P2X4 Receptor-Dependent Ca2+ Influx in Model Human Monocytes and Macrophages

Monocytes and macrophages express a repertoire of cell surface P2 receptors for adenosine 5′-triphosphate (ATP) a damage-associated molecular pattern molecule (DAMP), which are capable of raising cytoplasmic calcium when activated. This is achieved either through direct permeation (ionotropic P2X receptors) or by mobilizing intracellular calcium stores (metabotropic P2Y receptors). Here, a side-by-side comparison to investigate the contribution of P2X4 receptor activation in ATP-evoked calcium responses in model human monocytes and macrophages was performed. The expression of P2X1, P2X4, P2X5 and P2X7 was confirmed by qRT-PCR and immunocytochemistry in both model monocyte and macrophage. ATP evoked a concentration-dependent increase in intracellular calcium in both THP-1 monocyte and macrophages. The sarco/endoplasmic reticulum Ca2+-ATPase inhibitor thasigargin (Tg) responses to the maximal ATP concentration (100 μM) in THP-1 monocytes, and responses in macrophage were significantly attenuated. Tg-resistant ATP-evoked calcium responses in the model macrophage were dependent on extracellular calcium, suggesting a requirement for calcium influx. Ivermectin (IVM) potentiated the magnitude of Tg-resistant component and slowed the decay of response in the model macrophage. The Tg-resistant component was attenuated by P2X4 antagonists 5-BDBD and PSB-12062 but not by the P2X1 antagonist Ro0437626 or the P2X7 antagonist A438079. shRNA-mediated P2X4 knockdown resulted in a significant reduction in Tg-resistant ATP-evoked calcium response as well as reduced sensitivities towards P2X4-specific pharmacological tools, IVM and PSB-12062. Inhibition of endocytosis with dynasore significantly reduced the magnitude of Tg-resistant component but substantially slowed decay response. Inhibition of calcium-dependent exocytosis with vacuolin-1 had no effect on the Tg-resistant component. These pharmacological data suggest that P2X4 receptor activation contributed significantly towards the ionotropic calcium response evoked by ATP of the model human macrophage.


Introduction
The release of adenosine 5 -triphosphate (ATP) as a signaling molecule by activated leukocytes, platelets and apoptotic cells is crucial for the induction of a variety of physiological responses across different cell types [1,2]. ATP is able to mediate these biological processes through the activation of purinergic P2 receptors: P2X (ligand-gated ion channels) or P2Y (G-protein coupled receptors) [3,4]. To date, seven mammalian P2X receptor subunits (P2X1-P2X7) have been identified with widespread distribution throughout neuronal and non-neuronal tissues [5]. The expression of P2X receptors has been reported in immune cells such as monocytes and macrophages, with P2X4 and P2X7 being the pre-dominant subunits expressed [6]. Although significant research has been performed in the past decade to elucidate a role of P2X7 receptor in the immune system, the same cannot be said for the P2X4 receptor. Research on the P2X4 receptor has been greatly hampered by the lack of selective P2X4 receptor antagonists as well as the possible complicating presence of the P2X7 receptor [7]. In macrophages and microglia, P2X4 receptors have been shown to co-express with P2X7 receptors, although they do not appear to share the same downstream signaling pathways [8,9].
Although the role of P2X4 receptors in monocytes and macrophages is not clearly defined within the literature, P2X4 expression has been shown to be upregulated following nerve damage in microglia [10] and, in macrophages, P2X4 has been reported to mediate prostaglandin E2 (PGE2) release [11]. Studies by Li and Fountain (2012) in THP-1 monocytes also illustrated that P2X4 contributes towards the ATP-evoked Ca 2+ response and that its activity is suppressed by cholesterol lowering drug, fluvastatin [12].
In the present study, we have performed a side-by-side comparison of the expression of P2X receptors in model human THP-1 monocytes and macrophages through the use of qRT-PCR and immunocytochemistry. Though evidence of functional P2X4 can be revealed by whole-cell patch-clamps, studies evaluating the role of P2X receptors in intact cells are limited. To this end, we investigated the contribution of P2X4 receptor activation towards ATP-evoked intracellular Ca 2+ response, and this was assessed in the two cell models through intracellular Ca 2+ measurements. Our findings allowed the identification of P2X4 contribution of towards ATP-evoked Ca 2+ influx in model human monocytes and macrophages, while also assessing the reliability of commercially available pharmacological tools for the study of P2X4.

