Address correspondence and reprint requests to Dr Hitoshi Nakayama, Department of Pharmacology, Nara Medical University, 840 Shijo-Chou, Kashihara, Nara 634–8521, Japan. E-mail: email@example.com
We have investigated mechanisms of nicotine-induced phosphorylation of extracellular signal-regulated protein kinase (p42/44 MAP kinase, ERK) and cAMP response element binding protein (CREB) in PC12h cells. Nicotine transiently induced ERK phosphorylation at more than 1 µm. The maximal level of nicotine-induced ERK phosphorylation was lower than that of the membrane depolarization induced and, to a great extent, the nerve growth factor (NGF)-induced ERK phosphorylation. Nicotinic acetylcholine receptor (nAChR) α7 subunit-selective inhibitors had no significant effect on nicotine-induced ERK phosphorylation. l-Type voltage-sensitive calcium channel antagonists inhibited nicotine-induced ERK phosphorylation. Calcium imaging experiments showed that α7-containing nAChR subtypes were functional at 1 µm of nicotine in the nicotine-induced calcium influx, and non-α7 nAChRs were prominent in the Ca2+ influx at 50 µm of nicotine. An expression of dominant inhibitory Ras inhibited nicotine-induced ERK phosphorylation. A calmodulin antagonist, a CaM kinase inhibitor, a MAP kinase kinase inhibitor inhibited nicotine-induced ERK and CREB phosphorylation. The time course of the phosphorylation of CREB induced by nicotine was similar to that of ERK induced by nicotine. These results suggest that non-α7 nAChRs are involved in nicotine-induced ERK phosphorylation through CaM kinase and the Ras-MAP kinase cascade and most of the nicotine-induced CREB phosphorylation is mediated by the ERK phosphorylation in PC12h cells.
sodium dodecyl sulfate–polyacrylamide gel electrophoresis
voltage-sensitive calcium channels.
Nicotine has many acute and chronic pharmacological effects (Stolerman 1990; Brioni et al. 1997). Acute nicotine treatment stimulates neuronal nicotinic acetylcholine receptors (nAChRs) in many areas of the brain leading to a release of neurotransmitters, which are considered to cause many acute behavioral effects. Chronic nicotine treatment produces addiction and tolerance or sensitization to the acute effects of nicotine. Nicotine also affects cognitive performance (Changeux et al. 1998; Levin and Simon 1998). The mechanisms of the chronic effects remain unclear.
Neuronal nAChRs have been found to be expressed in adrenal medulla, PC12 cells and cell soma, dendrites and terminals of neurons in the peripheral and central nervous systems. Stimulation of the nAChRs with nicotine triggers an enhancement of Ca2+ influx into neurons. At the presynaptic terminals, enhanced Ca2+ influx is likely to be linked directly to the enhancement of the release of transmitters by stimulus-secretion coupling (Südhof 1995; Parnas et al. 2000). Nicotine stimulates transmitter release from synaptosomes prepared from brains in a Ca2+-dependent manner (Wonnacott et al. 1990), suggesting that nicotine-induced Ca2+ influx is linked directly to the transmitter release from terminals. Evidence has accumulated that presynaptic nAChRs in the brain mediate the release of dopamine, noradrenaline, acetylcholine, glutamate, serotonin, and γ-aminobutylate (Wonnacott et al. 1990; MacDermott et al. 1999). Thus a major function of nAChRs in the brain is to modulate synaptic transmission at presynaptic sites. However, until recently, the functional importance of nAChRs in the soma was not well characterized. A number of recent studies have shown that postsynaptic nAChRs in ciliary ganglion and hippocampus are involved in the fast and excitatory synaptic transmission (Zhang et al. 1996; Ullian et al. 1997; Frazier et al. 1998; Chang and Berg 1999), and suggested that postsynaptic nAChRs in neurons are functionally important.
