In endotherms, plasticity of internal heat production in response to environmental variability is an important component of thermoregulation. During embryogenesis endotherms cannot regulate their body temperature metabolically and are therefore similar to ectotherms. The transition from ectothermy to endothermy occurs by the development of metabolic capacity during embryogenesis. Here we test the hypothesis that the development of metabolism during embryogenesis in birds is under transcriptional control and that metabolic capacity is upregulated in colder environments. The peroxisome proliferator-activated receptor-γ (PPARγ) coactivator-1α (PGC-1α) is the major metabolic regulator in mammals. PGC-1α and its target PPARγ were significantly elevated during development in pectoral muscle and liver of chickens (Gallus gallus) compared with adults. However, the timing of upregulation of PGC-1α and PPARγ was not in synchrony. In cool incubation temperatures (35°C) both PGC-1α and PPARγ gene expression was increased in liver but not in skeletal muscle, compared with a 38°C incubation treatment. Cytochrome c oxidase and citrate synthase enzyme activities and ATP synthase gene expression increased during embryonic development in liver and muscle, and there was a significant effect of incubation temperature on these parameters. Our findings suggest that PGC-1α might be important for establishing endothermic metabolic capacity during embryogenesis in birds.
- peroxisome proliferator-activated receptor-γ
- oxidative capacity
- cytochrome c oxidase
evolution of endothermy in birds and mammals is associated with resting metabolic rates that are five- to tenfold higher than those of ectotherms (5, 6). According to the aerobic capacity model (6), avian and mammalian endothermy has evolved because elevated aerobic capacities favor high and sustained levels of activity. The heat produced by elevated metabolic rates has the additional advantage of uncoupling body temperature from environmental temperature fluctuations, and metabolic heat production is a principal mechanism of thermoregulation in mammals and birds (18). Maintaining a stable body temperature requires a complex regulatory system. Temperature is sensed in the preoptic-anterior hypothalamus (8, 14). This region senses changes in body temperature by receiving afferent sensory input from thermoreceptors distributed throughout the body, including the skin (19, 42, 52), as well as by hypothalamic neurons (62). The hypothalamus and medulla coordinate thermoregulatory responses by controlling the baroreflex (50, 64) and metabolic capacity via the hypothalamic-pituitary-thyroid and adrenal axes (14, 16, 30). Plasticity of internal heat production is an important component of thermoregulation to compensate for increased heat loss in cold environments (35, 36).
During embryogenesis endotherms cannot regulate their body temperature metabolically and are therefore similar to ectotherms. The transition from ectothermy to endothermy occurs with the development of metabolic capacity during embryogenesis (53). Environmental conditions experienced during embryogenesis may affect physiological traits so that physiological capacities are matched with environmental conditions experienced later in life (3, 22). Here we test the hypothesis that the development of metabolism during embryogenesis in birds is under transcriptional control and that metabolic capacity is upregulated in colder environments.
Precocial birds are a perfect model to study environmental influences on the development of metabolic capacity because embryogenesis occurs outside of the mother's womb. Although embryos and juvenile birds have only limited capabilities to regulate their body temperature, early development may be important in modulating metabolic capacity towards environmental demands (11, 56). During development of precocial birds, oxygen consumption increases exponentially until shortly before hatching, when it stabilizes as embryos reach their hatching mass (45, 59). The increase in oxidative metabolic capacity during embryogenesis is facilitated by increasing mitochondrial density and enzyme activity in oxidative pathways (25). Changes in quantity, thermosensitivity, and activity of rate-limiting enzymes of oxidative phosphorylation and citric acid cycle influence metabolic capacity (10, 25, 53).
During ontogeny in birds, temperature variation plays an important role in developmental processes (2, 10). Embryogenesis and hatchling survival depend on incubation temperature, and low incubation temperatures lead to delayed hatching or even death of the embryo. Also, birds exposed to low incubation temperatures during the last days of incubation may be cold acclimated after hatching, which is reflected in elevated thyroid hormone levels and increased heat production (11, 37). Interestingly, while reduced incubation temperatures will delay prenatal development, the development of metabolic capacity might be enhanced so that hatchlings perform better in colder temperatures (11).