Dependency of ATP-Evoked Ca 2+ Response on ER Calcium Store
As ATP evoked a global intracellular Ca 2+ response, mediated by both the activation of P2X and P2Y receptors, a sarco/endoplasmic reticulum Ca 2+ -ATPase inhibitor, Thapsigargin (Tg), was employed to eliminate metabotropic responses. Tg depletes intracellular stores of Ca 2+ resulting in a raise in cytosolic Ca 2+ level (F ratio 1.35 ± 0.0082 without Tg vs. 2.24 ± 0.065 with Tg; n = 6; p < 0.001; Figure 3). Pre-treatment of cells with 5 μM Tg completely attenuated ATP-evoked intracellular Ca 2+ response in THP-1 monocytes ( Figure 3A,B) while significantly inhibiting ATP-evoked intracellular Ca 2+ response in THP-1-differentiated macrophages (88.7 ± 3.29% inhibition; n = 6; p < 0.001) ( Figure 3C,D). Despite a significant inhibition with Tg, the results showed that 11.3 ± 3.29% (n = 6) of the ATP-evoked Ca 2+ response appeared to be resistant to Tg. In the absence of extracellular Ca 2+ , this Tg-resistant Ca 2+ response was completely abolished, implying that it is dependent on Ca 2+ influx ( Figure 3E).

Dependency of ATP-Evoked Ca 2+ Response on ER Calcium Store
As ATP evoked a global intracellular Ca 2+ response, mediated by both the activation of P2X and P2Y receptors, a sarco/endoplasmic reticulum Ca 2+ -ATPase inhibitor, Thapsigargin (Tg), was employed to eliminate metabotropic responses. Tg depletes intracellular stores of Ca 2+ resulting in a raise in cytosolic Ca 2+ level (F ratio 1.35 ± 0.0082 without Tg vs. 2.24 ± 0.065 with Tg; n = 6; p < 0.001; Figure 3). Pre-treatment of cells with 5 µM Tg completely attenuated ATP-evoked intracellular Ca 2+ response in THP-1 monocytes ( Figure 3A,B) while significantly inhibiting ATP-evoked intracellular Ca 2+ response in THP-1-differentiated macrophages (88.7 ± 3.29% inhibition; n = 6; p < 0.001) ( Figure 3C,D). Despite a significant inhibition with Tg, the results showed that 11.3 ± 3.29% (n = 6) of the ATP-evoked Ca 2+ response appeared to be resistant to Tg. In the absence of extracellular Ca 2+ , this Tg-resistant Ca 2+ response was completely abolished, implying that it is dependent on Ca 2+ influx ( Figure 3E).

Silencing of P2X4 Abolished Tg-Resistant Calcium Response and Reduced Sensitivities Towards IVM and PSB-12062
The data so far illustrated that P2X4 receptors contribute towards the Tg-resistant ATP-evoked intracellular Ca 2+ level. To provide further confirmation, a molecular-based approach to silence the P2X4 gene was employed. The success of the gene silencing technique was confirmed by the qRT-PCR approach (48.8 ± 9.64% knockdown; n = 3; p < 0.01; Figure 6A) and flow cytometry (57.7 ± 4.37% reduction in cells expressing extracellular P2X4; n = 3; p < 0.01; Figure 6B). Despite a reduction in cell surface expression of P2X4 receptor, no significant changes in 100 μM ATP-evoked intracellular Ca 2+ responses were observed in the P2X4-shRNA cells when compared to scrambled cells ( Figure 6C). However, silencing of P2X4 significantly reduced the Tg-resistant ATP-evoked Ca 2+ response in THP-1-differentiated macrophages (F ratio 0.14 ± 0.02 ATP + Tg scrambled shRNA vs. F ratio 0.02 ± 0.014 ATP + Tg P2X4 shRNA; n = 3; p < 0.001; Figure 6D). The effect of IVM and PSB-12062 was assessed on the P2X4 shRNA cells. When compared to scrambled negative control cells ( Figure 7A,B), P2X4 shRNA cells were found to be less sensitive towards both IVM and PSB-12062, as quantified using area under the curve ( Figure 7B). While IVM caused a potentiation towards the net calcium movement of scrambled shRNA cells by 2.06 ± 0.23 fold (n = 3; p < 0.01; Figure 7A,B), IVM was only

