A functional genomic study to identify differential gene expression in the preterm and term human myometrium.

The mechanisms that lead to the onset of human parturition are still unknown, although selected critical factors have been identified. To investigate the changes in myometrial gene expression associated with parturition, we used two macroarrays each containing 1176 different complementary human cDNA clones. Methods involving hierarchical clustering and conventional statistical analysis allowed us to generate a profile of genes expression at three stages of late pregnancy: preterm (29 wk amenorrhea); full term, not in labor (38 wk amenorrhea); and full term in labor (39 wk amenorrhea). Only 4% of the genes investigated were differentially expressed between the preterm and term groups (P < 0.05). These genes could be clustered as groups of either down-regulated or up-regulated transcripts. The changes in transcript abundance were particularly marked between the preterm and term stages of gestation, whereas the differences between term not in labor and term in labor were less pronounced. The parturition was characterized by a massive down-regulation of a large panel of developmental, cell adhesion molecule and proliferation-related genes, along with the up-regulation of inflammatory, contraction and apoptosis associated genes. We propose that the mechanisms of parturition consist primarily in the arrest of the processes of myometrial development, a step that might be essential to allow the uterus to recover appropriate contractile function before delivery.

Differential expression of inositol 1,4,5-trisphosphate receptor types 1, 2, and 3 in rat myometrium and endometrium during gestation.

The regulation of the phospholipase C (PLC) and the expression of inositol 1,4,5-trisphosphate receptors (IP(3)Rs) in terms of mRNA, proteins, and binding capacity were examined in the rat myometrium and endometrium at midgestation (Day 12) and at term (Day 21) comparatively to the estrogen-treated tissues (Day 0). In both uterine tissues, the production of inositol phosphates mediated by carbachol as well as by AlF(4)(-) was enhanced with advancing gestation. (3)[H]IP(3) binding sites in membranes also increased during pregnancy (Day 21 > Day 12 > Day 0). The mRNAs encoding for three isoforms of IP(3)R as well as their corresponding proteins, IP(3)R-1, IP(3)R-2, and IP(3)R-3 were coexpressed, albeit to different extents, in the myometrium and endometrium. The expression of IP(3)Rs increased with advancing gestation, except for IP(3)R-2 that increased only in the endometrium at term. Thus, the pregnancy-related upregulation of the PLC cascade coincided with an increase in the expression of IP(3)Rs. The difference noted between the two uterine tissues suggests that IP(3)Rs may have cell-specific functions.

Carbachol-induced desensitization of PLC-beta pathway in rat myometrium: downregulation of Gqalpha/G11alpha.

In the estrogen-treated rat myometrium, carbachol increased the generation of inositol phosphates by stimulating the muscarinic receptor-Gq/G11-phospholipase C-beta3 (PLC-beta3) cascade. Exposure to carbachol resulted in a rapid and specific (homologous) attenuation of the subsequent muscarinic responses in terms of inositol phosphate production, PLC-beta3 translocation to membrane, and contraction. Refractoriness was accompanied by a reduction of membrane muscarinic binding sites and an uncoupled state of residual receptors. Protein kinase C (PKC) altered the functionality of muscarinic receptors and contributed to the initial period of desensitization. A delayed phase of the muscarinic refractoriness was PKC independent and was associated with a downregulation of Gqalpha/G11alpha. Atropine failed to induce desensitization as well as Gqalpha/G11alpha downregulation, indicating that both events involve active occupancy of the receptor. Prolonged exposure to AlF-4 reduced subsequent AlF-4 as well as carbachol-mediated inositol phosphate responses and similarly induced downregulation of Gqalpha/G11alpha. Data suggest that a decrease in the level of Gqalpha/G11alpha is subsequent to its activation and may account for heterologous desensitization.

Modulation of phospholipase C pathway and level of Gq alpha/G11 alpha in rat myometrium during gestation.

The regulation of the receptor-G protein-phospholipase C (PLC) cascade was investigated in rat myometrium at midgestation (day 12) and at term (day 21) comparatively to the estrogen-treated tissue (day 0). Carbachol-mediated generation of [3H]inositol phosphates was insensitive to pertussis toxin and was enhanced at days 12 and 21 two- and threefold, respectively, with no alteration of muscarinic sites (M3 subtype). A similar increase could be detected in the production of inositol 1,4,5-trisphosphate, indicating the stimulation of a PLC degrading phosphatidylinositol 4,5-bisphosphate. AlF4- also enhanced PLC activation during gestation, suggesting pregnancy-related regulations that bypass receptor activation. Immunoreactive G proteins, Gq alpha and G11 alpha, and PLC-beta 3 were detected in all myometrial preparations. The amount of PLC-beta 3 was similar in day 0 and day 21 myometrium, although decreasing by 75% at midgestation. Of significance was the increased amount of Gq alpha in day 12 and day 21 myometrium (3- and 2-fold, respectively) which coincided with the enhanced phosphoinositide breakdown. The upregulation of Gq alpha may contribute to the enhanced PLC activity during pregnancy and at term.

