- Email: [email protected]
Placental Oxygen Consumption. Part I: In Vivo Studies—A Review
Placenta (2000), 21, Supplement A, Trophoblast Research, 14, S31–S37 doi:10.1053/plac.1999.0513, available online at http://www.idealibrary.com on
CELL BIOLOGY Placental Oxygen Consumption. Part I: In Vivo Studies—A Review A. M. Cartera Department of Physiology and Pharmacology, University of Southern Denmark, Winsloewparken 21, DK-5000 Odense, Denmark Paper accepted 6 December 1999
At term of pregnancy, oxygen consumption by the human or ovine placenta accounts for 40 per cent of total oxygen uptake by the gravid uterus. In the sheep, most oxygen is used for oxidative phosphorylation of glucose; the remainder is probably utilized for non-mitochondrial processes. The ATP yield is expended mainly in protein synthesis and cation transport. The fractional protein synthesis rate of ovine placenta is 60 per cent per day. Applying these data to man, protein synthesis is estimated to account for about 30 per cent of placental oxygen uptake. Probably this reflects the high rates of synthesis of peptide and steroid hormones. The Na + gradient is the basis for secondary active transport of amino acids and other substances, and the Na + –K + -pump probably accounts for 20–30 per cent of oxygen uptake, with a smaller contribution from Ca2+ -ATPase. Placental oxygen uptake remains constant during acute reductions in uterine oxygen supply and is maintained at the expense of the fetus. In the longer term, in experimental models of fetal growth restriction, placental oxygen consumption is reduced to a greater extent than fetal oxygen consumption. Placental oxygen consumption is greatly reduced under in vitro experimental conditions, due largely to an inadequate oxygen supply. This results in reduced protein synthesis and possibly inhibition of Na + –K + -ATPase. However, if the placenta is subjected to hyperoxia, by raising the P2 of the medium, there is an increase in anaerobic glycolysis and structural damage may ensue. Premature exposure of trophoblast to high oxygen tensions in vivo may result in reduced villous branching, but this is likely to be a cause, rather than a consequence, of reduced fetal growth and oxygen consumption. 2000 IFPA and Harcourt Publishers Ltd Placenta (2000), 21, Supplement A, Trophoblast Research, 14, S31–S37
INTRODUCTION The placenta has long been known to have a high rate of oxygen consumption (Campbell et al., 1966). This review will examine why the placenta requires so much oxygen for its metabolism and discuss how placental oxygen consumption affects transfer of oxygen to the fetus. Since oxygen uptake by the placenta decreases greatly under in vitro experimental conditions, it will be asked how this impacts on placental metabolism and function. This entails consideration of the practice of maintaining culture media and perfusion fluids at high levels of P2. Finally, since hyperoxia in vitro has been shown to alter placental metabolism and cause tissue damage, it will be asked if comparable conditions can arise in disease states. In examining the relationship between placental oxygen a
To whom correspondence should be addressed at: Department of Physiology and Pharmacology, Winsloewparken 21, 3rd Floor, DK-5000 Odense, Denmark. Fax: +45 6613 3479; E-mail: [email protected]
0143–4004/00/0A00S31+07 $35.00/0
consumption and metabolism, the approach adopted is that developed by Rolfe and Brown (1997) in their perspicuous analysis of standard metabolic rate.
PLACENTAL OXYGEN UPTAKE AND METABOLISM In sheep, uteroplacental oxygen uptake is calculated as the difference between total oxygen uptake by the gravid uterus and oxygen uptake by the fetus. This requires simultaneous measurement of the uterine and umbilical blood flows and of the arteriovenous differences in oxygen content across the two circulations (Meschia et al., 1980). In unanaesthetized sheep near term of pregnancy (term is about 147 days), uteroplacental oxygen uptake is 0.9 mmol/min (20 ml/min), or about 40 per cent of total uterine oxygen uptake (Meschia et al., 1980). In mid-pregnant sheep (71–81 days gestation), fetal oxygen consumption is smaller and uteroplacental oxygen uptake 2000 IFPA and Harcourt Publishers Ltd
Placenta (2000), Vol. 21, Supplement A, Trophoblast Research, Vol. 14
S32
accounts for more than 80 per cent of the total (Bell et al., 1986). To determine the oxygen consumption of the human placenta, Bonds et al. (1986) used indirect calorimetry to measure oxygen uptake in women undergoing elective caesarian section. Determinations were made before and after cord clamping and after removal of the placenta. This yielded values for fetal and placental oxygen uptake of 0.98 and 0.58 mmol/min, respectively (22 and 13 ml/min). Thus, as in the sheep, oxygen consumption by the human placenta is about 40 per cent of total uterine oxygen uptake.
Protein synthesis
Mitochondrial
ATP synthesis
Placental ATP synthesis
Other
Meschia et al. (1980) determined oxygen consumption, glucose utilization and lactate production by the ovine uterus and placenta. Glucose utilization was 0.205 mmol/min and total lactate production was 0.150 mmol/min. The overall reaction for anaerobic glycolysis is: Glucose+2Pi +2ADP<2 lactate+2ATP+2 H2O
Na+-K+ATPase
(1)
Thus the amount of glucose available for oxidative metabolism was 0.205(0.150/2)=0.130 mmol/min. The overall reaction for oxidative phosphorylation is: Glucose+31 ADP+31 Pi +31 H + +6O2<6 CO2 +31 (2) ATP+37 H2O Therefore, oxidative phosphorylation of 0.130 mmol/min glucose can account for 0.780 mmol/min oxygen or 87 per cent of the measured oxygen consumption, which was 0.896 mmol/ min. In the standard state, about 10 per cent of the body’s total oxygen uptake is non-mitochondrial and not, therefore, associated with ATP production (Rolfe and Brown, 1997). This fraction varies between tissues, reaching 20 per cent in the liver, and may well exceed 10 per cent in the placenta. Thus, oxidative metabolism of glucose can account for virtually all of the mitochondrial fraction of oxygen consumption in the ovine placenta at term. It was once thought that the reaction in equation (2) gave a net yield of 36 moles ATP for each mole of glucose utilized. However, the mechanistic ratio of ATP produced to oxygen consumed is now thought to be 2.6 rather than 3, with a yield of 30–32 moles ATP per mole of glucose (Rolfe and Brown, 1997). Additionally, under physiological conditions, glycolysis is partly uncoupled from ATP synthesis due to passive leakage of protons from the mitochondria (Rolfe and Brown, 1997). Estimates for the liver indicate that the proton leak amounts to 15 per cent of mitochondrial oxygen consumption. Therefore, the effective ratio of ATP to oxygen is closer to 2.2. Applied to the ovine placenta, this gives an approximate ATP synthesis rate of 3.4 mmol/min. In extrapolating to human placenta, it should be borne in mind that placental handling of glucose could differ from that
Nonmitochondrial
Proton leak
Figure 1. The estimated contribution of metabolic processes to placental energy consumption. The first column shows contribution of mitochondrial and non-mitochondrial oxygen consumption to total placental respiration. The second column shows the proportion of mitochondrial respiration used to drive ATP synthesis and proton leak and is based on estimates for whole body metabolism. The third column represents the contribution of protein synthesis and cation transport to total ATP consumption. Analysis based on Rolfe and Brown (1997).
in sheep. The weight specific oxygen consumption of human placenta is similar to that of the ovine placenta (37 versus 34 ml/min per kg; Bonds et al., 1986; Meschia et al., 1980), but the tissue mass is smaller. ATP synthesis by a 400 g human placenta should therefore be in the order of 2 mmol/min. How does the placenta dispose of this energy? ATP consumption is probably for protein synthesis and cation transport (Figure 1). The processes subserved by this energy consumption include: growth; transport and maintenance of ion gradients; and endocrine functions.
Contribution of protein synthesis to placental oxygen consumption The rate of protein synthesis by the human body is 347 g/day (James et al., 1976), accounting for 25–30 per cent of the ATP utilized under standard conditions (Rolfe and Brown, 1997). Protein turnover in individual tissues can be calculated by infusing a labelled amino acid, such as tyrosine, until the plasma concentration reaches a plateau, and measuring the rate at which the amino acid is incorporated into protein (Sender and Garlick, 1973). The fractional protein synthesis rate of the ovine placenta, determined in vivo, was 60 per cent per day (Young et al., 1982). This is five times faster than the rate determined for the liver of adult sheep (Buttery et al., 1975). Human placenta in vitro, perfused on the maternal side with diluted blood (haematocrit 20–25 per cent) at physiological P2, had a protein synthesis rate of 40 per cent per day (Carroll and Young, 1987).
