European Joumai of Pharmacology, 195 (1991) 19-25 0 1991 Elsevier Science Publishers B.V. ~1~2999/91/~3.50 ADONIS ~142~~1~217D

19

WP 51763

Effect of Gi protein ~~-~~~ylatio~ induced by pertwsis toxin on dopamine-mediated behaviors Concepcib Marin, Sotirios A. Parashos and Thomas N. Chase ~xpe~~e~t&~ ~eM~~tic~

Bruneir, Notional Institute of N~r~o~c~

Disordm and Stroke, Eethes&, MD, U.S.A.

Received 23 August 1990, revised MS received 22 November 1990, accepted 2 January 1991

The effect of Gi protein modification produced by intrastriatal pertussis toxin injection on dopzmine (DA)-mediated behaviors was studied. Ad~~st~tion of the selective Da agonist qu~p~ole induced ipsilateral rotation but the selective D, agonist SKF 38393 did not. However, SKF 38393 was able to increase the rotation induced by quinpirole. The selective Dz antagonist raclopride and the selective D, antagonist SCH 23390 both blocked the effect of quinpirole. Striatal levels of CAMP were measured in both intact and pertussis toxin injected striatum. SKP 38393 induced a significant increase in CAMP, but quinpirole had no effect. When both drugs were administered together, quinpirole attenuated the SKF 383934nduced increast in CAMP levels. Moreover, quinpirole-induced attenuation of SKF 38393 effect was greater in intact striatum. In pertussis toxin-injected striatum, quinpirole only attenuated SKF 38393-~du~ increase of CAMP to control levels. This imbalance between intact and injected striatum might be the cause of the rotation in pertussis toxin-injected rats suggesting an important role for G proteins in DA receptor interactions. Dopamine receptor interactions; Pertussis toxin; CAMP; Rotational behaviour

1. In~~uetion Characterization of two different dopamine (DA) receptors has been proposed on the basis of their biochemical, pharmacological and behavioral properties (Kebabian and Calne, 1979; Seeman, 1980; Waddington and G’Boyle, 1989). D, receptors are linked to activation of adenylate cyclase (AC) upon agonist stimulation whereas D2 receptors are either uncoupled or negatively coupled to this enzyme (Stoof and Kebabian, 1981). It is well established that AC activity is regulated by both stimulator (G,) and i~bito~ (Gi) guanine nucleotide binding regulatory proteins (Rodbell, 1980; Gilman, 1984). G proteins consists of three subunits: OL,/J and 6 (Codina et al., 1984). D, receptor stimulation is mediated by G, protein (Roufog~s et al., 1976) and D2 receptor i~bition by Gi protein (Fujita et al., 1985). Different mechanisms for the inhibitory control of adenylate cyclase have been postulated. Activation of Gi protein by an agonist-occupied inhibitory receptor leads to dissociation of the Gj protein into free a subunits (G,a) and a By subunits complex (Codina et al., 1984). Removal of free Gsa by association with released fly complex, direct Correspondence to: T.N. Chase, Experimental Therapeutics Branch, National Institute of Neurological Disorders and Stroke, Bethesda, MD, U.S.A.

i~bition of AC by by complex, and ~bitio~ of AC by Gia have all been implicated as possible mechanisms for inhibitory regulation (Katada et al., 1984; 1986). Certain toxins are capable of selectively modifying the function of these regulatory proteins. Cholera toxin ADP-ribosylates the G, protein causing a persistent activation of AC (Gill and Meren, 1978). Pertussis toxin ADP-ribosylates a cysteine residue on the Q subunit of Gi protein (Winslow et al., 1987), thus blocking the receptor-mediated inhibition of AC (Vi, 1984). Pertussis toxin may consequently be useful to study the role of Gi protein in DA r~eptor-mediate mechanisms. It has been shown that intrastriatal injection of pertussis toxin attenuates D, inhibition of AC activity (Olianas and Onali, 1987) and reduces apomorphine-induced stereotypy in rats (Fujita et al., 1985). Recently, it has been reported that the intrastriatal administration of pertussis toxin si~ficantly increases the spont~~us firing rates of caudate neurons (Boyajian et al., 1989). All these observations are in agreement with a loss of function of D, receptors. The recent use of selective agonists and antagonists of both DA receptor subtypes in biochemical, physiological and behavioral studies has resulted in clear evidence of a functional interaction between D, and DZ receptors. Biochemical studies have shown that the selective Dz agonist quinpirole attenuates CAMP accumulation induced by the selective D, agonist 5KF 38393

