Maternal Intake Restriction Programs the PKA-CREB Pathway to Regulate Energy Metabolism, CLOCK and mTOR Signals in the Skeletal Muscles of Goat Offspring

The biological mechanism about that maternal undernutrition increases the metabolic disorder risk of skeletal muscles in offspring is less known. We hypothesize that maternal intake restriction inuences metabolic signals in the skeletal muscles of offspring via a glucagon-mediated pathway. Twenty-four pregnant goats were assigned to the control (100% of the nutrients requirement) and restricted (60% of the control from pregnant days of 45 to 100) groups. Blood and longissimus thoracis muscle were sampled from dams, fetuses and kids in each group. Intake restriction reduced the total blood protein of dams and fetuses. Maternal restriction decreased the CREB1, CREBBP, PKA, BMAL1, AKT1, mTOR, and RPTOR mRNA expressions in the fetuses, reduced the CREBBP, NR1H3, DBP and PKA mRNA levels in the kids, but increased the PGC1α and TSC2 mRNA levels in the fetuses, while the mRNA expression of CLOCK and TSC2 genes was increased in the restricted kids. The protein expression of total PKA and phosphorylated PKA of the restricted fetuses and kids were downregulated, while the protein expression of total mTOR and phosphorylated mTOR were reduced in the restricted fetuses and kids. and PKA-CREB reveals a molecular pathway that maternal undernutrition leads to metabolic adaptation of skeletal muscle in offspring. in the offspring (27) and a reduction in the muscle mass in the kid goats (28). Glucagon inuences tissue metabolism through the classical cAMP-PKA-CREB pathway in peripheral tissues (29), and CREB is a key regulator in energy metabolism (30), protein synthesis and circadian control (31, 32). To test the hypothesis that whether maternal intake restriction programs the PKA-CREB pathway of the skeletal muscle in the offspring to affect metabolic signals, we investigated the effects of 40% maternal intake restriction during midgestation on the concentrations of blood metabolites, muscular cAMP and glycogen, and on the expression of genes involved in the mTOR, AMPK and CLOCK pathways. The ndings will give new insights into understanding the metabolic adaptation caused by maternal undernutrition in humans. the mRNA expression of AKT1, mTOR and RPTOR and the protein expression of mTOR and p-mTOR in the LT muscles of fetuses, but the mRNA expression of TSC2 was upregulated. The expression of TSC1 and TSC2 mRNA and p-TSC2 protein in the LT muscles of kids was upregulated by maternal restriction, while the protein expression of mTOR and p-mTOR was downregulated. These results showed that maternal restriction altered mTOR pathway signaling in both the fetuses and kids. Similarly, a decrease in the p-mTOR protein of the fetal LT muscle was observed under 50% maternal restriction during early to midgestation (21). Since the TP level was decreased, while the blood glucose, insulin, IGF1 (27), and muscular glycogen levels in the offspring of the present study were unaffected, the reason may be ascribed to the reduction in the overall protein or AA supply under intake restriction when energy is decient. A previous study in intrauterine growth restricted (IUGR) sheep has shown that AA perfusion reduces the rate of protein degradation and increases protein deposition by 150%, and the AA level independently regulates mTOR pathway signaling (49). In addition to energy, 40% maternal intake restriction may aggravate the lack of protein and AAs, upregulated the expression of genes involved in PKA-CREB and mTOR signals, downregulated the ACOX1 mRNA expression, and affected the CLOCK pathway in the LT muscles of nutrition-restricted fetuses and kids consistently. We inferred that maternal intake restriction during midgestation programmed the PKA-CREB pathway in the skeletal muscle of offspring to upregulate fat oxidation, downregulate protein synthesis, and alter the circadian clock. These results reveal the role of the PKA-CREB pathway in metabolic programming in the skeletal muscles of offspring exposed to intrauterine malnutrition, which deepened our understanding of the molecular mechanism of metabolic adaptation in skeletal muscle caused by maternal undernutrition in humans and mammals.

