Long-Term Gestational Hypoxia Modulates Expression of Key Genes Governing Mitochondrial Function in the Perirenal Adipose of the Late Gestation Sheep Fetus

Abstract

We previously reported that long-term hypoxia (LTH) increases expression of brown adipose tissue (BAT) genes in the perirenal adipose in the ovine fetus. The mechanisms with which hypoxia mediates the enhanced BAT phenotype are unresolved. This study was designed to examine the effects of LTH on (1) the expression of endothelial cell nitric oxide synthase (eNOS) and (2) indicators of mitochondrial biogenesis (transcription factors mitochondrial transcription factor A (mtTFA), nuclear respiratory factor (NRF) 1, and NRF-2; cytochrome c oxidase (COX) I, II, and IV and mitochondrial DNA content). Pregnant ewes were maintained at high altitude (3820 m) from ∼40 to 137 to 140 days of gestation and perirenal adipose was collected from normoxic control and LTH fetuses. There was no effect of LTH on fetal body weight or perirenal adipose mass. Long-term hypoxia increased (P < .05) perirenal eNOS and phospho-eNOS, messenger RNA (mRNA) for NRF1, NRF-2, mtTFA as well as COX-I, COX-II, and COX-IV mRNA. In contrast, mRNA for 2 markers for cellular proliferation (Ki67 and proliferating cell nuclear antigen [PCNA]) was lower in perirenal adipose from LTH fetuses compared to controls (P < .05), while mitochondrial to nuclear DNA ratio did not differ between groups. In conclusion, nitric oxide may function as a mechanism via which LTH enhances the BAT phenotype in fetal sheep prior to birth. Although there is an apparent increase in genes supporting mitochondrial function and adaptive thermogenesis in response to LTH, there does not appear to be an increased mitochondrial biogenesis per se. Such adaptive changes may provide a mechanism for the prominence of the BAT phenotype observed in the late gestation LTH fetus.

Keywords: adipose; fetus; hypoxia; mitochondria; ovine; sheep.

Figure 1.

Quantity of mRNA for eNOS (1A), iNOS (1B), and nNOS (1C) in perirenal adipose of late gestation control (CONT) and long-term hypoxic (LTH) fetal sheep. eNOS mRNA was approximately 10-fold more abundant compared to iNOS or nNOS. Both eNOS and iNOS mRNA were significantly elevated (P < .05) in the perirenal adipose of the LTH fetus. (n = 6 per group; mean ± SEM). eNOS indicates endothelial cell nitric oxide synthase; mRNA, messenger RNA; SEM, standard error of the mean.

Figure 2.

Western analysis for eNOS and phosphorylated serine 1177 eNOS in long-term hypoxic (LTH) and control (CONT) late gestation fetal sheep. Both eNOS and phosphorylated eNOS were significantly (P < .05) higher in the perirenal adipose of LTH fetuses. A representative sample for controls and LTH for both eNOS and phospho-eNOS are provided for comparison. eNOS indicates endothelial cell nitric oxide synthase.

Figure 3.

Quantity of mRNA for mtTFA, NRF-1, and NRF-2 (Figure 3A) in perirenal adipose of late gestation control (CONT) and long-term hypoxic (LTH) fetal sheep. All 3 transcription factors were significantly elevated (*P < .05) in the perirenal adipose of the LTH fetus. Messenger RNA for the housekeeping gene, cyclophilin (Figure 3B), was not different between the 2 groups. (n = 6 per group; mean ± SEM). mtTFA indicates mitochondrial transcription factor A; NRF, nuclear respiratory factor; SEM, standard error of the mean.

Figure 4.

Quantity of mRNA for COX-I (A), COX-2 (B), and COX-IV (C) in perirenal adipose of late gestation control (CONT) and long-term hypoxic (LTH) fetal sheep. All 3 cytochrome oxidases were significantly elevated (*P < .05) in the perirenal adipose of the LTH fetus. (n = 6 per group; mean ± SEM). COX indicates cytochrome c oxidase; SEM, standard error of the mean.

Figure 5.

Chromosomal DNA content (μg DNA/mg tissue) in perirenal adipose obtained from control (CONT) and long-term hypoxic (LTH) late gestation fetal sheep (A). Chromosomal DNA content was significantly (*P < .05) lower in the perirenal fat of LTH fetuses indicative of decreased cellularity. Mitochondrial and chromosomal DNA concentrations were not different in the perirenal adipose of CONT and LTH fetuses (5B) nor were the ratios of mitochondrial (mt) to chromosomal DNA (C), indicative of similar mitochondrial numbers in the 2 groups. (n = 6 per group; mean ± SEM). SEM indicates standard error of the mean.

Figure 6.

Quantity of mRNA for 2 markers for cell division (Ki67 and PCNA) were significantly lower (*P < .05) in the perirenal adipose of the late gestation LTH fetus (n = 6 per group; mean ± SEM). mRNA indicates messenger RNA; SEM, standard error of the mean; PCNA, proliferating cell nuclear antigen.

Figure 7.

Hematoxylin and eosin staining of perirenal adipose tissue from control and long-term hypoxic (LTH) fetuses (A; 20× magnification). No differences were observed in the number of dominant locules by locular area (μmol/L²; B) or percentage of total locular area (C) between control and LTH perirenal adipose samples.