Reductions in fetal plasma concentrations of certain amino acids and reduced amino acid transport in vesicle studies suggest impaired placental amino acid transport in human fetal growth restriction (FGR). In the present study, we tested the hypothesis of an impairment in amino acid transport in the ovine model of hyperthermia-induced FGR by determining transplacental and placental retention and total placental clearance of a branched-chain amino acid (BCAA) analog, the nonmetabolizable neutral amino acid aminocyclopentane-1-carboxylic acid (ACP), in singleton control (C) and FGR pregnancies at 135 days gestation age (dGA; term 147 dGA). At study, based on the severity of the placental dysfunction, FGR fetuses were allocated to severe (sFGR, n = 6) and moderate FGR (mFGR, n = 4) groups. Fetal (C, 3,801.91 ± 156.83; mFGR, 2,911.33 ± 181.35; sFGR, 1,795.99 ± 238.85 g; P < 0.05) and placental weights (C, 414.38 ± 38.35; mFGR, 306.23 ± 32.41; sFGR, 165.64 ± 28.25 g; P < 0.05) were reduced. Transplacental and total placental clearances of ACP per 100 g placenta were significantly reduced in the sFGR but not in the mFGR group, whereas placental retention clearances were unaltered. These data indicate that both entry of ACP into the placenta and movement from the placenta into fetal circulation are impaired in severe ovine FGR and support the hypothesis of impaired placental BCAA transport in severe human FGR.
- aminocyclopentane-1-carboxylic acid
- transport systems
investigations of fetal growth restriction (FGR) are based on the hypothesis that it represents a fetal adaptation to a stunted placental development, which has a reduced ability to supply oxygen and nutrients to the fetus. Experiments with human FGR syncytiotrophoblast vesicle preparations indicate that certain amino acids may be among the nutrients with the largest reduction in placental transport capacity (12, 13, 18). For example, the activities of the system A transporter, which transports small neutral amino acids, and the β- (taurine) transport systems in human syncytiotrophoblast microvillous membrane preparations have been reported to be significantly reduced in FGR (12, 18, 26). Additionally, reductions in the uptake of leucine in both microvillous and basal microvesicle preparations highlight possible alterations in the L transport system (13).
In vivo impairment of amino acid placental transport in human FGR has been difficult to quantify. Some studies have demonstrated a significant reduction in the fetal plasma concentration of several amino acids (7, 10). Although this observation appears to support the hypothesis of a reduced placental transport capacity, it is of uncertain value because fetal plasma concentrations depend on factors other than placental transport alone. Studies utilizing the maternal infusion of 13C-labeled leucine and phenylalanine in human FGR pregnancies have shown a reduced fetal/maternal plasma enrichment ratio compared with normal pregnancies (19, 27), in agreement with the hypothesis of a reduced transplacental flux of these amino acids. Furthermore, the magnitude of these reductions is correlated with the severity of the condition (19). However, to calculate changes in transplacental flux from enrichment ratios, measurements of fetal amino acid disposal rates need to be made, which are not feasible in humans. To overcome this limitation, in vivo studies have been conducted in a sheep model of FGR, in which the fetal/maternal enrichment ratio and the fetal disposal rate of an essential amino acid could be measured simultaneously (1, 30). These studies have confirmed a reduction of transplacental flux from maternal to fetal plasma for leucine in the ovine growth-restricted fetus, in agreement with the human studies hypothesis (19, 30). Specifically, in the ovine leucine experiments, the simultaneous infusion of two different leucine tracers into maternal and fetal circulation demonstrated that the direct flux of leucine from mother to fetus was significantly reduced in FGR pregnancies despite an insignificant change in umbilical leucine uptake per kilogram fetus (30). As with leucine, the direct threonine flux from maternal to fetal circulation was significantly reduced (1). It is uncertain, however, whether these changes in leucine and threonine direct flux solely represent changes in placental transporter capacity. An essential amino acid entering the placenta from its maternal surface is partitioned into placental protein turnover, placental oxidation, and direct transport into the umbilical circulation (30). Therefore, changes in placental metabolism could change the direct flux of a tracer essential amino acid into the fetus independently of changes in transport capacity.
