Uterine contractions are usually strengthened by oxytocin, prostaglandin E2 (PGE2) and prostaglandin F2$\alpha$ (PGF2$\alpha$) through their actions on oxytocin receptors (OXTR), PGE2 receptors (EP) and PGF2$\alpha$ receptors (FP), respectively [1, 2, 3]. At term, PGE2 mediates myometrial contraction mainly through the EP3 receptor subtype [4]. Generally recommended second-line agents for the treatment of uterine atony are misoprostol, a known EP3/EP2 agonist [2], and carboprost, which is a synthetic analog of PGF2$\alpha$ [5].

Postpartum hemorrhage (PPH) has an incidence of 4% and contributes to nearly one quarter of all maternal deaths worldwide [6], most of which occur due to uterine atony [7]. Oxytocin-induced or augmented labor contributes to uterine atony by causing desensitization of the myometrium to oxytocin [8, 9]. Currently, the treatment of uterine atony caused by oxytocin exposure is empirical [10, 11]. Therefore, the aim of the present study was to investigate changes in the receptors involved in myometrial contraction following OXTR desensitization, thus providing an experimental basis for the treatment of uterine atony.

Previous studies showed that OXTR expression was unable to increase at term in mice due to lack of the FP gene [12, 13]. Subsequent studies showed that treatment with PGF2$\alpha$ during pregnancy increased OXTR expression in human myometrial cells from the lower uterine segment and decreased it in the upper segment [14]. A recent study suggested that OXTR is an upstream regulator of cyclooxygenase-2, which facilitates the conversion of arachidonic acid to PGE2 and PGF2$\alpha$ [14]. OXTR antagonists inhibit PGF2$\alpha$-induced contractions and inflammatory responses in the human myometrium [15]. These findings suggest that OXTR and FP expression and function may interact directly or indirectly in the perinatal uterus.

We hypothesized that desensitization of OXTR would result in compensatory increases in the expression and activation of FP and EP3. We also hypothesized that the effects of prostaglandins and oxytocin on OXTR expression might be different in myometrium with OXTR desensitization. The primary goal of this in vitro study was therefore to explore changes in EP3, FP and OXTR expression in late-pregnancy rat myometrium following pretreatment with oxytocin, as well as their activation effects. The second goal was to investigate the effect of prostaglandins on OXTR expression.

The contractile activity of the myometrium was calculated as the average tension (g) $\times$ frequency (contractions/15 min). The ratio of contractility was used for statistical analysis and was calculated as the contractile activity at each drug concentration divided by the spontaneous activity. The spontaneous activity of each myometrium strip was calculated during 15 min of stable contractions prior to the addition of 10${}^{-6}$ M oxytocin/PSS.

All data are expressed as mean $\pm$ standard deviation (SD). The concentration-response curve was obtained by taking drug concentrations as the abscissa and ratio of contractility as the ordinate variable. The curve was fitted using Prism 8 (GraphPad Prism Software, San Diego, CA, USA) . The Student’s t-test was used to compare differences between the control and experimental groups for each uterotonic agent. The potency pEC50 (negative logarithm of the molar concentration required to elicit 50% maximum contraction response) and efficacy (maximum response [Emax(ratio)]) of each drug were compared between groups with one-way analysis of variance (ANOVA) and Dunnett’s post hoc (2-sided) test using SPSS Statistics 22 software (IBM, Armonk, NY, USA). In all cases, differences were considered significant at a P value of 0.05.

Myometrial contractile amplitude and frequency showed no significant changes during continuous exposure to PSS for 2 hours (Fig. 3A1). The response to 10${}^{-7}$ M oxytocin before and after this process also showed no difference. However, the baseline level, amplitude, and frequency of myometrial contraction gradually decreased during 2 hours of 10${}^{-6}$ M oxytocin treatment (Fig. 3A2). The contraction response of myometrium to 10${}^{-7}$ M oxytocin was significantly weaker than prior to the 10${}^{-6}$ M oxytocin pretreatment (P = 0.005) (Fig. 3B). The mRNA expression levels of EP3, FP and OXTR did not change significantly with and without oxytocin pretreatment (Fig. 3C1–C3).

Fig. 3.

