点击显示 收起
1 From the Rowett Research Institute, Aberdeen, Scotland, United Kingdom.
2 Supported by the Scottish Executive Rural Affairs Department.
3 Address reprint requests to AK Dutta-Roy, Rowett Research Institute, Aberdeen AB21 9SB, Scotland, United Kingdom. E-mail: adr{at}rri.sari.ac.uk.
ABSTRACT
To understand the placental role in the processes responsible for the preferential accumulation of maternal long-chain polyunsaturated fatty acids (LCPUFAs) in the fetus, we investigated fatty acid uptake and metabolism in the human placenta. A preference for LCPUFAs over nonessential fatty acids has been observed in isolated human placental membranes as well as in BeWo cells, a human placental choriocarcinoma cell line. A placental plasma membrane fatty acid binding protein (p-FABPpm) with a molecular mass of 40 kDa was identified. The purified p-FABPpm preferentially bound with essential fatty acids (EFAs) and LCPUFAs over nonessential fatty acids. Oleic acid was taken up least and docosahexaenoic acid (DHA) most by BeWo cells, whereas no such discrimination was observed in HepG2 liver cells. Studies on the distribution of radiolabeled fatty acids in the cellular lipids of BeWo cells showed that DHA is incorporated mainly into the triacylglycerol fraction, followed by the phospholipid fraction; the reverse is true for arachidonic acid (AA). The greater cellular uptake of DHA and its preferential incorporation into the triacylglycerol fraction suggests that both uptake and transport modes of DHA by the placenta to the fetus are different from those of AA. p-FABPpm antiserum preferentially decreased the uptake of LCPUFAs and EFAs by BeWo cells compared with preimmune serum. Together, these results show the preferential uptake of LCPUFAs by the placenta that is most probably mediated via the p-FABPpm.
Key Words: Human placenta • placental membrane fatty acid binding protein • p-FABPpm • long-chain polyunsaturated fatty acids • fetal development • fatty acid uptake • BeWo cells
INTRODUCTION
Dietary essential fatty acids (EFAs) are essential for cellular growth and metabolism, and membrane structure and function (1–3). EFAs and their long-chain polyunsaturated fatty acid (LCPUFA) derivatives are of critical importance in fetal and infant development (4, 5). Because of the fundamental roles of EFAs and LCPUFAs as structural elements and functional modulators (1, 2), it has been hypothesized that maternal, fetal, and neonatal LCPUFA status are important determinants of health and disease in infancy and later in life. The deposition of these LCPUFAs in the fetus is rapid during growth, and it has been suggested that inadequacy of LCPUFAs in critical membrane lipids may lead to a failure to accomplish a specific component of brain growth and irrevocable damage (4). Both the brain and the retina are rich in the LCPUFAs arachidonic acid (AA; 20:4n-6) and docosahexaenoic acid (DHA; 22:6n-3), and a sufficient supply of these fatty acids during pregnancy and the neonatal period is critical to proper function (4, 5). The fetal brain acquires 21 g DHA/wk during the last trimester of pregnancy (6). Evidence from various studies suggests that retinal function and learning ability may be permanently impaired if there is a reduction in the accumulation of sufficient DHA during intrauterine life (4, 5).
Overall maternal LCPUFA status declines steadily during pregnancy (7). Maternal DHA status was found to be significantly lower in multiparas than in primiparas, suggesting that in usual dietary conditions, pregnancy is associated with mobilization of DHA from a store that may not be readily replenished after delivery (7). Preterm infants have significantly lower EFA and LCPUFA statuses than do full-term neonates (8, 9). In fact, the potential for dietary EFA deficiency has become a significant issue for the nutrition of preterm infants, because they do not receive the third-trimester intrauterine supply of DHA and AA. The fatty acid pattern found in the walls of umbilical blood vessels indicates that the EFA status of a developing fetus is marginal or barely sufficient (10, 11). Supplementation of pregnant mothers with LCPUFAs has been shown to improve neonatal LCPUFA status (12). However, a better understanding of the biochemical processes involved in fetoplacental LCPUFA metabolism and transport is required before the optimum and safe levels of supplementation during pregnancy can be determined. This is particularly important for premature infants because they have low blood DHA concentrations, which affect their eye and brain functions as measured by electroretinogram, cortical visual evoked potential, and behavioral testing of visual acuity (9, 12, 13).
