Placental Glucose and Amino Acid Transport in Calorie-Restricted Wild-Type and Glut3 Null Heterozygous Mice

Abstract

Calorie restriction (CR) decreased placenta and fetal weights in wild-type (wt) and glucose transporter (Glut) 3 heterozygous null (glut3^+/−) mice. Because placental nutrient transport is a primary energy determinant of placentofetal growth, we examined key transport systems. Maternal CR reduced intra- and transplacental glucose and leucine transport but enhanced system A amino acid transport in wt mice. These transport perturbations were accompanied by reduced placental Glut3 and leucine amino acid transporter (LAT) family member 2, no change in Glut1 and LAT family member 1, but increased sodium coupled neutral amino acid transporter (SNAT) and SNAT2 expression. We also noted decreased total and active phosphorylated forms of mammalian target of rapamycin, which is the intracellular nutrient sensor, the downstream total P70S6 kinase, and pS6 ribosomal protein with no change in total and phosphorylated 4E-binding protein 1. To determine the role of placental Glut3 in mediating CR-induced placental transport changes, we next investigated the effect of gestational CR in glut3^+/− mice. In glut3^+/− mice, a key role of placental Glut3 in mediating transplacental and intraplacental glucose transport was established. In addition, reduced Glut3 results in a compensatory increase of leucine and system A transplacental transport. On the other hand, diminished Glut3-mediated intraplacental glucose transport reduced leucine transport and mammalian target of rapamycin and preserved LAT and enhancing SNAT. CR in glut3^+/− mice further reduced transplacental glucose transport and enhanced system A amino acid transport, although the increased leucine transport was lost. In addition, increased Glut3 was seen and preserved Glut1, LAT, and SNAT. These placental changes collectively protect survival of wt and glut3^+/− fetuses against maternal CR-imposed reduction of macromolecular nutrients.

Many epidemiological and animal studies have shown that nutritional stress during gestation results in low birth weight with increased susceptibility to chronic diseases of adulthood, e.g. obesity, diabetes, and cardiovascular disease (1–3). Intrauterine growth restriction (IUGR) is the result of perturbed placentofetal development. Fetal growth is also dependent on efficient transplacental nutrient supply from mother to fetus across the placental barrier. Abnormalities in fetal growth have been previously attributed to changes in placental nutrient transport function (4). We have previously demonstrated that diminished fetal rat growth was associated with a reduction in placental glucose transporter (Glut) concentrations (Glut1) (5). Additional studies inclusive of a decrease in placental glucose transporter expression reduced transplacental glucose transport causing a diminution in mouse placentofetal growth (6). Experimental studies have linked altered placental amino acid transporter activity with perturbed placental development and fetal growth (6, 7). Imprinted genes such as IgF2 when deleted specifically in trophoblasts led to a diminution in system A transport (8) but an increase in Glut3 expression, whereas the deletion of H19 and Grb.10 adversely affected placental size, embryonic development, and nutrient transporter gene expression (9).

In addition, the rapamycin-sensitive mammalian target of rapamycin (mTOR) signaling pathway is known to regulate nutrient transport, thereby affecting cellular growth and differentiation (10). In nonplacental human BJAB lymphoma cells, rapamycin treatment reduced several amino acid transporters (11). Furthermore, rapamycin-pretreated L6 myoblast cells demonstrated an inhibition of leucine induced system A transporter activity (12). In contrast, using mouse Fl.5.12 lymphoma cells, mTOR increased total amino acid transporter gene expression with enhanced surface expression (13). Given these in vitro studies in nontrophoblastic cells, ex vivo investigations in the postparturient human placenta associated with fetal growth restriction demonstrated down-regulation of rapamycin-induced leucine transport (14). No similar in vivo studies are possible within the intact human placenta. Hence, in vivo studies within late gestation placenta that is actively supplying the fetus with nutrients is feasible only in an animal model in which the placenta closely mimics the human hemochorial placenta. We therefore hypothesized that late-gestation calorie restriction in the mouse would alter transplacental glucose and amino acid transport by perturbing the respective placental transporter protein isoforms. Furthermore, late-gestation calorie restriction superimposed on a monoallelic deletion of a glucose transporter (glut3^+/−) background will help determine the role of glucose in regulating transplacental amino acid transporter and trophoblast nutrient sensing systems.

Materials and Methods

Animals

Glut3 heterozygous (glut3^+/−) and wild-type (glut3^+/+) C57/BL6 mice were housed in 12-h light, 12-h dark cycle with ad libitum access to a standard rodent chow (Harlan Teklad 7013; Harlan Laboratories, Indianapolis, IN) and water. At eight weeks of age both the glut3^+/− and glut3^+/+ female mice were mated with male counterparts. Presence of a vaginal plug was designated gestational d 1. At gestation d 10, the glut3^+/− and glut3^+/+ pregnant mice were either continued on an ad libitum feeding schedule or were restricted by 50% of their daily chow intake. The study protocol was approved by the Animal Research Committee of the University of California, Los Angeles, in accordance with the guidelines set by the National Institutes of Health.

Transplacental and intraplacental macronutrient transport

Glucose

Ten-week-old pregnant female mice (d 18.5 gestation) were ip injected with a bolus of 12 μCi of 2-deoxy-d-[1-¹⁴C] glucose (PerkinElmer, Boston, MA). An hour later, maternal blood was collected from the jugular vein, and glucose concentration and radioactivity were assessed. Soon afterward the mice were euthanized with pentobarbital sodium (100 mg/kg, ip), and the fetuses and placentas were collected separately. A small portion of preweighed placentas were used for genotyping, and the remaining placentas and whole fetuses were hydrolyzed separately in 1 m NaOH at 60 C for 45 min and then neutralized with 1 m HCl. One aliquot (200 μl) of the neutralized lysate was added to 1 ml of HClO₄, and another aliquot (200 μl) was added to Ba(OH)₂/ZnSO₄. After centrifugation, the supernatants (800 μl) were used to assess radioactivity in a liquid scintillation counter. Protein and glucose concentrations were estimated in placental/fetal lysates as well. Glucose uptake by the placenta and fetus was calculated as a ratio between specific activities of glucose in the placenta/fetus and maternal blood (15, 16). Placental and fetal total glucose uptake (HClO₄ precipitated supernatant) reflected both the intracellular glucose transport [Ba(OH)₂/ZnSO₄ precipitated supernatant] and phosphorylation [Ba(OH)₂/ZnSO₄ precipitate], whereas glucose uptake (transport+phosphorylation) by the fetus as a ratio to placental weight reflected transplacental glucose transport and phosphorylation (6, 17).

