Delineating Zinc Influx Mechanisms during Platelet Activation

Abstract

Zinc (Zn²⁺) is released by platelets during a hemostatic response to injury. Extracellular zinc ([Zn²⁺]_o) initiates platelet activation following influx into the platelet cytosol. However, the mechanisms that permit Zn²⁺ influx are unknown. Fluctuations in intracellular zinc ([Zn²⁺]_i) were measured in fluozin-3-loaded platelets using fluorometry and flow cytometry. Platelet activation was assessed using light transmission aggregometry. The detection of phosphoproteins was performed by Western blotting. [Zn²⁺]_o influx and subsequent platelet activation were abrogated by blocking the sodium/calcium exchanged, TRP channels, and ZIP7. Cation store depletion regulated Zn²⁺ influx. [Zn²⁺]_o stimulation resulted in the phosphorylation of PKC substates, MLC, and β3 integrin. Platelet activation via GPVI or Zn²⁺ resulted in ZIP7 phosphorylation in a casein kinase 2-dependent manner and initiated elevations of [Zn²⁺]_i that were sensitive to the inhibition of Orai1, ZIP7, or IP₃R-mediated pathways. These data indicate that platelets detect and respond to changes in [Zn²⁺]_o via influx into the cytosol through TRP channels and the NCX exchanger. Platelet activation results in the externalization of ZIP7, which further regulates Zn²⁺ influx. Increases in [Zn²⁺]_i contribute to the activation of cation-dependent enzymes. Sensitivity of Zn²⁺ influx to thapsigargin indicates a store-operated pathway that we term store-operated Zn²⁺ entry (SOZE). These mechanisms may affect platelet behavior during thrombosis and hemostasis.

1. Introduction

Dietary zinc (Zn²⁺) deficiency results in prolonged bleeding phenotypes in both rodents [1,2,3] and humans [4,5,6] that can be reversed by dietary supplementation. Plasma Zn²⁺ concentrations range from 10 to 20 µM; however, binding to plasma proteins (i.e., albumin and α₂ microglobulin) reduces the free (labile) concentration to around 0.5 µM [7,8,9]. Zn²⁺ concentrations are elevated within atherosclerotic plaques and at sites of vascular injury as a result of a release from platelet granules, and are likely to be much higher than the average plasma concentration [10,11].

At concentrations of 100 μM, Zn²⁺ acts as a platelet agonist, regulating intracellular signaling resulting in granule secretion, the tyrosine phosphorylation of protein kinase C (PKC), integrin α_IIbβ₃ activation, and platelet aggregation [12,13,14]. At low concentrations, Zn²⁺ is a coactivator, potentiating platelet activation to threshold concentrations of conventional agonists [12,15]. Previous research has demonstrated changes in platelet intracellular Zn²⁺ concentrations [Zn²⁺]_i following agonist stimulation, indicating a release from intracellular stores, which is consistent the a role of Zn²⁺ as an intracellular secondary messenger [14]. Agonist-evoked increases in [Zn²⁺]_i in the absence of [Zn²⁺]_o result in platelet activation, shape change, degranulation, and phosphatidyl-serine exposure. Similarly, Zn²⁺ regulates Ca²⁺ release from stores and may therefore be a central factor in regulating platelet activation [16,17].

In nucleated cells, Zn²⁺ permeability across the cell membrane is regulated by Zn²⁺ transporting proteins, including non-selective cation channels, transporters, and exchangers [18,19,20]. Cellular Zn²⁺ homeostasis is regulated by Zn²⁺ transporters (ZnTs) and Zrt-Irt-like proteins (ZIPs), of which transcripts are detected in megakaryocytes [20,21,22]. Although the Na²⁺/Ca²⁺ exchanger (NCX), operating in reverse mode, and transient receptor potential (TRP) channels have been reported to facilitate Zn²⁺ movement in nucleated cells, their role in platelet Zn²⁺ homeostasis has yet to be investigated [23,24]. [Zn²⁺]_i buffering systems are important mechanisms for regulating Zn²⁺ bioavailability [25]. In platelets, agonist-evoked increases in [Zn²⁺]_i are regulated by changes in the platelet redox state, suggestive of a role for redox-sensitive proteins, such as metallothioneins [26].

Intracellular platelet calcium concentrations ([Ca²⁺]_i) are maintained by store-operated Ca²⁺ entry (SOCE) and receptor-operated Ca²⁺ entry (ROCE) pathways, where the dense tubular system (DTS) is the primary Ca²⁺ store [27,28,29]. SOCE is induced via the activation of cation-selective CRAC channels [30,31,32,33]. The resting platelet Ca²⁺ concentration ([Ca²⁺]_rest) is maintained at approximately 100 nM by the action of channels and exchangers expressed along the plasma membrane and the surface of the DTS [34,35]. Interestingly, ryanodine receptors, expressed along the cardiomyocyte sarcoplasmic reticulum, have been shown to be regulated by the elevation of [Zn²⁺]_i; although, it is not clear whether the equivalent IP₃R in platelets is modulated in a similar manner [36,37].

