We tested the hypothesis that hypoxia diminishes the expression and transport of neutral amino acids by system A in full-term human trophoblasts. Cytotrophoblasts from normal human placentas were cultured in standard conditions of 20% O2 or in 1% and 3% O2 for 24 h before assay. Neutral amino acid transport for systems A, ASC, and L was assayed at 24 and 72 h by the cluster-tray technique. Hypoxia during the initial 24 h of culture reduced system A transport by 82% in 1% O2 and by 37% in 3% O2 (P < 0.01) compared with standard conditions. Hypoxia during the latter 24 h of the 72 h in culture reduced system A transport by 55% in 1% O2 and by 20% in 3% O2 (P < 0.05) compared with standard conditions at 72 h. Hypoxia (1% O2) also reduced total amino acid transport by 40% in the more differentiated syncytiotrophoblasts present at 72 h. Northern analysis of trophoblasts in standard conditions showed that subtypes of human amino acid transporter A (hATA1 and hATA2) were each expressed in cytotrophoblasts and syncytiotrophoblasts. Hypoxia decreased expression of hATA1 and hATA2 in both trophoblast phenotypes. We conclude that hypoxia downregulates system A transporter expression and activity in cultured human trophoblasts.
- system ASC
- system L
human fetal development requires an adequate supply of amino acids provided by placental transport (2, 3, 17). The trophoblast bilayer on placental villi is positioned to regulate maternal-to-fetal transport of amino acids. The level of most amino acids is higher in the fetal circulation than in maternal plasma (32), reflecting in part maternal-to-fetal transfer. Fifteen types of transporters for amino acids have been described in the human placenta, and most of these are expressed by the trophoblast (17). Among the multiple transporters for neutral amino acids, systems A, ASC, and L execute the bulk of the neutral amino acid transport across the human trophoblast epithelium.
Fetal growth restriction (FGR) is associated with reduced fetal plasma concentrations of a number of amino acids (4, 6), despite normal or higher maternal concentrations than in normal pregnancies (7, 11). System A activity is markedly decreased in the microvillous membranes of syncytiotrophoblasts in placentas from pregnancies with FGR (8, 9, 14, 25). Whether this defect is a primary contributor to placental dysfunction in FGR or secondary to the metabolic changes that accompany the growth-restricted state is unknown. FGR itself is not a specific disease but a clinical entity with multiple maternal and fetal factors that contribute to the reduced fetal growth profile (33). One set of risk factors for FGR is maldevelopment of the basal plate spiral arterioles, placental underperfusion, and villous hypoxia, which lead to suboptimal fetal growth in the second half of pregnancy (22). The objective of our study was to determine whether a hypoxic insult could reduce trophoblast amino acid transport. Therefore, we tested the hypothesis that hypoxia downregulates the expression and activity of system A transporters in cultures of full-term trophoblasts.
MATERIALS AND METHODS
Cell isolation and culture.
The use of human placentas for this study was approved by the Institutional Review Board of Washington University School of Medicine. Placentas (n = 16) were obtained immediately after a full-term singleton delivery from women with uncomplicated pregnancies. The gestational age was 37–41 wk, newborn weight was between the 10th and the 90th percentile for gestational age, and Apgar scores were ≥8. Villous cytotrophoblasts were isolated by the method described by Kliman et al. (23), with modifications (29). Isolated cytotrophoblasts were plated in medium 199 (Tissue Culture Facility, Washington University) containing 20 mM HEPES (Sigma, St. Louis, MO) and 2 mM l-glutamine (Sigma) supplemented with 10% fetal bovine serum (Hyclone, Logan, UT) in 5% CO2-95% air at 37°C. Nonadherent cells were removed after 4 h by washing three times with PBS, and the adherent cells were transferred to an anaerobic incubator (Forma Scientific, Marietta, OH) or maintained in standard conditions. The anaerobic incubator provided a hypoxic atmosphere, defined as <1% O2 (5% CO2-10% H2-85% N2), with the monitor indicating ≤15 mmHg Po 2. Standard conditions were defined as 5% CO2-95% air (i.e., 20% O2). Previous studies of cells cultured in standard conditions (34) showed that the cytotrophoblast phenotype predominates in the 1st day of culture under these conditions, and the cells differentiate hormonally and morphologically into differentiated syncytiotrophoblasts, which are prominent by 72 h. Culture conditions were modified in selected experiments to maintain the cytotrophoblast phenotype at 72 h by culturing in Ham's Waymouth medium (10) or to facilitate syncytiotrophoblast formation at 24 h (28) by culturing in standard conditions with 100 ng/ml epidermal growth factor (EGF; Upstate Biotechnology, Lake Placid, NY). For selected experiments, the anaerobic chamber's gas mixture was adjusted to 3% O2, with the monitor indicating 40 mmHg Po 2. Previous studies from our laboratory confirmed viability under standard and hypoxic conditions, reflected by trypan blue exclusion, uptake of difluoroacetate, and propidium iodide staining (29).
