Interaction of D2-dopamine receptor with two pertussis toxin sensitive G proteins in human placenta

Life Scicseer, Vol. 60, No. 14 pp. l365-1375.195’7 cnpydgbt 0 1997 Ehcvk.? scicaec Inc. Printed in the USA Au rights lucd ooz4-m/97 $17.00 t .w

ELSEVIER

PII SOON-3205(97)00082-9

INTERACDON OF D2-DOPAMINE RECEF’I’ORWITH TWO PERTUSSIS TOXIN SENSITIVE G PROTEINS IN HUMAN PLACENTA C. VaiIIancourt, A. Petit, and S. Belisle. Department of Obstetrics and gynecology, University of Montreal and Research Center, Ste-Justine Hospital, 3175 Cote Ste-Catherine, Montreal, Quebec, Canada, H3T lC5. (Received in final form January 17, 1997)

We have demonstrated the presence in human placenta of 9 dopamine receptors (DzR) which inhibit human placental lactogen (hPL) release. This inhibitory effect of dopamine (DA) was sensitive to pertussis toxin (PTX) indicating that it may be mediated by the GilGo family of G proteins. However, nothing is known on this G proteins/DzR interaction in human placenta. In this study, we demonstrate that DA (lo-4 M) inhibits by 39 96 the ADP-ribosylation by PTX of two G proteins of 40 and 41 kDa. This inhibition is receptor specific since it is reversed by spiperone, a D2R antagonist. Moreover we show that bromocriptine. a & agonist, inhibited the 1abeIing of these two proteins in a dose-dependent mamrer with a maximd inhibition of 37 96 at a concentration of lO-(jM. In order to understand the role of D2R in placental endocrinology, we have analyzed the interactions of these two PTXsensitive G proteins with D2R in normal and abnormal pregnancies. The autoradiographs of both PTX ADP-ribosylated placental proteins of 40 and 41 kDa showed differential labeling during normal pregnancy. Thus, the relative levels of ADP-ribosylation by PTX of both proteins were 2.5 and 3.0 fold lower at term than those observed during fist-tand second trimester whereas no difference was observed between the first and second trimester. Also, no significant change in the level of inhibition by DA was observed between 7-9 weeks and 18-40 weeks of pregnancies (3545 % inhibition). However, we observed a maximal inhibition between 10 to 17 weeks of pregnancy (64 % inhibition). In placentas from preeclamptic pregnancies, the levels of ADP-ribosyIation were similar to those observed in normal pregnancy, while the DA inhibition was increased by 24 8. The levels of ADP-ribosylation in molar placentas reached 20 96 of normal values, while no difference in DA inhibition was observed. This study demonstrates that two distinct PTX-sensitive G proteins are coupled to human placental D2R. The physiological significance of the variations in these ADP-ribosylated-G proteins/ D2R interaction during normal and preeclamptic pregnancies remains to be investigated.

Key Wor&: placenta, D1-dopamine,G proteins, pertussistoxin

Address requests for reprints to : Dr Serge BtlisIe, Departement of Obstetrics and Gynecology, University of Montreal, Quebec. Canada, H3C 357. tel : (514) 343-5841 ; fax : (514) 343-6971

