The mechanism of prostaglandin action on the pregnant human uterus

The concentrations of 15 methyl PGF2α, progesterone and estradiol in the peripheral plasma were assayed sequentially and the resting and active pressures of the uterus were quantitated in 10 first trimester pregnant patients, treated with a vaginal suppository containing 3 mg U-36,384. The purpose of the study was to determine the sequence of the prostaglandin induced changes in regulatory profile and uterine function and thus expose further the mechanism of prostaglandin action.The temporal relationships of the changes revealed that the primary action of exogenous prostaglandin is the disruption of the normal endocrine function of the conceptus and that the delayed oxytocic effect of this compound is secondary, a consequence of the primary action. Apparently prostaglandins are only effective as postconceptional agents if they convert the refractory normal pregnant uterus into a reactive organ. The academic and therapeutic significance of this finding is discussed.     

Physiological studies on nonpregnant and pregnant uterus in experimentally diabetic rats]

Spontaneous intraluminal pressure waves of diabetic nonpregnant uterus and contractile responses to oxytocin and prostaglandin F2 alpha (PGF 2 alpha) of both diabetic nonpregnant and diabetic pregnant uterus were investigated in vitro. Diabetes was induced by streptozotocin (STZ), 60 mg/kg for nonpregnant and 50 mg/kg for pregnant rats. Frequency of spontaneous intraluminal pressure waves of nonpregnant uterus was reduced in diabetic rats when compared with normal, but amplitude was slightly larger in diabetic than in normal uterus. Pressure-volume curves revealed that the compliance of nonpregnant diabetic uterus was remarkably reduced. Normal tubal side-circular muscle was significantly more sensitive to oxytocin and PGF 2 alpha than cervical one in contractile responses. This tendency was lost in diabetic nonpregnant uterus. Contractile responses of both tubal and cervical circular muscles to oxytocin were lower in nonpregnant diabetic than in normal rats, but those of longitudinal muscles were higher in diabetic nonpregnant than in normal rats. Cervical circular muscle of pregnant diabetic rats was more sensitive to both agents than those of normal. However, contractile responses of diabetic longitudinal muscle to both agents were higher than those of normal as in the case of nonpregnant uterus. The mechanism of diabetic changes of the nonpregnant and pregnant uterus was discussed.     

The effect of luteectomy-induced progesterone-withdrawal on the oxytocin and prostaglandin response of the first trimester pregnant human uterus

In 9 normal 1st trimester pregnant patients strong stimulation with a single intravenous dose of 250 mU oxytocin, or 100 mc prostaglandin F2alpha (PGF2alpha), only provoked barely detectable and transient contractures, rather than advanced cyclic intrauterine pressure (IUP). Evidently, the normal early pregnant human uterus is not an "inert" muscle organ, but is "refractory" to stimulants. Luteectomy-induced rapid and continued progesterone(P)-withdrawals in 5 of the 9 patients "converted" these "refractory" uteri to "responsive" organs. This "conversion" was revealed by advanced cyclic IUP during the OR and PGR within 24-28 hours after luteectomy. Abortion occurred in 60.6+ or - 7.6 hours. "Conversion" was delayed in those 2 patients whose P-withdrawal was slow and who only aborted after 140.5 + or - 13.5 hours. The remaining 2 patients, whose luteectomy-induced P-withdrawal was moderate and transient showed no "conversion" and failed to abort. The rate and degree of P-withdrawal in these 3 groups of patients was a function of gestational age. Apparently the normal pregnant human uterus is a suppressed organ, refractory to stimulants. Continued P-withdrawal suspends suppression, lowers threshold and thus "converts" the refractory uterus to a reactive organ. Once "converted" the pregnant human uterus becomes spontaneously active under endogenous stimulation and expels its contents, a process which can be hastened by exogenous stimulants.     

