Calcium and Calmodulin Regulate Mercury-induced Phospholipase D Activation in Vascular Endothelial Cells

Abstract

Earlier, we reported that mercury, the environmental risk factor for cardiovascular diseases, activates vascular endothelial cell (EC) phospholipase D (PLD). Here, we report the novel and significant finding that calcium and calmodulin regulated mercury-induced PLD activation in bovine pulmonary artery ECs (BPAECs). Mercury (mercury chloride, 25 μM; thimerosal, 25 μM; methylmercury, 10 μM) significantly activated PLD in BPAECs. Calcium chelating agents and calcium depletion of the medium completely attenuated the mercury-induced PLD activation in ECs. Calmodulin inhibitors significantly attenuated mercury-induced PLD activation in BPAECs. Despite the absence of L-type calcium channels in ECs, nifedipine, nimodipine, and diltiazem significantly attenuated mercury-induced PLD activation and cytotoxicity in BPAECs. This study demonstrated the importance of calcium and calmodulin in the regulation of mercury-induced PLD activation and the protective action of L-type calcium channel blockers against mercury cytotoxicity in vascular ECs, suggesting mechanisms of mercury vasculotoxicity and mercury-induced cardiovascular diseases.

Mercury (Hg), a heavy metal belonging to the transition metal series of the periodic table, is an established environmental pollutant with known toxicity in humans. Mercury is widely recognized for its cytotoxicity, neurotoxicity, and immunotoxicity, and it appears to play no known physiological role. ^1–3 Mercury usage in several devices causes accidental and occupational exposure to the metal among humans. ¹ Inorganic mercury, in the form of chloride, is toxic to many organisms, including humans, and can readily undergo microbial biomethlyation to form the highly toxic organic form, methylmercury. ⁴ Methylmercury has been shown to cause hypertension in rats. ⁵ It has also been documented that methylmercury generates reactive oxygen species (ROS), leading to cellular oxidative stress. ^6,7 Elemental mercury (Hg⁰) was commonly used in dental practices for the greater part of the 20th century, and mercury vapor released from amalgam surfaces in the mouth is the predominant source of mercury exposure in the general population. ⁸ Persistent use of thimerosal (an organic mercurial) as a preservative in vaccines remains a controversial topic, however, as yet there has been no conclusive report linking this mercury compound to neurological defects in infants and young children. ^9,10 Furthermore, increased exposure to mercury has been correlated with the increased risk of cardiovascular diseases in humans. ^11–15 Mercury concentrations in human toenail and urine samples have been found to be directly correlated with the increased risk of myocardial infarction and coronary heart disease. ^5,13,14 Despite the correlation between mercury and cardiovascular diseases, there remains a void in understanding the role of vascular endothelial cells (ECs) in the mechanism. Therefore, mercuryinduced cardiovascular diseases may conceivably be a result of the toxic effects of mercury on the vascular endothelium.

Recently, it has been shown in our laboratory that mercury activates phospholipases, which generate bioactive lipid second messengers in vascular ECs. ^7,16 One member of the phospholipase family, phospholipase D (PLD), is a ubiquitous lipid signaling enzyme present in mammalian cells that acts exclusively on the substrate phosphatidylcholine (PC). Phospholipase D hydrolyzes PC to form choline and phosphatidic acid (PA), which is subsequently metabolized to either 1,2-diacylglycerol (DAG) by phosphatidate phosphohydrolase or lysophosphatic acid (LPA) by phospholipase A₁/A₂, wherein PA, LPA, and DAG have been shown as potent cellular lipid signal mediators. ^17–20

