Prenatal hypoxia impairs cardiac mitochondrial and ventricular function in guinea pig offspring in a sex-related manner

Abstract

Adverse intrauterine conditions cause fetal growth restriction and increase the risk of adult cardiovascular disease. We hypothesize that intrauterine hypoxia impairs fetal heart function, is sustained after birth, and manifests as both cardiac and mitochondrial dysfunction in offspring guinea pigs (GPs). Pregnant GPs were exposed to 10.5% O₂ (HPX) at 50 days of gestation (full term = 65 days) or normoxia (NMX) for the duration of the pregnancy. Pups were allowed to deliver vaginally and raised in a NMX environment. At 90 days of age, mean arterial pressure (MAP) was measured in anesthetized GPs. NMX and prenatally HPX offspring underwent echocardiographic imaging for in vivo measurement of left ventricular cardiac morphology and function, and O₂ consumption rates and complex IV enzyme activity were measured from isolated cardiomyocytes and mitochondria, respectively. Prenatal HPX increased (P < 0.01) MAP (52.3 ± 1.3 and 58.4 ± 1.1 mmHg in NMX and HPX, respectively) and decreased (P < 0.05) stroke volume (439.8 ± 54.5 and 289.4 ± 15.8 μl in NMX and HPX, respectively), cardiac output (94.4 ± 11.2 and 67.3 ± 3.8 ml/min in NMX and HPX, respectively), ejection fraction, and fractional shortening in male, but not female, GPs. HPX had no effect on left ventricular wall thickness or end-diastolic volume in either sex. HPX reduced mitochondrial maximal respiration and respiratory reserve capacity and complex IV activity rates in hearts of male, but not female, GPs. Prenatal HPX is a programming stimulus that increases MAP and decreases cardiac and mitochondrial function in male offspring. Sex-related differences in the contractile and mitochondrial responses suggest that female GPs are protected from cardiovascular programming of prenatal HPX.

INTRODUCTION

Intrauterine hypoxia is one of the most significant complications during pregnancy, challenging normal fetal growth and development (6, 27, 33). The hypoxemic fetus adapts to its environment by redistributing its cardiac output (CO) to critical organs of need, such as the heart, brain, and adrenal glands (60, 62, 65, 66), resulting in asymmetric fetal growth restriction. Chronic exposure to intrauterine hypoxia may compromise fetal organ function, depending on the duration and severity of hypoxia and the gestational age at the time of exposure. Several studies have shown that chronic hypoxia induces asymmetric fetal growth restriction, accompanied by altered cardiovascular responses associated with aortic wall thickening, cardiac hypertrophy, and enhanced sympathetic innervation of peripheral arteries (for reviews see Refs. 32 and 37) and reduced cardiomyocyte endowment (13) in animal models. Long-term high-altitude hypoxia induces cardiac contractile dysfunction of fetal sheep associated with altered cardiac Ca²⁺ regulation and β-adrenergic receptor function (29). Furthermore, chronic maternal hypoxia decreases gene expression of PKCε and heat shock protein 70 and increases membrane type 1 matrix metalloproteinase in fetal rat hearts (58, 59, 79) and increases expression of inducible nitric oxide synthase, metalloproteinases, and proinflammatory cytokines in fetal guinea pig hearts (23, 26, 56, 78). Thus, increased gene expression attributed to oxidative/nitrative stress and cardiac inflammation, as well as cardiac remodeling and Ca²⁺ dysregulation, is implicated in several animal models as underlying causes of contractile dysfunction in the hypoxic fetal heart.

Programming effects of prenatal hypoxia on cardiac function in the offspring, addressing the developmental origins of health and disease (DOHaD) hypothesis that intrauterine stress impacts fetal growth and organ function and has lasting functional consequences in the offspring (11, 34), have also been reported (32, 55). Several studies support programming effects of prenatal hypoxia in offspring mammalian hearts, demonstrating increased risk for ischemia-reperfusion injury (44, 45, 84, 85), reduced cardiac contractile function (31, 68), reduced cardiac metabolism (69), and increased ventricular arrhythmias (54).

We recently focused on the effects of prenatal hypoxia on mitochondrial function of fetal and offspring hearts. We reported that prenatal hypoxia decreases cytochrome c oxidase activity in fetal (4) and offspring (5) guinea pig heart ventricles as evidence of prenatal hypoxia as a stimulus that impairs cardiac mitochondrial function. Attenuation of mitochondrial function is well documented to increase risk of heart dysfunction in the adult (16, 48). We propose that prenatal hypoxia inhibits mitochondrial respiratory activity and contributes to reduced ventricular performance postnatally. To test this idea, we measured contractile function using noninvasive echocardiography and indexes of mitochondrial respiration of heart cells of offspring exposed to prenatal hypoxia. The results of this study link the effects of intrauterine hypoxia in the fetal heart to mitochondrial and cardiac programming in the offspring.