1366

Placental D, Receptor/G Proteins Interaction

Vol. 60, No. 16, 1997

The D2 dopamine receptor (D2R) belongs to the guanine nucleotide (GTP)-binding protein (G protein)-coupled receptor superfamily (for review see 1,2). In most tissues, dopaminergic signalling depends on the coupling of D2R to several inhibitory G protein subtypes, resulting in the activation of multiple cellular messengers (l-3). An endocrine analogy between the human placenta and the hypothalamic-pituitary axis was proposed several years ago (for review see 45). In agreement with this hypoythesis, we previouly demonstrated that dopamine (DA) inhibition of human placental lactogen (hPL) release (6) is mediated by D2R (7) similar to the regulation of prolactin (PRL) in the pituitary (8-10). Also as observed in lactotroph cells of anterior pituitary (9,10), we recently demonstrated that this dopaminergic modulation of hPL release from trophoblastic cells may be mediated through both the adenylyl cyclase (AC) (11) and calcium influx pathways (12). All of these signaling events are pertussis toxin (PTX)-sensitive suggesting they are mediated by the Gi/G, family of G proteins. These PTX-sensitive G proteins have been shown to mediate coupling of D2R to various signalling pathways and are involved in DA inhibition PRL secretion in pituitary. We have demonstrated that human placenta contains two PTX ADP-ribosylable proteins of 40 and 41 kDa (6). We have also recently reported that the a-subunits of three PTX-sensitive G proteins (Gai2, Gai3 and Gao) were expressed in human placenta during pregnancy (13). Moreover, we have observed variations in the expression of the human placental &R (14) as well as the Gai2 and Gai3 proteins (13,15,16) during normal and abnormal pregnancies, namely preeclampsia and hydatidiform mole. In lactotroph cells, the coupling of D2R to Gi2, Gi3, and Go proteins was shown to mediate DA inhibition of adenylate cyclase (AC), calcium influx and PRL release (8-10). Moreover, in the pituiuuy, these interactions appear altered under some pathological situations (17,18). Contrary to the abundant literature concerning I.&R in the pituitary little is known on the D#/G proteins coupling in human placenta during normal and abnormal pregnancies. To extend our knowledge on placental endocrinology, using the ADP-ribcsylation with PTX technique, we have analyzed the interactions of the two placentaI FTX-sensitive G proteins with D2R at various trimesters of gestation from normal and from preeclamptic and molar pregnancies the two pathological situations am associated with relatively low maternal hPL levels (19.20)

Jbiaterials and Methods [sZP]-NAD (specific activity : 1000 Cilmmol) was purchased from Amersham International (Oakville, Ontario, Canada). Bromocriptine methanesulfonate. dopamine (DA), and pertussis toxin (PTX) were purchased from Research Biochemicals Inc (Natick, MA), while dithiotheitol (DTT) and Lubrol PX were obtained from Promega (Madison, WI). ATP, bovine serum albumin (BSA), guanosine triphosphate (GTP), L-a-phosphatidylcholine-dimyristoyl, N-ethylmaleimide (NEM), nicotinamine, spiperone and thymidine were bought from Sigma (St-Louis, MO). Electrophoretic reagents were obtained from Bio-Rad (Richmond, CA). All other chemicals (Fisher Scientifique Co., Pittsburgh, PA) were of reagent grade and were used without purification. Placental tissue This project was approved by the Human Subject Research Committee of our institution and participating hospitals. Human placentas were obtained at various stages of gestation from the obstetrical services of St-Luc (7-l 1 weeks), Maisonneuve-Rosemont (16-19 weeks) and %-Justine (28-40 weeks) Hospitals. Placentas from molar pregnancies were obtained from the Montreal Jewish General Hospital (3 cases) and Royal Victoria Hospital (2 cases) whereas placentas from preeclamptic mothers (presents clinically hypertension and proteinuria) were from Ste-Justine

Vol. 60, No. 16, 1997

Placental D, Receptor/G

Proteins Interaction

1367

Hospital. The villous layer of trophoblastic tissues were dissected, washed with ice-cold 150 mM NaCl, frozen on dry ice and stored at -80°C until time of membrane preparation. PeDariyion of dacend membm Placental membranes were prepared as previously described (21). Pieces of tissue (0.5 g) were resuspended in 5 ml of 50 mM Tris-HCl (pH 7.4), 0.25 M sucrose, 1 mM PMSF and homogenized in a Polytron (Brinkmann, Switzerland) set at 4-5 for four times, 10 set each burst, with 2 min cooling period between each burst. The homogenates were centrifuged at 600 x g for 10 min at 4°C to remove cellular debris. The supematants were centrifuged at 10, 000 x g for 30 min at 4”C, fortified to 0.1 M NaCl and 0.2 mM MgCl2. and further centrifuged at 40, 000 x g to obtain microsomal pellets. The pellets were washed in homogenization buffer and recentrifuged at 40, Ooo x g . The pellets were homogenized at 10-20 mg/ml in buffer and stored at -80°C until used. Protein concentration was measured according to Lowry et al (22) with BSA as standard. ADP-ribcsvlation assav ADP-ribosylation was performed essentially as described by Resine and Law (23) with minor modifications as previously reported (6). Briefly, placental membranes (30 pg proteins) were incubated for 60 min in 0.5 PM NAD+, 1 PCi [a-32P]NAD+, 1 mM ATP. 104 M GTP. 10 mM thymidine, 1mM EDTA, 0.05 mM MgCl2, 1mM L-a-phosphatidylcholine-dimyristoyl, 25 mM DTT, 10 mM nicotinamide. 70 mM Tris-HCl (pH 8.0), 0.1% Lubrol PX (nonionic detergent), and 100 ng of preactivated PI’X (30 min at 37°C with 10 mM DlT) in a 60 ~1 assay volume. The reaction was stopped by adding 20 )cl of 2 % sodium dodecylsulfate (SDS) with 100 @ml BSA, and proteins were precipitated overnight at 4°C with 10% trichloroacetic acid. After centrifugation for 15 min at 12. 000 X g, the pellets were washed twice with ethyl ether, solubilized and treated with NEM according to Toutant et al (24) before analysis by SDS-polyacrylamide gel electrophoresis (SDS-PAGE). The samples were separated by electrophoresis on a 10% gel according to the method of Iaemmli (25). Electrophoresis was performed at room temperature for about 90 min at 32 mamp and then for 4-5 h at 35 mamp. The electrophoresis was stopped 1 h after complete migration of the blue dye. Gels were dried by DryEase System from Novex (Scarborough, Ontario, Canada), and autoradiography was performed using Kodak X-Omat AR5 film (Cie Universelle Rayon X Canada, Dorval, Quebec, Canada). Band intensity was quantified using an LKB Ultrascan XL apparatus (Pharmacia. Dorval, Quebec, Canada).

