The ether lipid ET-18-OCH₃ increases cytosolic Ca²⁺ concentrations in Madin Darby canine kidney cells

Abstract

1. The effect of the ether lipid 1-O-octadecyl-2-O-methyl-sn-glycero-3-phosphorylcholine (ET-18-OCH₃) on the intracellular free Ca²⁺ concentration ([Ca²⁺]_i) in Madin Darby canine kidney (MDCK) cells was studied using fura-2 as the Ca²⁺ probe. In Ca²⁺ medium, ET-18-OCH₃ induced a significant rise in [Ca²⁺]_i at concentrations between 10–100 μM with a concentration-dependent delay of 45–175 s. The [Ca²⁺]_i signal was composed of a gradual rise and a sustained plateau.

2. In Ca²⁺-free medium, ET-18-OCH₃ (10–100 μM) induced a Ca²⁺ release from internal Ca²⁺ stores with a concentration-dependent delay of 45–175 s. This discharge of internal Ca²⁺ triggered capacitative Ca²⁺ entry in a concentration-dependent manner. This capacitative Ca²⁺ entry was not inhibited by econazole (25 μM), 1-[β-[3-(4-methoxyphenyl)propoxy]-4-methoxyphenethyl]-1H-imidazole hydrochloride (SKF96365; 50 μM), nifedipine (10 μM), verapamil (10 μM), diltiazem (10 μM) and cadmium (0.5 μM).

3. Methyl 2-(phenylthio)ethyl-1,4-dihydro-2,4,6-trimethylpyridine-3,5-dicarboxylate (PCA-4248), a platelet-activating factor (PAF) receptor antagonist, inhibited 25 μM ET-18-OCH₃-induced [Ca²⁺]_i rise in a concentration-dependent manner between 1–20 μM, with 20 μM exerting a complete block.

4. The [Ca²⁺]_i rise induced by ET-18-OCH₃ (25 μM) was not altered when the production of inositol 1,4,5-trisphosphate (IP₃) was suppressed by the phospholipase C inhibitor U73122 (2 μM), but was partly inhibited by the phospholipase D inhibitor propranolol (0.1 mM) or the phospholipase A₂ inhibitor aristolochic acid (20–40 μM).

5. In Ca²⁺-free medium, pretreatment with 25 μM ET-18-OCH₃ completely depleted the endoplasmic reticulum Ca²⁺ pump inhibitor thapsigargin-sensitive Ca²⁺ store. In contrast, pretreatment with thapsigargin abolished 0.1 mM ATP-induced [Ca²⁺]_i rise without altering the ET-18-OCH₃-induced [Ca²⁺]_i rise. This suggests that ET-18-OCH₃ depleted thapsigargin-sensitive Ca²⁺ stores and also released Ca²⁺ from thapsigargin-insensitive stores. The thapsigargin-insensitive stores involve mitochondria because the mitochondria uncoupler carbonylcyanide m-chlorophenylhydrazone (CCCP; 2 μM) induced a release of mitochondrial Ca²⁺ which was abolished by pretreatment with 25 μM ET-18-OCH₃.

6. ET-18-OCH₃ (25 μM) induced a significant Mn²⁺ quench of fura-2 fluorescence at 360 nm excitation wavelength confirming that ET-18-OCH₃ induced capacitative Ca²⁺ entry. La³⁺ (0.1 mM) or Gd³⁺ (50 μM) abolished the ET-18-OCH₃-induced Mn²⁺ quench and [Ca²⁺]_i rise.

7. Our data imply that ET-18-OCH₃ induced a [Ca²⁺]_i rise in MDCK cells by activating PAF receptors leading to an internal Ca²⁺ release followed by capacitative Ca²⁺ entry. Phospholipase D and phospholipase A₂, but not phospholipase C, might be involved in mediating the capacitative Ca²⁺ entry. La³⁺ abolished the ET-18-OCH₃-induced [Ca²⁺]_i rise presumably by inhibiting PAF receptors.

Introduction

A number of synthetic ether lipids with structure similar to platelet-activating factor (PAF) have been reported to be DNA-noninteractive anti-tumour drugs (Berdel, 1991). Among these drugs, 1-O-octadecyl-2-O-methyl-sn-glycero-3-phosphorylcholine (ET-18-OCH₃) is the most commonly used for the investigation of effects of this category of either lipids at the cellular level (Mollinedo et al., 1993). The primary cellular target for ET-18-OCH₃ was previously thought to be the plasma membrane proteins (Paltauf, 1994; Boggs et al., 1995). For example, ET-18-OCH₃ inhibits Na⁺-K⁺-ATP pump in bovine brain (Zheng et al., 1990), protein kinase C in HL-60 cells (Berkovic et al., 1994), and phosphatylcholine synthesis in macrophage-like cells (Boggs et al., 1995). Additionally, ET-18-OCH₃ also increases intracellular free Ca²⁺ concentration ([Ca²⁺]_i) in normal and tumour cells (Seewald et al., 1990; Lohmeyer & Workman, 1993; Bergmann et al., 1994; Brinkmeier et al., 1996; Alonso et al., 1997), and induces apoptosis in several human leukaemia cells (Mollinedo et al., 1993; 1997; Diomede et al., 1993). But how exactly ET-18-OCH₃ acts to trigger so many different cellular events is not clear at all.

