Obesity-induced changes in hepatic and placental clock gene networks in rat pregnancy

Obesity-induced changes in hepatic and placental clock gene networks in rat pregnancy Abstract Maternal obesity induces pregnancy complications and disturbs fetal development, but the specific mechanisms underlying these outcomes are unclear. Circadian rhythms are implicated in metabolic complications associated with obesity, and maternal metabolic adaptations to pregnancy. Accordingly, obesity-induced circadian dysfunction may drive adverse outcomes in obese pregnancy. This study investigated whether maternal obesity alters the rhythmic expression of clock genes and associated nuclear receptors across maternal, fetal, and placental tissues. Wistar rats were maintained on a cafeteria (CAF) diet prior to and throughout gestation to induce maternal obesity. Maternal and fetal liver and placental labyrinth zone (LZ) were collected at four-hourly time points across days 15–16 and 21–22 of gestation (term = 23 days). Gene expression was analyzed by RT-qPCR. Expression of the accessory clock gene Nr1d1 was rhythmic in the maternal and fetal liver and LZ of chow-fed controls, but in each case CAF feeding reduced peak Nr1d1 expression. Obesity resulted in a phase advance (approx. 1.5 h) in the rhythms of several clock genes and Ppar-delta in maternal liver. Aside from Nr1d1, expression of clock genes was mostly arrhythmic in LZ and fetal liver, and was unaffected by the CAF diet. In conclusion, maternal obesity suppressed Nr1d1 expression across maternal, fetal, and placental compartments and phase-advanced the rhythms of maternal hepatic clock genes. Given the key role of Nr1d1 in regulating metabolic, vascular, and inflammatory processes, our data suggest that disruptions to rhythmic Nr1d1 expression in utero may contribute to programmed health complications in offspring of obese pregnancies. Introduction Maternal obesity complicates approximately 20% of pregnancies in Western countries [1] and leads to a range of adverse maternal and fetal health outcomes. Obese women are more likely to experience pregnancy complications such as gestational diabetes and miscarriage [2, 3], and their infants are at a heightened risk for either macrosomia or intrauterine growth restriction [2, 4]. This disturbed development in utero can have lifelong health consequences, since offspring born to obese pregnancies often experience programmed metabolic disease in adulthood [5, 6]. While it is clear that obese pregnancy has negative effects on maternal and offspring health, the underlying mechanisms remain poorly understood. Circadian rhythms are endogenous timing processes that have evolved in response to the light-dark cycle in order to optimize the timing and efficiency of biological processes in effectively all living organisms. While the circadian system regulates daily oscillations in a range of physiological and behavioral functions, it is particularly crucial in the regulation of metabolism [7, 8]. Analysis of such oscillations is facilitated by the use of cosinor regression, whereby the characteristics of rhythms can be readily quantitated and thus statistically compared. At a molecular level, circadian rhythms are driven by positive and negative transcriptional feedback loops of clock genes (clock circadian regulator (Clock), aryl hydrocarbon receptor nuclear translocator like (gene symbol alias: Bmal1) (Arntl), period circadian regulator 1-3 (Per1-3), cryptochrome circadian regulator 1-2 (Cry1-2), nuclear receptor subfamily 1 group D member 1 (Nr1d1), and RAR-related orphan receptor A (gene symbol alias: Rora) (Nr1f1)) which then influence downstream metabolic pathways [9]. Since obesity is widely considered a metabolic disease, the circadian system is likely a key regulator of obesity-related patho-logies. Indeed, clock gene knockout models display altered metabolic phenotypes, including obesity and disturbed glucose metabolism [10, 11], while circadian disruption (e.g., shift work) leads to obesity and metabolic disease [12, 13]. Moreover, diet-induced obesity disturbs hepatic clock gene expression, and thereby alters expression of downstream genes and associated hepatic function [14–16]. Thus, there appears to be a reciprocal relationship between obesity and circadian dysfunction. Despite this evidence for obesity-induced changes to clock gene expression in the nonpregnant state, it is not known whether maternal obesity elicits similar effects during pregnancy. Maternal circadian rhythms change markedly across gestation [17], which likely facilitates the maternal metabolic adaptations required to sustain fetal growth. Moreover, these circadian adaptations appear essential for normal fetal development, since circadian disruption during pregnancy causes maternal and fetal metabolic dysfunction and programs adverse metabolic outcomes in offspring [18, 19]. Given a functional circadian system appears vital for healthy pregnancy outcomes, compromised circadian clocks could be key contributors to obesity-induced pregnancy complications. This may include not only maternal tissue clocks, but also those in the placental and fetal tissues, which also express clock genes [20]. Consequently, this study investigated the circadian expression profiles of clock genes and key downstream regulators of glucose and lipid metabolism (peroxisome proliferator activated receptor (Ppars) and solute carrier family 2 member 2 (previous gene symbol: Glut2) (Slc2a2)) in the maternal liver, placental labyrinth zone (LZ; the site of maternal–fetal exchange), and fetal liver in rat pregnancy. Maternal obesity was established by cafeteria (CAF) feeding [21], and tissues were analyzed across days 15–16 and 21–22 of gestation (term = 23 days), thus spanning the period of maximal fetal growth. It was hypothesized that maternal obesity alters rhythmicity of clock gene expression in hepatic and placental tissues, and that this has related effects on downstream metabolic genes. Materials and methods Animals and diets All animal procedures were approved by the Animal Ethics Committee of The University of Western Australia. Three-week-old albino Wistar rats were obtained from the Animal Resources Centre (Murdoch, Western Australia) and housed three to a cage. Animals were kept under a 12:12 h light-dark cycle at 22°C, with ad libitum access to water and standard rodent chow (14 KJ/g total energy; 12% energy as fat, 23% protein and 65% carbohydrate; Specialty Feeds, Glen Forrest, Western Australia). Following a week of acclimatization, animals were separated into control (CON) or CAF diet groups; CON animals were maintained on standard chow, while CAF animals were provided with four snack food items per day, in addition to chow. CAF items were given in excess to be essentially ad libitum and were changed daily at approximately ZT 11-12 (from a selection of 17 items) to maintain novelty. This resulted in an average energy density of 16.4 KJ/g (47% energy as fat, 8% as protein, and 44% as carbohydrate). For a detailed list of CAF diet constituents, see Crew et al. [21]. After 8 weeks of diet exposure, animals were mated overnight after determination of proestrus by an estrous cycle monitor (EC40; Fine Science Tools, Vancouver, British Columbia, Canada). The male rats were age-matched to the females, chow-fed, and maintained in identical environmental conditions to ensure minimal variation from paternal factors. Day 1 of pregnancy was confirmed by presence of spermatozoa in a vaginal smear the following morning. Pregnant animals were separated to individual housing and maintained on their respective diets throughout pregnancy. The timing of each tissue sampling is expressed relative to Zeitgeber time zero (ZT0), i.e., when lights were turned on in the animal facility (0700 h). Tissues were obtained at four-hourly intervals (ZT1, 0800 h; ZT5, 1200 h; ZT9, 1600 h; ZT13, 2000 h; ZT17, 2400 h; and ZT21, 0400 h) across days 15–16 and 21–22 of gestation (term occurs at day 23). At the appropriate collection time, animals (n = 7–8 per ZT for each diet) were anaesthetized using isoflurane/nitrous oxide. Maternal liver tissue was collected and fetal–placental pairs were removed via caesarean section. Placental LZ samples were isolated from whole placentas via blunt dissection, and liver tissue was collected from individual fetuses. Fetal sex was determined by amplification of the sex determining region Y (Sry) gene in fetal tail tissue at days 15–16 [22] or by anogenital distance at days 21–22 [23]. Although placental and fetal liver samples were collected for both sexes, only female tissues were analyzed for gene expression (one per litter). Notably, however, growth rates and steroid hormone responses to the CAF diet are similar in both sexes in this obesity model [24]. Maternal blood was taken from the descending aorta, and fetal blood was pooled from decapitated female fetuses within each litter at day 21. Blood glucose was measured immediately in maternal and fetal blood samples (Accu-Chek blood glucose monitor; Roche Diagnostics, Mannheim, Germany). Blood samples were then mixed with 10:1 (vol:vol) 0.6 M EDTA and centrifuged at 13 000× g for 6 min to isolate plasma. All collected tissue and plasma samples were snap frozen in liquid nitrogen and stored at –80°C until further analysis. Insulin measurement Insulin levels were measured in maternal and day 21 fetal plasma using a Milliplex MAP Rat Adipokine Magnetic Bead Panel (Cat. # RECYTMAG-65K; EMD Millipore Corporation, Billerica, MA, USA). The assay was performed according to the manufacturer's instructions and the plate was read on a Luminex Magpix (Luminex Corporation, Austin, Texas, USA). Data were analyzed with Magpix 4.2 Software. The intra-assay coefficient of variation was 8.4% and all quality controls provided with the Milliplex kit performed within expectations. RNA and cDNA sample preparation Total RNA was extracted from 50 to 100 mg of individual tissue samples using the QIAzol method (Qiagen Sciences, Maryland, USA), according to the manufacturer's instructions. RNA was assessed for concentration and purity with the Nanodrop ND-1000 spectrophotometer (Thermo Scientific, Wilmington, DE, USA) and 5 μg of RNA was reverse transcribed to cDNA by the mouse Moloney leukemia virus reverse transcriptase RNase H Point Mutant with random hexamers (Promega, Sydney, Australia) as previously reported [25]. The resultant cDNA was purified using the UltraClean PCR Cleanup Kit (MoBio Laboratories, Carlsband, CA, USA). Quantitative PCR The relative mRNA expression of clock genes (Clock, Arntl, Per1, Per2, Per3, Cry1, Cry2, Nr1f1, and Nr1d1) and genes that influence glucose and lipid metabolism (Ppar-alpha, Ppar-delta, Pgc1-alpha, Slc2a2) were analyzed via RT-qPCR on the Rotorgene Q (Qiagen, Hilden, Germany). All primers (see Table 1) were designed using Primer-BLAST [26] and were positioned to span an intron to prevent amplification from genomic DNA. Table 1. Primer details and PCR conditions for clock, metabolic, and reference genes. Gene  Primer sequence  Annealing Temp (°C)  Size (bp)  MgCl2 (mM)  Clock genes  Clock  F5΄ ACAGCGCACACACAGGCCTTC 3΄  60  175  2    R5΄ TGGCGGCGCCCTGTGATCTA 3΄        Arntl  F5΄ ACACTGCACCTCGGGAGCGA 3΄  60  100  2    R5΄ CGCCGAGCTCCAGAGCACAA 3΄        Per1  F5΄ CGCACTTCGGGAGCTCAAACTTC 3΄  60  169  2    R5΄ GTCCATGGCACAGGGCTCACC 3΄        Per2  F5΄ TGAGCTCCTTGGCGTTGCCG 3΄  60  147  2    R5΄ ACTCAGGCCCACTGGCCACA 3΄        Per3  F5΄ TTTTCCCCTTCAAGACATGG 3΄  60  167  2    R5΄ GAAAGAGAGGGCTGTTGTGC 3΄        Cry1  F5΄ AGCTGGCCACTGAGGCTGGT 3΄  60  158  2    R5΄ TGCTGGCATCTCCAGGGGCT 3΄        Cry2  F5΄ CTGCCCAGGAGCCACCAAGC 3΄  60  192  2    R5΄ GCATGCACACGCAAACGGCA 3΄        Nr1d1  F5΄ ATTGCCCACGGGGCGAGAGA 3΄  60  292  2    R5΄ GCCAAAAGAGCGGGCAGGGT 3΄        Nr1f1  F5΄ CCCAACCGTGTCCATGGCGG 3΄  60  113  2    R5΄ CCCGTCGATGCGTTTGGCGA 3΄        Ppars and Slc2a2  Ppar-alpha  F5΄ AATCCACGAAGCCTACCTGA 3΄  60  132  2.5    R5΄ GTCTTCTCAGCCATGCACAA 3΄        Ppar-delta  F5΄ GAGGGGTGCAAG GGCTTCTT 3΄  60  101  2.5    R5΄ CACTTGTTGCGGTTCTTCTTCTG 3΄        Pgc1-alpha  F5΄ TCTGGAACTGCAGGCCTAACTC 3΄  60  96  4    R5΄ GCAAGAGGGCTTCAGCTTTG 3΄        Slc2a2  F5΄ TAGGCGGAATGGTCGCCTCGT 3  61  102  2    R5΄ GGGCTCCAGTCAACGAGAGGCT 3΄        Reference genes  Ppia  F5΄ AGCATACAGGTCCTGGCATC 3΄  62  127  3    R5΄ TTCACCTTCCCAAAGACCAC 3΄        Sdha  F5΄ TGGGGCGACTCGTGGCTTTC 3΄  60  134  2    R5΄CCCCGCCTGCACCTACAACC 3΄        Ywhaz  F5΄ GACGGAGCTGAGGGACATCTGC 3΄  60  75  2    R5΄ GGCTGCGAAGCATTGGGGATCA 3΄        Gene  Primer sequence  Annealing Temp (°C)  Size (bp)  MgCl2 (mM)  Clock genes  Clock  F5΄ ACAGCGCACACACAGGCCTTC 3΄  60  175  2    R5΄ TGGCGGCGCCCTGTGATCTA 3΄        Arntl  F5΄ ACACTGCACCTCGGGAGCGA 3΄  60  100  2    R5΄ CGCCGAGCTCCAGAGCACAA 3΄        Per1  F5΄ CGCACTTCGGGAGCTCAAACTTC 3΄  60  169  2    R5΄ GTCCATGGCACAGGGCTCACC 3΄        Per2  F5΄ TGAGCTCCTTGGCGTTGCCG 3΄  60  147  2    R5΄ ACTCAGGCCCACTGGCCACA 3΄        Per3  F5΄ TTTTCCCCTTCAAGACATGG 3΄  60  167  2    R5΄ GAAAGAGAGGGCTGTTGTGC 3΄        Cry1  F5΄ AGCTGGCCACTGAGGCTGGT 3΄  60  158  2    R5΄ TGCTGGCATCTCCAGGGGCT 3΄        Cry2  F5΄ CTGCCCAGGAGCCACCAAGC 3΄  60  192  2    R5΄ GCATGCACACGCAAACGGCA 3΄        Nr1d1  F5΄ ATTGCCCACGGGGCGAGAGA 3΄  60  292  2    R5΄ GCCAAAAGAGCGGGCAGGGT 3΄        Nr1f1  F5΄ CCCAACCGTGTCCATGGCGG 3΄  60  113  2    R5΄ CCCGTCGATGCGTTTGGCGA 3΄        Ppars and Slc2a2  Ppar-alpha  F5΄ AATCCACGAAGCCTACCTGA 3΄  60  132  2.5    R5΄ GTCTTCTCAGCCATGCACAA 3΄        Ppar-delta  F5΄ GAGGGGTGCAAG GGCTTCTT 3΄  60  101  2.5    R5΄ CACTTGTTGCGGTTCTTCTTCTG 3΄        Pgc1-alpha  F5΄ TCTGGAACTGCAGGCCTAACTC 3΄  60  96  4    R5΄ GCAAGAGGGCTTCAGCTTTG 3΄        Slc2a2  F5΄ TAGGCGGAATGGTCGCCTCGT 3  61  102  2    R5΄ GGGCTCCAGTCAACGAGAGGCT 3΄        Reference genes  Ppia  F5΄ AGCATACAGGTCCTGGCATC 3΄  62  127  3    R5΄ TTCACCTTCCCAAAGACCAC 3΄        Sdha  F5΄ TGGGGCGACTCGTGGCTTTC 3΄  60  134  2    R5΄CCCCGCCTGCACCTACAACC 3΄        Ywhaz  F5΄ GACGGAGCTGAGGGACATCTGC 3΄  60  75  2    R5΄ GGCTGCGAAGCATTGGGGATCA 3΄        F: Forward primer, R: Reverse primer, bp: base pairs. View Large Each PCR reaction consisted of 10× Immolase buffer, 0.5 U Immolase DNA Polymerase (Bioline, Alexandria, NSW, Australia), 10 mM dNTPs, SYBR green (Molecular Probes, Eugene, OR, USA) at a 1/2000 dilution in DMSO, and gene-specific concentrations (see Table 1) of MgCl2 and forward and reverse primer. Complementary DNA template (1 μl) was added to give a total reaction volume of 10 μl. The PCR reaction cycle included an initial denaturing stage at 95°C for 10 min, followed by 45 cycles each comprising of 95°C for 1 s, a primer-specific annealing temperature (see Table 1) for 15 s and a 72°C extension period for 5 s. Samples were run in duplicate and no template controls were included in all PCR runs. Standard curves were generated in each PCR run from serial 10-fold dilutions of gel-extracted PCR product and were used to calculate the relative gene expression concentrations for each sample with Rotorgene Q series software. Common samples were included in each plate for comparison between PCR runs. All values were standardized against the reference genes peptidylprolyl isomerase A (Ppia), succinate dehydrogenase complex flavoprotein subunit A (Sdha), and tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein zeta (Ywhaz) using the GeNorm algorithm [27]. Statistical analysis All values are expressed as mean ± SEM, with n = 7–8 mothers per ZT group for each diet. Daily rhythms of gene expression or plasma hormone levels were assessed by nonlinear (cosinor) regression in Genstat 9.0 software (Hemel Hempstead, UK). This calculated the typical rhythm features of mesor (i.e., the rhythm-adjusted mean), amplitude (the difference between the mesor and peak), and acrophase (the time of the rhythm peak) for each daily profile. Comparisons of these cosine curve features between diet groups and gestational days were made by two-way analysis of variance (ANOVA) or t-test, as appropriate (GraphPad Prism version 6.00; La Jolla, California). Other comparisons were made by ANOVA using GenStat 9.0 software (Hemel Hempstead, UK) to test for differences due to diet, time of day, or stage of pregnancy. When significant (P < 0.05) interactions were observed between factors, differences were assessed by t-test or ANOVA, as appropriate. When the F-test reached statistical significance (P < 0.05), specific group comparisons were made by least significant difference (LSD) tests. Results Caloric intake, weight gain profiles, body composition, and pregnancy data (litter sizes and pregnancy success rates) for this animal cohort has been previously reported [24]. Briefly, CAF animals exhibited increased caloric intake and weight gain across the study period; this resulted in a 25% increase in body weight after 8 weeks of pre-pregnancy CAF feeding, and a 56% increase in total percentage adiposity in CAF mothers by day 21 of gestation. Fetal and placental weights were slightly reduced in the CAF group across both days of gestation, but total fetal and placental weight per mother did not differ between groups due to a small increase in litter size in the CAF animals. Profiles of blood glucose and plasma insulin As expected, maternal blood glucose levels were relatively stable across each of days 15 and 21, although a significant, low-amplitude rhythm became evident at day 21 (P = 0.006). There was also a Diet × Stage of Pregnancy interaction, with elevated blood glucose evident in CAF mothers at day 15 (P < 0.001; Figure 1A) but not at day 21. Fetal blood glucose levels increased 2.4-fold across day 21 (P < 0.001), but were unaffected by diet (Figure 1A). Figure 1. View largeDownload slide Daily profiles of blood glucose and plasma insulin in maternal and fetal plasma of control (CON) and cafeteria (CAF) diet groups. