Involvement of the glycoproteic Ib-V-IX complex in nickel-induced platelet activation.

We studied the effect of nickel ions on platelet function because hypernickelemia has been found in patients with acute myocardial infarction. We previously demonstrated that nickel can activate an intracellular pathway leading to cytoskeleton reorganization consequent to tyrosine phosphorylation of p60(src) in human platelets independently of integrin alpha-IIb-beta(3). Moreover, in von Willebrand factor-stimulated platelets, the tyrosine phosphorylation of pp60(c-src) is closely associated with the activation of phosphatidylinositol 3-kinase (PIK), and two adhesion receptors, glycoprotein (Gp)Ib and GpIIb/IIIa(alpha-IIb-beta(3)), are involved. In our study, 1 and 5 mM nickel in the presence of fibrinogen induced platelet aggregation (independently of protein kinase C activation) and secretion. The pretreatment with a PIK inhibitor, wortmannin, strongly decreased nickel-induced platelet aggregation. Platelet treatment with mocarhagin, a cobra venom metalloproteinase that cleaves GpIba, significantly reduced aggregation induced by 5 mM without affecting the response to other agonists such as adenosine diphosphate (ADP). Moreover, nickel caused PIK translocation to the cytoskeleton. Taken together, these observations suggest a partial involvement of both integrins alpha-IIb-beta(3) and GpIb-V-IX complex in Ni(2+)-induced platelet activation.

The mechanism of action of nickel compounds in the pathogenesis of several diseases linked to occupational exposure has been analyzed, but the toxicity, uptake, and mutagenicity of nickel are not fully understood. Nickel compounds exhibit differential activities at the level of cell surface; consequently, physicochemical surface interactions might contribute to cell injury in nickel-induced cytotoxicity (1). Moreover, nickel can induce in several human and rodent cell lines the expression of the receptor system Cap43, which is involved in a Ca 2+ -dependent process of signal transduction (2)(3)(4). Nickel induces CD69 expression in lymphocyte subpopulations of allergic patients with contact dermatitis (5). CD69 is constitutively expressed on human platelet surface where it can be involved in signal transduction (6). Hypernickelemia has been found in patients with acute myocardial infarction and has been thus related to the pathogenesis of myocardial ischemic injuries (7), in which platelet interactions with the exposed collagen fibers play a major role.
In previous studies we showed that NiCl 2 can enhance platelet aggregation induced by collagen in the presence of fibrinogen via a rapid cytoskeletal reorganization consequent to tyrosine phosphorylation of pp60 src , a signaling molecule that has been detected in a submembraneous location (8). However, nickel-induced platelet activation, expressed as p60 src phosphorylation, occurs even in the absence of fibrinogen, whereas aggregation requires fibrinogen binding to its receptor, integrin αIIbβ3 (8).
The pp60 src protein is not the only protein involved in regulating the formation of platelet cytoskeletal signaling complexes. Calpain, a thiol protease, is responsible for the cleavage of several adhesion structural proteins (talin and integrin αIIbβ3) (9,10) and signaling enzymes [focal adhesion kinase (FAK); phosphatidyl inositol 3-kinase (PIK)] (11,12) in spreading and aggregating platelets.
Von Willebrand factor (vWf)-induced platelet stimulation requires the coactivation of two different glycoproteins, the GpIb-V-IX complex and GpIIb-IIIa, or integrin αIIbβ3, the former being responsible for the initial contact and the latter leading to spreading and irreversible adhesion (13). However, vWf binding to GpIb-V-IX complex induces the activation and cytoskeletal association of pp60 src and PIK (14); this behavior resembles that observed in NiCl 2 -treated platelets.
Our aim was thus to verify whether the GpIb-V-IX complex has a role in NiCl 2induced platelet activation and to investigate the molecules involved in the signal transduction cascade that follows NiCl 2 binding to platelet membrane.
