Development of Muscarinic Cholinergic Inhibition of Adenylate Cyclase in Embryonic Chick Heart

Transcription

1 THE JOURNAL OF BIOLOGICAL CHEMISTRY by The American Society of Biological Chemists. Inc. Vol. 261, No. 19, hue of July 5, pp ,13@6 Printed in U.S.A. Development of Muscarinic Cholinergic Inhibition of Adenylate Cyclase in Embryonic Chick Heart ITS RELATIONSHIP TO CHANGES IN THE PROTEIN* INHIBITORY GUANINE NUCLEOTIDE REGULATORY (Received for publication, December 2,1985) Bruce T. Liang, Mark R. HellmichS, Eva J. Neer, and Jonas B. GalperQ From the Cardiovascular Division and Departments of Medicine, Brighnm and Women s Hospital and Harvard Medical School, Boston, Massachusetts Parasympathetic and sympathetic innervation of the embryonic chick heart proceed non-coordinately. 0- Adrenergic agonists mediate an increase in beating rate in embryonic chick heart prior to ingrowth of the vagus nerve (Culver, N. G., and Fishman, D. A. (1977) Am. J. Physiol. 232, R116-R123) while muscarinic agonists exert relatively little effect on beating rate in hearts 2-4 days in ouo (Papanno, A. J. (1979) Pharmacol. Rev. 29, 3-33). Studies of developmental changes in the ability of muscarinic agonists to inhibit adenylate cyclase activity and their relationship to the development of a physiologic response of the embryonic chick heart to muscarinic stimulation have been inconclusive. In the current studies the ability of isoproterenol to stimulate adenylate cyclase activity did not change during development. Maximum stimulation above basal was 760 pmol of camp/10 min/mg of proterin with an IC60 of 1.5 X log M for isoproterenol in homogenates of hearts 2%, 3H, and 10 days in ovo and 3 days posthatching. However, inhibition of isoproterenol-stimulated adenylate cyclase activity by carbamylcholine increased from 7.6% with a ICao for carbamylcholine of 16 f 5.0 p~ at day 2% in ovo to 29% with an IC,, of 0.4 f 0.1 PM at day 10 in ovo and to 43% with a IC,, of 0.6 f 0.1 PM at 3 days posthatching. Since previous data had demonstrated the presence of muscarinic receptors as early as 2% days in ovo (Galper, J. B., Klein, W., and Catterall, W. A. (1977) J. Biol. Chem. 252, ), studies of developmental changes in guanine nucleotide-coupling proteins were carried out to determine whether early in development muscarinic receptors were uncoupled from a physiologic response. Studies of pertussis toxincatalyzed ADP-ribosylation of homogenates of embry- onic chick heart with [SaP]NAD demonstrated the presence of two ADP-ribosylated proteins at 39,000 and 41,000 kda, respectively. Both ADP-ribosylation and immunoblotting of homogenates with an antibody to the 39-kDa guanine nucleotide-binding protein in bo- * This study was supported by a research grant from the National Heart Lung and Blood Institute, by National Institutes of Health Grant HL-22775, and by American Heart Association Grant 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. j:present address: Dept. of Physiology, Boston University School of Medicine, Boston, MA. 8 Established Investigator of the American Heart Association. To whom reprint requests should be sent. vine brain demonstrated that the 39-kDa a protein increased 1.8-fold between days 2% and 3% in ovo and another 1.8-fold from day 3% to 10 in ovo in parallel with the increase in the extent of muscarinic inhibition of adenylate cyclase activity. Between day 10 in ouo and 3 days posthatching the 39-kDa protein remained constant, but muscarinic receptor number measured by [SH]quinuclidinyl benzilate binding increased 2.6-fold from 200 fmol/mg of protein to 520 fmol/mg of protein in parallel with the increase in inhibition of adenylate cyclase activity by carbamylcholine. The data suggest that early in embryonic development coupling of muscarinic receptors to inhibition of adenylate cyclase activity may be limited by the availability of the 39-kDa guanine nucleotide regulatory protein but that later in development physiologic response to muscarinic stimulation might be limited by muscarinic receptor number. The ability of the embryonic chick heart to respond to muscarinic stimuli does not develop coordinately. Prior to ingrowth of the vagus nerve (days 4-5 in ouo) receptors, as demonstrated by the binding of antagonist (-)-[3H]dihydroalpranolol (1, 2), and muscarinic receptors, as measured by the binding of the muscarinic antagonist [3H]QNB (3, 4), are present on the cell membrane of the developing embryonic chick heart cell. However, while a 8-adrenergic agonist-mediated increase in beating rate (5) and adenylate cyclase activity (2) was demonstrated in hearts prior to ingrowth of the vagus nerve, muscarinic agonists exerted relatively little effect on beating rate in hearts 2-4 days in ouo (3,4). These data suggest that muscarinic receptors are not coupled to a physiologic response prior to vagal innervation of the heart. Hence the embryonic chick heart might offer a good model for the study of factors necessary for coupling of muscarinic receptors to a physiologic response. Guanine nucleotides have been shown to be necessary for coupling of 8-adrenergic agonist binding to stimulation of adenylate cyclase (7) and muscarinic agonist binding to inhibition of adenylate cyclase (6, 8) via guanine nucleotide regulatory proteins. This interaction with guanine nucleotides has also been shown to decrease the affinity of the muscarinic The abbreviations are: QNB, quinuclidinyl benzilate; RH, high affinity receptor; RL, low affinity receptor; N., a stimulatory guanine nucleotide regulatory protein complex; Ni, an inhibitory guanine nucleotide regulatory complex; NAD, nicotinamide adenine dinucleotide; IAP, islet-activating protein.

