Neuropeptides (1992) 22, 143-147 0 Longman Group UK Ltd 1992

Peptidergic Modulation of G-protein Coupled Cyclic-AMP Accumulation in the Rat Caudate Nucleus A. MOSER* and H. CRAMERt *Department of Neurology, Medical University of Liibeck, Ratzeburger Allee 160, D-2400 Liibeck, FRG. tDepattment of Neurology, University of Freiburg, Hansastr. 9, D-7800 Freiburg, FRG. (Reprint requests to AM)

Abstract - Somatostatin, substance P, and vasoactive intestinal polypeptide were incubated in an adenylate cyclase assay with a particulate fraction of caudate-putamen tissue of the rat in order to examine the effect of the neuropeptides on G-protein coupled adenylate cyclase in vitro. Somatostatin induced an enhancement of cyclic AMP formation in presence of guanine nucleotides and cholera toxin but inhibited pertussis toxin and forskolin enzyme stimulation. Pertussis toxin and cholera toxin also depressed forskolin-induced stimulation as described previously. Somatostatin was able to antagonize these inhibitory effects of both toxins. On the contrary, substance P reduced GTP and cholera toxin stimulated striatal adenylate cyclase, without affecting forskolin activation. In our preparation, VIP did not influence basal adenylate cyclase activity or the stimulation by guanine nucleotides, cholera toxin, and pertussis toxin. VIP potently inhibited the enhancement of cyclic AMP formation by forskolin and completely antagonized the inhibitory effect of cholera toxin on forskolin activation. These results suggest that neuromodulatory effects of somatostatin, substance P, and VIP are mediated by the inhibitory as well as stimulatory guanine nucleotide proteins G-i and G-s coupled to an adenylate cyclase system. Introduction Somatostatin, vasoactive intestinal polypeptide (VIP), and substance P are peptides widely distributed in the mammalian central nervous system (l-5). High amounts are present in the nigrostriatal pathway, where they may act as neuromodulators or neurotransmitters (2, 4, 6-8). Chneiweiss et al (9) ‘have found VIP to stimulate the production of cyclic

Date received 2 December 199 1 Date accepted 5 February 1992

AMP indicating the existence of a VIP-sensitive adenylate cyclase on striatal neurons. Immunocytochemically, VIP-like immunoreactivity has been demonstrated in the caudate nucleus (10). As previously described by Moser et al (11,12) somatostatin is able to inhibit dopamine-stimulated adenylate cyclase in the caudate-putamen of the rat and substance P, decreased in the striatum in Parkinson’s disease, was found to excite nigral neurons ( 13,14). G-proteins as heterotrimeric proteins composed of three subunits, are transducer between the receptor and its biochemical effector system (15). 143

144 Receptors of the D-l dopamine receptor type are coupled to an adenylate cyclase system (EC 4.6.1.1) that is regulated by stimulatory hormones through G-s and inhibitory hormones through G-i (16, 17). Both proteins are activated by guanine nucleotides, whereas nonhydrolysable guanine nucleotide analogues like guanylylimidodiphosphate (Gpp(NH)p) are more potent than GTP, the natural effector (18, 19). In both cases the regulatory proteins are affected by an ADP-ribosylating toxin that causes an increase in the cyclic AMP content of their target cells. In order to clarify a potential interaction between Gproteins and the neuropeptides somatostatin, substance P, and VIP, we used a particle preparation of the rat caudate nucleus to measure the effect of Gprotein modulated adenylate cyclase activity.

