Introduction

Top Abstract Introduction Procedures Results Discussion References

In the central nervous system and in peripheral tissues, many actions of acetylcholine (ACh) occur through the activation of muscarinic receptors. Molecular biology studies have led to the identification of five distinct molecular forms of the muscarinic receptors, named m1-m5, and the artificial expression of the cloned receptor subtypes in host cells has allowed the characterization of their signal transduction pathways (Bonner et al., 1987; Peralta et al., 1987; Hulme et al., 1990). Thus, it has been shown that m1, m3, and m5 receptors are predominantly coupled to phospholipase C through G proteins of G_q/11 type, whereas m2 and m4 receptors preferentially regulate adenylyl cyclase and ion channel activities by coupling to G_i/G_o proteins (Peralta et al., 1988). However, a great limitation to the study of the physiological role played by each receptor subtype in the different tissues is the lack of highly selective ligands. Indeed, the muscarinic receptor antagonists currently available display an affinity for one receptor subtype that is <10-fold greater than that for the other subtypes (Caufield, 1993). The limited selectivity of the drugs complicates the receptor characterization in tissues expressing a heterogeneous muscarinic receptor population. In addition, none of the classic antimuscarinic drugs bind with high selectivity to the m4 or m5 receptor subtype (Caufield, 1993). The need of m4-selective ligands is particularly critical because the m4 receptor is almost exclusively expressed in neurons (Wood et al., 1996; Mieda et al., 1997) and is preferentially localized in central neuronal pathways regulating motor and cognitive functions (Levey et al., 1991; Ferrari-Dileo et al., 1994).

Recently, Karlsson and collaborators (1994) reported the isolation of a new peptide toxin, named MT3, from green mamba venom. In radioligand binding studies using Chinese hamster ovary (CHO) cells separately expressing the five cloned muscarinic receptors, muscarinic toxin 3 (MT3) showed a high affinity for the m4 (pK_i = 8.70), a lower affinity for the m1 (pK_i = 7.11), and a very low affinity for the m2, m3, and m5 subtypes (pK_i < 6.0) (Jolkkonen et al., 1994). The distribution of m4 receptors in the rat brain has been studied by using radiolabeled [¹²⁵I]MT3 (Adem et al., 1995; Adem and Karlsson, 1997). MT3 was found to be a potent antagonist of muscarinic inhibition of rat striatal adenylyl cyclase activity, a putative m4-mediated response (Olianas et al., 1996). Thus, MT3 appeared to be a potent and selective antagonist of m4 receptors and a unique tool with which to investigate m4 receptor function.

In the present study, we examined the receptor subtype selectivity of MT3 in functional assays of the cloned human m1-m4 receptors and of the native m4 receptor expressed in two cell lines: the N1E-115 neuroblastoma and the NG108-15 neuroblastoma × glioma hybrid.

	Experimental Procedures

Top Abstract Introduction Procedures Results Discussion References

Materials. [-³²P]ATP (30-40 Ci/mmol), [2.8-H³]cAMP (25 Ci/mmol), and [³⁵S]guanosine-5'-O-(3-thio)triphosphate ([³⁵S]GTPS) (1306 Ci/mmol) were obtained from New England Nuclear-Du Pont (Bad Homburg, Germany). [³H]N-methylscopolamine (NMS) (83 Ci/mmol) was from Amersham (U.K.). MT3 was purified from the venom of Dendroaspis angusticeps as described previously (Jolkkonen et al., 1994). Forskolin and GTPS were from Calbiochem (La Jolla, CA). Himbacine was a generous gift of Prof. W. C. Taylor (Department of Organic Chemistry, University of Sydney, Australia). Pituitary adenylate cyclase activating polypeptide (PACAP) 38 was purchased from Peninsula Laboratories (Merseyside, U.K.). ACh, carbachol chloride (CCh), physostigmine hemisulfate, and the other reagents used were from Sigma Chemical (St. Louis, MO).

