Apelin Controls Fetal and Neonatal Glucose Homeostasis and Is Altered by Maternal Undernutrition

Apelin is a regulatory peptide, identified as an endogenous ligand of the apelin receptor named APJ (1). The apelin gene encodes a 77–amino-acid preproprotein that generates several molecular isoforms such as apelin-36, apelin-17, and apelin-13 by posttranslational processing (2,3). Moreover, the N-terminal glutamate of apelin-13 can be posttranslationally modified, thus creating the pyroglutamate apelin-13 ([pyr-1]-apelin-13), which is more protected from exopeptidase degradation (4). Apelin and APJ have a widespread distribution in the body and are involved in various physiological functions. Apelin and APJ mRNAs are expressed in heart, lung, placenta, mammary gland, several regions of the central nervous system, adipocytes, and the gastrointestinal tract (5,6). Depending on the cell type studied, APJ activation results in the activation of several intracellular effectors such as extracellular signal–regulated kinases, protein kinase B (PKB or Akt), and p70S6 kinase and in the inhibition of cAMP production (7,8).

Recently, apelin has been extensively described as a beneficial factor regarding glucose metabolism and as endowed with antidiabetes properties (9,10). Studies demonstrated that both short- and long-term apelin treatments improve insulin sensitivity in obese and insulin-resistant mice mainly by increasing glucose uptake in skeletal muscle (9–11). Apelin also modulates insulin secretion and increases pancreatic islet cell mass and β-cell insulin content in mice (12). During prenatal development, previous studies have revealed that APJ deficiency in mice causes early embryonic defects and leads to embryonic lethality due to growth retardation and cardiac malformations (13). However, the effect of apelin on glucose homeostasis in utero remains unknown. To gain further insight into apelin function during gestation and to study the effects of apelin on fetal and neonatal glucose homeostasis, we formulated the aims of the current study: to investigate in rats 1) the kinetics of apelin plasma levels in mother/fetus pairs, 2) the ex vivo placental apelin release, 3) the gene expression levels of apelin and APJ in the fetoplacental unit, 4) the effect of maternal apelin administration on transplacental glucose transfer, and 5) the effect of apelin administration to neonates on glucose uptake in several tissues. We subsequently studied the apelinergic (apelin and APJ) system in intrauterine growth–restricted (IUGR) fetuses from rat mothers that received only 30% of the food intake of control mothers (FR30 model).

Experiments were conducted in accordance with the institutional guidelines for the use of laboratory animals and were approved by the animal ethics committee (from Université de Lille). Adult Wistar rats (Janvier) were housed at 22°C with a 12-h light/dark cycle with free access to a chow diet (16% protein, 3% fat, and 60% carbohydrates). Females were mated with a male. Embryonic day 0 (E0) was defined the following day if spermatozoa were found in vaginal smears. Pregnant females were divided into two groups: 1) a control group (n = 50) in which dams were fed ad libitum and 2) a food-restricted group (n = 27) in which females received 30% of the food intake of control mothers from E1 to E21. Five virgin adult females were used for apelinemia determination.

Tail blood samples were collected at 9:00 a.m. in virgin and E7 females to measure apelinemia. At E13, E17, and E21, pregnant rats were killed by decapitation. Placentas, fetuses, and mesometrial triangles (Meso-tr) were collected by caesarean section. Maternal and fetal plasma aliquots were stored at −20°C. Placentas and Meso-tr were frozen in liquid N₂ and stored at −80°C. Some E21 fetuses (n = 5) were used to study the tissue distribution of apelin and APJ mRNAs using reverse transcriptase quantitative real-time PCR (RT-qPCR) analysis. In fetuses, 17 selected tissues were dissected, frozen, and stored at −80°C.

Commercially available ELISA/enzyme immunosorbent assay kits were used to measure plasma insulin (catalog no. EIA-2943; DRG International) and apelin (catalog no. EK-057-23, which assayed all isoforms of apelin from apelin-12 to apelin-36; Phoenix Pharmaceuticals) levels. Blood glucose was measured using a glucometer (Roche Diagnostics).

Methods for RT-qPCR analysis have been previously described (14). RT-qPCR was performed with LightCycler 480 SYBR Green I Master and a LightCycler 480 instrument (Roche). Primers for apelin, APJ, GLUT1, and GLUT3 genes reported in Supplementary Table 1 were designed using the Primer Premier software (Premier Biosoft International). Several housekeeping genes (Gapdh, Ppib, and Hprt1) were used for the normalization and were described previously (14).