The involvement of calcium in different functions of the cardiovascular system, including the vascular smooth muscle cells and ECs, is critical. ²¹ Calcium transport in the vascular smooth muscle cells is carried out through the voltage-gated L-type channels. ²² The presence or expression of L-type calcium channels in the vascular ECs is reported to be absent. ²³ Calcium influx in ECs occurs through different mechanisms involving the nonselective calcium-permeable cation channels and store-operated calcium channels. ²⁴ The candidates belonging to the family of transient receptor potential (TRP) channels are the primary permeability-coupled calcium channels in ECs. ²⁵ Mercury, especially methylmercury, exposure has been shown to increase intracellular calcium concentration in neuronal cells. ^26,27 Mercury chloride-and methylmercury-induced calcium influx has been documented in neuronal cells as being modulated by the L-type voltage-gated ion channels. ^28,29 Earlier, we reported that calcium is involved in the mechanism of activation of phospholipase A₂. ³⁰ In addition, the role of calcium in methylmercury-induced PLA₂ activation in neurons has been shown. ²⁷ Furthermore, the role of calcium in the agonist-mediated activation in PLD in mammalian cells has been reported. ^31–33 However, it is not known whether calcium plays a role in mercury-induced PLD activation in vascular ECs. Upon binding with calcium, calmodulin, the principal cellular calcium-binding protein, transforms into a calcium-calmodulin complex that transduces calcium-dependent signals at several cellular targets. ³⁴ Calcium/calmodulin-dependent protein kinase II (CaM kinase II), an important player in the calmodulin cascade, is emerging as a critical player in EC pathophysiology, and its involvement in mercury-induced activation of PLD in ECs has not been reported so far. ³⁵ Although ECs have been observed not to express L-type calcium channels, the L-type calcium channel blockers have been shown to offer protection against dysfunction and oxidative stress in ECs. ^36,37 On the other hand, the protective effect of L-type calcium channel blockers on mercury-induced PLD activation in vascular ECs has not been reported. Also, our earlier reported study demonstrated that mercury causes activation of PLD in lung ECs through oxidative stress, thereby establishing the premise to further investigate the role of calcium and calmodulin in enzyme activation. ⁷

Along these lines, in the current study, we tested our hypothesis that (1) calcium and calmodulin regulate mercury-induced PLD activation in vascular ECs and (2) despite the reports made on the absence of L-type calcium channels in vascular ECs, L-type calcium channel blockers attenuate mercury-induced PLD activation and protect against mercury cytotoxicity in vascular ECs. For the first time, the current study revealed new and significant findings demonstrating that calcium and calmodulin regulated mercury-induced PLD activation, and in spite of the reported absence of L-type calcium channels, L-type calcium channel blockers attenuated mercury-induced PLD activation and protected against mercury-induced cytotoxicity in lung vascular ECs. Furthermore, the current study offered interesting insights into the mechanism(s) and protection of mercury vasculotoxicity involving the roles of calcium, calmodulin, and PLD activation in ECs.

Materials and Methods

Materials

Bovine pulmonary artery ECs (BPAECs) (passage 2) were purchased from Cell Applications, Inc. (San Diego, CA). Nonessential amino acids, antibiotic-antimycotic (10 000 units/mL penicillin, 10 000 μg/mL streptomycin, 25 μg/mL amphotericin B), fetal bovine serum (FBS), and trypsin were purchased from Gibco Invitrogen Corp. (Grand Island, NY). Dulbecco’s Modified Eagle Medium (DMEM) phosphate-free medium, minimum essential medium (MEM), mercury chloride, methylmercury chloride, thimerosal, 1,2-bis-(o-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid tetra-(acetoxymethyl) ester (BAPTA-AM), ethyleneglycol-bis-(β-aminoethylether)-N,N, N′,N′-tetraacetic acid (EGTA), trifluoperazine dihydrochloride, lactate dehydrogenase (LDH) cytotoxicity assay kit, n-butanol, and Spinner Modification (without calcium chloride) MEM were obtained from Sigma Chemical Co. (St. Louis, MO). Nifedipine, nimodipine, diltiazem hydrochloride, and calmidazolium chloride were obtained from Calbiochem (San Diego, CA). [³²P]Orthophosphate and [⁴⁵Ca]calcium chloride were obtained from New England Nuclear (Wilmington, DE). Phosphatidylbutanol (PBt) was purchased from Avanti Polar Lipids (Alabaster, AL).

Cell Culture

Cell cultures were maintained in a humidified environment of 95% air-5% CO₂ and grown to contact inhibited monolayer with typical cobblestone morphology. Bovine pulmonary artery ECs were grown to confluence in MEM supplemented with 10% FBS, 100 units/mL penicillin and streptomycin, 1% nonessential amino acids, and 5 μg/mL endothelial growth factor, and grown as described previously. ¹⁷ Upon confluence, cells were trypsinized and subcultured in sterile 35-mm tissue culture dishes. Confluent cells showed cobblestone morphology under a light microscope and stained positive for Factor VIII. All experiments were conducted between 8 and 20 passages (75%–80% confluence).

