Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

The central nervous system consists of neurons and glial cells. Neurons play a central role in signal transduction by releasing neurotransmitters. One of the most important characteristics of neurons is that they cannot, in general, proliferate once they have begun the process of differentiation. For this reason, neurotrophic factors are essential for the functional maintenance and organization of neurons. In the central nervous system, neurons and glial cells secrete neurotrophic factors, such as nerve growth factor (NGF), brain-derived neurotrophic factor, neurotrophin 3 (Lessmann, 1998; Althaus and Richter-Landsberg, 2000), interleukin-6 (Schwaninger et al., 1997), and glial cell line-derived neurotrophic factor (Lin et al., 1993). NGF has pleiotrophic effects on neuronal differentiation and survival (prevention of apoptosis) in a variety of neurons (Levi-Montalcini, 1987).

Rat pheochromocytoma cells (PC-12) have been used as an in vitro model of neuronal differentiation. In response to NGF, these cells extend neurites and develop characteristics of sympathetic neurons (Greene and Tischler, 1976). We have previously reported that a new neurotrophic factor secreted from 1321N1 human astrocytoma cells caused the differentiation of PC-12 cells (Obara et al., 1998). In addition, we have shown that 1321N1 cell culture medium, conditioned by scabronines isolated from Sarcodon scabrosus, enhanced the differentiation of PC-12 cells (Obara et al., 1999a, 2001).

"So-jutsu" (Atractylodis lanceae rhizomas), an often prescribed and important group of Chinese medicines, has various health effects, including the normalization of stomach and intestinal functions and the promotion of diuresis. Sesquiterpenoids, such as -eudesmol, hinesol, and elemol, were isolated from So-jutsu and reported to have unique effects on the nervous system. A mixture of -eudesmol and hinesol was shown to act as a sedative, to prolong sleep induced by hexobarbital, and to have anti-convulsive effects in mice exposed to electrical stimulation (Yamahara et al., 1977). The mechanism by which -eudesmol mediated these effects has been partially clarified. Thus, Satoh et al. (1992) showed that -eudesmol inhibited Na⁺, K⁺-ATPase activity, whereas Kimura et al. (1991) demonstrated that it acted as a channel blocker for the nicotinic acetylcholine receptor in skeletal muscle cells. Furthermore, Tachikawa et al. (2000) showed that acetylcholine-induced catecholamine secretion from bovine adrenal chromaffin cells was suppressed by -eudesmol by inhibition of Ca²⁺ influx. -Eudesmol, a structural isomer of -eudesmol, was also shown to be a P/Q-type Ca²⁺ channel blocker (Asakura et al., 2000).

In light of their demonstrated beneficial effects, neurotrophic factors were suggested as potential therapeutic agents for the treatment of such neurological diseases as Alzheimer's and Parkinson's disease. Since these factors are polypeptides of high molecular weight, they cannot cross the blood-brain barrier and are easily proteolyzed when administrated peripherally. A useful strategy for addressing this drug delivery problem is the use of small organic compounds, which themselves either directly maintain neuronal function or up-regulate neurotrophic factors. To date, small molecules such as AIT-082 and SR57746A are known to be promising NGF-like agonists or NGF inducers (Xie and Longo, 2000). In the present study, we report the mechanism by which -eudesmol promotes neurite outgrowth in PC-12 cells.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. -Eudesmol, propidium iodide, and GF109203X were purchased from Wako Pure Chemicals (Tokyo, Japan). NGF, carbachol, W7, and adenosine deaminase were obtained from Sigma Chemical Co. (St. Louis, MO), and H89 was acquired from Seikagaku Corporation (Tokyo, Japan). Fura 2-acetoxymethylester and fluo 3 were purchased from Dojindo (Kumamoto, Japan), and U-73122 and U-73343 were purchased from Biomol (Plymouth Meeting, PA). [³H]inositol (23.4 Ci/mmol) was obtained from PerkinElmer Life Sciences (Boston, MA). The anion exchange column (AG1 × 8) and PD98059 were purchased from Bio-Rad (Hercules, CA) and Cell Signaling (Beverly, MA), respectively.

Cell Culture. PC-12 cells were obtained from the Japanese Cancer Research Bank (Tokyo, Japan) and were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% heat-inactivated fetal calf serum (Cell Culture Laboratory, Cleveland, OH), 5% heat-inactivated horse serum (Invitrogen, Grand Island, NY), penicillin (50 units/ml), and streptomycin (50 µg/ml) in a 5% CO₂ incubator at 37°C.