Nonmetabolizable amino acids represent a class of analogs that allow examination of placental amino acid transport in isolation without confounding variables introduced by placental and fetal metabolism. Aminocyclopentane-1-carboxylic acid (ACP) is a nonmetabolizable analog of neutral amino acids with hydrophobic side chains [the branched-chain amino acids (BCAA) and methionine and phenylalanine] (8, 16). Previously, we have reported that, in normal ovine pregnancy, ACP crosses the placenta rapidly from mother to fetus (14) and that this transport is a two-step process: 1) flux into the placenta, where ACP accumulates at a higher concentration than in maternal and fetal plasma, and 2) flux from the placenta into the umbilical circulation. The establishment of a concentration gradient between placenta and fetus indicates that ACP transport across the basal membrane of the trophoblast is a rate-limiting step. The present study was designed to measure the placental retention and transplacental clearances of ACP in the ovine FGR model, which has previously demonstrated a significant reduction in the direct fluxes from mother to fetus of leucine and threonine (1, 30). By measuring separately the placental retention and transplacental clearances, we sought to determine in vivo whether any reduction in transplacental transport could be attributed to specific reductions in either apical or basal membrane transport capacity.
MATERIALS AND METHODS
Animal care and surgery.
Seventeen time-mated 2- to 3-yr-old Columbia-Rambouillet ewes pregnant with a single fetus were studied. Animal care was in strict compliance with National Institutes of Health guidelines within an American Association for Accreditation of Laboratory Animal Care-certified facility, and the University of Colorado Health Sciences Center Animal Care and Use Committee approved the study. At ∼39 days gestational age (dGA; term ∼147 dGA), ewes were allocated to either a thermoneutral control (C) or hyperthermic treatment, which has been documented to produce a placental insufficiency-induced growth-restricted fetus (FGR) (31). A total of seven ewes were maintained in the control regime (20 ± 2°C, 24 h, 30% relative humidity), and 10 ewes were moved to individual pens in the environmental chamber. The temperature, humidity, and lighting regimens used in this treatment, the measurements of core body temperature (CBT) and breaths per minute (bpm), and the diet and feeding of alfalfa hay pellets to animals were as previously described (29). Ewes were maintained in the hyperthermic environment until 120 ± 1 dGA, ∼80 days in treatment. After this treatment, ewes were moved to the control environment and housed with age-matched control ewes, and feeding for both groups continued ad libitum.
At ∼128 dGA, ewes underwent surgery for placement of fetal and maternal catheters. After 24 h of fasting and 12 h without water, burenorphine (1 mg/kg sc) and diazepam (0.11 mg/kg iv) were administered ∼1 h before sedation with ketamine (4.4 mg/kg iv) and placement on isoflurane (1–3%) inhalation. The uterus was exposed through a midline abdominal incision, and the uterine vein (V) draining the gravid horn was catheterized. The fetus was exposed through a 6-cm incision of the uterine wall, and polyvinyl catheters (1.4 mm OD) were placed in the fetal pedal artery (α) and vein and in the umbilical vein (γ). An additional catheter was placed in the amniotic cavity to allow antibiotic prophylaxis delivery to the fetus. After closure of the uterus and abdominal wall, the maternal femoral artery (A) and vein (V) were catheterized. After surgery, animals were maintained as previously described (28).
Uteroplacental and fetal oxygenation and blood flow determination.
At ∼133 dGA, uteroplacental and fetal oxygenation and blood flow determinations were conducted at room temperature. All animal CBTs were ∼39.1°C. Ethanol was used to determine uterine and umbilical blood flow by the steady-state diffusion method (6). The ethanol infusion began 90 min before sampling commenced to ensure steady-state ethanol concentrations during the course of the four-sample collection study. Four sets of 1-ml blood samples were drawn at ∼20-min intervals from A, V, α, and γ for analysis of metabolites (glucose, lactate, and amino acids), maternal and fetal blood gases, and ethanol concentration (steady-state period).
Plasma glucose and lactate concentrations were measured in a glucose analyzer (Yellow Springs Instrument, Yellow Springs, OH; model 2700 Select). Blood samples were drawn into heparinized capillary syringes and analyzed for pH, Po2 and Pco2 (mmHg), oxygen capacity (mM/l), oxygen saturation (%), carboxyhemoglobin (%), methemoglobin (%) and bicarbonate concentration (mmol/l) by use of an ABL 520 blood gas analyzer (Radiometer, Copenhagen, Denmark).
Blood samples for ethanol determination were collected in EDTA-treated syringes, and ethanol concentrations were determined in triplicate aliquots with a quantitative enzymatic UV determination method (cat. no. Alcohol 332-UV; Sigma, St. Louis, MO). Hematocrit was determined as packed cell volume. Uterine and umbilical plasma flows were calculated using uterine or umbilical flow multiplied by (1 − fractional maternal or fetal hematocrit, respectively). Ethanol clearance rates, uterine and umbilical oxygen, and glucose uptake were calculated using the Fick principle, as described previously (6, 23).