Phospholipase D Activation in ECs

Bovine pulmonary artery ECs in 35-mm dishes (5 × 10⁵ cells/dish) were incubated with [³²P]orthophosphate (5 μCi/mL) in phosphate-free DMEM containing 2% FBS for 18 hours at 37°C in 95% air-5% CO₂. ¹⁷ The [³²P]orthophosphate radioactive medium was removed by aspiration, and the prelabeled cells were incubated in serum-free MEM or MEM containing the selected mercury compounds (mercury chloride, methylmercury chloride, or thimerosal) at the chosen concentrations in the presence of 0.05% butanol for the desired lengths of time (0–60 min), as described earlier. ⁷ When required, in studies to establish the role of calcium and calmodulin in enzyme activation, cells prelabeled with [³²P]orthophosphate were treated with MEM containing the selected pharmacological agents/inhibitors for 1 hour and then exposed to MEM containing the chosen mercury compounds at doses that could induce significant PLD activation, in the absence or presence of the pharmacological inhibitors and in the presence of 0.05% butanol for the desired length of time (30 min). The pharmacological inhibitors included calcium chelating agents, L-type calcium channel inhibitors (nifedipine, nimodipine, and diltiazem), and calmodulin inhibitors (calmidazolium and trifluoperazine). Incubations were terminated by the addition of 1 mL of methanol:HCl (100:1 vol/vol), and lipids were extracted in chloroform:methanol (2:1 vol/vol). ¹⁷ For experiments to investigate the role of calcium, BPAECs were incubated in the phosphate-free MEM containing [³²P]orthophosphate and calcium chelating agent EGTA for 18 hours to create a calcium-depleted environment. Cells were then exposed to mercury compounds in the presence of the desired concentrations of calcium chloride for 30 minutes in calcium-enriched and calciumdepleted or calcium-deficient (Spinner modification) MEM. [³²P]phosphatidylbutanol (PBt) formed as a result of PLD activation, and its associated transphosphatidylation reaction served as the index of PLD activity in intact cells. [³²P]PBt was separated by thin-layer chromatography (TLC), with the upper phase of ethyl acetate: iso-octane: glacial acetic acid: water (65:10:15:50 by vol) as the developing solvent system. Unlabeled PBt was added as a marker during TLC separation of lipids, and the lipids separated on the TLC plate were visualized upon exposure to iodine vapors. Spots of PBt were marked, carefully scraped, and the radioactivity associated with the [³²P]PBt spot was quantified using a liquid scintillation counter. All values were normalized to 1 million dpm in total lipid extract, and the [³²P]PBt formed was expressed as DPM/dish or percentage control. ⁷

Calcium Uptake Assay

Uptake of calcium by BPAECs was determined as previously established. ²⁷ Bovine pulmonary artery ECs (∼80% confluent in 35-mm dishes) were washed twice with 1 mL of phosphate-buffered saline (PBS) and treated with 1 mL MEM or MEM containing the mercury compound at the chosen dose for the designated length of time at 37°C in a humidified atmosphere of 95% air-5% CO₂. At the end of incubation, the medium was aspirated and the cells were incubated in calcium-free MEM containing 0.25 mCi/mL of [⁴⁵CaCl₂] for 30 minutes at 37°C under a humidified atmosphere of 95% air-5% CO₂. Following the incubation, the medium was aspirated and the cells were washed 4 times (15 sec each) with phosphate-buffered saline (PBS) containing 0.5 mM EGTA by gentle swirling. After the final wash, the adherent cells were solubilized with 1 mL of 1 M NaOH upon incubation at 37°C for 15 minutes, and the radioactivity was measured by liquid scintillation counting. The extent of [⁴⁵Ca] uptake was expressed as DPM/dish.

LDH Assay of Cytotoxicity

Cytotoxicity in BPAECs was determined by assaying the release of LDH from cells according to our previously published method. ¹⁶ Bovine pulmonary artery ECs grown up to 90% confluence in 35-mm dishes were pretreated with MEM alone or MEM containing the L-type calcium channel blocker (nifedipine, nimodipine, or diltiazem) for 1 hour, following which the cells were treated with MEM alone, MEM containing the L-type calcium channel blocker alone, MEM containing the selected mercury compound alone (mercury chloride or methylmercury chloride or thimerosal), or MEM containing the chosen L-type calcium channel blocker + mercury compound for 1 hour. At the end of treatment, the medium was collected, the experiment was terminated with the addition of 1 N HCl, and LDH released into the medium was determined spectrophotometrically according to the manufacturer’s recommendations (Sigma Chemical Co., St. Louis, MO).

Statistical Analysis of Data

All experiments were done in triplicate, and data were expressed as mean ± standard deviation (SD). One-way analysis of variance (ANOVA) and pairwise multiple statistical comparison were done by Dunnett’s method with P < .05 indicating statistical significance.

