Muscarinic Acetylcholine Receptors Activate Expression of the Egr Gene Family of Transcription Factors*

In order to search for genes that are activated by muscarinic acetylcholine receptors (mAChRs), we used an mRNA differential display approach in HEK293 cells expressing m1AChR. The zinc-finger transcription factor genes Egr-1,Egr-2, and Egr-3 were identified. Northern blot analyses confirmed that mRNA levels of Egr-1, Egr-2, and Egr-3 increased readily after m1AChR stimulation and that a maximum was attained within 50 min. At that time, Egr-4 mRNA was also detectable. Western blots and electromobility shift assays demonstrated synthesis of EGR-1 and EGR-3, as well as binding to DNA recognition sites in response to m1AChR activation. Activation of m1AChR increased transcription from EGR-dependent promoters, including the acetylcholinesterase gene promoter. Activity-dependent regulation of Egr-1 mRNA expression and EGR-1 protein synthesis was also observed in cells expressing m2, m3, or m4AChR subtypes. Increased EGR-1 synthesis was mimicked by phorbol myristate acetate, but not by forskolin, and receptor-stimulated EGR-1 synthesis was partially inhibited by phorbol myristate acetate down-regulation. Together, our results demonstrate that muscarinic receptor signaling activates the EGR transcription factor family and that PKC may be involved in intracellular signaling. The data suggest that transcription of EGR-dependent target genes, including the AChE gene, can be under the control of extracellular and intracellular signals coupled to muscarinic receptors.

Muscarinic acetylcholine receptors (mAChRs) 1 are members of a superfamily of G protein-coupled cell surface receptors with seven transmembrane domain topology. Five subtypes of mAChRs (m1-m5) can be divided according to their signaling mechanisms: m1, m3, and m5 AChRs preferentially couple to the pertussis toxin-insensitive Gq/G11 proteins that stimulate phosphoinositide hydrolysis, whereas m2 and m4 subtypes couple to the G i /G o proteins that predominantly inhibit adenylyl cyclase (1,2). In brain, mAChRs are involved in such functions as attention, learning, memory, and cognition (3,4). m1 and m3 AChR subtypes are localized to the somatodendritic cell surfaces of large pyramidal neurons throughout the cortex and the hippocampus, as well as on small cholinergic interneurons in the striatum. In contrast, m2 and m4 AChRs are predominantly present on axons of the large basal forebrain projection neurons that innervate cholinergic target cells throughout the cortex and the hippocampus. Activation of the postsynaptic m1/ m3/m5 AChR family by acetylcholine triggers a large variety of distinct signaling cascades including phospholipase D, adenylyl cyclase, phospholipase A2, the generation of diacylglycerol which activates protein kinase C (PKC) and couples mAChRs to the ERK-MAP-kinase signaling cascade, activation of endoplasmic reticulum IP3 receptors, stimulation of ligand-operated cell-surface Ca 2ϩ channels, and the activity of voltagegated potassium channels (5)(6)(7)(8)(9)(10)(11)(12). Cellular responses of mAChRs include the activation of neurite outgrowth, the finetuning of membrane potentials, and the regulation of mitogenic growth responses in cells that are not terminally differentiated (13). mAChRs are also involved in the activity-dependent regulation of posttranslational processing of the ␤-amyloid precursor protein of Alzheimer's disease (14 -16) by an unidentified protease associated with reduced generation of ␤-amyloid peptides (17)(18)(19), the principal component of amyloid plaques in Alzheimer's disease brain (20). mAChRs can activate the transcription of the immediateearly genes jun and fos (21,22). In order to identify additional genes that are regulated by mAChRs, we used a differential mRNA display screen of cDNA populations derived from m1AChR expressing cells, with or without receptor stimulation. We identified several differentially expressed clones, including the Egr family of transcription factors. This gene family encodes proteins with a zinc-finger containing DNA binding domain, and includes Egr-1 (23) (also named Krox-24 (24), zif/268 (25), NGFI-A (26), or TIS8 (27)), Egr-2 (28) (also named Krox-20 (29)), Egr-3 (30), and Egr-4 (also named NGFI-C (31), or pAT 133 (32)), as well as the related Wilms' tumor suppressor gene WT1 (33,34). EGR proteins bind DNA in a sequence-specific manner, recognizing the consensus element GCG(G/T)GGGCG (35,36), and they can either activate or suppress the transcription of associated genes, depending on the respective promoter (37)(38)(39)(40)(41)(42). We show here that the entire family of EGR transcription factors can be under the control of muscarinic acetylcholine receptors and their related signaling mechanisms.

