- Email: [email protected]
ATP Dependent Ca2+ Transport Across Basal Membrane of Human Syncytiotrophoblast in Pregnancies Complicated by Intrauterine Growth Restriction or Diabetes
Placenta (2003), 24, 445–452 doi:10.1053/plac.2002.0941
ATP Dependent Ca2+ Transport Across Basal Membrane of Human Syncytiotrophoblast in Pregnancies Complicated by Intrauterine Growth Restriction or Diabetes H. Stridade, E. Buchtc, T. Janssona, M. Wennergrenb and T. L. Powella a
Perinatal Center, Department of Physiology & Pharmacology and b Department of Obstetrics and Gynecology, Go¨teborg University, Sweden, c Department of Molecular Medicine, Karolinska Hospital, Karolinska Institute, Stockholm, Sweden and d Department of Clinical Medicine, O } rebro University, Sweden
Neonates born after pregnancies complicated by diabetes or intrauterine growth restriction (IUGR) have increased incidence of hypocalcaemia. Furthermore, IUGR is associated with reduced bone mineralization in infancy and osteoporosis in adult life. We tested the hypothesis that placental calcium transport is altered in these pregnancy complications. Transport of calcium into syncytiotrophoblast basal plasma membrane (BM) vesicles was studied by rapid filtration and protein expression of Ca2+ ATPase by Western blot. In IUGR Ca2+ ATPase activity was increased by 48 per cent (n=13; P<0.05) whereas protein expression was 15 per cent lower (n=13; P<0.05) than in controls (n=16). Basal membrane ATP dependent calcium transport was unaltered in gestational diabetes (GDM) but increased by 54 per cent in insulin dependent diabetes (IDDM) compared to controls (P<0.05; n=14). Diabetes did not affect Ca2+ ATPase expression in BM. We have previously shown that the mid-molecular fragment of parathyroid hormone related peptide (PTHrP midmolecule) stimulates BM Ca2+ ATPase in vitro. PTHrP midmolecule concentrations in umbilical cord plasma were measured using radioimmunoassay. The concentrations in umbilical cord plasma were increased in IUGR, but unaltered in diabetes. In conclusion, placental calcium pump is activated in IUGR and IDDM, which may be secondary to increased foetal calcium demand. We speculate that PTHrP midmolecule may be one mechanism for 2003 Published by Elsevier Science Ltd. activating BM Ca2+ ATPase in IUGR. Placenta (2003), 24, 445–452
INTRODUCTION Neonates born after pregnancies complicated by insulindependent diabetes (IDDM) or intrauterine growth restriction (IUGR) frequently show signs of disturbed calcium homeostasis. Umbilical blood levels of calcium are reduced at birth in association with IDDM (Cruikshank et al., 1983) and the infants are at risk to develop hypocalcaemia during the first days of life (Cruikshank et al., 1983; Mimouni et al., 1986; Mimouni et al., 1990; Demarini et al., 1994). Although one report has suggested a decreased bone mineralization in infants of mothers with IDDM (Mimouni et al., 1988) more recent studies using dual X-ray absorptiometry show that IDDM is associated with an increased whole body bone mineral content in the infant (Lapillonne et al., 1997). The incidence of neonatal hypocalcaemia is also increased in IUGR (Oh, 1977), however this may be related to birth asphyxia rather than IUGR per se (Tsang et al., 1975). Reduced bone mineral e
To whom correspondence should be addressed at: Department of Clinical Medicine, O } rebro University, 701 85 O } rebro, Sweden. Tel.: 46-19-602 66 63; Fax: 46-19-602 66 50; E-mail: [email protected] 0143–4004/03/$-see front matter
content is a common finding in the IUGR infant (Minton et al., 1983; Namgung et al., 1993). In addition, low birth weight has been shown to be associated with a reduced bone mineral content in the lumbar spine and femoral neck in woman and men aged 70 years (Gale et al., 2001). These findings raise the possibility that IUGR represents a risk factor for osteoporosis later in life. The factors causing the disturbed calcium homeostasis in IUGR infants and infants born to women with IDDM are not well established. The regulation of serum calcium concentrations in the foetus is complex and dependent on placental transport capacity as well as maternal and foetal calcium homeostasis. The possibility that placental transport of calcium is altered in pregnancy complications has been addressed in some detail in animal experimental models. Streptozotocin induced diabetes in the rat is associated with foetal hypocalcaemia and decreased maternal-to-foetal calcium transport (Husain et al., 1994), which may be due to reduced expression of placental calbindin, the primary cytosolic calcium binding protein in the rat placenta (Hamilton et al., 2000). Also IUGR induced by uterine artery ligation in the rat is characterized by a decreased placental transport of calcium (Mughal et al., 1989; 2003 Published by Elsevier Science Ltd.
446
Mimouni et al., 1995). However, placental transport of calcium in association with IUGR or diabetes has not been studied in the human. In the human placenta the syncytiotrophoblast cell (ST) represents the primary functional barrier for transplacental transport between maternal and foetal plasma. In particular, the rate of calcium transport depends on the transport characteristics of the two polarized plasma membranes of the ST and the expression of intracellular calcium binding proteins. The specific cytosolic calcium binding proteins involved in human placental calcium handling are not well defined. In contrast, calcium transport mechanisms across the microvillous (MVM) and basal plasma membrane (BM) have been studied in some detail. Both MVM and BM contain several ATP independent as well as ATP dependent transport mechanisms (Fisher et al., 1987; Brunette and Leclerc, 1992; Kamath et al., 1992; Kamath and Smith, 1994), including Na + /Ca2+ exchange (Kamath and Smith, 1994). In BM, Ca2+ ATPase is the primary transporter mediating the extrusion of calcium out of the cell (Fisher et al., 1987; Strid and Powell, 2000) representing a crucial step in the establishment of the foeto-maternal calcium gradient. PTHrP midmolecule may play a key role in regulating transport of calcium across the placenta since it has been shown to maintain the calcium gradient across the placenta in some animals (Care et al., 1990; Kronenberg et al., 1998) and we recently reported that PTHrP mid molecule stimulates Ca2+ ATPase in BM vesicles isolated from human placenta (Strid et al., 2002). In addition, expression of PTHrP in foetal membranes and placenta is increased in preterm IUGR (Curtis et al., 2000). In this study we tested the hypothesis that pregnancies complicated by IUGR and diabetes are associated with an altered ATP dependent calcium transport across the basal plasma membrane of the syncytiotrophoblast. Basal plasma membranes were isolated from placentae obtained from normal and complicated pregnancies. ATP dependent calcium transport was studied by rapid filtration techniques and the protein expression of plasma membrane Ca2+ ATPase (PMCA) using Western blot. Furthermore, we measured umbilical cord plasma concentrations of the calcium regulatory hormone PTHrP midmolecule.
MATERIAL AND METHODS Placental tissue The collection of placental tissue was approved by the Committee for Research Ethics at Go¨teborg University. Placentae were obtained from pregnancies complicated by diabetes (gestational diabetes, GDM and insulin-dependent diabetes, IDDM) or intrauterine growth restriction (IUGR) as well as from uncomplicated pregnancies (controls). Screening for GDM was carried out routinely in all pregnancies by measurements of capillary blood–glucose in the nonfasting state at least at four occasions starting in weeks
Placenta (2003), Vol. 24
8–12 of gestation. If non-fasting blood–glucose level was _7.0 mmol/l, patients were referred to a 75 g oral glucose tolerance test (OGTT) after overnight fasting. GDM was defined as a fasting blood–glucose level of 6.1 mmol/l or blood–glucose level of 9.0 mmol/l 2 h after OGTT in a pregnant woman without prior known diabetes. These pregnancies were not associated with any major complications other than GDM. Assessment of blood–glucose control was performed by self-monitoring fasting and postprandial blood– glucose levels at least twice a week. Patients with gestational diabetes were treated with diet adjustment only (n=20; Diabetes White A1), except in seven patients in which insulin was administered in a 5-dose regime (Diabetes White A2). Criteria for initiating insulin therapy was elevated blood– glucose levels (fasting glucose=6.1 and/or postprandial blood–glucose=8.0) despite consistent dietary adjustments. Patients in the IDDM group had no other major complication and were classified as White B, C or D (White, 1949). All women with IDDM attended the obstetric antenatal clinic at Sahlgren’s University Hospital, Go¨teborg. Patients carried out self-monitoring of capillary blood–glucose concentrations at least four times daily and glycosylated haemoglobin (Hb A1C) was measured every 2–6 weeks. All patients in the IDDM group received insulin in a 5-dose regime. Neonates large for gestational age (LGA) were defined as having a birth weight greater than mean plus 2 s.d. using intrauterine growth curves based on ultrasonically estimated foetal weight (Marsal et al., 1996). Similarly, IUGR was defined as a birth weight 2 s.d. below the mean for that gestational age (Marsal et al., 1996).