Efforts have been exerted to investigate whether the effect of ET-18-OCH₃ on [Ca²⁺]_i is correlated to its effect on apoptosis. The idea is that a prolonged elevation in [Ca²⁺]_i could lead to apoptosis (McConkey & Orrenius, 1996). Two studies performed in several normal and tumour cells suggest that these two effects of ET-18-OCH₃ appear to be dissociated (Brinkmeier et al., 1996; Alonso et al., 1997), i.e. under the experimental conditions that the ET-18-OCH₃-induced [Ca²⁺]_i rise is suppressed, the apoptosis effect of this drug is not affected.

The mechanism underlying the ET-18-OCH₃-induced [Ca²⁺]_i rise is not completely understood. Alonso and coworkers (1997) showed that in human neutrophils ET-18-OCH₃ induces a robust [Ca²⁺]_i transient with a rapid rise and decline through stimulation of PAF receptors, because the [Ca²⁺]_i rise is suppressed by a PAF receptor antagonist. The authors also demonstrated that internal Ca²⁺ release and external Ca²⁺ influx both contribute to the ET-18-OCH₃-induced [Ca²⁺]_i rise, but neither the source of the internal Ca²⁺ nor the identity of the Ca²⁺ influx pathway was investigated. In contrast, in a neuroblastoma cell line Brinkmeier et al. (1996) observed that ET-18-OCH₃ induces a Ca²⁺ signal with a delayed rise and a persistently elevated plateau, which is abolished by pretreatment with La³⁺ or Gd³⁺. Based on the lanthanide inhibition, the authors conclude that ET-18-OCH₃ induces a [Ca²⁺]_i rise by exclusively opening plasma membrane Ca²⁺ channels. However, it is not known which types of Ca²⁺ channels are involved. Further, the possibility that ET-18-OCH₃ might release internal Ca²⁺ is not examined.

In the present study we investigated the effect of ET-18-OCH₃ on [Ca²⁺]_i in Madin Darby canine kidney (MDCK) cells, a non-tumour cell line, in more detail than the previous studies. We found that ET-18-OCH₃ induced a rise in [Ca²⁺]_i by activating PAF receptors leading to a release of Ca²⁺ from both thapsigargin-sensitive and -insensitive internal Ca²⁺ stores. This discharge of internal Ca²⁺ triggered capacitative Ca²⁺ entry (Putney & Bird, 1993) which was blocked by lanthanides. Lanthanides might also directly inhibit PAF receptors. Phospholipase D and phospholipase A₂, but not phospholipase C, might be involved in mediating the [Ca²⁺]_i rise.

Methods

Cell culture

MDCK cells obtained from American Type Culture Collection (CRL-6253, MD, U.S.A) were cultured in Dulbecco's modified Eagle medium (DMEM) supplemented with 10% heat-inactivated foetal bovine serum, 100 U ml⁻¹ penicillin and 100 μg ml⁻¹ streptomycin at 37°C in 5% CO₂-containing humidified air.

Solutions

Ca²⁺ medium (pH 7.4) contained (in mM): NaCl 140; KCl 5; MgCl₂ 1; CaCl₂ 2; HEPES 10; glucose 10. Ca²⁺-free medium contained no Ca²⁺ plus 1 mM EGTA. ET-18-OCH₃ was dissolved in dimethyl sulphoxide (DMSO) as a 10 mM stock solution. PCA-4248 and thapsigargin were dissolved in ethanol as a 20 mM stock solution. The experimental solution contained 0–1% of solvent (DMSO or ethanol) which did not affect [Ca²⁺]_i (n=3).

Fluorescence measurements

Trypsinized cells (10⁶ ml⁻¹) were loaded with 2 μM 1-[2-(5-carboxyoxazol-2-yl ) -6-aminobenzofuran-5-oxy] -2- (2′-amino-5′-methylphenoxy)-ethane-N,N,N,N-tetraacetic acid pentaacetoxymethyl ester (fura-2/AM) for 30 min at 25°C in DMEM. Cells were washed and resuspended in Ca²⁺ medium and were washed every hour during experiments to minimize extracellular dye. Fura-2 fluorescence measurements were performed in a water-jacketed cuvette (25°C) with continuous stirring; the cuvette contained 1 ml of medium and 0.5 million of cells. Fluorescence was monitored with a Shimadzu RF-5301PC spectrofluorophotometer (Japan) by continuously recording excitation signals at 340 and 380 nm and emission signal at 510 nm at 1-s intervals. Maximal and minimal fluorescence values were obtained by adding TX-100 (0.1%) and EGTA (20 mM) sequentially at the end of an experiment. The ratio of excitation signals at 340 and 380 nm was used to calculate [Ca²⁺]_i as described previously (Grynkiewicz et al., 1985). Mn²⁺ quench experiments were performed in Ca²⁺ medium containing MnCl₂ (50 μM) by recording excitation signal at 360 nm and emission signal at 510 nm continuously at 1-s intervals. Our previous studies have shown that trypsinized cells prepared by our protocol respond to stimulation with ATP (Jan et al., 1998a), bradykinin (Jan et al., 1998b) or thapsigargin (Jan et al., 1999c) similarly to cells attached to coverslips. We decided to use trypsinized cells because this procedure is easier and less time-consuming. All experiments were performed at room temperature (25°C).

