Regulation of Myometrial Phospholipase C System and Uterine ...

BIOLOGY OF REPRODUCTION 70, 570–576 (2004) Published online before print 29 October 2003. DOI 10.1095/biolreprod.103.021535

Regulation of Myometrial Phospholipase C System and Uterine Contraction by b-Adrenergic Receptors in Midpregnant Rat Sakina Mhaouty-Kodja,1,2 Eric Houdeau,3 and Chantal Legrand2 Laboratoire de Physiologie et Physiopathologie,2 UMR-CNRS 7079, 75252 Paris CEDEX 05, France Unite´ de Neuro-Gastroente´rologie et Nutrition,3 INRA, 31931 Toulouse CEDEX 09, France

free Ca21 activates myosin light chain kinase, the key enzyme that promotes the interaction between actin and myosin, resulting in contraction. In myometrium, Ca21 entry through channels participates also in the intracellular free Ca21 rise [1]. In many mammalian species, pregnancy is associated with a decrease of signaling pathways involved in uterine contraction. Indeed, activation of the Gq/PLC system by OT receptors (OTR), a1-AR (adrenergic receptors), or muscarinic receptors (mR) was shown to be less effective during pregnancy [2–4]. This can be explained, at least in part, by the previously reported downregulation of receptors [3, 5] and Gaq protein [6] at this period. Changes in the type and/or expression level of PLC isoforms could also contribute to this pregnancy-dependent decrease of PLC-linked signaling pathways. To date, 11 mammalian PLC isoforms have been isolated, including four b, two g, four d, and a recently described e isoform [7]. Among the different subfamilies of PLCs, PLCb members are triggered by G protein-coupled receptors. PLC b 1, PLC b 2, PLC b 3, and PLCb4 are all stimulated by Ca21, but are differently regulated by G protein subunits [7]. PLCb3 can be activated by both Gaq/11 and Gbg subunits, whereas PLCb1 and PLCb4 are more sensitive to Gaq/11, and PLCb2 is strongly activated by the Gbg complex [8]. Inhibition of the Gbgstimulated PLCb2 and PLCb3 was reported to occur through phosphorylation by the cAMP-dependent protein kinase (PKA) [9–11]. In rat myometrium, only some fragmentary data are available on the expression of PLCb isoforms and their regulation throughout pregnancy. Pregnancy is also characterized by a potentiation of signaling pathways that maintain the uterus in a quiescent state. b-AR are the most studied receptors that allow such an effect. These receptors stimulate adenylyl cyclase through the Gsa protein, thereby enhancing myometrial intracellular concentrations of cAMP. The subsequent activation of PKA results in inhibition of myosin light chain kinase and activation of the plasma membrane and sarcoplasmic reticulum pumps that lower intracellular free Ca21 [12]. Furthermore, b-AR can also induce hyperpolarization of myometrial cells through activation of K1 channels. A number of tetraethylammonium (TEA)- and 4-aminopyridine (4-AP)-sensitive K1 channels have been described in myometrium, including Ca21-activated, ATP-sensitive, and voltage-gated channels [13]. Activation of b-AR was also shown to exert a negative regulation on the PLC pathway by reducing inositol phosphate (InsP) production of late pregnant myometrium through cAMP-dependent or -independent mechanisms [4, 14]. At midpregnancy, little is known about the cross-regulation between the myometrial b-AR and the PLC system. The present study was, therefore, undertaken to characterize and quantify by immunoblotting studies the myo-

ABSTRACT We investigated whether b-adrenergic receptors (b-AR) regulate the phospholipase C (PLC) system in midpregnant rat myometrium. PLCb isoforms were characterized, and the effect of isoproterenol (b-adrenergic agonist) was tested on myometrial inositol phosphate (InsP) production and uterine contraction. Using specific antibodies, we showed that rat myometrium expresses PLCb1, PLCb3, and PLCb4, and to a lesser degree PLCb2. Quantitative analysis revealed that PLCb isoforms are differentially expressed during pregnancy. Indeed, the amount of PLCb4 is increased at midpregnancy, whereas PLCb1, PLCb2, and PLCb3 are up-regulated at term. At midpregnancy, pretreatment of myometrial strips with isoproterenol significantly reduced basal and agonist-stimulated InsP production. Forskolin, a diterpene that increases cAMP accumulation by directly activating adenylyl cyclases, had no effect on InsP production. In contrast, two global potassium (K1) channel inhibitors, tetraethylammonium (TEA) and 4-aminopyridine (4-AP), prevented attenuation of InsP production by isoproterenol. Isoproterenol also significantly decreased spontaneous and agonist-induced contraction of the longitudinal layer of midpregnant rat myometrium. Preincubation of uterine strips with TEA plus 4-AP prior to b-AR activation blocked only partial uterine relaxation, whereas Forskolin was as potent as isoproterenol. This indicates that b-AR operate through both K1 channels and cAMP to induce uterine relaxation. In conclusion, we show for the first time that three myometrial PLCb isoforms (PLCb1, PLCb2, and PLCb3) are down-regulated at midpregnancy. At this period, bAR reduce basal and agonist-stimulated InsP production through activation of K1 channels. Altogether, these mechanisms could act to decrease responsiveness of the longitudinal layer of myometrium to contractant factors.