Cell Culture
Human HEK293 cells stably transfected with the muscarinic acetylcholine receptor subtypes m1, m2, m3 or m4 were grown in DMEM/F-12 medium supplemented with 10% fetal calf serum and 500 g/ml geneticin (G418, Life Technologies, Inc.). The cells were maintained in a 5% CO 2 atmosphere at 37°C. For experiments, the cells were subcultured * This study was funded by Bundesministerium fü r Bildung und Forschung, Deutsche Forschungsgemeinschaft, and Alzheimer Forschung International. 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.

Differential mRNA Display
Total RNA was prepared from the cells by using the RNeasy kit (Qiagen) according to the manufacturer's instructions. The RNA preparations were subjected to DNaseI digestion (Boehringer Mannheim), in the presence of the RNase inhibitor RNasin (Promega) for 30 min. RNA was extracted with phenol and precipitated in ethanol. Equal amounts of 0.2 g of RNA each were transcribed to cDNA with Expand TM Reverse Transcriptase (Boehringer Mannheim) by using onebase-anchor primers HT 11 A, HT 11 G and HT 11 C according to Liang et al. (43). The obtained cDNAs were subjected to polymerase chain reaction (PCR) employing the corresponding one-base-anchor oligonucleotides along with HAP-1 (TGCCGAAGCTTGATTGCC), HAP-3C (TGC-CGAAGCTTTGGTCAC), or HAP-3T (TGCCGAAGCTTTGGTCAT) as random primers with a HindIII restriction site by using PCR conditions described by Zhao et al. (44). PCR was performed with AmpliTaq polymerase (Perkin-Elmer Corp.) in the presence of [␣-35 S]dATP (2000 Ci/mmol, NEN Life Science Products) along with dCTP, dTTP, and dGTP (Amersham Pharmacia Biotech). The PCR products were separated on 6% polyacrylamide urea sequencing gels and dried on Whatman 3MM paper, and x-ray films (DuPont) were exposed for 8 -16 h. Differential bands were excised, and the contained cDNA was eluted in water, precipitated with glycogen in ethanol, and dialyzed against 10% glycerol. The obtained preparations were reamplified by 40 high-stringency cycles in 50-l PCR mixtures containing the corresponding composite primer pairs as described by Zhao et al. (44). Reamplified cDNAs were purified by agarose gel electrophoreses, gel extraction by using the QIAEX II kit (Qiagen, Hilden, Germany), and were ligated into pBluescript IIKSϩ (Stratagene) by using the HindIII restriction sites. Cloned cDNA fragments were sequenced on an automated ABI377 DNA sequencer (Perkin-Elmer Corp.) by using T3-and T7-derived sequencing primers.
Northern Blot Analysis 5-20 g of RNA each were separated by agarose formaldehyde gel electrophoresis and blotted onto nylon membranes (Hybond TM -Nϩ, Amersham Pharmacia Biotech) as described (45). Membrane filters were probed with ␣[ 32 P]dCTP random primed cDNA probes prepared from Egr-1, Egr-2, Egr-3, and Egr-4 cDNA clones or from PCR products obtained in the differential display. Hybridized filters were washed under high stringency conditions and were subjected to autoradiography. Equal RNA loading was confirmed by probing the identical blots with a glyceraldehyde 3-phosphate dehydrogenase (GAPDH) probe.

Preparation of Nuclear Extracts
Cells were washed twice with PBS, scraped from the plates in hypotonic buffer, swollen on ice, and lysed with 1% Nonidet P-40. Nuclei were pelleted and extracted in 200 l of ice-cold 10 mM HEPES (pH 7.9), 400 mM NaCl, 0.1 mM EDTA, 0.1 mM EGTA, 2 mM dithiothreitol, 1% Nonidet P-40 in the presence of pepstatin and aprotinin. Protein concentrations of nuclear extracts were determined by the Bradford protein assay (Bio-Rad).