The molecular mechanisms that underlie these observations in birds are unknown. In mammals aerobic metabolic capacity is controlled by several nuclear hormone receptors, such as thyroid receptor, retinoic acid receptor, peroxisome proliferator-activated receptors (PPARα, -γ and -δ) and, most importantly, their coactivators (PGC-1α and -β) (39, 47). PGC-1α is a metabolic key regulator in mammals (15) that mediates organisms' energetic responses to changing environments by inducing mitochondrial biogenesis, increasing cellular respiration rates, and facilitating energy substrate uptake and utilization (23, 24). PGC-1α coactivates transcription factors such as nuclear hormone receptors, which control the transcription of genes that encode enzymes involved in metabolic pathways and mitochondrial biogenesis (15, 32, 47, 63). In skeletal muscle of cold-exposed chickens, PGC-1α is upregulated and may be important in fiber-type expression (21, 57). The aim of this study was to investigate the effect of hypothermic incubation on the regulation and plasticity of metabolic capacity during embryonic development. In particular, we investigated whether PGC-1α acts as a metabolic key regulator in birds to adjust metabolic capacity in response to environmental challenges during development.
All procedures were approved by the University of Sydney Animal Ethics Committee (Approval No. L04/7-2006/2/4414). Fertilized chicken eggs (Gallus gallus) were obtained from a local supplier (Baiada Poultry, Sydney, Australia). Eggs were incubated either at a constant temperature of 38°C (normothermic group) or at 35°C (hypothermic group). Relative humidity was regulated at 60–67%, and eggs were turned several times a day. Incubation conditions were monitored by combined temperature and humidity data loggers (HOBO RH/Temp, Onset) throughout the experiment. In the normothermic group internal piping occurred on day 19/20 and hatching on day 20/21 of incubation. In the hypothermic group internal piping occurred on day 22/23 and hatching occurred on day 23/24 of incubation (7). After hatching, the hatchlings were transferred into an enclosure with a constant air temperature of 25°C and a 12:12-h light-dark cycle. Heating lamps were provided, and water and commercial chicken starter feed were available ad libitum.
Samples of both groups (n = 6/time point) were collected during incubation on embryonic days 15, 18, and 20 and on days 4 and 8 after hatching. Additionally, samples of the hypothermic group were harvested on embryonic days 21 and 23 so that developmental stages of the slower-developing hypothermic group could be matched with the faster-growing normothermic embryos. Comparison was made between incubation treatments 6, 3, and 1 days before hatching (referred to as −6, −3, and −1) before as well as 4 and 8 days (+4, and +8) after hatching. The sampling protocol and comparisons between incubation treatments are summarized in Fig. 1. Six adult chickens (452 days old) were obtained separately from a local supplier as a control. Animals were killed by cervical translocation. Breast muscle (musculus pectoralis) and liver was collected and frozen in liquid nitrogen for enzyme analyses or in RNAlater (Ambion) for gene expression analyses.
RNA was extracted from 40–100 mg of liver and muscle samples using TRIreagent (Molecular Research Centre), following the manufacturer's instructions. RNA quality and concentration were verified using a Bioanalyzer (Agilent Biotechnologies). One microgram of total RNA was treated with Dnase I (Sigma) and was reverse transcribed using RNAse H− Maloney murine leukemia virus reverse transcriptase (Bioscript, Bioline) and random hexamer primers (Bioline).
Quantitative RT-PCR was performed on an Applied Biosystems 7500 qRT-PCR machine according to published protocols (53) with the following modifications. Primer and dual-label probes were designed from sequences obtained from GenBank for the mitochondrial-encoded cytochrome c oxidase subunit 1 (COX), the nuclear-encoded ATP synthase H+ transporting mitochondrial F1 complex, β-subunit (Atp5b), the nuclear-encoded PGC-1α, and the nuclear-encoded PPARγ. Also RNA polymerase II polypeptide E (POLR2E) and cytochrome b (Cytb), both nuclear encoded, were used as housekeeping genes. The primer and probe sequences are shown in Table 1.
Real-time PCR reactions contained 1× Immomix (Bioline), 4.5 mM MgCl2, 50–900 nM primer and probe, 1× ROX reference dye (Invitrogen), and ∼50 ng cDNA. The cycle consisted of: 95°C for 7 min, 40 cycles of 95°C for 20 s, and 60°C for 1 min. Relative gene expression of the four target genes at each stage of development was calculated according to Pfaffl et al. (43) and normalized with POLR2E and Cytb. Adult chicken samples (n = 6) were used as control. Different stages of development and incubation treatments were compared by two-way ANOVA on l n transformed data followed by Tukey's post hoc tests. Data were tested for normality and homogeneity of variance using Levene's test. Gene expression was analyzed with the pairwise fixed reallocation randomization test using REST software (44) to compare gene expression during development with adult gene expression.