ian. 1981). In contrast, behavioral stic interaction between the two (Barone et al., 1956: Walters et D, and Dz dopamine receptor necessary for the expression of locomotor responses in various animal models: I agonist SKF 38393 facilitates stereotyped behavior or hyperactivity induced by the Dz agonist quinpirole 1987); a similar interaction between quinKF 38393 was found in experiments using avior in rats with unilateral quinolinic acid et al.. 1986). The mechanisms of this still unknown. In order to investigate a possible role for Gi proteins i/D? DA receptor interactions we have used a rotational model in rats with unilateral instrastrital pertussis toxin injection. In this model, rotational behavior induced by selective D, and D, agonists alone or in combination was studied.

tion of either behavioral or biochemical effects of pertussis toxin between 20 and 40 h after intrastriatal injection (data not shown) in agreement with previous reports (Fujita et al., 1985). Two hours of habituation was allowed before drug administration. All experiments were carried out between 9:00 a.m. and 4:00 p.m. in a quiet, dimly lit room. Rats were tested with the following drugs: the selective D, agonist SKF 38393 (10 and 20 mg/kg i-p.), the selective D2 agonist quinpirole (0.3, 1 and 3 mg/kg i.p.) and the mixed D,/D* agonist apomorphine (0.25 and 1 mg/kg s.c.). The selective D, antagonist SCH 23390 (0.05, 0.1 and 0.5 mg/kg i.p.) and the selective Dz antagonist raclopride (0.3, 1 and 3 mg/kg i.p.) were injected just before the Dz agonist in order to block quinpirole-induced behaviors (Barone et al., 1986). Statistical differences were determined using unpaired Student’s t-test for the comparison of two groups and ANOVA followed by Duncan Multiple Range test for the comparison of three or more groups, 2.3. Measurement of CAMP levels

2. I. intrastriatal itQecti0n.s Male Sprague-Dawley rats weighing 220-240 g were anesthetized with sodium pentobarbital(50 mg/kg i.p.) and placed in David Kopf stereotaxic frame. Unilateral intrastriatal injections of 1.5 pg of pertussis toxin in a volume of 3 ~1 were given at rate of 0.3 yl/min for 10 - through a 29 gauge cannula inserted in the left striatum (coordinates according to the Pellegrino-Cushman Atlas: A = 8.2, L = 2.6, V = 4.8) by means of a Harvard infusion pump (Harvard Apparatus, South Natick, MA). The dose of pertussis toxin used in this study is the EDso for inhibition of apomorphine-induced stereotypy (Fujita et al., 1985). Phosphate buffered saline (PBS) vehicle was injected to control rats. Injection cannula was left in place for 2 additional min after the infusion was completed to prevent leakage or diffusion from the site of injection. The postural asymmetry observed in pertussis toxin-injected rats was considered to indicate an accurate injection (Kelly et al., 1987) since in preliminary observations we have confirmed an accurate correlation between guide cannula position and the postural asymmetry that rats presented. No pathological changes in neuronal morphology were observed in cresyl violet-stained sections as has been previously reported (Boyajian et al., 1989). 2.2. Rotational behavior recording Turning behavior was recorded for 4 h in automated rotometers after 2426 h from pertussis toxin or vehicle injection. Preliminary observations showed no modifica-

Groups of animals that followed the same treatment protocol as for behavioral studies were killed 20 min after drug administration. Their brains were rapidly removed and striata dissected, homogenized in 300 ~1 of Tris/EDTA buffer (pH 7.5) and boiled for 3 min. A 50 ~1 aliquot of the homogenate was used for protein determination (Lowry et al., 1951). The homogenate was centrifuged at 16000 rpm for 20 min and duplicate aliquots of supematant (50 ~1) were removed for assay of CAMP content by RIA (CAMP assay system, Amersham, Arlington Heights, IL). In a preliminary experiment, a group of consrol animals received an i.p. injection of 3-isobutyl-1-methylxantine (IBMX, 10 mg/kg) 10 min prior to killing in order to inhibit cyclic nucleotide phosphodiesterase (Shickley and Krieger, 1984). Levels of CAMP in this group were not different from control animals which did not receive IBMX. Paired Student’s t-test was used for comparisons between injected and intact striatum. Differences between treatment groups were analyzed by a Dunn’s Multiple Comparisons test. 2.4. Drugs Pertussis toxin (Sigma, St. Louis, MO) was dissolved in phosphate buffered saline and apomorphine (Sigma, St. Louis, MO) was dissolved in a 0.2% ascorbic acid solution. SKF 38393 (Smith, Kline and French, U.S.A.) and quinpirole (Eli Lilly, U.S.A.) were dissolved in distilled water. SCH 23390 (Schering Plough Corp., Bloomfield, U.S.A.) was dissolved in a minimum amount of 0.5% lactic acid in normal saline. Raclopride (Astra, SBdertZlje, Sweden) was dissolved in normal saline.