To determine the effect of hypoxia on amino acid uptake in cytotrophoblasts, the cells were exposed to hypoxia from 4 to 28 h of culture before amino acid uptake assay (see below). To determine the effect of hypoxia on amino acid uptake by syncytiotrophoblasts, cells were cultured in standard conditions for 48 h, exposed to hypoxia from 48 to 72 h, and then assayed.
Uptake measurements in trophoblasts.
Alanine uptake by trophoblasts was measured by the cluster-tray technique (12, 13, 21). Briefly, media were removed and cultures were preincubated for 1 h in Earle's balanced salt solution to reduce amino acid pools. Uptake measurements in cells were begun by replacing Earle's solution with HEPES-buffered Krebs solution containing 1) 0.1 μM [3H]alanine (DuPont, Bloomington, DE) with sodium, 2) 0.1 μM [3H]alanine with sodium + 40 mM α-methyl-aminoisobutyric acid (MeAIB), 3) 0.1 μM [3H]alanine with sodium + 20 mM alanine, or4) 0.1 μM [3H]alanine without sodium.
After incubation for 5 min with sodium or for 20 min without sodium, cultures were washed four times with ice-cold PBS, and cells were disrupted by shaking for 30 min at 37°C in 0.5 ml of SDS. A 0.3-ml aliquot from each well was added to vials containing Optifluor (Packard Instruments, Downer's Grove, IL) and counted in a liquid scintillation counter (Pharmacia LKB Nuclear, Gaithersburg, MD). Alanine uptake per minute was expressed per milligram of DNA, determined in tandem cultures (20). Uptake with 20 mM alanine was extremely low and did not differ in the presence or absence of sodium. Using the four solutions described above, we calculated neutral amino acid uptake as1 − 2 for system A, 2 −4 for system ASC, and 4 − 3 for system L.
mRNA isolation and Northern analysis.
Total RNA was isolated from primary human trophoblasts using TRIreagent (MRC, Cincinnati, OH). Samples of total RNA (12–20 μg) were resolved by electrophoresis using a 1% agarose-1.5% formaldehyde gel. cDNA probes for the human amino acid transporters A1 (hATA1; see Ref.39) and A2 (hATA2; see Ref. 15) were subcloned in pSPORT1 vector and prepared as described elsewhere (15, 39). The probes were labeled with [32P]dCTP using a Prime-It II (Stratagene, La Jolla, CA) labeling kit. RNA was transferred to nylon membranes (Zeta-Probe, Hercules, CA) and hybridized overnight at 42°C. The blots were washed four times with 0.2× sodium chloride-sodium citrate buffer (1× buffer is 0.15 M sodium chloride and 0.015 M sodium citrate) with 0.1% SDS at 65°C. Blots were exposed to a PhosphorImager (Molecular Dynamics) screen for 2–24 h.
Each transport experiment on cells from a single placenta was performed in quadruplicate. To minimize the effect of outlier values, the median of the four uptake measurements for each placenta was determined by the Excel software package. The mean of the medians for standard conditions was compared by the paired t-test (Biostatistics software package; McGraw Hill, New York, NY) with the mean of the medians for primary cultures grown for 72 h in Ham's Waymouth medium (n = 6) or for 24 h in standard medium containing EGF (n = 5). Using separate primary cultures (n = 5 or 6) and similar calculations, we compared amino acid uptake in cells cultured in standard conditions with uptake in cells cultured in hypoxia. P < 0.05 determined significance.
Role of trophoblast differentiation in amino acid uptake.
In previous studies (12), system A transport was dominant in cultures grown for 72 h that contained abundant syncytiotrophoblasts. We questioned whether the enhanced system A transport during culture was secondary to differentiation or time in culture. We, therefore, first compared amino acid transport under standard conditions with two culture paradigms that are known to hinder (10) or enhance (28) differentiation of trophoblasts. Compared with standard conditions, cells cultured in Ham's Waymouth medium maintained the less differentiated cytotrophoblast phenotype during 72 h, showing low levels of human chorionic gonadotropin in medium and predominantly mononuclear cells in culture (data not shown). Amino acid transport studies showed system ASC transport to persist and to be similar to transport by system A, reflecting the phenotype typical of cytotrophoblasts (Fig.1 A). Conversely, when cells were cultured for 24 h in standard conditions without ligand or standard conditions with EGF to facilitate differentiation (28), human chorionic gonadotropin secretion by trophoblasts was markedly enhanced by EGF and syncytiotrophoblasts were abundant (data not shown). Importantly, system A was the dominant amino acid transport system in the ligand-treated cultures, and system ASC activity was diminished (Fig. 1 B). Both findings reflect the enhanced differentiation to syncytiotrophoblasts induced by EGF (28). These results show that amino acid transport characteristics of cultured trophoblasts, rather than time in culture, reflect the state of differentiation of the villous trophoblast.