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT) and data for each time point were derived from separate groups of animals. Note different scales for fetal blood glucose and plasma insulin. *P < 0.001 CON vs. CAF (overall diet effect in two-way ANOVA); †P < 0.05 compared to corresponding CON; ††P < 0.01 compared to corresponding CON (t-test following Diet × Time of Day interaction in two-way ANOVA). #P < 0.05 compared to peak ZT value for that diet and day (one-way ANOVA following Diet × Time of Day interaction in two-way ANOVA). Figure 1. View largeDownload slide Daily profiles of blood glucose and plasma insulin in maternal and fetal plasma of control (CON) and cafeteria (CAF) diet groups. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT) and data for each time point were derived from separate groups of animals. Note different scales for fetal blood glucose and plasma insulin. *P < 0.001 CON vs. CAF (overall diet effect in two-way ANOVA); †P < 0.05 compared to corresponding CON; ††P < 0.01 compared to corresponding CON (t-test following Diet × Time of Day interaction in two-way ANOVA). #P < 0.05 compared to peak ZT value for that diet and day (one-way ANOVA following Diet × Time of Day interaction in two-way ANOVA). Maternal hyperinsulinemia was evident in the CAF group across all time points at day 15 (P < 0.001; Figure 1B), whereas at day 21 there was a significant Diet × Time of Day interaction, with hyperinsulinemia evident in CAF mothers only at ZT13 (P = 0.02; Figure 1B). Fetal insulin varied with time of day across day 21 (P = 0.01) but was unaffected by diet (Figure 1B). Effects of the CAF diet on maternal, placental, and fetal clock gene expression Maternal hepatic clock gene expression The expression profiles of all clock genes in the maternal liver were clearly rhythmic (i.e., significant cosinor fit) at both gestational ages, the single exception being Nr1f1 at day 15, where rhythmicity was abolished by the CAF diet (see Table 2 for r2 and P values for cosine fit significance). Obesity reduced the mesor (i.e., the overall average expression across the full day) of Cry2 at both days of gestation (P < 0.05; Figure 2G and Table 3), and while the Per1 mesor appeared to be reduced in CAF mothers at day 15, this did not reach statistical significance (P = 0.08; Table 3). The CAF diet also reduced the amplitude (i.e., the difference between the mesor and peak expression level) of Per2 at both days, and those of Cry2 and Nr1d1 at day 21 (P < 0.05; Figure 2 and Table 3). Interestingly, the CAF diet induced a phase advance of approximately 1.2 h in the maternal hepatic expression of Arntl, Per2, Per3 at both days of gestation, Cry1 at day 15 and Nr1d1 at day 21 (all P < 0.05; Figure 2 and Table 3). Figure 2. View largeDownload slide Rhythmic expression profiles of maternal hepatic clock genes in control (CON) and cafeteria (CAF) diet groups at days 15 and 21 of gestation. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. The inset graph (different scale) is provided in panel I to highlight the shift in the Nr1d1 acrophase. Statistical differences for cosine curve features are summarized in Table 3. Figure 2. View largeDownload slide Rhythmic expression profiles of maternal hepatic clock genes in control (CON) and cafeteria (CAF) diet groups at days 15 and 21 of gestation. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. The inset graph (different scale) is provided in panel I to highlight the shift in the Nr1d1 acrophase. Statistical differences for cosine curve features are summarized in Table 3. Table 2. Cosinor rhythmicity (r2 and associated P values) for maternal hepatic gene expression profiles.   Maternal liver day 15  Maternal liver day 21    CON  CAF  CON  CAF  Clock  0.607 (P < 0.001)  0.419 (P < 0.001)  0.306 (P < 0.001)  0.284 (P < 0.001)  Arntl  0.925 (P < 0.001)  0.895 (P < 0.001)  0.880 (P < 0.001)  0.830 (P < 0.001)  Per1  0.602 (P < 0.001)  0.597 (P < 0.001)  0.553 (P < 0.001)  0.505 (P < 0.001)  Per2  0.826 (P < 0.001)  0.710 (P < 0.001)  0.789 (P < 0.001)  0.707 (P < 0.001)  Per3  0.825 (P < 0.001)  0.682 (P < 0.001)  0.842 (P < 0.001)  0.727 (P < 0.001)  Cry1  0.857 (P < 0.001)  0.841 (P < 0.001)  0.837 (P < 0.001)  0.763 (P < 0.001)  Cry2  0.486 (P < 0.001)  0.278 (P < 0.001)  0.733 (P < 0.001)  0.403 (P < 0.001)  Nr1d1  0.538 (P < 0.001)  0.600 (P < 0.001)  0.747 (P < 0.001)  0.777 (P < 0.001)  Nr1f1  0.271 (P < 0.001)  0.036 (NS)  0.161 (P = 0.006)  0.087 (P = 0.05)  Ppar-alpha  0.111 (P = 0.027)  0.025 (NS)  0.156 (P = 0.008)  0 (NS)  Ppar-delta  0.698 (P < 0.001)  0.559 (P < 0.001)  0.681 (P < 0.001)  0.578 (P < 0.001)  Pgc1-alpha  0 (NS)  0 (NS)  0.023 (NS)  0.07 (NS)  Slc2a2  0.542 (P < 0.001)  0.583 (P < 0.001)  0.684 (P < 0.001)  0.358 (P < 0.001)    Maternal liver day 15  Maternal liver day 21    CON  CAF  CON  CAF  Clock  0.607 (P < 0.001)  0.419 (P < 0.001)  0.306 (P < 0.001)  0.284 (P < 0.001)  Arntl  0.925 (P < 0.001)  0.895 (P < 0.001)  0.880 (P < 0.001)  0.830 (P < 0.001)  Per1  0.602 (P < 0.001)  0.597 (P < 0.001)  0.553 (P < 0.001)  0.505 (P < 0.001)  Per2  0.826 (P < 0.001)  0.710 (P < 0.001)  0.789 (P < 0.001)  0.707 (P < 0.001)  Per3  0.825 (P < 0.001)  0.682 (P < 0.001)  0.842 (P < 0.001)  0.727 (P < 0.001)  Cry1  0.857 (P < 0.001)  0.841 (P < 0.001)  0.837 (P < 0.001)  0.763 (P < 0.001)  Cry2  0.486 (P < 0.001)  0.278 (P < 0.001)  0.733 (P < 0.001)  0.403 (P < 0.001)  Nr1d1  0.538 (P < 0.001)  0.600 (P < 0.001)  0.747 (P < 0.001)  0.777 (P < 0.001)  Nr1f1  0.271 (P < 0.001)  0.036 (NS)  0.161 (P = 0.006)  0.087 (P = 0.05)  Ppar-alpha  0.111 (P = 0.027)  0.025 (NS)  0.156 (P = 0.008)  0 (NS)  Ppar-delta  0.698 (P < 0.001)  0.559 (P < 0.001)  0.681 (P < 0.001)  0.578 (P < 0.001)  Pgc1-alpha  0 (NS)  0 (NS)  0.023 (NS)  0.07 (NS)  Slc2a2  0.542 (P < 0.001)  0.583 (P < 0.001)  0.684 (P < 0.001)  0.358 (P < 0.001)  NS: not significant (P > 0.05). View Large Table 3. Rhythmic features (mesor, amplitude, and acrophase) of maternal hepatic gene expression profiles.   Day 15  Day 21    CON  CAF  CON  CAF  Clock  Mesor  100 ± 3  92 ± 4  95 ± 4  93 ± 5    Amplitude  39 ± 5  31 ± 5  24 ± 5  28 ± 7    Acrophase  0.2 ± 0.4  23.3 ± 0.7  23.1 ± 0.8  22.4 ± 0.8  Arntl  Mesor  100 ± 2  96 ± 3  84 ± 3†  80 ± 4†    Amplitude  98 ± 4  89 ± 5  82 ± 5†  79 ± 5†    Acrophase  0.1 ± 0.2  22.8 ± 0.2*  0.3 ± 0.2  23.1 ± 0.3*  Per1  Mesor  100 ± 6  85 ± 6  113 ± 8  110 ± 7    Amplitude  67 ± 8  59 ± 8  83 ± 11  62 ± 10    Acrophase  13.4 ± 0.4  13.1 ± 0.5  14.3 ± 0.5†  13.9 ± 0.5†  Per2  Mesor  100 ± 4  101 ± 4  90 ± 5  84 ± 5    Amplitude  86 ± 6  54 ± 5*  86 ± 7  66 ± 6*    Acrophase  16.3 ± 0.3  15.3 ± 0.4*  16.3 ± 0.3  15.3 ± 0.4*  Per3  Mesor  100 ± 6  103 ± 8  98 ± 5  100 ± 6    Amplitude  119 ± 8  108 ± 11  107 ± 7  90 ± 8    Acrophase  12.4 ± 0.3  11.2 ± 0.4*  12.9 ± 0.2  11.7 ± 0.3*  Cry1  Mesor  100 ± 3  95 ± 3  86 ± 3†  87 ± 4†    Amplitude  69 ± 4  69 ± 5  59 ± 4†  61 ± 5†    Acrophase  20.0 ± 0.2  18.9 ± 0.2*  19.5 ± 0.3  18.8 ± 0.3  Cry2  Mesor  100 ± 3  90 ± 4*  88 ± 3†  76 ± 3*,†    Amplitude  30 ± 5  22 ± 5  39 ± 4  25 ± 4*    Acrophase  14.9 ± 0.6  14.6 ± 0.9  15.6 ± 0.3  14.1 ± 07  Nr1d1  Mesor  100 ± 15  91 ± 12  29.7 ± 0.9†  25.8 ± 0.7†    Amplitude  157 ± 22  132 ± 17  43.3 ± 1†  34.1 ± 1*,†    Acrophase  8.9 ± 0.5  7.9 ± 0.5  10.0 ± 0.3†  8.5 ± 0.3*,†  Nr1f1  Mesor  100 ± 4  99 ± 7  101 ± 4  97 ± 5    Amplitude  27 ± 6  18 ± 10  17 ± 5  NS    Acrophase  20.9 ± 0.9  20.3 ± 2.0  19.8 ± 1.2  NS  Ppar-alpha  Mesor  100 ± 8  117 ± 9  190 ± 19  233 ± 24    Amplitude  30 ± 11  24 ± 14  NS  NS    Acrophase  12.1 ± 1.4  10.4 ± 2.2  NS  NS  Ppar-delta  Mesor  100 ± 7  83 ± 7  87 ± 5  84 ± 6    Amplitude  89 ± 11  71 ± 10  62 ± 7†  60 ± 8†    Acrophase  12.8 ± 0.4  11.3 ± 0.5*  13.5 ± 0.4  11.5 ± 0.5*  Pgc1-alpha  Mesor  100 ± 5  87 ± 5  158 ± 7  138 ± 8    Amplitude  NS  NS  NS  NS    Acrophase  NS  NS  NS  NS  Slc2a2  Mesor  100 ± 3  112 ± 5*  110 ± 3  120 ± 6    Amplitude  35 ± 5  50 ± 7  45 ± 5  43 ± 9    Acrophase  16.0 ± 0.5  15.2 ± 0.5  15.5 ± 0.4  14.8 ± 0.8    Day 15  Day 21    CON  CAF  CON  CAF  Clock  Mesor  100 ± 3  92 ± 4  95 ± 4  93 ± 5    Amplitude  39 ± 5  31 ± 5  24 ± 5  28 ± 7    Acrophase  0.2 ± 0.4  23.3 ± 0.7  23.1 ± 0.8  22.4 ± 0.8  Arntl  Mesor  100 ± 2  96 ± 3  84 ± 3†  80 ± 4†    Amplitude  98 ± 4  89 ± 5  82 ± 5†  79 ± 5†    Acrophase  0.1 ± 0.2  22.8 ± 0.2*  0.3 ± 0.2  23.1 ± 0.3*  Per1  Mesor  100 ± 6  85 ± 6  113 ± 8  110 ± 7    Amplitude  67 ± 8  59 ± 8  83 ± 11  62 ± 10    Acrophase  13.4 ± 0.4  13.1 ± 0.5  14.3 ± 0.5†  13.9 ± 0.5†  Per2  Mesor  100 ± 4  101 ± 4  90 ± 5  84 ± 5    Amplitude  86 ± 6  54 ± 5*  86 ± 7  66 ± 6*    Acrophase  16.3 ± 0.3  15.3 ± 0.4*  16.3 ± 0.3  15.3 ± 0.4*  Per3  Mesor  100 ± 6  103 ± 8  98 ± 5  100 ± 6    Amplitude  119 ± 8  108 ± 11  107 ± 7  90 ± 8    Acrophase  12.4 ± 0.3  11.2 ± 0.4*  12.9 ± 0.2  11.7 ± 0.3*  Cry1  Mesor  100 ± 3  95 ± 3  86 ± 3†  87 ± 4†    Amplitude  69 ± 4  69 ± 5  59 ± 4†  61 ± 5†    Acrophase  20.0 ± 0.2  18.9 ± 0.2*  19.5 ± 0.3  18.8 ± 0.3  Cry2  Mesor  100 ± 3  90 ± 4*  88 ± 3†  76 ± 3*,†    Amplitude  30 ± 5  22 ± 5  39 ± 4  25 ± 4*    Acrophase  14.9 ± 0.6  14.6 ± 0.9  15.6 ± 0.3  14.1 ± 07  Nr1d1  Mesor  100 ± 15  91 ± 12  29.7 ± 0.9†  25.8 ± 0.7†    Amplitude  157 ± 22  132 ± 17  43.3 ± 1†  34.1 ± 1*,†    Acrophase  8.9 ± 0.5  7.9 ± 0.5  10.0 ± 0.3†  8.5 ± 0.3*,†  Nr1f1  Mesor  100 ± 4  99 ± 7  101 ± 4  97 ± 5    Amplitude  27 ± 6  18 ± 10  17 ± 5  NS    Acrophase  20.9 ± 0.9  20.3 ± 2.0  19.8 ± 1.2  NS  Ppar-alpha  Mesor  100 ± 8  117 ± 9  190 ± 19  233 ± 24    Amplitude  30 ± 11  24 ± 14  NS  NS    Acrophase  12.1 ± 1.4  10.4 ± 2.2  NS  NS  Ppar-delta  Mesor  100 ± 7  83 ± 7  87 ± 5  84 ± 6    Amplitude  89 ± 11  71 ± 10  62 ± 7†  60 ± 8†    Acrophase  12.8 ± 0.4  11.3 ± 0.5*  13.5 ± 0.4  11.5 ± 0.5*  Pgc1-alpha  Mesor  100 ± 5  87 ± 5  158 ± 7  138 ± 8    Amplitude  NS  NS  NS  NS    Acrophase  NS  NS  NS  NS  Slc2a2  Mesor  100 ± 3  112 ± 5*  110 ± 3  120 ± 6    Amplitude  35 ± 5  50 ± 7  45 ± 5  43 ± 9    Acrophase  16.0 ± 0.5  15.2 ± 0.5  15.5 ± 0.4  14.8 ± 0.8  Values are the mean ± SEM and are expressed relative to CON mesor at day 15 (set to 100). Acrophase expressed in ZT. *P < 0.05 compared to CON at corresponding gestational day (t-test). †P < 0.05 overall effect compared to day 15, irrespective of diet (ANOVA). NS: Not significant for cosine fit. View Large Maternal hepatic clock gene expression also varied with gestational age; the mesor of Arntl, Per2, Cry1, Cry2, and Nr1d1 fell between days 15 and 21, while Per1 expression increased over this period (P < 0.001; Figure 2 and Table 3). The amplitudes of Arntl, Cry1, and Nr1d1 rhythms were also reduced between days 15 and 21 (P < 0.05; Figure 2 and Table 3). The acrophase (i.e., the time of the rhythm peak) for each maternal hepatic clock gene was largely unaffected by pregnancy stage, the only exception being that of Nr1d1, which was delayed by 0.8 h at day 21 compared to day 15 (P < 0.05; Figure 2I and Table 3). All of these gestational changes were similar in the two diet groups. Placental clock gene expression In contrast to the maternal liver, less than half of the placental clock genes measured were significant for cosinor fit (see Supplementary Table S1 for r2 and associated P values, and Supplementary Table S2 for cosine curve features). Accordingly, conventional ANOVA was used to examine the impact of CAF feeding on their placental expression. Labyrinth zone expression of Clock was reduced by the CAF diet at day 15 (P = 0.012 overall diet effect, Figure 3A), while Cry1 was increased in CAF at day 21 (P = 0.02 overall diet effect, Figure 3F). Labyrinth zone expression of Nr1d1 exhibited a Diet × Time of Day interaction in LZ tissue, whereby its peak level was reduced by the CAF diet at both gestational days (P < 0.05; Figure 3I). Labyrinth zone expression of all clock genes was increased from day 15 to 21 (P < 0.001; Figure 3), the sole exception being Nr1d1, overall expression of which fell between days 15 and 21 (P < 0.001; Figure 3I). Figure 3. View largeDownload slide Daily expression profiles of clock genes in the placental labyrinth zone at days 15 and 21 of gestation in control (CON) and cafeteria (CAF) diet groups. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. *P < 0.001 CON vs. CAF (overall diet effect in two-way ANOVA); **P < 0.001 compared to CON (t-test following Diet × Time of Day interaction in two-way ANOVA). Figure 3. View largeDownload slide Daily expression profiles of clock genes in the placental labyrinth zone at days 15 and 21 of gestation in control (CON) and cafeteria (CAF) diet groups. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. *P < 0.001 CON vs. CAF (overall diet effect in two-way ANOVA); **P < 0.001 compared to CON (t-test following Diet × Time of Day interaction in two-way ANOVA). Fetal hepatic clock gene expression As with the placenta, cosinor rhythmicity was not consistently observed for fetal hepatic clock gene profiles (10/18 profiles reached significance for cosinor fit; see Supplementary Table S3 for r2 and associated P values and Supplementary Table S4 for cosine curve features). Interestingly, there was consistency between the placenta and fetal liver with respect to cosinor rhythmicity (presence or absence) for all core clock gene profiles in CON pregnancies. Conventional ANOVA showed a significant Diet × Time of Day interaction for fetal hepatic Clock, Nr1d1, and Nr1f1 indicating that CAF effects on these genes were time-of-day-specific (see Figure 4). Most notable among the diet effects was a suppression of peak fetal hepatic Nr1d1 expression, similar to the effects of the CAF diet on Nr1d1 expression in both maternal liver (Figure 2) and the placental LZ (Figure 3). Figure 4. View largeDownload slide Daily expression profiles of fetal hepatic clock genes at day 21 of gestation in control (CON) and cafeteria (CAF) diet groups. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. *P < 0.001 CON vs. CAF (t-test following Diet × Time of Day interaction in two-way ANOVA). Figure 4. View largeDownload slide Daily expression profiles of fetal hepatic clock genes at day 21 of gestation in control (CON) and cafeteria (CAF) diet groups. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. *P < 0.001 CON vs. CAF (t-test following Diet × Time of Day interaction in two-way ANOVA). Effects of the CAF diet on maternal, placental, and fetal Slc2a2 and Ppars Maternal liver Rhythmic expression of Ppar-alpha and Slc2a2 was observed in maternal liver at both gestational ages in both diet groups (see Table 2 for r2 and P values). Ppar-delta expression was also rhythmic in CON mothers on both days, but in each case this rhythmicity was abolished by the CAF diet (Table 3 and Figure 5A). Figure 5. View largeDownload slide Rhythmic expression profiles of metabolic genes in maternal liver tissue of control (CON) and cafeteria (CAF) diet groups at days 15 and 21 of gestation. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. Cosinor curves are shown only for those genes that had significant cosinor rhythmicity. Statistical differences for cosine curve features are summarized in Table 3. Figure 5. View largeDownload slide Rhythmic expression profiles of metabolic genes in maternal liver tissue of control (CON) and cafeteria (CAF) diet groups at days 15 and 21 of gestation. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. Cosinor curves are shown only for those genes that had significant cosinor rhythmicity. Statistical differences for cosine curve features are summarized in Table 3. The CAF diet increased the mesor (P = 0.03) of maternal hepatic Slc2a2 at day 15 (when there was also a nonsignificant increase in Slc2a2 amplitude; P = 0.08), but not at day 21 (Table 3 and Figure 5D). While Pgc1-alpha and Ppar-alpha levels were unaffected by the CAF diet, expression of both genes increased with gestational age (P < 0.05), as did that of Slc2a2 (P < 0.05). In contrast, the amplitude of Ppar-delta fell between gestational days 15 and 21 (P < 0.05; Figure 5). Interestingly, the Ppar-delta rhythm was phase advanced (approximately 1.7 h) by the CAF diet on both gestational days (P < 0.01; Table 3 and Figure 5B), which was a comparable shift to that of several clock genes. To assess whether this Ppar-delta shift may be driven by clock genes, the relationships between Ppar-delta and key rhythm drivers, Arntl and Nr1d1 were assessed. This revealed a strong positive correlation between Ppar-delta and Arntl, and a negative association between Ppar-delta and Nr1d1 in each diet group across both days (see Figure 6 and Table 4). Figure 6. View largeDownload slide Relationship between (A) Ppar-delta and Nr1d1 and (B) Ppar-delta and Arntl in hepatic tissue of CAF mothers at day 21. Figure 6. View largeDownload slide Relationship between (A) Ppar-delta and Nr1d1 and (B) Ppar-delta and Arntl in hepatic tissue of CAF mothers at day 21. Table 4. R and associated P values for correlations between Ppar-delta and rhythmic drivers, Arntl and Nr1d1.   Day 15  Day 21    CON  CAF  CON  CAF  Ppar-delta – Arntl  0.780  0.789  0.800  0.850    (P < 0.001)  (P < 0.001)  (P < 0.001)  (P < 0.001)  Ppar-delta – Nr1d1  –0.380  –0.544  –0.440  –0.540    (P = 0.01)  (P < 0.001)  (P = 0.003)  (P < 0.001)    Day 15  Day 21    CON  CAF  CON  CAF  Ppar-delta – Arntl  0.780  0.789  0.800  0.850    (P < 0.001)  (P < 0.001)  (P < 0.001)  (P < 0.001)  Ppar-delta – Nr1d1  –0.380  –0.544  –0.440  –0.540    (P = 0.01)  (P < 0.001)  (P = 0.003)  (P < 0.001)  View Large Placental LZ Expression profiles for Pgc1-alpha and Ppar-delta were not rhythmic at either day of gestation, while Ppar-alpha expression was rhythmic only in CON placentas at day 15 and CAF placentas at day 21 (Supplementary Tables S1 and S2). Conventional ANOVA showed that the CAF diet increased LZ Pgc1-alpha expression at day 21 (P < 0.001 overall diet effect; Figure 7C). Labyrinth zone expression of Ppar-alpha and Pgc1-alpha both increased substantially between days 15 and 21 (P < 0.001; Figure 7A and C). Figure 7. View largeDownload slide Daily expression profiles of the Ppar genes in placental labyrinth zone in control (CON) and cafeteria (CAF) diet groups. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. *P < 0.001 CON vs. CAF (overall diet effect in two-way ANOVA). Figure 7. View largeDownload slide Daily expression profiles of the Ppar genes in placental labyrinth zone in control (CON) and cafeteria (CAF) diet groups. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. *P < 0.001 CON vs. CAF (overall diet effect in two-way ANOVA). Fetal liver Fetal hepatic Ppar-alpha, Ppar-delta, Pgc1-alpha, and Slc2a2 expression profiles all showed an increase across the day (P < 0.001; time of day effect in ANOVA; Figure 8), suggestive of a developmental change. Despite this time of day variation, none of these genes were affected by the CAF diet (Figure 8). Figure 8. View largeDownload slide Daily expression profiles of the Ppar genes in the fetal liver in control (CON) and cafeteria (CAF) diet groups. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. Figure 8. View largeDownload slide Daily expression profiles of the Ppar genes in the fetal liver in control (CON) and cafeteria (CAF) diet groups. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. Discussion This study demonstrates that maternal obesity in rat pregnancy suppresses the rhythmic expression of the accessory clock gene Nr1d1 in maternal liver, placental LZ, and fetal liver. CAF feeding also reduced hepatic expression of Cry2 and Per2 in the mother and phase-advanced the rhythms of Arntl, Per2, Per3, Cry1, and Nr1d1. Importantly, these latter effects did not extend to either the placenta or fetal liver, both of which exhibited largely arrhythmic clock gene expression profiles. Our data also show that opposite changes occur with gestational age in maternal hepatic (decreased) and placental (increased) clock gene expression, irrespective of obesity. Overall, these data provide important insights to the impact of maternal obesity on rhythmic hepatic function in the mother. They further suggest that the molecular clock networks in fetal and placental tissues remain relatively underdeveloped late in rat pregnancy. Nr1d1 is a vital component of the molecular clock network; its transcription is stimulated by the CLOCK: ARNTL heterodimer, and it acts in turn to repress Arntl and Clock transcription via RORE binding [28, 29]. Nr1d1 also plays a key role in the integration of the molecular clock with metabolic function [30], including the stimulation of adipocyte differentiation in white adipose tissue [31] and the regulation of hepatic and adipose lipid metabolism [32]. Maternal hepatic Nr1d1 expression was strongly rhythmic and was slightly reduced by the CAF diet (albeit significantly only for amplitude at day 21). But more significantly there was a marked decline in the mesor and amplitude of hepatic Nr1d1 expression with gestational age, indicative of a “flattening” of the overall rhythm. This change is consistent with recent observations in mouse pregnancy where a similar change appears linked to alterations in hepatic glucose metabolism [33]. It has been proposed that the physiological implications of reduced hepatic rhythmicity may relate to an increased capacity for substrate provision to meet the high demands of fetal growth [34]. Placental Nr1d1 profiles were also rhythmic in CON animals, with a distinct peak in expression that was suppressed by the CAF diet. Importantly, Nr1d1 also affects numerous other tissue functions not directly related to the molecular clock [35], including vascular [36] and inflammatory status [37], both of which change dramatically in rat placenta over the last week of pregnancy [38, 39]. Therefore, it will be of interest to determine whether the CAF-induced reduction in LZ Nr1d1 expression contributes to the placental and fetal growth restriction previously reported in this obesity model [21, 24]. The CAF diet also suppressed fetal hepatic Nr1d1 expression in a time-of-day-dependent manner, effectively blunting the transient peak evident in the control group. One previous study reported that maternal obesity actually increased fetal hepatic Nr1d1 in Rhesus macaques, but this was based on only a single time point [40]. Our data clearly show the importance of full circadian analyses, since CAF-induced suppression of Nr1d1 occurred only at a single time point. While the implications of altered Nr1d1 signaling in fetal tissues are unknown, it could potentially contribute to the adverse metabolic programming outcomes observed in offspring of obese pregnancies [6]. Several recent studies have noted that this effect of maternal obesity extends to offspring circadian biology; specifically, offspring of high-fat (HF) fed mothers have disrupted hepatic clock gene expression [41, 42], often in conjunction with symptoms of nonalcoholic fatty liver disease [43–45]. Interestingly, these effects were observed only in offspring weaned onto obesogenic diets, implying that postnatal feeding cues facilitate the development of these adverse metabolic outcomes. Taken together with these previous findings, our data suggest that altered Nr1d1 signaling in fetal life may also act as an underlying mechanism for the onset of programmed metabolic abnormalities. Although the alternative explanation that obesity per se is entirely responsible for altered clock gene expression in offspring remains a possibility, our observation that hepatic Nr1d1 is already altered in fetuses of obese mothers, well before they become obese, suggests otherwise. The slight CAF-induced reduction in maternal hepatic expression of several clock genes, and the approximately 1.5 h phase advance in their rhythms, are similar to previous observations in HF-fed, male mice [14, 16, 46]. Importantly, the hepatic clock is highly sensitive to food intake, with clock gene expression responding within 1 h of food consumption [47]. Moreover, both fat [16] and glucose [48] consumption alters hepatic clock gene expression in rodents independently of adiposity. As such, hepatic clock gene changes in CAF mothers may reflect CAF diet consumption per se, rather than adiposity-related parameters. The phase advances observed in the maternal hepatic clock may well reflect altered feeding times in CAF animals, since rodents with free access to an HF diet display increased daytime food intake [14, 46, 49]. Although the overall timing of food intake was not measured in this study, novel food items were provided to CAF mothers 1–2 h before lights off. As such, they are likely to have commenced daily eating earlier than the CON mothers, due to the novelty of the food items. Interestingly, because meal timing has major implications for the development of obesity in humans [50], atypical eating patterns may also contribute to the metabolic complications that occur in obese pregnant women. Consistent with the phase advance in maternal hepatic clock gene expression, hepatic Ppar-delta was also advanced by around 1.7 h in CAF mothers. While the molecular links between Ppar-delta and the core clock machinery are not completely understood, recent evidence in zebrafish indicates that the Ppar-delta promoter does not contain E-box response elements (i.e., the binding sites for CLOCK:ARNTL), but does have three Nr1d1 binding sites [51]. This suggests that Nr1d1 has transcriptional control over Ppar-delta, consistent with our observation of a strong negative correlation between Ppar-delta and Nr1d1. As expected, Ppar-delta also exhibited a strong positive correlation with Arntl expression in maternal hepatic tissue at both days of gestation, consistent with a previous report of a positive relationship between ARNTL and PPAR-delta in leukocyte mRNA transcripts of pregnant women [52]. Ppar-delta promotes insulin resistance by stimulating hepatic glucose utilization and impairing gluconeogenesis [53], and so the hepatic phase advance in Ppar-delta (likely driven by clock genes) may have downstream effects on glucose homeostasis in CAF mothers. In contrast to maternal hepatic clock genes, placental and fetal clock genes were largely unresponsive to the maternal obesity insult. While fetal liver may be shielded from rhythmic maternal feeding cues, this is not the case for the placenta. It seems more likely that the fetal and placental clock gene systems do not respond to the CAF diet because their molecular networks are underdeveloped. Indeed, both fetal and placental clock gene expression profiles were largely arrhythmic and did not exhibit the anti-phase expression patterns typical of specific clock gene pairs (e.g., Arntl and Per2) in a functional molecular clock [9]. Moreover, the presence or absence of cosinor rhythmicity of the core clock genes was consistent between placental and fetal liver expression profiles, suggestive of comparable levels of clock functionality in these tissues. Interestingly, unlike the reduction in maternal hepatic clock gene expression near term, most placental clock genes increased over the same period. As discussed above in relation to Nr1d1, it is possible that clock genes exert effects on placental function that are unrelated to the circadian clock. Importantly, there also appear to be differences between rodents and humans in terms of fetal and placental circadian development; rodents are altricial species, and evidence suggests that hepatic clock gene function does not become fully rhythmic in the rat until around postnatal day 30 [54]. In contrast, humans have a longer gestational length and are comparatively precocial at birth, and so rhythmicity may be more established in the human fetus and possibly the placenta. Indeed, recent evidence suggests that clock gene expression is rhythmic in the term human placenta [55], but it is unknown if this is disturbed by maternal obesity. CAF mothers were hyperglycemic and displayed increased hepatic Slc2a2 expression across day 15 of gestation, suggesting that obesity may elevate hepatic glucose flux at mid-gestation. The CAF-induced hyperglycemia occurred in conjunction with elevated insulin, suggestive of insulin resistance. By day 21 of gestation, there was no longer any dietary effect in blood glucose or hepatic Slc2a2 expression, and insulin was only elevated at the ZT13 time point in CAF mothers. This time-specific insulin elevation may relate to the provision of novel food items, since ZT13 is the first time point after these were introduced. Interestingly, the lack of an overall diet effect on either glucose or insulin at day 21 is consistent with previous reports showing metabolic convergence between lean and obese individuals in late gestation [56, 57]. In conclusion, this study provides novel evidence that obesity alters rhythmic Nr1d1 expression in maternal and fetal hepatic tissue, and in the LZ of the rat placenta. The CAF diet also reduced maternal hepatic expression of several clock genes and advanced their peak expression, and that of Ppar-delta. Other than Nr1d1, clock gene expression profiles in the placenta and fetal liver were largely arrhythmic, suggestive of functional immaturity of molecular clocks in these tissues. Supplementary data Supplementary data are available at BIOLRE online. Supplementary Table S1. Cosinor rhythmicity for LZ gene expression profiles. Values are r2 (P value) from nonlinear regression. Supplementary Table S2. Rhythmic features (mesor, amplitude, and acrophase) for LZ gene expression profiles. Supplementary Table S3. Cosinor rhythmicity for fetal hepatic gene expression profiles. Values are r2 (P value) from nonlinear regression. Supplementary Table S4. Rhythmic features (mesor, amplitude, and acrophase) for fetal hepatic gene expression profiles. Conflict of Interest: The authors have declared that no conflict of interest exists. † Grant Support: This research did not receive any specific funding in the public, commercial, or not-for-profit sector. References 1. Fisher SC, Kim SY, Sharma AJ, Rochat R, Morrow B. Is obesity still increasing among pregnant women? Prepregnancy obesity trends in 20 states, 2003–2009. Prev Med  2013; 56( 6): 372– 378. Google Scholar CrossRef Search ADS PubMed  2. Marchi J, Berg M, Dencker A, Olander EK, Begley C. Risks associated with obesity in pregnancy, for the mother and baby: a systematic review of reviews. Obes Rev  2015; 16( 8): 621– 638. Google Scholar CrossRef Search ADS PubMed  3. Martin KE, Grivell RM, Yelland LN, Dodd JM. The influence of maternal BMI and gestational diabetes on pregnancy outcome. Diabetes Res Clin Pract  2015; 108( 3): 508– 513. Google Scholar CrossRef Search ADS PubMed  4. Gardosi J, Francis A. A customized standard to assess fetal growth in a US population. Am J Obstet Gynecol  2009; 201( 25): 21– 27. 