Results

Mercury Activates PLD

Earlier, we reported that mercury chloride, thimerosal, and methylmercury chloride induced the activation of PLD in BPAECs in a dose- and time-dependent fashion. ⁷ In the current study, we again confirmed our earlier report on the activation of PLD in BPAECs induced by the 3 chosen mercury compounds (data not shown). More importantly, from these experiments, we chose optimal doses of the 3 chosen mercury compounds (25 μM mercury chloride, 25 μM thimerosal, and 10 μM methylmercury chloride) and optimal time of treatment of cells (30 min) for studies to establish the role of calcium and calmodulin in mercury-induced PLD activation in BPAECs.

Calcium Chelating Agents Attenuate Mercury-Induced PLD Activation

Calcium chelating agents complex with calcium and have been shown to inhibit its biological actions. Physiological calcium is pertinent in 2 forms, extracellular calcium and intracellular calcium. Therefore, here, the well-established and widely used extracellular calcium chelating agent EGTA (500 μM) and the intracellular calcium chelating agent BAPTA (5 μM) were used to determine the role of both extracellular and intracellular calcium in mercury-induced PLD activation. The procedure for the chelating agents differed with regard to pretreatment. Cells treated with EGTA were incubated for 30 minutes with MEM alone, MEM containing EGTA (500 μM), MEM containing the chosen mercury compound(s), or MEM containing both the chosen mercury compound(s) and EGTA. For experiments involving BAPTA, cells were first pretreated for 1 hour with MEM alone or MEM containing BAPTA and then exposed to the mercury compound(s) in the presence of BAPTA for 30 minutes. The extracellular calcium chelating agent EGTA significantly attenuated mercury chloride (25 μM)-, thimerosal (25 μM)-, and methylmercury chloride (10 μM)-induced PLD activation in BPAECs. Mercury chloride-, thimerosal-, and methylmercury-induced enzyme activation was also significantly inhibited (48%, 58%, and 52%, respectively) as compared to the untreated control cells (Figure 1A). The intracellular calcium chelating agent BAPTA also inhibited mercury chloride (25 μM)-, thimerosal (25 μM)-, and methylmercury chloride (10 μM)-induced PLD activation in BPAECs (21%, 17%, and 25% inhibition, respectively) (Figure 1B). Overall, these results showed that EGTA and BAPTA were significantly effective in causing the attenuation of mercury chloride-, thimerosal-, and methylmercury chloride-induced PLD activation in BPAECs. However, it should be noted that the extracellular calcium quencher EGTA was significantly more effective in attenuating mercury-induced PLD activation as compared to that offered by the intracellular quencher BAPTA.

Mercury-induced PLD Activation is Calcium Dependent

The earlier experiments of this study (Figures 1A and 1B) demonstrated that EGTA, the extracellular calcium quencher, was more effective than the intracellular calcium chelator BAPTA in attenuating mercury-induced PLD activation in BPAECs. As these results clearly demonstrated that extracellular calcium was a significant factor in contributing to mercury-induced PLD activation in the cells, here, the studies were conducted to establish the requirement for extracellular calcium in mercury-induced PLD activation in BPAECs. As shown in Figure 2, cells cultured in the calcium-depleted medium (created by prior EGTA treatment to sequester calcium in the medium) for 18 hours and then treated with mercury (25 μM mercury chloride, 25 μM thimerosal, and 10 μM methylmercury chloride) did not exhibit any activation of PLD, as the extent of enzyme activity in the mercury-treated cells was similar to that observed in the untreated control cells. In contrast, all 3 selected mercury compounds significantly induced PLD activation in the cells cultured in the calcium-enriched (containing ∼2.0 mM calcium) MEM. The results also revealed that the basal activity of PLD was also dependent on calcium in the medium, as the untreated control cells grown in the calcium-enriched medium exhibited a marked increase in PLD activation.

