Fetal growth and development are dependent on the delivery of amino acids from maternal amino acid pools to the fetal blood. This is accomplished via transfer across the apical and basal plasma membrane of the placental syncytiotrophoblast. The aim of this study was to determine whether inhibition of system A (amino acid transporter) was associated with a decrease in fetal weight in the rat. System A is a ubiquitous Na+-dependent amino acid transporter that actively transports small zwitterionic amino acids. In brief, system A was inhibited by infusing a nonmetabolizable synthetic amino acid analog, 2-(methylamino)isobutyric acid from days 7–20 of gestation. On day 20, the rats were killed and tissues (maternal liver, fetuses, and placentas) were collected for analysis. The degree of system A inhibition was determined, as was the impact of said inhibition on fetal and maternal weights, system A-mediated placental transport, and placental system A-mediated transporter expression. Our results suggest that when system A is inhibited, fetal weight is diminished [control group: −3.55 ± 0.04 g (n = 113), experimental group: −3.29 ± 0.04 g (n = 128)], implying an integral role for system A transport in fetal growth and development in the rat.
- intrauterine growth retardation
intrauterine growth retardation (IUGR), defined as birth weight less than two standard deviations below the median for gestational age, affects between 100,000 and 400,000 children each year in the United States (7). IUGR is associated with higher requirements for neonatal intensive care, as well as with specific birth defects and aberrant development. In addition, there is a strong link between IUGR and the incidence of hypertension and diabetes in adult life (2).
Given that an adequate maternal environment is essential if the fetus is to grow normally, a variety of perinatal factors may result in IUGR. Thus genetic endowment, hormonal regulation, maternal infections, maternal weight, and nutritional status all may impact fetal growth and well being, in many cases via alterations in placental function and nutrient supply (38).
The placenta fulfills a variety of essential functions during prenatal life. While its global role is to maintain a protected environment that facilitates optimal growth and development of the embryo and fetus, specific functions include 1) transfer of nutrients both to and from the fetus, 2) provision of metabolic byproducts (i.e., lactate) to the fetus, and 3) provision of hormonal support to both the mother and fetus (4). Twenty to forty percent of the energy supplied to the fetus is in the form of amino acids (4); thus adequate delivery of amino acids to the fetus is essential if normal fetal growth is to occur. Total uptake of amino acids by the fetus is in excess of that predicted by protein accretion rates, suggesting that amino acids also are used as a metabolic fuel source during fetal growth (8). Fetal concentrations of nearly all amino acids are greater than maternal concentrations, which suggests that the placenta actively mediates amino acid availability to the developing fetus (3). Amino acid concentrations are even higher within the placental trophoblast.
In the human hemomonochorial placenta, the maternal-facing (apical) membrane and the fetal-facing (basal) plasma membranes of the syncytiotrophoblast are the primary barriers to solute transfer between maternal and fetal circulations. A variety of energy-dependent amino acid transporters have been demonstrated on the apical and basal membrane (23). These transporter agencies have differing substrate affinities, kinetic properties, and membrane orientations (23). Among the Na+-dependent neutral amino acid transporters are systems A, ASC, and β, whereas the Na+-independent neutral, anionic, and cationic transporters include system L, system X , and systems y+, bo,+, and y+L, respectively (23). We have chosen to focus primarily on system A, responsible for the Na+-dependent transport of small neutral amino acids, because of both the central metabolic roles played by its substrates and the availability of a nonmetabolizable specific substrate, 2-(methylamino)isobutyric acid, with which it may be inhibited (24).
Maternal malnutrition has been shown to impact placental amino acid transport. Placental transport attributable to system A is diminished in both human IUGR pregnancies and experimental pregnancies complicated by malnutrition (30, 32). Although these studies have linked impaired fetal growth and decreased system A activity, causation has not been established. Specifically, published work has not established whether the decrease in amino acid transfer causes IUGR or is a result of IUGR. Consequently, our goal in this work was to test this hypothesis and to directly examine the impact of system A inhibition on fetal growth and development.