METHODS

Animal model.

All animal procedures were approved by the University of Maryland Institutional Animal Care and Use Committee in accordance with Association for Assessment and Accreditation of Laboratory Animal Care-accredited procedures (Animal Welfare Assurance No. A3200-01). The pregnant guinea pig is an excellent animal model of human pregnancy for study of the programming of intrauterine stress. The similarities of the pregnant guinea pig to human pregnancy include placental morphology, such as a hemomonochorial placenta with deep invasion, a luteoplacental shift and elevated progesterone levels at full term, fetal growth characteristics of similar fat content and epicardial depots, and timing of myogenesis, as well as prenatal maturation of fetal cardiovascular and nervous systems (22, 25, 35, 51, 52). This provides an excellent model for the study of cardiovascular programming of the offspring to intrauterine hypoxia because of the relative maturity of the full-term fetus and its adaptive responses to intrauterine stress.

Virgin female guinea pigs were mated in groups of three with a single male following visual evidence of vaginal membrane opening. Time-mated pregnant guinea pigs were randomly assigned to normoxia (NMX, room air) or hypoxia (HPX, 10.5% O₂) treatment and exposed to NMX during the entire gestation or to HPX during the last 14 days of pregnancy, a period of rapid fetal growth and increased O₂ utilization by the fetus. Chronic hypoxia does not reduce either maternal food or water intake and, therefore, represents a hypoxic model without nutrient restriction (80). Pregnant sows were allowed to deliver vaginally, and pups born into a hypoxic environment were removed to normal room air within 1 day of delivery. Newborn pups were housed in room air with their mothers until weaning (30 days) and then placed in individual cages. To generate offspring for the four experimental series (i.e., blood pressure measurement, echocardiography, cell isolation for extracellular flux analysis, and mitochondrial isolation for enzymatic assay), 89 sows were used to generate 135 offspring. A single male and a single female were selected from the same litter, if both were present, for comparison of sex differences. Litter size was selected based on two to four offspring per litter to avoid the influence of litter numbers on fetal body weight. Offspring body weight and food and water intake rates were monitored at 33 days of age and in 3-day intervals until 90 days of age, a time of established reproductive maturity and a steady growth profile. All measurements of blood pressure, left ventricular (LV) morphology and function, and indexes of mitochondrial function were acquired from 90-day-old guinea pig offspring. Food and water intake rates were obtained from animals being measured for arterial blood pressure.

Measurement of arterial blood pressure.

For measurement of arterial blood pressure, NMX or HPX male (n = 15) and female (n = 16) guinea pig offspring were anesthetized with ketamine (80 mg/kg sc)-xylazine (10 mg/kg sc). After lidocaine injection and a skin incision, the right brachial artery was cannulated with polyethylene tubing filled with heparinized saline. Systolic, diastolic, and mean arterial blood pressures were recorded (Power Laboratory 800, ADInstruments, Colorado Springs, CO) over a 5-min period and analyzed with Chart v. 4.2 software (ADInstruments).

Echocardiography.

LV morphology and function of NMX or HPX male (n = 18) and female (n = 18) offspring were measured by echocardiography. Animals were anesthetized with 1–5% isoflurane in balance with O₂ and studied using a Vevo 2100 imaging system (FUJIFILM-VisualSonics, Toronto, ON, Canada) equipped with a 20-MHz scan head. LV dimension and wall thickness at the level of the papillary muscle were measured using two-dimensional guided M-mode imaging via parasternal short-axis view, which was also used for calculation of LV fractional shortening [FS = (LVEDD − LVESD)/LVEDD × 100, where LVEDD is LV end-diastolic diameter and LVESD is LV end-systolic diameter], ejection fraction [EF = (EDV − SV)/EDV × 100, where EDV is end-diastolic volume and SV is stroke volume], and LV mass. Two-dimensional guided M-mode imaging of parasternal long- and short-axis views was performed for measurements of aortic diameter and LV chamber size/wall thickness, respectively, and an apical four-chamber view (a “long”-axis view in nature) was used for measurement of the E/A ratio [an index of diastolic function assessed by the velocity waveform during passive (E wave peak velocity) and active (A wave peak velocity during atrial contraction) filling of the LV during diastole] from the mitral inflow (42, 53). Under guidance of color Doppler imaging, aortic flow and mitral inflow were measured using pulse-wave Doppler imaging via suprasternal notch view and apical four-chamber view, respectively. Data were calculated according to the generally accepted formulas, as previously reported (17, 18).