Numerous studies have shown that LCPUFA concentrations are higher in fetal than in maternal circulation (4, 10, 11, 14), but the underlying biochemical mechanisms controlling this phenomenon are not understood. To examine the role of the placenta in the preferential accumulation of LCPUFAs in fetuses, we studied the interaction of EFAs and LCPUFAs with isolated human placental membranes and with human placental choriocarcinoma (BeWo) cells. The aim of this article is to bring together recent developments on the placental uptake and metabolism of LCPUFAs in humans and to advance concepts concerning the role of the newly discovered placental plasma membrane fatty acid binding proteins (p-FABPpm) in these processes (15, 16).
MATERNAL METABOLISM AND TRANSPORT OF EFAS AND THEIR LCPUFA DERIVATIVES
Linoleic acid (LA; 18:2n-6) and -linolenic acid (ALA; 18:3n-3) are the 2 main dietary EFAs. LA and ALA are not interchangeable, but can be further elongated and desaturated by the same enzyme systems to produce n-6 and n-3 LCPUFAs in the body (1, 17). Whereas dietary LA and ALA are primarily from vegetable oils, preformed LCPUFAs may also be consumed in foods of animal origin. The importance of LCPUFAs has been related to their structural action, their specific interaction with membrane proteins, and their ability to serve as precursors of second messengers (1, 18, 19). Therefore, LA and ALA must be converted to their metabolites, LCPUFAs, to exert their full range of biological actions (1). Maternal EFA metabolism is crucial for fetal growth and development because the fetus depends on maternal EFAs and LCPUFAs (4, 15). However, it is important to note that the accretion of LCPUFAs [eg, AA, eicosapentaenoic acid (EPA; 20:5n-3), DHA, and dihomo--linolenic acid (20:3n-6)] cannot be quantitatively explained by desaturation and elongation of dietary EFAs alone (1). Relations between dietary EFAs and fatty acid profiles of tissues are complex and appear to be dependent on many factors and underlying processes (1). Except for DHA, LCPUFAs are substrates for cyclooxygenase (prostaglandin-endoperoxide synthase) and lipoxygenases and produce a variety of compounds collectively called eicosanoids (1). These compounds have diverse biological functions in cell growth and development, inflammation, and the cardiovascular system (1, 3, 18, 19). DHA is involved mainly in retina and brain function, but not through the eicosanoid pathways (4, 20). The biological response elicited after eicosanoid release is dependent on the net balance of eicosanoids derived from n-6 and n-3 LCPUFAs. In the Western diet, AA is the most predominant precursor fatty acid of the highly biologically active eicosanoids of the 2 series (ie, eicosanoids with double bonds). AA thus provides adequate substrates for synthesis of circulating vasoactive factors such as prostacyclin and thromboxane A2 (1, 21). The ratio of thromboxane A2 to prostacyclin is an index of the relative activity of the opposing stimuli that modulate vascular tone and platelet activation (1, 21). The ratio of thromboxane A2 and prostacyclin in intervillous space may be involved in the mechanism of initiating myometrial contraction during labor. In addition, the endogenous formation of cyclooxygenase and noncyclooxygenase metabolites of AA has been implicated in the expression of the early response genes c-fos, Egr-1, c-myc, and c-jun in different cell types after mitogenic stimulation; EPA, however, inhibits the expression of these genes (3). Therefore, supplemental EPA not only decreases AA availability but also competes with AA for cyclooxygenase and lipoxygenase enzymes and thereby reduces the formation of the growth-stimulatory eicosanoids (1, 3).