We have previously shown that the exposure of platelets to [Zn²⁺]_o leads to its rapid and sustained accumulation within the platelet cytosol [14,26]. In this study, we investigate the mechanisms underpinning platelet Zn²⁺ influx, and the storage mechanisms that contribute to Zn²⁺-induced platelet activation (ZIPA).

2. Results

2.1. TRP Channels and Reverse-Mode NCX Regulate Zn²⁺ Influx during Zn²⁺-Induced Platelet Activation

Zn²⁺ acts in a dual manner to regulate platelet activation. When applied extracellularly, Zn²⁺ gains access to the platelet cytosol, whilst, in the absence of extracellular Zn²⁺, platelet activation via canonical receptors, such as GPVI, results in increases in intracellular Zn²⁺ as a result of a release from stores [12,14,26]. Increases in [Zn²⁺]_i via either pathway results in ZIPA, which is sensitive to a range of inhibitors and metal ion chelators [12]. However, the molecular identity and the contribution of cation channels underlying Zn²⁺ influx in platelets have not been investigated. TRP proteins are non-selective cation channel of which, TRPC6, TRPC1, TRPV1, and TRPC5 are expressed in platelets or megakaryocytes, and have been reported to facilitate cellular Zn²⁺ uptake in nucleated cells [24,38,39,40]. Additionally, the Na⁺/Ca²⁺ exchanger (NCX) substitutes Ca²⁺ for Zn²⁺ in order to mediate Zn²⁺ influx in intestinal epithelial cells [18,41].

We evaluated the roles of TRP channels and NCX during ZIPA. Platelets were stimulated by exposure to 100 µM ZnSO₄ in the presence of increasing concentrations of TRP channel blockers 2-APB (Figure 1a), FFA (Figure 1b), and SKF (Figure 1d), or KB-R7943 mesylate, which is a reverse-mode NCX inhibitor and has also been shown to be a potent blocker of TRP channels (KB-R, Figure 1c) [42]. ZIPA was reduced in a concentration-dependent manner following treatment with each inhibitor. Calculated IC₅₀ values (Figure 1e,

Table 1. IC₅₀ values for the inhibition of 100 µM Zn²⁺-induced platelet aggregation.

Drug	Target	IC50
2-APB	TRP channels/SOCE	34.5 ± 3.0 µM
FFA	TRP channels	76.7 ± 9.8 µM
KB-R	Reverse-mode NCX	28.9 ± 1.3 µM
SKF	TRP channels	23.9 ± 2.5 µM

) were comparable for 2-APB, KB-R, and SKF, whilst the curve for FFA was right-shifted. Platelet aggregation responses to 100 µM Zn²⁺ were reduced from 73.9 ± 12.1% in the presence of KB-R alone, and to 10.4 ± 5.2% by the combined blocking of KB-R and 2-APB (Figure 1f, p < 0.01). Residual ZIPA responses suggested that multiple Zn²⁺-permeable channels, and potentially Zn²⁺ transporters, contributed to platelet responses to Zn²⁺.

The stimulation of platelets with [Zn²⁺]_o leads to a concentration-dependent increase in [Zn²⁺]_i; however, the nature of the Zn²⁺ influx pathway or the relative contribution of the release of [Zn²⁺]_i stores remains unknown [12]. We assessed the role of NCX and TRP channels in platelet Zn²⁺ influx using FZ-3-loaded platelets. Platelets were stimulated by 100 µM Zn²⁺ following treatment with inhibitors, KB-R or SN-6, or 2-APB and the resultant Zn²⁺ signals were quantified using flow cytometry (Figure 2a). An inhibition of Zn²⁺ influx was observed upon pre-incubation with 30 µM KB-R (to 2611 ± 516 a.u.) or 0.5 µM SN-6, (to 2537 ± 204 a.u.) compared with 4882 ± 531 a.u. for the vehicle control, representing decreases of 53.4 ± 3.2% and 51.9 ± 1.8%, respectively (Figure 2a, p < 0.01). The pre-incubation of FZ-3-loaded platelets with 2-APB reduced the peak Zn²⁺ influx to 2035 ± 322 a.u. (41.7 ± 3.6% of vehicle control, Figure 2a, p < 0.001), and to 2662 + 420 a.u. following co-treatment with 30 µM KB-R and 2-APB, as observed in the platelet aggregation (54.5 ± 3.6% of vehicle control, Figure 2a, p < 0.001). These data are suggestive of a constitutively active Zn²⁺ entry pathway that allows platelets to sense changes in local Zn²⁺ concentrations.

As KB-R is an inhibitor of both NCX and TRP channels, we further investigated the roles of specific TRP channels in Zn²⁺ influx [42]. A range of TRP inhibitors were screened for their influence on the stimulation of FZ-3-loaded platelet suspensions with 100 μM Zn²⁺. Peak Zn²⁺ influx was reduced from 2612 ± 198 a.u (for the vehicle control) to 643 ± 103.9 a.u. in platelets treated with the TRPC6 inhibitor SAR7334 (Figure 2b, p < 0.001). Similarly, treatment with TRPC5 or TRPV1 blockers (AC1903 and AMG9810, respectively) reduced Zn²⁺ influx to 711 ± 136 and 1328 ± 173 a.u., respectively (Figure 2b, p < 0.001). The inhibition of TRPs that were not expressed on platelets (TRPM4 and TRPA1) [43,44] with AM0902 or 9-Phenanthrol, respectively, had no effect on Zn²⁺ influx (mean fluorescence intensities were 2462 ± 248 and 2766 ± 439 a.u. respectively; Figure 2b, ns). These data indicate that a component of observed [Zn²⁺]_o influx was insensitive to NCX and TRP blockers, suggesting that other Zn²⁺ entry pathways may also contribute to Zn²⁺ influx. Such pathways can include Zn²⁺ transporters or the release of Zn²⁺ into the cytosol from internal stores.

