Phenanthroline relaxes uterine contractions induced by diverse contractile agents by decreasing cytosolic calcium concentration

Abstract

Uterine contractions during labor and preterm labor are influenced by a complex interplay of factors, including hormones and inflammatory mediators. This complexity may contribute to the limited efficacy of current tocolytics for preterm labor, a significant challenge in obstetrics with 15 million cases annually and approximately 1 million resulting deaths worldwide. We have previously shown that the myometrium expresses bitter taste receptors (TAS2Rs) and that their activation leads to uterine relaxation. Here, we investigated whether the selective TAS2R5 agonist phenanthroline can induce relaxation across a spectrum of human uterine contractions and whether the underlying mechanism involves changes in intracellular Ca²⁺ signaling. We performed experiments using samples from pregnant women undergoing scheduled cesarean delivery, assessing responses to various inflammatory mediators and oxytocin with and without phenanthroline. Our results showed that phenanthroline concentration-dependently inhibited contractions induced by PGF2α, U46619, 5-HT, endothelin-1 and oxytocin. Furthermore, in hTERT-infected human myometrial cells exposed to uterotonics, phenanthroline effectively suppressed the increase in intracellular Ca²⁺ concentration induced by PGF2α, U46619, oxytocin, and endothelin-1. These results suggest that the selective TAS2R5 agonist may not only significantly reduce uterine contractions but also decrease intracellular Ca²⁺ levels. This study highlights the potential development of TAS2R5 agonists as a new class of uterine relaxants, providing a novel avenue for improving the management of preterm labor.

Graphical Abstract

1. Introduction

Appropriate uterine contraction is vital for normal reproduction, encompassing processes that include menstruation, embryo implantation, and parturition (Bulletti and de Ziegler, 2005; Bulletti et al., 2000; Rosen and Yogev, 2023; Smith, 2007). Abnormalities in uterine contraction underlie or contribute to a wide range of gynecological and obstetric disorders, including endometriosis, adenomyosis, dysmenorrhea, uterine dystocia, and preterm birth (LeFevre et al., 2021; Leyendecker and Wildt, 2011; MacGregor et al., 2023; Romero et al., 2014). Consequently, the identification of uterine contraction agonists and relaxants holds critical importance in both gynecology and obstetrics.

Preterm birth presents a significant challenge in obstetrics, with fifteen million cases annually and approximately one million resultant deaths worldwide (Blencowe et al., 2013; Chawanpaiboon et al., 2019). Tocolysis, the relaxation of uterine smooth muscle, is commonly employed to manage spontaneous preterm labor/birth (Arman et al., 2023; Coler et al., 2021; Di Renzo et al., 2018; Hanley et al., 2019; Munoz-Perez et al., 2019; Wilson et al., 2022; Wray et al., 2023), which accounts for 50% of all preterm births. While existing tocolytics do not directly improve neonatal outcomes, they serve three critical purposes: 1) providing the fetus additional developmental and maturation time in utero, 2) permitting time for the administration of prenatal corticosteroids, which facilitate fetal organ maturation, and 3) enabling the transfer of the pregnant woman to a hospital equipped for higher level neonatal care (Arman et al., 2023; Coler et al., 2021; Di Renzo et al., 2018; Hanley et al., 2019; Munoz-Perez et al., 2019; Wilson et al., 2022; Wray et al., 2023). Despite these benefits, current tocolytic agents demonstrate limited effectiveness, particularly beyond providing a 48-hour delay. This underscores a significant unmet need and offers an opportunity to identify new classes of uterine relaxants.

The development of effective tocolytics is challenged by the complex biological processes entailed in parturition. Successful parturition necessitates the activation of at least four common pathways: cervical ripening, membrane decidual activation, rupture of the fetal membranes, and the transition from a quiescent uterus during pregnancy to a powerfully contractile organ during labor (Challis et al., 2000; Leimert et al., 2021; Menon et al., 2016; Romero et al., 2014). Evidence suggests that premature activation of uterine transition can instigate preterm labor (Challis et al., 2000; Leimert et al., 2021; Menon et al., 2016; Romero et al., 2014). Moreover, research over recent decades has revealed that, alongside hormonal influences and mechanical stretch, intrauterine inflammation plays a pivotal role in catalyzing the uterine transition in both term and preterm labor (Bonney and Johnson, 2019; Gilman-Sachs et al., 2018; Gimeno-Molina et al., 2022; Leimert et al., 2021; Tersigni et al., 2020; Zhang and Wei, 2021).