Determination of amino acid concentrations.
Maternal and fetal arterial plasma samples for determination of the BCAA (leucine, isoleucine, and valine), as well as phenylalanine and methionine, were frozen within 5 min of collection at −70°C until the day of analysis, at which time they were thawed quickly and deproteinized with 10% sulfosalicylic acid containing 0.3 μmol/ml norleucine as an internal standard. The pH was adjusted to 2.2 with 1.5 N LiOH. After centrifugation, the supernatant was analyzed with a Dionex high-performance liquid chromatography (HPLC) amino acid analyzer (Dionex, Sunnyvale, CA). Amino acid concentrations were measured after reaction with ninhydrin at 570 nm. The same HPLC column was used for all samples from an individual animal. Reproducibility within the same column had a mean value of ±2%.
ACP experimental protocol.
The ACP experimental protocol was modified for use in growth-restricted fetuses from the protocol described previously (14). Briefly, after blood flow determination sample collection, 100 μCi (3.7 × 1012 Bq) of [14C]ACP (aminocyclopentane-1-[14C]carboxylic acid; American Radiolabeled Chemicals, St. Louis, MO) dissolved in 10 ml of saline was injected as a bolus over a 1-min period into the maternal femoral vein. During and after the bolus, blood samples (1 ml) were drawn simultaneously from the four vessels, A, V, γ, and α. The sequence for 16 sampling times was 0, 0.5, 1, 1.5, 2, 3, 4, 5, 10, 20, 30, 60, 90, 120, 150, and 180 min after the bolus injection. This sampling regimen ensured that the peak radioactivity in maternal plasma for ACP could be determined as shown in Fig. 1, based on previous studies, in which a similar design was utilized (14). Collected blood samples were centrifuged, and for radioactivity measurements plasma samples were solubilized in Biosafe II liquid scintillant (Research Products International, Mount Prospect, IL). 14C disintegrations per minute (dpm) were measured for 10 min in a Packard Tri-Carb 2300TR liquid scintillation analyzer.
Tissue collection and placental tissue preparation.
All ewes were euthanized after collection of the last blood sample, and the gravid uterus was removed and dissected into placental and fetal components. Placentomes were trimmed of endometrium and fetal membranes, weighed, and rinsed in ice-cold sterile saline. They were carefully dissected into maternal (caruncle) and fetal (cotyledon) components, snap frozen in liquid nitrogen, and stored at −70°C. Placental weight was recorded as the sum of the total cotyledon and caruncle weights. Approximately 6–10 g of sample were ground under liquid nitrogen and stored at −70°C pending later radioactivity content determination. Fetuses were towel dried and weighed, and crown rump length (CRL) was measured (22). Fetal liver and brain were removed and weighed. From these measurements, the fetal brain/liver ratio [brain weight (g)/liver wt (g)] and fetal ponderal index [fetal wt (g)/CRL3 (cm3)·100] were calculated (29).
For radioactivity content determinations, ∼50 mg of ground placental tissue were thawed and dissolved in 2 ml of Solvable (Packard Instrument, Meriden, CT) and incubated at 60°C overnight. Samples were allowed to cool, after which 200 μl of 30% peroxide were added and samples reheated at 60°C for 5 h. After being cooled, 20 ml of Ultima Gold (Packard Instruments) were added, and samples were left overnight. The radioactive content (14C dpm) was counted for 10 min in the Packard Tri-Carb 2300TR liquid scintillation analyzer.
ACP clearance calculations.
The transplacental clearance of ACP, or the clearance to the fetus from the mother (Cf,m, ml/min) was calculated according to the equation (1) where f is umbilical plasma flow, (γ − α) is the ACP concentration difference between umbilical venous and arterial plasma at time t (min), and At is the maternal arterial plasma ACP concentration at time t. The integral in the numerator of the function is the area under the curve (AUC) of a plot of (γ − α) vs. t for the 0- to 180-min period, with 0 being the time of the bolus ACP injection. The integral in the denominator is the AUC of a plot of A vs. t for the same time period. Areas were estimated using AUC determinations as provided by statistical software NCSS2000 (NCSS Statistical Software, East Kaysville, UT).
For the purpose of graphic representation, in each animal, the 16 sets of maternal and fetal ACP clearance measurements (Ct) were used to calculate 16 data points. The Ct values were then plotted against time. Each Ct value represents umbilical uptake of ACP at sampling time t, normalized for placental weight and the integral of maternal arterial ACP concentration: (2) where (3)
Total placental ACP clearance calculations.