Because both n-3 and n-6 fatty acids are present in vertebrate tissues owing only to dietary intake, important physiologic consequences follow the inadvertent selection of different average daily dietary supplies of these 2 types of EFAs. Because each type of EFA can interfere with the metabolism of the other, an excess of n-6 fatty acids will reduce the metabolism of ALA, possibly leading to a deficit of n-3 LCPUFA metabolites. This is a matter of concern in infant formulas, which may contain an excess of LA without balancing n-3 fatty acids. Therefore, a proper balance between the n-6 and n-3 fatty acids in the diet is important to maintain optimum growth and development, and there is an emerging consensus that infant formulas should be supplemented with DHA rather than EPA and have adequate amounts of AA. However, it is not known how the pregnant mother is able to provide an adequate supply of LCPUFAs, especially during the last trimester of pregnancy when the fetus requires large amounts of both AA (n-6) and DHA (n-3), but not much EPA (n-3).
Several studies (10, 11, 15, 22) of the fatty acid composition of fetal and maternal plasma showed that at birth, LA represents 10% of the total fatty acids in cord plasma compared with 30% in maternal plasma but, surprisingly, the AA concentration of cord plasma is twice (10%) that observed in the mother (5%). Similarly, the ALA concentration in newborns is half of that in the mother (0.3% and 0.6%, respectively), whereas the DHA concentration is double (3% and 1.5%, respectively). EPA concentrations were low, with no significant difference between fetuses and mothers. This situation, in which the relative plasma concentrations of the n-3 and n-6 LCPUFAs exceeds those of their precursors, is specific to newborns and has never been observed in adults. It is obviously an extremely favorable situation for the development of the newborn, especially at a time when large quantities of AA and, particularly, DHA are needed by the brain and retina. However, 3 important questions arise: 1) What are the underlying mechanisms leading to this situation at birth? 2) How does the situation evolve throughout gestation? and 3) How are the n-3 and n-6 LCPUFAs (DHA and AA) preferentially transferred?
Although the delivery of fatty acids from the maternal circulation to the developing fetus has been studied extensively (4, 23) and various suggestions have been put forward, none of the above questions have been answered satisfactorily. Because human placental tissue lacks both the 6- and 5-desaturase activities (4, 23–25), any LCPUFAs in the fetal circulation must be derived primarily from the maternal plasma. However, the situation has become more complicated by the discovery of desaturase and elongase activity in developing fetal brain and liver (24), although it is not certain to what extent desaturation and elongation of transferred linoleic and -linolenic acids may contribute to fetal LCPUFAs.
In the fetomaternal unit, free fatty acids (FFAs) originating in the maternal circulation are the major source of fatty acids for transport across the placenta (4, 23–25) because triacylglycerols are not transported intact (4, 26, 27). Maternal lipoprotein lipase must be active to facilitate placental uptake of FFAs from circulating triacylglycerols but not from chylomicrons (4, 15). Recently, we showed that purified placental microvillous membranes exhibit 2 distinct triacylglycerol hydrolase activities: a minor activity at pH 8.0 and a major activity at pH 6.0 (28). The triacylglycerol hydrolase activity at pH 8.0 appears to be lipoprotein lipase (identified by important criteria such as serum stimulation and salt inhibition), whereas the activity at pH 6.0 is unique because it is almost abolished by serum but is not affected by high sodium chloride concentrations. Therefore, it is possible that the inhibitory effect of serum on triacylglycerol hydrolase activity at pH 6.0 in placental microvillous membranes may not allow this enzyme to act on maternal lipoproteins but only on intracellular triacylglycerol stores in the placenta. If true, then this membrane-bound triacylglycerol hydrolase (optimum pH 6.0) may play an important role in releasing FFAs by hydrolyzing the placental triacylglycerol stores. This unique triacylglycerol hydrolase activity at pH 6.0 therefore may be involved in the packaging, release, or both of FFAs from the placenta. However, lipoprotein lipase, which is present mostly in placental macrophages, may be responsible for hydrolysis of maternal plasma lipoproteins, as suggested by Bonet et al (29). Lipoprotein receptors, mainly responsible for the uptake of cholesterol required for fetal tissue synthesis, have been detected on both placental macrophages and syncytiotrophoblasts (15). They may also be involved in fatty acid uptake by the placenta. However, the metabolic fate of fatty acids transported via lipoprotein receptors is uncertain and may not account for the preferential accumulation of LCPUFAs in fetal tissues.
