Effect of resveratrol on metabolic and cardiovascular function in male and female adult offspring exposed to prenatal hypoxia and a high‐fat diet

Key points

• Prenatal hypoxia, a common outcome of many pregnancy complications, predisposes offspring to chronic diseases in later life.

• We investigated the effect of prenatal hypoxia, on a background of a high‐fat diet, on metabolic and cardiac function in adult male and female rat offspring. We also examined the therapeutic role of resveratrol supplementation in preventing metabolic and cardiac dysfunction.

• Prenatal hypoxia impaired both metabolic and cardiac function in male and only cardiac function in female rat offspring. We also observed that male rat offspring were more susceptible to metabolic and cardiac dysfunction as compared with their female counterparts; this provides evidence of sexual dichotomy in the fetal programming of diseases due to prenatal hypoxia.

• Resveratrol supplementation in the diet improved metabolic and cardiac function independent of sex; this provides evidence of a possible therapeutic role of resveratrol in susceptible male and female rat offspring exposed to prenatal hypoxia.

Abstract

Prenatal hypoxia, a common outcome of pregnancy complications, predisposes offspring to the development of metabolic and cardiovascular disorders in later life. We have previously observed that resveratrol improved cardiovascular and metabolic health in adult male rat offspring exposed to prenatal hypoxia and a postnatal high‐fat (HF) diet; however, the effects of resveratrol in female rat offspring are not known. Our aim was to identify the mechanism(s) by which resveratrol may prevent metabolic and cardiac dysfunction in both male and female rat offspring exposed to prenatal hypoxia and a postnatal HF diet. Offspring that experienced normoxia or hypoxia in utero were fed a HF diet or a HF diet supplemented with resveratrol for 9 weeks following weaning. Body composition, metabolic function, in vivo cardiac function and ex vivo cardiac susceptibility to ischaemia–reperfusion (I/R) injury were assessed at 12 weeks of age. Prenatal hypoxia impaired metabolic function in male, but not female, rat offspring fed a HF diet and this was improved by resveratrol supplementation. Prenatal hypoxia also led to reduced recovery from cardiac I/R injury in male, and to a lesser extent in female, rat offspring fed a HF diet. Indices of cardiac oxidative stress after I/R were enhanced in both male and female rat offspring exposed to prenatal hypoxia. Resveratrol improved cardiac recovery from I/R injury and attenuated superoxide levels in both male and female rat offspring. In conclusion, prenatal hypoxia impaired metabolic and cardiac function in a sex‐specific manner. Resveratrol supplementation may improve metabolic and cardiovascular health in adult male and female rat offspring exposed to prenatal hypoxia.

Key points

• Prenatal hypoxia, a common outcome of many pregnancy complications, predisposes offspring to chronic diseases in later life.

• We investigated the effect of prenatal hypoxia, on a background of a high‐fat diet, on metabolic and cardiac function in adult male and female rat offspring. We also examined the therapeutic role of resveratrol supplementation in preventing metabolic and cardiac dysfunction.

• Prenatal hypoxia impaired both metabolic and cardiac function in male and only cardiac function in female rat offspring. We also observed that male rat offspring were more susceptible to metabolic and cardiac dysfunction as compared with their female counterparts; this provides evidence of sexual dichotomy in the fetal programming of diseases due to prenatal hypoxia.

• Resveratrol supplementation in the diet improved metabolic and cardiac function independent of sex; this provides evidence of a possible therapeutic role of resveratrol in susceptible male and female rat offspring exposed to prenatal hypoxia.

Abbreviations

ACC

acetyl CoA carboxylase

AMPK

AMP‐activated protein kinase

CVD

cardiovascular disease

DHE

dihydroethidium

HF

high‐fat

I/R

ischaemia–reperfusion

IUGR

intrauterine growth restriction

IVS

interventricular septal

LV

left ventricle

LVID

left ventricular internal diameter

LVPW

left ventricular posterior wall

EDV

end diastolic volume

ESV

end systolic volume

IVRT

isovolumic relaxation time

Introduction

Cardiovascular and metabolic diseases are leading causes of mortality and morbidity worldwide. Epidemiological and experimental studies have shown that populations who are exposed to prenatal hypoxia are more susceptible to developing metabolic and cardiovascular diseases (CVDs) in later life (Giussani & Davidge, 2013; Demicheva & Crispi, 2014), thereby linking prenatal hypoxia and postnatal metabolic and cardiovascular diseases. Many pregnancy complications result in fetal hypoxia, and offspring born from complicated pregnancies have been shown to have altered cardiac morphology and cardiac dysfunction in prenatal (Hecher et al. 1995; Crispi et al. 2008; Comas et al. 2010), neonatal (Fouzas et al. 2014) and postnatal life (Barker, 1993; Bjarnegard et al. 2013), and abnormal metabolic health in later life (Barker et al. 1993; Jaquet et al. 2000). One of the most common CVDs is acute myocardial infarction, treatment of which can also cause reperfusion‐induced myocardial injury, known as ischaemia–reperfusion (I/R) injury. We and others have previously reported that offspring exposed to prenatal hypoxia exhibit decreased cardiac function recovery after I/R injury (Rueda‐Clausen et al. 2009, 2011 b; Xue & Zhang, 2009) Further, our lab has also shown that male offspring exposed to prenatal hypoxia were also more susceptible to a secondary insult such as a high‐fat (HF) diet leading to metabolic syndrome at an early stage of life (Dolinsky et al. 2011; Rueda‐Clausen et al. 2012 a). Thus, while fetal programming by prenatal hypoxia predisposes offspring to cardiovascular diseases, additional metabolic dysfunction might be manifested in these susceptible offspring by exposure to an early postnatal stress, a so‐called second hit, which may include nutritional factors.

Several studies have shown that females are cardioprotected, an effect which might be attributed to the female sex hormone, oestrogen (Deschamps et al. 2010). In the isolated heart, it has been shown that females had improved cardiac recovery and reduced infarction after I/R injury (Gabel et al. 2005). However, both male and female rats exposed to prenatal hypoxia are more susceptible to I/R injury (Rueda‐Clausen et al. 2011 b). Whether a sexual dichotomy exists, the fetal programming of metabolic and cardiovascular diseases and the impact this would have on treatment regimens for the prevention or early intervention of these diseases is not clear.