PTX ADP-riboryldion of human trophoblartic membrane proteins from term plizcen&s. When membrane proteins from term placental tissue were ADP-ribosylated with PTX two ADP-ribosylated proteins of 40 and 41 kDa were observed as we previously described (6) by comparison of their migration on SDS-PAGE to the migration of G proteins of known molecular weight (39,40, and 41 kDa) obtained from bovine neuroblastoma cells. Figure 1 indicates that when these experiments were repeated in the presence of DA or bromocriptine we observed a reduction of the labelling of these two bands. A representative example of this dopaminergic-inhibition of ADPribosylation is shown in Figure 1A whereas the quantitative evaluation of the [s”P} ADP-ribose incorporation into these two bands is represented in Figures 1C. No change was observed between relative labelling of the 40 and 41 kDa substrates in placental membranes (data not shown). Figure 1B shows that bromocriptine, a D2 agonist, inhibits the PTX-catalyzed ADP-ribosylation of 40 and 41 kDa placental proteins in a dose.-dependent manner. The maximal inhibition reached 37 f 7 % with 10-6 M bromocriptine and was maintained at higher agonist concentrations. Similar doseresponse curves were obtained with DA instead of bromocriptine, the maximal inhibition reaching 39 f 2 % with lo-4 M DA (data not shown). To demonstrate that both DA and bromocriptine inhibitions of ADP-ribosylation by PTX were mediated by specific D2R we studied the effects of specific DA antagonists. As can be seen in Figure 1B spiperone (105 M), a potent antagonist of

l368

vd. 60, No. 141997

Placental D, Receptor/G Proteins Interaction

D2R Mocked completely the inhibitory effect of both DA and bromocriptine on the labeling of the 40 and 41 kDa placental PTX-substrates. Similar results were obtained with sulpiride (10 -5 M), anotherD 2 antagonist whereas, even at high concentration (10-5 M), scherring 23390 (specific Drantagonist) and propranolol ( &adrenergic antagonist) did not block the inhibitory effects of both DA and hromocriptine (data not shown).

