Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

5-Hydroxytryptamine (5-HT)_2A receptors are G_q-coupled receptors, expressed by a variety of cell types, that serve numerous functions. For example, receptors expressed on platelets play a role in stimulating aggregation. In contrast, on vascular smooth muscle cells, they stimulate contraction (Roth et al., 1986; Watts et al., 1996). In the central nervous system, 5-HT_2A receptor expression is heterogeneous. Receptors are expressed at high density in areas such as the neocortex, claustrum, anterior cingular cortex, mammillary nuclei, and basal ganglia. However, the receptors are expressed at low densities in the hippocampus, brainstem, and thalamus (Pazos et al., 1985). 5-HT_2A receptors have been implicated to be involved in both the pathophysiology and treatment of a number of psychiatric disorders. A role for the receptor in psychotic disorders is consistent with the observations that hallucinogens, such as lysergic acid diethylamide, act as partial receptor agonists (Glennon et al., 1984; Ferry et al., 1993; Roth et al., 1999), whereas the newer, "atypical" antipsychotics are 5-HT_2A receptor antagonists (Meltzer et al., 1989; Roth et al., 1999). A role for 5-HT_2A receptors in the pathophysiology of depression has also been proposed. Receptor expression in prefrontal cortex has been shown, in post-mortem studies, to be increased in suicide victims (Mann et al., 1989).

Although 5-HT_2A receptors seem to couple to signaling pathways that exert long-term neuronal changes, the pathway for receptor-coupling to ERK mitogen-activated protein kinases has not been delineated in neurons. These kinases have been shown to phosphorylate a number of transcription factors, including c-Jun, p62TCF/Elk-1, c-Fos, and c-Myc, and also seem to regulate translation of mRNA (Denton and Tavare, 1995). The ERK pathway is known to enhance cell survival and is required for normal neuronal functioning (Encinas et al., 1999; Erhardt et al., 1999). In fact, it seems to have a role in neurotrophin-stimulated neuronal differentiation and neuroprotection (Pang et al., 1995; Hetman et al., 1999). Coupling of 5-HT_2A receptors to activation of ERK has been demonstrated to occur in vascular and tracheal smooth muscle cells (Hershenson et al., 1995; Watts et al., 1996), as well as in mesangial cells (Greene et al., 2000). In both smooth muscle cells and mesangial cells, 5-HT_2A receptors were found to stimulate activation of ERK through the G protein G_q and consequent activation of protein kinase C (PKC) (Hershenson et al., 1995; Greene et al., 2000). Interestingly, findings from two studies by Zwiller and colleagues suggest that PC12 cells may serve as a useful model for studying the coupling of 5-HT₂ receptors to activation of ERK in neuronal cell types. Treatment of PC12 cells with 5-HT was reported to stimulate ERK (Esteve et al., 2001). In a separate study, a receptor with pharmacology consistent with 5-HT_2A/5-HT_2C receptors was found to mediate PC12 cell induction of TIS8/egr-1 and c-fos expression (Humblot et al., 1997). Together, these findings suggest the possibility that endogenous 5-HT₂ receptors couple to activation of ERK in PC12 cells. In the current studies we directly demonstrate that the activation of ERK is mediated through 5-HT_2A receptors. Unlike the findings reported for non-neuronal types of cells, we found that the activation of ERK, in PC12 cells, occurs through a PKC-independent pathway requiring Ca²⁺/calmodulin and tyrosine kinases.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Materials. MDL 100907 was kindly provided by Aventis (Bridgewater, NJ). Gö6983, Gö6976, H89, KN93, W13, bisindolylmaleimide I, Ro-31-8220, thapsigargin, ionomycin, PP1, AG 1478, and tyrphostin A9 (AG 17) were obtained from Calbiochem (San Diego, CA). 5-HT, ketanserin, R-()-2,5-dimethoxy-4-iodoamphetamine hydrochloride, KN62, SB206553, and W7 were obtained from Sigma-Aldrich (St. Louis, MO). Genistein was obtained from BIOMOL Research Laboratories (Plymouth Meeting, PA).