After cesarean section, E17 and E21 placentas (n = 24) were collected, rinsed in saline, and incubated for 24 h in dish plates containing 2 mL of DMEM (Gibco). Dish plates were placed at 37°C with 95% O₂ and 5% CO₂ and 95% humidity. Samples of the medium were collected after 2, 6, and 24 h of incubation for determination of apelin concentrations.

E17 pregnant females (n = 10) were anesthetized using isoflurane and 10 min later were implanted with a catheter in the jugular vein. An intravenous injection of 50 μCi 2-deoxy-D-[3H]glucose ([³H]-2DG) (NEN LifeScience) with either saline or 600 pmol/kg [pyr-1]-apelin-13 (Bachem [U.K.] Limited) was performed. After 30 min, placentas and fetuses were rapidly collected, weighed, and frozen. Total [³H]-2DG content in placental and fetal homogenates was measured.

E21 pregnant females (n = 9) were killed by decapitation. and fetuses were collected by cesarean section. Neonates were immediately weighed and injected with either 0.1 mL i.p. saline or different doses of [pyr-1]-apelin-13 (from 1 to 40 nmol/kg i.p.). After 30 min, blood glucose was measured. Plasma samples from neonates injected with saline and with 10, 15, 20, and 40 nmol/kg apelin were also used for insulinemia determination.

An injection of 50 μCi i.p. [³H]-2DG in saline or associated with [pyr-1]-apelin-13 (15 nmol/kg i.p.) was performed in rat neonates at birth (n = 14). After 30 min, rats were decapitated and seven selected tissues were collected for [³H]-2DG uptake determination.

Results are reported as mean ± SEM. Statistical analyses were performed using one-way ANOVA and the Dunnett test. A P value of <0.05 was considered significant.

Maternal plasma apelin concentrations were decreased at E7, increased from E7 to E17, and reduced at term (Fig. 1A). Fetal plasma apelin concentrations were twofold higher than maternal levels at E17, whereas they were similar at E21 (Fig. 1B). Apelin mRNA levels were higher in the Meso-tr compared with placenta at E13 and thereafter reduced until term of gestation in both tissues (Fig. 1C). APJ mRNA levels were similar in these two tissues at E13 (Fig. 1D). In the placenta, APJ expression was increased at E17 versus the E13 level and reduced at E21, whereas in the Meso-tr an opposite modulation was observed (Fig. 1D). Ex vivo, rat placentas released increasing quantities of apelin during 24 h, and this secretion was approximately twice as high at E17 than at E21 (Fig. 2A).

Using [³H]-2DG as a tracer, we showed that apelin-13 intravenous administration to E17 females did not significantly affect placental glucose content (Fig. 2B) but significantly increased fetal glucose uptake (Fig. 2C) without affecting placental GLUT1 and GLUT3 mRNA levels (Fig. 2D). In E21 fetuses, tissue distribution of apelin and APJ mRNAs demonstrated high expression levels in lung, heart, brain, kidney, stomach, muscle, and testis (Fig. 3A and B). A modest expression of this system was found in all other E21 tissues investigated (Fig. 3A and B). Neonates injected with 10–15 nmol/kg apelin-13 displayed a reduction in glycemia, whereas the injection of higher doses of apelin-13 caused an increase in glycemia (Fig. 3C) and a reduction of insulinemia (Fig. 3D). Concomitant administration of apelin-13 (at 15 nmol/kg) and [³H]-2DG demonstrated that apelin has a powerful glucose-lowering effect by enhancing glucose uptake in skeletal muscle and lung (Fig. 3E).

FR30 reduced both maternal and fetal body weights from E13 to E21 (Fig. 4A–C). FR30 reduced maternal plasma apelin concentrations from E13 to E21 (Fig. 4D) and fetal plasma apelin concentrations at E17 (Fig. 4E). Apelin mRNA levels were upregulated at E17 in FR30 placentas and Meso-Tr (Fig. 4F and G). APJ gene expression was increased in FR30 placentas at E13 and reduced at E21 in FR30 placentas and Meso-Tr (Fig. 4H and I).