Resveratrol, a natural polyphenol found in grape skins, has numerous health benefits including metabolic and cardiovascular effects. We have shown that postnatal resveratrol administration ameliorates increased susceptibility to diet‐induced obesity and prevents the deleterious cardiovascular effects of a HF diet in young adult male offspring that were exposed to prenatal hypoxia (Dolinsky et al. 2011; Rueda‐Clausen et al. 2012 a). Resveratrol is known to improve metabolic function via AMP‐activated protein kinase (AMPK)–acetyl CoA carboxylase (ACC) and AMPK/perioxysome proliferator‐activated receptor‐γ coactivator (PGC)‐1α pathways, resulting in increased fatty acid oxidation and mitochondrial biogenesis, respectively (Fullerton & Steinberg, 2010). In addition, resveratrol‐mediated cardioprotection after I/R injury has been shown to involve mitigation of oxidative stress (Ray et al. 1999; Dernek et al. 2004), which is one of the principal underlying mechanisms in the pathophysiology of cardiovascular diseases. Clinical studies have shown that growth‐restricted offspring resulting from pregnancy complications involving poor maternal nutrition had increased markers of oxidative stress in cord blood (Gupta et al. 2004). Further, offspring exposed to prenatal hypoxia exhibited increased cardiac oxidative stress (Evans et al. 2012; Rueda‐Clausen et al. 2012 b). However, the mechanism(s) involved in resveratrol‐mediated cardioprotection in male and female offspring exposed to the combined insult of prenatal hypoxia and a postnatal HF diet are not known. We hypothesized that prenatal hypoxia impairs metabolic and cardiac function after I/R injury in both male and female offspring. We further hypothesized that resveratrol ameliorates metabolic function via the AMPK–ACC pathway, reduces cardiac oxidative stress and improves cardiac function post‐I/R injury.

Methods

Animal ethics

All procedures described in this study were approved by the University of Alberta Health Sciences Animal Policy and Welfare Committee, and were in accordance with the guidelines of the Canadian Council on Animal Care. All experimental protocols conformed to the National Institutes of Health's Guide for the Care and Use of Laboratory Animals (NIH publication No. 85‐23, revised 1996).

Animal models

Female Sprague–Dawley rats, weighing 250–275 g, were obtained (Charles River, Quebec, Canada) and housed in a temperature controlled room with a 10 h:14 h light–dark cycle. After acclimatization for a week, they were mated overnight and pregnancy was confirmed (day 0) by the presence of sperm in a vaginal smear obtained the following morning. Pregnant dams were fed standard chow (Lab Diet, Ref. 5001; 3.02 kcal mg⁻¹; protein 23%, fat 4.5%, fibre 6%) throughout pregnancy. On day 15 of pregnancy, dams were randomly assigned to normoxia or maternal hypoxia protocols. Pregnant dams in the hypoxia protocol were individually housed in a Plexiglas chamber where the oxygen concentration was maintained at 11% by the continuous infusion of nitrogen gas during the last third of pregnancy (from day 15 to 21). This model results in fetal hypoxia ((Rueda‐Clausen et al. 2011 b). While it has been reported that food intake is decreased during hypoxic exposure (Camm et al. 2010), the lower nutrient supply may be appropriate for lower metabolic demand, as we have previously observed (Williams et al. 2005). Moreover, the body weight of the dams (minus the weight of the uterus and pups) is not altered by exposure to hypoxic conditions (data not shown), suggesting adequate maternal nutrition. However, this animal model may consist of an interaction between hypoxia and decreased nutrient intake. Both normoxic and hypoxic dams were allowed to give birth in normal housing conditions (21% oxygen).

After birth, pup body weight, crown‐to‐rump length and abdominal girth were recorded and litters were culled to eight pups (4 males and 4 females) to control the postnatal environment. At 3 weeks of age, offspring were weaned, single housed and randomly assigned to receive either a HF diet (D12451; Research Diet) or a HF diet with resveratrol supplementation (D120020402 4 g kg⁻¹; Research Diet) for 9 weeks. All experimental groups were fed on a HF diet, which was utilized as a tool (secondary stress factor) in order to establish a metabolic phenotype and hence allow assessment of the effect of resveratrol intervention. The dose of resveratrol was chosen based on previous studies (Lagouge et al. 2006; Dolinsky et al. 2011). Thus, the experimental study consisted of four groups: (1) normoxia–HF, (2) normoxia–HF + resveratrol, (3) prenatal hypoxia–HF and (4) prenatal hypoxia–HF + resveratrol. Male and female offspring from normoxic (n = 10) and hypoxic (n = 10) dams were utilized. Offspring were randomized to the above‐mentioned four groups such that no two animals in one group came from the same litter.

Body composition and adiposity

After weaning, body weight and food intake were recorded once per week. Body composition was determined using a whole body composition analyser based on time‐domain nuclear magnetic resonance technology (EchoMRI 4‐in‐1/1000; Echo Medical Systems, Houston, TX, USA). Abdominal circumference and crown to rump length were measured. Abdominal fat depots were dissected and weighed following killing for working heart preparation experiments at 12 weeks of age.

Glucose tolerance test

At 11 weeks of age and after 6 h of fasting, basal glucose levels were measured using an Accu‐Chek Aviva Nano blood glucose meter (Roche Diagnostics). A 50% glucose solution (2 g (kg body weight⁻¹)) was administered orally using a gavage needle. Blood glucose concentrations were measured via tail blood sampling at 15, 30, 60, 90 and 120 min following the glucose administration.

Echocardiography

Left ventricular morphology and function were assessed in male and female offspring at the age of 12 weeks using non‐invasive transthoracic echocardiography as previously described (Rueda‐Clausen et al. 2009). All echocardiographic assessments were performed by a single operator. The echocardiography was performed in the supine or semi‐left lateral decubitus position using a high‐resolution in vivo micro‐imaging system, Vevo 2100 (Visualsonics, Toronto, ON, Canada), equipped with a 13–23 MHz linear array transducer (Ram et al. 2011). Animals were anaesthetized (sedated with 4% isoflurane and 1 L min⁻¹ compressed air and maintained at 1.5% isoflurane and 1 L min⁻¹ compressed air) and echocardiography was performed in two‐dimensional guided M‐mode. The heart was imaged in the two‐dimensional mode in the parasternal and short axis view with a depth of 2 cm. From this view, an M‐mode cursor was positioned perpendicular to the interventricular septum and posterior wall of the LV at the level of the papillary muscles. Parameters of left ventricular morphology obtained were the following: left ventricular internal diameters during diastole (LVID_d) and systole (LVID_s), diastolic and systolic interventricular septal wall thickness (IVS_d and IVSs), and diastolic and systolic left ventricular posterior wall thickness (LVPW_d and LVPW_s). Ejection fraction (EF) and fractional shortening (FS) were calculated using ventricular diameters and thicknesses obtained from M‐mode images according to American Society of Echocardiography recommendations (Sahn et al. 1978) as follows:

$$\mathrm{EF}\left(\%\right)=\left[\left(\mathrm{LVEDV}-\mathrm{LVESV}\right)/\mathrm{LVEDV}\right]\times 100$$

$$\mathrm{FS}\left(\%\right)=\left[\left(\mathrm{LVI}{\mathrm{D}}_{\mathrm{d}}-\mathrm{LVI}{\mathrm{D}}_{\mathrm{s}}\right)/\mathrm{LVESV}\right]\times 100$$

where LVEDV is left ventricular end diastolic volume and LVESV is left ventricular end systolic volume.