catecholamines, pregnancy, signal transduction, uterus

INTRODUCTION

The phospholipase C (PLC) system plays a pivotal role in the regulation of uterine contractility. In response to neurotransmitters (catecholamines, acetylcholine) or hormones (oxytocin [OT], prostaglandins, endothelin), PLCs hydrolyze phosphatidylinositol 4,5-bisphosphate to produce the intracellular second messengers diacylglycerol and inositol 1,4,5-triphosphate. Diacylglycerol activates protein kinase C, and inositol 1,4,5-triphosphate releases calcium (Ca21) from internal stores. The following increase of intracellular Correspondence and current address: Sakina Mhaouty-Kodja, Institut Pasteur, Baˆtiment Jacques Monod. 25, rue du Docteur Roux, 75724 Paris CEDEX 15, France. FAX: 331 40 61 31 09; e-mail: [email protected]

1

Received: 24 July 2003. First decision: 20 August 2003. Accepted: 21 October 2003. Q 2004 by the Society for the Study of Reproduction, Inc. ISSN: 0006-3363. http://www.biolreprod.org

570

DIFFERENTIAL EXPRESSION OF MYOMETRIAL PLCb

metrial PLCb isoforms expressed throughout pregnancy. We also studied the effect of isoproterenol on myometrial InsP production and uterine activity of midpregnant rat in basal conditions and in response to OT, phenylephrine (Phe), and carbachol (CCh) (a1-adrenergic and muscarinic agonists, respectively). Contractile experiments were performed on both longitudinal and circular muscle layers. To determine the signaling pathway involved in the negative regulation exerted by b-AR on the PLC system, we compared the effects of Forskolin and two K1 channel inhibitors, TEA and 4-AP. The obtained data contribute to the better understanding of the molecular mechanisms underlying uterine quiescence during pregnancy and the initiation of labor at term. MATERIALS AND METHODS

Materials Krebs bicarbonate buffer was composed of the following: NaCl 117 mM, KCl 4.7 mM, MgSO4 1.1 mM, KH2PO4 1.2 mM, NaHCO3 24 mM, CaCl2 0.8 mM, glucose 1 mM, pH 7.4. Myo-[2-3H]inositol (10–25 Ci/ mmol) and polyvinylidene difluoride membranes were purchased from Perkin Elmer Life Sciences (Paris, France). Rabbit polyclonal antibodies directed against the carboxyl termini of PLCb were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA); horseradish peroxidase conjugated goat anti-rabbit antibody was from Jackson ImmunoResearch (West Grove, PA). TEA and 4-AP were kindly provided by Dr. P. Vincent, Paris, France. Isoproterenol, Forskolin, OT, Phe, and CCh were from Sigma (L’Isle d’Abeau, France); enhanced chemiluminescence reagent was from Amersham Pharmacia Biotech (Les Ulis, France); and AG1-X8 resin was from Bio-Rad Laboratories, Inc. (Marne la Coquette, France).

Animals and Tissues Sprague Dawley rats were obtained from Janvier (Le Genest, France). The females were caged with males overnight and successful mating was determined by the presence of spermatozoa in the vaginal smear (Day 1 of pregnancy). Animals were killed by cervical dislocation at estrus (nonpregnant), Day 15 of pregnancy (midpregnancy), or during the expulsion of fetoplacental units (term) following the guidelines laid down by the National Institutes of Health (NIH) Guide. The uterine horns were quickly isolated and cut open lengthwise, and the fetoplacental units were removed. For plasma membrane and cytosolic preparations and InsP production studies, the myometrium was freed of adherent endometrium.