Cell Culture. PC12 cells were obtained from American Type Culture Collection (Manassas, VA) and were routinely cultured in Dulbecco's modified Eagle's medium supplemented with L-glutamine, minimal essential medium nonessential amino acids, 15% dialyzed fetal bovine serum (dialyzed in membranes with 1000-Da molecular weight cutoffs against a 100-fold greater volume of 150 mM NaCl to remove endogenous 5-HT), and 100 units of penicillin-100 µg of streptomycin/ml (95% air/5% CO₂). A stable, tightly adherent cell population was obtained after several cycles of washing off loosely adherent cells.

Immunoblots. Monoclonal anti-phospho-ERK1/ERK2 (Thr202/Tyr204) and anti-phospho-pan protein kinase C (Ser equivalent to Ser 660 of PKC II) were obtained from Cell Signaling (Beverly, MA). Rabbit polyclonal anti-total ERK1/ERK2, monoclonal anti-phospho-tyrosine, and horseradish peroxidase-conjugated secondary antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). The day before use, cells were washed with phosphate-buffered saline and cultured overnight under low-serum (0.5%) conditions. Cells were stimulated with the specified concentrations of agonists, and routinely lysed with a 26-gauge needle in 25 mM HEPES pH 7.4, 150 mM NaCl, 1% Triton X-100, 1 mM -glycerolphosphate, 50 mM NaF, 5 mM EDTA, 1 mM sodium orthovanadate, 250 µM 4-(2-aminoethyl)-benzene-sulfonylfluoride hydrochloride, 0.1% aprotinin, and 10 µg/ml leupeptin. Proteins were separated on 12% resolving gels (Bio-Rad, Hercules, CA) and transferred to 0.45-µm Immobolin-P polyvinylidene difluoride membranes (Millipore Corporation, Bedford, MA). Bound antibodies were visualized using Enhanced Luminol Chemiluminescence Reagent (PerkinElmer Life Sciences, Boston, MA) and direct exposure to a Kodak Image Station 440CF with a cooled, full-frame-capture charge-coupled device camera (Eastman Kodak, Rochester, NY). Net intensity of bands was calculated directly from stored images using Kodak Digital Science 1D Image Analysis Software (version 3.5) on defined regions of interest.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

5-HT_2A Receptors Couple to Activation of ERK in PC12 Cells. Similar to a previous report by Esteve et al. (2001), we found that 5-HT stimulated activation of ERK mitogen-activated protein kinases in PC12 cells. Treatment of cells with 5-HT caused large increases in the level of activated, double-phosphorylated ERK1 and ERK2 (Fig. 1). Maximal activation by 5-HT occurred at 10 µM and could be seen at concentrations greater than 10 nM (Fig. 1A). Phosphorylation of ERK occurred within 2 min of treatment and was maximal at 5 min (Fig. 1B). By 15 min, the level of activated ERK approached, but did not entirely reach, basal levels. Ketanserin (100 nM), an antagonist for 5-HT_2A/5-HT_2C receptors, inhibited the activation of ERK (Fig. 1C). To determine whether 5-HT_2A or 5-HT_2C receptors were the relevant receptors, MDL 100907 and SB206553, antagonists selective for 5-HT_2A and 5-HT_2B/2C receptors, respectively, were tested. Pretreatment with 10 nM MDL 100907 (K_i = 0.9 nM for 5-HT_2A and 90 nM for 5-HT_2C; Kehne et al., 1996) completely inhibited the actions of 5-HT. In contrast, pretreatment with 10 nM SB206553 (K_i = 1.6 µM for 5-HT_2A, 1 nM for 5-HT_2B, and 13 nM for 5-HT_2C; Kennett et al., 1996) caused no inhibition, demonstrating that activation of ERK was mediated by 5-HT_2A receptors. R-()-2,5-dimethoxy-4-iodoamphetamine hydrochloride (1 µM), an agonist for 5-HT_2A/5-HT_2C receptors, stimulated activation of ERK, but consistent with it being a partial agonist, stimulated an increase in phosphorylated ERK only 50% of that stimulated by 1 µM 5-HT (data not shown). PC12 cells have been reported to express 5-HT₃ receptors (Hanna et al., 2000). However, 10 µM LY-278,584, a potent, selective 5-HT₃ receptor antagonist, did not inhibit 5-HT-stimulated activation of ERK, nor did 10 µM 1-(m-chlorophenyl)-biguanide, a potent, selective 5-HT₃ receptor agonist cause activation (data not shown).