At 180 min after ACP injection into the maternal circulation, an appreciable amount of ACP is present in the placenta, and at higher concentrations than in the 180-min maternal plasma samples (14). Therefore, the net uptake of ACP by the placenta from the maternal circulation has two components, i.e., the quantity crossing the basal membrane into the umbilical circulation (umbilical uptake) and the quantity retained by the placenta. The latter quantity was used to calculate a placental retention clearance (Cr): (4) where P180 is the quantity of ACP present in the whole placenta at 180 min. The total placental clearance of ACP from the maternal circulation (Ctot) was then calculated as the sum of Cr and Cf,m: (5) For the purpose of comparison between animals, the Ctot, Cr, and Cf,m clearances were expressed per 100 g placenta.
Twice weekly CBT measurements were pooled for 10-day periods, commencing upon entry into environmental chamber or control environment, to present average CBT for respective time periods. Differences between CBT, oxygen, glucose, lactate and amino acid concentrations, blood flow, ACP uptakes and clearances, and slaughter data were determined utilizing unpaired Student's two-tailed t-test. A P value of <0.05 was considered significant.
Rate of appearance of ACP and subclassification of FGR groups.
The time course of the changes in ACP concentrations in maternal and fetal circulation is presented in Fig. 1. The time point at which fetal ACP concentration was equivalent to the maternal ACP concentration was measured by calculating the time of intersection of the two curves, and was used as a crude measure of the velocity of the transplacental flux. In the control group (C), the average intersection of maternal and fetal ACP concentration curves occurred at 65.11 ± 4.86 min. On the basis of differences observed in time lapsed before equal maternal and fetal concentrations of ACP were reached, the FGR group was divided into a severe FGR group (sFGR, n = 6) and a moderate FGR group (mFGR, n = 4). The intersection of the ACP concentrations in the sFGR group occurred at 109.56 ± 12.11 min, significantly later than that observed in the C group (P < 0.01). The mFGR group reached equal concentrations faster than C with a time of 42.82 ± 5.85 min to equal concentrations (C vs. mFGR, P < 0.05; mFGR vs. sFGR, P < 0.005). One of the ewes assigned to the control group was found at necropsy to have a spontaneous FGR, and its data were kept separate from all other groups (spFGR; Table 1). This animal reached equal maternal and fetal concentrations at 122.86 min, comparable to the sFGR group (Fig. 1).
Table 1 presents the gestational ages, placental and fetal weights and fetal/placental weight ratios, fetal brain and liver weights and brain/liver ratios, CRLs, and Ponderal indexes for the three groups of animals (C, mFGR, and sFGR) and for the single spFGR animal. The sFGR and mFGR groups had mean fetal weights that were 47 and 77%, respectively, of the control fetal weight. Differences among these three groups were significant (C vs. mFGR, P < 0.01; C vs. sFGR, P < 0.001; mFGR vs. sFGR, P < 0.05; Table 1). Similarly, placental weights in both mFGR and sFGR groups were reduced, compared with C, with the sFGR placentas being significantly smaller than in mFGR (C vs. mFGR, P < 0.05; C vs. sFGR, P < 0.001; mFGR vs. sFGR, P < 0.05; Table 1). Fetal/placental weight ratios were not significantly different between groups (Table 1).
The sFGR group showed a significantly increased brain/liver ratio (P < 0.05), indicative of an asymmetrical growth restriction, whereas the brain/liver ratio of the mFGR group was not significantly different from that of C (Table 1). Furthermore, although the CRLs of both the mFGR and the sFGR groups were significantly shorter than those of C, the moderate group showed a higher Ponderal index than both C and sFGR (both P < 0.05), reflective of a greater body mass per length. The fetal and placental weights, CRL, and Ponderal and brain/liver ratio indexes of the single spFGR fetus were within the range of the sFGR group (Table 1).
The mFGR animals showed a significantly slower rise in maternal core temperatures during heat exposure than the sFGR ewes, with CBTs not significantly different compared with C until 60 dGA (Fig. 2). Furthermore, the mFGR group demonstrated a significantly lower CBT compared with the sFGR group at two later stages of pregnancy: 100 and 120 dGA (P < 0.01). The sFGR group displayed a significantly elevated CBT compared with C at all stages of treatment except at 60 dGA (Fig. 2). The ewe pregnant with the spFGR fetus demonstrated an elevated CBT at arrival at the laboratory (35 dGA; CBT 40.04°C), which remained elevated throughout the experimental period (Fig. 2), possibly an indication of an infectious disease in pregnancy.