As the calcium concentration of the calcium-rich MEM (calcium-enriched medium) is ∼2 mM, a study was conducted to establish the stoichiometric (dose-dependent) relationship between extracellular calcium and activation of PLD in BPAECs. Bovine pulmonary artery ECs were grown in commercial calcium-deficient medium (Spinner modification) and were exposed to mercury (25 μM mercury chloride, 25 μM thimerosal, and 10 μM methylmercury chloride) in the same calcium-deficient medium with a gradual increase in the extracellular calcium (0–2.5 mM), following which mercury-induced PLD activation was studied. Although mercury chloride (25 μM) caused a 6-fold increase in PLD activation at 30 minutes of treatment in BPAECs in the calcium-enriched MEM as compared to the untreated control cells, mercury chloride (25 μM) did not induce activation of PLD in the calcium-deficient medium (Figure 3A). However, upon addition of calcium (0.1–2.5 mM) to the calcium-deficient medium surrounding the cells, a significant and dose-dependent activation of PLD resulted in BPAECs treated with mercury chloride as compared to the same in the untreated control cells in the calcium-deficient medium alone. When 2.5 mM of calcium was added to the calcium-deficient medium of ECs, only a 4-fold increase in the mercury chloride-induced activation of PLD in BPAECs was observed, as compared to the same in the cells exposed to mercury chloride in the calcium-enriched medium (Figure 3A). In the calcium-enriched medium, thimerosal (25 μM) resulted in a 4-fold activation of PLD after 30 minutes of treatment in BPAECs as compared to the same in the untreated control cells under identical conditions (Figure 3B). In contrast, after 30 minutes of treatment, thimerosal (25 μM) did not cause any activation of PLD in BPAECs in the calcium-deficient medium as compared to the same in the untreated control cells under identical conditions. Following the introduction of calcium (0.1–2.5 mM) to the calcium-deficient medium of cells, a significant and dose-dependent increase in thimerosal-induced PLD activation was evident. When calcium, at concentrations of 1.5–2.5 mM, was added to the calcium-deficient medium of cells, the extent of thimerosal-induced PLD activation in BPAECs was similar to that noticed in the cells treated with thimerosal in the calcium-enriched medium (Figure 3B). Although methylmercury (10 μM) caused an 8-fold increase in PLD activation after 30 minutes of treatment in BPAECs in the calcium-enriched MEM as compared to the same in the untreated control cells, methylmercury (10 μM) did not cause activation of PLD in cells in the calcium-deficient medium (Figure 3C). Following addition of calcium (0.1–2.5 mM) to the calcium-deficient medium of BPAECs, a significant and dose-dependent activation of PLD resulted in cells exposed to methylmercury for 30 minutes as compared to the same in the untreated control cells in the calcium-deficient medium alone. After a 2.5 mM dose of calcium was added to the calcium-deficient medium, only a 4-fold increase in methylmercury-induced PLD activation in BPAECs was observed as compared to the cells treated with methylmercury in the calcium-enriched medium (Figure 3C).

Calmodulin Inhibitors Attenuate Mercury-induced PLD Activation

Calmodulin is the chief ligand for cellular calcium, and upon binding with calcium, the calcium/calmodulin complex is formed. This complex has properties of modulating calcium/calmodulin-dependent cellular processes at specific targets. ³⁴ In vascular ECs, one such important target is the calcium/calmodulin-dependent protein kinase II (CaM Kinase II), which is emerging as a key player in EC pathophysiological cell signaling. ³⁵ As our earlier experiments of the current work showed that both intracellular and extracellular calcium regulated mercury-induced PLD activation in BPAECs, we hypothesized here that calmodulin could possibly play a regulatory role in mercury-induced PLD activation in ECs. To study the effect of calmodulin inhibition, we used the calmodulin antagonists calmidazolium chloride and trifluoperazine. Prior to exposure of BPAECs to the selected mercury compound(s) (25 μM mercury chloride, 25 μM thimerosal, and 10 μM methylmercury chloride), cells were pretreated for 1 hour with MEM or MEM containing the calmodulin antagonists calmidazolium 50 μM and/or trifluoperazine 50 μM, and then co-treated for 30 minutes with the mercury compound(s) and calmodulin antagonist(s). Trifluoperazine, a commonly used antipsychotic agent and calmodulin antagonist, offered effective and significant attenuation of mercury chloride-, thimerosal-, and methylmercury chloride-induced PLD activation (80%, 65%, and 38%, respectively) in BPAECs (Figure 4A). Similarly, calmidazolium significantly inhibited mercury chloride-, thimerosal-, and methylmercury chloride-induced PLD activation (76%, 94%, and 89%, respectively) (Figure 4B).

Mercury Stimulates Cellular Uptake of Calcium

As the earlier experiments of this study revealed that both intracellular and extracellular calcium and calmodulin regulated mercury-induced activation of PLD in ECs, here, the effect of mercury chloride on the calcium uptake (influx) by BPAECs was investigated. As shown in Figure 5, mercury chloride (25 μM), following 30 minutes of exposure, caused a significant increase in the cellular influx of calcium as compared to the same in the untreated control cells. These results clearly revealed that mercury elevated the intracellular calcium in ECs by enhancing cellular calcium uptake.

