Biochemical Journal

Research article

NAADP regulates human platelet function

Carmen H. Coxon, Alexander M. Lewis, Amanda J. Sadler, Sridhar R. Vasudevan, Andrew Thomas, Kirsten A. Dundas, Lewis Taylor, R. Duncan Campbell, Jonathan M. Gibbins, Grant C. Churchill, Katherine L. Tucker


Platelets play a vital role in maintaining haemostasis. Human platelet activation depends on Ca2+ release, leading to cell activation, granule secretion and aggregation. NAADP (nicotinic acid–adenine dinucleotide phosphate) is a Ca2+-releasing second messenger that acts on acidic Ca2+ stores and is used by a number of mammalian systems. In human platelets, NAADP has been shown to release Ca2+ in permeabilized human platelets and contribute to thrombin-mediated platelet activation. In the present study, we have further characterized NAADP-mediated Ca2+ release in human platelets in response to both thrombin and the GPVI (glycoprotein VI)-specific agonist CRP (collagen-related peptide). Using a radioligand-binding assay, we reveal an NAADP-binding site in human platelets, indicative of a platelet NAADP receptor. We also found that NAADP releases loaded 45Ca2+ from intracellular stores and that total platelet Ca2+ release is inhibited by the proton ionophore nigericin. Ned-19, a novel cell-permeant NAADP receptor antagonist, competes for the NAADP-binding site in platelets and can inhibit both thrombin- and CRP-induced Ca2+ release in human platelets. Ned-19 has an inhibitory effect on platelet aggregation, secretion and spreading. In addition, Ned-19 extends the clotting time in whole-blood samples. We conclude that NAADP plays an important role in human platelet function. Furthermore, the development of Ned-19 as an NAADP receptor antagonist provides a potential avenue for platelet-targeted therapy and the regulation of thrombosis.

  • calcium
  • cell signalling
  • Ned-19
  • nicotinamide acid–adenine dinucleotide phosphate (NAADP)
  • platelet


Platelets are small anucleate cells that circulate in high numbers in the blood. Under normal physiological conditions, platelets respond to vascular injury to induce the formation of a thrombus which serves to limit blood loss and maintain haemostasis. Platelets also play a role in the formation of pathological thrombi, resulting in myocardial infarction or stroke. Understanding the regulation of platelet function will provide new opportunities to effectively manage platelet activation and reduce thrombotic risk.

NAADP (nicotinic acid–adenine dinucleotide phosphate) is a potent mediator of Ca2+ release in a number of mammalian systems, resulting in a rapid release of Ca2+ from an internal acidic store [13]. Originally identified as a Ca2+-releasing messenger in sea urchin egg homogenates [4], there have since been a series of reports describing the actions of NAADP in a broad range of mammalian cells and tissues, including pancreatic acinar cells [510], MIN-6 cells [11], T-lymphocytes [12], heart microsomes [13], hepatocyte microsomes [14], neurons [1517] and platelets [1821].

Ca2+ release is a vital process for human platelet activation and hence the function of platelets in maintaining haemostasis. Human platelets contain two Ca2+ stores: the DTS (dense tubular system) and a potentially lysosome-like acidic store [1820,22]. Ins(1,4,5)P3 is known to act upon receptors in the DTS following PLC (phospholipase C) activation in response to agonist [23], leading to a rise in intracellular Ca2+ concentration. The luminal Ca2+ concentration of the DTS is maintained by the SERCA (sarcoplasmic/endoplasmic reticulum Ca2+-ATPase) 2b Ca2+ pump, and this store displays a high sensitivity to thapsigargin [19,22,24] and is insensitive to TBHQ [2,5-di-(t-butyl)-1,4-hydroquinone] [19]. Previously, it was thought that platelets do not contain ryanodine receptors and are hence insensitive to the actions of cADPR (cADP-ribose) [22]. Recent work, however, has shown the generation of cADPR in mouse platelets in response to thrombin, as well as demonstrating a requirement for CD38 activity for murine platelet activation and thrombus formation both in vitro and in vivo [21].

