The signal transduction pathway underlying ion channel gene regulation by SP1-C-Jun interactions.

During neuronal differentiation, an exquisitely controlled program of signal transduction events takes place, leading to the temporally and spatially regulated expression of genes associated with the differentiated phenotype. A critical class of genes involved in this phenomenon is that made up of genes encoding neurotransmitter-gated ion channels that play a central role in signal generation and propagation within the nervous system. We used the well established PC12 cell line to investigate the molecular details underlying the expression of the neuronal nicotinic acetylcholine receptor class of ion channels. Neuronal differentiation of PC12 cells can be induced by nerve growth factor, leading to an increase in neuronal nicotinic acetylcholine receptor gene expression. Nerve growth factor initiates several signal transduction cascades. Here, we show that the Ras-dependent mitogen-activated protein kinase and phosphoinositide 3-kinase pathways are critical for the nerve growth factor-mediated increase in the transcriptional activity of a neuronal nicotinic acetylcholine receptor gene promoter. In addition, we show that a component of the Ras-dependent mitogen-activated protein kinase pathway, nerve growth factor-inducible c-Jun, exerts its effects on receptor gene promoter activity most likely through protein-protein interactions with Sp1. Finally, we demonstrate that the target for nerve growth factor signaling is an Sp1-binding site within the neuronal nicotinic acetylcholine receptor gene promoter.

During neuronal differentiation, an exquisitely controlled program of signal transduction events takes place, leading to the temporally and spatially regulated expression of genes associated with the differentiated phenotype. A critical class of genes involved in this phenomenon is that made up of genes encoding neurotransmitter-gated ion channels that play a central role in signal generation and propagation within the nervous system. We used the well established PC12 cell line to investigate the molecular details underlying the expression of the neuronal nicotinic acetylcholine receptor class of ion channels. Neuronal differentiation of PC12 cells can be induced by nerve growth factor, leading to an increase in neuronal nicotinic acetylcholine receptor gene expression. Nerve growth factor initiates several signal transduction cascades. Here, we show that the Ras-dependent mitogen-activated protein kinase and phosphoinositide 3-kinase pathways are critical for the nerve growth factor-mediated increase in the transcriptional activity of a neuronal nicotinic acetylcholine receptor gene promoter. In addition, we show that a component of the Ras-dependent mitogen-activated protein kinase pathway, nerve growth factor-inducible c-Jun, exerts its effects on receptor gene promoter activity most likely through protein-protein interactions with Sp1. Finally, we demonstrate that the target for nerve growth factor signaling is an Sp1-binding site within the neuronal nicotinic acetylcholine receptor gene promoter.
Nerve growth factor (NGF) 1 is critical for the survival and differentiation of sensory and sympathetic neurons in the peripheral nervous system (1) and of basal forebrain and hippocampal cholinergic neurons in the central nervous system (2). NGF signals through binding to its receptors, a high affinity tyrosine kinase receptor, TrkA, and a low affinity receptor, p75 (3). Binding of NGF to its receptors initiates several signal transduction cascades, including, among others, ras-dependent mitogen-activated protein kinase (MAPK), phosphoinositide 3-kinase (PI3-K), and the cAMP-dependent protein kinase pathways (3). Some of the downstream targets of the activated protein kinase pathways are transcription factors such as AP-1, cAMP responsive element-binding protein, and NF-B.
