全文

Abstract

Objective

This study analyzed the ability of montelukast, a cysteinyl-leukotrienes receptor antagonist and anti-inflammatory agent, to produce a consistent tocolytic effect alone or in combination with nifedipine, a calcium (Ca2+) channel blocker currently used in clinical practice.

 

Study design

Uterine biopsies were obtained from consenting women undergoing elective cesarean sections at term (n = 20). Myometrial microsomal fractions were analyzed by immunoblotting to quantify relative cysteinyl leukotrienes receptor 1 (CysLTR1) levels. Isometric tension measurements were performed in vitro on human myometrial strips (n = 120) in isolated organ baths in order to establish concentration–response curves to montelukast and to quantify changes in Ca2+ sensitivity on β-escin permeabilized tissues.

 

Results

Immunodetection analysis revealed the presence of CysLTR1 receptor in uterine tissues, fetal membranes and placenta. A significant increase in area under the curve (AUC) was quantified following the addition of leukotriene D4 (LTD4) (0.01–0.3 μM), an end-product of the lipoxygenase pathway. Conversely, addition of montelukast produced a significant tocolytic effect by decreasing the frequency and AUC (IC50 = 1 μM). Moreover, addition of montelukast also resulted in a reduced Ca2+ sensitivity as compared to control tissues (EC50 values of 654 and 403 nM; p = 0.02 at pCa 6), while an additive effect was observed in combination with 0.1 nM nifedipine (p = 0.004).

 

Conclusion

This original study demonstrates the potency of montelukast as a tocolytic agent in an in vitro human uterine model. Montelukast, in combination with nifedipine, could represent a therapeutic approach to reduce inflammation associated with prematurity while facilitating the inhibition of preterm labor.

 

Abbreviations

5-LO, 5-lipoxygenase; 12-LO, 12-lipooxygenase; AA, arachidonic acid; CysLTR1, cysteinyl leukotrienes receptor 1; COX, cyclooxygenase; CYP450, cytochrome P450; DAG, diacylglycerol; EETs, epoxy-eicosatrienoic acid regioisomers; IP3, inositol 1,4,5-triphosphate; LO, lipoxygenase; LTD4, leukotriene D4; PLC, phospholipase C
 

Keywords

5-LO; Leukotrienes; Montelukast; Tocolytic; Uterine contractile activity

 

1. Introduction

Preterm birth is a major public health problem and currently remains the leading cause of perinatal deaths in developed countries despite sustained research efforts in the last 20 years [1]. Recent studies related to prematurity all agree that a new therapeutic approach is needed. For instance, tocolysis maintenance for more than 48 h has failed to display any beneficial effects on either perinatal morbidity or mortality [2]. The exact mechanisms underlying preterm labor are still poorly understood, but the presence of an inflammatory condition is a likely risk factor [1] and [3], and is thus rapidly becoming a logical pharmacological target.

 

The arachidonic acid metabolic pathway allows the production of various inflammatory lipids mediators that are likely involved in the control of uterine contraction and relaxation [4]. Indeed, prostaglandins, synthesized by cyclooxygenase (COX) from free arachidonic acid precursor are known to initiate labor and to increase in vitro uterine contractility [5] and [6]. Moreover, among tocolytic agents, the inhibition of prostaglandin production by indomethacin (a non-selective COX inhibitor) remains the most efficient but leads to major fetal secondary effects, both hemodynamic and renal [7] and [8]. Two alternative arachidonic acid metabolic pathways known to be involved in the modulation of uterine contractions are the cytochrome P450 (CYP450) pathway, producing relaxing mediators [4], and the lipoxygenase (5-LO) pathway producing leukotrienes (LTs).