Evaluation of Neurite Outgrowth. PC-12 cells (1 × 10⁵ cells/ml) were seeded onto 24-well plates and were cultured for 1 day, after which time drugs were added and the cells cultured for an additional 1 to 2 days. The cells were then fixed with 2% paraformaldehyde (Wako Pure Chemicals) in phosphate-buffered saline (PBS), and cell morphology was assessed under a phase-contrast microscope. Neurite extension from PC-12 cells was regarded as an index of neuronal differentiation (Obara et al., 1998). Processes with a length equivalent to one or more diameters of the cell body were regarded as neurites. The differentiation of PC-12 cells was evaluated by examining the proportion of neurite-bearing cells to total cells in randomly selected fields. The mean differentiation score was obtained for more than a 100 PC-12 cells in each well. Data were expressed as the means ± S.E.M. of three different wells from a single culture.

Measurement of [³H]Thymidine Incorporation. PC-12 cells were seeded onto 24-well plates (1 × 10⁵ cell/ml), and the following day, drugs together with [³H]thymidine (1 µCi/well) (Amersham Biosciences, Piscataway, NJ) were added, and the cells were cultured for an additional 1 to 2 days. After washing twice with ice-cold PBS, the reaction was terminated by the addition of 5% trichloroacetic acid, after which time the plates were kept at 4°C for 60 min. After three washes with ethanol, the radioactivity of [³H]thymidine in the cell lysates was determined by scintillation counting. Data were expressed as the percentage of control.

MTT Assay. PC-12 cells were seeded into 96-well plates (200 µl/well) at a density of 5 × 10⁶ cells/ml. At the end of the drug incubation period, MTT (0.1 mg) (Dojindo) was added to each well, and the plates incubated for an additional 4 h at 37°C. After centrifugation at 350g for 5 min, the medium was replaced with dimethyl sulfoxide. The absorbance of reduced MTT at 595 nm was measured with a plate reader.

Measurement of Intracellular Free-Ca²⁺ Concentration by Fura 2. Intracellular free-Ca²⁺ concentration ([Ca²⁺]_i) was measured by the fura 2 assay, as previously described (Obara et al., 1999b). Briefly, semiconfluent PC-12 cells cultured in a 150-mm dish were collected by gentle pipetting and were washed with a modified tyrode buffer (137 mM NaCl, 2.7 mM KCl, 1 mM MgCl₂, 0.18 mM CaCl₂, 5.6 mM glucose, 10 mM HEPES, pH 7.4). The cells were then incubated with fura 2-acetoxymethylester (2 µM) at 37°C for 30 min, after which time the cells were washed twice with buffer and the fluorescence intensity of fura 2 (excitation wavelength at 340 and 380 nm and emission wavelength at 510 nm) was measured with a fluorescence spectrophotometer (F-2000; Hitachi, Tokyo, Japan). In our preliminary experiments, the Ca²⁺ responsiveness to -eudesmol in these suspended PC-12 cells was nearly equivalent to that seen in attached cells (data not shown).

Measurement of Ca²⁺ Release from Sarcoplasmic Reticulum Vesicles by Fluo 3. Sarcoplasmic reticulum vesicles were prepared from rabbit skeletal muscle, according to the method described by Kim et al. (1983), in the presence of the protease inhibitors aprotinin (76.8 µM) (Sigma Chemical Co.) and benzamidine (0.83 mM) (Sigma Chemical Co.). Briefly, rabbit white muscle was homogenized in 5 volumes of 5.0 mM Tris-maleate, pH 7.0, and centrifuged at 5000g for 15 min. The supernatant was centrifuged at 12,000g for 30 min. The pellet was suspended in 5.0 mM Tris-maleate containing 90 mM KCl and was centrifuged at 70,000g for 40 min. The pellet obtained after this high-speed spin, which contained the heavy fraction of sarcoplasmic reticulum, was resuspended and stored at 80°C. The protein concentration of the pellet was determined by the method previously described by Bradford (1976), using bovine serum albumin (BSA) as the standard.

