Distribution and pharmacological characterization of muscarinic-cholinergic receptors in the cockroach brain.код для вставкиСкачать
Archives of Insect Biochemistry and Physiology 16:107-122 (1991) Distribution and Pharmacological Characterization of Muscarin ic-Cholinergic Receptors in the Cockroach Brain Gregory L. Om, Nailah Orr, and Robert M. Hollingworth Pesticide Research Center, Michigun State University (NeuroscienceProgram), East Lansing, Michigan The binding of [3H]quinuclidinyl benzilate to a cockroach brain preparation was investigated. Specific binding was saturable with a K d of 0.25 nM and of 604 pmollmg protein. Kinetic analyScatchard analysis indicated a ,B sis indicated that the ligand is binding in a complex fashion while dissociation followed a simple kinetic process. The pharmacology of the site was typical of muscarinic receptors but the site cannot be characterized in terms of vertebrate muscarinic-receptor subtypes. Affinity of the receptor for agonists was modulated by Mg2+ and guanylylimidodiphosphate but not by pertussis toxin indicating the involvement of a pertussis-toxin insensitive G-protein. Carbamylcholine did not inhibit basal or forskolin-stimulated adenylate cyclase activity. The binding site was localized autoradiographically and was restricted to the median and lateral calyces of the brain. Key words: [3H]quinuclidinyl benzilate, radio-ligand binding, autoradiography INTRODUCTION ACH is an important neurotransmitter in the insect CNS and, as is the case in vertebrates, two types of ACH-receptor are found in the insect CNS: nicotinic and muscarinic [1,2]. However, in contrast to vertebrate neural tissue, Acknowledgments: Dr. G.L. Orr was supported by a Natural Sciences and Engineering Research Council of Canada Postdoctoral Fellowship and a Barnett Rosenberg Fellowship from Michigan State University. We wish to thank Dr. M. Weiss and Ms. I . Smithson for their assistance in the production of the photomicrographs. Received June25,1990; accepted September 16,1990. Address reprint requests to Dr. R.M. Hollingworth, Director, Pesticide Research Center, Michigan State University, East Lansing, MI 48824-1311. 0 1991 Wiley-Liss, Inc. 108 Orretal. insect nACHRs‘ are present at about 10 times the concentration of mACHRs and this may be one reason for the relative lack of information concerning insect mACHRs. Nevertheless, recent studies have succeeded in isolating and expressing in vertebrate cells a cDNA clone encoding the mACHR found in Drasophila CNS [ 3 ] . In vertebrates, different subtypes of neural mACHR can be defined pharmacologically (Ml, M2, M3 types) but genes encoding five mACHR types have been isolated [4,5]. These various subtypes are associated with different transmembrane signaling processes such as inhibition of adenylate cylase AC, activation of PLC, and/or the regulation of K + channels and the receptors are linked to these systems through different guanine-nucleotide-binding regulatory proteins (G-proteins) [ 6 ] .In insects the presence of different mACHR subtypes has not been clearly demonstrated nor has the scope of second-messenger systems associated with mACHRs been established. Studies in locust CNS indicate that insect mACHRs may inhibit AC [7-9] and activate PLC but direct activation of PLC has only been shown using Drosophila mACHRs expressed in vertebrate cell cultures [ 3 ] .The only physiological role of insect mACHRs so far described involves the inhibition of ACH release from synaptosomal preparations  although electrophysiological evidence suggests a postsynaptic role inManduca [lo]. The distribution of these receptors appears to encompass both cell body and synaptic sites  however, visualization of mACHRs in intact tissue is restricted to a single autoradiographic study in cockroach metathoracic ganglia [ll]. In the present study we have identified a very high density population of mACHRs in the cockroach brain using [3H]QNBand have performed an extensive pharmacological investigation to determine if these receptors can be classified in terms of the vertebrate mACHR subtypes. In addition, we have demonstrated the role of G-proteins in regulating the receptor and investigated the nature of the second-messenger system associated with it. Finally, we present the first autoradiographic study of insect mACHRs using [3H]QNB and clearly establish the morphological distribution of these sites in the brain. MATERIALS AND METHODS I3H1QNB Binding The methodology followed was similar to that described for the characterization of a-BGT binding sites in the cockroach brain [ E l . Brains (without optielobes) were dissected from adult (1-3 months after final molt) male American cockroaches, Peeviplaneta americma, and placed in ice cold buffer A (50 mM Tris*Abbreviations used: AC = adenylate cyclase; ACH = acetylcholine; BGT = bungarotoxin; B, = maximal binding-site concentration; 4-DAMP = 4-diphenylacetoxy-N-rnethylpiperidine methiodide; DlT = dithiothreitol; GppNHp = guanylylimidodiphosphate; bbs = experimental association rate; ba = off-rate constant; bn= on-rate constant; & = equilibrium dissociation constant; Ki = inhibition