INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

OXYTOCIN (OT), a cyclic nonapeptide produced primarily in the hypothalamic supraoptic and paraventricular nuclei, stimulates contraction of mammary myoepithelial and uterine smooth muscle cells by virtue of its interaction with a specific high-affinity G protein-coupled receptor (2, 6, 7, 13). OT is one of the most potent uterotonic agents identified, and during late gestation, OT receptors (OTR) are significantly induced in the myometrium in all mammalian species (3, 6, 17). Despite these associations, the physiological importance of OT in modulating the initiation and progression of labor remains unclear. Whereas OT infusion augments uterine contractions and hastens the progression of labor in both human and animal studies (1, 3, 7), mice rendered OT deficient by gene targeting demonstrate no defect in parturition (9, 11, 16). On the other hand, mice with defects in either prostaglandin synthesis or action, which manifest as delayed or absent labor, initiate labor when a fall in progesterone and induction of uterine OTR occur (9, 14). We have recently shown that mice with cyclooxygenase (COX)-1 deficiency have markedly impaired generation of PGF₂ during pregnancy that results in delayed luteolysis and labor, whereas mice with combined COX-1 and OT deficiency initiate luteolysis and labor at the normal time (9). These mice with combined COX-1 and OT deficiency, however, exhibit prolongation in the duration of labor. On the basis of these findings, we hypothesize that OT exerts a direct effect on the corpus luteum to maintain progesterone production in late murine gestation and, together with prostaglandins, facilitates uterine contractions and the progression of labor at term. Because the rise in prostaglandin production resulting in luteolysis is the primary determinant of labor initiation, the effects of OT deficiency are only evident in the setting of compromised prostaglandin production.

In this current study, we test the hypothesis that OT exerts direct trophic effects on the corpus luteum during murine pregnancy. We demonstrate that differential regulation of OTR in the corpus luteum and myometrium results in the ability of OT to either inhibit or promote the progression of labor depending on the relative expression of OTR in these tissues.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animal husbandry. Wild-type (WT) and OT knockout (KO) mice (C. E. Luedke and L. J. Muglia, unpublished data) were maintained on an outbred (129/Sv × Black Swiss) genetic background. Mice were housed on a 12:12 light-dark cycle with ad libitum access to rodent chow. All mouse protocols were in accordance with National Institutes of Health guidelines and approved by the Animal Care and Use Committee of Washington University School of Medicine, St. Louis, MO. Natural mating of estrous females with stud males was confirmed by detection of a copulation plug the morning following introduction of the female into the male cage. Females were subsequently isolated until the time of analysis to ensure accurate gestation timing, with the morning of copulation plug detection designated 0.5 days gestation.

Continuous infusion of OT. On day 15.5 of gestation, gravid OT KO or WT female mice (mated with OT KO or WT males, respectively) underwent implantation of osmotic minipumps (Alza, Palo Alto, CA) under isoflurane anesthesia. Pumps were loaded with either sterile PBS or OT (Sigma, St. Louis, MO) dissolved in PBS at concentrations necessary to provide 0, 0.12, 1, 2.5, or 5 U/day. Pumps were primed by overnight incubation at 37°C in a sterile tube containing PBS, then implanted subcutaneously. In some OT KO pregnancies (n = 3), gravid females received 150 µg indomethacin (5 mg/kg) via gavage 30 min before pump implantation. Time until the onset of labor, defined as delivery of the first pup, was subsequently measured in all pregnancies (n = 3-7 per genotype and dose) by observation of gravid females in the mornings and evenings (9) following pump implantation. Statistical analysis was by ANOVA with significance taken for P < 0.05. In unmanipulated and vehicle- or OT pump-implanted pregnancies (n = 3-7), OT KO and WT females underwent retroorbital phlebotomy at 16.0 or 19.0 days postcoitus for measurement of serum progesterone concentration by radioimmunoassay (Diagnostic Products, Los Angeles, CA). In the indomethacin-treated OT KO females, phlebotomy was performed at the time of delivery of the first pup on day 16 for progesterone measurement, and uteri were harvested for PGF₂ measurement as we have previously described (9). Statistical analysis was by ANOVA, with significance taken for P < 0.05.

In situ hybridization. Ovaries and uteri were fixed by immersion in diethylpyrocarbonate-treated 4% paraformaldehyde in PBS for 24 h at 4°C. Samples were then cryopreserved in 10% sucrose in PBS and embedded in OCT compound (Miles, Elkhart, IN) for sectioning on a cryostat. Ten-micrometer sections were thaw mounted onto Superfrost Plus slides (Fisher Scientific, Pittsburgh, PA) and hybridized to an -[³³P]UTP-labeled 800-base OTR or a 1.2-kb rat 20-hydroxysteroid dehydrogenase antisense riboprobe (10) as previously described (9). After washing, slides were exposed to Kodak NTB-2 emulsion (Eastman Kodak, Rochester, NY) for 3-10 days, developed, and then counterstained with hematoxylin and eosin.