Further experiments were performed to further confirm the role of TRP channels in [Zn²⁺]_o influx. OAG and the PKC inhibitor, GO6983, have previously been used to potentiate TRPC6 activity in HEK293 and smooth muscle cells [45,46]. FZ-3-loaded platelets were pre-treated with OAG, GO6983, or a combination of the two prior to stimulation by Zn²⁺. Zn²⁺ influx was increased in dual-treated platelets (F/F₀ for dual-treated platelets was 9.3 ± 0.2 a.u., compared to untreated platelets: 5.9 ± 0.4 a.u.; Figure 2c,d, p < 0.01.), supporting the evidence that TRP channels are a route for Zn²⁺ influx.

2.2. Zn²⁺-Induced Platelet Activation Is Associated with the Phosphorylation of Substrates of Ca²⁺-Dependent Enzymes

Whilst ZIPA produces a distinct pattern of tyrosine phosphorylation, the effect on other phosphoproteins has not been studied [14]. It has been reported that Zn²⁺ can substitute for Ca²⁺ at binding sites on calmodulin and other Ca²⁺-dependent enzymes [47,48]. Therefore, we investigated the association of ZIPA with the phosphorylation and activation of key activatory proteins, including the Ca²⁺-dependent myosin light chain kinase (MLCK) and PKC. Previous work using PKC inhibitors highlighted a role for MLCK and PKC during ZIPA [14], with Zn²⁺ possibly substituting for Ca²⁺ at the active site of these enzymes, leading to their activation [17]. Platelet lysates were prepared at 0, 2, 5, and 10 min intervals following stimulation by activatory (1 mM) or sub-activatory (30 µM) concentrations of Zn²⁺ (Figure 3). Increases in the protein phosphorylation of PKC and MLCK substrates alongside the phosphorylation of the integrin β3 chain occurred in a temporal manner and peaked within 5 min of stimulation with 1 mM Zn²⁺ (Figure 3), consistent with the time course for ZIPA. The phosphorylation of these substrates was observed in platelet populations treated with the integrin α_IIbβ₃. antagonist integrilin, demonstrating that phosphorylation was not attributable to outside–in signaling through α_IIbβ₃.

2.3. Zn²⁺ Influx Is Mediated by Ca²⁺ Store Depletion

Upon receptor-mediated activation, platelets generate IP₃ that activates IP₃R leading to the release of Ca²⁺ from intracellular stores into the cytosol. This process is supported by casein kinase 2 (CK2), which has also been shown to be involved in ZIP7 phosphorylation in nucleated cells [49,50]. SOCE or ROCE both act to increase [Ca²⁺]_o uptake, and we hypothesized that these mechanisms might also be responsible for Zn²⁺ influx. Washed platelet suspensions were loaded with FZ-3 or Fluo-4, and thrombin- or thapsigargin (TG)-induced fluctuations of [Ca²⁺]_i or [Zn²⁺]_i were investigated using flow cytometry (Figure 4). As shown previously, thrombin caused a rapid increase in [Ca²⁺]_i, which was enhanced in the presence of 2 mM [Ca²⁺]_o, consistent with SOCE (F/F₀ peaked at 5.3 ± 0.4 a.u. compared to 4.4 ± 0.2 a.u for thrombin alone, Figure 4a,b). In the presence of subactivatory concentrations (30 μM) of [Zn²⁺]_o, thrombin-mediated [Ca²⁺]_i increases were significantly reduced (to 2.7 ± 0.1 a.u., p < 0.001). This effect was partially restored with the addition of 2 mM [Ca²⁺]_o (3.2 ± 0.0 a.u., p < 0.001, Figure 4a,b), indicating that [Zn²⁺]_o has a negative effect on Ca²⁺ influx. Fluctuations in Ca²⁺ were sensitive to pre-treatment with the non-specific cation chelator BAPTA-AM (10 µM), confirming the nature of the [Ca²⁺]_i response. Pre-treatment with the intracellular Zn²⁺ chelator, TPEN (50 µM), had no significant effect on thrombin-induced [Ca²⁺]_i signals, but abolished the reductive effect of [Zn²⁺]_o on [Ca²⁺]_i signals (F/F₀ was 4.0 ± 0.1 a.u compared to 4.4 ± 0.2 a.u; Figure 4a,b, ns), further confirming the role of Zn²⁺ in regulating platelet [Ca²⁺]_i responses.