5. Thornburg KL. The programming of cardiovascular disease. J Dev Orig Health Dis  2015; 6( 05): 366– 376. Google Scholar CrossRef Search ADS PubMed  6. Blackmore HL, Ozanne SE. Maternal diet-induced obesity and offspring cardiovascular health. J Devel Orig Health Dis  2013; 4( 05): 338– 347. Google Scholar CrossRef Search ADS   7. Bailey SM, Udoh US, Young ME. Circadian regulation of metabolism. J Endocrinol  2014; 222( 2): R75– R96. Google Scholar CrossRef Search ADS PubMed  8. Froy O. Metabolism and circadian rhythms—implications for obesity. Endocr Rev  2010; 31( 1): 1– 24. Google Scholar CrossRef Search ADS PubMed  9. Reppert SM, Weaver DR. Coordination of circadian timing in mammals. Nature  2002; 418( 6901): 935– 941. Google Scholar CrossRef Search ADS PubMed  10. Kennaway DJ, Varcoe TJ, Voultsios A, Boden MJ. Global loss of Bmal1 expression alters adipose tissue hormones, gene expression and glucose metabolism. PLoS One  2013; 8( 6): e65255. Google Scholar CrossRef Search ADS PubMed  11. Turek FW, Joshu C, Kohsaka A, Lin E, Ivanova G, McDearmon E, Laposky A, Losee-Olson S, Easton A, Jensen DR, Eckel RH, Takahashi JS et al.   Obesity and metabolic syndrome in circadian clock mutant mice. Science  2005; 308( 5724): 1043– 1045. Google Scholar CrossRef Search ADS PubMed  12. Figueiro MG, White RD. Health consequences of shift work and implications for structural design. J Perinatol  2013; 33: S17– S23. Google Scholar CrossRef Search ADS PubMed  13. Knutsson A. Health disorders of shift workers. Occup Med  2003; 53( 2): 103– 108. Google Scholar CrossRef Search ADS   14. Kohsaka A, Laposky AD, Ramsey KM, Estrada C, Joshu C, Kobayashi Y, Turek FW, Bass J. High-fat diet disrupts behavioral and molecular circadian rhythms in mice. Cell Metab  2007; 6( 5): 414– 421. Google Scholar CrossRef Search ADS PubMed  15. Hsieh MC, Yang SC, Tseng HL, Hwang LL, Chen CT, Shieh KR. Abnormal expressions of circadian-clock and circadian clock-controlled genes in the livers and kidneys of long-term, high-fat-diet-treated mice. Int J Obes Relat Metab Disord  2010; 34( 2): 227– 239. Google Scholar CrossRef Search ADS   16. Eckel-Mahan KL, Patel VR, de Mateo S, Orozco-Solis R, Ceglia NJ, Sahar S, Dilag S, Dyar KA, Baldi P, Sassone-Corsi P. Reprogramming of the circadian clock by nutritional challenge. Cell  2013; 155( 7): 1464– 1478. Google Scholar CrossRef Search ADS PubMed  17. Wharfe MD, Mark PJ, Wyrwoll CS, Smith JT, Yap C, Clarke MW, Waddell BJ. Pregnancy-induced adaptations of the central circadian clock and maternal glucocorticoids. J Endocrinol  2016; 228( 3): 135– 147. Google Scholar CrossRef Search ADS PubMed  18. Varcoe TJ, Wight N, Voultsios A, Salkeld MD, Kennaway DJ. Chronic phase shifts of the photoperiod throughout pregnancy programs glucose intolerance and insulin resistance in the rat. PLoS One  2011; 6( 4): e18504. Google Scholar CrossRef Search ADS PubMed  19. Varcoe TJ, Boden MJ, Voultsios A, Salkeld MD, Rattanatray L, Kennaway DJ. Characterisation of the maternal response to chronic phase shifts during gestation in the rat: Implications for fetal metabolic programming. PLoS One  2013; 8( 1): e53800. Google Scholar CrossRef Search ADS PubMed  20. Wharfe MD, Mark PJ, Waddell BJ. Circadian variation in placental and hepatic clock genes in rat pregnancy. Endocrinology  2011; 152( 9): 3552– 3560. Google Scholar CrossRef Search ADS PubMed  21. Crew RC, Waddell BJ, Mark PJ. Maternal obesity induced by a ‘cafeteria’ diet in the rat does not increase inflammation in maternal, placental or fetal tissues in late gestation. Placenta  2016; 39: 33– 40. Google Scholar CrossRef Search ADS PubMed  22. Jones ML, Mark PJ, Keelan JA, Barden A, Mas E, Mori TA, Waddell BJ. Maternal dietary omega-3 fatty acid intake increases resolvin and protectin levels in the rat placenta. J Lipid Res  2013; 54( 8): 2247– 2254. Google Scholar CrossRef Search ADS PubMed  23. Faber KA, Hughes CL. Anogenital distance at birth as a predictor of volume of the sexually dimorphic nucleus of the preoptic area of the hypothalamus and pituitary responsiveness in castrated adult rats. Biol Reprod  1992; 46( 1): 101– 104. Google Scholar CrossRef Search ADS PubMed  24. Crew RC, Mark PJ, Clarke MW, Waddell BJ. Obesity disrupts the rhythmic profiles of maternal and fetal progesterone in rat pregnancy. Biol Reprod  2016; 95( 55): 51– 10. Google Scholar PubMed  25. Hewitt DP, Mark PJ, Dharmarajan AM, Waddell BJ. Placental expression of secreted frizzled related protein-4 in the rat and the impact of glucocorticoid-induced fetal and placental growth restriction. Biol Reprod  2006; 75: 75– 81. Google Scholar CrossRef Search ADS PubMed  26. Rozen S, Skaletsky H. Primer3 on the WWW for general users and for biologist programmers. In: Krawetz S, Miserer S (eds.), Bioinformatics Methods and Protocols: Methods in Molecular Biology . Humana Press, Totowa, NJ; 2000: 365– 386. Google Scholar CrossRef Search ADS   27. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A, Speleman F. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol  2002; 3: 1– 11. Google Scholar CrossRef Search ADS   28. Preitner N, Damiola F, Luis Lopez M, Zakany J, Duboule D, Albrecht U, Schibler U. The orphan nuclear receptor REV-ERB; controls circadian transcription within the positive limb of the mammalian circadian oscillator. Cell  2002; 110: 251– 260. Google Scholar CrossRef Search ADS PubMed  29. Cho H, Zhao X, Hatori M, Yu RT, Barish GD, Lam MT, Chong L-W, DiTacchio L, Atkins AR, Glass CK, Liddle C, Auwerx J et al.   Regulation of circadian behaviour and metabolism by Rev-erb-α and Rev-erb-β. Nature  2012; 485: 123– 127. Google Scholar CrossRef Search ADS PubMed  30. Duez H, Staels B. Rev-erbα gives a time cue to metabolism. FEBS Lett  2008; 582: 19– 25. Google Scholar CrossRef Search ADS PubMed  31. Fontaine C, Dubois G, Duguay Y, Helledie T, Vu-Dac N, Gervois P, Soncin F, Mandrup S, Fruchart J-C, Fruchart-Najib J, Staels B. The orphan nuclear receptor Rev-Erbα is a peroxisome proliferator-activated receptor (PPAR) γ target gene and promotes PPARγ-induced adipocyte differentiation. J Biol Chem  2003; 278: 37672– 37680. Google Scholar CrossRef Search ADS PubMed  32. Delezie J, Dumont S, Dardente H, Oudart H, Gréchez-Cassiau A, Klosen P, Teboul M, Delaunay F, Pévet P, Challet E. The nuclear receptor REV-ERB is required for the daily balance of carbohydrate and lipid metabolism. FASEB J  2012; 26: 3321– 3335. Google Scholar CrossRef Search ADS PubMed  33. Wharfe MD, Wyrwoll CS, Waddell BJ, Mark PJ. Pregnancy-induced changes in the circadian expression of hepatic clock genes: implications for maternal glucose homeostasis. Am J Physiol Endocrinol Metab  2016; 311: E575– E586. Google Scholar CrossRef Search ADS PubMed  34. Mark PJ, Crew RC, Wharfe MD, Waddell BJ. Rhythmic three-part harmony: the complex interaction of maternal, placental and fetal circadian systems. J Biol Rhythms  2017, doi: 10.1177/0748730417728671. 35. Everett LJ, Lazar MA. Nuclear receptor Rev-Erbα: up, down, and all around. Trends Endocrinol Metab  2014; 25: 586– 592. Google Scholar CrossRef Search ADS PubMed  36. Li M, Gallo D, Csizmadia E, Otterbein LE, Wegiel B. Carbon monoxide induces chromatin remodelling to facilitate endothelial cell migration. Thromb Haemost  2014; 111: 951– 959. Google Scholar CrossRef Search ADS PubMed  37. Gibbs JE, Blaikley J, Beesley S, Matthews L, Simpson KD, Boyce SH, Farrow SN, Else KJ, Singh D, Ray DW, Loudon ASI. The nuclear receptor REV-ERBα mediates circadian regulation of innate immunity through selective regulation of inflammatory cytokines. Proc Natl Acad Sci USA  2012; 109: 582– 587. Google Scholar CrossRef Search ADS PubMed  38. Hewitt DP, Mark PJ, Waddell BJ. Placental expression of peroxisome proliferator-activated receptors in rat pregnancy and the effect of increased glucocorticoid exposure1. Biol Reprod  2006; 74: 23– 28. Google Scholar CrossRef Search ADS PubMed  39. Mark PJ, Lewis JL, Jones ML, Keelan JA, Waddell BJ. The inflammatory state of the rat placenta increases in late gestation and is further enhanced by glucocorticoids in the labyrinth zone. Placenta  2013; 34: 559– 566. Google Scholar CrossRef Search ADS PubMed  40. Suter M, Bocock P, Showalter L, Hu M, Shope C, McKnight R, Grove K, Lane R, Aagaard-Tillery K. Epigenomics: maternal high-fat diet exposure in utero disrupts peripheral circadian gene expression in nonhuman primates. FASEB J  2011; 25: 714– 726. Google Scholar CrossRef Search ADS PubMed  41. Wang D, Chen S, Liu M, Liu C. Maternal obesity disrupts circadian rhythms of clock and metabolic genes in the offspring heart and liver. Chronobiol Int  2015; 32: 615– 626. Google Scholar CrossRef Search ADS PubMed  42. Borengasser SJ, Kang P, Faske J, Gomez-Acevedo H, Blackburn ML, Badger TM, Shankar K. High fat diet and in utero exposure to maternal obesity disrupts circadian rhythm and leads to metabolic programming of liver in rat offspring. PLoS One  2014; 9: e84209. Google Scholar CrossRef Search ADS PubMed  43. Bruce KD, Szczepankiewicz D, Sihota KK, Ravindraanandan M, Thomas H, Lillycrop KA, Burdge GC, Hanson MA, Byrne CD, Cagampang FR. Altered cellular redox status, sirtuin abundance and clock gene expression in a mouse model of developmentally primed NASH. Biochim Biophys Acta of Lipids  2016; 1861: 584– 593. Google Scholar CrossRef Search ADS   44. Mouralidarane A, Soeda J, Sugden D, Bocianowska A, Carter R, Ray S, Saraswati R, Cordero P, Novelli M, Fusai G, Vinciguerra M, Poston L et al.   Maternal obesity programs offspring non-alcoholic fatty liver disease through disruption of 24-h rhythms in mice. Int J Obes Relat Metab Disord  2015; 39: 1339– 1348. Google Scholar CrossRef Search ADS   45. Carter R, Mouralidarane A, Soeda J, Ray S, Pombo J, Saraswati R, Novelli M, Fusai G, Rappa F, Saracino C, Pazienza V, Poston L et al.   Non-alcoholic fatty pancreas disease pathogenesis: A role for developmental programming and altered circadian rhythms. PLoS One  2014; 9: e89505. Google Scholar CrossRef Search ADS PubMed  46. Pendergast JS, Branecky KL, Yang W, Ellacott KLJ, Niswender KD, Yamazaki S. High-fat diet acutely affects circadian organisation and eating behavior. Eur J Neurosci  2013; 37: 1350– 1356. Google Scholar CrossRef Search ADS PubMed  47. Oike H, Nagai K, Fukushima T, Ishida N, Kobori M. Feeding cues and injected nutrients induce acute expression of multiple clock genes in the mouse liver. PLoS One  2011; 6: e23709. Google Scholar CrossRef Search ADS PubMed  48. Iwanaga H, Yano M, Miki H, Okada K, Azama T, Takiguchi S, Fujiwara Y, Yasuda T, Nakayama M, Kobayashi M, Oishi K, Ishida N et al.   Per2 gene expressions in the suprachiasmatic nucleus and liver differentially respond to nutrition factors in rats. JPEN J Parenter Enteral Nutr  2005; 29: 157– 161. Google Scholar CrossRef Search ADS PubMed  49. Kentish SJ, Vincent AD, Kennaway DJ, Wittert GA, Page AJ. High-fat diet-induced obesity ablates gastric vagal afferent circadian rhythms. J Neurosci  2016; 36: 3199– 3207. Google Scholar CrossRef Search ADS PubMed  50. Garaulet M, Gómez-Abellán P. Timing of food intake and obesity: a novel association. Physiol Behav  2014; 134: 44– 50. Google Scholar CrossRef Search ADS PubMed  51. Kopp R, Billecke N, Legradi J, den Broeder M, Parekh SH, Legler J. Bringing obesity to light: Rev-erb[alpha], a central player in light-induced adipogenesis in the zebrafish? Int J Obes Relat Metab Disord  2016; 40: 824– 832. Google Scholar CrossRef Search ADS   52. Pappa KI, Gazouli M, Anastasiou E, Iliodromiti Z, Antsaklis A, Anagnou NP. Circadian clock gene expression is impaired in gestational diabetes mellitus. Gynecol Endocrinol  2013; 29: 331– 335. Google Scholar CrossRef Search ADS PubMed  53. Lee C-H, Olson P, Hevener A, Mehl I, Chong L-W, Olefsky JM, Gonzalez FJ, Ham J, Kang H, Peters JM, Evans RM. PPARδ regulates glucose metabolism and insulin sensitivity. Proc Natl Acad Sci USA  2006; 103: 3444– 3449. Google Scholar CrossRef Search ADS PubMed  54. Sládek M, Jindráková Z, Bendová Z, Sumová A. Postnatal ontogenesis of the circadian clock within the rat liver. Am J Physiol Regul Integr Comp Physiol  2007; 292: R1224– R1229. Google Scholar CrossRef Search ADS PubMed  55. Pérez S, Murias L, Fernández-Plaza C, Díaz I, González C, Otero J, Díaz E. Evidence for clock genes circadian rhythms in human full-term placenta. Syst Biol Reprod Med  2015; 61: 360– 366. Google Scholar CrossRef Search ADS PubMed  56. Forbes S, Barr SM, Reynolds RM, Semple S, Gray C, Andrew R, Denison FC, Walker BR, Norman JE. Convergence in insulin resistance between very severely obese and lean women at the end of pregnancy. Diabetologia  2015; 58: 2615– 2626. Google Scholar CrossRef Search ADS PubMed  57. King V, Dakin RS, Liu L, Hadoke PWF, Walker BR, Seckl JR, Norman JE, Drake AJ. Maternal obesity has little effect on the immediate offspring but impacts on the next generation. Endocrinology  2013; 154: 2514– 2524. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. Published by Oxford University Press on behalf of Society for the Study of Reproduction. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Biology of Reproduction Oxford University Press

Obesity-induced changes in hepatic and placental clock gene networks in rat pregnancy

Loading next page...
 
/lp/ou_press/obesity-induced-changes-in-hepatic-and-placental-clock-gene-networks-tdw60cw0ZL
Publisher
Oxford University Press
Copyright
© The Author(s) 2017. Published by Oxford University Press on behalf of Society for the Study of Reproduction. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
0006-3363
eISSN
1529-7268
D.O.I.