A CTL

BROMO

DA

BROMO

DA

Sh

S’PI

kDa

41L

so**

7-r

~o~Bd

~~~*

-3

4,, . CTL

***

-8

,

,

,

,

-7

-6

-5

-4

Log Bromoeriptine

6 a

(M)

0.4 CTl. BROMO

DA

BROMO

DA

SPI

Sh

+

Fig. 1 Effects of DA and bromocriptine on ADP-ribosylation by PTX of human trophoblastic membrane proteins fhm term placenta. Placental membranes (30 pg) were preincubated 15

min at 37’C without (CTL) or with 10-s M DA, or 10-e M bromocriptine (BROMO) in combination with 105 M spipcrone (SPI) (A and C) or increasing concentrations of BROMO (B). Membranes were then washed by centrifugation and incubated 60 min at 30 “C in the presence of P2P]-NAD (1 Fi) and preactivated FTX (100 rg/ml). After protein precipitation and washing, the samples were solubilized, denatured and electrophorectically separated by 10 96 SDS-PAGE. The dried gels were exposed to autoradiography without intensifying screen and quantification of autoradiographs was realized as described in Materials and Methods. A ) Representative examples of autoradiographs. B) The figure shows the dose-response curve for bromocriptine-inhibited ADP-ribosylation by FTX. C) The quantification of 40 and 41 kDA bands were measured by densitometry and expressed in arbitrary units. Control values arc arbitrarily set at 1.0 and results of quantification (mean f SE) expressed as % of controls. Results are representative of 3 independent autoradiographs obtained from 34 term placentas. * p c 0.01 vs ffL, ** p < 0.01 vs BROMO. *** p < 0.01 vs DA by ANOVA followed by Fisher’s PLSD comparison test.

EjJiwtr of DA on PTX ADP-ribosylation of human placental membrane proteins throughout normal pregnancy. The ADP-rihosylation by PTX of human placental membrane proteins from different trimesters of pregnancy is shown in Figure 2. A representative example of autoradiographs is shown in Figure 2A whereas the quantitative evaluation by scan analysis is presented in Figure 2B. The M’X-catalyzed ADP-ribosylation in human placental tissues varies during normal pregnancy, with the lowest levels observed at term. Thus, the relative levels of ADP-riboylation of the 40 and 41 kDa proteins in first and second trimesters of pregnancy were about 2.5 at 3.0 fold greater than at term, Also, Figure 2 shows the profile of interactions of these

Vol. 60, No. 16, 1997

Placental D, Receptor/G Proteins Interaction

1369

two PTX-sensitive G proteins with D2R throuhout pregnancy. No significant change in the inhibitory effect of DA on this PTX-catalyzed ADP-ribosylsation was observed between 7 - 9 weeks and 18 - 40 weeks of pregnancy; the levels reached 35 f 4 8 and 47rt 3 % of inhibition respectively. However, we observed a significant increase of this inhibitoty effect of DA between 10 to 17 weeks of pregnancy. Thus, this inhibitory effect at this stage of gestation was about 20% greater than the levels obtained between 7-9 weeks and 18-40 week (p < 0.01).

A kDa

Week of pregnancy 7

,

10

11

,

18

,

teml

B

:I*

*

73

IO-17

T

18.19

teml

Week of pregnancy

Fig. 2 Effects of DA on ADP-ribosylatlon by PTX of human trophoblaslic membrane proteins thronghoot pregnancy. Membrane proteins (30 pg) were extracted from trophoblastic tissues obtained from different weeks of pregnancy and preincubated 15 min at 37 “C without (CTL; 1,3,5,7.9 lanea) or with 104 M DA (2.4.6.8.10 lanes). Membranes were then washed, subjected to PTX-catalyzed ADP-ribosylation in the presence of [32P]-NAD, solubilized, and electrophorectically separated by 10 8 SDS-PAGE as described in Figure 1. Detection and quantification were carried out as described in Materials and methods. A) Representative example of autoradiographs. B) The quantification of both 40 and 41 kDa bands was measured by densitometry and expressed in arbitraryunits. Control values are arbitrarily set at 1.0 and results (mean * SE) expressed as 8 of controls. Results are representative of 3 independent autoradiographs obtained from 3 to 4 different placentas for each stage of pregnancy. * p < 0.01 vs term mL, ** p < 0.01 vs term DA by ANOVA followed by Fisher’s PLSD comparison test. Eflectx of DA on PTX ADP-ribosylation of human placental membrane The ADP-ribosylation by PTX of human placental proteins from abnormul pregnancies. membranes from preeclamptic and molar pregnancies, is shown in Figures 3 and 4. Representative examples of autoradiographs are shown in Figure 3A and 4A. Scan analysis of the automdiographs

