- Split View
-
Views
-
Cite
Cite
J.L. James, P.R. Stone, L.W. Chamley, The regulation of trophoblast differentiation by oxygen in the first trimester of pregnancy, Human Reproduction Update, Volume 12, Issue 2, March/April 2006, Pages 137–144, https://doi.org/10.1093/humupd/dmi043
- Share Icon Share
Abstract
In the first trimester of human pregnancy villous cytotrophoblasts are able to differentiate to form either the overlying syncytiotrophoblast layer or, in anchoring villi, extravillous trophoblasts which grow out from the villi and invade into the maternal decidua, acting to both physically attach the placenta to the decidua, and modify the maternal spiral arteries to sustain pregnancy. During the first 10–12 weeks of gestation, extravillous trophoblast plugs block the spiral arteries and prevent maternal blood flow entering the intervillous space, thereby creating an environment of physiological hypoxia in which placental and fetal development occur. As extravillous trophoblasts migrate away from the villus they differentiate from a proliferative to an invasive phenotype. The hypoxic environment of the first trimester is believed to play an important role in the regulation of trophoblast differentiation. However, there is currently a large body of conflicting experimental evidence concerning this topic. This review examines the experimental evidence to date on the role of oxygen in trophoblast differentiation.
Submitted on February 2, 2005; resubmitted on July 21, 2005; accepted on September 9, 2005
Introduction
Placentation is initiated when the blastocyst makes contact with the epithelial lining of the uterus shortly after implantation. Placental villi develop which consist of a mesenchymal core surrounded by a monolayer of mononuclear villous cytotrophoblast stem cells which either fuse to form the overlying multinucleated syncytiotrophoblast or, in anchoring villi, differentiate into extravillous trophoblasts which grow out from the villous and spread laterally around the placenta (Irving et al., 1995). As extravillous trophoblasts move away from the placenta they differentiate into an invasive phenotype in a process which is tightly regulated both spatially and temporally and is essential for the success of pregnancy. Phenotypically there are key differences between invasive extravillous trophoblasts and villous cytotrophoblasts. Such differences can be seen in the expression of cell surface molecules such as adhesion molecules, as well as in the secretion of cytokines, growth factors and proteases (Norwitz et al., 2001).
Invasive extravillous trophoblasts play an important role in adapting the decidua to sustain pregnancy. Extravillous trophoblasts invade the walls of the uterine spiral arteries and adapt these vessels into large bore conduits capable of delivering the increased blood supply required in the second and third trimesters and ensure that this blood supply is independent of maternal vasoconstriction (Robertson et al., 1967; Zhou et al., 1997). As the extravillous trophoblasts invade the spiral arteries early in pregnancy they form plugs which occlude the spiral arteries and prevent maternal blood from entering the intervillous space, thereby creating a physiologic hypoxic environment (Hustin and Schaaps, 1987; for review, see Jaffe et al., 1997; Burton et al., 1999). From 8 to 10 weeks of gestation direct measurements of oxygen tension showed that the partial pressure of oxygen in the placenta (17.9 mm Hg) was significantly lower than the endometrium (39.6 mm Hg) (Rodesch et al., 1992). Furthermore, from 12 to 13 weeks of gestation the placental oxygen tension increased to levels which were not significantly different from the endometrium, consistent with the loosening of the trophoblast plugs at that time (Rodesch et al., 1992). More recently, Jauniaux and colleagues measured respiratory gases and acid-base values in 30 early pregnancies and confirmed that before 11 weeks of gestation the partial pressure of oxygen in the placenta was 2.5 times lower than that in the decidua (Jauniaux et al., 2001). Near the end of the first trimester the trophoblast plugs progressively loosen, exposing the developing placenta to maternal blood flow from approximately 10 weeks of gestation (Jauniaux et al., 1992, 2003; Jaffe and Woods, 1993). However, it is important to note that while there is a mounting body of evidence to support a low oxygen environment during the first trimester, not all experimental evidence supports the ability of trophoblast plugs to block maternal blood flow into the intervillous space beyond 6 weeks of gestation, with one study showing that the majority of spiral arteries are not plugged, as well as sonographic evidence that maternal blood flow into the intervillous space occurs early in the first trimester (Kurjak et al., 1993; Kurjak and Kupesic, 1997; Meekins et al., 1997). Nevertheless the current weight of evidence supports the concept that the placenta and fetus develop in a hypoxic environment in the first trimester.