As 2-APB and KB-R were implicated in Zn²⁺ influx (Figure 1), these pathways were assessed to further explore the mechanisms that contributed to the Zn²⁺ regulation of SOCE. In 2 mM [Ca²⁺]_o, 2-APB caused a significant decrease in thrombin-evoked [Ca²⁺]_i responses (from 5.3 ± 0.4 a.u to 3.0 ± 0.3, p < 0.01; Figure 4a,b). As 2-APB affected both TRP channels and IP₃R, these findings implicate Zn²⁺ in the regulation of cation entry via TRP channels or Ca²⁺ or Zn²⁺ release via DTS. KB-R pre-treatment had no significant effect on thrombin-evoked [Ca²⁺]_i fluctuations, irrespective of the cation conditions tested. Interestingly, in 2-APB-treated platelets, thrombin-evoked Ca²⁺ signals in the presence of [Zn²⁺]_o did not differ to the vehicle controls. Thus, 2-APB-sensitive pathways were not involved in the Zn²⁺-mediated regulation of [Ca²⁺]_i.

To investigate whether [Zn²⁺]_o influenced Ca²⁺ influx following store depletion, Fluo-4-loaded platelets were stimulated by TG in the presence and absence of [Zn²⁺]_o, and [Ca²⁺]_I fluctuations were quantified using flow cytometry. TG (1 µM) stimulation elevated [Ca²⁺]_i to 2.0 ± 0.1 a.u (Figure 4c,d), reflecting Ca²⁺ release from intracellular stores. This was increased significantly in the presence of [Ca²⁺]_o, consistent with SOCE (peak Fluo-4 fluorescence was 4.5 + 0.5 a.u.; Figure 4c,d, p < 0.05). In the presence of [Zn²⁺]_o, Ca²⁺ influx was reduced by 72% (to 1.3 + 0.1; Figure 4c,d, p < 0.05). In the presence of both [Ca²⁺]_o and [Zn²⁺]_o, peak TG-induced [Ca²⁺]_i signals were reduced to 3.0 + 0.3 a.u. (Figure 4c,d, p < 0.01) suggesting a competition between Zn²⁺ and Ca²⁺ for cation entry mechanisms. As expected, BAPTA pre-treatment abolished Ca²⁺ responses to TG treatment, irrespective of extracellular cation status (Figure 4c,d, p < 0.05). The peak signal evoked by 1 μM TG upon TPEN (25 μM) pre-treatment was 1.8 ± 0.1 a.u, compared to the vehicle control (2.0 ± 0.1 a.u; Figure 4c,d, ns) indicating that increases in [Zn²⁺]_i, either as a result of its release from stores or via Zn²⁺ entry, did not affect SOCE. KB-R pre-treatment did not affect Ca²⁺ influx, whilst 2-APB pre-treatment significantly reduced Ca²⁺ influx in the presence of 2 mM [Ca²⁺]_o (peak fluorescence was 3.3 + 0.4 following 2-APB treatment, compared to 4.4 + 0.3 a.u for untreated platelets; Figure 4c,d, p < 0.05). These findings implicate Zn²⁺ in the regulation of cation influx.

We hypothesized that cation channels involved in SOCE could also facilitate [Zn²⁺]_o influx. To test this, platelets were loaded with FZ-3 and stimulated with thrombin or TG, and Zn²⁺ influx was quantified using flow cytometry. In the absence of [Zn²⁺]_o, thrombin stimulation did not affect FZ-3 fluorescence. However, with 30 μM [Zn²⁺]_o, thrombin stimulation led to an increase in FZ-3 fluorescence (to 1.7 ± 0.1 a.u for [Zn²⁺]_o and 1.5 ± 0.1 a.u for [Zn²⁺]_o/[Ca2]_o; Figure 4e,f, p < 0.0001) demonstrating an activation-dependent [Zn²⁺]_o influx. FZ-3 fluorescence was unaffected by [Ca²⁺]_o. Further experiments employing the cation chelators BAPTA and TPEN abrogated increases in FZ-3 fluorescence, confirming that Zn²⁺ was responsible for the FZ-3 signal. The pre-treatment of platelets with KB-R or 2-APB had no significant effect on the [Zn²⁺]_i signal evoked in the presence of [Zn²⁺]_o or [Ca²⁺]_o, or a combination of both cations (Figure 4e,f).

TG was used to deplete internal stores, and Zn²⁺ influx into FZ-3-loaded platelets quantified using flow cytometry. In the absence of [Zn²⁺]_o, TG stimulation did not result in FZ-3 fluorescence in the absence of external cations (Figure 4g,h), and [Ca²⁺]_o did not effect FZ-3 responses (1.2 ± 0.1 a.u; Figure 4g,h, ns). However, peak FZ-3 fluorescence increased in response to TG stimulation. F/F₀ was 3.3 + 0.9 a.u or 2.9 + 0.1 a.u. in the presence of [Zn²⁺]_o or [Zn²⁺]_o/[Ca²⁺]_o, respectively (Figure 4g,h, p < 0.001). TPEN or BAPTA abolished TG-evoked increases in FZ-3 fluorescence (Figure 4g,h). TG-mediated F/F₀ signals did not differ upon pre-treatment with 30 μM KB-R in comparison to the vehicle control (Figure 4e,f, ns). However, in 2-APB-treated platelets, the Zn²⁺ signal was significantly reduced in the presence of [Zn²⁺]_o (2.1 ± 0.1 a.u) compared to the vehicle control (3.3 ± 0.1 a.u; Figure 4g,h, p < 0.0001). This was also the case for [Zn²⁺]_o/Ca²⁺]_o, where the peak signal was 1.9 ± 0.1 a.u (Figure 4g,h; p < 0.001).