Carter: Placental Oxygen Consumption (Part 1)
S33
These values suggest that protein synthesis accounts for a substantial fraction of placental oxygen consumption. Assuming a placental weight of 400 g (Bonds et al., 1986), of which 12 per cent is protein (Widdowson and Spray, 1951), the rate of protein synthesis is 29 g/day (at a protein synthesis rate of 60 per cent per day). Taking the average molecular weight per peptide bond to be 110 g/mol (Rolfe and Brown, 1997), this amounts to 0.18 mmol/min. The equivalent oxygen consumption can be derived by multiplying the protein synthesis rate by the ATP requirement (four ATP equivalents per peptide bond) and dividing by 2.2, the effective ratio of ATP synthesized to oxygen consumed (see above). A value of 0.17 mmol/min O2 or 3.7 ml/min is found. This is approximately 30 per cent of the total oxygen consumption of a 400 g placenta (13 ml/min; Bonds et al., 1986). The placenta synthesizes a wide range of structural proteins and enzymes and a range of products with putative endocrine, metabolic, haemostatic and immunological functions (Chard, 1986). Placental proteins are secreted primarily to the maternal circulation and their effects are probably confined to the mother. They include human chorionic gonadotrophin (hCG), human placental lactogen (hPL), pregnancy-associated placental protein-A [PAPP-A; insulin-like growth factor binding protein-4 protease (Lawrence et al., 1999)], pregnancy-specific 1-glycoprotein (PSG) and placental growth hormone (Chard, 1986; Sørensen et al., 1995; Frankenne et al., 1988). Secretion of hCG peaks at 8–10 weeks of pregnancy, but secretion of other proteins continues to rise (Sørensen et al., 1995). At 17 weeks, daily secretion of hPL, PAPP-A and PSG amounts to about 350 mg (Sørensen et al., 1995). The rate of secretion of hPL towards term, about 1 g/day, exceeds that of any other human peptide hormone (Grumbach et al., 1968). The placenta also secretes quantities of steroids, principally oestriol and progesterone, into the fetal and maternal circulations. Other placental products that probably have far reaching effects on fetal physiology are prostaglandin E2 (Thorburn, 1995) and adenosine (Ball et al., 1995).
for 20–30 per cent of total oxygen uptake in fetal placenta and 18–24 per cent in maternal placenta. However, the oxygen uptake of placenta in these experiments in vitro was only 15 per cent of in vivo values (Vatnick and Bell, 1992) and it is not known whether Na + –K + -ATPase activity was reduced proportionately. The energy cost of Ca2+ cycling by the Ca2+ -ATPase of the sarcoplasmic reticulum of skeletal muscle accounts for 20– 50 per cent of the oxygen consumption during contraction and ca. 5 per cent in resting muscle cells. In the brain, where Ca2+ serves as a signal for neurotransmitter release, Ca2+ cycling across the plasma membrane also contributes significantly to energy turnover. The energy cost of maintaining Ca2+ gradients in the placenta is not known, but probably does not exceed that in the liver, where it accounts for 1–10 per cent of energy consumption (Rolfe and Brown, 1997).
Contribution of cation transport to placental oxygen consumption
Reactions that uncouple metabolism
Cation transport makes a significant contribution to basal oxygen consumption. It has been calculated that oxygen consumption by the Na + –K + -pump accounts for about 20 per cent of the basal metabolic rate (Clausen, van Hardeveld and Everts, 1991). The Na + gradient is the basis for secondary active transport of amino acids and other substances via co-transport systems. Active Na + –K + transport is, for example, essential for the absorptive function of the intestine, where it answers for a substantial fraction of total energy turnover (ca. 30 per cent). An indication of the contribution of Na + –K + -ATPase to placental energy utilization is given by the ouabain-sensitive oxygen uptake. This was determined for ovine placental tissues at three stages of gestation by Vatnick and Bell (1992). Ouabain-sensitive oxygen uptake accounted
Contribution of other reactions to placental oxygen consumption The standard metabolic rate of the whole body receives major contributions from gluconeogenesis, ureagenesis, actinomyosin ATPase (chiefly in cardiac muscle), substrate recycling and RNA synthesis (Rolfe and Brown, 1997). Gluconeogenesis makes a substantial contribution to the metabolic rate and may account for as much as 30 per cent of oxygen consumption by the human liver. Human placenta certainly exhibits phosphoenolpyruvate carboxykinase activity (Diamant et al., 1975) and may produce glucose (Prendergast et al., 1999), but there are no quantitative estimates of placental gluconeogenesis. They may be hard to come by experimentally, as phosphoenolpyruvate carboxykinase activity cannot be detected in the sheep placenta (Edwards et al., 1977). Neither human nor ovine placenta is capable of ureagenesis (Diamant et al., 1975; Edwards et al., 1977) and the activity of actinomyosin ATPase can be discounted in this organ.
Cellular energy metabolism is uncoupled by processes in the same cell or elsewhere in the body; if it were not, metabolism would cease. Protein synthesis is uncoupled by protein degradation, the Na + pump by Na + channels and Na + -coupled transport, the Ca2+ pump by Ca2+ channels and gluconeogenesis by glycolysis.
OXYGEN TRANSFER TO THE FETUS Factors that determine oxygen transfer across the placenta include: the oxygen affinity and oxygen capacity (haemoglobin concentration) of maternal and fetal blood; rates of blood flow in the uterine and umbilical circulations; the permeability and surface area of the placental barrier; and placental oxygen
Fetal O2 consumption (ml O2 /min/kg)
consumption (reviewed in Carter, 1989; Longo, 1987). To a great extent, fetal oxygen consumption can be maintained despite fluctuations in oxygen delivery. Thus, fetal and maternal placental blood flows and blood oxygen capacities can be altered by as much as 50 per cent without any major change occurring in fetal oxygen uptake. The oxygen content of umbilical venous blood will fall, and with it oxygen delivery to the fetus, but fetal oxygen consumption will be maintained by increasing the fractional extraction of oxygen from the blood. The resultant fall in oxygen content of the umbilical arterial blood will maintain the arteriovenous difference across the umbilical circulation and placental oxygen transfer will remain the same. This has been verified in a wide range of experimental settings, supporting theoretical predictions made by Longo et al. (1972) on the assumption that compensatory mechanisms would intervene to maintain fetal oxygen consumption. There is, however, a critical threshold below which fetal oxygen uptake becomes dependent on oxygen delivery. In the sheep this corresponds to an oxygen delivery of about 0.6 mmol/min per kg fetal weight (13 ml/min per kg). Fetal oxygen uptake tends to fall when oxygen delivery is reduced below this level (Figure 2). This can be shown by decreasing maternal placental blood flow (Gu et al., 1985; Wilkening and Meschia, 1983); raising or lowering fetal haematocrit (Edelstone et al., 1985); decreasing maternal oxygen capacity (Christensen et al., 1986; Paulone et al., 1987); or decreasing fetal oxygen affinity (Edelstone et al., 1989). The resultant tissue hypoxia and inability to maintain oxidative metabolism is reflected in a lowering of fetal arterial pH and base excess. Clearly, if the uterine oxygen supply is limited, the large oxygen consumption of the placenta will impose a limitation on oxygen transfer to the fetus. Placental oxygen consumption remains remarkably constant following acute alterations in uterine oxygen supply. This is true both when the oxygen supply is reduced, by partial occlusion of the uterine artery (Gu et al., 1985; Wilkening and Meschia, 1983), and when it is increased, by raising the oxygen saturation of maternal arterial blood (Battaglia et al., 1968). Thus, reduction of uterine blood flow by 30–50 per cent for 60 min depressed fetal oxygen uptake, because oxygen consumption by the placenta was maintained at the expense of the fetus [Figure 3(A); Gu et al., 1985]. At the same time, fetal glycogen stores were mobilized to supply the placenta with glucose and lactate (Gu et al., 1985). Fetal supply of nutrients to the placenta cannot be sustained indefinitely. Thus, some form of adaptation can be expected to occur during long-term reductions in uterine blood flow. In fetal growth restriction following carunclectomy (Owens et al., 1987a), where the supply of oxygen and nutrients was restricted throughout gestation, uteroplacental oxygen consumption was reduced to a greater extent than fetal oxygen consumption [Figure 3(B)]. Consequently, it accounted for a smaller proportion of total uterine oxygen uptake (28 per cent) than during normal growth (53 per cent). Uteroplacental consumption of glucose was also reduced compared to normal growth controls (Owens et al., 1987b).