Western Blot Analysis
50 g of nuclear protein were separated by SDS-PAGE, transferred to nitrocellulose filters, blocked with 5% nonfat dry milk in PBST, and probed with polyclonal antisera against either EGR-1, EGR-2, or EGR-3 (Santa Cruz Biotechnology) at dilutions of 1:3000. Primary antisera were visualized with horseradish peroxidase conjugated to protein A and by enhanced chemiluminescence (Amersham Pharmacia Biotech) using x-ray films.

Electrophoretic Mobility Shift Assay
A radiolabeled double-stranded oligonucleotide containing two copies of the consensus binding motif for EGR proteins (5Ј-GATCCAGC-GGGGGCGAGCGGGGGCGAGATC-3Ј) was prepared with Klenow polymerase and by using ␣[ 32 P]dCTP. 0.5 ng of the 32 P-labeled DNA was incubated with 10 g of nuclear protein for 20 min at room temperature in a 20-l reaction mixture containing 10 mM HEPES (pH 8.0), 5% glycerol, 100 mM NaCl, 5 mM MgCl 2 , 2 mM dithiothreitol, 0.1 mM EDTA, 10 M ZnCl, and 1 g of poly(dI-dC) as nonspecific competitor. In some cases, protein extracts were preincubated with polyclonal antisera against EGR proteins (Santa Cruz Biotechnology). Protein-DNA complexes were separated from free DNA on 4% polyacrylamide gels using 0.5ϫ TBE (Tris-borate-EDTA) as running buffer.

Luciferase Assays
HEK293 cells stably expressing m1AChR, grown on poly-D-lysine, were transfected at 30 -40% density with 4 g of expression constructs precipitated by calcium phosphate. 24 -48 h after transfection, cells were serum deprived and then treated with or without test substances; 19 h following stimulation, luciferase activity was quantitated with a luminometer (Lumat LB 9501, Berthold) (46).
EGR Assay-Expression constructs pTATA-luc/EGR and pTATA-luc/ EGRmut encode an EGR responsive promoter fused to a luciferase reporter gene. These plasmids were constructed by inserting one copy of either the EGR binding site 5Ј-GATCCAGCGGGGGCGAGCGGGGGC-GACTAG-3Ј or a mutant EGR binding site 5Ј-GATCCAGCTAGGGC-GAGCTAGGGCGACTAG-3Ј into pTATA-luc that includes the luciferase gene under control of the ␤-globin minimal promoter. pCMV/Krox24 together with either pTATA-luc/EGR or pTATA-luc/EGRmut was used as a positive control for the transfection.
AChE Assay-Expression construct pAChE-luc (construct A) (37) consists of approximately 1.1 kb of DNA sequence upstream of the region similar to the mouse AChE gene exon 1 in addition to exon 1, and 1.5 kb of the putative intronic region between sequences corresponding to exons 1 and 2 in the AChE gene. The 3Ј end includes the first 16 bases of exon 2 upstream of the ATG start codon.