black goat, local breed) were selected and randomly assigned to a control [C, 100% of the nutrients requirement suggested in the feeding standard of meat-producing sheep and goats of China (2004), n=12] or a feed intake-restricted (R, 60% of the control, n=12) group. The restriction period was from 45 to 100 d of gestation. The dam weight, animal management and diet composition were the same as those in our previous study (27). At 100 d of gestation, after examination with ultrasonography (Aloka SSD-500 with a 5-MHzlinear probe Aloka, Shanghai, China), six dams from each group were randomly chosen for slaughtering, and the samples from dams and their fetuses were collected. The remaining dams were realimented to 100% of nutrients requirement after 101 d of pregnancy. After delivery, the newborn goats were naturally breastfed until 50 d postpartum. Between 50 and 60 d of age, the kids were preweaned. After complete weaning at 60 d of age, the diet containing milk replacer and fresh grass was given until 90 d of age. The management measures and diet composition for kids were described previously (28). At 90 d of age, sixteen kids (8 of each group) were slaughtered and samples were collected.
Blood and tissue sampling After fasting overnight, the dams and kids were electrically shocked, and blood was collected from the jugular veins. Plasma was anticoagulated with heparin sodium and separated through centrifugation at 1200 × g for 10 min at 4°C and stored at -80°C for subsequent analysis. Following exsanguination and ventrotomy of the dams, fetal blood samples were collected from the umbilical cord, and the plasma was separated as described above. The hot carcass weight of animals were weighed after removal of the skin, head, hoofs, tail and visceral organs except kidneys, and samples of the longissimus thoracis (LT) muscle between the 5 th and 8 th ribs were sliced and snap-frozen in liquid nitrogen and then stored at -80°C until further analysis.

Blood biochemical parameters and free amino acid (AA) analysis
After plasma samples were thawed at 4°C, the total protein (TP) concentration was determined using assay kits (Beijing Leadman Biochemistry Company Limited, Beijing, China) with an automatic biochemical analyser (Hitachi 7600, Hitachi, Ltd., Tokyo, Japan). The concentrations of non-esteri ed fatty acid (NEFA) were measured using assay kits (Nanjing SenBeiJia Biological Technology Co., Ltd., Nanjing, China). The concentrations of triiodothyronine (T3) and thyroxine (T4) were measured using an enzyme-linked immunosorbent assay according to the manufacturer's instructions (Cusabio Biotech Company Limited, Wuhan, China). The intra-and inter-assay coe cient variations of T3 and T4 were all below 10%.
The analysis of free AAs in blood was determined as described earlier (33). Brie y, 500 μL of plasma sample was taken and mixed with 500 μL of 0.30 mol/L sulfosalicylic acid solution. The mixed solution was placed at 4°C overnight and centrifuged at 10000 × g for 10 min at 4°C. The supernatant was collected and ltrated using a 0.22 μm microporous membrane. The ltrate was measured using an L-8800 automatic amino acid analyser (L-8800, Hitachi, Tokyo, Japan).

Glycogen and cAMP determination
The glycogen (Nanjing Jiancheng Bioengineering Research Institute, Nanjing, China) and cAMP (NewEast Biosciences, Malvern, USA) in the LT muscle were determined according to the manufacturer's instructions. The preparation procedure was performed as previously described (27), and the results were expressed as mg/g protein in the fresh tissue.

Quantitative RT-PCR
The total RNA of sample was extracted using precooled TriQuick Reagent (Solarbio, Beijing, China) according to the manufacturer's instructions. The expression of the target mRNAs was analysed using the SYBR green method with a LightCycler® 480 system (Roche Applied Science, Basel, Switzerland), and the detailed method was described previously (27). The speci c gene primers are listed in Table 1. Relative gene expression levels were normalized to the levels of the reference gene ACTG1 using the 2-ΔΔCt method (34), where Ct denoted the threshold cycle.

Statistical analysis
The data were analysed using the linear MIXED model of SPSS 19.0 (IBM SPSS, Inc., 2010). The data from dams were tested with the nutritional level as a xed effect and the initial weight of the dams as a covariate. The data from the offspring were analysed with the nutritional level, litter size and offspring sex as xed factors. After there are signi cant differences among xed factors, the multiple comparison method of SIDAK was applied. The signi cant difference was set at P < 0.05, and a signi cant trend was considered at 0.05 ≤ P ≤ 0.10. The results were expressed as the means and standard errors of the means (SEM).