Placenta (2000), Vol. 21, Supplement A, Trophoblast Research, Vol. 14
10
5
0 10
Base excess (meq/l)
S34
0
–10
–20
0
10
20
30
Fetal O2 delivery (ml O2 /min/kg) Figure 2. Fetal oxygen consumption and arterial base excess as a function of oxygen delivery in sheep. Fetal oxygen uptake remained constant until oxygen delivery was reduced below ca. 13 ml/min per kg fetal weight (0.6 mmol/min per kg). There was then an increase in anaerobic metabolism with a resultant fall in arterial base excess. Data are from experiments in which both oxygen affinity (haemoglobin type) and oxygen capacity (haemoglobin concentration) were varied. Circles: fetal haemoglobin; squares: adult haemoglobin; filled circles and squares: normal haematocrit; half-filled circles and squares: moderate anaemia; open circles and squares: severe anaemia. Reproduced with permission from Edelstone, Darby, Bass & Miller (1989) Effects of reductions in hemoglobin-oxygen affinity and hematocrit level on oxygen consumption and acid-base state in fetal lambs. American Journal of Obstetrics and Gynecology, 160, 820–828.
PLACENTAL OXYGEN CONSUMPTION IN VITRO The high rates of oxygen consumption measured in vivo usually are not maintained under in vitro experimental conditions. Indeed, oxygen consumption by the placenta is greatly reduced, largely because it does not receive an adequate oxygen supply. As shown by Challier, Schneider and Dancis (1976), and detailed in this volume (Schneider, 2000), oxygen uptake by the perfused placental cotyledon is linearly related to oxygen delivery. In their experiments, the preparation was first perfused on the fetal or maternal side, or both, with buffered salt solution. The highest values for oxygen supply and oxygen consumption were obtained when the placenta was perfused on both sides with buffer equilibrated with 95 per cent oxygen and 5 per cent carbon dioxide. However, even under these
Carter: Placental Oxygen Consumption (Part 1)
S35
A 1600
1200
Oxygen uptake (µmol/min)
800
400
0 B 2400
Fetus
Placenta
Control
Fetus
Placenta
Arterial occlusion
2000 1600 1200 800 400 0
Fetus
Placenta
Normal growth
Fetus
Placenta
Growth restricted
Figure 3. (A) Effect of an acute 60 min reduction in uterine blood flow on oxygen uptake of the ovine fetus and uteroplacenta. Placental oxygen consumption is maintained at the expense of the fetus. Data from Gu et al. (1985). (B) Effect of long-term restriction in uterine oxygen and substrate supply, following carunclectomy, on oxygen uptake of the ovine fetus and uteroplacenta. Placental oxygen consumption is reduced to a greater extent than that of the fetus. Data from Owens et al. (1987a).
conditions, the supply of oxygen was much less than it would be in vivo. The amount of oxygen dissolved in aqueous solution depends on the solubility coefficient () and is about 23 ml/l solution (at 760 mm Hg and 37C). A far greater quantity of oxygen can be carried in the blood, due to the high oxygen affinity of haemoglobin. Thus blood with a haemoglobin concentration of 150 g/l can carry 200 ml/l oxygen as oxyhaemoglobin. It can be estimated that the placental oxygen supply under physiological conditions is in the order of 160 ml/min/ kg placenta. In experiments where the cotyledon was perfused either with whole blood or with red blood cells suspended in buffer, equilibrated with room air or 30 per cent oxygen, placental oxygen consumption was much higher than in those where the perfusate was buffered salt solution (Challier, Schneider and Dancis, 1976; Schneider, 2000). What does this mean in terms of placental metabolism? Glucose uptake was greater when the placenta was perfused with blood equilibrated with air than when the perfusate was buffered salt solution equilibrated with 95 per cent oxygen (Schneider et al., 1989). Moreover, the amount of glucose converted to lactate was 22 per cent when the placenta was
perfused with blood, about what would be expected by extrapolation from the sheep studies, but 77 per cent of glucose uptake when the placenta was perfused with buffer. As we have seen, it is not possible to achieve physiological levels of oxygen delivery with buffer as the perfusate, and this explains the higher rate of anaerobic glycolysis. In a third group of experiments, the buffer was initially equilibrated with 95 per cent nitrogen and 5 per cent carbon dioxide although, because of the gas permeability of the perfusion system, the measured P2 of the solution was 34 mmHg. These placentae had a higher rate of oxygen uptake than those perfused with buffer equilibrated with 95 per cent oxygen, suggesting that hyperoxia has a deleterious effect on placental metabolism (Schneider et al., 1989). How do these metabolic changes affect placental function? It seems inevitable that placental protein synthesis will decrease as a consequence of reduced oxidative metabolism and, with it, the rate of secretion of steroids and protein hormones. There is a marked increase in production of hPL and hCG when the placenta is perfused with diluted blood (haematocrit 22 per cent) compared to buffered salt solution (Schneider, 2000). Furthermore, hCG production by villous fragments perifused with buffer and tissue culture medium has been shown to vary with oxygen tension (Kay et al., 1997). The highest rate of secretion was associated with 20 per cent oxygen (measured P2 167 mmHg). A significantly lower secretion rate was observed under hyperoxic conditions, following equilibration of the medium with 95 per cent oxygen (measured P2 543 mmHg). There is indirect evidence that a reduced oxygen supply affects the activity of the sodium pump. Ouabain decreased the oxygen consumption of first trimester villous fragments by 22 per cent (Birdsey et al., 1997), much as would be expected by extrapolation from the sheep data (Vatnick and Bell, 1992). However, ouabain was without effect on the oxygen consumption of term villous fragments (Birdsey et al., 1997). In these experiments, the culture medium was gassed with 5 per cent CO2 in air, and thus had an acceptable P2, but a low oxygen content. Perhaps the placenta is able to shut down the sodium pump under these conditions.
PLACENTAL ‘HYPEROXIA’ In the context of in vitro experiments, raising P2 achieves little in the way of improving the oxygen supply, whilst non-physiological levels of oxygen tension may impair metabolism and cause tissue damage. First trimester syncytiotrophoblast is especially susceptible to hyperoxia. Following just 4 h of culture with 21 per cent oxygen, the syncytiotrophoblast of 9 week old placental tissue showed mitochondrial swelling and loss of microvilli (Watson et al., 1998). Can the placenta experience hyperoxia under in vivo conditions? It has been suggested that this occurs in some pathological pregnancies and is the root cause of defective villous branching (Kingdom and Kaufmann, 1997). In fetal
S36
growth restriction with absent end diastolic flow, for example, it was suggested that failure of the fetus to extract oxygen from the intervillous space leads to an elevation in the P2 to which the villi are exposed and removes the hypoxic drive for normal branching angiogenesis (Kingdom and Kaufmann, 1997). However, the P2 of maternal arterial blood does not exceed 100 mmHg and the normal intervillous P2 may be close to 60 mmHg (Rodesch et al., 1992). It is difficult to conceive of a change in P2 within this narrow range being sufficient to halt villous development. It could be argued that villous branching occurs quite early in placental development, in the second trimester, when intervillous P2 is lower, possibly <20 mm Hg (Rodesch et al., 1992). However, fetal consumption of oxygen at this time is much smaller. In sheep, for example, the fetal share of uterine oxygen uptake at mid gestation is only 15–20 per cent of the total (Bell et al., 1986). It seems unlikely that a reduction in what is already a small share of uterine oxygen uptake would alter trophoblast P2 sufficiently to affect angiogenesis. If premature exposure of trophoblast to high oxygen tensions does lead to reduced villous branching, this is more likely to be a cause than a consequence of reduced fetal growth. CONCLUSIONS Surprisingly little is known about the metabolism of the placenta. Most of the oxygen consumed is likely to be used for oxidative phosphorylation and the effective ratio of ATP generated to oxygen consumed is probably about 2.2, as in other tissues. There is no reason to believe that uncoupling occurs to a greater extent in the placenta, but for a dissenting view see Schneider (2000). Based largely on sheep data, it can be estimated that >50 per cent of placental oxygen consumption is for protein synthesis and cation pumping, similar to other metabolically active tissues. As discussed further by Schneider (2000), these functions are cut back when the oxygen supply is reduced, perhaps in favour of preserving structural integrity. This occurs in most in vitro settings, because only a small amount of oxygen can be dissolved in aqueous solution compared to that bound in the blood. Increasing the P2 of perfusion fluids much above physiological levels is counterproductive. It may further depress the metabolism and lead to tissue damage. We do not fully comprehend how placental metabolism impacts fetal oxygen consumption when the oxygen supply is reduced in an in vivo setting. If placental oxygen uptake is reduced in favour of the fetus, as it is in fetal growth restriction, what is the trade off in terms of reduced placental function, including secretion of protein hormones into the maternal circulation? Clarification of these important issues is worthy of a renewed effort to understand the metabolism of the placenta. REFERENCES Ball KT, Gunn TR, Power GG, Asakura H & Gluckman PD (1995) A potential role for adenosine in the inhibition of nonshivering thermogenesis in the fetal sheep. Pediatr Res, 37, 303–309.