Measurement of Ca²⁺-ATPase Activity. Ca²⁺-ATPase activity in sarcoplasmic reticulum vesicles was measured as follows: Sarcoplasmic reticulum vesicles (40 µg/ml) in 500 µl of basic solution (100 mM MOPS-Tris, pH 7.4, 200 mM KCl, 2.0 mM MgCl₂, and 0.4 mM CaCl₂) were preincubated in the absence of drugs and ATP at 30°C for 5 min. This was followed by a 5-min incubation in the presence of 400 µl of drugs. The reaction was started by the addition of 100 µl of ATP (10 mM), which was left on the cells for 3 min (final volume, 1.0 ml). ATPase activity was determined by measurement of the amount of liberated phosphates, using the malachite green method as described by Chan et al. (1986).

Measurement of Inositol Phosphates. The accumulation of total inositol phosphates was measured as follows: PC-12 cells were cultured in 12-well plates for 2 days at a density of 4 × 10⁵ cells/ml/well. Cultures were then incubated overnight with [³H]inositol (2 µCi/ml), after which time they were washed twice with serum-free DMEM-HEPES buffer, pH 7.4. They were then preincubated in DMEM-HEPES containing LiCl (10 mM) for 10 min, after which they were incubated with drugs for additional 10 min. The reaction was terminated by the addition of 1 ml of ice-cold 5% trichloroacetic acid after aspiration of the medium. The extracts were washed three times with diethyl ether to remove the trichloroacetic acid. Diethyl ether in the sample was removed by storage at 47°C for 30 min. Total [³H]inositol phosphates were separated by an anion exchange column (AG-1X8, formate form, 100-200 mesh), as previously described (Nakahata et al., 1989).

SDS-Polyacrylamide Gel Electrophoresis and Immunoblotting for Mitogen-Activated Protein Kinase (MAPK) and cAMP-Responsive Element Binding Protein (CREB). Samples used for phospho-MAPK and phospho-CREB detection were prepared as follows: PC-12 cells were seeded onto 6-well plates at a density of 4 × 10⁵ cells/ml. The cells were cultured overnight in serum-free DMEM, after which drugs were added for varying periods of time. The incubation medium was aspirated after the reaction, and the cells were dissolved in Laemmli sample buffer (final concentration, 75 mM Tris-HCl, 2% SDS, 15% glycerol, 3% 2-mercaptoethanol, pH 6.8) and boiled at 95°C for 5 min.

Immunostaining of Phospho-MAPK and Phospho-CREB. PC-12 cells were seeded onto 24-well plates at a density of 2 × 10⁵ cells/ml. The medium was replaced with serum-free DMEM and the cells were cultivated overnight, after which they were incubated in serum-free DMEM-HEPES containing drugs for 5 min. The cells were then fixed with 2% paraformaldehyde/PBS for 15 min, permeated with 0.5% Triton X-100/PBS for 5 min, and then washed with PBS. After blocking with 5% BSA for 3 h at 37°C, the cells were incubated with anti-phospho-MAPKs antibody (1:500 dilution) or anti-phospho-CREB antibody (1:500 dilution) at 4°C overnight, followed by goat fluorescein-labeled anti-rabbit IgG antibody (1:80 dilution) (Kirkegaard & Perry Laboratories, Inc., Gaithersburg, MD) at 37°C for 60 min. The cells were then incubated with propidium iodide (50 ng/ml) for 30 min to stain the nuclei. The cells were then visualized under a confocal laser microscope (DMRB/E, TCS NT; Leica, Wetzlar, Germany).

Statistical Methods. Data were expressed as the mean values ± S.E.M., and the significant differences were analyzed using an analysis of variance. Tukey's method was used for carrying out the multiple comparisons shown in Fig. 9.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Effects of -Eudesmol on Neurite Outgrowth in PC-12 Cells. To investigate the effect of -eudesmol on PC-12 cells (Fig. 1), cells were cultured with it at a concentration of 10 to 150 µM for 48 h, and morphological changes were observed under the phase-contrast microscope. -Eudesmol promoted neurite extension in PC-12 cells (Fig. 2A). When the percentage of neurite-bearing cells was evaluated, -eudesmol, at concentrations of 100 and 150 µM, was found to have significantly promoted neurite extension (EC₅₀, >70 µM) to a degree that was similar to that seen with NGF (Fig. 2B). When cell growth was investigated by [³H]thymidine incorporation, a 48-h incubation with -eudesmol (150 µM) was found to have significantly suppressed proliferation (IC₅₀, >105 µM) (Fig. 2C). Since the concentration range that caused neurite outgrowth was similar to the range suppressing proliferation, it was assumed that the differentiation of PC-12 cells was due to the arrest of cell growth. A cytotoxic effect was not revealed in this concentration range, as determined by the MTT assay (data not shown).