l370

Placental D, Receptor/G Proteins Interaction

Vol. 60, No. 16, 1997

are shown in Figure 3B and 4B. In placentas from preeclamptic pregnancies, the levels of ADP-

ribosylation were equivalent to the levels seen in normal placental tissues at the same period of pregnancy (see Figure 2), while the inhibitory effect of DA (10-W) was increased by 24 96 (Fig. 3B) in preeclampsia, when compared with normal pregnancy. The inhibition in normal term and preeclatnptic placentas reached respectively 60 f 2 % and 36 f 4 % of control values (p < 0.05

Contrary to what we observed in preeclamptic pregnancies, in molar placentas the level of ADPribosylation was about 2(k 1 % (p < 0.05) of normal values at comparable periods of gestation (lo17 weeks), while no difference in DA inhibition was observed (60 f 2 96 inhibition) (Fig 4).

A PE

N

B

0

P 'I

c-m DA

1.2

F $ _

0.8

*

8 4

0.4

1 $

0.0

~ N

Effects of DA on ADP-ribosylation by PTX of hman tropboblastlc membrane pdelna fhml normd (Iv) and preechllpac (PE) pqnanck Membrane proteins (30 fig) obtained from normal (38-40 weeks) and preeclamptic (35-39 weeks) placentas were preincubated wothout (CTL: 1,3,5.7,9 lanes) or with 104M DA (2.4.6.8.10 lanes). Membranes were then washed and incubated in the presence of [s2P]-NAD (1 j&i) and preaetivated PlX (100 pglml). and processed as explained in Figure 1. A) Representative example of autoradiographs. The dried gel was autoradiographied by a 7 to 8 days exposure witbout intensifying screen. B) Quantification of autoradiograpbs by densitometry was carried out as described in Materials and methods and expressed in arbitrary units. Controls values are arbitrarily set at 1.0 and results (mean f SE) expressed as 8 of controls. The data is representative of 2 independent autoradiographs obtained from 6 and 5 different placentas from normal and preeclamptic pregnancies respectively. * p < 0.01 vs normal CTL. ** p < 0.01 vs normal DA by ANOVA

followed by Fisher’s PLSD comparison test.

vol. 60, No. 16.15~97

krh

1371

Placental D, Receptor/G Proteins Interaction

M(ll)

I

M(16)

,

M(W

,

NW)

I

N (19)

B c

a

I

T

N (1047)

M (11-M)

N (1.319)

DA

Mm

Fig.4 Effects of DA ou ADP-~Ibosylatiou by PTX of huum trophoblastic membrane proteius preguueiw. Membrane proteins (30 pg) obtained from normal (lo-19 weeks) and molar (1 l-20 weeks) placentas were preincubated without (CTL: uneven lanes) or with 104 M DA (even lanes). Membranes were then washed and incubated in the presence of [s2P]-NAD (1 &!i) aud preactivated PTX (100 pglml), and processed as explained in Figure 1. A) Representative example of autoradiographs. The dried gel was autoradiograpbied by a 7 to 8 days exposure without intensifying screen. B) Quantification of autorndiograpbs by densitometry was carried out as described in Materials and methods and expressed in arbitrary units. Controls values are arbitrarily set at 1.0 and results (mean f SE) expressed as 8 of controls. The data is representative of 2 independent autoradiographs obtained from 6 and 4 different placentasfrom normal and molar pregnancies respectively. * p < 0.01 vs normal CTL. ** p c 0.01 vs normal DA by ANOVA followed by Fisher’s PLSD comparison test.

f’rom normal (N) amd molar (M)

Discussion In the present study we have examined the association of human placental D2R with G proteins using the PTX catalyzed ADP-ribosylation technique. during normal and abnormal pregnancies. Our results demonstrnte for the first time that at least two distinct RX-sensitive G proteins are coupled to human placental D2R. This reinforces the functional correlation between

l372

Placental D, Receptor/G Proteins Interaction

Vol. 60, No. 16, 1997