Carcass weight and blood biochemical indices
The hot carcass weight of the R dams was less than that of the C dams (P = 0.009, Table 2). The hot carcass weight of the R fetuses was not affected by maternal restriction (P = 0.182). The hot carcass weight was less in the R kids than that in the C kids (P = 0.017). The NEFA concentrations of the R dams increased (P = 0.007). The TP concentrations of R dams and R fetuses decreased (P < 0.05). The T4 and T3 concentrations in the dams, fetuses, and kids were not affected by the treatment. The glucagon concentrations were previously reported (27), which increased in the R dams (P = 0.002). Sex and litter size affected the TP of the fetuses and kids (P < 0.05). The TP concentration of the male fetuses (28.87 g/L) were higher (P = 0.045) than the females (27.58 g/L), as well as in kids (68.37 g/L in males vs. 62.89 g/L in females, P = 0.009). TP concentration of fetuses was decreased (P = 0.007) in sequence in singleton (32.03 g/L), twins (27.56 g/L) and triplets (25.09 g/L), while its concentration of kids was lower in twins (59.81 g/L) than in singleton (69.81 g/L) and triplets (67.26 g/L). Litter size affected the glucagon of fetuses (P = 0.018), which increased in order in singleton (31.53 ng/L), twins (32.80 ng/L) and triplets (39.10 ng/L).

Blood amino acid pro le
In the R dams (Table 3), the Ser concentration was increased (P = 0.046). In the R kids, the concentration of Thr increased (P = 0.011). Other detected AAs in the dams and kids were not in uenced by treatment (P > 0.05). Litter size affected the Val, Met and Tyr concentrations of the kids (P < 0.05), which were all higher (P < 0.05) in triplets than those in singleton and twins, but sex had no in uence (P > 0.05) on the AA pro le of the kids. Glycogen and cAMP in the muscles of offspring Maternal restriction did not alter the cAMP and glycogen concentrations in the LT muscles of the offspring (Table 4). Litter size affected (P = 0.006) the cAMP concentration of the kids, which was higher (P < 0.05) in triplets than those in singleton and twins. Sex and litter size did not in uence (P > 0.05) the cAMP and glycogen levels of the fetuses. The mRNA and protein expression of the energy metabolism-related genes in the LT muscles The mRNA expression of ACOX1, NR1H3, GR and PRKAB1 in the LT muscles of R dams was upregulated (P < 0.05, Fig. 1A), while the PCK2 mRNA tended to be decreased (P = 0.075). The expression of AMPKa and STK11 proteins in the R dams was not affected (P > 0.05, Fig. 1B).
Maternal intake restriction upregulated the fetal PGC1A mRNA (P = 0.018, Fig. 2A) and tended to increase the fetal ACOX1 mRNA (P < 0.10). The expression of NR1H3 mRNA was downregulated in the R kids (P < 0.05, Fig. 2B), but the ACOX1 mRNA in the R kids tended to be increased (P = 0.092). Sex and litter size did not affect the mRNA expression of these detected genes in the fetuses (P > 0.05). Sex affected (P < 0.05) the mRNA expression of PCK1 and PRKAB1 in the kids, while litter size had no effect (P > 0.05) on the mRNA expression of these genes. Maternal intake restriction tended to increase (P < 0.10) the p-STK11 protein abundance in the fetuses, and the ratio of p-STK11/STK11 tended to increase (P < 0.10) in the restricted fetuses and kids ( Fig.  2C and D). The p-AMPKα and AMPKα protein expression of the R fetuses and R kids were not affected (P > 0.05). Litter size affected the p-AMPKα/AMPKα level in the fetuses (P = 0.028), which was lower in triplets than those in singleton and twins. Sex had no effect (P > 0.05) on the protein expression in the offspring.
Maternal intake restriction reduced the CREB1, CREBBP and PKA mRNA expression in the fetuses (P < 0.05, Fig. 3A). The expression of CREBBP and PKA mRNA were downregulated in the R kids (P < 0.05, Fig. 3B), and the CREB1 mRNA in the R kids tended to be reduced (P = 0.099). The expression of the p-PKA and PKA proteins in both the fetuses and kids was downregulated (P < 0.05, Fig. 3C and D) by maternal intake restriction. Litter size affected the p-PKA protein expression in the kids (P = 0.019), which was lower in singleton than those in twins and triplets. Sex had no effect (P > 0.05) on the protein expression in the offspring.
The mRNA expression of circadian CLOCK pathway genes in the LT muscles Maternal intake restriction reduced the fetal BMAL1 mRNA expression (P = 0.043, Table 5), tended to reduce (P = 0.051) the fetal CRY2 mRNA level, and tended to increase (P = 0.068) the fetal DBP mRNA level. Maternal intake restriction decreased (P = 0.049) the DBP mRNA expression but increased (P = 0.019) the CLOCK mRNA expression in the kids. Litter size affected the BMAL1 mRNA expression in the kids (P = 0.023), which was higher in triplets than that in twins. Sex had no in uence (P > 0.05) on the mRNA expression of these circadian genes in the fetuses and kids. The mRNA and protein expression of the mTOR pathway Maternal intake restriction downregulated (P < 0.05, Fig. 3A) the AKT1, mTOR, and RPTOR mRNA expression in the LT muscles of the fetuses and upregulated (P = 0.049) the fetal TSC2 mRNA expression. Maternal intake restriction upregulated (P = 0.01, Fig. 3B) the mRNA expression of TSC2 in the kids and tended to increase (P = 0.069) the mRNA expression of TSC1. Sex and litter size did not affect (P > 0.05) the mRNA expression of the above genes in the fetuses and kids.
Maternal intake restriction reduced (P < 0.05, Fig. 3C and 3D) the protein expression of mTOR and p-mTOR in the LT muscles of the fetuses and kids. The protein expression of p-TSC2 and the ratio of p-TSC2/TSC2 in the R kids tended to be upregulated (P < 0.10), while the ratio of p-mTOR/mTOR tended to be decreased (P = 0.095). Sex affected (P = 0.019) fetal p-mTOR protein expression, litter size affected (P < 0.05) fetal mTOR and p-mTOR/mTOR expression, fetal mTOR level in the triplets was higher (P = 0.017) than those in the singleton and twins, while fetal p-mTOR/mTOR in the singleton was higher (P = 0.014) than those in the twins and triplets. Sex also affected (P = 0.046) the mTOR protein expression in the kids, which were higher in the female kids than that in the males. Litter size affected (P < 0.05) the expression of mTOR and p-mTOR, and they were both lower in the triplets than those in the singleton, while they were intermediate in the twins.