Isolation of mitochondria.

Tissue was ground with a mortar and pestle in liquid N₂ to isolate mitochondria from frozen heart LV of NMX or HPX male and female offspring. Frozen tissue was homogenized in 1× homogenization buffer (0.25 M sucrose, 5 mM HEPES, and 1 mM EDTA, pH 7.2) with zirconium oxide beads in a VWR multitube vortexer at high speed for 10 min at 4°C. Samples were centrifuged twice at 600 g for 10 min at 4°C to remove cellular debris. The supernatant was carefully transferred to a clean 1.5-ml tube, and the mitochondria were pelleted by centrifugation at 12,500 g for 10 min at 4°C. The mitochondrial pellet was resuspended in 100 μl of homogenization buffer with 0.1 mM N-dodecyl β-d-maltoside (Sigma, St. Louis, MO). Mitochondrial protein of each sample was determined by the Bio-Rad protein assay. Purity of the mitochondrial-enriched fraction of the heart tissue is ∼100% (98.2–99.6%), similar to that reported for placenta mitochondria (75). This percent contamination is based on low contamination from nuclear (1.8%) or cytosolic (0.4%) proteins by targeting specific nuclear (lamin A/C), cytoplasmic (α-tubulin), and mitochondrial [voltage-dependent anion channel (VDAC)] proteins with primary antibodies using Western blot analysis. Percent contamination was quantified by normalization of band density to VDAC in the mitochondrial fraction.

Mitochondrial complex activity assay.

Respiratory complex IV contains cytochrome c oxidase, which reduces O₂ to H₂O in the respiratory chain. Enzyme activity was measured in isolated, enriched cardiac mitochondrial fractions of frozen LV of NMX or HPX male (n = 16) and female (n = 16) offspring. Complex IV activity is measured colorimetrically by monitoring the oxidation rate of cytochrome c. Briefly, 3 μg of mitochondrial protein were added to each well of a 96-well plate containing the assay buffer [10 mM Tris·HCl (pH 7.0) and 120 mM KCl + reduced cytochrome c] (76). Optical density was measured at 550 nm in a 96-well plate reader (BioTek, Winooski, VT) for 30 min at 10-s intervals. Our assay was optimized using a range of protein amounts (2–5 μg) that generated a minimum and a maximum rate of oxidation; a 3-μg amount was selected based on the kinetics profile for oxidation that fell between these curves and was used as the standardized amount of total mitochondrial protein for determination of the slope in the linear range.

The rate of the reaction was calculated in the linear range according to the following formula: cytochrome c oxidase activity (units/mg mitochondrial protein) = (ΔOD/Δt) ÷ [ε * protein (mg)], where OD is optical density, t is time, and ε is the extinction coefficient (7.04 mM⁻¹·cm⁻¹).

Isolation of cardiomyocytes of male and female offspring hearts.

After thoracotomy, hearts were excised from anesthetized and heparinized NMX or HPX male (n = 19) and female (n = 17) guinea pigs for isolation of cardiac cells. Hearts were immediately placed in iced physiological buffer solution (PBS) and mounted via the aorta onto a glass cannula of a modified Langendorff heart perfusion apparatus. Guinea pig cardiomyocytes were isolated using a modified procedure reported for isolation of fetal sheep cardiac cells (38). Hearts were retrograde-perfused at 37°C with a low-Ca²⁺ (no Ca²⁺ added) Tyrode solution (in mM: 140 NaCl, 5 KCl, 10 HEPES, 10 glucose, and 1 MgCl₂, pH 7.35) without enzymes for 5 min and then with Tyrode solution containing enzymes [collagenase (80 U/ml), protease (0.59 U/ml), and albumin (1 mg/ml)] for 12 min. Hearts were then perfused with Kraft-Brühe (KB) buffer (in mM: 30 KCl, 10 HEPES, 10 glucose, 74 potassium glutamate, 20 taurine, 1.5 MgSO₄·7H₂O, 0.5 EGTA, and 30 KH₂PO₄, pH 7.37) for 5 min to wash out the enzymes. Hearts were removed from the apparatus and placed in a beaker containing warmed KB buffer for gentle mincing to release cells from the heart. Cells were filtered through 150-μm nylon mesh and flushed into a beaker containing 37°C KB buffer. Cells were pelleted twice by centrifugation (200–250 g) and resuspended in KB buffer, and the supernatant was discarded. The final cell pellet was resuspended in DMEM containing 1× penicillin (100 U)-streptomycin (100 μg) (PenStrep, catalog no. 15140-122, Life Technologies) and then transferred to a 50-ml tube containing ~10–15 ml of DMEM with d-glucose (1 g/l). After isolation, cells were counted on an Invitrogen Countess automated cell counter, and percent cell viability (~95%) was determined using Trypan blue. An aliquot volume was calculated for transfer of 20,000 cells/100 μl to laminin-coated 96-well microplates and incubated for 2–4 h at 37°C and 5% CO₂ in a humidified incubator.