The platelet acidic Ca2+ store expresses both the SERCA3 Ca2+ pump and the vacuolar H+-ATPase, and exhibits low sensitivity to thapsigargin and high sensitivity to TBHQ and nigericin [19,22,24]. NAADP can induce Ca2+ release in permeabilized platelets and can be inhibited by selective depletion of the acidic store with TBHQ or the L-type Ca2+ channel inhibitor nimodipine [19]. It has been shown that the thrombin GPCRs (G-protein-coupled receptors) PAR (protease-activated receptor)-1 and PAR-4 release Ca2+ from both the DTS and acidic stores. PAR-1 mainly induces Ca2+ release from the DTS, whereas PAR-4 induces Ca2+ release from both the DTS and the acidic store [18,19]. Thrombin-mediated Ca2+ release can also occur through the activation of the von Willebrand factor GP (glycoprotein) Ib–IX–V receptor complex. Thrombin can induce Ca2+ release from the acidic store though GPIb–IX–V in platelets that have been desensitized to both PAR-1 and PAR-4 [18]. CRP (collagen-related peptide) is a selective GPVI agonist that leads to increases in intracellular Ca2+ via PLCγ2 to mediate platelet activation, but its dependency on NAADP for this process has not yet been investigated.

In the present study, we have employed a selective NAADP receptor antagonist, Ned-19 [25,26]. Elucidating the role of NAADP-mediated Ca2+ release in cellular activation has been hampered by the poor availability of suitable tools [18,19,27,28]. The emergence of new cell-permeant small-molecule modulators of NAADP-mediated Ca2+ release has been of great value in the study of NAADP as a second messenger [25,2933]. We investigated the contribution of NAADP in agonist-induced human platelet activation via both GPCR and tyrosine kinase receptor activation. Using radioligand-binding assays, we identified an NAADP-binding site in human platelets, strongly suggestive of an NAADP receptor, for which the novel NAADP receptor antagonist Ned-19 can compete. We also found that NAADP was able to release Ca2+ from permeabilized platelets loaded with 45Ca2+. Using the acidic store inhibitors TBHQ and nigericin, we demonstrated that both CRP and thrombin rely on the acidic store for the mobilization of Ca2+. Ned-19 inhibits CRP and thrombin-induced Ca2+ release as well as aggregation, α-granule secretion and cell spreading. We conclude that NAADP has an important role in agonist-induced human platelet activation. Furthermore, these findings provide a new avenue for the development of novel anti-platelet drugs and for the management of cardiovascular disease.



CRP, a GPVI-selective agonist, was synthesized by, and purchased from, Richard Farndale (Department of Biochemistry, University of Cambridge, Cambridge, U.K.). PE (phycoerythrin)-conjugated anti-P-selectin (CD62P) (RB40.34) was from BD Biosciences. All other materials were purchased from Sigma unless stated otherwise.

Preparation of human platelets

Human platelets were isolated as described previously [34]. Platelets were prepared from whole blood by differential centrifugation and resuspended in modified Tyrode's/Hepes buffer (134 mM NaCl, 0.34 mM Na2HPO4, 2.9 mM KCl, 12 mM NaHCO3, 20 mM Hepes, 5 mM glucose and 1 mM MgCl2, pH 7.3) to a density of 4×108 cells/ml.

Radioreceptor-binding assay

Freshly prepared washed platelets were sonicated on ice (40% power, 5 s on, 9.9 s off for 10 min) in Tyrode's/Hepes buffer in the presence of a protease inhibitor cocktail (Roche). Specific binding was determined in a manner similar to that described previously [35]. Briefly, standard concentrations of NAADP or Ned-19 were prepared in intracellular medium or DMSO as required. Sonicated platelets in Tyrode's/Hepes buffer were then added to give a final protein concentration of approximately 1 mg/ml. [32P]NAADP was synthesized from [32P]NAD (PerkinElmer) as described previously [35] and added to give a final concentration of ~2 nM. The tubes were then incubated for 1 h at room temperature (25°C), before filtration using a Brandel cell harvester and Whatmann GF/B filter papers, washing three times with ice-cold intracellular medium. Radioactivity was determined using a storage phosphor screen and Typhoon 9400 scanner (GE Healthcare).