These data indicate a mechanism where the depletion of cation stores in platelets following SERCA inhibition results in increased [Zn²⁺]_i as a result of Zn²⁺ influx from the extracellular medium in a similar manner to SOCE. We described this mechanism as store-operated Zn²⁺ entry (SOZE).

2.4. [Zn²⁺]_o Regulates Intracellular Signaling via the Regulation of Zn²⁺ Stores

Zn²⁺ signaling may play a regulatory role in canonical Ca²⁺ signaling responses upon platelet activation. Zn²⁺ regulates Ca²⁺ efflux from platelet stores and it is likely that distinct intracellular Zn²⁺ stores contribute to Ca²⁺-dependent platelet responses [16,17]. In nucleated cells, PLC-dependent generation of IP₃ causes cation store depletion resulting in STIM1 and Orai1 clustering with a resultant extracellular Ca²⁺ influx by SOCE-mediated activation [29,51]. Exogenous Zn²⁺ induces ZIP7-mediated Zn²⁺ influx leading to tyrosine kinase activation in a process that is regulated by CK2 [50]. We investigated the role of the Zn²⁺ transporter, ZIP7, in elevating [Zn²⁺]_i by facilitating Zn²⁺ entry in platelets.

The activation of platelets with CRP-XL (1 µg/mL) or Zn²⁺ (100 µM) led to an increased expression of ZIP7 on the platelet surface demonstrated by flow cytometry (Figure 5a,b; p < 0.001). The stimulation of platelets with CRP-XL, but not the TP agonist U46619, resulted in increases in ZIP7 phosphorylation consistent with the role of GPVI in Zn²⁺ signaling (Figure 5c; p < 0.001) [14]. CRP-XL- and Zn²⁺-mediated ZIP7 phosphorylation was abrogated by pre-treatment with the CK2 inhibitor CX-4945 (Figure 5c), consistent with the roles of CK2 and ZIP7 in Zn²⁺ release from stores. The role of ZIP7 in Zn²⁺ influx was further investigated using the ZIP7 inhibitor NVS-ZP7-4. The stimulation of FZ-3-loaded platelets with 100 µM Zn²⁺ following NVS-ZP7-4 inhibition (5 μM) resulted in reduced FZ-3 fluorescence in a concentration-dependent manner (Figure 5d, p < 0.001). These findings demonstrate the role of ZIP7 in transiently elevating [Zn²⁺]_i in response to extracellular signals, and the role of ZIP7 phosphorylation involving CK2.

The relative contributions of [Zn²⁺]_o influx or [Zn²⁺]_i release from stores to [Zn²⁺]_i signals remains unclear. Therefore, we utilized flow cytometry to investigate the influence of Ca²⁺- or Zn²⁺-release mechanisms on activation-dependent [Zn²⁺]_i signals in the presence and absence of [Zn²⁺]_o, to isolate the contribution of [Zn²⁺]i to intracellular stores.

As expected, a higher FZ-3 signal was observed following [Zn²⁺]_o stimulation compared to CRP-XL (7375.6 ± 1394.2 a.u. and 3962 ± 328.6, respectively; Figure 6), consistent with contributions to [Zn²⁺]_i from both extracellular and intracellular sources, but with a greater relative contribution from [Zn²⁺]_o. The effects of the inhibitions of IP₃R (with Teriflunomide, 100 μM) [39], Orai1 (AnCOA4, 5 μM) [52], or ZIP7 (NVS-ZP7-4, 5 μM) were assessed. The inhibition of IP₃R, Orai1, or ZIP7 reduced Zn²⁺-mediated FZ-3 increases to 2533.4 ± 385.2 a.u., 1549 ± 240.5 a.u., and 1289.4 ± 155.6 a.u. respectively, compared to the vehicle control (7375.6 ± 1394.2 a.u.; Figure 6, p < 0.001). CRP-XL stimulation (100 μg/mL) in the absence of extracellular [Zn²⁺]_o resulted in FZ-3 signals that were insensitive to the inhibition of IP₃R, ZIP7, or Orai1 (Figure 6, ns) indicating that Zn²⁺ was not being released from intracellular stores through IP₃R (as is the case with Ca²⁺ signaling) or ZIP7.

3. Discussion

Here, we demonstrate for the first time that Zn²⁺ influx into platelets and subsequent platelet activation are partially blocked by the conventional TRP channel and NCX blockers and an inhibitor of ZIP-7, indicating multiple pathways for Zn²⁺ entry into platelets.

In nucleated cells, [Zn²⁺]_o uptake and the elevation of [Zn²⁺]_i are mediated by the involvement of Zn²⁺-permeable ion channels, exchangers, and Zn²⁺ transporters, including NCX and TRP channels [20,39,53]. Our data show that KB-R (a reverse-mode NCX inhibitor) caused substantial reductions in ZIPA (Figure 1c), suggesting a distinct role for these exchangers in Zn²⁺ influx into platelets. Reductions in ZIPA were associated with the reduced fluorescence of FZ-3-loaded platelets, consistent with a reduction in reduced [Zn²⁺]_o-induced Zn²⁺ influx (Figure 2a,b). These observations are consistent with a previous study that shows NCX’s role in mediated Zn²⁺ influx in neuronal cells and that Zn²⁺ allosterically inhibits Na²⁺-independent Ca²⁺ release by inhibiting NCX activity [18,41].