constant; mACHR = muscarinic-acetylcholine receptor; McN-A-343 = (4-hyd roxy-2-buty nyI)trime t hylammon i u m chloride ; MLA = methyl Iycacon iti ne cit rate; nACHR = nicotinic-acetylcholine receptor; OXA-22 = cis-2-methyl-5-trimethylammoniummethyl-l,3-oxathiolane iodide; PLC = phospholipase C; PT = pertussis toxin; [3HlQNB = L-[benzili~-4,4’-~H( N)J-quinuclidinylbenzilate; SD-35651 = tetrahydro-2-(nitromethylene)-2H-II 3-thiazine. Muscarink Receptors in the Cockroach Brain 109 HCI, 120 mM NaCl, pH 7.4). The tissue was initially homogenized in a glass/ glass hand homogenizer (10 strokes) followed by further homogenization using a motorized glassheflon homogenizer (15 strokes, 550 rpm). The homogenate was centrifuged (28,00Og, 10 min) and the resulting pellet was resuspended in 3 ml buffer A and recentrifuged. The final pellet was resuspended in buffer A for use in the binding assay. Tissue aliquots (30-40 pg protein) were incubated at 24°C for the times indicated in a reaction medium containing 50 mM Tris-HCI, 2 mg/ml BSA, 120 mM NaC1, (39 Ci/mmol) [3H]QNB(New England Nuclear Research Products, Wilmington, DE) and any competing ligands (final volume 200 p1, pH 7.4). In experiments using ACH the tissue was preincubated for 10 min with 40 cl.M neostigmine to inhibit acetylcholinesterase. The reaction was initiated by the addition of tissue and, following incubation, bound and free ligand were separated by vacuum filtration of the medium over Whatman GF/C filters (Whatman, Clifton, NJ) presoaked with buffer A containing 10 mg/ml BSA using a Brandel cell harvester (model M-30R; Brandel Ltd., Gaithersburg, MD). The filters were rapidly washed twice with 3 ml buffer B (buffer A containing 2 mgiml BSA). Filter disks were removed and placed in scintillation vials with Safety-Solve (Research Products Int., Mount Prospect, IL) scintillation fluid and bound radioactivity determined using a TM Analytical, Inc. (Elk Grove Village, IL) scintillation counter (model 6895). Binding to tissue was determined with duplicate tissue aliquots and nonspecific binding was defined as binding in the presence of 0.1 mM atropine. Reported values are the means of data from two to four separate experiments. In experiments assessing the effect of PT the toxin was activated prior to use by incubation for 1 h at 37°C in the presence of 50 mM DTT. The initial tissue pellet described above was incubated for 1 h at 37°C in a reaction medium containing 20 mM Tris-HCI, 1 mM nicotinamide-adenine dinucleotide NAD, 1 mM ATP, 1 pg/ml BSA, 6.25 mM DTT, 1mM EDTA, 10 mM thymidine, and 15 pg/ml PT (pH 7.5). The reaction was stopped by the addition of 5 ml ice cold Tris-HCl(20mM final concentration) and the solution was centrifuged at 28,0009 for 10 min. The resulting pellet was resuspended in buffer A and used in the binding experiments. Analysis of saturation, Competition, and ordoff rate experiments were performed using EBDAiLIGAND and KINETIC computer software programs (G.A. McPherson, Elsevier-BIOSOFT, Cambridge, England) which utilize nonlinear least squares curve fitting techniques. These programs compare models applied to the data by testing for a significant reduction in weighted sum of squares (i.e., an improved fit) using an F-test. r3H1QNB Autoradiography Brains (cerebral ganglia) from adult male American cockroaches were dissected free of any adhering tissue and placed in ice cold buffer C (75 mM sodium phosphate, 140 mM sucrose, pH 7.4) and kept on ice for up to 30 min. The tissue was covered with Ames O.C.T. compound (Ames, Elkhart, IN) and mounted on a specimen holder of an IEC-CTF cryostat (International Equipment Co., Needham Heights, MA). The block was trimmed and sectioned at 8 pm. Near serial sections were thaw-mounted on chrome-aluminum gelatin- 110 Orretal. coated glass slides and frozen at - 20°C for not more than 3 weeks. Slides were preincubated on ice, in buffer D (120 mM NaCl, 1.8 mM K2HP04, 0.2 mM KH2P04,1.8 mM CaC12,40 mM Tris-HC1,4.8 mM Tris-base, 0.2% BSA, pH 7.4) for 10 min, prior to incubation with labeled ligand. Sections were then incubated for 40 min at 24°C in 2.5 nM [3H]QNB. For nonspecific binding, sections were coincubated with either 0.1 mM or 1 p.M atropine. The reaction was terminated by rapidly dipping the sections in ice cold buffer D. Sections were fixed on ice for 30 min with 2% glutaraldehyde in 75 mM sodium phosphate buffer, pH 7.4, rinsed with distilled water, cold-air dried, stored overnight at 4°C with desiccant, and prepared for autoradiography by one of two methods. Briefly, glass coverslips were coated with Kodak NTB3 emulsion (diluted 1:l with water) and apposed to the tissue sections as described by Young and Kuhar  or the sections were dipped in a 1:l diluted solution of Kodak NTB3. In both cases, the autoradiograms were exposed for 6-8 weeks at 4°C in a light-proof container in the presence of a desiccant. Both types of autoradiograms were then developed with Kodak D-19 developer, fixed, stained with toluidine blue, and mounted in DPX (Fluka, Ronkonkoma, Switzerland). Sections were photographed in brightfield and darkfield, with a Zeiss Standard-18 microscope using