Throughout late gestation and labor, the uterus experiences an influx of immune cells that release a variety of cytokines and chemokines (Bonney and Johnson, 2019; Gilman-Sachs et al., 2018; Gimeno-Molina et al., 2022; Tersigni et al., 2020; Zhang and Wei, 2021). These mediators activate signaling cascades that augment the expression of contractile proteins, such as connexin43, prostaglandin F2 alpha (PGF2α) receptor, and oxytocin receptor. Additionally, certain inflammatory mediators can directly induce uterine contractions. Compounds such as prostaglandins, endothelin-1, and serotonin or 5-hydroxytryptamine (5-HT) have been implicated as key contributors to both term and preterm labor (Breuiller-Fouche et al., 2005; Bytautiene et al., 2008; Chibbar et al., 1993; Cordeaux et al., 2008; Fischer et al., 2008; Garfield et al., 2006; Hirst et al., 1995; Kim et al., 2017; Koren et al., 1965; Olund et al., 1980; Osada et al., 1997; Peiris et al., 2021; Romero et al., 1994; Romero et al., 1987; Uvnas-Moberg et al., 2019; Wang et al., 2008; Yallampalli and Garfield, 1994).

Given the pivotal role of inflammatory responses in parturition, it has become an attractive target for research into the prevention of preterm labor prevention. Hypothetically, blocking uterine contractions instigated by inflammatory mediators could halt the progression of spontaneous preterm labor. However, due to the array of inflammatory mediators that induce uterine contractions, preventing preterm uterine activity would likely necessitate an agent capable of reversing or preventing uterine contractions caused by various inflammatory mediators.

Bitter taste, one of five basic taste qualities, is critical to the survival of animals including humans, as it promotes avoidance of harmful toxins and noxious substances (Behrens and Meyerhof, 2009; Chandrashekar et al., 2000). It was long believed that bitter taste receptors, also known as Taste 2 receptors (TAS2Rs) which initiate the sensation of bitterness, were only present in the specialized epithelial cells in the taste buds of the tongue (Chandrashekar et al., 2000; Wong et al., 1996). However, these receptors have increasingly been found in extra-oral organs and participate in a variety of biological activities (Avau and Depoortere, 2016; Behrens and Lang, 2022; Bloxham et al., 2020; Carey et al., 2016; Harmon et al., 2021; Lu et al., 2017; Tuzim and Korolczuk, 2021).

We recently discovered that TAS2Rs and their downstream signaling components are expressed in uterine smooth muscle (Zheng et al., 2017). Strikingly, we found that chloroquine, an agonist of multiple TAS2Rs, more effectively relax mouse and human uterine smooth muscle - which has been pre-contracted by membrane depolarization or oxytocin - than other clinically-used tocolytics like the non-steroidal anti-inflammatory drug indomethacin and the calcium channel blocker nifedipine. In mice, chloroquine also provides better prevention of preterm labor induced by lipopolysaccharide and the progesterone receptor antagonist RU486 than other tocolytics (Zheng et al., 2017).

However, bitter tastants, such as chloroquine which activates multiple TAS2Rs (Meyerhof et al., 2010), might lead to increased side effects. This is because TAS2Rs are widely expressed across various organs, encompassing both reproductive and non-reproductive systems (Avau and Depoortere, 2016; Behrens and Lang, 2022; Bloxham et al., 2020; Carey et al., 2016; Harmon et al., 2021; Lu et al., 2017; Tuzim and Korolczuk, 2021). Interestingly, we have found that phenanthroline, a bitter tastant that selectively activates TAS2R5 agonist (Meyerhof et al., 2010), can reverse human uterine contractions induced by membrane depolarization and oxytocin (Zheng et al., 2017). In this study, we tested whether phenanthroline can relax uterine contraction mediated by different inflammatory mediators, considering their significant role in uterine contraction during parturition and/or preterm labor. If this selective TAS2R5 agonist proves effective, we aimed to determine whether a common process underpins this broad uterine relaxation effect. We focused on PGF2α, oxytocin, U46619 (a thromboxane A2 (TP) receptor agonist), 5-HT, and endothelin-1 as uterotonics due to the well-established roles these compounds or their receptors play in the term and/or preterm labor pathway. Furthermore, we delved into Ca²⁺ signaling as a potential unified mechanism for the uterine relaxation effect of bitter tastants, given that Ca²⁺ is the primary signal for uterine contraction (Aguilar and Mitchell, 2010; Sanborn, 2000; Wray and Arrowsmith, 2021).