PREFERENTIAL UPTAKE OF LCPUFAS BY PLACENTAL MEMBRANES
The preferential accumulation of LCPUFAs in fetal tissues was initially thought to be the result of the combined effects of placental fatty acid uptake from the maternal plasma, selective fatty acid oxidation, lipid synthesis in the placenta, and LCPUFA synthesis by the fetal liver (4, 15). Evidence on the role of the placenta in preferential accumulation of LCPUFAs in fetal tissues has emerged from research in our laboratory in which the fatty acid binding characteristics of human placental membranes were determined by using 4 different radiolabeled fatty acids [LA, ALA, AA, and the nonessential fatty acid oleic acid (18:1n-9)] (30).
The binding of both EFAs and LCPUFAs to human placental membranes was highly reversible and specific compared with that of oleic acid (30). In addition, oleic acid binding was inhibited strongly by the LCPUFAs [AA, -linolenic acid (18:3n-6), and EPA], followed by the relevant parent EFAs (LA and ALA). The lack of strong inhibition of the binding of EFAs and LCPUFAs to placental membranes by oleic acid suggested the existence of stronger affinities for EFAs and LCPUFAs compared with oleic acid (30). EPA and its eicosanoid metabolites are considered to be growth inhibitors because they reduce the availability of AA and its metabolites by competing for cyclooxygenase, lipoxygenase, and elongation and desaturation pathways of EFA metabolism (1, 3). EPA has little inhibitory effect on AA transport by placental membranes. In contrast with EPA, its parent fatty acid, ALA, strongly inhibited the binding of both AA and LA. (Table 1). Non–fatty acid ligands such as bromsulfophthalein and -tocopherol did not inhibit the binding of any of the radiolabeled fatty acids, indicating that binding sites are specific to fatty acids. The competition experiments also suggested that the binding sites have heterogeneous binding affinities for different fatty acids. Binding sites seem to have a strong preference for LCPUFAs: the order of competition was AA >> LA > ALA >> oleic acid. Our finding on the preferential binding of LCPUFAs by placental membranes was supported by Haggarty et al (31), who also showed that the human placenta can selectively transfer LCPUFAs and EFAs to the fetal circulation in preference to the nonessential fatty acids.
View this table:
TABLE 1.. Fatty acid binding specificity of human placental membranes1
Our data also suggested that trans fatty acids compete with LCPUFAs for fatty acid binding sites in human placental membranes, thereby inhibiting the transport of EFA and LCPUFAs to the placenta (30). trans Fatty acids are not synthesized in situ but are present in both fetal and placental tissues, which suggests that they are transported through the placenta from the maternal circulation (32, 33). Our studies support these observations that maternal trans fatty acids can be transported to the fetal circulation via the placenta. Once trans fatty acids are in the fetoplacental unit they may exert detrimental effects on growth and development by disturbing LCPUFA metabolism (eicosanoid production) and membrane structure and function (32, 33).