Mitochondrial respiration studies of isolated cardiomyocytes.

Mitochondrial respiration of isolated heart cells was measured using an extracellular flux analyzer (Seahorse XF24, Agilent Technologies, Santa Clara, CA), as previously described (64) and modified for guinea pig cardiac cells. In a separate series of experiments, conditions were optimized (20,000 cells/well) for cardiac cell density by seed density experiments (15,000–25,000 cells/well) and drug concentrations of oligomycin (1 µg/ml, an ATP synthase inhibitor), carbonyl cyanide-4-(trifluoromethoxy) phenylhydrazone [FCCP, 1 μM, a protonophore that uncouples respiration, inducing maximal O₂ consumption rate (OCR)], and antimycin A (10 μM, an inhibitor of complex III for determination of cellular nonmitochondrial O₂ consumption). OCRs of cardiac cells obtained from a single guinea pig heart were measured at baseline and following separate injections of oligomycin, FCCP, and antimycin A. Maximal OCR is the difference between OCR measured after FCCP and nonmitochondrial OCR measured following antimycin A. Reserve capacity is the difference between maximal OCR and baseline OCR. Stock concentrations of oligomycin (catalog no. AB674, Sigma), FCCP (catalog no. C2920, Sigma), and antimycin A (catalog no. 75351, Sigma) were prepared in ethanol.

Statistics.

Values are means ± SE. Body weight and food and water intake rates were compared between NMX and prenatally HPX groups by three-way repeated-measures analysis of variance, with days, sex, and treatment as independent variables. Post hoc tests were not performed if there were no significant differences between the groups. Mitochondrial indexes (e.g., OCR and complex IV activities) and cardiac function were compared between NMX and prenatally HPX groups by two-way analysis of variance, with treatment and sex as variables. P < 0.05 was considered significantly different. Statistical analyses were carried out using Systat software (San Jose, CA).

RESULTS

Body weight and food and water intake rates.

Figure 1 illustrates the temporal profile of body weight and food and water intake rates of male (n = 8 NMX and 8 HPX) and female (n = 7 NMX and 8 HPX) guinea pig offspring. Birth weights of neonates exposed to prenatal hypoxia were significantly (P < 0.05) decreased at 1 day of age [male: 91.3 ± 5.5 g (NMX) and 80.7 ± 3.5 g (HPX); female: 97.9 ± 5.2 g (NMX) and 84.9 ± 2.4 g (HPX)] but were not different at 90 days of age (Fig. 1). Prenatal hypoxia did not affect food or water intake rates over the 90-day period. Furthermore, there were no differences in heart, liver, or kidney weight in HPX offspring compared with their NMX controls, regardless of sex. Exposure to prenatal hypoxia reduced (P < 0.05) birth weights in male and female offspring.

Fig. 1.

Effect of prenatal hypoxia on body weight (BW) and food and water intake rates of guinea pig offspring. Guinea pig offspring were exposed to prenatal normoxia (NMX) or hypoxia (HPX, 10.5% O₂ for 14 days) and raised in a normoxic environment until 90 days of age. Values are means ± SE; n = 8 NMX and 8 HPX males and 7 NMX and 8 HPX females.

Mean arterial blood pressure.

