AMPK activation in pregnant human myometrial arteries from high-altitude and intrauterine growth-restricted pregnancies

Abstract

High-altitude (>2,500 m) residence increases the incidence of intrauterine growth restriction (IUGR) due, in part, to reduced uterine artery blood flow and impaired myometrial artery (MA) vasodilator response. A role for the AMP-activated protein kinase (AMPK) pathway in protecting against hypoxia-associated IUGR is suggested by genomic and transcriptomic studies in humans and functional studies in mice. AMPK is a hypoxia-sensitive metabolic sensor with vasodilatory properties. Here we hypothesized that AMPK-dependent vasodilation was increased in MAs from high versus low-altitude (<1,700 m) Colorado women with appropriate for gestational age (AGA) pregnancies and reduced in IUGR pregnancies regardless of altitude. Vasoreactivity studies showed that, in AGA pregnancies, MAs from high-altitude women were more sensitive to vasodilation by activation of AMPK with A769662 due chiefly to increased endothelial nitric oxide production, whereas MA responses to AMPK activation in the low-altitude women were endothelium independent. MAs from IUGR compared with AGA pregnancies had blunted vasodilator responses to acetylcholine at high altitude. We concluded that 1) blunted vasodilator responses in IUGR pregnancies confirm the importance of MA vasodilation for normal fetal growth and 2) the increased sensitivity to AMPK activation in AGA pregnancies at high altitude suggests that AMPK activation helped maintain MA vasodilation and fetal growth. These results highlight a novel mechanism for vasodilation of MAs under conditions of chronic hypoxia and suggest that AMPK activation could provide a therapy for increasing uteroplacental blood flow and improving fetal growth in IUGR pregnancies.

NEW & NOTEWORTHY Intrauterine growth restriction (IUGR) impairs infant well- being and increases susceptibility to later-in-life diseases for mother and child. Our study reveals a novel role for AMPK in vasodilating the myometrial artery (MA) from women residing at high altitude (>2,500 m) with appropriate for gestational age pregnancies but not in IUGR pregnancies at any altitude.

INTRODUCTION

Intrauterine growth restriction (IUGR) is approximately threefold more common in residents of high altitudes (>2,500 m) compared with those at sea level and is principally due to the reduction in oxygen and other nutrient availability (4, 13, 30). IUGR at low altitudes is marked by a lesser rise in uterine artery blood flow across gestation, which is largely the result of impaired myometrial artery (MA) vasodilation and incomplete remodeling of uterine spiral arteries (17, 24, 29, 33). Women residing at high altitude in Colorado have reduced uterine artery blood flow and fetal growth relative to their low-altitude (<1,700 m) counterparts (13). Whereas vasoconstrictor responses in MAs were similar in women living at high or low altitude, we have recently shown that MAs from high-altitude women had reduced vasodilation in response to acetylcholine (ACh) and less ACh- or bradykinin (BK)-stimulated nitric oxide (NO) production (22).

Our prior work has shown that multigenerational Andean high-altitude residents are protected from altitudinal reductions in uterine artery blood flow and fetal growth compared with their shorter term (European) counterparts due to genetic factors (14–16). In Andeans, we identified multiple single nucleotide polymorphisms (SNPs) showing evidence of recent positive selection and, among these, one SNP (rs1345778) located near PRKAA1, which encodes the catalytic subunit of the AMP-activated kinase (AMPK), was positively associated with birth weight and uterine artery diameter at high altitude (2), the latter in accordance with the well-known vasodilator effects of AMPK activation (8, 27, 31). Suggesting a functional link, evidence shows that maternal PRKAA1 SNP genotype (rs1345778) was also associated with gene-expression patterns in key pathways that have been proposed to play a role in the pathophysiology of fetal growth restriction (2).

AMPK is a metabolic sensor that is activated by ATP depletion, hypoxia, and other stressors (11, 34). AMPK also has potent vasodilator effects, promoting endothelial cell NO production (27) and relaxing vascular smooth muscle cells (8) through the modulation of sarcoplasmic Ca²⁺ drive and Ca²⁺-activated K⁺ channels (31). Suggesting that AMPK is involved in the regulation of uteroplacental blood flow under hypoxic conditions, evidence shows that pharmacologic AMPK activation vasodilated isolated murine uterine arteries, an effect that was increased under hypoxic conditions (32). Furthermore, we have recently reported that in vivo AMPK activation protected against hypoxia-associated reductions of fetal growth in mice (6).