Rat pheochromocytoma PC12 cells have been extensively used to study the molecular mechanisms involved in NGF signaling. PC12 cells are chromaffin-like cells that in response to NGF treatment withdraw from the cell cycle and differentiate into sympathetic-like neurons, a process accompanied by neurite outgrowth, increased electrical excitability, and changes in neurotransmitter synthesis (4 -6). Among the signal transduction cascades activated by NGF in PC12 cells, the MAPK and the PI3-K pathways are thought to play a central role, although other pathways are clearly important (7)(8)(9)(10). Activated TrkA promotes conversion of the membrane-anchored Ras-GDP into Ras-GTP. Ras then cooperates in activation of a serine/threonine kinase, Raf-1, which phosphorylates the dual specificity MAPK/extracellular signal-regulated kinase (ERK) kinase (MEK), which in turn phosphorylates and activates the ERKs, ERK1 and ERK2. Recently, it was demonstrated that activation of ERK in response to NGF in both primary sensory neurons and in PC12 cells also requires activation of the PI3-K pathway (11,12). PI3-K is required for internalization of TrkA and for its signaling to ERK via a Ras family member, Rap 1 (11,12). Rapid activation of ERK in response to NGF is mediated through Ras, whereas the sustained activation is Rap 1-dependent (11). Activation of ERK is thought to be required for PC12 responsiveness to NGF, because NGF-induced differentiation can be blocked by specific kinase inhibitors, by antibodies against Ras and MEK, or by expression of dominant negative mutants of Ras and MEK (13)(14)(15)(16). In addition, constitutively active forms of Ras, Raf-1, and MEK1 can induce neuronal differentiation of PC12 cells (15,(17)(18)(19). Activated ERKs translocate to the nucleus, where they phosphorylate and activate the existing transcription factor c-Jun, as well as induce expression of c-Jun (20). Interestingly, overexpression of activated c-Jun in PC12 cells induces neurite outgrowth, suggesting that c-Jun plays an important role in promoting PC12 neural differentiation (20).
Among the neuronal genes whose expression is up-regulated in response to NGF in PC12 cells are the genes encoding several subunits (␣3, ␣5, ␣7, ␤2, and ␤4) of neuronal nicotinic acetylcholine receptors (nAChRs) (21)(22)(23). nAChRs are pentameric ligand-gated ion channels important for synaptic transmission in the nervous system (24). Previously, we demonstrated that the NGF-mediated increase in nAChR gene expression is independent of cAMP-dependent protein kinase signaling (21) and that NGF treatment of PC12 cells increases the transcriptional activity of the ␤4 gene promoter (25). In this study we investigated the involvement of the other NGF-sig-* This work was supported in part by grants from the National Institutes of Health and the Smokeless Tobacco Research Council, Inc. (to P. D. G.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
‡ Supported by a POWRE grant from The National Science Foundation.
§ To whom correspondence should be addressed: Brudnick Neuropsychiatric Research Inst., Dept. of Psychiatry, University of Massachusetts Medical School, 55 Lake Ave. North, P.O. Box 2795, Worcester, MA 01613. Tel.: 508-856-4035; Fax: 508-856-4130; E-mail: paul.gardner@umassmed.edu. 1 The abbreviations used are: NGF, nerve growth factor; MAPK, ras-dependent mitogen-activated protein kinase; PI3-K, phosphoinositide 3-kinase; ERK, extracellular signal-regulated kinase; MEK, MAPK/ ERK kinase; ACh, acetylcholine; nAChR, neuronal nicotinic acetylcholine receptor; bp, base pair. naling pathways, namely MAPK, PI3-K, protein kinase C, and phospholipase C, in the regulation of nAChR gene expression. We show that in PC12 cells, activation of PI3-K and one of the components of the MAPK pathway, MEK, is important for up-regulation of ␤4 subunit gene promoter activity in response to NGF. We also show that NGF-inducible c-Jun can transactivate the promoter of the ␤4 gene through cooperation with Sp1 protein bound to the ␤4 promoter at a previously identified Sp1-binding site, a CA box. We further demonstrate that this Sp1-binding site is necessary for the responsiveness of the ␤4 promoter to NGF.