View larger version (22K): [in this window] [in a new window]   Fig. 1.   5-HT stimulates activation of ERK although 5-HT2A receptors. A, PC12 cells were treated for 5 min with the indicated concentrations of 5-HT and then lysed. B, cells were treated with 1 µM 5-HT for the indicated periods of time and then lysed. C, cells were treated for 5 min with 1 µM 5-HT in the absence or presence of 10 nM MDL 100907, 10 nM SB206553, or 100 nM ketanserin and then lysed. Total lysate was analyzed by immunoblotting with antibody to phospho-ERK1/ERK2 (p-ERK). Membranes were then stripped and analyzed with antibody to total ERK1/ERK2 (Total). Net intensities of bands were calculated from three or more separate experiments, performed in duplicate, and expressed as the means ± S.E.M. (× 103). , p < 0.001 versus 5-HT, analysis of variance, Bonferroni analysis. Representative immunoblots from one of the three experiments are shown.

Activation of ERK Requires an Increase in Intracellular [Ca²⁺]. Agonists for 5-HT_2A receptors have been shown to use G_q/11 type G proteins to stimulate phosphoinositide (PI) turnover and consequently increase the level of intracellular Ca²⁺ (Roth et al., 1986; Pritchett et al., 1988; Ferry et al., 1993). We therefore examined the role of Ca²⁺ in mediating 5-HT_2A receptor-stimulated activation of ERK. Activation of ERK was found to be inhibited by 50% when the extracellular [Ca²⁺] was reduced from 2 mM to 200 nM (a concentration similar to the intracellular [Ca²⁺] seen in many types of resting cells) by pretreatment with 2 mM EGTA (Fig. 2A). A similar magnitude of reduction in 5-HT-stimulated activation of ERK was seen when extracellular [Ca²⁺] was reduced further to 100 nM (data not shown). Pretreatment of cells with 30 nM thapsigargin, to slowly deplete intracellular stores of Ca²⁺ before treatment with 5-HT, caused a 63% reduction in activated ERK. Therefore, maximal 5-HT-stimulated activation of ERK requires an increase in intracellular [Ca²⁺] originating both from influx of extracellular Ca²⁺ and release from intracellular stores.

View larger version (24K): [in this window] [in a new window]   Fig. 2.   5-HT2A receptor-mediated activation of ERK requires Ca2+ but not PKC. A, PC12 cells were pretreated with 2 mM EGTA (to reduce extracellular [Ca2+] to 200 nM) or with 30 nM thapsigargin for 30 min before treatment with 1 µM 5-HT for 5 min. B, cells were treated with 1 µM 5-HT for the indicated periods of time and then lysed. C, PC12 cells were pretreated for 30 min with either 1 µM Ro-31-8220, 1 µM bisindolylmaleimide I, or 1 µM Gö6976 before treatment for 5 min with 1 µM 5-HT. Total lysate was analyzed by immunoblotting with antibody to phospho-ERK1/ERK2 (p-ERK) (A and C) or phospho-PKC (p-PKC) (B). Membranes were then stripped and analyzed with antibody to total ERK1/ERK2 (Total). Net intensities of bands were calculated from three or more separate experiments, performed in duplicate, and expressed as the means ± S.E.M. (× 103). , p < 0.01; , p < 0.001 versus 5-HT or analysis of variance, Bonferroni analysis. Representative immunoblots from one of the three experiments are shown.

