Maternal high-fat diet impairs cardiac function in offspring of diabetic pregnancy through metabolic stress and mitochondrial dysfunction

Abstract

Offspring of diabetic pregnancies are at risk of cardiovascular disease at birth and throughout life, purportedly through fuel-mediated influences on the developing heart. Preventative measures focus on glycemic control, but the contribution of additional offenders, including lipids, is not understood. Cellular bioenergetics can be influenced by both diabetes and hyperlipidemia and play a pivotal role in the pathophysiology of adult cardiovascular disease. This study investigated whether a maternal high-fat diet, independently or additively with diabetes, could impair fuel metabolism, mitochondrial function, and cardiac physiology in the developing offspring's heart. Sprague-Dawley rats fed a control or high-fat diet were administered placebo or streptozotocin to induce diabetes during pregnancy and then delivered offspring from four groups: control, diabetes exposed, diet exposed, and combination exposed. Cardiac function, cellular bioenergetics (mitochondrial stress test, glycolytic stress test, and palmitate oxidation assay), lipid peroxidation, mitochondrial histology, and copy number were determined. Diabetes-exposed offspring had impaired glycolytic and respiratory capacity and a reduced proton leak. High-fat diet-exposed offspring had increased mitochondrial copy number, increased lipid peroxidation, and evidence of mitochondrial dysfunction. Combination-exposed pups were most severely affected and demonstrated cardiac lipid droplet accumulation and diastolic/systolic cardiac dysfunction that mimics that of adult diabetic cardiomyopathy. This study is the first to demonstrate that a maternal high-fat diet impairs cardiac function in offspring of diabetic pregnancies through metabolic stress and serves as a critical step in understanding the role of cellular bioenergetics in developmentally programmed cardiac disease.

Keywords: cardiac bioenergetics; developmental programming; diabetic cardiomyopathy; diabetic pregnancy; maternal high-fat diet; mitochondrial dysfunction; proton production rate.

Fig. 1.

Lipid droplet analysis. A: Oil Red O staining of newborn rat hearts from each group (10-μm left ventricular sections at ×60) shows red lipid droplets indicated by arrowheads. CD-CB, control; CD-STZ, diabetes exposed; HF-CB, high-fat (HF) diet exposed; HF-STZ, combination exposed. Note that in HF diet-exposed hearts, especially in combination with diabetes, the lipid droplets increase in both number and size. B: point counting of lipid droplets revealed a significant interaction effect by 2-way ANOVA (^P = 0.01). Offspring from HF-CB and HF-STZ groups had significantly more droplets compared with controls by 1-way ANOVA posttest (*P < 0.05).

Fig. 2.

Mitochondrial stress test. Oxygen consumption rate (OCR) trace (A) and bar graphs (B and C) illustrate mitochondrial respiration under various conditions. Diabetes-exposed cardiomyocytes have lower basal, maximal, spare, nonmitochondrial, ATP-linked, and proton leak OCR. HF diet-exposed cardiomyocytes have significantly lower basal OCR. Combination-exposed cardiomyocytes consistently display the lowest mitochondrial respiration of all groups. Values are means ± SE; n = 5–7 litters/group. Significant differences: *dietary effect, ^±diabetes effect (P ≤ 0.05).

Fig. 3.

Glycolytic stress test. Extracellular acidification (ECAR) trace (A) and bar graphs (B) illustrate the glycolytic response of newborn cardiomyocytes to glucose, oligomycin, and 2-deoxyglucose from CD-CB, CD-STZ, HF-CB, and HF-STZ offspring. Diabetes-exposed offspring had lower basal, glucose-stimulated, and oligomycin-stimulated ECAR regardless of diet exposure. Data are expressed as means ± SE; n = 5–7 litters/group. ^±P < 0.05 by 2-way ANOVA.

Fig. 4.

Proton production rate (PPR). A: calculated (43) PPR for neonatal cardiomyocytes under basal, glucose-stimulated, oligomycin-stimulated, and 2-deoxyglucose-inhibited conditions is represented for CD-CB, CD-STZ, HF-CB, and HF-STZ offspring. B and C: the contributions of both lactate from anaerobic glycolysis and CO₂ production from cellular respiration during glucose stimulation (B) and oligomycin stimulation (C) confirm a decreased glycolytic capacity in diabetes-exposed neonatal cardiomyocytes. Diabetes exposure was associated with lower lactate production from anaerobic glycolysis under glucose stimulation and with lower CO₂ production from aerobic respiration under oligomycin stimulation consistent with impaired cardiac fuel flexibility. Data are expressed as means ± SE; n = 5–7 litters/group. ^±P < 0.05 by 2-way ANOVA.

Fig. 5.

Palmitate oxidation test. OCR trace (A) and bar graphs depict mitochondrial respiration at basal (B) and palmitate-stimulated (C) rates from CD-CB, CD-STZ, HF-CB, and HF-STZ offspring. Etomoxir, a carnitine-fatty acid transport inhibitor, was given and repeated to ensure maximal inhibition of exogenous fatty acid oxidation. Diabetes-exposed neonatal cardiomyocytes had lower basal (B) and palmitate-stimulated (C) OCR. In HF-STZ offspring maximal OCR remains lower than all other groups despite a response to exogenous palmitate (A). Data are expressed as means ± SE; n = 5–7 litters/group. ^±P < 0.05 by 2-way ANOVA.

Fig. 6.

Lipid peroxidation in newborn cardiomyocytes. Malondialdehyde (MDA) levels, as a marker of oxidative stress, were higher in cardiomyocytes from offspring exposed to a maternal HF diet (HF-CB and HF-STZ) but not diabetes alone (CD-STZ). Data are expressed as means ± SE; n = 4 or 5 litters/group. *P ≤ 0.0001 by 2-way ANOVA.

Fig. 7.

Mitochondrial histology. Isolated neonatal cardiomyocytes from CD-CB, CD-STZ, HF-CB, and HF-STZ offspring were stained with MitoTracker (mitochondrial identification), tetramethylrhodamine ethyl ester (TMRE, mitochondrial membrane potential/charged mitochondria), and Hoechst (DNA). When images are merged, healthy mitochondria appear long, thin, and well polarized (yellow). Combination-exposed (HF-STZ) mitochondria are fragmented and depolarized (green). Inset shows a magnified view of individual mitochondria for each group (arrowheads).

Fig. 8.

Mitochondrial copy number. Relative mitochondrial DNA copy number was determined by real-time PCR quantitation of mitochondrion-specific cytochrome-c oxidase I (MT-CO1; A) and the control region (D-loop; B) of rat mitochondrial DNA for CD-CB, CD-STZ, HF-CB, and HF-STZ offspring. For both genes, a significant interaction effect was noted by 2-way ANOVA, and all exposed groups had significant differences compared with control by 1-way ANOVA and Dunnett's posttest. Mitochondrial DNA copy number was decreased in diabetes-exposed hearts and increased in HF diet-exposed hearts. Values are means ± SE; n = 16 pups/group. Significant differences: *dietary effect, ^±diabetes effect, ^interaction effect (P ≤ 0.05).