human placenta and the pituitary. Moreover, our data show that these associations varied throughout pregnancy and that different coupling profiles exists between normal and preeclamptic pregnancies. No change in this interaction was observed in placentas from hydatidiform moles. These results confirm our previous observations that two PTX-sensitive G proteins were ADP-ribosylated in human trophoblastic membranes (6). Thus, two PTX-ADP-ribosylated proteins of 40 and 41 kDa were present in human trophoblastic membranes from each stage of gestation analyzed (Fig 2) as well as in placentas from abnormal pregnancies (Fig 3 and 4). However, even if two ADP-ribosylated proteins were constantly present in human trophoblastic membrane, in some cases (see Fig 3A and 4A) only one large band was visualised since these two proteins have similar molecular weight (40 and 41 kDa) and are difficult to separate using onedimensional SDS-PAGE (26,27). The results imply that the visualized band corresponds to both 40 and 41 kDa. We have recently demonstrated, by immunoblot analysis, that human placental membranes expressed at least three known PTX-sensitive a-subunits; Gai2, Gai3 and Gao (13). The antibodies used in this study recognized bands with apparent molecular weight of 37.7 kDa for Gao, 39.5 for Gai2, and 41.5 kDa for Gai3. In light of their electrophorectic mobilities on SDSPAGE, the two placental ADP-ribosylated proteins detected appear to represent Gai2, and Gai3. Moreover, previous studies have shown that Gai are more sensitive to PTX actions than Gao (28). We observed (Fig 2) that the levels of these PTX-substrates were lower in term placentas when compared to tissues from first trimester pregnancies. This is in agreement with our previous report showing that the levels of placental Gai2 and Gai3 but not Gao, decrease during normal pregnancy (13). Together, these observations strongly suggest that, as reported in other tissues (29), the placental Gai2 and Gai3 are substrates for PTX. Finally, the differences we observed in molecular weight were probably due to shifts in the mobilities of a-subunits as a result of ADP-ribosylation (30) and intervariability in SDS-PAGE separation. It is also possible that the placental Gao protein was also ADP-ribosylated but not visualized under the conditions used in our laboratory. The intrinsic property of PTX to ADP-ribosylate the a-subunit of the intact Gafiy heterotrimer protein and not thefree a (with or without bound nucleotide) (3 1) allows one to study the receptor/G protein coupling. The receptor/G protein interaction induces the dissociation of asubunit to pr and blocks the ADP-ribosylation of Gi/Go proteins by PTX. The major finding of our investigation is the substantial inhibition of the ADP-ribosylation by PTX of placental 40 and 41 kDa proteins by DA as well as by bromocriptine, a specific D2-dopaminergic agonist (Fig l), suggesting that the human placental D2R was coupled with these PTX-sensitive G proteins. These proteins were then probably involved in the DA inhibitory effects on hPL release (6), CAMP production (11) and calcium influx (12). The DA-inhibitory effect of ADP-ribosylation by PTX increased between 10 and 18 weeks of pregnancy (Fig 2), suggesting a parallel increase of D$/G protein interactions. However, we have recently reported that placental D2 is differentially expressed during normal pregnancy (14). Moreover, by immunoblot analysis with D2R antibody, we have observed that the relative level of D2R was at its minimal value between 10 to 18 weeks with a maximal value at term (14). The inverse correlation between D2R expression and PTX-ADP-ribosylation inhibition by DA we found is difficult to explain. In central nervous systems (32) and in the pituitary (33,34), it has been shown that the expression ratio of the D2R isoforms, D2(longf12(short), is regulated by sex steroids. Therefore, our results might represent a difference in the ratio of expression of the two D2R isoforms throughout normal pregnancy. The two isoforms of the D2R are encoded by splice variants of a single gene and differ only by the presence of an additional 29 amino acids in the third intracellular loop of D2(long)R (for reviews see 35). Amino acid residues within the third intracellular loop of the D2R isoforms play a role in the coupling of these receptors to G proteins (35). Moreover, Senogle (36) has recently reported that these isoforms of D2R appear to be coupled to distinct inhibitory G proteins and with a different potential in GH4Cl pituitary cells lines. Therefore, we can speculate that an increase in the form of D2R more potently coupled with the two