Placenta (2000), Vol. 21, Supplement A, Trophoblast Research, Vol. 14 Battaglia FC, Meschia G, Makowski EL & Bowes W (1968) The effect of maternal oxygen inhalation upon fetal oxygenation. J Clin Invest, 47, 548–555. Bell AW, Kennaugh JM, Battaglia FC, Makowski EL & Meschia G (1986) Metabolic and circulatory studies of fetal lamb at midgestation. Am J Physiol, 250, E538–E544. Birdsey TJ, Boyd RDH, Sibley CP & Greenwood SL (1997) Microvillous membrane potential (Em) in villi from first trimester human placenta: comparison to Em at term. Am J Physiol, 273, R1519–R1528. Bonds DR, Crosby LO, Cheek TG, Ha¨gerdal M, Gutsche BB & Gabbe G (1986) Estimation of human fetal-placental unit metabolic rate by application of the Bohr principle. J Dev Physiol, 8, 49–54. Buttery PJ, Beckerton A, Mitchell RM, Davies K & Annison CG (1975) The turnover rate of muscle and liver protein in sheep. Proc Nutri Soc, 34, 91A–92A. Campbell AGM, Dawes GS, Fishman AP, Hyman AI & James GB (1966) The oxygen consumption of the placenta and foetal membranes in the sheep. J Physiol, 182, 439–464. Carroll MJ & Young M (1987) Observations on the energy and redox state and protein synthetic rate in animal and human placentas. J Perinat Med, 15, 21–30. Carter AM (1989) Factors affecting gas transfer across the placenta and the oxygen supply to the fetus. J Dev Physiol, 12, 305–322. Challier J-C, Schneider H & Dancis J (1976) In vitro perfusion of human placenta. V. Oxygen consumption. Am J Obstet Gynecol, 126, 261–265. Chard T (1986) Placental synthesis. Clin Obstet Gynaecol, 13, 447–467. Christensen P, Grønlund J & Carter AM (1986) Placental gas exchange in the guinea-pig: fetal blood gas tensions following the reduction of maternal oxygen capacity with carbon monoxide. J Dev Physiol, 8, 1–9. Clausen T, van Hardeveld C & Everts ME (1991) Significance of cation transport in control of energy metabolism and thermogenesis. Physiol Rev, 71, 733–774. Diamant YZ, Mayorek N, Neumann S & Shafrir E (1975) Enzymes of glucose and fatty acid metabolism in early and term human placenta. Am J Obstet Gynecol, 121, 58–61. Edelstone DI, Caine ME & Fumia FD (1985) Relationship of fetal oxygen consumption and acid-base balance to fetal hematocrit. Am J Obstet Gynecol, 151, 844–851. Edelstone DI, Darby MJ, Bass K & Miller K (1989) Effects of reductions in hemoglobin-oxygen affinity and hematocrit level on oxygen consumption and acid-base state in fetal lambs. Am J Obstet Gynecol, 160, 820–828. Edwards EM, Rattenbury J, Varnam GC, Dhand UK, Jeacock MK & Shepherd DA (1977) Enzyme activities in the sheep placenta during the last three months of pregnancy. Biochim Biophys Acta, 497, 133–143. Frankenne F, Closset J, Gomez F, Scippo ML, Smal J & Hennen G (1988) The physiology of growth hormones (GHs) in pregnant women and partial characterization of the placental GH variant. J Clin Endocrinol Metabol, 66, 1171–1180. Grumbach MM, Kaplan SL, Sciarra JJ & Burr IM (1968) Chorionic growth hormone-prolactin (CGP): secretion, disposition, biologic activity in man, and postulated function as the ‘growth hormone’ of the second half of pregnancy. Ann NY Acad Sci, 148, 501–531. Gu W, Jones CT & Parer JT (1985) Metabolic and cardiovascular effects on fetal sheep of sustained reduction of uterine blood flow. J Physiol, 368, 109–129. James WPT, Garlick PJ, Sender PM & Waterlow JC (1976) Studies of amino acid and protein metabolism in normal man with -[U-14C]tyrosine. Clin Sci Mol Med, 50, 525–532. Kay HH, Robinette B, Shin YY, Siew P, Shellhaas CS & Tyrey L (1997) Placental villous glucose metabolism and hormone release respond to varying oxygen tensions. J Soc Gynecol Invest, 4, 241–246. Kingdom JCP & Kaufmann P (1997) Oxygen and placental villous development: origins of fetal hypoxia. Placenta, 18, 613–621. Lawrence JB, Oxvig C, Overgaard MT, Sottrup-Jensen L, Gleich GJ, Hays LG, Yates JR & Conover CA (1999) The insulin-like growth factor (IGF)-dependent IGF binding protein-4 protease secreted by human fibroblasts is pregnancy-associated plasma protein-A. Proc Natl Acad Sci USA, 96, 3149–3153. Longo LD (1987) Respiratory gas exchange in the placenta. In Handbook of Physiol. Section 3. The Respiratory System. Volume IV. Gas Exchange, pp. 351–401. Bethesda: American Physiological Society. Longo LD, Hill EP & Power GG (1972) Theoretical analysis of factors affecting placental O2 transfer. Am J Physiol, 222, 730–739.
Carter: Placental Oxygen Consumption (Part 1) Meschia G, Battaglia FC, Hay WW & Sparks JW (1980) Utilization of substrates by the ovine placenta in vivo. Federation Proc, 39, 245–249. Owens JA, Falconer J & Robinson JS (1987a) Effect of restriction of placental growth on oxygen delivery to and consumption by the pregnant uterus and fetus. J Dev Physiol, 9, 137–150. Owens JA, Falconer J & Robinson JS (1987b) Effect of restriction of placental growth on fetal and utero-placental metabolism. J Dev Physiol, 9, 225–238. Paulone ME, Edelstone DI & Shedd A (1987) Effects of maternal anemia on uteroplacental and fetal oxidative metabolism in sheep. Am J Obstet Gynecol, 156, 230–236. Prendergast CH, Parker KH, Gray R, Venkatesan S, Bannister P, Castro-Soares J, Murphy KW, Beard RW, Regan L, Robinson S, Steer P, Halliday D & Johnston DG (1999) Glucose production by the human placenta in vivo. Placenta, 20, 591–598. Rodesch F, Simon P, Donner C & Jauniaux E (1992) Oxygen measurements in endometrial and trophoblastic tissues during early pregnancy. Obstet Gynecol, 80, 283–285. Rolfe DFS & Brown GC (1997) Cellular energy utilization and molecular origin of standard metabolic rate in mammals. Physiol Rev, 77, 731–758. Schneider H (2000) Placental Oxygen Consumption. Part II, in vitro studies—A review. Placenta, 21 (Supp. A), 38–44. Schneider H, Malek A, Duft R & Bersinger N (1989) Evaluation of an in vitro dual perfusion system for the study of placental proteins: energy metabolism. In Placenta as a Model and Source (Eds) Genbacev O, Klopper A & Beaconsfield R, pp. 39–50. New York: Plenum Press.
S37 Sender PM & Garlick PJ (1973) Synthesis rates of protein in the Langendorff-perfused rat heart in the presence and absence of insulin, and in the working heart. Biochem J, 132, 603–608. Sørensen S, Momsen G, Ruge S & Pedersen JF (1995) Differential increase in the maternal serum concentrations of the placental proteins human chorionic gonadotrophin, pregnancy-specific 1-glycoprotein, human placental lactogen and pregnancy-associated plasma protein-A during the first half of normal pregnancy, elucidated by means of a mathematical model. Hum Reprod, 10, 453–458. Thorburn GD (1995) The placenta and the control of fetal breathing movements. Reprod Fertil Dev, 7, 577–594. Vatnick I & Bell AW (1992) Ontogeny of fetal hepatic and placental growth and metabolism in sheep. Am J Physiol, 263, R619–R623. Watson A, Palmer M, Jauniaux E & Burton G (1998) Evidence for oxygen-derived free radical mediated damage to first trimester human trophoblast in vitro. Trophoblast Research, 11, 259–276. Widdowson EM & Spray CM (1951) Chemical development in utero. Arch Dis Childhood, 26, 205–214. Wilkening RB & Meschia G (1983) Fetal oxygen uptake, oxygenation, and acid-base balance as a function of uterine blood flow. Am J Physiol, 244, H749–H755. Young M, Stern MDR, Horn J & Noakes DE (1982) Protein synthetic rate in the sheep placenta in vivo: the influence of insulin. Placenta, 3, 159–164.