EXPERIMENTAL PROCEDURES
Cell Culture and Transfections-SN17 cells (26) and Drosophila melanogaster Schneider SL2 cells were maintained as described (27,28). PC12 cells (29) were cultured and differentiated with NGF (Upstate Biotechnology, Lake Placid, NY) as previously described (25). Twentyfour h prior to transfections, SN17 cells were plated at a density of 250,000 cells per 35-mm dish. Transfections were performed by a calcium phosphate method using a commercially available kit (Eppendorf-5 Prime, Inc., Boulder, CO). The wild type rat ␤4-luciferase expression plasmid, pX1B4FH, containing a 226-bp FokI/HindIII fragment spanning nucleotides Ϫ89 to ϩ137, relative to the ␤4 transcription initiation site, and the construct pX1B4FHmut4, containing a mutation in the CA box of the ␤4 promoter, were described previously (25,30). Cells were transfected with 2.5 g of test DNA (pX1B4FH or pX1B4FHmut4), 2.5 g of effector DNA (the empty pCMV5 vector (Invitrogen, Carlsbad, CA) or pCMV-jun constructs), and 2.5 g of a ␤-galactosidase expression vector, RSV-␤gal. In some cases, no effector DNA was included in the transfections. To ensure that the calcium phosphate/DNA precipitates had equal amounts of DNA, appropriate quantities of pBluescript II SK DNA (Stratagene, La Jolla, CA) were added to each sample. PC12 cells were transfected in 60-mm dishes at a density of 10 6 cells/ml using 30 l of LipofectAMINE (2 mg/ml; Life Technologies, Inc.) and 2.5 g of each DNA. After a 5-h incubation in the lipid/DNA mix, cells were washed twice with Dulbecco's modified Eagle's medium and fed with growth medium. For experiments involving the pharmacological kinase inhibitors, growth media were supplemented with or without 100 ng/ml NGF and with or without inhibitors, as indicated. All inhibitors were purchased from Calbiochem (San Diego, CA) and were used at the following concentrations: MEK inhibitor PD98059 at 100 M (31), PI3-K inhibitor LY294002 at 50 M (12, 32), phospholipase C inhibitor U-73122 at 1 M (33), and protein kinase C inhibitors calphostin C (33) and bisinodylmalemide (BIM) (34) at 400 nM and 1 M, respectively. SL2 cells were transfected as previously described (28), except that 100 ng of the effector DNAs (pActSp1 or pPacJun) were used. The pPacJun plasmid was a kind gift of Dr. J. Noti and is described elsewhere (35). Forty-eight h following transfection, cells were harvested and assayed for luciferase activity using a commercially available kit (Promega Corp., Madison, WI) and an Autolumat LB953 luminometer (EG&G Berthold, Gaithersburg, MD). All transfections were done a minimum of two times with two different preparations of plasmid DNAs. To correct for differences in transfection efficiencies between dishes, the luciferase activity in each sample was normalized to the ␤-galactosidase activity in the same sample, which was measured using a commercially available kit (Galacto-Light; Tropix, Inc., Bedford, MA).

PI3-K and MEK Activation
Are Necessary for the Induction of the ␤4 Promoter in Response to NGF-NGF-dependent upregulation of neuronal nAChR gene expression in PC12 cells is thought to occur partly at the level of transcription (25). To investigate which of the NGF-activated pathways is involved in the transcriptional regulation of neuronal nAChR gene expression in PC12 cells, we examined the effects of blocking several of these pathways on the transcriptional activity of the ␤4 gene promoter. We previously identified a 226-bp promoter fragment of the ␤4 gene that can confer neuron-specific expression to a reporter gene in transient transfection assays (27). We have also demonstrated that treatment of PC12 cells with NGF results in a significant increase in ␤4 promoter activity and that the 226-bp region is sufficient to mediate this response (25,27). We therefore investigated whether MEK, PI3-K, protein kinase C, or phospholipase C is involved in the regulation of transcription of the ␤4 gene through elements in the 5Ј regulatory region of the gene. The 226-bp ␤4 fragment fused upstream of the luciferase gene (pX1B4FH) was transiently transfected into PC12 cells. Following transfection cells were treated with NGF for 36 h. A parallel set of transfected cells was left untreated and served as a control. Consistent with our previous results, 36 h of NGF treatment stimulated ␤4 promoter activity and resulted in induction of reporter gene activity (Fig. 1). When NGF treatment of the transfected cells was carried out in the presence of PD98059, a selective pharmacological agent that blocks activation of MEK, or LY294002, a specific pharmacological inhibitor of PI3-K, stimulation of the ␤4 promoter was significantly reduced, as judged by reporter gene activity (Fig. 1). In contrast, inhibitors of phospholipase C, U-73122, and protein kinase C, calphostin C, and bisinodylmalemide, had little effect on the activity of the ␤4 promoter in transfected cells in response to NGF (Fig. 1). These results suggest that activation of the ␤4 promoter in response to NGF is mediated in part via the MAPK-and PI3-K-dependent pathways in PC12 cells. This is consistent with a model suggesting that these two pathways share overlapping functions in neuronal cells (12). In addition, these data indicate that the 226-bp fragment of the ␤4 promoter contains the necessary elements to mediate the response to ERK activation, a common downstream target of MEK and PI3-K (12).