Isolation of syncytiotrophoblast plasma membranes Placentae were placed on ice immediately after delivery and the membrane isolation procedure was started within 30 min. Basal plasma membranes (BM) were prepared according to a well characterized protocol (Illsley et al., 1990; Jansson and Illsley, 1993; Jansson et al., 1993) with some modifications (Strid and Powell, 2000). Briefly, after initial homogenization and centrifugation steps (at 4C in 250 m sucrose, 10 m Hepes/Tris, protease inhibitors, pH 7.4) BM were separated from MVM by Mg2+ precipitation and further purified on a sucrose step gradient. Samples were frozen in liquid nitrogen and stored at 80C. BM enrichment was assessed using standard activity assays for adenylate cyclase (Schultz, 1984) and alkaline phosphatase (Bowers and McComb, 1966). The production of cAMP by adenylate cyclase was measured by radioimmunoassay (New England Nuclear, Boston, MA, USA). Orientation assay for BM vesicles was carried out by measuring alkaline phosphatase activity prior to and after permeabilization of membranes with 1 per cent Triton-X, an approach similar to the technique used by Glazier and coworkers (Glazier et al., 1988). The ratio of alkaline phosphatase activity before and after permeabilization represents a measure of right-side-out vesicles.
Strid et al.: Placental Ca2+ ATPase in IUGR and Diabetes
Western blot Gel electrophoresis (20 g membrane protein/lane) and transfer were carried out as described elsewhere (Strid and Powell, 2000). The monoclonal primary antibody used (Clone 5F10,ABR) is directed to all four known isoforms of the plasma membrane Ca2+ ATPase. Detection was performed using the ECL-system (Amersham, Buckinghamshire, UK) according to the manufacturers directions. Densitometry of the autoradiography for PMCA expression was performed using IP Lab gel (Signaling Analytics Corporation, Vienna, VA, USA). Only the prominent 140 kDA band was analysed when comparing PMCA expression between groups. The densitometric values fell within the linear range of detection. Each gel contained nine samples, two of which were the same on all the gels. The mean density of these two samples was used to normalize all the gels. No bands were observed in control experiments in the absence of primary antibody.
Calcium uptake The uptake of calcium by basal membrane vesicles was measured by standard rapid filtration techniques (Fisher et al., 1987) with some modification (Strid and Powell, 2000). Briefly, the vesicles were resuspended in vesicle buffer (10 m Hepes/ Tris; 250 m sucrose; pH 7.4 at 37C) and centrifuged. After centrifugation the vesicles were resuspended in vesicle buffer. Calcium uptake measurements were started by the addition of 20 l of vesicles (100–200 g total membrane protein) to 500 l calcium transport buffer (240 m sucrose; 5 m MgCl2; 0.2 m EGTA; 0.2 m CaCl2; 1 Ci 45CaCl2; 10 m Hepes/ Tris, pH 7.4) in the presence or absence of 5 m ATP–MgCl2. Calcium uptake was allowed to proceed for 10 min at 37C and stopped by the addition of ice cold wash solution (10 m Hepes/Tris; 250 m Sucrose; 4 m EGTA; pH 7.4 at 4C) to approximate initial rate (Strid and Powell, 2000). Subsequently, the reaction mixture was filtered under vacuum on 0.45 m filters (HAWO, Millipore). The radioactivity on the filter was measured by liquid scintillation. The ATP dependent calcium uptake was calculated as the difference between Ca2+ uptake in presence and absence of ATP. Calcium uptake was expressed as nmoles of Ca2+ /mg total membrane protein.
Protein determination Protein concentration was determined according to the method of Bradford (Bradford, 1976), using the Bio Rad (Hercules, CA, USA) assay procedure and BSA as a standard.
Umbilical cord blood collection Umbilical cord blood (10 ml pooled venous and arterial blood) was collected for PTHrP analysis, within 10 min after delivery, in tubes containing protease inhibitors (PTHrP Cocktail
447
Table 1. Clinical characteristics of IUGR group and corresponding control group
n Gestational age (wks) Placental weight (g) Birth weight (g)
Controls
IUGR
16 37.20.5 64643 3103254
13 36.10.8 34923* 1665132*
Means. *P<0.05 as compared with controls. IUGR, intrauterine growth restriction.
Tubes; Nicholas Institute). Samples were obtained from normal term pregnancies as well as from pregnancies complicated by diabetes or IUGR. The collected blood was centrifuged for 15 min at 2000 rpm at 4C, plasma was separated and frozen at 80C until radioimmunoassays were performed. Radioimmunoassay A synthetic midmolecular fragment of PTHrP 63–77 was used for antibody production in rabbits, as reference standard and for radioiodination. The samples were extracted by Sep-PAK C18 cartridge (Waters, Milford, MA, USA) as described previously with the RIA (Bucht et al., 1992). The detection limit of the assay, based on 2 s.d. below maximal binding, was after extraction 0.8 pg/ml. Intraassay variation was <9 per cent, and interassay variation was <15 per cent at all concentrations. Data analysis Statistical significance of differences between groups was determined utilizing t-test or ANOVA followed by Dunnett test. Data are expressed as mean unless otherwise specified. RESULTS Clinical data Gestational age, placental weight and birth weight for all groups are presented in Tables 1 and 2. Both birth weight and placental weight were lower in the IUGR group as compared to controls (P<0.05; n=13–16), however gestational age was not different between groups (Table 1). In the IDDM group the birth weight was significantly higher than in the control group (Table 2; P<0.05; n=14–27). Characterization of basal membrane preparations In BM isolated from control placentae, the enrichment in adenylate cyclase activity was 29–32 fold while the enrichment
448
Placenta (2003), Vol. 24
Table 2. Clinical characteristics of diabetes groups and corresponding control groups
n Gestational age (wks) Placental weight (g) Birth weight (g)
Control
GDM
IDDM
37 38.80.3 71331 351499
27 39.00.3 66732 3618103
14 38.30.4 85575 397213*
Means. *P<0.05 as compared to controls. GDM, gestational diabetes mellitus; IDDM, insulin dependent diabetes mellitus.
Table 3. Marker enzyme activities in diabetes and control groups Adenylate cyclase (pmol/minmg)
Control GDM IDDM
Alkaline phosphatase (nmol PO4/smg)
n
BM
P2
n
BM
Hom
18 13 13
570.783.8 (28.8) 495.658.3 (30.3) 488.261.7 (31.8)
17.94.4 16.43.9 16.96.5
18 24 12
9.00.9 (2.6) 8.50.8 (2.3) 8.20.6 (2.1)
3.50.5 3.60.4 3.90.4
Values are given as means. Enzyme enrichments (in parentheses) were calculated as vesicle enzyme specific activity relative to that for the homogenate (alkaline phosphatase) or the post nuclear membrane pellet (P2; adenylate cyclase).
in alkaline phosphatase activity was 2.5 fold (n=8–9). Activities of marker enzymes were not significantly different in BM isolated from complicated pregnancies. In BM isolated from placentae collected from pregnancies complicated with IUGR, the enrichment in adenylate cyclase activity was 28 fold while the enrichment in alkaline phosphatase activity was 2.4 fold (n=10–11). Furthermore, enrichments of marker enzymes in IDDM and GDM groups were similar to BM obtained from control placentae (Table 3). We have previously shown that the degree of contamination by endoplasmic reticulum and mitochondria is low in our BM preparation and that enrichments in BM preparations from premature placentae are not different from term placentae (Strid and Powell, 2000). The ratio of alkaline phosphatase activity before and after permeabilization was 0.800.08 (n=6) in control BM, indicating that approximately 80 per cent of vesicles were oriented right-side-out. This value is similar to the 7312 per cent reported previously for BM prepared according to the protocol of Illsley and coworkers (Illsley et al., 1990). Furthermore, the fraction of BM vesicles oriented right-side-out in IUGR (7612 per cent, n=5), GD (859 per cent, n=5) and IDDM groups (876 per cent, n=7) was not significantly different from control vesicles.
Ca2+ transport As shown previously (Strid and Powell, 2000) the ATPindependent Ca2+ uptake was approximately 100 times lower than the ATP-dependent transport under these experimental conditions and did not differ between groups (data not shown). The ATP dependent transport of calcium in BM isolated from
Figure 1. ATP dependent calcium transport in IUGR BM (n=13) in comparison with control BM (n=16; *P<0.05; t-test).
IUGR placentae was 48 per cent higher as compared to control (P<0.05; n=13–16; Figure 1). In BM isolated from GDM pregnancies ATP dependent calcium transport was unaltered. In contrast, IDDM was associated with a 54 per cent increase in ATP dependent transport compared to controls (P<0.05; n=14–27; Figure 2). Further subgrouping of the diabetic placentae showed no significant differences in the calcium transport between IDDM and IDDM/LGA or GDM and GDM/LGA (data not shown). PMCA protein expression We used Western blot to investigate possible changes in Ca2+ ATPase expression per mg total protein in BM isolated from
Strid et al.: Placental Ca2+ ATPase in IUGR and Diabetes
449
Figure 5. Relative expression of PMCA in BM isolated from pregnancies complicated by GDM (n=12) and IDDM (n=13) in comparison with BM isolated from control group (n=18; ANOVA). Figure 2. ATP dependent calcium transport into BM isolated from control (n=37), GDM (n=27) or IDDM (n=14) groups (*P<0.05 as compared to controls; ANOVA).
Figure 3. Autoradiography of Western blot of membrane Ca2+ ATPase (PMCA) in BM isolated from IUGR pregnancies (I) and BM isolated from normal pregnancies (C).
Figure 6. Plasma concentrations of PTHrP midmolecule in umbilical cord blood, measured by radioimmunoassay using an antibody directed towards residues 63–77. Blood was collected from infants born after uncomplicated pregnancies (control; n=13), pregnancies complicated by growth restriction (IUGR; n=6) or maternal diabetes (n=5) *P<0.05 as compared to controls ANOVA.