CELL BIOLOGY Placental Oxygen Consumption. Part I: In Vivo Studies—A Review A. M. Cartera Department of Physiology and Pharmacology, University of Southern Denmark, Winsloewparken 21, DK-5000 Odense, Denmark Paper accepted 6 December 1999
At term of pregnancy, oxygen consumption by the human or ovine placenta accounts for 40 per cent of total oxygen uptake by the gravid uterus. In the sheep, most oxygen is used for oxidative phosphorylation of glucose; the remainder is probably utilized for non-mitochondrial processes. The ATP yield is expended mainly in protein synthesis and cation transport. The fractional protein synthesis rate of ovine placenta is 60 per cent per day. Applying these data to man, protein synthesis is estimated to account for about 30 per cent of placental oxygen uptake. Probably this reflects the high rates of synthesis of peptide and steroid hormones. The Na + gradient is the basis for secondary active transport of amino acids and other substances, and the Na + –K + -pump probably accounts for 20–30 per cent of oxygen uptake, with a smaller contribution from Ca2+ -ATPase. Placental oxygen uptake remains constant during acute reductions in uterine oxygen supply and is maintained at the expense of the fetus. In the longer term, in experimental models of fetal growth restriction, placental oxygen consumption is reduced to a greater extent than fetal oxygen consumption. Placental oxygen consumption is greatly reduced under in vitro experimental conditions, due largely to an inadequate oxygen supply. This results in reduced protein synthesis and possibly inhibition of Na + –K + -ATPase. However, if the placenta is subjected to hyperoxia, by raising the P2 of the medium, there is an increase in anaerobic glycolysis and structural damage may ensue. Premature exposure of trophoblast to high oxygen tensions in vivo may result in reduced villous branching, but this is likely to be a cause, rather than a consequence, of reduced fetal growth and oxygen consumption. 2000 IFPA and Harcourt Publishers Ltd Placenta (2000), 21, Supplement A, Trophoblast Research, 14, S31–S37
INTRODUCTION The placenta has long been known to have a high rate of oxygen consumption (Campbell et al., 1966). This review will examine why the placenta requires so much oxygen for its metabolism and discuss how placental oxygen consumption affects transfer of oxygen to the fetus. Since oxygen uptake by the placenta decreases greatly under in vitro experimental conditions, it will be asked how this impacts on placental metabolism and function. This entails consideration of the practice of maintaining culture media and perfusion fluids at high levels of P2. Finally, since hyperoxia in vitro has been shown to alter placental metabolism and cause tissue damage, it will be asked if comparable conditions can arise in disease states. In examining the relationship between placental oxygen a
To whom correspondence should be addressed at: Department of Physiology and Pharmacology, Winsloewparken 21, 3rd Floor, DK-5000 Odense, Denmark. Fax: +45 6613 3479; E-mail: [email protected]
0143–4004/00/0A00S31+07 $35.00/0
consumption and metabolism, the approach adopted is that developed by Rolfe and Brown (1997) in their perspicuous analysis of standard metabolic rate.
PLACENTAL OXYGEN UPTAKE AND METABOLISM In sheep, uteroplacental oxygen uptake is calculated as the difference between total oxygen uptake by the gravid uterus and oxygen uptake by the fetus. This requires simultaneous measurement of the uterine and umbilical blood flows and of the arteriovenous differences in oxygen content across the two circulations (Meschia et al., 1980). In unanaesthetized sheep near term of pregnancy (term is about 147 days), uteroplacental oxygen uptake is 0.9 mmol/min (20 ml/min), or about 40 per cent of total uterine oxygen uptake (Meschia et al., 1980). In mid-pregnant sheep (71–81 days gestation), fetal oxygen consumption is smaller and uteroplacental oxygen uptake 2000 IFPA and Harcourt Publishers Ltd
Placenta (2000), Vol. 21, Supplement A, Trophoblast Research, Vol. 14
S32
accounts for more than 80 per cent of the total (Bell et al., 1986). To determine the oxygen consumption of the human placenta, Bonds et al. (1986) used indirect calorimetry to measure oxygen uptake in women undergoing elective caesarian section. Determinations were made before and after cord clamping and after removal of the placenta. This yielded values for fetal and placental oxygen uptake of 0.98 and 0.58 mmol/min, respectively (22 and 13 ml/min). Thus, as in the sheep, oxygen consumption by the human placenta is about 40 per cent of total uterine oxygen uptake.
Protein synthesis
Mitochondrial
ATP synthesis
Placental ATP synthesis
Other
Meschia et al. (1980) determined oxygen consumption, glucose utilization and lactate production by the ovine uterus and placenta. Glucose utilization was 0.205 mmol/min and total lactate production was 0.150 mmol/min. The overall reaction for anaerobic glycolysis is: Glucose+2Pi +2ADP<2 lactate+2ATP+2 H2O
Na+-K+ATPase
(1)
Thus the amount of glucose available for oxidative metabolism was 0.205(0.150/2)=0.130 mmol/min. The overall reaction for oxidative phosphorylation is: Glucose+31 ADP+31 Pi +31 H + +6O2<6 CO2 +31 (2) ATP+37 H2O Therefore, oxidative phosphorylation of 0.130 mmol/min glucose can account for 0.780 mmol/min oxygen or 87 per cent of the measured oxygen consumption, which was 0.896 mmol/ min. In the standard state, about 10 per cent of the body’s total oxygen uptake is non-mitochondrial and not, therefore, associated with ATP production (Rolfe and Brown, 1997). This fraction varies between tissues, reaching 20 per cent in the liver, and may well exceed 10 per cent in the placenta. Thus, oxidative metabolism of glucose can account for virtually all of the mitochondrial fraction of oxygen consumption in the ovine placenta at term. It was once thought that the reaction in equation (2) gave a net yield of 36 moles ATP for each mole of glucose utilized. However, the mechanistic ratio of ATP produced to oxygen consumed is now thought to be 2.6 rather than 3, with a yield of 30–32 moles ATP per mole of glucose (Rolfe and Brown, 1997). Additionally, under physiological conditions, glycolysis is partly uncoupled from ATP synthesis due to passive leakage of protons from the mitochondria (Rolfe and Brown, 1997). Estimates for the liver indicate that the proton leak amounts to 15 per cent of mitochondrial oxygen consumption. Therefore, the effective ratio of ATP to oxygen is closer to 2.2. Applied to the ovine placenta, this gives an approximate ATP synthesis rate of 3.4 mmol/min. In extrapolating to human placenta, it should be borne in mind that placental handling of glucose could differ from that
Nonmitochondrial
Proton leak
Figure 1. The estimated contribution of metabolic processes to placental energy consumption. The first column shows contribution of mitochondrial and non-mitochondrial oxygen consumption to total placental respiration. The second column shows the proportion of mitochondrial respiration used to drive ATP synthesis and proton leak and is based on estimates for whole body metabolism. The third column represents the contribution of protein synthesis and cation transport to total ATP consumption. Analysis based on Rolfe and Brown (1997).
in sheep. The weight specific oxygen consumption of human placenta is similar to that of the ovine placenta (37 versus 34 ml/min per kg; Bonds et al., 1986; Meschia et al., 1980), but the tissue mass is smaller. ATP synthesis by a 400 g human placenta should therefore be in the order of 2 mmol/min. How does the placenta dispose of this energy? ATP consumption is probably for protein synthesis and cation transport (Figure 1). The processes subserved by this energy consumption include: growth; transport and maintenance of ion gradients; and endocrine functions.
Contribution of protein synthesis to placental oxygen consumption The rate of protein synthesis by the human body is 347 g/day (James et al., 1976), accounting for 25–30 per cent of the ATP utilized under standard conditions (Rolfe and Brown, 1997). Protein turnover in individual tissues can be calculated by infusing a labelled amino acid, such as tyrosine, until the plasma concentration reaches a plateau, and measuring the rate at which the amino acid is incorporated into protein (Sender and Garlick, 1973). The fractional protein synthesis rate of the ovine placenta, determined in vivo, was 60 per cent per day (Young et al., 1982). This is five times faster than the rate determined for the liver of adult sheep (Buttery et al., 1975). Human placenta in vitro, perfused on the maternal side with diluted blood (haematocrit 20–25 per cent) at physiological P2, had a protein synthesis rate of 40 per cent per day (Carroll and Young, 1987).