RESULTS
m1AChR Stimulated Transcription of Egr-1, Egr-2, Egr-3, and Egr-4 -Cloning and sequence analysis of several clones derived from differential bands generated by the reverse transcription primer HT 11 G along with the PCR primer HAP1, and by the reverse transcription primer HT 11 A, along with the PCR primers HAP-3C and HAP-3T, revealed the zinc-finger domain transcription factor genes Egr-1 (48), Egr-2 (28), and Egr-3 (30). By using a probe generated from the reamplified differential band corresponding to Egr-1, we identified on Northern blots a message at 3.5 kb that was differentially up-regulated within 10 -60 min of m1AChR stimulation. This message was undetectable in RNA preparations obtained from unstimulated cells. To verify that this message corresponded to the zincfinger transcription factor Egr-1, Northern blot analyses of identical RNA preparations were done with a probe generated from a cDNA fragment of Egr-1. Again, we detected a major transcript at 3.5 kb (Fig. 1A) in the stimulated condition but not in the unstimulated control condition. Within 10 min of stimulation, a weak signal was detectable, and a maximum was attained after 50 min (Fig. 1B). The time course of m1induced Egr-1 expression was biphasic, and a second peak was observed after 240 min and after a minimum was reached 100 min after receptor stimulation.
Northern blot analyses with probes generated from Egr-2 or Egr-3 cDNAs verified that m1AChR stimulation activated transcription of the Egr-2 and Egr-3 genes as suggested by the differential display (Fig. 2, A and B). Expression levels of Egr-1 were the highest, and these of Egr-3 were one order of magnitude higher than those of Egr-2. In addition, a weak signal with an Egr-4-specific probe (32) just above the detection limit showed that Egr-4, also, was up-regulated by m1AChR (Fig.  2C). Together, these results demonstrate that all four members of the Egr transcription factor gene family are under the control of the m1AChR, with different induction levels.
Muscarinic Receptor Stimulation Increases Cellular Levels of Functional EGR-1 Protein-Stimulation of m1AChR with car-bachol readily increased the synthesis of EGR-1 protein within 120 min, and protein levels remained high until 180 min after stimulation (Fig. 3A). Receptor-induced EGR-1 synthesis was completely blocked by atropine, indicating a specific and receptor-mediated action of carbachol. EGR-1 synthesis was also stimulated with PMA, a potent activator of PKC, suggesting that PKC may couple m1AChR to EGR-1 synthesis (Fig. 3A). Down-regulation of PKC by 16 h of PMA pretreatment, however, was associated with only a partial block of the receptormediated induction, suggesting additional signaling mechanisms that are not responsive to PMA in the coupling of m1AChR to EGR-1 synthesis in HEK293 cells. To exclude cAMP as a signaling molecule for this response, we demonstrated that forskolin failed to influence EGR-1 synthesis (Fig.  3A). In agreement with the absence of Egr-1 message, no EGR-1 protein was detectable on Western blots of nuclear proteins extracted from unstimulated cells (Fig. 3A).
EGR-3 protein was also detectable on Western blots of m1AChR cells 120 min after stimulation with carbachol or with PMA (Fig. 3B). No EGR-3 protein was detectable in unstimulated cells or in cells treated with atropine/carbachol or 8-bromo-cAMP. Levels of EGR-3 protein were lower than those of EGR-1 with possible differences in affinity were even considering that different antibodies used for the detection of EGR-1 and EGR-3. EGR-2 protein was undetectable on Western blots using two different EGR-2 antibodies (data not shown).
In order to establish that m1AChR activation leads to the binding of EGR-1 protein to its specific DNA recognition site, we used electrophoretic mobility shift assays (Fig. 4A). We found that a nuclear protein from both carbachol-and PMAtreated cells, but not from control cells, formed a specific complex with an EGR recognition site and with the expected mobility of EGR-1. Carbachol-induced binding of this nuclear protein was completely blocked by atropine. It was partially abolished and supershifted by anti-EGR-1 antibodies, confirming specificity for the binding of EGR-1 to its DNA recognition sequence (Fig. 4B). Down-regulation of PKC by 16 h of pretreatment with PMA resulted only in a partial inhibition of the mobility shift, suggesting redundant signaling pathways in coupling the response to the surface receptors. Again, elevating levels of cellular cAMP by forskolin failed to induce the mobil-ity shift (Fig. 4A).
Stimulation of the AChR subtypes m2 and m3 clearly induced a mobility shift with the EGR recognition site, and this interaction was also supershifted by EGR-1-specific antibodies (Fig. 5A). Stimulation of the AChR subtype m4 induced a mobility shift with a considerably lower magnitude as compared with the other mAChR subtypes (Fig. 5A). These responses were also receptor-mediated, as evidenced by the complete atropine block in all cell lines. In the nontransfected parent HEK293 control cells, carbachol was ineffective in inducing a mobility shift (Fig. 5A). Western blot analyses of nuclear proteins confirmed that stimulation of AChR subtype m2, m3, or m4 increased EGR-1 protein synthesis in an atropine-sensitive manner (Fig. 5B). Again, carbachol was ineffective in nontransfected parent HEK293 control cells.
m1AChR Activate Transcription via an EGR-Dependent Promoter-Transfection experiments with luciferase reporter constructs fused to an EGR-dependent minimal promoter demonstrated that both m1AChR and PMA effectively induced transcription from this promoter, suggesting downstream activation of genes that contain EGR responsive promoters (Fig.  6A). Again, the m1-induced response was blocked by atropine, and it was completely absent when control constructs were assayed that contain a mutated EGR binding domain (Fig. 6B). 8-Bromo-cAMP failed to increase luciferase activity from the EGR-dependent promoter, confirming that cAMP is ineffective in its regulation.
Transcriptional Activity of the AChE Gene Promoter Is Increased by m1AChR Stimulation-Stimulation of m1AChR in cells transfected with the AChE gene promoter (37,49) fused to a luciferase reporter gene resulted in an activation of the transcription via the AChE gene promoter, as indicated by increased luciferase activity (Fig. 7A). Atropine blocked the increase in luciferase activity, indicating that this effect was caused specifically by m1AChR activation. Again, PMA mimicked the m1AChR activation of the AChE gene promoter, whereas 8-bromo-cAMP failed to induce transcription. Cotransfection of the pAChE-luc construct together with Krox24 demonstrated that EGR effectively induced the AChE gene promoter in our experimental system (Fig. 7B).