Liver and muscle tissue were homogenized in 9 volumes of 50 mM imidazole/HCl, 2 mM MgCl2, 5 mM EDTA, 1 mM reduced glutathione and 1% Triton X-100. Assays for COX and citrate synthase (CS) were determined in duplicate at 35°C and 40°C. Enzyme activities were determined according to published protocols (51). Different stages and treatments were compared by using ANOVA with repeated measures (RM-ANOVA), where experimental assay temperatures were set as repeated measure and environmental temperature treatment and stages of development as factors, on ln transformed data followed by Tukey's post hoc test. Data were tested for normality and homogeneity of variance by using Levene's test. Enzyme activity during development was compared with adult samples by RM-ANOVA on ln transformed data following post hoc Tukey's test, and both incubation treatments were analyzed separately.
Transcriptional regulators in liver.
During late embryogenesis (days −3 and −1) PGC-1α expression was significantly elevated (F4,59 = 8.38, P < 0.001) compared with earlier stages (day −6) and after hatching, as well as compared with adults (REST randomization test, P < 0.002; Fig. 2A). Furthermore, there was a significant interaction between incubation temperature and stages of development (F4,59 = 2.78, P < 0.04), and PGC-1α expression was higher at day −3 in the 35°C incubation treatment compared with the 38°C treatment group. Interestingly, PPARγ gene expression did not vary significantly during prenatal embryogenesis, but it increased significantly (F4,59 = 22.67, P < 0.001) after hatching (days +4 and +8; Fig. 2B). In addition, between days −1 to +4, PPARγ gene expression was significantly higher in the 35°C incubation treatment compared with the 38°C treatment group (F1,59 = 3.25, P < 0.02). Furthermore, there was a significant interaction between incubation temperature and stages of development (F4,59 = 3.25, P < 0.02).
Metabolic capacity parameters in liver.
Gene expression levels of Fo/F1 Atp5b were significantly elevated (F4,59 = 3.48, P < 0.015) during early prenatal development (day −6) compared with day +8 and adults (P < 0.002; Fig. 3A). Also, mRNA levels of Atp5b (F1,59 = 10.76, P < 0.003) decreased in the 35°C incubation treatment 8 days after hatching. There was no significant interaction between incubation temperature and stages of development (F4,59 = 1.07, P > 0.1).
COX mRNA expression remained constant regardless of incubation temperature or developmental stage (F1,59 = 0.71, P > 0.05; F1,59 = 0.062, P > 0.1, respectively; Fig. 3B). Interestingly, COX enzyme activity followed a different pattern to its gene expression (Fig. 3C). Activity increased significantly (F4,50 = 3.18, P < 0.025) between days −6 to +8, when it reached adult levels. Although the 35°C incubation treatment had no effect on COX gene expression (F4,59 = 0.062, P > 0.1), COX activity was significantly lower in the 35°C incubation group compared with the 38°C treatment (F1,50 = 4.40, P < 0.05). There was no significant interaction between incubation temperature and stages of development (F4,50 = 0.31, P > 0.1).
CS activity was significantly greater (F4,50 = 12.27, P < 0.001) between days −1 and +8 compared with earlier developmental stages and to adults (Fig. 3D). Incubation at 35°C resulted in significantly greater CS activity after hatching (F4,50 = 4.04, P < 0.007), compared with the 38°C incubation treatment. There was no significant interaction between incubation temperature and stages of development (F4,50 = 0.35, P > 0.1).
There was a significant positive relationship between CS activity and PPARγ gene expression in the liver of the 38°C treatment group (linear regression, F1,4 = 81.13, P < 0.004, r2 = 0.96) and the 35°C treatment group (linear regression, F1,4 = 37.05, P < 0.01, r2 = 0.93; Fig. 4). However, there were no significant relationships between PPARγ gene expression and PGC-1α gene expression, COX gene expression or COX activity, or between PGC-1α gene expression and COX gene expression, COX activity or CS activity (all F1,4 < 5.00, P > 0.1).
Transcriptional regulators in skeletal muscle.
PGC-α gene expression was elevated (F4,59 = 10.47, P < 0.001) during prenatal development (days −3 to −1) compared with after hatching (days +4 to +8; Fig. 5A). However, unlike in liver, incubation temperature had no effect on PGC-α gene expression (F1,59 = 0.31, P > 0.05). There was no significant interaction between incubation temperature and stages of development (F4,59 = 2.54, P > 0.05). PPARγ mRNA levels were significantly elevated during prenatal development from day −6 to day −1 compared with adults (P < 0.002; Fig. 5B), but incubation temperature had no influence on PPARγ gene expression (F1,59 = 0.21, P > 0.05). There was no significant interaction between incubation temperature and stages of development (F4,59 = 1.54, P > 0.05). PGC-1α and PPARγ gene expression are not related to each other during development in skeletal muscle (linear regression F1,4 < 2.00, P > 0.2, r2 = 0.39).