45Ca2+-release assay

Regenerative platelet medium was prepared with Tyrode's/Hepes buffer containing 2 mM ATP, 10 units/ml creatine kinase, 100 mM creatine phosphate, 50 μM digitonin and 2 μM 45Ca2+. Uptake of 45Ca2+ into platelets was initiated by adding platelets to a final concentration of 0.1–0.2 mg of protein/ml to the regenerative platelet medium prepared above. Uptake was performed for 60 min. At the end of the uptake period, NAADP was added and, after 10 s, the whole preparation was washed through a cell harvester manifold on to Whatmann GF/B filter papers. Radioactivity was determined using a storage phosphor screen and Typhoon 9400 scanner.

Cytotoxicity assay

Cytotoxicity was assessed using the alamarBlue® cell health and viability assay (Invitrogen). Experiments were carried out according to the manufacturers' instructions. In brief, washed human platelets were incubated with Tyrode's/Hepes buffer, DMSO or 1, 10, 50 or 100 μM Ned-19 for 1 h before fluorescence assessment of resazurin conversion.


Platelets (450 μl) were stimulated with thrombin or CRP in a final volume of 500 μl at 37°C with continuous stirring (1200 rev./min) in an optical aggregometer. For Ned-19 studies, platelets were incubated with Ned-19 for 30 s. After this time, agonist was added and stirring was initiated. Aggregations were followed for 5 min following the addition of agonist.

α-Granule secretion

α-Granule release was measured by cell-surface exposure of P-selectin using a PE-conjugated anti-P-selectin antibody. In a total volume of 50 μl, 2×106 washed platelets were stimulated with either 0.5 unit/ml thrombin or 1 μg/ml CRP in the presence of 1 μl of PE-conjugated anti-P-selectin antibody. Reactions were terminated by the addition of 950 μl of modified Tyrode's/Hepes buffer. Cell-surface exposure of P-selectin was determined by the mean fluorescence intensity of the anti-P-selectin antibody using a FACScan flow cytometer (Becton Dickinson) and analysed using CELLQuest software (Becton Dickinson).

Measurement of [Ca2+]i by spectrofluorimetry

Washed human platelets (at 4×108 cells/ml) were incubated in nominally Ca2+-free Tyrode's/Hepes buffer with 3 μM fura 2/AM (acetoxymethyl ester) for 45 min as described previously [36]. Briefly, platelets were washed once and resuspended at 4×108 cells/ml in modified Tyrode's/Hepes buffer. Platelets (450 μl) were stimulated with either 250 ng/ml CRP or 0.1 unit/ml thrombin in the presence or absence of 1 mM EGTA with constant stirring at 37°C. Fura 2 fluorescence was measured on a LS-50B spectrophotometer (PerkinElmer) with excitation wavelengths of 340 and 380 nm. Fluorescence emission was measured at a wavelength of 510 nm. The ratio of emission values (excitation at 340/380 nm) was calculated and converted into Ca2+ concentration using FLWinLab software (PerkinElmer) using the equation [Ca2+]i=Kd×(RRmin)/(RmaxR)×SFB, where Kd=224 nM, R is the 340/380 nm ratio, Rmin is R under Ca2+-free conditions, Rmax is R under Ca2+-saturated conditions and SFB is the ratio of baseline fluorescence at 380 nm under Ca2+-free and Ca2+-bound conditions.