RNase protection. Five or ten micrograms of total RNA from ovary or uterus, respectively, were subjected to RNase protection assay (RPA) by RPA II Kit (Ambion, Austin, TX) according to the manufacturer's specifications. The OTR antisense RNA probe was generated by digestion of a 1.0-kb mouse OTR cDNA fragment from exon 2 in pBluescript SK II+ with ApaL I and labeled with -[³²P]CTP (800 Ci/mmol) by transcription with T7 polymerase (Promega, Madison, WI). The product expected from protection of OTR mRNA was 0.38 kb. Each OTR mRNA hybridization signal was corrected for loading and recovery by normalization to mouse cyclophilin A intensity (Ambion, Austin, TX) in the same hybridization reaction. After hybridization and nuclease digestion, samples were subjected to electrophoresis through 6% acrylamide/8 M urea denaturing gels. After electrophoresis, the denaturing gels were transferred to Whatmann 3MM paper and quantitated on a Molecular Dynamics Phosphorimager (Sunnyvale, CA).

Northern blot hybridization. RNA was prepared by the guanidinium thiocyanate-cesium chloride method (4). Ten micrograms of total RNA from uterus and ovary were subjected to electrophoresis through 1.2% agarose-formaldehyde gels and transferred to nitrocellulose membranes. RNA probes specific for mouse OTR or cyclophilin A (as a loading control) mRNAs labeled with -[³²P]UTP were generated and hybridized as previously described (9).

Western blot analysis. Cell membrane protein extracts from uterus and ovary of WT mice and thymus from mice harboring an inducible OTR transgene regulated by the tetracycline transactivator (T. Imamura and L. J. Muglia, unpublished data) were prepared by sonication in 5 ml of 10 mM Tris (pH 7.5), 10 mM NaCl, 1 µg/ml pepstatin A, 2 µg/ml aprotinin, 5 µg/ml leupeptin, and 200 µM phenylmethanesulfonyl fluoride. The supernatant, after centrifugation at 800 g for 20 min at 4°C, was subjected to centrifugation at 46,000 g for 2 h at 4°C. The subsequent pellet containing membrane proteins was resuspended in 1 ml of 100 mM Na₂CO₃ (pH 11.5), 1 µg/ml pepstatin A, 2 µg/ml aprotinin, 5 µg/ml leupeptin, and 200 µM phenylmethanesulfonyl fluoride for 30 min at 4°C. The resulting suspension was subjected to a second centrifugation at 46,000 g for 2 h at 4°C followed by resuspension of the pellet in 10 mM HEPES-potassium hydroxide (pH 7.5), 1 µg/ml pepstatin A, 2 µg/ml aprotinin, 5 µg/ml leupeptin, and 200 µM phenylmethanesulfonyl fluoride. Twelve-microgram aliquots of the extracts were subjected to 10% PAGE and transferred to nitrocellulose. The anti-OTR monoclonal antibody O-2F8 (15) at 1 µg/ml (Rohto Pharmaceuticals, Japan) diluted in 0.44 M sodium chloride, 25 mM Tris (pH 8.0), 3 mM potassium chloride, 0.1% Tween-20, and 5% nonfat dry milk was used to detect OTR with visualization using an enhanced chemiluminescent detection kit (Amersham Life Sciences, Arlington Heights, IL) followed by exposure to BioMax film (Eastman Kodak, Rochester, NY). Specific bands of 56 and 69 kDa were found in uterus from WT mice and thymus from transgenic mice with induced OTR expression, corresponding closely in size to products detected in human myometrial cells with this antibody (12). Equivalent protein loading and transfer was confirmed by Ponceau S staining of the membrane.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Effect of OT infusion in mice. To determine if modulation of plasma OT concentration alters the timing for initiation of labor, we evaluated the consequences of continuous OT infusion in WT and OT KO mice. In OT KO mice, possible alterations in endogenous OT secretion are removed. Moreover, changes in OT sensitivity as a consequence of chronic OT deficiency can be assessed. WT and OT KO mice at day 15.5 of gestation underwent implantation of osmotic minipumps delivering OT at infusion rates of 0 (vehicle only), 0.12, 1.0, 2.5, or 5 U/day. WT mice receiving OT at rates of 0 or 0.12 U/day showed the onset of labor at the expected time following pump implantation. A significant delay in the onset of labor was observed in WT mice in response to OT infused at 1.0 U/day (onset of labor 4.6 ± 0.38 days after pump placement with 1.0 U/day OT vs. 3.8 ± 0.32 days for vehicle infused, P < 0.05, Fig. 1). When WT mice received OT at the higher infusion rates of 2.5 or 5 U/day, labor occurred prematurely, most often within 24 h of pump implantation. OT KO mice showed a similar pattern of response but at a lower dose range than the WT mice, such that the onset of labor was delayed at the infusion rate of 0.12 U/day and was premature at the rate of 1.0 U/day or above (Fig. 1). Thus at lower doses, OT infusion delays labor in both genotypes, whereas at higher doses, OT infusion precipitates preterm labor. Interestingly, chronic OT deficiency in the OT KO mice results in increased sensitivity to OT administration.

