Introduction

Top Abstract Introduction Methods Results Discussion References

L--Lysophosphatidylcholine (palmitoyl) (LPC) is an amphipathic metabolite of membrane phosphatidylcholine; high concentrations of LPC have been found in ischemic hearts (Sobel et al., 1978; Shaikh and Downer, 1981; Corr et al., 1982; Otani et al., 1989; Sedlis et al., 1990). LPC applied externally increases intracellular Ca²⁺ concentration ([Ca²⁺]_i) in isolated cardiac myocytes (Sedlis et al., 1983; Woodley et al., 1991; Ver Donck et al., 1992; Hashizume and Abiko, 1996). Therefore, LPC is thought to be one of the causes of Ca²⁺ overload and arrhythmia during cardiac ischemia and reperfusion (Hoque et al., 1995; Hashizume et al., 1998). Magishi et al. (1996) reported that exogenous LPC induces a nonselective cation current (I_NSC) in cardiac myocytes, but the mechanism involved has not been elucidated.

Recently, Yokoyama et al. (2002) found that LPC increases [Ca²⁺]_i and membrane current in cultured human endothelial cells in a manner similar to that found in cardiac myocytes. Furthermore, they found that 3-hydroxy-3-methylglutaryl (HMG)-CoA reductase inhibitors such as fluvastatin, cerivastatin, and pravastatin inhibit the LPC-induced increase in [Ca²⁺]_i. HMG-CoA reductase inhibitors prevent the synthesis of mevalonic acid (MVA) from HMG-CoA, thereby reducing cholesterol synthesis. Therefore, it is used clinically to prevent the development of atherosclerosis. Recently, it has been shown that MVA is not only a precursor of cholesterol but also of numerous metabolites that are involved in important cell functions, including endothelial function, inflammatory responses, plaque stability, and thrombus formation (Corsini et al., 1999). We tested fluvastatin in cardiac myocytes and found that it inhibited LPC-induced I_NSC in guinea pig ventricular myocytes. We investigated the mechanism of LPC-induced I_NSC by using fluvastatin under the whole-cell voltage clamp.

	Methods

Top Abstract Introduction Methods Results Discussion References

Isolation of Cells. All experiments were performed according to the regulations of the Animal Research Committee of Fukushima Medical University. Single cardiac ventricular cells were isolated essentially by the method of Yazawa et al. (1990). Guinea pigs weighing 250 to 400 g were anesthetized by intraperitoneal injection of pentobarbital (30 mg/kg). The chest was opened under artificial ventilation, the aorta was cannulated in situ, and the heart was removed. Blood was washed out with Tyrode-HEPES solution and the heart was mounted in a Langendorff perfusion system. Tyrode-HEPES solution contained 140 mM NaCl, 5.4 mM KCl, 1.8 mM CaCl₂, 1 mM MgCl₂, 0.33 mM NaH₂PO₄, 5.5 mM glucose, and 5 mM HEPES, pH 7.4. The perfusate was changed to Ca²⁺-free Tyrode solution to stop the heartbeat and then to one containing 0.01% (w/v) collagenase (Wako, Osaka, Japan) and 0.002% (w/v) alkaline protease (Nagase, Tokyo, Japan). After about 20 min, the collagenase solution was washed out by perfusing with a high-K⁺/low-Cl solution [modified KB solution (Isenberg and Klöckner, 1982)]. The modified KB solution contained 70 mM KOH, 50 mM l-glutamic acid, 40 mM KCl, 20 mM taurine, 20 mM KH₂PO₄, 3 mM MgCl₂, 10 mM glucose, 0.2 mM EGTA, and 10 mM HEPES, pH 7.2. Incisions were made in the cardiac ventricular tissue in the modified KB solution and the tissue was shaken gently to isolate the cells. The cell suspension was stored at 4°C for later use.