Preparation of Myometrial Plasma Membrane and Cytosolic Fractions Myometrial tissue was homogenized in 0.5 mM EDTA and 10 mM Tris pH 7.4 supplemented with a cocktail of protease inhibitors (Sigma). After 10 min centrifugation at 48C, supernatants were collected and submitted to 100 000 3 g centrifugation at 48C for 1 h to separate plasma membranes from cytosol. Pellet containing plasma membranes was resuspended in homogenization buffer, and protein concentration of plasma membrane and cytosolic fractions was determined according to Bradford [15], with bovine serum albumin as standard. Samples were stored at 2808C until use.

Immunoblotting Twenty micrograms of myometrial plasma membrane and cytosolic fractions were subjected to SDS-Page in 7.5% gels and transferred to polyvinylidene difluoride membranes. The blots were blocked overnight at 48C in Tris-buffered saline containing 5% nonfat dried milk and incubated for 1 h at room temperature with anti-PLCb (diluted 1:200). Incubation with secondary antibody (diluted 1:10 000) was carried out for 45 min at room temperature. Immunoreactive bands were visualized by the chemiluminescence detection system.

Myometrial InsP Production Myometrial production of InsP was measured as briefly described. Myometrial strips (20 mg) were incubated at 378C for 4 h with 7 mCi

571

FIG. 1. Immunodetection of myometrial PLCb isoforms. Representative immunoblots for PLCb1, PLCb2, PLCb3, and PLCb4 isoforms with rabbit polyclonal antibodies in plasma membranes (pm) and cytosol (cyt) of nonpregnant (NP), midpregnant (MP), and term (T) rat myometrium.

myo-[3H]inositol (0.4 mM) in 1 ml of Krebs bicarbonate buffer in the presence of 5% CO2/95% O2. Increasing concentrations of OT, Phe, or CCh were added after 10 min incubation of myometrial strips with 10 mM LiCl in Krebs bicarbonate buffer. Assays were stopped 15 min later by freezing the strips in liquid N2. When used, isoproterenol 100 nM or Forskolin 100 mM were added 15 min before the addition of OT, Phe, or CCh. Preincubation of strips with TEA (10 mM) and 4-AP (2 mM) was conducted for 5 min before adding isoproterenol. [3H]InsP were measured as previously described [16].

Isometric Contraction Measurements Uterine strips of 4 mm long were prepared from pregnant rats and mounted in tissue baths containing 8 ml of Krebs bicarbonate buffer bubbled continuously with 5% CO2/95% O2 and warmed to 308C. Depending on the orientation of the strips, we measured tissue tension of the circular inner or the longitudinal outer layer of uterine muscle using Bioscience UF1 tension transducer (Phymep, Paris, France). Strips were washed twice with Krebs buffer and allowed to equilibrate for 30 min under 0.7 g resting force. Response to cumulative doses of OT, Phe, or CCh (0.1 nM–1 mM) were examined either in the absence or presence of isoproterenol (100 nM) or Forskolin (100 mM). When tested, TEA (10 mM) plus 4-AP (2 mM) or KCl (30 mM) were added to the tissue bath for 5 min prior to the application of isoproterenol (100 nM). All drugs used represented 1: 1000 of total volume. The concentration-response curves were recorded by computerized calculation of the integral under the tension/time curve for 3 min.

Statistical Studies Results are expressed as means 6 SEM. Statistical significance was assessed by Student t test for unpaired data. A P value less than 0.05 was considered to be significant.

RESULTS

Characterization and Quantification of Myometrial Plasma Membrane and Cytosolic PLCb Isoforms During Pregnancy

We performed immunoblotting studies on rat myometrial plasma membrane and cytosolic fractions using specific antibodies directed against PLCb1, PLCb2, PLCb3, and PLCb4. Rat brain, where the expression of the four PLCb isoforms has been previously reported, was used as positive control. As we have previously shown [16], PLCb1 (150 kDa), PLCb3 (150 kDa), and PLCb4 (130 kDa), but not PLCb2, are present in plasma membrane preparations (Fig. 1). PLCb1 and PLCb3 were detected in the cytosolic compartment as well, with a relatively high expression compared with plasma membranes. Interestingly, a specific band of 145 kDa corresponding to PLCb2 was also seen in the cytosol after a long time exposure, whereas no signal was detected for PLCb4 in this compartment.