The early gestation placenta is poorly protected against oxidative damage, as the antioxidant enzymes copper/zinc superoxide dismutase and mitochondrial superoxide dismutase are not expressed by the syncytiotrophoblast until approximately 8–9 weeks of gestation, rendering the syncytiotrophoblast acutely sensitive to oxygen mediated damage (Watson et al., 1997). The expression of these protective enzymes increases significantly after this point as the trophoblast plugs loosen and the placenta becomes exposed to gradually increasing levels of oxygen and consequently experiences oxidative stress (Watson et al., 1997, 1998; Jauniaux et al., 2000).
To be able to respond to increasing levels of oxygen, as the maternoplacental circulation is established at the end of the first trimester and as invasive trophoblast progress along the spiral arteries, trophoblasts must be able to accurately sense oxygen tension (Caniggia and Winter, 2002). The exact mechanism by which trophoblasts sense oxygen tension is currently unclear; however, several potential pathways have been identified. Many of these pathways utilize the formation of reactive oxygen species (ROS), but it is currently unclear whether hypoxia results in an increase or decrease in cellular levels of ROS (for review see DeMarco and Caniggia, 2002). In hypoxic conditions, trophoblast oxygen sensing mechanisms utilize several different pathways to control gene expression. These pathways often utilize redox-sensitive transcription factors, of which the hypoxia inducible factor (HIF) family are the best characterized in trophoblasts.
The HIF family of transcription factors
To date, three members of the HIF family have been identified, but it is possible that others exist. All the members of the HIF family consist of an inducible alpha subunit (HIF-α) and a constitutively expressed beta subunit (HIF-β, also known as aryl hydrocarbon receptor nuclear translocator (ARNT)). Under physiologically normoxic conditions HIF-α is rapidly broken down following binding of von Hippel-Lindau tumour suppressor protein (pVHL), which targets HIF-α for degradation by the ubiquitin-proteasome pathway (Figure 1) (Maxwell et al., 1999; Cockman et al., 2000; Ohh et al., 2000). pVHL binds to HIF-α following hydroxylation of proline residues in the oxygen-dependent degradation domain of HIF-α (Jaakkob et al., 2001; Masson et al., 2001). However, under hypoxic conditions this hydroxylation of the proline residues is blocked, resulting in stabilization and accumulation of HIF-α in the cytoplasm (Figure 1) (Ivan et al., 2001; Lando et al., 2002). The stabilized HIF-α can translocate to the nucleus where it dimerizes with HIF-β. The HIF-α–β dimer is then able to bind to DNA and induce gene expression (Wang et al., 1995). Hypoxia further enhances the activity of HIF-α by promoting the ability of the HIF-α C-terminal trans-activation domain (CAD) to interact with coactivators such as p300 and thereby amplify HIF-α induced gene transcription (Ema et al., 1999).
HIF-1
The predominant member of the HIF family is HIF-1, which has been shown to regulate the expression of more than 60 genes in a variety of cell types (Semenza, 2003). In the placenta in vivo, HIF-1α has been shown to be strongly expressed from 5 weeks of gestation, predominantly in the extravillous trophoblast, syncytiotrophoblast and villous cytotrophoblast (Caniggia et al., 2000b; Rajakumar and Conrad, 2000). However, expression decreases at around 9 weeks of gestation, and by 12 weeks of gestation immunoreactivity for HIF-1α is weak or absent (Caniggia et al., 2000b; Caniggia and Winter, 2002).