Patch-Clamp Recording. Membrane currents were recorded by the whole-cell patch-clamp method using pCLAMP (ver. 8.0, Axon Instruments, Union City, CA) and custom RAM5 software and patch-clamp amplifiers [TM-1000 (ActME, Tokyo, Japan) and EPC-7 (List, Darmstadt, Germany)]. Single cardiac ventricular cells were placed in a recording chamber (0.5-ml volume) attached to an inverted microscope (Nikon ECLIPSE TE200, Tokyo, Japan) and superfused with the Tyrode solution at a rate of 5 ml/min. The temperature of the bath solution was maintained at 35 ± 0.5°C with a water jacket. Patch pipettes were forged from 1.5-mm diameter glass capillaries (Nihon Rikagaku Kikai, Tokyo, Japan) with a microelectrode puller (pp-83; Narishige, Tokyo, Japan). The pipette resistance was 2~4 M when filled with the pipette solution. The pipette solution contained 120 mM CsOH, 20 mM CsCl, 60 mM aspartic acid, 3 mM MgCl₂, 5 mM MgATP, 10 mM BAPTA, and 20 mM HEPES, pH 7.2, with aspartic acid. BaCl₂ (0.5 mM) in Tyrode solution was added to block K⁺ currents. The series resistance was compensated. Current signals were filtered at a 2.5-kHz bandwidth and were stored on a PC (XPS T450; Dell Computer Corp., Round Rock, TX).

Drugs. LPC (Sigma-Aldrich, St. Louis, MO) was dissolved in chloroform to make a 1 mM stock solution, which was evaporated to dryness under a stream of N₂ gas and stored at 20 °C. After thawing the LPC was diluted to appropriate concentrations with the external solution. DL-Mevalonic acid lactone, geranylgeranylpyrophosphate (GGP) and Clostridium botulinum C3 exoenzyme (C3) were purchased from Sigma-Aldrich. Fluvastatin was a kind gift from Novartis (Basel, Switzerland) and Y-27632, a Rho kinase inhibitor, was from Welfide (Osaka, Japan).

Data Analysis. All the values are presented as mean ± S.E. (number of experiments). Student's t test and analysis of variance were used for statistical analyses. P values less than 0.05 were considered significant. The concentration-response data were fitted and IC₅₀ and Hill coefficient values were obtained using Delta Graph Professional (Polaroid Computing, Tokyo, Japan).

	Results

Top Abstract Introduction Methods Results Discussion References

Effects of LPC on Membrane Current. Figure 1 shows the effect of 10 µM LPC on the membrane current in a ventricular myocyte. LPC in the external solution induced I_NSC with a lag and it did not increase steadily but oscillated. The current response was reversible and the second application of 10 µM LPC induced I_NSC with an amplitude similar to the first application. The lag before inducing I_NSC was longer after the first than the second application. The control I-V curves before the application of LPC and the largest current induced by 10 µM LPC are plotted in Fig. 1, B and C. The LPC-induced current crossed with the control at 0 mV, confirming that the LPC-induced current is I_NSC (Magishi et al., 1996).

View larger version (26K): [in this window] [in a new window]   Fig. 1.   Effect of LPC on membrane current. A, current recording versus time. LPC at 10 µM was added twice as indicated. B and C, I-V curves of control (a, c) and maximum current (b, d) induced by the initial and the second applications of LPC. D, concentration dependence of LPC-induced INSC density. E, concentration dependence of the lag. The method of current density measurement is described in the text.

Effect of Fluvastatin on LPC-Induced I_NSC. Figure 2 shows the effect of fluvastatin on LPC-induced current. As shown in Fig. 1, repetitive application of LPC induces I_NSC with each application. However, as shown in Fig. 2, the initial application of 10 µM LPC induced I_NSC in the absence of fluvastatin but did not in response to the second application of LPC, when 5 µM fluvastatin was present in the external solution. After washing out fluvastatin, LPC again induced I_NSC (not shown). These results indicate that fluvastatin reversibly inhibits LPC-induced I_NSC. When fluvastatin was applied after I_NSC had developed in response to LPC, it did not inhibit the current (not shown), indicating that fluvastatin does not directly inhibit I_NSC channels.