Materials

The reagents for cell culture were from Gibco (NY, U.S.A.). Fura-2/AM was from Molecular Probes (OR, U.S.A.). ET-18-OCH₃, (±)1-O-hexadecyl-2-O-methylglycero-3-phosphorylcholine (2-O-methyl PAF), and PCA-4248 were from Biomol (Plymouth Meeting, PA, U.S.A.). The other reagents were from Sigma (MO, U.S.A.).

Statistics

All values are reported as means±s.e.mean of 3–4 experiments. Statistical comparisons were determined by using the Student's paired t-test, and significance was accepted when P<0.05.

Results

ET-18-OCH₃ at concentrations between 10–100 μM induced a delayed, gradual rise in [Ca²⁺]_i in the presence of extracellular Ca²⁺. A concentration of 1 μM had no effect. Representative traces are shown in Figure 1A. The concentration-response plot is shown in Figure 1B. The response did not saturate at 100 μM. Because the 100 μM ET-18-OCH₃-induced [Ca²⁺]_i rise reached a peak value as great as ∼1.5 μM followed by a sustained plateau, higher concentrations were not tested to avoid cell damage. The rise of the Ca²⁺ signal was slower (i.e. delay time was shorter) in response to lower concentrations of ET-18-OCH₃, and the relationship between concentration and the delay time before [Ca²⁺]_i rises is plotted in Figure 1C. For example, in the [Ca²⁺]_i rise induced by 100 μM ET-18-OCH₃, the delay time was 48±3 s (n=3; P<0.05), while that induced by 10 μM ET-18-OCH₃ had a delay time of 171±4 s (n=3; P<0.05). At a concentration of 25 μM, 2-O-methyl PAF, another PAF agonist, induced a rise in [Ca²⁺]_i with a similar shape and magnitude as that induced by 50 μM ET-18-OCH₃ (n=5; data not shown).

Figure 1.

(A) Concentration-dependent effects of ET-18-OCH₃ on [Ca²⁺]_i. Concentration of ET-18-OCH₃ was 100 μM in trace a, 50 μM in trace b, 25 μM in trace c, 10 μM in trace d and zero in trace e. ET-18-OCH₃ was applied at 30 s. The experiments were performed in Ca²⁺ medium. Traces are typical of 3–4 experiments. (B) A concentration-response plot of the ET-18-OCH₃-induced Ca²⁺ signals shown in (A). The y axis is the area under the curve (percentage of maximum response). The x axis is the concentration of ET-18-OCH₃ in logarithmic scale. The data are means±s.e.mean of 3–4 experiments. (C) Relationship between the delay time (s) before [Ca²⁺]_i (nM) rises. The data are means±s.e.mean of 3–4 experiments.

We next examined whether ET-18-OCH₃ releases Ca²⁺ from internal stores. Figure 2A shows that in Ca²⁺-free medium (no added Ca²⁺ plus 1 mM EGTA) 10–100 μM ET-18-OCH₃ induced significant [Ca²⁺]_i rises. In MDCK cells, a release of Ca²⁺ from the endoplasmic reticulum (ER) store often induces capacitative Ca²⁺ entry (Jan et al., 1998a1998b1998c; 1999a1999b1999c1999d). Thus, we next examined whether ET-18-OCH₃ triggers capacitative Ca²⁺ entry. Figure 2A shows that CaCl₂ (3 mM) added at the time point of 360 s when the internal Ca²⁺ stores had been substantially discharged by ET-18-OCH₃ pretreatment induced significant capacitative Ca²⁺ entry with a magnitude proportionally correlating to the concentration of ET-18-OCH₃, i.e. a larger concentration of ET-18-OCH₃ induces a larger capacitative Ca²⁺ entry. We examined the effects of several Ca²⁺ channel blockers on the ET-18-OCH₃-induced capacitative Ca²⁺ entry. The blocker was added 30 s prior to CaCl₂. These blockers were 10 μM of nifedipine, diltiazem and verapamil, 25 μM econazole, 50 μM SKF96365, and 0.5 mM cadmium. None of the drugs tested had any inhibition (n=3; data not shown). The effect of La³⁺ (0.1 mM) on capacitative Ca²⁺ entry could not be tested using this experimental protocol because La³⁺ is chelated by EGTA. Addition of EGTA (1 mM) is needed for measuring capacitative Ca²⁺ entry because the ET-18-OCH₃-induced [Ca²⁺]_i rise remained elevated at ∼400 nM for a sustained period of time (>6 min) in nominally Ca²⁺-free (no EGTA) medium, thus addition of 3 mM CaCl₂ did not induce a further rise in [Ca²⁺]_i (data not shown). The concentration-response plot of the capacitative Ca²⁺ entry suggests that the response saturates at a concentration of 50 μM (Figure 2B) with an EC₅₀ of about 32 μM. Note that the [Ca²⁺]_i rises induced by all concentrations of ET-18-OCH₃ have a similar magnitude in terms of the area under the curve and the net peak value (∼40 nM). However, these responses are different in the delay time before [Ca²⁺]_i rises. Thus, similar to the responses observed in the presence of extracellular Ca²⁺, in the absence of extracellular Ca²⁺ the ET-18-OCH₃-induced internal Ca²⁺ release also has a delay time depending on the concentration of ET-18-OCH₃, i.e. a larger concentration of ET-18-OCH₃ induces a Ca²⁺ release with a shorter delay time. The relationship between delay time and concentration is plotted in Figure 2C.