Left ventricle (LV) mass (g) was also automatically calculated using M‐mode images by the uncorrected cube function formula:

$$\begin{array}{ccc}\hfill \mathrm{LV}\mathrm{mass}& =& 1.053\times \left(\left(\mathrm{LVI}{\mathrm{D}}_{\mathrm{d}}\right.\right.\hfill \\ & & +\left.{\left.\mathrm{LVP}{\mathrm{W}}_{\mathrm{d}}+\mathrm{IV}{\mathrm{S}}_{\mathrm{d}}\right)}^3-\mathrm{LVI}{{\mathrm{D}}_{\mathrm{d}}}^3\right)\hfill \end{array}$$

where 1.053 is the specific gravity of myocardium.

In the apical four‐chamber view, the transmitral flow velocity profile (E wave and A wave velocities) was obtained and used to assess diastolic function parameters including isovolumic relaxation time (IVRT) and isovolumic contraction time, E/A ratio and deceleration time of E wave. Tissue Doppler imaging was used to obtain mitral annular velocities (E′ and A′).

Isolated working heart preparation

At 12 weeks of age, isolated working hearts were prepared using a previously described method (Rueda‐Clausen et al. 2009). In brief, the heart was rapidly excised from anaesthetized rats, kept in ice‐cold modified Krebs–Henseleit solution (mm: 120 NaCl, 25 NaHCO₃, 5.5 glucose, 4.7 KCl, 1.2 KH₂PO₄, 1.2 MgSO₄, 2.5 CaCl₂, pH 7.4) and inserted onto a cannula, and the aorta was ligated with silk. Hearts were perfused with oxygenated Krebs–Henseleit solution for 10 min in retrograde Langendorff mode. During this time, excess pericardial tissues were removed and the left atrium was cannulated. After stabilization, hearts were perfused in anterograde working mode by opening of the atrial inflow and aortic outflow lines. Cardiac function was reported as cardiac power (J min⁻¹ (g dry wt)⁻¹), calculated as: [(peak systolic pressure (mmHg) – maximal preload (mmHg)) × cardiac output (mL min⁻¹) × 0.13]/dry weight (g)]. Cardiac parameters were recorded using an HSE data acquisition system and isoheart software for Windows 2000 (Harvard Apparatus, Canada). Measurements of cardiac function were carried out every 10 min for a period of 80 min, consisting of 30 min stabilization (aerobic), 10 min no‐flow ischaemia and 40 min reperfusion. Hearts which experienced non‐reversible arrthymia were excluded from the study.

Western blot analysis

At 12 weeks of age, liver tissue was collected following killing for working heart preparation experiments and LV tissues were collected at the end of working heart experiments. Tissues were snap frozen and stored at −80ºC. Frozen tissues were homogenized in lysis buffer (mm: 20 Tris (pH 7.4), 5 EDTA, 10 sodium pyrophosphate tetrabasic, 100 sodium fluoride, and 1% NP‐40, and protease inhibitor cocktail of 1× Halt 200 (Thermo Fisher Scientific, Waltham, MA, USA) and 1 mm PMSF (Sigma‐Aldrich, St. Louis, MO, USA)) and phosphatase inhibitors (2 mm sodium orthovanadate, Sigma‐Aldrich). The protein concentration of the lysate was determined using the bicinchoninic acid assay (Pierce/Thermo Fisher Scientific); 100 μg of protein was loaded and separated by SDS‐PAGE on a 7.5% or 10% polyacramide gel and transferred to a nitrocellulose membrane. The membrane was incubated with 50% blocking reagent for 1 h at room temperature. After washing with phosphate buffered saline solution, the membrane was incubated overnight at 4ºC with primary antibodies for: p‐AMPK (1:2000, Cell Signaling Technology, Danvers, MA, USA), p‐ACC (1:2000, Cell Signaling) or β‐actin (1:1000; Santa Cruz Biotechnology, Dallas, TX, USA). The membrane was incubated with secondary antibody conjugated with fluorescent tag and blots were visualized with a LI‐COR Odyssey Bioimager and quantified by densitometry with software (Odyssey v3.0, LI‐COR Biosciences, Lincoln, NE, USA).

Dihydroethidium staining

Evidence of superoxide anions in cardiac tissue after I/R injury was assessed using dihydroehidium (DHE) fluorescence as described elsewhere (Zanetti et al. 2005). After I/R, LV tissue was embedded in OCT and sliced (10 μm thickness) using a cryostat. Sections were thawed, washed three times with Hanks’ balanced salt solution (HBSS), and incubated with DHE (200 μmol l⁻¹) for 30 min at 37ºC. The image was obtained using a fluorescence microscope (IX81 Olympus, Japan) with a 546 nm filter attached to an image analysis system (Cell Sense, Olympus, Japan). Fluorescence intensity was quantified using ImageJ 1.48 software.

Statistical analysis

Data are presented as means ± standard error of the mean. Student's t test was used to assess differences between birth outcomes from normoxic dams and dams exposed to hypoxia. Statistical significance of differences among the four treatment groups was assessed using two‐way ANOVA with prenatal hypoxia and resveratrol as sources of variation, followed by a Bonferroni multiple comparison post hoc test using Prism 6 software (GraphPad Software, San Diego, CA, USA). A P value of ˂ 0.05 was considered statistically significant.

Results

Effect of prenatal hypoxia on fetal phenotype at birth

We have established an animal model of low birth weight using prenatal hypoxia as an in utero insult and have extensively characterized the fetal outcomes at birth in our previous studies (Dolinsky et al. 2011; Rueda‐Clausen et al. 2011 b, 2012 a). Exposure of pregnant dams to hypoxia during the last third of pregnancy did not significantly alter the gestational length. In both male and female offspring, prenatal hypoxia decreased birth weight (Fig. 1 A and D), and abdominal girth (Fig. 1 B and E), without affecting crown‐to‐rump length (Fig. 1 C and F). There were no significant differences between the groups in litter size (normoxia 14 ± 0.7 vs. prenatal hypoxia 15 ± 0.6 pups, P = 0.3) or sex distribution (proportion of male offspring: normoxia 47.64% vs. prenatal hypoxia 51.44%, P = 0.7).

Figure 1. Effect of prenatal hypoxia on phenotypes at birth in male (upper panels) and female (lower panels) offspring .

A and D, body weight; B and E, abdominal girth; C and F, crown‐to‐rump length. A t test was performed to compare between normoxia and prenatal hypoxia (P‐hypoxia)‐exposed offspring. *P < 0.05, **P < 0.01 vs. normoxia (n = 10 per group).