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Effect of fluvastatin on LPC-induced INSC. A, Current recording versus time. Fluvastatin was applied 3 min before the second application of LPC. B, I-V curves of control (a) and the maximum response to LPC. C, I-V curves before (c) and in the presence of fluvastatin (d) and in fluvastatin and LPC (e).

Effect of MVA on LPC-Induced Current. If fluvastatin inhibited LPC-induced I_NSC by inhibiting HMG-CoA reductase, MVA should prevent this effect of the drug and this was examined in the experiment shown in Fig. 3 in which 100 µM MVA was perfused in the external solution for 3 min before adding 5 µM fluvastatin. In the presence of MVA and fluvastatin, 10 µM LPC induced I_NSC. The I-V curve of the current induced by LPC (Fig. 3B, d) was identical to that induced in the absence of the two agents. This result indicates that fluvastatin inhibits LPC as a result of its inhibition of an HMG-CoA reductase.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Effect of fluvastatin on LPC-induced current in the presence of MVA. A, chart recording of current. MVA was applied 3 min before adding fluvastatin. B, I-V curves of the corresponding labels in A. Control I-V curve before (a) and during (b) MVA application, in the presence of MVA and fluvastatin (c), and maximum response to LPC in the presence of MVA and fluvastatin (d).

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Averaged data of LPC-induced iNSC density (A) and lag (B) under the conditions indicated. *, significant difference. Mev, MVA.

Involvement of Geranylgeranylpyrophosphate. Geranylgeranylpyrophosphate (GGP) is a downstream isoprenoid product of MVA in the cholesterol synthesis pathway (Glomset et al., 1990). To pursue the mechanism of LPC-induced I_NSC further, GGP was included in the pipette solution and the effect of fluvastatin was examined. Inclusion of GGP alone did not induce I_NSC. Fluvastatin did not inhibit I_NSC induced by 10 µM LPC in the presence of GGP (Fig. 5). This indicates that GGP is necessary for the current induction by LPC. Because GGP is necessary for membrane association of the small GTP binding protein, Rho (Goldstein and Brown, 1990; Maltese, 1990; Ridley and Hall, 1992), this result suggests the possible involvement of Rho in the effect of LPC.

View larger version (22K): [in this window] [in a new window]   Fig. 5.   Effect of fluvastatin on LPC-induced INSC in the presence of GGP in the pipette solution. A, chart recording. B, I-V curves of the corresponding labels in A.

Effect of C3 Exoenzyme on LPC-Induced I_NSC. Rho is selectively inactivated by C3 through ADP-ribosylation (Sah et al., 2000). To assess whether Rho is involved in the effect of LPC, we included 4 µg/ml C3 in the pipette solution and applied LPC (Fig. 6). LPC did not induce any significant change in current in the presence of C3 whether or not MVA was present in the bath solution (Fig. 6, B and C). Heat-inactivated C3 did not inhibit LPC-induced current. These results indicate that the small GTP binding protein Rho is involved in the process of LPC-induced I_NSC.

View larger version (22K): [in this window] [in a new window]   Fig. 6.   Effect of C3 on LPC-induced INSC. A, current recording versus time. C3 is included in the pipette solution. B, I-V curves before (a) and during (b) LPC application. C, I-V curves before (c) and during (d) MVA application and further addition of LPC (e).

Effect of a Rho-Kinase Inhibitor on LPC-Induced Current. Y-27632 is a selective inhibitor of Rho-kinase (Uehata et al., 1997). Because Rho may activate Rho-kinase in the process of LPC-induced I_NSC, we tested the effect of LPC with 10 µM Y-27632 in the pipette solution. Figure 7 shows that LPC induced I_NSC, but it was significantly smaller and the lag was longer in the presence of Y-27632 than in its absence.