2 9012 Muscarinic Inhibition of Adenylate Cyclase receptor for agonists (9, 10) during conversion of the muscarinic receptor from a high affinity form (RH) to a low affinity form (RL) (11, 12). Recent studies have demonstrated that hormonal stimulation and inhibition of adenylate cyclase are mediated by two distinct guanine nucleotide-binding proteins, N. and Ni, respectively (13). Both N. and Ni are heterotrimers composed of subunits cy. By. The (Y subunits of N, and Ni are structurally and functionally different while the (? and y subunits are very similar (14-16). The (Y subunit of Ni, molecular weight 41,000, is ADP-ribosylated by a toxin isolated from Bordetella pertussis with the loss of hormonal and GTP-dependent inhibition of adenylate cyclase and conversion of the muscarinic receptor to a low affinity form (14, 17, 18). Recently, bovine brain has been shown to contain two proteins, a 39- and a 41- kda protein which are substrates for ADP-ribosylation by pertussis toxin (19, 20). The 41-kDa protein is the species found in most membranes and is felt to be the (Y subunit of Ni, ai (17,18); the role of the 39-kDa proteins in brain referred to as a0 is not well understood. We demonstrated previously that the absence of a physiologic response to muscarinic stimulation in cultures of hearts studied prior to ingrowth of the vagus nerve was associated with the inability of guanine nucleotides to convert high affinity muscarinic receptors (RH) to a low affinity form (RL). The appearance of a physiologic response during incubation of these cells with specific lots of horse serum was associated with the development of guanine nucleotide sensitivity of RH (21). In the current studies of embryonic chick heart homogenates, we demonstrate that prior to ingrowth of the vagus nerve, another GTP-dependent function, muscarinic inhibition of adenylate cyclase activity, is also markedly depressed. One interpretation of these data is that prior to vagal inner- vation of the heart, a guanine nucleotide-coupling protein might be absent or present in an inactive form. Data are presented which support the hypothesis that the appearance of a physiologic response to muscarinic stimulation during maturation of the embryonic chick heart is associated with an increase in the levels of an a 39-kDa guanine nucleotidebinding protein in the heart. at each embryonic age at which muscarinic inhibition of adenylate cyclase activity was to be studied; 2%, 3%, and 10 days in ouo and 3 days posthatching. In the experiments summarized in Fig. 1, the dose-response curve for isoproterenol-stimulated adenylate cyclase activity is compared for hearts of various embryonic ages. There is no statistically significant difference between the extent of stimulation by isoproterenol (750 pmol of cyclic AMP/mg/lO min), the concentration of isoproterenol required to elicit a maximal stimulation M), and the concentration of isoproterenol at which stimulation is half-maximal (1.5 X M) for all embryonic ages studied. Although data for hatched chick hearts appear to fall below the data for embryonic hearts, the scatter in the hatched chick data is more marked and hence this difference is not statistically significant. Hence between days 2% in ouo and hatching, no developmental change in the ability of (?-adrenergic agonists to stimulate adenylate cyclase activity can be detected by these methods. Inhibition of Adenylate Cyclase Activity by Muscarinic Agonists in Homogenates of Embryonic Chick Hearts-In order to study muscarinic cholinergic inhibition of adenylate cyclase in embryonic chick heart, it was first necessary to establish conditions for the assay of muscarinic inhibition of adenylate cyclase in chick heart homogenates. In these studies adenylate cyclase activity is separated into 3 types: 1) activity in the absence of added GTP (referred to as basal activity); 2) activity in the presence of M GTP; 3) activity in the presence of M GTP plus isoproterenol. In homogenates of chick hearts 10 days in ouo 10 PM isoproterenol caused a 1.6-fold increase (220 pmol/mg of protein/lo min) in adenylate cyclase activity over basal levels (Table I, column 2). Addition of M carbamylcholine decreased isoproterenolstimulated adenylate cyclase activity by 115 pmol/mg/lo min or 52% (Table I, column 2). Inhibition was blocked by M atropine consistent with a specific muscarinic effect. The dependence of muscarinic inhibition of adenylate cyclase activity on the concentration of carbamylcholine in homogenates of chick hearts 10 days in ouo is depicted in the experiment summarized in Fig. 2. A maximal inhibition of MATERIALS AND METHODS AND RESULTS~ Developmental Changes in Isoproterenol Stimulation of Adenylate Cycluse Activity-In order to study developmental changes in muscarinic inhibition of isoproterenol-stimulated adenylate cyclase activity at various embryonic ages, it was first necessary to determine whether p-adrenergic stimulation of adenylate cyclase was altered during embryonic development of the chick heart. Although it had previously been demonstrated that the maximal response of adenylate cyclase to isoproterenol stimulation did not change between days 2 and 12 in ouo and that the K, for isoprenaline stimulation of adenylate cyclase activity was the same at days 5 and 9 in ouo (2), in the current studies of muscarinic inhibition of adenylate cyclase, we chose a wider range of embryonic ages. For this reason it was necessary to determine whether the maximal response and the concentration dependence for isopro- Isoproterenol Concentration. M terenol stimulation of adenylate cyclase activity was the same FIG. 1. Effect of varying concentrations of isoproterenol on adenylate cyclase activity at various developmental stages of Portions of this paper (including Materials and Methods, part embryonic chick heart. Assays of adenylate cyclase activity were of Results, and Fig. 3) are presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are available from the Journal of Biologperformed at 37 C with homogenates of chick hearts of the indicated embryonic ages as described under Materials and Methods. Final concentrations of GTP and NaCl were 100 pm and 80 mm, respecical Chemistry, 9650 Rockville Pike, Bethesda, MD Request tively. Ordinate is the increase in adenylate cyclase activity above Document No. 85M-3912, cite the authors, and include a check or levels in the presence of GTP alone. Error bars were omitted to avoid money order for $2.40 per set of photocopies. Full size photocopies overcrowding the figure. 0,2?h days in ouo; A, 3% days in ouo; 0, 10 are alsoincludedin the microfilm edition of the Journal that is days in ouo; V, 3 days posthatching. Each point is the mean of four available from Waverly Press. determinations carried out in triplicate.

3 TABLE I Inhibition of adenylate cyclase by carbamylchline in hmgenutes of embryonic chick hearts Five hearts from embryos 10 days in ouo werehomogenized in buffer as described under Materials and Methods. 5O-pl aliquots containing approximately 25 pg of protein were assayed for adenylate cyclase activity in the presence of the protease inhibitors, alamethecin, and the indicated concentrations of GTP, isoproterenol, carbamvlcholine. or atrouine as described under Materials and Methods. Muscarinic Inhibition of Adenylate Cyclase 9013 Adenylate cyclase activitf (I)* (2) -Na +Na (80 mm) pmollmg~l0 min k S.E. fn = 7) GTP (10-4 M) 340 f k 16 GTP plus isoproterenol (lo4 M) 590 f f 11 GTP plus isoproterenol plus car- 550 f f 15 bamylcholine ( M) plus atropine (io+ M) f 12 Inhibition by carbamylcholine The differences between successive pairs of numbers in column 2 were significant at the p = level. Incubation buffer 80 mm in glucamine added to maintain ionic strength. Carbamylcholine, M FIG. 2. The effect of various concentrations of carbamylcholine on the inhibition of isoproterenol-stimulated adenylate cyclase activity. Assays of adenylatecyclaseactivity were performed with homogenates of chick hearts from embryos 10 days in ouo as described under Materials and Methods in the presence of 10 PM isoproterenol, 100 p~ GTP, and 80 mm NaCland the indicated concentrations of carbamylcholine. Data were plotted as pmol/mg of protein/lo min of inhibition of isoproterenol-stimulated adenylate cyclase activity. Each point is the mean of 3 replicate determinations and is typical of 5 separate determinations. 