Effect of resveratrol on body weight gain, food intake and abdominal circumference

In male offspring exposed to prenatal hypoxia, there was no difference in body weight gain and food intake throughout the dietary intervention period (Fig. 2 A and B). Interestingly, in female offspring exposed to prenatal hypoxia, resveratrol supplementation significantly decreased body weight gained from 10 weeks onwards despite the same amount of food consumption throughout the dietary intervention period compared with the normoxia group (Fig. 2 D and E). There were no significant differences in abdominal circumferences among the groups after 9 weeks of dietary intervention in male or female offspring (Fig. 2 C and F).

Figure 2. Body weight gain, food intake and abdominal circumference .

In male (A and B) and female offspring (D and E), body weight gain and food intake were recorded every week from 3 to 12 weeks of age. Abdominal circumference for male (C) and female (F) was measured at 12 weeks of age. All groups were compared using two‐way ANOVA followed by a Bonferroni post hoc analysis (n = 10 per group). *P < 0.05 for respective source of variation (prenatal hypoxia or Resv). HF, high fat; Resv, resveratrol; P‐hypoxia, prenatal hypoxia.

Effect of resveratrol on body composition, abdominal fat and adipocyte morphometry

In vivo assessment of whole body composition using echo magnetic resonance imaging showed that there were no significant differences in whole body fat and lean tissue composition among the groups in male (Fig. 3 A) or female (Fig. 3 D) offspring following intervention with a HF diet, indicating no differences in general fat deposition. Abdominal fat measurement, an indicator of central fat depots, was recorded after mechanical extraction, which revealed that there was an overall effect of resveratrol treatment on reducing percent abdominal fat in only male offspring (Fig. 3 B). Furthermore, histological analysis of visceral fat revealed that adipocyte diameter was larger in both male (Fig. 3 C) and female (Fig. 3 F) offspring exposed to prenatal hypoxia and a postnatal HF diet. Consistent with the autopsy data, resveratrol treatment had an overall effect on decreasing adipocyte diameter in only male offspring (Fig. 3 C) without affecting adipocyte diameter in female offspring (Fig. 3 F).

Figure 3. Body composition, abdominal fat and adipocyte morphometry .

At 12 weeks of age, body composition and abdominal fat pads were measured. In male (A and B) and female (D and E) offspring, body composition data and abdominal fat were expressed as a percentage of the body weight. Average adipocyte diameter in male (C) and female (F) offspring after quantification of haematoxylin–eosin stained images of adipose tissue section. All groups were compared using two‐way ANOVA followed by a Bonferroni post hoc analysis (n = 5–8). *P < 0.05, ***P < 0.001 for respective source of variation (prenatal hypoxia or Resv). ††P < 0.01 vs. prenatal hypoxia–HF. #P < 0.05 vs. normoxia–HF. HF, high fat; Resv, resveratrol; P‐hypoxia, prenatal hypoxia.

Effect of resveratrol on oral glucose tolerance

Being born growth restricted due to exposure to prenatal hypoxia significantly increased the fasting plasma glucose levels in male offspring fed a HF diet (Fig. 4 A), indicating impaired fasting glucose handling compared with normoxia male offspring on the same diet. The ability to dispose of glucose from the blood after a 2 g kg⁻¹ glucose load was tested using an oral glucose tolerance test, which showed significantly impaired glucose tolerance in male offspring exposed to prenatal hypoxia (Fig. 4 B). Interestingly, resveratrol supplementation did not affect fasting basal plasma glucose (Fig. 4 A) but significantly improved the glucose tolerance in male offspring exposed to prenatal hypoxia and fed the HF diet (Fig. 4 B and C). In contrast, fasting plasma glucose levels in female offspring fed a HF diet were not affected by either prenatal hypoxia exposure or resveratrol treatment (Fig. 4 D). Furthermore, the glucose tolerance was similar among the groups in female offspring (Fig. 4 E and F), indicating better glucose handling by female offspring and signifying a sex dichotomy in the fetal programming of metabolic diseases in this model.

Figure 4. Glucose tolerance test .

After 6 h of fasting, basal plasma glucose levels were measured in male (A) and female (D) offspring. A glucose tolerance test (GTT) in male (B) and female (E) offspring was performed for 120 min after 2 g kg⁻¹ glucose load. GTT area under the curve (AUC) was calculated in male (C) and female (F) offspring. All groups were compared using two‐way ANOVA followed by a Bonferroni post hoc analysis (n = 5 per group). *P < 0.05 for respective source of variation (prenatal hypoxia or Resv), †P < 0.05 vs. prenatal hypoxia–HF. HF, high fat; Resv, resveratrol; P‐hypoxia, prenatal hypoxia.

Effect of resveratrol on AMPK–ACC signalling

We have previously shown that resveratrol activates the AMPK–ACC axis in liver tissue in normoxia and prenatal hypoxia‐exposed male offspring fed a HF diet (Dolinsky et al. 2011). Considering the intact metabolic function observed in female offspring in the present study, despite the offspring being fed a HF diet, we investigated the effect of resveratrol on the AMPK–ACC axis in the liver tissue in female offspring. Neither prenatal hypoxia nor resveratrol supplementation affected the AMPK–ACC axis in the liver tissue in female offspring (Fig. 5 A and B).

Figure 5. Representative Western blots for phosphorylation of AMP‐activated protein kinase (AMPK) and acetyl CoA carboxylase (ACC) .

In female (Fig. 5 A and B) offspring, proteins were measured in the liver tissue lysates after 9 weeks of dietary intervention. Blots were quantified using Odyssey software and depicted as p‐AMPK/β‐actin or p‐ACC/β‐actin ratio (Fig. 5 A and B). All groups were compared using a two‐way ANOVA. Data were obtained from 3 independent experiments. HF, high fat; Resv, resveratrol; P‐hypoxia, prenatal hypoxia.

Effect of resveratrol on cardiac morphometry and in vivo cardiac function

We have previously demonstrated that prenatal hypoxia alters cardiac morphometry (cardiac hypertrophy) at 12 months of age, but not at 4 months of age, in male but not female offspring (Rueda‐Clausen et al. 2009). Consistent with this finding, in young adult offspring (12 weeks of age), ex vivo data revealed that there were no significant differences in heart weight or left ventricular weight in either male (Fig. 6 A and B) or female (Fig. 6 D and E) offspring, indicating that prenatal hypoxia combined with a postnatal HF diet had no effect on cardiac morphometry at a young age. In agreement with the ex vivo data, echocardiography data revealed that cardiac morphometry (LV septal and posterior wall thickness, LV mass and LV internal diameters) was unaltered by prenatal hypoxia and postnatal HF diet in male (Table 1) or female (Table 2) offspring. There was an interaction effect in LV end systole septal thickness and end diastole posterior wall thickness with opposing effects of resveratrol in normoxia vs. prenatal hypoxia‐exposed offspring (P < 0.01). However, the effect of prenatal hypoxia or resveratrol treatment on these parameters did not reach statistical significance (Table 2). To further validate this finding at the cellular level, a histological analysis of LV tissue was performed and cardiac myocyte diameter was determined. Prenatal hypoxia did not alter myocyte diameter among the groups in either male (Fig. 6 C) or female offspring (Fig. 6 F). However, there was an overall effect of resveratrol on myocyte diameter in both male (Fig. 6 C) and female (Fig. 6 F) offspring. In male offspring exposed to prenatal hypoxia, resveratrol significantly decreased the myocyte diameter (Fig. 6 C).