View larger version (10K): [in this window] [in a new window]   Fig. 7.   Effect of Y-27632 on LPC-induced INSC. A, chart recording of current. B, I-V curves of control (a) and the maximum response to LPC (b) with Y-27632 in the pipette solution.

View larger version (18K): [in this window] [in a new window]   Fig. 8.   Summary of results in Figs. 6 and 7. *, significant difference. Current density was measured as described in the legend for Fig. 1.

Effect of Pertussis Toxin on LPC-Induced Current. To examine the possible involvement of heterotrimeric G-proteins in the LPC-mediated pathway, pertussis toxin, an inhibitor of G_i/G_o, was included in the pipette solution. In the presence of pertussis toxin (1 µg/ml), the LPC (10 µM)-induced current was 1.62 pA/pF (n = 4) at 100 mV (figure not shown). This was significantly smaller than that of the control, indicating that the Gi/Go-coupled membrane receptor is involved in the LPC-mediated pathway.

	Discussion

Top Abstract Introduction Methods Results Discussion References

In this study, we demonstrated that LPC induces I_NSC in guinea pig ventricular myocytes through a mechanism that involves an activation of the small G-protein Rho. The LPC-induced effect must be mediated by a receptor that interacts with the PTX sensitive G-proteins. Figure 9 shows a proposed signal transduction pathway for LPC-induced I_NSC based on our results.

View larger version (22K): [in this window] [in a new window]   Fig. 9.   Proposed mechanisms of the LPC-induced signal transduction pathway. The LPC receptor is shown as Gi/o-coupled receptor. Rho is a small G-protein.

LPC activated I_NSC in a concentration-dependent manner with an EC₅₀ of approximately 20 µM. There was a lag before the development of I_NSC with an average of 115 ± 32 s (n = 10) with 10 µM LPC, and it was shortened to 40 ± 11 s (n = 7) with 30 µM LPC, suggesting that the induction of I_NSC by LPC involves a distinct signal transduction process that takes a few minutes. We demonstrated that fluvastatin, a HMG-CoA reductase inhibitor, inhibited LPC-induced I_NSC. Fluvastatin inhibits MVA synthesis from HMG-CoA. With MVA in the external solution, fluvastatin did not inhibit LPC-induced i_NSC. Therefore, it seems that MVA is essential for the induction of I_NSC by LPC. Recently, it has been shown that the MVA-derived isoprenoids, GGP and farnesylpyrophosphate, modify several proteins by covalent attachment or prenylation, which is necessary for their association with membranes and thus for their function (Glomset et al., 1990; Goldstein and Brown, 1990; Maltese, 1990). When GGP was included in the pipette solution, fluvastatin did not inhibit LPC-induced I_NSC. Therefore, it seems that some protein that is isoprenylated by GGP is involved in the LPC effect. Candidate proteins for isoprenylation by GGP are Rho, Rac, Cdc42, Rab, Rap, and the -subunit of heterotrimeric G-proteins (Corsini et al., 1999). As shown in Fig. 6, botulinum toxin C3 exoenzyme, but not heat-inactivated C3, in the pipette solution almost completely inhibited LPC-induced I_NSC. C3 is an ADP-ribosyltransferase that inactivates Rho subfamily proteins (RhoA, RhoB, and RhoC) more specifically than other small GTP binding proteins, such as Rac or Cdc42 (Ridley and Hall, 1992). The result with C3 indicates that Rho is involved in the induction of I_NSC by LPC (Fig. 9).

Many effectors of Rho-binding proteins have been identified but the best characterized are Rho kinases, which are serine/threonine directed and include Rho kinase/ROK/Rock-II and p160ROCK/ROCK (Sah et al., 2000). Y-27635 specifically inhibits Rho-kinase Rock, by competing for ATP binding (Uehata et al., 1997). Y-27635 at 10 µM inhibited LPC-induced I_NSC significantly but not completely (Fig. 8), indicating that Rho-kinase might be involved in the effect of LPC (Fig. 9).