Materials and Methods Female wistar rats of 150-200 g body weight were used. Rats were decapitated and caudate-putamen were dissected on ice. For each pellet preparation identical CP (caudate-putamen) from two rats were combined to reduce inter-individual differences. The tissues were homogenized ( 10 strokes) in an ice cold calcium-free solution of 10 mM TRIS-HCl and 4 mM EDTA, pH 7.6 (previously perfused with 95% oxygen, 5% carboxygen), centrifuged at 1500 g, resuspended and recentrifuged 3 times at 0°C. The resulting particle suspension was used as a source of enzyme immediately after preparation. Protein was determined according to Bensadoun and Weinstein (20). According to our previous studies (11) the incubation medium was composed of approximately 10 l,tgmembrane protein, 0.4 mM magnesium chloride, 0.1 mM papaverine, 0.1 mM 5-adenylyl-imidodiphosphate, approximately 10 uCi a-32-P-ATP (1000 000 cpmtube), 10 ccg/ml adenosine deaminase, 0.1 M glycylglycine buffer, pH 7.4 and the indicated substances. The assay was linear with protein concentrations up to 100 ~18and with incubation times up to 30 min. The standard incubation time was 8 min at 30” C. To preactivate cholera toxin and pertussis toxin, 5 l.tM of each substance was mixed in 16 mM dithiothreitol and 10 @4 GTP in 100 mM glycylglycine buffer, pH 7.4 and preincubated at 30°C for 40 min. The reaction was termi-

NEUROPEPTIDES

nated by the addition of a solution of 10 mM ATP, 10 mM cyclic AMP, sodium dodecylsulfate 2% in 50 mM TRIS-HCl, pH 7.4 . The formation of 32-Pcyclic AMP was measured according to Salomon (21) by separation of the reaction product through sequential chromatography on Dowex 50 cation exchanger and on neutral alumina. Within each experiment adenylate cyclase measurements were replicated 3 times for each experimental condition. Values were expressed in pmoles cyclic AMP/min/mg protein f standard deviation (SD). For statistical analysis Student’s t-test was employed.

Results As shown in Table1 , basal adenylate cyclase activity was not influenced by the neuropeptides. Both pertussis toxin (0.1 l&Q and cholera toxin (0.1 @i) were able to enhance cyclic AMP formation that was previously described by Moser and Schuster (32) (Table 1). The stimulation induced by pertussis toxin was only depressed by somatostatin (0.1 pM). On the other hand, somatostatin (0.1 u.M) activated adenylate cyclase in presence of cholera toxin, while substance P (0.1 @4) inhibited the enzyme activity (Table 1). VIP (0.1 l&l) did not modify the toxin induced stimulations (Table 1). When the guanine nucleotide GTP (10 @I) was present in the incubation medium, substance P (0.1 uM) showed an inhibitory effect but stimulated in presence of Gpp(NH)p (10 @I) (Table 2). After employment of GTP or Gpp(NH)p, somatostatin (0.1 @4) slightly stimulated the adenylate cyclase

Table 1 Effects of substance P (SP, 0.1 pM), somatostatin (SST, 0.1 CLM),and vasoactive intestinal polypeptide (VIP, 0.1 pM) inthe absence andpresence ofpertussis toxin (PT, 0.1 ClM)and cholera toxin (CT, 0.1 l&l) on adenylate cyclase activity in pmoles cyclic AMP/min/mg protein f SD. SD calculated for several experiments, n = 3 for each experimental condition PT control SP SST VIP

*p < 0.05

6.0 7.2 6.2 5.9 significant

f f k f

0.5 0.8 0.3 0.2

compared

17.6 18.1 14.3 16.7

f 1.5 f2 f 0.3* f 2

to control.

CT 16.5 14.0 23.6 16.2

f 1.1 f 0.5* zk 1.2* k 3

PEPTIDERGIC

MODULATION

I

OF G-PROTEIN COUPLED CYCLIC-AMP

I

I

C

I

I

VIP

SST

SP

Fig. 1 Percentual GTP ( 10 pM) and Gpp (NH) induced adenylate cyclase activity after incubation stance P (SP, 0.1 @I), somatostatiu (0.1 pM>, and intestinal polypeptide (0.1 ph4). GTP + Gpp(NH)p 100% in each experimental condition.