Effect of hypoxia on amino acid uptake by cytotrophoblasts.
We next determined whether hypoxia decreased neutral amino acid transport in full-term cytotrophoblasts cultured for 24 h (Fig.2 A). Transport of amino acids by system A was significantly lower (by 82%) in cells cultured in 1% O2 than in cells cultured in standard conditions. There was no difference in transport by system ASC or system L in cells cultured in the presence or absence of low O2. Similarly, there was no difference in total amino acid transport between cells grown in 20% O2 and those grown in 1% O2, likely because system A contributes a smaller percentage of total amino acid transport in undifferentiated than in differentiated trophoblasts (12).
We questioned whether the effect of hypoxia on amino acid transport was dependent on the O2 concentration. We used the above paradigms, utilizing exposure to 3% O2 for comparison to standard conditions (Fig. 2 B). In trophoblasts cultured for 24 h in 3% O2, system A transport was higher than in cells cultured for 24 h in 1% O2 but significantly lower (by 37%) than in cells cultured for 24 h in standard conditions.
Effect of hypoxia on amino acid transport by syncytiotrophoblasts.
We next examined the effect of hypoxia on amino acid transport by syncytiotrophoblasts (Fig.3 A). Similar to cytotrophoblasts cultured for 24 h, differentiated cells cultured for 48 h in standard conditions before exposure to 1% O2 for 24 h showed 55% lower amino acid transport by system A than cells cultured for 72 h in standard conditions. There was again no difference in transport by system ASC or system L under the two conditions at 72 h. Importantly, total amino acid transport in the differentiated cells assayed at 72 h was significantly lower (by 40%) after exposure to hypoxia than in standard conditions (Fig. 3 A). Furthermore, cells cultured for 48 h in standard conditions and then for an additional 24 h in 3% O2 had higher system A transport than cells cultured in 1% O2 but had significantly lower (by 20%) system A transport than trophoblasts cultured for 72 h in standard conditions (Fig. 3 B). Cells cultured in standard conditions or 3% O2 had no differences in system ASC, system L, or total amino acid transport.
Hypoxia and system A transporter expression.
The expression of system A transporters has not previously been examined in human trophoblasts (39). We cultured cells for 24 and 72 h in standard conditions and used Northern analysis to examine the expression of hATA1 and hATA2 in the two trophoblast phenotypes (Fig. 4). Each transporter was expressed in cytotrophoblasts and syncytiotrophoblasts.
We next examined expression of the two transporters in trophoblasts cultured under hypoxic conditions (1% O2) and compared expression of the transporters in cells cultured in standard conditions. Hypoxia decreased expression of hATA1 and hATA2 in each of the villous trophoblast phenotypes.
The objective of our study was to determine whether hypoxia reduces amino acid uptake by cultured human trophoblasts. Compared with standard conditions, the data show that low Po 2decreases the activity and expression of system A transport. The hypoxia-induced reduction in system A transport was concentration dependent and affected the cytotrophoblast and the syncytiotrophoblast phenotypes. Importantly, a hypoxia-induced reduction in total amino acid transport was observed in the differentiated trophoblast phenotype, but not in the cytotrophoblast phenotype. This finding reflects the effect of hypoxia on the normally greater contribution of system A transport activity to total amino acid uptake in syncytiotrophoblasts (12). Hypoxia reduced the level of detectable mRNA for the hATA1 and the hATA2 system A transporters in each of the two trophoblast phenotypes. This suggests that hypoxia reduces the transcription of system A transporters or alters the posttranscriptional processing of the system A transporter mRNAs.
The human fetus requires amino acids at ∼10–25 mmol · kg−1 · day−1to supply the anabolic and catabolic requirements for normal growth (1). Some of the nonessential amino acids are newly produced in the placenta, the fetus, or both (2, 3). Other amino acids are delivered by maternal blood perfusing the intervillous space. Studies in vivo in uncomplicated human pregnancies by means of maternal stable isotope infusions show differences in rates of transport of nonessential and essential amino acids (5). However, studies in vivo of pregnancies with FGR have not found a difference between the rate of transfer of nonessential amino acids, e.g., glycine and proline, and essential amino acids, e.g., leucine and phenylalanine (30).