Prostaglandin production by the pregnant and non-pregnant rabbit uterus

We have studied the prostaglandin synthesis of the pregnant and non-pregnant rabbit uterus in a microsomal membrane preparation, and in an ex vivo perfused uterus preparation which retains agonist stimulated prostaglandin production. In both the microsomal and isolated perfused system, prostacyclin was the major arachidonic acid metabolite produced; PGE2 was also produced in substantial quantities while TxB2 and PGF2 alpha were not detectable. Moreover, oxytocin was a specific stimulus of PGE2 release. The steroid hormone milieu influenced the level of agonist stimulated prostaglandin release; in general, ovariectomized, estrogen treated animals were more responsive to agonist stimulation than those treated with estrogen followed by progesterone. The microsomal studies indicated that the pregnant animal had a greatly enhanced capacity to metabolize arachidonic acid when compared with the non-pregnant animal. However, this was not reflected in the ability of agonists to stimulate prostaglandin release in the ex vivo perfused preparation.     

Gestational changes in the progesterone and prostaglandin F levels of the guinea-pig

In 87 guinea-pigs the gestational changes were measured in the progesterone (P) and prostaglandin F (PGF) levels of the peripheral and uterine vein plasmas, ovaries, uterus, placenta, fetal membranes and amniotic fluid. In the ovaries, the peripheral and uterine vein plasma, placenta and uterus, P-concentrations increase during early pregnancy and after a plateau decrease significantly as term approaches. In contrast, the uterine-vein PGF-levels remain low throughout pregnancy and only increase near term. Thus, in the guinea-pig, as in the classic species of P-action, normal pregnancy is characterized by high P and low PGF levels and labor by low P and high PGF levels. Of special interest are the additional findings that in the guinea-pig the uterine tissue P-levels are only a fraction of the peripheral plasma levels and the placental PGF-levels far exceed those of the uterus and fetal membranes. To promote the biological interpretation of the endogenous changes in the regulatory profile of the pregnant guinea-pig, current studies examine the functional consequences of the experimentally induced changes in P and PGF-levels.     

Catecholamines and contractility of the human myometrium at term: a possible role for prostaglandins

The contractile response of small myometrial specimens from the term pregnant human uterus was investigated using a superfusion technique. Adrenaline, noradrenaline and dopamine all had a stimulatory effect on the contractility. It was also demonstrated that this stimulatory effect was alpha-adrenoceptor mediated. If the tissue was pretreated with the prostaglandin synthetase inhibitor indomethacin or the arachidonic acid analogue eicosa-5,8,11,14-tetraenoic acid the effect of catecholamines was significantly reduced. This suggests a specific role of prostaglandins in the mechanism of action of catecholamines on the human myometrium.     

On the mechanism of midtrimester abortions induced by prostaglandin "impact"

Using Csapo's technique, a single dose of 24.3 +or- 1.1 mg prostaglandin F2alpha (PGF2alpha) had been delivered intraamniotically to 20 sedated pregnant patients (15.9 +or- 0.6 weeks pregnant) in order to provoke a PG impact (PGI), a consequent progesterone (P) withdrawal, and a conversion of the pharmacologically refractory normal pregnant uterus into a reactive organ. The side effects were occasional and acceptable and no further PGF2alpha treatment was needed except in 4 cases (5-10 mg). Only after the Oxytocin test showed that the uterus is becoming reactive, was 50 mU/minute oxytocin infused intravenously to facilitate the evolution of IUP to 93 +or- 3 mmHg and thus promote clinical progress. All the 20 patients aborted both the fetus and the placenta in 16.5 +or- 2.1 hours, but 8 women retained small placental residues to be removed by curettage. The Csapo score was a high 92 +or- 2. As early as 3 hours after PGI, the plasma P levels already decreased significantly. They continued to decline throughout the IAT and reached a 72% withdrawal when the fetus was aborted. 15 patients whose P withdrawal was rapid, aborted before the mean IAT, while those 5 women whose P withdrawal was slow aborted after this time. Thus, the rate of P withdrawal was direct while parity and gestational age indirectly related to the IAT. Studies are in progress to elucidate further the abortifacient action of PGF2alpha and through this knowledge promote predictable therapy.     