Platelet spreading

Washed platelets were prepared as described above and resuspended at 2×107 cells/ml in Tyrode's/Hepes buffer. Platelets were allowed to spread over glass coverslips coated with either 100 μg/ml CRP or 100 μg/ml fibrinogen. Coverslips were washed and cells were fixed in 4% (w/v) paraformaldehyde/PBS for 15 min at room temperature. Cells were washed and permeabilized in 0.2% Triton X-100/PBS. F-actin (filamentous actin) was visualized by staining cells with a phalloidin–Alexa Fluor® 488 conjugate (Molecular Probes, Invitrogen). Images were captured on a Zeiss Axioskop2 microscope in conjunction with AxioVision software (Zeiss). Cell area was determined using the area function in the AxioVision software.

Acid extraction of NAADP

Washed platelets were incubated with agonist for the desired time, after which samples were removed and treated with stop solution (0.75 M HClO4). Samples were sonicated briefly to disrupt the cells and centrifuged at 15115 g for 10 min at 4°C. The supernatant was removed and neutralized with an equal volume of 2 M KHCO3, after which samples were stored at 4°C. NAADP levels were determined using a radioreceptor-binding assay.

NAADP radioreceptor-binding assay

Quantification of NAADP levels in agonist-stimulated platelet samples was determined as described previously [35]. Briefly, 25 μl of NAADP standard/sample was incubated with 125 μl of 1% (v/v) sea urchin egg homogenate, diluted in Glu-IM (gluconate intracellular-like medium) (250 mM N-methylglucamine, 250 mM potassium glucanoate, 20 mM Hepes and 1 mM MgCl2) for 10 min at room temperature. Then, 100 μl of [32P]NAADP, diluted to allow for 0.5 μl/tube in Glu-IM, was added and the reaction mixture was incubated for a further 10 min at room temperature. Bound NAADP was harvested on to filter paper using a Brandel cell harvester. Tubes were washed three times with Hepes wash buffer (20 mM Hepes and 500 mM potassium acetate, pH 7.4). Filter papers were wrapped in clingfilm and placed in a cassette to expose a storage phosphor screen (GE Healthcare) for at least 2 h. The screen was scanned using a Typhoon 9400 scanner at a resolution of 100 μm and the image was analysed using ImageQuant software (GE Healthcare). Standards of known NAADP concentration were used to construct a standard curve, which was then used to determine unknown NAADP concentrations in platelet samples.

Whole-blood assessment of platelet function

To assess whole-blood effects of Ned-19, measurements were carried out using ADP/collagen and ADP/adrenaline cartridges for a PFA-100® instrument (Siemens Healthcare Diagnostics), according to the manufacturer's instructions following 60 s of incubation with 50 μM Ned-19.


Human platelets express an NAADP receptor

In order to establish whether NAADP is able to function as a messenger in human platelets, we investigated the presence of a binding site for NAADP. We used a radioreceptor-binding assay [35] to determine the presence of platelet NAADP receptors. We found that NAADP binds to sonicated platelet homogenate with a Kd of 56 nM (Figure 1A), comparable with previous studies in mammalian tissue [16]. This strongly suggests the presence of an NAADP receptor in platelets. Ned-19 has been shown to be a selective antagonist in sea urchin egg homogenates [25] and was recently reported to inhibit thrombin-mediated Ca2+ release in mouse platelets [26]. We carried out competitive binding assays with Ned-19 and NAADP in platelet homogenates to determine whether Ned-19 was able to compete for the NAADPbinding site and found that Ned-19 competed for the NAADP-binding site with an IC50 of 100 μM (Figure 1A). This value is higher than that reported for the sea urchin, but consistent with results in mouse pancreatic β-cells [25], verifying further the mechanism of action for this antagonist as well as validating its use as an NAADP receptor antagonist in these studies.

Figure 1 NAADP-induces Ca2+ release in human platelets

(A) Displacement curve demonstrating that NAADP binds to an endogenous receptor in human platelets with a Kd of 56 nM. Ned-19 displaces NAADP from its receptor with an IC50 of 100 μM. (B) NAADP releases Ca2+ from an intracellular store in human platelets with an EC50 of 48 nM.

