The authors have declared that no competing interests exist.
Conceived and designed the experiments: JM IA SD. Performed the experiments: JM IA. Analyzed the data: JM IA. Contributed reagents/materials/analysis tools: PYC PB SD. Wrote the paper: JM IA PYC PB SD.
Danshen, in particular its derivative tanshinone IIA (TS), is a promising compound in the treatment of cardiovascular diseases and has been used for many years in traditional Chinese medicine. Although many actions of TS have been researched, its vasodilator effects in pregnancy remain unknown. There have been a few studies that have shown the ability of TS to reduce blood pressure in women with hypertensive pregnancies; however, there are no studies which have examined the vascular effects of TS in the pregnant state in either normal or complicated pregnancies. Our aim was to determine the vasoactive role of TS in multiple arteries during pregnancy including: rat resistance (mesenteric and uterine) and conduit (carotid) arteries. Further, we aimed to assess the ability of TS to improve uterine blood flow in a rodent model of intrauterine growth restriction. Wire myography was used to assess vascular responses to the water-soluble derivative, sodium tanshinone IIA sulphonate (STS) or to the endothelium-dependent vasodilator, methylcholine. At mid-pregnancy, STS caused direct vasodilation of rat resistance (pEC50 mesenteric: 4.47±0.05 and uterine: 3.65±0.10) but not conduit (carotid) arteries. In late pregnancy, human myometrial arteries responded with a similar sensitivity to STS (pEC50 myometrial: 3.26±0.13). STS treatment for the last third of pregnancy in eNOS-/- mice increased uterine artery responses to methylcholine (Emax eNOS-/-: 55.2±9.2% vs. eNOS-/- treated: 75.7±8.9%, p<0.0001). The promising vascular effects, however, did not lead to improved uterine or umbilical blood flow
To date, there have been three clinical studies which showed positive maternal effects of TS on hypertension in pregnancy [
In light of its effects on oxidative stress, inflammatory state and vasodilation, (S)TS may provide a potential treatment strategy for compromised pregnancies which involve pathological changes in these areas. We hypothesized that STS would cause vasodilation of maternal resistance arteries, including the mesenteric and uterine vasculature, in pregnancy. Further, we hypothesized that treatment with STS would improve uterine artery vasodilation in an animal model of complicated pregnancy; the eNOS-/- mouse model of intrauterine growth restriction, via actions on non-NO mediated vasodilator pathways.
All protocols were approved by the University of Alberta Health Sciences Animal Policy and Welfare Committee in accordance with the guidelines of the Canadian Council on Animal Care and the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health.
Three month old female Sprague-Dawley rats (Charles River, Wilmington, MA) were maintained on
Arteries were cleaned of all surrounding adipose and connective tissues and 2 mm sections were mounted on two 40 μm wires attached to a wire myograph (DMT, Copenhagen, SV, Denmark) to allow isometric tension recordings. Vessels were normalized through a series of stepwise increases in diameter to determine their optimal resting tension, set to 0.8 x IC100 (the internal circumference equivalent to a transmural pressure of 100 mmHg).
Following a 30-minute equilibration period, vessels were twice exposed to a single dose of phenylephrine (PE, 10 μmol/l) followed by a single dose of methylcholine (MCh, 3 μmol/l) to check functional endothelial and smooth muscle integrity. A cumulative concentration-response curve to the adrenergic agonist phenylephrine (PE, 0.1 to 30 μmol/l, mesenteric and uterine arteries) or the thromboxane mimetic U46619 (U19, 0.01 to 3 μmol/l, carotid arteries) was performed. The EC80 (the effective concentration producing 80% of the maximum response) for each vasoconstrictor/artery combination was then determined.
Vascular responses to STS (1 to 100 μmol/l) were investigated following preconstriction with the EC80 concentration of the relevant vasoconstrictor. STS responses were performed in the absence or presence of inhibitors to investigate some of the possible vasodilator mechanisms. The three main endothelium-dependent vasodilator pathways; nitric oxide, prostaglandins and endothelial derived hyperpolarisation (EDH), were investigated using inhibitors of nitric oxide synthase (NOS) (
In a separate series of experiments, the endothelium of mesenteric and uterine arteries was denuded using a knotted, human hair. The lack of endothelium was confirmed by assessment of responses to MCh (3 μmol/l). Vascular responses to STS (1 to 100 μmol/l) were then assessed following preconstriction with the EC80 concentration of the relevant vasoconstrictor.