2. Materials and Methods

2.1. Study design and cohort

Pregnant women who underwent a scheduled cesarean delivery for clinical indications at UMass Memorial Medical Center were screened and consented to participate in uterine specimen donation. Patients with blood-borne infections, including hepatitis B, hepatitis C, and HIV, were excluded. At the time of surgery, none of the women had used tocolytics. This study was approved by the UMass Chan Medical School Institutional Review Board (IRB#: H00014516).

2.2. Specimen collection

Uterine tissues, approximately 5 cm × 1.5 cm × 3 cm in size, were excised from the upper edge of the lower segment transverse uterine incision following the delivery of the infant and placenta, and before the closure of the hysterotomy. These uterine biopsies were immediately placed in DMEM/F-12 medium (Thermo Fisher Scientific) kept at 4°C, and transported to the laboratory within 2 hours of collection.

2.3. Measurement of uterine smooth muscle contraction

Once the samples were transferred to the laboratory, the endometrium and connective tissue were removed under a dissecting microscope in a petri dish containing ice-cold, oxygenated Krebs physiological solution (KPS). The KPS consisted of (mM): 118.07 NaCl, 4.69 KCl, 2.52 CaCl₂, 1.16 MgSO₄, 1.01 NaH₂PO₄, 25 NaHCO₃, and 11.10 glucose. The remaining myometrium was then placed in the oxygenated and ice-cold KPS until processed for force measurement. Experiments were conducted on the same day, typically within 8 hours of surgery. Previous studies have demonstrated that uterine samples stored, up to 48 hours, under similar conditions generate robust contractile responses to agonists (Ko et al., 1990; Yu et al., 1995). Hence, the uterine tissues used in this study were likely healthy without any significant deterioration.

Uterine specimens were cut into longitudinal strips (5 mm × 1.5 mm × 1.0 mm), which were then transferred to 5 ml muscle baths containing oxygenated KPS at 37°C. The strips were mounted on a wire myograph chamber (610-M; Danish Myo Technology, Aarhus, Denmark), and tension was measured with a PowerLab recorder (ADInstruments, Colorado Springs, CO, USA). Each smooth muscle strip was equilibrated for 60 minutes, following which a 0.1 g load was applied.

To assess the contractile response, each strip was stimulated twice with KCl (60 mM), with each stimulation lasting approximately 5 minutes. There was a 10-minute interval between these KCl stimuli before proceeding with other treatments. Each treatment involving a test compound had a duration exceeding 5 minutes to ensure the full development of the response. The order and duration of the test compound treatments are indicated in the accompanying figures and their respective captions.

To analyze the contractile response to each test compound, we used the response produced by the second KCl treatment as the reference. This approach was necessitated by three factors: (1). KCl and contractile agonists did not consistently induce sustained, uniform contractions across strips obtained from different donors. (2). There were variations in treatment durations between KCl and contractile agonists. (3). Variations in treatment times occurred due to manual administration and washout procedures.

To address these considerations, we first calculated the average force produced by each treatment by taking the area under the force curve (AUC) (which has units of millinewtons (mN) x minutes) and dividing it by the treatment duration in minutes (resulting in final units of mN). Subsequently, we normalized the average force for each test compound to the average force produced by the second KCl treatment in the same strip x100. We call the final (unitless) result the 'unitary AUC’.

2.4. hTERT-HM cell culture

The hTERT-infected human myometrial (hTERT-HM) cells, an immortalized cell line that maintains the characteristics of primary human uterine smooth muscle cells (as they express smooth muscle cell contractile markers and functional receptors including TAS2Rs and the oxytocin receptor (Condon et al., 2002; Zheng et al., 2017), were cultured in DMEM/F-12 containing 10% (v/v) fetal bovine serum (Thermo Fisher Scientific). Cells were maintained at 37°C in a humidified atmosphere of 5% CO₂ and 95% air. The culture medium was changed every other day. When the cells reached 90% confluence, they were detached with 0.25% (w/v) trypsin and then reseeded at a density of 1 × 10⁴ cells/cm².