 

Cysteinyl leukotrienes (LTC4, LTD4 and LTE4) are synthesized stepwise by 5-lipoxygenase (5-LO) from the arachidonic acid metabolic pathway via an unstable intermediate metabolite, the leukotriene A4 (LTA4) [9]. These metabolites are major inflammatory mediators in smooth muscle such as vascular and bronchial walls [9]. Leukotrienes may also be involved in the onset of labor given that leukotriene plasma levels increase at the end of pregnancy and at the beginning of labor [10], [11] and [12]. Since LTC4 was found to increase uterine contractility in an in vitro model [13], [14] and [15], it has been proposed that LO inhibitors may display tocolytic effects. In a previous study, we explored the inhibition of an alternative arachidonic acid pathway, the lipoxygenase (LO) pathway. The inhibition of 5- and 12-LO was found to block leukotriene production resulting in a tocolytic effect and therefore of potential interest as pharmacological agents in the treatment of preterm labor [16]. More specifically, results showed that both 5- and 12-LO were present in subcellular fractions and that specific inhibition of these isoforms resulted in a decrease in the area under the curve with a major effect observed on mean amplitude of uterine phasic contractions.

 

5-LO metabolites (such as LTD4) are natural ligands for cysteinyl leukotriene receptor 1 (CysLTR1). The actions of these inflammatory metabolites are mediated by activation of the corresponding receptor which is known to produce a positive inotropic effect [9]. The inhibition of this signaling pathway by montelukast, a highly selective antagonist of CysLTR1, is currently used as a second line treatment for asthma since it can reduce inflammatory responses and result in the inhibition of contractile responses. The use of montelukast is safe during pregnancy [17] and [18], which suggests that this specific CysLTR1 antagonist may be of pharmacological interest in this particular instance. To date, no study has validated the effect of montelukast on uterine contractility in vitro.

 

The presence of CysLT receptor (type 1) in uterine tissues was detected prior to assessing the functional effects of montelukast and nifedipine on spontaneous myometrial contractile activities in vitro. This study therefore analyzes the tocolytic effect of montelukast on spontaneous in vitro uterine contractions, either alone or in combination with nifedipine, a calcium (Ca2+) channel blocker currently used in clinical practice. The effect of montelukast after LTD4 treatment, mimicking partial inflammatory conditions, was also quantified.

 

2. Materials and methods

2.1. Subjects and sample collection

All biopsy specimens were obtained from patients admitted for an elective cesarean section. The study was approved by our institutional Ethics Committee for research on human subjects (project # 09-040; ClinicalTrials.gov Identifier: NCT00939744) and all volunteers gave written informed consent. The inclusion criteria were (1) a gestational age between 37+0 and 40+0 weeks of gestation, (2) a singleton gestation, (3) no labor and (4) signed informed consent. Exclusion criteria included (1) a history of preterm labor, (2) the use of pharmacological agents to induce labor and/or montelukast during pregnancy, (3) the presence of infections (chorioamnionitis, HIV, genital herpes, hepatitis B and C) or (4) vaginal bleeding after the third trimester. Medical data were obtained from the patients’ medical files.

 

During the cesarean section, immediately after delivery of the baby, all myometrium biopsies were excised from the upper lip of the lower uterine segment incision in the midline as previously described [4]. Placental biopsies (1 cm3) from the maternal side were obtained immediately after removal and the membranes excised free of placenta in a systematic manner by the same investigator. Once collected, all tissue biopsies were placed in Krebs–Heinseleit physiological salt solution (Krebs) of the following composition (mmol/l): 4.7 potassium chloride, 118 sodium chloride, 1.2 magnesium sulfate, 2.5 calcium chloride, 1.2 potassium phosphate, 25 sodium bicarbonate and 11.1 glucose (Sigma–Aldrich, St. Louis, MO) at pH 7.4. Tissues were stored at 4 °C and used within 8 h of collection or rapidly rinsed in DMEM-F12 + 10% glycerol before snap freezing in liquid nitrogen and subsequently stored at −80 °C until analysis.

 

2.2. Western blot analysis

Subcellular fractions (cytosolic and microsomal) were prepared from myometrium, fetal membranes and placenta, and separated on SDS PAGE as previously described [4]. For Western blot analyses, membranes were blocked for 2 h with 5% non-fat dry milk in Tris-buffered saline with 0.1% Tween at room temperature. Blots were incubated overnight at 4 °C with rabbit antiserum raised against CysLTR1 receptor (Assay Biotech, CA, USA). After washing, the membranes were incubated in a solution containing peroxidase-conjugated donkey anti-rabbit IgG antiserum (Amersham, QC, Canada). An enhanced chemiluminescence kit (Roche, QC, Canada) was used to detect protein labeling. Immunostainings were digitized and analyzed with Lab-Image software (Kaplan GmbH, Germany).