MATERIALS AND METHODS
14C-labeled 2-(methylamino)isobutyric acid (MeAIB) was obtained from American Radiolabeled Chemicals (St. Louis, MO). ScintiSafe scintillation fluid was obtained from Fisher Scientific (Pittsburgh, PA). Dr. Jeffery Rothstein (The Johns Hopkins University, Baltimore, MD) generously supplied antibodies against rat [GLAST1 (EAAT1), GLT1 (EEAT2), and EAAT4], and anti-EAAC1 (EAAT3) was developed as previously described (33). All other chemicals were of reagent grade or of the highest grade commercially available.
Timed-pregnant Sprague-Dawley rats (with jugular and carotid cannulations and skull-mounted ports) were obtained on the fifth postimpregnation day from Zivic-Miller (Zellenople, PA). The rats were housed in a temperature-controlled room with 12:12-h light-dark cycles. This study was approved by the Laboratory Animal Medical Ethics Committee of the University of Florida. Upon arrival, rats were weighed, matched by weight, and separated into two groups, control and experimental. On day 6 of gestation, both groups were given 35 g of food; subsequently, control animals were pair fed with their experimental counterparts. Also on day 6, experimental rats received a 300-μl bolus of MeAIB (1.18 mmol MeAIB/100 g body wt) in saline intravenously; control animals were given an equivalent volume of normal saline. Immediately after these boluses were administered, tethers were affixed to skull-mounted ports, and MeAIB (40 μmol/h in 100 μl of saline) or saline (100 μl/h) was infused continuously until day 20 of gestation. On day 20, 0.5 ml of blood was withdrawn from a previously placed carotid catheter and utilized for determination of serum amino acid levels. The animals were then weighed individually, and boluses of 10 μCi/100 g body wt of [14C]MeAIB and 10 μCi/100 g body wt ofd-[3H]aspartate were administered intravenously. The animals were subsequently anesthetized with pentobarbital sodium, and tissues were collected from each animal 20 min after initial injection of radioactivity. This time point was chosen on the basis of initial studies demonstrating relatively linear uptake of radioactive tracer into fetuses through this time point (data not shown). Serum, placentas, and fetuses were collected and snap frozen in liquid nitrogen, except those utilized for placental membrane preparations. Frozen tissues were then individually powdered under liquid nitrogen in a mortar, and each tissue was then deproteinized in four volumes of 6% perchloric acid. Tissue was then processed utilizing a Virtashear (Gardinier, NY) and subsequent manual douncing. After centrifugation at 150,000 g for 15 min at 4°C, the supernatants were neutralized with 30% KOH to a pH of 7.0. After neutralization, the supernatants were centrifuged at 2,000 gfor 10 min at 4°C (pellet discarded). Subsequently, experimental and control samples were assayed for radioactivity by adding 100 μl of supernatant to 4.5 ml of scintillation fluid with subsequent shaking (1 h) on an orbital shaker. The samples were then placed in a scintillation spectrophotometer to determine disintegrations per minute (dpm). Relative inhibition of system A was then calculated by comparing concentrations of [14C]MeAIB between the control and experimental groups.
Preparation of placental apical and basal predominant membrane vesicle/protein and enzyme determinations.