Metabolic capacity parameters in skeletal muscle.
ATP synthase mRNA level was elevated during development compared with adults (P < 0.002; Fig. 6A), but gene expression of Atp5b was significantly lower in the 35°C incubation treatment compared with 38°C (F1,59 = 15.25, P < 0.001). Also there was a significant interaction between incubation temperature and stages of development (F4,59 = 6.88, P < 0.001), and ATP5b gene expression level was higher in the normothermic treatment group compared with hypothermic treatment group (F4,59 = 6.88, P < 0.001) 8 days after hatching.
COX gene expression remained constant during development (F1,59 = 1.53, P > 0.05); there was no effect of incubation temperature (F1,59 = 1.49, P > 0.05; Fig. 6B) and also no significant interaction between incubation temperature and stages of development (F4,59 = 0.87, P > 0.1). Unlike gene expression, COX enzyme activity increased significantly (F4,50 = 34.98, P < 0.001) after hatching (days +4 to +8) compared with earlier stages of development (Fig. 6C). Also there was no significant interaction between incubation temperature and stages of development (F4,50 = 1.24, P > 0.1). CS showed a similar pattern in activity as COX (Fig. 6D), and activity increased after hatching compared with earlier stages (F4,50 = 9.49, P < 0.001). However, incubation temperature did not affect CS enzyme activity in skeletal muscle (F1,50 = 1.99, P > 0.05), and there was no significant interaction between incubation temperature and stages of development (F4,50 = 1.15, P > 0.1).
Furthermore, there were no significant relationships between PPARγ and COX gene expression, COX activity or CS activity, or between PGC-1α gene expression and COX gene expression, COX activity or CS activity (all F1,4 < 5.00, P > 0.1).
We report a novel expression pattern of the mammalian metabolic regulator PGC-1α in the embryonic development of a bird. In liver, one of the main sites for metabolic heat production, PGC-1α gene expression is upregulated during late embryogenesis. At that developmental stage the transcriptional machinery needs to coordinate the switch from anaerobic metabolism to oxidative metabolism which precedes hatching (53). Additionally, the embryo is not growing during the last 20% of development in the egg (59), so that PGC-1α upregulation is unlikely to be in response to growth. In mammals, PGC-1α plays a central role in mitochondrial biogenesis and coordinates nuclear and mitochondrial-encoded gene expression (29, 47, 49).
It is known that glucagon and catecholamines activate PGC-1α gene expression in hepatocytes via cAMP and cAMP response element binding protein (CREB) signaling to regulate gluconeogenic gene expression (20, 65). Gluconeogenesis, mostly from lactate (13), is highly activated during embryogenesis in birds to produce glucose. In contrast, in mammals glucose is transferred from the mother to the embryo (61). Hence, PGC-1α gene expression may activate gluconeogenesis during embryogenesis in birds.
An incubation temperature of 35°C leads to increased PGC-1α gene expression. In rats, cold stimulates PGC-1α gene expression to increase mitochondrial biogenesis (48). Similar to stimulation by glucagon and catecholamines, cellular responses to cold are mediated via β3-adrenergic receptors at the cell membrane to stimulate cAMP and CREB signaling pathways, which activate transcription of key regulators like PGC-1α (48). However, gene expression patterns of PPARγ differ from those of PGC-1α, which indicates that it is unlikely that these molecules interact during development in liver. PPARγ gene expression strongly increases after hatching. This transcription factor is known to be involved in several metabolic regulatory pathways especially in lipid and glucose metabolism (12, 60). Lipid metabolism in birds differs from mammals. In chicken, 90% of fatty acid synthesis occurs in liver, and adipose tissue only functions as storage tissue. In mammals, however, fatty acid synthesis takes place in liver and adipose tissue (31, 41). Furthermore, the diet of the developing embryo is 90% fat-based yolk, and it changes after hatching to a carbohydrate-based diet (54). Hence, key enzymes of glycolysis show low activities during embryogenesis but strongly increases after hatching caused by glucose availability (53, 61). PPARγ is involved directly in the glucose-sensing apparatus in liver by stimulating transcription of glucosetransporter 2 and glucokinase, two key proteins in hepatic glucose metabolism and lipogenesis in liver (26–28). Hence, the increase in PPARγ gene expression after hatching would confirm its role in hepatic glucose metabolism in birds. In mammals other members of the PPAR family, especially PPARα, play a prominent role in lipid and carbohydrates metabolism, as well as in mitochondrial biogenesis in hepatocytes (34). Also, PGC-1α may interact directly with PPARα in liver to coordinate gene expression during development in birds rather than with PPARγ. Therefore, future research could examine the importance of PPARα during hepatic development in chicken.