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Effect of continuous oxytocin (OT) infusion on the time until the onset of labor. Osmotic minipumps were implanted at day 15.5 of gestation into gravid wild-type (WT) or knockout (KO) females, and the duration until the onset of labor was determined. Vehicle administration (0 U/day) did not alter the timing of labor compared with that in unmanipulated mice. The number of mice in each group is indicated within the corresponding bar. Data are presented as means ± SE. * P < 0.05 vs. OT KO (0 U/day OT); ** P < 0.005 vs. OT KO (0 U/day OT); # P < 0.05 vs. WT (0 U/day OT); ## P < 0.0001 vs. WT (0 U/day OT).

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Effect of OT infusion on corpus luteum function. Control WT mice (no pump) or WT mice implanted at day 15.5 of gestation with osmotic minipumps delivering PBS or 1 U/day OT underwent retroorbital phlebotomy for analysis of plasma progesterone concentration at 19.0 days of gestation. Also shown for reference is plasma progesterone concentration in WT mice at day 16.0 of gestation before the onset of luteolysis. Data are presented as means ± SE. * P < 0.05 vs. no pump or PBS pump-implanted mice.

2

2

2

2

OTR expression in the ovary and uterus. To determine whether the maintenance of serum progesterone could be a direct effect of OT on the corpus luteum, we performed in situ hybridization analysis of ovary and uterus for detection of OTR mRNA. OTR mRNA was expressed at high levels in the corpus luteum at 16.0 days gestation, with little expression in the uterine myometrium (Fig. 3). In contrast, just before the onset of labor at 19.0 days gestation, less OTR mRNA was detected in the corpus luteum, whereas significant expression was apparent in the myometrium (Fig. 3). The decreased expression of OTR in the corpus luteum at 19.0 days gestation cannot be attributed to a general cessation of transcription during luteolysis, because other genes, such as 20-hydroxysteroid dehydrogenase (10), are induced in the corpus luteum at this time (Fig. 3).

View larger version (73K): [in this window] [in a new window]   Fig. 3.   Reciprocal regulation of the OT receptor mRNA in late-gestation ovary and uterus. Bright-field, hematoxylin- and eosin-stained sections (left) and the corresponding dark-field views for visualization of silver grains after emulsion autoradiography (right) are shown for days 16.0 (D 16.0, top) and 19.0 (D 19.0, bottom) of gestation. In situ hybridization using an antisense OT receptor (OTR) probe (middle) detected abundant expression in the corpora lutea (CL) and little expression in the myometrium (M) and decidua (D) at 16.0 days of gestation. In contrast to the pattern of expression at 16.0 days of gestation, in situ hybridization at 19.0 days of gestation showed little OTR mRNA in the CL and increased expression within the M. To confirm tissue integrity and hybridization specificity, an antisense 20-hydroxysteroid dehydrogenase (20-HSD) probe was hybridized to adjacent sections (right). 20-HSD mRNA increased in abundance in the corpus luteum as OTR mRNA was declining.

View larger version (55K): [in this window] [in a new window]   Fig. 4.   Quantitation of OTR mRNA in uterus and ovary by RNase protection assay. A: representative autoradiogram of products resulting after RNase treatment of uterine RNA hybridized simultaneously with OTR and cyclophilin (CYC) antisense probes. Nongravid (NG), 16.0-day gestation, and 19.0-day gestation protection products from WT and OT KO mice are shown with 2 separate animals per genotype and time point. Positions for the undigested OTR and CYC probes are shown in the first and last lanes, respectively. B: average OTR mRNA/CYC mRNA of duplicate samples from NG, 16-day gestation, and 19.0-day gestation uterus and ovary is shown, with 16-day WT normalized to 1.

View larger version (96K): [in this window] [in a new window]   Fig. 5.   Northern blot analysis of uterus and ovary OTR mRNA. Total uterine or ovarian RNA from NG or pregnant females of the gestational age (in days) and genotype given above the corresponding lanes were hybridized to OTR or CYC radiolabeled probes. Sizes of the uterine and ovarian transcripts appear the same.