Carter: Placental Oxygen Consumption (Part 1)
S33
These values suggest that protein synthesis accounts for a substantial fraction of placental oxygen consumption. Assuming a placental weight of 400 g (Bonds et al., 1986), of which 12 per cent is protein (Widdowson and Spray, 1951), the rate of protein synthesis is 29 g/day (at a protein synthesis rate of 60 per cent per day). Taking the average molecular weight per peptide bond to be 110 g/mol (Rolfe and Brown, 1997), this amounts to 0.18 mmol/min. The equivalent oxygen consumption can be derived by multiplying the protein synthesis rate by the ATP requirement (four ATP equivalents per peptide bond) and dividing by 2.2, the effective ratio of ATP synthesized to oxygen consumed (see above). A value of 0.17 mmol/min O2 or 3.7 ml/min is found. This is approximately 30 per cent of the total oxygen consumption of a 400 g placenta (13 ml/min; Bonds et al., 1986). The placenta synthesizes a wide range of structural proteins and enzymes and a range of products with putative endocrine, metabolic, haemostatic and immunological functions (Chard, 1986). Placental proteins are secreted primarily to the maternal circulation and their effects are probably confined to the mother. They include human chorionic gonadotrophin (hCG), human placental lactogen (hPL), pregnancy-associated placental protein-A [PAPP-A; insulin-like growth factor binding protein-4 protease (Lawrence et al., 1999)], pregnancy-specific 1-glycoprotein (PSG) and placental growth hormone (Chard, 1986; Sørensen et al., 1995; Frankenne et al., 1988). Secretion of hCG peaks at 8–10 weeks of pregnancy, but secretion of other proteins continues to rise (Sørensen et al., 1995). At 17 weeks, daily secretion of hPL, PAPP-A and PSG amounts to about 350 mg (Sørensen et al., 1995). The rate of secretion of hPL towards term, about 1 g/day, exceeds that of any other human peptide hormone (Grumbach et al., 1968). The placenta also secretes quantities of steroids, principally oestriol and progesterone, into the fetal and maternal circulations. Other placental products that probably have far reaching effects on fetal physiology are prostaglandin E2 (Thorburn, 1995) and adenosine (Ball et al., 1995).
for 20–30 per cent of total oxygen uptake in fetal placenta and 18–24 per cent in maternal placenta. However, the oxygen uptake of placenta in these experiments in vitro was only 15 per cent of in vivo values (Vatnick and Bell, 1992) and it is not known whether Na + –K + -ATPase activity was reduced proportionately. The energy cost of Ca2+ cycling by the Ca2+ -ATPase of the sarcoplasmic reticulum of skeletal muscle accounts for 20– 50 per cent of the oxygen consumption during contraction and ca. 5 per cent in resting muscle cells. In the brain, where Ca2+ serves as a signal for neurotransmitter release, Ca2+ cycling across the plasma membrane also contributes significantly to energy turnover. The energy cost of maintaining Ca2+ gradients in the placenta is not known, but probably does not exceed that in the liver, where it accounts for 1–10 per cent of energy consumption (Rolfe and Brown, 1997).
Contribution of cation transport to placental oxygen consumption
Reactions that uncouple metabolism
Cation transport makes a significant contribution to basal oxygen consumption. It has been calculated that oxygen consumption by the Na + –K + -pump accounts for about 20 per cent of the basal metabolic rate (Clausen, van Hardeveld and Everts, 1991). The Na + gradient is the basis for secondary active transport of amino acids and other substances via co-transport systems. Active Na + –K + transport is, for example, essential for the absorptive function of the intestine, where it answers for a substantial fraction of total energy turnover (ca. 30 per cent). An indication of the contribution of Na + –K + -ATPase to placental energy utilization is given by the ouabain-sensitive oxygen uptake. This was determined for ovine placental tissues at three stages of gestation by Vatnick and Bell (1992). Ouabain-sensitive oxygen uptake accounted
Contribution of other reactions to placental oxygen consumption The standard metabolic rate of the whole body receives major contributions from gluconeogenesis, ureagenesis, actinomyosin ATPase (chiefly in cardiac muscle), substrate recycling and RNA synthesis (Rolfe and Brown, 1997). Gluconeogenesis makes a substantial contribution to the metabolic rate and may account for as much as 30 per cent of oxygen consumption by the human liver. Human placenta certainly exhibits phosphoenolpyruvate carboxykinase activity (Diamant et al., 1975) and may produce glucose (Prendergast et al., 1999), but there are no quantitative estimates of placental gluconeogenesis. They may be hard to come by experimentally, as phosphoenolpyruvate carboxykinase activity cannot be detected in the sheep placenta (Edwards et al., 1977). Neither human nor ovine placenta is capable of ureagenesis (Diamant et al., 1975; Edwards et al., 1977) and the activity of actinomyosin ATPase can be discounted in this organ.
Cellular energy metabolism is uncoupled by processes in the same cell or elsewhere in the body; if it were not, metabolism would cease. Protein synthesis is uncoupled by protein degradation, the Na + pump by Na + channels and Na + -coupled transport, the Ca2+ pump by Ca2+ channels and gluconeogenesis by glycolysis.
OXYGEN TRANSFER TO THE FETUS Factors that determine oxygen transfer across the placenta include: the oxygen affinity and oxygen capacity (haemoglobin concentration) of maternal and fetal blood; rates of blood flow in the uterine and umbilical circulations; the permeability and surface area of the placental barrier; and placental oxygen
Fetal O2 consumption (ml O2 /min/kg)
consumption (reviewed in Carter, 1989; Longo, 1987). To a great extent, fetal oxygen consumption can be maintained despite fluctuations in oxygen delivery. Thus, fetal and maternal placental blood flows and blood oxygen capacities can be altered by as much as 50 per cent without any major change occurring in fetal oxygen uptake. The oxygen content of umbilical venous blood will fall, and with it oxygen delivery to the fetus, but fetal oxygen consumption will be maintained by increasing the fractional extraction of oxygen from the blood. The resultant fall in oxygen content of the umbilical arterial blood will maintain the arteriovenous difference across the umbilical circulation and placental oxygen transfer will remain the same. This has been verified in a wide range of experimental settings, supporting theoretical predictions made by Longo et al. (1972) on the assumption that compensatory mechanisms would intervene to maintain fetal oxygen consumption. There is, however, a critical threshold below which fetal oxygen uptake becomes dependent on oxygen delivery. In the sheep this corresponds to an oxygen delivery of about 0.6 mmol/min per kg fetal weight (13 ml/min per kg). Fetal oxygen uptake tends to fall when oxygen delivery is reduced below this level (Figure 2). This can be shown by decreasing maternal placental blood flow (Gu et al., 1985; Wilkening and Meschia, 1983); raising or lowering fetal haematocrit (Edelstone et al., 1985); decreasing maternal oxygen capacity (Christensen et al., 1986; Paulone et al., 1987); or decreasing fetal oxygen affinity (Edelstone et al., 1989). The resultant tissue hypoxia and inability to maintain oxidative metabolism is reflected in a lowering of fetal arterial pH and base excess. Clearly, if the uterine oxygen supply is limited, the large oxygen consumption of the placenta will impose a limitation on oxygen transfer to the fetus. Placental oxygen consumption remains remarkably constant following acute alterations in uterine oxygen supply. This is true both when the oxygen supply is reduced, by partial occlusion of the uterine artery (Gu et al., 1985; Wilkening and Meschia, 1983), and when it is increased, by raising the oxygen saturation of maternal arterial blood (Battaglia et al., 1968). Thus, reduction of uterine blood flow by 30–50 per cent for 60 min depressed fetal oxygen uptake, because oxygen consumption by the placenta was maintained at the expense of the fetus [Figure 3(A); Gu et al., 1985]. At the same time, fetal glycogen stores were mobilized to supply the placenta with glucose and lactate (Gu et al., 1985). Fetal supply of nutrients to the placenta cannot be sustained indefinitely. Thus, some form of adaptation can be expected to occur during long-term reductions in uterine blood flow. In fetal growth restriction following carunclectomy (Owens et al., 1987a), where the supply of oxygen and nutrients was restricted throughout gestation, uteroplacental oxygen consumption was reduced to a greater extent than fetal oxygen consumption [Figure 3(B)]. Consequently, it accounted for a smaller proportion of total uterine oxygen uptake (28 per cent) than during normal growth (53 per cent). Uteroplacental consumption of glucose was also reduced compared to normal growth controls (Owens et al., 1987b).