IDENTIFICATION AND CHARACTERIZATION OF A PLACENTAL MEMBRANE FATTY ACID BINDING PROTEIN
The tissue uptake of FFAs has been studied extensively and a variety of mechanisms of FFA uptake have been proposed, including passive diffusion, specific binding to plasma membrane associated fatty acid binding proteins such as ubiquitous FABPpm, fatty acid translocase, and fatty acid transport protein (FATP) (34–38). There are now several reports providing evidence of the involvement of these membrane proteins in the uptake of FFAs into various mammalian tissues (34). Therefore, the presence of a membrane FABP in the placenta was sought. We identified and purified p-FABPpm from human and sheep placentas by using a technique similar to that used for ubiquitous FABPpm (16, 39). The apparent molecular mass of the p-FABPpm was estimated to be 40 kDa, as determined by gel permeation chromatography and by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (16). The isoelectric point (pI) value and amino acid composition of p-FABPpm (40 kDa) are different from those of ubiquitous FABPpm (40 kDa), fatty acid translocase (88 kDa), and FATP (62 kDa) (34–39). In addition, unlike ubiquitous FABPpm (40), p-FABPpm does not have glutamic oxaloacetic transaminase activity. Therefore, despite their similar size and membrane location (both are peripheral-membrane-bound proteins), p-FABPpm and ubiquitous FABPpm differ in both structure and function. The autoradioblotting technique, which was initially developed to detect and estimate cellular retinol-binding proteins, was used to investigate the [14C]fatty acid binding activity of the purified membrane protein and also to detect p-FABPpm in solubilized placental membranes (16). Evidence for the involvement of p-FABPpm in fatty acid uptake came from the trypsin-treated placental membranes, in which specific [14C]fatty acid binding decreased compared with untreated membranes (30). Furthermore, preincubation of human placental membranes with polyclonal antiserum against the human p-FABPpm inhibited the binding of fatty acids but the degree of inhibition varied depending on the type of fatty acid (30). Antibody-mediated inhibition was much more effective for the LCPUFA and EFA binding than the oleic acid binding. This suggests again that fatty acid binding is heterogeneous, and that p-FABPpm may be preferentially involved in the uptake of EFAs and their LCPUFA derivatives.
ROLE OF p-FABPpm IN PREFERENTIAL UPTAKE OF LCPUFAS BY THE PLACENTA
To elucidate further the role of p-FABPpm in the preferential transfer of maternal plasma LCPUFAs across the human placenta, we investigated the direct binding of the purified p-FABPpm with various fatty acids (41). Binding of radiolabeled fatty acids to the purified p-FABPpm revealed that the placental protein had higher affinities (Kd) and binding capacities (Bmax) for LCPUFAs than for other fatty acids. The apparent Bmax values for oleic acid, LA, AA and DHA were 2.0, 2.1, 3.5, and 4.0 mol/mol FABPpm, whereas the apparent Kd values were 1.0, 0.73, 0.45, and 0.40 µmol/L, respectively (41). There was no significant difference in Kd and Bmax values between oleic acid and LA or between AA and DHA. However, there were significant differences in these values between the LCPUFAs (AA and DHA) and oleic acid and LA. Human serum albumin did not show such variations in binding capacities and affinities for these fatty acids; the Kd and Bmax values for these fatty acids were 1 µmol/L and 5 mol/mol FABPpm, respectively (41).
During the last trimester of pregnancy, plasma FFA concentrations increase rapidly to support the fetal demand for LCPUFAs (4, 23–25). Therefore, we also examined the binding of a fatty acid mixture taken from women during the last trimester of pregnancy that mimicked that of plasma with the purified p-FABPpm. The total FFA concentration in the plasma pool ( ± SD, in mmol/L) was 0.71 ± 0.015, of which 0.22 ± 0.07 was palmitic acid, 0.14 ± 0.08 was stearic acid, 0.19 ± 0.06 was oleic acid, 0.12 ± 0.021 was LA, 0.011 ± 0.004 was ALA, 0.016 ± 0.005 was AA, and 0.015 ± 0.003 was DHA (42). Thus, 76.7% of total fatty acids were nonessential (palmitic, stearic, and oleic acids) and 23% were essential and, of these, LCPUFAs amounted to 4.5%. AA and DHA were present in similar concentrations in the plasma FFA pool of pregnant women despite a 10-fold difference in concentrations of their n-6 and n-3 precursors, LA and ALA, respectively. This suggests a mechanism that increases the availability of unesterified DHA in the plasma of pregnant women and thus increase the fetal supply of DHA.