The effects of progesterone, prostaglandin F2α and oxytocin on the calcium-activation of the uterus

The relationship between “activator-calcium” (A-Ca), progesterone (P), prostaglandin F2α (PGF2α) and oxytocin (Oxy) has been examined in 100 uterine strips of 34 pregnant and 100 strips of 34 post partum rabbits. At the 25th day of gestation, uterine P was 13.9±1.3 ng/g, while within 3–12 hours post partum 3.3±0.3 ng/g tissue (P<0.001). Uterine strips, mounted isometrically in Krebs' solution, sustained maximum excitability in a steady state when exposed every 30 seconds for 4 seconds to an electric field of 12 V/5 cm (a.c.). The maximally contracting muscles were then rinsed at intervals of 6 minutes with Ca-free Krebs.In Ca-free Krebs, the post partum uterus lost 31% of its Ca and 96% of its excitability in a short 25 minutes, while the pregnant uterus lost 30% of its Ca and 93% of its excitability in 50 minutes (P<0.001). Since the extracellular space is 30% in the uterus, this 30% Ca, lost by both muscles, most probably was extracellular Ca and the small A-Ca fraction which is presumably “bound” more strongly at the membrane systems of the P-dominated pregnant, than the non-dominated post partum uterus. The significantly faster and more complete recovery from Ca-deficiency and inexcitability of the pregnant than the post partum uterus (P<0.001), at different levels of external Ca, further substantiates this premise. So does the demonstration that exposure to Ca-free Krebs increases ⁴⁵Ca-efflux 400% in the post partum and only 110% in the pregnant uterus (P<0.001). Exposure to 100 ng/ml PGF2α in normal Krebs has a similar effect on the ⁴⁵Ca-efflux of the post partum uterus, while the response of the pregnant uterus is indistinct (P<0.001).These highly significant differences between the post partum and the pregnant uteri in their Ca-efflux explain the higher threshold (P<0.001) and lower “sensitivity” to PGF2α and Oxy (P<0.001) of the pregnant than the post partum uterus. The already very highly significant differences between the two muscles, in threshold and sensitivity to these two most potent oxytocics, were increased still further by rendering the uterine strips Ca-deficient. All together, these findings substantiate the early contention (1–7,18,19) that uterine function at the cellular level is regulated by opposing actions of the suppressor P and the intrinsic stimulant PG or other oxytocic agents on threshold, excitability and the Ca-activation of the contractile process.     

Prostaglandin production by the pregnant and non-pregnant rabbit uterus

We have studied the prostaglandin synthesis of the pregnant and non-pregnant rabbit uterus in a microsomal membrane preparation, and in an perfused uterus preparation which retains agonist stimulated prostaglandin production. In both the microsomal and isolated perfused system, prostacyclin was the major arachidonic acid metabolite produced; PGE₂ was also produced in substantial quantities while TxB₂ and PGF_2α were not detectable. Moreover, oxytocin was a specific stimulus of PGE₂ release. the steroid hormone milieu influenced the level of agonist stimulated prostaglandin release; in general, ovariectomized, estrogen treated animals were more responsive to agonist stimulation than those treated with estrogen followed by progesterone. The microsomal studies indicated that the pregnant animal had a greatly enhanced capacity to metabolize arachidonic acid when compared with the non-pregnant animal. However, this was not reflected in the ability of agonists to stimulate prostaglandin release in the perfused preparation.     