Apical membrane vesicles were prepared by using the method of Booth et al. (6), and the basal predominant membrane preparation was prepared by using the method first described by Kelley et al. (25). Both of these preparations were subsequently modified for use in rats and verified at specific times of gestation (31, 36). Timed-pregnant rats, after infusion of either placebo or MeAIB as denoted in Animals, were killed on day 20. Placentas were removed and finely minced, manually homogenized in a small Dounce-type homogenizer, and saved as homogenate. Approximately 30 g of placental tissue were then utilized for each preparation. After crude dissection of decidual tissue, the remaining tissue was minced and filtered through a 210-μm nylon mesh. The filtrate was then refiltered and centrifuged at 800g for 10 min at 4°C (pellet discarded), and the supernatant was again centrifuged at 10,000 g for 10 min (pellet discarded). The subsequent supernatant was centrifuged at 150,000 g for 25 min, and the pellet was homogenized in a motor-driven Teflon glass homogenizer and brought up in 10 mM MgCl2(supernatant discarded). After subsequent homogenization, the solution was centrifuged at 2,000 g for 12 min, the pellet was discarded, and the supernatant was again centrifuged at 150,000g for 25 min. The apical membrane pellet was then collected, and the supernatant was discarded. Basal membranes were isolated utilizing tissue previously withheld on the 210-μm nylon mesh, beginning by washing the tissue three times with 50 mM Tris · HCl (pH 6.9). Three additional washes were performed with 5 mM Tris · HCl (pH 6.9). The tissue was added to Tris-sucrose-EDTA buffer (300 mM sucrose, 2 mM Tris base, and 10 mM Na-EDTA, pH 6.9), incubated at room temperature for 10 min, and again filtered through 210-μm nylon mesh. After the further addition of 10 mM EDTA solution, the tissue was sonicated for 20 s at 4°C and filtered, and the filtrate was centrifuged at 4,000 g for 10 min at 4°C. The resultant supernatant was then centrifuged at 10,000g for 2 min at 4°C. After the pellet was discarded, basal membranes were sedimented by centrifugation at 150,000 g for 25 min at 4°C. Both apical membrane and basal membranes were resuspended in sucrose buffer (300 mM sucrose and 10 mM HEPES-Tris base, pH 7.4). Membranes were stored in liquid nitrogen until use.
Membrane protein content was determined by the method of Lowry (28), utilizing bovine serum albumin as standard. In general, enzyme assays were performed on fresh (unfrozen) samples within 24 h of isolation. Activities of alkaline phosphatase, a marker for the apical membrane, and [3H]dihydroalprenol binding, a marker for the basal membrane, were determined by methods used previously in this laboratory (31). Basal membrane vesicles were enriched 25- to 40-fold in [3H]dihydroalprenol binding compared with apical membrane vesicles, whereas apical membrane vesicles were enriched 15- to 21-fold in alkaline phosphatase compared with basal membrane vesicles. These values are similar to those we have reported previously (31,36). Samples were stored at 4°C until use.
Timed uptakes of tritiated amino acid compounds were performed as described previously (37). In brief, 20-μl aliquots of membrane vesicles (20–70 μg protein/sample) suspended in buffer containing 0.2 mM CaCl2, 10 mM MgCl2, 10 mM HEPES-KOH buffer, and 280 mM sucrose, pH 7.5, were preincubated in a glass culture tube at 37°C for 2–4 min. Uptake studies were initiated by the addition of 80 μl of incubation solution (10 mM HEPES-KOH, 125 mM sodium or potassium thiocyanate, 10 mM MgCl2, 0.2 mM CaCl2, pH 7.5, and indicated concentrations of radioactively labeled substrate, with additional sucrose to maintain osmolarity at 310 mol) to the preincubated membrane suspension. System ASC activity was assessed as the Na+-dependent uptake of 50 μM serine in the presence of 3 mM arginine, 3 mM leucine, and 10 mM MeAIB. System y+activity was determined by determining the Na+-independent uptake of 50 μM arginine in the presence of 2 mM 2-amino-2-norbornane carboxylic acid. System A activity was determined by assaying the Na+-dependent uptake of 50 μM MeAIB. Each of the above activities was measured at 10 s. System X activity was measured after preloading of vesicles with 50 mM KCl, as we have described previously (33), utilizing the uptake (30 s) of 1 μM glutamate in the presence or absence of 500 μM d-aspartate. Timed uptakes were terminated by the rapid addition of 3 ml of ice-cold stop solution (100 mM sodium or potassium thiocyanate, 100 mM sucrose, and 10 mM HEPES-KOH, pH 7.5) and filtered through a 0.45-μm nitrocellulose filter prewetted with 3 ml of 1 mM unlabeled MeAIB (or other test amino acids) to reduce nonspecific binding. The tube was rinsed with an additional 3 ml of stop solution, and the filter was washed three additional times. The filter was then placed into a vial and dissolved in 4.5 ml of scintillation fluid, and dpm were determined in a scintillation spectrophotometer. Uptakes were corrected with a membrane blank in which 3 ml of ice-cold stop solution were added to the membrane sample before the addition of radiolabeled substrate. This blank value was then subtracted from each experimental uptake. Initial rates of uptake were estimated from uptakes at 10 s. All incubations (blanks and uptakes) were performed at least in triplicate. System A activity, defined as the Na+-dependent component of MeAIB uptake, was determined by subtracting uptake in the presence of an inwardly directed K+ gradient from that in the presence of an inwardly directed Na+ gradient (41).