Uterine and umbilical blood flows, ethanol clearance, uteroplacental oxygenation, and glucose and lactate concentrations.
The blood flows are presented in Table 2. Compared with C, there was a significant increase in uterine blood flow per unit placental weight in the sFGR group (C 352.6 ± 26.5 vs. sFGR 517.1 ± 58.5 ml·min−1·100 g placenta−1, P < 0.05) as well as a significant reduction in umbilical blood flow (C 198.4 ± 17.0 vs. sFGR 146.2 ± 11.6 ml·min−1·100 g placenta−1, P < 0.05). These changes led to a striking increase in the uterine flow/umbilical flow ratio (C 1.82 ± 0.14 vs. sFGR 3.62 ± 0.45, P < 0.005). In the mFGR group, there was also a significant increase of uterine flow per unit placental weight (mFGR 517.1 ± 48.5 ml·min−1·100 g placenta−1, P < 0.05). However, umbilical flow was not reduced significantly, leading to a smaller increase in the uterine flow/umbilical flow ratio (mFGR 3.0 ± 0.4, P < 0.05; Table 2).
Table 2 also summarizes the ethanol clearances for the C, mFGR, and sFGR groups and the single spFGR fetus. Although ethanol clearances (both absolute and expressed per kg fetal wt) were significantly reduced in sFGR compared with C and mFGR (absolute clearance: C vs. sFGR, P < 0.001; mFGR vs. sFGR, P < 0.005; and clearance per kg fetal wt: C vs. sFGR and mFGR vs. sFGR, P < 0.05), ethanol clearances per 100 g placental weight were not significantly different between the groups (Table 2).
Maternal hematocrit and O2 capacity, arterial O2 content, O2 saturation, and Po2 were similar between control and treatment animals. However, uterine vein O2 saturation was significantly increased in the sFGR group (P < 0.05; Table 3), whereas Po2 tended to be increased. The umbilical vein and fetal artery O2 content, O2 saturation, and Po2 were reduced significantly in sFGR compared with C and mFGR, with comparable values in the spFGR animal. In the mFGR group, these variables were not significantly different from those in C. The pH values were similar to those in C in both treatment groups, with only one fetus in the sFGR group displaying an acidosis (arterial pH 7.18). Uterine O2 uptakes were not significantly different in all groups when expressed for placental weight. However, uterine and umbilical O2 uptakes per kilogram fetus were significantly reduced in sFGR compared with C and mFGR (Table 3).
Table 4 displays the glucose and lactate concentrations and glucose uptakes. Maternal arterial glucose concentrations were significantly lower in mFGR compared with C and sFGR. However, fetal arterial glucose concentrations were significantly reduced in both mFGR and sFGR compared with C. Umbilical glucose uptakes, when expressed per kilogram fetal weight, were unaltered in mFGR and sFGR. The sFGR group displayed significantly elevated lactate levels. The spFGR fetus had a reduced arterial glucose concentration, with a glucose uptake comparable to that of the sFGR group.
ACP concentrations, uptakes, and clearances.
As discussed in materials and methods, the products of umbilical venoarterial concentration difference and umbilical plasma flow per kilogram fetus or per 100 g placenta were used to calculate the ACP umbilical uptake over the 180 min of the study. Because the fetal uptake is a function of the maternal concentration, to compare data in the different groups, the umbilical uptakes were divided by the mean of the maternal ACP concentration (Eq. 2). These “normalized” uptakes for each of the individual sampling points are presented in Fig. 3. The spFGR fetus and the sFGR group were virtually indistinguishable, and these two groups had significantly less umbilical ACP uptake per 100 g placenta than C, whereas the mFGR group demonstrated a higher normalized uptake compared with C (Fig. 3 and Table 5).
The ACP concentrations in venous plasma and placental tissues at tissue collection are summarized in Fig. 4. In every group, the ACP concentration in placental tissues, caruncle, and cotyledon was higher than in maternal plasma and higher in the cotyledon than in fetal plasma. Although the concentration ratio between cotyledon and γ trended to be increased in sFGR compared with C (P = 0.05), there was a significant decrease in the ACP concentration ratio of cotyledon and V in sFGR compared with mFGR (Table 5).