Experiment 1: Receptor mRNA expression in myometrium from the experimental group (with oxytocin pretreatment) and from the control group (without oxytocin pretreatment). (A) Representative isometric tension recordings for the control group (A1) and for the experimental group (A2). The time scale was suitable for Fig. A1 and Fig. A2. (B) Myometrial contractile activity before and after desensitization with 10${}^{-6}$ M oxytocin or PSS for 2 hrs of contraction, stimulated by 10${}^{-7}$ M oxytocin. # statistically significant, P = 0.005 (paired-samples T test). (C) mRNA expression levels for the PGE2 receptor (EP3; C1), PGF2$\alpha$ receptor (FP; C2), and oxytocin receptor (OXTR; C3) were measured and normalized to the housekeeping gene GAPDH. No statistical differences in the mRNA levels of these receptors were observed between the control and experiment groups. Data are mean $\pm$ SD (whiskers). OT, oxytocin; PSS, physiological salt solution.

Without oxytocin pretreatment, the concentration-response curve for misoprostol was flat. However, with oxytocin pretreatment the myometrial contractility increased rapidly between misoprostol concentrations of 10${}^{-10}$ to 10${}^{-7}$ M (Fig. 4). Misoprostol had a more obvious contractile effect on the myometrium after oxytocin pretreatment compared to the control (mean (SD) Emax(ratio): 4.44 $\pm$ 1.47 vs 1.32 $\pm$ 0.22, P = 0.02). There was no significant difference in the maximum myometrial contraction effect produced by carboprost with or without oxytocin pretreatment. However, carboprost significantly enhanced the contractile potency of myometrium pretreated with oxytocin compared to the control (pEC50: 7.74 $\pm$ 0.56 vs 6.81 $\pm$ 0.25, P = 0.03). At any oxytocin concentration, myometrial contractility induced by oxytocin in the experimental group was significantly lower than that observed in the control group (Emax(ratio): 1.62 $\pm$ 0.27 vs 2.82 $\pm$ 0.98, P = 0.015). Myometrial contractile activity in the control group was greater than that of the oxytocin pretreatment group at any time in PSS (Table 1).

In the control group, the maximum contractile effect of oxytocin on myometrium was significantly greater than for misoprostol, but was not significantly different to that of carboprost. However, the contractile potency (pEC50) of oxytocin on myometrium was significantly greater than that of carboprost. In the experimental group, the potency of oxytocin on myometrial contraction was significantly greater than that of both misoprostol and carboprost. However, there was no significant difference in the maximum contractile effect between the three groups.

After completing the contraction tests stimulated by uterotonics, the myometrial strips for the experimental group were evaluated for OXTR expression at the mRNA and protein levels. Compared to continuous oxytocin exposure, OXTR expression increased significantly following equilibration in PSS. There was no significant difference in OXTR expression in myometrium between the misoprostol, carboprost and oxytocin treatment groups. Moreover, there was no statistical difference in OXTR expression between the PSS treatment and misoprostol or carboprost treatment groups (Fig. 5).

Fig. 4.

Experiment 2. Concentration-response curves for misoprostol (A), carboprost (B), oxytocin (C) and PSS (D) in late-pregnancy rat myometrium, with or without oxytocin pretreatment. Contractile responses are shown as ratios of contractility. The ratio of contractility was calculated as the contractile activity at each drug concentration divided by the baseline spontaneous activity. Data are mean $\pm$ SD (whiskers). The difference in ratio between the control group (n = 6) and oxytocin pretreatment group (n = 6) was compared using T-test at each drug concentration (at each period in the PSS group). “*” represents 0.01 P 0.05, “**” represents 0.001 P 0.01, “***” represents P 0.001.

Fig. 5.

Experiment 2. Oxytocin receptor (OXTR) expression. Expression of the housekeeping gene GAPDH was used as an internal control to assess the expression of OXTR mRNA following treatment of myometrium with oxytocin, misoprostol, or carboprost (A). Whole-tissue lysates were subjected to Western blotting. GAPDH served as a loading control, and blots were scanned for densitometric analysis (B). Representative Western blots are also shown. Data are mean $\pm$ SD (whiskers). Significant differences (A,B) between groups are illustrated by different lowercase letters above each bar; groups sharing the same letter did not differ (one-way analysis of variance, Bonferroni post hoc comparison). PSS, physiological salt solution.