572

MHAOUTY-KODJA ET AL.

FIG. 2. Quantification of immunoblotting signals corresponding to plasma membrane (pm) and cytosolic (cyt) PLCb1 (A) and PLCb3 (C), cytosolic PLCb2 (B), and plasma membrane PLCb4 (D) in nonpregnant (NP), midpregnant (MP), and term (T) rat myometrium. Data are expressed as percentage of T and are the mean of three to five independent experiments. aSignificantly different (P , 0.05) from T.

We then compared the expression of each myometrial PLCb isoform between nonpregnant, midpregnant, and term rats in both plasma membrane and cytosolic fractions. As illustrated in Figures 1 and 2A–C, the level of PLCb1, PLCb2, and PLCb3 declined at midpregnancy. Indeed, at this period PLCb1 and PLCb2 isoforms were hardly detectable, and the amount of PLCb3 decreased in both plasma membrane (270%) and cytosolic (250%) fractions. At term, the expression of these isoforms increased to reach a similar level to that observed in nonpregnant myometrium, at least for PLCb1 and PLCb3. Interestingly, PLCb4 presented a completely different pattern of expression (Figs. 1 and 2D). Indeed, this PLCb isoform was undetectable in nonpregnant myometrium. Its amount increased at midpregnancy and declined by 300% at term. These results show, for the first time, that rat myometrium differentially expresses PLCb isoforms during pregnancy. Effect of Agonists on Myometrial InsP Production and Uterine Contraction in Midpregnant Rat

We compared the effect of OTR, a1-AR, or mR activation on PLC activity at midpregnancy. As illustrated in Figure 3A, OT, Phe, and CCh elicited a dose-dependent increase of myometrial InsP production. However, the efficacy and potency of responses were different between the three tested agonists. Indeed, the calculated EC50 values indicated the following rank order of efficacy: OT . Phe

FIG. 3. Inositol phosphates (InsP) production (A) and contraction of the longitudinal (B) and the circular (C) layers of myometrium in response to increasing concentrations (0.1 nM to 1 mM) of oxytocin (OT), phenylephrine (Phe), or carbachol (CCh). Data are represented as percentage of basal InsP production (A) or spontaneous contraction (B and C) and are the mean of five independent experiments done in triplicate.

. CCh (3 nM, 47 nM, and 1.2 mM for OT, Phe, and CCh, respectively). In contrast, the maximal response was observed for CCh. Since myometrial PLC activation is linked to uterine contraction, we compared the responses of both longitudinally and circularly mounted strips to OT, Phe, and CCh. All uterine strips exhibited spontaneous rhythmic contractions a few minutes after being mounted in the bath (data not shown). The addition of increasing concentrations of


573

agonists revealed a higher efficacy of OTR to increase uterine contraction compared with the other receptors, whatever the muscle layer studied (Fig. 3, B and C). However, OT was more efficient in the longitudinal muscle (EC50 values were of 3 and 100 nM in the longitudinal and circular muscles, respectively). Furthermore, while CCh similarly contracted both uterine muscle layers, a specific effect of Phe was observed only in the circular muscle (Fig. 3, B and C). In the longitudinal muscle, we noted a significant decrease of uterine contraction at high concentrations of Phe, probably due to its unspecific interaction with b-AR as previously shown in mice [16].

FIG. 4. Effect of isoproterenol (iso) on myometrial InsP production and uterine contraction in midpregnant rat. A) Basal and OT-, Phe-, and CChstimulated myometrial InsP production were measured in the absence (control) or presence of iso pretreatment. Results are the mean of four independent experiments. aSignificantly different (P , 0.05) from control. b Significantly different (P , 0.05) from basal InsP production without iso pretreatment. cSignificantly different (P , 0.05) from basal InsP production with iso pretreatment. B) Representative recordings of spontaneous longitudinal (LM) and circular muscle (CM) tension following iso pretreatment. C) Average results from five experiments done in triplicate. aSignificantly different (P , 0.001) from control.