System L amino acids (leucine)

Five microcuries per 100 μl L-[U-¹⁴C]leucine (GE Healthcare Life Sciences, Indianapolis, IN; CFB67, specific activity 318 mCi/mmol) in sterile saline was administered via tail vein to d 18.5 gestation, 10-wk-old pregnant mice. After the administration of isotope, the mouse was placed in a cage in which it was free to ambulate for 5 min when a venous blood sample was withdrawn for determination of plasma radioactivity. Mice were euthanized with sodium pentobarbital (100 mg/kg, ip) (18), placentas and fetuses were collected separately and weighed individually and then lysed overnight at 55 C in 4 ml of Biosol for the fetus and 0.5 ml volume for the placenta. Radioactivity in samples was assessed and expressed per gram (wet weight) of tissue.

System A amino acids

Three and a half microcuries per 100 μl of [¹⁴C]methylaminoisobutyric acid (MeAIB; NEN Life Science Products, Boston, MA; NEC-671; specific activity of 50.5 mCi/mmol) in sterile saline was administered via the tail vein to d 18.5 gestation, 10-wk-old pregnant mice. Five minutes after the tracer administration, placentas and fetuses were dissected after hysterotomies. Tissue lysis was carried out overnight at 55 C in Biosol (National Diagnostics, Atlanta, GA) in 4 ml volume for the fetus and 0.5 ml volume for the placenta. Measured aliquots were used to assess radioactivity that was expressed per gram (wet weight) of tissue (6, 19).

In the case of leucine and system A amino acid (MeAIB), intraplacental transfer signified the amino acid uptake by the placenta from maternal blood, which was expressed as total placental uptake per unit placental weight, whereas transplacental transfer was amino acid uptake by the fetus from placenta expressed as total fetal uptake per unit placental weight.

Glucose transporters: Glut1 and Glut3 and system L [leucine amino acid transporter family member 2 (LAT2)]

Placentas and fetuses at d 18.5 gestation were individually collected, weighed (accuracy of 0.01 mg), snap frozen, and stored at −80 C. Placentas were homogenized, solubilized in a lysate buffer, and protein concentration determined by the Bio-Rad dye-binding assay (Bio-Rad Laboratories, Hercules, CA). Western blots were performed as described previously (20). For studies involving placental Glut1 (also fetal) and Glut3 expression (20 μg) and LAT2 (50 μg), homogenates were subjected to 10% SDS-PAGE and transferred to nitrocellulose membranes (Transblot; Bio-Rad Laboratories). The primary antibody consisted of an affinity-purified rabbit antimouse Glut3 or Glut1 IgG that was generated and previously characterized by us (20, 21). The primary antimouse Glut3, the antimouse Glut1, and anti-LAT2 (Sc 27581; Santa Cruz Biotechnology, Santa Cruz, CA) antibody incubations were done at a 1:1000 dilution for 2 h at room temperature. The secondary antibody consisted of a horseradish peroxidase-conjugated antirabbit antibody (Glut1, Glut3 = 1:2500; LAT2 = 1:1000) that allowed detection of immunoreactive protein bands by enhanced chemiluminescence (GE Healthcare Bio-Sciences, Uppsala, Sweden).

mTOR and downstream effectors and AMP-activated protein kinase (AMPK)

Fifty micrograms of protein homogenates were subjected to electrophoresis on 7% sodium dodecyl sulfate polyacrylamide gels. After transfer of separated proteins, the membranes were blocked with 5% milk in PBS-Tween 20 for 2 h. To detect mTOR (total and phosphor forms), the membranes were incubated with the anti-mTOR antibody (ser 2448) diluted 1:1000 in PBS-Tween 20 overnight at 4 C. The secondary peroxidase-labeled goat antirabbit antibody (Vector Laboratories, Burlingame, CA) was diluted 1:2500 and used in combination with enhanced chemiluminescence. To examine protein expression of total and phosphorylated downstream targets of mTOR, equal amounts of protein (20–25 μg) were electrophoresed in 10% sodium dodecyl sulfate polyacrylamide gels. After an overnight transfer of separated proteins, membranes were blocked with 5% milk in Tris-buffered saline (wt/vol) plus 0.1% Tween 20 (vol/vol) for 1 h at room temperature and then incubated overnight at 4 C with antibodies diluted in 5% BSA-Tris-buffered saline (wt/vol) plus 0.1% Tween 20 (vol/vol) against phospho-P70S6 kinase (S6K) (Thr-389), total S6K, phospho-4E-binding protein 1 (4E-BP1; Thr-37/46), total 4E-BP1, phospho-S6 ribosomal protein (ser 235/236), total S6 ribosomal protein, and anti-AMPK (SC2532; Santa Cruz Biotechnology) diluted 1:1000 except in the case of antiphospho-S6K1 antiphosphorylated AMPK (Ser-485; Santa Cruz Biotechnology) in which the dilution was 1:500. All the remaining mTOR signaling pathway antibodies were purchased from Cell Signaling Technology (Danvers, MA). Finally, the blots were incubated for 1 h with a peroxidase-labeled goat antirabbit antibody (Vector Laboratories) that was diluted 1:1000 in 5% milk solution, and the protein signal was detected by the enhanced chemiluminescence system.