300 pmol/mg/lo min was demonstrated at a carbamylcholine concentration of M. Inhibition was half-maximal at 0.48 & 0.11 PM. For further characterization of muscarinic inhibition of adenylate cyclase activity see the Miniprint. Developmental Changes in Muscarinic Cholinergic Inhibition of Basal and GTP-stimulated Adenylate Cyclase Actiuity- In order to study developmental changes in muscarinic cholinergic inhibition of adenylate cyclase activity, we compared the ability of various concentrations of carbamylcholine to inhibit basal, GTP-stimulated, and isoproterenol-stimulated adenylate cyclase activity in homogenates of hearts from embryos 2%, 3%, and 10 days in ouo and 3 days posthatching. Initially we studied developmental changes in basal adenylate cyclase activity. As described under Methods, hearts of each embryonic age were homogenized and assayed directly for adenylate cyclase activity in the presence of alamethicin and protease inhibitors. The addition of alamethicin increased basal adenylate cyclase activity and the reproducibility of these measurements. Although no statistically significant difference in basal adenylate cyclase activity could be demonstrated between days 2% and 10 in ouo, basal activity increased 4.7-fold from 317 f 18 (n = 3) pmol/mg/lo min to 1690 f 95 (n = 3) pmol/ mg/lo between day 10 and hatching (Table 11, column 1). Inhibition of basal adenylate cyclase activity by carbamylcholine increased with embryonic age from 6 pmol/mg/lo min in hearts 3% days in ouo to 376 pmol/mg/lo min in hatched chicks (Table 11, column 2). Although inhibition expressed as a percentage of basal activity increased more than 9-fold during development, maximum inhibition of basal activity was only 22% in hatched chicks. Since carbamylcholine inhibition of adenylate cyclase is GTP dependent, one explanation for the increase in carbamylcholine inhibition of basal adenylate cyclase activity during embryonic development is that crude homogenates of hearts of different embryonic ages contain different levels of endogenous GTP. To rule out this explanation, we compared the inhibition of adenylate cyclase at various embryonic ages in the presence of a large excess M) of GTP. In the presence of alamethicin, GTP had no effect on basal adenylate cyclase activity (Table 11, compare columns 1 and 3), but the percentage of muscarinic inhibition of adenylate cyclase increased with embryonic age in a manner similar to that for the development of inhibition of adenylate cyclase in the absence of GTP (Table 11, compare columns 2 and 4). Inhibition increased from 20 f 6 (n = 8) pmol/mg/lo min or 7% at 2% days in ouo to 33 f 2 (n = 25) pmol/mg/lo min or 11% at 3% days in ouo to 66 f 16 (n = 20) pmol/mg/lo min or 20% in hearts 10 days in ouo to 1000 f 16 (n = 10) pmol/mg/ 10 min or 45% in hearts 3 days posthatching. Developmental Changes in Muscarinic Inhibition of Isopro- terenol-stimulated Adenylate Cyclase Activity-In order to determine whether muscarinic cholinergic inhibition of isoproterenol-stimulated adenylate cyclase activity also increased with embryonic age, we compared the ability of various concentrations of carbamylcholine to inhibit isoproterenol-stimulated adenylate cyclase activity in homogenates of hearts from embryos 2Y2, 3%, and 10 days in ouo and 3 days posthatching. The data summarized in Table 11, columns 5 and 6, and plotted in Fig. 4 indicate that prior to ingrowth of the vagus nerve (days 2%-3% in ouo) adenylate cyclase activity in embryonic chick hearts is relatively insensitive to inhibition by carbamylcholine. However, maximal inhibition of isoproterenol-stimulated adenylate cyclase activity and sensitivity to carbamylcholine increased continuously even after ingrowth of the vagus nerve. Hence, maximal inhibition increased from 80 f 16 (n = 7, S.E.) pmol/lo min/mg of protein or 7.6% of the isoproterenol-stimulated level at 2% days in ouo to 120 f 12 (n = 7) pmol/lo min/mg or 11% of the isoproterenol-stimulated level at day 3% to 320 f 24 (n = 8) pmol/lo min/mg or 29% of the isoproterenol-stimulated level at day 10 in ouo to 1040 f 28 pmol/lo min/mg (n = 5) or 46% of the isoproterenol-stimulated level at 3 days posthatching. The total increase in the extent of inhibition during development was 13-fold. Furthermore, the sensitivity to carbamylcholine increased with embryonic age (Fig. 4). The concentration required to mediate a half-maximal inhibition of adenylate cyclase activity (ICSO) in the presence of 10 PM isoproterenol in homogenates of hearts 2% days in ouo was 16.5 f 5 FM (S.E., n = 7) carbamylcholine compared to 0.6 & 0.1 PM (n = 7) carbamylcholine in homogenates of hearts 3 days posthatching (Table 111). This represents a 26-fold increase in sensitivity of adenylate cyclase activity to inhibition

4 9014 Muscarinic Inhibition of Adenylate Cyclase TABLE I1 Age dependence for carbamylcholine inhibition of basal, GTP-stimulated, and isoproterenol-stimulated adenylate cyclase activity Assays carried out as described in Table I and under "Materials and Methods." Concentrations of carbamylcholine were M; GTP, 10" M; and isoproterenol, M. GTP was present in the isoproterenol-stimulated assays. Basal" GTP Isoproterenol' Age (1)' (2)' (3)' (4)' (5)' (6)' Activity" Inhibitiond % Activity" Inhibitiond % Activity"a Inhibitiond days 299 in ouo 2% 295 (7) (7.6) 3% (2) (n = 8) (11) 1046 (11) (6) (n = 25) (20) (29) (n = 20) Hatched (22) 1700 (56) (43) (n = 11) Data for basal adenylate cyclase activity and carbamylcholine inhibition of adenylate cyclase represent the mean of 5 determinations carried out in triplicate. Basal adenylate cyclase activity for 2%, 3%, and 10 days in OLIO was not significantly different while the difference between inhibition of basal adenylate cyclase was significantly different between any two ages by Student's t test. Minus carbamylcholine. Plus carbamylcholine. dactivity is expressed as pmol of cyclic AMP formed/lo min/mg of protein. Inhibition is calculated as the difference between adenylate cyclase activity in the presence and absence of carbamylcholine. Per cent inhibition is the ratio of inhibition to total activity X 100. For all embryonic ages, standard deviation did not exceed 10%. For hatched chick hearts, standard deviations did not exceed 15%. e Data derived from experiments similar to those for determination of basal adenylate cyclase activity with the addition of lo-' M GTP. 'Data derived from experiments shown in Fig. 4. #Data derived from maximal adenylate cvclase activity at each age (basal plus the isoproterenol stimulated values in Fig. 1). % by carbamylcholine during development. The inhibition of adenylate cyclase by muscarinic agonist was totally abolished by lo-' M atropine at all ages studied. One explanation of these data is that prior to ingrowth of the vagus nerve the affinity of the inhibitory system for GTP, Na+, or Mg2+ is decreased. However, we demonstrated that increasing GTP levels up to M and/or increasing Na' concentration to 100 or 150 mm or increasing M$+ levels from 10 to 24 mm had no effect on the extent of inhibition of adenylate cyclase by carbamylcholine in hearts from embryos 3% days in ouo. Hence, stimulation of adenylate cyclase is relatively stable during embryonic development (Fig. l), muscarinic inhibition of basal GTP-stimulated and 8-adrenergic stimulated adenylate cyclase activity develops continuously throughout embryonic life. Developmental Changes in Guanine Nucleotide-coupling Proteins Demonstrated by ADP-ribosylatwn by Pertussis Toxin of a 39- and 41-kDa Protein in Chick Heart Homogenates-Data presented here and in previous studies (21) suggest that prior to ingrowth of the vagus nerve, two processes which require the interaction of guanine nucleotides with Ni, 1) inhibition of adenylate cyclase and 2) guanine nucleotidemediated conversion of high affinity muscarinic receptors to a low affinity form (21), either cannot be demonstrated at all or are markedly attenuated. One explanation of these findings is that early in embryonic development, Ni is either absent or present in an inactive form. In order to study developmental changes in Ni crude membrane preparations from embryonic chick hearts of various embryonic ages were incubated with islet-activating protein in the presence of 32P NAD+ and high levels of protease inhibitors. Following polyacrylamide gel electrophoresis of membranes solubilized in SDS, autoradiography of the dried gels demonstrated two specifically labeled bands, 41 and 39 kda (Fig. 5) at each embryonic age. Densitometry scanning of the autoradiographs demonstrated that the intensity of the 39-kDa band increased 1.7-fold from day 2% to 3% in ouo, 1.5-fold from day 3% to 10 in ovo, and was unchanged from day 10 until hatching (Figs. 5b and 6). The 41-kDa band increased 1.5-fold from day 2% to 3% in ouo, was unchanged from day 3% to 10 in ouo, and may have decreased slightly from day 10 until hatching (Fig. 5b). One explanation for these differences in ADP-ribosylation might be that at various embryonic ages, differences in ribosylation reflect differences in accessibility of pertussis toxin and/or NAD' to the substrate. To rule out this possibility, experiments similar to those shown in Fig. 5 were carried out with membranes incubated for 5 min with low concentrations of cholate ( %) prior to incubation with IAP. These experiments demonstrated no effect of detergent on the relative labeling of the 39,000- and 41,000-kDa proteins seen in Fig. 5 and suggested that accessibility of the substrate is not responsible for differences in ADP-ribosylation. A second possible explanation is that the differences be- tween 32P labeling of the two peaks might be due to differences in the relative rates of ribosylation of these two proteins. To rule out this possibility the time course for the ADP-ribosylation of the 39- and 41-kDa bands from hearts 3% days in ouo and in hatched chick heart membranes was determined. Ribosylation was maximal for both bands and at both ages by 15 min and did not change for as long as 90 min. Although these data strongly support developmental increases in the 39-kDa IAP substrate, ADP-ribosylation of this protein by IAP is an indirect method of quantitation and may be limited as a tool for quantitation because of incomplete ribosylation. For example, accessibility may be limited even in the presence of detergent, or protease activity may interfere