Effect of b-Adrenergic Activation on Myometrial InsP Production and Uterine Contraction in Midpregnant Rat Pretreatment of myometrial strips with isoproterenol (100 nM, 15 min) decreased basal InsP production by 35% (Fig. 4A). Total InsP response to maximally effective OT (1 mM), Phe (1 mM), or CCh (10 mM) was also diminished to the same extent (Fig. 4A). However, the net response (InsP production above basal) was not affected by isoproterenol (160%, 1100%, and 1160% for OT, Phe, and CCh in the absence or presence of isoproterenol). The EC50 values calculated for the three agonists were also unaltered by such pretreatment (data not shown). We also investigated whether activation of b-AR alters uterine contraction. Pretreatment of uterine strips with isoproterenol significantly decreased spontaneous contractions in the longitudinal but not in the circular uterine muscle (Fig. 4, B and C). This is in agreement with the previously reported predominant localization of b-AR in the longitudinal layer of myometrium [17]. Thus, all of the subsequent experiments were conducted in the longitudinal muscle. Pretreatment of uterine strips with isoproterenol did not block responses to OT or CCh, which were still able to increase contractile activity in a dose-dependent manner (Fig. 5, A and B). However, the obtained sigmoid doseresponse curves for OT revealed a significant 30%–40% decrease in uterine tension in the presence of isoproterenol (Fig. 5C). The calculated EC50 values were not significantly altered by isoproterenol pretreatment (3.5 and 6 nM in the absence or presence of isoproterenol, respectively). Similar results were obtained for CCh (data not shown). Together these data indicate that activation of the b-adrenergic pathway alters both InsP production and contractile activity of the longitudinal layer of myometrium at midpregnancy. Mechanisms Involved in the b-Adrenergic-Induced Decrease of Myometrial InsP Production and Uterine Contraction in Midpregnant Rat It is well known that b-AR increase cAMP production in pregnant rat myometrium. b-AR have also been impliFIG. 5. A) and B) Representative recordings of the effect of isoproterenol (iso) pretreatment on longitudinal muscle tension induced by increasing concentrations of OT or CCh (0.1 nM to 1 mM). C) OT-induced contractile activity in the absence (control) or presence of iso pretreatment. Results are the mean of five independent experiments done in triplicate.

574


FIG. 7. Representative recordings of the effects of K1 channel inhibitors TEA plus 4-AP (A), KCl (B), and Forskolin (C) on uterine longitudinal muscle activity of midpregnant rat. Uterine strips were preincubated with TEA (10 mM) plus 4-AP (2 mM) or KCl (30 mM) before adding isoproterenol (iso). The effect of Forskolin (FK) was tested on spontaneous contraction.

was as potent as isoproterenol in inducing uterine relaxation (Fig. 7C). Altogether these data reveal that reduction of InsP production by b-AR involves activation of K1 channels through a cAMP-independent mechanism. This could contribute to the induction of relaxation of the longitudinal layer of myometrium at midpregnancy. DISCUSSION FIG. 6. Effects of Forskolin and K1 channel inhibitors on myometrial InsP production of midpregnant rat. A) Basal InsP production was measured in the absence (control) or presence of 100 nM isoproterenol (iso) or 100 mM Forskolin (FK). B) The effect of TEA (10 mM) plus 4-AP (2 mM) on iso pretreatment. Data are the mean of three independent experiments. ns, not significantly different from control. aSignificantly different (P , 0.001) from control. bSignificantly different (P , 0.001) from iso.

cated in the activation of K1 channels in this tissue [13]. Therefore, we investigated which of these pathways mediates isoproterenol-induced effects at midpregnancy. Forskolin (100 mM), a diterpene that acts downstream b-AR by directly activating adenylyl cyclases, had no effect on basal InsP production (Fig. 6A). Interestingly, preincubation of myometrial strips with K1 channel inhibitors TEA (10 mM) plus 4-AP (2 mM) completely blocked the isoproterenol-induced decrease of InsP production (Fig. 6B). When incubated with uterine strips, TEA plus 4-AP evoked an immediate increase in uterine contraction, probably due to their ability to induce myometrial cell depolarization (Fig. 7A). Addition of isoproterenol suppressed this increase, but a remaining activity was still observed (Fig. 7A). Currently used to trigger myometrial membrane depolarization, KCl (30 mM) also elicited a rapid increase of uterine contraction (Fig. 7B). However, in this case no remaining activity was observed after isoproterenol application (Fig. 7B). This indicates that TEA and 4-AP partially blocked isoproterenol-induced relaxation at midpregnancy. Thus, a part of isoproterenol’s effects involves the coupling of b-AR receptors to K1 channels. When used, Forskolin