RNA extraction and RT-PCR

Tissues were homogenized in Trizol reagent (Invitrogen, Carlsbad, CA), and total RNA was extracted according to manufacturer's instructions. Total RNA (1 μg) was reverse transcribed to cDNA at 50 C using 5 μg/μl random hexamers by the SuperScript first-strand synthesis system for RT-PCR according to the manufacturer's instructions. Only those genes lacking an adequate antibody targeting its protein product were assessed. The leucine amino acid transporter family member 1 (LAT1) cDNA was amplified by PCR using primers specific for mouse LAT1 (forward: 5′-GAGACCCTAGAGATGGAACCC-3′, reverse: 5′-TCACATCACACTGGTGACAGAG-3′) (NM 001161413) and mouse LAT2 (forward: 5′-GCCCATGGTCAAGGTCAATGC-3′, reverse: 5′-GGCTTGGCTTCTGCAGTCT) (NM 001253679). Similarly, the SNAT1 and SNAT2 cDNA were amplified using primers specific for mouse SNAT1 (forward: 5′-ACGACTCTAATGACTTCACAG-3′, reverse: 5′-ACTGACTGTCGAGTTCTGCTGC-3′) (NM 134086) and mouse SNAT2 (forward: 5′-AAAGGTGCCGTTCACAGTTTC-3′ and reverse: 5′-AACTACTCACCCAACCAAG-3′) (NM 175121.3). PCR was catalyzed by Platinum Taq DNA polymerase (Invitrogen) under the following conditions: initially at 95 C for 2 min; 34 (LAT1) or 31 (SNAT1 and SNAT2) cycles of denaturing (94 C for 30 sec), annealing (LAT1 = 55 C; SNAT1 and SNAT2 = 56.5 C for 30 sec) and extension (72 C for 2 min). The PCR-amplified products were separated by 2% agarose gel electrophoresis and in the presence of ethidium bromide were visualized under UV light. The relative intensities of the gel electrophoresis-separated amplification products were assessed.

Quantification of protein or mRNA bands

Protein or mRNA bands were visualized in a Typhoon 9410 phosphor imager (GE Healthcare Biosciences, Piscataway, NJ). Quantification was performed using the Image Quant 5.2 software (GE Healthcare Biosciences) after subtracting the background from a comparable area. All proteins were normalized to vinculin (internal control) and mRNA to cyclophillin (internal control; NM 001252444) before being expressed as a percent of the wild-type (wt) control values. Signaling molecules were also expressed as ratios between the phosphorylated to total amounts after normalizing each one to vinculin before expressing them as a percent of the wt control values.

Statistical analysis

Data are shown as mean ± se. All statistical analyses were performed using Sigmastat 3.5 software (Systat Point, Richmond, CA) Differences between four experimental groups (control (C) (wt; +/+), caloric restriction (CR) (wt; +/+), glut3 (+/−)-C, and glut3 (+/−)-CR) were simultaneously analyzed by one-way ANOVA. Normal distribution of data was confirmed and the Fisher protected least significant difference (PLSD) method was applied for post hoc intergroup comparisons. When only two groups (C vs. corresponding CR or wt-C vs. glut3 (+/−)-C) were compared, the Student's t test was used. A P < 0.05 was assigned significance.

Results

Macronutrient transport

Glucose

Transplacental (TP) and intraplacental (IP) total glucose uptake (TP: P = 0.0037 and IP: < 0.0001) and glucose-6-phosphate (TP: P = 0.0028 and IP: 0.031) decreased in control glut3 (+/−) mice when compared with wt (+/+) mice, both maintained on ad libitum chow diet (C). When exposed to mid- to late gestation maternal CR, the wt and glut3 (+/−) mice demonstrated a reduction in TP and IP total glucose uptake (wt: TP, P = 0.0037, IP, P < 0.0001; glut3 (+/−): TP, P = 0.018, IP, P < 0.0001) and glucose-6-phosphate (wt: TP, P < 0.0001, IP, P = 0.031; glut3 (+/−): TP, P < 0.0001, IP, P = 0.0008) as well. Although comparing the CR-induced effect on both TP and IP total glucose uptake (TP, P = 0.018; IP, P = 0.05) and glucose-6-phosphate (TP, <0.0001; IP, P = 0.0008), in general a more severe decrease was noted in the glut3 (+/−) genotype as opposed to the wt (Fig. 1A). Thus, overall an approximately 70% reduction in total glucose uptake and an approximately 50% decrease in glucose-6-phosphate was observed in wt, whereas an approximately 80% in total glucose uptake and an approximately 88% in glucose-6-phosphate was evident in the glut3 (+/−) genotype in response to maternal calorie restriction.

Fig. 1.