Amino acid transporters in the microvillous membrane of the syncytiotrophoblast mediate uptake from maternal blood, and the basal plasma membrane mediates egress from the syncytiotrophoblast for transfer to the fetal circulation (16, 19, 26, 36). Transporter activities in the two membranes differ, with system A the predominant transporter in the microvillous membrane and system ASC and system L, as well as system A, in the basal plasma membrane. We measured alanine uptake in the presence or absence of sodium and MeAIB to distinguish the three major transporters for neutral amino acids in the trophoblast (12, 13, 21). System A transport is preferential for alanine, serine, and proline, is sodium dependent, and is inhibited by MeAIB. System ASC transports alanine, serine, and cysteine, and this transporter is sodium dependent but not inhibited by MeAIB. System L transports leucine preferentially and is neither dependent on sodium nor inhibited by MeAIB. The data show that the reduction in system A transport is not related to a generalized reduction of membrane transporters, since the activity of system ASC was not decreased in our paradigms. This result mimics the selective effect on nutrient transporters observed in FGR in vivo (8, 9,14, 25). Despite the multiple maternal and fetal risk factors for FGR (33), the system A transport in placental membranes from pregnancies with FGR is decreased, yet expression of glucose transporters and several other amino acid transporters is unchanged (17, 18). Hypoxia may not be the only factor to reduce system A transport in the placenta, but our data suggest that it is one factor that contributes to a reduced system A transport in the trophoblast.
Transporters for different groups of amino acids have been cloned from other epithelia, and most are also expressed in the trophoblast epithelium (17). Our studies indicate that hATA1 and hATA2 are expressed in trophoblasts, although we cannot determine whether there is a differential expression of hATA1 and hATA2 in microvillous and basal plasma membranes of the syncytiotrophoblast. Hypoxia diminishes the expression of both system A transporters. The mediators of the hypoxia effect remain to be determined, but we speculate that hypoxia inducible factor 1α is the prime candidate on the basis of the important role of this transcription factor in cellular responses to hypoxia (35). The effect of hypoxia cannot be attributed to a reduction in trophoblast viability, because the activity of neither system ASC nor system L is reduced. Importantly, previous studies from our laboratory have shown that the cells are viable under the hypoxic conditions employed and that hypoxia upregulates the expression of some proteins, e.g., p53 (24), cyclooxygenase 2 (27), and nuclear factor-κB (unpublished data).
Amino acid transport depends on the villous trophoblast phenotype (12, 27). In primary cultures and in a BeWo cell line, system ASC was highest in the less differentiated trophoblast, and during differentiation, there was increased activity of system A in cultures containing syncytiotrophoblasts. We used conditions that limit (10) or enhance (28) differentiation of cultured trophoblasts to determine whether the changes in transport by system ASC and system A observed during 72 h of culture were related to differentiation or time in culture. The paradigm using Ham's Waymouth medium maintained the cytotrophoblast phenotype for 72 h, as previously described (10). There was enhanced system ASC and reduced system A amino acid transport activity compared with control cells allowed to differentiate in standard conditions. Complementing these results, the use of EGF to facilitate differentiation at 24 h (28) showed that cultures containing syncytiotrophoblasts expressed higher system A activity and lower system ASC activity, independent of time in culture.
Some cases of FGR are associated with placental hypoperfusion and villous hypoxia (22). Umbilical cord blood gas analysis in fetuses with severe growth restriction indicates Po 2 between 15 and 30 mmHg (31, 37,38). These O2 levels are similar to the low Po 2 used in our study. However, our in vitro model of acute hypoxia to study amino acid transport in trophoblasts may not reflect the responses of villi to the chronic state of FGR. Importantly, the reduction in system A transport activity was dependent on O2 concentration. On the basis of our results, we speculate that hypoxia in vivo may diminish the expression of selected amino acid transporters in placentas from pregnancies complicated by FGR. This mechanism would explain the lower fetal blood levels of amino acids transported by system A in such pregnancies (4, 7).
We thank E. Sadovsky for excellent technical assistance and M. Dioneda for help in manuscript preparation.
Address for reprint requests and other correspondence: D. M. Nelson, Dept. of Obstetrics and Gynecology, Washington University School of Medicine, Campus Box 8064, 4566 Scott Ave., St. Louis, MO 63110-1094 (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.
First published September 25, 2002;10.1152/ajpcell.00253.2002
- Copyright © 2003 the American Physiological Society