Estrogen and progesterone receptors in the oviduct during egg transport in cyclic and pregnant rats

We investigated the temporal relationships between ovum transport and changes in the concentration of nuclear steroid receptors in the oviduct of cyclic and pregnant rats. A lack of parallelism between estrogen and progesterone fluctuations in plasma and their respective nuclear receptor concentrations in the oviduct predominated during egg transport. In pregnant animals, oviductal egg transport took 24 h longer than in nonpregnant animals. In both conditions, transport was initiated while the action of estrogen and progesterone on the oviduct--measured as nuclear receptor accumulation--was decreasing. Three or four days later, depending on whether the animal was pregnant, the eggs entered the uterus shortly after an increase in the nuclear receptor accumulation of both hormones. Treatment with RU486, a progesterone receptor-blocking agent known to cause premature arrival of eggs in the uterus, advanced estrogen receptor accumulation in the oviduct of pregnant rats. These data suggest that the arrival of eggs in the uterus is timed by a transitory increase in nuclear estrogen receptor in the oviduct that does not necessarily reflect a similar change of circulating estradiol. Moreover, in pregnant rats, the onset of this estrogenic action is delayed by a progesterone receptor-mediated effect that hinders nuclear estrogen receptor accumulation.     

Changes in cholinergic and noradrenergic nerves in the pregnant and postpartum uterus of the albino rat and guinea pig

The present study was undertaken to investigate the structural changes in both cholinesterase (ChE)-positive nerve fibers and adrenergic nerves with formaldehyde-induced fluorescence in pregnant and postpartum uteri of both the albino rat and guinea pig. Particular attention was directed to the relationship between these changes and the local factors associated with the growing fetus. ChE reaction was absent in the control and pregnant uterus of the guinea pig. In the albino rat, there were signs of degeneration in pregnancy. These were evidenced by vacuolation of large nerve trunks and the presence of focal segments with very faint reaction along the course of the nerve bundles. Myometrial segments from fetus-containing horns showed some fragmented nerve fibers, but at the same time some other normal ones. Most of the fine nerve bundles gave a weak reaction. Three weeks after delivery, multiple ChE fibers were found in the uterus of the albino rat. The normal appearance was, however, not regained and some nerve fibers were still fragmented. Noradrenergic (NA) nerve fibers were disintegrated and markedly reduced in number in the myometrium of the pregnant uterus of both the guinea pig and albino rat, particularly in the uterine horns that were distended by fetuses. The number of NA fibers was not significantly reduced in the tubal ends of the albino rat uterus. Three weeks after delivery, normal NA fibers were seen in the myometrium of both the albino rat and guinea pig uterus. Nerves with reduced fluorescence reaction were observed less frequently.     

Phospholipid Containing Choline Histochemistry of Mouse Uterine Epithelia during Preimplantation Stage

The physiological role of high lipid content in endometrial cells during pregnancy has not been well established. In the present work we used histochemical techniques to analyze the total lipids and phospholipid containing choline (PCC) in the mouse uterine glandular and luminal epithelia during preimplantation stage. Sudan black histochemistry showed the highest intensity during the second day of pregnancy in both the basal and apical portions of luminal epithelium. Peaks of PCC staining were seen both in the luminal and glandular epithelia at the second and fifth days of pregnancy. Changes in localization and in the amount of lipid in the uterine epithelia suggest high mobility and metabolic rates of this substance, which maybe related to morphological and/or functional changes occurring at the same time in the pregnant uterus. The increase and depletion timing of PCC content in the uterine epithelia during preimplantation stage, when uterine prostaglandin is also oscillating, suggest a possible involvement of PCC in prostaglandin biosynthesis. Therefore, the fate of lipid droplets found in the uterine epithelia may be related to critical changes of the pregnant endometrium, rather than the nourishment of developing embryos alone.     

Fetal stimulation of maternal immunoglobulin production in mice

Within 12-24 h of parturition in mice, there was a dramatic increase in the number of immunoglobulin secreting cells in the paraaortic lymph nodes (PALN) draining the pregnant uterus. Compared with stimulation with lipopolysaccharide the ratio of IgG:IgM forming cells was very high in PALN draining a pregnant uterus. The response was eliminated when fetectomy (ablating the embryo but leaving the placenta intact) was carried out on the 12th day of pregnancy. With unilateral fetectomy the uterine horn with intact fetal/placental units can be used as a positive control since lymphoid drainage is laterally confined. Neither healthy (gross and histological criteria) nor partly necrotic placentae stimulated Ig secreting cells in the PALN. The placentae of bilaterally fetectomized females were delivered apparently normally and at about the same time as normal (control) fetuses. Injection of prostaglandin E-2 or F-2 alpha into the tail base led to the appearance of Ig-forming cells in the PALN of normal (virgin) female mice. Indomethacin fed to the pregnant female greatly reduced the numbers of these cells in the PALN. We conclude that the observed local stimulation of maternal Ig production by the fetus may be involved in the transplacental transfer of Ig from mother to fetus.     