a Nikon microflex UFX-I1 photogTaphic attachment. Sections shown are representative of at least 10-15 similar treatments. cAMP Determination Brains were dissected and homogenized in ice cold Tris-EDTA buffer (10 mM Tris, 1.0 mM DTT, 1.0 mM EDTA, pH 7.0) using a glass-teflon homogenizer. The resulting homogenate was centrifuged at 28,0009 for 10 min at 5°C and the pellet was resuspeitded in Tris-DTT buffer (10 mM Tris, 1.0 mM DTT, pH 7.0). The suspension was centrifuged again at 28,OOOg for 10 min at 5°C and resuspended in Tris-DTT buffer for determination of cAMP production. AMP production was determined in an incubation mixture containing 75 mM Tris-HC1(pH 7.2), 0.1 mM guanosine triphosphate GTP, 0.1 mM 3-isobutyll-methylxanthine, 2 mM MgC12, 0.5 mM ATP, 0.3 mM EGTA, 100 mM NaCl, test compounds, and tissue preparation in a final volume of 200 p.1. The reaction was initiated by the addition of ATP and conducted at 30°C in a shaking water bath. The reaction was terminated by immersion of the tubes in a boiling water bath for 2 min. The mixture was diluted with 50 mM sodium acetate buffer (pH 6.2), centrifuged at 2,8009 for 10 rnin and a 100 p1 aliquot assayed for AMP content using a modification  of the RIANEN radioimmunoassay kit (New England Nuclear Research Products). Chemicals $-DAMP, S( + )-dexetamide hydrochloride, R( - )-levetamide hydrochloride, McN-A-343, methoctramine hydrochloride, OXA-22, oxotremorine methiodide, oxotremorine sesquifumara te, ( 2 )-pilocarpine hydrochloride, ( - )quinuclidinyl benzilate, ( fr )-quinuclidinyl benzilate, and ( - )-scopolamine hydrobromide were purchased from Research Biochemicals Inc., Natick, MA. Lobeline hydrochloride, nicotine, atropine sulfate, ipratropium bromide, ACH chloride, carbamylcholine chloride, neostigmine methyl sulphate, forskolin, 5'-GppNHp, and PTwere purchased from Sigma Chemical Co.,St. Louis, MO. Muscarinic Receptors in the Cockroach Brain 11 1 The following compounds were generously provided free of charge: pirenzipine and AF-DX-116, Boehringer Ingelheim Pharmaceuticals Inc., Ridgefield, CT; morantel and pyrantel tartrate, Dr. K. Gration, Pfizer Central Research, Sandwich, Kent, England; MLA, Dr. M.H. Benn, University of Calgary; and SD-35651, Shell Development Co., Modesto, CA. All other chemicals were of reagent grade. RESULTS [jHIQNB Binding Specific binding of [3H]QNB was linear with respect to sample protein between 10-90 kg protein and equilibrium binding conditions were reached following a 20-30 min incubation with 2.5 nM [3H]QNBand after 60-90 min with 0.3 nM (data not shown). Specific binding was saturable under equilibrium conditions at concentrations of [3H]QNBabove 2 nM with very low levels of nonspecific binding (Fig. 1A). No additional binding sites were observed at ligand concentrationsas high as 16 nM. Scatchard analysis of these data yielded a linear plot (Fig. 1B) which results in a Kd and B,, of 0.25 2 0.02 nM and 604 f 31 fmol/mg protein, respectively. The Hill plot (Fig. 1C) calculated from these data was linear and the coefficient (0.97 ? 0.04) approximates unity indicating the involvement of a single, noncooperative population of binding sites. These parameters are summarized and compared with similar data obtained in studies of the a-BGT binding site in the cockroach brain (Table 1). The second-order association rate was simplified to a pseudo-first order process by ensuring that less than 10% of the ligand was bound at equilibrium (Fig. 2A). Assuming a simple kinetic model following the law of mass action, the k,,bs was calculated as 0.23 2 0.02 min-". However, the association rate was best fit by a bi-exponential model ( P < 0.001) and yielded b b s values of 0.15 k 0.01 and1.6 k 0.63min-l. Whenexpressedasapseudo-firstorderprocess [ln(Beq/Beq-B)vs. time] the plot was curvilinear (Fig. 2A, inset) which is indicative of a complex association process . Dissociation was measured by allowing binding to reach equilibrium with 2.5 nM 13H]QNB, adding 10 p M atropine, and monitoring binding at subsequent times (Fig, 2B). Dissociation was slow with 50% dissociation of ligand occurring after approximately 45 min incubation with atropine. Analysis of these data indicated that the process was best fit by a mono-exponential model min-l. This conwhich yielded an off rate constant Kd of 1.4 2 0.17 X clusion is supported by the linear first-order plot [ln(LR/LRo)vs. time] of the data (Fig. 2B, inset) which is typical of a simple first-order reaction . Assuming a sim le kinetic model, the on-rate constant is calculated to be 9.1 x lo7 M-'m- (hbs-k&ligand concentration), The Kd calculated using this rate constant is 0.15 nM (kff/kn). Using the two-site model for association two on-rate constants can be calculated, 5.9 x 107and6.6 X lo8 M-'rn-'. The Kd values calculated from these rate constants for the two putative sites identified by the association experiments equal 0.23 and 0.02 nM with only 11.7% 2 2.1 existing in the high affinity form. Calculations are summarized in Table 2 and compared with data obtained for the a-BGT binding site in the same tissue. P 112 Orretal. 