HIF-2 and -3
HIF-2 and -3 arise from combinations of HIF-2α and -3α subunits, respectively, with the common HIF-β chain. HIF-2α is expressed in the syncytiotrophoblast, cytotrophoblasts and mesenchymal cells of the first trimester villous placenta (Rajakumar and Conrad, 2000). HIF-2α mRNA increases significantly with advancing gestational age, but conversely HIF-2α protein decreases with gestational age, indicating that as for the degradation of HIF-1α by pVHL, the main point of regulation for HIF-2a is post-transcriptional (Rajakumar and Conrad, 2000). In situ, pVHL is expressed in villous cytotrophoblast cells and sites of invasive trophoblast column initiation, corresponding to cytoplasmic HIF-2α staining (Genbacev et al., 2001). Confusingly, it has been reported that as trophoblasts invade into the decidua, pVHL is down-regulated and HIF-2α is localized to the nucleus (Genbacev et al., 2001). The up-regulation of trophoblast pVHL in response to hypoxia has been confirmed in vitro (Genbacev et al., 2001). Anchoring villi with attached trophoblast columns show strong staining for pVHL when cultured in 2% oxygen, whereas in 10% or 20% oxygen staining for pVHL is weak or absent (Genbacev et al., 2001). These results are contrary to expectation as higher oxygen conditions are proposed to result in the degradation of HIF-2α by pVHL, therefore pVHL would be expected to be down-regulated in hypoxic conditions. However, it seems likely that the ratio of HIF-2α to pVHL is more important than the absolute levels of either protein in determining a cell’s response to low oxygen. HIF-2α would also be expected to be targeted for degradation when oxygen tension increased, not localized to the nucleus as observed in this study (Genbacev et al., 2001). The function of HIF-2α in the placenta is yet to be elucidated.
Human HIF-3α was first identified in 2001; however, to date its placental expression pattern and functional role have not been reported (Hara et al., 2001).
Regulation of HIF
HIF-1α is able to be stabilized under normoxic conditions by a variety of growth factors and cytokines including epidermal growth factor (EGF), insulin, heregulin, insulin-like growth factors 1 and 2, transforming growth factor β1 and interleukin-1β (Zelzer et al., 1998; Feldser et al., 1999; Hellwig-Burgel et al., 1999; Laughner et al., 2001; Fukuda et al., 2002; Stiehl et al., 2002). However, stabilization of the HIF-1α subunit alone does not yield transcriptionally active HIF-1 (Salceda and Caro, 1997). The activation of HIF-1 involves multiple steps including, post-translational phosphorylation of the HIF-1α chain, nuclear translocation, HIF-β heterodimerization, DNA binding, recruitment of tissue-specific transcriptional cofactors and target gene trans-activation, which are under the control of a range of factors including oxygen concentration, growth factors, hormones, nutrients and cross communication with other signalling pathways allowing cell specific fine tuning of the hypoxic response (Figure 1) (Wenger, 2002; Lee et al., 2004).
The protein expression and transcriptional activity of HIF-1α is able to be regulated independently of hypoxia through the phosphatidylinositol 3 kinase (PI3K) and/or mitogen-activated protein kinase (MAPK) signal transduction pathways allowing cell or stimulus specific regulation of HIF-1α, which is independent of the oxygen-mediated regulation of HIF-a levels (Figure 1) (Conrad et al., 1999; Richard et al., 1999; Jiang et al., 2001; Fukuda et al., 2002). The MAPK pathway, in particular, is known to play an important role in the signal transduction of a range of cellular responses including proliferation, differentiation and stress responses (Kita et al., 2003). In human trophoblasts, the PI3K signalling pathway is activated by interleukin-12, whereas the MAPK signalling pathway is known to mediate the effects of insulin, IGF-II, insulin-like growth factor binding protein-I (IGFBP-I), leptin, endothelin-1, GnRH and EGF (Kang et al., 2000; Gleeson et al., 2001; McKinnon et al., 2001; Kong et al., 2002; Bifulco et al., 2003; Chakraborty et al., 2003; Mackova et al., 2003). The MAPK pathway can also be activated by the formation of ROS in hypoxic conditions (Kulisz et al., 2002).