While the incidence of IUGR is increased at high altitude, not all babies are affected. Hence, here we tested the hypothesis that AMPK-dependent vasodilation was increased in MAs from high altitude compared with low altitude from women with appropriate-for-gestational age (AGA) fetuses but, conversely, decreased in IUGR versus AGA pregnancies regardless of altitude. We compared the vasodilator responses to pharmacologic AMPK activation in isolated MAs obtained at Cesarean section from women residing at low or high altitude in Colorado with AGA or IUGR pregnancies. The present study highlights an as-yet undescribed mechanism for maintaining uterine blood flow under conditions of chronic hypoxia and suggests that AMPK activation could provide a novel therapy for dilating MAs and improving fetal growth in IUGR pregnancies.

MATERIALS AND METHODS

Human subjects.

Pregnant women residing at low (<1,700 m, n = 41) or high altitude (>2,500 m, n = 26) in Colorado were recruited after signing consent forms approved by the University of Colorado Multiple Institutional Review Board (Approval No. 14-2178) and the Catholic Health Initiative Institute for Research and Innovation Institutional Review Board (Approval No. 1310). Inclusion criteria for all women were maternal age between 18 and 45 yr; a prepregnant body mass index less than 30 kg/m²; the absence of known risk factors for IUGR (i.e., diabetes, chronic hypertension, gestational hypertension, preeclampsia, fetal congenital anomalies, and IUGR or preeclampsia in a prior pregnancy); residence at either low or high altitude throughout pregnancy, with less than 2 wk total time spent visiting low altitudes for the high-altitude group; and no known highland ancestry. AGA and IUGR pregnancies were defined by an estimated fetal weight either >10th percentile or ≤10th percentile by ultrasound throughout pregnancy, respectively (10). Low-altitude subjects were recruited at the University of Colorado Hospital (Aurora, CO, elevation 1,640 m) and high-altitude subjects at St. Anthony’s Summit Medical Center (Frisco, CO, elevation 2,793 m). No women at either altitude were diagnosed with preeclampsia or any other hypertensive disorder of pregnancy, and their predelivery blood pressures were within normal ranges (<140/90 mmHg). All low-altitude women delivered at the University of Colorado Hospital in Aurora, CO, and all high-altitude subjects delivered at St. Anthony’s Summit Medical Center in Frisco, CO. In addition, information regarding place of birth from the high-altitude subjects, demographic information, health and obstetric history, labor and delivery information, and newborn characteristics were obtained from medical records or self-reported questionnaire. None of the high-altitude subjects was born at high altitude, and only 4 out of 26 were raised at high altitude.

Myometrial samples.

Myometrial biopsies from the upper lip of the lower uterine segment were obtained from nonlaboring women undergoing elective Cesarean section at term (≥37 wk of gestation) under spinal anesthesia. Myometrial tissue was washed with ice-cold phosphate buffered saline (PBS; Thermo Fisher Scientific, Waltham, MA), a 0.5-cm³ section was immediately fixed in 4% paraformaldehyde (PFA; Affymetrix, Cleveland, OH), and the rest was stored in ice-cold PBS until experiments were performed. High-altitude specimens were stored in ice-cold PBS during their ~2-h transport to University of Colorado Denver for study; tissues obtained from the University of Colorado Hospital were also stored for an equivalent period in ice-cold PBS. MA vasoreactivity is intact after such storage (7, 22), and transport from high to low altitudes does not affect altitude-related changes in vasoreactivity (12, 25).

Immunohistochemistry.

The myometrial samples were fixed in 4% PFA for at least 24 h at room temperature and then embedded in paraffin. Sections (5 µm) were adhered to slides, deparaffinized with xylenes, and rehydrated. Slides were then treated with an antigen unmasking, citrate-based solution (Vector Laboratories, Burlingame, CA) for antigen retrieval and stained with either anti-AMPKα antibody (1:100; rabbit monoclonal, No. ab32047, Abcam, Cambridge, MA), anti-phosphorylated-AMPKα antibody (Thr-172, 1:50; rabbit monoclonal, No. CST50081, Cell Signaling Technology, Danvers, MA), or anti-CD31 antibody (1:50; rabbit polyclonal, No. ab28364, Abcam) for endothelial cell staining and then mounted and imaged with a confocal microscope (Olympus, Waltham, MA). These antibodies have been used in immunohistochemistry studies to stain human tissue and cells (5, 20, 21). Five randomly chosen ×400 magnification images per subject were obtained for small myometrial arterioles using identical camera settings and lengths of exposure. Staining was quantified as the number of positive pixels within an area encompassing the tunica intima and tunica media, excluding the luminal area, using Slidebook software (Intelligent Imaging Innovations, Denver, CO). Positive pixels were expressed as percentage of the total number of pixels. The identity (i.e., low vs. high altitude) of each slide was blinded to the investigators taking the microscope images.