Calcium signaling and its stimulation of gene expression are two of the most important issues in neuroscience (Impey et al. 1999; Ahn et al. 2000). Many studies on mechanisms of activity-dependent synaptic plasticity have focused on an increase in cytosolic and intracellular Ca2+ signaling in neurons. In particular, ionotropic glutamate receptor-mediated synaptic plasticity has been extensively investigated in connection with calcium signaling. In the Ca2+ signaling, activation of protein kinases including calmodulin-dependent protein kinase (CaM kinase), MAP kinase and cAMP-dependent protein kinase, and a transcription factor, CREB, have been particularly well studied (Shaywitz and Greenberg 1999; Soderling and Derkach 2000; Nestler 2001). Nicotine also activates CREB and MAP kinases, ERK1 and ERK2, in PC12 cells (Hiremagalur and Sabban 1995; Tang et al. 1998). Nerve growth factor (NGF) and cAMP have been suggested to regulate the level of nAChR α3 subunit mRNA via Ras-MAP kinase cascade and CREB in PC12 cells (Nakayama et al. 2000). Although nicotine stimulates activation of a number of genes, the calcium signaling pathways involved in nicotine-stimulated gene expression are not well characterized.
In the present study, we have investigated the signaling mechanisms of nicotine-induced phosphorylation of ERK and CREB in PC12 cells. Furthermore, we have investigated nAChR subtypes involved in the ERK phosphorylation.
Materials and methods
The following were purchased from commercial sources: U0126 from Promega (Madison, WI, USA); methyllycaconitine (MLA) from Calbiochem (San Diego, CA, USA); α-bungarotoxin (αBgt), nifedipine and diltiazem from Sigma (St Louis, MO, USA); fluo-3AM from Molecular Probes (Eugene, OR, USA); Block Ace from Dainihonseiyaku Co. (Osaka, Japan); BCA protein assay kit from Pierce (Rockford, IL, USA); antibody against MAP kinase (p44/42 MAP kinase), phospho-MAP kinase, CREB and phospho-CREB from New England Biolabs Inc. (Beverly, MA, USA); horseradish peroxidase-conjugated anti-rabbit IgG from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA); Dulbecco's modified Eagle's medium, Ham's F12 medium and horse serum from Gibco BRL, Life Technologies, Inc. (Rockville, MD, USA); and precolostrum newborn calf serum from Mitsubishi Kagaku (Tokyo, Japan). Electrochemiluminescent (ECL) western blot detection reagents and [3H]αBgt (1.52 TBq/mmol) were purchased from Amersham Pharmacia Biotech (Buckinghamshire, UK). All other chemicals were purchased from Nacalai Tesque (Kyoto, Japan) or Wako (Osaka, Japan).
Three kinds of PC12 cell lines (PC12h, a dominant inhibitory Ras mutant (M-M17–26) and the parental PC12 cell line) were maintained in 75 cm2 flasks using a 1 : 1 mixture (5/5/DF) of Dulbecco's modified Eagle's medium and Ham's F12 medium containing 5% precolostrum newborn calf serum and 5% heat-inactivated horse serum in a humidified atmosphere of 95% air and 5% CO2 at 37°C. For western blot analysis of phosphorylated proteins, all kinds of PC12 cells were plated at a density of 4–6 × 104 cells/cm2 onto collagen-coated 60-mm dishes in a 5/5/DF medium and then cultured either overnight or for 2 days. M-M17–26 cells and their parental PC12 cells were kindly supplied by Dr Geoffrey M. Cooper (Harvard Medical University) (Szeberényi et al. 1990).
Western blot analysis
On the next day, or 2 days later, after the plating of PC12 cells, the medium was replaced with serum-free DF medium. After being cultured for approximately 3 h, PC12h cells were stimulated with a given concentration of nicotine, 10 ng/mL NGF or 30 mm KCl for the times indicated in the figure legends. The cells were then washed once in 5 mL of phosphate-buffered saline, lysed in a buffer containing 62.5 mm Tris-HCl (pH 6.8), 2% sodium dodecyl sulfate (SDS) and 1% glycerol, and sonicated twice for 1 min. Proteins were identified by using a BCA protein assay kit. After an addition of 5% mercaptoethanol and heat treatment, lysate proteins (10 µg) were subjected to SDS–polyacrylamide gel electrophoresis (SDS–PAGE) on a 7.5% polyacrylamide gel. The electrophoresed proteins were transferred to polyvinylidene difluoride (PVDF) membranes. Detection of phosphorylated ERK and CREB was as described previously (Nakayama et al. 2000). Blots were developed using ECL detection reagents. The density of the band was quantified with a scanner and NIH Image software. Phosphorylated ERK was calculated by comparing the value obtained for phosphorylated ERK with the value obtained for ERK.