Cell Culture and Membrane Preparation. CHO cells stably expressing the cloned human m1-m4 receptors were kindly provided by Prof. A. D. Strosberg (Institut Cochin de Genetique Moleculaire, Paris, France). The cells were grown as a monolayer culture in Ham's F-12 medium (GIBCO BRL) supplemented with 10% fetal calf serum (GIBCO BRL) in a humidified atmosphere (5% CO₂) at 37°C. Cells were grown to ~80% confluency in plastic Petri dishes (Falcon), the medium was removed, and the cells were washed with ice-cold phosphate-buffered saline. The cells were then scraped into an ice-cold buffer containing 10 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES)/NaOH and 1 mM EDTA (pH 7.4) and lysed with a Dounce tissue grinder. The cell lysate was centrifuged at 1000g for 2 min at 4°C. The supernatant was collected and centrifuged at 32,500g for 30 min at 4°C. The pellet was resuspended in homogenization buffer to a protein concentration of ~3.0 mg/ml. The membrane preparations were either used immediately or stored at 70°C.

Assay of [³⁵S]GTPS Binding. CHO cell membranes were diluted 10-fold in 10 mM HEPES/NaOH, 1 mM EDTA, and 0.1% bovine serum albumin (BSA) (pH 7.4); centrifuged; and resuspended in the same buffer. The binding of [³⁵S]GTPS was assayed in a reaction mixture (final volume, 100 µl) containing 25 mM HEPES/NaOH (pH 7.4), 10 mM MgCl₂, 1 mM EDTA, GDP (0.1 µM for m1 and m3 and 1 µM for m2 and m4 receptor activities), 100 mM NaCl, 10 kallikrein inhibitor units (KIU) of aprotinin, and 1.0 to 1.5 nM [³⁵S]GTPS. The incubation was started by adding the membrane suspension (1.5-2.0 µg of protein) and was carried out at 30°C for 60 min. The incubation was terminated by adding 5 ml of ice-cold buffer containing 10 mM HEPES/NaOH (pH 7.4) and 1 mM MgCl₂, immediately followed by rapid filtration through glass-fiber filters (Whatman GF/C) presoaked in the same buffer. The filters were washed twice with 5 ml of buffer, and the radioactivity trapped was determined by liquid scintillation spectrometry. Nonspecific binding was determined in the presence of 100 µM GTPS. Assays were performed in duplicate.

Assay of Adenylyl Cyclase Activity. The enzyme activity was assayed in a reaction mixture (final volume, 100 µl) containing 50 mM HEPES/NaOH (pH 7.4), 2.3 mM MgCl₂, 0.3 mM ethylene glycol bis(-aminoethyl ether)-N,N,N',N'-tetraacetic acid, 0.05 mM [-³²P]ATP (150-200 cpm/pmol), 0.5 mM [³H]cAMP (160 cpm/nmol), 100 µM GTP, 1 mM 3-isobutyl-1-methylxanthine, 5 mM phosphocreatine, 50 units/ml creatine kinase, 50 µg of BSA, 10 µg of bacitracin, 10 KIU of aprotinin, and 10 µM physostigmine. The incubation was started by adding the tissue preparation (30-40 µg of protein) and carried out at 25°C for 20 min. [³²P]cAMP was isolated according to Salomon et al. (1974).

Assay of [³H]NMS. The binding of [³H]NMS to NG108-15 cell membranes was performed in an incubation mixture (final volume, 0.5 ml) containing 50 mM Tris·HCl, 0.5 mM EDTA, 100 KIU/ml aprotinin, and 0.1% BSA (pH 7.4). Membrane protein concentration was 100 to 150 µg/ml. In saturation binding assay, the [³H]NMS concentration ranged from 10 pM to 3 nM, whereas in competition experiments, the radioligand concentration was 0.5 nM. Binding assays were performed at 32°C for 120 min. Nonspecific binding was determined in the presence of 1 µM atropine.