View larger version (11K): [in this window] [in a new window]   Fig. 1.   The chemical structure of -eudesmol.

View larger version (31K): [in this window] [in a new window]   Fig. 2.   Effect of -eudesmol on neurite outgrowth in PC-12 cells. A, morphological change in PC-12 cells induced by -eudesmol. PC-12 cells were cultured in the presence or absence of -eudesmol (150 µM) for 48 h, and their morphological appearance was then observed under phase-contrast microscopy. Scale bar is 100 µm. B, evaluation of neurite outgrowth in PC-12 cells induced by -eudesmol. PC-12 cells were cultured with -eudesmol at indicated concentrations or NGF (50 ng/ml) for 48 h, and the percentage of neurite-bearing cells was then calculated as described under Experimental Procedures. Values are the means ± S.E.M. of three different wells from a single culture. -Eudesmol (100 and 150 µM) significantly promoted neurite outgrowth from PC-12 cells. *, P < 0.05 versus control without drug. C, effect of -eudesmol on [3H]thymidine incorporation in PC-12 cells. PC-12 cells were cultured in the presence of -eudesmol (10-150 µM) or NGF (50 ng/ml) together with [3H]thymidine for 48 h, and then incorporated [3H]thymidine was measured. Values are the means ± S.E.M. of three different wells from a single culture. -Eudesmol (150 µM) significantly suppressed [3H]thymidine incorporation. *, P < 0.05 versus control without drug.

View larger version (12K): [in this window] [in a new window]   Fig. 3.   Effect of -eudesmol on [Ca2+]i in PC-12 cells. The [Ca2+]i level was monitored by using the fura 2 assay, as described under Experimental Procedures. A, typical responses to carbachol (100 µM) and -eudesmol (10-150 µM). B, concentration-dependent increase in [Ca2+]i in response to -eudesmol. Data are expressed as the -eudesmol-induced increase in [Ca2+]i ([Ca2+]i). Values are the means ± S.E.M. of three independent determinations. -Eudesmol at 100 µM and 150 µM significantly increased [Ca2+]i. *, P < 0.05 versus control without drug.

[Ca²⁺]_i Increase by -Eudesmol Mediated by an Activation of Phosphoinositide-Specific Phospholipase C. To examine the mechanism of -eudesmol-signaling, [Ca²⁺]_i was measured by the fura 2 assay. -Eudesmol caused an [Ca²⁺]_i elevation in a concentration-dependent manner in PC-12 cells. The effects of -eudesmol at concentrations of 100 and 150 µM were significant, and the maximum response of -eudesmol was as large as that seen with carbachol (100 µM). When extracellular Ca²⁺ was removed by EGTA, the increase in [Ca²⁺]_i was partially inhibited (this inhibition was not statistically significant), although the preponderance of the [Ca²⁺]_i increase remained (Fig. 4). These results suggest that the [Ca²⁺]_i elevation induced by -eudesmol primarily resulted from Ca²⁺ release from intracellular Ca²⁺ stores in the PC-12 cells.

View larger version (29K): [in this window] [in a new window]   Fig. 4.   Effect of -eudesmol on [Ca2+]i under extracellular Ca2+-free conditions in PC-12 cells. A, -eudesmol (150 µM) was added in the presence or absence of EGTA (0.5 mM), and the [Ca2+]i level was monitored using the fura 2 assay as shown in Fig. 3. B, effect of removal of extracellular Ca2+ on -eudesmol (150 µM)-induced [Ca2+]i elevation. Values are the means ± S.E.M. of three independent determinations.