 

2.3. Isolated organ bath experiments

Longitudinal myometrial strips (measuring approximately 2 mm × 2 mm × 10 mm) were dissected, cleansed of serosa, fibrous tissue and blood vessels, and mounted for isometric recordings under 2 g of resting tension in an organ bath system as previously described [4]. The tissue baths contained 7 ml of Krebs solution maintained at 37 °C, pH 7.4, and were continuously gassed with a mixture of 95% oxygen/5% carbon dioxide. Myometrial strips were allowed to equilibrate for at least 2 h, after which a 30 min period was allotted to achieve spontaneous phasic contractions. Each strip was compared to vehicle control. Passive and active tensions were assessed using transducer systems (Radnoti Glass Tech., Monrovia, CA) coupled to Polyview software (Grass-Astro-Med Inc., West Warwick, RI) for facilitating data acquisition and analysis.

 

2.4. β-Escin permeabilization and Ca2+ sensitivity of the myometrial contraction machinery

Ca2+ sensitivity was measured exactly as previously reported [19]. Tension developed by permeabilized myometrial strips was measured in activating solutions containing 10 mM EGTA and step increases of CaCl2 to yield the desired free Ca2+ concentration.

 

2.5. Drugs and chemical reagents

Montelukast, LTD4 (Sigma–Aldrich, St. Louis, MO) and nifedipine (Sigma–Aldrich, St. Louis, MO) were dissolved in 100% ethanol (EtOH) and stored as 1 and 10 mM stock solutions, respectively. Final bath concentration of EtOH never exceeded 0.1%. Exogenous inhibitors were added separately to the tissue bath or in a cumulative manner at increasing concentrations (10 nM to 10 μM) in 30-min intervals. Fresh Krebs solution was prepared daily.

 

2.6. Data analysis and statistics

The effect of pharmacological agents and respective controls were assessed by calculation of the amplitude or area under the curve (AUC) for each 30-min interval. Values were normalized as a percentage of the AUC obtained in the 30-min basal activity period and were corrected for the reduction in contractile activity observed with the vehicle control using Sigma Plot 11.0 (SPSS-Science, Chicago, IL). Data were not normally distributed and were therefore analyzed with nonparametric tests. The Wilcoxon signed rank test was used for paired results and the Mann–Whitney test for unpaired results. Differences were considered significant when p < 0.05.

 

3. Results

3.1. Study population

Demographic characteristics of all patients who participated in the study are summarized in Table 1. The study group was comprised of 20 healthy Caucasian pregnant women with a mean age of 28.9 years (range 19–40). Participants underwent cesarean delivery between 37+0 and 40+0 weeks of gestation with a mean pre-pregnancy BMI of 26.5 (range 18–35). Indications for cesarean section are described in Table 1.

 

 

3.2. Immunodetection of CysLTR1 in uterine tissues

Western blots were performed to detect the presence of CysLTR1 in uterine tissues using a specific primary antibody, with standard molecular weight proteins as negative control (data not shown). An immunoreactive band was consistently detected at 38 kDa in all tested microsomal fractions (Fig. 1A) and was slightly increased in fetal membranes. Fig. 1B reveals that the CysLTR1 receptor was mainly detected in fetal membranes compared to relative detection levels in both myometrial and placental microsomal fractions (p < 0.05; n = 5).

 

 

3.3. Effect of montelukast, a CysLTR1 receptor antagonist, on contractile activity

Control recordings revealed rhythmic activities with a frequency of 9 contractions per hour and contractile amplitudes of up to 2 g. No significant inhibitory effect was observed when comparing the area under the curve (AUC) after a 4 h time control (p = 1.00; data not shown) and the addition of 0.3% ethanol (p = 0.31; Fig. 2A).