Activation of ERK Is Independent of Protein Kinase C. PKC has been shown in studies of smooth muscle cells and mesangial cells to be required for coupling of 5-HT_2A receptors to activation of ERK (Hershenson et al., 1995; Greene et al., 2000). In fact, using an antibody directed toward PKC isoforms phosphorylated at the equivalent to Ser 660 of PKC II, we found that 5-HT stimulates autophosphorylation of PKC in PC12 cells (Fig. 2B). We therefore studied the role of PKC in mediating activation of ERK by 5-HT_2A receptors in these cells. Our finding that activation of ERK required Ca²⁺ suggested that the conventional Ca²⁺/diacylglycerol-dependent PKCs (, 1/2, and ) would be the potentially relevant kinases. However, pretreatment with 1 µM Gö6976, an inhibitor of the Ca²⁺/diacylglycerol-dependent PKCs, as well as the Ca²⁺-insensitive PKC µ (Martiny-Baron et al., 1993; Zang et al., 1994; Gschwendt et al., 1996) caused no inhibition of 5-HT-stimulated activation of ERK (Fig. 2C). To the contrary, an increase in 5-HT-stimulated activation was observed. Because PC12 cells express multiple isoforms of PKC, including Ca²⁺/diacylglycerol-dependent , 1/2, and ; Ca²⁺ insensitive/diacylglycerol-dependent , , and µ (protein kinase D); and Ca²⁺/diacylglycerol-independent  and / (Wooten et al., 1997), we also tested the effects of protein kinase C inhibitors with broader spectrums of inhibition. Concentrations (1 µM) of Ro-31-8220 that inhibit , 1/2, , , , , , and / isoforms of PKC (Standeart et al., 1997; Anthonsen et al., 2001) and bisindolylmaleimide I, which inhibits , 1/2, , , and  (Martiny-Baron et al., 1993), caused no inhibition of 5-HT-stimulated activation of ERK. To the contrary, each PKC inhibitor caused a 1.6-fold potentiation. Basal levels of activated ERK were not altered by treatment with the PKC inhibitors (data not shown). The 1 µM concentrations of inhibitors used were relatively high and were sufficient to inhibit PKC, as demonstrated in studies in which phorbol 12-myristate 13-acetate (PMA) was used to directly activate PKC. Although the activation of ERK stimulated by PMA was significantly greater than that stimulated by 5-HT, it was almost completely inhibited by 1 µM concentrations of the broad-spectrum PKC inhibitors Ro-31-8220 and bisindolylmaleimide I (Fig. 3A). As might be expected, Gö6976, with a profile of inhibition limited to the Ca²⁺/diacylglycerol-dependent PKCs and PKC µ, caused only partial inhibition. In contrast, even when the concentration of Ro-31-8220 was increased to 10 µM, no inhibition of 5-HT-stimulated ERK activation was observed (Fig. 3B). Similarly, 10 µM Gö6983, which inhibits all isoforms of PKC (including the atypical PKC ) except PKC µ (Gschwendt et al., 1996), caused complete inhibition of PMA-stimulated activation of ERK, but caused no inhibition of 5-HT-stimulated activation (Fig. 3C). Therefore, the activation of ERK stimulated by 5-HT_2A receptors was not mediated by any of the known isoforms of PKC.

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Unlike 5-HT-stimulated activation of ERK, the activation by PMA is sensitive to all PKC inhibitors. A, PC12 cells were pretreated for 30 min with either 1 µM Ro-31-8220, 1 µM bisindolylmaleimide I, or 1 µM Gö6976 before treatment for 5 min with 100 nM PMA. PC12 cells were pretreated with either 10 µM Ro-31-8220 (B) or 10 µM Gö6983 (C) for 30 min before treatment with 1 µM 5-HT or 100 nM PMA for 5 min and then lysed. Total lysate was analyzed by immunoblotting with antibody to phospho-ERK1/ERK2 (p-ERK). Membranes were then stripped and analyzed with antibody to total ERK1/ERK2 (Total). Net intensities of bands were calculated from three or more separate experiments, performed in duplicate, and expressed as the means ± S.E.M. (× 103). , p < 0.001 versus PMA, analysis of variance, Bonferroni analysis. Representative immunoblots from one of the three experiments are shown.