Transplacental and intraplacental macronutrient transfer. A, Glucose uptake consisting of intracellular glucose transport and phosphorylation (glucose-6-phosphate) in ad libitum C and 50% CR wt and glut3 (+/−) d 18.5 gestation placentas is depicted as transplacental and intraplacental glucose uptake. Transplacental glucose uptake (n = 6, 8, 8, 6) is as follows: F value = 9.749, p(ANOVA) = 0.0002, Fisher's PLSD significance is shown as follows: *, vs. wt-C; #, vs. glut3 (+/−)-C; †, vs. wt-CR. *, wt-CR (P = 0.0037), glut3 (+/−)-C (P = 0.0037), glut3 (+/−)-CR (P < 0.0001) vs. wt-C; #, glut3 (+/−)-CR (P = 0.018) vs. glut3 (+/−)-C; †, glut3 (+/−)-CR (P = 0.018) vs. wt-CR. Transplacental glucose-6-phosphate (n = 6, 8, 8, 11) is as follows: F value = 26.158, p(ANOVA) < 0.0001. Fisher's PLSD is as follows: *, wt-CR (P < 0.0001), glut3 (+/−)-C (P = 0.0028), glut3 (+/−)-CR (P < 0.0001) vs. wt-C; #, wt-CR (P = 0.0005), glut3 (+/−)-CR (P < 0.0001) vs. glut3 (+/−)-C. Intraplacental glucose uptake (n = 6, 8, 6, 10) is as follows: F value = 36.898, p(ANOVA) < 0.0001. Fisher's PLSD is as follows: *, wt-CR (<0.0001), glut3 (+/−)-C (<0.0001), glut3 (+/−)-CR (<0.0001) vs. wt-C; #, wt-CR (0.01), glut3 (+/−)-CR (<0.0001) vs. glut3 (+/−)-C; †, glut3 (+/−)-CR (0.05) vs. wt-CR. Intraplacental glucose-6-phosphate (n = 6, 6, 6, 9) is as follows: F value = 14.495, p(ANOVA) < 0.0001. Fisher's PLSD is as follows: *, wt-CR (0.031), glut3 (+/−)-C (0.031), glut3 (+/−)-CR (<0.0001) vs. wt-C; #, glut3 (+/−)-CR (0.0008) vs. glut3 (+/−)-C; †, glut3 (+/−)-CR (0.0008) vs. wt-CR. B, System L (leucine) transfer in ad libitum chow fed (C) and 50% CR wt and glut3 (+/−) d 18.5 gestation placentas is depicted as transplacental and intraplacental leucine transfer. Transplacental leucine transfer (n = 7, 6, 4, 8) is as follows: F value = 4.682, p(ANOVA) = 0.012. Fisher's PLSD is as follows: *, wt-CR (0.045), glut3 (+/−)-C (0.07; Student's t test 0.009) vs. wt-C; #, wt-CR (0.0014) vs. glut3 (+/−)-C; †, glut3 (+/−)-CR (0.03) vs. wt-CR. Intraplacental leucine transfer (n = 7, 6, 4, 8) is as follows: F value = 13.859, p(ANOVA) < 0.0001. Fisher's PLSD is as follows: *, wt-CR (<0.0001), glut3 (+/−)-C (0.0001), glut3 (+/−)-CR (<0.0001) vs. wt-C. C, System A (methylaminoisobutyric acid) transfer in ad libitum chow fed (C) and 50% CR wt and glut3 (+/−) d 18.5 gestation placentas is depicted as transplacental and intraplacental MeAIB acid transfer. Transplacental MeAIB transfer (n = 5, 7, 6, 5) is as follows: F value = 6.330, p(ANOVA) = 0.0037. Fisher's PLSD is as follows: *, glut3 (+/−)-C (0.07, Student's t test 0.05), glut3 (+/−)-CR (0.0007) vs. wt-C; #, glut3 (+/−)-CR (0.0076) vs. glut3 (+/−)-C; †, glut3 (+/−)-CR (0.0027) vs. wt-CR. Intraplacental MeAIB transfer (n = 5, 7, 6, 4) is as follows: F value = 3.185, p(ANOVA) = 0.049. Fisher's PLSD is as follows: *, wt-CR (0.013) vs. wt-C; #, wt-CR (0.027) vs. glut3 (+/−)-C.

System L amino acids (leucine)

Transplacental leucine transfer decreased by approximately 35% in the wt calorie-restricted group vs. the control (P = 0.045) along with a similar reduction seen in the glut3 (+/−) calorie-restricted group vs. the respective control group, although not different from that of the wt control group. An approximately 41% increase in TP leucine transfer was observed in glut3 (+/−) vs. wt ad libitum calorie-exposed mice (Fisher's PLSD, P = 0.07; Student's t test = 0.009) (Fig. 1B). When comparing the effect of calorie restriction on wt vs. glut3 (+/−) genotypes, leucine transfer in glut3 (+/−) mice was greater than in the wt genotype (P = 0.03). Glut3 (+/−) ad libitum fed genotype alone reduced IP leucine transport when compared with the wt ad libitum-fed genotype (P = 0.0001). The exposure to calorie restriction led to a significant reduction in the IP leucine transport only in the wt genotype (P < 0.0001). However, unlike the wt genotype, calorie restriction in the glut3 (+/−) genotype did not further reduce the IP leucine transport beyond that which occurs in the ad libitum-fed glut3 (+/−) genotype.

System A amino acids

Although an increase in TP system A transport is observed in glut3 (+/−) vs. wt ad lib fed groups (Student's t test P = 0.05), a further compensatory increase was observed in the glut3 (+/−) calorie-restricted group vs. the ad libitum-fed respective control (P < 0.0076). No similar effect of calorie restriction was noted in the wt mice (Student's t test, P = 0.06). In contrast, an increase in IP system A amino acid transport was observed only in the wt genotype when exposed to calorie restriction (P = 0.013) (Fig. 1C).

Placental and fetal weights

The wt fetal weight at gestation d 18.5 decreased by approximately 47.9% (P < 0.0001) and placental weight decreased by approximately 37.5% (P < 0.0001) secondary to maternal calorie restriction vs. ad libitum control. Maternal calorie restriction in glut3 (+/−) mice displayed an approximately 28.6% decrease in fetal weight (P < 0.0001) and an approximately 34% decrease in placental weight (P < 0.0001) vs. the respective control (+/−) with access to ad libitum calories (Table 1). The glut3 (+/−) genotype alone did not affect the fetal weight (P = 0.069) or placental weight (P = 0.85) when compared with wt. The impact of caloric restriction on placental weight (P = 0.0011) and fetal weight (P = 0.0077) in the glut3 (+/−) genotype was also seen similar to that of wt (Table 1).

Table 1.

Placental and fetal body weights

Group (n)	Placental weight (g)	Fetal body weight (g)
wt C (n = 22)	0.08 ± 0.002	1.15 ± 0.02
wt CR (n = 27)	0.05 ± 0.002a,b	0.6 ± 0.02a,b
glut3 (+/−) C (n = 18)	0.09 ± 0.003	1.05 ± 0.05
glut3 (+/−) CR (n = 26)	0.06 ± 0.002a,b,c	0.75 ± 0.04a,b,c

Placental and fetal body weights of CON and 50% CR wt and glut3 (+/−) d 18.5 gestation mice are shown as mean ± sem. Placenta weight was as follows: n = 22 (wt-C); n = 27 (wt-CR); n = 18 [glut3(+/−)-C]; n = 26 [glut3(+/−)-CR]; F value = 33.79, p(ANOVA) less than 0.0001. Fetal body weight was as follows: n = 22, 27, 19, 25; F value = 48.96, p(ANOVA) less than 0.0001.