PTHrP midmolecule concentration measurements in umbilical cord blood plasma
Figure 4. Relative expression of plasma membrane Ca2+ ATPase (PMCA) in BM isolated from IUGR (n=13) pregnancies in comparison with BM isolated from normal pregnancies (n=13; *P<0.05; t-test)
pregnancies complicated with IDDM, GDM or IUGR compared to controls. Figure 3 illustrates a Western blot with IUGR (I) and control (C) samples. PMCA protein expression in IUGR was decreased by 15 per cent compared to controls (Figure 4; P<0.05; n=13). However, there was no change in the PMCA expression in the two diabetic groups compared to control (Figure 5).
The PTHrP midmolecule concentration in umbilical cord blood plasma from infants born after uncomplicated term pregnancies was 5.80.6 pg/ml (n=13). PTHrP midmolecule concentrations in umbilical cord blood plasma collected from infants born after pregnancies complicated by diabetes were not significantly different compared to the control group. In contrast, babies born growth restricted had a 67 per cent increase in umbilical plasma PTHrP concentrations compared to controls (P<0.05, n=6). Figure 6 illustrates the differences in plasma concentrations between the three groups.
DISCUSSION Pregnancies complicated by IUGR or IDDM are associated with foetal/neonatal hypocalcaemia (Oh, 1977; Cruikshank et al., 1983; Mimouni et al., 1986; Mimouni et al., 1990; Demarini et al., 1994) and IUGR infants show signs of reduced bone mineralization (Minton et al., 1983; Namgung et al., 1993). Somewhat surprising, the main findings in this
450
study were that ATP dependent Ca2+ transport across syncytiotrophoblast basal plasma membrane vesicles is increased in IUGR and IDDM consistent with an increased transplacental transport of calcium in these pregnancy complications. Therefore these changes are unlikely to be the primary cause of the disturbance in calcium homeostasis in IUGR and IDDM but may represent a compensatory mechanism in response to an imbalance between calcium requirement and calcium supply. In addition, our data indicate that these alterations are due to an activation of existing transporters since protein expression of the calcium pump was unchanged or reduced. Furthermore, we report increased umbilical plasma concentrations of PTHrP midmolecule in association with IUGR and we suggest that PTHrP midmolecule may mediate the activation of Ca2+ ATPase in BM isolated from IUGR placentae. The increased BM Ca2+ ATPase activity in the IUGR placenta does not explain the impaired foetal bone mineralization in this condition. One interpretation of these findings is that the increased BM Ca2+ ATPase activity is secondary to an insuffient calcium supply due to decreased activity or expression of other components of the placental calcium transporting systems, such as cytosolic calcium binding proteins in the syncytiotrophoblast. Furthermore, reduced bone mineralization in IUGR may be caused by an insufficiency of the placental calcium transport system that is relative rather than absolute. This speculation relates to the fact that IUGR foetuses commonly exhibit asymmetric growth restriction, i.e. body weight is reduced much more than body length. Placental weight, and therefore placental exchange area, is correlated to foetal weight rather than length whereas skeletal mass is related to length. Consequently, the demands for calcium of the relatively unaffected skeletal mass in IUGR may be greater than can be provided by the placental supply system, which could lead to stimulation of BM Ca2+ ATPase. If this speculation is correct, the efforts to enhance placental calcium transport are insufficient in many cases of IUGR, explaining the impaired bone mineralization observed in the IUGR infant. IUGR has been shown to be associated with a number of alterations in the expression and/or activity of key placental nutrient and ion transporters. In particular amino acid transporters appear to be affected since IUGR is characterized by a reduced activity of system A (Mahendran et al., 1993) and taurine transporters (Norberg et al., 1998) in MVM but not in BM (Norberg et al., 1998; Jansson et al., 2002), a lower activity of cationic amino acid transporters in BM but not in MVM (Jansson et al., 1998) whereas the activity for transporters for leucine has been demonstrated to be reduced in both syncytiotrophoblast plasma membranes (Jansson et al., 1998). In contrast, the activity and expression of syncytiotrophoblast glucose transporters are unaffected by IUGR (Jansson et al., 1993). The impact of IUGR on BM PMCA activity in the current study is unique since this is the first demonstration of increased activity of a placental transporter in association with this pregnancy complication. This is an important finding, strongly suggesting that the changes observed in placental
Placenta (2003), Vol. 24
nutrient and ion transporters are specific rather than a consequence of a general detrimental effect of the IUGR condition on the placenta. Collectively, these observations indicate that syncytiotrophoblast transporters are subjected to regulation by mechanisms unique for each transporter and are consistent with the possibility that some transporters in the two polarized plasma membranes of the syncytiotrophoblast are regulated independently. Insulin-dependent diabetes is often complicated by accelerated foetal growth as illustrated by the higher birth weights in the IDDM group in our study. Accelerated foetal growth in IDDM is correlated with foetal hyperinsulinaemia (Sosenko et al., 1979; Brans et al., 1983; Kainer et al., 1997). Since insulin stimulates foetal bone formation directly (Yano et al., 1994) as well as mediated via IGF-1 release from the liver (Canalis, 1980) it is possible that the hyperinsulinaemia causes the increased whole body bone mineral content of the large-forgestational age foetus in IDDM pregnancy as discussed by Lapillonne and coworkers (Lapillonne et al., 1997). Therefore the increased activity of Ca2+ ATPase in the placental barrier in association with IDDM demonstrated in this study may reflect a compensatory mechanism to increase placental transport of calcium in order to meet the increased foetal demands. The increased calcium transport across BM in IDDM in vitro is consistent with an increased transplacental transport of calcium in vivo, however firm conclusions must await direct measurements in vivo. Nevertheless, the findings of the present study are in contrast to the observations of decreased placental transport of calcium in experimental models of diabetes in the rat (Husain et al., 1994; Hamilton et al., 2000). One explanation to this discrepancy could be that our interpretation of the increased calcium transport across BM in IDDM is incorrect. Our findings might instead be a consequence of an impairment of other components of the placental calcium transport system, such as a decreased expression of cytosolic calcium binding proteins in the syncytiotrophoblast. Or alternatively, the streptozotocin model for diabetes may not resemble the human condition with respect to placental transport functions. Several observations support this possibility. For example, the STZ model in the rat usually results in IUGR rather than accelerated growth (Mughal et al., 1989; Mimouni et al., 1995). The total concentration of calcium and EGTA used in our studies results in free calcium concentrations well above saturating concentrations (Fisher et al., 1987). Our uptake rates of 3–4 nmol/min/mg in control vesicles are similar to the Vmax reported by Fisher et al. (Fisher et al., 1987). Therefore, it appears likely that the increase in transport observed in IUGR and IDDM can be attributed to an increase in Vmax. We have previously shown that ATP-dependent calcium transport in BM increases during third trimester (Strid and Powell, 2000) and, similar to the changes in IUGR and IDDM in the present study, this increase was due to an activation of existing transporters rather than an increased PMCA protein expression. These observations suggest that transporter modification, typically changes in phophorylation, represent an important
Strid et al.: Placental Ca2+ ATPase in IUGR and Diabetes
mechanism of regulation for BM Ca2+ ATPase. The alterations in BM Ca2+ ATPase activity are likely to be mediated by calcium regulating hormones. One potential candidate is PTHrP midmolecule, which has been shown to stimulate BM Ca2+ ATPase (Strid et al., 2002). Furthermore, it has been shown that PTHrP midmolecule is responsible for the maintenance of the calcium gradient across the placenta both in sheep and mouse (Care et al., 1990; Kronenberg et al., 1998). Expression of PTHrP in foetal membranes and placenta is increased in preterm IUGR (Curtis et al., 2000) compatible with a role for this hormone in the increased BM Ca2+ ATPase activity in IUGR. In this study we have, for the first time, measured the PTHrP midmolecule concentrations in plasma of umbilical cord blood from infants born after pregnancies complicated with IUGR or maternal diabetes. We found a marked increase in PTHrP midmolecule plasma concentrations in IUGR. In order to increase BM Ca2+ ATPase activity by 38 per cent in vitro (Strid et al., 2002) much higher
451
concentrations of PTHrP midmolecule were required than measured in IUGR umbilical plasma in the current study. However, this discrepancy does not exclude the possibility that PTHrP midmolecule may mediate the activation of Ca2+ ATPase in BM isolated from IUGR placentae. First, PTHrP midmolecule signalling pathways may be partially inactivated in vitro. Second, PTHrP (38–94) is produced locally, e.g., in the placenta, and umbilical plasma concentrations are unlikely to directly reflect concentrations close to the placental barrier. Therefore, the local concentrations of PTHrp midmolecule in the proximity of the syncytiotrophoblast basal plasma membrane in vivo may be substantially higher than measured in the umbilical circulation. In contrast to IUGR, diabetes in pregnancy was not associated with significant changes in umbilical plasma concentrations of PTHrP midmolecule suggesting that other factors are involved in stimulating BM ATP dependent calcium transport in IDDM.