Fetal viability does not affect the onset of stretch-induced labor and the increase in amniotic fluid prostaglandin F2 alpha and plasma prostaglandin F2 alpha metabolite levels.

The role of the fetus in the onset and progress of stretch-induced labor and in the change in amniotic fluid prostaglandin F2 alpha and plasma prostaglandin F2 alpha metabolite levels was evaluated in six normal pregnant women (group 1) and six women whose fetuses had been dead for more than one week (group 2). The uterus was distended by a balloon inflated with physiologic saline. Regular uterine contractions occurred, and increased in all patients. Within 21 hours, all patients delivered a normal baby in group 1 and a macerated fetus in group 2. There was no significant difference in induction-delivery interval between the two groups. Both groups showed a significant and similar range of increases in the levels of amniotic fluid prostaglandin F2 alpha and plasma prostaglandin F2 alpha metabolite during treatment (P less than 0.001). Thus, the fetus has no functional role in the onset and progress of stretch-induced labor or in the rise of amniotic fluid prostaglandin F2 alpha and plasma prostaglandin F2 alpha metabolite levels.     

Uterine vein prostaglandin levels in late pregnant dogs.

We studied the uterine venous plasma concentrations of prostaglandins E2, F2 alpha, 15 keto 13,14 dihydro E2 and 15 keto 13,14 dihydro F2 alpha in late pregnant dogs in order to evaluate the rates of production and metabolism of prostaglandin E2 and F2 alpha in pregnancy in vivo. We used a very specific and sensitive gas chromatography-mass spectrometry assay to measure these prostaglandins. The uterine venous concentrations of prostaglandin E2 and 15 keto 13,14 dihydro E2 were 1.35 +/- .27 ng/ml and 1.89 +/- .37 ng/ml, respectively; however, we could not find any prostaglandin F2 alpha and very little of its plasma metabolite in uterine venous plasma. Since uterine microsomes can generate prostaglandin F2 alpha and E2 from endoperoxides, prostaglandin F2 alpha production in vivo must be regulated through an enzymatic step after endoperoxide formation. Prostaglandin E2 is produced by pregnant canine uterus in quantities high enough to have a biological effect in late pregnancy; however, prostaglandin F2 alpha does not appear to play a role at this stage of pregnancy.     

Microarray analysis of uterine gene expression in mouse and human pregnancy

Improved care of infants born prematurely has increased their survival. However, the incidence of preterm labor has not changed. To understand the processes involved in preterm labor, we used oligonucleotide microarrays to study gene expression in murine and human uterus during pregnancy. The induction of enzymes for prostaglandin synthesis was used as a marker for important changes during pregnancy because prostaglandins strongly contribute to both human and murine labor. We identified 504 genes that changed at least 2-fold between d 13.5 and 19.0 in the gravid mouse uterus. In the pregnant human myometrium, we found 478 genes that changed at least 2-fold in either term or preterm labor compared with preterm nonlabor specimens and 77 genes that significantly varied in both preterm and term labor. Patterns of gene regulation within functional groups comparing human preterm and term labor were similar, although the magnitude of change often varied. Surprisingly, few genes that changed significantly throughout pregnancy were the same in the mouse and human. These data suggest that functional progesterone withdrawal in human myometrium may not be the primary mechanism for labor induction, may implicate similar mechanisms for idiopathic preterm and term labor in humans, and may identify novel targets for further study.     

