maternal blood vessels and sinuses (Ramsey, 1975;. Hogan et al. ..... S;E 40. S o. Control -ATP +PHE. Fig. 11. Calcium uptake by cell-free placental membranes.
505
Development 110, 505-513 (1990) Printed in Great Britain © The Company of Biologists Limited 1990
Ca
-activated ATPase of the mouse chorioailantoic placenta:
developmental expression, characterization and cytohistochemical localization
ROCKY S. TUAN and NEIL BIGIONI Department of Biology, University of Pennsylvania, Philadelphia, PA 19104, USA and Department of Orthopaedic Surgery, Thomas Jefferson University, Philadelphia, PA 19107, USA
Summary
A membrane-associated, Ca2+-activated, Mg2+-dependent ATPase activity has been identified in the mouse chorioailantoic placenta. The enzyme activity is expressed and increases as a function of gestation. Biochemical characterization shows that the enzyme is highly specific for Ca 2+ and nucleotide triphosphates, with a Km of 0.97 mM [Ca2+] and a V ^ of 1.05 nmol P, released mg"1 placental protein min" 1 . The mouse placental Ca2+-ATPase activity has a pi of approximately 6.8, and corresponds to two apparent MT values of 118 and 150 xlO3, based on Ferguson analysis of nondenaturing electrophoretograms. Enzyme activity is in-
hibited by phenothiazin (suggesting a calmodulin dependence), vanadate, erythrosin B and quercetin, but not by ouabain or levamisole. Enzyme cytohistochemistry revealed that the Ca2+-ATPase is localized topolyploid trophoblastic cells of the mouse inner placenta. These results suggest that the enzyme may be a functional component of transplacental calcium transport during mouse embryonic development.
Introduction
cantly in tissue architecture. The former is characterized by well-defined chorionic villi enveloped in the maternal circulation, whereas the latter consists of a labyrinth of polyploidal trophoblasts intertwined with maternal blood vessels and sinuses (Ramsey, 1975; Hogan et al. 1986). The human and mouse CaBPs are structurally related since they are immuno-crossreactive (Tuan and Cavanaugh, 1986) and the mouse CaBP cDNA cross-hybridizes with RNAs from both tissues (Tuan and Kirwin, 1988). Interestingly, human and mouse placental CaBPs are both expressed as a function of gestation (Tuan, 1982; Tuan and Cavanaugh, 1986), corresponding to the increased fetal need for calcium during development (Widdowson, 1968; Garel, 1983). Furthermore, the distribution of both CaBPs is localized to the trophoblasts (Tuan, 1985; Tuan and Cavanaugh, 1986), the cell type believed to be responsible for nutrient transport in the placenta (Dearden and Ockleford, 1983). These findings thus suggest that calcium transport by human and mouse placentae may indeed involve a similar, developmentally regulated mechanism. In this study, we have further analyzed the mechanistic aspect of calcium transport in the mouse placenta by examining its Ca2+-ATPase activity, based on the rationale that such an activity is probably also a
During mammalian embryonic development, the fetoplacental structure transfers nutrients from the maternal to the fetal circulation. Although some nutrients are translocated passively (e.g. gases) or diffuse in a facilitated manner (e.g. sugars) by virtue of concentration gradients, most of the nutrients are probably translocated actively, i.e. in an energy-dependent manner, across the placental barrier (e.g. see reviews by Boyd, 1987; Munro et al. 1983; Shennan and Boyd, 1987; Truman and Ford, 1984). Our laboratory has been studying the mechanism of calcium transport in human and mouse placentae by characterizing the biochemical components that may be functional in the transport pathway. Our studies have identified high-Mr, Ca -binding proteins (CaBPs) in human (Tuan, 1982 and 1985) and mouse (Tuan and Cavanaugh, 1986) placentae, and a Ca2+-activated ATPase in the human placenta (Tuan and Kushner, 1987). In addition, the functional involvement of the CaBP and Ca2+-ATPase in calcium transport has been demonstrated in vitro using cell-free membrane vesicles isolated from term human placenta (Tuan, 1985). Although human and mouse placentae are both hemochorial in nature (Ramsey, 1975), they differ signifi-
Key words: embryonic development, membrane transport, calcium homeostasis, trophoblast.