Myography.

Isolated radial MAs were mounted in a small-vessel wire myograph (Multi Wire Myograph System 610M, DMT-USA, Ann Arbor, MI) with two wires (40-µm diameter) threaded through the vessel lumen and connected to either a tension transducer or micrometer in a chamber containing oxygenated (95% O₂-5% CO₂) and warmed (37°C) Krebs buffer (118 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl₂, 1.2 mM MgSO₄, 1.2 mM KH₂PO₄, 25 mM NaHCO₃, and 11 mM d-glucose). MAs were normalized to an internal diameter of 0.9 of L_13.3kPa. After at least 45 min of equilibration, MAs were constricted with 60 mM KCl to establish viability; vessels failing to constrict at least 1 mN (estimated as an appropriate signal-to-noise ratio for tension measurement) were excluded from further study. The effect of AMPK activation on opposing vasoconstriction was assessed by applying increasing concentrations of phenylephrine (PE; 1 nM to 100 µM, Sigma-Aldrich) in the presence or absence of the AMPK agonist A769662 (30 µM, pretreated for 20 min; Cayman Chemical, Ann Arbor, MI) with each PE concentration being added for 3–5 min before the next concentration. MA vasodilation was assessed in vessels preconstricted with 10 µM PE; the vasodilator responses to either A769662 (0.1 nM to 100 µM), acetylcholine (ACh; 0.1–100 µM, Thermo Fisher Scientific), or bradykinin (BK; 0.1–1 µM, Thermo Fisher Scientific) were measured for 3–5 min before addition of the next concentration; and the value after a stable plateau was reached was recorded. To assess the contributions of endothelial factors to the AMPK vasodilator response observed, vessels were incubated with either the NO synthase inhibitor N^G-nitro-l-arginine methyl ester (l-NAME; 10 µM, BioVision, Inc., Milpitas, CA) or l-NAME (10 µM) plus the cyclooxygenase inhibitor indomethacin (Indo; 10 µM, Sigma-Aldrich) for 20 min before PE-constriction and the application of A769662. In some MAs, the endothelial cell layer was mechanically disrupted by rolling a hair through the vessel lumen; loss of endothelial function was judged as <30% relaxation to 10 µM ACh or 10 nM BK. The force elicited by PE was normalized to that evoked by 60 mM KCl (K_max), whereas A769662 relaxation was calculated as percentage of 10 µM PE contraction. Half-maximal effective concentration (EC₅₀), maximal force, and area under the curve (AUC) were calculated using Graph Pad 8 software (San Diego, CA). Due to the limited number of vessels from each subject, it was not possible to apply all the treatments to the same vessel or vessels from the same subject.

Statistical analyses.

Sample sizes in the figures and tables reflect the number of subjects. On average, 3.2 ± 0.2 vessels per subject (range from 1 to 8) were used for vasoreactivity studies. Values were averaged within a subject in cases where more than one vessel per subject was used for a given experimental condition. Quantification of histological data was subjected to two-way analysis of variance (ANOVA) followed by Sidak's multiple comparisons (Graph Pad 8 software). The effects of drug treatment or altitude were determined using Kruskal-Wallis or two-way ANOVA followed by Dunn’s or Sidak's multiple comparisons, respectively, when comparing concentration-response curves EC₅₀, maximal force, or AUC (Graph Pad 8). Demographic characteristics were analyzed by nonparametric (Mann-Whitney U or Fisher’s) tests as appropriate using Graph Pad 8 software. Birth weights adjusted for gestational age at the time of delivery and infant sex were compared between groups using analysis of covariance (ANCOVA) within SPSS v26 (IBM, Chicago, IL). P < 0.05 was considered significant.

RESULTS

Maternal and infant characteristics.