In the current study, we demonstrate that maternal apelinemia is augmented during gestation and that ex vivo the placenta releases high amount of apelin at E17, a period that coincided with a peak secretion of apelin in both maternal and fetal plasmas. Moreover, we show that maternal intravenous apelin administration increases the transplacental transport of glucose and that, in neonates, acute intraperitoneal apelin injection has a powerful glucose-lowering effect associated with enhanced glucose uptake in lung and muscle.

We found that maternal apelin levels are decreased during the first week of gestation, increased gradually from E7 to E17, and reduced at term. This decline of apelinemia at term is related to an increase in apelin clearance by the placental ACE-related carboxypeptidase-2, which catabolized apelin in late gestation (15). In fetuses, we observed that apelin levels are twice as high as in mothers at E17, whereas they are similar at E21. This is consistent with studies in humans that show a similar doubled apelinemia in umbilical cord blood compared with maternal plasma apelin concentrations (16). In newborn babies, a drop in apelin levels was found at neonatal day 1 (16), suggesting that the placenta may be a source of apelin. In accordance, the apelinergic system is expressed in several placental compartments in both rats (15) and humans (17). In our study, we analyzed the gene expression level of this system in rat placentas and in Meso-tr from E13 to E21. We observed that placental apelin gene expression is closely correlated with maternal plasma apelin levels, pointing to the potential placental origin of maternal apelin. To test this hypothesis, we studied the apelin release from placenta ex vivo. We demonstrated that rat placentas are able to release a significant amount of apelin and that this secretion is twice as high at E17 than at term (E21). This secretion coincides with a peak secretion of apelin in both maternal and fetal plasmas at E17, suggesting that this organ is a source of circulating apelin. This interpretation is in accordance with data from Van Mieghem et al. (15) showing that in rats, a fetoplacental reduction significantly reduced maternal apelin levels.

Here, we report that the apelinergic system is expressed at the fetomaternal interface. Of note, we observed a drastic increase in APJ placental gene expression between E13 and E17. We postulated that apelin may enhance transplacental nutrients transfer, more especially glucose transport. We found that apelin injection to E17 mothers increases transplacental glucose transport, suggesting that maternal apelin levels control fetal glucose supply. This was observed without a change in placental glucose transporters expressions. In rat placentas, apelin was detected in perivascular smooth muscle of the labyrinth, suggesting that apelin may exert vasoactive action in the placenta (15), as it has been extensively reported for apelin in the periphery (18–20). We propose that apelin may induce a vasodilatation of placental vessels, which would enhance transplacental glucose transfer, or indirectly may modulate maternal blood pressure, resulting in this enhanced placental transport.

We observed high apelin levels in fetuses, which suggests that apelin may be important for fetal development. In E21 fetuses, the apelin and APJ tissue distribution was closely comparable with that in adults (5,6,21), with higher expressions found in lung, heart, muscle, brain, kidney, stomach, and testis. Injection of apelin into neonates was able to reduce glycemia at a low dose and, inversely, to increase glycemia at a high dose by inhibiting insulin release. These results are in accordance with the hypoglycemic effect of physiologic low doses of apelin in mice (11) and with the inhibitory role of high pharmacologic doses of apelin on insulin release (22). We demonstrate that the hypoglycemic effect of apelin is associated with an increase in lung and muscle glucose uptakes. Similarly, in mice, apelin was shown to exert a powerful glucose-lowering effect associated with enhanced glucose utilization in skeletal muscle (11).

Finally, we studied the apelinergic system in IUGR fetuses from FR30 mothers. FR30 drastically reduces maternal apelin levels from E13 to E21 as well as fetal apelin levels at E17. In accordance, plasma apelin concentrations were found to be reduced by food restriction in adult rodents and humans (9,21). In our FR30 model, we found that at the fetomaternal interface, apelin gene expression is upregulated at E17. This phenomenon may be a compensatory response of IUGR fetuses to increase their glucose supply to improve their growth. Altogether, our data demonstrate that apelin is a new hormone implicated in fetal and neonatal glucose homeostasis.

Acknowledgments. The authors thank Anne Dickes-Coopman, Valérie Montel, and Yseult Lesage for technical help.

Funding. This work was supported by grants from the French Ministry of Education and the French National Research Agency (project no. ANR-06-PNRA-022) and by a grant from the Appert Institute.

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. S.M., J.-S.W., M.-A.L., S.L., L.B., A.D., B.B., C.L., C.K., and J.L. designed the experiments, researched data, and wrote the manuscript. D.E., L.S., D.V., and C.B. interpreted the results and drafted the manuscript. J.L. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.