Figure 2.

(A) ET-18-OCH₃-induced capacitative Ca²⁺ entry. These experiments were performed in Ca²⁺-free medium. Capacitative Ca²⁺ entry was induced by depleting intracellular Ca²⁺ stores with different concentrations of ET-18-OCH₃ (10–100 μM) in Ca²⁺-free medium (no added Ca²⁺ plus 1 mM EGTA) followed by addition of 3 mM CaCl₂. Trace e is the control CaCl₂ effect without ET-18-OCH₃ pretreatment. (B) A concentration-response plot of the ET-18-OCH₃-induced capacitative Ca²⁺ entry shown in (A). The y axis is the peak height of the [Ca²⁺]_i rise induced by addition of 3 mM CaCl₂. The x axis is the concentration of ET-18-OCH₃ in logarithmic scale. The data are means±s.e.mean of 3–4 experiments. (C) Relationship between delay time (s) before [Ca²⁺]_i rises and the concentration of ET-18-OCH₃. The data are means±s.e.mean of 3–4 experiments.

Because ET-18-OCH₃ is a synthetic analogue of platelet-activating factor (PAF), we tested whether the ET-18-OCH₃-induced [Ca²⁺]_i rise is downstream to activation of PAF receptors. We used PCA-4248, a PAF receptor antagonist (Ortega et al., 1990), to see if it could affect the ET-18-OCH₃-induced [Ca²⁺]_i rise. Figure 3A shows that pretreatment with PCA-4248 for 30 s inhibited the 25 μM ET-18-OCH₃-induced [Ca²⁺]_i rise in a concentration-dependent manner between 1–20 μM. Figure 3B shows the concentration-inhibition plot of the PCA-4248 effect. The areas under the curves of the five traces in Figure 3A were computed and compared. Maximum inhibition occurs at 20 μM PCA-4248 which inhibited 95±5% (n=3; P<0.05) of the control ET-18-OCH₃ response (in the absence of PCA-4248). The plot indicates an IC₅₀ of about 2.5 μM.

Figure 3.

Effect of PCA-4248 on the ET-18-OCH₃-induced [Ca²⁺]_i rise. (A) ET-18-OCH₃ (25 μM) was added at 50 s. PCA-4248 at a concentration between 1–20 μM was added at 20 s. Control trace represents the ET-18-OCH₃ response without PCA-4248 pretreatment. The experiments were performed in Ca²⁺ medium. Traces are typical of 3–4 experiments. (B) A concentration-response plot of the PCA-4248 inhibition of the ET-18-OCH₃-induced Ca²⁺ signal shown in (A). The y axis is the percentage of maximum inhibition which is exerted by 20 μM PCA-4248 (95±5% inhibition of control ET-18-OCH₃ response in terms of the area under the curve). The x axis is the concentration of PCA-4248 in logarithmic scale. The data are means±s.e.mean of 3–4 experiments.

We next examined whether ET-18-OCH₃ releases internal Ca²⁺ by first elevating cytosolic levels of IP₃. We have previously shown that the phospholipase C inhibitor U73122 blocks IP₃ production leading to an inhibition of ATP- or bradykinin-induced release of Ca²⁺ from thapsigargin-sensitive ER Ca²⁺ stores (Jan et al., 1998c). Because U73122 induces significant Ca²⁺ influx (Jan et al., 1998c), the following experiments were performed in Ca²⁺-free medium. Shown in Figure 4A is the [Ca²⁺]_i transient induced by ATP (0.1 mM). Figure 4B (solid trace) shows that pretreatment with U73122 (2 μM) for 320 s did not alter the resting [Ca²⁺]_i, but substantially reduced the ATP-induced [Ca²⁺]_i rise by 91±5% in net peak height (31±5 vs 350±16 nM; n=3; P< 0.05). This result most likely suggests that the IP₃ production via phospholipase C was significantly inhibited under this condition. When ET-18-OCH₃ (25 μM) was added subsequently at 440 s there occurred a [Ca²⁺]_i rise which was indistinguishable from the control ET-18-OCH₃ response (without U73122/ATP pretreatment; dashed trace). We also examined whether phospholipase D and phospholipase A₂ are involved in mediating the ET-18-OCH₃-induced [Ca²⁺]_i rise. We used propranolol to inhibit phospholipase D (Billah, 1989) and aristolochic acid to inhibit phospholipase A₂ (Rosenthal et al., 1989). Table 1 shows that in the presence of extracellular Ca²⁺, propranolol (0.1 mM) and aristolochic acid (20–40 μM) both significantly inhibited the [Ca²⁺]_i rise induced by 25 μM ET-18-OCH₃. However, these two inhibitors did not alter the [Ca²⁺]_i rise induced by ET-18-OCH₃ in the absence of extracellular Ca²⁺ (n=3; data not shown).