Reverse Transcription-Polymerase Chain Reaction Analysis of Muscarinic Receptor Subtypes. Total RNA was extracted from NG108-15 and N1E-115 cells by using TRIzol (GIBCO BRL) according to the manufacturer's protocol. The RNAs were first treated with 5 units of RNase-free DNase (Boehringer-Mannheim) for 30 min at 37°C in the presence of RNase inhibitor (GIBCO BRL). First-strand cDNA was synthesized by using 2 µg of total RNA and SuperScript II reverse transcriptase (GIBCO BRL) in a final volume of 20 µl containing 0.5 µg of oligo(dT)_12-18 primers, 10 mM dithiothreitol, 0.5 mM deoxynucleotide triphosphates, 50 mM Tris·HCl (pH 8.3), 3 mM MgCl₂, and 75 mM KCl. The amplification reaction mixture (final volume, 50 µl) contained 20 mM Tris·HCl (pH 8.4), 50 mM KCl, 2 mM MgCl₂, 0.2 mM concentration of deoxynucleotide triphosphates, 20 pmol of each primer, 2 µl of cDNA, and 1.25 units of Taq DNA polymerase (GIBCO BRL). The primers used to identify the m2 and m4 receptor subtypes were taken from Drescher et al. (1992) and corresponded to sequences 633 to 653 and 1184 to 1164 of the rat m2 receptor DNA and to sequences 543 to 566 and 1052 to 1029 of the rat m4 receptor cDNA. Polymerase chain reaction (PCR) was performed by an initial denaturation at 94°C for 2 min, followed by 35 cycles at 94°C for 45 s, 55°C for 45 s, 72°C for 60 s, and a final extension at 72°C for 10 min. PCR products (10 µl) were resolved on 1.6% agarose gels and visualized by UV illumination after ethidium bromide staining.

Statistical Analysis. Results are given as mean ± S.E. Concentration-response curves were analyzed by a least-squares curve-fitting computer program (GraphPAD Prism, San Diego, CA). Antagonist effects of MT3 were examined according to Arunlakshana-Schild analysis (Arunlakshana and Schild, 1959), and the potencies were determined from the ratios between the EC₅₀ values of the agonist in the absence and in the presence of different concentrations of the antagonist. The pA₂ values were calculated by using the PHARM/PCS program of Tallarida and Murray (1987). In other experiments where the effect of a single concentration of antagonist was examined, the inhibition constant (K_i) was calculated from the equation:

(1)

where EC_50a and EC_50b are the concentrations of the agonist producing half-maximal effect in the absence and in the presence of the antagonist, respectively, and I is the antagonist concentration. MT3 was also examined for its ability to completely reverse the agonist effect. Experiments were therefore performed in which the effects of multiple concentrations of MT3 on the response elicited by a fixed concentration of the agonist were determined. The data were analyzed as competition curves by nonlinear regression analysis for models of one or two noninteracting sites. The MT3 inhibition constants were calculated according to the equation (Cheng and Prusoff, 1973):

(2)

where IC₅₀ is the concentration of antagonist producing half-maximal inhibition, A is the agonist concentration, and EC₅₀ is the concentration of the agonist producing half-maximal effect. For comparison with pA₂ values, the K_i values were converted to negative logarithmic form (pK_i). Statistical significance of the difference between means was determined by Student's t test.