Placentas from both the experimental and control rats were homogenized as described previously (33). SDS-PAGE (7.5%) was used to separate placental homogenate proteins (27), which were then electrotransferred to a 0.45-μm nitrocellulose membrane. The presence of immunoreactive GLAST1 (EAAT1), GLT1 (EAAT2), EAAC1 (EAAT3), or EAAT4 protein was evaluated by immunoblot analysis and visualization with a chemiluminescence kit, as described previously (33). We demonstrated previously the specificity of the antibodies for their respective substrates in rat placental tissue (33). Densitometric analyses were performed on all immunoreactive bands using a ScanJet 6300C (Hewlett-Packard, Palo Alto, CA) and ScionImage software (Scion, Frederick, MD). Control experiments were performed to ensure that the absorbance values were in the linear range of the densitometer and the film.
ATA2 cDNA was the kind gift of Dr. V. Ganapathy (University of Georgia, Athens, GA). Placental tissue (1 g) was snap frozen and ground to a powder in a mortar and pestle under liquid nitrogen The placental powder was then added to denaturing solution (4 M guanidinium thiocyanate, 0.5% N-lauryl sarcosine, 25 mM sodium citrate, pH 7.0, and 100 mM β-mercaptoethanol) to inhibit RNase activity. The solution was homogenized with 10 strokes in a glass homogenizer with a motor-driven Teflon pestle. Total RNA was isolated by the method of Chomczynski and Sacchi (9), and poly(A)+selected mRNA was isolated from 500 μg of total RNA using the Poly ATract System (Promega, Madison, WI). Equal amounts of poly(A)+ selected mRNA were loaded per lane (3 μg) and subjected to 1% agarose gel electrophoresis in the presence of 0.02 M formaldehyde. In later isolations, given the observation that ATA2 RNA was sufficiently abundant to be easily detected in samples of total RNA, we employed the RNeasy mini-protocol for isolation of total RNA from animal tissue, as provided by Qiagen (Valencia, CA). For these blots, 15 μg of RNA were loaded per lane. The RNA was then capillary transferred to nylon membrane and hybridized with32P-labeled cDNA probe prepared by random priming extension (GIBCO BRL, Gaithersburg, MD). The resulting autoradiograph was quantified by densitometry (ScanJet 6300C and ScionImage). To correct for loading differences, the levels of mRNA were normalized to constitutively expressed cathepsin B (33). Preliminary experiments were performed to ensure that the autoradiographic exposures were within the linear range of the film and densitometer. Arbitrary densitometric values corrected as noted above for loading were compared between groups utilizing paired Student'st-tests, as performed using GraphPad Instat (GraphPad Software, San Diego, CA).
Serum amino acid determinations.
Serum amino acid levels were performed in the laboratory of Dr. R. Webb (Virginia Tech University, Blacksburg, VA), utilizing triethlyamine derivatization of samples for HPLC analysis (5).
Preliminary studies were conducted to determine the amount of MeAIB required to maintain serum MeAIB levels in the 2–4 mM range. These levels were chosen on the basis of the experimentally determined Michaelis-Menten constant for system A of ∼500 μM in placental apical membrane (36). On the basis of these data, as well as published total concentrations of system A substrates in pregnant female rats of 0.6–1.5 mM, we hypothesized an ∼60–80% inhibition of system A activity within the apical membrane if maternal MeAIB levels could be maintained in the 2–4 mM range (1, 35,36). Samples were obtained on various days of infusion (40 μmol/h) in preliminary experiments employing four pregnant female rats (Fig. 1). Average MeAIB levels in maternal serum were 4.9 mM on day 7 (postbolus) and 5.7 mM on day 20.
Maternal glucose values.