Finally the transcriptional activity of HIF-1 can be regulated by factor inhibiting HIF-1 (FIH-1), an Fe(II)-dependent enzyme that uses molecular oxygen to modify its substrate, potentially allowing FIH-1 to function directly as a cellular oxygen sensor (Mahon et al., 2001). FIH-1 acts as co-repressor by forming a complex with both pVHL and HIF-1 (Lando et al., 2002). However, to date no data on the placental expression of FIH-1 has been reported.
Other transcription factors involved in regulating the response of trophoblasts to hypoxia
Several other transcription factors involved in trophoblast differentiation are responsive to hypoxia. The transcription factors Id1, Mash2 and the helix-loop-helix transcription factors upstream stimulatory factor-1 and -2 (USF1 and USF2) which mediate the effects of Mash2 are all up-regulated in 2% oxygen in comparison to 20% oxygen (Jiang et al., 2000; Jiang and Mendelson, 2003). The up-regulation of Mash2, USF1 and USF2 may inhibit cytotrophoblast fusion into syncytiotrophoblast (Jiang et al., 2000; Jiang and Mendelson, 2003).
The elevation of intracellular Ca2+ is believed to activate an HIF-1-independent signalling pathway that involves the transcription factor activator protein-1 (AP-1), with cooperation between the HIF-1 and AP-1 pathways allowing fine regulation of hypoxic gene expression (Laderoute et al., 2002; Salnikow et al., 2002) (Figure 2). AP-1 is a dimeric transcription factor composed from the products of the Jun and Fos proto-oncogenes (c-Jun, JunB, JunD, c-Fos, FosB, Fra-1 and Fra-2) (Dakour et al., 1999). AP-1 transcription factors are believed to play an important role in trophoblast differentiation. In the villus, AP-1 transcription factor expression is limited; however, extravillous trophoblasts express c-Jun, JunB, c-Fos, FosB and Fra-2 both in the first trimester and later in gestation (Bamberger et al., 2004). To date, no data on trophoblast intracellular calcium elevation or the role of the AP-1 pathway in mediating the effects of hypoxia in trophoblasts has been published. However, it is feasible that, as hypoxia is able to influence the AP-1 pathway in other cell types, this pathway may play a role in the response of trophoblasts to hypoxia (Schorpp-Kistner et al., 1999).
The trophoblast response to hypoxia in the first trimester of pregnancy
Transcription factors facilitate the expression of a wide range of genes in response to hypoxia in the first trimester of pregnancy. Gene responses to hypoxia can be broadly divided into two groups—those that promote cell survival by increasing oxygen delivery, decreasing oxygen consumption and adapting cellular metabolic responses and are therefore a common response to hypoxia in many tissues and those that play a specific role in the regulation of implantation and placentation. Changes in gene expression in hypoxia that promote trophoblast survival include increased glucose consumption and the expression of proteins involved in glycolysis (Hoang et al., 2001; Baumann et al., 2002). The remainder of this review will focus specifically on the second group and how hypoxia regulates trophoblast differentiation in the first trimester.
The effect of low oxygen on trophoblast differentiation
In vivo, only the extravillous trophoblasts most proximal to the villi proliferate (Irving et al., 1995; Vivovac et al., 1995). As extravillous trophoblasts migrate away from the villi and invade into the maternal decidua they progressively develop an invasive phenotype and are no longer able to proliferate (Genbacev et al., 1997, 2000). There are two opposing schools of thought on the effect of hypoxia on trophoblast differentiation in the first trimester of human pregnancy. Evidence exists to demonstrate that hypoxia promotes a proliferative trophoblast phenotype, which would create a large pool of trophoblasts, thereby providing sufficient numbers to invade into the maternal decidua. However, contradictory results have also shown hypoxia to promote an invasive trophoblast phenotype which may be important in achieving sufficient depth and extent of trophoblast invasion.