Figure 4.

Effect of inhibition of IP₃ production on the ET-18-OCH₃-induced internal Ca²⁺ release. (A) 0.1 mM ATP was added at 30 s. (B) U73122 (2 μM) was added at 30 s followed by ATP (0.1 mM) and ET-18-OCH₃ (25 μM) at 340 s and 440 s, respectively. These experiments were performed in Ca²⁺-free medium. Traces are typical of 3–4 experiments.

Table 1.

Effects of propranolol and aristolochic acid on the [Ca²⁺]_i rise induced by ET-18-OCH₃ (25  μM)

We went on to investigate whether ET-18-OCH₃ releases Ca²⁺ from the thapsigargin-sensitive ER store, the major internal Ca²⁺ store in MDCK cells (Jan et al. 1998a1998b1998c; 1999a1999b1999c1999d). Thapsigargin is a substance known to cause a passive leak of Ca²⁺ from the ER store by inhibiting the ER Ca²⁺ pump (Thastrup et al., 1990). Figure 5A shows that in the absence of extracellular Ca²⁺, after pretreatment with 25 μM ET-18-OCH₃ for 320 s thapsigargin hardly induced a rise in [Ca²⁺]_i (see Figure 5B for control thapsigargin response). This suggests that ET-18-OCH₃ completely discharged the thapsigargin-sensitive ER Ca²⁺ store. However, other Ca²⁺ stores also contribute to the ET-18-OCH₃ response because, as shown in Figure 5B, after thapsigargin had completely depleted the thapsigargin-sensitive ER Ca²⁺ store (evidenced by the observation that 0.1 mM ATP added after thapsigargin did not increase [Ca²⁺]_i), ET-18-OCH₃ still induced a rise in [Ca²⁺]_i normally (compared to Figure 5A).

Figure 5.

Effect of ET-18-OCH₃ on internal Ca²⁺ stores. (A) ET-18-OCH₃ (25 μM) was added at 30 s followed by thapsigargin (1 μM) at 350 s. (B) Thapsigargin (1 μM) was added at 30 s, followed by ATP (0.1 mM) and ET-18-CH₃ (25 μM) at 630 s and 760 s, respectively. These experiments were performed in Ca²⁺-free medium. Traces are typical of 3–4 experiments.

The thapsigargin-insensitive Ca²⁺ store in MDCK cells which could be examined by pharmacological tools is mitochondria. We have previously shown that the mitochondrial uncoupler carbonylcyanide m-chlorophenylhydrazone (CCCP; 2 μM) induced a rise in [Ca²⁺]_i (Jan et al., 1998c; 1999b), suggesting that in this cell mitochondria contain a significant amount of Ca²⁺ which could be released when mitochondria are uncoupled by CCCP. Thus, we investigated whether ET-18-OCH₃ releases Ca²⁺ from mitochondria. Figure 6A shows that in the absence of extracellular Ca²⁺, pretreatment with 25 μM ET-18-OCH₃ prevented CCCP (2 μM) from inducing a rise in [Ca²⁺]_i. However Figure 6B clearly demonstrates that CCCP induced a significant rise in [Ca²⁺]_i with a net peak height of 51±7 nM (n=4; P<0.05). Thus, these results suggest that ET-18-OCH₃ depletes the mitochondrial Ca²⁺ store. Figure 6B also shows that the ET-18-OCH₃-induced rise in [Ca²⁺]_i was not much altered after the mitochondrial Ca²⁺ had been depleted.

Figure 6.

Effect of ET-18-OCH₃ on mitochondrial Ca²⁺ stores. (A) In Ca²⁺-free medium, ET-18-OCH₃ (25 μM) was added at 30 s followed by CCCP (2 μM) at 410 s. (B) Similar to (A) except that CCCP was added at 30 s followed by ET-18-OCH₃ at 390 s. Traces are typical of 3–4 experiments.