 

 

Fig. 2B and C displays a typical recording of myometrial strip contractile activity during the basal activity period, upon addition of 0.1 to 1 nM nifedipine (a L-type calcium channel blocker), and 0.1–1 μM montelukast (a specific antagonist of CysLTR1) at 30-min intervals. Both pharmacological agents displayed a tocolytic effect as quantified by the cumulative concentration–response curves (CCRC; Fig. 2D). As can be seen in Fig. 3A, the combined use of montelukast and nifedipine reduced the amplitude of the phasic contractions, along with an efficient recovery after a 30-min washout period (Fig. 3A, right panel). The concentrations used correspond to the relative IC25 values which were determined by data curve fitting from the cumulative concentration response curve (CCRC) to both pharmacological agents (Fig. 2D). Combined inhibition of in vitro myometrial contractile activity by nifedipine and montelukast are reported in Fig. 3B, resulting in an additive effect (p = 0.004). The additive effect was obtained for independent experiments, regardless of the order of administration of the pharmacological compounds.

 

 

3.4. Effect of montelukast on Ca2+ sensitivity

Comparative analyses were therefore performed on β-escin-permeabilized tissues to quantify putative modifications in Ca2+-sensitivity either in the absence (control) or presence of 1 μM montelukast treatment. This concentration corresponds to the relative montelukast IC50 value determined from the cumulative concentration response curve (CCRC) described above (Fig. 2D). Fig. 4A illustrates a typical tension recording obtained during pre-calibrated Ca2+ step increases under control conditions on a myometrial strip, whereas Fig. 4B illustrates a typical recording following a 2-h pretreatment with 1 μM montelukast on Ca2+ induced responses. Quantitative analysis revealed an apparent Ca2+ sensitivity following montelukast pretreatment with an EC50 value of 654 nM as compared to an EC50 value of 403 nM for free Ca2+ concentration in control condition (Fig. 4C). However, despite a shift of 251 nM, this difference in Ca2+ sensitivity reached statistical significance only with 1 μM CaCl2.

 

 

3.5. Effect of LTD4 pretreatment

In order to assess the functional integrity of the CysLTR1 receptor as well as the effect of montelukast in a partial inflammatory condition, uterine strips were treated with increasing concentrations of the leukotriene LTD4 (0.01–0.3 μM). A typical recording (Fig. 5A) displays an increased contractile response. LTD4 alone significantly increased the amplitude of phasic spontaneous contractions (Fig. 5B) as well as the area under the curve (Fig. 5C). Note that since the initiation time of the first contraction was not reduced upon addition of LTD4, this compound was not used to trigger uterine contractions. Fig. 5D shows that these increases were completely abolished by the addition of 0.1 μM montelukast. The addition of 0.1 μM montelukast after a 30-min basal activity produced a 25% decrease of the AUC whereas 0.1 μM montelukast administered after the addition of 0.3 μM LTD4 produced a 60% relaxation of the LTD4-induced rise in AUC (p = 0.006). Combined addition of both tocolytic agents (montelukast and nifedipine) following LTD4 pretreatment did not induce an additive effect ( Fig. 5D; right histogram bar). This effect does not differ from the effect of montelukast alone in presence of LTD4.

 

 

4. Comment

4.1. Principal findings of the study

This is the first study detecting the presence of the CysLTR1 receptor in myometrial tissues. In addition, the study clearly brings to light the tocolytic effect of montelukast, either alone or in combination with nifedipine, a reference tocolytic according to current guidelines [8]. Finally, montelukast was found to reverse uterine smooth muscle reactivity upon LTD4 treatment.

 

4.2. Presence of the CysLTR1 receptor

The CysLTR1 receptor has previously been detected in numerous organs such as lung, heart, vessels as well as in the placenta [20] and [21], but this is the first report of the presence of this receptor in myometrial tissues. Results obtained tend to demonstrate that the receptor is mainly localized in fetal membranes compared to relative detection levels in both human myometrium and placenta. Interestingly, other receptors such as prostaglandin receptors are also highly expressed in fetal membranes, including the amnion and choriodecidua, compared to expression in maternal tissues [22] and [23]. These results hence support an effector signaling between fetal membranes and myometrium via the choriodecidual space in the initiation of myometrial contractility and the onset of labor [24] and [25]. Indeed, the choriodecidual space is known to be enriched in signaling molecules [26].