Proof of principle studies were performed in a small number of human arteries (n = 12 artery sections from n = 3 patients) to determine if STS has vasodilator effects in human tissues from a relevant vascular bed. All human tissues were obtained from patients attending the Royal Alexander Hospital. All patients provided informed written consent to participate and the procedure was approved following full ethics review by the Alberta Health Services Ethics Committee, Edmonton. Signed consent forms were retained as documentation of participant consent. A myometrial biopsy from the uterus was obtained during scheduled caesarean sections of healthy pregnancies, mean gestational age 38 weeks and 6 days (st.dev. 4 days), without labour. The biopsy was immediately placed into ice-cold modified Kreb’s solution for collection and transportation. The tissue arrived in the laboratory and myometrial vessels were isolated within 60 minutes of delivery. Isolation of vessels and subsequent experimental protocols were performed in PSS as per animal vessels. Responses to the thromboxane mimetic U46619 (U19, 0.01 to 3 μmol/l) were investigated to assess the EC80 following which vascular responses to STS (1 to 100 μmol/l) were determined.
eNOS-/- (strain B6.129P2-
Systolic blood pressure was measured by a validated tail-cuff system (IITC Life Science, CA, USA) on GD 17. Uterine and umbilical artery blood flow velocities were assessed
On GD 18, mice were euthanized by exsanguination via cardiac puncture under inhaled isoflurane anaesthesia and the pregnancy outcome was determined by weighing and measuring the pups. The main uterine branch arteries were dissected in ice-cold PSS and prepared for experimental procedures on a wire myograph as described for studies in rat vessels. Following a 30-minute equilibration period, vessels were exposed to PE (10 μmol/l) and MCh (3 μmol/l) to check functional endothelial and smooth muscle integrity. Responses to the adrenergic agonist PE (0.0001 to 10 μmol/l) were performed to determine the EC80. Following constriction with PE, responses to the endothelium-dependent vasodilator MCh (0.0001 to 10 μmol/l) or the nitric oxide donor, sodium nitroprusside (SNP, 0.0001 to 10 μmol/l) in the absence or presence of L-NAME (100 μmol/l) were investigated.
All vascular function data were presented as mean ± standard error of the pEC50 (negative log of the effective concentration that will produce 50% of the maximum response) or the Emax (maximum response). Phenotypical parameters from ultrasound biomicroscopy, blood pressure measurements and offspring biometrics were presented as mean ± standard error. All data were normally distributed as assessed by the Kolmogorov-Smirnov test for Gaussian distribution. The significance of the difference in mean values of continuous variables between groups was determined by a one- or two-way analysis of variance (ANOVA), with Bonferroni post-test for multiple groups. A p value < 0.05 was considered statistically significant.
In mid-gestation rats, STS caused vasodilation of mesenteric resistance arteries (
Vascular responses to STS in the carotid (open squares, n = 8), uterine (open circles, n = 10), and mesenteric arteries (closed circles, n = 8) from the pregnant rat (GD 10.9±0.2).
Inclusion of either the NOS inhibitor L-NAME or inhibition of COX using meclofenamate did not significantly affect STS-induced vasodilation of mesenteric arteries from mid-gestation rats. The presence of the potassium channel blockers apamin and TRAM-34, however, significantly inhibited STS-induced vasodilation compared to the control group (p<0.0001,
Mechanisms of vascular responses to STS in mesenteric (A&C) and uterine (B&D) arteries from the pregnant rat (GD 10.9±0.2). Vasodilator responses to STS in the absence (closed circles) or presence of inhibitors of nitric oxide synthase (NOS) (
Endothelium removal in mesenteric arteries resulted in a significant reduction in maximal vasodilator responses to MCh of 62.1 ± 7.4% (p<0.0001). Despite a 2.6-fold reduction in endothelial function, there was no significant change in sensitivity or maximal responses to STS following endothelium removal (
Vascular responses to STS in A: mesenteric arteries and B: uterine arteries from the pregnant rat (GD 10.9±0.2) with intact endothelium (closed circles) or following endothelial removal using a knotted human hair (open circles). Mesenteric intact n = 6; mesenteric denuded n = 5; uterine intact n = 4; uterine denuded n = 5.
Human myometrial arteries responded to STS with a vasodilator response that was of similar potency (pEC50: 3.26 ± 0.13) and efficacy (Emax: 89.07 ± 3.42%) to that observed in female pregnant rat uterine arteries. Human arteries, however, did not demonstrate the contractile component of the biphasic response observed in rat uterine arteries.