DISCUSSION
In brain, mAChRs are involved in long-term potentiation, synaptic plasticity, and higher cognitive functions, including learning and memory (50 -52). Such plastic alterations in neuronal structure and function are associated with rapid and transient transcription of activity-dependent genes (4,53,54), such as the immediate-early genes c-fos, jun-B, Egr-1, and Egr-2 (21,49,55,56). The results of our study show that mAChRs can increase the expression of the zinc-finger domain transcription factor genes Egr-1, Egr-2, Egr-3, and Egr-4, albeit to different extents.
Egr-1 message was detectable within 10 min of m1AChR stimulation, a maximum was reached within 50 min, and EGR-1 protein was present within 120 and 180 min of stimulation. m1AChR-induced EGR-1 protein bound to an EGRspecific DNA recognition domain, as evidenced by mobility shift assays. In addition, luciferase reporter assays showed that m1AChR induced transcription from EGR dependent promoters. The carbachol-induced effects on transcriptional regulation were m1AChR-specific, as evidenced by complete atropine blocks, and they were mimicked by direct activation of PKC with PMA. Down-regulation of PKC, however, was associated with only partial inhibitions of the receptor responses, indicating that activation of PMA-dependent forms of PKC can be sufficient, but is not necessary, for coupling m1AChR to transcriptional activation of the Egr-1 gene. After PKC inhibition in PC12 cells, the remaining induction of Egr-1 by mAChR could be blocked by chelation of extracellular calcium with EGTA (57), suggesting the involvement of calcium influx as an additional signaling mechanism. To exclude cAMP in the coupling of m1AChR to Egr-1 transcription, we used the potent adenylyl cyclase stimulators forskolin and 8-bromo-cAMP, and we found that they failed to increase Egr-1 expression.
Our data suggest Egr-1 as a major target among members of the Egr gene family of m1 receptor, because competition experiments with EGR-1-specific antibodies almost completely blocked the binding of nuclear extracts to the EGR recognition sequence that is known to interact with all members of the Egr family. Higher amounts of EGR-1 protein, corresponding to the much higher expression level than that found in Egr-2 and Egr-3, may account for this result. This interpretation is underscored by the failure of EGR-2-and EGR-3-specific antibod- FIG. 2. m1AChR increase levels of Egr-2, Egr-3, and Egr-4 mRNA. Northern blot analysis of total RNA from unstimulated (control) and carbachol stimulated (min of stimulation) m1 HEK293 cells. mRNA levels of Egr-2 (A) were lower than those of Egr-3 (B), but both Egr-2 and Egr-3 were readily activated by m1AChR stimulation with carbachol. mRNA levels of Egr-4 (C) were just above the detection limit. Exposure times of the blots were 11 days (Egr-2), 7 days (Egr-3), and 4 weeks (Egr-4). Northern blot analysis of Egr-3 (B) was done with the same RNA blot used for the analysis of Egr-1 (Fig. 1).

