The impact of maternal obesity on iron status, placental transferrin receptor expression and hepcidin expression in human pregnancy.

BACKGROUND: Obesity is associated with decreased iron status, possibly due to a rise in hepcidin, an inflammatory protein known to reduce iron absorption. In animals, we have shown that maternal iron deficiency is minimised in the foetus by increased expression of placental transferrin receptor (pTFR1), resulting in increased iron transfer at the expense of maternal iron stores. OBJECTIVE: This study examines the effect of obesity during pregnancy on maternal and neonatal iron status in human cohorts and whether the placenta can compensate for decreased maternal iron stores by increasing pTFR1 expression. SUBJECTS/METHODS: A total of 240 women were included in this study. One hundred and fifty-eight placentas (Normal: 90; Overweight: 37; Obese: 31) were collected at delivery. Maternal iron status was measured by determining serum transferrin receptor (sTFR) and ferritin levels at 24 and 34 weeks and at delivery. Hepcidin in maternal and cord blood was measured by ELISA and pTFR1 in placentas by western blotting and real-time RT-PCR. RESULTS: Low iron stores were more common in obese women. Hepcidin levels (ng ml(-1)) at the end of the pregnancy were higher in obese than normal women (26.03±12.95 vs 18.00±10.77, P<0.05). Maternal hepcidin levels were correlated with maternal iron status (sTFR r=0.2 P=0.025), but not with neonatal values. mRNA and protein levels of pTFR1 were both inversely related to maternal iron status. For mRNA and all women, sTFR r=0.2 P=0.044. Ferritin mRNA levels correlated only in overweight women r=-0.5 P=0.039 with hepcidin (r=0.1 P=0.349), irrespective of maternal body mass index (BMI). CONCLUSIONS: The data support the hypothesis that obese pregnant women have a greater risk of iron deficiency and that hepcidin may be a regulatory factor. Further, we show that the placenta responds to decreased maternal iron status by increasing pTFR1 expression.

The Effect of Iron Deficiency on Osmotic Sensitivity of Red Blood Cells from Neonatal Rats and Their Mothers.

Iron deficiency during pregnancy has many effects on both the mother and her developing foetus. These can be both short and long term. One effect is an alteration in fatty acid metabolism and we hypothesised that these changes may result in alterations in membrane function and structure. In order to test this hypothesis, we measured osmotic sensitivity in red blood cells isolated from neonates and their mothers at different times following birth. We fed female rats control or iron-deficient diets for 4 weeks prior to mating and kept them on the same diet until term. At that time, we returned one group of deficient dams to the control diet. The others were kept on the same diet. We showed that iron deficiency results in a decrease in osmotic sensitivity in the mothers but not in their neonates. Returning the dams to the control diet resulted in a return of their red cell osmotic sensitivity to control levels. In the neonates, there was no recovery in haematocrit or in any other parameter, though they did not get any worse, in contrast to the pups being suckled by deficient mothers. The data show two things. The first is that following birth, the mother restores her own iron stores at the expense of the pups, and secondly, there are differences in properties and sensitivities between red cells from mothers and their neonates. This latter observation cannot be explained by differences in the membrane fatty acid profiles, which were not significantly different.

The relationship between dietary supplement use in late pregnancy and birth outcomes: a cohort study in British women.

OBJECTIVE: To examine the relationship between dietary supplement use during pregnancy and birth outcomes. DESIGN: A prospective birth cohort. SETTING: Leeds, UK. SAMPLE: One thousand two hundred and seventy-four pregnant women aged 18-45 years. METHODS: Dietary supplement intake was ascertained using three questionnaires for the first, second and third trimesters. Dietary intake was reported in a 24-hour dietary recall administered by a research midwife at 8-12 weeks of gestation. Information on delivery details and antenatal pregnancy complications was obtained from the hospital maternity records. MAIN OUTCOME MEASURES: Birthweight, birth centile and preterm birth. RESULTS: Reported dietary supplement use declined from 82% of women in the first trimester of pregnancy to 22% in the second trimester and 33% in the third trimester. Folic acid was the most commonly reported supplement taken. Taking any type of daily supplement during any trimester was not significantly associated with size at birth taking into account known relevant confounders. Women taking multivitamin-mineral supplements in the third trimester were more likely to experience preterm birth (adjusted OR = 3.4, 95% CI 1.2, 9.6, P = 0.02). CONCLUSIONS: Regular multivitamin-mineral supplement use during pregnancy, in a developed country setting, is not associated with size at birth. However, it appears to be associated with preterm birth if taken daily in the third trimester. The mechanism for this is unclear and our study's findings need confirming by other cohorts and/or trials in developed countries.