Evidence for low oxygen promoting a proliferative phenotype in villous trophoblast
Several different lines of evidence support the hypothesis that hypoxia promotes trophoblast proliferation. In vivo, there is a high level of mitoses in the villous cytotrophoblast between 6 and 10 weeks of gestation, but the mitotic index decreases significantly between 10 and 12 weeks of gestation when the placenta is exposed to maternal blood (Tedde and Tedde Piras, 1978). This observation is supported by in vitro evidence. Firstly, HTR-8/SVneo cells, a human first trimester cytotrophoblast cell line, show both increased proliferation and reduced invasion through Matrigel when cultured in 2% oxygen conditions (Kilburn et al., 2000). Secondly, isolated first trimester cytotrophoblasts show increased rates of DNA synthesis in 2% oxygen in comparison to 20% oxygen (Jiang et al., 2000). Thirdly, in vitro the ratio of cytotrophoblast : syncytiotrophoblast nuclei abruptly declines in placentae over 8 weeks of gestation despite the number of cytotrophoblast nuclei per unit area remaining constant suggesting that cytotrophoblast proliferation is greater in the hypoxic conditions of early pregnancy (Bose et al., 2003). Finally, in comparison to explants cultured in 20% oxygen, first trimester villous explants cultured in 2 or 3% oxygen show increased 5-bromo-2"-deoxyuridine (BrdU) (thymidine analogue) incorporation, increased extravillous trophoblast outgrowth and an increase in the total number of cells in this outgrowth (Genbacev et al., 1997; Caniggia et al., 2000b). In support of these results, first trimester villous explants cultured in 1 and 5% oxygen formed 55 and 40% more outgrowths, respectively, than explants cultured in 20% oxygen (Sferruzzi-Perri and Roberts, 2003). In contrast, our group has observed a 27% decrease in the number of extravillous trophoblast outgrowths formed from first trimester villous explants cultured in 1.5% oxygen in comparison to those cultured in 8% oxygen (James et al., 2004). Furthermore, in our study, extravillous trophoblast outgrowths formed under hypoxic conditions contained fewer cells than those produced in 8% oxygen (James et al., 2004).
Hypoxia has also been shown to reduce the invasive capacity of trophoblasts and the expression of molecules associated with an invasive trophoblast phenotype such as α1 integrin and matrix metalloprotease-2 (MMP-2) (Genbacev et al., 1997; Caniggia et al., 2000a; Kilburn et al., 2000; Crocker et al., 2003). Culture in 3% oxygen also reduces the invasion of isolated cytotrophoblasts from the vessel lumen into the walls of dissected spiral arteries in comparison to culture in 20% oxygen (Crocker et al., 2003).
Further indirect evidence for the stimulation of a proliferative trophoblast phenotype in response to hypoxia is provided by the regulation of key cytokines that promote this phenotype. Transforming growth factor β3 (TGFβ3) acts under the control of HIF-1α to inhibit trophoblast invasion (Caniggia et al., 2000b; Schaffer et al., 2003; Nishi et al., 2004). TGFβ3 expression in placental villi has been reported to increase from approximately 6 to 9 weeks of gestation, with a failure in its down-regulation around 9 weeks of gestation reported to be associated with shallow trophoblast invasion, which may predispose the pregnancy to pre-eclampsia (Caniggia et al., 1999, 2000a). Therefore, induction of HIF-1α by hypoxia in the first trimester may up-regulate TGFβ3 expression, inhibiting trophoblast differentiation to an invasive phenotype (Caniggia et al., 2000a). However, these temporal changes in TGFβ3 expression were not detected by Simpson and colleagues, who could only detect low levels of TGFβ3 immunostaining between 7 and 19 weeks of gestation (Simpson et al., 2002).