A potential confounding variable in the studies was the impact of MeAIB infusion on maternal serum glucose values, given that system A substrates are generally gluconeogenic. Serum glucose values did not differ on day 20 of pregnancy between control (90 ± 19 mg/dl) and experimental animals (109 ± 10 mg/dl). These values are similar to those published previously (12). Glucose levels earlier in pregnancy also were similar between groups (data not shown). Glucose levels in the fetus were not measured because of the lack of gluconeogenic activity in the fetal liver (13,18).
Serum amino acid levels.
Maternal serum amino acid levels were determined to define whether MeAIB infusion was associated with significant alterations in amino acid availability to the fetus. We hypothesized that the concentrations of system A substrates would be higher in the experimental rats compared with the control rats. Indeed, levels of alanine (78 ± 13 vs. 39 ± 12 mg/l, P < 0.01), proline (39 ± 7 vs. 17 ± 3 mg/l, P < 0.01), and valine (22 ± 2 vs. 16 ± 1 mg/l, P < 0.01;n = 4 for each group, in duplicate) were significantly enhanced in the experimental compared with control groups. We hypothesize that these increases may have resulted from system A inhibition with subsequent limitation of efflux from the maternal circulation into maternal and fetal cellular compartments. Interestingly, the serum level of glycine, another potential system A substrate, was unaffected, as was the serum level of all other amino acids.
Maternal, fetal, and placental weights.
Maternal weight gain was examined in both the experimental and pair-fed control group. Given the ubiquitous distribution of system A in mammalian cells, we had hypothesized that maternal weight gain would be impaired by MeAIB infusion. In fact, as shown in Table1, maternal weight gain was unaffected by MeAIB infusion (P ≤ 0.29). Placental weight also was unaffected (P ≤ 0.19), in contrast to our previous findings in nutritionally induced IUGR (32). Fetal weight, however, was impacted. MeAIB infusion was associated with a significant decrement (P < 0.001) in fetal weight of 7.3%. There was no significant difference between litter size in control compared with experimental groups (P ≤ 0.775), and no morphological abnormalities were noted in experimental fetuses; thus neither factor appeared to account for the difference in fetal weights noted above.
System A inhibition in vivo.
Inhibition of transfer attributable to system A across the placenta was evaluated by determining (on day 20 of gestation) the accumulation of [14C]MeAIB in the fetus after maternal injection. d-[3H]aspartate, transferred by system X exclusive of system A (17), also was injected in an effort to determine whether potential changes in system A activity might impact the activity of other unrelated transfer proteins. As demonstrated in Fig.2, transfer of radiolabeled MeAIB to the fetus was inhibited by 55% (P ≤ 0.05). Conversely, transfer of radiolabeled d-aspartate was unaffected.
Transport studies were conducted to determine whether prolonged exposure to millimolar concentrations of a specific system A substrate impacted the activity of system A in isolated apical membrane and basal membrane vesicles. Controls included determination of system ASC activity, which in part overlaps in substrate specificity with system A, and system X and y+ activities, responsible for the Na+-dependent transfer of anionic amino acids and the Na+-independent transfer of cationic amino acids, respectively, which do not overlap with system A (20). System A, ASC, X , and y+ activities were assayed in apical membrane vesicles, whereas system A and ASC activities were assayed in basal membrane vesicles. We demonstrated previously that the apical membrane preparation utilized represents primarily apical layer two of the rat hemotrichorial placenta, whereas the basal membrane preparation represents predominately basal layer three of the rat placenta. Because of the fenestrated nature of layer one of the hemotrichorial placenta, these membranes are believed to represent the first and last membrane layers encountered by substrate as it moves from the maternal to the fetal circulation and, thus, are conceptually analogous to the human microvillous and basolateral membrane (31, 36). We anticipated that the activities of systems X and y+ would be independent of the activity of system A. Transport was assessed in both apical and basal membrane vesicles, although measurements in basal membrane vesicles were limited because of tissue availability. Measurements were normalized to control for interexperiment variability. Transport attributable to systems ASC and A was not significantly different in control compared with experimental basal membrane (Fig. 3 B). Conversely, in apical membranes, whereas serine transport attributable to system ASC and arginine transport attributable to system y+ were not different between control and experimental groups, transport activity in system A was diminished by ∼40% and that in system X was diminished by >50% in experimental compared with control animals (Fig. 3 A). These data suggest that excess system A substrate in the maternal circulation results in diminished system A transport activity at the maternal fetal interface. This effect is not entirely specific, as documented by the unexpected diminution noted in apical membrane system X activity.