NAADP releases Ca2+ from an acidic store in human platelets in response to CRP

In order to ascertain whether NAADP is able to release Ca2+ from a store in human platelets, permeabilized platelets were loaded with 45Ca2+, and the degree of Ca2+ release in response to various concentrations of NAADP was determined. NAADP released Ca2+ with an EC50 of 48 nM (Figure 1B), consistent with previous studies on rabbit heart microsomes [13] and with the binding affinity (Figure 1A). Previous studies have shown that thrombin-mediated activation of PAR-4 and the GPIb–IX–V complex releases Ca2+ from an acidic store [19,20,22]. To assess whether CRP was also reliant on Ca2+ release from the acidic store, we performed aggregations in the presence of the SERCA3 inhibitor TBHQ and the proton carrier nigericin. In accordance with previous studies [20], we observed a significant inhibition of thrombin-induced Ca2+ release at both 90 s (TBHQ n=6, P=0.031; nigericin n=6, P=0.008) and 5 min (TBHQ n=6, P=0.035; nigericin n=6, P=0.025) after the addition of agonist (Figures 2A and 2B). Inhibition of CRP-induced platelet aggregation in the presence of both TBHQ (P=0.004, n=6) and nigericin (P<0.001, n=6) was observed at the 5 min time point, whereas only TBHQ significantly reduced CRP-induced aggregation at 90 s (P=0.013, n=6) (Figures 2C and 2D). These results suggest that CRP relies, in part, on the acidic store to induce platelet aggregation. These data also support previous observations that thrombin acts on the acidic store to induce Ca2+ release.

Figure 2 NAADP-mediated Ca2+ release in human platelets

Washed platelets were incubated in the presence of store inhibitors for 15 min before assessment of aggregation by light-transmission aggregometry in response to 250 ng/ml CRP or 0.1 unit/ml thrombin. Disruption of the acidic store by both TBHQ (20 μM) and nigericin (10 μM) resulted in reduced aggregation at both 90 s (A and C) and 5 min (B and D). Results are means±S.E.M. (n=6). *P<0.05, Student's t test.

The selective NAADP receptor antagonist Ned-19 inhibits Ca2+ release in human platelets. The radioreceptor-binding assay demonstrates the presence of an NAADP-binding site in platelets, and the 45Ca2+ assay demonstrates the ability of this messenger to release Ca2+ from an intracellular store. Both thrombin and CRP release Ca2+ from an acidic store. To assess the contribution of NAADP-mediated Ca2+ release to platelet function, we used the NAADP receptor antagonist Ned-19. Washed platelets, in the presence of 1 mM EGTA, were incubated with Ned-19 for 30 s before the addition of either 0.1 unit/ml thrombin (Figure 3A) or 250 ng/ml CRP (Figure 3B). Ned-19 inhibited agonist-induced Ca2+ release in a dose-dependent manner, showing greatest inhibition at 100 μM (Figures 3C and 3D). This is in agreement with previous reports of Ned-19 activity, showing a maximal inhibitory effect against NAADP-mediated Ca2+ release at 100 μM in sea urchin egg homogenates [25] and insulin-induced ERK (extracellular-signal-regulated kinase) phosphorylation in the mouse pancreatic β-cell cell line MIN6 [32]. Recent studies have also shown that Ned-19 reduces Ca2+ release in response to thrombin in mouse platelets [26]. However, the authors reported inhibition with 1 mM Ned-19 following 25 min of pre-incubation, suggesting that murine platelets may be less sensitive to NAADP-mediated effects. To rule out cytotoxicity as an explanation for the effects of Ned-19 on platelet function, we assessed cell viability at a range of concentrations (1–100 μM) and found no cytotoxic effects (results not shown).