Activation of alpha-1A adrenoceptors mobilizes calcium from the intracellular stores in myocytes from rat portal vein.

Intracellular free Ca++ concentration ([Ca++]i) was monitored using the fluorescence from the dye fura-2-acetoxymethylester in single myocytes from rat portal vein. In the presence of oxodipine (a L-type Ca++ channel inhibitor), norepinephrine (10 microM) evoked transient increases in [Ca++]i which were related to release of Ca++ from intracellular stores. The alpha-1 adrenoceptors mediating intracellular Ca++ release and inositol phosphate accumulation were identified by using subtype-selective agonists and antagonists. Pretreatment with chloroethylclonidine had little effect on the norepinephrine-induced increase in [Ca++]i and inositol phosphate accumulation. In contrast, prazosin, 2-(2,6-dimethoxyphenoxyethyl)aminomethyl-1,4-benzodioxane and alpha-ethyl-3,4,5-trimethoxy-alpha-(3-((2-(2-methoxyphenoxy)ethyl)-amino )- propyl)benzeneacetonitrile fumarate produced a concentration-dependent inhibition of both intracellular Ca++ release and inositol phosphate accumulation. The rank of potency was prazosin > 2-(2,6-dimethoxyphenoxyethyl)aminomethyl-1,4-benzodioxane > alpha-ethyl-3,4,5-trimethoxy-alpha-(3-((2-(2-methoxyphenoxy)ethyl)-amino - propyl) benzeneacetonitrile fumarate. Methoxamine was as effective as norepinephrine but was less potent as shown by the rightward shift of the concentration-response curves. These results indicate that myocytes from rat portal vein express alpha-1A adrenoceptors whose activation stimulates phosphoinositide turnover and release of Ca++ from intracellular stores. The alpha-1A adrenoceptor stimulation of [Ca++]i and subsequent activation of Ca(++)-activated Cl- current was insensitive to intracellular applications of pertussis toxin, but concentration-dependently blocked by intracellular dialysis with a pipette solution containing anti-alpha q/alpha 11 antibody (whole cell recording mode).(ABSTRACT TRUNCATED AT 400 WORDS)

Diverse prostaglandin receptors activate distinct signal transduction pathways in rat myometrium.

Attempts were made to identify prostaglandin (PG) receptors in rat myometrium, according to the differential rank order of potencies displayed by the natural PGs and their analogues, both at the level of second messenger generation and contraction. In estrogen-treated rat myometrium, PGs [iloprost = PGI2 greater than PGE2 much greater than 16,16-dimethyl (DM)-PGE2; sulprostone = misoprostol = 0] induced adenosine 3',5'-cyclic monophosphate generation, indicating the contribution of a PGI2 receptor. The generation of inositol phosphates was stimulated by PGs (PGF2 alpha greater than PGD2 much greater than PGE2 = DM-PGE2 much greater than iloprost greater than sulprostone = misoprostol = 0), reflecting a PGF2 alpha-receptor-mediated process, which was insensitive to pertussis toxin (PTX). Contractions caused by PGF2 alpha were closely correlated to PGF2 alpha-receptor activation associated with the phospholipase C pathway. By contrast, contractions evoked by PGE2, equally mimicked by sulprostone and misoprostol, were abolished by PTX and were independent of phospholipase C activation. In the pregnant myometrium (day 21), the latter PGE-receptor-mediated mechanism also contributed to contractions caused by PGE2 (less than microM concn). Phospholipase C activation was coupled not only to PGF2 alpha but also to PGE receptors and could be correlated with contractions induced by PGF2 alpha and PGE2 greater than microM concn). All PGs tested were coupled to inhibitory G protein-mediated adenylate cyclase inhibition, displaying an equipotency that did not allow characterization of the inhibitory PG receptors.

Characterization of G proteins in rat myometrium. A differential modulation of Gi2 alpha and Gi3 alpha during gestation.