Differential Expression of Myometrial PLCb Isoforms During Pregnancy

Using specific antibodies, we showed the presence of the four members of the PLCb family (PLCb1, PLCb2, PLCb3, and PLCb4) in rat myometrium. PLCb1 and PLCb3 are present in both plasma membrane and cytosolic fractions, whereas PLCb2 was detectable in the cytosol after a long time exposure. The latter finding can explain why many studies on plasma membrane preparations, including ours, failed to detect the PLCb2 isoform in nonpregnant and pregnant myometrium [16, 18]. Expression of PLCb4 (130 kDa) was observed only in myometrial plasma membranes, indicating that this isoform corresponds to PLCb4a, the splice that can be activated by the Gaq protein [19]. Our quantitative analysis revealed for the first time that rat myometrium differentially expresses PLCb isoforms during the course of pregnancy. At midpregnancy, when the uterus is under progesterone dominance, the amounts of PLCb1, PLCb2, and PLCb3 are decreased, whereas that of PLCb4 is increased. At the opposite, when the myometrium is under estrogen dominance (estrus and term), the expression of PLCb1, PLCb2, and PLCb3 is upregulated, whereas that of PLCb4 is downregulated. Therefore, it is tempting to suggest that progesterone and/or estradiol directly or indirectly regulate the expression of myometrial PLCb isoforms. Downregulation of PLCb1, PLCb2, and PLCb3, together with that of Gaq [6], can explain, at least in part, the low ability of many G protein-coupled receptors to induce phosphoinositide hydrolysis and uterine contraction during


pregnancy [2–4]. The physiological meaning of PLCb4 upregulation at this period remains to be determined. A possible explanation is that PLCb4 is less efficient than the other members of the PLCb family to hydrolyze phosphatidylinositol 4,5-bisphosphate and induce contraction. Alternatively, PLCb4 could be involved in other myometrial responses such as cell growth and/or hypertrophy during pregnancy. In neonatal rat cardiomyocytes, PLCb expression seems to be required for the induction of early genes by hypertrophic factors like growth hormone and insulinlike growth factor I [20]. Regulation of Myometrial InsP Production and Uterine Contraction by Agonists in Midpregnant Rat

We report in the present work that OTR, a1-AR, and mR are linked to myometrial PLC activation and uterine contraction in midpregnant rat. Responses differ, however, depending on the type of activated receptor. Indeed, we noted a higher efficacy of OTR compared with a1-AR and mR to stimulate InsP production and induce contraction, whatever the uterine muscle layer studied. Whether or not these receptors are differently coupled to myometrial Gaq protein and PLCb4 should be further examined in future studies. Uterine responses to agonists are also dependent on the type of muscle layer. Indeed, whereas both layers responded similarly to CCh, longitudinal muscle was more sensitive to OT. In addition, a specific effect of a1-AR was found in the circular muscle only. This probably reflects a layer-dependent distribution of OTR and a1-AR. We have previously shown that a1-AR are predominantly localized in the rat circular muscle [17], and a recent work reported that OTR density is five-fold more important in the longitudinal muscle layer of swine uterus [21]. Regulation of PLC System and Uterine Contraction by b-AR in Midpregnant Rat

Previous studies reported that activation of myometrial b-AR significantly reduced OT- and CCh-stimulated InsP production in late pregnant rat [4, 14]. No data were available at midpregnancy, a different situation in terms of hormonal environment and expression of signaling transduction entities. Furthermore, none of these groups compared the effects of isoproterenol on both InsP production and uterine contraction. The present work shows that activation of b-AR decreases both basal and OT- and CCh-stimulated InsP production and uterine contraction in midpregnant rat. Interestingly, neither the net response nor the EC50 values for OTR and mR were affected. This indicates that isoproterenol pretreatment does not alter the ability of receptors to activate PLC. In the present report, we also show evidence that b-AR exert their inhibitory effects on the PLC signaling pathway through a cAMP-independent pathway. Indeed, Forskolin, at a concentration that increases maximally myometrial adenylyl cyclase activity at midpregnancy [22], failed to cause any attenuation of InsP production. Interestingly, two global K1 channel inhibitors, TEA and 4-AP, prevented the effect of isoproterenol. Our interpretation of these data is that hyperpolarization induced by activation of K1 channels indirectly affects InsP production by reducing the intracellular concentrations of Ca21. First, PLC enzymes are highly sensitive to Ca21 [7]. Second, activation of K1 channels by b-AR results in the closure of voltage-gated Ca21 channels in pregnant rat myometrium [14]. Therefore, a b-adrenergic-induced decrease of InsP production is probably due to