Figure 3 Ned-19 inhibits Ca2+ release in human platelets

Representative traces of Ca2+ release in response to 0.1 unit/ml thrombin (A) and 250 ng/ml CRP (B). Ned-19 inhibits the Ca2+ release in a dose-dependent manner. (C) Summary of Ned-19-mediated inhibition of Ca2+ release by 0.1 unit/ml thrombin. (D) Summary of Ned-19-mediated inhibition of Ca2+ release by 250 ng/ml CRP. Results are means±S.E.M. (n=4). *P<0.05, Student's t test.

Ned-19 inhibits thrombin- and CRP-induced aggregation

On the basis of the observation that Ned-19 inhibits Ca2+ release induced by both CRP and thrombin, we next assessed the inhibitory effects of Ned-19 with regard to gross platelet function. Using light-transmission aggregometry [3739], we found that pre-incubation of washed platelets with Ned-19 led to a decrease in their ability to aggregate in response to agonist (Figure 4). Thrombin-induced platelet aggregation (0.1 unit/ml) was inhibited by 30% when cells were pre-treated with 100 μM Ned-19 (Figures 4A and 4C). A similar inhibition of aggregation was observed using 250 ng/ml CRP (Figures 4B and 4D).

Figure 4 Ned-19 can inhibit thrombin- and CRP-mediated aggregation

(A) Representative trace showing the inhibition of thrombin (0.1 unit/ml) -mediated aggregation by 100 μM Ned-19. (B) Representative trace showing the inhibition of CRP (250 ng/ml) -mediated aggregation by 100 μM Ned-19. (C) Summary of Ned-19-induced inhibition platelet aggregation in response to 0.1 unit/ml thrombin. (D) Summary of Ned-19-induced inhibition platelet aggregation in response to 250 ng/ml CRP. Results are means±S.E.M. (n=3–4). *P<0.05, Student's t test.

Ned-19 inhibits α-granule secretion

To assess further the inhibitory effects of Ned-19 on platelet function, we measured externalization of P-selectin, a marker for α-granule release, by flow cytometry. Cells were incubated with Ned-19 and anti-P-selectin antibody before the addition of agonist. Ned-19 had no effect on P-selectin exposure in resting platelets. Platelets showed a reduction in cell-surface presentation of P-selectin in response to both thrombin (0.1 unit/ml) and CRP (250 ng/ml) when pre-treated with Ned-19 (Figure 5). Externalization of P-selectin induced by both thrombin (P=0.018) and CRP (P=0.087) was reduced by ~75% (71.8 and 78.7% respectively); however, the decrease in P-selectin exposure with CRP was not significant.

Figure 5 Ned-19 inhibits α-granule release

P-selectin externalization by Ned-19 on the platelet surface was measured by flow cytometry. Washed platelets were incubated with either 100 μM Ned-19 or DMSO control in the presence of PE-conjugated anti-P-selectin antibody, before the addition of 0.5 unit/ml thrombin or 1 μg/ml CRP. Results are mean±S.E.M. fluorescence intensity (n=3). *P<0.05, Student's t test. a.u., arbitrary units.

Ned-19 reduces platelet spreading

Functional assessment of platelet function revealed an inhibitory effect of Ned-19 on both aggregation and secretion. We next assessed cell spreading over agonist-coated surfaces. Platelets were allowed to adhere and spread over CRP- or fibrinogen-coated coverslips for 45 min. Cells were fixed, stained for F-actin with phalloidin–Alexa Fluor® 488, imaged and assessed for cell size (Figure 6). We observed reduced platelet spreading over both CRP (Figures 6A and 6B) (58.7±5.6% reduction, n=3, mean±S.E.M.) and fibrinogen (Figures 6A and 6C) (47.3±3.2% reduction, n=3, mean±S.E.M.) at 100 μM Ned-19.