In rats gestational iron deficiency does not change body fat or hepatic mitochondria in the aged offspring.

Mitochondrial dysfunction and resulting changes in adiposity have been observed in the offspring of animals fed a high fat (HF) diet. As iron is an important component of the mitochondria, we have studied the offspring of female rats fed complete (Con) or iron-deficient (FeD) rations for the duration of gestation to test for similar effects. The FeD offspring were ~12% smaller at weaning and remained so because of a persistent reduction in lean tissue mass. The offspring were fed a complete (stock) diet until 52 weeks of age after which some animals from each litter were fed a HF diet for a further 12 weeks. The HF diet increased body fat when compared with animals fed the stock diet, however, prenatal iron deficiency did not change the ratio of fat:lean in either the stock or HF diet groups. The HF diet caused triglyceride to accumulate in the liver, however, there was no effect of prenatal iron deficiency. The activity of the mitochondrial electron transport complexes was similar in all groups including those challenged with a HF diet. HF feeding increased the number of copies of mitochondrial DNA and the prevalence of the D-loop mutation, however, neither parameter was affected by prenatal iron deficiency. This study shows that the effects of prenatal iron deficiency differ from other models in that there is no persistent effect on hepatic mitochondria in aged animals exposed to an increased metabolic load.

Altered muscle development and expression of the insulin-like growth factor system in growth retarded fetal pigs.

We have used a porcine model of spontaneous differential fetal growth to investigate the effects of fetal size on muscle development. We hypothesized that altered muscle development may occur in small fetuses as a consequence of modified expression of selected genes of the insulin-like growth factor system. We examined the development of the Longissimus muscle (m. Longissimus) in small fetuses and their average sized littermates. We collected small for gestational age fetuses and their average sized sibling on days 45, 65 and 100 of gestation (term is 113-116 days). Small fetuses had significantly lower body weight at all three stages of gestation (p<0.05) and significantly reduced secondary to primary muscle fibre ratio in m. Longissimus on day 100 (p<0.05) compared to their littermates. On day 65, the expression of insulin-like growth factor receptor 1 and insulin-like growth factor binding protein 3 were significantly higher (p<0.05) in m. Longissimus of the small fetuses compared with their average sized littermates. On day 100, the expression of insulin-like growth factor receptor 1 remained significantly higher (p=0.001), in addition to significantly higher levels of insulin-like growth factor receptor 2 and insulin-like growth factor binding protein 5 in the small fetuses (p<0.05). No difference in levels of myogenin was observed between the small and average sized littermates. In conclusion, we demonstrate that reduced fetal muscle development is associated with an increased expression of several genes of the insulin-like growth factor system in small fetuses in mid to late gestation.

Fetal programming: causes and consequences as revealed by studies of dietary manipulation in rats -- a review.

During pregnancy, the developing fetus is dependent on its mother for all nutritional requirements. It is not surprising, therefore, that variations in maternal nutrition can be reflected in alterations in fetal health and well-being. Interestingly, the changes can persist into adulthood and may result in increased risk of diseases such as diabetes, obesity and cardiovascular disease. The first observations of these phenomena resulted in the development of hypotheses collectively brought under the heading of "fetal" or, more recently, "developmental" programming. In this review, we will examine some of the animal models used to understand the mechanisms involved and attempt to determine whether there are common, "gatekeeper", pathways or genes, altered by the different nutritional stresses. We will concentrate primarily on nutrition related to post-natal development of hypertension and will restrict the review to studies in rodents, since that is where most of the mechanistic studies are being undertaken. Our conclusions are that, while there may well be some common gatekeeper pathways, there is also some diversity of mechanism which may contribute to the generation of the same or similar phenotypes.

Maternal Body Weight and Gestational Diabetes Differentially Influence Placental and Pregnancy Outcomes.