Imaging of intracellular Ca2+ in PC12h cells cultured for 2–3 days on cover glasses (Matsunami, Sizuoka, Japan) coated with collagen was performed. Cells were incubated at 37°C with 10 µm fluo-3AM diluted in KRH for 1 h. The dye intensity was monitored using a confocal laser microscope (RCM 8000: Nikon, Tokyo). Cells were irradiated with an excitation blue light beam (488 nm) produced by an argon ion laser at a scanning frequency of 1/30 s. The emitted fluorescence was guided through a 40 water-immersion objective to a pinhole diaphragm at 520 nm using a diachronic mirror. The intensity of emission from each cell targeted was scanned at 1/30 s intervals with a monitor video enhancer. Cells were treated with nicotine by bath application at concentrations of 1 and 50 µm. Inhibitors of nAChRs were added 20 min prior to application of nicotine. We then performed the imaging assay in the presence of these inhibitors. The Ca2+ imaging experiments were performed at room temperature. All experiments were performed four to seven times, and data were confirmed to be reproducible.
[3H]αBgt binding assay
PC12h cells were washed once with phosphate-buffered saline and collected by centrifugation. The collected cells were suspended in 10 mL of homogenizing medium (buffer A) containing 50 mm sodium phosphate buffer, 5 mm EDTA, 5 mm EGTA, 0.5 mm iodoacetamide, and 0.2 mm phenylmethylsulfonyl fluoride at pH 7.4, and then homogenized with a glass homogenizer fitted with a Teflon pestle. The homogenate was centrifuged at 60 000 g for 40 min. The pellet was washed once by suspension in the above homogenizing buffer, rehomogenization and recentrifugation. The obtained membrane fraction was suspended in buffer B containing 50 mm sodium phosphate buffer, 5 mm EDTA, 5 mm EGTA and a protease inhibitor cocktail. The protease inhibitor cocktail contained 100 mm 4-(2-aminoethyl)benzenesulfonyl fluoride, 80 µm aprotinin, 1.5 mm E-64, 2 mm leupeptin, 5 mm pepstatin and 1 mm pepstatin A, and was diluted 1/100 in the membrane suspension. Male Wistar rats (200∼250 g) were used for the [3H]αBgt binding experiments. Preparation of the membrane fraction was essentially as described previously (Nakayama et al. 1990) except for the above homogenizing buffer (buffer A) and suspending buffer (buffer B). [3H]αBgt binding assay was performed by rapid filtration. The assay mixture contained 10 nm[3H]αBgt (1.52 TBq/mmol), membrane proteins and a 1/2 dilution of buffer B in a total volume of 0.1 mL. After incubation of 2 h at 37°C, rapid filtration was performed using a Whatman GF/B filter presoaked in 0.3% polyethyleneimine as described previously (Nakayama et al. 1995). Specific binding was defined as the difference between the binding of [3H]αBgt in the absence and presence of 5 µmαBgt. In these assay conditions, the binding of [3H]αBgt to the PC12h or brain membrane fractions increased linearly with an increase to at least 70–200 µg of protein.
ERK phosphorylation induced by nicotine, high K+ and NGF in PC12h cells
Nicotine, NGF and membrane depolarization with high K+ induced ERK phosphorylation, but the level and time course of the nicotine-induced phosphorylation were different from those of high K+-and NGF-induced ERK phosphorylation (Fig. 1). Nicotine-induced ERK phosphorylation reached a maximum level within 5 min and returned to the basal level in 20 min. The level of high K+-induced ERK phosphorylation was higher than that of the nicotine-induced one and was sustained for up to 1 h. The level of ERK phosphorylation by NGF was much higher than that by nicotine and high K+. Nicotine induced ERK phosphorylation in a dose-dependent manner (Fig. 2). No ERK phosphorylation was found at less than 1 µm of nicotine.