Inhibition of both Zn²⁺ influx and ZIPA were observed following pre-treatment with 2-APB, SKF, or FFA, which suggests the involvement of TRP channels (Figure 2a,b,d). The lack of the full inhibition of Zn²⁺ influx as a result of NCX or TRP inhibitions indicates the existence of alternative Zn²⁺ entry pathways. In addition to being a TRP channel inhibitor, 2-APB also inhibits IP₃R, and therefore could potentially have inhibited [Ca²⁺]_i release from IP₃-sensitive stores. Among the TRP channels, TRPC1, TRPC3, TRPC4, TRPC5, TRPC6, and TRPV1 are expressed on the platelets [54,55]. Specific inhibition identified roles for TRPC6, TRPC5, and TRPV1 in mediating Zn²⁺ influx (Figure 2b), which is consistent with previous studies that show TRPC6 to be involved in Zn²⁺ influx in nucleated cells [24,56]. Further evidence for the involvement of TRP channels comes from the observation that OAG and GO6983 treatments (both TRP channel activators) resulted in increased Zn²⁺ influx (Figure 2c,d).

We observed Zn²⁺-induced phosphorylation of substrates of Ca²⁺-dependent kinases, PKC, and MLCK (Figure 3). The activation of MLCK requires interactions with the Ca²⁺-binding protein calmodulin (CaM), whilst PKC has a Ca²⁺-binding domain [57,58]. Zn²⁺ was shown to substitute for Ca²⁺ at the EF hand domains of CaM, which may lead to the conformational change and activation of CaM-dependent substrates [47,59]. Additionally, Zn²⁺ binds to and activates PKC isoforms [13,60]. Thus, Zn²⁺ may influence platelet activation by the direct activation of Ca²⁺-dependent proteins. This process may act as a mechanism for priming the platelet for activation in response to low concentrations of conventional agonists [14].

Depletion of Ca²⁺ stores leads to the opening of membrane Ca²⁺ channels in a variety of cell types, including platelets, in a process known as store-dependent Ca²⁺ entry (SOCE). In platelets, the site of the Ca²⁺ store is the DTS. Here, we demonstrate that thrombin stimulation or the depletion of intracellular cation stores using thapsigargin results in increases in [Zn²⁺]_i (Figure 4). As this signal was not observed in the absence of subactivatory (30 μM) [Zn²⁺]_o, we conclude that this signal is due to Zn²⁺ influx, in a process we describe as store-operated Zn²⁺ entry (SOZE). Conversely, thrombin or thapsigargin stimulation results in increases in [Ca²⁺]_i in the absence of any extracellular cation, consistent with Ca²⁺ from intracellular stores. Therefore, SOZE does not result in the liberation of Zn²⁺ from intracellular stores.

SOZE, but not SOCE, was sensitive to TPEN pre-treatment, both confirming the nature of the Zn²⁺ signal and providing evidence that SOCE is not dependent on Zn²⁺. As neither SOCE nor SOZE were affected by KB-R, NCX is not involved in these processes. This is consistent with the previous data which indicate that NCX is not involved in the regulation of [Ca²⁺]_i in neuronal cells [41]. Interestingly, thapsigargin-, but not thrombin-mediated cation influx, was influenced by 2-APB. As 2-APB is also known to inhibit IP₃R, this indicates the absence of IP₃R activity in SOCE [61]. The extent of Ca²⁺ influx in SOCE was reduced in the presence of [Zn²⁺]_o, consistent with competition for cation entry mechanisms. Conversely, the rate and extent of Zn²⁺ influx were unaffected by [Ca²⁺]_o, suggesting that Zn²⁺ entry mechanisms, whilst being cation non-specific, favored Zn²⁺. Competition between Ca²⁺ and Zn²⁺ during intestinal absorption has been previously described where the two ions share a common transport pathway [62]. Hence, we suggest the possibility of competition between the two ions occurring via Zn²⁺ entry mechanisms, including TRP channels and NCX. Additionally, inhibition by 2-APB did not affect TG-induced Zn²⁺ influx in the presence of exogenous 2 mM/30 μM [Ca²⁺]_o/[Zn²⁺]_o (Figure 4), suggesting that NCX mediates Zn²⁺ influx, but does not influence store-operated Ca²⁺ release.

