Gestational disruptions in metabolic rhythmicity of the liver, muscle, and placenta affect fetal size

Abstract

Maternal metabolic adaptations are essential for successful pregnancy outcomes. We investigated how metabolic gestational processes are coordinated, whether there is a functional link with internal clocks, and whether disruptions are related to metabolic abnormalities in pregnancy, by studying day/night metabolic pathways in murine models and samples from pregnant women with normally grown and large-for-gestational age infants. In early mouse pregnancy, expression of hepatic lipogenic genes was up-regulated and uncoupled from the hepatic clock. In late mouse pregnancy, rhythmicity of energy metabolism-related genes in the muscle followed the patterns of internal clock genes in this tissue, and coincided with enhanced lipid transporter expression in the fetoplacental unit. Diurnal triglyceride patterns were disrupted in human placentas from pregnancies with large-for-gestational age infants and this overlapped with an increase in BMAL1 expression. Metabolic adaptations in early pregnancy are uncoupled from the circadian clock, whereas in late pregnancy, energy availability is mediated by coordinated muscle-placenta metabolic adjustments linked to internal clocks. Placental triglyceride oscillations in the third trimester of human pregnancy are lost in large-for-gestational age infants and may be regulated by BMAL1. In summary, disruptions in metabolic and circadian rhythmicity are associated with increased fetal size, with implications for the pathogenesis of macrosomia.-Papacleovoulou, G., Nikolova, V., Oduwole, O., Chambers, J., Vazquez-Lopez, M., Jansen, E., Nicolaides, K., Parker, M., Williamson, C. Gestational disruptions in metabolic rhythmicity of the liver, muscle, and placenta affect fetal size.

Keywords: circadian clock; macrosomia; metabolism; pregnancy; triglycerides.

Figure 1.

Serum lipid oscillations during the LDC in mouse pregnancy. Serum from d 7 and d 14 pregnant and nonpregnant female mice was assessed for total cholesterol (A), FFAs (B), and triglycerides (C). Data are means ± sem (n ≥ 5 per group per time point). ^a–cP < 0.05; nonpregnant (a), d 7 (b), d 14 (c) for fluctuations during LDC within the same stage of pregnancy. ^*P < 0.05 for d 7 vs. 14, ^#P < 0.05 for nonpregnant vs. d 7, ^$P < 0.05 for nonpregnant vs. d 14 for comparisons at the same ZT point in different stages of pregnancy.

Figure 2.

Metabolic and circadian gene expression and endogenous FFA levels of pregnancy during the LDC in the liver. A) Hepatic transcriptional profile of early pregnant (d 7), late pregnant (d 14), and nonpregnant control mice for Fas, Scd2, Hmgcr, Ppara, and Cpt1a genes. B) Endogenous FFA levels in the liver. C) Gene expression patterns of clock genes. Data are means ± sem (n ≥ 5 per group per time point). ^a–cP < 0.05; nonpregnant (a), d 7 (b), d 14 (c) for fluctuations during LDC within the same stage of pregnancy. ^*P < 0.05 for d 7 vs. 14, ^#P < 0.05 for nonpregnant vs. d 7, ^$P < 0.05 for nonpregnant vs. d 14 for comparisons at the same ZT point in different stages of pregnancy.

Figure 3.

Metabolic and circadian gene expression and endogenous FFA levels of pregnancy during the LDC in the muscle. A) Transcriptional profile of the energy homeostasis genes Cpt1b, Fabp3, and Pdk4 in muscle of early pregnant (d 7), late pregnant (d 14), and nonpregnant control mice. Endogenous FFA levels (B). Gene expression of clock genes (C). Data are means ± sem (n ≥ 5 per group per time point). ^a–cP < 0.05; nonpregnant (a), d 7 (b), d 14 (c) for fluctuations during LDC within the same stage of gregnancy. ^*P < 0.05 for d 7 vs. 14, ^#P < 0.05 for nonpregnant vs. d 7, ^$P < 0.05 for nonpregnant vs. d 14 for comparisons at the same ZT point in different stages of pregnancy.

Figure 4.

Transplacental nutrient transport during the LDC. A) FFA and TG concentrations in placenta and fetal liver on d 14 of pregnancy. B) Gene expression profile of lipases and fatty acid transport on d 14 of pregnancy. C) Gene expression of clock genes during LDC in placenta. Data are means ± sem (n ≥ 5 per group per time point). ^{a, b}P < 0.05; fluctuations of FFA (a) and TG (b) during LDC.

Figure 5.

Triglyceride levels and clock gene expression patterns during the day in human pregnancy. A) Triglyceride levels in the serum of pregnant and nonpregnant women after an overnight fast followed by a standardized high-calorie meal. Mean gestational age, 33.1 ± 1.13 wk. Data are means ± sem. *P < 0.05 for fluctuations during the day; ^#P < 0.05 for pregnant vs. nonpregnant women. B) Diurnal fluctuations of triglyceride levels in normal pregnancy are not maintained in LGA pregnancy. C) Clock gene mRNA expression profile in human placenta. BMAL1 (left) has increased mRNA levels in LGA pregnancy compared to controls. No changes were observed in CLOCK (right) or PER1 (bottom) mRNA. Data are means ± sem (n = 4–8 per group per time point). *P < 0.05 for fluctuations during the day; ^#P < 0.05 for differences in gene expression levels.

Figure 6.

Daily rhythms in circadian and metabolic processes in pregnancy. A) The liver–muscle–placental gestational switch. Metabolic adaptations are tightly programmed in mouse pregnancy. Hepatic genes involved in metabolic processes show constantly higher expression levels on d 7 of pregnancy, followed by a drop in gene expression levels on d 14. Day 7 hepatic metabolism is uncoupled from the circadian clock (represented by the melted clock image), whereas on d 14 hepatic genes exhibit rhythmicity during the LDC, consistent with negative-feedback oscillations of Rev-erb-a and Rev-erb-b mRNA. Muscle appears to coordinate energy availability for transfer in the fetoplacental unit on d 14 of pregnancy, with lower gene expression levels and absence of rhythmicity on d 7 of pregnancy. The switching between d 7 and 14 in the muscle is regulated by Rev-erb-b. Muscle activities coincide with a peak of TG/FA levels and lipid transport genes in the fetoplacental unit from ZT12 onward, consistent with a peak expression of placental clock genes toward the end of the light phase or during the dark phase. TG, triglycerides; FA, fatty acids. B) Placental lipid homeostasis in human pregnancy. Despite the absence of placental rhythmicity in both early (CVS) and term pregnancies, diurnal fluctuations of triglycerides during the day of normal pregnancy are lost in pregnancies with LGA infants where triglycerides are consistently increased. The melted-clock image denotes uncoupling of metabolic actions from the circadian clock machinery, whereas the normal light and dark phase clocks represent synchronization of metabolic responses with the circadian clocks.