^aFisher's PLSD for wt-CR was less than 0.0001; glut3 (+/−)-CR, 0.0003 vs. wt-C; wt-CR, less than 0.0001, glut3 (+/−)-C, 0.069, not significant; glut3 (+/−)-CR, less than 0.0001 vs. wt-C.

^bFisher's PLSD for wt-CR was less than 0.0001; glut3 (+/−)-CR, less than 0.0001 vs. glut3 (+/−)-C; wt-CR, less than 0.0001; glut3 (+/−)-CR, less than 0.0001 vs. glut3 (+/−)-C.

^cFisher's PLSD for glut3 (+/−)-CR vs. wt-CR was 0.0011; glut3 (+/−)-CR, 0.0077 vs. wt-CR.

Glucose transporters: Glut3 and Glut1

Placental Glut3 protein concentrations decreased by approximately 50% in the wt calorie-restricted group vs. the respective ad libitum control (P = 0.024). In contrast, calorie restriction in glut3 (+/−) mice demonstrated an approximately 6-fold increase in placental Glut3 protein concentrations (P < 0.0001) (Fig. 2A) vs. the glut3 (+/−) controls exposed to ad libitum calories. As expected, the glut3 (+/−) genotype alone expressed approximately 75% lower placental Glut3 protein concentrations when compared with the wt counterpart (P = 0.0031). Placental Glut1 protein concentrations did not significantly change in response to maternal calorie restriction in wt mice. No change from the wt ad libitum-fed group was seen in response to the glut3 (+/−) genotype alone. However, ad libitum-fed and caloric restriction in glut3 (+/−) mice resulted in an approximately 35% (0.0178) and approximately 43% (P = 0.0045) decrease, respectively, in placental Glut1 protein concentrations when compared with the wt calorie-restricted group (Fig. 2A). Again, although the glut3 (+/−) genotype alone did not make a difference, calorie restriction significantly increased fetal Glut1 protein concentrations in both wt (P = 0.03) and glut3 (+/−) mice (P = 0.0013) (Fig. 2B).

Fig. 2.

Glucose transporters. Representative Western blots are seen in the top panels depicting Glut3 and Glut1 protein bands in C and CR wt (+/+) and glut3 (+/−) d 18.5 gestation placentas (A) and Glut1 protein bands in d 18.5 gestation embryo (B), with vinculin serving as the internal loading control. Quantification of Glut3 and Glut1 as a ratio to the vinculin protein are expressed as a percent of the wt ad libitum-fed control values in the corresponding bottom panels. Placental Glut3 protein (n = 5/group) (left lower panel) is as follows: F value = 17.240, p(ANOVA) < 0.0001. Fisher's PLSD is as follows: *, wt-CR (0.02), glut3 (+/−)-C (0.003), glut3 (+/−)-CR (0.0069) vs. wt-C; #, glut3 (+/−)-CR (<0.0001) vs. glut3 (+/−)-C; †, glut3 (+/−)-CR (<0.0001) vs. wt-CR. Placental Glut1 protein (n = 5/group) (middle lower panel) is as follows: F value = 4.098, p(ANOVA) = 0.0246. Fisher's PLSD is as follows: #, wt-CR (0.0178) vs. glut3 (+/−)-C; †, glut3 (+/−)-CR (0.0045) vs. wt-CR. Fetal Glut1 protein (n = 5/group) (C) is as follows: F value = 7.133, p(ANOVA) = 0.0029. Fisher's PLSD is as follows: *, wt-CR (0.03), glut3 (+/−)-CR (<0.0021) vs. wt-C; #, wt-CR (0.0194), glut3 (+/−)-CR (0.0013) vs. glut3 (+/−)-C.

Amino acid transporter expression

LAT1 and LAT2

Both LAT1 (250 bp) and LAT2 (280 bp) mRNA concentrations in placenta demonstrated a trend toward an increase in glut3 (+/−) vs. the wt ad libitum calorie-exposed mice, although not statistically significant (Fig. 3A). No difference was observed between the calorie-restricted and ad libitum calorie-exposed groups in the wt genotype in the case of LAT1 (only a trend in increase) and LAT2 mRNA. In contrast, no change in LAT1 but a reduction in LAT2 mRNA (P = 0.043) was noted in response to calorie restriction vs. ad libitum calorie intake in the glut3 (+/−) genotype (Fig. 3A). An approximately 30% reduction in placental LAT2 protein concentration was, however, noted only in the calorie-restricted wt mice when compared with the ad libitum-exposed counterpart (P < 0.05, Student's t test), with no similar difference observed between the two glut3 (+/−) groups. Glut3 (+/−) ad libitum-fed genotype, although demonstrating a trend, did not demonstrate a significant reduction in LAT2 protein when compared with the wt ad libitum-fed control group (Fig. 3B).

Fig. 3.

Amino acid transporters. A, Detection of LAT1 and LAT2 mRNA in C and CR d 18.5 placentas of wt and glut3 (+/−) mice by RT-PCR using specific primers (top panel). Cyclophilin served as the internal control. Relative band intensity was quantified and expressed as a percentage of the wt ad libitum control (bottom panel). A, LAT1 mRNA (n = 5/group): F value = 0.639, p(ANOVA) = 0.6008; Fisher's PLSD was not significant; LAT2 mRNA (n = 6, 5, 5, 5): F value = 2.68, p(ANOVA) = 0.0797; Fisher's PLSD is as follows: #, glut3 (+/−)-CR (0.015) vs. glut3 (+/−)-C. B, LAT2 protein is shown in C and CR wt and glut3 (+/−) d 18.5 gestation placenta (n = 7/group). F value = 0.708, p(ANOVA) = 0.5564. Fisher's PLSD was not significant; *, Student's t test is as follows: *, wt-CR (0.047) vs. wt-C. C, SNAT1 and SNAT2 mRNA in C and CR wt and glut3 (+/−) placentas. SNAT1 mRNA (n = 6, 5, 5, 5): F value = 4.473, p(ANOVA) = 0.0183. Fisher's PLSD is as follows: *, wt-CR (0.0047), glut3 (+/−)-C (0.032) and glut3 (+/−)-CR (0.013) vs. wt-C. SNAT2 mRNA (n = 6, 5, 5, 5): F value = 2.713, p(ANOVA) = 0.0795. Fisher's PLSD is as follows: *, wt-CR (0.0397), glut3 (+/−)-C (0.04), and glut3 (+/−)-CR (0.036) vs. wt-C.