Activation of ERK Requires Calmodulin, but Is Independent of Calmodulin (CaM)-Dependent Protein Kinase. In the next set of studies, a role for the Ca²⁺ binding protein calmodulin was examined. Pretreatment of PC12 cells with 50 µM W7 or W13, two selective inhibitors of calmodulin, caused a 70% inhibition of 5-HT-stimulated activation of ERK (Fig. 4A). At this concentration, W7 and W13 have been shown to not alter the kinetics of Ca²⁺ entry into PC12 cells (Egea et al., 1999). Demonstrating that isolated increases in intracellular [Ca²⁺] would be sufficient to explain the actions of 5-HT, we found that directly increasing intracellular [Ca²⁺] with the Ca²⁺ ionophore ionomycin similarly stimulated activation of ERK (Fig. 4B). Pretreatment of cells with 50 µM W7 inhibited ionomycin-stimulated activation by approximately 60%. Increasing the concentration of W7 to 100 µM caused complete inhibition of both 5-HT- and ionomycin-stimulated ERK activation (data not shown) but also seemed to cause nonspecific effects in that the cells became more loosely attached to the culture dishes. The requirement for calmodulin suggested a possible role for CaM-dependent protein kinase. However, pretreatment of PC12 cells with 10 µM KN62 or KN93, two selective CaM kinase inhibitors with K_i values of 0.9 and 0.37 µM, respectively (Tokumitsu et al., 1990; Sumi et al., 1991), caused no inhibition of ERK activity (Fig. 4C).

View larger version (26K): [in this window] [in a new window]   Fig. 4.   5-HT2A receptor-mediated activation of ERK is sensitive to inhibitors of calmodulin, but not CaM kinase. PC12 cells were pretreated with 50 µM W7 or W13 for 30 min before treatment with 1 µM 5-HT (A) or 1 µM ionomycin (B) for 5 min and then lysed. C, cells were pretreated with either 10 µM KN62 or KN93 for 30 min before treatment with 1 µM 5-HT and then lysed. Total lysate was analyzed by immunoblotting with antibody to phospho-ERK1/ERK2 (p-ERK). Membranes were then stripped and analyzed with antibody to total ERK1/ERK2 (Total). Net intensities of bands were calculated from three or more separate experiments, performed in duplicate, and expressed as the means ± S.E.M. (× 103). , p < 0.001 versus 5-HT (A and C) or ionomycin (B), analysis of variance, Bonferroni analysis. Representative immunoblots from one of the three experiments are shown.

Activation of ERK Is Mediated by Tyrosine Kinases but Is Independent of Protein Kinase A. In PC12 cells, activation of Ca²⁺/calmodulin-activated forms of adenylyl cyclase has been reported to cause activation of ERK (Grewal et al., 2000). Therefore, the role of protein kinase A (PKA) in mediating 5-HT_2A receptor-stimulated activation of ERK was assessed using the PKA inhibitor H89. At the high concentration of 10 µM, H89 (K_i for PKA = 48 nM; Chijiwa et al., 1990) caused minimal inhibition of 5-HT-stimulated activation of ERK (Fig. 5A). In contrast, pretreatment with the tyrosine kinase inhibitor genistein caused almost complete inhibition of 5-HT-stimulated activation of ERK, suggesting a role for one or more tyrosine kinases. In fact, treatment of PC12 cells with 5-HT was found to stimulate tyrosine phosphorylation of several proteins (Fig. 5B). A protein doublet with a weight slightly greater than the 40-kDa molecular weight marker likely represented activated tyrosine-phosphorylated ERK1 (p44) and ERK2 (p42) (full activation of ERK requires phosphorylation of threonine 202 and tyrosine 204). Pretreatment with W7, to inhibit calmodulin, caused inhibition of 5-HT-induced tyrosine phosphorylation of all discernible proteins. However, phosphorylation of a protein migrating with a molecular weight below the 199-kDa marker was inhibited the least. Interestingly, treatment with W7 actually caused increased phosphorylation of a protein migrating just below the 87-kDa marker. However, phosphorylation of this protein was not increased by treatment with 5-HT, alone, relative to control.