ACKNOWLEDGEMENTS This work was supported by the Swedish Research Council (14555 and 10838), the Swedish Diabetes Association, the Torsten and Ragnar So¨derbergs Foundation Frimurare-Barnhus-direktionen, the A r hlens Foundation, the General Maternity Hospital Foundation, the Willhelm & Martina Lundgrens Foundation, Samariten Foundation and the Emma and Erik Granes Foundation.
REFERENCES Bowers GN Jr & McComb RB (1966) A continuous spectrophotometric method for measuring the activity of serum alkaline phosphatase. Clin Chem, 12, 70–89. Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem, 72, 248–254. Brans YW, Shannon DL & Hunter MA (1983) Maternal diabetes and neonatal macrosomia. II. Neonatal anthropometric measurements. Early Hum Dev, 8, 297–305. Brunette MG & Leclerc M (1992) Ca2+ transport through the brush border membrane of human placenta syncytiotrophoblasts. Can J Physiol Pharmacol, 70, 835–842. Bucht E, Eklund A, Toss G, Lewensohn R, Granberg B, Sjostedt U, Eddeland R & Torring O (1992) Parathyroid hormone-related peptide, measured by a midmolecule radioimmunoassay, in various hypercalcaemic and normocalcaemic conditions. Acta Endocrinol (Copenh), 127, 294–300. Canalis E (1980) Effect of insulinlike growth factor I on DNA and protein synthesis in cultured rat calvaria. J Clin Invest, 66, 709–719. Care AD, Abbas SK, Pickard DW, Barri M, Drinkhill M, Findlay JB, White IR & Caple IW (1990) Stimulation of ovine placental transport of calcium and magnesium by mid-molecule fragments of human parathyroid hormone-related protein. Exp Physiol, 75, 605–608. Cruikshank DP, Pitkin RM, Varner MW, Williams GA & Hargis GK (1983) Calcium metabolism in diabetic mother, fetus, and newborn infant. Am J Obstet Gynecol, 145, 1010–1016. Curtis NE, King RG, Moseley JM, Ho PW, Rice GE & Wlodek ME (2000) Preterm fetal growth restriction is associated with increased parathyroid hormone-related protein expression in the fetal membranes. Am J Obstet Gynecol, 183, 700–705. Demarini S, Mimouni F, Tsang RC, Khoury J & Hertzberg V (1994) Impact of metabolic control of diabetes during pregnancy on neonatal hypocalcemia: a randomized study. Obstet Gynecol, 83, 918–922. Fisher GJ, Kelley LK & Smith CH (1987) ATP-dependent calcium transport across basal plasma membranes of human placental trophoblast. Am J Physiol, 252, C38–46. Gale CR, Martyn CN, Kellingray S, Eastell R & Cooper C (2001) Intrauterine programming of adult body composition. J Clin Endocrinol Metab, 86, 267–272.
Glazier JD, Jones CJP & Sibley CP (1988) Purification and Na + uptake by human placental microvillus membrane vesicles prepared by three different methods. Biochim Biophys Acta, 945, 127–134. Hamilton K, Tein M, Glazier J, Mawer EB, Berry JL, Balment RJ, Boyd RD, Garland HO & Sibley CP (2000) Altered calbindin mRNA expression and calcium regulating hormones in rat diabetic pregnancy. J Endocrinol, 164, 67–76. Husain SM, Birdsey TJ, Glazier JD, Mughal MZ, Garland HO & Sibley CP (1994) Effect of diabetes mellitus on maternofetal flux of calcium and magnesium and calbindin9K mRNA expression in rat placenta. Pediatr Res, 35, 376–381. Illsley NP, Wang ZQ, Gray A, Sellers MC & Jacobs MM (1990) Simultaneous preparation of paired, syncytial, microvillous and basal membranes from human placenta. Biochim Biophys Acta, 1029, 218–226. Jansson T & Illsley NP (1993) Osmotic water permeabilities of human placental microvillous and basal membranes. J Membr Biol, 132, 147–155. Jansson T, Powell TL & Illsley NP (1993) Non-electrolyte solute permeabilities of human placental microvillous and basal membranes. J Physiol (Lond), 468, 261–274. Jansson T, Scholtbach V & Powell TL (1998) Placental transport of leucine and lysine is reduced in intrauterine growth restriction. Pediatr Res, 44, 532–537. Jansson T, Wennergren M & Illsley NP (1993) Glucose transporter protein expression in human placenta throughout gestation and in intrauterine growth retardation. J Clin Endocrinol Metab, 77, 1554–1562. Jansson T, Ylve´n K, Bjorn C, Wennergren M & Powell TL (2002) Glucose transport and system A activity in syncytiotrophoblast microvillous and basal membranes in intrauterine growth restriction. Placenta, 23, 392–399. Kainer F, Weiss PA, Huttner U, Haas J & Reles M (1997) Levels of amniotic fluid insulin and profiles of maternal blood glucose in pregnant women with diabetes type-I. Early Hum Dev, 49, 97–105. Kamath SG, Kelley LK, Friedman AF & Smith CH (1992) Transport and binding in calcium uptake by microvillous membrane of human placenta. Am J Physiol, 262, C789–794. Kamath SG & Smith CH (1994) Na + /Ca2+ exchange, Ca2+ binding, and electrogenic Ca2+ transport in plasma membranes of human placental syncytiotrophoblast. Pediatr Res, 36, 461–467. Kronenberg HM, Lanske B, Kovacs CS, Chung UI, Lee K, Segre GV, Schipani E & Juppner H (1998) Functional analysis of the PTH/PTHrP network of ligands and receptors. Recent Prog Horm Res, 53, 283–301.
452 Lapillonne A, Guerin S, Braillon P, Claris O, Delmas PD & Salle BL (1997) Diabetes during pregnancy does not alter whole body bone mineral content in infants. J Clin Endocrinol Metab, 82, 3993–3997. Mahendran D, Donnai P, Glazier JD, D’Souza SW, Boyd RD & Sibley CP (1993) Amino acid (System A) transporter activity in microvillous membrane vesicles from the placentas of appropriate and small for gestational age babies. Pediatr Res, 34, 661–665. Marsal K, Persson PH, Larsen T, Lilja H, Selbing A & Sultan B (1996) Intrauterine growth curves based on ultrasonically estimated foetal weights. Acta Paediatr, 85, 843–848. Mimouni F, Loughead J, Miodovnik M, Khoury J & Tsang RC (1990) Early neonatal predictors of neonatal hypocalcemia in infants of diabetic mothers: an epidemiologic study. Am J Perinatol, 7, 203–206. Mimouni F, Mughal Z, Tsang RC & Hammond G (1995) Effect of maternal oxygen therapy on placental calcium transport in intrauterine growth retarded rats. J Am Coll Nutr, 14, 165–168. Mimouni F, Steichen JJ, Tsang RC, Hertzberg V & Miodovnik M (1988) Decreased bone mineral content in infants of diabetic mothers. Am J Perinatol, 5, 339–343. Mimouni F, Tsang RC, Hertzberg VS & Miodovnik M (1986) Polycythemia, hypomagnesemia, and hypocalcemia in infants of diabetic mothers. Am J Dis Child, 140, 798–800. Minton SD, Steichen JJ & Tsang RC (1983) Decreased bone mineral content in small-for-gestational-age infants compared with appropriate-forgestational-age infants: normal serum 25-hydroxyvitamin D and decreasing parathyroid hormone. Pediatrics, 71, 383–388. Mughal MZ, Ross R & Tsang RC (1989) Clearance of calcium across in situ perfused placentas of intrauterine growth-retarded rat fetuses. Pediatr Res, 25, 420–422.
Placenta (2003), Vol. 24 Namgung R, Tsang RC, Specker BL, Sierra RI & Ho ML (1993) Reduced serum osteocalcin and 1,25-dihydroxyvitamin D concentrations and low bone mineral content in small for gestational age infants: evidence of decreased bone formation rates. J Pediatr, 122, 269–275. Norberg S, Powell TL & Jansson T (1998) Intrauterine growth restriction is associated with a reduced activity of placental taurine transporters. Pediatr Res, 44, 233–238. Oh W (1977) Considerations in neonates eith intrauterin growth retardation. In Clinical Obstetrics and Gynecology (Ed.) Frigoletto FD Jr, p. 989. Hagerstown: Haper & Row. Schultz GJK (1984) Adenylate cyclase. In Methods of Enzymatiic Analysis (Ed.) Bergmeyer H, pp. 369–378. Weinham: Verlag Chemie. Sosenko IR, Kitzmiller JL, Loo SW, Blix P, Rubenstein AH & Gabbay KH (1979) The infant of the diabetic mother: correlation of increased cord C-peptide levels with macrosomia and hypoglycemia. N Engl J Med, 301, 859–862. Strid H, Care AF, Jansson T & Powell TL (2002) Parathyroid hormonerelated Peptide (38-94) amide stimulates ATP-dependent calcium transport in the Basal plasma membrane of the human syncytiotrophoblast. J Endocrinol, 175(2), 517–524. Strid H & Powell TL (2000) ATP-dependent Ca2+ transport is up-regulated during third trimester in human syncytiotrophoblast basal membranes. Pediatr Res, 48, 58–63. Tsang RC, Gigger M, Oh W & Brown DR (1975) Studies in calcium metabolism in infants with intrauterine growth retardation. J Pediatr, 86, 936–941. White P (1949) Pregnancy complicating diabetes. Am J Med, 7, 609. Yano H, Ohya K & Amagasa T (1994) Effects of insulin on in vitro bone formation in fetal rat parietal bone. Endocr J, 41, 293–300.