21

3.1. Effect animals

of DA

agonists on

pertussis toxin-injected

The selective D, agumst SKF 38393 did not induce rotation in pertussis toxin-injected rats at any of doses tested (10 and 20 mg/kg) (fig. la). Low doses of the mixed D,/D, agonist apomorphine (0.25 and 1 mg/kg) showed a tendency to produce ipsilateral rotation which did nut achieve statistical significance. At higher dose f3 mg/kg), apomo~~ne caused stereotyped behaviur such as sniffing, gnawing and rearing that resulted in a a

QUINPRoLE AlDNE

TURNS

o-05

0.5

SCH 23390 (mglkg)

400 7

Fig. 2. Effect of the seWive D, antagonist SCH 23390 on rotational behavior induced by quinp~ole (I mg/kg) in pertussis toxin-injected rats. Quinpirole-induced rotation was blocked by all doses tested (0.05, 0.1 and 0.5 mg/kg). Points represent the mean number of turns+ S.E.M. for n = 5-8 anin& per group. * P +z0.01 versus quinpirole alone.

340 280 220 -

100 40 -20 ‘0”

SALfNE

10

20

SKF 38393

0.25

1

~~PH~NE

decreased locomotion (data not shown). The selective D2 agunist quinpirole induced ipsilateral rotation in a dose-dependent manner (P K 0.01). !KF 38393 (10 mg/kg) increased the number of rotations induced by quinpirole (1 mg/kg) when both dopamine agonists were administered in combin&iun (P c 0.01) (fig. lb). 3.2. Effect of DA antagonists on quinpiroie-induced rotation

(mg/kg)

Img/kg)

b

0.1

TURNS 400 340 1

Rutatiunal behavior induced by the selective D, agonist quinpirole was completely blocked by concur-

## #

# *

l

l

TURNS 120 too

IT

loo 80 40 SALINE

0.3

1

QUINPIROLE Fig. 1. Behavioral effect of dopamine

w 3

-I1 +SKF 33393

(10 mgikg) (mg/kg)

agonists in pertussis totin-injeeted rats. Twenty-four to twenty-six hours after unilateral intrastriatal injection of 1S p&/3 ~1 of pertussis toxin or vehicle rats were tested for rotational response to an acute challenge with SKF 38393, apomorpbine (a), quinpirok or SKF 33393 +quinpirole (b). Values represent the mean number of turnsf S.E.M. for n = 5-8 per group. * P c 0.05, * * P < 0.01 versus respective control; @P -z 0.05 versus pertussis toxin alone; o11P c 0.05 versus quinpiroie I mg/kg; {Cl)control, (W)pertussis toxin.

60 40 20 n ” QUINPIROLE ALUNE

g 3

1 ~A~L~P~l~E

3 ( mglkg)

Fig. 3. Effect of the selective Dz antagonist raclcpride on rotational behavior induced by quinpirole (1 mg/kg) in pertussis toxin-injected rats. Quinpirale-induced rotation was blocked at the highest dose tested (3 mg/kg). Pomts represent the mean number of turns f SE-M. for n = 5-g animals per group. * P c 0.01 versus Cjuinpirole alone.

E&Y& of ~~~~~ agonists on striatsl cAMP levek in pcrtussis t~~~~-~~_~t~~ pr$ts.~~~~t~-f~~r to fmrty hcms after unilateral in3 ~1 of percussis toxin, rats received an ‘r,quinpirole or their combination at tered alone. CAMP levels were measun& 20 min after drug injection. Each value represents mean of striata! cAMP Ievek* SEM. for n i?i:5-6 per group. ~~at~~t

Dose ~~~~g~

Saline

SKF 38-93 ~~~~pi~

10 1 10 I

cAMP (pmol/mg protein) Intact striatum

Injected striatum

3.4 + 0.3 4.6kO.7’ X5+0.09

3.3kO.l 4.5 rto.2 il 3.5 zko.03

2.4-co.2 hd

3.5 fU.2 c.=

h p -z&O1 ~%rsUS saline: E PcO.05. SKF f8393: ’ P -z 0.05 versus contralateral side.

= P