p (10 pM) with subvasoactive activity =

activity in both cases (Table 2). In contrast to this, somatostatin (0.1 l&I) inhibited the stimulatory effect of forskolin (10 @I) (Table 2). Addition of VIP (0.1 pM) did not alter the stimulation of adenylate cyclase produced by GTP or Gpp(NH)p but markedly reduced the forskolin-induced enhancement of cyclic AMP formation (Table 2). As shown in Figure 1, the portion of GTP and Gpp(NI-I)p-induced effects on adenylate cyclase activity after modification by the neuropeptides (data of Table 2), related to total guanine nucleotide-induced activation (GTP + Gpp(NH)p = lOO%), was changed after incubation with substance P (0.1 @I). Herein substance P appeared to shift the GTP/Gpp(NH)p activity ratio of the enzyme raising Gpp(NH)p effects (Fig. 1). Cholera toxin (0.1 u.IvI)as well as pertussis toxin (0.1 @l) were able to decrease forskolin-induced

145

ACCUMULATION

enhancement of cyclic AMP formation to 68% and 75%, respectively (Fig. 2). Figure 3 shows that the inhibitory effects of the toxins increased almost linearly with preincubation times from 1O-40 min. (pertussis toxin data not shown). The inhibitory effects of pertussis toxin (0.1 l&I) and cholera toxin (0.1 @I) on forskolin-stimulated enzyme activity were completely antagonized by somatostatin (0.1 u.IvI) (Fig. 2). Similarly to somatostatin, vasoactive intestinal polypeptide (VIP, 0.1 @vI) was able to reduce cholera toxin-inhibition (Fig. 2). However, VIP did not modify the pertussis toxin effect on forskolin-activated adenylate cyclase (Fig. 2).

Discussion

The neuropeptides investigated are ail present invarious neurons in the caudate nucleus of several species and man (2, 10,22). The highest concentrations of vasoactive intestinal polypeptide were formed in the cerebral cortex (23) but VIP-like immunoreactivity has also been demonstrated in the corpus striatum, where such cells were identified as spiny neurons (10). In contrast to VIP, substance P seems localized in spiny I or spiny II efferent neurons of the striatum (22). Earlier evidence suggested a possible role of somatostatin in the modulation of dopaminergic activity in the caudate nucleus (11,

C

BP

Table 2

Effects of substance P (SP, 0.1 @I), somatostatin (SST, 0.1 pM), and vasoactive intestinal polypeptide (VIP, 0.1 pM) in the presence of GTP (10 ph4), Gpp(NH)p (10 l&l), or forskolin (10 ph$) on the adenylate cyclase activity in pmoles cyclic AMFVmin/mg protein * SD. SD calculated for 3 experiments, n = 3 for each experimental condition GTP

control SP SST VIP

13.5 * 7.8 f 20.3 f 14.0 f

0.2 0.5* 5* 2

GPP Ovf!) P

39.3 61.2 50.3 35.1

f 3 f3* * 3* f2

*p < 0.01 significant compared to control.

Forskolin

56.8 53.5 46.0 38.3

f 3 f 2 f 2* f: 3*

VIP

100

50

60 40 a0 % INHIBITION

0

a0

40

40

40

100

Fig. 2 Effects of substance P (SP, 0.1 pM), somatostatin (SST, 0.1 pM), and vasoactive intestinal polypeptide (VIP, 0.1 pM) on pertussis toxin (PT, 0.1 pM) and cholera toxin (CT, 0.1 @I) induced inhibition of forskolin (FO, 10 p&I) stimulated adenylate cyclase activity expressed as in percent of forskolin inhibition + SD. SD calculated for several experiments, n = 3 for each experimental condition. *p < 0.01 significant compared to control.

146

NEUROPEPTIDES

XINUIBlTION

10 1 40 i

a0 -

a0IO-

min 30 min 20 min

40

10 mln I

0’ -8

-7

-6

log M CHOLERA TOXIN

Fig. 3 Effect of different preincubation times on cholera toxin (10 nM - 1 @I) induced inhibition of forskolin (10 @I) activated adenylate cyclase expressed as in percent of enzyme inhibition.