Although the lack of a suitable antibody recognizing ATA2, the protein most likely responsible for placental system A activity (42), limited our investigation of protein expression in control and experimental groups, we performed immunoblot analysis utilizing antibodies against the four glutamate transport proteins (EAAC1, GLT1, GLAST, and EAAT4), known to be present in rat placenta (33), because of the unexpected decrease in system X activity noted in Transport studies(Fig. 4). Homogenates were utilized for these analyses because of the limited supply of apical and basal membrane available to us. Each homogenate represented a pool of placentas (10–50) subsequently utilized for apical and basal membrane preparations. Densitometry was performed, and data were normalized to those from the experimental group. Relative amounts of each protein in control compared with experimental homogenates were EAAC1: 1.89 ± 0.28, n = 3; GLT1: 1.51 ± 0.59, n = 3; GLAST: 1.2 ± 0.7, n= 3; and EAAT4: 1.6 ± 0.44, n = 2. These results are concordant with the aforementioned transport data and demonstrate a generalized downregulation of glutamate transfer proteins in the experimental group.
Northern analysis was performed utilizing a cDNA coding for rat ATA2. As shown in Fig. 5, steady-state levels of ATA2 mRNA were not different between control and experimental groups (control: 0.86 ± 0.12; experimental: 0.80 ± 0.29,P > 0.05; results expressed in arbitrary densitometric units normalized to cathepsin B, ±SD).
A variety of previous studies have demonstrated links between intrauterine growth retardation and diminished placental amino acid transfer. These studies, conducted in rodent, ovine, and human subjects, both in vivo and in vitro, have examined amino acid transport as it varies with maternal malnutrition, heat stress, and vascular insufficiency, among others (15, 19, 23, 26, 30, 32, 40). A relatively constant finding, regardless of the system examined, has been the diminution of system A amino acid transport activity. Unfortunately, the significance of these findings has been uncertain, especially given that other transport systems may be downregulated in these models in addition to system A, raising the question of whether effects noted are of importance in the pathogenesis of in utero growth failure or merely an associated epiphenomenon. The current study addresses this issue directly through the in vivo inhibition of system A, utilizing the nonmetabolizable system A substrate MeAIB. Christensen and colleagues (10, 11, 39) developed the concept of utilizing N-methylated amino acid derivatives in vivo to examine the physiological impact of single amino acid transport system inhibition; more recently, Freeman et al. (16) used a single dose of MeAIB in vivo to inhibit the uptake of system A substrates during liver regeneration. Our study is the first to use these compounds throughout the major portion of gestation via intravenous infusion, allowing relatively constant serum levels and, thus, relatively constant degrees of inhibition to be maintained despite ongoing renal excretion. Despite initial concerns about the potential for maternal wasting during infusion, pregnant animals thrived throughout the treatment period, gaining as much or more weight than animals receiving saline infusions. These results indicate that the presence of overlapping systems for amino acid transfer into maternal cells compensates for blockade of system A-mediated uptake. The presence of elevated serum levels of system A substrates in the treated maternal animals may have facilitated this effect and also demonstrate the effectiveness of the MeAIB infusion.
The major finding of this study was that partial blockade of system A-mediated uptake throughout the majority of gestation resulted in impaired fetal growth. The fact that this impairment occurred in the face of at least equivalent maternal dietary intake, growth, and serum amino acid levels, and in the absence of obvious fetal anomalies, suggests that growth impairment was, in fact, the result of this inhibition, either in the placenta (as we demonstrated) or fetus. Our data also suggest that other amino acid transfer systems with overlapping specificities were unable to compensate for system A inhibition. Unexpectedly, the activity of system X also was diminished in apical membrane vesicles. Conversely, system X -mediated transfer in vivo from mother to fetus was not impacted. As a result, the contribution of the observed in vitro system X inhibition to the observed decrement in fetal growth is difficult to assess. Our studies also were unable to differentiate between placental and fetal effects, data perhaps obtainable only through the use of organ-specific “knockout” models.