Maternal and infant characteristics were similar between low- and high-altitude residents with AGA pregnancies, with the exception of birth weight, length and head circumference, which were lower at high altitude compared with low altitude (Table 1). Apart from the expected reductions in birth weight, length, and head circumference in IUGR compared with AGA at both altitudes, maternal and other fetal characteristics were similar between AGA and IUGR women at both altitudes (Table 1). Gestational age was modestly but clinically insignificantly shorter in high-altitude AGA versus low-altitude AGA pregnancies and between AGA and IUGR at both altitudes (Table 1). Mean arterial pressures before Cesarean section were slightly higher in low altitude IUGR compared with AGA (Table 1) but below 140/90 mmHg in all cases.

Table 1.

Maternal and fetal characteristics

	AGA			IUGR
Characteristic	Low altitude	High altitude	P Value Low vs. High Altitude	Low altitude	High altitude	P Value Low vs. High Altitude	P Value AGA vs. IUGR
n	36	23		5	3
Maternal
Age at delivery, yr	32.5 ± 0.8	32.2 ± 1.3	0.6845	33.6 ± 2.1	33.7 ± 4.8	0.5179	0.3095 low 0.4827 high
Height, cm	164.1 ± 1.5	165.6 ± 1.4	0.2782	162.2 ± 2.5	160 ± 1.7	0.3929	0.2939 low 0.0443 high
Pregravid BMI, kg/m2	24.4 ± 0.5	23.4 ± 0.8	0.0909	25.25 ± 1.8	20.9 ± 0.8	0.0714	0.3426 low 0.3635 high
Parity (No. of live births)	2.3 ± 0.1	2.1 ± 0.2	0.1452	1.8 ± 0.2	1.67 ± 0.3	0.6429	0.0761 low 0.3289 high
Ethnicity, %Hispanic	13.9	21.7	0.4902	20	33.3	>0.9999	0.5668 low >0.9999 high
Altitude of residence, m	1,667 ± 10	2,878 ± 35	<0.0001	1605 ± 26	2667 ± 238	0.0179	0.0202 low 0.4504 high
MAP at term, mmHg	86 ± 1.0	84.9 ± 2.1	0.8253	93.3 ± 4.5	82.9 ± 4.0	0.0714	0.0343 low 0.6347 high
Infant
Gestational age, wk	39.2 ± 0.1	38.8 ± 0.1	0.001	37.5 ± 0.4	35.5 ± 0.5	0.4464	0.0018 low 0.0351 high
Birth weight, g	3,561 ± 81	3,036 ± 69	<0.0001	2,599 ± 76	2,250 ± 31	0.0357	0.0002 low 0.0015 high
Birth weight, adjusted, g	3,509 ± 83	3,078 ± 111	0.004	2,745 ± 245	2,354 ± 265	0.230	0.009 low 0.014 high
Length, cm	50.4 ± 0.4	48.8 ± 0.4	0.0172	47.3 ± 0.8	45.2 ± 0.4	0.1786	0.0109 low 0.0008 high
Head circumference, cm	35.4 ± 0.2	34.5 ± 0.2	0.0044	33.9 ± 0.5	32.6 ± 0.2	0.1071	0.0204 low 0.0035 high
Ponderal Index	2.8 ± 0.04	2.6 ± 0.08	0.1261	2.5 ± 0.1	2.4 ± 0.04	0.9643	0.0451 low 0.1797 high
Sex, %female	52.8	36.8	0.3947	20	66.7	0.4643	0.3433 low 0.5442 high

Values are means ± SE unless noted. AGA, appropriate for gestational age; IUGR, intrauterine growth restriction; BMI, body mass index; MAP, mean arterial pressure. P values were estimated by nonparametric Mann-Whitney U test or Fisher’s test. Birth weights were adjusted for gestational age at the time of delivery and fetal sex and compared between groups using analysis of covariance (ANCOVA). Boldfaced values indicate significant values.

AMPK is expressed and phosphorylated in myometrial vessels.

Immunohistological analysis showed distinct AMPK and phosphorylated-AMPK (p-AMPK) protein expression in smooth muscle cells of the myometrial stroma and the vascular smooth muscle layer of myometrial vessels. Vascular smooth muscle cells in MAs were identified by their proximity to endothelial CD31 staining. Staining with CD31 suggests that AMPK and p-AMPK were expressed in both myometrial vascular endothelial and smooth muscle cells (Fig. 1A). Quantitative analysis did not indicate differences in AMPK or p-AMPK staining intensity in MAs between altitudes (Fig. 1, B and C).

Fig. 1.