L-type Calcium Channel Blockers Attenuate Mercury-induced PLD Activation

The elevation of intracellular calcium levels through the activation of calcium channels as a mechanism of mercury-induced toxicity in several mammalian cell systems has been established. ^26,38,39 The cytotoxicity of mercury in cellular models including the neurons has been shown to be protected by voltage-gated calcium channel blockers. ^38–41 Earlier, we reported that mercury-induced PLD activation is also regulated by ROS and thiol-redox perturbation in ECs. ⁷ The earlier experiments in the current study revealed the role of calcium in mercury-induced PLD activation in BPAECs. Despite the reports made on the absence and/or expression of voltage-gated L-type calcium channels in vascular ECs, the documented antioxidant properties of L-type calcium channel blockers in cardiovascular protection have directed us to investigate whether these channel blockers would attenuate mercury-induced PLD activation in BPAECs. ^23,42 Therefore, here, the effect of 3 well-established L-type calcium channel blockers (nifedipine, nimodipine, and diltiazem) on mercury-induced PLD activation was investigated. Prior to exposure of cells to the chosen mercury compound(s) (25 μM mercury chloride, 25 μM thimerosal, and 10 μM methylmercury chloride), cells were pretreated for 1 hour with MEM or MEM containing the selected L-type calcium channel blocker (10 μM) and then treated with the mercury compound(s) in the presence of the channel blocker(s) for 30 minutes. Nifedipine, a widely used angina therapeutic and L-type calcium channel blocker, offered effective and significant attenuation of mercury chloride-, thimerosal-, and methylmercury chloride induced PLD activation (58%, 25%, and 37%, respectively) in BPAECs (Figure 6A). Nimodipine also offered effective and significant inhibition of thimerosal- and methylmercury chloride-induced PLD activation (43% and 52%, respectively), but it only caused a marginal attenuation of mercury chloride-induced enzyme activation (23%) in BPAECs (Figure 6B). Similarly, diltiazem also offered effective and significant inhibition of thimerosal- and methylmercury chloride-induced PLD activation (70% and 67%, respectively), but it caused only a minimal attenuation of mercury chloride-induced enzyme activation (22%) in BPAECs (Figure 6C).

L-type Calcium Channel Blockers Protect Against Mercury-induced Cytotoxicity

The protective role of L-type calcium channel blockers as antioxidants against oxidative stress in vascular ECs has been documented. ^21,23,43,44 Earlier, we demonstrated that mercury-induced PLD activation in vascular ECs is regulated by ROS and depletion of thiols, which confirms that mercury induces oxidative stress and thiol-redox imbalance in vascular ECs. ⁷ In addition, the earlier experiments of the current study demonstrated that L-type calcium channel blockers attenuated mercury-induced PLD activation in BPAECs. Taking these findings into account, we investigated whether L-type calcium channel blockers would protect against mercury-induced cytotoxicity (LDH release) in BPAECs. As shown in Figures 7A–7C, 1 of the most commonly used L-type calcium channel blockers, diltiazem (10 μM), significantly protected against cytotoxicity induced by mercury chloride (10 μM), thimerosal (10 μM), and methymercury chloride (5 μM). Diltiazem offered ∼50% protection against methymercury-induced cytotoxicity in BPAECs (Figure 7C). Nifedipine (10 μM) offered ∼80% protection against methylmercury-induced cytotoxicity in BPAECs (Figure 7D).

Discussion

The present study revealed that mercury in 3 different forms (inorganic mercury chloride, pharmaceutical thimerosal, and organic methylmercury) activated PLD in lung vascular ECs through calcium and calmodulin regulation. The study further demonstrated that L-type calcium channel blockers attenuated mercury-induced activation of PLD and protected against the cytotoxicity of mercury in ECs. Phospholipase D, which belongs to the class of phospholipid hydrolases, is a lipid signaling enzyme, ubiquitously present in mammalian cells, that preferentially hydrolyzes PC to generate PA and choline. Phosphatidic acid is further metabolically converted to either DAG by phosphatidate phospho-hydrolase or LPA by phospholipase A₁/A₂, wherein PA, LPA, and DAG are potent cellular bioactive lipid second messengers. ¹⁷

Oxidants have been shown to stimulate PLD activity through upstream regulation by several protein kinases in vascular ECs. ^17,20 Nevertheless, the regulation of agonist-induced activation of PLD appears to be complicated, involving complex signaling pathways. ^7,17 Earlier, we reported that mercury induced the activation of PLD in vascular ECs through ROS generation and thiol-redox perturbation. ⁷ Although the current study did not focus on the role of protein kinases and G proteins, their involvement in mercury-induced PLD activation in BPAECs is not ruled out.