Placenta (2000), Vol. 21, Supplement A, Trophoblast Research, Vol. 14
10
5
0 10
Base excess (meq/l)
S34
0
–10
–20
0
10
20
30
Fetal O2 delivery (ml O2 /min/kg) Figure 2. Fetal oxygen consumption and arterial base excess as a function of oxygen delivery in sheep. Fetal oxygen uptake remained constant until oxygen delivery was reduced below ca. 13 ml/min per kg fetal weight (0.6 mmol/min per kg). There was then an increase in anaerobic metabolism with a resultant fall in arterial base excess. Data are from experiments in which both oxygen affinity (haemoglobin type) and oxygen capacity (haemoglobin concentration) were varied. Circles: fetal haemoglobin; squares: adult haemoglobin; filled circles and squares: normal haematocrit; half-filled circles and squares: moderate anaemia; open circles and squares: severe anaemia. Reproduced with permission from Edelstone, Darby, Bass & Miller (1989) Effects of reductions in hemoglobin-oxygen affinity and hematocrit level on oxygen consumption and acid-base state in fetal lambs. American Journal of Obstetrics and Gynecology, 160, 820–828.
PLACENTAL OXYGEN CONSUMPTION IN VITRO The high rates of oxygen consumption measured in vivo usually are not maintained under in vitro experimental conditions. Indeed, oxygen consumption by the placenta is greatly reduced, largely because it does not receive an adequate oxygen supply. As shown by Challier, Schneider and Dancis (1976), and detailed in this volume (Schneider, 2000), oxygen uptake by the perfused placental cotyledon is linearly related to oxygen delivery. In their experiments, the preparation was first perfused on the fetal or maternal side, or both, with buffered salt solution. The highest values for oxygen supply and oxygen consumption were obtained when the placenta was perfused on both sides with buffer equilibrated with 95 per cent oxygen and 5 per cent carbon dioxide. However, even under these
Carter: Placental Oxygen Consumption (Part 1)
S35
A 1600
1200
Oxygen uptake (µmol/min)
800
400
0 B 2400
Fetus
Placenta
Control
Fetus
Placenta
Arterial occlusion
2000 1600 1200 800 400 0
Fetus
Placenta
Normal growth
Fetus
Placenta
Growth restricted
Figure 3. (A) Effect of an acute 60 min reduction in uterine blood flow on oxygen uptake of the ovine fetus and uteroplacenta. Placental oxygen consumption is maintained at the expense of the fetus. Data from Gu et al. (1985). (B) Effect of long-term restriction in uterine oxygen and substrate supply, following carunclectomy, on oxygen uptake of the ovine fetus and uteroplacenta. Placental oxygen consumption is reduced to a greater extent than that of the fetus. Data from Owens et al. (1987a).
conditions, the supply of oxygen was much less than it would be in vivo. The amount of oxygen dissolved in aqueous solution depends on the solubility coefficient () and is about 23 ml/l solution (at 760 mm Hg and 37C). A far greater quantity of oxygen can be carried in the blood, due to the high oxygen affinity of haemoglobin. Thus blood with a haemoglobin concentration of 150 g/l can carry 200 ml/l oxygen as oxyhaemoglobin. It can be estimated that the placental oxygen supply under physiological conditions is in the order of 160 ml/min/ kg placenta. In experiments where the cotyledon was perfused either with whole blood or with red blood cells suspended in buffer, equilibrated with room air or 30 per cent oxygen, placental oxygen consumption was much higher than in those where the perfusate was buffered salt solution (Challier, Schneider and Dancis, 1976; Schneider, 2000). What does this mean in terms of placental metabolism? Glucose uptake was greater when the placenta was perfused with blood equilibrated with air than when the perfusate was buffered salt solution equilibrated with 95 per cent oxygen (Schneider et al., 1989). Moreover, the amount of glucose converted to lactate was 22 per cent when the placenta was
perfused with blood, about what would be expected by extrapolation from the sheep studies, but 77 per cent of glucose uptake when the placenta was perfused with buffer. As we have seen, it is not possible to achieve physiological levels of oxygen delivery with buffer as the perfusate, and this explains the higher rate of anaerobic glycolysis. In a third group of experiments, the buffer was initially equilibrated with 95 per cent nitrogen and 5 per cent carbon dioxide although, because of the gas permeability of the perfusion system, the measured P2 of the solution was 34 mmHg. These placentae had a higher rate of oxygen uptake than those perfused with buffer equilibrated with 95 per cent oxygen, suggesting that hyperoxia has a deleterious effect on placental metabolism (Schneider et al., 1989). How do these metabolic changes affect placental function? It seems inevitable that placental protein synthesis will decrease as a consequence of reduced oxidative metabolism and, with it, the rate of secretion of steroids and protein hormones. There is a marked increase in production of hPL and hCG when the placenta is perfused with diluted blood (haematocrit 22 per cent) compared to buffered salt solution (Schneider, 2000). Furthermore, hCG production by villous fragments perifused with buffer and tissue culture medium has been shown to vary with oxygen tension (Kay et al., 1997). The highest rate of secretion was associated with 20 per cent oxygen (measured P2 167 mmHg). A significantly lower secretion rate was observed under hyperoxic conditions, following equilibration of the medium with 95 per cent oxygen (measured P2 543 mmHg). There is indirect evidence that a reduced oxygen supply affects the activity of the sodium pump. Ouabain decreased the oxygen consumption of first trimester villous fragments by 22 per cent (Birdsey et al., 1997), much as would be expected by extrapolation from the sheep data (Vatnick and Bell, 1992). However, ouabain was without effect on the oxygen consumption of term villous fragments (Birdsey et al., 1997). In these experiments, the culture medium was gassed with 5 per cent CO2 in air, and thus had an acceptable P2, but a low oxygen content. Perhaps the placenta is able to shut down the sodium pump under these conditions.
PLACENTAL ‘HYPEROXIA’ In the context of in vitro experiments, raising P2 achieves little in the way of improving the oxygen supply, whilst non-physiological levels of oxygen tension may impair metabolism and cause tissue damage. First trimester syncytiotrophoblast is especially susceptible to hyperoxia. Following just 4 h of culture with 21 per cent oxygen, the syncytiotrophoblast of 9 week old placental tissue showed mitochondrial swelling and loss of microvilli (Watson et al., 1998). Can the placenta experience hyperoxia under in vivo conditions? It has been suggested that this occurs in some pathological pregnancies and is the root cause of defective villous branching (Kingdom and Kaufmann, 1997). In fetal
S36
growth restriction with absent end diastolic flow, for example, it was suggested that failure of the fetus to extract oxygen from the intervillous space leads to an elevation in the P2 to which the villi are exposed and removes the hypoxic drive for normal branching angiogenesis (Kingdom and Kaufmann, 1997). However, the P2 of maternal arterial blood does not exceed 100 mmHg and the normal intervillous P2 may be close to 60 mmHg (Rodesch et al., 1992). It is difficult to conceive of a change in P2 within this narrow range being sufficient to halt villous development. It could be argued that villous branching occurs quite early in placental development, in the second trimester, when intervillous P2 is lower, possibly <20 mm Hg (Rodesch et al., 1992). However, fetal consumption of oxygen at this time is much smaller. In sheep, for example, the fetal share of uterine oxygen uptake at mid gestation is only 15–20 per cent of the total (Bell et al., 1986). It seems unlikely that a reduction in what is already a small share of uterine oxygen uptake would alter trophoblast P2 sufficiently to affect angiogenesis. If premature exposure of trophoblast to high oxygen tensions does lead to reduced villous branching, this is more likely to be a cause than a consequence of reduced fetal growth. CONCLUSIONS Surprisingly little is known about the metabolism of the placenta. Most of the oxygen consumed is likely to be used for oxidative phosphorylation and the effective ratio of ATP generated to oxygen consumed is probably about 2.2, as in other tissues. There is no reason to believe that uncoupling occurs to a greater extent in the placenta, but for a dissenting view see Schneider (2000). Based largely on sheep data, it can be estimated that >50 per cent of placental oxygen consumption is for protein synthesis and cation pumping, similar to other metabolically active tissues. As discussed further by Schneider (2000), these functions are cut back when the oxygen supply is reduced, perhaps in favour of preserving structural integrity. This occurs in most in vitro settings, because only a small amount of oxygen can be dissolved in aqueous solution compared to that bound in the blood. Increasing the P2 of perfusion fluids much above physiological levels is counterproductive. It may further depress the metabolism and lead to tissue damage. We do not fully comprehend how placental metabolism impacts fetal oxygen consumption when the oxygen supply is reduced in an in vivo setting. If placental oxygen uptake is reduced in favour of the fetus, as it is in fetal growth restriction, what is the trade off in terms of reduced placental function, including secretion of protein hormones into the maternal circulation? Clarification of these important issues is worthy of a renewed effort to understand the metabolism of the placenta. REFERENCES Ball KT, Gunn TR, Power GG, Asakura H & Gluckman PD (1995) A potential role for adenosine in the inhibition of nonshivering thermogenesis in the fetal sheep. Pediatr Res, 37, 303–309.