Arterial blood pressures were measured in male (n = 8 NMX and 7 HPX) and female (n = 8 NMX and 8 HPX) NMX and HPX guinea pig offspring. There were no differences in arterial blood pressure between males and females born to NMX mothers. In contrast, prenatal HPX significantly increased (P < 0.01) arterial blood pressure by 12% in male (50.7 ± 1.7 and 57.4 ± 1.6 mmHg in NMX and HPX, respectively), but not female (50.5 ± 2.5 and 52.0 ± 2.0 mmHg in NMX and HPX, respectively), offspring, with no effect on heart rate (male: 161 ± 4 and 171 ± 6 beats/min in NMX and HPX, respectively; female: 169 ± 5 and 163 ± 6 beats/min in NMX and HPX, respectively). The HPX-induced increase in mean arterial blood pressure in males was accompanied by a significant increase (P < 0.01) in systolic (60.3 ± 2.1 vs. 68.3 ± 2.3 mmHg in NMX and HPX, respectively) and diastolic (41.4 ± 1.3 and 46.9 ± 1.1 mmHg in NMX and HPX, respectively) blood pressures, with no difference in pulse pressure (18.8 ± 1.0 and 21.4 ± 1.5 mmHg in NMX and HPX, respectively). In females, systolic pressure (59.7 ± 2.9 and 61.8 ± 2.4 mmHg in NMX and HPX, respectively), diastolic pressure (41.2 ± 2.1 and 42.4 ± 1.7 mmHg in NMX and HPX, respectively), pulse pressure (18.5 ± 1.1 and 19.4 ± 0.9 mmHg in NMX and HPX, respectively), and heart rate (169.1 ± 5.4 vs 163.2 ± 6.1 beats/min in NMX and HPX, respectively) were similar between NMX and HPX offspring.

Cardiac morphology and function.

LV morphology and function were measured by echocardiography from 90-day-old male (n = 9 NMX and 9 HPX) and female (n = 9 NMX and 9 HPX) offspring exposed to NMX and prenatal HPX (Fig. 2). No significant differences between NMX and HPX groups were found in males or females for LV dimension, LV wall thickness, or interventricular septal wall thickness at systole or diastole. In addition, there were no differences between NMX and HPX offspring in total heart mass (male: 1.94 ± 0.11 and 1.90 ± 0.05 g in NMX and HPX, respectively; female: 1.94 ± 0.11 and 1.90 ± 0.05 g in NMX and HPX, respectively), LV weight (male: 1.50 ± 0.10 and 1.41 ± 0.04 g in NMX and HPX, respectively; female: 1.50 ± 0.10 and 1.41 ± 0.04 g in NMX and HPX, respectively), and LV weight-to-total weight ratio (male: 0.76 ± 0.01 and 0.74 ± 0.01 in NMX and HPX, respectively, female: 0.75 ± 0.01 and 0.75 ± 0.01 in NMX and HPX, respectively). Thus there were no signs of LV hypertrophy in this model, despite the increase in mean arterial pressure.

Fig. 2.

Effect of prenatal hypoxia on morphological indexes of guinea pig offspring. Guinea pig offspring were exposed to prenatal normoxia (NMX) or hypoxia (HPX, 10.5% O₂ for 14 days) and raised in a normoxic environment until 90 days of age. Morphological indexes of left ventricular (LV) diameter, LV wall thickness, and interventricular septal (IVS) thickness were measured using M-mode imaging (Vevo 2100) Doppler ultrasound biomicroscopy during systole and diastole of the cardiac cycle. Values are means ± SE; n = 9 NMX and 9 HPX males and 9 NMX and 9 HPX females.

Male HPX offspring demonstrated compromised LV global performance (SV and CO) and systolic function (EF and FS) (Fig. 3). Prenatal HPX significantly (P < 0.05) decreased SV, CO, and cardiac index (CO/body weight) but had no effect on heart rate (250 ± 14 and 237 ± 11 beats/min in NMX and HPX, respectively) in male offspring. In females, there was no effect of HPX on heart rate (222 ± 2 and 222 ± 6 beats/min in NMX and HPX, respectively), SV, CO, or cardiac index compared with NMX controls. Furthermore, prenatal HPX decreased LV EF and FS in male, not female, hearts. Prenatal HPX had no effect on the E/A ratio of the 90-day-old male offspring (1.65 ± 0.09 and 1.72 ± 0.07 in NMX and HPX, respectively) but decreased (P < 0.05) the E/A ratio in female offspring (1.88 ± 0.14 and 1.43 ± 0.08 in NMX and HPX, respectively), which was attributed to a decrease in the maximal peak velocity of the E wave.

Fig. 3.

Effect of prenatal hypoxia on functional indexes of guinea pig offspring. Guinea pig offspring were exposed to prenatal normoxia (NMX) or hypoxia (HPX, 10.5% O₂ for 14 days) and raised in a normoxic environment until 90 days of age. Functional indexes of stroke volume, cardiac output, ejection fraction, and fractional shortening of left ventricular diameter (LVD) were measured using pulse-wave Doppler ultrasound. Values are means ± SE; n = 9 NMX and 9 HPX males and 9 NMX and 9 HPX females. *P < 0.05 vs. NMX.

Mitochondrial respiratory function and complex IV activity.