10.1093/biolre/iox158
Publisher site
See Article on Publisher Site

Abstract

Abstract Maternal obesity induces pregnancy complications and disturbs fetal development, but the specific mechanisms underlying these outcomes are unclear. Circadian rhythms are implicated in metabolic complications associated with obesity, and maternal metabolic adaptations to pregnancy. Accordingly, obesity-induced circadian dysfunction may drive adverse outcomes in obese pregnancy. This study investigated whether maternal obesity alters the rhythmic expression of clock genes and associated nuclear receptors across maternal, fetal, and placental tissues. Wistar rats were maintained on a cafeteria (CAF) diet prior to and throughout gestation to induce maternal obesity. Maternal and fetal liver and placental labyrinth zone (LZ) were collected at four-hourly time points across days 15–16 and 21–22 of gestation (term = 23 days). Gene expression was analyzed by RT-qPCR. Expression of the accessory clock gene Nr1d1 was rhythmic in the maternal and fetal liver and LZ of chow-fed controls, but in each case CAF feeding reduced peak Nr1d1 expression. Obesity resulted in a phase advance (approx. 1.5 h) in the rhythms of several clock genes and Ppar-delta in maternal liver. Aside from Nr1d1, expression of clock genes was mostly arrhythmic in LZ and fetal liver, and was unaffected by the CAF diet. In conclusion, maternal obesity suppressed Nr1d1 expression across maternal, fetal, and placental compartments and phase-advanced the rhythms of maternal hepatic clock genes. Given the key role of Nr1d1 in regulating metabolic, vascular, and inflammatory processes, our data suggest that disruptions to rhythmic Nr1d1 expression in utero may contribute to programmed health complications in offspring of obese pregnancies. Introduction Maternal obesity complicates approximately 20% of pregnancies in Western countries [1] and leads to a range of adverse maternal and fetal health outcomes. Obese women are more likely to experience pregnancy complications such as gestational diabetes and miscarriage [2, 3], and their infants are at a heightened risk for either macrosomia or intrauterine growth restriction [2, 4]. This disturbed development in utero can have lifelong health consequences, since offspring born to obese pregnancies often experience programmed metabolic disease in adulthood [5, 6]. While it is clear that obese pregnancy has negative effects on maternal and offspring health, the underlying mechanisms remain poorly understood. Circadian rhythms are endogenous timing processes that have evolved in response to the light-dark cycle in order to optimize the timing and efficiency of biological processes in effectively all living organisms. While the circadian system regulates daily oscillations in a range of physiological and behavioral functions, it is particularly crucial in the regulation of metabolism [7, 8]. Analysis of such oscillations is facilitated by the use of cosinor regression, whereby the characteristics of rhythms can be readily quantitated and thus statistically compared. At a molecular level, circadian rhythms are driven by positive and negative transcriptional feedback loops of clock genes (clock circadian regulator (Clock), aryl hydrocarbon receptor nuclear translocator like (gene symbol alias: Bmal1) (Arntl), period circadian regulator 1-3 (Per1-3), cryptochrome circadian regulator 1-2 (Cry1-2), nuclear receptor subfamily 1 group D member 1 (Nr1d1), and RAR-related orphan receptor A (gene symbol alias: Rora) (Nr1f1)) which then influence downstream metabolic pathways [9]. Since obesity is widely considered a metabolic disease, the circadian system is likely a key regulator of obesity-related patho-logies. Indeed, clock gene knockout models display altered metabolic phenotypes, including obesity and disturbed glucose metabolism [10, 11], while circadian disruption (e.g., shift work) leads to obesity and metabolic disease [12, 13]. Moreover, diet-induced obesity disturbs hepatic clock gene expression, and thereby alters expression of downstream genes and associated hepatic function [14–16]. Thus, there appears to be a reciprocal relationship between obesity and circadian dysfunction. Despite this evidence for obesity-induced changes to clock gene expression in the nonpregnant state, it is not known whether maternal obesity elicits similar effects during pregnancy. Maternal circadian rhythms change markedly across gestation [17], which likely facilitates the maternal metabolic adaptations required to sustain fetal growth. Moreover, these circadian adaptations appear essential for normal fetal development, since circadian disruption during pregnancy causes maternal and fetal metabolic dysfunction and programs adverse metabolic outcomes in offspring [18, 19]. Given a functional circadian system appears vital for healthy pregnancy outcomes, compromised circadian clocks could be key contributors to obesity-induced pregnancy complications. This may include not only maternal tissue clocks, but also those in the placental and fetal tissues, which also express clock genes [20]. Consequently, this study investigated the circadian expression profiles of clock genes and key downstream regulators of glucose and lipid metabolism (peroxisome proliferator activated receptor (Ppars) and solute carrier family 2 member 2 (previous gene symbol: Glut2) (Slc2a2)) in the maternal liver, placental labyrinth zone (LZ; the site of maternal–fetal exchange), and fetal liver in rat pregnancy. Maternal obesity was established by cafeteria (CAF) feeding [21], and tissues were analyzed across days 15–16 and 21–22 of gestation (term = 23 days), thus spanning the period of maximal fetal growth. It was hypothesized that maternal obesity alters rhythmicity of clock gene expression in hepatic and placental tissues, and that this has related effects on downstream metabolic genes. Materials and methods Animals and diets All animal procedures were approved by the Animal Ethics Committee of The University of Western Australia. Three-week-old albino Wistar rats were obtained from the Animal Resources Centre (Murdoch, Western Australia) and housed three to a cage. Animals were kept under a 12:12 h light-dark cycle at 22°C, with ad libitum access to water and standard rodent chow (14 KJ/g total energy; 12% energy as fat, 23% protein and 65% carbohydrate; Specialty Feeds, Glen Forrest, Western Australia). Following a week of acclimatization, animals were separated into control (CON) or CAF diet groups; CON animals were maintained on standard chow, while CAF animals were provided with four snack food items per day, in addition to chow. CAF items were given in excess to be essentially ad libitum and were changed daily at approximately ZT 11-12 (from a selection of 17 items) to maintain novelty. This resulted in an average energy density of 16.4 KJ/g (47% energy as fat, 8% as protein, and 44% as carbohydrate). For a detailed list of CAF diet constituents, see Crew et al. [21]. After 8 weeks of diet exposure, animals were mated overnight after determination of proestrus by an estrous cycle monitor (EC40; Fine Science Tools, Vancouver, British Columbia, Canada). The male rats were age-matched to the females, chow-fed, and maintained in identical environmental conditions to ensure minimal variation from paternal factors. Day 1 of pregnancy was confirmed by presence of spermatozoa in a vaginal smear the following morning. Pregnant animals were separated to individual housing and maintained on their respective diets throughout pregnancy. The timing of each tissue sampling is expressed relative to Zeitgeber time zero (ZT0), i.e., when lights were turned on in the animal facility (0700 h). Tissues were obtained at four-hourly intervals (ZT1, 0800 h; ZT5, 1200 h; ZT9, 1600 h; ZT13, 2000 h; ZT17, 2400 h; and ZT21, 0400 h) across days 15–16 and 21–22 of gestation (term occurs at day 23). At the appropriate collection time, animals (n = 7–8 per ZT for each diet) were anaesthetized using isoflurane/nitrous oxide. Maternal liver tissue was collected and fetal–placental pairs were removed via caesarean section. Placental LZ samples were isolated from whole placentas via blunt dissection, and liver tissue was collected from individual fetuses. Fetal sex was determined by amplification of the sex determining region Y (Sry) gene in fetal tail tissue at days 15–16 [22] or by anogenital distance at days 21–22 [23]. Although placental and fetal liver samples were collected for both sexes, only female tissues were analyzed for gene expression (one per litter). Notably, however, growth rates and steroid hormone responses to the CAF diet are similar in both sexes in this obesity model [24]. Maternal blood was taken from the descending aorta, and fetal blood was pooled from decapitated female fetuses within each litter at day 21. Blood glucose was measured immediately in maternal and fetal blood samples (Accu-Chek blood glucose monitor; Roche Diagnostics, Mannheim, Germany). Blood samples were then mixed with 10:1 (vol:vol) 0.6 M EDTA and centrifuged at 13 000× g for 6 min to isolate plasma. All collected tissue and plasma samples were snap frozen in liquid nitrogen and stored at –80°C until further analysis. Insulin measurement Insulin levels were measured in maternal and day 21 fetal plasma using a Milliplex MAP Rat Adipokine Magnetic Bead Panel (Cat. # RECYTMAG-65K; EMD Millipore Corporation, Billerica, MA, USA). The assay was performed according to the manufacturer's instructions and the plate was read on a Luminex Magpix (Luminex Corporation, Austin, Texas, USA). Data were analyzed with Magpix 4.2 Software. The intra-assay coefficient of variation was 8.4% and all quality controls provided with the Milliplex kit performed within expectations. RNA and cDNA sample preparation Total RNA was extracted from 50 to 100 mg of individual tissue samples using the QIAzol method (Qiagen Sciences, Maryland, USA), according to the manufacturer's instructions. RNA was assessed for concentration and purity with the Nanodrop ND-1000 spectrophotometer (Thermo Scientific, Wilmington, DE, USA) and 5 μg of RNA was reverse transcribed to cDNA by the mouse Moloney leukemia virus reverse transcriptase RNase H Point Mutant with random hexamers (Promega, Sydney, Australia) as previously reported [25]. The resultant cDNA was purified using the UltraClean PCR Cleanup Kit (MoBio Laboratories, Carlsband, CA, USA). Quantitative PCR The relative mRNA expression of clock genes (Clock, Arntl, Per1, Per2, Per3, Cry1, Cry2, Nr1f1, and Nr1d1) and genes that influence glucose and lipid metabolism (Ppar-alpha, Ppar-delta, Pgc1-alpha, Slc2a2) were analyzed via RT-qPCR on the Rotorgene Q (Qiagen, Hilden, Germany). All primers (see Table 1) were designed using Primer-BLAST [26] and were positioned to span an intron to prevent amplification from genomic DNA. Table 1. Primer details and PCR conditions for clock, metabolic, and reference genes. Gene  Primer sequence  Annealing Temp (°C)  Size (bp)  MgCl2 (mM)  Clock genes  Clock  F5΄ ACAGCGCACACACAGGCCTTC 3΄  60  175  2    R5΄ TGGCGGCGCCCTGTGATCTA 3΄        Arntl  F5΄ ACACTGCACCTCGGGAGCGA 3΄  60  100  2    R5΄ CGCCGAGCTCCAGAGCACAA 3΄        Per1  F5΄ CGCACTTCGGGAGCTCAAACTTC 3΄  60  169  2    R5΄ GTCCATGGCACAGGGCTCACC 3΄        Per2  F5΄ TGAGCTCCTTGGCGTTGCCG 3΄  60  147  2    R5΄ ACTCAGGCCCACTGGCCACA 3΄        Per3  F5΄ TTTTCCCCTTCAAGACATGG 3΄  60  167  2    R5΄ GAAAGAGAGGGCTGTTGTGC 3΄        Cry1  F5΄ AGCTGGCCACTGAGGCTGGT 3΄  60  158  2    R5΄ TGCTGGCATCTCCAGGGGCT 3΄        Cry2  F5΄ CTGCCCAGGAGCCACCAAGC 3΄  60  192  2    R5΄ GCATGCACACGCAAACGGCA 3΄        Nr1d1  F5΄ ATTGCCCACGGGGCGAGAGA 3΄  60  292  2    R5΄ GCCAAAAGAGCGGGCAGGGT 3΄        Nr1f1  F5΄ CCCAACCGTGTCCATGGCGG 3΄  60  113  2    R5΄ CCCGTCGATGCGTTTGGCGA 3΄        Ppars and Slc2a2  Ppar-alpha  F5΄ AATCCACGAAGCCTACCTGA 3΄  60  132  2.5    R5΄ GTCTTCTCAGCCATGCACAA 3΄        Ppar-delta  F5΄ GAGGGGTGCAAG GGCTTCTT 3΄  60  101  2.5    R5΄ CACTTGTTGCGGTTCTTCTTCTG 3΄        Pgc1-alpha  F5΄ TCTGGAACTGCAGGCCTAACTC 3΄  60  96  4    R5΄ GCAAGAGGGCTTCAGCTTTG 3΄        Slc2a2  F5΄ TAGGCGGAATGGTCGCCTCGT 3  61  102  2    R5΄ GGGCTCCAGTCAACGAGAGGCT 3΄        Reference genes  Ppia  F5΄ AGCATACAGGTCCTGGCATC 3΄  62  127  3    R5΄ TTCACCTTCCCAAAGACCAC 3΄        Sdha  F5΄ TGGGGCGACTCGTGGCTTTC 3΄  60  134  2    R5΄CCCCGCCTGCACCTACAACC 3΄        Ywhaz  F5΄ GACGGAGCTGAGGGACATCTGC 3΄  60  75  2    R5΄ GGCTGCGAAGCATTGGGGATCA 3΄        Gene  Primer sequence  Annealing Temp (°C)  Size (bp)  MgCl2 (mM)  Clock genes  Clock  F5΄ ACAGCGCACACACAGGCCTTC 3΄  60  175  2    R5΄ TGGCGGCGCCCTGTGATCTA 3΄        Arntl  F5΄ ACACTGCACCTCGGGAGCGA 3΄  60  100  2    R5΄ CGCCGAGCTCCAGAGCACAA 3΄        Per1  F5΄ CGCACTTCGGGAGCTCAAACTTC 3΄  60  169  2    R5΄ GTCCATGGCACAGGGCTCACC 3΄        Per2  F5΄ TGAGCTCCTTGGCGTTGCCG 3΄  60  147  2    R5΄ ACTCAGGCCCACTGGCCACA 3΄        Per3  F5΄ TTTTCCCCTTCAAGACATGG 3΄  60  167  2    R5΄ GAAAGAGAGGGCTGTTGTGC 3΄        Cry1  F5΄ AGCTGGCCACTGAGGCTGGT 3΄  60  158  2    R5΄ TGCTGGCATCTCCAGGGGCT 3΄        Cry2  F5΄ CTGCCCAGGAGCCACCAAGC 3΄  60  192  2    R5΄ GCATGCACACGCAAACGGCA 3΄        Nr1d1  F5΄ ATTGCCCACGGGGCGAGAGA 3΄  60  292  2    R5΄ GCCAAAAGAGCGGGCAGGGT 3΄        Nr1f1  F5΄ CCCAACCGTGTCCATGGCGG 3΄  60  113  2    R5΄ CCCGTCGATGCGTTTGGCGA 3΄        Ppars and Slc2a2  Ppar-alpha  F5΄ AATCCACGAAGCCTACCTGA 3΄  60  132  2.5    R5΄ GTCTTCTCAGCCATGCACAA 3΄        Ppar-delta  F5΄ GAGGGGTGCAAG GGCTTCTT 3΄  60  101  2.5    R5΄ CACTTGTTGCGGTTCTTCTTCTG 3΄        Pgc1-alpha  F5΄ TCTGGAACTGCAGGCCTAACTC 3΄  60  96  4    R5΄ GCAAGAGGGCTTCAGCTTTG 3΄        Slc2a2  F5΄ TAGGCGGAATGGTCGCCTCGT 3  61  102  2    R5΄ GGGCTCCAGTCAACGAGAGGCT 3΄        Reference genes  Ppia  F5΄ AGCATACAGGTCCTGGCATC 3΄  62  127  3    R5΄ TTCACCTTCCCAAAGACCAC 3΄        Sdha  F5΄ TGGGGCGACTCGTGGCTTTC 3΄  60  134  2    R5΄CCCCGCCTGCACCTACAACC 3΄        Ywhaz  F5΄ GACGGAGCTGAGGGACATCTGC 3΄  60  75  2    R5΄ GGCTGCGAAGCATTGGGGATCA 3΄        F: Forward primer, R: Reverse primer, bp: base pairs. View Large Each PCR reaction consisted of 10× Immolase buffer, 0.5 U Immolase DNA Polymerase (Bioline, Alexandria, NSW, Australia), 10 mM dNTPs, SYBR green (Molecular Probes, Eugene, OR, USA) at a 1/2000 dilution in DMSO, and gene-specific concentrations (see Table 1) of MgCl2 and forward and reverse primer. Complementary DNA template (1 μl) was added to give a total reaction volume of 10 μl. The PCR reaction cycle included an initial denaturing stage at 95°C for 10 min, followed by 45 cycles each comprising of 95°C for 1 s, a primer-specific annealing temperature (see Table 1) for 15 s and a 72°C extension period for 5 s. Samples were run in duplicate and no template controls were included in all PCR runs. Standard curves were generated in each PCR run from serial 10-fold dilutions of gel-extracted PCR product and were used to calculate the relative gene expression concentrations for each sample with Rotorgene Q series software. Common samples were included in each plate for comparison between PCR runs. All values were standardized against the reference genes peptidylprolyl isomerase A (Ppia), succinate dehydrogenase complex flavoprotein subunit A (Sdha), and tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein zeta (Ywhaz) using the GeNorm algorithm [27]. Statistical analysis All values are expressed as mean ± SEM, with n = 7–8 mothers per ZT group for each diet. Daily rhythms of gene expression or plasma hormone levels were assessed by nonlinear (cosinor) regression in Genstat 9.0 software (Hemel Hempstead, UK). This calculated the typical rhythm features of mesor (i.e., the rhythm-adjusted mean), amplitude (the difference between the mesor and peak), and acrophase (the time of the rhythm peak) for each daily profile. Comparisons of these cosine curve features between diet groups and gestational days were made by two-way analysis of variance (ANOVA) or t-test, as appropriate (GraphPad Prism version 6.00; La Jolla, California). Other comparisons were made by ANOVA using GenStat 9.0 software (Hemel Hempstead, UK) to test for differences due to diet, time of day, or stage of pregnancy. When significant (P < 0.05) interactions were observed between factors, differences were assessed by t-test or ANOVA, as appropriate. When the F-test reached statistical significance (P < 0.05), specific group comparisons were made by least significant difference (LSD) tests. Results Caloric intake, weight gain profiles, body composition, and pregnancy data (litter sizes and pregnancy success rates) for this animal cohort has been previously reported [24]. Briefly, CAF animals exhibited increased caloric intake and weight gain across the study period; this resulted in a 25% increase in body weight after 8 weeks of pre-pregnancy CAF feeding, and a 56% increase in total percentage adiposity in CAF mothers by day 21 of gestation. Fetal and placental weights were slightly reduced in the CAF group across both days of gestation, but total fetal and placental weight per mother did not differ between groups due to a small increase in litter size in the CAF animals. Profiles of blood glucose and plasma insulin As expected, maternal blood glucose levels were relatively stable across each of days 15 and 21, although a significant, low-amplitude rhythm became evident at day 21 (P = 0.006). There was also a Diet × Stage of Pregnancy interaction, with elevated blood glucose evident in CAF mothers at day 15 (P < 0.001; Figure 1A) but not at day 21. Fetal blood glucose levels increased 2.4-fold across day 21 (P < 0.001), but were unaffected by diet (Figure 1A). Figure 1. View largeDownload slide Daily profiles of blood glucose and plasma insulin in maternal and fetal plasma of control (CON) and cafeteria (CAF) diet groups. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT) and data for each time point were derived from separate groups of animals. Note different scales for fetal blood glucose and plasma insulin. *P < 0.001 CON vs. CAF (overall diet effect in two-way ANOVA); †P < 0.05 compared to corresponding CON; ††P < 0.01 compared to corresponding CON (t-test following Diet × Time of Day interaction in two-way ANOVA). #P < 0.05 compared to peak ZT value for that diet and day (one-way ANOVA following Diet × Time of Day interaction in two-way ANOVA). Figure 1. View largeDownload slide Daily profiles of blood glucose and plasma insulin in maternal and fetal plasma of control (CON) and cafeteria (CAF) diet groups. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT) and data for each time point were derived from separate groups of animals. Note different scales for fetal blood glucose and plasma insulin. *P < 0.001 CON vs. CAF (overall diet effect in two-way ANOVA); †P < 0.05 compared to corresponding CON; ††P < 0.01 compared to corresponding CON (t-test following Diet × Time of Day interaction in two-way ANOVA). #P < 0.05 compared to peak ZT value for that diet and day (one-way ANOVA following Diet × Time of Day interaction in two-way ANOVA). Maternal hyperinsulinemia was evident in the CAF group across all time points at day 15 (P < 0.001; Figure 1B), whereas at day 21 there was a significant Diet × Time of Day interaction, with hyperinsulinemia evident in CAF mothers only at ZT13 (P = 0.02; Figure 1B). Fetal insulin varied with time of day across day 21 (P = 0.01) but was unaffected by diet (Figure 1B). Effects of the CAF diet on maternal, placental, and fetal clock gene expression Maternal hepatic clock gene expression The expression profiles of all clock genes in the maternal liver were clearly rhythmic (i.e., significant cosinor fit) at both gestational ages, the single exception being Nr1f1 at day 15, where rhythmicity was abolished by the CAF diet (see Table 2 for r2 and P values for cosine fit significance). Obesity reduced the mesor (i.e., the overall average expression across the full day) of Cry2 at both days of gestation (P < 0.05; Figure 2G and Table 3), and while the Per1 mesor appeared to be reduced in CAF mothers at day 15, this did not reach statistical significance (P = 0.08; Table 3). The CAF diet also reduced the amplitude (i.e., the difference between the mesor and peak expression level) of Per2 at both days, and those of Cry2 and Nr1d1 at day 21 (P < 0.05; Figure 2 and Table 3). Interestingly, the CAF diet induced a phase advance of approximately 1.2 h in the maternal hepatic expression of Arntl, Per2, Per3 at both days of gestation, Cry1 at day 15 and Nr1d1 at day 21 (all P < 0.05; Figure 2 and Table 3). Figure 2. View largeDownload slide Rhythmic expression profiles of maternal hepatic clock genes in control (CON) and cafeteria (CAF) diet groups at days 15 and 21 of gestation. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. The inset graph (different scale) is provided in panel I to highlight the shift in the Nr1d1 acrophase. Statistical differences for cosine curve features are summarized in Table 3. Figure 2. View largeDownload slide Rhythmic expression profiles of maternal hepatic clock genes in control (CON) and cafeteria (CAF) diet groups at days 15 and 21 of gestation. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. The inset graph (different scale) is provided in panel I to highlight the shift in the Nr1d1 acrophase. Statistical differences for cosine curve features are summarized in Table 3. Table 2. Cosinor rhythmicity (r2 and associated P values) for maternal hepatic gene expression profiles.   Maternal liver day 15  Maternal liver day 21    CON  CAF  CON  CAF  Clock  0.607 (P < 0.001)  0.419 (P < 0.001)  0.306 (P < 0.001)  0.284 (P < 0.001)  Arntl  0.925 (P < 0.001)  0.895 (P < 0.001)  0.880 (P < 0.001)  0.830 (P < 0.001)  Per1  0.602 (P < 0.001)  0.597 (P < 0.001)  0.553 (P < 0.001)  0.505 (P < 0.001)  Per2  0.826 (P < 0.001)  0.710 (P < 0.001)  0.789 (P < 0.001)  0.707 (P < 0.001)  Per3  0.825 (P < 0.001)  0.682 (P < 0.001)  0.842 (P < 0.001)  0.727 (P < 0.001)  Cry1  0.857 (P < 0.001)  0.841 (P < 0.001)  0.837 (P < 0.001)  0.763 (P < 0.001)  Cry2  0.486 (P < 0.001)  0.278 (P < 0.001)  0.733 (P < 0.001)  0.403 (P < 0.001)  Nr1d1  0.538 (P < 0.001)  0.600 (P < 0.001)  0.747 (P < 0.001)  0.777 (P < 0.001)  Nr1f1  0.271 (P < 0.001)  0.036 (NS)  0.161 (P = 0.006)  0.087 (P = 0.05)  Ppar-alpha  0.111 (P = 0.027)  0.025 (NS)  0.156 (P = 0.008)  0 (NS)  Ppar-delta  0.698 (P < 0.001)  0.559 (P < 0.001)  0.681 (P < 0.001)  0.578 (P < 0.001)  Pgc1-alpha  0 (NS)  0 (NS)  0.023 (NS)  0.07 (NS)  Slc2a2  0.542 (P < 0.001)  0.583 (P < 0.001)  0.684 (P < 0.001)  0.358 (P < 0.001)    Maternal liver day 15  Maternal liver day 21    CON  CAF  CON  CAF  Clock  0.607 (P < 0.001)  0.419 (P < 0.001)  0.306 (P < 0.001)  0.284 (P < 0.001)  Arntl  0.925 (P < 0.001)  0.895 (P < 0.001)  0.880 (P < 0.001)  0.830 (P < 0.001)  Per1  0.602 (P < 0.001)  0.597 (P < 0.001)  0.553 (P < 0.001)  0.505 (P < 0.001)  Per2  0.826 (P < 0.001)  0.710 (P < 0.001)  0.789 (P < 0.001)  0.707 (P < 0.001)  Per3  0.825 (P < 0.001)  0.682 (P < 0.001)  0.842 (P < 0.001)  0.727 (P < 0.001)  Cry1  0.857 (P < 0.001)  0.841 (P < 0.001)  0.837 (P < 0.001)  0.763 (P < 0.001)  Cry2  0.486 (P < 0.001)  0.278 (P < 0.001)  0.733 (P < 0.001)  0.403 (P < 0.001)  Nr1d1  0.538 (P < 0.001)  0.600 (P < 0.001)  0.747 (P < 0.001)  0.777 (P < 0.001)  Nr1f1  0.271 (P < 0.001)  0.036 (NS)  0.161 (P = 0.006)  0.087 (P = 0.05)  Ppar-alpha  0.111 (P = 0.027)  0.025 (NS)  0.156 (P = 0.008)  0 (NS)  Ppar-delta  0.698 (P < 0.001)  0.559 (P < 0.001)  0.681 (P < 0.001)  0.578 (P < 0.001)  Pgc1-alpha  0 (NS)  0 (NS)  0.023 (NS)  0.07 (NS)  Slc2a2  0.542 (P < 0.001)  0.583 (P < 0.001)  0.684 (P < 0.001)  0.358 (P < 0.001)  NS: not significant (P > 0.05). View Large Table 3. Rhythmic features (mesor, amplitude, and acrophase) of maternal hepatic gene expression profiles.   Day 15  Day 21    CON  CAF  CON  CAF  Clock  Mesor  100 ± 3  92 ± 4  95 ± 4  93 ± 5    Amplitude  39 ± 5  31 ± 5  24 ± 5  28 ± 7    Acrophase  0.2 ± 0.4  23.3 ± 0.7  23.1 ± 0.8  22.4 ± 0.8  Arntl  Mesor  100 ± 2  96 ± 3  84 ± 3†  80 ± 4†    Amplitude  98 ± 4  89 ± 5  82 ± 5†  79 ± 5†    Acrophase  0.1 ± 0.2  22.8 ± 0.2*  0.3 ± 0.2  23.1 ± 0.3*  Per1  Mesor  100 ± 6  85 ± 6  113 ± 8  110 ± 7    Amplitude  67 ± 8  59 ± 8  83 ± 11  62 ± 10    Acrophase  13.4 ± 0.4  13.1 ± 0.5  14.3 ± 0.5†  13.9 ± 0.5†  Per2  Mesor  100 ± 4  101 ± 4  90 ± 5  84 ± 5    Amplitude  86 ± 6  54 ± 5*  86 ± 7  66 ± 6*    Acrophase  16.3 ± 0.3  15.3 ± 0.4*  16.3 ± 0.3  15.3 ± 0.4*  Per3  Mesor  100 ± 6  103 ± 8  98 ± 5  100 ± 6    Amplitude  119 ± 8  108 ± 11  107 ± 7  90 ± 8    Acrophase  12.4 ± 0.3  11.2 ± 0.4*  12.9 ± 0.2  11.7 ± 0.3*  Cry1  Mesor  100 ± 3  95 ± 3  86 ± 3†  87 ± 4†    Amplitude  69 ± 4  69 ± 5  59 ± 4†  61 ± 5†    Acrophase  20.0 ± 0.2  18.9 ± 0.2*  19.5 ± 0.3  18.8 ± 0.3  Cry2  Mesor  100 ± 3  90 ± 4*  88 ± 3†  76 ± 3*,†    Amplitude  30 ± 5  22 ± 5  39 ± 4  25 ± 4*    Acrophase  14.9 ± 0.6  14.6 ± 0.9  15.6 ± 0.3  14.1 ± 07  Nr1d1  Mesor  100 ± 15  91 ± 12  29.7 ± 0.9†  25.8 ± 0.7†    Amplitude  157 ± 22  132 ± 17  43.3 ± 1†  34.1 ± 1*,†    Acrophase  8.9 ± 0.5  7.9 ± 0.5  10.0 ± 0.3†  8.5 ± 0.3*,†  Nr1f1  Mesor  100 ± 4  99 ± 7  101 ± 4  97 ± 5    Amplitude  27 ± 6  18 ± 10  17 ± 5  NS    Acrophase  20.9 ± 0.9  20.3 ± 2.0  19.8 ± 1.2  NS  Ppar-alpha  Mesor  100 ± 8  117 ± 9  190 ± 19  233 ± 24    Amplitude  30 ± 11  24 ± 14  NS  NS    Acrophase  12.1 ± 1.4  10.4 ± 2.2  NS  NS  Ppar-delta  Mesor  100 ± 7  83 ± 7  87 ± 5  84 ± 6    Amplitude  89 ± 11  71 ± 10  62 ± 7†  60 ± 8†    Acrophase  12.8 ± 0.4  11.3 ± 0.5*  13.5 ± 0.4  11.5 ± 0.5*  Pgc1-alpha  Mesor  100 ± 5  87 ± 5  158 ± 7  138 ± 8    Amplitude  NS  NS  NS  NS    Acrophase  NS  NS  NS  NS  Slc2a2  Mesor  100 ± 3  112 ± 5*  110 ± 3  120 ± 6    Amplitude  35 ± 5  50 ± 7  45 ± 5  43 ± 9    Acrophase  16.0 ± 0.5  15.2 ± 0.5  15.5 ± 0.4  14.8 ± 0.8    Day 15  Day 21    CON  CAF  CON  CAF  Clock  Mesor  100 ± 3  92 ± 4  95 ± 4  93 ± 5    Amplitude  39 ± 5  31 ± 5  24 ± 5  28 ± 7    Acrophase  0.2 ± 0.4  23.3 ± 0.7  23.1 ± 0.8  22.4 ± 0.8  Arntl  Mesor  100 ± 2  96 ± 3  84 ± 3†  80 ± 4†    Amplitude  98 ± 4  89 ± 5  82 ± 5†  79 ± 5†    Acrophase  0.1 ± 0.2  22.8 ± 0.2*  0.3 ± 0.2  23.1 ± 0.3*  Per1  Mesor  100 ± 6  85 ± 6  113 ± 8  110 ± 7    Amplitude  67 ± 8  59 ± 8  83 ± 11  62 ± 10    Acrophase  13.4 ± 0.4  13.1 ± 0.5  14.3 ± 0.5†  13.9 ± 0.5†  Per2  Mesor  100 ± 4  101 ± 4  90 ± 5  84 ± 5    Amplitude  86 ± 6  54 ± 5*  86 ± 7  66 ± 6*    Acrophase  16.3 ± 0.3  15.3 ± 0.4*  16.3 ± 0.3  15.3 ± 0.4*  Per3  Mesor  100 ± 6  103 ± 8  98 ± 5  100 ± 6    Amplitude  119 ± 8  108 ± 11  107 ± 7  90 ± 8    Acrophase  12.4 ± 0.3  11.2 ± 0.4*  12.9 ± 0.2  11.7 ± 0.3*  Cry1  Mesor  100 ± 3  95 ± 3  86 ± 3†  87 ± 4†    Amplitude  69 ± 4  69 ± 5  59 ± 4†  61 ± 5†    Acrophase  20.0 ± 0.2  18.9 ± 0.2*  19.5 ± 0.3  18.8 ± 0.3  Cry2  Mesor  100 ± 3  90 ± 4*  88 ± 3†  76 ± 3*,†    Amplitude  30 ± 5  22 ± 5  39 ± 4  25 ± 4*    Acrophase  14.9 ± 0.6  14.6 ± 0.9  15.6 ± 0.3  14.1 ± 07  Nr1d1  Mesor  100 ± 15  91 ± 12  29.7 ± 0.9†  25.8 ± 0.7†    Amplitude  157 ± 22  132 ± 17  43.3 ± 1†  34.1 ± 1*,†    Acrophase  8.9 ± 0.5  7.9 ± 0.5  10.0 ± 0.3†  8.5 ± 0.3*,†  Nr1f1  Mesor  100 ± 4  99 ± 7  101 ± 4  97 ± 5    Amplitude  27 ± 6  18 ± 10  17 ± 5  NS    Acrophase  20.9 ± 0.9  20.3 ± 2.0  19.8 ± 1.2  NS  Ppar-alpha  Mesor  100 ± 8  117 ± 9  190 ± 19  233 ± 24    Amplitude  30 ± 11  24 ± 14  NS  NS    Acrophase  12.1 ± 1.4  10.4 ± 2.2  NS  NS  Ppar-delta  Mesor  100 ± 7  83 ± 7  87 ± 5  84 ± 6    Amplitude  89 ± 11  71 ± 10  62 ± 7†  60 ± 8†    Acrophase  12.8 ± 0.4  11.3 ± 0.5*  13.5 ± 0.4  11.5 ± 0.5*  Pgc1-alpha  Mesor  100 ± 5  87 ± 5  158 ± 7  138 ± 8    Amplitude  NS  NS  NS  NS    Acrophase  NS  NS  NS  NS  Slc2a2  Mesor  100 ± 3  112 ± 5*  110 ± 3  120 ± 6    Amplitude  35 ± 5  50 ± 7  45 ± 5  43 ± 9    Acrophase  16.0 ± 0.5  15.2 ± 0.5  15.5 ± 0.4  14.8 ± 0.8  Values are the mean ± SEM and are expressed relative to CON mesor at day 15 (set to 100). Acrophase expressed in ZT. *P < 0.05 compared to CON at corresponding gestational day (t-test). †P < 0.05 overall effect compared to day 15, irrespective of diet (ANOVA). NS: Not significant for cosine fit. View Large Maternal hepatic clock gene expression also varied with gestational age; the mesor of Arntl, Per2, Cry1, Cry2, and Nr1d1 fell between days 15 and 21, while Per1 expression increased over this period (P < 0.001; Figure 2 and Table 3). The amplitudes of Arntl, Cry1, and Nr1d1 rhythms were also reduced between days 15 and 21 (P < 0.05; Figure 2 and Table 3). The acrophase (i.e., the time of the rhythm peak) for each maternal hepatic clock gene was largely unaffected by pregnancy stage, the only exception being that of Nr1d1, which was delayed by 0.8 h at day 21 compared to day 15 (P < 0.05; Figure 2I and Table 3). All of these gestational changes were similar in the two diet groups. Placental clock gene expression In contrast to the maternal liver, less than half of the placental clock genes measured were significant for cosinor fit (see Supplementary Table S1 for r2 and associated P values, and Supplementary Table S2 for cosine curve features). Accordingly, conventional ANOVA was used to examine the impact of CAF feeding on their placental expression. Labyrinth zone expression of Clock was reduced by the CAF diet at day 15 (P = 0.012 overall diet effect, Figure 3A), while Cry1 was increased in CAF at day 21 (P = 0.02 overall diet effect, Figure 3F). Labyrinth zone expression of Nr1d1 exhibited a Diet × Time of Day interaction in LZ tissue, whereby its peak level was reduced by the CAF diet at both gestational days (P < 0.05; Figure 3I). Labyrinth zone expression of all clock genes was increased from day 15 to 21 (P < 0.001; Figure 3), the sole exception being Nr1d1, overall expression of which fell between days 15 and 21 (P < 0.001; Figure 3I). Figure 3. View largeDownload slide Daily expression profiles of clock genes in the placental labyrinth zone at days 15 and 21 of gestation in control (CON) and cafeteria (CAF) diet groups. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. *P < 0.001 CON vs. CAF (overall diet effect in two-way ANOVA); **P < 0.001 compared to CON (t-test following Diet × Time of Day interaction in two-way ANOVA). Figure 3. View largeDownload slide Daily expression profiles of clock genes in the placental labyrinth zone at days 15 and 21 of gestation in control (CON) and cafeteria (CAF) diet groups. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. *P < 0.001 CON vs. CAF (overall diet effect in two-way ANOVA); **P < 0.001 compared to CON (t-test following Diet × Time of Day interaction in two-way ANOVA). Fetal hepatic clock gene expression As with the placenta, cosinor rhythmicity was not consistently observed for fetal hepatic clock gene profiles (10/18 profiles reached significance for cosinor fit; see Supplementary Table S3 for r2 and associated P values and Supplementary Table S4 for cosine curve features). Interestingly, there was consistency between the placenta and fetal liver with respect to cosinor rhythmicity (presence or absence) for all core clock gene profiles in CON pregnancies. Conventional ANOVA showed a significant Diet × Time of Day interaction for fetal hepatic Clock, Nr1d1, and Nr1f1 indicating that CAF effects on these genes were time-of-day-specific (see Figure 4). Most notable among the diet effects was a suppression of peak fetal hepatic Nr1d1 expression, similar to the effects of the CAF diet on Nr1d1 expression in both maternal liver (Figure 2) and the placental LZ (Figure 3). Figure 4. View largeDownload slide Daily expression profiles of fetal hepatic clock genes at day 21 of gestation in control (CON) and cafeteria (CAF) diet groups. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. *P < 0.001 CON vs. CAF (t-test following Diet × Time of Day interaction in two-way ANOVA). Figure 4. View largeDownload slide Daily expression profiles of fetal hepatic clock genes at day 21 of gestation in control (CON) and cafeteria (CAF) diet groups. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. *P < 0.001 CON vs. CAF (t-test following Diet × Time of Day interaction in two-way ANOVA). Effects of the CAF diet on maternal, placental, and fetal Slc2a2 and Ppars Maternal liver Rhythmic expression of Ppar-alpha and Slc2a2 was observed in maternal liver at both gestational ages in both diet groups (see Table 2 for r2 and P values). Ppar-delta expression was also rhythmic in CON mothers on both days, but in each case this rhythmicity was abolished by the CAF diet (Table 3 and Figure 5A). Figure 5. View largeDownload slide Rhythmic expression profiles of metabolic genes in maternal liver tissue of control (CON) and cafeteria (CAF) diet groups at days 15 and 21 of gestation. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. Cosinor curves are shown only for those genes that had significant cosinor rhythmicity. Statistical differences for cosine curve features are summarized in Table 3. Figure 5. View largeDownload slide Rhythmic expression profiles of metabolic genes in maternal liver tissue of control (CON) and cafeteria (CAF) diet groups at days 15 and 21 of gestation. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. Cosinor curves are shown only for those genes that had significant cosinor rhythmicity. Statistical differences for cosine curve features are summarized in Table 3. The CAF diet increased the mesor (P = 0.03) of maternal hepatic Slc2a2 at day 15 (when there was also a nonsignificant increase in Slc2a2 amplitude; P = 0.08), but not at day 21 (Table 3 and Figure 5D). While Pgc1-alpha and Ppar-alpha levels were unaffected by the CAF diet, expression of both genes increased with gestational age (P < 0.05), as did that of Slc2a2 (P < 0.05). In contrast, the amplitude of Ppar-delta fell between gestational days 15 and 21 (P < 0.05; Figure 5). Interestingly, the Ppar-delta rhythm was phase advanced (approximately 1.7 h) by the CAF diet on both gestational days (P < 0.01; Table 3 and Figure 5B), which was a comparable shift to that of several clock genes. To assess whether this Ppar-delta shift may be driven by clock genes, the relationships between Ppar-delta and key rhythm drivers, Arntl and Nr1d1 were assessed. This revealed a strong positive correlation between Ppar-delta and Arntl, and a negative association between Ppar-delta and Nr1d1 in each diet group across both days (see Figure 6 and Table 4). Figure 6. View largeDownload slide Relationship between (A) Ppar-delta and Nr1d1 and (B) Ppar-delta and Arntl in hepatic tissue of CAF mothers at day 21. Figure 6. View largeDownload slide Relationship between (A) Ppar-delta and Nr1d1 and (B) Ppar-delta and Arntl in hepatic tissue of CAF mothers at day 21. Table 4. R and associated P values for correlations between Ppar-delta and rhythmic drivers, Arntl and Nr1d1.   Day 15  Day 21    CON  CAF  CON  CAF  Ppar-delta – Arntl  0.780  0.789  0.800  0.850    (P < 0.001)  (P < 0.001)  (P < 0.001)  (P < 0.001)  Ppar-delta – Nr1d1  –0.380  –0.544  –0.440  –0.540    (P = 0.01)  (P < 0.001)  (P = 0.003)  (P < 0.001)    Day 15  Day 21    CON  CAF  CON  CAF  Ppar-delta – Arntl  0.780  0.789  0.800  0.850    (P < 0.001)  (P < 0.001)  (P < 0.001)  (P < 0.001)  Ppar-delta – Nr1d1  –0.380  –0.544  –0.440  –0.540    (P = 0.01)  (P < 0.001)  (P = 0.003)  (P < 0.001)  View Large Placental LZ Expression profiles for Pgc1-alpha and Ppar-delta were not rhythmic at either day of gestation, while Ppar-alpha expression was rhythmic only in CON placentas at day 15 and CAF placentas at day 21 (Supplementary Tables S1 and S2). Conventional ANOVA showed that the CAF diet increased LZ Pgc1-alpha expression at day 21 (P < 0.001 overall diet effect; Figure 7C). Labyrinth zone expression of Ppar-alpha and Pgc1-alpha both increased substantially between days 15 and 21 (P < 0.001; Figure 7A and C). Figure 7. View largeDownload slide Daily expression profiles of the Ppar genes in placental labyrinth zone in control (CON) and cafeteria (CAF) diet groups. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. *P < 0.001 CON vs. CAF (overall diet effect in two-way ANOVA). Figure 7. View largeDownload slide Daily expression profiles of the Ppar genes in placental labyrinth zone in control (CON) and cafeteria (CAF) diet groups. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. *P < 0.001 CON vs. CAF (overall diet effect in two-way ANOVA). Fetal liver Fetal hepatic Ppar-alpha, Ppar-delta, Pgc1-alpha, and Slc2a2 expression profiles all showed an increase across the day (P < 0.001; time of day effect in ANOVA; Figure 8), suggestive of a developmental change. Despite this time of day variation, none of these genes were affected by the CAF diet (Figure 8). Figure 8. View largeDownload slide Daily expression profiles of the Ppar genes in the fetal liver in control (CON) and cafeteria (CAF) diet groups. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. Figure 8. View largeDownload slide Daily expression profiles of the Ppar genes in the fetal liver in control (CON) and cafeteria (CAF) diet groups. Shaded areas represent the dark period. Values are the mean ± SEM (n = 7–8 per diet group at each ZT), and data for each time point were derived from separate groups of animals. Discussion This study demonstrates that maternal obesity in rat pregnancy suppresses the rhythmic expression of the accessory clock gene Nr1d1 in maternal liver, placental LZ, and fetal liver. CAF feeding also reduced hepatic expression of Cry2 and Per2 in the mother and phase-advanced the rhythms of Arntl, Per2, Per3, Cry1, and Nr1d1. Importantly, these latter effects did not extend to either the placenta or fetal liver, both of which exhibited largely arrhythmic clock gene expression profiles. Our data also show that opposite changes occur with gestational age in maternal hepatic (decreased) and placental (increased) clock gene expression, irrespective of obesity. Overall, these data provide important insights to the impact of maternal obesity on rhythmic hepatic function in the mother. They further suggest that the molecular clock networks in fetal and placental tissues remain relatively underdeveloped late in rat pregnancy. Nr1d1 is a vital component of the molecular clock network; its transcription is stimulated by the CLOCK: ARNTL heterodimer, and it acts in turn to repress Arntl and Clock transcription via RORE binding [28, 29]. Nr1d1 also plays a key role in the integration of the molecular clock with metabolic function [30], including the stimulation of adipocyte differentiation in white adipose tissue [31] and the regulation of hepatic and adipose lipid metabolism [32]. Maternal hepatic Nr1d1 expression was strongly rhythmic and was slightly reduced by the CAF diet (albeit significantly only for amplitude at day 21). But more significantly there was a marked decline in the mesor and amplitude of hepatic Nr1d1 expression with gestational age, indicative of a “flattening” of the overall rhythm. This change is consistent with recent observations in mouse pregnancy where a similar change appears linked to alterations in hepatic glucose metabolism [33]. It has been proposed that the physiological implications of reduced hepatic rhythmicity may relate to an increased capacity for substrate provision to meet the high demands of fetal growth [34]. Placental Nr1d1 profiles were also rhythmic in CON animals, with a distinct peak in expression that was suppressed by the CAF diet. Importantly, Nr1d1 also affects numerous other tissue functions not directly related to the molecular clock [35], including vascular [36] and inflammatory status [37], both of which change dramatically in rat placenta over the last week of pregnancy [38, 39]. Therefore, it will be of interest to determine whether the CAF-induced reduction in LZ Nr1d1 expression contributes to the placental and fetal growth restriction previously reported in this obesity model [21, 24]. The CAF diet also suppressed fetal hepatic Nr1d1 expression in a time-of-day-dependent manner, effectively blunting the transient peak evident in the control group. One previous study reported that maternal obesity actually increased fetal hepatic Nr1d1 in Rhesus macaques, but this was based on only a single time point [40]. Our data clearly show the importance of full circadian analyses, since CAF-induced suppression of Nr1d1 occurred only at a single time point. While the implications of altered Nr1d1 signaling in fetal tissues are unknown, it could potentially contribute to the adverse metabolic programming outcomes observed in offspring of obese pregnancies [6]. Several recent studies have noted that this effect of maternal obesity extends to offspring circadian biology; specifically, offspring of high-fat (HF) fed mothers have disrupted hepatic clock gene expression [41, 42], often in conjunction with symptoms of nonalcoholic fatty liver disease [43–45]. Interestingly, these effects were observed only in offspring weaned onto obesogenic diets, implying that postnatal feeding cues facilitate the development of these adverse metabolic outcomes. Taken together with these previous findings, our data suggest that altered Nr1d1 signaling in fetal life may also act as an underlying mechanism for the onset of programmed metabolic abnormalities. Although the alternative explanation that obesity per se is entirely responsible for altered clock gene expression in offspring remains a possibility, our observation that hepatic Nr1d1 is already altered in fetuses of obese mothers, well before they become obese, suggests otherwise. The slight CAF-induced reduction in maternal hepatic expression of several clock genes, and the approximately 1.5 h phase advance in their rhythms, are similar to previous observations in HF-fed, male mice [14, 16, 46]. Importantly, the hepatic clock is highly sensitive to food intake, with clock gene expression responding within 1 h of food consumption [47]. Moreover, both fat [16] and glucose [48] consumption alters hepatic clock gene expression in rodents independently of adiposity. As such, hepatic clock gene changes in CAF mothers may reflect CAF diet consumption per se, rather than adiposity-related parameters. The phase advances observed in the maternal hepatic clock may well reflect altered feeding times in CAF animals, since rodents with free access to an HF diet display increased daytime food intake [14, 46, 49]. Although the overall timing of food intake was not measured in this study, novel food items were provided to CAF mothers 1–2 h before lights off. As such, they are likely to have commenced daily eating earlier than the CON mothers, due to the novelty of the food items. Interestingly, because meal timing has major implications for the development of obesity in humans [50], atypical eating patterns may also contribute to the metabolic complications that occur in obese pregnant women. Consistent with the phase advance in maternal hepatic clock gene expression, hepatic Ppar-delta was also advanced by around 1.7 h in CAF mothers. While the molecular links between Ppar-delta and the core clock machinery are not completely understood, recent evidence in zebrafish indicates that the Ppar-delta promoter does not contain E-box response elements (i.e., the binding sites for CLOCK:ARNTL), but does have three Nr1d1 binding sites [51]. This suggests that Nr1d1 has transcriptional control over Ppar-delta, consistent with our observation of a strong negative correlation between Ppar-delta and Nr1d1. As expected, Ppar-delta also exhibited a strong positive correlation with Arntl expression in maternal hepatic tissue at both days of gestation, consistent with a previous report of a positive relationship between ARNTL and PPAR-delta in leukocyte mRNA transcripts of pregnant women [52]. Ppar-delta promotes insulin resistance by stimulating hepatic glucose utilization and impairing gluconeogenesis [53], and so the hepatic phase advance in Ppar-delta (likely driven by clock genes) may have downstream effects on glucose homeostasis in CAF mothers. In contrast to maternal hepatic clock genes, placental and fetal clock genes were largely unresponsive to the maternal obesity insult. While fetal liver may be shielded from rhythmic maternal feeding cues, this is not the case for the placenta. It seems more likely that the fetal and placental clock gene systems do not respond to the CAF diet because their molecular networks are underdeveloped. Indeed, both fetal and placental clock gene expression profiles were largely arrhythmic and did not exhibit the anti-phase expression patterns typical of specific clock gene pairs (e.g., Arntl and Per2) in a functional molecular clock [9]. Moreover, the presence or absence of cosinor rhythmicity of the core clock genes was consistent between placental and fetal liver expression profiles, suggestive of comparable levels of clock functionality in these tissues. Interestingly, unlike the reduction in maternal hepatic clock gene expression near term, most placental clock genes increased over the same period. As discussed above in relation to Nr1d1, it is possible that clock genes exert effects on placental function that are unrelated to the circadian clock. Importantly, there also appear to be differences between rodents and humans in terms of fetal and placental circadian development; rodents are altricial species, and evidence suggests that hepatic clock gene function does not become fully rhythmic in the rat until around postnatal day 30 [54]. In contrast, humans have a longer gestational length and are comparatively precocial at birth, and so rhythmicity may be more established in the human fetus and possibly the placenta. Indeed, recent evidence suggests that clock gene expression is rhythmic in the term human placenta [55], but it is unknown if this is disturbed by maternal obesity. CAF mothers were hyperglycemic and displayed increased hepatic Slc2a2 expression across day 15 of gestation, suggesting that obesity may elevate hepatic glucose flux at mid-gestation. The CAF-induced hyperglycemia occurred in conjunction with elevated insulin, suggestive of insulin resistance. By day 21 of gestation, there was no longer any dietary effect in blood glucose or hepatic Slc2a2 expression, and insulin was only elevated at the ZT13 time point in CAF mothers. This time-specific insulin elevation may relate to the provision of novel food items, since ZT13 is the first time point after these were introduced. Interestingly, the lack of an overall diet effect on either glucose or insulin at day 21 is consistent with previous reports showing metabolic convergence between lean and obese individuals in late gestation [56, 57]. In conclusion, this study provides novel evidence that obesity alters rhythmic Nr1d1 expression in maternal and fetal hepatic tissue, and in the LZ of the rat placenta. The CAF diet also reduced maternal hepatic expression of several clock genes and advanced their peak expression, and that of Ppar-delta. Other than Nr1d1, clock gene expression profiles in the placenta and fetal liver were largely arrhythmic, suggestive of functional immaturity of molecular clocks in these tissues. Supplementary data Supplementary data are available at BIOLRE online. Supplementary Table S1. Cosinor rhythmicity for LZ gene expression profiles. Values are r2 (P value) from nonlinear regression. Supplementary Table S2. Rhythmic features (mesor, amplitude, and acrophase) for LZ gene expression profiles. Supplementary Table S3. Cosinor rhythmicity for fetal hepatic gene expression profiles. Values are r2 (P value) from nonlinear regression. Supplementary Table S4. Rhythmic features (mesor, amplitude, and acrophase) for fetal hepatic gene expression profiles. Conflict of Interest: The authors have declared that no conflict of interest exists. † Grant Support: This research did not receive any specific funding in the public, commercial, or not-for-profit sector. References 1. Fisher SC, Kim SY, Sharma AJ, Rochat R, Morrow B. Is obesity still increasing among pregnant women? Prepregnancy obesity trends in 20 states, 2003–2009. Prev Med  2013; 56( 6): 372– 378. Google Scholar CrossRef Search ADS PubMed  2. Marchi J, Berg M, Dencker A, Olander EK, Begley C. Risks associated with obesity in pregnancy, for the mother and baby: a systematic review of reviews. Obes Rev  2015; 16( 8): 621– 638. Google Scholar CrossRef Search ADS PubMed  3. Martin KE, Grivell RM, Yelland LN, Dodd JM. The influence of maternal BMI and gestational diabetes on pregnancy outcome. Diabetes Res Clin Pract  2015; 108( 3): 508– 513. Google Scholar CrossRef Search ADS PubMed  4. Gardosi J, Francis A. A customized standard to assess fetal growth in a US population. Am J Obstet Gynecol  2009; 201( 25): 21– 27. 