5 Muscarinic Inhibition of Adenylate Cyclase 9015 i chick Corbornylcholine, M FIG. 4. Effect of various concentrations of carbamylcholine on the inhibition of isoproterenol-stimulated adenylate cyclase activity at various developmental stages of embryonic chick heart. Assays of adenylate cyclase activity were carried out on homogenate5 of chick hearts from embryos of the ages indicated in the presence of 100 PM GTP, 80 mm Na+, 10 p~ isoproterenol, and the indicated concentrations of carbamylcholine. Data were plotted as inhibition of isoproterenol-stimulated adenylate cyclase activity (control isoproterenol-stimulated adenylate cyclase activity minus isoproterenol-stimulated adenylate cyclase activity in the presence of carbamylcholine). The data were the mean of n independent experiments; n = 3,3,5, and 3 for 2% days in ouo, 3% days in ouo, 10 days in ouo, and hatched chick, respectively. The levels of inhibition of adenylate cyclase by carbamylcholine were significantly different between any two ages by Student's t test. Those values for hatched chick without error bars represent single determinations. V, 2% days in ouo; 0, 3% days in ouo; A, 10 days in OVO; 0, 3 days posthatched chick. TABLE I11 Age dependence of the IC, for carbamylchaline inhibition of isoproterenol-stimulnted adenylate cyclase activity pm S.E. 2% 16 f 5 (n = 3)b 3% 2.9 f 0.9 (n = 5) f 0.1 (n = 7) Posthatched 0.6 f 0.1 (n = 6) Data derived from experiments similar to those shown in Fig. 4. A11 values for IC, were significantly different ( p = 0.01) except the difference between day 10 and posthatched with the pertussis toxin substrate activity of Furthermore, ADP-ribosylation of these proteins has been shown to be dependent on the presence of the j3 subunit of Ni (19). Hence differences in ADP-ribosylation with embryonic age may reflect developmental differences in j3 levels. For these reasons determination of levels of j3 and the 39-kDa IAP substrate by immunoblotting constitute a more accurate method for quantitating the levels of these proteins in the developing heart. Immunoblotting of the j3 and a 39 Subunits of Ni at Varwus Embryonic Ages-Antibodies with a high degree of specificity and a39 from bovine brain have been generated and carefully characterized in our laboratory (26). No antibody to all is presently available. In order to determine developmental changes and as9, membranes from hearts 2%, 3Y2, and 10 days in ouo and 3 days posthatching were subjected to polyacrylamide gel electrophoresis, electrotransferred from the gel to a nitrocellulose filter, incubated either with anti-@ or anti- ( Y antibody ~ ~ followed by lz5i protein A as described under "Materials and Methods" and the specific binding of '=I protein A determined by autoradiography. Coomassie Blue staining of nitrocellulose filters demonstrated that the electrotransfer was complete (data not shown). The antibody to bovine j3 cross-reacts with the chick heart proteins as demonstrated by the comigration of bands from chick heart homogenates with j3 protein purified from bovine brain (Fig. 7a). Comparison of the relative intensity of j3 bands in autoradiographs of immunoblots of proteins from chick hearts of various embryonic ages reveals that levels of /3 were unchanged between day 2% to 3% in ouo, increased 2- old between day 3% and 10 in ouo, but were unchanged between day 10 and 3 days posthatching (Fig. 7, a and b). The absence of a change in j3 levels between days 2% and 3% in ouo when the extent of ADP-ribosylation of a39 and (Y41 increased suggests that a developmental increase could not account for the increase in ADP-ribosylation of either a39 or ff41 between day 2% to 3% in ouo. Furthermore, if levels of j3 were limiting during development, ADP-ribosylation of ff41 might be greater or increase more rapidly during development than ( Y since ~ ~ ff41 from bovine brain has a somewhat higher affinity for /3 than as9 (19). However, these data do not rule out the possibility that the increase in j3 between days 3% and 10 in ouo could contribute at least in part to the increased ADP- ribosylation of a39 between these ages. Measurement of levels of a39 in embryonic chick heart with antibody to bovine brain ( ~ demonstrated 3 ~ cross-reactivity between from these species (Fig. 8, a and b). Studies of developmental changes in a39 levels measured by immunoblotting with anti-a39 demonstrated that a39 increased 2-fold between day 2% and 3% in ouo, increased 2-fold further between day 3% and day 10 in ouo, but was unchanged between day 10 in ouo and 3 days posthatching. Comparison of the quantitation of relative levels of a39 by ADP-ribosylation and immunoblotting (Fig. 9) demonstrated that both methods gave remarkably similar results. By both methods, cy3$ increased a total of fold between day 2% in ouo and day 10 in ouo with a half-maximal increase at day 5 in ouo. Differences between a39 determined by the two methods are significant only at day 2% in ouo. These data support the conclusion that relative levels of a39 measured by ADP-ribosylation accurately reflect developmental changes in asg. Furthermore, since the availability of,8 and the relative accessibility of ff41 and a39 to IAP for ADP-ribosylation appear not to play a role in the extent of ADP-ribosylation, the ADPribosylation of ff41 may give a good estimation of the relative level of (~41. Developmental Changes in the Number of Muscarinic Re- ceptors-in order to determine whether the increase in muscarinic inhibition of adenylate cyclase activity seen during embryonic development reflected changes in the number of muscarinic receptors, we studied levels of muscarinic receptors as a function of embryonic age. Although we had previously demonstrated that muscarinic receptor number as meas- ured by the binding of [3H]QNB was unchanged between day 2% to day 18 in ouo (41, studies of receptor number in embryos older than 18 days in ouo had not been carried out. Furthermore, over the last several years, several factors, including

6 9016 Muscarinic Inhibition of Adenylate Cyclase a kd 0139 I I 1 I I I MIGRATION (mm) FIG. 5. a, autoradiography following polyacrylamide gel electrophoresis of embryonic chick heart proteins ADPribosylated with pertussis toxin in the presence of [32P]NAD+. Homogenates were prepared and incubated for 30 min at 37 "C with pertussis toxin and t3'p]nad+, solubilized, and subjected to polyacrylamide gel electrophoresis and autoradiography as described under "Materials and Methods." Each lane contains 100 pgof protein as determined by the method of Lowry (33). Molecular weights were determined by plotting the distance migrated in millimeters for standard proteins stained with Coomassie Blue versus the logarithm of their molecular weights. The standards included trypsinogen, carbonic anhydrase, glyceraldehyde-3-phosphate dehydrogenase, egg albumin, bovine albumin, molecular weights 24,000, 29,000, 36,000, 45,000, and 66,000, respectively. Calculated molecular weights for the two bands were 39 & 0.4 kda and 41 k 0.3 LDa (S.E., n = 9). The first lane contains a39 kda standard purified from bovine brain; lanes 1, 3, 5, and 7 contain homogenates of hearts 2% days in ovo, 3% days in ouo, 10 days in ouo, and 3 days posthatching, respectively. Even-numbered lanes contain the same homogenates as in the preceding lane incubated in the absence of islet-activating protein. b, densitometry tracings of autoradi- ographs of ( Y kda ~ ~ and ai1 kda from a. Autoradiographs were scanned on an LKB laser densitometer and the base line for scans at each embryonic age superimposed. Data is typical of scans of 10 different determinations. -, control base line; - - -, 2% days in ovo;...., 3% days in ouo; - - -, 10 days in ovo; - 3 days posthatched. kd, kilodaltons.