SNAT1 and SNAT2

A 2.5- to 3-fold increase in placental SNAT1 (700 bp) expression was observed in both the glut3 (+/−) genotype alone (P = 0.032) and in response to maternal calorie restriction in the wt genotype (P = 0.0047) (Fig. 3C). However, no change in response to maternal calorie restriction was noted in the glut3 (+/−) genotype. All three groups, namely the calorie-restricted wt (P = 0.0397) and glut3 (+/−) (P = 0.036), and glut3 (+/−) exposed to ad libitum calories (P = 0.04), demonstrated an increase in placental SNAT2 (750 bp) expression compared with wt control exposed to ad libitum calories. Again, no difference in the glut3 (+/−) genotype in response to maternal calorie restriction was observed beyond the respective control (Fig. 3C).

mTOR signaling molecules and AMPK

Total mTOR decreased in calorie restricted wt placentas by approximately 60% (P = 0.0005) and glut3 (+/−) placentas by approximately 22% (P = 0.0021) when compared with wt controls. glut3 (+/−) genotype alone also reduced placental total mTOR concentrations (P = 0.0183); however, calorie restriction in the glut3 (+/−) genotype did not further reduce total mTOR when compared with the respective glut3 (+/−) ad libitum-fed group. Phosphorylated mTOR (ser-2448) concentrations demonstrated a decrease (∼47%) in wt placentas in response to maternal calorie restriction (P = 0.035) with a trend toward a decrease but no statistical change in the glut3 (+/−) ad libitum-exposed placental phosphorylated mTOR concentrations (P = 0.12). A reduction in glut3 (+/−) calorie-restricted group was observed vs. the wt ad libitum group (P = 0.0086) but not vs. the glut3 (+/−) ad libitum group (Fig. 4A). The ratio between phosphorylated mTOR to total mTOR revealed a trend toward an increase in wt calorie-restricted vs. the ad libitum-fed group that was not statistically different (P = 0.25).

Fig. 4.

mTOR signaling molecules and AMPK. Representative Western blots shown in top panels demonstrate total and phosphorylated mTOR (A), S6K (Thr-389) (B), 4E-BP1 (C), S6 ribosomal (ser 235/236) (D), and AMPK (E) protein bands, with vinculin as the internal loading control. Quantification is shown in the corresponding bottom panels. A, Total mTOR (n = 5/group): F value = 7.312, p(ANOVA) = 0.0026. Fisher's PLSD is as follows: *, wt-CR (0.0005), glut3 (+/−)-C (0.018), glut3 (+/−)-CR (0.0021) vs. wt-C. pmTOR (n = 5/group): F value = 3.284, p(ANOVA) = 0.0481. Fisher's PLSD is as follows: *, wt-CR (0.035), glut3 (+/−)-CR (0.0086) vs. wt-C. p-mTOR/mTOR (n = 5/group): F value = 0.63, p(ANOVA) = 0.6050. Fisher's PLSD was not significant. B, Total S6K (n = 5/group): F value = 2.179, p(ANOVA) = 0.13. Fisher's PLSD is as follows: *, wt-CR (0.0275) vs. wt-C. p-S6K (n = 6/group): F value = 0.121, p(ANOVA) = 0.9466. Fisher's PLSD was not significant. p-S6K/S6K (n = 5/group): F value = 1.822, p(ANOVA) = 0.1838. Fisher's PLSD is as follows: *, wt-CR (0.0335) vs. wt-C. C, Total 4E-BP (n = 6/group): F value = 2.382, p(ANOVA) = 0.0998. Fisher's PLSD is as follows: †, glut3 (+/−)-C (0.025) vs. wt-CR; #, glut3 (+/−)-CR (0.0447) vs. glut3 (+/−)-C. p4E-BP1 (n = 6/group): F value = 1.199, p(ANOVA) = 0.3355. Fisher's PLSD was not significant. p4E-BP1/4E-BP1 (n = 6/group): F value = 0.893, p(ANOVA) = 0.4616. Fisher's PLSD was not significant. D, Total S6-r (n = 5/group). F value = 1.064, p(ANOVA) = 0.3922. Fisher's PLSD was not significant [wt-CR vs. glut3 (+/−)-C, P = 0.096]. p-S6-r (n = 5/group): F value = 1.829, p (ANOVA) = 0.1825. Fisher's PLSD is as follows: #, wt-CR (0.0393) vs. glut3 (+/−)-C. p-S6-r/S6-r (n = 5/group): F value = 0.118, p(ANOVA) = 0.9481. Fisher's PLSD was not significant. E, Total AMPK (n = 6/group): F value = 0.519, p(ANOVA) = 0.6741. Fisher's PLSD was not significant. pAMPK (n = 6/group): F value = 0.278, p(ANOVA) = 0.8404. Fisher's PLSD was not significant. pAMPK/AMPK (n = 6/group): F value = 0.728, p(ANOVA) = 0.5473. Fisher's PLSD was not significant.

Total S6K (Thr-389) concentrations also decreased by approximately 46% in wt (P = 0.028) with a trend toward a decrease that was not statistically different in glut3 (+/−) (P = 0.22) placentas in response to maternal calorie restriction when compared with the wt ad libitum group. No further change in calorie-restricted glut3 (+/−) vs. the corresponding glut3 (+/−) ad libitum group was seen. In contrast, no change in any of the groups was observed in the case of phosphorylated (p) S6K when compared with their respective ad libitum calorie-exposed groups (Fig. 4B). However, an increase in the pS6K to total S6K ratio was evident between the wt ad libitum and calorie-restricted groups (P = 0.0335) with no similar change in the glut3 (+/−) genotype.