IGF-II has also been identified as a cytokine which is able to mediate the effects of hypoxia on extravillous outgrowth (Sferruzzi-Perri and Roberts, 2003). Addition of exogenous IGF-II to first trimester villous explant cultures increases the formation of extravillous trophoblast outgrowth in 20% oxygen conditions, but not in 1 or 5% oxygen conditions in which outgrowth production is elevated independent of exogenous IGF-II (Sferruzzi-Perri and Roberts, 2003).
Evidence for low oxygen promoting an invasive trophoblast phenotype
In contrast to the above discussion, a smaller body of indirect evidence exists that suggests that hypoxia may induce extravillous trophoblast differentiation into an invasive trophoblast phenotype. One mechanism of trophoblast invasion involves urokinase-type plasminogen activator (uPA), which is secreted as an inactive pro-enzyme that is activated upon binding to its highly specific receptor (Blasi, 1993). Receptor bound uPA is able to convert plasminogen into plasmin, which is then able to degrade several ECM components, activating growth factors and latent metalloproteases required for invasion (Saksela, 1985; Petersen et al., 1988; Mayer, 1990). In direct contrast with the results of Kilburn and colleagues discussed previously, Graham and colleagues found that culture of HTR-8/SVneo trophoblast cells in 1% oxygen increased the invasion of these cells through Matrigel in comparison to culture in 20% oxygen, by up-regulating uPA receptors (Graham et al., 1998). The uPA system is regulated by PAI-1 and PAI-2, which inhibit both free and bound uPA by forming irreversible covalent complexes. HTR8/SVneo cells cultured in hypoxia show elevated PAI-1 expression (Fitzpatrick and Graham, 1998). Therefore, Graham and colleagues propose a model whereby hypoxia stimulates trophoblast invasion by increasing uPA receptor expression at the leading edge of the invading cell, as well as stimulating the secretion of PAI-1 at the receding edge where proteolytic activity is no longer required but detachment of the cell is needed for migration (Graham et al., 2000).
Leptin is a hormone primarily produced in humans by adipocytes and has a role in the regulation of food intake and energy expenditure. However significant amounts of leptin are also produced by the placenta during pregnancy, with levels peaking in the second trimester (Masuzaki et al., 1997). Leptin is expressed homogenously throughout extravillous cell columns, the leptin receptor is expressed in a clear gradient along the extravillous trophoblast columns with the strongest expression at the distal cells of the columns (Castellucci et al., 2000). Accordingly the binding of leptin to its receptor is believed to stimulate an invasive trophoblast phenotype (Castellucci et al., 2000). In vitro, hypoxia activates the human leptin gene promoter through HIF-1 in BeWo choriocarcinoma cells (Grosfeld et al., 2002). Therefore, leptin may provide another mechanism by which hypoxia could stimulate an invasive trophoblast phenotype. However, the role of hypoxia in the stimulation of leptin production and activity in the first trimester of human pregnancy has yet to be elucidated.
Alterations in the trophoblast adhesion molecule repertoire and production of extracellular matrix components by hypoxia have been linked with extravillous trophoblast differentiation into an invasive phenotype. HTR-8/SVneo cells cultured in hypoxic conditions show increased trophoblast fibronectin production, but decreased expression of α5 integrin which forms part of the α5β1 integrin fibronectin receptor and consequently decreased trophoblast adhesion to fibronectin (Lash et al., 2001; Chen and Aplin, 2003). It is possible that increased trophoblast fibronectin production is an indirect result of the acidic environment that hypoxia creates (Gaus et al., 2002). However, despite the above evidence being used by several authors as support for an invasive phenotype, fibronectin blocking antibodies do not affect trophoblast invasion in vitro, and consequently α5β1 integrin is not believed to play a significant role in trophoblast invasion (Damsky et al., 1994). The uPA receptor may also interact directly with specific integrins, such as β1 integrin, to modulate trophoblast adhesion to, and migration towards, ECM vitronectin (Wei et al., 1996; Lash et al., 2001). However, PAI-1 is also able to act as an antagonist by binding to vitronectin and preventing its association with the uPA receptor (Lash et al., 2001).