7 I FIG. 6. Histogram of relative levels of am determined from densitometry tracing of autoradiograms. Peaks from densitometry tracings from n experiments similar to those shown in Fig. 5b were traced individually and the overlapping edge of each peak approximated by symmetry with the free edge. Peaks were cut out and the relative weight at each age determined. Data were normalized to the value at 2% days in ouo and the highest value plotted here as 100%. Values shown are the means of n determinations. n = 11 at day 2% in ouo, n = 9 at day 3% in ouo, n = 9 at day 10 in ouo, n = 3 at 3 days posthatching. Error bars represent? S.E. decreased gestational time of chick embryos, have suggested that the maturation processes in this highly inbred line of chickens might be taking place earlier than previously described. For these reasons we studied [3H]QNB binding between day 2% in ouo and 3 days posthatching in the present flock of chicks. Muscarinic receptor number was constant at 200 fmol/mg of protein between days 2% and 10 in ouo. However, unlike our previous findings (4), receptor number increased 2.6-fold between day 10 and 3 days posthatching: from 200 f 10 (n = 5) fmol/mg of protein to 270 f 15 (n = 3) fmol/mg of protein at day 14 in ouo, to 315 f 15 (n = 3) fmol/mg of protein at day 18 in ouo, and to 520 f 30 (n = 3) fmol/mg of protein on the third day after hatching (see normalized data, Fig. 10). All these determinations were significantly different from each other at the p < 0.01 level. Correlations between Changes in the Ability of Muscarinic Agonists to Inhibit Adenylate Cyclase Activity and Changes in a39 and Receptor Number-In order to determine whether a relationship existed between developmental changes in the number of muscarinic receptors, levels of a39, and the appearance of increased sensitivity of adenylate cyclase to inhibition by muscarinic agonists, measurements of each of these parameters were normalized to its value at 2% days in ouo (Fig. 10). This allowed data for each measurement to be expressed as the extent of increase since day 2Y2 in ouo. Hence the inhibition of adenylate cyclase increased 2.7-fold between day 2% and day 10 in ouo while muscarinic receptor number was constant. However, during that same time period, a39 increased 2.4-fold with a time course which closely paralleled the increase in muscarinic inhibition of adenylate cyclase activity. Between day 10 in ouo and 3 days posthatching, inhibition of adenylate cyclase increased 2.2-fold and musca- Muscarinic Inhibition of Adenylate Cyclase 9017 rinic receptor number increased 2.6-fold with a time course which closely paralleled the increase in inhibition of adenylate cyclase activity. During that time period, the levelof a39 remained constant. These data suggest that early in embryonic development muscarinic inhibition of adenylate cyclase activity increases in parallel with the levels of a39, while between day 10 in ouo and 3 days posthatching inhibition of adenylate cyclase activity increases in parallel with receptor number. One explanation of these data is that early in embryonic development, muscarinic receptors are present but uncoupled from the muscarinic response because levels of a39 in the cell are limiting. Subsequently a39 increases to levels sufficient to couple existing receptors to a physiologic response, and the number of muscarinic receptors appears to limit the physiologic response. Since a41 increases only 1.5-fold between day 2% and 3% in ouo and remains unchanged or even decreases slightly from day 3% in ouo until 3 days posthatching (see x on Fig. lo), it appears not to parallel the change in physiologic responsiveness. Hence these data suggest that a39 may be the form of a which couples the muscarinic receptor to inhibition of adenylate cyclase in the embryonic chick heart. DISCUSSION Adenylate cyclase in hearts 2%-31/~ days in ouo demonstrates a markedly decreased response to muscarinic inhibition compared to hearts 10 days in ouo or 3 days posthatching. This relative unresponsiveness could be due to 1) the absence of muscarinic receptors in these hearts, 2) the absence of a specific subtype of muscarinic receptors necessary for physiologic activity, 3) the absence of an essential subunit of the muscarinic receptor or post-translational modification of the muscarinic receptor necessary for interaction with Ni, 4) the absence or abnormality of a subunit of Ni either (a) alr the guanine nucleotide-binding protein which couples formation of an agonist-receptor complex to the inhibition of adenylate cyclase activity or (b)/3, which appears to dissociate from Ni to mediate inhibition of adenylate cyclase, or 5) decreased sensitivity of the inhibitory process to GTP, Na+, or M$+. Several of these possible explanations can be easily eliminated. Although muscarinic receptors are present in the embryonic chick heart as early as 2% days in ouo, the number of receptors does not increase between day 2% and 10 in ouo when adenylate cyclase inhibition by muscarinic agonists increases 2.6-fold. Hence at least early in embryonic development, a change in the number of muscarinic receptors in the heart does not explain the increased responsiveness of adenylate cyclase in these cells to muscarinic agonists. Data presented in this study also demonstrated that increasing GTP levels up to M and/or increasing Na+ concentration to 100 or 150 mm or increasing Mg+ levels from 10 to 24 mm had no effect on the extent of inhibition of adenylate cyclase by carbamylcholine in hearts from embryos 3% days in ouo. These findings rule out a decreased sensitivity of the inhibitory process to GTP, Na+, or Mg+ as the cause of the decreased inhibitory response of adenylate cyclase activity to muscarinic agonists. Our previously reported data demonstrate that cells cultured from embryonic chick hearts prior to ingrowth of the vagus nerve remained unresponsive to muscarinic stimulation for up to 9 days in culture (21). In these cultures M carbamylcholine had no effect on beating rate and only a small effect on K+ permeability as measured by the rate of 42 K + efflux. Guanine nucleotides did not mediate the conversion of high affinity receptors (RH) to a low affinity form (RL) in homogenates of these cultures (21). Following growth for 3

8 9018 Muscarinic Inhibition of Adenylate Cyclase FIG. 7. a, immunoblotting assay at various embryonic ages. Proteins were prepared, solubilized, and polyacrylamide gel electrophoresis, immunoblotting, and autoradiography carried out as described under Materials and Methods. Equal amounts of protein, 110 pg, were loaded into each lane. Lane 1, immunoblot of purified p from bovine brain; lane 2, immunoblot of samples from hearts 2% days in ouo; lane 3, 3% days in ouo; lane 4, 10 days in ouo; and lane 5, 3 days posthatching. This experiment is typical of 5 similar determinations. b, histogram of relative levels of binding of lz5i protein A to immunoblots of p. The area of the nitrocellulose which corresponded to the bands seen in the autoradiographs in a was cut out and 1251 determined in a y- counter by the method of Shrenkoff. All values were normalized to the value at 2% days in ouo and plotted as a percentage of the value in the hatched chick taken as 100%. Values are the mean of n determinations; errors are S.E. IO H b I 80- r 3 Z 70- X Y- o 50- o\ IOtl - E H n=5 n=5 n=5 n=5 b FIG. 8. a, immunoblot assay of a39. Proteins were prepared, solubilized, and polyacrylamide gel electrophoresis and immunoblotting carried out as described under Materials and Methods. Equal amounts of protein (100 pg) were loaded onto each lane. Lane 1, immunoblot of purified a39 from bovine brain; lane 2, immunoblot of samples from hearts 2% days in ouo; lane 3, 3% days in ouo; lane 4, 10 days in ouo; and lane 5, 3 days posthatched. Data are typical of 10 experiments. b, histogram of relative levels of 1251 protein A binding to immunoblots of 01~~. The areas of the nitrocellulose blot which corresponded to the bands on the autoradiographs in a were cut out and lz5i determined in a y-counter by the Shrenkoff method. All values were normalized to the value at day 2% in ouo and the value in the hatched chick taken as 100%. Data represent the mean of n determinations. Errors are the S.E. 2; 3+ IO t r_ n=4 n=5 n=5 - H n=4 days in media supplemented by specific lots of horse serum, nucleotide to convert RH to RL. These data suggested that these cells developed a K permeability response to musca- high affinity receptors are necessary for a physiologic response rinic agonists. This development of a physiologic response to muscarinic stimulation and that inadequate numbers of was associated with a l%-fold increase in the number of high high affinity receptors may be present in hearts 2% to 3% affinity receptors and the appearance of the ability of guanine days in ovo (21).