FIG. 3. Stimulated m1AChR increase cellular concentration of EGR-1 and EGR-3 protein.
Western blots of nuclear extracts from m1 HEK293 cells were done with EGR-1-specific (A) and EGR-3-specific (B) antibodies. The carbachol-induced increase of the EGR proteins was completely blocked by atropine, and they were mimicked by direct activation of PKC with PMA. Increasing cellular cAMP levels by forskolin or 8-bromo-cAMP failed to elevate cellular levels of EGR-1 or EGR-3, and down-regulation of PKC prior to m1AChR stimulation partially blocked the responses. The bands at 60 and 96 kDa in A and at 68 and 60 kDa in B are known nonspecific signals. Exposure time of the Egr-1 blot was 2 min; exposure time of the GAPDH blot was 25 min.
ies to induce a detectable supershift with the above preparation. Nevertheless, EGR-3 protein was detected unequivocally on Western blots, indicating that the transcription of the Egr-3 gene led to the generation of translation products.
These results confirm and extend the observation that muscarinic AChRs increase Egr-1 message in PC12 (42,57,58), NG108 -15 (59), and 3T3 cells transfected with m1 AChR cells (22). Our data show that in addition to Egr-1, the expression of Egr-2, Egr-3, and Egr-4 is under the control of muscarinic AChRs. Moreover, our results show binding to, and activation of, EGR promoter sequences followed by the synthesis of functional protein as a result of mAChR stimulation.
The muscarinic AChR subtypes used in our assays induced EGR-1 binding to EGR recognition domain with different efficacy: whereas m1, m2, and m3 AChRs caused a very clear induction of EGR-1 binding, m4AChR caused only subtle activation of EGR-1 binding in our cells. Stimulation of all analyzed mAChR receptor subtypes, including m4, increased EGR-1 protein synthesis, as verified by Western blot analyses of the same nuclear proteins used for the EGR binding assays. These results demonstrate that transcribed Egr-1 gene induced by mAChRs is followed by the synthesis of functional EGR-1 proteins. m2 and m4 AChRs efficiently inhibit adenylyl cyclase activity but activate phosphoinositide hydrolysis with only onefifth of the efficacy of m1 and m3 AChRs (60). These levels of phosphoinositide turnover induced by m2 and m4 AChRs may be sufficient to mediate Egr-1 expression. Alternatively, other second messenger pathways, including the inhibition of adenylyl cyclase, may be involved in signaling.
The ability of different muscarinic AChR subtypes to stimu-late Egr-1 expression suggests that similar genes are controlled by acetylcholine in both pre-and postsynaptic neuronal populations. It is possible that these use distinct signaling mechanisms in the coupling of the surface receptors to Egr expression.
In vivo experiments with intact neuroanatomical structures, however, are required to test this hypothesis. Egr-1 expression is up-regulated also by nicotinic acetylcholine receptors in a mammalian skeletal muscle cell line (61), indicating that EGR transcription factors can be under the control of several distinct muscarinic and nicotinic surface receptor subtypes. Together, our data suggest that target genes with EGRresponsible promoters can be regulated by m1AChR. Candidate genes for EGR-mediated regulation include the AChE gene (37), the platelet-derived growth factor A and B chains (38,39), the transforming growth factor ␤1 (40), luteinizing hormone (41), phenylethanolamine N-methyltransferase (42), and human interleukin 2 (62). Thus, the regulation of multiple cellular functions, including signaling, growth, and metabolism, may be coupled to muscarinic AChR activity via EGR transcription factors.
EGR-1 increases the promoter activity of the AChE gene, a serine hydrolase that catalyzes the hydrolysis of acetylcholine (49). Our data, generated by using the AChE gene promoter fused to a luciferase reporter, show that stimulated m1AChR specifically increased AChE gene promoter activity. Even though basal AChE gene promoter activity was found in all cell types examined so far, AChE message could be determined in FIG. 4. Stimulated m1AChR lead to interaction of EGR-1 with an EGR-specific DNA recognition site. A, electrophoretic mobility shift assays were performed by using ␣-32 P-labeled oligonucleotide and with nuclear extracts from m1 HEK293 cells. The EGR-oligonucleotide complex is indicated by the arrow. Nuclear extracts were corrected for protein concentration prior to the assay. EGR interaction with the specific DNA recognition site was stoichiometrically related to the amounts of EGR-1 protein shown on Western blots. Carbachol-induced EGR binding to the DNA recognition site was completely blocked by atropine. EGR-DNA interaction was mimicked by direct activation of PKC with PMA, and down-regulation of PKC prior to receptor stimulation only partially blocked the interaction. B, supershift analysis using nuclear extract of cells exposed to carbachol for 2 h. The binding reactions contained specific antibodies to EGR-1, EGR-2, or EGR-3.