AMPK is expressed and phosphorylated in myometrial vessels from women at low and high altitude. A: representative microscope pictures of myometrial tissue from appropriate for gestational age (AGA) or intrauterine growth-restricted (IUGR) pregnant women residing at low or high altitude showing staining of AMPK, p-AMPK, and CD31. Yellow arrowheads show blood vessels, orange arrowheads show blood vessels zoomed-in for each inset, and green staining shows localization of endothelial cells. Scale bars = 50 µm (10 µm, insets). B and C: Quantification of AMPK and p-AMPK staining in myometrial arteries (MAs) from low (closed symbols)- and high-altitude (open symbols) subjects, expressed as percentage of stained-positive pixels within the vessels. Symbols are averaged values per subject; bars are means ± SE. Same italicized letters represent no statistical differences among groups with a P < 0.05 by two-way ANOVA.

AMPK activation similarly opposed PE vasoconstriction in MAs from AGA low- and high-altitude pregnancies.

Isolated MAs from low- and high-altitude women with AGA pregnancies constricted to PE in a similar manner (untreated vessels in Fig. 2, A and B, and Table 2). AMPK activation with A769662 reduced the vasoconstrictor response to PE at both altitudes (Fig. 2, A and B). At low altitude, both PE maximum response and sensitivity were reduced by A769662 (P < 0.05, Fig. 2, C and D, and Table 2), whereas at high altitude A769662-treatment diminished PE maximal response (P < 0.05, Fig. 2E and Table 2) but not sensitivity (Fig. 2F and Table 2).

Fig. 2.

AMPK activation reduced vasoconstriction in both low- and high-altitude myometrial arteries (MAs). Concentration-response curves of phenylephrine (PE) expressed as percentage of maximal K⁺ constriction (K_max) in the presence (+ A769662, gray symbols) or absence (untreated, black symbols) of 30 µM A760662 in MAs from low (A; closed symbols)- and high-altitude (B; open symbols) women. Number in parentheses are number of subjects per group. Symbols are mean values ± SE. Half-maximal effective concentration (EC₅₀) and maximal responses (Max. response) analysis of PE concentration-response curves at low (C and D) and high altitude (E and F). Symbols are averaged vessels per subject; bars are means ± SE. Different italicized letters represent statistical differences among EC₅₀ or maximal responses with a P < 0.05 by Kruskal-Wallis.

Table 2.

PE half-maximal concentrations and maximal responses in MAs

	EC50, µM		P Value		Maximal Response, %Kmax		P Value
Vasoactive Agent + Treatment	Low altitude	High altitude	Low vs. High altitude	Untreated vs. Treated	Low altitude	High altitude	Low vs. High altitude	Untreated vs. Treated
PE (untreated)	1.0 ± 0.3 (11)	1.5 ± 0.3 (13)	0.3458		94.8 ± 16.4	75.0 ± 17.4	0.7222
+A769662	3.6 ± 1.2 (14)	3.1 ± 0.8 (10)	>0.9999	0.0112 low 0.2960 high	39.1 ± 6.9	37.2 ± 9.7	>0.9999	0.0153 low 0.0446 high

Values are means ± SE; n, number of subjects in parentheses. EC₅₀, half-maximal concentrations; MAs, myometrial arteries; PE, phenylephrine. P values were estimated by Kruskal-Wallis test. Boldfaced values indicate significant values.

Greater AMPK-dependent vasodilation in MAs from high-altitude compared with low-altitude AGA women.

Increasing doses of the pharmacologic AMPK agonist A769662 vasodilated PE-preconstricted MAs from both altitude groups but did so to a greater extent at high altitude as illustrated by the downward shift in the concentration-response curve and reduced AUC (P < 0.05, Fig. 3, A and B, and Table 3). To evaluate the contribution of endothelial factors in the vasodilator response to A769662, we compared vessels treated with l-NAME or l-NAME plus indomethacin or that had been denuded of their endothelium. In low-altitude vessels, AMPK-dependent vasodilation was largely endothelium independent as demonstrated by the lack of an effect of incubation with l-NAME, l-NAME plus Indo, or mechanical endothelium removal (Fig. 3, C and D, and Table 3). In contrast, AMPK-dependent vasodilation in high-altitude vessels was partially reduced by l-NAME or l-NAME plus Indo (P < 0.05 and P < 0.01, respectively), whereas endothelium removal did not change A769662 responses (Fig. 3, E and F, and Table 3).

Fig. 3.