9 0 t P s a 2- I " IN OVO Muscarinic Inhibition of Adenylate Cyclase 9019 appears to rule out low levels as the cause of a decreased inhibitory effect of muscarinic agonists on adenylate cyclase early in embryonic development. The level of a41 does increase during embryonic development (Fig. 5, a and b), and data presented here do not rule out a role of increased levels of a41 in coupling muscarinic receptors to inhibition of adenylate cyclase. However, of all these parameters, changes in the level of ( Y appear ~ ~ to correlate best with the increase in muscarinic inhibition of adenylate cyclase seen between days 2% and 10 AGE (day*) in ouo. FIG. 9. Comparison of a39 determined by ADP-ribosylation The finding that CY might be limiting early in embryonic with pertussis toxin and immunoblotting. Data from Figs. 6 and development is intriguing especially in view of the finding 86 are normalized to the values at 2% days in ouo. 0, ADP-ribosyla- that inhibitory guanine nucleotide regulatory proteins appear tion; A, immunoblotting. to be present in high concentrations in tissues such as brain (26). Determination of the relative numbers of a39 and muscarinic receptors would allow direct demonstration of whether the number of receptors present actually exceeds the number of CY^^ molecules between days 2% and 10 in ouo. However, such a quantitation is not possible because the extent of crossreactivity of the antibody to bovine brain CY^^ with the chick P i 1- "HATCHED' AGE (days) FIG. 10. Comparison between changes in muscarinic inhibition of adenylate cyclase, muscarinic receptor number, and a39 during embryonic development of the chick heart. Physiologic responsiveness, muscarinic receptor number, and levels of a39 and ai1 at each embryonic age were normalized by setting the value for each parameter at 2% days in DUO at one and determining the ratio of values at each embryonic age to the value at 2% days in ouo. The ratios for carbamylcholine inhibition of adenylate cyclase were determined from the data in Table I1 by dividing per cent inhibition (column 6) by 7.6. The ratio for inhibition of adenylate cyclase activity in hearts 16 days in ouo represents the ratio of the mean of 3 separate determinations carried out in triplicate. Ratios for receptor number were determined by dividing the number of [3H]QNB-binding sites at each age by 200 fmol/mg of protein, the value at 2% days in ouo. Ratios for as9 and a41 were determined by dividing the relative area under the curve in Fig. 5 for a41 at each embryonic age by the value for ai1 at 2% days in ouo and the value for at each embryonic age by the value for a39 at 2% days in ouo. A, inhibition of cyclase; 0, [3H]QNB-binding sites; 0, a39; X, arl. Work by Kurose et al. (17) suggests that ADP-ribosylation ofni decreases the affinity of the muscarinic receptor for agonist while a more recent study suggests that reconstitution of solubilized muscarinic receptors with ~~41s~ or a3987 results in an increase in affinity of the receptor for agonist and the development of sensitivity of the receptor to guanine nucleotides (31, 32). Although we cannot rule out the absence of a subunit of the muscarinic receptor or a post-translational modification of the receptor which prevents coupling of the receptor to a guanine nucleotide regulatory protein, the presence of low levels of RH and the absence of a response to guanine nucleotides in heart cells cultured from hearts 3% days in ouo could reflect the presence of insufficient levels of Ni in heart cells early in embryonic development (21). The finding that developmental changes in the amount of the /3 subunit of Ni present in these hearts do not correspond to increases in the extent of inhibition of adenylate cyclase heart a39 is not known. Furthermore, although ADP-ribosylation of a39 gave relative levels of a39 in close agreement with those found by immunoblotting, ADP-ribosylation may not accurately quantitate a39 since the reaction may not go to completion; hence both methods could underestimate (~3~. Recently Halvorsen and Nathanson (33) and Malbun et al. (34) have presented data which suggest that a39 and a41 can both be detected in embryonic chick heart and rat heart, respectively. Unlike the data presented here, Halvorsen and Nathanson demonstrated that a39 appears to be absent as late as day 4-5 in ouo and appears at day 6 in ouo. Data on changes in ( ~ after 3 ~ day 8 in ouo are not presented. Furthermore, these authors were unable to demonstrate an increase in muscarinic inhibition of adenylate cyclase activity during that phase of embryonic development which they studied. The close parallel between the increase in the number of muscarinic receptors and the increase in inhibition of adenylate cyclase between day 10 in ouo and 3 days posthatching suggests that there are few "spare" receptors at this point in development. This finding is in agreement with our prior observation (35) that the decrease in muscarinic receptors as measured by the binding of [3H]methylscopolamine following prolonged exposure to agonist closely paralleled the decrease in the ability of agonists to inhibit beating and to increase K+ permeability, suggesting that the number of receptors was critical for the mediation of a physiologic response. A review of data in Fig. 4 and Table I11 reveals that although the extent of muscarinic inhibition of adenylate cyclase activity increases continuously throughout development, the IC, for carbamylcholine which decreases 26-fold between day 2% and day 10 in ouo is constant between day 10 in ouo and hatching. One possible explanation for these findings is that increased levels of CY^^ cause both an increase in coupling of the receptor to inhibition of adenylate cyclase and an increase in the affinity of the receptor for agonist. This might increase both extent of inhibition and IC, for inhibition. An increase in total receptor number after day 10 in ouo in the presence of levels of a39 adequate to couple these new receptors to inhibition of adenylate cyclase might increase only the extent of inhibition. Recently, Hosey et al. (36) presented data which suggested that between day 18 in ouo and hatching, [3H]QNB binding increased. Although these authors saw an increase in the extent of oxotremorine inhibition of basal adenylate cyclase activity from 17 to 26% between day 18 in ouo and hatching, they saw no increase in inhibition of isoproterenol-stimulated