To confirm that ET-18-OCH₃ induced capacitative Ca²⁺ entry, we next measured whether ET-18-OCH₃ caused Mn²⁺ quench of fura-2 fluorescence. Mn²⁺ enters cells through similar pathways as Ca²⁺, but quenches fura-2 fluorescence at all excitation wavelengths (Merritt et al., 1989). Thus, Ca²⁺ influx can be estimated by Mn²⁺ quench of fura-2 fluorescence at the Ca²⁺-insensitive excitation wavelength of 360 nm and emission wavelength of 510 nm. Figure 7A illustrates that 25 μM ET-18-OCH₃ induced a significant Mn²⁺ quench of fura-2 fluorescence after a delay of 120±8 s (n =3; P<0.05) (control; solid trace) which closely matches the delay time observed in Figure 1C. This result suggests that ET-18-OCH₃ induced considerable Ca²⁺ influx, in keeping with our data in Figure 2 that ET-18-OCH₃ induced significant capacitative Ca²⁺ entry. Because we have observed that La³⁺ is a potent blocker of capacitative Ca²⁺ entry in MDCK cells (Jan et al., 1998a, 1998c; 1999c, 1999d), we next examined the effect of La³⁺ on the ET-18-OCH₃-induced Mn²⁺ quench of fura-2 fluorescence. Figure 7A shows that the ET-18-OCH₃-induced Mn²⁺ quench of fura-2 fluorescence was abolished by pretreatment with 0.1 mM La³⁺ (dashed trace). Consistently, Figure 7B shows that pretreatment with 0.1 mM La³⁺ for 20 s totally suppressed 25 μM ET-18-OCH₃-induced [Ca²⁺]_i rise (trace f). La³⁺ (0.1 mM) also abolished the [Ca²⁺]_i rise induced by 25 μM 2-O-methyl PAF (n=3; data not shown). When La³⁺ is added during the ET-18-OCH₃ response, the inhibition of La³⁺ depends on the time it is added. Thus, when added at time points of 180 s (trace e), 205 s (trace d), 220 s (trace c) and 265 s (trace b), La³⁺ immediately stopped the rise of [Ca²⁺]_i and maintained the [Ca²⁺]_i at a persistently elevated phase. However, when added at the late phase of the [Ca²⁺]_i rise, for example at 475 s, La³⁺ had little effect on the [Ca²⁺]_i rise (trace a). Substitution of 50 μM Gd³⁺ for 0.1 mM La³⁺ produced identical results (data not shown).

Figure 7.

(A) The ET-18-OCH₃-induced Ca²⁺ influx detected by Mn²⁺ quench measurements. The experiments were performed in Ca²⁺ medium plus 50 μM MnCl₂. Solid trace: control response of ET-18-OCH₃ (25 μM), applied at 30 s. Dashed trace: La³⁺ (0.1 mM) was added at 0 s followed by ET-18-OCH₃ (25 μM) at 30 s. Excitation signal at 360 nm and emission signal at 510 nm were recorded. (B) Effect of La³⁺ on the ET-18-OCH₃-induced [Ca²⁺]_i rise. ET-18-OCH₃ was added at 30 s in traces a–f and La³⁺ (0.1 mM) was added at 475 s in trace a, 265 s in trace b, 220 s in trace c, 205 s in trace d, 180 s in trace e and 0 s in trace f. Traces are typical of 3–4 experiments.

Discussion

Although ET-18-OCH₃ has been found to cause a rise in [Ca²⁺]_i in a number of cells, its effect on Ca²⁺ signalling is not fully characterized and the precise underlying mechanisms are not completely clear. In the present study we investigated how ET-18-OCH₃ affects [Ca²⁺]_i in MDCK cells. Our results suggest the following mechanism for the ET-18-OCH₃-induced [Ca²⁺]_i rise: ET-18-OCH₃ activates PAF receptors leading to a release of Ca²⁺ from both thapsigargin-sensitive and -insensitive Ca²⁺ stores. This discharge of internal Ca²⁺ subsequently activates capacitative Ca²⁺ entry which is blocked by lanthanides. Lanthanides might also directly inhibit PAF receptors. Phospholipase D and phospholipase A₂, but not phospholipase C, might be involved in triggering the [Ca²⁺]_i rise.

We have found that ET-18-OCH₃ induces a significant rise in [Ca²⁺]_i between 10–100 μM. A similar effective concentration range was found for a neuroblastoma cell line (Brinkmeier et al., 1996) but a lower range was effective in human neutrophils (10 nM–10 μM) (Alonso et al., 1997). This discrepancy might be due to differences in cell type. We propose that ET-18-OCH₃ acts through activating PAF receptors because the ET-18-OCH₃-induced [Ca²⁺]_i rise is inhibited by PCA-4248, a PAF receptor antagonist, in a concentration-dependent manner. Our results are supported by the observations that 2-O-methyl PAF, another PAF receptor agonist, also induces a [Ca²⁺]_i rise in MDCK cells similarly to ET-18-OCH₃ (data not shown), and that in U937 cells and human neutrophils WEB-2170, a PAF receptor antagonist (Heuer et al., 1990), also blocks the ET-18-OCH₃-induced [Ca²⁺]_i rise (Alonso et al., 1997).