2.5. Measurement of intracellular Ca²⁺ concentration ([Ca²⁺]_i)

hTERT-HM cells, whose intracellular [Ca²⁺]_i can be increased by various contractile agonists (Condon et al., 2002; Zheng et al., 2017), were loaded with 2 μM Cal-520 acetoxymethyl ester calcium indicator (AAT Bioquest, Pleasanton, CA) in DMEM/F-12 for 40 minutes at 37°C, followed by 30 minutes of de-esterification in indicator-free culture medium. Fluorescence Ca²⁺ images in hTERT-HM cells were captured using a custom-built wide-field digital imaging system (Qu et al., 2021). The camera was interfaced with an IX71, Olympus inverted microscope with a 20x 1.3 NA objective (Nikon, Tokyo, Japan). The 488 nm line of an argon ion laser served as a fluorescence excitation source, with a shutter controlling exposure time; emission of the Ca²⁺ indicator was observed at wavelengths greater than 500 nm. Images were captured at a rate of 1 Hz, with subsequent image processing and analysis conducted using ImageJ software (National Institutes of Health, Bethesda, MD). [Ca²⁺]_i was expressed as (Ft − F0)/F0*100, i.e., ΔF/F0(%), where Ft represents the fluorescence intensity of the entire cell at a given time point and F0 represents the average Ft at rest, determined by averaging the Ft of the first 20 images before the application of the test agents.

2.6. Protocols of phenanthroline treatment in Ca²⁺ measurement

Our preliminary studies indicated that oxytocin, U46619, and PGF2α could repeatedly increase [Ca²⁺]_i following a 15-minute recovery period. However, endothelin-1 only increased [Ca²⁺]_i upon its first application, and 5-HT did not result in a detectable increase in [Ca²⁺]_i. Consequently, we employed two different protocols to evaluate the effect of phenanthroline on contractile agonist-induced increases in [Ca²⁺]_i. For oxytocin, U46619, and PGF2α, hTERT-HM cells were stimulated with two pulses of an agonist, with phenanthroline applied for 15 minutes prior to the second pulse. In the absence of phenanthroline, the calcium response to the second pulse of the agonist was not as substantial as the response to the first pulse - this decreased response is referred to as a “rundown.” The effect of phenanthroline was quantified by comparing the peak [Ca²⁺]_i induced by the second pulse to that induced by the first pulse. For endothelin-1, hTERT-HM cells were stimulated with a pulse of the agonist, either with or without a 15-minute pretreatment with phenanthroline. The effect of phenanthroline on the endothelin-1-induced increase in [Ca²⁺]_i is expressed as the ratio of the peak [Ca²⁺]_i. with and without phenanthroline, respectively.

2.7. Statistical analysis

Data are reported as mean ± SEM, with ‘n’ indicating the number of uterine sample donors or myometrial cells. All concentration-response curves were subjected to ANOVA. When results reached statistical significance, post hoc Dunnett’s tests were performed. To assess the effect of phenanthroline on basal [Ca²⁺]_i, we used a paired t-test. The significance level was set at P ≤ 0.05.

3. Results

3.1. Characteristics of human subjects and uterine samples

In this study, we obtained consent from 14 pregnant women, with gestations ranging from 37 to 40 weeks, to donate their uterine samples. On the day of their scheduled cesarean delivery, we collected uterine tissues, approximately 5 cm × 1.5 cm × 3 cm in size, excised from the upper edge of the lower uterine segment transverse hysterotomy incision from each participant. These patients were not in labor at the time of cesarean delivery and tissue collection. Uterine tissues were promptly delivered to the laboratory within two hours following the procedure. All specimens were processed and assessed for contractile responses on the same day as their respective surgeries.

3.2. Effects of the TAS2R5 agonist phenanthroline on human uterine contractions induced by inflammatory contractile mediators and the hormone oxytocin

To minimize potential variability in contractions arising from factors such as differences in uterine strip size, patient health conditions, and tissue decomposition, we conducted contraction experiments with two reference control components. First, we normalized agonist-induced contractions relative to KCl-induced contractions in the same strip. Second, we tested each contractile agonist for its response in the presence of phenanthroline in one strip and in its absence in another strip from the same patient.

When normalizing relative to KCl, each strip underwent two treatments with 60 mM KCl before being tested with contractile agonists. We normalized the contractile responses relative to the second KCl treatment because it produced a stable contraction compared to the first KCl treatment.

As detailed in the results below, KCl and contractile agonists induced either phasic or biphasic contractions, with the latter involving a peak contraction followed by a sustained contraction. As previously mentioned, the treatment durations with KCl and contractile agonists differed, and there were variations in the treatment durations for contractile agonists due to manual administration and washout. Therefore, to better quantify these dynamic changes in contraction and minimize errors, we compared unitary AUC for each test compound, as described in the Methods section.