575

the suppression of the Ca21-stimulated component of PLC enzymes in midpregnant myometrium. Among the different TEA- and 4-AP-sensitive currents that have been described in pregnant myometrium, Ca21-activated and ATP-sensitive K1 channels were reported to mediate b-adrenergic-induced hyperpolarization of myometrial cells [23, 24]. Activation of these channels seems to occur not only through cAMPdependent phosphorylation [24], but also directly via a G protein [24, 25]. Further studies should be addressed to determine the precise mechanism underlying b-adrenergic activation of K1 channels in midpregnant rat myometrium. This regulation may have a supportive role in the cAMPdependent actions of b-AR at midpregnancy. Indeed, our contraction experiments showed that TEA and 4-AP only partially blocked b-adrenergic-mediated effects, whereas Forskolin was as potent as isoproterenol in inducing uterine relaxation. Our data contrast with those obtained at late pregnancy, where attenuation of rat myometrial OT-stimulated InsP production by isoproterenol was found to involve inhibition of PLCb3 by a cAMP/PKA pathway [4]. In support of the present work, the potential PKA targets PLCb2 and PLCb3 [9–11] are downregulated at midpregnancy (immunoblotting studies). Our interpretation of such a discrepancy is that, depending on the type of PLCb expressed during pregnancy, b-AR use different mechanisms to counteract the InsP increase. Thus, b-AR may operate through suppression of Ca21-dependent stimulation of PLCb4 at midpregnancy, and both suppress Ca21-dependent stimulation of PLCb1 and inhibit PLCb2 and PLCb3 via phosphorylation by PKA at late pregnancy. At this time, these complementary mechanisms would help to inhibit myometrial activity during uterine preparation for parturition. Indeed, up-regulation of many contraction-associated proteins occurs while the uterus is still quiescent. For instance, the expression of PLCb1, PLCb2, and PLCb3 is maximal from Day 19 of pregnancy (data not shown). In summary, we show that the amount of myometrial PLCb1, PLCb2, and PLCb3 declines at midpregnancy. This, together with the downregulation of Gaq protein, can explain, at least in part, the decreased responsiveness of myometrium to activation of PLC-linked signaling pathways at this period. Concomitant stimulation of b-AR exerts a negative regulation on PLC activity through activation of K1 channels in the longitudinal layer of myometrium. Altogether, these mechanisms could act to reduce responses to contractant hormones and neurotransmitters during pregnancy in order to permit development of fetoplacental units. At term, upregulation of signaling transduction entities of the PLC pathway (Gaq, PLCb1, PLCb2, and PLCb3) together with the desensitization of b-AR could help to initiate uterine contraction. The increased expression of PLCb4 at midpregnancy is intriguing. Future studies should evaluate whether this isozyme is involved in other responses such as myometrial cell growth and/or hypertrophy. ACKNOWLEDGMENTS We thank Dr. P. Vincent for the generous gift of TEA and 4-AP, Dr. J.G. Barbara for helpful discussions, and R. Monnerie for technical assistance.

REFERENCES 1. Sanborn BM. Relationship of ion channel activity to control of myometrial calcium. J Soc Gynecol Investig 2000; 7:4–11.