Myometrial membranes, obtained from estrogen-dominated (day 0) rat uteri, were immunoblotted with antiserum (SG1), which recognizes the alpha subunits of both Gi1 and Gi2, with antiserum (LE2) specific for Gi2 alpha, and with I3B antiserum, specific for Gi3 alpha. The data revealed the absence of detectable levels of Gi1 alpha and the simultaneous presence of Gi2 alpha and Gi3 alpha as Gi subunits in rat myometrium. The expression of Gi proteins during gestation (days 0, 12, 21) was studied with the above antibodies. No qualitative change in the nature of Gi alpha species was observed during gestation: Gi1 alpha remained undetectable, Gi2 alpha and Gi3 alpha were both present on days 12 and 21. Of significance was the increase (160%) in the amount of Gi2 alpha at midgestation (day 12) compared to days 0 and 21. A different pattern was observed with Gi3 alpha, which decreased with advancing gestation (day 0 greater than 12 greater than 21). Immunodetection of beta subunits of G proteins indicated the presence of a 35/36 kDa doublet on days 0, 12 and 21, with an increase at midgestation. The simultaneous increase in Gi2 alpha and beta subunits may provide an explanation for the previously demonstrated alteration in adenylate cyclase stimulability detected at midgestation.

Prostaglandins and muscarinic agonists induce cyclic AMP attenuation by two distinct mechanisms in the pregnant-rat myometrium. Interaction between cyclic AMP and Ca2+ signals.

In pregnant-rat myometrium (day 21 of gestation), isoprenaline-induced cyclic AMP accumulation, resulting from receptor-mediated activation of adenylate cyclase, was negatively regulated by prostaglandins [PGE2, PGF2 alpha; EC50 (concn. giving 50% of maximal response) = 2 nM] and by the muscarinic agonist carbachol (EC50 = 2 microM). PG-induced inhibition was prevented by pertussis-toxin treatment, supporting the idea that it was mediated by the inhibitory G-protein Gi through the inhibitory pathway of the adenylate cyclase. Both isoprenaline-induced stimulation and PG-evoked inhibition of cyclic AMP were insensitive to Ca2+ depletion. By contrast, carbachol-evoked attenuation of cyclic AMP accumulation was dependent on Ca2+ and was insensitive to pertussis toxin. The inhibitory effect of carbachol was mimicked by ionomycin. Indirect evidence was thus provided for the enhancement of cyclic AMP degradation by a Ca2(+)-dependent phosphodiesterase activity in the muscarinic-mediated effect. The attenuation of cyclic AMP elicited by carbachol coincided with carbachol-stimulated inositol phosphate (InsP3, InsP2 and InsP) generation, which displayed an almost identical EC50 (3 microM) and was similarly unaffected by pertussis toxin. Both carbachol effects were reproduced by oxotremorine, whereas pilocarpine (a partial muscarinic agonist) failed to induce any decrease in cyclic AMP accumulation and concurrently was unable to stimulate the generation of inositol phosphates. These data support our proposal for a carbachol-mediated enhancement of a Ca2(+)-dependent phosphodiesterase activity, compatible with the rises in Ca2+ associated with muscarinic-induced increased generation of inositol phosphates. They further illustrate that a cross-talk between the two major transmembrane signalling systems contributed to an ultimate decrease in cyclic AMP in the pregnant-rat myometrium near term.

Identification of both Gi2 and a novel, immunologically distinct, form of Go in rat myometrial membranes.

Immunoblotting of rat myometrial membranes with an antiserum (SG1) which recognizes the alpha-subunits of both Gi1 and Gi2 indicated the presence of detectable levels of an apparently single form of some 40 kDa. A second antiserum (LE2) specific for Gi2 also recognized this protein, confirming its identity. Immunoblotting of the myometrial membranes with a series of antipeptide (OC1, IM1, ON1) and polyclonal (RV3) antisera, all of which recognize rat brain Go, produced a more complex pattern. Antisera OC1 and ON1 recognized a single polypeptide which essentially comigrated with rat brain Go. In contrast, antisera RV3 and IM1 did not recognize the myometrial protein. These data provide evidence for the presence of Gi2 and of a novel G-protein, related immunologically to Go, in rat myometrial membranes.

Heterologous regulations of cAMP responses in pregnant rat myometrium. Evolution from a stimulatory to an inhibitory prostaglandin E2 and prostacyclin effect.