Following pretreatment in vitro with 10${}^{-6}$ M oxytocin for 2 hours, the contractile response of myometrium to oxytocin decreased significantly, although mRNA expression for OXTR, FP and EP3 did not change. After oxytocin pretreatment, the sensitivity of myometrium to carboprost increased, but the maximal contraction induced by carboprost did not change. Misoprostol had an obvious contractile effect on myometrium pretreated with oxytocin, but was inactive for myometrial contraction in the control group. Following the desensitization of OXTR, continuous oxytocin exposure reduced OXTR expression compared to equilibrium in PSS. Furthermore, there were no statistically significant differences between the PSS, carboprost and misoprostol groups for OXTR mRNA expression.

Oxytocin acts directly on phosphoinositidase C-linked G protein coupled receptors (OXTR) to increase cytosolic Ca${}^{2+}$ to strengthen myometrial contraction. Many literatures as well as our experimental results demonstrated that oxytocin exposure would induce desensitization of OXTR. In the research of Phaneuf et al. [21], oxytocin exposure decreased the binding of oxytocin to cell membranes, however flow cytometry experiments demonstrated that OXTR were not internalized during this treatment. The second messengers calcium, inositol phosphates (InsPs) and cyclic nucleotides play decisive roles in uterine contractility [22]. In a previous study, oxytocin-induced desensitization did not change the ability of PGF2$\alpha$ to increase intracellular free calcium, did but change such ability for oxytocin [17]. The present study confirms that FP expression was not affected by oxytocin pretreatment, and that the myometrial contractile potency of carboprost became stronger. One possible explanation is that oxytocin-induced desensitization acts only on OXTR levels without affecting post-receptor signaling. Earlier studies involving human and rat myometrial tissues also showed that pretreatment with oxytocin decreased the response to subsequent oxytocin exposure, but the myometrium consistently responded to PGF2$\alpha$ stimulation [17, 18].

As shown in Fig. 3, treatment with 10${}^{-6}$ M oxytocin for 2 hours caused OXTR desensitization and the response to 10${}^{-7}$ M oxytocin before and after this process showed significant differences (A2, B). However, the level of OXTR mRNA did not change (C3). In the oxytocin pretreatment group, OXTR expression decreased significantly following continuous exposure to oxytocin compared with equilibration in PSS (Fig. 5). We speculate that continuous use of oxytocin initially decreases the response of myometrium to oxytocin, while the decreased expression of OXTR mRNA needs to further prolong the time of oxytocin treatment. Although it is believed that oxytocin/OXTR signaling is essential, numerous studies have shown that OXTR expression does not correlate with the oxytocin-induced uterine contraction effect. A recent study showed that peaks in oxytocin level during labor did not correlate with the time of uterine contractions [23]. Another study showed that late-pregnancy uterine contractions in mice are mainly controlled by modification of the contractile signal machinery rather than by the level of OXTR [24]. The decrease in OXTR mRNA in myometrium accompanied by the decline in response to oxytocin might be related to the longer time of oxytocin pretreatment [21].

A previous study showed that EP3 receptor-deficient mice have normal parturition [25]. Similarly, the present study found the EP3 agonist misoprostol had no contractile effect on late-pregnancy rat myometrium in the control group. However, we also found that misoprostol had a significant myometrial contractile effect following oxytocin-pretreatment, even though EP3 mRNA did not increase. Misoprostol is a mixed EP3/EP2 receptor agonist [26]. During human pregnancy, myometrial EP3 receptors are excitatory while EP2 receptors are inhibitory [27]. Therefore, we speculate the myometrial contractile effect of misoprostol following oxytocin-pretreatment may be due to inhibition of EP2 expression or activity. In contrast to the present results, Balki et al. [28] showed that responses to PGF 2$\alpha$ and to misoprostol were not affected by labour or by prior exposure to oxytocin. A possible reason for the discordant results is that the myometrium in their research was balanced in PSS for 2 hours before administering uterotonics and was not exposed to oxytocin, thus causing the myometrium to re-sensitize to oxytocin [29].