Discussion
Food deprivation in humans and mammals is common in underdeveloped areas, and nutrient restriction during gestation programs the muscle development and metabolism of the offspring (3,8), but the mechanism remains unclear. In the present study, the 40% intake restriction reduced the carcass weight of dams and kids, and reduced the plasma TP contents of dams and fetuses. Maternal malnutrition did not affect the cAMP and glycogen contents of the offspring muscles, while the mRNA expression of genes associated with fatty acid oxidation and regulation in the mother and offspring were upregulated, and the mRNA and protein expression of genes involved in the mTOR pathway in offspring were downregulated. Moreover, the mRNA expression of the CLOCK pathway was affected in offspring.
Intake restriction reduced the hot carcass weight of the dams, indicating that the maternal muscle mass declined. The TP concentrations in the maternal and fetal plasma was reduced, whereas the venous AA pro le of dams and kids, the AA pro le in the umbilical cord blood of fetal goats that was reported previously (33), and the glucose and insulin concentrations that we have reported (27) were not affected. These results suggest an insu cient supply of protein in the dams and fetuses. However, the 40% maternal restriction did not reduce the weight and birth weight of the fetuses, while the body weight and carcass weight of the kids after birth were reduced (28). These results suggest in utero maternal compensation and defective muscle growth and development programming after birth.
ACOX1 is the rst rate-limiting enzyme of fatty acid β-oxidation, which catalyzes acyl-CoA into 2-trans-enoyl-CoA. NR1H3 (also known as LXRα) regulates fat homeostasis and plays an important role in cholesterol metabolism and lipid synthesis (35,36). Glucocorticoid receptor (GR) is associated with the stress response caused by a high level of cortisol under starvation, which is a common target regulator of intrauterine metabolic programming (37). PRKAB1 is a regulatory subunit in the AMPK protein complex that monitors cellular energy status (38). The increase in the blood NEFA concentration and the ACOX1, NR1H3, GR and PRKAB1 mRNA expression in the restricted dams indicated the change in lipid metabolism and the upregulation of fatty acid oxidation in the LT muscle. PCK2 in muscle tissues acts to catalyze the conversion of oxaloacetate into phosphoenolpyruvate, by which the oxaloacetate availability of the tricarboxylic acid (TCA) cycle and glucose homeostasis are regulated (39). Downregulation of PCK2 mRNA in muscle implied the reduced glucose availability in the muscle tissue and the downregulated TCA cycle oxidation-energy pathways. Therefore, intake restriction probably resulted in the stimulation of fatty acid oxidation and the suppression of glucose oxidative degradation in the LT muscles of the dams.
As one of the target organs for intrauterine metabolic programming (3,40), muscle tissue is involved in blood glucose uptake and clearance and the regulation of glucose and lipid metabolism homeostasis. In our study, maternal restriction did not alter the cAMP and glycogen concentrations of LT muscles in the offspring. The mRNA expression of G6PD, PCK1, PCK1, PRKAA2 and PRAKB1 and the protein expression of AMPKα in the offspring were also unaffected in our study. These results were in line with the ndings in lambs after 50% intrauterine restriction during early to midgestation (21). We speculated that 40% maternal intake restriction during midgestation did not alter the glucose storage capacity and consumption in the LT muscles of the fetuses and kids. However, maternal intake restriction increased the mRNA expression of ACOX1 in the LT muscles of the fetuses and kids, and these results were consistent with the mRNA expression of ACOX1 in the dams. Moreover, the mRNA expression of PGC1A in the fetuses and the protein expression of p-STK11/STK11 in the fetuses and kids were upregulated, while the NR1H3 mRNA expression was downregulated in the kids. PGC1A promotes mitochondrial oxidative metabolism (41), while p-STK11 stimulates the phosphorylation of ACOX1 (42) and inhibits fat synthesis. The increase in the ACOX1 and PGC1A mRNA and the p-STK11/STK11 protein in the restricted fetuses and kids indicate that maternal restriction upregulated the lipolysis of muscle tissue and fatty acid β-oxidation as an energy source. Similarly, a 40% maternal feed restriction also upregulated the mRNA expression of PGC1A in the LT muscles of fetal calves during mid-to late gestation (43).