An increase in cyclooxygenase-2 expression and the resulting increase in prostaglandin levels are key processes in the uterine transition from quiescence to parturition (Fischer et al., 2008; Hirst et al., 1995; Olund et al., 1980; Peiris et al., 2021; Romero et al., 1994; Romero et al., 1987). Therefore, we examined the effect of the TAS2R5 agonist phenanthroline on prostaglandin-induced contractions. Figure 1 shows that 3 μM PGF2α generated a force similar to that produced by 60 mM KCl, and phenanthroline reduced PGF2α-induced contraction in a concentration-dependent manner. Specifically, 100 μM phenanthroline reduced PGF2α-induced contraction by approximately 50%, whereas 1 mM phenanthroline essentially abolished it (Figure 1B and 1C). U46619, a TP receptor agonist, at 1 μM caused a ~1.7-fold increase in contraction compared to 60 mM KCl. Phenanthroline at 100 μM suppressed ~65% of the contraction induced by U46619, and at 1 mM, it abolished it (Figure 2).

Figure 1. TAS2R5 agonist phenanthroline (PA) concentration-dependently inhibits PGF2α-induced human uterine contractions.

Representative traces of uterine strip tension changes in response to 3 μM PGF2α in the absence of PA (A), and in its presence at 1 mM (B). (C) Summarized results of PA’s inhibition of PGF2α-induced contraction. Data are represented as means ± SEM (n= 6 strips from 6 donors) and expressed as unitary AUC, i.e., AUC/treatment duration for PGF2α /AUC/treatment duration for 2^nd KCl x 100. **P < 0.01; ****P < 0.0001 by the post hoc dunnett's test following ANOVA at the statistically significant level.

Figure 2. Phenanthroline (PA) concentration-dependently inhibits U46619-induced human uterine contractions.

Representative traces of uterine strip tension changes in response to 1 μM U46619 in the absence of PA (A), and in its presence at 1 mM (B). (C) Summarized results of PA’s inhibition of U46619-induced contraction. Data are represented as means ± SEM (n= 6 strips from 6 donors) and expressed as unitary AUC as defined in the Methods and Figure 1. **P < 0.01; ***P < 0.001 by the post hoc dunnett's test following ANOVA at the statistically significant level.

Oxytocin (OT) plays a significant role at the onset of labor and during parturition (Chibbar et al., 1993; Fuchs et al., 1982; Kim et al., 2017; Uvnas-Moberg et al., 2019). As shown in Figure 3, 0.3 μM oxytocin induced a contraction approximately 3-fold greater than that elicited by 60 mM KCl. Phenanthroline at 100 μM reduced the contraction induced by OT by around 20%, although this reduction was not statistically significant. However, 300 μM phenanthroline significantly suppressed the OT-induced contraction, and at 1 mM, it suppressed approximately 90% of the contraction.

Figure 3. Phenanthroline (PA) concentration-dependently inhibits oxytocin (OT)-induced human uterine contractions.

Representative traces of uterine strip tension changes in response to 0.3 μM OT in the absence of PA (A), and in its presence at 1 mM (B). (C) Summarized results of PA’s inhibition of OT-induced contraction. Data are means ± SEM (n= 5 strips from 5 donors) and expressed as unitary AUC as defined in the Methods and Figure 1. NS, P>0.05; *P < 0.05; ***P < 0.001 by the post hoc dunnett's test following ANOVA at the statistically significant level.

Endothelin is significantly elevated and causes uterine contraction during labor in humans (Breuiller-Fouche et al., 2005; Heluy et al., 1995; Osada et al., 1997; Yallampalli and Garfield, 1994) and may play a critical role in inflammation-associated preterm labor based on animal studies (Wang et al., 2008). We found that 0.1 μM ET-1 induced uterine contraction by 1.7-fold compared to 60 mM KCl. This contraction was significantly inhibited by 300 μM phenanthroline and abolished by 1 mM phenanthroline (Figure 4).

Figure 4. Phenanthroline (PA) concentration-dependently inhibits ET-1-induced human uterine contractions.

Representative traces of uterine strip tension changes in response to 0.1 μM ET-1 in the absence of PA (A), and in its presence at 1 mM (B). (C) Summarized results of PA’s inhibition of ET-1-induced contraction. Data are means ± SEM (n= 5 strips from 5 donors) and expressed unitary AUC as defined in the Methods and Figure 1. NS, P>0.05; *P < 0.05; ***P < 0.001 the post hoc dunnett's test following ANOVA at the statistically significant level.

5-HT contracts human uterine smooth muscle at term and aberrant serotonin signaling has been implicated as a contributor to preterm labor (Bytautiene et al., 2008; Cordeaux et al., 2009; Garfield et al., 2006; Koren et al., 1965). Figure 5 shows that 3 μM 5-HT robustly contracted uterine smooth muscle, and phenanthroline concentration-dependently suppressed 5-HT-induced contraction and abolished it at both 300 μM and 1 mM.