CONTEXT: Maternal obesity and gestational diabetes mellitus (GDM) can both contribute to adverse neonatal outcomes. The extent to which this may be mediated by differences in placental metabolism and nutrient transport remains to be determined. OBJECTIVE: Our objective was to examine whether raised maternal body mass index (BMI) and/or GDM contributed to a resetting of the expression of genes within the placenta that are involved in energy sensing, oxidative stress, inflammation, and metabolic pathways. METHODS: Pregnant women from Spain were recruited as part of the "Study of Maternal Nutrition and Genetics on the Foetal Adiposity Programming" survey at the first antenatal visit (12-20 weeks of gestation) and stratified according to prepregnancy BMI and the incidence of GDM. At delivery, placenta and cord blood were sampled and newborn anthropometry measured. RESULTS: Obese women with GDM had higher estimated fetal weight at 34 gestational weeks and a greater risk of preterm deliveries and cesarean section. Birth weight was unaffected by BMI or GDM; however, women who were obese with normal glucose tolerance had increased placental weight and higher plasma glucose and leptin at term. Gene expression for markers of placental energy sensing and oxidative stress, were primarily affected by maternal obesity as mTOR was reduced, whereas SIRT-1 and UCP2 were both upregulated. In placenta from obese women with GDM, gene expression for AMPK was also reduced, whereas the downstream regulator of mTOR, p70S6KB1 was raised. CONCLUSIONS: Placental gene expression is sensitive to both maternal obesity and GDM which both impact on energy sensing and could modulate the effect of either raised maternal BMI or GDM on birth weight.

Copper uptake by cultured trophoblast cells isolated from human term placenta.

This paper has examined copper uptake from CuHis2 complexes by cytotrophoblast cells isolated from term human placenta. Uptake is time-dependent, reaching equilibrium after about 90 min, and saturable, with a calculated apparent Km of 0.174 +/- 0.061 microM and Vmax, measured over 30 min, of 0.721 +/- 0.092 pmol/min/micrograms DNA. To determine whether ATP was required for uptake, cells were incubated with inhibitors of glycolysis (iodoacetate) and the TCA cycle (sodium azide and cyanide). Iodoacetate and sodium azide had no effect on uptake, but cyanide decreased the initial rate of uptake. This effect was due to copper binding to the inhibitor and decreasing the effective substrate concentration rather than inhibition of uptake through ATP depletion. Ouabain and monensin had no effect, showing that neither the Na+ gradient nor endocytosis were involved in uptake. The monovalent ion chelator, bathocuproine sulphonate, had no effect on uptake but buthionine sulfoximine, an inhibitor of glutathione synthesis, did decrease both the rate of uptake and equilibrium copper levels, suggesting that copper may bind to glutathione within the cell. The data show that copper is taken up by a passive carrier-mediated transporter and, following uptake, binds to glutathione within the cell.

A plasma membrane NADH oxidase is involved in copper uptake by plasma membrane vesicles isolated from rat liver.

The accumulation of copper (Cu) by hepatocytes is initiated by the binding of Cu in either a CuHis2 complex or as a CuHisAlb ternary complex, followed by transfer of the metal alone across the cell membrane. In this paper, we provide evidence that the transfer involves reduction of cupric (Cu(II)) copper to cuprous (Cu(I)) copper and further we show that membrane-bound NADH oxidase can provide the electron required for the reduction. 64Cu uptake by rat liver plasma membrane vesicles is stimulated by the addition of NADH, but not NAD+. The stimulation increases the Vmax from 4.75 +/- 0.02 to 8.38 +/- 0.40 nmol Cu/mg protein per min (P < 0.05, mean +/- S.E., n = 3) without significantly altering the K0.5 (1.52 +/- 0.17 and 2.10 +/- 0.22 mumol/l; with n values of 1.30 +/- 0.01 and 1.43 +/- 0.10, respectively; analysing by the Hill equation). Correspondingly, addition of CuHis2 stimulated NADH-oxidase activity by a maximum of 7.4 +/- 2.1 nmol/mg protein per min (P < 0.01, mean +/- S.E., n = 5) at 5 mumol/l and a NADH concentration of 150 mumol/l. Ascorbic acid also stimulated copper uptake, and points to a reductive dissociation of copper prior to its movement into the cell. Our data indicate that membrane bound enzymes can provide an electron for the reduction of copper prior to uptake and suggest a physiological role for the plasma membrane NADH oxidase.