0 1 2 5 4 5 6 20 3H-IIN8 (nu) 0.00 4o 100 200 300 400 500 3H-QNB Bound (fmoles/mg protein) -0.80-10.5 . -10.0 -9.5 -9.0 -8.5 Log (M) free 3H-ONB Fig. 1. Binding of L3H1QNB to cockroach brain membranes. A: Saturation of binding with nonspecific binding (a),and speincreasing ligand concentration showing total binding (O), Data are representative of three separate experiments. B: Scatchard analysis cific binding (A). of the specific binding in A. C: Hill plot of the specific binding in A. Muscarinic Receptors in the Cockroach Brain 113 TABLE 1. Comparison of Binding Parameters Obtained in Equilibrium Studies of Muscarinicand Nicotinic-Binding Sites in Cockroach Brain Ligand Kd (nM) Brnax 0.25 2 0.02c 1.09 2 0.17 [3H]QNB" ['251]~-BGTb Hi11 coefficient (fmoVmF9 604 8926 * 31 k 492 0.97 2 0.04 0.90 ? 0.04 'Present study. bOrr et al. 1990. 'Values equal the mean 2 SEM for three separate experiments. : : m .U Ej ._ 200 0.10 / 100 w 0 m 4 3 0 0 p 0 00 a 0 , , , , , , , , , , 1 2 3 4 5 1 7 B 9 1 0 1 1 1 2 1 J 1 . ~ 5 TlML (mi?) o 0 5 10 15 25 20 30 Time (rnin) P 3 m w 70 60 50- $ 40: 30 20 . 10- -1.ool 1 0 10 20 I 10 20 M nme (mi"] 40 30 50 60 40 50 I Time (min) Fig. 2. Rates of association and dissociation of ['HIQNB in cockroach brain membranes. A Association rate determined with 2.5 nM ['HIQNB. Values are representativeof three separate experiments. Inset: the pseudo first-order rate plot of the data in (A). Beq = specific binding at equilibrium; B = specific binding at the time of determination. 6: Dissociation rate determined by incubating tissue aliquots with 2.5 nM ['HIQNB for 25 min followed by the addition of 10 pM atropine. Bound ligand was then determined at the appropriate times. Values are representative of two separate experiments. Inset: data in (B) plotted as a first-order reaction. LR,, = concentration of ligand-receptor complex at equilibrium; LR = concentration of ligand-receptor complex at times following addition of atropine. 114 Orretal. TABLE 2. Binding Parameters Obtained in Kinetic Studies of Muscarinic and Nicotinic Binding Sites in Cockroach Brain kclbsl Ligand (M-') QNB" &-BGT~ 0.15' 0.24 kbs2 konl (M-') (M-lrn-') 1.6 - 5.9 X lo7 2.5 x 10' L l 2 lhffl kff2 (M-lrn-') (m-') (m-') 6.6 x 10' 1.4 x 2.5 x 107 7.2 x 1 0 - 4 5.9 x 10-3 &I1 Kd2 nM nM 0.23 0.03 0.02 0.25 aPresentstudy. bOrret al. 1990. 'Values are the mean of two to three separate experiments. Pharmacological Characterization The pharmacological nature of the 13H]QNBbinding site(s)was investigated by performing displacement studies with a number of selective and general cholinergic agents (Table 3). Muscarinic compounds were much more effective than nicotinic agents (MLA, lobeline, and nicotine) and the most potent compounds were found amongst those which do not discriminate between the various muscarinic classes. The biologically active enantiomer of QNB, (-)QNB, had the highest affinity observed (Ki = 0.17 nM) which compared well with the affinity of the labeled ligand. A racemic mixture of QNB was 4.4-fold less effective than the active isomer alone. Similar potency was observed for dexetamide with scopolamine, atropine, and ipratropium displaying affinities in the nanomolar range (1-7 nM). The muscarinic agonist OXA-22 and ACH had comparatively low affinities (5-7 pM) as did levetamide, the inactive isomer of dexetamide. Somewhat lower potency (10-40 p M ) was displayed by the muscarinic agonists oxotremorine, pilocarpine, carbamylcholine, and the class A muscarinic agonist, oxotremorine-M. The anthelmintics, morantel and pyrantel, and the insecticide SD-35651 were poor displacing agents in this system. In general, antagonists were associated with a Hill coefficient approximating unity while agonists had Hill values significantly less than one. Exceptions to this rule were seen with the agonists oxotremorine and pilocarpine and the racemic mixture of QNB. To compare the pharmacology of these binding sites with that described for vertebrate muscarinic receptor classes a variety of compounds were tested which discriminate between vertebrate M1, M2, and M3 receptor types. Similar affinities (0.2-0.5 FM) were found for pirenzipine, methoctramine, and 4-DAMP which have a high affinity for M1, M2, and M3 receptors, respectively. AF-DX-116, an M2 selective antagonist, was 17-45-fold less effective than the compounds mentioned above and the M1 selective agonist McN-A-343was 3-fold less effective than AF-DX-116. All of these agents had Hill values approximating unity. The displacement curves for these selective agents and atropine are seen in Figure 3. The regulation of these sites by G-proteins was investigated by studying the displacement of [3H]QNBby the cholinergic agonist carbamylcholine (Table 4). Under standard conditions the displacement curve (Ki = 40 pM) was best fit by a two-site model (P < 0.01) with the Kd for the high and low affinity states of 7.5 and 149 pM, respectively. The distribution of sites in these two states was approximately equal and the Hill coefficient was significantly Muscarinic Receptors in the Cockroach Brain 11 5 TABLE 3. Pharmacology of i3H]QNB Binding Sites in the Cockroach Brain Ligand Selective agents