OCRs were measured in isolated cardiomyocytes derived from single hearts of male (n = 9 NMX and 10 HPX) and female (n = 9 NMX and 8 HPX) offspring exposed to NMX or prenatal HPX. OCRs of baseline, in association with ATP synthesis, electron leak, maximal rate, and reserve capacity, were measured as absolute OCR and OCR normalized to baseline (Fig. 4). Baseline OCR and leak OCR were similar between cardiomyocytes of NMX and HPX hearts, regardless of sex. There was a significant increase (P < 0.05) in OCR associated with ATP synthesis in male hearts exposed to prenatal HPX but no difference when OCR was normalized to baseline (Fig. 5). In contrast, HPX significantly reduced (P < 0.05) both maximal OCR and reserve capacity (percent baseline) in males, but not females, compared with their respective NMX controls. Exposure to prenatal HPX reduced complex IV activity in hearts of males (n = 8 NMX and 8 HPX), but not females (n = 8 NMX and 8 HPX), compared with NMX controls (Fig. 6).

Fig. 4.

Representative traces of O₂ consumption rates (OCR) measured in freshly isolated guinea pig cardiomyocytes. Cells were obtained at 90 days of age from single hearts of male or female offspring exposed to normoxia (NMX) or prenatal hypoxia (HPX). Baseline OCR was measured at 3 consecutive time points followed by addition of the ATPase inhibitor oligomycin (Oligo, 1 μg/ml). Maximal OCR was obtained in the presence of 1 μM carbonyl cyanide-4-(trifluoromethoxy) phenylhydrazone (FCCP). Antimycin A (AA, 10 μM) was added to inhibit mitochondrial OCR, allowing subtraction of nonmitochondrial OCR. Values are means (SD) of 3 identically treated wells.

Fig. 5.

Effect of prenatal hypoxia on O₂ consumption rates (OCR) of isolated cardiac cells from guinea pig offspring. Hearts were obtained from 90-day-old guinea pig offspring exposed to prenatal normoxia (open bars) or hypoxia (hatched bars, 10.5% O₂ for 14 days). Mitochondrial respiration (OCR) was measured (pmol·min⁻¹·20,000 cells⁻¹ or percentage of baseline OCR) at baseline (Base), following oligomycin-induced inhibition of ATP synthesis (ATP), and following carbonyl cyanide-4-(trifluoromethoxy) phenylhydrazone (FCCP)-induced maximal OCR (Max), and respiratory reserve capacity (ResCap) was calculated. Values are means ± SE; n = 9 NMX and 10 HPX males and 9 NMX and 8 HPX females. *P < 0.05 vs. NMX.

Fig. 6.

Effect of prenatal hypoxia on complex IV activity of heart ventricles of guinea pig offspring. Hearts were obtained from 90-day-old guinea pig offspring exposed to prenatal normoxia (open bars) or hypoxia (hatched bars, 10.5% O₂ for 14 days). Values are means ± SE; n = 8 NMX and 8 HPX males and 8 NMX and 8 HPX females. *P < 0.05 vs. NMX.

DISCUSSION

This study showed that exposure to prenatal HPX increases mean arterial blood pressure and decreases cardiac contractile and mitochondrial function in male compared with female offspring. Prenatal HPX caused fetal growth restriction in both sexes, although differences in body weight were absent at 90 days of age. There was no effect of prenatal HPX on food or water intake rates between treatment groups to account for any nutrient intake differences after birth.

Prenatal HPX increased mean arterial blood pressure in a sex-related manner in this guinea pig model of fetal growth restriction. Previous studies showed that prenatal HPX or placental insufficiency increased arterial blood pressure in the offspring rat (1, 3, 14, 20, 43), mouse (83), and guinea pig (60), while other studies showed no effect of gestational hypoxia on blood pressure in adult rats (49, 61, 70) or chickens (71). Fetal programming of hypertension varies depending on the etiology of fetal growth restriction, with possible causes including maternal nutrient and protein restriction (24, 77), reduced uterine blood flow (2, 20), and gestational hypoxia (33). Depending on the animal model, systemic hypertension may be mediated by reduced nephrogenesis, increased sympathetic hyperinnervation of blood vessels, endothelial dysfunction, altered regulation of the renin-angiotensin system, and/or increased angiotensin II responsiveness (2, 20, 24, 32, 67). Since CO was reduced with HPX, we attribute the hypertension in the offspring to increased systemic vascular resistance, likely mediated by programming of vascular mechanisms in utero (2, 20, 32, 67).