View larger version (23K): [in this window] [in a new window]   Fig. 6.   Regulation of OTR protein in uterus and ovary during pregnancy. Western blot analysis of cell membrane proteins from uterus and ovary of NG or gravid WT mice of the indicated gestational ages (in days) is shown. Included as controls are membrane protein extracts from transgenic (Tg) mice with inducible expression of OTR in the thymus (I, induced; B, basal).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In this study, we demonstrate that OT can either inhibit or hasten the onset of labor. The findings that OTR localizes to the corpus luteum near the end of mouse pregnancy and that low-dose OT infusion augments progesterone production are in accord with our previous demonstration that mice with combined COX-1 and OT deficiency undergo a fall in plasma progesterone at term without prostaglandin-induced luteolysis (9). The physiological consequence of the augmentation in luteal function with OT infusion was prolonged gestation both in WT and OT KO mice. At higher doses, even without the 10- to 100-fold increase in OTRs that occurs at term (17), the uterotonic action of OT predominated such that preterm labor was induced. This uterotonic action of OT before the onset of normal term labor is likely a direct effect on low numbers of OTRs, because concomitant inhibition of prostaglandin biosynthesis did not abrogate this effect, and plasma progesterone remained elevated during labor.

To assess OTR signaling in the absence of endogenous OT, we examined the relative sensitivity of OT KO and WT mice to the OT infusion. Indeed, OT KO mice showed alteration in the timing of parturition at lower doses of OT than WT mice, suggesting increased receptor signaling. The altered dose response to OT in OT KO mice is not due to increased transcription of the OTR gene as demonstrated by our RNase protection assay. Whether the increase in signaling is due to increased OTR protein synthesis, alteration in posttranslational receptor modification, increased receptor numbers on the membrane due to lack of ligand binding, or increased signaling by an unchanged number of receptors is the subject of ongoing studies in our laboratory.

Examination of OTR expression in the ovary and uterus of WT mice near the end of gestation showed abundant expression of OTR mRNA in the corpus luteum at day 16.0 of gestation (before luteolysis) and very little at day 19.0 (during luteolysis). The uterine myometrium showed the opposite pattern, with a low level of OTR mRNA at day 16.0 and a high level on day 19.0. Whereas the OTR protein profile of the uterus closely reflected mRNA changes, the OTR protein level in the ovary at day 16.5 was not as high as might be predicted by the relative amount of OTR mRNA at this time, though the level of expression decreased in a fashion consistent with the mRNA expression. The discordance between OTR mRNA and protein levels in the ovary could reflect regulation of OTR protein expression by translational or posttranslational mechanisms. Additionally, alteration in migration of OTR after induction at term in uterus was detected by Western analysis. The OTR species migrating between 56 and 69 kDa are consistent with alteration in phosphorylation, which could result from receptor modification by a G protein receptor kinase (5) after OT-induced activation as part of the process of augmentation of uterine contractions during labor. Future studies using OT KO and inducible OTR transgenic mice will seek to determine the etiology of these altered OTR species. OT infused at the lower doses was capable of significantly modulating ovarian function, despite the modest receptor protein level, with the result of promoting survival of the corpus luteum and prolonging gestation. At the same time, this low-dose infusion did not effectively activate the low levels of uterine OTRs. The sustained progesterone production would further serve to directly inhibit OTR signaling in the uterus and delay parturition (8). On the other hand, at higher doses, OT was able to precipitate labor, demonstrating that augmentation of uterine contractions before normal term labor is achievable despite low uterine receptor levels and elevated plasma progesterone concentrations if the level of OT is high enough.

We conclude that whereas the sole absence of OT does not significantly impact the timing of normal term labor in mice, abnormal modulation of OT action can affect the timing for the onset of labor. The lack of measurable alteration in the timing of term labor in OT KO mice may be, in part, explained by the elimination of OT's opposing roles in delaying and promoting labor, which are, in turn, determined by both OT and OTR concentrations. In this light, we propose that one role of the OT/OTR system during normal gestation may be to focus the timing for the onset of labor. Before the induction of uterine OTRs immediately preceding the onset of labor, the primary consequence of OT action would be to augment corpus luteum function and avoid premature labor. During luteolysis, OTR expression is reduced in the corpus luteum and rapidly induced in the uterine myometrium. This allows increased uterine sensitivity to OT without the further augmentation of luteal progesterone production that would occur if OT concentration simply increased globally. Thus labor would begin promptly and progress efficiently without further antagonism from progesterone.

Perspectives