M1-muscarinic Pirenzipine (ant.)c McN-A-343 (ag.) M2-muscarinic Methoctramine (ant.) AF-DX-116 (ant.) M3-muscarinic 4-DAMP (ant.) Nicotinic Methyllycaconitine Lobeline Nicotine General cholinergic agents ( - ) QNB (ant.) Dexetamide (ant.) ( 2 )QNB (ant.) Scopolamine (ant.) Atropine (ant.) Ipratropium (ant.) Oxa-22 (ag.) A C H ~(ag.) Levetamide (ant.) Oxotremorine-M (ag.) Oxotremorine (ag.) Pilocarpine (ag.) Carbamylcholine (ag.) Morantel Pyrantel SD-35651 Ki (MPb Hill coefficientb 2.2 x 10-7 2.8 x 10-5 1.05 0.96 5.0 x 10-7 8.6 X 0.96 0.97 1.9 x 10-7 1.20 > I x 10 * >1 x 10-4 >I x 10-4 - 1.7 X 5.0 x 7.5 x 1.4 x lo-"' lo-'' 10-1" 3.6 X 6.6 X 4.8 x 5.0 x 6.8 X 1.0 x 1.3 x 1.6 x 4.0 x >i x >1 x >I x lo-' 10-9 0.94 1.00 0.73 0.85 0.98 0.85 0.78 0.72 1.40 10-5 10-5 10-5 10-5 10-4 10 10-4 0.66 0.99 1.10 0.67 ~ aKi = K&/[1 -I-L/Kd] where ICs0 = the concentration of ligand required to inhibit 50% of specific binding, L = the concentration of labeled ligand, and Kd = the affinity of the labeled ligand. bValues are the mean of two to three separate experiments. antagonist, ag. = agonist. 'ant. dTissuepreincubated 10 min with 40 pM neostigmine. L less than one (0.63). Inclusion of MgC12 (5 mM) in the reaction media with carbamylcholineimproved the Ki value (7.2 pM) by 5.5-fold and shifted the Kd for the high and low affinity sites to 1.7 and 66.3 p M (Hill coefficient = 0.58). In the presence of both Mg2+ and GppNHp the Ki for carbarnylcholine was reduced to 57 pM and while the data were still best fit by a two-site model (P < 0.03), the distribution of sites was predominantly (76%) in the higher affinity state. This redistribution of sites was also reflected in the Hill coefficient (0.79) which is closer to unity than in the absence of GppNHp. Preincubation of the membranes with PT prior to conducting displacement studies had no significant effect on displacement. The displacement curves are shown in Figure 4. The possibility that these sites were acting through the inhibition of AC was investigated in membrane preparations of the cockroach brain (Table 5). Orretal. 116 V c 0 3 m 50 40 30 20 --,- 10 Ol, 10 9 7 8 6 5 4 3 Ligond Concentration (-log M) Fig. 3. Displacement of L3H1QNBby various selective agents and atropine. (W) atropine, pirenzipine, (a)methoctramine, (U)AF-DX-116, (A)McN-A-343. Valuesare rep(AII-DAMP,(0) resentative of two to three separate experiments. Carbamylcholine had no significant effect on basal or forskolin-stimulated cyclic AMP production in this tissue. Localization The distribution of [3H]QNBbinding sites was shown using autoradiographic techniques (Fig. 5A-C). Use of wet emulsion (dipping) to coat the sections after exposure to the ligand resulted in excessive diffusion of ligand from the sections making localization impossible. However, use of a dry emulsion coated coverslip closely apposed to the labeled section resulted in excellent grain development over areas of binding with little or no diffusion. Furthermore, this technique allowed staining of the same section which produced the autoradiographic image, thus providing a more accurate determination of the morphological distribution of the grains. Figure 5A shows a stained section in which the major neuropile areas of the brain can be seen. Grain development was limited almost exclusively to the areas which correspond to the median and lateral calyces (Fig. 58). No grain development above background was apparent in the nonspecific binding section which was exposed to t3H]QNB in the presence of 1.0 pM or 100 pM (data not shown) atropine. Similar techniques were used to visualize [3H]QNBbinding sites in sixth abdominal ganglia. TABLE 4. Effects of Mg" ,Guanine Nucleotides, and PT on CarbamylcholineBinding Mg2' Carbamylcholine Ki (+MI Hill coefficient KdHa (FM) (FM) % Hiah 40 1 7' (5 mM) Mg2+ + GppNHp Mg2+ + PT (5 mM + 100 pMj (5 mM + 15 p,g/mlj 0.63 2 0.05 7.5 f 1.8 149 22 * 7.2 2 0.6 0.58 & 0.02 1.7 f 0.9 66.3 r 6.3 57 f 2.9 0.79 k 0.03 33.0 2 3.3 492 f 170 50 60 76 dissociation constant for the high affinity site. dissociation constant for the low affinity site. "Values are the mean t SEM for two to four separate experiments. a K d = ~ b&t = 13.2 f 2.1 0.60 2 0.02 1.7 ? 0.6 57.6 t 15.0 46 Muscarinic Receptors in the Cockroach Brain 7 6 5 4 3 117 2 Carbornylcholine (-log M) Fig. 4. Displacement of 13H]QNB by carbamylcholine in the presence and absence of Mg2' and/or CppNHp. (0) carbamylcholine, (0)carbarnylcholine + 5 mM Mg2+,(A) carbarnylcholine t 5 mM M$+ + 100 pM GppNHp. Values are representative of three separate experiments. DISCUSSION Scatchard and Hill analyses of the [3H]QNB binding sites in the cockroach brain indicate that this tissue has a single population of noninteracting muscarinic binding sites. These results are similar to those from other studies of insect neural mACHRs [ l l , 16-19] but differ from reports in Schistocerca neural tissue  which indicate the presence of at least two separate populations of [3H]QNBbinding sites. The high concentration of muscarinic binding sites in the cockroach brain correlates well with the high level of nicotinic sites reported in this tissue  and, with the