Characterisation of the copper uptake mechanism and isolation of the ceruloplasmin receptor/copper transporter in human placental vesicles.

In this paper we have studied copper (Cu) uptake by microvillar vesicles isolated from human term placenta. We have characterised Cu uptake from CuHis2 complexes and shown that ceruloplasmin (Cp) inhibits uptake. Inhibition is complex and variable; in one series of experiments, the Vmax for uptake drops from 31.3 +/- 1.2 nmol/min per mg vesicle protein without added Cp to 11.3 +/- 1 nmol/min per mg vesicle protein at 91 micrograms/ml Cp. Similarly, the K0.5 increases from 0.35 +/- 0.08 microM to 1.35 +/- 0.25 microM, while the n value (the Hill coefficient) falls from 1.9 +/- 0.23 in the absence of Cp to 1.1 +/- 0.13 In another series, Cp had no effect below concentrations of about 100 micrograms/ml and in a third series only increased K0.5. The variability in effect seems to be related to the specific activity of the ceruloplasmin, which in turn is related to the copper complexes of the protein. The effect is specific for Cp; apotransferrin and a2-macroglobulin have no effect. 67Cu-labelled ceruloplasmin binds specifically to vesicles of term placenta with an affinity of 2.8 microU/mg vesicle protein and a Bmax of 79 microU/mg vesicle protein. CuHis2, but not histidine alone, can block the uptake. The data can be reconciled by proposing that the binding site of the transporter is relatively small and recognises a Cu-dihistidine structure common to the low-molecular-weight complex and to the Type I and Type II coppers of ceruloplasmin. We have used these observations to develop an isolation method for the transporter and have identified it as a protein of M(r) 90,000 which is closely associated with alkaline phosphatase. There are also two proteins of M(r) 45,000 and 40,000 which may be breakdown products of the larger complex. Antibodies to the 45,000 protein block Cu binding and uptake from CuHis2 complexes, strongly implicating it as the copper transporter/ceruloplasmin receptor of human term placenta.

A non-radioactive method for measuring Cu uptake in HepG2 cells.

At present, all data on Cu uptake and metabolism have been derived from radioactive uptake experiments. These experiments are limited by the availability of the radioactive isotopes 64Cu or 67Cu, and their short half-life (12.5 and 62 h, respectively). In this paper, we investigate an alternative method to study the uptake of Cu with natural isotopes in HepG2 cells, a liver cell line used extensively to study Cu metabolism. In nature, Cu occurs as two stable isotopes, 63Cu and 65Cu (63Cu/65Cu = 2.23). This ratio can be measured accurately using inductively coupled plasma mass spectrometry (ICP-MS). In initial experiments, we attempted to measure the time course of Cu uptake using 65Cu. The change in the 63Cu/65Cu ratio, however, was too small to allow measurement of Cu uptake by the cells. To overcome this difficulty, the natural 63Cu/65Cu ratio in HepG2 cells was altered using long-term incubation with 63Cu. This had a significant effect on Cu concentration in HepG2 cells, changing it from 81.9 +/- 9.46 pmol microg DNA(-1) (week 1) to 155 +/- 8.63 pmol microg DNA(-1) (week 2) and stabilising at 171 +/- 4.82 pmol microg DNA(-1) (week 3). After three weeks of culture with 2 microM 63Cu the 63Cu/65Cu changed from 2.18 +/- 0.05 to 15.3 +/- 1.01. Cu uptake was then investigated as before using 65Cu. Uptake was linear over 60 min, temperature dependent and consistent with previous kinetics data. These observations suggest that stable isotope ICP-MS provides an alternative technique for the study of Cu uptake by HepG2 cells.

Ca2+-uptake by tissue sections and biochemical characteristics of sarcoplasmic reticulum isolated from fish fast and slow muscles.