Figure 6. Ex vivo cardiac morphometry .

Average heart weight and left ventricular weight in male (A and B) and female (D and E) offspring were measured. Values were normalized to body weight/100. Average cardiomyocyte diameter in male (C) and female (F) offspring after quantification of haematoxylin–eosin stained images of heart tissue section. All groups were compared using two‐way ANOVA followed by a Bonferroni post hoc analysis (n = 5 per group). *P < 0.05, **P < 0.01 for respective source of variation (prenatal hypoxia or Resv), †P < 0.05 vs. prenatal hypoxia–HF. HF, high fat; Resv, resveratrol; P‐hypoxia, prenatal hypoxia.

Table 1.

Echocardiographic assessments in male normoxia and prenatal hypoxia‐exposed offspring

	Normoxia		Prenatal hypoxia
Measurement	HF diet	HF diet + Resv	HF diet	HF diet + Resv	P‐hypoxia	Diet	Int
Cardiac morphometry
End‐diastole septal thickness (mm)	2.15 ± 0.10	2.17 ± 0.17	2.15 ± 0.03	2.17 ± 0.11
End‐systole septal thickness (mm)	3.93 ± 0.22	3.83 ± 0.05	3.50 ± 0.25	3.64 ± 0.13
LV end‐diastole internal diameter (mm)	8.29 ± 0.24	8.24 ± 0.28	8.11 ± 0.22	7.82 ± 0.26
LV end‐systole internal diameter (mm)	3.69 ± 0.5	4.01 ± 0.42	3.92 ± 0.5	4.01 ± 0.27
LV end‐diastole posterior wall thickness (mm)	2.34 ± 0.12	2.16 ± 0.08	2.17 ± 0.06	2.34 ± 0.08
LV end‐systole posterior wall thickness (mm)	3.92 ± 0.22	3.67 ± 0.20	3.44 ± 0.17	3.79 ± 0.12
Systolic function
Body weight (g)	652.6 ± 26.4	648.8 ± 28.7	638.8 ± 27.3	603.3 ± 31
Basal heart rate (beats min−1)	364 ± 11	363 ± 20.8	368 ± 13	365 ± 12.9
LV end‐diastole volume (μl)	397.8 ± 24.8	406.2 ± 31.9	380.9 ± 23.0	359.8 ± 26.6
LV end‐systole volume (μl)	61.1 ± 14.4	70.9 ± 19.1	67.0 ± 17.2	70.0 ± 14.1
Cardiac output (ml min−1)	123.1 ± 6.2	128.4 ± 19.1	115.7 ± 7.2	105.9 ± 8.1
LV ejection fraction (%)	83.4 ± 3.7	80.0 ± 4.9	80.0 ± 4.4	78.1 ± 3.0
LV shortening fraction (%)	55.7 ± 5.0	51.3 ± 4.7	52.2 ± 5.3	48.8 ± 2.9
Diastolic function
Mitral E max velocity (mm s−1)	1101.8 ± 30.6	892.7 ± 56.9	900.4 ± 75.6	880.4 ± 79.5
Mitral A max velocity (mm s−1)	964.8 ± 44.0	732.8 ± 64.2	685.6 ± 78.6	720.7 ± 50.9	*		*
Mitral E/A index	1.15 ± 0.03	1.23 ± 0.04	1.34 ± 0.1	1.22 ± 0.06
Mitral deceleration time (ms)	41.0 ± 3.7	42.2 ± 3.3	38.5 ± 2.3	47.4 ± 2.9
IVRT (ms)	18.5 ± 0.6	18.9 ± 1.4	19.0 ± 0.5	17.9 ± 1.0
Myocardial performance (Tei) index	0.56 ± 0.04	0.62 ± 0.10	0.60 ± 0.03	0.57 ± 0.05
Tissue Doppler mitral E′ wave (mm s−1)	45.2 ± 4.8	54.1 ± 2.4	41.2 ± 2.9	44.6 ± 3.2
Tissue Doppler mitral A′ wave (mm s−1)	44.1 ± 4.1	54.3 ± 2.7	43.4 ± 4.0	45.2 ± 3.6
Tissue Doppler E′/A′ index	1.02 ± 0.02	1.00 ± 0.01	0.96 ± 0.04	0.99 ± 0.02

Values are expressed as means ± SEM (n = 5 per group). All groups were compared using a two‐way ANOVA followed by a Bonferroni post hoc test. *P < 0.05, for the respective source of variation (prenatal hypoxia or Resv). HF, high fat; Resv, resveratrol; P‐hypoxia, prenatal hypoxia; Int, interaction; LV, left ventricle; IVRT, isovolumic relaxation time.

Table 2.

Echocardiographic assessments in female normoxia and prenatal hypoxia‐exposed offspring