Levels of calcium in the cytosol of ECs play a unique role in EC signal transduction, leading to the proliferation and motility of ECs. ⁴⁵ Diverse EC functions in response to agonist stimulation (hormones and transmitters) are governed by calcium signaling due to elevated cytosolic calcium levels. ⁴⁶ Elevated cytosolic calcium levels and associated calcium signaling in ECs are linked to the macromolecular hyperpermeability and endothelial barrier dysfunction. ^24,25 Elevated calcium levels in lung ECs due to the activation of calcium channels are linked to perturbations in lung function. ⁴⁷

A role for calcium in the action of phospholipase D in vascular endothelium has been reported. ⁴⁸ In Chinese hamster ovary (CHO) cells expressing the cholecystokinin-A receptor, receptor-mediated PLD activation has been shown to be dependent upon calcium increase. ³³ Phospholipase D activation induced by synthetic peptide (WKYMVM) in mouse dendritic cells has been observed to be regulated by an increase in calcium and PKC activation. ³² The activation of P2X7 receptors has been shown to cause calcium influx and activation of calcium-dependent PLD in RBA-2 astrocytes. ³¹ Our laboratory has shown the involvement of calcium in the activation of PLD in ECs by various agonists, including pro-oxidants such as vitamin C. ⁴⁹ The results of the current study clearly demonstrate that extracellular calcium, in contrast to intracellular calcium, played a greater role in mercury-induced activation of PLD in BPAECs. The results of the present study are also in agreement with those reported by others, suggesting that calcium plays a pivotal role in PLD activation in ECs induced by agonists including mercury.

Recently, we have shown that calcium is also involved in mercury-induced activation of phospholipase A₂ (PLA₂) and in cyclooxygenase (COX)-catalyzed formation of arachidonic acid (AA) metabolites in ECs. ³⁰ We have also shown that the pro-oxidant vitamin C-induced activation of PLD is regulated by both intracellular and extracellular calcium and also upstream activation of calcium-dependent cytosolic PLA₂ (cPLA₂), COX, lipoxygenase, and the formation of AA metabolites. ⁴⁹ Therefore, it is conceivable that upstream activation of cPLA₂, through calcium regulation and formation of AA metabolites, might have additionally played a signaling role in mercury-induced activation of PLD in ECs, as observed in the present study.

An increase in the influx of calcium has been shown to be associated with agonist-induced activation of PLD in ECs. ⁴⁸ Methylmercury has been identified to increase calcium uptake by the neuronal cells, which is associated with the activation of PLA₂. ²⁷ However, neither the mechanism of the increase of calcium influx nor the association of calcium influx, especially with PLD activation in cellular systems due to mercury exposure, has been established. In the present study, mercury chloride was shown to increase calcium influx in BPAECs. Environmental toxicants such as lead and mercury have been shown to block calcium channels and use those channels to gain access into the cells, thus leading to toxicity. ⁵⁰ Methylmercury has been shown to alter calcium homeostasis in the rat cerebellar granule cells through interaction with the L-, N-, and Q-type calcium channels, which apparently facilitate the uptake of the toxicant by neurons. ⁵¹ Reports suggest that heavy metals such as mercury and lead bind to the inositol polyphosphate (IP) receptor, which may lead to altered intracellular calcium concentrations and thus to perturbations in neuronal activity. ⁵² Agonists (hormones and chemicals) are known to elevate levels of cytosolic calcium in vascular ECs, which leads to calcium signaling events. ⁴⁶