ATP Dependent Ca2+ Transport Across Basal Membrane of Human Syncytiotrophoblast in Pregnancies Complicated by Intrauterine Growth Restriction or Diabetes H. Stridade, E. Buchtc, T. Janssona, M. Wennergrenb and T. L. Powella a
Perinatal Center, Department of Physiology & Pharmacology and b Department of Obstetrics and Gynecology, Go¨teborg University, Sweden, c Department of Molecular Medicine, Karolinska Hospital, Karolinska Institute, Stockholm, Sweden and d Department of Clinical Medicine, O } rebro University, Sweden
Neonates born after pregnancies complicated by diabetes or intrauterine growth restriction (IUGR) have increased incidence of hypocalcaemia. Furthermore, IUGR is associated with reduced bone mineralization in infancy and osteoporosis in adult life. We tested the hypothesis that placental calcium transport is altered in these pregnancy complications. Transport of calcium into syncytiotrophoblast basal plasma membrane (BM) vesicles was studied by rapid filtration and protein expression of Ca2+ ATPase by Western blot. In IUGR Ca2+ ATPase activity was increased by 48 per cent (n=13; P<0.05) whereas protein expression was 15 per cent lower (n=13; P<0.05) than in controls (n=16). Basal membrane ATP dependent calcium transport was unaltered in gestational diabetes (GDM) but increased by 54 per cent in insulin dependent diabetes (IDDM) compared to controls (P<0.05; n=14). Diabetes did not affect Ca2+ ATPase expression in BM. We have previously shown that the mid-molecular fragment of parathyroid hormone related peptide (PTHrP midmolecule) stimulates BM Ca2+ ATPase in vitro. PTHrP midmolecule concentrations in umbilical cord plasma were measured using radioimmunoassay. The concentrations in umbilical cord plasma were increased in IUGR, but unaltered in diabetes. In conclusion, placental calcium pump is activated in IUGR and IDDM, which may be secondary to increased foetal calcium demand. We speculate that PTHrP midmolecule may be one mechanism for 2003 Published by Elsevier Science Ltd. activating BM Ca2+ ATPase in IUGR. Placenta (2003), 24, 445–452
INTRODUCTION Neonates born after pregnancies complicated by insulindependent diabetes (IDDM) or intrauterine growth restriction (IUGR) frequently show signs of disturbed calcium homeostasis. Umbilical blood levels of calcium are reduced at birth in association with IDDM (Cruikshank et al., 1983) and the infants are at risk to develop hypocalcaemia during the first days of life (Cruikshank et al., 1983; Mimouni et al., 1986; Mimouni et al., 1990; Demarini et al., 1994). Although one report has suggested a decreased bone mineralization in infants of mothers with IDDM (Mimouni et al., 1988) more recent studies using dual X-ray absorptiometry show that IDDM is associated with an increased whole body bone mineral content in the infant (Lapillonne et al., 1997). The incidence of neonatal hypocalcaemia is also increased in IUGR (Oh, 1977), however this may be related to birth asphyxia rather than IUGR per se (Tsang et al., 1975). Reduced bone mineral e
To whom correspondence should be addressed at: Department of Clinical Medicine, O } rebro University, 701 85 O } rebro, Sweden. Tel.: 46-19-602 66 63; Fax: 46-19-602 66 50; E-mail: [email protected] 0143–4004/03/$-see front matter
content is a common finding in the IUGR infant (Minton et al., 1983; Namgung et al., 1993). In addition, low birth weight has been shown to be associated with a reduced bone mineral content in the lumbar spine and femoral neck in woman and men aged 70 years (Gale et al., 2001). These findings raise the possibility that IUGR represents a risk factor for osteoporosis later in life. The factors causing the disturbed calcium homeostasis in IUGR infants and infants born to women with IDDM are not well established. The regulation of serum calcium concentrations in the foetus is complex and dependent on placental transport capacity as well as maternal and foetal calcium homeostasis. The possibility that placental transport of calcium is altered in pregnancy complications has been addressed in some detail in animal experimental models. Streptozotocin induced diabetes in the rat is associated with foetal hypocalcaemia and decreased maternal-to-foetal calcium transport (Husain et al., 1994), which may be due to reduced expression of placental calbindin, the primary cytosolic calcium binding protein in the rat placenta (Hamilton et al., 2000). Also IUGR induced by uterine artery ligation in the rat is characterized by a decreased placental transport of calcium (Mughal et al., 1989; 2003 Published by Elsevier Science Ltd.
446
Mimouni et al., 1995). However, placental transport of calcium in association with IUGR or diabetes has not been studied in the human. In the human placenta the syncytiotrophoblast cell (ST) represents the primary functional barrier for transplacental transport between maternal and foetal plasma. In particular, the rate of calcium transport depends on the transport characteristics of the two polarized plasma membranes of the ST and the expression of intracellular calcium binding proteins. The specific cytosolic calcium binding proteins involved in human placental calcium handling are not well defined. In contrast, calcium transport mechanisms across the microvillous (MVM) and basal plasma membrane (BM) have been studied in some detail. Both MVM and BM contain several ATP independent as well as ATP dependent transport mechanisms (Fisher et al., 1987; Brunette and Leclerc, 1992; Kamath et al., 1992; Kamath and Smith, 1994), including Na + /Ca2+ exchange (Kamath and Smith, 1994). In BM, Ca2+ ATPase is the primary transporter mediating the extrusion of calcium out of the cell (Fisher et al., 1987; Strid and Powell, 2000) representing a crucial step in the establishment of the foeto-maternal calcium gradient. PTHrP midmolecule may play a key role in regulating transport of calcium across the placenta since it has been shown to maintain the calcium gradient across the placenta in some animals (Care et al., 1990; Kronenberg et al., 1998) and we recently reported that PTHrP mid molecule stimulates Ca2+ ATPase in BM vesicles isolated from human placenta (Strid et al., 2002). In addition, expression of PTHrP in foetal membranes and placenta is increased in preterm IUGR (Curtis et al., 2000). In this study we tested the hypothesis that pregnancies complicated by IUGR and diabetes are associated with an altered ATP dependent calcium transport across the basal plasma membrane of the syncytiotrophoblast. Basal plasma membranes were isolated from placentae obtained from normal and complicated pregnancies. ATP dependent calcium transport was studied by rapid filtration techniques and the protein expression of plasma membrane Ca2+ ATPase (PMCA) using Western blot. Furthermore, we measured umbilical cord plasma concentrations of the calcium regulatory hormone PTHrP midmolecule.
MATERIAL AND METHODS Placental tissue The collection of placental tissue was approved by the Committee for Research Ethics at Go¨teborg University. Placentae were obtained from pregnancies complicated by diabetes (gestational diabetes, GDM and insulin-dependent diabetes, IDDM) or intrauterine growth restriction (IUGR) as well as from uncomplicated pregnancies (controls). Screening for GDM was carried out routinely in all pregnancies by measurements of capillary blood–glucose in the nonfasting state at least at four occasions starting in weeks
Placenta (2003), Vol. 24
8–12 of gestation. If non-fasting blood–glucose level was _7.0 mmol/l, patients were referred to a 75 g oral glucose tolerance test (OGTT) after overnight fasting. GDM was defined as a fasting blood–glucose level of 6.1 mmol/l or blood–glucose level of 9.0 mmol/l 2 h after OGTT in a pregnant woman without prior known diabetes. These pregnancies were not associated with any major complications other than GDM. Assessment of blood–glucose control was performed by self-monitoring fasting and postprandial blood– glucose levels at least twice a week. Patients with gestational diabetes were treated with diet adjustment only (n=20; Diabetes White A1), except in seven patients in which insulin was administered in a 5-dose regime (Diabetes White A2). Criteria for initiating insulin therapy was elevated blood– glucose levels (fasting glucose=6.1 and/or postprandial blood–glucose=8.0) despite consistent dietary adjustments. Patients in the IDDM group had no other major complication and were classified as White B, C or D (White, 1949). All women with IDDM attended the obstetric antenatal clinic at Sahlgren’s University Hospital, Go¨teborg. Patients carried out self-monitoring of capillary blood–glucose concentrations at least four times daily and glycosylated haemoglobin (Hb A1C) was measured every 2–6 weeks. All patients in the IDDM group received insulin in a 5-dose regime. Neonates large for gestational age (LGA) were defined as having a birth weight greater than mean plus 2 s.d. using intrauterine growth curves based on ultrasonically estimated foetal weight (Marsal et al., 1996). Similarly, IUGR was defined as a birth weight 2 s.d. below the mean for that gestational age (Marsal et al., 1996).