	Normoxia		Prenatal hypoxia
Measurement	HF diet	HF diet + Resv	HF diet	HF diet + Resv	P‐hypoxia	Diet	Int
Cardiac morphometry
End‐diastole septal thickness (mm)	1.90 ± 0.06	1.99 ± 0.12	1.96 ± 0.14	1.86 ± 0.09
End‐systole septal thickness (mm)	3.09 ± 0.23	3.50 ± 0.27	3.47 ± 0.11	2.75 ± 0.32			*
LV end‐diastole internal diameter (mm)	6.72 ± 0.46	6.58 ± 0.28	6.88 ± 0.16	6.83 ± 0.25
LV end‐systole internal diameter (mm)	3.66 ± 0.5	3.07 ± 0.32	3.23 ± 0.2	3.56 ± 0.28
LV end‐diastole posterior wall thickness (mm)	1.94 ± 0.05	2.03 ± 0.10	2.09 ± 0.11	1.82 ± 0.06			*
LV end‐systole posterior wall thickness (mm)	3.25 ± 0.24	3.44 ± 0.10	3.06 ± 0.27	3.17 ± 0.11
M‐mode trace LV end‐diastole volume (μl)	287.1 ± 30.8	263.3 ± 14.1	268.1 ± 14.3	266.7 ± 21.6
M‐mode trace LV end‐sysstole volume (μl)	35.7 ± 5.3	32.9 ± 7.5	37.7 ± 6.4	42.5 ± 6.8
Systolic function
Body weight (g)	373.8 ± 10.5	388.8 ± 10.2	390.0 ± 16.1	357.8 ± 18.3
Basal heart rate (beats min−1)	384 ± 14	386 ± 16.0	353 ± 15	351 ± 14.1	*
Cardiac output (ml min−1)	95.9 ± 9.5	89.8 ± 8.6	80.7 ± 4.2	79.7 ± 10.0
LV ejection fraction (%)	87.6 ± 1.1	87.6 ± 2.5	86.0 ± 2.1	83.8 ± 2.5
LV shortening fraction (%)	59.2 ± 1.5	59.9 ± 3.6	57.3 ± 2.5	54.9 ± 3.2
Diastolic function
Mitral E max velocity (mm s−1)	915.6 ± 77.7	879.1 ± 42.0	760.1 ± 48.1	787.3 ± 37.7	*
Mitral A max velocity (mm s−1)	738.6 ± 50.4	676.5 ± 55.0	607.8 ± 50.7	582.3 ± 85.8
Mitral E/A index	1.14 ± 0.0	1.37 ± 0.14	1.27 ± 0.1	1.51 ± 0.27
Mitral deceleration time (ms)	38.7 ± 2.5	35.7 ± 3.7	39.2 ± 2.9	37.5 ± 2.4
IVRT (ms)	19.9 ± 1.8	16.7 ± 0.5	19.1 ± 1.2	19.2 ± 1.3
Myocardial performance (Tei) index	0.66 ± 0.07	0.58 ± 0.02	0.54 ± 0.03	0.56 ± 0.02
Tissue Doppler mitral E′ wave (mm s−1)	43.7 ± 2.7	40.8 ± 4.5	35.9 ± 2.4	36.9 ± 2.7
Tissue Doppler mitral A′ wave (mm s−1)	43.8 ± 2.5	39.2 ± 2.6	36.4 ± 3.6	37.8 ± 2.1
Tissue Doppler E′/A′ index	1.00 ± 0.04	1.03 ± 0.08	1.01 ± 0.06	0.97 ± 0.03

Values are expressed as means ± SEM (n = 5 per group). All groups were compared using a two‐way ANOVA followed by a Bonferroni post hoc test. *P < 0.05, for the respective source of variation (prenatal hypoxia or Resv). HF, high fat; Resv, resveratrol; P‐hypoxia, prenatal hypoxia; Int, interaction; LV, left ventricle; IVRT, isovolumic relaxation time.

In male (Table 1) and female (Table 2) offspring, cardiac output, LV ejection fraction and LV shortening fraction were not significantly different among the groups, indicating the absence of systolic dysfunction in offspring exposed to prenatal hypoxia and a postnatal HF diet compared with their normoxic counterparts. A similar result was observed for diastolic function, as indicated by similar mitral E/A index, Tei index and tissue Doppler E′/A′ index among the groups, in male (Table 1) and female (Table 2) offspring. However, there was an overall effect of prenatal hypoxia on mitral A max velocity and mitral E max velocity in male and female offspring, respectively.

Effect of resveratrol on ex vivo cardiac function

In both male (Fig. 7 A and B) and female (Fig. 7 D and E) offspring, baseline aerobic cardiac power was comparable among the groups. After inducing I/R injury, however, the recovery of cardiac power following mild ischaemia (10 min of global no‐flow ischaemia) was significantly decreased throughout the reperfusion period in both male (Fig. 7 A and C) and female (Fig. 7 D and F) offspring exposed to prenatal hypoxia and a postnatal HF diet compared with normoxia offspring exposed to a postnatal HF diet, suggesting a susceptibility to I/R injury in offspring exposed to prenatal hypoxia. In male offspring exposed to prenatal hypoxia, recovery of cardiac power was almost completely abolished (an ∼90% reduction; Fig. 7 C). Interestingly, in female offspring exposed to prenatal hypoxia, the ability of heart to recover after I/R injury was reduced by ∼50% (Fig. 7 F), implying a greater cardioprotection in the susceptibility to I/R injury in female offspring fed a HF diet. Following this observation, we performed a two‐way ANOVA considering sex and prenatal hypoxia as variables, which revealed that there was a significant overall effect of sex (P < 0.05) and prenatal hypoxia (P < 0.0001) among the groups, providing evidence of a sex dichotomy in cardiac power recovery after I/R injury. Furthermore, resveratrol significantly improved recovery of cardiac power after I/R in both males (Fig. 7 A and C) and females (Fig. 7 D and F) exposed to prenatal hypoxia and a postnatal HF diet, suggesting that resveratrol itself had a cardioprotective effect following mild I/R injury that was sex‐independent. Following resveratrol treatment in the male and female groups, there was no significant overall effect of prenatal hypoxia (P = 0.60) or sex (P = 0.65) among the groups.

Figure 7. Ex vivo cardiac function .

Cardiac function was assessed in isolated hearts using a Langendorff heart system. After stabilization, aerobic cardiac power was recorded for 30 min as a baseline. After 10 min of mild ischaemia, reperfusion cardiac power was recorded for 40 min in male (A) and female (D) offspring. Absolute aerobic cardiac power and percentage change of reperfusion cardiac power are shown in male (B and C) and female (E and F) offspring. All groups were compared using a two‐way ANOVA followed by a Bonferroni post hoc test (n = 5–7). *P < 0.05, for the respective source of variation (prenatal hypoxia or Resv), #P < 0.05 vs. normoxia–HF, †P < 0.05 vs. prenatal hypoxia–HF. HF, high fat; Resv, resveratrol; P‐hypoxia, prenatal hypoxia.

Effect of resveratrol on cardiac oxidative stress

Although the female normoxic offspring had lower superoxide levels than their male counterparts, the pattern of change with prenatal hypoxia and mitigation by resveratrol treatment was similar between the sexes (Fig. 8). In both male and female offspring (Fig. 8 A–E), evidence of cardiac oxidative stress after I/R injury was significantly greater in offspring exposed to prenatal hypoxia than in offspring exposed to normoxia (Fig. 8 F–J). Resveratrol treatment significantly decreased the cardiac oxidative stress after I/R in male (Fig. 8 A–E) and female (Fig. 8 F–J) offspring exposed to prenatal hypoxia.

Figure 8. Oxidative stress in cardiac tissue .

Light microscopic images of heart tissue sections stained with dihydroethidium in male (A–E) and female (F–J) offspring. Data were expressed as mean fluorescence after quantification of images using a fluorescence microscope. All groups were compared using a two‐way ANOVA followed by a Bonferroni post hoc test (n = 5 per group). *P < 0.05, for the respective source of variation (prenatal hypoxia or Resv), #P < 0.05 vs. normoxia–HF. †P < 0.05 vs. prenatal hypoxia–HF. HF, high fat; Resv, resveratrol; P‐hypoxia, prenatal hypoxia.