Frozen sections (10 micrometer) were cut from fast, slow and cardiac muscles of rainbow trout, frog, and rat. Rates of 45Ca2+-uptake by thin sections were compared at each animal's normal body temperature. Initial rates of Ca2+-uptake were 1.8 and 2.4 times higher, respectively, in trout and rat fast than slow muscles. In spite of a lower body temperature (10 degrees C) rates of Ca2+-uptake by trout fast muscles were 2.8 times higher than for rat Extensor digitorum longus at 37 degrees C. It is suggested that the high functional capacity of fish sarcoplasmic reticulum (SR) is related to adaptations associated with the need for rapid cycles of contraction and relaxation during high speed swimming. The biochemical characteristics of SR isolated from trout fast and slow muscles has also been investigated. The ratio of Ca2+-ATPase activity between fast and slow fibres (2:1) was similar to that obtained for Ca2+-uptake by whole muscle sections. No evidence was obtained for modulation of calcium transport by cAMP dependent protein kinases. The protein composition of highly purified trout SR was investigated by polyacrylamide gel electrophoresis. The concentration of 95 000 to 100 000 dalton (calcium pump protein) was significantly reduced in slow compared to fast muscle SR. Slow muscle SR contains a high proportion of additional protein bands of 48 000 and 31 000 molecular weight.

Physiologic function of the Wilson disease gene product, ATP7B.

The genes responsible for Wilson disease and Menkes syndrome have been cloned and identified as copper ATPases. These enzymes form part of a large family of transporters, the P-type ATPases. Although copper ATPases share strong structural similarities with these other pumps, comparatively little is known about their physiologic function. In this review, we examine data relating to the Wilson disease protein, ATP7B, in the liver. We present evidence suggesting that ATP7B is located intracellularly, together with data suggesting that, at least in part, ATP7B may also be found on the canalicular membrane. We also examine the form of copper that the transporter recognizes. We then review data on the Long-Evans Cinnamon rat, a model for Wilson disease, and discuss what effect the Wilson disease mutation has on copper transport. Finally, we conclude that, although we have made major advances in our understanding of copper metabolism in the liver, there are still many questions awaiting answers.

Transferrin binding by microvillar vesicles isolated from rat placenta.

Microvillar vesicles have been isolated from rat placentae. The vesicles have been characterized with respect to their microscopic appearance, their capacity to accumulate amino acids, and their enzyme activities. Alkaline phosphatase activity in the vesicle fraction was threefold greater than in the homogenate, and cytochrome c reductase activity was decreased fourfold. The vesicles were used to characterize transferrin binding to the trophoblast membrane in vitro. Transferrin binding was specific (K alpha = 8 X 10(7) M-1), and could not be replaced by apotransferrin. Apotransferrin binding showed lower affinity than the diferric transferrin, and did not appear to saturate over the range 0 to 500 micrograms/ml.

The effect of ceruloplasmin on iron release from placental (BeWo) cells; evidence for an endogenous Cu oxidase.

The mechanism of iron release from the placenta into the fetal circulation is not well understood. Ceruloplasmin, a plasma ferroxidase, has been implicated in iron efflux from a variety of cell types. The hypothesis is that circulating ceruloplasmin facilitates iron efflux by oxidizing the released Fe(II) to Fe(III) for incorporation into transferrin. We tested whether this mechanism mediates iron release from placental cells into the fetal circulation, using the BeWo cell line, a choriocarcinoma which can differentiate into a syncytium.(59)Fe release from undifferentiated or differentiated cells and from cells grown on porous filters was not stimulated by extracellular ceruloplasmin. Instead, we found that BeWo cells express an endogenous ferroxidase. The protein is membrane bound and cross-reacts with an anti-ceruloplasmin antibody, but has a different size; 100 and 140 kDa. Similar immunoreactivity was identified in first- and third-trimester human placentae. In BeWo cells, the protein has a perinuclear localization but does not entirely co-localize with markers for the endoplasmic reticulum or Golgi apparatus. We propose that this oxidase performs the same function as serum ceruloplasmin and is involved in iron release into the fetal circulation.

Characterization of glucose transport and glucose transporters in the human choriocarcinoma cell line, BeWo.