View larger version (23K): [in this window] [in a new window]   Fig. 5.   Involvement of PI-PLC in -eudesmol-induced Ca2+ release in PC-12 cells. A, inhibitory effect of U-73122 on -eudesmol-induced Ca2+ release. PC-12 cells were preincubated with U-73343 (2 µM) (a) or U-73122 (2 µM) (b) for 1 min before the addition of -eudesmol (150 µM) under extracellular Ca2+-free conditions. B, effect of -eudesmol on PI-hydrolysis. PC-12 cells were incubated with carbachol (100 µM) or -eudesmol (50-150 µM) for 10 min in DMEM containing 1.8 mM Ca2+, and total inositol phosphate accumulation was measured as described. Values are the means ± S.E.M. of three different wells from a single culture. The maximum y-values in these -eudesmol and carbachol graphs are 4000 and 8000, respectively. -Eudesmol and carbachol significantly increased the accumulation of total inositol phosphates. *, P < 0.05 versus control without drug.

Involvement of MAPK and CREB Activation in -Eudesmol-Signaling in PC-12 Cells. Since MAPK plays a central role in NGF-induced neurite outgrowth (Cowley et al., 1994), we wished to determine whether MAPK activation was involved in -eudesmol-induced neurite outgrowth in PC-12 cells. After incubation with -eudesmol (150 µM) for 5 min, phospho-MAPK was clearly detected in the nuclei of PC-12 cells by immunostaining (Fig. 6A). In addition, -eudesmol (150 µM) promoted the phosphorylation of MAPKs (ERK1 and 2) in a time-dependent manner (Fig. 7). Since CREB regulates the expression of genes that play a role in neuronal functions, such as those encoding catecholamine synthesizing enzymes and choline o-acetyltransferase, the phosphorylation of CREB was also examined. After the cells were incubated with -eudesmol (150 µM) for 5 min, phospho-CREB was detected in their nuclei (Fig. 6B). -Eudesmol caused the phosphorylation of CREB in a time-dependent manner (Fig. 7). The phosphorylation of both MAPK and CREB that was induced by -eudesmol was transient. Both phosphorylations were blocked by PD98059 (50 µM) and U-73122 (2 µM) (Fig. 8A). PD98059 inhibited MAPK (ERK 1, 2) and CREB phosphorylation by 68.7, 59.0, and 43.9%, respectively, whereas U-73122 reduced the degree of phosphorylation by 42.7, 43.8, and 38.2%, respectively. These data indicate that the -eudesmol-induced MAPK phosphorylation resulted from activation of PI-PLC and that CREB phosphorylation was dependent on MAPK activation. Furthermore, both the calmodulin (CaM) inhibitor W7 and the protein kinase A (PKA) inhibitor H89 also reversed the phosphorylation of MAPK and CREB in a concentration-dependent manner (1-10 µM) (Fig. 8, B and C). W7 (10 µM) inhibited MAPK (ERK 1, 2) and CREB phosphorylation by 80.9, 69.5, and 71.0%, respectively, whereas H89 (10 µM) reduced the degree of phosphorylation by 67.2, 67.7, and 55.0%, respectively. For determinations of involvements of these molecules, the concentrations of inhibitors used in this study were referred to these articles (Wang et al., 1998; Grewal 2000a,b). On the other hand, the protein kinase C (PKC) inhibitor GF109203X (1 µM) did not affect -eudesmol-induced MAPK activation (data not shown). In response to [Ca²⁺]_i elevation, neuronal cells secrete adenine nucleotides such as ATP and adenosine, which promote cAMP accumulation through the adenosine A_2a/b receptor (Ohkubo et al., 2000; Kasai et al., 2001). For this reason, we examined the involvement of adenine nucleotides in MAPK activation by using adenosine deaminase (1 unit/ml). However, MAPK phosphorylation induced by eudesmol, was not inhibited by adenosine deaminase (data not shown). These results suggest that the phosphorylation of both MAPK and CREB were mediated through activation of CaM and PKA.

View larger version (36K): [in this window] [in a new window]   Fig. 6.   Effect of -eudesmol on the phosphorylation of MAPK and CREB in PC-12 cells. PC-12 cells were incubated with -eudesmol (150 µM) for 5 min, which was followed by their processing for phospho-MAPK (A) and phospho-CREB (B) immunoreactivity as described. Nuclei were stained with propidium iodide.

View larger version (42K): [in this window] [in a new window]   Fig. 7.   Time-dependent effect of -eudesmol on the phosphorylation of MAPK and CREB. PC-12 cells were incubated with NGF (50 ng/ml) for 5 min or -eudesmol (150 µM) for 2 to 30 min under serum-free conditions, and bands of phosphorylated MAPK and CREB were detected in cell lysates by Western blotting. Lanes: 1, control; 2, NGF (5 min); 3, -eudesmol (2 min); 4, -eudesmol (5 min); 5, -eudesmol (10 min); 6, -eudesmol (30 min).