Figure 5. Phenanthroline (PA) concentration-dependently inhibits 5-HT-induced human uterine contractions.

Representative traces of uterine strip tension changes in response to 3 μM 5-HT in the absence of PA (A), and in its presence at 1 mM (B). (C) Summarized results of PA’s inhibition of 5-HT-induced contraction. Data are means ± SEM (n= 5 strips from 5 donors) and expressed unitary AUC as defined in the Methods and Figure 1. **P < 0.01; ****P < 0.0001 by the post hoc dunnett's test following ANOVA at the statistically significant level.

3.3. Effects of phenanthroline on increase in intracellular Ca²⁺ induced by inflammatory contractile mediators and oxytocin

Having characterized the phenanthroline-induced relaxation of contractions induced by various inflammatory contractile mediators and oxytocin, we sought to identify a common cellular signal underlying this diverse relaxation. Because Ca²⁺ is the primary determinant of uterine smooth muscle contraction (Aguilar and Mitchell, 2010; Sanborn, 2000; Wray and Arrowsmith, 2021), we investigated whether phenanthroline interferes with Ca²⁺ signaling as a possible common mechanism. We employed hTERT-HM cells for this purpose because they are an immortalized cell line derived from human uterine smooth muscle cells (Condon et al., 2002), and express abundant TAS2R5 mRNA (Zheng et al., 2017).

Because activation of TAS2R is likely to increase [Ca²⁺]_i (Chandrashekar et al., 2000), we first examined the effect of phenanthroline on [Ca²⁺]_i in hTERT-HM cells. We observed that 1 mM phenanthroline did not alter the basal [Ca²⁺]_i (0.4±0.5% ΔF/F0(%); n=10; P>0.05 in a paired t-test comparing with vs without phenanthroline). Using a dual-pulse protocol, we noted a robust increase in [Ca²⁺]_i upon the first stimulation of PGF2α, with approximately a 20% reduction upon the second stimulation (Figure 6). Phenanthroline did not alter the [Ca²⁺]_i following the washout of PGF2α (Figure 6B-C) or other agonists (see Figures 7 and 8 below). However, when hTERT-HM cells were pretreated with phenanthroline, there was a concentration-dependent inhibition of the Ca²⁺ increase triggered by the second PGF2α application. At 1 mM, phenanthroline completely abolished the PGF2α-induced increase in [Ca²⁺]_i.

Figure 6: Phenanthroline (PA) concentration-dependently inhibits PGF2α-induced increase in intracellular calcium in hTERT-HM cells.

Time courses of calcium concentration changes in response to 1 μM PGF2α in the absence of PA (A), and in its presence at 0.031 mM (B), 0.1 mM (C), and 0.31 mM (D). The gaps between two PGF2α treatments were 15 min. (E) Summarized results of PA’s inhibition of PGF2α-induced increase in cytosolic calcium level. Data are means ± SEM (n = 20-28 cells) and expressed as the ratio of calcium increase peak induced by the second PGF2α application to that induced by the first PGF2α in A-D. NS, P >0.05; ****P < 0.0001 by the post hoc dunnett's test following ANOVA at the statistically significant level.

Figure 7: Phenanthroline (PA) concentration-dependently inhibits U46619-induced increase in intracellular calcium in hTERT-HM cells.

Time courses of calcium concentration changes in response to 1 μM U46619 in the absence of PA (A), and in its presence at 0.031 mM (B), 0.1 mM (C), and 0.310 mM (D). The gaps between two U46619 treatments are 15 min. (E) Summarized results on PA’s inhibition of U46619-induced increase in cytosolic calcium level. Data are means ± SEM (n = 20-30 cells) and expressed as the ratio of calcium increase peak by the second U46619 application to that by the first U46619 application in A-D. ****P < 0.0001 by the post hoc dunnett's test following ANOVA at the statistically significant level.

Figure 8: Phenanthroline (PA) concentration-dependently inhibits oxytocin (OT)-induced increase in cytosolic calcium concentration in hTERT-HM cells.

Time courses of calcium concentration changes in response to 0.1 μM OT in the absence of PA (A), and in its presence at 0.031 mM (B), 0.1 mM (C), and 0.31 mM (D). (E) Summarized results of PA’s inhibition of OT-induced increase in cytosolic calcium level. Data are means ± SEM (n = 20-28 cells) and expressed as the ratio of calcium increase peak induced by the second OT application to that induced by the first OT application in A-D. ****P < 0.0001 by the post hoc dunnett's test following ANOVA at the statistically significant level.