10 ~~.. ~~ 9020 Muscarinic Inhibition of Adenylate Cyclase adenylate cyclase activity between day 13 in ouo and hatching. The studies presented here demonstrate a gradual increase of muscarinic receptor number from day 12 in ouo until 3 days after hatching and parallel a increase in muscarinic inhibition of both basal and isoproterenol-stimulated adenylate cyclase activity. The finding that muscarinic inhibition stimulated adenylate cyclase activity is a non-competitive process rules out the possibility that the ligands compete directly at the receptor or that the two guanine nucleotide regulatory proteins compete for the same site on the catalytic unit of adenylate cyclase. Hence they exert their effects on adenylate cyclase activity either by independent actions directly on the catalytic subunit of adenylate cyclase or by an interaction between N. and Ni, or both. These observations are in close agreement with the model presented by Katada et al. (37) which suggests that agonist binding to an inhibitory receptor in the presence of GTP results in the release of /3 from Ni which can then associate with N, to deactivate it. The functions of a39 and a41 have not been well established. However, in many systems, a41 has been the predominant ADP-ribosylated protein detected when pertussis toxin blocks hormonal inhibition of adenylate cyclase (17, 19). Several laboratories have reported the presence of both and a41 in the heart (33, 34). The data presented here do not establish which of these a proteins couple muscarinic stimulation to inhibition of adenylate cyclase. Data have been presented by other groups suggesting that purified a39 and a41 reconstituted with solubilized receptors both couple muscarinic receptors to effects of GTP on affinity for agonist (32). However, the close correlation of an increase of a39 with increased adenylate cyclase inhibition suggests that in the chick heart, a39 might be responsible for this interaction with adenylate cyclase. Recent evidence suggests that guanine nucleotide coupling proteins may be involved in a number of functions unrelated to cyclic AMP production. These proteins may be involved in stimulation of phosphoinositide turnover (38), Ca2+ movements in guinea pig neutrophils, and release of arachadonic acid (39). Hence, a39 and a41 may interact not only with adenylate cyclase but may also serve several independent functions in the embryonic chick heart. The decreased level of muscarinic responsiveness of the embryonic chick heart and its association with decreased levels of a suggest that the developing chick heart could offer an interesting model for the study of the interaction of a proteins with the adenylate cyclase system and other regulatory processes in the heart. REFERENCES 1. McCarty, L. P., Lee, W. C., and Sheidman, F. E. (1960) J. Pharmacol. Exp. Ther. 129, Alexander, R. W., Galper, J. B, Neer, E. J., and Smith, T. W. (1982) Bwchem. J. 204, Papanno, A. J. (1979) Pharmacol. Rev Galper, J. B., Klein, W., and Catterall, W. A. (1977) J. Biol. Chem. 252, Culver, N. G., and Fishman, D. A. (1977) Am. J. Physiol. 232, R116-R Jakobs, K. A. (1979) Mol. Cell. Endmr. 16, Ross, E. A,, and Gilman, A. G. (1980) Annu. Rev. Bwchem. 49, Jakohs, K. A., Aktories, K., and Schultz, G. (1979) Naunyn-Schmiedekg's Arch. Pharmol. 310, Berrie, C. D., Birdsall, N. J. M., Burgen, A. S. V., and Hulme, E. C. (1979) Biochem. Biophys. Res. Commun. 87, Rosenberger, L. B., Yamamura, H. I., and bske, W. R. (1980) J. Biol. Chem. 255, Galper, J. B., Dziekan, L. C., OHara, D. S., and Smith, T. W. (1982) J. Biol. Chem. 257, Hulme, E. C., Berrie, C. P., Birdsall, N. J. M., and Burgen, A. S. V. (1981) in Drug Receptors and Their Effectors (Birdsall, N. J. M., ed) pp , Macmillan Journals Ltd., London 13. Gilman, A. G. (1984) Cell 36, Bokoch, G. M., Katada, T., Northup, J. K., Ui, M., and Gilman, A. G. (1984) J. BwL Chem. 259, Manning, D. R., and Gilman, A. G. (1983) J. Bid. Chem. 258, Hildebrandt, J. D., Codina, J., Risinger, R., and Birnbaumer, L. (1984) J. Biol. Chem. 259, Kurose, H., Katada, T., Amano, T., and Ui, M. (1983) J. Biol. Chem. 258, Burns, D. L., Hewlett, E. L., Moss, J., and Vaughan, M. (1983) J. Biol. Chem. 258, Neer, E. J., Lok, J. M., and Wolf, L. G. (1984) J. Biol. Chem. 259, Sternweiss, P. L., and Robishaw, J. D. (1984) J. Biol. Chem. 259, Galper, J. B., Dziekan, L. C., and Smith, T. W. (1984) J. Biol. Chem. 259, Hamberger,-U., and Hamilton, H. L. (1945) J. Morphol. 88, Neer, E. J. (1978) J. Biol. Chern. 253, Stadel, J. M., DeLean, A,, and Lefkowitz, R. J. J. Biol. Chem. 255, Laemmli, U. K. (1980) Nature 227, Huff, R.M., Axton, J. M., and Neer, E. J. (1985) J. Bwl. Chern. 260, Towbin. H.. Staehelin. T.. and Gordon.. J. (1979).. Proc. Natl. Acad. Sci. U. S. A. 76,' ' 28. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., Randall, R. J. (1951) J. Biol. Chem. 193, Blume, A. J., Lichtshtein, D., and Boone, G. (1979) Proc. Natl. Acad. Sci. U. S. A. 76, Dixon, M., and Webb, E. C. (1979) jn The Enzymes, (Boyer, P. D., ed) Chapter VIII, pp , Academic Press, New York 31. Marayama, T and Ui, M. (1983) J. Bwl. Chem. 258, Florio, V. A., grid Sternweiss, P. C. (1985) J. Biol. Chem. 260, Halvorsen, S. W., and Nathanson, N. M. (1984) Biochemistry Malbun. C. C.. Maneano. T. J.. and Watkins. D. C. (1985).. Biochem. BwDhvs. " Res. Commun. 1%,& Galper, J. B., Dziekan, L. C., Miura, D. S., and Smith, T. W. (1982) J. Cen. Physiol. 80, Hosey, M.M., McMahon, K. K., Danckers, A. M., OCallahan, C. M., Wong, J., and Green, R. D. (1985) J. Pharmacol. Exp. Ther. 232, 795-!Wl Ka;ya'da. ~~.~, T.. Nnrthun. d. ~~ K.. Bokoch. G. M..Ui.M..,,, and Gilman. A. G. (1984) J.'B~~L c&. 269' Cockroft, S., and Gomperts B. B. (1985) Nature 314, Bokoch, G. M., and Gilmai, A. G. (1984) Cell 39, SUPPLEMENTARY MATERIAL TO: Development of Muscarinic Cholinergic Inhibition of Adenylate Cyclase in Embryonic Chick Heart: Its Relationship to Changes in the Inhibitory Guanine Nucleotide Regulatory Protein Bruce T. Liang, Mark R. Hellmich,+ ** Eva J. Neer and Jonas 8. Galper METHODS CT). (Norwich, determined according were ages Embryonic description the to of Hamberger and Hamilton (22). Alamethecin was a gift of Mr. H.L. Keene Materials Co., Upjohn the of Kalamzoo. MI. ATP, GTP, creatine phosphate, creatine kinase, lima bean trypsin inhibitor, soybean trypsin inhibitor; 1.10-phenanthroline, L-isoproterenol. Adenylate Cyclase Assa atropine sulfate, carbamycholine, N-ethylmaleimide, dithiothreitol were ah1 Adenylate cyclase'activity was determined as described by Neer (23). ATP prepared by phosphorylation of adenine and hence low in contaminating obtained from Sigma; IlZ5 protein A was from Amersham. ['HI-ATP was from.. GTP was used substrate. a as The reaction mixture was 0.4 mm?hi-atp Amersham/Searle. Cyclic [14C] AMP was from New England Nuclear. 3ZP-NAD was from ICN and Islet activating protein (IAP) was from List Biochemicals, (specific activity mci/mm), 12 mm MgC12, 10 mm phosphocreatine,.84 u Campbell, CA. Bovine albumin was from Miles Scientific Naperville, IL. Leupetin was generously supplied by Dr. A.L. Goldberg, (Harvard Medical creatine phosphokinase, 1.0 d4 CAMP, 1 mm dithiothreitol, 75 mm NaCl, 100 School). Embryonated chicken eggs were purchased from Spafas, Inc. um GTP, 50 nm Tris HCl, 1 d4 EDTA, 0.2 M sucrose, and protease inhibitors