Cooxidation of steroidal and non-steroidal estrogens by purified prostaglandin synthase results in a stimulation of prostaglandin formation

Estrone (E1), estradiol (E2), the catechol estrogens 2-OHE1 and 2-OHE2, and diethylstilbestrol (DES) were incubated with purified prostaglandin synthase (PHS) in vitro in the presence of arachidonic acid and their PHS-catalyzed cooxidation was determined. 2-OHE1, 2-OHE2, and DES were extensively metabolized by PHS peroxidase activity, E1 and E2 to a lesser extent. The cooxidation of the estrogens is accompanied by an increased prostaglandin formation and an increase in cyclooxygenase activity in vitro; progesterone and nylestriol are without effect. Prostaglandins have been proposed to play a role in events related to early estrogen action in tissues such as the uterus. The cooxidation of estrogens and their metabolites by prostaglandin hydroperoxidase might represent one type of interaction between the hormones and the arachidonic acid cascade that could lead to changes in prostaglandins.     

Decrease of utero-placental blood flow during prostaglandin F2α induced abortion

Pregnancy had been terminated in 6 normal midtrimester pregnant patients by the extraovular injection of 10 mg prostaglandin F2α (PGF2α). In these 6 Experimental and 3 Control patients utero-placental blood flow had been measured, by changes in the density of radioactive Indium, distributed over the uterine area, as a function of time. In comparison with Controls utero-placental blood flow decreased in the Experimental patients already at 5 minutes after PG-treatment, long before advanced cyclic IUP evolved. This finding substantiates the conclusion (1–3), based on experiments in animal “models”, that decrease in utero-placental blood flow is an early step in the mechanism of PG action.     

Pregnancy-induced alterations of vascular function in mouse mesenteric and uterine arteries

Normal pregnancy involves dramatic changes to maternal vascular function, while abnormal vascular adaptations may contribute to pregnancy-associated diseases such as preeclampsia. Many genetic mouse models have recently emerged to study vascular pathologies of pregnancy. However, vascular adaptations to pregnancy in normal mice are not fully understood. Thus, we studied changes in vascular reactivity during normal mouse pregnancy. We hypothesized that pregnant mice will have enhanced endothelial-dependent vasodilation compared with nonpregnant mice, via an enhancement of the nitric oxide synthase (NOS) prostaglandin H synthase (PGHS), and other endothelial-derived hyperpolarizing pathways. Late pregnant (Day 17-18) C57BL/6J mice (n = 10) were compared with nonpregnant mice (n = 7). Uterine and mesenteric arteries were mounted on a wire myograph system and assessed for endothelium-dependent (methacholine) and -independent (sodium nitroprusside; SNP) relaxation responses. Endothelial-dependent relaxation was enhanced in pregnant uterine and mesenteric arteries, which was blunted after the addition of inhibitors of the PGHS or NOS pathways. In nonpregnant mice, these pathways had no effect in modulating relaxation in uterine arteries, whereas vasodilation in mesenteric arteries was reduced only by NOS inhibition. Both uterine and mesenteric vessels had nonnitric oxide- and nonprostaglandin-mediated relaxation, but this relaxation was not enhanced during pregnancy. Endothelial-independent relaxation was also enhanced in pregnant uterine but not mesenteric arteries. Our data indicate that uterine and mesenteric arteries from pregnant mice have enhanced vasodilation. Understanding vascular adaptations to normal mouse pregnancy is crucial for interpreting changes that may occur in genetic mouse models.     

Microsomal lipid peroxidation in human pregnant uterus and placenta

A study was made of the microsomal lipid peroxidation of the pregnant human uterus and placenta. It was found that the lipid peroxidation of the microsomal fraction of the uterus is specific for prostaglandin formation: the lipid peroxidation was enhanced by arachidonic acid, and inhibited by anti-prostaglandins. Accordingly, it is suitable as a screening test for the pharmacological examination of anti-prostaglandin effects. The lipid peroxidation in the placenta is not specific. In both tissues examined the lipid peroxidation is linked to ascorbic acid.