Figure 6 Ned-19 inhibits cell spreading

Washed human platelets were allowed to adhere to CRP- or fibrinogen-coated surfaces. Cells were fixed and stained for F-actin using phalloidin–Alexa Fluor® 488. (A) Representative images of cell spreading. A reduction in cell size is observed when cells are pre-incubated with 100 μM Ned-19 before spreading over both CRP (B) and fibrinogen (C). Results are means±S.E.M. (n=3). *P<0.05, Student's t test.

CRP increases NAADP levels in human platelets

NAADP receptor antagonism using Ned-19 inhibits various aspects of platelet function induced by both thrombin and CRP, pointing to a role for NAADP as a second messenger for cell activation by these agonists. It has been reported previously that thrombin can increase NAADP levels in human platelets [26], with levels peaking at ~75 s after addition of agonist. We investigated whether CRP was also able to increase levels of NAADP in washed human platelets. As shown in Figure 7, addition of CRP increased NAADP levels above control samples (Figures 7A and 7B) with levels reaching three times that of the resting cell (Figure 7C) and peaking at ~60 s. These results strongly indicate that GPVI signalling leads to generation of NAADP.

Figure 7 CRP-induced NAADP levels in human platelets

NAADP levels were measured in washed human platelets treated with either Tyrode's/Hepes buffer (A) or 1 μg/ml CRP (B) over time (n=3). (C) CRP treatment significantly increased NAADP levels (results are for all points after zero time). Results are means±S.E.M. *P<0.05, Student's t test.

Ned-19 inhibits whole-blood aggregation

The results shown above demonstrate a role for NAADP in platelet activation. To determine whether it may also play a role in thrombus formation, the effect of Ned-19 on platelet function in whole blood was assessed using the PFA-100® system to measure time to occlusion. Assessment of whole-blood aggregation revealed that 50 μM Ned-19 increased the time to occlusion in collagen/ADP cartridges from 81±4.8 s (DMSO control) to 98±11.2 s (n=5, mean±S.E.M., P<0.05), suggesting that Ned-19 can be inhibitory in the presence of factors that activate the coagulation cascade. These findings show that the development of selective NAADP receptor antagonists may provide effective anti-thrombotic agents for in vivo use.


Ca2+ release is vital for human platelet activation, secretion and aggregation. Traditionally, Ca2+ release in human platelets has been mainly attributed to the generation of Ins(1,4,5)P3 from the membrane by PLC [22,40,41], leading to Ca2+ release from the DTS. The present study, in addition to previous work [18,19,22], shows that human platelets are able to mobilize Ca2+ from an acidic store [18,19,22]. Full irreversible human platelet activation is marked by granule secretion, secondary aggregation [42] and inside-out signalling, leading to an increase in the affinity of integrin αIIbβ3 for its ligand [41], a process that is partially dependent on Ins(1,4,5)P3-mediated Ca2+ release via the activation of PLCγ2 and PLCβ [40,43]. Shattil and Brass [44] reported that thrombin-induced inside-out signalling could be attributed, in part, to both Ins(1,4,5)P3 and arachidonic acid pathways, but also that there was “an additional pathway activated by high concentrations of thrombin” (0.1 unit/ml). One might speculate that this additional pathway is, in fact, NAADP-mediated Ca2+ release from the acidic store, contributing to an agonist-induced increase in intracellular Ca2+, promotion of integrin αIIbβ3 activation and progression to full irreversible platelet activation.