vol. 60, No. 16, 1997

Placental D, Receptor/G

Proteins Interaction

1373

PTX ADP-ribosylated proteins of 40 and 41 kDa may explain, in part, the increase of DA-inhibited ADP-ribosylation we observed between 10 and 18 weeks of gestation. However, nothing is known about the expression of these two receptor subtypes in human placenta. We have recently demonstrated that the expression of the placental D2R proteins decreased in preeclamptic and molar pregnancies (14). We have also previouly shown that Gai2 and Gai3 were expressed at low levels in human placentas from hydatidiform moles (15) whereas no change in these G proteins was observed in pregnancies complicated by preeclampsia (16). In agreement with these results, we observed a decrease in PTX-catalyzed ADP-ribosylation of the 40 and 41 kDa proteins in molar placentas (Fig 4), while no change was observed in pregnancies complicated by preeclampsia (Fig 3). No change in the levels of DA-inhibited ADP-ribosylation was observed in placentas from hydatidiform moles. However, we have previouly decribed that the levels of both D2R and G proteins were very low in this pathology suggesting that even in the presence of an adequate coupling, the DA effect may be lower in hydatidiform mole than in normal placentas. Similar phenomenon have been described in mammosomatotroph tumors. In such tumoral tissue the normal DzR/effector coupling is retained (37). Our results also demonstrate that, contrary to what we observed in hydatidiform moles an increase of DA-inhibited ADP-ribosylation was observed in pregnancies complicated by preeclampsia. This increase was similar to what we observed between 10 and 18 weeks in nomal placentas. However, we have observed by immunoblot that the relative level of D2R proteins decreased in preeclamptic placentas (14). We found it difficult to explain this inverse correlation between D2R expression and DA inhibited PTX-ADP-ribosylation. Bhansali and Eugere (38), have recently reported an abnormal concentration of estradiol and progesterone in preeclamptic placentas, suggesting the differential regulation of mRNA splicing by these steroid hormones may occur in this condition. Therefore, we can speculate that it might represent, as we previously mentionned, a difference in the ratio of expression of the two isoforms of the D2R. Determining whether this increase of DA-inhibitory effect is relevant physiologically will require further studies. In summary, the results reported here clearly demonstrate the association of the human placental D;?R with at least two PTX-sensitive G proteins during normal pregnancy as well as in preeclamptic and molar pregnancies. Even if our results cannot clearly determine the identity and the degree of functionnal coupling of G proteins to human placental D2R, they strongly suggest a regulatory role for Gi2 and Gi3 in the effects of DA in human placental endocrine activity. Also, the significant changes in placental DzR/G protein coupling in preeclamptic pregnancies, in contrast to what we observed in placentas from hydatidiform moles, suggest a specific alteration in the signal tranduction machinery associated with D2R in pathological placentas. The physiological significance of these ADP-ribosylated G protein,&R interactions throughout normal and abnormal pregnancies remains to be investigated.

The authors thank Dr Michele Brochu and Dr Philippe Vernier for support and helpful discussion and Dr Kenneth Chan for his assistance in obtaining molar placentas. This study was supported by Biopedia Inc and Studentship from Medical Research Council Canada to Cathy Vaillancourt.

1.

DK. GRANDY,

JR BUNZOW,

and 0. CIVELLI.

The Dopamine-D2

(Guest Ed) pp. 151-164 Marcel Dekker Inc., New York (19%).

Receptor. L FIB. Niznik