Discussion

Several studies, including our own, have shown that prenatal hypoxia during pregnancy poses a risk to offspring for long‐term adverse health outcomes including metabolic and cardiovascular health (reviewed in Rinaudo & Wang, 2012; Giussani & Davidge, 2013). In the present study, a postnatal HF diet was used to induce a metabolic phenotype in offspring that had been exposed to prenatal hypoxia and its developmental programming effects. This HF diet was given to all groups, including offspring exposed to a normoxic in utero environment, in order that the effects of prenatal environment and postnatal supplementation with resveratrol could be assessed. We made the interesting finding that female offspring exposed to prenatal hypoxia and a postnatal HF diet displayed impaired cardiovascular susceptibility to I/R injury without any effect on their metabolic health when compared with their normoxia, HF fed controls. However, consistent with previous findings by ourselves and others (Xue & Zhang, 2009; Rueda‐Clausen et al. 2011 a,b), in male offspring prenatal hypoxia impaired metabolic health and increased cardiac susceptibility to I/R injury.

Further we observed that resveratrol supplementation in the diet prevented cardiovascular susceptibility to I/R injury in both male and female offspring and improved metabolic health in male offspring. Taken together, the interesting findings of our study are twofold. Firstly, we demonstrated sex‐specific fetal programming of CVD and susceptibility of offspring to a secondary stressor such as a HF diet. In particular, susceptibility to metabolic disease was greater in prenatal hypoxia‐exposed offspring with females exhibiting greater protection of metabolic and cardiovascular health in later life. Secondly, we demonstrated that resveratrol had similar metabolic and cardioprotective effects, independent of sex, indicating its therapeutic potential in the treatment of metabolic and cardiovascular diseases in both sexes.

The effects of prenatal hypoxia on long‐term adverse metabolic and cardiovascular outcomes have been previously characterized in adult male offspring (Li et al. 2003; Dolinsky et al. 2011; Giussani et al. 2012; Al‐Hasan et al. 2014). However, the female counterpart has been relatively less investigated with only a few studies available in the literature. We therefore specifically aimed to investigate the long‐term effects of prenatal hypoxia on metabolic and cardiovascular function and cardiac susceptibility to I/R injury in female offspring born following IUGR due to prenatal hypoxia exposure and possible early life therapeutic intervention to prevent any manifestation of metabolic and cardiac dysfunction using resveratrol as a dietary supplement.

A population born with a low birth weight or IUGR has been shown to develop metabolic diseases in later life (Jaquet et al. 2000; Xiao et al. 2010). However, the fetal programming of metabolic diseases in later life may depend on the sex, age and specific insult experienced by the offspring (Alexander et al. 2014). Our current findings on the fetal programming of impaired glucose homeostasis due to prenatal hypoxia in male offspring exposed to prenatal hypoxia are consistent with our previous study (Dolinsky et al. 2011). Interestingly, in contrast to studies that have shown metabolic dysfunction in female IUGR offspring induced by diet restriction during pregnancy, we demonstrated that prenatal hypoxia did not programme metabolic phenotypes such as impaired fasting plasma glucose and impaired glucose tolerance in female offspring, despite being born following IUGR and fed a postnatal HF diet. In adult male offspring, prenatal hypoxia has been shown to promote molecular markers of insulin resistance (Camm et al. 2011), which may be one of the mechanisms for predisposing adult male offspring to metabolic diseases. In fact, the mechanism for metabolic dysfunction in male offspring exposed to prenatal hypoxia was shown to be due to impaired insulin signalling, i.e. increased phosphorylation of insulin receptor substrate‐1 (IRS‐1) at Ser1101 and subsequently decreased activation of protein kinase B (Akt) in liver tissue isolated after 15 min of insulin injection (Dolinsky et al. 2011). The normal metabolic function in female offspring exposed to prenatal hypoxia and a postnatal HF diet might be due to intact insulin signalling, which needs further investigation. The present finding, along with other studies, provides evidence for a sex dichotomy for susceptibility to developing metabolic dysfunction in a prenatal hypoxia‐induced IUGR population. Thus this finding may be relevant in a clinical setting as it imparts knowledge about the importance of sex consideration while dealing with an IUGR population for their susceptibility to impaired metabolic health in later life. Also, the nature of the insult occurring during pregnancy, such as maternal protein restriction, placental insufficiency, or prenatal hypoxia, should be taken into consideration to address the long‐term metabolic complications in a specific at‐risk population to create a personalized therapeutic approach (Dessi et al. 2012).

We further demonstrated that resveratrol could potentially be used as an early life therapeutic intervention to prevent metabolic dysfunction in our animal model of prenatal hypoxia‐induced IUGR. It has been shown that AMPK activation may play an adaptive role to maintain utero‐placental blood flow in a hypoxic pregnancy by causing vasodilatation of the uterine artery in a murine model (Skeffington et al. 2016), which signifies the protective role of AMPK during a hypoxic pregnancy. We have previously shown that resveratrol activates the AMPK–ACC axis in the liver tissue in both normoxia and prenatal hypoxia‐exposed male offspring (Dolinsky et al. 2011). In contrast, we did not observe an activation of the AMPK–ACC axis by resveratrol in the liver tissue in either normoxia or prenatal hypoxia‐exposed female offspring in the present study. In fact, several studies have shown that resveratrol did not improve metabolic function in normal rodents (Turrens et al. 1997; Juan et al. 2002; Jeon et al. 2012). Similarly, it was reported that resveratrol supplementation neither improved in vivo metabolic function nor activated AMPK in skeletal muscle tissue in non‐obese women with normal metabolic function (Yoshino et al. 2012). These observations indicate the differential role of resveratrol to activate its downstream metabolic pathways, which may be dependent on the sex and metabolic status of the individual.

Many cardiovascular complications lead to cardiac remodelling in order to compensate for compromised cardiac function and this may ultimately lead to heart failure. Several clinical and experimental studies have demonstrated a clear link between being born growth restricted and susceptibility to CVD, which may be sex dependent (Rueda‐Clausen et al. 2011 b; Intapad et al. 2014). Surprisingly, neither our previous studies (Rueda‐Clausen et al. 2011 b) nor the current study showed any changes in basal cardiac function when assessed using in vitro or in vivo techniques. In fact, in many instances, CVDs may undergo a subclinical phase for decades before the first clinical manifestation appears (Berenson, 2002). Therefore, an IUGR population, with a subclinical phase of cardiovascular manifestation in prenatal and postnatal life, may exhibit cardiac dysfunction with the presence of secondary stressors such as HF diet and/or I/R injury (Rueda‐Clausen et al. 2011 b). These observations may suggest that the cardiovascular status of an IUGR population is in a compensatory physiological state, which could be triggered to a pathological state by postnatal stressors. In vivo echocardiographic assessment revealed no differences in systolic or diastolic function among the groups. However, we observed a lower heart rate in female offspring exposed to prenatal hypoxia. We have previously shown that aged animals had a decrease in heart rate (4 months vs. 12 months of age) while IUGR animals had no changes in heart rate when compared with their respective controls in male and female offspring (Rueda‐Clausen et al. 2009). In contrast, the present study showed a lower heart rate only in female offspring. Consistent with the present finding, Jansson & Lambert (1999) have shown that there was a significant positive correlation between birth weight and heart rate in female, but not male, IUGR rats (created by uterine artery ligation). The lower heart rate observed in female IUGR rats could be because of an increased parasympathetic tone to the heart in female IUGR offspring in order to maintain mean arterial pressure in a situation of elevated sympathetic activity observed in IUGR animals (Jansson & Lambert, 1999). The determination of the mechanism for the lower heart rate in female offspring requires further investigation in our animal models.