It has been shown previously that thrombin relies, in part, on NAADP to mediate platelet activation. In the present study, we provide the first demonstration that human platelet activation by CRP also involves NAADP-mediated Ca2+ release and that this is also mediated, in part, by the release of Ca2+ from an acidic store. Ned-19 significantly inhibits α-granule release and aggregation by thrombin as well as inhibiting CRP-induced aggregation and cell spreading in washed platelets. Ned-19 also has inhibitory effects in whole human blood, inhibiting clot retraction and reducing occlusion time under physiological shear rates. It is interesting that NAADP appears to be involved in Ca2+ release by two very different pathways, one being GPCR-mediated signalling, and the other a tyrosine kinase-dependent pathway [45]. Both of these pathways lead to Ca2+ release via the generation of Ins(1,4,5)P3 by PLC isoforms. However, the findings of the present study indicate that there is an additional component to the Ca2+ response in both pathways: that mediated by NAADP. This may appear at odds with the near complete inhibition of both ADP- and vasopressin-induced Ca2+ release by U73122 [19], a PLC inhibitor, and that observed with the PKC (protein kinase C) inhibitor Ro31-8220. A possible explanation for these observations is that these inhibitors are having off-target effects. U73122 has been reported to inhibit ADP-ribosyl cyclase [46], the proposed synthetic enzyme for NAADP, whereas Ro31-8220 targets GRK-5 (GPCR kinase 5), which has been suggested to have a role in TxA2 (thromboxane A2) receptor signalling in platelets [47]. Another possibility is that the mechanism for NAADP synthesis is selectively coupled to thrombin and CRP signalling, but not to that of ADP or vasopressin, as thrombin can still evoke a diminished Ca2+ response in the presence of U73122.

The present paper is the first to report Ned-19 competing with NAADP binding in a mammalian system. We report a Kd of 100 μM, which is higher than that reported in the sea urchin, although it is consistent with the IC50 reported in a mammalian system [41,42,44]. Ned-19 has been shown to inhibit NAADP-induced Ca2+ signals in the sea urchin and in mammalian systems with IC50 values of 2 and 32 μM respectively. It has also been shown to compete for NAADP binding in the sea urchin with a Kd of 6 μM [25]. There are a number of potential reasons for these differences. NAADP binding in the sea urchin is relatively well characterized. The binding observed is at a high-affinity site, although there is also believed to be a second lower-affinity site [4]. In fact, different analogues of Ned-19 are able to distinguish between the different sites in this system [48]. In mammalian systems, the binding of NAADP is less well characterized. In the present study, we observe only a single site for NAADP binding, although others report the presence of two sites [9,13,49]. In this system, Ned-19 may compete predominantly at a binding site that we have not revealed in the assay. Alternatively, one diastereomer in the mixture employed in the present study may be able to compete at a higher affinity, but this effect is not revealed in the presence of the other diastereomers.

The results of the present study show a functional role for NAADP in human platelets for the first time. The use of the recently discovered selective NAADP antagonist, Ned-19, allowed detailed functional studies that, until recently, would not have been possible. The results of the present study demonstrate a role for NAADP in human platelet activation in response to two important agonists: thrombin and CRP. These findings therefore not only demonstrate a physiological role for NAADP in human haemostasis, but also provide a potential therapeutic target for cardiovascular disease.


Carmen Coxon and Alexander Lewis carried out experiments, performed data analysis and wrote the paper. Amanda Sadler, Sridhar Vasudevan, Lewis Taylor, Andrew Thomas and Kristen Dundas conducted experiments. Duncan Campbell, Grant Churchill and Jonathan Gibbins provided resources and helped with revision of the paper. Katherine Tucker performed experiments and helped with the paper. All authors read and approved the final paper before submission.


This work was funded by the Medical Research Council [grant number G0600681 (to C.H.C., A.J.S. and R.D.C.)], the Wellcome Trust [grant number 082338 (to K.L.T.)] and the Biotechnology and Biological Sciences Research Council [grant number BB/G008523/1 (to A.M.L.)].

Abbreviations: cADPR, cADP-ribose; CRP, collagen-related peptide; DTS, dense tubular system; F-actin, filamentous actin; Glu-IM, glucanoate intracellular-like medium; GP, glycoprotein; GPCR, G-protein-coupled receptor; NAADP, nicotinic acid–adenine dinucleotide phosphate; PAR, protease-activated receptor; PE, phycoerythrin; PLC, phospholipase C; SERCA, sarcoplasmic/endoplasmic reticulum Ca2+-ATPase; TBHQ, 2,5-di-(t-butyl)-1,4-hydroquinone


View Abstract