5. Thornburg KL. The programming of cardiovascular disease. J Dev Orig Health Dis  2015; 6( 05): 366– 376. Google Scholar CrossRef Search ADS PubMed  6. Blackmore HL, Ozanne SE. Maternal diet-induced obesity and offspring cardiovascular health. J Devel Orig Health Dis  2013; 4( 05): 338– 347. Google Scholar CrossRef Search ADS   7. Bailey SM, Udoh US, Young ME. Circadian regulation of metabolism. J Endocrinol  2014; 222( 2): R75– R96. Google Scholar CrossRef Search ADS PubMed  8. Froy O. Metabolism and circadian rhythms—implications for obesity. Endocr Rev  2010; 31( 1): 1– 24. Google Scholar CrossRef Search ADS PubMed  9. Reppert SM, Weaver DR. Coordination of circadian timing in mammals. Nature  2002; 418( 6901): 935– 941. Google Scholar CrossRef Search ADS PubMed  10. Kennaway DJ, Varcoe TJ, Voultsios A, Boden MJ. Global loss of Bmal1 expression alters adipose tissue hormones, gene expression and glucose metabolism. PLoS One  2013; 8( 6): e65255. Google Scholar CrossRef Search ADS PubMed  11. Turek FW, Joshu C, Kohsaka A, Lin E, Ivanova G, McDearmon E, Laposky A, Losee-Olson S, Easton A, Jensen DR, Eckel RH, Takahashi JS et al.   Obesity and metabolic syndrome in circadian clock mutant mice. Science  2005; 308( 5724): 1043– 1045. Google Scholar CrossRef Search ADS PubMed  12. Figueiro MG, White RD. Health consequences of shift work and implications for structural design. J Perinatol  2013; 33: S17– S23. Google Scholar CrossRef Search ADS PubMed  13. Knutsson A. Health disorders of shift workers. Occup Med  2003; 53( 2): 103– 108. Google Scholar CrossRef Search ADS   14. Kohsaka A, Laposky AD, Ramsey KM, Estrada C, Joshu C, Kobayashi Y, Turek FW, Bass J. High-fat diet disrupts behavioral and molecular circadian rhythms in mice. Cell Metab  2007; 6( 5): 414– 421. Google Scholar CrossRef Search ADS PubMed  15. Hsieh MC, Yang SC, Tseng HL, Hwang LL, Chen CT, Shieh KR. Abnormal expressions of circadian-clock and circadian clock-controlled genes in the livers and kidneys of long-term, high-fat-diet-treated mice. Int J Obes Relat Metab Disord  2010; 34( 2): 227– 239. Google Scholar CrossRef Search ADS   16. Eckel-Mahan KL, Patel VR, de Mateo S, Orozco-Solis R, Ceglia NJ, Sahar S, Dilag S, Dyar KA, Baldi P, Sassone-Corsi P. Reprogramming of the circadian clock by nutritional challenge. Cell  2013; 155( 7): 1464– 1478. Google Scholar CrossRef Search ADS PubMed  17. Wharfe MD, Mark PJ, Wyrwoll CS, Smith JT, Yap C, Clarke MW, Waddell BJ. Pregnancy-induced adaptations of the central circadian clock and maternal glucocorticoids. J Endocrinol  2016; 228( 3): 135– 147. Google Scholar CrossRef Search ADS PubMed  18. Varcoe TJ, Wight N, Voultsios A, Salkeld MD, Kennaway DJ. Chronic phase shifts of the photoperiod throughout pregnancy programs glucose intolerance and insulin resistance in the rat. PLoS One  2011; 6( 4): e18504. Google Scholar CrossRef Search ADS PubMed  19. Varcoe TJ, Boden MJ, Voultsios A, Salkeld MD, Rattanatray L, Kennaway DJ. Characterisation of the maternal response to chronic phase shifts during gestation in the rat: Implications for fetal metabolic programming. PLoS One  2013; 8( 1): e53800. Google Scholar CrossRef Search ADS PubMed  20. Wharfe MD, Mark PJ, Waddell BJ. Circadian variation in placental and hepatic clock genes in rat pregnancy. Endocrinology  2011; 152( 9): 3552– 3560. Google Scholar CrossRef Search ADS PubMed  21. Crew RC, Waddell BJ, Mark PJ. Maternal obesity induced by a ‘cafeteria’ diet in the rat does not increase inflammation in maternal, placental or fetal tissues in late gestation. Placenta  2016; 39: 33– 40. Google Scholar CrossRef Search ADS PubMed  22. Jones ML, Mark PJ, Keelan JA, Barden A, Mas E, Mori TA, Waddell BJ. Maternal dietary omega-3 fatty acid intake increases resolvin and protectin levels in the rat placenta. J Lipid Res  2013; 54( 8): 2247– 2254. Google Scholar CrossRef Search ADS PubMed  23. Faber KA, Hughes CL. Anogenital distance at birth as a predictor of volume of the sexually dimorphic nucleus of the preoptic area of the hypothalamus and pituitary responsiveness in castrated adult rats. Biol Reprod  1992; 46( 1): 101– 104. Google Scholar CrossRef Search ADS PubMed  24. Crew RC, Mark PJ, Clarke MW, Waddell BJ. Obesity disrupts the rhythmic profiles of maternal and fetal progesterone in rat pregnancy. Biol Reprod  2016; 95( 55): 51– 10. Google Scholar PubMed  25. Hewitt DP, Mark PJ, Dharmarajan AM, Waddell BJ. Placental expression of secreted frizzled related protein-4 in the rat and the impact of glucocorticoid-induced fetal and placental growth restriction. Biol Reprod  2006; 75: 75– 81. Google Scholar CrossRef Search ADS PubMed  26. Rozen S, Skaletsky H. Primer3 on the WWW for general users and for biologist programmers. In: Krawetz S, Miserer S (eds.), Bioinformatics Methods and Protocols: Methods in Molecular Biology . Humana Press, Totowa, NJ; 2000: 365– 386. Google Scholar CrossRef Search ADS   27. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A, Speleman F. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol  2002; 3: 1– 11. Google Scholar CrossRef Search ADS   28. Preitner N, Damiola F, Luis Lopez M, Zakany J, Duboule D, Albrecht U, Schibler U. The orphan nuclear receptor REV-ERB; controls circadian transcription within the positive limb of the mammalian circadian oscillator. Cell  2002; 110: 251– 260. Google Scholar CrossRef Search ADS PubMed  29. Cho H, Zhao X, Hatori M, Yu RT, Barish GD, Lam MT, Chong L-W, DiTacchio L, Atkins AR, Glass CK, Liddle C, Auwerx J et al.   Regulation of circadian behaviour and metabolism by Rev-erb-α and Rev-erb-β. Nature  2012; 485: 123– 127. Google Scholar CrossRef Search ADS PubMed  30. Duez H, Staels B. Rev-erbα gives a time cue to metabolism. FEBS Lett  2008; 582: 19– 25. Google Scholar CrossRef Search ADS PubMed  31. Fontaine C, Dubois G, Duguay Y, Helledie T, Vu-Dac N, Gervois P, Soncin F, Mandrup S, Fruchart J-C, Fruchart-Najib J, Staels B. The orphan nuclear receptor Rev-Erbα is a peroxisome proliferator-activated receptor (PPAR) γ target gene and promotes PPARγ-induced adipocyte differentiation. J Biol Chem  2003; 278: 37672– 37680. Google Scholar CrossRef Search ADS PubMed  32. Delezie J, Dumont S, Dardente H, Oudart H, Gréchez-Cassiau A, Klosen P, Teboul M, Delaunay F, Pévet P, Challet E. The nuclear receptor REV-ERB is required for the daily balance of carbohydrate and lipid metabolism. FASEB J  2012; 26: 3321– 3335. Google Scholar CrossRef Search ADS PubMed  33. Wharfe MD, Wyrwoll CS, Waddell BJ, Mark PJ. Pregnancy-induced changes in the circadian expression of hepatic clock genes: implications for maternal glucose homeostasis. Am J Physiol Endocrinol Metab  2016; 311: E575– E586. Google Scholar CrossRef Search ADS PubMed  34. Mark PJ, Crew RC, Wharfe MD, Waddell BJ. Rhythmic three-part harmony: the complex interaction of maternal, placental and fetal circadian systems. J Biol Rhythms  2017, doi: 10.1177/0748730417728671. 35. Everett LJ, Lazar MA. Nuclear receptor Rev-Erbα: up, down, and all around. Trends Endocrinol Metab  2014; 25: 586– 592. Google Scholar CrossRef Search ADS PubMed  36. Li M, Gallo D, Csizmadia E, Otterbein LE, Wegiel B. Carbon monoxide induces chromatin remodelling to facilitate endothelial cell migration. Thromb Haemost  2014; 111: 951– 959. Google Scholar CrossRef Search ADS PubMed  37. Gibbs JE, Blaikley J, Beesley S, Matthews L, Simpson KD, Boyce SH, Farrow SN, Else KJ, Singh D, Ray DW, Loudon ASI. The nuclear receptor REV-ERBα mediates circadian regulation of innate immunity through selective regulation of inflammatory cytokines. Proc Natl Acad Sci USA  2012; 109: 582– 587. Google Scholar CrossRef Search ADS PubMed  38. Hewitt DP, Mark PJ, Waddell BJ. Placental expression of peroxisome proliferator-activated receptors in rat pregnancy and the effect of increased glucocorticoid exposure1. Biol Reprod  2006; 74: 23– 28. Google Scholar CrossRef Search ADS PubMed  39. Mark PJ, Lewis JL, Jones ML, Keelan JA, Waddell BJ. The inflammatory state of the rat placenta increases in late gestation and is further enhanced by glucocorticoids in the labyrinth zone. Placenta  2013; 34: 559– 566. Google Scholar CrossRef Search ADS PubMed  40. Suter M, Bocock P, Showalter L, Hu M, Shope C, McKnight R, Grove K, Lane R, Aagaard-Tillery K. Epigenomics: maternal high-fat diet exposure in utero disrupts peripheral circadian gene expression in nonhuman primates. FASEB J  2011; 25: 714– 726. Google Scholar CrossRef Search ADS PubMed  41. Wang D, Chen S, Liu M, Liu C. Maternal obesity disrupts circadian rhythms of clock and metabolic genes in the offspring heart and liver. Chronobiol Int  2015; 32: 615– 626. Google Scholar CrossRef Search ADS PubMed  42. Borengasser SJ, Kang P, Faske J, Gomez-Acevedo H, Blackburn ML, Badger TM, Shankar K. High fat diet and in utero exposure to maternal obesity disrupts circadian rhythm and leads to metabolic programming of liver in rat offspring. PLoS One  2014; 9: e84209. Google Scholar CrossRef Search ADS PubMed  43. Bruce KD, Szczepankiewicz D, Sihota KK, Ravindraanandan M, Thomas H, Lillycrop KA, Burdge GC, Hanson MA, Byrne CD, Cagampang FR. Altered cellular redox status, sirtuin abundance and clock gene expression in a mouse model of developmentally primed NASH. Biochim Biophys Acta of Lipids  2016; 1861: 584– 593. Google Scholar CrossRef Search ADS   44. Mouralidarane A, Soeda J, Sugden D, Bocianowska A, Carter R, Ray S, Saraswati R, Cordero P, Novelli M, Fusai G, Vinciguerra M, Poston L et al.   Maternal obesity programs offspring non-alcoholic fatty liver disease through disruption of 24-h rhythms in mice. Int J Obes Relat Metab Disord  2015; 39: 1339– 1348. Google Scholar CrossRef Search ADS   45. Carter R, Mouralidarane A, Soeda J, Ray S, Pombo J, Saraswati R, Novelli M, Fusai G, Rappa F, Saracino C, Pazienza V, Poston L et al.   Non-alcoholic fatty pancreas disease pathogenesis: A role for developmental programming and altered circadian rhythms. PLoS One  2014; 9: e89505. Google Scholar CrossRef Search ADS PubMed  46. Pendergast JS, Branecky KL, Yang W, Ellacott KLJ, Niswender KD, Yamazaki S. High-fat diet acutely affects circadian organisation and eating behavior. Eur J Neurosci  2013; 37: 1350– 1356. Google Scholar CrossRef Search ADS PubMed  47. Oike H, Nagai K, Fukushima T, Ishida N, Kobori M. Feeding cues and injected nutrients induce acute expression of multiple clock genes in the mouse liver. PLoS One  2011; 6: e23709. Google Scholar CrossRef Search ADS PubMed  48. Iwanaga H, Yano M, Miki H, Okada K, Azama T, Takiguchi S, Fujiwara Y, Yasuda T, Nakayama M, Kobayashi M, Oishi K, Ishida N et al.   Per2 gene expressions in the suprachiasmatic nucleus and liver differentially respond to nutrition factors in rats. JPEN J Parenter Enteral Nutr  2005; 29: 157– 161. Google Scholar CrossRef Search ADS PubMed  49. Kentish SJ, Vincent AD, Kennaway DJ, Wittert GA, Page AJ. High-fat diet-induced obesity ablates gastric vagal afferent circadian rhythms. J Neurosci  2016; 36: 3199– 3207. Google Scholar CrossRef Search ADS PubMed  50. Garaulet M, Gómez-Abellán P. Timing of food intake and obesity: a novel association. Physiol Behav  2014; 134: 44– 50. Google Scholar CrossRef Search ADS PubMed  51. Kopp R, Billecke N, Legradi J, den Broeder M, Parekh SH, Legler J. Bringing obesity to light: Rev-erb[alpha], a central player in light-induced adipogenesis in the zebrafish? Int J Obes Relat Metab Disord  2016; 40: 824– 832. Google Scholar CrossRef Search ADS   52. Pappa KI, Gazouli M, Anastasiou E, Iliodromiti Z, Antsaklis A, Anagnou NP. Circadian clock gene expression is impaired in gestational diabetes mellitus. Gynecol Endocrinol  2013; 29: 331– 335. Google Scholar CrossRef Search ADS PubMed  53. Lee C-H, Olson P, Hevener A, Mehl I, Chong L-W, Olefsky JM, Gonzalez FJ, Ham J, Kang H, Peters JM, Evans RM. PPARδ regulates glucose metabolism and insulin sensitivity. Proc Natl Acad Sci USA  2006; 103: 3444– 3449. Google Scholar CrossRef Search ADS PubMed  54. Sládek M, Jindráková Z, Bendová Z, Sumová A. Postnatal ontogenesis of the circadian clock within the rat liver. Am J Physiol Regul Integr Comp Physiol  2007; 292: R1224– R1229. Google Scholar CrossRef Search ADS PubMed  55. Pérez S, Murias L, Fernández-Plaza C, Díaz I, González C, Otero J, Díaz E. Evidence for clock genes circadian rhythms in human full-term placenta. Syst Biol Reprod Med  2015; 61: 360– 366. Google Scholar CrossRef Search ADS PubMed  56. Forbes S, Barr SM, Reynolds RM, Semple S, Gray C, Andrew R, Denison FC, Walker BR, Norman JE. Convergence in insulin resistance between very severely obese and lean women at the end of pregnancy. Diabetologia  2015; 58: 2615– 2626. Google Scholar CrossRef Search ADS PubMed  57. King V, Dakin RS, Liu L, Hadoke PWF, Walker BR, Seckl JR, Norman JE, Drake AJ. Maternal obesity has little effect on the immediate offspring but impacts on the next generation. Endocrinology  2013; 154: 2514– 2524. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. Published by Oxford University Press on behalf of Society for the Study of Reproduction. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

Journal

Biology of ReproductionOxford University Press

Published: Jan 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off