In this study we have characterized 2-deoxyglucose (2DG) transport and hexose transporter expression in the human choriocarcinoma cell line, BeWo. 2DG uptake in BeWo cells displayed saturable kinetics (V(max), 29+/-1.5 nmol/min/mg protein;K(m), 1.5+/-0.02 m m) and was significantly inhibited in the presence of 2-deoxyglucose, mannose and 3- O -methyl glucose (all at a competing concentration of 30 m m) by up to 97 per cent, but not by galactose or fructose. Glucose uptake was not Na(+)-dependent, but was inhibited by cytochalasin B (by approx 85 per cent) indicating that hexose uptake was mediated via a facilitative glucose transport mechanism. Northern and immunoblot analyses revealed that BeWo cells expressed GLUT1 and GLUT5, but not GLUT2 or GLUT3. On immunoblots, GLUT1 migrated as a broad protein band on SDS-gels (average M(r)of 55 kDa) and treatment with N -glycanase resulted in a significant shift in its electrophoretic mobility; the core protein migrating as a 40 kDa band indicating that the carrier was heavily glycosylated. GLUT5 was detected as a discrete 60 kDa band and like GLUT1, the observed immunoreactive signal was lost when using antiserum that had been pre-adsorbed with the antigenic peptide. Our findings indicate that BeWo cells express a facilitative glucose transport system with characteristics broadly similar to those reported in isolated human placental membrane vesicles and that they are likely to serve as a useful experimental system for studying the regulation of placental glucose transport and transporter expression.

Effects of metallothionein on the observed copper distribution in cell extracts.

Systematic studies have been undertaken to compare the effects of cell lysis and chromatography conditions on the observed distribution of Cu amongst Cu-binding proteins in cultured cells. The variables included rate of centrifugation, presence or absence of non-ionic detergent, and presence or absence of dithiothreitol. The application of an improved FPLC gel filtration system has permitted us to examine the effects of the addition of exogenous metallothionein (MT) to cell extracts. When the cell extract contains low levels of endogenous MT, the addition of MT in the presence of dithiothreitol causes a shift of copper to the MT peak. High levels of MT can therefore remove copper from other Cu-binding ligands during cell homogenization, hence producing artifactual Cu distribution results. The use of an anaerobic buffer system has greatly reduced the observed level of Cu exchange, and has allowed comparison of Cu distribution in normal cells and cells from patients with Menkes' disease.

Albumin has no role in the uptake of copper by human fibroblasts.

The mechanism of copper uptake by cells has been the subject of controversy for some time. This paper examines the possibility of a role for albumin in the uptake of copper by fibroblasts. Although the cells could accumulate copper from a copper-albumin complex, there was no evidence for either copper-albumin or albumin receptors on the cell surface. The possibility of a surface exchange mechanism for copper was examined. While copper uptake showed saturation with increasing concentrations of labelled copper-albumin, adding unlabelled copper to the incubation medium did not inhibit uptake. Adding albumin or histidine to the copper-albumin complex resulted in an inhibition of copper uptake. The results can only be explained by the cell taking up free copper from the incubation medium, with the albumin then releasing its copper to maintain the equilibrium between free and bound metal. Since, in vivo there is essentially no free copper in serum, it is concluded that albumin is most unlikely to play a role in the uptake of copper by fibroblasts.

The effect of D-penicillamine on metallothionein mRNA levels and copper distribution in mouse hepatocytes.

Penicillamine increases the levels of metallothionein (MT) mRNA in a time and concentration dependent manner without altering either the rate of copper uptake or the amount of copper within the cell. The effect is dependent on the presence of intracellular copper, however, since depletion of copper by chelators blocks the effect, and does not alter the ability of dexamethasone to stimulate mRNA production. Penicillamine did not alter the distribution of 64Cu in the hepatocytes, as measured by fast protein liquid chromatography (FPLC), although the pattern may be affected by the amount of MT present. The data indicates that penicillamine removes copper from some intermediary ligand, thereby making it available to induce metallothionein. It is possible that this is part of the therapeutic action of the chelator in the treatment of Wilson's disease.

Cu metabolism in the liver.

This paper has, given some idea of our concepts of the processes involved in the transport of Cu across cell membranes in the liver, which we have summarised in Fig 1. Cu(II)His2 is reduced to Cu(I). This is transported across the membrane, re-oxidised, either before or after binding to glutathione (Freedman et al., 1989) or HAH1 (Klomp et al., 1997), binds to SAHH, and donates Cu(II) to the ATPase. It is very interesting that cells which are very diverse from an evolutionary point of view still use very similar methods to handle the metal. Whether regulation of transport is also the sam remains to be seen. We would guess that, although there will be strong similarities, there will also be very significant differences, reflecting the different environments seen by different tissues in mammalian cells and given the different requirements of the tissues.