Although no differences in total 4E-BP1 or p4E-BP1 protein concentrations were observed in the wt between ad libitum and calorie restriction (P = 0.24), the glut3 (+/−) genotype demonstrated an increase in total 4E-BP1 in glut3 (+/−) ad libitum vs. the wt calorie-restricted group (P = 0.025), and calorie restriction reduced total 4E-BP1 vs. the ad libitum calorie group in the glut3 (+/−) genotype (P = 0.045) (Fig. 4C). However, although mimicking a similar pattern, no statistical change was seen with either p4E-BP1 or the p4E-BP1 to total 4E-BP1 ratio between the four experimental groups (Fig. 4C).

Similarly, the S6 ribosomal protein concentrations trended toward a decrease in the wt calorie-restricted group vs. the wt ad libitum group but was not statistically different for both total and phosphorylated forms (P = 0.096). In contrast, although no change is seen between the calorie-restricted and ad libitum-fed glut3 (+/−) groups with respect to the total and phosphorylated forms, the latter was increased in glut3 (+/−) ad libitum vs. the wt calorie-restricted groups (P = 0.039). No change in the phosphorylated to total S6 ribosomal protein ratio was observed between the four groups (Fig. 4D).

There was no change in total or phosphorylated AMPK in calorie-restricted wt and glut3 (+/−) placentas compared with their respective ad libitum controls. In the case of the phosphorylated to total ratio, again no statistical differences were noted except a trend toward an increase in the wt calorie-restricted and glut3 (+/−) ad libitum and calorie-restricted groups vs. the wt ad libitum calorie group, supporting a relative placental energy deficit state in all three experimental groups vs. the wt control group (Fig. 4E).

Discussion

Glucose and amino acid transport in mouse wt placenta

We have demonstrated that maternal calorie restriction during late gestation in mice reduces placental and fetal weights. This occurs in light of perturbations in placental nutrient transport in which TP and IP glucose and leucine transport are reduced with a compensatory increase in system A amino acid transport.

Glucose transporters in wt placentas

The mechanisms underlying these observations reveal a decrease in placental Glut3 protein with no change in Glut1 protein glucose transporter isoform expression. Both these isoforms are facilitative in type transporting glucose passively down a concentration gradient (22, 23). These observations in mice mimic prior rat studies in which maternal under nutrition with intrauterine growth restriction was associated with a reduction in the placental exchange surface and Glut3 protein concentration (24, 25). In contrast, maternal glucocorticoid administration resulting in fetal growth restriction demonstrated an increase in rat placental Glut3 concentration (26). This difference in placental Glut3 concentrations may stem from a reduction in maternofetal glucose supply in the former vs. a glucocorticoid induced reduction in placental vascularization and concomitant hypoxic/ischemia in the latter (27). Our present study in mice demonstrates that maternal caloric restriction reduces TP and IP glucose transport that parallels the decrease in Glut3 protein concentrations despite the normal concentrations of Glut1. Thus, placental Glut1 concentrations are unable to compensate for the decreased Glut3 in placenta. In contrast, fetal Glut1 concentrations demonstrate a compensatory increase in response to maternal caloric restriction displaying a protective phenomenon against the adverse intrauterine nutritional environment.

System L amino acid transporters in wt placentas

System L amino acid transporter is sodium independent and energy dependent and mainly transports neutral branched chain amino acids like leucine. In the human pregnancy associated with IUGR, reduced activity of the system L amino acid transporter has been reported in the microvillous and basal membranes (28) of the postparturient placenta. In the ewe IUGR fetus, the leucine flux across the epitheliochorial placenta to the fetus is also significantly reduced (29). In keeping with these observations, our present investigations in mice demonstrated a reduction in TP and IP leucine transport in response to maternal calorie restriction. This reduction in TP leucine transport could be explained by the reduction in placental LAT2 protein concentrations.

System A amino acid transporters in wt placentas

System A amino acid transporter transports small neutral amino acids such as serine, alanine, and glutamine. In mice, we observed that maternal calorie restriction demonstrated a trend toward enhanced TP MeAIB transport via the system A amino acid transporter isoforms SNAT1 and SNAT2. Similar to our studies, rat maternal under nutrition was associated with reduced placental surface area but higher MeAIB transport due to increased SNAT2 expression (30). In contrast, a low protein diet in rat mothers reduced placental system A transport (31). In the human postparturient placenta, extreme IUGR was associated with reduced syncytiotrophoblast microvillous plasma membrane surface (32) and system A transporter (33).

mTOR nutrient sensing mechanisms in wild-type placentas

These observations cumulatively support the important role played by Glut3 and LAT2 in mediating the maternal calorie-restricted reduction in murine TP glucose and system L amino acid transport. Paralleling the Glut3- and LAT2-mediated reduction in placentofetal glucose and system L amino acid transport, a reduction in total and phosphorylated (active) mTOR was evident along with a reduction in its downstream key molecule total S6K. Although no change in total or phosphorylated 4E-BP1, S6-r, or AMPK was noted, there was a tendency for the phosphorylated form to proportionately increase within the total amount of mTOR, S6K, and 4E-BP1, thereby augmenting activation in the face of a reduction in total amounts (e.g. mTOR, P70S6K). This tendency was also seen with pAMPK/AMPK, reflecting an energy-deficient state due to caloric restriction in the wild-type genotype (34).