c-Jun Transcativates the ␤4 Promoter in PC12 and SN17 Cells-It is well established that in PC12 cells, activation of the ERK cascade by NGF induces c-Jun expression and phosphorylation of its gene product (20). c-Jun is a transcription factor that has been implicated in regulation of gene expression in response to multiple extracellular stimuli (39). To address the question whether c-Jun is capable of regulating the activity of the ␤4 promoter, PC12 cells were transiently transfected with the 226-bp ␤4 promoter-luciferase construct, pX1B4FH, alone or in combination with an expression construct for c-Jun, pCMV-jun, in which the c-Jun gene is under the control of the CMV promoter, or with an empty parental vector, pCMV5. As shown in Fig. 2A (left side), transfections of the empty expression vector pCMV5 had no effect on reporter gene activity. However, when the ␤4-luciferase construct was cotransfected with c-Jun, a dramatic increase in reporter gene activity was observed ( Fig. 2A). Similar results were observed in the cho-linergic cell line SN17 (Fig. 2B). Together, these data indicate that, indeed, c-Jun is capable of transactivating the ␤4 promoter and suggest that c-Jun-responsive elements are present within this 226-bp fragment of the ␤4 promoter.
Jun proteins bind to the DNA sequence 5Ј-TGACTCA-3Ј, referred to as an AP-1 site, as homodimers or as heterodimers with Fos proteins (39). Interestingly, sequence analysis of the ␤4 promoter fragment used in these experiments indicates that there are no AP-1-binding sites within this region. It is therefore possible that c-Jun transactivates the ␤4 promoter through protein-protein interactions with a factor that can bind to ␤4 regulatory elements.
c-Jun and Sp1 Physically Interact in PC12 Cells-c-Jun has been shown to functionally interact with a number of transcription factors, including members of the Sp family, Sp1 and Sp3 (35, 40 -42). Moreover, c-Jun can directly interact with Sp1 in vitro (41). This is particularly relevant to the ␤4 promoter, because we previously reported that both Sp1 and Sp3 can efficiently transactivate this promoter (28,30,36). Therefore, it is possible that c-Jun activates the ␤4 promoter through protein-protein interactions with Sp1 and/or Sp3. To test this hypothesis, we first wanted to address the question of whether c-Jun and Sp1 are physically associated in PC12 cells. To that end, we performed immunoprecipitation experiments followed by Western blot analysis. Anti-Sp1 or anti-c-Jun antiserum was used to immunoprecipitate proteins from nuclear extracts prepared from NGF-treated PC12 cells. The precipitated proteins were separated via SDS-polyacrylamide gel electrophoresis and blotted onto a nitrocellulose membrane that was subsequently incubated with anti-Sp1 or anti-c-Jun antiserum. As shown in Fig. 3, preimmune serum did not precipitate either Sp1 or c-Jun in this experiment. However, complexes containing Sp1 and c-Jun were detected by immunoprecipitations using antibodies directed toward one or the other protein (Fig. 3). It should be noted that the levels of Sp1 found in c-Jun immunoprecipitates were somewhat lower when compared with those precipitated with anti-Sp1 antibody (Fig. 3), suggesting that not all cellular Sp1 is engaged in a complex with c-Jun. The same was true with c-Jun as well (Fig. 3). These data provide compelling evidence for direct physical interactions between Sp1 and c-Jun in PC12 cells and support the idea that c-Jun may regulate the ␤4 promoter through its association with Sp1.
c-Jun Transcativates the ␤4 Promoter through an Sp1-binding Site-Previously, we showed that Sp1 can specifically bind to a CA box in the 226-bp region of the ␤4 promoter (30,36) and that this interaction is critical for Sp-mediated transactivation of the ␤4 promoter (36). If c-Jun activates the ␤4 promoter through its association with Sp1, then the Sp1-binding site within this promoter, the CA box, should be important for the ability of c-Jun to regulate the ␤4 promoter. To test this hypothesis, PC12 cells were cotransfected with a wild type ␤4 promoter-luciferase expression construct, pX1B4FH, or with a construct in which the CA box is mutated at three nucleotide positions, pX1B4FHmut4, and with an expression construct for

FIG. 3. Immunoprecipitations of c-Jun-Sp1 complexes from NGF-treated PC12 cell lysates.