AMPK-dependent vasodilation was increased by altitude in myometrial arteries (MAs) in an endothelium-dependent manner. A: A769662 concentration-response curves expressed as percentage of 10 µM phenylephrine (PE) in MAs from low (closed circles)- or high-altitude (open symbols) women. C and E: before A769662 curves were performed, MAs from low (C, closed symbols) or high altitude (E, open symbols) were pretreated with 10 µM N^G-nitro-l-arginine methyl ester (l-NAME (squares) or 10 µM l-NAME + 10 µM Indo (diamonds), or their endothelium was removed (−endothelium, triangles). Numbers in parentheses are number of subjects per group. Symbols are means values ± SE. B, D, and F: area under the curve (AUC) analysis of A769662 concentration curves in A, C, and E, respectively. Symbols are averaged vessels per subject; bars are means ± SE. Different italicized letters represent statistical differences among AUC with a P < 0.05 by two-way ANOVA.

Table 3.

Area under the curve analysis of MA vasoreactivity studies

	AUC, AU
Vasoactive Agent + Treatment	Low altitude	High altitude	P Value Low vs. High Altitude	P Value Untreated vs. Treated or AGA vs. IUGR
AGA
A769662 (untreated)	342.5 ± 18.2 (17)	273.1 ± 22.1 (11)	0.0459
+l-NAME	347.9 ± 20.31 (11)	343.5 ± 25.2 (9)	0.9775	0.9959 low 0.0144 high
+l-NAME + Indo	342.6 ± 13.5 (7)	400.6 ± 36.3 (7)	0.4080	>0.9999 low 0.0009 high
−Endothelium	360.5 ± 21.7 (4)	308.8 ± 32.4 (7)	0.6607	0.9537 low 0.6415 high
IUGR
A769662	374.8 ± 4.5 (4)	318.3 ± 24.6 (3)	0.4878	0.6305 low 0.5418 high

Values are means ± SE; n, number of subjects in parentheses. MA, myometrial artery; AUC, area under the curve; AU, arbitrary units; AGA, appropriate for gestational age; IUGR, intrauterine growth restriction; l-NAME, N^G-nitro-l-arginine methyl ester. P values were estimated by two-way ANOVA. Boldfaced values indicate significant values.

MAs from high-altitude IUGR pregnancies had reduced ACh-dependent vasodilation.

Because MA vasodilation has previously been shown to be impaired in IUGR pregnancies (29, 33), we assessed whether MA vasodilation from IUGR pregnancies at either low or high altitude was also reduced. In both low- and high-altitude vessels, the vasodilatory response to AMPK activation did not change in IUGR compared with AGA pregnancies (Fig. 4 and Table 3), although there was a nonsignificant reduction of AMPK vasodilation at high altitude. The MA vasodilator response to AMPK activation was also similar between low- and high-altitude IUGR pregnancies (Fig. 4 and Table 3). However, similar to the high altitude-dependent reduction of ACh responses observed in our previous study (22), we observed a blunted response to ACh in MAs from IUGR pregnancies at low altitude (P < 0.05, Fig. 5, A and B), whereas the MA vasodilator response to BK was unchanged in IUGR pregnancies at both altitudes (Fig. 5, C and D).

Fig. 4.

Myometrial artery (MA) vasodilator response to AMPK activation was similar in intrauterine growth restriction (IUGR) pregnancies at low altitude but not at high altitude. A769662 concentration-response curves expressed as percentage of 10 µM phenylephrine (PE) in MAs from women with appropriate for gestational age (AGA; black symbols) or IUGR (gray symbols) pregnancies residing at low (A; closed symbols) or high altitude (B; open symbols). Numbers in parentheses are number of subjects per group. Symbols are means values ± SE. C: area under the curve (AUC) analysis of A769662 concentration curves at low and high altitude. Symbols are averaged vessels per subject; bars are means ± SE. Different italicized letters represent statistical differences among AUC with a P < 0.05 by two-way ANOVA.

Fig. 5.

Acetylcholine (ACh) but not bradykinin (BK) vasodilator responses were reduced in intrauterine growth restriction (IUGR) pregnancies at both altitudes. ACh (A) and BK (C) concentration-response curves expressed as percentage of 10 µM phenylephrine (PE) in myometrial artery (MA) from IUGR (gray symbols) pregnant women residing at low (closed symbols) or high altitude (open symbols). Numbers in parentheses are number of subjects per group. Symbols are means values ± SE. B and D: area under the curve (AUC) analysis of ACh and BK concentration-curves in A and C, respectively. Symbols are averaged vessels per subject, bars are means ± SE. Same italicized letters represent no statistical differences among groups with a P < 0.05 by two-way ANOVA.