Similar to PGF2α, U46619 robustly increased cytosolic Ca²⁺ during its first stimulation but did not cause as much Ca²⁺ rundown in its second application (Figure 7). As with PGF2α, phenanthroline concentration-dependently inhibited the Ca²⁺ increase induced by the second application of U46619, and at 1 mM, it abolished the U46619-induced increase in [Ca²⁺]_i.

Figure 8 shows that oxytocin robustly increased [Ca²⁺]_i during its first stimulation, with the response experiencing approximately a 10% rundown during its second stimulation. Phenanthroline concentration-dependently inhibited the oxytocin-induced rise in [Ca²⁺]_i, and it was abolished at a phenanthroline concentration of 1 mM.

ET-1 increased cytosolic Ca²⁺ (Figure 9) but failed to increase it when reapplied after a 15-minute recovery. Therefore, to assess the effect of phenanthroline on the ET-1-induced increase in [Ca²⁺]_i, we compared responses to a single application of ET-1 either with or without phenanthroline. Figure 9 shows that phenanthroline prevented ET-1 from increasing cytosolic Ca²⁺ in a concentration-dependent manner.

Figure 9: Phenanthroline (PA) concentration-dependently inhibits ET-1-induced increase in cytosolic calcium concentration in hTERT-HM cells.

Time courses of calcium concentration changes in response to 0.1 μM ET-1 in the absence of PA (A), and in its presence at 0.031 mM (B), 0.1 mM (C), and 0.310 mM (D). (E) A comparison of the averaged calcium peaks calculated from the cells similar to those shown in A-D. Data are means ± SEM; n = 20 cells. *P<0.05, ****P < 0.0001 by the post hoc dunnett's test after ANOVA at the statistically significant level.

4. Discussion

Phenanthroline possesses two distinct known properties: it acts as a selective TAS2R5 agonist (Meyerhof et al., 2010) and inhibits Zn²⁺ metallopeptidases (Granato et al., 2015). The latter property stems from its ability to bind metal ions in enzymes. Our experiments indicated that phenanthroline neither affects basal Ca²⁺ levels in myometrial cells at rest nor after the washout of contractile agonists (see Figures 6-8). This is consistent with the fact that this compound has a much higher stability constant for Zn²⁺ (K= 2 x 10⁶ M) compared to Ca²⁺ (K = 5 M) (Frederick et al., 1984). Moreover, the inhibition of metallopeptidases has been shown to potentiate uterine contraction induced by oxytocin and bradykinin (Ottlecz et al., 1991; Schriefer and Molineaux, 1993). This stands in contrast to our findings, where phenanthroline relaxed precontracted myometrial strips. Thus, it is more plausible that phenanthroline activates TAS2R5, leading to uterine relaxation.

Recent studies have highlighted the widespread distribution of TAS2Rs and their downstream components within the reproductive system, suggesting their potential importance in reproductive functions. In the male reproductive system, TAS2Rs have been identified in testicular tissues and sperm cells. Their activation has been implicated in the regulation of sperm chemotaxis, motility, and spermatogenesis (Fehr et al., 2007; Governini et al., 2020; Liu et al., 2022; Luddi et al., 2019; Roemer et al., 2012). In the female reproductive system, TAS2Rs have been found in the myometrium, cervix, vagina, ovary, and placenta (Liu et al., 2017; Semplici et al., 2021; Wolfle et al., 2016; Zheng et al., 2017). Their physiological roles in female reproduction remain largely unknown. Our previous study (Zheng et al., 2017) suggested that chloroquine, an agonist of several TAS2Rs, can relax uterine strips pre-contracted by membrane depolarization and oxytocin in both mice and humans. Moreover, chloroquine can prevent lipopolysaccharide- or RU486-induced preterm labor in mice, suggesting that TAS2R activation may be a promising strategy for treating preterm labor. Our present study provides additional evidence in support of this concept by showing that the selective TAS2R5 agonist phenanthroline can also induce significant relaxation in pre-contracted uterine strips. This finding is significant as it underscores the potential of targeting a single TAS2R. This may result in fewer side effects compared to a multi-TAS2R agonist, given the widespread distribution of different TAS2Rs in extra-oral tissues throughout the entire human body including the respiratory system, gastrointestinal tract, and urinary system (Avau and Depoortere, 2016; Behrens and Lang, 2022; Bloxham et al., 2020; Carey et al., 2016; Lu et al., 2017; Tuzim and Korolczuk, 2021).

