Extracellular chloride is required for efficient platelet aggregation

Abstract Anion channels perform a diverse range of functions and have been implicated in ATP release, volume regulation, and phosphatidylserine exposure. Platelets have been shown to express several anion channels but their function is incompletely understood. Due to a paucity of specific pharmacological blockers, we investigated the effect of extracellular chloride substitution on platelet activation using aggregometry and flow cytometry. In the absence of extracellular chloride, we observed a modest reduction of the maximum aggregation response to thrombin or collagen-related peptide. However, the rate of aggregation was substantially reduced in a manner that was dependent on the extracellular chloride concentration and aggregation in the absence of chloride was noticeably biphasic, indicative of impaired secondary signaling. This was further investigated by targeting secondary agonists with aspirin and apyrase or by blockade of the ADP receptor P2Y12. Under these conditions, the rates of aggregation were comparable to those recorded in the absence of extracellular chloride. Finally, we assessed platelet granule release by flow cytometry and report a chloride-dependent element of alpha, but not dense, granule secretion. Taken together these data support a role for anion channels in the efficient induction of platelet activation, likely via enhancement of secondary signaling pathways.

Given the lack of specific anion channel blockers, we focus on the effect of extracellular Cl − ([Cl − ] o ) substitution on platelet activation. Our experiments highlight a role for anion channels in modulating the rate of platelet aggregation.

Materials
Aspirin, apyrase, and thrombin were from Sigma (Poole, UK). AR-C66096 was from Tocris Bioscience (Bristol, UK). Cross-linked collagen-related peptide (CRP-XL) was prepared as described previously [16] and supplied by R. Farndale (Cambridge, UK). Unless indicated, all other reagents were from Sigma.

Washed platelet preparation
This study was approved by the local Ethics Committee at Anglia Ruskin University. Human blood was collected from healthy volunteers following informed consent in accordance with the Declaration of Helsinki. Blood was collected into 11 mM sodium citrate and washed platelets were prepared as described previously [17]. Platelets were resuspended in a nominally calciumfree buffer containing (in mM) 145 NaCl, 5 KCl, 1 MgCl 2 , 10 glucose, 10 HEPES, titrated to pH 7.35 with NaOH. Where indicated, [Cl − ] o was substituted by equimolar gluconate.

Aggregometry
Platelet aggregation was monitored as described previously using an AggRam aggregometer (Helena Biosciences, Gateshead, UK) [17]. In the experiments of Figure 1c, aliquots of 151 and 1 mM [Cl − ] o -containing platelet suspensions were mixed 5 min prior to agonist addition. Platelets were preincubated with each drug(s) for 5 min at 37°C.

Granule release
Thrombin-evoked alpha and dense granule release was assessed by flow cytometry using fluorescently conjugated CD62P and CD63 antibodies (BD Biosciences, Oxford, UK), respectively. Antibody binding was monitored for 5 min using an Accuri C6 Flow cytometer (BD Biosciences) and the percentage of positive cells was calculated within FlowJo (V10.2, Oregon, USA).

Data analysis and statistics
Maximum aggregation (%) and the initial rate of aggregation (% s −1 ) were calculated in Excel (Microsoft, Redmond, Washington, USA), where rate was determined as the change in aggregation (%) in the first 30 s following shape change. Data were analyzed in GraphPad Prism by two-way ANOVA or Student's t test as indicated and are representative of a minimum of four independent experiments. ***, **, *, and ns denote P < 0.001, P < 0.01, P < 0.05, and not significant, respectively.

Conclusions
Here we demonstrate that [Cl − ] o enhances the rate of platelet aggregation in a concentration-dependent manner (Figure 1). This effect was equivalent to blockade of secondary mediators and P2Y12 inhibition (Figure 2). One possible explanation of these data is that [Cl − ] o is required for efficient release of ATP and/or ADP from the platelet, but we failed to observe a change in dense granule secretion (Figure 2c). Pannexin-1 has been shown to activate in response to elevation of intracellular Ca 2+ ([Ca 2+ ] i ) [20], facilitating cytosolic ATP release [14]. It has been suggested that ADP release may occur via a similar mechanism [21]. Given that release of alpha granule cargo (e.g., fibrinogen, thrombin, and Zn 2+ ) is required for aggregation and is enhanced by P2Y12 signaling [22][23][24], it is possible that [Cl − ] o enhances platelet activation by mediating efficient alpha granule secretion.
Reduction of [Cl − ] o has been shown to substantially reduce thrombin-plus-CRP-XL-mediated elevation of [Ca 2+ ] i in a similar manner to that of Cl − channel blockers [25]. It has been suggested that Cl − currents hyperpolarize the cell, increasing driving force for Ca 2+ influx [9]. However, the platelet Cl − equilibrium is ≈35 mV in the platelet [7], meaning activation of a Cl − conductance would depolarize rather than hyperpolarize platelets. Reduced Ca 2+ influx may be due to reduced secondary signaling, rather than altered membrane potential.
We have focused on the contribution of [Cl − ] o by way of ionic substitution experiments because of the paucity of specific pharmacological tools to study anion channels. This may also explain why anion channels have previously received much less attention than cation channels. It is worth noting that anion channels have been associated with cystic fibrosis, bleeding phenotypes, and inflammatory conditions [12,13,15,26,27] and may represent valuable therapeutic targets, as demonstrated by clinical use of CFTR modulators [28]. Further work will be required to investigate the contribution(s) by the cohort of platelet anion channels during platelet activation.

Declaration of interest
The authors report no declarations of interest.