Placenta (2000), Vol. 21, Supplement A, Trophoblast Research, Vol. 14 Battaglia FC, Meschia G, Makowski EL & Bowes W (1968) The effect of maternal oxygen inhalation upon fetal oxygenation. J Clin Invest, 47, 548–555. Bell AW, Kennaugh JM, Battaglia FC, Makowski EL & Meschia G (1986) Metabolic and circulatory studies of fetal lamb at midgestation. Am J Physiol, 250, E538–E544. Birdsey TJ, Boyd RDH, Sibley CP & Greenwood SL (1997) Microvillous membrane potential (Em) in villi from first trimester human placenta: comparison to Em at term. Am J Physiol, 273, R1519–R1528. Bonds DR, Crosby LO, Cheek TG, Ha¨gerdal M, Gutsche BB & Gabbe G (1986) Estimation of human fetal-placental unit metabolic rate by application of the Bohr principle. J Dev Physiol, 8, 49–54. Buttery PJ, Beckerton A, Mitchell RM, Davies K & Annison CG (1975) The turnover rate of muscle and liver protein in sheep. Proc Nutri Soc, 34, 91A–92A. Campbell AGM, Dawes GS, Fishman AP, Hyman AI & James GB (1966) The oxygen consumption of the placenta and foetal membranes in the sheep. J Physiol, 182, 439–464. Carroll MJ & Young M (1987) Observations on the energy and redox state and protein synthetic rate in animal and human placentas. J Perinat Med, 15, 21–30. Carter AM (1989) Factors affecting gas transfer across the placenta and the oxygen supply to the fetus. J Dev Physiol, 12, 305–322. Challier J-C, Schneider H & Dancis J (1976) In vitro perfusion of human placenta. V. Oxygen consumption. Am J Obstet Gynecol, 126, 261–265. Chard T (1986) Placental synthesis. Clin Obstet Gynaecol, 13, 447–467. Christensen P, Grønlund J & Carter AM (1986) Placental gas exchange in the guinea-pig: fetal blood gas tensions following the reduction of maternal oxygen capacity with carbon monoxide. J Dev Physiol, 8, 1–9. Clausen T, van Hardeveld C & Everts ME (1991) Significance of cation transport in control of energy metabolism and thermogenesis. Physiol Rev, 71, 733–774. Diamant YZ, Mayorek N, Neumann S & Shafrir E (1975) Enzymes of glucose and fatty acid metabolism in early and term human placenta. Am J Obstet Gynecol, 121, 58–61. Edelstone DI, Caine ME & Fumia FD (1985) Relationship of fetal oxygen consumption and acid-base balance to fetal hematocrit. Am J Obstet Gynecol, 151, 844–851. Edelstone DI, Darby MJ, Bass K & Miller K (1989) Effects of reductions in hemoglobin-oxygen affinity and hematocrit level on oxygen consumption and acid-base state in fetal lambs. Am J Obstet Gynecol, 160, 820–828. Edwards EM, Rattenbury J, Varnam GC, Dhand UK, Jeacock MK & Shepherd DA (1977) Enzyme activities in the sheep placenta during the last three months of pregnancy. Biochim Biophys Acta, 497, 133–143. Frankenne F, Closset J, Gomez F, Scippo ML, Smal J & Hennen G (1988) The physiology of growth hormones (GHs) in pregnant women and partial characterization of the placental GH variant. J Clin Endocrinol Metabol, 66, 1171–1180. Grumbach MM, Kaplan SL, Sciarra JJ & Burr IM (1968) Chorionic growth hormone-prolactin (CGP): secretion, disposition, biologic activity in man, and postulated function as the ‘growth hormone’ of the second half of pregnancy. Ann NY Acad Sci, 148, 501–531. Gu W, Jones CT & Parer JT (1985) Metabolic and cardiovascular effects on fetal sheep of sustained reduction of uterine blood flow. J Physiol, 368, 109–129. James WPT, Garlick PJ, Sender PM & Waterlow JC (1976) Studies of amino acid and protein metabolism in normal man with -[U-14C]tyrosine. Clin Sci Mol Med, 50, 525–532. Kay HH, Robinette B, Shin YY, Siew P, Shellhaas CS & Tyrey L (1997) Placental villous glucose metabolism and hormone release respond to varying oxygen tensions. J Soc Gynecol Invest, 4, 241–246. Kingdom JCP & Kaufmann P (1997) Oxygen and placental villous development: origins of fetal hypoxia. Placenta, 18, 613–621. Lawrence JB, Oxvig C, Overgaard MT, Sottrup-Jensen L, Gleich GJ, Hays LG, Yates JR & Conover CA (1999) The insulin-like growth factor (IGF)-dependent IGF binding protein-4 protease secreted by human fibroblasts is pregnancy-associated plasma protein-A. Proc Natl Acad Sci USA, 96, 3149–3153. Longo LD (1987) Respiratory gas exchange in the placenta. In Handbook of Physiol. Section 3. The Respiratory System. Volume IV. Gas Exchange, pp. 351–401. Bethesda: American Physiological Society. Longo LD, Hill EP & Power GG (1972) Theoretical analysis of factors affecting placental O2 transfer. Am J Physiol, 222, 730–739.
Carter: Placental Oxygen Consumption (Part 1) Meschia G, Battaglia FC, Hay WW & Sparks JW (1980) Utilization of substrates by the ovine placenta in vivo. Federation Proc, 39, 245–249. Owens JA, Falconer J & Robinson JS (1987a) Effect of restriction of placental growth on oxygen delivery to and consumption by the pregnant uterus and fetus. J Dev Physiol, 9, 137–150. Owens JA, Falconer J & Robinson JS (1987b) Effect of restriction of placental growth on fetal and utero-placental metabolism. J Dev Physiol, 9, 225–238. Paulone ME, Edelstone DI & Shedd A (1987) Effects of maternal anemia on uteroplacental and fetal oxidative metabolism in sheep. Am J Obstet Gynecol, 156, 230–236. Prendergast CH, Parker KH, Gray R, Venkatesan S, Bannister P, Castro-Soares J, Murphy KW, Beard RW, Regan L, Robinson S, Steer P, Halliday D & Johnston DG (1999) Glucose production by the human placenta in vivo. Placenta, 20, 591–598. Rodesch F, Simon P, Donner C & Jauniaux E (1992) Oxygen measurements in endometrial and trophoblastic tissues during early pregnancy. Obstet Gynecol, 80, 283–285. Rolfe DFS & Brown GC (1997) Cellular energy utilization and molecular origin of standard metabolic rate in mammals. Physiol Rev, 77, 731–758. Schneider H (2000) Placental Oxygen Consumption. Part II, in vitro studies—A review. Placenta, 21 (Supp. A), 38–44. Schneider H, Malek A, Duft R & Bersinger N (1989) Evaluation of an in vitro dual perfusion system for the study of placental proteins: energy metabolism. In Placenta as a Model and Source (Eds) Genbacev O, Klopper A & Beaconsfield R, pp. 39–50. New York: Plenum Press.
S37 Sender PM & Garlick PJ (1973) Synthesis rates of protein in the Langendorff-perfused rat heart in the presence and absence of insulin, and in the working heart. Biochem J, 132, 603–608. Sørensen S, Momsen G, Ruge S & Pedersen JF (1995) Differential increase in the maternal serum concentrations of the placental proteins human chorionic gonadotrophin, pregnancy-specific 1-glycoprotein, human placental lactogen and pregnancy-associated plasma protein-A during the first half of normal pregnancy, elucidated by means of a mathematical model. Hum Reprod, 10, 453–458. Thorburn GD (1995) The placenta and the control of fetal breathing movements. Reprod Fertil Dev, 7, 577–594. Vatnick I & Bell AW (1992) Ontogeny of fetal hepatic and placental growth and metabolism in sheep. Am J Physiol, 263, R619–R623. Watson A, Palmer M, Jauniaux E & Burton G (1998) Evidence for oxygen-derived free radical mediated damage to first trimester human trophoblast in vitro. Trophoblast Research, 11, 259–276. Widdowson EM & Spray CM (1951) Chemical development in utero. Arch Dis Childhood, 26, 205–214. Wilkening RB & Meschia G (1983) Fetal oxygen uptake, oxygenation, and acid-base balance as a function of uterine blood flow. Am J Physiol, 244, H749–H755. Young M, Stern MDR, Horn J & Noakes DE (1982) Protein synthetic rate in the sheep placenta in vivo: the influence of insulin. Placenta, 3, 159–164.