In regard to sex dimorphisms, several studies have observed cardioprotection after I/R in females using the isolated heart technique (Gabel et al. 2005; Wang et al. 2006). In regard to effects due to prenatal environments, Xue & Zhang (2009) have shown that prenatal hypoxia increased cardiac susceptibility to I/R injury in male offspring caused by a downregulation of protein kinase C ε expression in adult hearts. However, in adult female offspring, they have shown that prenatal hypoxia had no effect on cardiac function recovery after I/R (Xue & Zhang, 2009). In contrast, we have previously shown that prenatal hypoxia decreased cardiac function recovery after I/R in female offspring (Rueda‐Clausen et al. 2011 b). The discrepancy between these studies may be due to differences in the age of the offspring, the hypoxic environment and experimental protocol used in these studies. Consistent with our previous finding, the present study also showed decreased cardiac function recovery after I/R in female offspring fed a HF diet. Interestingly, compared with male offspring, we also observed better protection from cardiac susceptibility to I/R injury in female IUGR offspring, indicating a sexual dimorphism in the susceptibility to I/R injury in offspring born after exposure to prenatal hypoxia and in line with previous literature supporting a cardioprotective effect in females. Furthermore, it is possible that the sex‐dependent differences in cardiac susceptibility to I/R injury might be due to the metabolic status of the IUGR offspring. It is plausible that impaired metabolic health in male IUGR offspring predisposes them to greater susceptibility to cardiac dysfunction. In contrast, the intact metabolic health observed in female IUGR offspring may have provided better cardiac tolerance to I/R injury. In fact, metabolic syndrome and its components are well recognized risk factors for CVD, and epidemiological studies have demonstrated that an IUGR population developed metabolic syndromes such as glucose intolerance, insulin resistance and dyslipidaemia ultimately leading to CVD (Ma & Hardy, 2012).

Increased oxidative stress has also been linked to CVDs such as hypertension (Sinha & Dabla, 2015) and coronary artery disease (Pennathur et al. 2001; Khatri et al. 2004). In particular, cardiac oxidative stress causes hypertrophy, fibrosis and apoptosis leading to impaired cardiac function (Sugden & Clerk, 1998; Shiomi et al. 2004; Giordano, 2005). Sex‐dependent differences in cardiac function recovery after I/R injury, therefore, may be due to the heightened cardiac oxidative stress observed in male offspring compared with female offspring exposed to prenatal hypoxia. In fact, we have previously shown that prenatal hypoxia was associated with an increase in oxidative stress after I/R injury only in the adult male offspring (Rueda‐Clausen et al. 2012 b). However, it is important to note that the pattern of change of superoxide levels due to prenatal hypoxia as well as mitigation by resveratrol treatment in the current study was similar between the sexes. Future studies to further delineate mechanisms for both sexes are needed.

CVD, in particular ischaemic heart disease, remains one of the most prevalent causes of morbidity and mortality despite substantial advancement in its management. One of the complications in the treatment of ischaemic heart disease includes reperfusion‐induced cardiac dysfunction. Animal studies have shown that cardiac oxidative stress was increased during I/R injury. Therefore, one of the therapeutic strategies to reduce cardiac dysfunction during I/R injury may be the reduction of cardiac oxidative stress. Studies have shown that treatment of offspring exposed to prenatal hypoxia with an antioxidant during prenatal or postnatal life prevented cardiac dysfunction (Kane et al. 2013) and cardiac susceptibility to I/R injury in adult life (Patterson et al. 2012). Resveratrol, a known antioxidant, has been shown to improve cardiac dysfunction after I/R via the reduction of cardiac oxidative stress (Ray et al. 1999; Dernek et al. 2004). Consistent with this finding, we observed a decrease in cardiac oxidative stress in reperfused heart tissue in resveratrol‐treated male and female offspring exposed to prenatal hypoxia. This indicates that the improved contractile function post I/R in the resveratrol treated groups might be due to a decreased level of cardiac oxidative stress. However, considering the multiple molecular targets of resveratrol to protect against ischaemic damage and I/R injury (Zordoky et al. 2015), it is very plausible that the cardioprotective effects of resveratrol in our animal model involve multiple molecular pathways.

In conclusion, the present study has shown that prenatal hypoxia programs a susceptibility to metabolic and cardiovascular disorders that may manifest in male and female offspring, even at a young age, following exposure to secondary stressors such as a HF diet. Furthermore, male offspring are more susceptible to a manifestation of metabolic and cardiovascular disorders compared with their female counterparts; however, females were not completely protected in regard to recovery from cardiac I/R injury. Our findings indicate a sex dichotomy in the fetal programming of diseases that highlights the importance of sex consideration in the clinical setting while assessing long‐term risks of diseases in susceptible populations. In addition, the results of the present study suggest that early intervention with resveratrol supplementation could be a potential therapeutic approach to preventing a manifestation of metabolic and cardiovascular diseases in susceptible adult male and female populations.

Additional information

Competing interests

None.

Author contributions

A.S. contributed to the conception and design of the study, collection of data, and analysis and interpretation of data, and drafted the manuscript. L.M.R., D.F. and J.S. contributed to collection and interpretation of data. S.T.D. and J.S.M. contributed to the conception and design of the study, analysis and interpretation of data and critical revision of the manuscript. All authors have approved the final version of the manuscript and agree to be accountable for all aspects of the work. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed. Studies were conducted at University of Alberta, Alberta, Canada.

Funding

This work is funded by grants from the Canadian Institutes of Health Research (CIHR); and the Women and Children's Health Research Institute (WCHRI) through the generous contributions of the Stollery Children's Hospital Foundation (SCHF) and the Royal Alexandra Hospital Foundation (RAHF). A.S. is supported by fellowships from the Heart and Stroke Foundation of Canada; L.R., J.S. and D.F. are supported by studentships from Alberta Innovates – Health Solutions (AIHS); S.D. is a Canada Research Chair in Maternal and Perinatal Cardiovascular Health.