Isolated mouse uterine arteries from GD 18 demonstrated reduced PE-induced vasoconstriction in eNOS-/- mice that was increased following six days of STS treatment (Emax control untreated: 2.04 ± 0.31 mN/mm vs. treated: 2.94 ± 0.15 mN/mm; eNOS-/- untreated 1.49 ± 0.13 mN/mm vs. treated 1.88 ± 0.19 mN/mm; group genotype effect p<0.001, group treatment effect p<0.01). MCh-induced vasodilation was also significantly reduced in eNOS-/- mice compared to control mice (p<0.0001,
A: Vascular responses of mouse uterine arteries to MCh from control (circles) and eNOS-/- (squares) late pregnant (GD 18) mice, untreated (closed symbols) or treated (open symbols) with STS from GD 12 to 18. STS was given in drinking water at a dose of approx. 27 mg/kg/day. Summary data of B: maximal responses and C: sensitivity (negative log of the effective concentration producing 50% of the maximal response) to MCh. Data analysed by two-way ANOVA with a Bonferroni post-test; **: p<0.01 group treatment effect, ****: p<0.0001 group genotype effect, ††: p<0.01 vs. eNOS-/- untreated, †††: p<0.001 vs. control untreated. Control untreated n = 5; control treated n = 5; eNOS-/- untreated n = 6; eNOS-/- treated n = 4.
Following the observation of direct STS-induced vasodilation and STS-mediated upregulation of MCh-induced vasodilation of the uterine vasculature, the ability of STS to improve uterine and umbilical artery blood flow and the outcomes of compromised pregnancy in an animal model of IUGR was tested.
Echocardiographic parameters of umbilical blood flow from control (white bars) and eNOS-/- (black bars) late pregnant (GD 18) mice, untreated (open bars) or treated (hatched bars) with STS from GD 12 to 18. STS was given in drinking water at a concentration of approx. 27 mg/kg/day. A: end diastolic velocity (EDV), B: peak systolic velocity (PSV) and C: calculated resistance index (RI). Data analysed by two-way ANOVA with a Bonferroni post-test; **: p<0.01, ***: p<0.001, group genotype effect. Control untreated n = 4; control treated n = 8; eNOS-/- untreated n = 6; eNOS-/- treated n = 8.
In terms of pregnancy outcomes, the eNOS-/- genotype caused a reduction in offspring birth weight (p<0.0001,
Pregnancy outcomes in control (white bars) and eNOS-/- (black bars) late pregnant (GD 18) mice, untreated (open bars) or treated (hatched bars) with STS from GD 12 to 18. STS was given in drinking water at a concentration of approx. 27 mg/kg/day. A: pup weight, B: pup crown to rump length, C: maternal systolic blood pressure and D: placental weight. Data analysed by two-way ANOVA with a Bonferroni post-test; **: p<0.01, ***: p<0.001, group genotype or treatment effects; ††: p<0.01 vs. control untreated. Control untreated n = 5(33); control treated n = 8(60); eNOS-/- untreated n = 6(30); eNOS-/- treated n = 7(43); where n = dams(pups).
This study demonstrated a direct vasodilator effect of STS in vascular resistance arteries from pregnant rats. We have also demonstrated that STS was effective not only in dilating uterine arteries from the pregnant rat but also similarly dilates human myometrial arteries; providing proof of principle that STS may be effective in human pathology. Further, we showed that STS was able to improve the vasodilator capacity of uterine arteries when given
STS caused non-NO dependent vasodilation of rat resistance (mesenteric and uterine) but not conduit (carotid) arteries. This was in line with our own studies in male rat mesenteric arteries [
The effect of STS on the uterine vasculature during the pregnant state was of greater interest. Responses of rat uterine arteries to STS had a distinctly biphasic character that has also been observed in a previous study of rat pulmonary arteries [
The effect of STS was also assessed in the closest available human equivalent to the rodent uterine artery, the myometrial artery. Excitingly, not only were vasodilator responses observed, these responses in the myometrial artery were consistent in efficacy and potency to those observed in the rat uterine artery suggesting that the effects of STS were translatable from rat to human arteries.
A vasodilator of the uterine vasculature during pregnancy has potential clinical applicability, particularly for disorders that involve a decrease in placental and fetal perfusion. In particular, the endothelium-independence of STS responses provides a potential therapeutic approach in conditions characterized by endothelial dysfunction, such as preeclampsia or IUGR. Therefore, we next tested the ability of
Several concerns regarding the use of STS in pregnancy were raised in regards to its effect on the control animals. STS treatment caused a biphasic response in rat uterine arteries which included almost 30% constriction prior to the vasodilator response. Further, adrenergic-mediated vasoconstriction in mouse uterine arteries was increased by STS treatment. These observations might suggest a risk for maternal hypertension or decreased uterine blood flow as a result of a hyper-constrictive phenotype. The increased adrenergic constriction following STS treatment, however, normalized the reduced PE constriction observed in untreated eNOS-/- mice compared with untreated control mice. In addition, in our
In summary, we have demonstrated vasodilation to the Danshen derivative, STS, in resistance arteries that was consistent between rodent and human arteries. The