Maternal restriction simultaneously reduced the mRNA expression of CREB1 and CREBBP in both the fetuses and kids. Recent studies have found that CREB interacts with PGC1A, NR1H3 and DBP to regulate glucose synthesis, gluconeogenesis (30), and fat metabolism (44) in mouse livers. CREB also alters the main clock in the supraoptic nucleus of mice by phosphorylation at Ser133 (31), while CREB expression is regulated by the rhythm clock genes to maintain metabolic rhythms. Study in the mouse muscle had revealed that knockout of the rhythmic gene BMAL1 leads to disturbances in the transcription of glucose, fat and protein metabolism-related genes (45). Consistent with the change in the CREB and CREBBP regulators in the offspring of this study, maternal restriction had a tendency to increase DBP mRNA in the fetal muscle, while the DBP mRNA expression was decreased in the kids. Moreover, maternal restriction reduced the mRNA expression of BMAL1 and CRY2 in the fetuses, while the mRNA expression of CLOCK was increased in the kids. In the rhythmic CLOCK signaling pathway, DBP activates the transcription of PER, while PER inhibits the expression of CLOCK and BMAL1 by forming a dimer with CRY (46). The mRNA expression patterns of DBP, BMAL1 and CLOCK in the fetuses and kids in this study were consistent with the CLOCK signaling transmission pathway. Of note, CLOCK plays a role in histone acetylation (47), and this epigenetic mechanism is closely related to developmental programming. We proposed that maternal intake restriction programmed the CREB-CREBBP regulatory factors to alter the mRNA expression of energy metabolism and the CLOCK pathway in the offspring.
The AKT-TSC-mTORC1 signaling pathway is one of the main regulators of muscle protein synthesis, which occurs in response to cellular energy, AA, insulin and insulin-like growth factor 1 (IGF1) signals (15,48). Maternal restriction downregulated the mRNA expression of AKT1, mTOR and RPTOR and the protein expression of mTOR and p-mTOR in the LT muscles of fetuses, but the mRNA expression of TSC2 was upregulated. The expression of TSC1 and TSC2 mRNA and p-TSC2 protein in the LT muscles of kids was upregulated by maternal restriction, while the protein expression of mTOR and p-mTOR was downregulated. These results showed that maternal restriction altered mTOR pathway signaling in both the fetuses and kids. Similarly, a decrease in the p-mTOR protein of the fetal LT muscle was observed under 50% maternal restriction during early to midgestation (21). Since the TP level was decreased, while the blood glucose, insulin, IGF1 (27), and muscular glycogen levels in the offspring of the present study were unaffected, the reason may be ascribed to the reduction in the overall protein or AA supply under intake restriction when energy is de cient. A previous study in intrauterine growth restricted (IUGR) sheep has shown that AA perfusion reduces the rate of protein degradation and increases protein deposition by 150%, and the AA level independently regulates mTOR pathway signaling (49). In addition to energy, 40% maternal intake restriction may aggravate the lack of protein and AAs, leading to the downregulation of the mTOR pathway in offspring, which affects muscle tissue protein synthesis and muscle mass. Moreover, recent studies found that rhythmic per protein regulates the mTORC1 signaling pathway by recruiting TSC1 in the mouse (32), while the rhythmic factor BMAL1 is also tightly linked to mTOR pathway protein synthesis via S6K (50). These ndings are consistent with the change in the CLOCK and mTOR pathways in this study; and these results suggest the connection among intrauterine malnutrition, rhythm disruption, and protein synthesis in the skeletal muscles of the offspring.
Furthermore, both the mRNA and protein expression levels of PKA were decreased in the offspring from the restricted group in this study. CREB is located downstream of the PKA factor (51), and PKA phosphorylates raptor to regulate mTORC1 (52). It is reasonable to conclude that maternal intake restriction alters the PKA-CREB pathway to regulate energy metabolism, CLOCK signaling and protein synthesis and leads to metabolic programming in the LT muscles of the offspring. The classical pathway for the regulation of glucose metabolism under energy-de cient conditions is the glucagon-cAMP-PKA pathway (29). Elevated glucagon caused by intake restriction in dams was observed, but the cAMP and glycogen concentrations in the LT muscles of offspring were not altered. Intermediate mediators between the high level of maternal glucagon and the downstream PKA-CREB pathway need to be identi ed.
The effects of sex and litter size on the phenotype and metabolism of the skeletal muscles of mammals has been reported (53-56). In this study, the effects of litter size on blood and tissue metabolites, and genes expression of offspring were also observed, such as TP, glucagon, Val, Met, Tyr and cAMP concentrations, and mTOR protein expression. Intake restriction of dams apparently leads to a more severe protein de ciency in triplets than those of singleton and twins. Sex also in uenced the blood TP concentration and mTOR (or p-mTOR) protein expression in offspring. The effects of sex and litter size on protein metabolism in the IUGR offspring need further investigation.