p-FABPpm bound only 14.58% of the total fatty acids when the purified protein was incubated with a fatty acid mixture as described above (Table 2). Despite the high concentrations of the nonessential fatty acids in the assay mixture, little stearic or oleic acid bound to the p-FABPpm (0.57 ± 0.07 and 4.05 ± 0.40 nmol, respectively; n = 3) whereas binding of palmitic acid was not detectable. For the 2 EFAs, binding of ALA was not detected, whereas almost 23% of the LA present in the mixture bound to the protein. In contrast, 98% of the AA and 83% of the DHA present in the assay mixture was bound to the protein (1.62 ± 0.04 and 1.24 ± 0.03 nmol, respectively; n = 3). Human serum albumin, with a similar ratio of fatty acid to protein, bound 2.5-fold more fatty acids than did the p-FABPpm with no preference for any particular type of fatty acid. The percentage of each fatty acid bound to the human serum albumin was greatest for stearic acid (71%) followed by oleic acid (34%), palmitic acid (33%), and least for LA (18%) and AA (23%); binding of DHA was not detectable. These data clearly show that p-FABPpm preferably binds with LCPUFAs, whereas human serum albumin does not show any preference for particular fatty acids present in the plasma FFA pool of pregnant women during the last trimester of pregnancy. It has been generally accepted that serum albumin serves as a reservoir of fatty acids for replenishment of those taken up by cells. The binding experiments provide direct evidence for the role of p-FABPpm in preferential sequestration of maternal LCPUFAs by the placenta for transport to the fetus. With the exclusive location of the p-FABPpm on the on the maternal-facing membranes of the placenta (43) and its preference for LCPUFAs in the plasma FFA pool, a unidirectional flow of maternal LCPUFAs to the fetus is favored.
View this table:
TABLE 2.. Binding of mixture of fatty acids as present in the plasma free fatty acid pool of pregnant women to placental plasma membrane fatty acid binding protein (p-FABPpm) and in human serum albumin1
UPTAKE AND METABOLISM OF FATTY ACIDS IN A HUMAN PLACENTAL CHORIOCARCINOMA CELL LINE
All the above-mentioned studies were performed using isolated human placental membranes and therefore no information was available on intact trophoblast cells with metabolic activity. We therefore examined the uptake and metabolism of oleic acid, linoleic acid, AA, and DHA in the BeWo human placental choriocarcinoma cell line (44). We also compared the uptake and metabolism of oleic acid and LA by BeWo cells with the carrier-mediated uptake by HepG2 liver cells (45). These 2 cell types are derived from organs with specialized roles in lipid metabolism. For example, the liver is involved in a multiplicity of lipid metabolic roles including synthesis, esterification, lipoprotein export, and oxidation for energy, whereas, as already noted, the placenta is primarily concerned with sequestering the maternal LCPUFAs and transporting them to the fetus.
The kinetic data and temperature-dependent nature of fatty acid uptake by BeWo cells indicated a saturable process. Oleic acid was taken up least and DHA most by these cells. Moreover, competitive studies indicated a preferential uptake compared with oleic acid in the order of DHA, AA, and LA by BeWo cells (44), supporting previous studies on the placental transport of DHA (46). In contrast, HepG2 cells did not exhibit a differential affinity for the fatty acids studied (Table 3).
View this table:
TABLE 3.. Effect of fatty acid addition on uptake of oleic and linoleic acids by human placental choriocarcinoma (BeWo) and HepG2 liver cells1
With the use of a polyclonal antibody to p-FABPpm, Western blot analyses showed the presence of a p-FABPpm in BeWo cells, but not in HepG2 cells. Ubiquitous FABPpm in HepG2 cells did not react with p-FABPpm, further showing the difference between ubiquitous and p-FABPpm (44). Cells pretreated with p-FABPpm antiserum inhibited uptake of oleic acid, LA, AA, and DHA to different degrees than did untreated cells (45). With the same amount of antiserum, oleic acid uptake was inhibited least (32%), whereas uptake of AA (68%) and DHA (64%) was inhibited most, followed by LA (50%) (45). These results clearly show that p-FABPpm may be involved in the preferential uptake of EFAs and their LCPUFA derivatives by these cells.