l374

2. 3.

4. 5. 6.

7. 8. 9. 10.

11. 12. 13. 14. 15. 16.

17.

18. 19.

20. 21. 22. 23.

Placental D, Receptor/G Proteias Interaction

Vol. 60, No. 16,19!37

P. SOKOLOFF and J.-C SCHWARTZ. Trends pharmacol Sci. 16 270-275 (1995). AM. LEW and HP. ELSHOLTZ. Dopaminergic Signaling and Regulation of Pituitary Hormone Genes. In : Dotqmine Rece~ rs and Tralk~porters * ~harmacol~n~. structure. and function. HB. Niznik (Guest Ed) pp. 4g-492 Marcel Dekker Inc., New York (1994). F. PETRAGLIA. Mol. Cell. Endocrinol. zC109-Cl12 (1991). F. PETRAGLIA, A. VOLDE, AR. GENAZZANI, J. RIVIER, PE. SAWCHENKO, and W. VALE. Frontiers in Neurocndocrinology. u 6-37 (1990). A. PETIT, G. GUILLON, C. PANTALONI, M. TENCE, N. GALLO-PAYET, D. BELLABARBA, J.-G. LEHOUX AND S. BELISLE. J. Clin. Endocrinol. Metab. a 1573-1580 (1990). C. VAILLANCOURT, A. PETIT and S. BELISLE. J Recept Res. 14 1 l-22 (1994). L. VALLAR and J. MELDOLESI. Trends Pharmacol Sci. u 7477 (1989) PM., LLEDO, P. LEGENDRE, JM. ISRAEL, and JD. VINCENT. Endocrinology. 127 990-1001(1990). A. ENJALBERT, F. SLADECZEK, G. GUILLON, P. BERTRAND, C. SHU, J. EPELBAUM, A. GARCIA-SAINZ, S. JARD, C. LOMBARD, C. KORDON, and J. BOCKAERT. J Biol Chem. 262 4071-4075 (1986). C. VAILLANCOURT. A. PETIT and S. BELISLE. Life Sci. s 1545-1552 (1994). A. PETIT, N. GALLO-PAYET, C. VAILLANCOURT, D. BELLABARBA, J.-G. LEHOUX and S. B&ISLE. J Clin Endocrind Metab. n 670-676 (1993). A. PETIT, P. GEOFFROY, P. BESSETTE, J. PRI?VOST and S. BELISLE. J Sot Gyn Invest. 2 678-685 (1995) C. VAILLANCOURT, A. PETIT, and S. BELISLE. Mol Cell Endocrinol. Submitted. A. PETIT, P. GEOFFROY, P. BESSETTE, and S. BELISLE. placenta 1z. 337-343 (1996) A. PETIT, P. GEOFFROY, P. BESSETTE, and S. BELISLE. Expression of human placental G proteins in pregnancies complicated by preeclampsia. Abstract #331 43 rd Annual Meeting of the Society for GynecoIogic Investigation, Philadelphia, PA, March 2023, (1996). L. CACCAVELLI, F. FERON, I. MORANGE, E. ROUER. R. BENAROUS, D. DEWAILLY, P. JAQUET, C. KORDON. and A. ENJALBERT. Neuroendccrinology. 60 3 14-322 (1994). DE WOOD, JM. JOHNSTON, and DG. JOHNSTON. Clin ndocrinol. a 455-466 (1991). AT. LETCHWORTH and T. CHARD. J Obstet Gyneacol Br Common. 79 680-683 (1972). K. BENIRSCHKE and P. KANFMANN. Molar pregnancies. In : Patholoev of the ta. 2nd edition. Springer-Verglad. pp. 782-815 (1990). and A. PETIT. Mol Cell Endocrinol. a 11 l- 119 (1989). vM. TENC O.H. LOWRY, N.J. ROSEBROUGH, A.L. FARR and R.J. RANDALL. J Biol Chem

=265-275 (1951). T . RESINE and F.S. LAW. Pertussis Toxin in analysis of receptor mechanisms. In: Methods in Neuroscience Edited by P.M. Michael Conn vol. 8 Neurotoxins. Academic

24.

press Inc, San Diego, CA. pp.358-367 (1992). M. TOUTANT, D. AUNIS, J. BOCKAERT and B. ROUOT.

FEBS Lett. 215339-344

25. 26.

(1987). U.K. LAEMMLI. Nature 22?. 680-685 (1970). D. STRASSHEIM, GT. MILLIGAN and MD. HOUSLAY

Biochem

27. 28. 29. 30. 31.

J 266 521-526

(1990). SJ McCLUE and G. MILLIGAN. Mol Pharmacol 40 627-632 (1991). R.M.HUFF and ENEER. J Biol Chem. al105 (1986). MI. SIMON, MP. STRATHMANN, N. GAUTAM. Science a 802-808 (1991) SC. ROERIG, HH. LOH and PY LAW. Biochem Biophys Res Commun m 12271232 (1991). G.M. BoKOCH, T. KATADA, J.K. NORTHRUP, M. UI and A.G. GILMAN. J BIOL Chem 259 3560-3567 (1984).

Vol. 60, No. 16, 19!X7

32. 33. 34. 35.

36. 37. 38.

Placental D2 Receptor/G Proteins Interaction

1375

D. GUIVARCH, P. VERNIER, and JD. VINCENT. Neuroscience. 69 159-166 (1995). LA. KUKSTAS, C. DOMEC, L. BASCLES, J. BONNET, D. VERRIER. JM. ISRAEL, and JD. VINCENT. Endocrinology 129 1101-1103 (1991). J. ZABAVNIK, WX. WU, KA. EIDNE, and AS. McNEILLY. Mol Cell Endocrinol. 95 121-128 (1993). P. FALARDEAU. Functional Distinctions of Dopamine &bng and Dzsbrt Receptors. In: boamine ReceDtors and Tmuworters * Dh ev. structure. and function. HB. Niz.nik (Guest Ed) pp. 323-342 Marcel Dekke; Inc?z’tork (1994). SE. SENOGLE. J Biol Chem. m 23120-23 127 (1994). A. SPADA, M. BASSETTI, F. REZA-ELAHI, M. AROSIO, P. GIL-DEL-ALAMO, and L. VALLAR. J Clin Endocrinol Metab. 28 411-417 (1994). KG. BHANSALI and EJ. EUGERE Res Comm Chem Pathology Pharmacol. a 161-169 (1992)