 

4.3. Tocolytic effect of montelukast

Despite a lower potency than nifedipine, montelukast nevertheless induced a significant tocolytic effect in vitro. However, the apparent additive effect of the association between nifedipine and montelukast is particularly noteworthy. This association could minimize side effects by using low doses of efficient compounds [27], since severe maternal hypotension and headaches can occur upon nifedipine treatment [28]. From a mechanistic standpoint, montelukast is known to inhibit the LTD4 pathway through CysLTR1 receptors, which are coupled to Gq and PLC in order to produce IP3 (an internal Ca2+ release agonist) and DAG [9] whereas the mode of action of nifedipine is strictly through the blockade of the L-type Ca2+ channels. Thus, the combined use of these compounds with two different modes of action may result in an additive effect. Such additive effect between montelukast and nifedipine could potentially constitute a complementary treatment with both a direct effect on uterine contraction crisis and inhibition of the inflammatory cascade. Note that we have not tested montelukast in concert with a COX1/2 inhibitor (indomethacin) due to its lack of specificity and its clinical secondary effects [7] and [8].

 

4.4. Modulation of the Ca2+ sensitivity

In the present study, our data also revealed that montelukast reduce the calcium-sensitivity of myometrial strips at 1 μM free-Ca2+ (p = 0.02), which would partially explain the tocolytic effect of the cysteinyl leukotriene receptor antagonist on basal or LTD4-treated contractile activities. Further experiments are necessary, however, to strengthen this basic observation.

 

4.5. LTD4 effects

In vivo, leukotrienes can trigger inflammatory processes [29] and contractile responses [14]. LTD4 treatments are known to induce a partial inflammatory condition and enhance contractile responses in various smooth muscle tissues [30] and [31]. Moreover, LTD4 is the metabolite with the highest intrinsic activity toward CysLTR1 [32]. Herein, we demonstrate that montelukast modulates basal uterine activity and abolishes the effects of LTD4 on the amplitude of phasic contractions. In this study, the concentrations of LTD4 used are slightly supra-physiologic to mimic an “acute” inflammatory status. These results thus validate the effect of LTD4 on contractility. which is consistent with published data [15] and further demonstrate that montelukast is able to reverse the effects of LTD4 pretreatment. Taken together, these results suggest that LTD4 enhances either Ca2+ entry into myometrial smooth muscle cells (through L-type Ca2+ channels), Ca2+ release from internal calcium stores, or enhances the Ca2+-sensitivity of the contractile machinery via the phosphorylation of regulatory proteins.

 

In conclusion, we demonstrate for the first time the presence of the CysLTR1 receptor in myometrial tissues. Results also show that montelukast displays a significant tocolytic effect on myometrial contractile activity by reducing both amplitude and AUC, in addition to exhibiting an additive effect with low concentrations of nifedipine. This effect of montelukast was further enhanced in the presence of the leukotriene LTD4. These original findings are consistent with the detection of CysLTR1 in all uterine tissues. The present preclinical data could represent a valuable strategy to reduce contractile activity and uterine inflammation, the latter being associated with prematurity. Our data also suggest the potential benefit of a closer diagnostic investigation of inflammatory status in the management of preterm labor.

 

Conflict of interest statement

The authors report no conflict of interest.

 

Financial support

FRQS, Fondation des Étoiles and PAFI.

 

Poster presentation # 528 at SMFM 33th Annual Meeting, The Annual Meeting, San Francisco, CA February 15th, 2013.

 

Acknowledgements

The authors wish to thank Mr. Pierre Pothier for critical review of the manuscript and for helpful insights. We also thank all of the obstetrical staff of the Centre Hospitalier Universitaire de Sherbrooke who participated in the sample collection procedure. S.C. was a recipient of a PhD studentship from the FRQS (Fond de Recherche du Québec-Santé).