Effects of nicotinic and calcium channel antagonists on nicotine-induced ERK phosphorylation in PC12h cells
Hexamethonium, a non-selective nAChR inhibitor, inhibited the ERK phosphorylation induced by nicotine, but not by high K+(Fig. 3a), showing that nAChRs mediate the phosphorylation. To investigate nAChR subtypes involved in nicotine-induced ERK phosphorylation, we examined the effects of selective inhibitors of nAChR subtypes (Fig. 3b). αBgt and MLA inhibit selectively α7-containing nAChR subtypes in the nanomolar range (Couturier et al. 1990; Ward et al. 1990; Palma et al. 1996), whereas d-tubocurarine (d-TC) is a non-selective antagonist of nAChR subtypes. No significant inhibition of nicotine-induced ERK phosphorylation was found at up to 5 µm of αBgt. MLA did not affect nicotine-induced ERK phosphorylation at 50 nm, and slightly inhibited the phosphorylation at 500 nm. MLA has been reported to inhibit α3- and α4-containing nAChRs at high concentrations of ACh (Drasdo et al. 1992; Palma et al. 1996). No or only a slight inhibitory effect of αBgt and MLA suggests no appreciable involvement of α7-containing nAChRs in nicotine-induced ERK phosphorylation in PC12h cells. d-TC completely inhibited the phosphorylation. These results suggest an involvement of non-α7-containing nAChRs in nicotine-induced ERK phosphorylation in PC12h cells. The α7-containing homo-oligomer expressed in Xenopus oocytes has a high permeability to Ca2+ (Séguéla et al. 1993). The Ca2+ permeability of nAChRs in PC12 cells is much higher than that of muscle type nAChRs (Sand and Barish 1991). However it has been shown that no αBgt-sensitive nAChRs contribute to the Ca2+ influx through neuronal nAChRs in adrenal chromaffin cells (Vermino et al. 1994). Then we investigated whether Ca2+ influx through nAChR channels or voltage-sensitive calcium channels (VSCCs) is crucial for nicotine-induced ERK phosphorylation in PC12h cells. Nifedipine and diltiazem are l-type VSCC antagonists. Both VSCC antagonists strongly inhibited nicotine-induced ERK phosphorylation in PC12h cells (Fig. 4). These results show that the nicotinic stimulation leads to an activation of VSCC, which induces ERK phosphorylation in PC12h cells.
Effects of a MAP kinase kinase (MEK) inhibitor and dominant inhibitory Ras on ERK phosphorylation by nicotine and high K+ depolarization in PC12 cells
ERK phosphorylation with high K+ is mediated by the Ras-MAP kinase cascade (Rosen et al. 1994). Then, we investigated whether the Ras-MAP kinase cascade also mediates nicotine-induced ERK phosphorylation (Fig. 5). For this purpose, we examined the effects of an MEK inhibitor, U0126 (Favata et al. 1998), and expression of dominant-negative Ras (Szeberényi et al. 1990) on nicotine-induced ERK phosphorylation. U0126 inhibited both nicotine- and high K+-induced ERK phosphorylation in PC12h cells (Fig. 5a). M-M17–26 cells highly express dominant-negative Ras which strongly inhibits NGF-induced neurite extension and Ras-mediated pathways (Szeberényi et al. 1990; Nakayama et al. 2000). In PC12 wild cells, the parent cells of M-M17–26 cells, U0126 inhibited nicotine-, high K+- and NGF-induced ERK phosphorylation (Fig. 5b, left panel), consistent with the results obtained in PC12h cells (Fig. 5a). Expression of dominant-negative Ras protein inhibited nicotine-, high K+-and NGF-induced ERK phosphorylation (Fig. 5b, right panel).