Blood pressure values in the guinea pig are similar to those reported in other studies for guinea pigs. For example, mean arterial blood pressures of ketamine-anesthetized nonpregnant guinea pigs were 54 mmHg (73) and 47.1 ± 6.8 mmHg (15) compared with 50.7 ± 1.7 mmHg in male guinea pigs in the current study. In awake guinea pigs, blood pressures varied from 53 to 70 mmHg [59 ± 4 mmHg (72), 43 mmHg (63), 70 ± 2 mmHg (73), 53.1 ± 4.2 mmHg (15), 64 ± 1.38 mmHg (28), and 59.7 ± 7.2 mmHg (81)]; yet, despite a well-known cardiodepressive effect of anesthesia, our values fell within this range. However, heart rates of our ketamine-exposed animals were significantly reduced (161–171 beats/min) compared with those in awake guinea pigs (205–267 beats/min) (15, 28, 63, 72, 73, 81), although heart rates measured during echocardiography in urethane-exposed animals were similar (222–250 beats/min).

Previous studies using ex vivo techniques to assess cardiac function in offspring exposed to prenatal HPX report a decrease in recovery following ischemia-reperfusion injury (32, 58), cardiac β-adrenergic receptor signaling (46), and diastolic function (7). In addition, gestational hypoxia has variable effects on ventricular wall mass (32), causing increased hypertrophic growth or no change, depending on the animal model. In the present study, prenatal HPX had no effect on cardiac morphology (i.e., relative heart weight, cardiac wall thickness, and chamber diameter) in either sex, despite inhibition of functional indexes of SV, CO, EF, and FS in male guinea pigs only. Furthermore, there was no LV hypertrophy or increased heart mass, despite a significant increase in mean arterial blood pressure in male offspring exposed to prenatal HPX. This may be related to an increased afterload that is below threshold for inducing cardiac hypertrophy.

The decreased CO may be mediated by several factors, including decreased preload and contractility, increased afterload. While preload was not measured, E/A ratios provide an index of diastolic function of passive and active filling of the ventricle. Prenatal HPX had no effect on the E/A ratio in male offspring, whereas the E/A ratio was decreased in females. This was attributed to a decrease in E wave velocity and a decreased passive filling during diastole. Yet, no other functional indexes in female hearts were altered by HPX, indicating an unlikely significant effect on cardiac performance. The decreases in EF and FS in males are consistent with decreased contractility. In fetal sheep studies of high-altitude hypoxia, reduced CO was attributed to reduced Ca²⁺ release mechanisms associated with sarcoplasmic reticulum (29) and reduced myofibrillar Mg²⁺-activated ATPase (30). In the current study, a role for altered Ca²⁺ handling by cardiomyocytes (21, 48) cannot be ruled out. Rather, we propose mitochondrial dysfunction as an additional mechanism inhibiting LV contractility in prenatal HPX-exposed male offspring by altering respiratory efficiency and ATP availability.

Mitochondria account for ~95% of the O₂ consumed in cardiac cells, which rely predominantly on oxidative phosphorylation for generation of the ATP supply (47). In fetal and adult hearts, HPX reduces respiratory enzyme activities (4, 5, 36), which reduces mitochondrial function and may contribute to cardiomyopathies and cardiac failure (16, 48, 49). Prenatal HPX had no effect on baseline OCR or normalized OCR associated with ATP synthesis in isolated heart cells of male or female offspring, although H⁺ leak was slightly decreased in males. Yet, maximal OCR and the respiratory reserve capacity were reduced by prenatal HPX in cardiomyocytes of male, but not female, hearts. The lack of effect of prenatal HPX on baseline OCR may be attributed to plated cardiomyocytes at rest. However, the HPX-mediated decreases in maximal OCR and respiratory reserve capacity measured ex vivo suggest a limited capacity of cardiomyocytes to increase their energy supply in response to energy demand. The HPX-induced reduction of complex IV activity may contribute to reduced electron flux along the respiratory chain and impair mitochondrial efficiency. Thus the mitochondrial deficiencies may confer an underlying risk to secondary cardiac challenges, such as systemic hypertension, exercise, or diet, when a higher level of respiration is required.