	Results

Top Abstract Introduction Procedures Results Discussion References

Effects of MT3 on [³⁵S]GTPS Binding to CHO/m1-m4 Cell Membranes. In membranes of CHO/m1, m2, m3, and m4 cells, ACh elicited a concentration-dependent increase of [³⁵S]GTPS binding with EC₅₀ values of 8.5 ± 0.8, 0.28 ± 0.03, 9.0 ± 0.6, and 1.0 ± 0.08 µM, respectively (Fig. 1). The maximal stimulatory effects corresponded to 71 ± 8%, 205 ± 15%, 25 ± 1.2%, and 248 ± 18% increase in basal value, respectively. In CHO/m1 cells, MT3 (0.1-2.0 µM) antagonized the ACh response with a pA₂ value of 6.78 ± 0.08 and a slope of 0.97 ± 0.05. In CHO/m2 cells, the toxin failed to produce a significant shift in the agonist curve at concentrations up to 0.5 µM. Conversely, the M₂ antagonist himbacine (100 nM) increased the EC₅₀ of ACh by 9-fold, a shift yielding a pK_i of 7.91 ± 0.1. In CHO/m3 cells, MT3 (0.1-2.0 µM) was a weak antagonist of the ACh effect with a pA₂ value of 6.30 ± 0.03 and a slope of 0.96 ± 0.08. In contrast, in CHO/m4 cells, the toxin, tested at concentrations ranging from 15 to 500 nM, potently antagonized the ACh stimulation with a pA₂ of 8.33 ± 0.05 and a slope value of 1.01 ± 0.08. At each muscarinic receptor subtype investigated, MT3 per se failed to affect [³⁵S]GTPS binding.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Effects of MT3 on ACh stimulation of [35S]GTPS binding to membranes of CHO cells expressing the cloned human m1 to m4 receptors. [35S]GTPS binding was determined at the indicated concentrations of ACh in the absence () and in the presence of: MT3 100 (), 500 (), and 2000 () nM for the m1; MT3 100 () and 500 () nM and himbacine 100 nM () for the m2; MT3 100 (), 500 (), and 1500 () nM for the m3; and MT3 15 (), 100 (), and 500 () nM for the m4 receptor. Data are the mean ± S.E. of four experiments for m1 and m3 and three experiments for m2 and m4 receptors.

Effects of MT3 on Native Muscarinic Receptors Coupled to Adenylyl Cyclase in N1E-115 and NG108-15 Cells. In N1E-115 cells, 10 nM PACAP 38 stimulated basal adenylyl cyclase activity by 10-fold. CCh inhibited the PACAP-stimulated enzyme activity in a concentrationdependent manner with an EC₅₀ value of 1.6 ± 0.02 µM. The maximal inhibitory effect corresponded to a 25.0 ± 3.5% reduction of control activity (P < .001, n = 8). The addition of 3, 30, and 100 nM MT3 shifted to the right the CCh curve by 6.2-, 40.4-, and 130-fold, respectively, without affecting the maximal inhibitory effect (Fig. 2A). Arunlakshana-Schild analysis of the MT3 antagonism yielded a pA₂ value of 8.81 ± 0.1 with a slope of 0.94 ± 0.04. Increasing concentrations of MT3 completely antagonized the inhibition of PACAP-stimulated cAMP formation elicited by 30 µM CCh (Fig. 2B). The competition curve was monophasic with a Hill coefficient of 0.96 ± 0.07. The pK_i value of MT3 calculated according to eq. 2 was 8.33 ± 0.08. 

View larger version (12K): [in this window] [in a new window]   Fig. 2.   MT3 antagonism of muscarinic inhibition of PACAP 38-stimulated adenylyl cyclase activity in N1E-115 cells. A, the enzyme activity stimulated by 10 nM PACAP 38 was assayed at the indicated concentrations of carbachol (CCh) in the absence () and in the presence of 3 (), 30 (), and 100 () nM MT3. Data are expressed as percentage of the maximal enzyme inhibition elicited by CCh and represent the mean ± S.E. of three experiments. B, the enzyme activity stimulated by 10 nM PACAP 38 was assayed at the indicated concentrations of MT3 in the absence and in the presence of 30 µM CCh. Data are expressed as percentage of the CCh inhibitory effect observed in the absence of MT3 and represent the mean ± S.E. of three experiments. At the concentrations used, MT3 per se failed to affect the enzyme activity.

View larger version (12K): [in this window] [in a new window]   Fig. 3.   MT3 antagonism of muscarinic inhibition of forskolin-stimulated adenylyl cyclase activity in NG108-15 cells. A, the enzyme activity stimulated by 10 µM forskolin was assayed at the indicated concentrations of carbachol (CCh) in the absence () and in the presence of 3 (), 30 (), 100 (), and 300 () nM MT3. Data are expressed as percentage of the maximal enzyme inhibition elicited by CCh and represent the mean ± S.E. of three experiments. B, the enzyme activity stimulated by 10 µM forskolin was assayed at the indicated concentrations of MT3 in the absence and in the presence of 30 µM CCh. Data are expressed as percentage of the CCh inhibitory effect observed in the absence of MT3 and represent the mean ± S.E. of three experiments. At the concentrations used, MT3 per se failed to affect the enzyme activity.