Effects of antagonists for calmodulin and calmodulin-dependent protein kinases (CaM kinases) on nicotine-induced ERK phosphorylation in PC12h cells
Recently, membrane depolarization with high K+ has been shown to induce the activation of at least two calcium signaling pathways in PC12 cells (Lev et al. 1995; Egea et al. 1999; Zwick et al. 1999). One is a CaM kinase-mediated pathway and the other is mediated by a cytosolic protein kinase, PYK2. Both pathways converge at the Ras-MAP kinase cascade. To investigate the involvement of the CaM kinase-mediated pathway in nicotine-induced ERK phosphorylation, we used two inhibitors, W7 and KN93, which have been reported to inhibit the actions of calmodulin and CaM kinases (Sumi et al. 1991; Nebigil and Malik 1993), respectively. Both inhibitors, which are structurally and functionally unrelated compounds, significantly inhibited nicotine-induced ERK phosphorylation in PC12h cells (Fig. 6). No significant inhibition of nicotine-induced Ca2+ influx in PC12h cells was obtained by 2 and 5 µm of KN93 or 2 µm of W7 (data not shown). Addition of 5 µm of W7 inhibited the Ca2+ influx less than 50%. Another CaM kinase inhibitor, KN62, also inhibited nicotine-induced ERK phosphorylation (data not shown). These results suggest that CaM kinases mediate nicotine-induced ERK phosphorylation in PC12h cells.
CREB phosphorylation induced by nicotine, high K+ and NGF in PC12h cells
It has been shown that a substantial part of the calcium influx-mediating activation of genes depends on CREB phosphorylation at Ser133 in neuronal cells and PC12 cells (Ahn et al. 2000). Although CREB phosphorylation on Ser133 can be potentially catalyzed by CaM kinases and RSK2, it has been shown that in PC12 cells, ERK phosphorylation by Ca2+ influx induces an activation of RSK2, which in turn directly phosphorylates CREB (Impey et al. 1998). CaMKIV, which is localized in the nuclear compartment of the neurons, was not detected in PC12 cells, whereas CaMKI and II were detected in the cytosol of PC12 cells (Finkbeiner et al. 1997; Impey et al. 1998). Then, we investigated signaling pathways of nicotine-induced CREB phosphorylation (Fig. 7). The time course of CREB phosphorylation by nicotine, high K+ and NGF was very similar to that of ERK phosphorylation by these effectors as shown in Fig. 1. U0126, KN93 and W7 strongly inhibited nicotine-induced CREB phosphorylation. Taken together with an absence of CaMKIV in PC12 cells (Finkbeiner et al. 1997; Impey et al. 1998), these results suggest that nicotine induces CREB phosphorylation via activation of CaMKI and/or CaMKII and the Ras-MAP kinase cascade in PC12h cells. Since U0126 completely inhibited nicotine-induced CREB phosphorylation, RSK2 seems to be a candidate for a CREB kinase in CREB phosphorylation induced by nicotine in PC12h cells.
Nicotine-induced Ca2+ influx in PC12h cells
Figure 3(b) strongly suggests that non-α7-containing nAChRs are involved in nicotine-induced ERK phosphorylation in PC12h cells. This led us to the question of whether α7-containing nAChRs are poorly expressed or α7-mediating calcium signaling does not induce ERK phosphorylation in PC12h cells. In chick ciliary ganglion neurons, the relative contribution of α7-containing nAChR subtypes to nicotine-induced Ca2+ influx depends on nicotine concentrations added to the culture medium; that is, many kinds of nAChR subtypes are expressed in the neuron and the contribution of α7-containing nAChR subtypes to nicotine-induced Ca2+ influx is larger at lower nicotine concentrations (Vijayaraghavan et al. 1992). Thus we examined the involvement of α7-containing nAChRs in nicotine-induced calcium influx in PC12h cells (Fig. 8). For this purpose, we performed imaging of intracellular Ca2+ in cell culture with fluo-3 followed by confocal laser microscopy. At 1 and 50 µm, nicotine produced a rapid detectable response (Fig. 8a). The increase in fluorescence at 50 µm of nicotine was approximately two-folds higher than that at 1 µm(Fig. 8b). d-TC completely inhibited both the 1 and 50 µm nicotine-induced increase in fluorescence. In these calcium imaging experiments, αBgt and MLA were used at 0.5 µm, which had no or only a slight effect on nicotine-induced ERK phosphorylation (Fig. 3b). At 1 µm of nicotine, αBgt almost completely inhibited the nicotine-induced increase in fluorescence (Figs 8b and c). MLA also significantly inhibited the nicotine-induced increase in fluorescent at 1 µm of nicotine. In contrast, at 50 µm of nicotine, no significant effect of αBgt or MLA on the nicotine-induced increase in fluorescence was found. These results strongly suggest that functional α7-containing nAChR subtypes are involved in nicotine-induced Ca2+ influx in PC12h cells at less than 1 µm of nicotine, but Ca2+ influx mediated by α7-containing nAChR subtypes is not likely to induce ERK phosphorylation in PC12h cells. Ca2+ influx induced by 50 µm of nicotine seems to be mediated by non-α7-containing nAChRs in PC12h cells.