View larger version (79K): [in this window] [in a new window]   Fig. 8.   Involvement of PI-PLC, CaM and PKA in -eudesmol-signaling. A, inhibitory effects of PD98059 and U-73122 on -eudesmol-induced MAPK and CREB phosphorylation. PC-12 cells were preincubated with PD98059 (50 µM) or U-73122 (2 µM) for 10 min, followed by a 5-min incubation with -eudesmol (150 µM). The bands of phosphorylated MAPK and CREB in the cell lysates were detected by Western blotting. Lanes: 1, control; 2, PD98059 (50 µM); 3, U-73122 (2 µM); 4, -eudesmol (150 µM); 5, -eudesmol (150 µM) + PD98059 (50 µM); 6, -eudesmol (150 µM) + U-73122 (2 µM). B, inhibitory effect of W7 on -eudesmol-induced MAPK and CREB phosphorylation. PC-12 cells, preincubated with W7 (1-10 µM) for 10 min, were incubated with -eudesmol (150 µM) for 5 min. Lanes: 1, control; 2, -eudesmol (150 µM); 3, W7 (10 µM); 4, -eudesmol (150 µM) + W7 (1 µM); 5, -eudesmol (150 µM) + W7 (3 µM); 6, -eudesmol (150 µM) + W7 (10 µM). C, inhibitory effect of H89 on -eudesmol-induced MAPK and CREB phosphorylation. PC-12 cells, preincubated with H89 (1-10 µM) for 10 min, were incubated with -eudesmol (150 µM) for 5 min. Lanes: 1, control; 2, -eudesmol (150 µM); 3, H89 (10 µM); 4, -eudesmol (150 µM) + H89 (1 µM); 5, -eudesmol (150 µM) + H89 (3 µM); 6, -eudesmol (150 µM) + H89 (10 µM).

View larger version (34K): [in this window] [in a new window]   Fig. 9.   Involvement of PI-PLC and MAPK in -eudesmol-induced neurite outgrowth in PC-12 cells. A, effects of U-73122 and PD98059 on the -eudesmol-induced neurite outgrowth in PC-12 cells. PC-12 cells, cultured in medium containing -eudesmol (150 µM) for 24 h in the presence or absence of U-73122 (2 µM) or PD98059 (30 µM), were scored according to their degree of neurite outgrowth. Values are the means ± S.E.M. of three different wells from a single culture. -Eudesmol significantly promoted neurite outgrowth from PC-12 cells. This effect was significantly inhibited by U-73122 and PD98059. In addition, -eudesmol significantly enhanced neurite outgrowth in the presence of U-73122. *, P < 0.05 versus control without drug; , P < 0.05; #, P < 0.05. B, effects of U-73122 and PD98059 on the -eudesmol-suppressed [3H]thymidine incorporation in PC-12 cells. PC-12 cells were cultured in medium containing -eudesmol (150 µM) and [3H]thymidine for 24 h in the presence or absence of U73122 (2 µM) or PD98059 (30 µM). The radioactivity of incorporated [3H]thymidine was measured as described. Values are the means ± S.E.M. of three different wells from a single culture.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

In the present study, we showed that -eudesmol caused neurite outgrowth from PC-12 cells and that this effect was mediated by MAPK activation. We also demonstrated that the [Ca²⁺]_i elevation was triggered by the promotion of PI-hydrolysis, followed by the phosphorylation of MAPK and CREB.

The concentration range of -eudesmol that resulted in neurite outgrowth enhancement was similar to the concentration that suppressed [³H]thymidine incorporation, suggesting that the arrest of proliferation triggered cellular differentiation. Furthermore, the fact that the concentrations of -eudesmol that produced statistically significant changes in neurite outgrowth and [Ca²⁺]_i elevation were also equivalent suggested that Ca²⁺-signaling stimulated neurite outgrowth. With regard to [Ca²⁺]_i elevation, the concentration range of -eudesmol that caused inositol phosphate accumulation was lower than the concentration that promoted the [Ca²⁺]_i increase. Therefore, it was assumed that the part of [Ca²⁺]_i elevation was due to other effects of -eudesmol, such as its inhibition of Na⁺, K⁺-ATPase activity. This speculation was consistent with the data showing that U-73122 did not completely inhibit the [Ca²⁺]_i increase induced by -eudesmol (Fig. 5A).