View larger version (30K): [in this window] [in a new window]   Fig. 5.   Activation of ERK is mediated by tyrosine kinases but is independent of protein kinase A. PC12 cells were pretreated, as indicated with 10 µM H89, 100 µM genistein (gen) (A), or 50 µM W7 (B) for 30 min before treatment with 1 µM 5-HT for 5 min and then lysed. Total lysate was analyzed by immunoblotting with antibody to phospho-ERK1/ERK2 (p-ERK) (A) or to phospho-tyrosine (p-Tyr) (B). Membranes were then stripped and analyzed with antibody to total ERK1/ERK2. Net intensities of bands were calculated from three or more separate experiments, performed in duplicate, and expressed as the means ± S.E.M. (× 103). , p < 0.001 versus 5-HT, analysis of variance, Bonferroni analysis. Representative immunoblots from one of the three experiments are shown. Molecular weight markers (M) are shown in the left lane of blot in B.

View larger version (22K): [in this window] [in a new window]   Fig. 6.   Activation of ERK is sensitive to Src and EGF receptor kinase inhibitors. PC12 cells were pretreated, as indicated with 10 µM PP1, 10 µM tyrphostin A9 (tyr), or 100 nM AG 1478 (AG) for 30 min before treatment with 1 µM 5-HT for 5 min and then lysed. Total lysate was analyzed by immunoblotting with antibody to phospho-ERK1/ERK2 (p-ERK). Membranes were then stripped and analyzed with antibody to total ERK1/ERK2. Net intensities of bands were calculated from three or more separate experiments, performed in duplicate, and expressed as the means ± S.E.M. (× 103). , p < 0.001 versus 5-HT, analysis of variance, Bonferroni analysis. A representative immunoblot from one of the three experiments is shown.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Our finding that coupling of 5-HT_2A receptors to activation of ERK was independent of PKC was somewhat surprising. In tracheal smooth muscle cells and mesangial cells, 5-HT_2A receptors have been reported to stimulate activation of ERK through pathways sensitive to PKC inhibitors (Hershenson et al., 1995; Greene et al., 2000). Our finding of an absence of a role for PKC was not the result of lack of expression of PKC in PC12 cells. The cells have been shown to express multiple isoforms of PKC, including conventional isoforms (, 1/2, and ), novel isoforms (, , but not ), atypical isoforms ( and /), and PKC µ (also referred to as protein kinase D) (Wooten et al., 1997). In fact, we found that 5-HT stimulated activation of one or more PKC isoforms. A time-dependent autophosphorylation of PKC was observed using an antibody detecting phosphorylation at Ser 660 of PKC II and the homologous serines on other PKC isoforms. Interestingly, in contrast to the inhibition of PMA-stimulated ERK activation induced by pretreatment with all of the tested PKC inhibitors, a potentiation of 5-HT-stimulated activation was observed. This increase may have been the result of an attenuation of PKC-mediated inhibition of receptor-stimulated PI turnover, resulting in increased levels of intracellular Ca²⁺. Phorbol esters have been found to inhibit 5-HT_2A receptor-mediated PI turnover in rat aorta (Roth et al., 1986) and platelets (Kagaya et al., 1990), whereas the PKC inhibitors staurospine (Berg et al., 1998) and mezerein (Kagaya et al., 1990) have been shown to enhance 5-HT-stimulated PI turnover. Because 5-HT was found to cause activation of PKC in PC12 cells (demonstrated by autophosphorylation), it is not clear why this activation did not contribute to activation of ERK. It is possible that the specific PKC isoforms responsible for PMA-stimulated activation of ERK were not activated by 5-HT. Future studies will be required to determine whether activation of ERK in PC12 cells is mediated by only specific isoforms of PKC and whether the increases in Ca²⁺ and diacylglycerol stimulated by 5-HT_2A receptor occupancy are sufficient to activate these relevant isoforms.