Anti-c-Jun, anti-Sp1, or preimmune rabbit serum (PIRS) was used to immunoprecipitate (IP) proteins from a PC12 cell extract. Western analysis of the immunoprecipitated material was carried out using anti-Sp1 or anti-c-Jun antiserum as indicated.
c-Jun, pCMV-jun. As mentioned earlier, c-Jun was able to strongly transactivate the wild type ␤4 promoter, pX1B4FH ( Fig. 2A). However, when the ␤4 promoter-luciferase reporter containing mutations in the CA box, pX1B4FHmut4, was used in this experiment, c-Jun activation of the promoter was marginal ( Fig. 2A). Similar results were obtained from transfections performed in SN17 cells (Fig. 2B). Thus, the CA box appears to be critical for c-Jun-mediated transactivation of the ␤4 promoter.
c-Jun and Sp1 Synergistically Activate the ␤4 Promoter in Drosophila SL2 Cells-To further investigate the importance of c-Jun-Sp1 interactions for the transactivation of the ␤4 promoter, we performed transfection experiments using the Drosophila SL2 cell line. These cells lack endogenous Sp1 activity (43)(44)(45)(46), as well as endogenous c-Jun activity (35), making them a useful system to study the functional consequences of the protein-protein interactions mentioned above. We previously reported that Sp1 can strongly transactivate the ␤4 promoter when transfected into SL2 cells (28,30,36). To determine whether c-Jun activates the ␤4 promoter through its interactions with Sp1, SL2 cells were cotransfected with a wild type ␤4 promoter-luciferase expression construct (pX1B4FH) and pActSp1 or pPacJun or both. To confirm that Sp1 and c-Jun expression was involved in transactivation, the reporter DNA was also transfected with the parental vectors devoid of Sp1 and c-Jun coding sequences, pAct and pPacO, alone. As expected, Sp1 was capable of transactivating the wild type ␤4 promoter-luciferase construct ϳ20-fold (Fig. 4). Consistent with the observation that the ␤4 promoter fragment used in these experiments does not contain an AP-1-binding site, c-Jun by itself did not have any effect on reporter gene activity (Fig.  4). However, when Sp1 and c-Jun were cotransfected with the ␤4 promoter-luciferase construct, an ϳ120-fold increase in reporter gene activity was observed, indicating a synergistic effect of the two regulatory factors (Fig. 4). These results are consistent with the hypothesis that c-Jun activates the ␤4 promoter through direct interactions with Sp1.
To investigate the importance of the Sp1-binding site for the ability of Sp1 and c-Jun to synergistically activate the ␤4 promoter, similar transfection experiments were performed using a ␤4 promoter-luciferase construct in which the CA box is mutated at three nucleotide positions (pX1B4FHmut4). Muta-tion in this site reduces the ability of Sp1 to transactivate the ␤4 promoter in SL2 cells (Fig. 4) (36). The residual activation seen with the mutated promoter is most likely a result of Sp1 binding to downstream low affinity Sp-binding sites (30,36). As in the case of the wild type ␤4 promoter, c-Jun was unable to transactivate the mutated promoter (Fig. 4). Synergistic transactivation of the wild type ␤4 promoter by Sp1 and c-Jun was dramatically reduced when the mutated promoter was used. The observed reporter gene activity was comparable with that of Sp1 alone on the wild type promoter (Fig. 4). Together, these results indicate that Sp1 and c-Jun can cooperate to activate the ␤4 promoter in Drosophila cells and that an intact Sp1binding site is important for this cooperativity.