exception of cricket ganglia , is the highest muscarinic-receptor density reported in insects. In addition, the ratio of nicotinic to muscarinic binding sites in the cockroach brain (Z15:l) is consistent with reports in other insect studies [ l ] and underscores a fundamental difference between insect and vertebrate neural tissue. The affinity of this binding site is at least an order of magnitude higher than that reported in studies of insect nerve cord [11,18,19] but is comparable to similar studies in insect brain [16,17,21]. This may reflect a fundamental difference between muscarinic receptors in the brain and nerve cord but additional studies of the gene(s) encoding these receptors will be required to pinpoint the nature of any putative structural differences. The comparative ease with which such a high density, high affinity binding TABLE 5. Effect of Various Compounds on Cyclic AMP Production in the Cockroach Brain Cyclic AMP increase (pmol/min/mg) Treatment Basal 940 t 94" Carbamylcholine (100 pM) Forskolin (1 KM) Forskolin + carbamylcholine Forskolin + carbamylcholine + atropine (100 pM) 842 aValuesare the mean 2 SEM for three separate experiments. * 62 1921 f 105 1739 * 119 2198 k 161 118 Orretal. Fig. 5. A: Toluidine-blue stained frontal section of the cockroach brain showing regions of neuropile: A, alpha lobe; 6,beta lobe; CB, central body; GA, glomeruli of the antenna1 lobe; GC, globuli cells; LC, lateral calyx; MC, medial calyx. B: Darkfield autoradiograph showing total ['HIQNB binding in the calyx regions of the neuropile. C: A similar section to (B) showing binding in the presence of 1.O FM atropine (nonspecific binding). Scale bar = 0.05 rnrn. Muscarink Receptors in the Cockroach Brain 119 site can be studied makes this tissue ideal for future investigation of insect mACHRs. A similar situation has previously been described for nACHRs in cockroach brain (121 thus demonstrating that cockroach brain could be a useful model system for studying cholinergic receptor structure, function, and pharmacology in insects. Assuming simple binding kinetics, the Kd calculated from the rate constants for [‘HIQNB binding correlates well with the Kd calculated at equilibrium. Havever, careful analysis of the association rate indicates that [3H]QNBis binding in a complex fashion while dissociation is proceeding via a simple kinetic process. The lower affinity Kd calculated using the two resulting on-rate constants is virtually identical to that calculated at equilibrium while the higher affinity receptor population is not apparent under equilibrium conditions. This situation could reflect the very low concentration of this high affinity site and the difficulties in quantifying such low levels of binding but it may also involve the isomerization of the receptor by QNB itself. The latter possibility is well documented in vertebrate neural tissue where it is revealed by complex binding kinetics although a single population of receptors is detected under equilibrium conditions . Such an effect has not been previously reported in insects but could be expected considering the similarities between insect and vertebrate mACHRs [1,2]. Complex binding kinetics were also observed in a study of cockroach brain nACHRs  indicating the need for more detailed studies of ligand-receptor interactions at insect cholinergic receptors. The pharmacological nature of the [3H]QNBbinding site is typical of both insect and vertebrate mACHRs with muscarinic agents being much more effective than nicotinic compounds in displacement studies. In this regard, the lack of effect of the insecticidal plant alkaloid MLA and the anthelmintics morantel and pyrantel, all of which have been shown to have high affinity for nACHRs in cockroach brain , indicate that these compounds effectively distinguish between nicotinic and muscarinic receptors. As was seen with the brain nicotinic receptor [121, the insecticidal nitromethylene heterocycle, SD-35651, was of limited effectiveness suggesting that its disruption of cholinergic systems may involve sites removed from those which bind a-BGT and QNB. The stereospecificity of the site is shown by the increased effectiveness of the active enantiomer of QNB over that of the racemic mixture and the 1,000-fold greater efficacy of dexetamide over its less active isomer, levetamide. Therefore, the pharmacology of the brain [3H]QNBbinding site is consistent with it being a mACHR. Subtypes of muscarinic receptors in vertebrates are classified on a pharmacological, functional, and regional basis as M1, M2, and M3 and on a genetic basis as ml-m5 . A variety of agents are available which have a well defined selectivity for these sites and can be used to define the pharmacological nature of a given receptor population. However, none of these compounds has a high affinity for only one receptor subtype and, ideally, a number of different compounds should be used to establish the classification of a particular site. Studies using the M1 selective compound pirenzipine [7,8] and the M2 selective agent AF-DX-116  as displacing agents have identified binding sites in insects with similarities to