Table 5 also presents the transplacental ACP clearance (Cf,m, Eq. 1), the placental ACP retention clearance (Cr, Eq. 4), the total placental ACP clearance (Ctot, Eq. 5), and the concentrations of the five amino acids that share transport systems with ACP, i.e., valine, leucine, isoleucine, phenylalanine, and methionine. The mFGR group demonstrated a significantly increased Ctot due to an increase in Cf,m (P < 0.05; Table 5). The sFGR group, however, displayed reduced Ctot and Cf,m; these differences were significant compared with C and mFGR (Table 5). Although Cr was not different between groups, the percentage of ACP retained in placenta was increased in the sFGR group (Table 5). The single spFGR animal displayed clearances comparable to those in the sFGR group.
In association with maternal hypoglycemia, the ewes in the mFGR group displayed a maternal hypoaminoacidaemia, with significantly reduced maternal arterial concentrations, compared with C, of valine, leucine, isoleucine, and methionine (Table 5). The ewes in the sFGR group had reduced circulating levels of leucine and methionine, whereas the concentrations of the other three amino acids were not significantly different compared with those in C. The total maternal arterial concentration of the five amino acids studied was significantly reduced in mFGR but not in sFGR. Fetal arterial amino acid concentrations were not different from those in C in the sFGR group; in the mFGR group, however, fetal concentrations of valine and isoleucine were significantly reduced compared with C (Table 5). There were no differences in concentrations of leucine, phenylalanine, and methionine or in the total concentration of the five amino acids measured in fetal arterial plasma (Table 5).
The reductions in maternal glucose (Table 4) and amino acid (Table 5) concentrations can be explained by reductions in feed intake in the FGR animals. In treatment, there were no significant differences in feed intake between the treatment groups (C, 1.24 ± 0.11; mFGR, 1.21 ± 0.19; sFGR, 1.26 ± 0.12; and spFGR, 1.09 kg/day). At study, however, C ewes increased their intake (C at study, 1.50 ± 0.09 kg/day; at study vs. in treatment, P < 0.05), and FGR ewes decreased their feed intakes (mFGR, 0.58 ± 0.06; sFGR, 0.67 ± 0.09 kg/day, in treatment, vs. at study, P < 0.005 and P < 0.0001, respectively). These differences resulted in significantly lower feed intakes at study in both mFGR and sFGR ewes, even though the ewes had been housed in normal, control temperatures for ∼13 days (C vs. mFGR and C vs. sFGR, P < 0.0001).
ACP placental transport may show competitive inhibition by maternal concentrations of amino acids that are transported by similar transport systems. To test this hypothesis, the reciprocal of the total placental ACP clearance was plotted as a function of the maternal BCAA, methionine, and phenylalanine concentration. This plot is presented in Fig. 5. The control and mFGR groups appear to fit a single regression line, with the clearance decreasing as maternal BCAA, methionine, and phenylalanine concentration increased (R2 = 0.6145). This finding supports the hypothesis of competitive inhibition of ACP transport by these five amino acids. The sFGR group and spFGR fetus had an altered relationship between ACP clearance and BCAA concentration, displaying a reduced placental ACP clearance at any given maternal amino acid concentration compared with C and mFGR.
In human pregnancy, FGR affects 3–7% of fetuses, which have an increased risk of morbidity and mortality. Interest in the development of the FGR fetus has been enhanced by retrospective clinical follow-up studies (3, 4) and animal studies (9, 33) that suggest long-term consequences of FGR into adult life, including a higher incidence of diabetes and hypertension. Within FGR, there is a subgroup of fetuses with a clearly abnormal physiological state, including hepatic growth restriction with brain sparing, hypoxia, and increased pulsatility indexes of the umbilical arteries. Although the exact mechanism of these changes is unknown, this subgroup of FGR is commonly associated with increased transplacental Po2 and glucose concentration gradients (28, 31), which are indicative of a placental insufficiency.
In the present study, exposing sheep to elevated ambient temperatures has led to two distinct types of FGR: a group of severe FGR (sFGR) and a moderate FGR group (mFGR). Whereas the first group of fetuses displayed hypoxia, hypoglycemia, increased brain/liver ratios, and severe reductions in placental and fetal weights (40 and 47% of control, respectively), the latter group of fetuses displayed only moderate reductions in placental and fetal weights (74 and 77% of control, respectively), had normal brain/liver ratios, and were normoxic with normal oxygen and glucose uptakes per kilogram fetus. These observations have led us to define a subgroup of FGR (sFGR) that is most likely caused by placental insufficiency.