Following OXTR desensitization, both the contractile potency of carboprost and the maximum contractility of misoprostol increased significantly. This suggests it was reasonable and necessary to use FP and EP3 receptor agonists to enhance myometrial contraction after OXTR desensitization. Following pretreatment with oxytocin, OXTR expression was significantly higher in the PSS group compared to the oxytocin exposure group, while OXTR expression in the carboprost and misoprostol groups was not significantly different compared to the PSS group. A previous study on human myometrium also demonstrated that oxytocin combined with carboprost produced a better contractile effect compared to oxytocin alone following oxytocin pretreatment [30]. Recent literature has also reported that misoprostol plus oxytocin was a more effective strategy for preventing PPH than oxytocin alone [10, 31]. Therefore, combinations or alternate uses of oxytocin, FP and EP3 receptor agonists may be more effective at strengthening myometrial contraction.

Balki et al. [30] reported there was incomplete information on the plasma levels of uterotonics following parenteral administration in the setting of PPH. They suggested that oxytocin levels vary significantly during pregnancy, labor and postpartum from 10${}^{-12}$ to 10${}^{-8}$ M and may not accurately reflect the local myometrial concentration [32, 33, 34, 35]. Because the release of oxytocin during physiological labor is pulsatile, the measured values of serum oxytocin concentration vary greatly depending on the sampling intervals. Oxytocin levels doubled in response to a doubling of the infusion rate of exogenous oxytocin [23]. Thus, accurate plasma levels of oxytocin are difficult to define. The approximate peak serum level of carboprost after intramuscular injection of 250 $\mu$g was approximately 10${}^{-8}$ M in term pregnant women [36]. Misoprostol is commonly administered into the vagina and hence the plasma level of this drug may differ significantly from the concentrations to which the uterine smooth muscle is exposed.

Although the in vitro drug concentrations used in the present study may not directly reflect serum levels in vivo, the concentration range (10${}^{-10}$ to 10${}^{-5}$ M) investigated here may well include the physiologic serum levels of these drugs. Morrison et al. [20] provided a detailed pharmacodynamic analysis of uterotonics on isolated myometrium from women undergoing elective cesarean delivery at term. Their results showed similar pEC50 for oxytocin and carboprost to the present study, while misoprostol was also inactive in myometrial contraction. Hence, the previous literature suggests that rat and human myometrial tissues have similar pharmacodynamic responses to uterotonics.

The present study of the receptor activation effect confirmed our hypothesis that carboprost and misoprostol enhance myometrial contraction following OXTR desensitization. However, we did not further explore the possible mechanism of OXTR desensitization, such as the oxytocin-induced effect of increased cytoplasmic free calcium concentration before and after receptor desensitization. Pretreatment with oxytocin for 2 hours increased EP3 mRNA expression, although this did not reach statistical significance. It is unclear whether prolonging the duration of oxytocin exposure would lead to significantly increased EP3 mRNA expression. Misoprostol also acts on EP2 to inhibit myometrial contraction [27], but we did not further investigate the effect of oxytocin pretreatment on EP2. In vitro experiments based on animal tissues have inherent limitations, such as species variation and differences between in vivo and in vitro environments.

In summary, we found that following pretreatment with 10${}^{-6}$ M oxytocin for 2 hours, the response ability of myometrium to subsequent oxytocin decreased significantly. However, the mRNA expression levels of EP3, FP and OXTR were not affected by pretreatment with oxytocin, whereas the contractile potency of carboprost and the maximal contractile effect of misoprostol both increased significantly. No significant difference in OXTR expression was observed between oxytocin and prostaglandin treatments. Misoprostol was inactive in normal myometrial contraction. These experimental results suggest that prostaglandin uterotonics, especially misoprostol, may have desirable therapeutic effects on uterine atony caused by long-time oxytocin exposure. Further studies are required to confirm this hypothesis.

LL and SH designed the research study. LL and JH performed the research. TW analyzed the data. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.

All animal experiments were approved by the Animal Ethics Committee of the Shanghai Medical College, Fudan University [Approval number: 201907007Z].

Thanks for the department of physiology & pathophysiology of Fudan University Shanghai Medical College for providing laboratory. Thanks to the researchers of this laboratory for their guidance on experimental technology.

This study was funded by the General Foundation of Shanghai Municipal Health Commission (201840327).

The authors declare no conflict of interest.