Arrest of mouse preterm labor until term delivery by combination therapy with atosiban and mundulone, a natural product with tocolytic efficacy

Abstract

There is a lack of FDA-approved tocolytics for the management of preterm labor (PL). In prior drug discovery efforts, we identified mundulone and mundulone acetate (MA) as inhibitors of in vitro intracellular Ca²⁺-regulated myometrial contractility. In this study, we probed the tocolytic potential of these compounds using human myometrial samples and a mouse model of preterm birth. In a phenotypic assay, mundulone displayed greater efficacy, while MA showed greater potency and uterine-selectivity in the inhibition of intracellular-Ca²⁺ mobilization. Cell viability assays revealed that MA was significantly less cytotoxic. Organ bath and vessel myography studies showed that only mundulone exerted inhibition of myometrial contractions and that neither compounds affected vasoreactivity of ductus arteriosus. A high-throughput combination screen identified that mundulone exhibits synergism with two clinical-tocolytics (atosiban and nifedipine), and MA displayed synergistic efficacy with nifedipine. Of these combinations, mundulone+atosiban demonstrated a significant improvement in the in vitro therapeutic index compared to mundulone alone. The ex vivo and in vivo synergism of mundulone+atosiban was substantiated, yielding greater tocolytic efficacy and potency on myometrial tissue and reduced preterm birth rates in a mouse model of PL compared to each single agent. Treatment with mundulone after mifepristone administration dose-dependently delayed the timing of delivery. Importantly, mundulone+atosiban permitted long-term management of PL, allowing 71% dams to deliver viable pups at term (>day 19, 4-5 days post-mifepristone exposure) without visible maternal and fetal consequences. Collectively, these studies provide a strong foundation for the development of mundulone as a single or combination tocolytic for management of PL.

Keywords: Drug combination; Mundulone; Natural products; Preterm birth; Preterm labor; Tocolytic; U46619 (Pubchem CID: 5311493); Uterus; atosiban (Pubchem CID: 5311010); indomethacin (Pubchem CID: 3715); mifepristone (pubchem CID: 55245); mundulone (Pubchem CID: 4587968); mundulone acetate (Pubchem CID: 6857790); nifedipine (Pubchem CID: 4485); oxytocin acetate (Pubchem CID: 5771).

Fig. 1.

In vitro uterine-selectivity of mundulone and mundulone acetate (MA) to inhibit intracellular Ca²⁺-release. Real-time recording of concentration-dependent inhibition of intracellular Ca²⁺-release by mundulone (A) or MA (C) from myometrial cells and aorta VSMCs in 384-well format. Concentration-response curves of mundulone (B) or MA (D) showing selectivity towards myometrial cells compared to the aorta VSMCs due to either a significant shift in potency (IC₅₀) or efficacy (E_max), respectively. Non-linear regression was used to fit the data (mean + SEM) and to calculate the IC₅₀ and E_max., which are provided in Table 1, along with p-values. A 2-way ANOVA with a post-hoc Fisher’s LSD test was used to compare the E_max values (shown).

Fig. 2.

Ex vivo tocolytic ability and uterine-selectivity. Representative recording of isometric spontaneous contractility (measured in grams of tension) of mouse or human myometrial tissue (A and C, respectively) prior to the addition of increasing concentrations (10 pm - 100 μM) of vehicle control, mundulone or mundulone acetate (MA). Recordings were analyzed for contractile AUC (B and D). Non-linear regression was used to fit the data (mean + SEM) and to calculate IC₅₀ and E_max. C. Representative tracing of fetal mouse ductus arteriosus (DA) tone (measured by lumen diameter) prior to and after the addition of increasing concentrations (1 nm - 10 μM) of vehicle control, mundulone or MA. D. Recordings were analyzed for % change from baseline lumen diameter. A 2-way ANOVA with a post-hoc Tukey analysis was used to compare the E_max values (shown).

Fig. 3.

Identification of mundulone and mundulone acetate (MA) synergistic combinations with clinical tocolytics. A. Plate map used for the high-throughput combination Ca²⁺-assay in which one 384-well plate allows testing of six different combinations of two-compounds. Controls (no compounds, white squares) and individual compounds “A” (columns 1, 9 and 17) and “B” on (rows H and I, respectively) were included on each plate. The direction of the gradient shows the increasing concentrations of compound “A” and “B”. The %response data for each compound concentration was averaged and then analyzed with Combenefit software to provide synergy scores using the Bliss-independence model. Heat maps of synergy scores are shown (B-D). Red or black boxes indicate the three fixed ratios chosen for CRC analysis to confirmation synergistic potency or efficacy. Concentration-response curves of combinations at fixed ratio indicated in red box is shown (E-G). A 2-way ANOVA with a post-hoc Tukey analysis was used to compare the E_max values (shown).

Fig. 4.

Ex vivo tocolytic effect of mundulone and atosiban combination. Representative recording of isometric spontaneous contractility (measured in grams of tension) of mouse or human myometrial tissue (A and C, respectively) prior to the addition of increasing concentrations (10 pm - 100 μM) of mundulone +atosiban at a fixed ratio (3.7:1), as well as their single-compound controls (mundulone or atosiban). Recordings were analyzed for contractile AUC (B and D). Non-linear regression was used to fit the data (mean + SEM) and to calculate EC₅₀ and E_max. A 2-way ANOVA with a post-hoc Tukey analysis was used to compare the E_max values (shown).

Fig. 5.

In vivo therapeutic efficacy of mundulone and atosiban combination to delay delivery in a preclinical mouse model of PTB. A. Experimental timeline for induction of preterm labor and birth using mifepristone (MIF) on day 15 of pregnancy, followed by administration of either vehicle, test or control drugs in mice. A. Determining the minimal effective dose of mifepristone to induce PTB in mice. Dose of mifepristone versus timing of delivery (B) or rate of PTB (C) in CD-1 pregnant mice. Timing of delivery (D) and PTB rates (E) for mice receiving either 30 μg MIF (s.c.) prior to administration of either vehicle (Veh), mundulone + atosiban combination (Comb), or mundulone (Mund), or atosiban as single agents (Atos). Nifedipine (Nif) was used as a control tocolytic. Doses listed are mg/kg. Dotted line represents PTB, defined as 24hrs prior to average timing of term delivery of untreated mice.