The differences between the two groups of FGR are highlighted by the differences in placental ACP transport. The sFGR group displayed a significant reduction in placental ACP transport to ∼58% of control, when expressed per 100 g placental weight. Given the fact that there is an additional reduction of placental weight to 40% of control, placental ACP transport, in absolute terms, in the sFGR group was reduced to ∼23% of control. In contrast, the mFGR group displayed a reduction in placental weight to 74% of control, whereas the placental clearance of ACP expressed per 100 g placenta was increased to 131% of control. This increase in ACP clearance is probably the result of decreased competition, caused by a lower maternal concentration of amino acids utilizing the same transport systems. After correcting for the lower concentration of these competing amino acids, as represented in Fig. 5, 1 g of mFGR placenta appears to function as well as a normal placenta, resulting in a normoxic fetus with normal O2 and glucose consumption. Therefore, the reduction in fetal weight in the mFGR group may be due to a reduction in placental mass but cannot be attributed to a weight-specific reduction in placental function.
A possible reason why similar exposures to elevated environmental temperatures lead to two distinct types of FGR probably lies in the onset and duration of the placental insult, secondary to differences in maternal body temperature regulation. Although the ewes in the sFGR group displayed a rapid increase in CBT that remained significantly elevated throughout most of treatment, from 40 to 120 dGA, the mFGR group showed a significantly slower rise followed by an earlier decrease in CBT, with temperatures significantly higher than control only at 70 and 90 dGA. The observation of a moderate reduction in fetal and placental weight in concurrence with a less prolonged exposure to maternal hyperthermia is in agreement with earlier studies in the hyperthermia-induced ovine FGR model (5, 11, 32). These studies suggest that severe heat-induced reductions in fetal growth are caused by an early reduction in placental growth and development. It is of interest to note that the ewe pregnant with the spontaneous FGR fetus displayed a high CBT upon entry to the facility at 35 dGA, a fever that remained throughout most of gestation and probably was caused by an infectious disease, and produced a fetus with weight and ACP transport characteristics similar to those in the sFGR group.
The reduction in placental transport of ACP to 23% of control in the sFGR group is comparable to the reductions in transplacental fluxes of leucine and threonine in previous studies of this ovine model of FGR (1, 30). The direct flux of leucine from the mother to the fetus, per 100 g placenta, was ∼60% of control, whereas placental size was reduced to ∼36% of control, leading to an absolute flux of leucine ∼22% of control (30). Similarly, the direct flux of threonine from mother to fetus per unit placental weight in FGR was ∼64% of control, with a placental size of ∼41% control, resulting in absolute threonine flux to the fetus of ∼26% of control (1). The similar percentage reductions in transplacental amino acid flux in this FGR model highlight the severity of the placental impairment in leucine, threonine, and ACP transfer, due to a reduction in placental size as well as placental transport capacity per unit placental weight. It is of further interest to note that, despite the large reduction in placental transport capacity for these amino acids, the fetal/placental weight ratios for the three groups were not significantly different. Therefore, in sFGR, the fetus is relatively large for a placenta that has a normal weight relative to fetal size, but has a severely impaired transport capacity. This impaired transport capacity is demonstrated by the significant 50% decrease in ACP clearance per kilogram fetus. Similarly, leucine and threonine FGR transplacental fluxes, expressed per kilogram fetus, were 47 and 64%, respectively, of control (1, 30).
The placental retention clearance, or amount of ACP retained in the placenta normalized for the maternal concentration, did not display any significant differences among the three groups. Therefore, we postulate that the reduction in transport capacity on one side of the placenta is concomitant with a reduction in transport capacity on the other side of the placenta, resulting in relatively stable intracellular amino acid concentrations in vivo. It is possible, for example, that a reduction in the density of transporters on either the apical or basal membrane occurs first and is then balanced with a reduction in transporter density on the opposite membrane to prevent abnormal amino acid concentrations within the trophoblast.
These changes in transport on both membranes of the trophoblast are not the consequence of changes in placental blood flows. Although the absolute uterine blood flow is reduced in sFGR, when expressed per unit placental weight, uterine blood flow is increased. Furthermore, the ratio of uterine to umbilical blood flow is increased almost twofold in the sFGR group, from 1.82 to 3.62, similar to earlier reports (28, 31). Despite this increase in uterine blood flow, total placental ACP clearance is reduced, suggesting a decrease in ACP transport across the maternal facing membrane of the placenta in severe FGR compared with control.