12, 24, 25). A known physiological action of somatostatin is the inhibition of release of hormones from the adenohypophysis, which appears to be due to an effect of somatostatin on the inhibitory guanine nucleotide regulatory protein G-i of an adenylate cyclase in this organ (26). Jakobs and coworkers (27) have also found somatostatin induced inhibition mediated by G-i in S49 lymphoma cells. Levitzki (15) has proposed a G-protein model of adenylate cyclase modulation, in which the regulatory guanine nucleotide proteins are capable of hydrolyzing GTP, where the nonhydrolysable guanine nucleotides are more active than GTP itself. Further, Nomura et al (28 ) could demonstrate that in striatal membranes of rats GTP binds to the stimulatory guanine nucleotide protein G-s with higher affinity than to G-i. Guanylylimidodiphosphate (Gpp(NH)p) showed a potent stimulatory effect on G-s but seemed to act additionally on G-i (28,29). Pertussis toxin, which was found to block certain classes of guanine nucleotide binding proteins (30, 3 l), is able to stimulate the adenylate cyclase acting via the inhibitory guanine nucleotide regulatory subunit in different membrane preparations (19,26,32). On the other hand, cholera toxin was described as being functionally related to G-s (19, 33). In our experiments, basal adenylate cyclase activity was not influenced by the neuropeptides. In agreement with Macdonald and Boyd (34) who have found a substance P effect mediated by a G-protein required for high affinity binding in peripheral tissues, substance P was able to antagonize the stimu-

latory effect of cholera toxin itself and to activate the inhibitory effect of cholera toxin on forskolininduced stimulation in the caudate nucleus. In addition, the effect of the hydrolysable analogue GTP was inhibited by substance P, suggesting an interaction between substance P and the GTPase system of the stimulatory nucleotide protein G-s. In contrast to substance P, somatostatin was found to potentiate the stimulatory effect of cholera toxin but antagonized the cholera toxin-induced inhibition of forskolin-stimulated enzyme activity. Consistently, somatostatin reduced cyclic AMP formation induced by forskolin that was previously described acting predominantly on the stimulatory guanine nucleotide regulatory subunit G-s (35,36). This is in agreement with the results of Markstein et al. (37). Since somatostatin was also able to antagonize the stimulatory effect of pertussis toxin itself and the pertussis toxin induced inhibition of forskolin activation, the effect of somatostatin seems to be associated with both G-s and G-i. In our system, VIP reduced the forskolin stimulated adenylate cyclase activity as well as the cholera toxin effect induced on forskolin stimulation. These results are compatible with a site of action of VIP at the stimulatory guanine nucleotide protein G-s. In this respect an effect of VIP was only found in potently stimulated enzyme activity. In conclusion, we demonstrate that the neuropeptides distinctively regulate G-protein coupled adenylate cyclase activity. Each peptide was shown to have a characteristic attribute and profile of Gprotein related action. In certain circumstances, in future studies it should be possible to classify other neuropeptides according to their peptide-like Gprotein modulating pattern. Acknowledgement This study was supported by the Central Research Association of the Medical University of Ltlbeck.

References Eckernits S. A., Aquilonius S. M., Lundquist G. and Roose B. E. (1978). Regional distribution of choline@ and monoaminergic markers, somatostatin and substance P in the human brain. Acta Neurol. Stand. 57: 225-226. Hiikfelt, T., Johanson, C., Ljungdahl, J. M. and Schultzberg, M. (1980). Peptidergic neurons. Nature 284: 515-521. Loren, I., Emson, P. C., Fahrenkrug, J., Bj&klund, A., Alumets, J., Hakason, R. and Sundler, F. (1979). Distribution

PEPTIDERGIC

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15. 16.

17.

18. 19.

20.

