Holmsen, H. (1998) Elevation o f cyclic-AMP decreases phosphoinositide turnover and inhibits thrombin-induced secretion in human platelets. Biochim. Biophys. Acta 1394, 235–248. 45. Missiaen, L., De Smedt, H., Parys, J.B & Casteels, R. (1994) Co-activation of inositol trisphosphate-induced Ca 2+ release b y cytosolic C a 2+ is loading- dependent. J . Biol. Che m. 269, 7238– 7242. 46. Keularts, I.M.L.W., van Gorp, R .M.A., Feijge, M .A.H., Vuist, W.M.J & Heemskerk, J.W.M. (2000) a 2A -Adrenergic receptor stimulation potentiates calcium release in platelets by m odulating cAMP levels. J. Biol. C hem . 275, 1 763–1772. 47. Van de Put, F.H.M.M., de Pont, J.J.H.H.M & Willems, P.G.H.M. (1994) Heterogeneity between intracellular Ca 2+ stores as the underlying p rinciple of quantal C a 2+ release by inositol 1,4,5-trisphosphate in permeabilized pancreatic acinar cells. J. Biol. Chem. 269, 12438–12443. 48. Marshall, I.C.B & Taylor, C .W. (1993) Biphasic effects of c yto- solic Ca 2+ on Ins (1,4,5) P 3 -stimulated Ca 2+ mobilization in hepatocytes. J. Biol. Chem. 268, 13214–13220. 49. Tertyshnikova, S & Fein, A. (1998) Inhibition of inositol 1,4,5-trisphosphate-induced Ca 2+ release by cAMP-dependent protein kinase in a living cells. Proc. Natl Acad. Sci. USA 95, 1613–1617. 50. Smith, P.M & Gallacher, D.V. (1994) Thapsigargin-induced Ca 2+ mobilization in acutely isolated m ouse lacrimal acinar cells is dependent on a basal level of Ins (1,4,5) P 3 and i s inhibited by heparin. Bioche m. J. 299, 37–40. 51. Thorn, P., Moreton, R & Berridge, M. (1996) Multiple, coordinated Ca 2+ release events underlie the inositol trisphos- phate-induced Ca 2+ spikes in mouse pancreatic acinar cells. EMBO J. 15, 999–103. 1552 R. M. A. van Gorp et al. (Eur. J. Biochem. 269) Ó FEBS 2002 . Irregular spiking in free calcium concentration in single, human platelets Regulation by modulation of the inositol trisphosphate receptors Roosje. characteristics of [Ca 2+ ] i spiking in single platelets To determine the involvement of different Ca 2+ stores in the [Ca 2+ ] i spiking process in single platelets,