SOCE involves the detection of store Ca²⁺ levels by Orai, with subsequent signaling to the membrane cation, STIM1. Whether SOZE is regulated by the detection of Ca²⁺ or Zn²⁺ depletion is not known. Platelets have a Zn²⁺ store that is released into the cytosol following agonist stimulation, in a manner that is consistent with a secondary messenger [14,26]. However, the nature of the Zn²⁺ store and the mechanism by which Zn²⁺ exerts its actions has received little attention. Previous studies indicate that the store is sensitive to platelet stimulation via GPVI and TP, but not PARs, and is sensitive to the redox state [14]. A further mechanism by which cells regulate Zn²⁺ levels is through Zn²⁺ transporters, of which ZIP7 is present in the platelet proteome [63]. In nucleated cells, the activation of tyrosine kinase pathways depend on the release of Zn²⁺ from the ER into the cytosol, mediated by the Zn²⁺ transporter, ZIP7 [64,65,66]. We confirm that ZIP7 is present on the platelet surface (Figure 5) and that platelet activation via Zn²⁺ or CRP-XL leads to increases in surface expression. This is consistent with the presence of ZIP7 on intracellular membranes of the open canalicular system, which becomes externalized upon platelet activation. Platelet activation using Zn²⁺ results in an increase in ZIP7 phosphorylation, which is sensitive to inhibition of CK2, indicating an activatory pathway. These results complement the findings on nucleated cells, where ZIP7 is regulated via CK2-mediated phosphorylation in MCF-7 cells. In this work, Zn²⁺ stimulation resulted in an increase in [Zn²⁺]_I, which was inhibited upon ZIP7 knockdown via siRNA [65].

Previous work has demonstrated increases in [Zn²⁺]_i, both as a result of Zn²⁺ influx and release from stores [12,14,26]. To determine the relative contributions of each pathway to [Zn²⁺]_i increases, we stimulated FZ-3-loaded platelets with [Zn²⁺]_o or CRP-XL (in the absence of [Zn²⁺]_o, precluding Zn²⁺ influx). Both pathways result in increases in [Zn²⁺]_i with Zn²⁺ stimulation generating a stronger signal, consistent with a greater contribution from Zn²⁺ influx. Other published works have implicated IP₃R in the persistent, rapid [Zn²⁺]_I increases in cortical neurons [67]. To investigate the influence of IP₃R on Zn²⁺ store release, we inhibited IP₃R and observed reduced [Zn²⁺]_i increases in the response to [Zn²⁺]_o, but not CRP-XL. This profile is consistent with the roles of IP₃R, ZIP7, and Orai1 in Zn²⁺ influx, but not for agonist-dependent Zn²⁺ release in the absence of [Zn²⁺]_o. Therefore, Zn²⁺ is not stored and released in a similar manner to Ca²⁺, and ZIP7 was only effective in Zn²⁺ influx following activation-mediated externalization. This is consistent with the observation that [Zn²⁺]_i release, but not Ca²⁺ release, is sensitive to the platelet redox state [14,26] and is supportive of a role for redox-sensitive cation storage proteins, such as metallothioneins. Further experiments are required to further elucidate the nature of the Zn²⁺ store in platelets. The reason why IP₃R inhibition reduces [Zn²⁺]_o influx is unclear; however, it is suggestive of a role for Ca²⁺ in regulating Zn²⁺ entry. This is consistent with the observation that both TG- and thrombin-mediated increases in [Ca²⁺]_i were reduced in the presence of [Zn²⁺]_o (Figure 4). This observation may suggest cross-talk between Zn²⁺- and Ca²⁺-handling mechanisms during platelet activation.

In conclusion, this study provides evidence that platelets are able to detect and respond to increases in [Zn²⁺]_o via influx through TRP channels, the NCX exchanger, and ZIP7 (Figure 7). Store depletion facilitates Zn²⁺ entry via Orai1, which we have termed store-operated Zn²⁺ entry (SOZE). ZIP7 is externalized following platelet activation. Both influx and Zn²⁺ release from stores both contribute to increases in [Zn²⁺]_i. It is likely that other cell systems employ similar mechanisms to detect and respond to changes in [Zn²⁺]_o.

4. Materials and Methods

Materials: 2-Aminoethoxydiphenyl borate (2-APB), SKF 96365 hydrochloride (SKF; 1-[2-(4-Methoxyphenyl)-2-[3-(4-methoxyphenyl)propoxy]ethyl-1H-imidazole hydrochloride), KB-R7943 mesylate (KB-R; 2-[2-[4-(4-Nitrobenzyloxy)phenyl]ethyl]isothiourea mesylate), SN-6 (2-[[4-[(4-Nitrophenyl)methoxy]phenyl]methyl]-4-thiazolidinecarboxylic acid ethyl ester), SAR7334 (4-[[(1R,2R)-2-[(3R)-3-Amino-1-piperidinyl]-2,3-dihydro-1H-inden-1-yl] oxy]-3-chlorobenzonitrile dihydrochloride), 9-Phenanthrol, AM0902 (1-[[3-[2-(4-Chlorophenyl)ethyl]-1,2,4-oxadiazol-5-yl]methyl]-1,7-dihydro-7-methyl-6H-purin-6-one), Teriflunomide (2-Cyano-3-hydroxy-N-[4-(trifluoromethyl)phenyl]-2-butenamide), and integrilin were from Tocris (Bristol, UK). Flufenamic acid (FFA), 1-Oleoyl-2-acetyl-sn-glycerol (OAG), thrombin, calcium chloride, and zinc sulphate were supplied by Sigma (Poole, UK), and thapsigargin (TG) was from Calbiochem (Nottingham, UK). AC1903 and NVS-ZP7-4 were from MedChem express (Monmouth Junction, NJ, USA); AMG9810 ((2E)-N-(2,3-Dihydro-1,4-benzodioxin-6-yl)-3-[4-(1,1-dimethylethyl)phenyl]-2-propenamide) was from Abcam (Cambridge, UK). AnCoA4 was from Merck (Watford, UK). CX-4945 was from Stratetech, Ely, UK). Unless stated, all other reagents were from Sigma Aldrich.