Glucose and amino acid transporters in glut3 (+/−) null mice

Because maternal calorie restriction reduced both the glucose (nonenergy dependent) and system L amino acid (energy dependent) transport and spared system A amino acid transport, we argued that these two transport processes, i.e. glucose and system L amino acid transport were interrelated. In particular, system L amino acid transport was dependent on TP or IP glucose transport for energy necessary for its function against the TP concentration gradient. Similar to our present observations, we had previously also demonstrated that in a glut3 (+/−) genotype, a significant reduction in TP and IP glucose transport was encountered (6). Hence, we used this mouse model of gene mutation-induced reduction in TP and IP glucose transport to uncover the in vivo interaction between glucose and system L amino acid transport systems in the placenta.

Our present studies in the glut3 (+/−) background revealed that despite the observed decrease in transplacental and intraplacental glucose transport and the trend toward an increase in placental pAMPK/AMPK (reflecting a diminished energy state), an increase in transplacental system L and system A amino acid transport occurs when compared with the wild-type counterpart. These changes are associated with a trend toward an increase, although not significant in placental LAT1 expression and increased expression of placental SNAT1 and SNAT2. The lack of a concomitant change in placental LAT2 expression and protein concentration is associated with a decrease in intraplacental leucine transport. The placental glucose and system A amino acid transport changes seen in glut3 (+/−) mice exposed to ad libitum calories mimics that seen in response to calorie restriction in the wt genotype, except that in glut3 (+/−) control mice, the compensatory increase in transplacental leucine transport accompanies the increased system A transport that collectively targets protecting the glucose-deficient fetus. However, this increase in transplacental leucine transport occurs at the expense of intraplacental leucine transport, which is reduced in glut3 (+/−) mice paralleling that seen with reduced glucose transport. Hence, it is possible that intraplacental transport of leucine may be dependent on glucose transport, whereas transplacental leucine transport can compensatorily increase when the transplacental supply of fetal glucose is low.

Caloric restriction imposed on glut3 (+/−) placentas

We also stressed the glut3 (+/−) mice by superimposing calorie restriction. This stress on the placental cellular energy state was evident in the observed trend toward an increase in pAMPK/AMPK concentrations similar to that seen with caloric restriction in the wild-type genotype, although not different from the increase of the ad libitum-fed glut3 (+/−) genotype. In contrast to the wild-type background, we observed that an increase in placental Glut3 expression was observed, and although Glut1, LAT1 mRNA, and LAT2 protein did not change, SNAT1 and SNAT2 showed a similar increase in the calorie-restricted group. Thus, it appears that the placental transport system compensated for the reduced maternal supply of glucose and amino acids toward protecting fetal health. This was accomplished by increased expression of placental Glut3 and system A amino acid transporters. Despite this molecular compensation, an exaggerated reduction in transplacental glucose transport was encountered along with the increase in system A amino acid transport. Furthermore, although transplacental but not intraplacental leucine transport decreased compared with the ad libitum-fed glut3 (+/−) counterpart, the transplacental component was no different from the wild-type ad libitum-fed group. In the absence of such compensatory changes in placental transporter expression and function, a further reduction of macromolecular transport may have significantly compromised the survival of the fetus. In addition, the increase in fetal Glut1 with caloric restriction imposed on the glut3 (+/−) genotype further contributed toward protecting fetal survival.

mTOR nutrient-sensing mechanisms in glut3 (+/−) null placentas

Growing evidence supports that mTOR, a serine/threonine protein kinase, not only acts as a cell growth/development marker (35) but it also acts as a nutrient sensor (36). mTOR is down-regulated in human IUGR with a decrease in system L amino acid transporter but not system A transporter (36). Yung et al. (37) have also recently shown that in the human IUGR placenta, mTOR is down-regulated. mTOR is known to exert its effect on placental amino acid transporter activity (38). In both situations, namely the wt calorie-restricted and the glut3 (+/−) ad libitum-fed group, the nutrient sensing system reflected the system L transport system by expressing a tendency of reduction in mTOR and S6K. However, imposition of calorie restriction on the glut3 (+/−) genotype did not further change either mTOR or S6K from that of the glut3 (+/−) ad libitum-fed group. In contrast, 4E-BP1 increased, whereas S6 ribosomal protein did not change in glut3 (+/−) genotype vs. the wild-type with a reduction in 4E-BP1 and no further change in pS6-r upon exposure to caloric restriction in the glut3 (+/−) genotype. Thus, the absence of one allele of glut3 with resultant reductions in transplacental and intraplacental glucose transport and associated energy deficiency reduced the proximal limb (mTOR, P70S6K) and enhanced the distal limb (4E-BP1) of the placental mTOR sensing system. However, when a reduction in glucose transport was combined with a reduction in system L amino acid transport encountered with superimposed maternal calorie restriction, no further change in the mTOR-S6K-S6-r system was observed, except for a reduction in total 4EBP1. Thus, it appears that the transplacental compensatory increments in amino acid transport seen in the glut3 (+/−)-control group come into play in protecting the survival of the fetus, even in the face of additional caloric restriction.

Interplay between placental glucose and amino acid transporters and their function

Our present study demonstrates that in vivo transplacental and intraplacental glucose transport has divergent effects on system L amino acid transport. A reduction in transplacental glucose transport initiates a compensatory increase in the system L and system A amino acid transport systems associated with preservation of p-S6K and p-S6-r and enhanced placental 4E-BP1 (mediators of protein translation) toward protecting the fetus. In contrast, decreased intraplacental glucose transport may have reduced intraplacental leucine without affecting system A amino acid transport, with a concomitant reduction in the mTOR nutrient sensor in both the wild-type caloric restriction and glut3 (+/−) ad libitum groups.

Summary

We have shown the in vivo impact of maternal calorie restriction on murine hemochorial placental glucose and amino acid transport systems and the mTOR nutrient sensing pathway. We have also demonstrated that glut3-dependent reduction of placental and fetal glucose transport increases transplacental system L and system A amino acid transport due to the interplay of placental compensatory and protective mechanisms. These protective mechanisms promote survival of the fetus under adverse nutritional conditions, albeit slowing in utero growth. This diminution in fetal growth has long-lasting consequences that affect the ultimate phenotype of the offspring for multiple generations (39). Thus, transplacental glucose and amino acid transport bear major implications to the health during the entire life trajectory of the offspring.