FIG. 5. EGR-1 induction by specific muscarinic receptor subtypes.
Stimulation of the AChR subtypes m2, m3, and m4 induced a mobility shift, completely blocked by atropine and abolished by specific antibody to EGR-1 in all cell lines. In addition, EGR-1 protein synthesis was activated by stimulation of the receptors. In the nontransfected parent HEK293 control cells (wt), carbachol was ineffective in inducing either a mobility shift or EGR-1 protein synthesis. A, electrophoretic mobility shift assays were performed employing an ␣-32 P-labeled oligonucleotide carrying the EGR binding sequence and nuclear extracts from HEK293 cells stably transfected with m2, m3, and m4 AChR, as well as untransfected HEK293 cells. Nuclear extracts were prepared 2 h after incubation of cells with test substances. B, Western blots of the same nuclear extracts as used for the mobility shift assays were done with EGR-1-specific antibodies.
only a small faction of these cells. HEK293 cells, in particular, have a very low endogenous promoter activity and no detectable AChE enzyme activity (37). We were unable to detect AChE mRNA in either unstimulated or stimulated m1AChR cells, possibly reflecting the absence of AChE mRNA stabilization known to occur during neuronal differentiation (47). If confirmed for the subcortical cholinergic projection system in brain, EGR-dependent regulation of AChE gene transcription may be involved in a receptor-coupled feedback control of cholinergic transmission.
Cholinergic signaling in Alzheimer's disease brain is heavily impaired as a result of the early and massive degeneration of the long basal forebrain projection neurons to brain hippocampus and cortex. Inasmuch as EGR-dependent genes in postsynaptic cholinergic target cells are regulated by muscarinic AChR activity, expression of such genes may be decreased in Alzheimer's disease brains. Postmortem studies are required to test this hypothesis. Drugs designed to activate muscarinic AChRs, including AChE inhibitors and m1 agonists, that are currently being tested in clinical trials for the treatment of Alzheimer's disease may be expected to stimulate transcription of Egr genes along with EGR-dependent target genes. In vivo studies are required to test whether pharmacological treatments designed to stimulate brain muscarinic AChRs increase AChE gene expression, along with AChE enzyme activity and accelerated breakdown of acetylcholine. FIG. 6. Increased amount of EGR by stimulation of muscarinic AChRs results in functionally active EGR in transfected HEK293 cells. HEK293 cells stably transfected with m1AChR were transiently cotransfected with either pTATA-luc/EGR (A), carrying the EGR binding site in front of the luciferase gene under control of the ␤-globin promoter, or pTATA-luc/EGRmut (B), the EGR binding site of which was mutated. Luciferase assays revealed that carbachol and PMA induced transcription from the pTATA-luc/EGR promoter, which was blocked by atropine. No induction was achieved with an inactive, mutated EGR promoter (pTATA-luc/EGRmut). 8-Bromo-cAMP failed to induce transcription. Columns represent means Ϯ S.D. of n ϭ 5 dishes per treatment group. Similar results were determined in four independent experiments. The carbachol-induced increases were significantly different from control (p Ͻ 0.0001, analysis of variance). PMA was also significantly different from control (p Ͻ 0.05).

FIG. 7. The AChE gene promoter is inducible by stimulation of muscarinic AChRs in transfected HEK293 cells.
A, HEK293 cells stably transfected with m1AChR were transiently transfected with pAChE-luc, carrying the promoter region of the AChE gene in front of the luciferase reporter gene. Stimulation of transfected cells with carbachol or PMA induced transcription from the transfected AChE gene promoter. Atropine blocked induction of transcription by carbachol, whereas 8-bromo-cAMP failed to induce transcription. Columns represent means Ϯ S.D. of n ϭ 3 dishes per treatment group. Similar results were obtained in three independent experiments. The carbachol-and PMA-induced increases were significantly different from the control (p Ͻ 0.001, analysis of variance). B, in a control experiment, HEK293 cells were cotransfected with pCMV-Krox24 and pAChE-luc to verify responsiveness of pAChE-luc to Egr expression. As a negative control, cells were transfected with either pCMV-Krox24 or pAChE-luc alone.