ATP synthase gene expression is constantly elevated relative to adults during embryonic growth, reflecting the demands to sustain mitochondrial proliferation as well as to increase mitochondrial densities within cells. COX is a rate-limiting enzyme in the electron transport chain (55) and, interestingly, COX gene expression does not correlate with the increase in enzyme activity, which illustrates that this enzyme is controlled by other mechanism than transcription. COX is a membrane-bound enzyme and therefore depends on membrane fatty acid composition. Differences in membrane composition influence enzyme activities independently from enzyme concentration (25). Shortly before hatching, polyunsaturated fatty acids and phospholipid concentrations increases in liver membranes (40), which may influence membrane-bound enzymes, like COX. Additionally, the increase in arachidonic and docosahexaenoic acids (40) promote membrane proton flux and therefore endothermic heat production (9).
Eight days after hatching, CS activity increased in the cold-treatment group, while COX activity and ATP synthase mRNA levels decreased compared with the normothermic treatment group. The TCA cycle not only provides the cell with FADH2 and NADH, which are well-known substrates of the electron transport chain, but also with precursors for amino acid synthesis. Hence, upregulation of the TCA cycle may be indicative of elevated protein synthesis in the 35°C treatment group. Colder incubation temperature leads to greater growth after hatching (11), which would also require elevated rates of protein synthesis. The putative role of CS in protein synthesis is speculative, but it may warrant further investigation.
The upregulation of PGC-1α during prenatal development in muscle may be linked to orchestrating mitochondrial biogenesis and fiber-type determination (4, 33, 49). In mammalian muscle, PGC-1α−/− mice show a 30%–50% reduction in the expression of genes important in oxidative phosphorylation, fatty acid oxidation, and ATP synthesis, which clearly demonstrate the prominence of this molecule to establish metabolic capacity in skeletal muscle (1). The lack in correlation between the PGC-1α and metabolic enzymes in liver and skeletal muscle indicates that this coactivator does not influence transcription of COX1 and Atp5b directly. Also, posttranslational modification of the PGC-1α protein may lead to increased activity, independently of mRNA upregulation (46).
PPARγ gene expression in muscle is consistently elevated during embryogenesis. As in liver, this expression pattern suggests that PPARγ is not a direct target of PGC-1α during development in birds. However, transcriptional processes are highly complex and may interact indirectly. Different transcription factors or coactivators might be transcribed at different times, and protein concentrations may be altered by translational control, processing, and posttranslational modifications or activation (17).
As in liver, ATP synthase gene expression is elevated, which reflects the demands to perpetuate mitochondrial proliferation and increasing cristae density during development (55). Unlike in liver, however, COX and CS activity pattern increase in parallel after hatching. Hence, metabolic development in muscle is complete only after hatching, at a time when hatchlings need to use skeletal muscles for locomotion. Muscle fiber-type composition, contractile proteins, and metabolic proteins change with activity level to ensure the muscle energy supply (4). Accordingly, cold temperature treatment had no effect on enzyme activity in muscle because plasticity towards metabolic demands may start after hatching.
Cold incubation influences gene expression, which affects the development and plasticity of metabolic capacity in birds. Our findings that qualitative markers of metabolic capacity are decreased in the cold are in agreement with earlier studies that constant cold incubation leads towards decreased thermogenic responses in embryos and 1-day hatchlings (7, 38). But we also found that transcriptional parameters, as well as CS activity, are elevated, which may indicate that hatchlings initially risk a lower thermogenic capacity but boost their metabolic capacity in response to environmental demands later during postnatal growth.
In conclusion, even though endothermy has developed independently in mammals and birds, molecular mechanisms that control development of metabolic capacity are conserved. The differential expression of PGC-1α during development and in response to cold indicates that, as in mammals, it plays a role in regulating metabolism in birds. It will be an interesting field for future research to investigate potential candidates, other than PPARγ, which interact with PGC-1α to coordinate the establishment and plasticity of endothermic metabolic capacity in birds. Interestingly, the upregulation of PGC-1α coincides with the upregulation of plasma triiodothyronine (T3) levels (58). Also, it is known that plasma T3 concentration increases in response to cold incubation during development in chicken (11). Hence, elevated PGC-1α mRNA during the 35°C incubation treatment may be associated with increased T3 level. To understand the mechanisms that lead to the development of endothermy, it will be important to determine how thyroid hormone regulates PGC-1α gene expression in birds.
The work was supported by an Australian Research Council Discovery Grant (to F. Seebacher).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society