Alternatively, umbilical blood flow was significantly reduced to ∼74% of control when expressed per 100 g placenta in severe FGR. When the concentration ratios between the placental tissues and fetal plasma are compared, the ratio between placenta and the fetal circulation (sFGR cot/γ 2.711) is much greater than the ratio within the fetal placental circulation, before and after perfusion of the cotyledon (sFGR arteriovenous concentration ratio γ/α 1.035). Therefore, blood flow is sufficiently rapid to maintain a relatively small umbilical arteriovenous concentration difference. This is a similar situation to that observed in control pregnancies, and we postulate that the uptake of ACP is not limited by flow, but rather by the transport capacity of the placental membrane.
However, transport of a substrate across a membrane is affected not only by membrane properties or flow but by concentrations of other amino acids that utilize the same transport systems as the substrate under study. Compared with control, the placental clearance of ACP from the maternal circulation is significantly increased in the mFGR placenta, which could be explained by the lower concentration of amino acids competing for the same transport systems in this group. In the umbilical circulation, however, the total concentrations of these amino acids are not significantly different between groups. Therefore, it is not likely that competitive inhibition plays a role in the reduced transplacental clearance of ACP across the basal membrane of the trophoblast in severe FGR.
ACP, also known as ACPC or cycloleucine, is a nonmetabolizable amino acid analog of the BCAA and other amino acids that have large apolar side chains, such as methionine and phenylalanine (8, 21, 25). ACP, in contrast to other nonmetabolizable amino acids, is not specific for a single transporter (20, 21). Although this may seem a disadvantage, it is important to note that the transplacental flux of an amino acid may require affinity for at least two transport systems, one on the maternal and one on the fetal surface of the placenta, transporting substrates from the placenta into fetal circulation. A highly specific amino acid might enter the placenta but be prevented from exiting, as has been demonstrated in the sheep placenta for the nonmetabolizable amino acid analog α-methylaminoisobutyric acid.
Although the majority of ACP and BCAA transport occurs through the high-affinity, sodium-independent system L (34), some evidence exists that ACP is transported by other systems as well, such as system A (21) and possibly system ASC (17). For example, high concentrations of ACP are able to inhibit cellular accumulation of many naturally occurring amino acids, such as alanine, serine, proline, and lysine, in cultured human fibroblasts (24). Furthermore, within the BCAA, phenylalanine, and methionine group of amino acids, the umbilical uptake of one amino acid can be inhibited by raising the concentration of other amino acids (15). In the present study, the placenta transports ACP more rapidly when concentrations of BCAA, phenylalanine, and methionine are low. This indicates that the transport of ACP is competitive with these amino acids in vivo; i.e., it utilizes their transport systems.
Several studies utilizing membrane vesicles from human FGR placenta have demonstrated a reduction in amino acid transport in vitro (12, 13, 18). Reductions in the uptake of leucine in both microvillous and basal membrane preparations have been reported, highlighting possible changes in the L transport system (13). Furthermore, system A transport activity is reduced in FGR vesicles of the maternal facing membrane of the human trophoblast (18), with the activity of system A inversely correlated to the severity of FGR (12). Additionally, System y+L-mediated basal membrane uptake of lysine is reduced (13), although microvillous membrane system y+ and y+L activity is unaltered in FGR vesicles (2, 13).
This in vivo study highlights that placentas associated with some FGR pregnancies are not only reduced in size but also functionally impaired. Human in vivo studies confirm this impaired placental function, demonstrating that the placental transport and fetoplacental metabolism of essential amino acids, such as leucine and phenylalanine, is most likely reduced in FGR (27). Further studies in the pregnant sheep highlight alterations in placental transport as well as changes in fetoplacental metabolism of leucine and threonine (1, 30). The present study provides further evidence of a placental functional impairment in FGR, in terms of amino acid transport, and highlights the fact that, for an amino acid with a preference for system L, transplacental clearance is significantly reduced in situations of severe FGR but not in situations of moderate FGR. These data suggest that, in severe FGR, both the uptake from maternal circulation and release into fetal circulation may be impaired for other amino acids utilizing the same transport systems.
This work was supported through National Institute of Child Health and Human Development Grants RO1-HD-20761 and HD-41505; and the Ter Meulen Fund, Royal Dutch Academy of Arts and Sciences, supported B. de Vrijer.
We are grateful to Karen Trembler, Willie Jones, and Larry Toft, to David Caprio and laboratory staff, and to I-Da and Yu-Ching Fan for technical support. We also thank Drs. Eric A. P. Steegers and Didi D. M. Braat for support of B. de Vrijer's fellowship.
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 © 2004 by American Physiological Society