The CA Box Is Important for the NGF Responsiveness of the ␤4 Promoter in PC12 Cells-Collectively, the data presented above suggest that the CA box of the ␤4 promoter might be a target for NGF signaling in PC12 cells. To address this question, we transfected PC12 cells with a wild type ␤4 promoterluciferase expression construct, pX1B4FH, or with the construct in which the CA box is mutated, pX1B4FHmut4. Consistent with our previous findings (30), the basal activity of the mutant promoter was significantly lower than that of the wild type in untreated PC12 cells (Fig. 5). Thirty-six h of NGF treatment of the transfected cells resulted in a dramatic increase in the activity of the wild type ␤4 promoter, as judged by reporter gene activity (Fig. 5). However, activity of the mutated promoter was increased only slightly, resulting in reporter gene activity similar to that produced by the wild type ␤4 promoter in unstimulated PC12 cells (Fig. 5). These data suggest that the CA box is necessary, but probably not sufficient, to mediate the response of the ␤4 promoter to NGF in PC12 cells.

DISCUSSION
To date, twelve genes (␣2-␣10, ␤2-␤4) encoding subunits of neuronal nAChRs have been identified. Assembly of the different subunits into pentameric homo-or heteromeric receptors results in formation of ion channels with distinct physiological and pharmacological profiles (24). This diversity is thought to be a consequence of differential expression of the subunit genes and subsequent incorporation of their respective products into mature receptors (47). Whereas the implications of the func- tional diversity of nAChR are beginning to be understood, relatively little is known regarding regulation of their gene expression in general and in response to neurotrophin signaling in particular. NGF-driven differentiation of PC12 cells has been extensively used as a model system to study the physiological, biochemical, and molecular mechanisms of neurotrophin signaling. In regard to neuronal nAChR, it has been demonstrated that in PC12 cells, NGF increases ACh-induced channel activity and ACh-induced macroscopic current density (4,21,48), phenomena that are thought to be a consequence of an increase in the number of functional nAChR (21). This hypothesis is consistent with observed increases in the binding of [ 3 H]nicotine (49) and of anti-nAChR antibodies (50) that occur in PC12 cells in response to NGF. Whereas the exact mechanisms of the increase in the number of nAChRs are unknown, NGF-driven induction of nAChR gene expression is clearly important (21). Previously, we showed that the NGFinduced up-regulation of nAChR expression and increases in ACh-induced current density are independent of cAMPdependent protein kinase activity (21). In this study, using specific chemical inhibitors, we demonstrate that the MAPK and PI3-K signal transduction cascades, but not the phospholipase C or protein kinase C pathways, are important for NGFmediated induction of the promoter of the ␤4 subunit gene in PC12 cells. Significantly, positive stimulation of ␤4 promoter activity in response to NGF was considerably reduced by the addition of a pharmacological inhibitor of MEK, PD98059, and was reduced to a similar extent by addition of a PI3-K inhibitor, LY294002. These data suggest that activation of the MAPK and PI3-K pathways is clearly necessary for the induction of nAChR gene expression. Moreover, they indicate that MAPK and PI3-K pathways might have common downstream targets in PC12 cells. Originally, PI3-K and MAPK were thought to initiate two distinct signal transduction cascades in neuronal cells. However, in a recent report York et al. (12) convincingly demonstrated that inhibitors of PI3-K block well known targets of the MAPK pathway, namely ERK and B-Raf, therefore indicating that the MAPK and PI3-K pathways are indeed interconnected.