Platelet isolation. We collected human platelets from drug-free, healthy donors in acid/citrate/dextrose (ACD)-containing tubes. Platelet-rich plasma (PRP) was obtained after centrifugation at 180 × g for 15 min and concentrated at 800 × g for 20 min. The platelet pellet was resuspended in one-third volume of autologous plateletpoor plasma (PPP) and incubated with 1 mM aspirin for 15 min at 37°C. The platelets were then washed twice in Tyrode's buffer (137 mM NaCl, 2.68 mM KCl, 0.42 mM NaH 2 PO 4 , 1.7 mM MgCl 2 ) containing 10 mM HEPES (pH 6.5) and resuspended in Tyrode's buffer containing 0.2% BSA, 0.1% glucose, and 10 mM HEPES (pH 7.35). The final platelet suspension was adjusted to 2.5 × 10 8 cells/mL. Platelet aggregation. In vitro platelet aggregation was performed in a PACKS-4 aggregometer (Helena Laboratories, Beaumont, TX, USA) using siliconized glass cuvettes at 37°C under continuous stirring. NiCl 2 (1 mM and 5 mM), ADP (5 µM), and the thromboxane A2 analog U46619 (1 µM) were used as platelet agonists. Fibrinogen (1 mg/mL) was added before the agonists. ATP release. Platelet activation was stopped after 2 min by adding formaldehyde/EDTA according to Costa and Murphy (15). After centrifugation at 10,000 × g for 30 sec, we measured the ATP concentration in the supernatant in an LKB 1251 luminometer (LKB, Turku, Finland) after adding luciferin (40 mg/mL) and luciferase (880 U/mL). The results were expressed as the percentage of ATP released relative to the total ATP present in cells lysed by means of digitonin (50 µM) (16).
When washed platelets were treated with a calpain selective inhibitor, calpeptin (50 µg/mL), platelet aggregation in response to the thromboxane synthetic agonist U46619 (1 µM) was completely abolished (Figure 4, trace a), whereas platelet response to NiCl 2 was unmodified ( Figure 4, trace b). Figure 5 shows that when washed platelets were treated with mocarhagin (10 µg/mL), a cobra venom metalloprotease that cleaves GpIbα, the platelet response to the specific agonist of GpIbα (Ristocetin) was completely inhibited (data not shown), whereas platelet aggregation induced by 5 mM NiCl 2 in the presence of fibrinogen was significantly reduced (37.8 ± 18.6 vs. 78 ± 8.7 maximum % of aggregation), and the response to ADP, whose action is not exerted through GpIb-V-IX, was unaffected (70.8 ± 12.3 vs. 71.4 ± 10.6 maximum % of aggregation; Figure 5).
When washed platelets were treated with a PIK inhibitor, wortmannin (10 µM), platelet aggregation induced by 1 mM NiCl 2 was completely abolished, and the aggregation induced by 5 mM NiCl 2 was strongly decreased (36.6 ± 15.9 maximum % of aggregation; Figure 6). Figure 7 shows the immunoblot analysis of the low-speed Triton insoluble cytoskeletal fraction of nickel-stimulated platelets using a polyclonal antibody directed against the p85 subunit of PIK, closely correlated with PtdIns 3-kinase enzymatic activity in platelets (18). The figure shows that a faint band, corresponding to p-85, already detectable after 3 min NiCl 2 stimulation, became marked after 5 min stimulation of aspirinated, RGDS-treated platelets.

Discussion
In the present study we provided evidence that nickel can cause platelet aggregation and ATP release from the internal stores. NiCl 2 exerts these actions without entering the cell; in fact, in unpublished observations, we found that the addition of NiCl 2 to Fura-2-treated platelets did not quench Fura-2 fluorescence, but an evident reduction was obtained after cellular lysis. This result was obtained comparing the reduction of Fura-2 fluorescence induced by NiCl 2 to the capability of Mn 2+ to totally quench the Fura-2 fluorescence after the addition of digitonin (50 µM), according to the method described by Sage (19).   Moreover, although fibrinogen is essential for NiCl 2 -induced platelet aggregation, it is not required for NiCl 2 -induced platelet activation, expressed as p60 src phosphorylation, as demonstrated by the fact that such response is not inhibited by the incubation with the tetrapeptide RGDS (120 µg/mL), which prevents fibrinogen binding to its receptor, the integrin αIIbβ3 (8).