Macrovascular ECs have been determined not to express L-type calcium channels. ⁵³ It has been documented that voltage-dependent calcium channels are not important in ECs as compared to their existence and role in vascular smooth muscle cells. ⁵⁴ Calcium entry and elevation of cytosolic calcium in vascular ECs is operated by different mechanisms than the action of voltage-gated channels, which include the highly calcium-selective ion channels and nonselective calcium-permeable cation channels. ⁵⁵ Those permeability-coupled calcium channels belong to the family of transient receptor potential channels. ²⁵ Thimerosal has been shown to cause the release of calcium from intracellular reserves and enhance the influx of extracellular calcium into the cells involving the thiols. ⁵⁶ It is conceivable to argue that calmodulin could also be a target for mercury, probably through thiol involvement in leading to elevated intracellular calcium levels and associated PLD activation in ECs by taking into consideration the role of calcium/calmodulin-dependent CaM kinase II in ECs. ³⁵ Cytotoxicity of inorganic mercury in lymphocytes has been shown to be associated with ROS and alterations in calcium homeostasis. ⁴⁰ We have also shown that mercury activates PLD in ECs through alterations in the thiol redox and induction of oxidative stress. ⁷ It can be postulated that mercury exposure might have caused oxidative stress in addition to direct binding of mercury to thiols of unique calcium channels and thus, might have led to the activation of these calcium channels, elevated intracellular calcium levels, and activation of PLD in vascular ECs.

The current study revealed 2 striking observations: (1) L-type calcium channel blockers attenuated mercury-induced PLD activity in BPAECs and (2) L-type calcium channel blockers protected against mercury-induced cytotoxicity in BPAECs. Calcium channel blockers have been shown to offer protection against methylmercury toxicity. ⁴¹ Nifedipine has been shown to attenuate the methylmercury-induced influx of calcium through the plasma membrane in NG108-15 cells. ³⁸ Calcium channel blockers such as flunarizine, nifedipine, nicardipine, and verapamil have been shown to protect against methylmercury-induced neurological disorders in rats in vivo. ⁴¹ Mercury-induced cytotoxicity in neurons has been shown to be protected by flunarizine. ³⁹ In cells other than ECs, which possess L-type calcium channels, the reported protection offered by the L-type calcium channel blockers against mercury toxicity is linked with activation of L-type calcium channels and elevation of cytosolic calcium.

Calcium channel antagonists used in the treatment of hypertension and angina pectoris as powerful vasodilators are broadly divided to in 3 categories: (1) dihydropyridines (nifedipine, amlodipine, and felodepine), (2) phenylalkylamines (verapamil), and (3) benzothiazepines (dilitiazem). ⁵⁷ Nifedipine has been shown to protect against apoptosis in ECs and vascular inflammation and offer improvement of endothelial function in patients with hypertension and diabetes. ²³ Different clinical investigations have revealed that calcium channel blockers protect against endothelial dysfunction. ⁵⁸ Treatment, combined with rennin-angiotensin system inhibitors and calcium channel blockers, has been recommended for improved endothelial functions. ³⁷ Thus, these reports support the idea that L-type calcium channel blockers are beneficial to the vascular endothelium.

L-type calcium channel blockers have been shown to possess antioxidant properties. ^23,43,58 Our current observation that L-type calcium channel agonists protected against mercury-induced PLD activation and cytotoxicity in BPAECs, which are reported to lack L-type calcium channels, could be attributed to the antioxidant actions exhibited by L-type calcium channel blockers. One of the remarkable properties of calcium channel blockers is their metal-complexing/chelating property. ^42,59 The ability of calcium channel blockers to bind/complex with transition metals is also suggested to facilitate their antioxidant effects. ⁴² Recently, nifedipine has been shown to form a mononuclear complex with the transition metal copper. ⁶⁰ These reports support that calcium channel blockers do chelate the transition metals and further suggest that they also protect against heavy metal toxicity, including mercury cytotoxicity and PLD activation in ECs by metal complexation (Schema I). Thus, the L-type calcium channel blockers appear as promising therapeutic agents for protection against the mercury-induced toxic effects on vascular endothelium leading to vasculotoxicity.

Agonists—including the hormones bradykinin, angiotensin II, serotonin, acetylcholine, inositol triphosphate, and diacylglycerol—have been shown to cause the initial elevation of intracellular calcium levels in ECs, which are involved in the downstream signaling cascades. ⁴⁶ The regulation of EC barrier function by store-operated calcium entry and calcium signaling has been emphasized. ^61,62 The role of calcium channels and entry of calcium from the extracellular stores has been implicated in pulmonary EC permeability and chronic pulmonary hypertension. ⁴⁷ Several toxicants, including mercury, have been shown to cause vasculotoxicity, endothelial dysfunction, and pulmonary hypertension. ⁶³ We have also shown that mercury causes toxicity to pulmonary ECs. ¹⁶ The results of the present study highlight the importance of calcium and calmodulin in the mercury-induced activation of PLD and generation of the bioactive lipid signal mediator PA in the ECs as a mechanism of vasculotoxicity of mercury, with an emphasis on mercury-induced endothelial dysfunction and pulmonary hypertension (Schema I).