Given the complex and interactive nature of the mechanisms underlying both preterm and term labor, resulting from parallel and sequential changes in the endocrine system, uterine mechanics, and inflammation (Challis et al., 2000; Leimert et al., 2021; Menon et al., 2016; Romero et al., 2014; Vidal et al., 2022), it is improbable that a single treatment will suffice to prevent or reverse preterm labor (Keelan et al., 1997). Many agents, including beta-2 agonists and oxytocin receptor antagonists, have been studied but were found to be clinically ineffective in preventing or treating preterm birth (Wilson et al., 2022). The available tocolytics, such as nifedipine and indomethacin, only delay delivery by a few days (Wilson et al., 2022). We need to explore new classes of agents, beyond calcium channel blockers and non-steroidal anti-inflammatory drugs, to expand our understanding of possible treatments for preterm labor. In light of our findings, TAS2R5 emerges as a potential target for the development of a new uterine relaxant.

Interestingly, the TAS2R5 agonist phenanthroline can inhibit uterine contractions induced by various contractile agonists, including membrane depolarization (Zheng et al., 2017), oxytocin, and various contractile inflammatory mediators. This suggests that potential activation of TAS2R5 by phenanthroline may stimulate a common molecular mechanism that can counteract different receptor-mediated contractions. Our study reveals that one such mechanism is a decrease in cytosolic Ca²⁺ concentration. However, further experiments are needed to determine the molecular basis for this decrease in the cytosolic Ca²⁺ level. It is plausible that this change is mediated by multiple mechanisms such as the activation of large conductance Ca²⁺-activated K⁺ channels (Deshpande et al., 2010), inhibition of inositol 1,4,5 trisphosphate receptors (Tan and Sanderson, 2014), enhanced uptake of Ca²⁺ by mitochondria (Talmon et al., 2019), or beta-gamma subunit-induced inhibition of voltage-gated Ca²⁺ channels (Zhang et al., 2013), because each of these mechanisms has been proposed to mediate relaxation in other smooth muscle tissues.

5. Strengths and limitations

This study has several strengths. We utilized human uterine biopsies from term gravidas, thereby deepening our understanding of human uterine physiology and pharmacology. Moreover, by employing a human myometrial cell line that expresses TAS2Rs and the corresponding downstream signaling components present in fresh human uteri, we were able to uncover the underlying Ca2+ signal by which the TAS2R5 agonist phenanthroline induces relaxation of uterine contractions elicited by oxytocin and various inflammatory mediators. It is crucial to investigate TAS2R5 in human tissues, given this receptor is absent in other mammals (Grau-Bove et al., 2022). This unique expression does not rule out the potential use of animal models in future research. Tissue and cell-specific knock-in techniques could be applied to introduce this receptor into animals, enabling in vivo studies of its reproductive role.

Yet, our study has certain limitations. We focused on myometrial samples from term gravidas. Although we anticipate that phenanthroline can induce relaxation in preterm uteri, its efficacy or potency could vary if the expression of TAS2R5 and its downstream signaling components is reduced or altered. Moreover, our conclusion concerning TAS2R5’s involvement in phenanthroline-induced responses in myometrial cells is based on findings that phenanthroline activates only this receptor within the TAS2R family (Meyerhof et al., 2010). To solidify our conclusion, it would be beneficial to replicate our results in myometrial cells with TAS2R5 knocked out using advanced techniques like CRISPR-Cas9 gene editing. Lastly, although hTERT-HM cells enable us to identify Ca²⁺ signaling as a common mechanism for phenanthroline's broad-spectrum relaxation, this cell line lacks functional serotonin receptors or their downstream signaling components. Future studies should explore and validate Ca²⁺ hypothesis using freshly dissociated uterine smooth muscle cells from both term and preterm uteri.

6. Conclusion

In summary, the selective TAS2R5 agonist phenanthroline effectively inhibits contractions of term uterine strips in response to inflammatory contractile mediators and the contractile hormone oxytocin. Furthermore, this agonist is capable of suppressing the Ca²⁺ increase elicited by these bioactive contractile substances. This suggests that TAS2R agonists may operate through a common mechanism of interfering with the Ca²⁺ signaling pathway to produce uterine relaxation. Considering the critical role of various contractile bioactive substances in both parturition and preterm labor, a comprehensive understanding of TAS2R5 function and its agonists could significantly contribute to the development of innovative and effective broad-spectrum uterine relaxants aimed at preventing preterm birth.