MODULATION

OF G-PROTEIN

COUPLED

CYCLIC-AMP

of vasoactive intestinal polypeptide in the rat and mouse brain, Neuroscience 4: 1953-1976. Emson, P. C. and Marley, P. D. Cholecystokinin and vasoactive intestinal peptide. In: Iversen, L. L., Iversen, S. D. and Snyder, S. H. (eds) Handbook of psychopharmacology, Vol 16. Plenum Press, New York, 255-306. Said, S. I. and Mutt, V. (1979). Polypeptide with broad biological activity: isolation from small intestine. Science 169: 1217-1218. Hong, J. S., Hang, H. Y. T. and Costa, E. (1977). Projections of substance P containing neurons from neostriamm to substantia nigra. Brain Res. 122: 541-544. Kanazawa. I., Emson. P. C. and Cuello, A. C. (1977). Evidence for the existence of substance P containing fibres in striatonigral and pallidonigral pathways in rat brain. Brain Res. 119: 447-453. Mroz, E. A., Brownstein, M. J. and Leemann, S. E. (1977). Evidence for substance P in the striatonigral tract. Brain Res. 125: 305-311. Chneiweiss, H., Glowinski, J. and Premont, J. (1985). Vasoactive intestinal polypeptide receptors linked to an adenylate cyclase, and their relationship with biogenic amine- and somatostatin-sensitive adenylate cyclases on central neuronal and glial cells in primary cultures. J. Neurochem. 44: 779-786. Theriault, E. and Landis, D. M. D. (1987). Morphology of striatal neurons containing VIP-like immunoreactivity. J. Comp. Neurol. 256: 1-13. Moser, A., Reavill, C., Jenner, P., Marsden, C. D. and Cramer, H. (1986). Effects of somatostatin on dopamine sensitive adenylate cyclase activity in the caudate-putamen of the rat. Exp. Brain. Res. 62: 567-571. Moser, A., Reavill, C., Liebetrau, A., Jenner, P., Marsden, C. D. and Cramer, H. (1987). Dopamine and somatostatin modulated adenylate cyclase activity in the rat caudate-putamen following unilateral cortical ablation. Exp. Brain Res. 68: 406-410. Cheramy, A., Nieoullon, A., Michelot, R. and Glowinski, J. (1977). Effects of intranigral application of dopamine and substance P on the in vivo release of newly sythesized ‘Hdopamine in the ipsilateral caudate nucleus of the rat. Neurosci. Lett. 4: 105-109. Lembeck, F., Mayer, N. and Schindler, G. (1977). Substance P in rat brain synaptosomes. Naunyn-Scbmiedeberg’s Arch. Phatmacol. 301: 17-22. Levitzki, A. (1987). Regulation of adenylate cyclase by hormones and G-proteins. FEBS Lett 211: 113-l 18. Codina, J., Hildebrandt, J., Sunyer, T., Sekura, R. D., Manclark, C. R., Iyengar, R. and Bimbaumer, L. (1984). Mechanisms in the vectorial receptor-adenylate cyclase signal transduction. Adv. Cyclic Nucl. Res. 17: 11 l-125. Rodbell, M., Lad, P. M., Nielson, T. B., Cooper, D. M. F., Schlegel, W., Preston, M. S., Londos, C. and Kempner, E. S. (198 1). The structure of adenylate cyclase systems. Adv. Cyclic Nucl. Res. 14: 3-14. Sunyer, T., Codina, J. and Bimbaumer, L. ( 1984). GTP hydrolysis by pure Ni, the inhibitory regulatory component of adenylyl cyclases. J. Biol. Chem. 259: 15447-15451. Hildebrandt, J. D., Sekura, R. D., Codina, J., Iyengar, R., Manclark, C. R. and Bimbaumer, L. (1983). Stimulation and inhibition of adenylyl cyclases mediated by distinct regulatory proteins. Nature 302: 707-709. Bensadoun, A. and Weinstein, D. (1976). Assay of protein in the presence of interfering materials. Anal. Biochem. 70: 241-250.