11 Muscarinic Inhibition of Adenylate Cyclase 9021 athe concentrations indicated. 1-isoproterenol, carbamylcholine, and atropine were added in the order and the concentrations indicated in the text. Following incubation for 10 min at 37OC diluting solution containing unlabeled CAMP, ATP and l4 C-CAMP was added and the reaction mixture boiled for two min., loaded onto Dowex AG50wXZ (Biorad. Inc.) columns and the CAMP peak collected. Further separation of remaining adenine and guanine nucleotides was accomplished by Ba(OH)Z, and &SO4 precipitation. Following Ba(OH)2 and ZnS04 precipitation recovery of [3H]cAMP was determined by measurement.of the ratio of C3H] to 14C] in an aliquot of the supernatant in a Beckman liquid scintillation iounter. Recoveries of C3H]cAMP were usually 40%. Preparation of tissue: Hearts of chick embryos of the indicated gestational age were suspended in TMSD buffer (50 mm Tris-HC1, 1 mm EDTA, 0.2 M sucrose and 0.6 dl dithiothreitol, ana 75 mm NaCl), Dounce homogenized and filtered through gauze at 4 C. The filtrate was adjusted approximately 0.5 mg protein/cc and 50 ul samples assayed in each tube. In some cases the filtrate was centrifuged at 30,OOOg for 15 min and resuspended in the same buffer at the concentration indicated Effect of protease inhibitors and almethicin on muscarinic inhibition of adenylate cyclase activity. Inhibition of adenvlate cvclase activitv bv carbamvlcholine varied significantly from prephation-to preparation. -The addition of a mixture of protease inhibitors 16 ugh1 soy bean trypsin inhibitor, 16 ugh1 lima bean trypsin inhibitor, 9 ugh1 leupeptin, 0.18 mg/ml, 1,lO phenanthroline and 1.4 dl EGTA markedly imporved reproducibility. Several groups have indicated that hormone mediated inhibition of adenylate cyclase activity was sensitive to trypsin and other proteases (24). The extent of inhibition of isoproterenol-stimulated adenylate cyclase activity was more reproducible (40-50%). especially in older age embryos, in the presence of the surface active antiobiotic alamethicin (14 ug/ml). Al experiments were carried out in the presence of protease inhibitors and alamethecin unless otherwise indicated. significant in Na+-free medium and increased nearly 3-fold in BO mm Na+. The concentration of GTP required for half-maximal inhibition of adenylate cyclase activity has been shown to be 5-fold greater than that for stimulation of adenylate cyclase (8). Studies of GTP dependence of inhibition of adenylate cyclase activity in homogenates of embryonic chick heart were complicated by the presence of high levels of endogenous guanine nucleotides in the crude samples studied here. Efforts to wash membrane pellets and to fractionate the membrane preparations led to large losses in responsiveness of adenylate cyclase to inhibition by muscarinic agonists. However, studies carried out on curde homogenates of embryonic chick heart demonstrated a 3-fold increase in carbamylcholine mediated inhibition of adenylate cyclase as GTP concentration was varied from 0 to 10-4M with a half-maximal effect at 7 um GTP. Kinetics of inhibition of isodrotereno1 stimulated adenylate cyclase activit. Alfhough the finding that muscarinic inhibition of adenylate cyclase is only partial even at high concentrations of muscarinic agonist suggests that inhibition is non-competitive, the mechanism of inhibition has not been rigorously established by classical analysis of competition data by Lineweaver-Burke plots. The effect of various doses of carbamylcholine on the concentration effect curve for isoproterenol stimulation of adenylate cyclase activity is presented in the double reciprocal plot shown in Fig. 3. The finding that these curves intersect at a single point on the x-axis is consistent with the interpretation that carbamylcholine decreased the extent of stimulation by isoproterenol, without affecting the K for the binding of isoproterenol at the /-adrenergic receptor. However, ken at the highest concentrations studied carbamylcholine was not capable of totally inhibiting isoproterenol stimulated adenylate cyclase activity (Figs. 2 and 3). Hence carbamylcholine is a partial non-competitive inhibitor of isoproterenol stimulated adenylate cyclase activity. The effectiveness of an inhibitor is expressed by a constant, KI. KI for a partial non-competitive inhibitor is given by the equation ADP-ribos lation with Pertussis Toxin ADP-:ibosylation was carried out for the times indicated at 37OC. The assay mixture contained 10 um NAD+, uci (32P)-NAD, 2.5 dl ATP. 2 mm GTP, 10 dl isoniazid, 10 M thymidine. 10 mm MgC12, 25 ng pertussis toxin, 10 mm phosphocreatine, 84u, creatine phospokinase in a total volume of 25 ul. Tissue homogenized in TMSD buffer was added to the reaction mixture at ug/assay. Protease inhibitors, soybean trypsin inhibitor, lima bean trypsin inhibitor and leupetin were present at the concentrations described previously from the time of initial homogenazation of the tissue. The reaction was stopped by the addition of 2% SDS in Laemnli sample buffer (25) followed by boiling for 1 min. Analysis of PePtideS following ADP-ribosylation was performed on 9% acrylamide gels prepared according to LaemnIl (25). Urled gels were exposgd to Kodak XAR film with or without enhancing screens for 1-2 days at -70 C. p Imnunoblottin : Antibodies were prepared and characterized as described previously 26). Imnunoblotting was carried out by a modification of the method of Towbin et at. (27). Heart proteins were solubilized in 2% SDS in Laemnli sample buffer as described under ADP-ribosylation and separated by SDS-polyacrylamide gel electrophoresis on 9% acrylamide gels. The gels were equilibrated in transfer buffer 20 mm Tris-OH, 150 dl glycine, 0.015% SDS and 20% methanol for 30 min. The proteins were then electrophoretically transferred to nitrocellulose at 30V overnight. The dried filters were ingubated with 3% BSA, in 10 dl Tris, ph 7.6, 150 mm NaCl for 60 min at 23 C tg decrease nonspecifc binding, followed by incubation overnight at 4 C with antiserum diluted ]:ZOO with the same buffer. Nitrocellulose was then washed two times with 10 mm Tris, 150 dl NaCl. for 10 min, once with 10 dl Tris., 150 mm NaCl and 0.05% nonidet P-40 for 20 min, and again with 10 dl Tris, 150 dl NaCl for 10 min. lz5i protein A (100, cpm/ml) in 10 mm Tris, 150 mm NaCl, 3% BSA was incubated with blots for 1 hr and the wash procedure repeated. After drying, blop were exposed to Kodak XAR film with enhancing screens for 1-5 days at -70 C. where V is the velocity in the absence of inhibition, V is the velocity in the presence of inhibitor at concentration I. V' is the'velocity in the presence of a large excess of inhibitor (30). The values of I/", I/V and I/V' are given by the intersection with the vertical axis of a plot OF l/velocity versus l/isoproterenol concentration in the presence of the appropriate concentration of inhibitor. The value for KI derived form the data in Fig. 3 was 1x10'6M for I = 10-6M carbamylcholine, and V- =the velocity in the presence of 10-3M carbamylcholine. J Protein Assay Proteins were determined according to the method of Lowry et al. using bovine serum albumin as the standard (28). RESULTS Further characterization of muscarinic inhibition activity of adenylate cyclase Effects of Na+ and GTP on muscarinic inhibition of 1 -adrenergic stimulated adenylate cyclase in the embryonic chick heart. Hormonal inhibition of adenylate cyclase activity mediated via opiate receptors in brain, adrenergic receptors in the platelet (29) and muscarinic receptop in the heart (6) has been shown to be dependent on Na' ions. Although Na had little effect on GTP-stimulated or isoproterenol-stimulated adenylate cyclase activity in chick heart homogenates (Table I, Column (1)). inhibition of carbamylcholine was barely I/ [ISOPROTERENOL] Fig. 3. Stimulation of adenylate cyclase by various concentrations of isoproterenol in the presence of several concentrations of carbamylcholine. Hearts from embryos 10 days ME were homogenized and adenylate cyclase activity assayed at the indicated concentrations of isoproterenol as described in Methods. GTP and Na+ were 100 um and 80 mm respectively. Carbamylchaline was either absent, 0; or present at 1 um.4 ; or 1 my.8. Data were plotted as the inverse of the concentration of isoproterenol (M) x ~ D - versus ~ the inverse of the stimulation of adenylate cyclase activity above GTP-stimulated levels in pmol/lomin/mg protein x lo3.