576


2. Lajat S, Tanfin Z, Guillon G, Harbon S. Modulation of phospholipase C pathway and level of Gq alpha/G11 alpha in rat myometrium during gestation. Am J Physiol 1996; 271:C895–C904. 3. Limon-Boulez I, Mhaouty-Kodja S, Coudouel N, Benoit de Coignac A, Legrand C, Maltier JP. The alpha1B-adrenergic receptor subtype activates the phospholipase C signalling pathway in rat myometrium at parturition. Biol Reprod 1997; 57:1175–1182. 4. Dodge KL, Carr DW, Yue C, Sanborn BM. A role for AKAP (A kinase anchoring protein) scaffolding in the loss of a cyclic adenosine 39,59-monophosphate inhibitory response in late pregnant rat myometrium. Mol Endocrinol 1999; 13:1977–1987. 5. Alexandrova M, Soloff MS. Oxytocin receptors and parturition. I. Control of oxytocin receptor concentration in the rat myometrium at term. Endocrinology 1980; 106:730–735. 6. Cohen-Tannoudji J, Mhaouty S, Elwardy-Merezak J, Lecrivain JL, Robin MT, Legrand C, Maltier JP. Regulation of myometrial Gi2, Gi3, and Gq expression during pregnancy. Effects of progesterone and estradiol. Biol Reprod 1995; 53:55–64. 7. Rhee SG. Regulation of phosphoinositide-specific phospholipase C. Annu Rev Biochem 2001; 70:281–312. 8. Rebecchi MJ, Pentyala SN. Structure, function, and control of phosphoinositide-specific phospholipase C. Physiol Rev 2000; 80:1291– 1335. 9. Liu M, Simon MI. Regulation by cAMP-dependent protein kinase of a G-protein-mediated phospholipase C. Nature 1996; 382:83–87. 10. Ali H, Fisher I, Haribabu B, Richardson RM, Snyderman R. Role of phospholipase Cbeta3 phosphorylation in the desensitization of cellular responses to platelet-activating factor. J Biol Chem 1997; 272: 11706–11709. 11. Yue C, Dodge KL, Weber G, Sanborn BM. Phosphorylation of serine 1105 by protein kinase A inhibits phospholipase Cbeta3 stimulation by Galphaq. J Biol Chem 1998; 273:18023–18027. 12. Wray S. Uterine contraction and physiological mechanisms of modulation. Am J Physiol 1993; 264:C1–C18. 13. Khan RN, Matharoo-Ball B, Arulkumaran S, Ashford MLJ. Potassium channels in the human myometrium. Exp Physiol 2001; 86:255–264. 14. Khac LD, Arnaudeau S, Lepretre N, Mironneau J, Harbon S. Beta adrenergic receptor activation attenuates the generation of inositol

15. 16.

17. 18. 19. 20. 21. 22.

23. 24. 25.

phosphates in the pregnant rat myometrium. Correlation with inhibition of Ca11 influx, a cAMP-independent mechanism. J Pharmacol Exp Ther 1996; 276:130–136. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976; 72:248–254. Mhaouty-Kodja S, Houdeau E, Cohen-Tannoudji J, Legrand C. Catecholamines are not linked to myometrial phospholipase C and uterine contraction in late pregnant and parturient mouse. J Physiol 2001; 536: 123–131. Legrand C, Vivat V, Rigolot C, Maltier JP. Selective distribution of alpha-1 and beta adrenoceptors in pregnant rat uterus visualized by autoradiography. J Pharmacol Exp Ther 1991; 256:767–772. Ku CY, Qian A, Wen Y, Anwer K, Sanborn BM. Oxytocin stimulates myometrial guanosine triphosphatase and phospholipase-C activities via coupling to G alpha q/11. Endocrinology 1995; 136:1509–1515. Kim MJ, Min DS, Ryu SH, Suh PG. A cytosolic, galphaq- and betagamma-insensitive splice variant of phospholipase C-beta4. J Biol Chem 1998; 273:3618–3624. Schnabel P, Mies F, Nohr T, Geisler M, Bohm M. Differential regulation of phospholipase C-beta isozymes in cardiomyocyte hypertrophy. Biochem Biophys Res Commun 2000; 275:1–6. Kitazawa T, Kajiwara T, Kiuchi A, Hatakeyama H, Taneike T. Muscle layer- and region-dependent distributions of oxytocin receptors in the porcine myometrium. Peptides 2001; 22:963–974. Mhaouty-Kodja S, Bouet-Alard R, Limon-Boulez I, Maltier JP, Legrand C. Molecular diversity of adenylyl cyclases in human and rat myometrium. Correlation with global adenylyl cyclase activity during mid- and term pregnancy. J Biol Chem 1997; 272:31100–31106. Anwer K, Toro L, Oberti C, Stefani E, Sanborn BM. Ca21-activated K1 channels in pregnant rat myometrium: modulation by a beta-adrenergic agent. Am J Physiol 1992; 263:C1049–C1056. Hamada Y, Nakaya Y, Hamada S, Kamada M, Aono T. Activation of K1 channels by ritodrine hydrochloride in uterine smooth muscle cells from pregnant women. Eur J Pharmacol 1994; 288:45–51. Toro L, Ramos-Franco J, Stefani E. GTP-dependent regulation of myometrial Kca channels incorporated into lipid bilayers. J Gen Physiol 1990; 96:373–394.