Effects of MT3 on [³H]NMS Binding. Saturation experiments indicated that [³H]NMS bound to CHO/m4 cell membranes with a K_D of 0.20 ± 0.04 nM. Increasing concentrations of MT3 completely displaced the binding of 0.5 nM [³H]NMS with a pK_i value of 8.68 ± 0.09 and a Hill slope of 1.0 (Fig. 4A). In NG-108-15 cell membranes, [³H]NMS binding displayed a K_D of 0.17 ± 0.02 nM and a B_max of 169 ± 25 fmol/mg protein. MT3 displaced 85% of the specific [³H]NMS binding with a pK_i of 8.4 ± 0.2 (Fig. 4B). The remaining fraction of [³H]NMS binding sites was unaffected by toxin concentrations up to 1 µM.

View larger version (10K): [in this window] [in a new window]   Fig. 4.   Concentration-dependent inhibition of [3H]NMS binding. A, the binding of [3H]NMS to CHO/m4 cell membranes was determined as described in Experimental Procedures at the indicated concentrations of MT3. Data are the mean ± S.E of three experiments. B, the binding of [3H]NMS to NG108-15 cell homogenates was determined as described in Experimental Procedures at the indicated concentrations of MT3. Data are the mean ± S.E. of three experiments.

Reverse Transcription-PCR Analysis of m2 and m4 Receptor Expression. Amplification of cDNA transcribed from total RNA of NG108-15 cells using primers specific for the m2 subtype yielded a band of the expected size of 552 bp (Fig. 5, lane 2). This product was not detected in N1E-115 cells (lane 4). On the other hand, reverse transcription (RT)-PCR analysis using primers specific for the m4 subtype yielded a band of the expected size of 510 bp in both NG108-15 (lane 3) and in N1E-115 cell samples (lane 5). No amplification product was obtained in samples of both cell lines when the reverse transcriptase step was omitted (lanes 6 and 7 for NG108-15 and N1E-115, respectively).

View larger version (48K): [in this window] [in a new window]   Fig. 5.   Agarose gel electrophoresis with ethidium bromide staining of PCR products amplified from NG108-15 (lanes 2 and 3) and N1E-115 (lanes 4 and 5) cells, using primers specific for either the m2 (lanes 2 and 4) or the m4 (lanes 3 and 5) receptor. The expected size of the amplification product is 552 bp for the m2 and 510 bp for the m4 receptor. Lane 1, DNA standard; lanes 6 and 7, control samples of NG108-15 and N1E-115 cells, respectively, without reverse transcriptase addition.

	Discussion

Top Abstract Introduction Procedures Results Discussion References

In the present study, the muscarinic receptor selectivity of MT3 has been evaluated in functional assays of cloned and native receptors. The agonist-induced stimulation of [³⁵S]GTPS binding to CHO cell membranes expressing the cloned muscarinic receptor subtypes was used to characterize the pharmacological activity of the toxin. This assay has previously been shown to constitute a valid method for the analysis of the receptor response to a variety of muscarinic agonists and antagonists (Lazareno and Birdsall, 1993; Olianas and Onali, 1996). At each receptor subtype, MT3 failed to stimulate [³⁵S]GTPS binding, indicating a lack of agonist activity. However, the toxin was able to antagonize the receptor activation induced by ACh with a marked subtype selectivity. In fact, MT3 blocked the m4-mediated [³⁵S]GTPS binding with a potency (pA₂ = 8.33) that was 38-fold higher than that displayed at the m1 subtype (pA₂ = 6.78). The toxin failed to block the m2 receptor subtype at concentrations up to 500 nM and displayed a quite low potency (pA₂ = 6.3) in counteracting the activation of the m3 subtype. Both in terms of absolute values and affinity differences, the selectivity profile determined in the functional studies agrees well with that previously observed in radioligand binding studies (Jolkkonen et al., 1994). At the concentrations used, the inhibitory activity of MT3 was characterized by a progressive displacement to the right of the agonist concentration-response curve without depression of the maximal response. Moreover, the slopes of the Schild plots were not significantly different from unity. These data indicate that the toxin behaved as a competitive antagonist.