Many authors suggested that fibrinogen receptor exposure involves protein kinase C (20). However, NiCl 2 -induced platelet aggregation does not seem to depend on protein kinase C activation, as demonstrated by the observation that the treatment with a protein kinase C inhibitor, Ro 31-8220, only slightly modified the aggregometrical response to NiCl 2 .
In a previous study (8), we demonstrated that NiCl 2 induces a rapid cytoskeletal reorganization consequent to tyrosine phosphorylation of the signaling molecule p60 src . Among the key proteins regulating the formation of platelet cytoskeletal signaling process, calpain is not involved in NiCl 2 -induced triggering of signal transduction. In fact, the treatment with the calpain inhibitor calpeptin did not reduce NiCl 2 -induced response.
Because it has been demonstrated that pp60 src translocation to the cytoskeleton is triggered by vWf binding to GpIb-V-IX complex, independently of ligand binding to αIIbβ3 (14), and that neither vWf nor αIIbβ3 requires calpain activation, it was conceivable that a direct involvement of GpIb-V-IX complex in Ni 2+ evoked platelet activation. To verify this hypothesis, we performed a platelet treatment with the cobra venom metalloprotease mocarhagin, which cleaves GpIba. The observation of a significant reduction of platelet aggregation in response to NiCl 2 led us to hypothesize that NiCl 2 exerts its action mainly through GpIb-V-IX. Moreover, this seems confirmed by the fact that the cytoskeletal translocation of activated pp60 src , similar to what was observed after vWf stimulation (14), was demonstrated in NiCl 2 -treated platelets (8). Because the association of p60 src and PIK to cytoskeleton occurs once platelets aggregate (14), we tested whether the activation of PIK was required in NiCl 2 -induced response. The finding that the treatment with PIK inhibitor wortmannin strongly decreased platelet aggregation in response to NiCl 2 indicated a role for PIK in NiCl 2 -induced response.
PtdIns 3-kinase is known to translocate from the cytosol to the membrane cytoskeleton in the absence of fibrinogen only after vWf stimulation of platelets, with the ensuing activation of GpIb-V-IX complex, whereas stimulation with agonists such as ADP, epinephrine, or collagen has no such effect (14). In fact, vWf-induced cytoskeletal association of PIK and pp60 src occurs despite treatment with RGDS or EDTA, which disrupts the ligand-binding capacity of αIIbβ3 but does not affect the ability of vWf to bind GpIb-V-IX (21,22). Moreover, this last binding is specifically blocked by an anti-GpIb monoclonal antibody, and it is not observed in platelets lacking the glycoprotein Ib/IX complex (Bernard Soulier syndrome) (14). To verify whether NiCl 2 -induced platelet activation proceeds through its binding to platelet GpIb-V-IX, we studied the cytoskeletal localization of PIK after NiCl 2 addition in the absence of exogenous fibrinogen. The results demonstrated that nickel was able to cause PIK translocation in a time-dependent manner.
The fact that RGDS had no inhibitory effect on NiCl 2 -induced cytoskeletal association of PtdIns 3-kinase strongly supports the idea that NiCl 2 acts in a vWf-like manner. Our data confirm that this step, far from being dependent on αIIbβ3 binding to fibrinogen, involves other integrins, possibly GpIb.
Taken together, these observations suggest the existence of two phases in platelet response to NiCl 2 stimulation: a first step represented by the initial contact and binding to GpIb-V-IX complex, associated with cytoskeletal reorganization and translocation of PIK, and a second leading to the conversion of GpIIb-IIIa to an activated state necessary to support platelet aggregation. To the best of our knowledge, this is the first evidence of a direct role of NiCl 2 in inducing receptor activation in platelets.