Concerning the mechanism of neurite outgrowth in PC-12 cells, -eudesmol triggered the phosphorylation of MAPK (Fig. 6 and 7), which has generally been thought to play a central role in the NGF-induced differentiation of neurons (Cowley et al., 1994). MAPK activation was also assumed to be due to PI-breakdown since U-73122 suppressed the phosphorylation of MAPK (Fig. 8A). Furthermore, neurite outgrowth induced by -eudesmol was prevented by PD98059 (Fig. 9). Hence, it was conceivable that MAPK activation was responsible for -eudesmol-induced neurite outgrowth. Nevertheless, it is generally accepted that sustained MAPK phosphorylation is crucial for the differentiation of PC-12 cells that is induced by NGF (Harada et al., 2001). Although our observation (i.e., the transient MAPK phosphorylation induced by -eudesmol) is inconsistent with this notion, we speculate that other signaling molecules that also affect neurite extension, such as c-Jun N-terminal kinase or phosphatidylinositol-3-kinase, may account for the transient MAPK phosphorylation.

Starikova et al. (2000) demonstrated that KCl (40 mM) triggered the differentiation of PC-12 cells by inducing Ca²⁺ influx through a voltage-dependent Ca²⁺ channel. This result indicated that the [Ca²⁺]_i increase was involved in the differentiation of the cells since an [Ca²⁺]_i increase can induce MAPK activation (Grewal et al., 2000b). In spite of that the Ras/MAPK pathway is an important target for Ca²⁺-signaling in many cell types and activated via the Ca²⁺-sensitive tyrosine kinase Pyk2 (Lev et al., 1995), it has been clearly shown that the PKA-dependent mechanism of Rap-1/B-Raf-signaling pathway predominates in Ca²⁺-induced MAPK activation in PC-12 cells (Grewal et al., 2000a,b). In our study, the CaM inhibitor W7 and the PKA inhibitor H89 reversed -eudesmol-induced phosphorylation of both MAPK and CREB (Fig. 8, B and C), suggesting the involvement of Ca²⁺/CaM and PKA. Since adenylate cyclase types I and VIII are CaM-sensitive, the possibility exists that the increased [Ca²⁺]_i induced by -eudesmol stimulated the MAPK cascade via PKA as a result of cAMP accumulation. This speculation is consistent with the report that increased [Ca²⁺]_i promoted phosphorylation of MAPK mediated through the CaM/PKA/Rap-1/B-Raf pathway (Grewal et al., 2000a,b). Further study is needed to confirm the involvement of such molecules (i.e., Rap-1 and B-Raf) in MAPK phosphorylation induced by -eudesmol. Regarding Ca²⁺/CaM-insensitive signaling, it has been shown that PKC is also involved in neurite outgrowth in response to NGF in PC-12 cells, which is mediated through MAPK phosphorylation (Brodie et al., 1999). Although classical and novel PKCs are assumed to be activated as a result of PI-hydrolysis induced by -eudesmol, the PKC/MAPK cascade pathway does not play a central role in -eudesmol-signaling since PKC inhibitor did not inhibit -eudesmol-induced MAPK phosphorylation (Y. Obara, T. Aoki, and M. Kusano, unpublished observation).