Does the physiological hypoxia of early pregnancy promote an invasive or a proliferative trophoblast phenotype?
The confusion as to the effect of hypoxia on trophoblast differentiation in the first trimester of human pregnancy is confounded by the fact that the cellular response to hypoxia is not an isolated pathway, but a combination of multiple sensing mechanisms, transcription factors, oxidative stress responses and cytokine production, not to mention a large number of discordant results in the literature. Many of the studies providing evidence for hypoxia promoting an invasive trophoblast phenotype use the extravillous trophoblast hybrid cell line HTR-8/SVneo, as a substitute for extravillous trophoblasts which are difficult to propagate in culture, and whether this cell line truly reflects the behaviour of normal trophoblasts is open to debate. The use of explant cultures to study extravillous trophoblast differentiation provides significant advantages in that they allow outgrowth in a manner similar to that which occurs in vivo, but results from individual explants tend to be variable and therefore a high number of replicates are required to create meaningful data. It is interesting to note that the gene expression profile of first trimester villous explants cultured in 3% oxygen is remarkably similar to that of term placentae from high-altitude and pre-eclamptic pregnancies indicating that, although a low oxygen environment is the physiological norm in the first trimester, this model may provide insights into how a decreased oxygen supply may affect the placenta later in gestation (Soleymanlou et al., 2005). The interpretation of what extravillous trophoblast outgrowth from explants signifies in different explant models is also often unclear, with some reports concluding that this is indicative of extravillous trophoblast proliferation and clearly separate it from extravillous migration (Caniggia et al., 2000b), whereas others conclude that extravillous trophoblast outgrowth represents trophoblast invasion (Irving et al., 1995; James et al., 2004). Finally, as the acquisition of an invasive phenotype is a progressive phenomenon we should not assume that the proliferative and invasive trophoblast phenotypes are necessarily mutually exclusive, and it is possible that the effect of hypoxia on trophoblast phenotype may vary with the specific properties of various cell groups.
It is important to note that many of the above experiments used physiologically superoxic conditions of 20% oxygen (pO2 = 140 mm Hg) as a control. However, once the maternofetal circulation is established the blood in the intervillous space is a mixture of arterial and venous blood, and blood sampled from the intervillous space at term has a pO2 of 40 mm Hg (Howard et al., 1961). Therefore, a more physiologically relevant control concentration of around 6–8% oxygen may more accurately represent conditions in vivo once the maternal circulation in the intervillous space is established. Furthermore, the range of experimental oxygen concentrations (from 1 to 3% oxygen) used to represent hypoxia may also contribute to the experimental differences observed. Finally, the gestation and mode of collection of placentae may also be important as exposure to maternal blood or significant time delay at the time of tissue collection leading to oxidative stress is able to affect immunoreactivity of not only HIF-1α, but the induction of downstream factors such as TGFβ3 (Dyson et al., 2003).
In conclusion, trophoblast differentiation is essential for the success of human pregnancy, and despite some conflicting experimental evidence, hypoxia appears to play a vital role in regulating trophoblast differentiation in the first trimester. The regulation of trophoblast differentiation by hypoxia is a result of complex interactions between factors associated with oxidative stress, oxygen sensing mechanisms and the release of inflammatory cytokines. Therefore, aberrations in any one of these factors, along with the temporal and spatial regulation of blood flow in the intervillous space has the potential to result in altered gene expression and trophoblast phenotype.
Acknowledgements
J.James receives a University of Auckland Doctoral Scholarship.