One of the unique characteristics of the ET-18-OCH₃-induced [Ca²⁺]_i rise in MDCK cells is that it has a concentration-dependent delay before the [Ca²⁺]_i starts to rise, which is not observed in the [Ca²⁺]_i responses induced by other agents such as ATP (Jan et al., 1998a), bradykinin (Jan et al., 1998b), the phospholipase C inhibitor U73122 (Jan et al., 1998c), econazole (Jan et al., 1999a), SKF96365 (Jan et al., 1999b) and thapsigargin (Jan et al., 1998c) in MDCK cells. This delay time is not altered by removal of extracellular Ca²⁺ (compare Figure 1C to Figure 2C), implying that this delay time is needed for Ca²⁺ to be released from internal stores in response to ET-18-OCH₃ stimulation. Because the ET-18-OCH₃-induced [Ca²⁺]_i rise is blocked by lanthanides and PCA-4248, the possibility that ET-18-OCH₃ induces a [Ca²⁺]_i rise by entering the cell and directly acting on intracellular sites is unlikely.

Our findings suggest that activation of phospholipase D and phospholipase A₂, but not phospholipase C, might be involved in mediating the [Ca²⁺]_i rise induced by ET-18-OCH₃ because the [Ca²⁺]_i rise is partly inhibited by the phospholipase D inhibitor propranolol and the phospholipase A₂ inhibitor aristolochic acid, but not the phospholipase C inhibitor U73122. Indeed, in Swiss 3T3 fibroblasts ET-18-OCH₃ was found to inhibit the IP₃ production induced by platelet-derived growth factor (PDGF) and [AlF₄]¹⁻ (Seewald et al., 1990). We found that phospholipase D and phospholipase A₂ appear to be involved in the regulation of the ET-18-OCH₃-induced extracellular Ca²⁺ influx instead of intracellular Ca²⁺ release because propranolol and aristolochic acid only inhibit the [Ca²⁺]_i rise induced by ET-18-OCH₃ in the presence, but not in the absence, of extracellular Ca²⁺. How ET-18-OCH₃ releases internal Ca²⁺ is not clear.

We have previously shown that U73122 (Jan et al., 1998c), econazole (Jan et al., 1999a) and SKF96365 (Jan et al., 1999b) all release Ca²⁺ from the thapsigargin-sensitive ER Ca²⁺ store presumably via inhibition of the ER Ca²⁺ pump. However, the action of ET-18-OCH₃ differs from that of these three substances at least in two aspects. First, the latter three substances are not known to act through a receptor on the plasma membrane. Second, ET-18-OCH₃ not only depletes the thapsigargin-sensitive ER Ca²⁺ store but also releases Ca²⁺ from mitochondria and possibly other stores, because in the absence of extracellular Ca²⁺, pretreatment with ET-18-OCH₃ prevents thapsigargin or CCCP from releasing Ca²⁺. Consistently, the [Ca²⁺]_i rise induced by ET-18-OCH₃ is not altered by pretreatment with either thapsigargin or CCCP, suggesting that both the thapsigargin-sensitive ER store and the CCCP-sensitive mitochondrial store contribute to the internal Ca²⁺ release induced by ET-18-OCH₃. The involvement of other stores cannot be excluded.

It is interesting that econazole (25 μM) and SKF96365 (50 μM) do not inhibit the capacitative Ca²⁺ entry induced by ET-18-OCH₃. We have recently shown that econazole partly inhibits the capacitative Ca²⁺ entry induced by thapsigargin (Jan et al., 1999c), cyclopiazonic acid (Jan et al., 1999a) and U73122 (Jan et al., 1998c); and that SKF96365 partly inhibited the capacitative Ca²⁺ entry induced by thapsigargin (Jan et al., 1999c) and 2,5-di-tert-butylhydroquinone (Jan et al., 1999d). Thus, the capacitative Ca²⁺ entry induced by ET-18-OCH₃ and that induced by the other agents appear to be different in nature.