The results of our studies are consistent with previous reports that activation of calmodulin is sufficient to stimulate activation of ERK in PC12 cells. For example, cell depolarization (Egea et al., 1998, 1999; Grewal et al., 2000) and agonists for nicotinic acetylcholine receptors (Nakayama et al., 2001) have been shown to induce activation of ERK through calmodulin-dependent pathways. However, the steps subsequent to calmodulin seem to vary, depending on the stimulus. For example, the activation of ERK resulting from depolarization has been reported to be inhibited by the protein kinase A inhibitors H89 and KT5720 (Grewal et al., 2000), suggesting a requirement for calcium/calmodulin-activated adenylyl cyclase. In contrast, we found that H89 did not inhibit the activation of ERK stimulated by 5-HT_2A receptors. Similarly, nicotinic acetylcholine receptor-stimulated activation of ERK in PC12 cells has been reported to be sensitive to CaM kinase inhibitors (Nakayama et al., 2001), whereas we found no inhibition of 5-HT_2A receptor-stimulated activation.

G_q-coupled M₁-muscarinic acetylcholine receptors have also been reported to couple to activation of ERK, in PC12 cells, through a pathway requiring Ca²⁺, but independent of PKC. Interestingly, that pathway required the Ca²⁺/diacylglycerol-regulated guanine nucleotide exchange factor I (Guo et al., 2001). Although our findings demonstrate that the activation of ERK stimulated by 5-HT_2A receptors requires activation of calmodulin by Ca²⁺, we cannot rule out an additional role for Ca²⁺ in activating Ca²⁺/diacylglycerol-regulated guanine nucleotide exchange factor I. It will be interesting to study the role of this guanine nucleotide exchange factor when inhibitors become available.

5-HT was found to stimulate tyrosine phosphorylation of several proteins. A role for one or more tyrosine kinases in the receptor-mediated activation of ERK was demonstrated by 1) inhibition of ERK activation by the tyrosine kinase inhibitor genistein, and 2) by inhibition of tyrosine phosphorylation by the calmodulin inhibitor W7. Although it is not currently clear which tyrosine kinase, or kinases, are the relevant kinase, a number of possibilities are suggested by findings from studies of other G_q-coupled receptors. For example, transfected -_1A adrenergic receptors have been shown to cause activation of PYK2 in transfected PC12 cells (Berts et al., 1999). Additionally, tyrosine phosphorylation of the epidermal growth factor receptor has been reported to be induced by depolarization (Egea et al., 1998, 1999). However, unlike depolarization-stimulated activation of ERK, depolarization-stimulated phosphorylation of epidermal growth factor receptor is not inhibited by calmodulin inhibitors (Egea et al., 1998, 1999). Src represents another possible tyrosine kinase. In adrenal chromaffin cells, angiotensin II receptors activate ERK through a pathway inhibited by the Src inhibitor PP1 (10 µM) (Cammarota et al., 2001).

We have begun preliminary experiments to determine which kinases are relevant to 5-HT_2A receptor-mediated activation of ERK. We have determined that activation of ERK is sensitive to the Src inhibitor PP1 and to the EGF receptor kinase inhibitor AG 1478. In contrast, the platelet-derived growth factor receptor kinase inhibitor AG 17, which has been shown in neutrophils to additionally selectively inhibit PYK2 (Avdi et al., 2001; Fuortes et al., 1999), caused no inhibition. Therefore, it seems that activation of Src and transactivation of the EGF receptor are required for 5-HT_2A receptor-mediated activation of ERK in PC12 cells.