Isolation of syncytiotrophoblast plasma membranes Placentae were placed on ice immediately after delivery and the membrane isolation procedure was started within 30 min. Basal plasma membranes (BM) were prepared according to a well characterized protocol (Illsley et al., 1990; Jansson and Illsley, 1993; Jansson et al., 1993) with some modifications (Strid and Powell, 2000). Briefly, after initial homogenization and centrifugation steps (at 4C in 250 m sucrose, 10 m Hepes/Tris, protease inhibitors, pH 7.4) BM were separated from MVM by Mg2+ precipitation and further purified on a sucrose step gradient. Samples were frozen in liquid nitrogen and stored at 80C. BM enrichment was assessed using standard activity assays for adenylate cyclase (Schultz, 1984) and alkaline phosphatase (Bowers and McComb, 1966). The production of cAMP by adenylate cyclase was measured by radioimmunoassay (New England Nuclear, Boston, MA, USA). Orientation assay for BM vesicles was carried out by measuring alkaline phosphatase activity prior to and after permeabilization of membranes with 1 per cent Triton-X, an approach similar to the technique used by Glazier and coworkers (Glazier et al., 1988). The ratio of alkaline phosphatase activity before and after permeabilization represents a measure of right-side-out vesicles.
Strid et al.: Placental Ca2+ ATPase in IUGR and Diabetes
Western blot Gel electrophoresis (20 g membrane protein/lane) and transfer were carried out as described elsewhere (Strid and Powell, 2000). The monoclonal primary antibody used (Clone 5F10,ABR) is directed to all four known isoforms of the plasma membrane Ca2+ ATPase. Detection was performed using the ECL-system (Amersham, Buckinghamshire, UK) according to the manufacturers directions. Densitometry of the autoradiography for PMCA expression was performed using IP Lab gel (Signaling Analytics Corporation, Vienna, VA, USA). Only the prominent 140 kDA band was analysed when comparing PMCA expression between groups. The densitometric values fell within the linear range of detection. Each gel contained nine samples, two of which were the same on all the gels. The mean density of these two samples was used to normalize all the gels. No bands were observed in control experiments in the absence of primary antibody.
Calcium uptake The uptake of calcium by basal membrane vesicles was measured by standard rapid filtration techniques (Fisher et al., 1987) with some modification (Strid and Powell, 2000). Briefly, the vesicles were resuspended in vesicle buffer (10 m Hepes/ Tris; 250 m sucrose; pH 7.4 at 37C) and centrifuged. After centrifugation the vesicles were resuspended in vesicle buffer. Calcium uptake measurements were started by the addition of 20 l of vesicles (100–200 g total membrane protein) to 500 l calcium transport buffer (240 m sucrose; 5 m MgCl2; 0.2 m EGTA; 0.2 m CaCl2; 1 Ci 45CaCl2; 10 m Hepes/ Tris, pH 7.4) in the presence or absence of 5 m ATP–MgCl2. Calcium uptake was allowed to proceed for 10 min at 37C and stopped by the addition of ice cold wash solution (10 m Hepes/Tris; 250 m Sucrose; 4 m EGTA; pH 7.4 at 4C) to approximate initial rate (Strid and Powell, 2000). Subsequently, the reaction mixture was filtered under vacuum on 0.45 m filters (HAWO, Millipore). The radioactivity on the filter was measured by liquid scintillation. The ATP dependent calcium uptake was calculated as the difference between Ca2+ uptake in presence and absence of ATP. Calcium uptake was expressed as nmoles of Ca2+ /mg total membrane protein.
Protein determination Protein concentration was determined according to the method of Bradford (Bradford, 1976), using the Bio Rad (Hercules, CA, USA) assay procedure and BSA as a standard.
Umbilical cord blood collection Umbilical cord blood (10 ml pooled venous and arterial blood) was collected for PTHrP analysis, within 10 min after delivery, in tubes containing protease inhibitors (PTHrP Cocktail
447
Table 1. Clinical characteristics of IUGR group and corresponding control group
n Gestational age (wks) Placental weight (g) Birth weight (g)
Controls
IUGR
16 37.20.5 64643 3103254
13 36.10.8 34923* 1665132*
Means. *P<0.05 as compared with controls. IUGR, intrauterine growth restriction.
Tubes; Nicholas Institute). Samples were obtained from normal term pregnancies as well as from pregnancies complicated by diabetes or IUGR. The collected blood was centrifuged for 15 min at 2000 rpm at 4C, plasma was separated and frozen at 80C until radioimmunoassays were performed. Radioimmunoassay A synthetic midmolecular fragment of PTHrP 63–77 was used for antibody production in rabbits, as reference standard and for radioiodination. The samples were extracted by Sep-PAK C18 cartridge (Waters, Milford, MA, USA) as described previously with the RIA (Bucht et al., 1992). The detection limit of the assay, based on 2 s.d. below maximal binding, was after extraction 0.8 pg/ml. Intraassay variation was <9 per cent, and interassay variation was <15 per cent at all concentrations. Data analysis Statistical significance of differences between groups was determined utilizing t-test or ANOVA followed by Dunnett test. Data are expressed as mean unless otherwise specified. RESULTS Clinical data Gestational age, placental weight and birth weight for all groups are presented in Tables 1 and 2. Both birth weight and placental weight were lower in the IUGR group as compared to controls (P<0.05; n=13–16), however gestational age was not different between groups (Table 1). In the IDDM group the birth weight was significantly higher than in the control group (Table 2; P<0.05; n=14–27). Characterization of basal membrane preparations In BM isolated from control placentae, the enrichment in adenylate cyclase activity was 29–32 fold while the enrichment
448
Placenta (2003), Vol. 24
Table 2. Clinical characteristics of diabetes groups and corresponding control groups
n Gestational age (wks) Placental weight (g) Birth weight (g)
Control
GDM
IDDM
37 38.80.3 71331 351499
27 39.00.3 66732 3618103
14 38.30.4 85575 397213*
Means. *P<0.05 as compared to controls. GDM, gestational diabetes mellitus; IDDM, insulin dependent diabetes mellitus.
Table 3. Marker enzyme activities in diabetes and control groups Adenylate cyclase (pmol/minmg)
Control GDM IDDM
Alkaline phosphatase (nmol PO4/smg)
n
BM
P2
n
BM
Hom
18 13 13
570.783.8 (28.8) 495.658.3 (30.3) 488.261.7 (31.8)
17.94.4 16.43.9 16.96.5
18 24 12
9.00.9 (2.6) 8.50.8 (2.3) 8.20.6 (2.1)
3.50.5 3.60.4 3.90.4
Values are given as means. Enzyme enrichments (in parentheses) were calculated as vesicle enzyme specific activity relative to that for the homogenate (alkaline phosphatase) or the post nuclear membrane pellet (P2; adenylate cyclase).
in alkaline phosphatase activity was 2.5 fold (n=8–9). Activities of marker enzymes were not significantly different in BM isolated from complicated pregnancies. In BM isolated from placentae collected from pregnancies complicated with IUGR, the enrichment in adenylate cyclase activity was 28 fold while the enrichment in alkaline phosphatase activity was 2.4 fold (n=10–11). Furthermore, enrichments of marker enzymes in IDDM and GDM groups were similar to BM obtained from control placentae (Table 3). We have previously shown that the degree of contamination by endoplasmic reticulum and mitochondria is low in our BM preparation and that enrichments in BM preparations from premature placentae are not different from term placentae (Strid and Powell, 2000). The ratio of alkaline phosphatase activity before and after permeabilization was 0.800.08 (n=6) in control BM, indicating that approximately 80 per cent of vesicles were oriented right-side-out. This value is similar to the 7312 per cent reported previously for BM prepared according to the protocol of Illsley and coworkers (Illsley et al., 1990). Furthermore, the fraction of BM vesicles oriented right-side-out in IUGR (7612 per cent, n=5), GD (859 per cent, n=5) and IDDM groups (876 per cent, n=7) was not significantly different from control vesicles.
Ca2+ transport As shown previously (Strid and Powell, 2000) the ATPindependent Ca2+ uptake was approximately 100 times lower than the ATP-dependent transport under these experimental conditions and did not differ between groups (data not shown). The ATP dependent transport of calcium in BM isolated from
Figure 1. ATP dependent calcium transport in IUGR BM (n=13) in comparison with control BM (n=16; *P<0.05; t-test).
IUGR placentae was 48 per cent higher as compared to control (P<0.05; n=13–16; Figure 1). In BM isolated from GDM pregnancies ATP dependent calcium transport was unaltered. In contrast, IDDM was associated with a 54 per cent increase in ATP dependent transport compared to controls (P<0.05; n=14–27; Figure 2). Further subgrouping of the diabetic placentae showed no significant differences in the calcium transport between IDDM and IDDM/LGA or GDM and GDM/LGA (data not shown). PMCA protein expression We used Western blot to investigate possible changes in Ca2+ ATPase expression per mg total protein in BM isolated from
Strid et al.: Placental Ca2+ ATPase in IUGR and Diabetes
449
Figure 5. Relative expression of PMCA in BM isolated from pregnancies complicated by GDM (n=12) and IDDM (n=13) in comparison with BM isolated from control group (n=18; ANOVA). Figure 2. ATP dependent calcium transport into BM isolated from control (n=37), GDM (n=27) or IDDM (n=14) groups (*P<0.05 as compared to controls; ANOVA).
Figure 3. Autoradiography of Western blot of membrane Ca2+ ATPase (PMCA) in BM isolated from IUGR pregnancies (I) and BM isolated from normal pregnancies (C).
Figure 6. Plasma concentrations of PTHrP midmolecule in umbilical cord blood, measured by radioimmunoassay using an antibody directed towards residues 63–77. Blood was collected from infants born after uncomplicated pregnancies (control; n=13), pregnancies complicated by growth restriction (IUGR; n=6) or maternal diabetes (n=5) *P<0.05 as compared to controls ANOVA.