ACCUMULATION

147

2 1. Salomon, Y. (1979). Adenylate cyclase assay. Adv. Cyclic Nucl. Res. 10: 35-55. 22. Groves, P. M. (1983). A theory of the functional organization of the neostriatum and the neostriatal control of voluntary movement. Brain Res. Rev. 122: 109-132. 23. Besson, J., Rotsztejn, W., Laburthe, M., Epelbaum, J., Beaudet, H., Kordon, C. and Rosselin, G. (1979). Vasoactive intestinal peptide (VIP): brain distribution, subcellular localization and effect of differentiation of the hypothalamus in male rats. Brain Res. 165: 79-85. 24. Cohn, M. L. and Cohn, M. (1975). ‘Barrel rotation’ induced by somatostatin in the non-lesioned rat. Brain Res. 96: 138141. 25. Garcia-Sevilla, J. A., Magnusson. T., Carlsson. A. (1978). Effect of intracerebroventricularly administered somatostatin on brain monamine turnover. Brain Res. 155: 159164. 26. Reisine, T. D., Zhan, Y. and Sekura, R. D. (1983). Pertussis toxin blocks somatostatin’s inhibition of stimulated cyclic AMP accumulation in anterior pituitary tumor cells. Biochem. Biophys. Res. Commun. 115: 794-799. 27. Jakobs, K. H., Aktories, K., Schultz, G. (1983). A nucleotide regulatory site for somatostatin inhibition of adenylate cyclase in S49 lymphoma cells. Nature 303: 177-178. 28. Nomura, Y., Arima, T. and Swgawa, T. (1987). Inthtences of pertussis toxin, guanine nucleotides and forskolin on adenylate cyclase in striatal membranes of infant, adult and senescent rats. Int. J. Dev. Neurosci. 5: 271-279. 29. Seamon, K. B. and Daly, J. W. (1982). Guanosine 5’-(S, yimido) triphosphate inhibition of forskolin-activated adenylate cyclase is mediated by the putative inhibitory guanine nucleotide regulatory protein. J. Biol. Chem. 257: 1159111596. 30. Spuler, A. and Grafe, P. (1989). Adenosine, pertussis-sensitive G-proteins, and K+ conductance in central mammalian neurons under energy deprivation. Neurosci. Len 98: 280284. 3 1. Silinsky, M., Solsona, C. and Hirsh, J. K. (1989). Pertussis toxinprevents the inhibitory effect ofadenosineandunmasks adenosine-induced excitation of mammalian motor nerve endings. Br. J. Pharmacol. 97: 16-18. 32. Moser, A., Schuster, 0. (1990). The effect of oxiferriscorbone on striatal adenylate cyclase ofthe rat. Neuropharmacol. 29: 73 l-734. 33. Bimbaumer, L., Swartz, T. L., Abramowitz, J., Mintz, P. W. and Iyengar, R. (1980). Transient and steady state kinetics or the interaction of guanyl nucleotides with the adenylate cyclase system from rat liver plasma membranes. J. Biol. Chem. 255: 3542-355 l. 34. Macdonald, S. G. and Boyd, N. D. ( 1989). Regulation of substance P receptor affiity by guanine nucleotide-binding proteins. J. Neurochem. 53: 264-272. 35. Seamon, B.. Padgett, W. and Daly, J. W. (1981). Forskolin: unique diterpene activator of adenylate cyclase in membranes and in intact cells. Proc. Natl. Acad. Sci. 78: 33633367. 36. Battaglia, G.. Norman, A. B., Hess, A. J. and Creese, I. (1986). Forskolin potentiates the stimulation of rat striatal adenylate cyclase mediated by D-l dopamine receptors, guanine nucleotides, and sodium fluoride. J. Neurochem. 46: 1180-1185. 37. Markstein, R., Stockli, K. A. and Reubi, J. C. (1989). Differential effects of somatostatin on adenylate cyclase as functional correlate for different brain somatostatin receptor subpopulations. Neuroscience Letters 104: 13-18.