The regulation of cAMP generation in rat myometrium during gestation was investigated comparatively to the stimulations evoked in the estrogen-dominated myometrium. At the onset of gestation, there was a progressive attenuation in both tissue cAMP accumulation and adenylate cyclase activation in response to isoproterenol and prostaglandins (PGs) (PGE2 and prostacyclin, PGI2), as well as to cholera toxin and to forskolin. Minimal responses were observed at midgestation (day 12). The decline in isoproterenol-mediated cAMP accumulation could not be ascribed to alterations in either beta-adrenergic receptor density or coupling properties. Treatment of day 12 myometrium with islet-activating protein (IAP), pertussis toxin, caused a reversal of the attenuated beta-adrenergic--as well as cholera toxin--responses, suggesting that the inhibitory regulatory protein, Gi, was involved. However IAP treatment did not improve the PGE2-mediated stimulatory effect. In membranes from day 12 myometrium, the amount of [32P]NAD incorporated by IAP into the Mr = 41,000 peptides (alpha i, subunit of Gi was markedly increased compared to membranes from day 0 tissue. There was also a consistent decrease (25%) of the label incorporated by cholera toxin into the Mr = 52,000 (major) and 45,000/53,000 (minor) peptides (alpha s, subunit of Gs). The advanced stages of gestation were associated with a progressive restoration of cAMP responses to isoproterenol, cholera toxin, and forskolin with a full responsiveness before parturition (day 22). By contrast, the inability of PGE2 to generate any active Gs (stimulatory regulatory protein)-C complex persisted and coincided with the development of PGE2-induced inhibition of cAMP generation due to isoproterenol. The inhibitory effect, which was similarly evoked by PGE2 as well as by PGI2 and PGF2 alpha, was prevented by IAP. The data suggest that alterations of the stimulatory Gs-mediated pathway, which culminated at midgestation, is due at least in part to an increase in the abundance of Gi relative to Gs. Additional alterations operate at the level of the prostaglandin receptor(s) with a conversion from a stimulatory (Gs-mediated) cAMP response in the estrogen-dominated myometrium to an inhibitory (Gi-mediated) effect, fully expressed in the final stage of gestation.

Forskolin modulates cyclic AMP generation in the rat myometrium. Interactions with isoproterenol and prostaglandins E2 and I2.

In the intact rat myometrium, forskolin stimulated cyclic AMP generation and markedly potentiated increases in cyclic AMP caused by isoproterenol, prostaglandin E2 and prostacyclin. The diterpene increased the maximal responses and lowered the EC50 for both isoproterenol- and prostaglandin-stimulated cyclic AMP accumulation. Forskolin did not modify the Ki for the beta-adrenergic antagonist propranolol. Activation of cyclic AMP generation by forskolin was biphasic with respect to concentration; the major response being mediated by a low affinity interaction (Kapp 28 microM) and a minor effect being due to an interaction with a high affinity site (Kapp 0.5 microM). By contrast, the synergistic effect of the diterpene with isoproterenol, prostaglandin E2 as well as with cholera toxin, involved a single component of high affinity (Kapp 0.5 to 2 microM), which was thus considered to be associated with the activated complex of the cyclase catalytic subunit and the guanine nucleotide regulatory protein. Forskolin could further partially maintain isoproterenol-mediated synergism in a beta-adrenergic desensitized tissue. In myometrial membrane preparations, forskolin stimulated adenylate cyclase activity but failed to potentiate isoproterenol- and prostaglandin E2-mediated activation.

Differential role of microtubules in the control of prostaglandin E2 and beta-adrenergic stimulation of cyclic AMP accumulation in the rat myometrium.

A possible modulatory role of microtubules was investigated for the beta-adrenergic and prostaglandin E2 (PGE2)-induced cyclic AMP accumulation in the estrogen-treated rat myometrium. Colchicine, vinblastine and nocodazole, three different antitubulin drugs, enhanced cyclic AMP accumulation caused by PGE2. The effect of inhibitors was dose-(0.1-5 microM) and time-dependent, increased maximal responses without changing EC50 for PGE2, did not occur with trimethyl-colchicinic acid, which does not bind to tubulin, and was totally prevented in tissues pretreated with taxol, an agent that enhances polymerization and stabilization of microtubules. Concomitantly, colchicine reduced the rate and extent of PGE2-induced refractoriness in terms of cyclic AMP. In contrast, the antitubulin drugs failed to affect the rise in cyclic AMP evoked by isoproterenol and cholera toxin but enhanced the response to prostacyclin (PGI2), which is assumed to share common receptors with PGE2. Colchicine and vinblastine also failed to alter the ability of the beta-adrenergic agonist to provoke a cyclic AMP refractory state. Stimulations induced by all effectors were totally insensitive to cytochalasin B. The data suggest a relation between the constraints associated with the microtubules and/or membrane tubulin of the myometrium and the efficacy of PGE2 and PGI2 (but not the beta-adrenergic agonist or cholera toxin) to interact with the cyclic AMP forming system.