System A activity in apical membranes derived from the placentas of experimental animals was diminished by ∼40% compared with that of control animals. The mechanism underlying this observed regulation cannot be deduced from these studies, although maternal serum amino acid levels, nutritional status, or uteroplacental blood flow would seem unlikely factors under the conditions of this study. Conversely, degree of transporter saturation or alterations of intraplacental amino acid concentrations may play significant roles. System A activity on the basal membrane was not significantly decreased in experimental animals compared with controls. System ASC activity, the Na+-dependent transport system sharing the most substrate specificity with system A, was not altered on either the apical or basal membranes. This finding does not preclude that, under the conditions present in vivo, system ASC may play a larger role in the transfer of system A substrates in the presence of MeAIB excess. Similarly, the activity of system y+, a Na+-independent transporter of cationic amino acids, was unaffected under the conditions of these studies. The finding that system X activity on the apical membrane was diminished by 50% in the experimental group was unexpected. System X , associated with the Na+-dependent transport of anionic amino acids, has been demonstrated previously on the apical membrane of both human and rodent placenta (33,34). In the ovine in vivo model, glutamate transfer does not occur from the maternal to the fetal circulations (43). Such data have not been conclusively replicated in the human or rodent. The current study suggests an effective transfer ofd-aspartate, a system X substrate, from the maternal to the fetal circulations. Unlike MeAIB, however, which clearly has been shown to be nonmetabolizable (11),d-aspartate may be metabolized by d-aspartate oxidase, responsible for the conversion of d-aspartate to oxaloacetate, ammonia, and hydrogen peroxide. This enzyme is present to some degree in rat tissues (44); thus some degree of metabolism may occur before placental transfer. Conversely, the possibility that placental transfer occurs is suggested by the presence of d-aspartate in the normal developing (fetal) central nervous system, as well as in a variety normal adult tissues including the placenta (21). The observed relationship between system A and system X is more difficult to discern. Alanine (a system A substrate) in conjunction with α-ketoglutarate may, by the action of alanine aminotransferase, be converted to pyruvate and glutamate. Glutamate uptake in neurons has been linked to glucose uptake, with subsequent production of lactate, utilized as a fetal fuel (14, 29). Thus, intuitively, diminished alanine uptake into the placenta or fetus might be expected to increase, rather than decrease, needs for glutamate. The decrease in system X activity demonstrated here is thus, on the surface, counterintuitive. Conversely, transfer of tritium associated with d-aspartate from mother to fetus was not diminished in the experimental group, suggesting either the presence of other anionic amino acid transfer agencies (i.e., Na+-independent system X ) or a change in driving forces, effecting increased anionic amino acid transfer in vivo despite the diminished transfer shown in vitro. To further examine these issues, we examined the expression of amino acid transport proteins previously shown to be present within the rodent placenta. These results, demonstrating an overall downregulation of transport protein expression within the placenta, corroborate our measurements of transport within the apical membrane, especially given our previous demonstration of these proteins on the placental apical and basal membranes (33). Unfortunately, the lack during these studies of an available antibody against the recently cloned system A transport protein ATA2 made study of system A protein expression problematic. Steady-state mRNA levels were unaffected by our treatment protocol.
In summary, inhibition in vivo of the system A amino acid transport system utilizing the nonmetabolizable amino acid analog MeAIB is associated with suboptimal fetal growth. This finding suggests that alterations in the activity of placental amino acid transport systems may cause, rather than be merely associated with, alterations in fetal growth.
This work was supported by a grant from the March of Dimes, as well as by National Institute of Child Health and Human Development Award RO1-HD-29934.
Address for reprint requests and other correspondence: D. Novak, Box 100296, Univ. of Florida College of Medicine, Gainesville, FL 32610-0296 (E-mail:).
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 © 2002 the American Physiological Society