ΑBgt binding to membranes of PC12h and cerebral cortex
Next we determined [3H]αBgt binding activity in the membrane fraction prepared from PC12h cells. For comparison, we also detected [3H]αBgt binding activity in the membrane fraction prepared from rat cerebral cortex. When 10 nm[3H]αBgt was used, αBgt binding activity was 0.943 ± 0.182 pmol/mg protein (n = 4) in PC12h cells. In the same assay condition, the binding activity was 0.0682 ± 0.0227 pmol/mg protein (n = 4) in the cerebral cortex. These results support the expression of α7-containing nAChRs in PC12h cells.
The present study provides new information on nicotine-induced signaling pathways in PC12h cells. First, nicotine-induced ERK phosphorylation was transient and its level was lower than that of ERK phosphorylation induced by high K+ and NGF. Second, non-α7-containing nAChR subtypes have been suggested to be involved in ERK phosphorylation based on the following: (i) no ERK phosphorylation was found at less than 1 µm of nicotine; (ii) selective inhibitors of α7-containing nAChRs have only a slight or no effect on nicotine-induced ERK phosphorylation; and (iii) l-type calcium channel antagonists almost completely inhibited nicotine-induced ERK phosphorylation. Furthermore, calcium imaging and α-Bgt-binding experiments show that α7-containing nAChRs are functional in PC12 cells, but their contribution to nicotine-induced ERK phosphorylation is negligible under the conditions of the present experiments. Third, CaM kinases and the Ras-MAP kinase cascade seems to be involved in nicotine-induced ERK phosphorylation. Fourth, nicotine-induced CREB phosphorylation is suggested to be mediated by ERK phosphorylation.
Presynaptic α7-containing nAChRs have been shown to be involved in fast excitatory synaptic transmission at less than 1 µm of nicotine in brains (MacGehee et al. 1995; Gray et al. 1996). Similarly, much evidence has accumulated that nicotine at less than 1 µm has physiological effects. These findings are connected with smoking. Smoking causes a rapid increase in nicotine concentrations in serum and brain and then the levels of nicotine fall rapidly but remain low for a long time (Benowitz 1996). Acute nicotinic actions reflect the rapid increase in nicotine in the target organs, but it remains unclear whether chronic effects depend on nicotine concentrations in the initial phase (high nicotine levels) or in the latter phase (prolonged lower nicotine levels). Furthermore, although the relationships between nAChR subtypes and acute physiological effects have been extensively investigated, the nAChR subtypes involved in chronic effects are poorly characterized. Chronic nicotinic effects can persist after nicotine withdrawal (Levin and Simon 1998), suggesting that acute and/or chronic nicotine treatment lead to the activation of a number of genes. Thus it is important to investigate nicotine-activating signaling pathways inducing the activation of gene expression and nAChR subtypes involved in the signaling pathways. The data presented here show no involvement of α7-containing nAChRs in nicotine-induced ERK phosphorylation in PC12h cells. Recently, levels of α7-containing nAChRs have been shown to differ markedly among populations of PC12 cells (Blumenthal et al. 1997). Considering a report that the relative contribution of α7-containing nAChRs to nicotine-induced calcium influx is decreased with an elevation of nicotine concentration in cultured chick ciliary ganglion (Vijayaraghavan et al. 1992), we investigated the effects of αBgt and MLA on nicotine-induced Ca2+ influx at low and high nicotine concentrations and found that these selective inhibitors