Preparation of washed human platelets: ethical approval for this study was obtained from the Faculty Research Ethics Committee at Anglia Ruskin University. Human blood was collected from healthy volunteers who had not taken medication for two weeks, following informed consent in accordance with the Declaration of Helsinki. Blood was collected in 11 mM of sodium citrate and washed platelets were prepared as described previously [12]. For studies of Ca²⁺ or Zn²⁺ mobilizations, aliquots of platelet-rich plasma (PRP) were retained. Platelets were resuspended using a calcium-free Tyrodes buffer (CFT) containing, in mM, 140 NaCl, 5 KCl, 10 HEPES, 5 Glucose, 0.42 NaH₂PO₄, and 12 NaHCO₃, titrated to pH 7.4 with NaOH.

Light transmission aggregometry: platelet aggregation was monitored as described previously using an AggRam light transmission aggregometer (Helena Biosciences, Gateshead, UK) [12]. Washed platelet suspensions in CFT were stimulated under stirring conditions at 37 °C and aggregation traces were acquired digitally using proprietary software (HemoRam v1.3, Helena Biosciences). Platelets were stimulated by 100 µM ZnSO₄ following pre-incubation (up to 15 min) with specified channel blockers or vehicle control (0.1% DMSO).

Fluorometry: platelets in PRP were loaded with Fluozin-3 (FZ-3, 1 µM, Invitrogen, Paisley, UK) or Fluo-4 (1 µM, Invitrogen) for 30 min, at 37 °C. Platelets were collected by centrifugation (350× g, 15 min), resuspended in CFT, and rested at 37 °C for 30 min prior to use. FZ-3 fluorescence data were acquired under 494 nm excitation and 516 nm emission using a Fluoroskan Ascent Fluorometer (ThermoScientific, Oxford, UK).

Flow cytometry: PRP was loaded with Fluo-4 or FZ-3 (1 µM) and elevations of [Ca²⁺]_i or [Zn²⁺]_i were monitored over a 4 min period using an Accuri C6 flow cytometer (BD Biosciences, Oxford, UK). Platelets were stimulated by thrombin (1 U/mL) in the presence of 2 mM [Ca²⁺]_o, 30 μM [Zn²⁺]_o, or a combination of both. Where stated, platelets were incubated with 1 µM TG, a SERCA pump inhibitor, for 5 min in the absence of [Ca²⁺]_o to assess Ca²⁺ release from the DTS. For experiments assessing the role of NCX or TRP channels, platelets were incubated with inhibitors for 30 min at 37 °C, before stimulation with 100 µM Zn²⁺, and the FZ-3 signal was measured for 5 min. For experiments assessing [Zn²⁺] release from internal stores, platelets were pre-treated with inhibitors against Orai1, IP₃R, or ZIP7 for 30 min at 37 °C, prior to stimulation with 100 µM Zn²⁺ or 1 μg/mL CRP-XL (Cambcol, Ely, UK).

Western blotting: Western blotting was performed as described previously [12]. Briefly, washed platelet suspensions were pre-treated with the α_IIbβ₃ inhibitor integrilin (5 μM) for 5 min prior to stimulation with 30 μM or 1 mM Zn²⁺ for the given time periods, where platelets were then lysed with RIPA buffer (150 mM NaCl, 50 mM Tris-HCL, pH 8.0, 1% nonidet P-40, 0.5% Na deoxycholate, 0.15 SDS). Lysates were separated on 8% SDS-PAGE or 4–12% Gradient NuPAGE Bis-Tris (BioRad) gels and transferred to PVDF membranes. Zn²⁺-induced protein phosphorylation was assessed using antibodies against phospho-β3 and Y759 (AbCam, Cambridge, UK), PKC substrate (AbCam), or p-MLC S19 (Cell Signalling Technology, Danvers, MA, USA), or anti-Phospho-ZIP7 antibodies (Cambridge Bioscience, Cambridge, UK), followed by anti-mouse, anti-rabbit, or anti-goat HRP-conjugated secondary antibodies (AbCam. Antibodies against total β3 or β-actin (Santa Cruz, Heidelberg, Germany) were included as loading controls. Blots were representative of 3–4 platelet preparations.

Data analysis and statistics: maximum aggregation and mean fluorescence values were calculated using Microsoft Excel 365 (Microsoft, Redmond, WA, USA). Representative traces of FZ-3 or Fluo-4 fluorescence responses were generated using Flow Jo (V10.2, Ashland, OR, USA). Western blots were analyzed using ImageJ (v1.45, NIH, Bethesda, MD, USA). Data were analyzed in GraphPad Prism by a two-way ANOVA of Student’s t-test, where p < 0.0001 (****), p < 0.001 (***), p < 0.01 (**), p < 0.05 (*), and p > 0.05 (ns) were indicated.