We also compared the metabolism of oleic acid, LA, AA, and DHA in BeWo cells and observed marked differences (47). BeWo cells were found to rapidly incorporate exogenous AA and DHA into the total cellular lipid pool. The extent of esterification was more rapid for DHA than for AA, although this difference abated with time, leaving only a small percentage of the fatty acids in their unesterified form. In the cellular lipids, these fatty acids were esterified mainly into the phospholipid and the triacyglycerol fractions; smaller amounts were also detected in the diacylglycerol and cholesteryl ester fractions. Almost 60% of the total DHA taken up by cells was esterified into triacylglycerol whereas 37% was in phospholipid fractions. The reverse was true for AA; 60% of the total uptake was incorporated into phospholipid fractions whereas <35% was in triacyglycerol fractions. Marked differences were also found in the distribution of the fatty acids into individual phospholipid classes. The higher incorporations of DHA and AA were found in phosphatidylcholine and phosphatidylethanolamine, respectively. The greater cellular uptake of DHA and its preferential incorporation into triacyglycerols suggests that the uptake and transport modes of DHA by the placenta to the fetus are different from those of AA. The preferential incorporation of DHA into triacylglycerol suggests that triacylglycerol may play an important role in the placental transport of DHA to the fetal circulation. Together, these results show the preferential uptake of LCPUFAs by BeWo cells, which is most probably mediated via p-FABPpm. The uptake and subsequent esterification of DHA was the greatest and the esterification of DHA into triacylglycerol was more efficient than that of the other 3 fatty acids. One of the main functions of the placenta is to deliver DHA into the fetal circulation; therefore, the triacylglycerol form may favor the transport of DHA to the fetal circulation. However, at present it is not known how and in what form DHA is released from the placenta into the fetal circulation.
A multistep process of cellular uptake and intracellular translocation of FFAs facilitated by various membrane-associated and cytoplasmic FABPs has been proposed (36). We therefore examined the presence of FABPs in primary trophoblasts, BeWo cells, and placental membranes (both microvillous and basal). Western blot analysis showed the presence of multiple membrane-associated and cytoplasmic FABPs in human placental cells (48). In addition to p-FABPpm, fatty acid translocase (88 kDa) and FATP (62 kDa) were detected in both microvillous and basal membranes of the human placenta (48). Among the cytoplasmic FABPs, heart- (H) and liver- (L) type FABPs were detected in the cytosol of human placental primary trophoblasts as well as in BeWo cells (48). Immunoreactivity of epidermal- (E) type FABP was not detected in trophoblasts or BeWo cells despite its presence in human placental cytosol. The location of fatty acid translocase and FATP on both sides of bipolar placental cells may favor transport of FFA pools in both directions, ie, from the mother to the fetus and vice versa. However, p-FABPpm, because of its exclusive location on the microvillous membranes, may favor the unidirectional flow of maternal plasma LCPUFAs present in the plasma FFA pool to the fetus because of the binding specificity for LCPUFAs. Although the roles of these proteins in placental fatty acid uptake and metabolism are yet to be fully understood, their complex interaction may be involved in the uptake of maternal FFAs by the placenta for delivery to the fetus.
CONCLUSIONS
Among the various biofactors released at the fetomaternal interface, LCPUFAs play major roles in fetoplacental development and pregnancy outcome (36, 49). LCPUFAs are important for optimal growth because they are the precursors of eicosanoids and also maintain cell membrane structure and function (1). The preferential accumulation of maternal LCPUFAs in fetal tissues has been thought to be the result of combined processes occurring in both the mother and the fetoplacental unit. Our data indicate that p-FABPpm, located exclusively on the maternal-facing membranes of the placenta, may be involved in the sequestration of maternal LCPUFAs and therefore may play a critical role in fetal growth and development (36, 49). However, further studies are required on the fatty acid–binding domain of the p-FABPpm, which allows the preferential binding of LCPUFAs over nonessential fatty acids. The preferential incorporation of DHA into the triacylglcyerol fraction suggests that triacylglcyerols may play an important role in the placental transport of DHA to the fetal circulation. The major membrane lipase activity at pH 6.0 may play an important role in packaging placental-stored DHA for fetal transport.