vertebrate Ml and M2 receptors. In the present study, the order of effectiveness of pirenzipine, AF-DX-116, the M2 selective 120 Orretal. compound methoctramine, and the M3 selective agent 4-DAMP is similar to that seen with vertebrate M1 receptors  but the Ki values for pirenzipine, methoctramine, and 4-DAMP are at least an order of magnitude poorer than would be expected of a typical M1 receptor. Therefore, we are reluctant to classify this as an M1 receptor based solely on our pharmacological data. Establishing the degree of homology will ultimately require isolation and expression of the gene encoding the brain mACHR. Interestingly, the Drosophilu mACHR gene has recently been sequenced and expressed and although it has certain similarities to the vertebrate M1 receptor it has a relatively low affinityfor pirenzipine (0.57 pM) . The Hill coefficients calculated from displacement curves for the selective agents all approximate unity suggesting that these compounds are displacing QNB from a single site. This differs significantly from the previously mentioned studies in Schistocercu brain where multiple binding sites for pirenzipine and AF-DX-116 are suggested by complex displacement curves with Hill coefficients between 0.5-0.6. Furthermore, high and low affinity binding sites for pirenzipine have been reported to be present in cell bodies and synaptosomal fractions, respectively, of Locusta nervous tissue . This discrepancy suggests the potential for species variation in mACHR distribution and function which deserves further investigation, In this context it should be noted that only a single gene encoding a mACHR could be detected in Drosophilu . Muscarinic receptor effects are mediated by G-proteins which, in the presence of divalent cations (e.g., Mg2+)and guanine nucleotides, modulate the affinity of the receptor for agonists but not antagonists . The receptor population will exist as a mixture of high and low affinity sites dependent on the degree of receptor/G-protein interaction. A shift in agonist affinity in the presence of these compounds has been demonstrated in insects and supports the premise that insect mACHRs are coupled to G-proteins [23,24]. To extend these observations we monitored the distribution of receptors in various affinity states when using carbamylcholine as a "typical" agonist to displace [3H]QNB. Evidence was obtained for the existence of at least three affinity states for the receptor. In the presence of Mg2+ the receptor is distributed betwe'en a high (60%, Kd 2 pM) and an intermediate (40%,Kd 30-60 FM) affinity state. The addition of GppNHp abolishes the high affinity condition and the receptor is found predominantly in the intermediate affinity (76%)but a low affinity (24%,Kd 500 pM) state is also observed. This multiplicity of agonist affinity states has previously been observed with vertebrate mACHRs  but has not been reported in insects. These observations are consistent with the involvement of G-proteins in mACHR actions in the cockroach brain. Many G-proteins are ribosylated by PT thereby inhibiting their ability to interact with the receptor. Therefore, if the mACHR is being modulated by a PT-sensitive G-protein, pretreatment of the tissue with PT should shift the receptor to a low affinity state similar to that seen in the presence of GppNHp [22,26]. However, PT pretreatment had no effect on this system suggesting that a PT-insensitive G-protein is involved. PT-insensitive G-proteins are associated with mACHR-mediated activation of PLC in vertebrates [6,27,28] and it is possible that the cockroach brain receptors are also associated with this response. While PT-sensitive inhibition of AC by mACHRs is well documented in vertebrates, Muscarinic Receptors in the Cockroach Brain 121 and muscarinic agonists have been reported to inhibit AC in insects [7-91, we saw no effect of carbamylcholine on basal or forskolin-stimulaled cyclic AMP production in our system ruling out the involvement of a PT-insensitive G-protein mediating AC inhibition. Work is proceeding to determine if these receptors are coupled to activation of PLC and the production of inositol triphosphate and diacylglycerol as has been suggested by studies of cockroach nerve cord , locust brain , and the Drosophila mACHR . The cellular localization of insect rnACHRs has been demonstrated only in the cockroach metathoracic ganglia using [N-methyl-3H]scopolamine[111. In this ganglia the receptors appear to be dispersed throughout the neuropile. The present study shows for the first time the distribution of [3H]QNBbinding sites in insect nervous tissue and confirms the presence of mACHRs in ganglionic neuropile (data not shown). In comparison to these ganglionic sites it is striking that the brain sites are concentrated in one particular area of the neuropile, the median, and lateral calyces. Unlike the distribution of nACHRs  there appears to be little or no binding in the other neuropile regions of the brain. The physiological role of these receptors is unclear but mACHRs in locust brain appear to act presynaptically to regulate ACH release . Therefore, since the calyces receive a great deal of sensory input from the antenna1 lobes , these receptors may be involved in regulating the release of ACH from the afferent neurons and thereby coordinating the flow of information to the brain. LITERATURE CITED 1. Sattelle DB:Acetylcholine receptors. In: Comprehensive Insect Physiology, Biochemistry and Pharmacology. Kerkut GA, Gilbert, LI eds. Pergamon Press, Oxford, vol 11, pp 395-434 (1985). 