Previous studies provide insight into several possible mechanisms by which intrauterine hypoxia may mediate decreased mitochondrial and contractile function in the heart of offspring: 1) epigenetic programming of hypermethylation of specific promoters associated with contractile protein function (58, 59, 85), 2) altered cardiac cell remodeling of the fetal heart in response to intrauterine stress (12, 13, 50), 3) transcriptional regulation of mitochondrial biogenesis and cardiac cell proteins (5, 13), and 4) indirect effects of systemic hypertension and afterload on heart remodeling in the offspring (32). First, PKCε plays an important role in cardioprotection through its interaction with both contractile proteins and targeting of mitochondria (74). Previous studies identified hypermethylation of the PKCε promoter and reduced PKCε mRNA/protein levels in hearts from prenatal HPX-exposed male rat offspring (58, 59, 85), reducing the cardioprotective effect of contractile function against ischemia-reperfusion injury. Additionally, PKCε plays an important role in cardioprotection at the mitochondria via translocation of PKCε to the inner mitochondrial membrane and phosphorylation of targeted proteins such as cytochrome c oxidase, VDAC, adenine nucleotide transporter, ATP synthase, and the mitochondrial ATP-sensitive K⁺ channel (9, 74). The interaction of PKCε with mitochondrial proteins has been demonstrated to be integral to metabolic function associated with glycolysis, the tricarboxylic acid cycle, β-oxidation, and ion transport signaling pathways (74). Thus, reduced PKCε expression and/or translocation to the mitochondria by intrauterine hypoxia may inhibit cardiac function via contractile and metabolic pathways. Second, decreased gene expression of mitochondrial respiratory proteins, such as cytochrome c oxidase (4) and metabolic enzymes associated with fatty acid oxidation and transport (13), in hearts of prenatal HPX-exposed guinea pig offspring has been reported, indicating a role for gene repression by epigenetic mechanisms. Third, chronic hypoxia reduces the number of cardiomyocytes in both fetal (12) and offspring guinea pig hearts in a sex-related manner that favors females (13), suggesting a decrease in cardiac cell endowment that is sustained into adulthood and may limit the heart’s functional capacity against secondary challenges such as increased afterload (32). These studies indicate that intrauterine hypoxia may impart lasting perturbations in heart function of the offspring by altering mechanisms associated with contractile and metabolic function.

The sex-related differences in contractile and mitochondrial function uncovered by the current study may indicate a functional advantage in female compared with male hearts. An extensive literature has identified cardioprotection in females over males attributed to differences in estrogen levels (40, 41, 57), longevity in aging (41), and increased cardiac tolerance to ischemic injury (57). Sex-related differences in mitochondrial function favoring females are attributed to an enhanced mitochondrial antioxidant capacity (10), contributing to reduced reactive oxygen species and increased levels of estrogen, which regulate gene expression of mitochondrial proteins via activation of estrogen receptor-α/β (41, 82) in reproductively mature offspring. In addition, mitochondrial morphology of 6-wk-old male mice exhibited increased fragmented, circular and smaller mitochondria compared with age-matched female mitochondria (40), indicating a reduced biogenesis and an altered dynamic process favoring fission over fusion in males (40). Other sex-related differences include lower mitochondrial Ca²⁺ uptake (8), mitochondrial content (19), and efficiency, as measured by H₂O₂ generation (19), in females than males. There are also sex differences in mechanisms associated with Ca²⁺-handling and -release mechanisms (i.e., L-type Ca²⁺ channels, Na⁺/Ca²⁺ exchangers, and ryanodine Ca²⁺-release channels) of the cardiomyocyte favoring females, with a higher Ca²⁺ efflux from cells (reviewed in Ref. 57). Thus, while there are advantages in females over males in terms of cardioprotective mechanisms, our understanding of sex-related differences in the adaptive responses to programming requires further evaluation. For example, our study does not attribute sex differences to mitochondrial or contractile function under control conditions but exhibits a sex difference in response to hypoxia-induced programming.

In conclusion, this is the first study to identify a sexual dimorphism linking decreased cardiac contractility and mitochondrial deficits of respiratory enzyme activity and cellular respiration in hearts of guinea pig offspring. Investigation of the specific sex differences in mitochondrial mechanisms associated with biogenesis, electron transport, or dynamics is important in identifying the programming response of heart function to intrauterine hypoxia. We propose that prenatal HPX may increase the vulnerability to cardiac and mitochondrial dysfunction later in life in a sex-dependent manner, increasing the risk of heart disease and/or failure. Given the sex differences in heart failure, with the incidence lower in women than men (20), further study is needed to understand the underlying contribution of mitochondrial and contractile programming by prenatal HPX to heart dysfunction.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

L.P.T., L.C., and B.M.P. conceived and designed the research; L.P.T., L.C., B.M.P., G.P., and H.S. analyzed data; L.P.T., L.C., B.M.P., and H.S. interpreted results of experiments; L.P.T., B.M.P., and H.S. prepared figures; L.P.T. drafted manuscript; L.P.T., L.C., B.M.P., G.P., and H.S. edited and revised manuscript; L.P.T., L.C., B.M.P., G.P., and H.S. approved final version of manuscript; L.C., B.M.P., G.P., and H.S. performed experiments.