PTHrP midmolecule concentration measurements in umbilical cord blood plasma
Figure 4. Relative expression of plasma membrane Ca2+ ATPase (PMCA) in BM isolated from IUGR (n=13) pregnancies in comparison with BM isolated from normal pregnancies (n=13; *P<0.05; t-test)
pregnancies complicated with IDDM, GDM or IUGR compared to controls. Figure 3 illustrates a Western blot with IUGR (I) and control (C) samples. PMCA protein expression in IUGR was decreased by 15 per cent compared to controls (Figure 4; P<0.05; n=13). However, there was no change in the PMCA expression in the two diabetic groups compared to control (Figure 5).
The PTHrP midmolecule concentration in umbilical cord blood plasma from infants born after uncomplicated term pregnancies was 5.80.6 pg/ml (n=13). PTHrP midmolecule concentrations in umbilical cord blood plasma collected from infants born after pregnancies complicated by diabetes were not significantly different compared to the control group. In contrast, babies born growth restricted had a 67 per cent increase in umbilical plasma PTHrP concentrations compared to controls (P<0.05, n=6). Figure 6 illustrates the differences in plasma concentrations between the three groups.
DISCUSSION Pregnancies complicated by IUGR or IDDM are associated with foetal/neonatal hypocalcaemia (Oh, 1977; Cruikshank et al., 1983; Mimouni et al., 1986; Mimouni et al., 1990; Demarini et al., 1994) and IUGR infants show signs of reduced bone mineralization (Minton et al., 1983; Namgung et al., 1993). Somewhat surprising, the main findings in this
450
study were that ATP dependent Ca2+ transport across syncytiotrophoblast basal plasma membrane vesicles is increased in IUGR and IDDM consistent with an increased transplacental transport of calcium in these pregnancy complications. Therefore these changes are unlikely to be the primary cause of the disturbance in calcium homeostasis in IUGR and IDDM but may represent a compensatory mechanism in response to an imbalance between calcium requirement and calcium supply. In addition, our data indicate that these alterations are due to an activation of existing transporters since protein expression of the calcium pump was unchanged or reduced. Furthermore, we report increased umbilical plasma concentrations of PTHrP midmolecule in association with IUGR and we suggest that PTHrP midmolecule may mediate the activation of Ca2+ ATPase in BM isolated from IUGR placentae. The increased BM Ca2+ ATPase activity in the IUGR placenta does not explain the impaired foetal bone mineralization in this condition. One interpretation of these findings is that the increased BM Ca2+ ATPase activity is secondary to an insuffient calcium supply due to decreased activity or expression of other components of the placental calcium transporting systems, such as cytosolic calcium binding proteins in the syncytiotrophoblast. Furthermore, reduced bone mineralization in IUGR may be caused by an insufficiency of the placental calcium transport system that is relative rather than absolute. This speculation relates to the fact that IUGR foetuses commonly exhibit asymmetric growth restriction, i.e. body weight is reduced much more than body length. Placental weight, and therefore placental exchange area, is correlated to foetal weight rather than length whereas skeletal mass is related to length. Consequently, the demands for calcium of the relatively unaffected skeletal mass in IUGR may be greater than can be provided by the placental supply system, which could lead to stimulation of BM Ca2+ ATPase. If this speculation is correct, the efforts to enhance placental calcium transport are insufficient in many cases of IUGR, explaining the impaired bone mineralization observed in the IUGR infant. IUGR has been shown to be associated with a number of alterations in the expression and/or activity of key placental nutrient and ion transporters. In particular amino acid transporters appear to be affected since IUGR is characterized by a reduced activity of system A (Mahendran et al., 1993) and taurine transporters (Norberg et al., 1998) in MVM but not in BM (Norberg et al., 1998; Jansson et al., 2002), a lower activity of cationic amino acid transporters in BM but not in MVM (Jansson et al., 1998) whereas the activity for transporters for leucine has been demonstrated to be reduced in both syncytiotrophoblast plasma membranes (Jansson et al., 1998). In contrast, the activity and expression of syncytiotrophoblast glucose transporters are unaffected by IUGR (Jansson et al., 1993). The impact of IUGR on BM PMCA activity in the current study is unique since this is the first demonstration of increased activity of a placental transporter in association with this pregnancy complication. This is an important finding, strongly suggesting that the changes observed in placental
Placenta (2003), Vol. 24
nutrient and ion transporters are specific rather than a consequence of a general detrimental effect of the IUGR condition on the placenta. Collectively, these observations indicate that syncytiotrophoblast transporters are subjected to regulation by mechanisms unique for each transporter and are consistent with the possibility that some transporters in the two polarized plasma membranes of the syncytiotrophoblast are regulated independently. Insulin-dependent diabetes is often complicated by accelerated foetal growth as illustrated by the higher birth weights in the IDDM group in our study. Accelerated foetal growth in IDDM is correlated with foetal hyperinsulinaemia (Sosenko et al., 1979; Brans et al., 1983; Kainer et al., 1997). Since insulin stimulates foetal bone formation directly (Yano et al., 1994) as well as mediated via IGF-1 release from the liver (Canalis, 1980) it is possible that the hyperinsulinaemia causes the increased whole body bone mineral content of the large-forgestational age foetus in IDDM pregnancy as discussed by Lapillonne and coworkers (Lapillonne et al., 1997). Therefore the increased activity of Ca2+ ATPase in the placental barrier in association with IDDM demonstrated in this study may reflect a compensatory mechanism to increase placental transport of calcium in order to meet the increased foetal demands. The increased calcium transport across BM in IDDM in vitro is consistent with an increased transplacental transport of calcium in vivo, however firm conclusions must await direct measurements in vivo. Nevertheless, the findings of the present study are in contrast to the observations of decreased placental transport of calcium in experimental models of diabetes in the rat (Husain et al., 1994; Hamilton et al., 2000). One explanation to this discrepancy could be that our interpretation of the increased calcium transport across BM in IDDM is incorrect. Our findings might instead be a consequence of an impairment of other components of the placental calcium transport system, such as a decreased expression of cytosolic calcium binding proteins in the syncytiotrophoblast. Or alternatively, the streptozotocin model for diabetes may not resemble the human condition with respect to placental transport functions. Several observations support this possibility. For example, the STZ model in the rat usually results in IUGR rather than accelerated growth (Mughal et al., 1989; Mimouni et al., 1995). The total concentration of calcium and EGTA used in our studies results in free calcium concentrations well above saturating concentrations (Fisher et al., 1987). Our uptake rates of 3–4 nmol/min/mg in control vesicles are similar to the Vmax reported by Fisher et al. (Fisher et al., 1987). Therefore, it appears likely that the increase in transport observed in IUGR and IDDM can be attributed to an increase in Vmax. We have previously shown that ATP-dependent calcium transport in BM increases during third trimester (Strid and Powell, 2000) and, similar to the changes in IUGR and IDDM in the present study, this increase was due to an activation of existing transporters rather than an increased PMCA protein expression. These observations suggest that transporter modification, typically changes in phophorylation, represent an important
Strid et al.: Placental Ca2+ ATPase in IUGR and Diabetes
mechanism of regulation for BM Ca2+ ATPase. The alterations in BM Ca2+ ATPase activity are likely to be mediated by calcium regulating hormones. One potential candidate is PTHrP midmolecule, which has been shown to stimulate BM Ca2+ ATPase (Strid et al., 2002). Furthermore, it has been shown that PTHrP midmolecule is responsible for the maintenance of the calcium gradient across the placenta both in sheep and mouse (Care et al., 1990; Kronenberg et al., 1998). Expression of PTHrP in foetal membranes and placenta is increased in preterm IUGR (Curtis et al., 2000) compatible with a role for this hormone in the increased BM Ca2+ ATPase activity in IUGR. In this study we have, for the first time, measured the PTHrP midmolecule concentrations in plasma of umbilical cord blood from infants born after pregnancies complicated with IUGR or maternal diabetes. We found a marked increase in PTHrP midmolecule plasma concentrations in IUGR. In order to increase BM Ca2+ ATPase activity by 38 per cent in vitro (Strid et al., 2002) much higher
451
concentrations of PTHrP midmolecule were required than measured in IUGR umbilical plasma in the current study. However, this discrepancy does not exclude the possibility that PTHrP midmolecule may mediate the activation of Ca2+ ATPase in BM isolated from IUGR placentae. First, PTHrP midmolecule signalling pathways may be partially inactivated in vitro. Second, PTHrP (38–94) is produced locally, e.g., in the placenta, and umbilical plasma concentrations are unlikely to directly reflect concentrations close to the placental barrier. Therefore, the local concentrations of PTHrp midmolecule in the proximity of the syncytiotrophoblast basal plasma membrane in vivo may be substantially higher than measured in the umbilical circulation. In contrast to IUGR, diabetes in pregnancy was not associated with significant changes in umbilical plasma concentrations of PTHrP midmolecule suggesting that other factors are involved in stimulating BM ATP dependent calcium transport in IDDM.
ACKNOWLEDGEMENTS This work was supported by the Swedish Research Council (14555 and 10838), the Swedish Diabetes Association, the Torsten and Ragnar So¨derbergs Foundation Frimurare-Barnhus-direktionen, the A r hlens Foundation, the General Maternity Hospital Foundation, the Willhelm & Martina Lundgrens Foundation, Samariten Foundation and the Emma and Erik Granes Foundation.