DISCUSSION

Failure of MA vessels to increase their vasodilator response during pregnancy is an important characteristic of pregnancies complicated either by IUGR or preeclampsia at low altitude (18, 23, 29, 33), both of which have been associated with reduced uterine artery blood flow and fetal growth (4, 17). Similarly, women residing at high altitude show reduced MA vasodilation and uterine artery blood flow and give birth to lower birth-weight infants (13, 22). While birth weights statewide in Colorado are lower at high altitude than low altitude and the frequency of IUGR increased, not all babies born at high altitude are growth restricted. Since a protective role for AMPK activation was suggested from our prior human and mouse studies (2, 32), we hypothesized that an increased MA vasodilator response to AMPK protected against IUGR under conditions of chronic hypoxia. To test this hypothesis we studied MAs from AGA as well as IUGR pregnancies at both altitudes, expecting that the MA vasodilator response to AMPK would be greater in AGA pregnancies at high altitude than low altitude but blunted in IUGR pregnancies regardless of altitude. Consistent with our hypothesis, we found that MAs from high- compared with low-altitude women with AGA pregnancies dilated more in response to pharmacologic AMPK activation. However, although not statistically significant, it is worth noting that the vasodilatory response to AMPK activation in IUGR cases at low or high altitude was qualitatively lower compared with AGA cases. While birth weights were, as expected, lower (14%) at high altitude than low altitude, the more profound reduction in birth weight seen in IUGR babies was consistent with the possibility that AMPK-dependent vasodilatory effects were able to protect against modest but not severe reductions in fetal growth.

We undertook the present studies given our previous work in highland Andean populations that identified the AMPK pathway as playing a potentially important role in the preservation of a normal pregnancy rise in uterine artery blood flow and fetal growth (2). Additionally, our functional studies in uterine arteries from pregnant mice showed that AMPK activation acted as a potent vasodilator and that this vasodilation was increased under hypoxic conditions (32). Furthermore, we recently reported that in vivo treatment with an AMPK agonist prevented half the hypoxia-associated reduction in fetal weight in mice (6). These observations suggest that AMPK activation played a previously unrecognized and potentially important role in the maintenance of adequate uteroplacental blood flow to support normal fetal growth.

AMPK is a metabolic sensor that is activated by an increased ratio of intracellular AMP to ATP and cellular stressors, including hypoxia (11, 34). It has been shown to have potent vasodilator effects in isolated aorta and mesenteric vessels that are evoked through both endothelium-dependent and independent mechanisms (8, 27, 31). The present study showed that vasoconstriction to PE alone did not differ in low- versus high-altitude MAs. The increased vasodilator effect of AMPK activation at high altitude appeared due to an effect on increasing vasodilation per se rather than being one of opposing vasoconstriction given that pretreatment with A769662 reduced the vasoconstrictor response to PE similarly at low and high altitude. Furthermore, the role of endothelial cell products in the vasodilator effects of AMPK activation differed at high versus low altitude; namely, vasodilator effects in low-altitude MAs were largely endothelium independent since they were unchanged by endothelium removal or NO or cyclooxygenase blockade, whereas about half the vasodilator response in high-altitude MAs was due to greater production of NO. Since AMPK vasodilation in high-altitude denuded vessels was indistinguishable from that which was observed following l-NAME treatment, endothelial NO production appeared to be the primary cause of the increased AMPK-dependent vasodilation in high-altitude MAs. Of interest, such a central role for increased NO production in the vasodilator response to AMPK at high altitude contrasts with our previous work showing the lack of a NO contribution in the vasodilator responses to ACh and BK in high- versus low-altitude MAs (22). In other words, the present study combined with our previous study indicates that the specific mechanisms by which chronic hypoxia alters MA vasoreactivity are agonist dependent.