Brinkmeier and coworkers (1996) also found a 10–25 s delay time before [Ca²⁺]_i starts to rise in response to ET-18-OCH₃ in a neuroblastoma cell line, but whether this delay depends on concentration is not shown. However, there is no delay time in the ET-18-OCH₃-induced [Ca²⁺]_i rise in other cells such as HL-60 cells, U937 cells and human neutrophils (Alonso et al., 1997). This discrepancy may be caused by differences in cell type. Brinkmeier and coworkers (1996) proposed that the delay time is required for ET-18-OCH₃ or its derivative(s) to integrate into the plasma membrane and open a Ca²⁺ channel based on the fact that La³⁺ and Gd³⁺ abolish the ET-18-OCH₃-induced [Ca²⁺]_i rise. However, the authors did not investigate whether ET-18-OCH₃ could release internal Ca²⁺ by removing extracellular Ca²⁺ as we did in this study. As we previously discussed, La³⁺ is unlikely to enter MDCK cells (Jan et al., 1998a), and because we have demonstrated that La³⁺ abolishes the capacitative Ca²⁺ entry induced by ATP (Jan et al., 1998a), thapsigargin (Jan et al., 1999c), 2,5-di-tert-butylhydroquinone (Jan et al., 1999d) and U73122 (Jan et al., 1998c) in MDCK cells, it is reasonable to believe that La³⁺ might also abolish the ET-18-OCH₃-induced capacitative Ca²⁺ entry. But this cannot explain why La³⁺ completely inhibits the ET-18-CH₃-induced Mn²⁺ entry and [Ca²⁺]_i rise because La³⁺ would not be expected to inhibit internal Ca²⁺ release. The inhibition of La³⁺ is not specific for ET-18-OCH₃ alone because La³⁺ also abolishes the [Ca²⁺]_i rise induced by another PAF agonist, 2-O-methyl PAF (data not shown). Thus, one possible interpretation is that La³⁺ might inhibit PAF receptors. This hypothesis gains support from the fact that La³⁺ might also inhibit ATP receptors in MDCK cells (Jan et al., 1998a).

Figure 7B shows that the La³⁺-sensitive capacitative Ca²⁺ entry is triggered early during the ET-18-OCH₃-induced Ca²⁺ signal because when added 40 s after the [Ca²⁺]_i starts to rise, La³⁺ is able to inhibit the rise of the signal. We have found a similar effect of La³⁺ on the phospholipase C inhibitor U73122-induced [Ca²⁺]_i rise in MDCK cells (Jan et al., 1999c). These results suggest that capacitative Ca²⁺ entry is initiated when internal Ca²⁺ stores are only partly depleted.

In HL-60 cells, Lohmeyer & Workman (1993) proposed that the ether lipid (ET-18-OCH₃ and its analogue SRI 62-834)-induced [Ca²⁺]_i rise should be transient (i.e. with an initial peak followed by a fast decline), and the previously reported elevated plateau of the [Ca²⁺]_i rise induced by SRI 62-834 (Lazenby et al., 1990) is due to external Ca²⁺ influx and leakage of internal quin-2 caused by extensive cell damage. In view of our results, it is unlikely that the elevated plateau in the 25 μM ET-18-OCH₃-induced [Ca²⁺]_i rise in MDCK cells could be attributed to cell damage-induced fura-2 leakage and external Ca²⁺ influx, because if it were the case, the ET-18-OCH₃-induced [Ca²⁺]_i rise would not be abolished by pretreatment with La³⁺ or PCA-4248.

Taken together, we have found several unique characteristics about the ET-18-OCH₃-induced [Ca²⁺]_i rise which have not been reported previously: (1) The delay time before the [Ca²⁺]_i rise is dependent on the concentration of ET-18-OCH₃. A higher concentration needs a shorter delay time. (2) ET-18-OCH₃ releases internal Ca²⁺ from both thapsigargin-sensitive ER stores and thapsigargin-insensitive, CCCP-sensitive mitochondrial stores. (3) The ET-18-OCH₃-induced [Ca²⁺]_i rise in the presence of external Ca²⁺ is composed of two sources. First, an IP₃-independent internal Ca²⁺ release followed by econazole- and SKF96365-insensitive capacitative Ca²⁺ entry. This capacitative Ca²⁺ entry might be modulated by phospholipase D and phospholipase A₂, but not by phospholipase C. And lastly, we suspect that lanthanides might directly inhibit PAF receptors. What remains to be answered is how ET-18-OCH₃ activation of PAF receptors causes a release of internal Ca²⁺ via a mechanism independent of the activities of phospholipases C, D and A₂. The possibility that lanthanides might inhibit PAF receptors needs further investigation.

Abbreviations

ATP

(adenosine 5′-triphosphate)

[Ca²⁺]_i

intracellular free Ca²⁺ concentration

DMEM

Dulbecco's modified Eagle medium)

ER

endoplasmic reticulum

ET-18-OCH₃

1-O-octadecyl-2-O-methyl-sn-glycero-3-phosphorylcholine

fura-2/AM

1-[2-(5-carboxyoxazol-2-yl)-6-aminobenzofuran-5-oxy]-2-(2′-amino-5′-methylphenoxy)-ethane-N,N,N,N-tetraacetic acid pentaacetoxymethyl ester

IP₃

inositol 1,4,5-trisphosphate

MDCK cells

Madin Darby canine kidney cells

2-O-methyl PAF

(±)1-O-hexadecyl-2-O-methylglycero-3-phosphorylcholine

PAF

platelet-activating factor

PCA-4248

methyl 2-(phenylthio)ethyl-1,4-dihydro-2,4,6-trimethylpyridine-3,5-dicarboxylate

SKF96365

1-[β-[3-(4-methoxyphenyl)propoxy]-4-methoxyphenethyl]-1H-imidazole hydrochloride

U73122

1-(6-((17β-3-methoxyestra-1,3,5(10)-trien-17-yl)amino)hexyl)-1H-pyrrole-2,5-dione