REFERENCES Bowers GN Jr & McComb RB (1966) A continuous spectrophotometric method for measuring the activity of serum alkaline phosphatase. Clin Chem, 12, 70–89. Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem, 72, 248–254. Brans YW, Shannon DL & Hunter MA (1983) Maternal diabetes and neonatal macrosomia. II. Neonatal anthropometric measurements. Early Hum Dev, 8, 297–305. Brunette MG & Leclerc M (1992) Ca2+ transport through the brush border membrane of human placenta syncytiotrophoblasts. Can J Physiol Pharmacol, 70, 835–842. Bucht E, Eklund A, Toss G, Lewensohn R, Granberg B, Sjostedt U, Eddeland R & Torring O (1992) Parathyroid hormone-related peptide, measured by a midmolecule radioimmunoassay, in various hypercalcaemic and normocalcaemic conditions. Acta Endocrinol (Copenh), 127, 294–300. Canalis E (1980) Effect of insulinlike growth factor I on DNA and protein synthesis in cultured rat calvaria. J Clin Invest, 66, 709–719. Care AD, Abbas SK, Pickard DW, Barri M, Drinkhill M, Findlay JB, White IR & Caple IW (1990) Stimulation of ovine placental transport of calcium and magnesium by mid-molecule fragments of human parathyroid hormone-related protein. Exp Physiol, 75, 605–608. Cruikshank DP, Pitkin RM, Varner MW, Williams GA & Hargis GK (1983) Calcium metabolism in diabetic mother, fetus, and newborn infant. Am J Obstet Gynecol, 145, 1010–1016. Curtis NE, King RG, Moseley JM, Ho PW, Rice GE & Wlodek ME (2000) Preterm fetal growth restriction is associated with increased parathyroid hormone-related protein expression in the fetal membranes. Am J Obstet Gynecol, 183, 700–705. Demarini S, Mimouni F, Tsang RC, Khoury J & Hertzberg V (1994) Impact of metabolic control of diabetes during pregnancy on neonatal hypocalcemia: a randomized study. Obstet Gynecol, 83, 918–922. Fisher GJ, Kelley LK & Smith CH (1987) ATP-dependent calcium transport across basal plasma membranes of human placental trophoblast. Am J Physiol, 252, C38–46. Gale CR, Martyn CN, Kellingray S, Eastell R & Cooper C (2001) Intrauterine programming of adult body composition. J Clin Endocrinol Metab, 86, 267–272.
Glazier JD, Jones CJP & Sibley CP (1988) Purification and Na + uptake by human placental microvillus membrane vesicles prepared by three different methods. Biochim Biophys Acta, 945, 127–134. Hamilton K, Tein M, Glazier J, Mawer EB, Berry JL, Balment RJ, Boyd RD, Garland HO & Sibley CP (2000) Altered calbindin mRNA expression and calcium regulating hormones in rat diabetic pregnancy. J Endocrinol, 164, 67–76. Husain SM, Birdsey TJ, Glazier JD, Mughal MZ, Garland HO & Sibley CP (1994) Effect of diabetes mellitus on maternofetal flux of calcium and magnesium and calbindin9K mRNA expression in rat placenta. Pediatr Res, 35, 376–381. Illsley NP, Wang ZQ, Gray A, Sellers MC & Jacobs MM (1990) Simultaneous preparation of paired, syncytial, microvillous and basal membranes from human placenta. Biochim Biophys Acta, 1029, 218–226. Jansson T & Illsley NP (1993) Osmotic water permeabilities of human placental microvillous and basal membranes. J Membr Biol, 132, 147–155. Jansson T, Powell TL & Illsley NP (1993) Non-electrolyte solute permeabilities of human placental microvillous and basal membranes. J Physiol (Lond), 468, 261–274. Jansson T, Scholtbach V & Powell TL (1998) Placental transport of leucine and lysine is reduced in intrauterine growth restriction. Pediatr Res, 44, 532–537. Jansson T, Wennergren M & Illsley NP (1993) Glucose transporter protein expression in human placenta throughout gestation and in intrauterine growth retardation. J Clin Endocrinol Metab, 77, 1554–1562. Jansson T, Ylve´n K, Bjorn C, Wennergren M & Powell TL (2002) Glucose transport and system A activity in syncytiotrophoblast microvillous and basal membranes in intrauterine growth restriction. Placenta, 23, 392–399. Kainer F, Weiss PA, Huttner U, Haas J & Reles M (1997) Levels of amniotic fluid insulin and profiles of maternal blood glucose in pregnant women with diabetes type-I. Early Hum Dev, 49, 97–105. Kamath SG, Kelley LK, Friedman AF & Smith CH (1992) Transport and binding in calcium uptake by microvillous membrane of human placenta. Am J Physiol, 262, C789–794. Kamath SG & Smith CH (1994) Na + /Ca2+ exchange, Ca2+ binding, and electrogenic Ca2+ transport in plasma membranes of human placental syncytiotrophoblast. Pediatr Res, 36, 461–467. Kronenberg HM, Lanske B, Kovacs CS, Chung UI, Lee K, Segre GV, Schipani E & Juppner H (1998) Functional analysis of the PTH/PTHrP network of ligands and receptors. Recent Prog Horm Res, 53, 283–301.
452 Lapillonne A, Guerin S, Braillon P, Claris O, Delmas PD & Salle BL (1997) Diabetes during pregnancy does not alter whole body bone mineral content in infants. J Clin Endocrinol Metab, 82, 3993–3997. Mahendran D, Donnai P, Glazier JD, D’Souza SW, Boyd RD & Sibley CP (1993) Amino acid (System A) transporter activity in microvillous membrane vesicles from the placentas of appropriate and small for gestational age babies. Pediatr Res, 34, 661–665. Marsal K, Persson PH, Larsen T, Lilja H, Selbing A & Sultan B (1996) Intrauterine growth curves based on ultrasonically estimated foetal weights. Acta Paediatr, 85, 843–848. Mimouni F, Loughead J, Miodovnik M, Khoury J & Tsang RC (1990) Early neonatal predictors of neonatal hypocalcemia in infants of diabetic mothers: an epidemiologic study. Am J Perinatol, 7, 203–206. Mimouni F, Mughal Z, Tsang RC & Hammond G (1995) Effect of maternal oxygen therapy on placental calcium transport in intrauterine growth retarded rats. J Am Coll Nutr, 14, 165–168. Mimouni F, Steichen JJ, Tsang RC, Hertzberg V & Miodovnik M (1988) Decreased bone mineral content in infants of diabetic mothers. Am J Perinatol, 5, 339–343. Mimouni F, Tsang RC, Hertzberg VS & Miodovnik M (1986) Polycythemia, hypomagnesemia, and hypocalcemia in infants of diabetic mothers. Am J Dis Child, 140, 798–800. Minton SD, Steichen JJ & Tsang RC (1983) Decreased bone mineral content in small-for-gestational-age infants compared with appropriate-forgestational-age infants: normal serum 25-hydroxyvitamin D and decreasing parathyroid hormone. Pediatrics, 71, 383–388. Mughal MZ, Ross R & Tsang RC (1989) Clearance of calcium across in situ perfused placentas of intrauterine growth-retarded rat fetuses. Pediatr Res, 25, 420–422.
Placenta (2003), Vol. 24 Namgung R, Tsang RC, Specker BL, Sierra RI & Ho ML (1993) Reduced serum osteocalcin and 1,25-dihydroxyvitamin D concentrations and low bone mineral content in small for gestational age infants: evidence of decreased bone formation rates. J Pediatr, 122, 269–275. Norberg S, Powell TL & Jansson T (1998) Intrauterine growth restriction is associated with a reduced activity of placental taurine transporters. Pediatr Res, 44, 233–238. Oh W (1977) Considerations in neonates eith intrauterin growth retardation. In Clinical Obstetrics and Gynecology (Ed.) Frigoletto FD Jr, p. 989. Hagerstown: Haper & Row. Schultz GJK (1984) Adenylate cyclase. In Methods of Enzymatiic Analysis (Ed.) Bergmeyer H, pp. 369–378. Weinham: Verlag Chemie. Sosenko IR, Kitzmiller JL, Loo SW, Blix P, Rubenstein AH & Gabbay KH (1979) The infant of the diabetic mother: correlation of increased cord C-peptide levels with macrosomia and hypoglycemia. N Engl J Med, 301, 859–862. Strid H, Care AF, Jansson T & Powell TL (2002) Parathyroid hormonerelated Peptide (38-94) amide stimulates ATP-dependent calcium transport in the Basal plasma membrane of the human syncytiotrophoblast. J Endocrinol, 175(2), 517–524. Strid H & Powell TL (2000) ATP-dependent Ca2+ transport is up-regulated during third trimester in human syncytiotrophoblast basal membranes. Pediatr Res, 48, 58–63. Tsang RC, Gigger M, Oh W & Brown DR (1975) Studies in calcium metabolism in infants with intrauterine growth retardation. J Pediatr, 86, 936–941. White P (1949) Pregnancy complicating diabetes. Am J Med, 7, 609. Yano H, Ohya K & Amagasa T (1994) Effects of insulin on in vitro bone formation in fetal rat parietal bone. Endocr J, 41, 293–300.