-Eudesmol also caused the phosphorylation of CREB at Ser 133, which is a key regulatory site for the control of its transcription activity (Fig. 6 and 7). CREB phosphorylation was suppressed by both U-73122 and PD98059 (Fig. 8A), implicating the involvement of PI-hydrolysis and MAPK activation in this suppression. In addition, both W7 and H89 also clearly reversed CREB phosphorylation (Fig. 8B and 8C). So far, it has been shown that the phosphorylation of CREB at Ser 133 occurs via p90 ribosomal S6 kinase (RSK) downstream of MAPK, CaM-kinase, and PKA (Gonzalez and Montminy, 1989; Sheng et al., 1991; Xing et al., 1996). Recently, Impey et al. (1998) showed that the Ca²⁺-mediated phosphorylation of CREB in PC-12 cells occurred via MAPK-dependent activation of the CREB kinase RSK2. PKA was also required for regulation of the nuclear translocation of MAPK-RSK2, a prerequisite for CREB phosphorylation. Based on this report, we speculate that the phosphorylation of CREB induced by -eudesmol is dependent on the MAPK/RSK pathway, which is assumed to be activated via the CaM/PKA pathway, as described above. However, we do not deny the possibility of direct phosphorylation of CREB by CaM-kinase or PKA. It has been shown that NGF potentiates neuronal functions in PC-12 and bovine adrenal chromaffin cells, which is accompanied by the induction of the key enzymes in catecholamine biosynthesis, such as tyrosine hydroxylase and dopamine--hydroxylase, and of ion channels, such as voltage-gated Na⁺ and Ca²⁺ channels (Acheson et al., 1984; Bouron et al., 1999). Since one physiological role of CREB in NGF-signaling is assumed to be the induction of a series of these catecholamine-synthesizing enzymes (Ghee et al., 1998), then it follows that these enzymes are up-regulated by -eudesmol (i.e., functional differentiation). Further study is necessary to examine the role of -eudesmol in the functional differentiation of PC-12 cells.

The concentration-dependence curve of the -eudesmol-evoked increase in [Ca²⁺]_i showed a biphasic response in PC-12 cells (Fig. 3). In addition, the [Ca²⁺]_i increase was not completely inhibited by PI-PLC inhibition (Fig. 5A). These results suggest that there are multiple mechanisms responsible for this [Ca²⁺]_i regulation. Previously, Satoh et al. (1992) demonstrated that -eudesmol suppressed Na⁺, K⁺-ATPase and Ca²⁺-ATPase activity, with IC₅₀ values of 160 µM and 1.1 mM, respectively. Although the concentration-range of -eudesmol within which these inhibitory effects were mediated was higher than that used in the present study, its inhibitory effects on these Ca²⁺-regulating ion pumps are thought to be one of other mechanisms in the [Ca²⁺]_i increase induced by -eudesmol since ouabain caused an increase in [Ca²⁺]_i by inhibition of Na⁺, K⁺-ATPase (Hochstrate and Schlue, 2001). In contrast to the results of our study, Tachikawa et al. (2000) demonstrated that -eudesmol prevented the Ca²⁺ influx induced by both acetylcholine and high KCl in a concentration-dependent manner in cultured bovine adrenal chromaffin cells and that the acetylcholine-induced secretion of catecholamines was consequently abolished by -eudesmol. One interpretation of these conflicting findings is that high concentrations (100 µM) of -eudesmol affect at least two distinct sites that promote Ca²⁺ release and inhibit Ca²⁺ influx, respectively, although its effects on Ca²⁺ release predominates in PC-12 cells. In other words, since Ca²⁺ release was lacking in bovine adrenal chromaffin cells, it was assumed that the [Ca²⁺]_i increase was not observed (Tachikawa et al., 2000).

PI-PLC is classified into three major groups (, , and ) on the basis of molecular mass, deduced amino sequence, and immunological cross-activity. So far, 10 different mammalian PI-PLC isoenzymes (_1-4, _1-2, and _1-4) have been characterized (Rebecchi and Pentyala, 2000). PLC- has been shown to be regulated by G-proteins, such as the -subunits of a pertussis toxin-insensitive G_q family of G-proteins and -subunits of G-proteins, whereas PLC- was shown to be regulated by tyrosine phosphorylation by the action of tyrosine kinases of growth factor receptors or members of the src family, such as Src and Lyn (Rebecchi and Pentyala, 2000). In contrast to PLC- and - isoenzymes, the physiological role and regulation of PLC- family isoenzymes have not been determined clearly, despite their wide distribution. It has been shown that PC-12 cells express at least these three PLC isoenzymes (, , and ) in sufficient quantities to be detected by Western blot (Ryu et al., 1990); however, the detailed expression patterns of the isoenzymes in each family have not been clearly defined. Which isoenzymes of PLC are involved in -eudesmol-signaling and whether -eudesmol directly activates PLC also remain to be investigated.

In conclusion, our data suggest that -eudesmol induces neurite outgrowth in PC-12 cells, which is accompanied by MAPK activation. -Eudesmol, being a small molecule, may be a promising lead compound for potentiating neuronal function. In addition, the drug may be useful in helping to clarify the mechanisms underlying neuronal differentiation.