As the m4 selectivity of MT3 was assessed in radioligand binding (Jolkkonen et al., 1994) and functional (present work) studies using the cloned receptor subtypes overexpressed in host cells, it was of interest to investigate whether the toxin was capable of recognizing the m4 receptor with high affinity in cells expressing the receptor in a native membrane environment. We have therefore considered two cell lines, the N1E-115 and NG108-15 cells, which have previously been shown to express the mRNA encoding the m4 receptor (Fukuda et al., 1988; Peralta et al., 1988; McKinney et al., 1991). In addition, in both cell lines, the activation of the m4 receptor has been demonstrated to cause inhibition of cAMP accumulation (Baumgold and White, 1989; McKinney et al., 1991). Accordingly, we found that CCh elicited a significant inhibition of PACAP 38- and forskolin-stimulated adenylyl cyclase activities in membranes of N1E-115 and NG108-15 cells, respectively. MT3 antagonized the CCh inhibitory effects in the two cell lines with a different pattern. In N1E-115 cells, the toxin counteracted the muscarinic response with high potency (pA₂ = 8.81) and generated monophasic inhibitory curves with a complete reversal of the CCh effect. These data are consistent with the involvement of a homogeneous population of m4 receptors. Conversely, in NG108-15 cells, the Schild plot of the MT3 antagonism yielded a pA₂ of 7.6, which appeared too low for a selective action on m4 receptors. The analysis of the toxin inhibitory curve revealed the presence of two components in the CCh effect: one blocked by the toxin with a potency (pK_i = 8.46) comparable to the affinity for the m4 subtype, and another not reversed by MT3 at a concentration as high as 3 µM. A possible explanation of these findings is that in NG108-15 cells, the muscarinic inhibition of cAMP is mediated not only by m4 but also by another receptor subtype, possibly the m2. The presence of a receptor population relatively insensitive to MT3 was also demonstrated by the data obtained in radioligand binding studies showing that MT3 was unable to completely displace [³H]NMS bound with high affinity. To investigate the possibility that NG108-15 cells express the m2 in addition to the m4 receptor, RT-PCR analysis was conducted using primers specific for the two receptor subtypes. The results indicate that the NG108-15 cells contain the mRNA for both m2 and m4 subtypes, whereas the N1E-115 cells lack the mRNA for the m2 subtype. The failure of previous studies (Peralta et al., 1987; Fukuda et al., 1988) to detect the expression of m2 mRNA in NG108-15 cells by Northern blot analysis may be attributed to the lower sensitivity of the latter assay compared with RT-PCR. It is noteworthy that immunological studies using subtype-selective antisera have previously demonstrated the expression of both m4 and m2 receptors in NG108-15 cells (Yasuda et al., 1993). In addition, Akiyama et al. (1984) reported that pirenzepine, a drug with higher affinity for M₁ and M₄ than for M₂ receptors (Caufield, 1993), inhibited [³H]quinuclidinyl benzilate binding to NG108-15 cell lysate according to a two-site model, with 72% of the labeled sites displaying a high affinity for pirenzepine. This heterogeneity of binding sites was not observed in other studies (Evans et al., 1984; Baumgold and White, 1989; Michel et al., 1989). However, the presence in NG108-15 cells of M₂ sites was postulated by Lazareno et al. (1990) on the basis of kinetic and equilibrium radioligand binding data.

In conclusion, the present study shows that in functional studies, MT3 is capable of recognizing both the cloned and the native m4 receptors with high affinity and selectivity. Moreover, the data demonstrate that in cells where both m4 and m2 receptors may regulate a common response, the toxin constitutes a powerful tool for determining the relative contribution by each receptor subtype.