The MAPK pathway, through a series of phosphorylation events (that include phosphorylation of ERK), is thought to relay signals received by NGF receptors in the plasma membrane to specific transcription factors in the nucleus (51). These activated transcription factors are required for expression of the delayed early and late genes, some of which, such as neuronal nAChR, contribute to the neuronal phenotype developed by PC12 cells in response to NGF. c-Jun is believed to be one of the key factors activated by the MAPK signaling cascade (39). c-Jun is an inducible transcription factor that is activated in response to multiple extracellular stimuli, including growth factors, cytokines, neurotransmitters, T cell activators, various forms of stress, and UV radiation (39). Regulation of c-Jun activity occurs at the level of transcription, resulting in an increase in c-Jun mRNA, as well as post-transcriptionally, resulting in an enhancement of its transactivation potential through phosphorylation (39). Leppa et al. (20) recently demonstrated that in PC12 cells, activation of the MAPK pathway, more specifically ERK, results in stimulation of both c-Jun synthesis and phosphorylation. Moreover, they showed that an exogenous activated form of c-Jun induced marked neurite outgrowth and that overexpression of wild type c-Jun potentiated differentiation induced by MEK1, whereas dominant-negative mutants of c-Jun inhibited it (20). Thus, proper c-Jun activation appears to be necessary for neuronal differentiation of PC12 cells. We therefore investigated whether c-Jun is capable of regulating the promoter of the ␤4 subunit gene. Using the same fragment of the ␤4 promoter that confers neuronspecific expression to a reporter gene (24) and that is sufficient to mediate responses to NGF in PC12 cells (25), we showed that c-Jun can strongly transactivate the ␤4 promoter in both PC12 and SN17 cells. Surprisingly, this 226-bp region of the ␤4 promoter does not contain "classic" Jun-binding sites, AP-1 elements, to which c-Jun normally binds as a dimer with Jun-Fos proteins (52). Thus, we hypothesized that c-Jun transactivates the ␤4 promoter through protein-protein interactions with a nuclear factor bound to ␤4 regulatory elements. c-Jun has been shown to directly interact with a number of transcription factors, for example, members of the cAMP responsive element-binding protein family, glucocorticoid receptor (53), Smad proteins (54,55), and members of the Sp family, Sp1 and Sp3 (35, 40 -42). Interestingly, we have previously demonstrated that both Sp1 and Sp3 can strongly transactivate the ␤4 promoter (28,30,36). Sp1 and Sp3 are zinc finger DNA-binding proteins that bind a CA box element in the 226-bp region of the ␤4 promoter (30,36). Originally, Sp family members were thought to be ubiquitous factors contributing to core promoter activities (56). However, more recently, they have been shown to participate in regulated and cell typespecific gene expression, including regulation of transcription of a number of neuron-specific genes (30,36,(57)(58)(59)(60)(61)(62). Moreover, in PC12 cells, NGF treatment has been shown to stimulate Sp1-dependent transcription (63). Although the exact mechanism of this stimulation is unknown, it is thought that posttranslational modifications and/or association with other factors may play a role (63). Hence, NGF-induced activation of ␤4 gene expression may occur through functional interactions between c-Jun and Sp1. Consistent with this idea are results of our immunoprecipitation/Western blot analysis demonstrating that c-Jun and Sp1 can physically associate in NGF-treated PC12 cells. Furthermore, the Sp1-binding site in the 226-bp region of the ␤4 promoter, a CA box, is important for the ability of c-Jun to transactivate this promoter in both PC12 and SN17 cells. In contrast to PC12 and SN17 cells, c-Jun by itself had no effect on reporter gene activity in Drosophila SL2 cells that were devoid of endogenous Sp proteins. c-Jun was able to transactivate the ␤4 promoter in SL2 cells only when cotransfected with Sp1, providing additional support for the hypothesis that it does so through direct interactions with the latter protein. Interestingly, transactivation by c-Jun and Sp1 occurred in a synergistic manner and did so in a CA box-dependent manner. Together, these data suggest that Sp1 and c-Jun can cooperate to activate the ␤4 promoter and that an intact Sp1-binding site is important for this cooperativity. Moreover, the CA box of the ␤4 promoter appears to be critical for NGF signaling, because the responsiveness of the ␤4 promoter to NGF is dramatically reduced when the CA box is mutated. Previous studies demonstrated that c-Jun and Sp1 functionally interact on both synthetic and natural promoters (40,41); however, to our knowledge, this study is the first demonstration of cooperativity of these proteins in neuronal cells. Whether, similar to Sp1, Sp3 can functionally interact with c-Jun to regulate nAChR gene expression remains to be elucidated.
It should be noted that the ␤4 subunit, together with the ␣3 and ␣5 subunits, forms the predominant neuronal nAChR subtype expressed in the peripheral nervous system (64,65). These three genes are tightly clustered in mammalian genomes, raising the possibility that they are coordinately expressed via a common regulatory mechanism. Indeed, it has been demonstrated that Sp1 is important for the regulation of expression of all three genes (30, 36, 61, 62). Therefore, it is possible that through interactions with Sp1, c-Jun might regulate the promoters of the ␣3 and ␣5 subunit genes as well.