Silencing of P2X4 Abolished Tg-Resistant Calcium Response and Reduced Sensitivities Towards IVM and PSB-12062
The data so far illustrated that P2X4 receptors contribute towards the Tg-resistant ATP-evoked intracellular Ca 2+ level. To provide further confirmation, a molecular-based approach to silence the P2X4 gene was employed. The success of the gene silencing technique was confirmed by the qRT-PCR approach (48.8 ± 9.64% knockdown; n = 3; p < 0.01; Figure 6A) and flow cytometry (57.7 ± 4.37% reduction in cells expressing extracellular P2X4; n = 3; p < 0.01; Figure 6B). Despite a reduction in cell surface expression of P2X4 receptor, no significant changes in 100 µM ATP-evoked intracellular Ca 2+ responses were observed in the P2X4-shRNA cells when compared to scrambled cells ( Figure 6C). However, silencing of P2X4 significantly reduced the Tg-resistant ATP-evoked Ca 2+ response in THP-1-differentiated macrophages (F ratio 0.14 ± 0.02 ATP + Tg scrambled shRNA vs. F ratio 0.02 ± 0.014 ATP + Tg P2X4 shRNA; n = 3; p < 0.001; Figure 6D). The effect of IVM and PSB-12062 was assessed on the P2X4 shRNA cells. When compared to scrambled negative control cells ( Figure 7A,B), P2X4 shRNA cells were found to be less sensitive towards both IVM and PSB-12062, as quantified using area under the curve ( Figure 7B). While IVM caused a potentiation towards the net calcium movement of scrambled shRNA cells by 2.06 ± 0.23 fold (n = 3; p < 0.01; Figure 7A,B), IVM was only able to potentiate the net calcium movement of P2X4 shRNA cells by 1.53 ± 0.055 fold (n = 3; p < 0.01; Figure 7C,D). These data provide further evidence that P2X4 receptor contributes towards Ca 2+ influx in THP-1-differentiated macrophages and that both IVM and PSB-12062 are useful tools for the study of P2X4. able to potentiate the net calcium movement of P2X4 shRNA cells by 1.53 ± 0.055 fold (n = 3; p < 0.01; Figure 7C,D). These data provide further evidence that P2X4 receptor contributes towards Ca 2+ influx in THP-1-differentiated macrophages and that both IVM and PSB-12062 are useful tools for the study of P2X4.