Although the exact roles of the membrane-associated FABPs (fatty acid translocase, FATP, and FABPpm) in complex FFA uptake and metabolism in the human placenta are not yet understood, studies in other tissues suggest that these proteins alone or in tandem may be involved in effective uptake of FFAs (50). Location of fatty acid translocase and FATP on both sides of the bipolar placental cells may allow bidirectional flow of all fatty acids (nonessential, essential, and long-chain polyunsaturated) across the placenta, whereas the exclusive location of p-FABPpm on the maternal side may favor the unidirectional flow of maternal LCPUFAs to the fetus by virtue of preference for these fatty acids (48).
Cytoplasmic heart- and liver-type FABPs in the placenta have been observed (48). There are significant differences in the properties and regulation of these 2 FABP types. Heart FABP binds only long-chain fatty acids whereas liver FABP binds heterogeneous ligands such as bile salts, heme, peroxisome proliferators, selenium, lysophosphatidic acid, and eicosanoids (36). By virtue of its binding of eicosanoids, liver FABP may play a critical role in eicosanoid synthesis in the fetoplacental unit. Liver FABP increases fatty acid uptake and targets fatty acids for esterification into phospholipids. It has been suggested that FABPs may also interact with several fatty acid–mediated cellular processes, such as control of cell growth, cell signaling, and regulation of gene expression. Although there are differences between the function and binding activity of liver and heart FABP types, their individual and coordinated roles and expression in trophoblasts are yet to be established. A complex interaction between these proteins may be required for the effective uptake of maternal FFAs by the placenta and their subsequent transport to the fetus. The putative roles of the membrane-associated and cytoplasmic fatty acid binding proteins in placental fatty acid uptake and metabolism are summarized in Figure 1.
FIGURE 1. . Putative roles of the placental-membrane-associated and cytoplasmic fatty acid binding proteins (FABPs) in placental fatty acid uptake and metabolism. The location of fatty acid translocase and FAPBs on both sides of the bipolar placental cells and the lack of specificity for particular types of free fatty acids (FFAs) allow transport by FAPBs of all FFAs (nonessential, essential, and long-chain polyunsaturated) bidirectionally, ie, from the mother to the fetus and vice versa. However, by virtue of its exclusive location on the maternal-facing placental membranes and preference for long-chain polyunsaturated fatty acids (LCPUFAs), placental plasma membrane FAPB (p-FABPpm) sequesters maternal plasma LCPUFAs to the placenta. Cytoplasmic FABPs may be responsible for transcytoplasmic movement of FFAs to their sites of esterification, ß-oxidation, or to the fetal circulation via placental basal membranes. FAT, fatty acid translocase; FATP, fatty acid transport protein; L-, liver-type; H-, heart-type; PG, prostaglandins; LT, leukotrienes; TG, triacylglycerols, PL, phospholipids; CE, cholesteryl esters; HEPTE, hydroperoxyeicosatraenoic acid. Reprinted from Placenta, Vol 19, (5/6), Campbell FM, Bush PG, Veerkamp JH, Dutta-Roy AK, Detection and cellular localization of plasma membrane–associated and cytoplasmic fatty acid binding proteins in human placenta, pp409–15, 1998, by permission of the W B Saunders Company Limited (48).
Recent studies have suggested the beneficial effects of maternal dietary supplementation with EFAs in both normal and complicated pregnancies (12, 51). However, to fully exploit the benefit of dietary supplemention of EFAs and LCPUFAs to pregnant women it is important to understand the metabolism and transport of these fatty acids in the mother and the fetoplacental unit. The discovery of a new p-FABPpm in the human placenta provides a unique opportunity to study the role of the placenta in the supply of EFAs and LCPUFAs in both physiologic and clinical conditions, eg, intrauterine growth retardation, low weight-for-gestational age, gestational diabetes, and preterm birth.
REFERENCES