of α7-containing nAChRs almost completely inhibited nicotine-induced Ca2+ influx at 1 µm of nicotine but had no effect on the Ca2+ influx at 50 µm. These results suggest that α7-containing nAChRs are involved in nicotine-induced Ca2+ influx at less than 1 µm of nicotine. αBgt binding activities for membrane fractions prepared from PC12h cells and rat cerebral cortex support the calcium imaging experiments, though the binding activity in the membrane surface was not determined. Since α3-containing nAChRs are highly expressed in PC12 cells (Rogers et al. 1992; Ishiguro et al. 1997), it is possible that at high concentrations of nicotine, α3-containing nAChRs dominate nicotine-induced calcium influx and ERK phosphorylation. Since α7-containing nAChRs are very rapidly desensitized (Séguéla et al. 1993) and many kinds of nAChR subunits are highly expressed in PC12h cells (Rogers et al. 1992; Henderson et al. 1994; Ishiguro et al. 1997), selective inhibitors of α7-containing nAChRs may have no effect on nicotine-induced Ca2+ influx at 50 µm of nicotine in PC12h cells. α7-containing nAChRs have high permeability for Ca2+ (Séguéla et al. 1993). Taken together, it is possible that at 1 µm of nicotine, α7-containing nAChRs and/or l-type VSCCs are important to Ca2+ influx, whereas at 50 µm of nicotine, l-type VSCCs activated by non-α7 nAChRs-dependent depolarization induce the Ca2+ influx. Recently it has been reported that continual exposure of PC12 cells to nicotine induces a two-phase increase in the intracellular Ca2+ concentration (Gueorguiev et al. 1999, 2000). The initial increase is followed by a return to the basal level in 10 min, but after 5–10 min, a second increase in Ca2+ occurs and lasts for several hours. The second phase is mediated by α7-containing nAChRs. In the present study, we did not detect nicotine-induced ERK phosphorylation after 2 h of nicotine treatment. Thus the second increase in Ca2+ may induce ERK phosphorylation in PC12h cells, although we did not examine the second increase in Ca2+ in PC12h cells.
Recent reports have shown that in PC12 cells, membrane depolarization with high K+ induces activation of CaM kinase followed by sequential activation of EGFR and the Ras-MAP kinase cascade (Egea et al. 1999; Zwick et al. 1999). The present report also suggests nicotine-induced activation of CaM kinases and the Ras-MAP kinase cascade in PC12h cells. We tried to detect phosphorylation of EGFR but did not obtain clear results. The activation of EGFR by nicotine may be weak. A nicotinic agonist and membrane depolarization induce PYK2-mediated ERK phosphorylation via the Ras-MAP kinase cascade in PC12 cells (Lev et al. 1995; Zwick et al. 1999). In the present study, we did not try to detect activation of PYK2 and thus the possibility that nicotine also activates PYK2 is not excluded. NMDA receptors and membrane depolarization with high K+ mediate Ca2+ signaling to regulate gene expression through two different Ca2+ signaling pathways in hippocampal neurons (Bading et al. 1993). In this case, the high Ca2+ permeability of NMDA receptors may induce activation of Ca2+ signaling pathways other than membrane depolarization-mediated pathways. The present study suggests that in PC12h cells, both nicotine and high K+ activate VSCCs to induce Ca2+ influx and activate similar Ca2+ signaling pathways. Nevertheless, the present findings, that the levels and duration of ERK phosphorylation after nicotinic stimulation were different from those after membrane depolarization, raise the possibility that two different stimulations lead to different effects on gene expression via phosphorylation of ERK and CREB.