Discussion
The expression of P2X receptors is widespread within the immune system, in particular P2X4 and P2X7 [6]. Despite this observation, there has been no clear evidence of the contribution of P2X4 receptor activation towards ATP-evoked Ca 2+ influx in monocytes and macrophages. Since THP-1 monocytes and THP-1 differentiated macrophages are frequently used as an experimentally amenable model of human monocytes and macrophages [14][15][16], we utilized these model systems to investigate the aforementioned issue. Through qRT-PCR analysis, the present study illustrated that both THP-1 and THP-1 differentiated macrophages expresse all P2X receptors at the mRNA level with the exception of P2X2 for THP-1 and P2X2 and P2X3 for THP-1 differentiated macrophages. It was also interesting to observe that PMA differentiation of THP-1 into the model macrophage caused a downregulation of P2X1 and P2X2 mRNA expression while an upregulation of P2X4 mRNA expression was observed. To date, there is no reported comparison of mRNA expression level between THP-1 monocytes and THP-1 differentiated macrophages. However, our findings corroborated previously reported evidence by Myrtek et al. (2008) illustrating the expression of all P2X receptors except for P2X2, P2X3 and P2X6 [17]. Protein expression of P2X1, P2X4, P2X5 and P2X7 was further confirmed by immunocytochemistry analysis for both cell models.
To investigate the contribution of P2X4 receptor activation towards ATP-evoked intracellular Ca 2+ responses in THP-1 monocytes and THP-1 differentiated macrophages, we employed intracellular Ca 2+ measurements with the aid of various pharmacological tools. ATP elicited a concentration-dependent increase in intracellular Ca 2+ level in both cells but with significantly different response characteristics. First, the peak magnitude of ATP-evoked Ca 2+ responses was significantly larger in THP-1 monocytes, compared to THP-1 differentiated macrophages. Secondly, decay kinetic response in THP-1 cells was found to be significantly slower than that observed in THP-1 differentiated macrophages. Although the factors underlying these differences were not investigated further, several speculations can be made. The difference in peak magnitude may be a result of the different expression of P2Y receptors expressed on the cell surface of the two cell types. This is a plausible explanation as P2Y receptor activation has previously been reported as being responsible for mediating the first initial intracellular Ca 2+ peak response [18,19]. A second plausible explanation would be the difference in ectoenzyme CD39 expression, responsible for the breakdown of ATP into ADP and AMP. Although monocytes and macrophages are reported to express ectoenzymes CD39 and CD73, there has been no direct comparison on the level of their expression [20,21]. On the other hand, the difference in decay kinetics of the two cells in response to ATP stimulation may be a result of the different buffering capacity of Ca 2+ of the two cells. Previous studies revealed that antibody/FcR generated phagocytosis in macrophages is a process that relies on increase in cytosolic Ca 2+ concentration [22,23]. Therefore, as macrophages are professional phagocytes, they may be more equipped to adapt to the changes in Ca 2+ level.
As ATP elicits global P2 responses constituting of P2X and P2Y receptors, we utilized a SERCA inhibitor, Thapsigargin (Tg), to deplete the intracellular stores of Ca 2+ to eliminate mobilization of Ca 2+ from intracellular stores by P2Y receptors [24,25]. Interestingly, pre-treatment of THP-1 monocytes with 5 µM Tg completely abolished the ATP-evoked Ca 2+ response, suggesting that it is entirely dependent on metabotropic P2Y receptor activation. The same treatment in THP-1 differentiated macrophages resulted in a significant inhibition of the ATP-evoked Ca 2+ response, leaving a small Tg-resistant component, which was dependent on Ca 2+ influx. Due to the lack of Tg-resistant Ca 2+ response, we did not pursue investigation into THP-1 monocytes and focused on THP-1 differentiated macrophages instead. Our study provides evidence regarding the contribution of P2X4 receptor towards ATP-evoked Ca 2+ response in THP-1 differentiated macrophages. This was evident from the effect observed by the use of positive allosteric modulator ivermectin (IVM) [26] and selective P2X4 receptor antagonists 5-BDBD and PSB-12062 [27][28][29][30]. Pre-treatment of THP-1 differentiated macrophages with 3 µM IVM caused a significant potentiation towards the peak magnitude of Tg-resistant Ca 2+ response, accompanied by a significant delay in decay kinetics. It has been previously reported that the main characteristics of the IVM effect on P2X4 receptors include potentiation in magnitude and delay in decay current, although studies by Norenberg et al. (2012) illustrated that only the effect on decay kinetics is specific to P2X4 receptors [26,28]. The effects of 5-BDBD and PSB-12062 on Tg-resistant Ca 2+ response was also consistent, in which they both significantly reduced the peak magnitude as well as abolished the second slower response. Despite similar effect in inhibition, PSB-12062 appeared to be more potent in causing an inhibition towards the Tg-resistant Ca 2+ response, and this may be due to the nature of the antagonists. While PSB-12062 has been described as an allosteric antagonist [28], 5-BDBD is thought to act competitively, which implies that its effect may be masked due to the high agonist concentration used throughout the study (100 µM ATP). Finally, the contribution of P2X4 receptor activation towards ATP-evoked Ca 2+ response was confirmed with a knockdown approach, which illustrated that P2X4 knockdown THP-1 differentiated macrophages lacked a Tg-resistant ATP-evoked Ca 2+ response as well as reduced sensitivities towards IVM and PSB-12062.
As P2X1, P2X4 and P2X7 are all expressed in the immune cells [31,32], we also investigated the contribution of P2X1 or P2X7 towards the Tg-resistant Ca 2+ response. Selective P2X1 antagonist Ro0437626 and P2X7 antagonist A438079 had no significant inhibitory effect towards the Tg-resistant ATP-evoked Ca 2+ response in THP-1 differentiated macrophages. The lack of contribution of P2X1 in ATP-evoked Ca 2+ response in these cells contradict previous studies reported by Sim et al. and Wareham et al. illustrating functional evidence of P2X1 receptor in mouse macrophages [33] and human lung mast cells [31], respectively. Our qRT-PCR data illustrated that mRNA expression of P2X1 receptors in THP-1 differentiated macrophages is downregulated compared to THP-1 monocytes. It may therefore be interesting to investigate if this downregulation translates at the protein level, therefore being responsible for the lack of P2X1 contribution in these cells. Unlike P2X1, the lack of P2X7 receptor contribution towards ATP-evoked Ca 2+ response in THP-1 differentiated macrophages is not as surprising. It is well established that 100 µM ATP used through this study is significantly below the activation threshold for P2X7 receptors (>500 µM) [34][35][36].
Altogether, the data provides evidence that P2X4 is functionally expressed in THP-1 differentiated macrophages as reflected from their contribution towards ATP-evoked Ca 2+ response, but their functional evidence in THP-1 monocyte is lacking. Studies by Li et al. have previously reported functional P2X4 in THP-1 monocytes, although, in this case, PLC inhibitor U-73122 was used instead of Tg [12]. Furthermore, through intracellular Ca 2+ measurements, we have showed that IVM and PSB-12062 are reliable tools for the study of P2X4, supported with findings from P2X4 knockdown cells.

Intracellular Ca 2+ Measurements
Cells at 2 × 10 5 /well were loaded for 1 h at 37 • C with 2 µM Fura-2 AM. Cells were washed twice with physiological saline (SBS) buffer. Measurements were performed at 37 • C on a 96-well plate reader (FlexStation III, Molecular Devices, Sunnyvale, CA, USA). The change in intracellular