LPC may stimulate some membrane receptor through which a signal transduction pathway is activated and this pathway requires Rho and Rock to open I_NSC channels. In this study, PTX inhibited the effect of LPC, indicating that guinea pig ventricular myocytes possess an LPC-receptor that is coupled to the pertussis toxin-sensitive heterotrimeric G-protein, G_i/G_o (Fig. 9). Recently, two G-protein-coupled orphan receptors, G2A (Kabarowski et al., 2001) and GPR4 (Zhu et al. 2001), were identified as the LPC receptors. Both of these receptors are coupled to the PTX-sensitive G protein, and mediate the LPC-induced increase in intracellular Ca²⁺ concentration. In addition, GPR4 was shown to be expressed in the heart, and its effects were inhibited by both PTX and botulinus toxin C3 (Zhu et al. 2001). Involvement of GPR4 and/or G2A in LPC-induced I_NSC should be investigated in cardiac ventricular cells.

G-protein-coupled lysophopholipid receptors have also been found for sphingosine-1-phosphate (SPP), sphingosylphosphocholine, and lysophosphatidic acid (Fukushima et al., 2001). Bünemann et al. (1995) found that SPP activates I_K(Ach) in guinea pig atrial myocytes via Gi-coupled receptors. Recently, they detected Edg1, Edg3, Edg5, and Edg8 mRNAs and OGR1 lysophospholipid receptor protein in rat atrium and ventricle (Liliom et al., 2001). More recently, Muraki and Imaizumi (2001) succeeded in recording PTX-sensitive nonselective cation channels induced by SPP in human endothelial cells.

Some important questions about the LPC-induced signal transduction pathway need to be addressed. One is how LPC activates the Rho signaling pathway. Although G₁₂ and G₁₃ have recently been identified as the G-proteins that transduce signals from various receptors to the Rho GTPase by stimulating the GDP-GTP exchange, our results suggest that certain PTX-sensitive G-proteins can initiate the Rho signaling pathway in a GGP-dependent manner. Because GGP alone did not induce I_NSC, it seems that LPC triggers geranylgeranylation of Rho. It is important to identify what types of PTX-sensitive G-proteins and which of the - or -subunits are associated with the LPC receptors.

Another important question is how fluvastatin affects the intracellular levels of GGP so rapidly (i.e., within a few minutes). It has already been demonstrated that inhibition of HMG-CoA reductase by statins blocks the Rho-dependent cellular responses by preventing isoprenyl modification of Rho. However, this effect of statin was generally observed between 6 and 24 h (Laufs and Liao, 1998). The present results demonstrated that the inhibition of LPC-induced I_NSC by statin was very rapid, occurring within a few minutes. Furthermore, mevalonate or GGP prevented the inhibitory effect of statin, suggesting that statin exerted its effect by lowering intracellular GGP. We recently demonstrated similar results with LPC-induced I_NSC in human aortic endothelial cells (Yokoyama et al, 2002). In the endothelial cells, LPC-induced I_NSC was accompanied by a rapid translocation of GTP-bound RhoA into membrane within one min. Furthermore, a brief treatment with fluvastatin (3 min) prevented the LPC-induced I_NSC and translocation of RhoA to the membrane. This effect of statin also disappeared in the presence of MVA or GGP. These results suggest that LPC stimulates I_NSC through a similar mechanism in both ventricular myocytes and endothelial cells. Given these results, we suggest that, at least in these cells, intracellular GGP levels may be controlled dynamically through metabolic processes. Further studies with direct measurements of intracellular GGP levels are required to evaluate our hypothesis.

Fluvastatin and various other statins are used clinically for hyperlipidemia to lower serum cholesterol levels. There are increasing reports on various effects of statins. The inhibitory effect of fluvastatin on LPC-induced I_NSC provides a new insight into the signal transduction mechanism and may have important clinical implications.