2. Breer H, Sattelle DB: Molecular properties and functions of insect acetylcholine receptors. J Insect Physiol33, 771 (1987). 3. Shapiro RA, Wakimoto BA, Subers EM, Nathanson N: Characterization and functional expression in mammalian cells of genomic and cDNA clones encoding a Drosophila muscarinic receptor. Proc Natl Acad Sci USA 86,9039 (1989). 4. Mei L, Roeske WR, Yamamura HI: Molecular pharmacology of muscarinic receptor heterogeneity. Life Sci 45, 1831 (1989). 5. Bonner T New subtypes of muscarinic acetylcholine receptors. Trends Pharmacol Sci Suppl, Dec 1989, p 11. 6. Lechleiter J, Peralta E, Clapham D: Diverse functions of rnuscarinic acetylcholine receptor subtypes. Trends Pharmacol Sci Suppl, Dec 1989, p 34. 7. Duggan MJ, Lunt GG: Coupling of muscarinic receptors to second messenger systems in locust ganglia. In:Neurotox '88 Molecular Basis of Drug and Pesticide Action. Lunt GG, ed. Elsevier Science Publishers, Amsterdam, pp 245-253 (1988). 8. Knipper M, Breer H Subtypes of muscarinic receptors in insect nervous system. Comp Biochem Physiol9UC, 275 (1988). 9. Knipper M, Breer H: Muscarinic receptors modulating acetylcholine release from insect synaptosomes. Comp Biochem Physiol93C, 287 (1989). 10. Trimmer BA,Weeks JC: Effects of nicotinic and muscarinic agents on an identified motoneurone and its direct afferent inputs in larval Munducu sextu. J Exp Biol144,303 (1989). 11. Lummis SCR, Sattelle DB: [N-methyl-3HH]Scopolaminebinding sites in the central nervous system of the cockroach Periplaneta americuna. Arch Insect Biochem Physiol3,339 (1986). 12. Orr GL, Orr N, Hollingworth RM: Localization and pharmacological characterization of nicotinic-cholinergic binding sites in cockroach brain using a- and neuronal bungarotoxin. Insect Biochem 20,557 (1990). 13. Young WS, 111, Kuhar MJ: A new method for receptor autoradiography: [3H]opioidreceptors in rat brain. Brain Res 279,255 (1979). 122 ' Orr et al. 14. Orr GL, Hollingworth RM: Agonist-induced desensitization of an octopamine receptor. Insect Biochem 20,239 (1990). 15. Weiland GA, Molinoff PB: Quantitative analysis of drug-receptor interactions: I. determination of kinetic and equilibrium properties. Life Sci 29,313 (1981). 16. Breer H:Properties of putative nicotinic and muscarinic cholinergic receptors in the central nervous system of Locusta migratoriu. Neurochem Int 3,43 (1981). 17. Jones SW, Sumikawa K: Quinuclidinyl benzilate binding in house fly heads and rat brain. J Neurochem 36,454 (1981). 18. Lummis SCR, Sattelle DB: Binding of N-[pr~pionyl-~HH]propionylated or-bungarotoxin and L-[benzili~-4,4'-~HH]quinuclidinyl benzilate to cns extracts of the cockroach Peviplaiieta arnericana. Comp Biochem Physiol BOC, 75 (1985). 19. Meyer RM, Reddy G R Muscarinic and nicotinic cholinergic binding sites in the terminal abdominal ganglion of the cricket (Acketa domesticus). J Neurochem 45,1101 (1985). 20. Aguilar JS, Lunt GG: Cholinergicbinding sites with muscarinic properties on membranes from the supraoesophagealganghon of the locust (Schistocercagreguriu). Neurochem Int 6,501 (1984). 21. Haim N, Nahum S, Dudai Y: Properties of a putative muscarinic cholinergic receptor from Drosophila rnelanogaster. J Neurochem 32,543 (1979). 22. Nathanson N: Molecular properties of the muscarinic acetylcholine receptor. Annu Rev Neurosci 10, 195 (1987). 23, Dudai Y: Modulation of a putative muscarinic receptor from Drosophilu melnnognstet.by ions and guanyl nucleotide. Comp Biochem Physiol69C, 387 (1981). 24. Whyte J, Lunt GG: The influence of guanine nucleotides on muscarinic receptor binding in the locust supra-oesophageal ganglion. Biochem Soc Trans 14,690 (1986). 25. McMahon KK, Hosey MM: Agonist interactions with cardiac muscarinic receptors, effects of Mg2', guanine nucleotides and monovalent cations. Mol Pharmacol28,400 (1985). 26. Lucchesi PA, Romano FD, Scheid CR, Yamaguchi H, Honeyman TW: Interaction of agonists and selectiveantagonists with gastric smooth muscle receptors. Naunyn-schmiedeberg's Arch Pharmacol339,145 (1989). 27. Fong HKW, Yoshimoto K, Eversole-Cire P, Simon MI: Identification of a GTP-binding protein a subunit that lacks an apparent ADP-ribosylation site for pertussis toxin. Proc Natl Acad Sci USA 85,3066 (1988). 28. Ashkenazi A, Peralta EG, Winslow JW, Ramachandran J, Capon DJ: Functional diversity of muscarinic receptor subtypes in cellular signal transduction and growth. Trends Pharmacol Sci Suppl, Dec 1989,16 (1989). 29. Trimmer BA, Berridge MJ: Inositol phosphates in the insect nervous system. Insect Biochem 25, 811 (1985). 30. Schurmann F-W: The architecture of the mushroom bodies and related neuropils in the insect brain. In: Arthropod Brain. Gupta AP, ed. John Wiley & Sons, New York, pp 231-264 (1987).