Although CaM kinases catalyze in vitro phosphorylation of CREB (Dash et al. 1991; Sheng et al. 1991), CaM kinases did not catalyze NGF-induced CREB phosphorylation in PC12 cells (Ginty et al. 1994). Recent studies have shown that membrane depolarization with high K+ induces CREB phosphorylation via sequential activation of ERK and RSK2, a CREB kinase, in cultured hippocampal neurons and PC12 cells: RSK2 activated by ERK translocates to the nucleus and then catalyzes phosphorylation of CREB (Impey et al. 1998). In addition, CaMKIV may contribute to CREB phosphorylation by membrane depolarization in the hippocampal neurons, whereas no CaMKIV is expressed in PC12 cells (Finkbeiner et al. 1997; Impey et al. 1998). In accordance with these reports, the present results obtained using inhibitors of calmodulin, CaM kinases and MEK suggest that CaM kinase II, Ras-MAP kinase cascade and RSK2 mediate nicotine-induced CREB phosphorylation. In particular, the complete inhibition of nicotine-induced CREB phosphorylation by a MEK inhibitor strongly suggests that ERK signaling is necessary for nicotine-induced CREB phosphorylation, and CaMK II acts upstream of ERK. Interestingly, CREB phosphorylation induced by nicotine has been shown to be sustained for between 0.5 and 6 h in rat hippocampus (Hiremagalur and Sabban 1995), which contrasts with the transient phosphorylation of CREB by nicotine in PC12h cells. What causes the prolonged CREB phosphorylation is to be investigated.
It has been shown that nicotine-induced calcium influx causes the activation of a number of genes. Acute nicotine treatment induces the activation of immediate early genes including c-fos and c-jun in PC12 cells, adrenal medulla, peripheral ganglion tissues and many regions of the brain (Greenberg et al. 1986; Koistinaho et al. 1993; Matta et al. 1993; Pang et al. 1993; Kiba and Jayaraman 1994; Clegg et al. 1995; Pelto-Huikko et al. 1995; Panagis et al. 1996), though the signaling pathways have not been well characterized. The activation of the c-fos gene is mediated by activation of ERK and/or CREB in neurons. Thus it is possible that nicotine-induced ERK and CREB phosphorylation results in the activation of immediate early genes followed by the activation of late-response genes. In the present study, the nicotine-induced ERK phosphorylation was transient and its level was much lower than the level of NGF-induced ERK phosphorylation. NGF and BDNF induce the activation of a number of genes, which are thought to be important to synaptic plasticity (Schinder and Poo 2000). Since the ERK phosphorylation induced by nicotine is much lower than that induced by NGF, the weak activation of ERK by nicotine may have little effect on the activation of genes in PC12h cells. NGF and cAMP induce ERK phosphorylation and affect the level of nAChR α3 subunit mRNA in PC12 cells (Nakayama et al. 2000). Nicotine-induced phosphorylation pathways of ERK and CREB are potentially feedback pathways for the α3 expression. However, both the level and duration of nicotine-induced phosphorylation of ERK and CREB are much lower than those of the NGF- and cAMP-induced one. It has been reported that although nicotine treatment of PC12 cells induced a transient or slight decrease in the α3 subunit mRNA (Madhok et al. 1995; Takahashi et al. 1999), long-term treatment of PC12 cells with nicotine had no effect on the α3 nAChR subunit mRNA level (Ishiguro et al. 1997). This may at least in part be due to weak activation of ERK by nicotine in PC12h cells. In support of this assumption, the nicotine-induced ERK phosphorylation was lower than that of the membrane depolarization-induced one and membrane depolarization results in no up-regulation of α3 nAChR subunit mRNA. It seems likely that the level and/or duration of ERK phosphorylation is important to changes in the α3 subunit mRNA.
We thank Professor Geoffrey M. Cooper for the gift of M-M17–26 cells and the parental PC12 cells. This work was supported in part by a grant-in-aid from the Ministry of Education, Science and Culture, Japan and a research grant from the Smoking Research Foundation, Japan. TN is a JSPS Research Fellow.
1Present address: Laboratory of Neurobiology, Faculty of Engineering, Kansai University, Suita, Osaka 564-8680, Japan.