IUGR pregnancies show a reduced rise in uterine artery blood flow, which is evident before a slowing of fetal growth can be detected, suggesting that reduced uterine artery blood flow plays an important role in the etiology of the disease (13, 17). This reduced blood flow could be due to less vasodilation, greater vasoconstriction, or impaired remodeling of the downstream uterine resistance arteries such as the MAs and spiral arteries. Although the reduction in oxygen or other nutrient delivery may not be sufficient to account for the slowing of fetal growth at high altitude, it may act as a trigger that, in turn, inhibits mammalian target of rapamycin pathways or other pathways involved in regulating protein synthesis and hence fetal growth (3, 28, 36). The present study showed that MAs from IUGR versus AGA pregnancies had reduced vasodilator responses to AMPK or ACh at high altitude, the latter compared with ACh curves in AGA subjects from our previous studies (22), supporting an important role for MA vasodilation for fetal growth at high altitude consistent with that previously suggested at low altitude (17, 23, 29, 33). While a reduced vasodilator response to BK has been observed in MAs from IUGR pregnancies at low altitude (29), contrasting the IUGR data from present study with AGA data from our previous study (22) indicate that BK responses did not differ in AGA and IUGR. The differences between ours and the previous report could be due to the vasoconstrictor agents used to preconstrict MAs (PE in our study vs. the thromboxane mimetic U46619), a slightly lesser degree of IUGR present here, other differences between the women being compared, or the relatively small sample sizes; future efforts will aim to increase sample collections to elucidate such differences. This is the first study to show that the MA vasodilator response to ACh is reduced as well. Since studies in animal models of IUGR also have shown similar reductions in uterine artery response to cholinergic stimulation (1, 35), future studies are required to determine whether cholinergic mechanisms are involved in decreasing the vasodilation of uterine vessels characteristic of human IUGR. AMPK-dependent MA vasodilation was greater in AGA pregnancies at high altitude than low altitude and apparently enhanced in AGA versus IUGR pregnancies at high altitude; our results are consistent with the possibility that AMPK activation at high altitude helps support MA vasodilation and maintain uteroplacental blood flow in AGA pregnancies whereas impaired ACh-dependent, and possibly AMPK-dependent, vasodilation reduces MA vasodilation in IUGR.

Limitations of our study include the possibilities that our mechanical removal of the endothelium was incomplete, that other endothelium-independent vasodilator factor(s) were activated by AMPK (e.g., the sarcoplasmic/endoplasmic Ca²⁺-ATPase and/or the large-conductance Ca²⁺-activated K⁺ channel) (31) and revealed after endothelium denudation, or that the smaller sample size for the denuded vessels limited our statistical power to detect differences. Future studies will aim to elucidate the endothelium-independent mechanisms of AMPK vasodilation in MAs by measuring AMPK-dependent changes in both intracellular Ca²⁺ and K⁺ currents in smooth muscle cells. One consideration regarding the interpretation of our data is the comparatively small number of IUGR subjects included for study. To avoid other confounding factors in our IUGR cases (e.g., no overlapping preeclampsia or other maternal comorbidities, no labor, term deliveries, etc.), we were extremely conservative in our patient selection. While we recognize the limitations this potentially introduces, we viewed this conservative sampling approach as a strength as having a slightly larger sample size derived from a more clinically diverse population may confound our findings. Even with the comparatively small number of IUGR cases, effect size for AMPK AUC for HA-AGA versus HA-IUGR shows a medium-to-large effect size of d = 0.75. In addition, although A769662 has been described as a selective activator of AMPK (9), one study reported AMPK-independent effects of A769662 in cultured mouse embryonic fibroblasts, incubated with A769662 for longer durations than the treatments employed here (26); we cannot exclude the possibility of nonspecific effects of A769662 in the present study.

In summary, our study has shown an important role for AMPK in the vasodilation of MAs in high-altitude AGA pregnancies and that the apparent reduction of vasodilator responses to AMPK may contribute to uteroplacental ischemia characteristic of IUGR. Further research is required to understand the mechanisms by which AMPK activation causes MA vasodilation and its association with fetal growth, such efforts could lead to novel therapies for treating hypoxia-related pregnancy complications such as IUGR since drugs that activate AMPK (e.g., metformin) are already approved for pregnancy use (19).

GRANTS

This study was supported by National Institute of Child Health and Human Development Grant R01-HD-088590 (to L.G.M. and C.G.J.) and Center for Women’s Health Research Junior Faculty Seed grant (to R.A.L.).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

R.A.L. and L.G.M. conceived and designed research; R.A.L., C.J.M., E.S.B., and J.A.H. performed experiments; R.A.L., C.J.M., D.J.O., A.G.E., C.G.J., and L.G.M. analyzed data; R.A.L., C.G.J., and L.G.M. interpreted results of experiments; R.A.L. prepared figures; R.A.L. drafted manuscript; R.A.L., C.G.J., and L.G.M. edited and revised manuscript; R.A.L., C.J.M., E.S.B., J.A.H., D.J.O., A.G.E., C.G.J., and L.G.M. approved final version of manuscript.