Conclusions
Using the goat model in this study, maternal 60% energy-protein undernutrition upregulated the expression of genes involved in PKA-CREB and mTOR signals, downregulated the ACOX1 mRNA expression, and affected the CLOCK pathway in the LT muscles of nutrition-restricted fetuses and kids consistently. We inferred that maternal intake restriction during midgestation programmed the PKA-CREB pathway in the skeletal muscle of offspring to upregulate fat oxidation, downregulate protein synthesis, and alter the circadian clock. These results reveal the role of the PKA-CREB pathway in metabolic programming in the skeletal muscles of offspring exposed to intrauterine malnutrition, which deepened our understanding of the molecular mechanism of metabolic adaptation in skeletal muscle caused by maternal undernutrition in humans and mammals.

Declarations
Ethics approval and consent to participate All the protocols used in this study were approved by the Animal Care Committee according to the Animal Care and the Use Guidelines of the Institute of Subtropical Agriculture, Chinese Academy of Sciences, Changsha, China (No. KYNEAAM-2015-0009).

Consent for publication
Not applicable.

Availability of data and material
The datasets supporting the conclusions of this article are included within the article.

Competing interests
The authors declare that they have no competing interests. The mRNA (A) and protein expression (B) of energy metabolism-related genes in the Longissimus thoracis muscles of the dams (n = 6). *P < 0.05 and #0.05≤ P <0.10 for the effect of treatment.