506
R. S. Tuan and N. Bigioni
functional component of the transport machinery, as in the human placenta. We report here that a developmentally expressed Ca2+-activated ATPase activity is indeed present in the mouse placenta that shares many common properties with the human placental enzyme. In addition, the enzyme has been localized histochemically to the trophoblastic cells of the mouse chorioallantoic placenta, consistent with its possible functional involvement in placental calcium transport. Materials and methods Animals and placental tissues Mice (females, C57BL; males SJL) were obtained from the Jackson Laboratory (Bar Harbor, ME, USA). Mating was carried out randomly, and the first presence of a vaginal plug in the female was counted as day 1 of gestation. Animals were killed by cervical dislocation, and the chorioallantoic placenta was dissected and cleared from the amniotic sac and uterine wall, rinsed thoroughly in cold 0.9% NaCl, and used immediately for membrane preparation or extraction. Occasionally, the tissues were kept frozen at — 80°C and then extracted. Tissue extraction and membrane preparation Placental tissues were either extracted directly or processed for microsomal membranes as follows. For direct extraction, the placenta was homogenized (1:4, w/v) in a Triton-Tris buffer (13.7 mM Tris-HCl, pH 7.4, 0.12 M NaCl, 4.74 mM KC1, 98.5/iM glucose, 0.02% NaN3, and 1% Triton X-100; Tuan and Knowles, 1984). The soluble placental extract was defined as the supernate after centrifugation of the homogenate at 31000 g for 30 min. Microsomal membranes were prepared as described previously for human term placenta (Tuan, 1985; Tuan and Kushner, 1987). Briefly, a freshly prepared, isoosmotic homogenate of mouse placentae was fractionated by differential centrifugation, and microsomes were obtained by centrifugation (80000g, 80 min) of the post-mitochondrial fraction. Microsomal membrane vesicles were suspended in 10mM imidazole-HCl, pH7.0, containing 5min MgCl2 and 0.1M KC1 to a concentration of 5-10 mg of protein m l . Assay of Co2*-ATPase activity The ATPase enzyme activity was determined using tissue extract or solubilized microsomal membranes at pH8.0 as described previously (Tuan and Knowles, 1984) using the phosphomolybdate-Malachite Green colorimetric reaction, and based on differential activities in the presence of Ca2+ and ethylene glycol bis(amino-ethyl ether)tetra-acetate (EGTA). Levamisole (lmM) was included in most assays to inhibit placental alkaline phosphatase activity (van Belle, 1972). Enzyme activity was expressed as moles of phosphate released per minute. Protein determination Protein was estimated by the method of Lowry et al. (1951) with bovine serum albumin as standard. Co2*-ATPase cytohistochemistry The procedure used was as described previously (Tuan and Knowles, 1984; Tuan and Kushner, 1987), using 10^m cryosections of frozen-embedded mouse placenta. The incubation mixture contained ATP, lead citrate, CaQ 2 and levamisole to inhibit alkaline phosphatase. The reaction product of lead phosphate was visualized by incubation with (NH^S. Con-
trols included omission (or substitution) of ATP (ADP), CaCl2 (EGTA), or lead citrate from the incubation mixture. All stained sections were mounted in glycerine for microscopic observation. Co2*-ATPase zymogram Solubilized samples were subjected to non-denaturing polyacrylamide gel electrophoresis (Tuan and Knowles, 1984), or isoelectric focusing in polyacrylamide gel using the LKB Multiphor unit as described previously (Ono and Tuan, 1986). For isoelectric focusing, ampholytes (IsoLytes, IsoLab) covering the range of pH3-10 were used. After electrophoretic or isoelectric fractionation, the gels were rinsed and reacted histochemically for Ca2+-ATPase as described previously (Tuan and Knowles, 1984). The pH gradient of the focusing gel was determined using a flat-surface pH electrode across a full-length portion of the gel. Placental membrane calcium uptake The procedure used was as described previously (Tuan, 1985: Tuan et al. 1986). Microsomal membranes were isolated from an isosmotic homogenate of day-14 mouse placenta by differential centrifugation (80000^, 80 min) of the post-mitochondrial fraction. Membrane vesicles were obtained by suspension in 10mM imidazole-HCl, pH7.0, containing 5mM MgCl2, and 0.1M KC1 to a concentration of 3—4 mg of protein ml"1. The calcium uptake assay conditions were: imidazole buffer, 0.1 mM CaCl2 (containing approximately 0.5/iCi of ^CaClzmT 1 ), 5mM ATP, and incubation at 37°C. At the end of incubation, membranes were separated from the solution phase by centrifugation through a layer of silicone oil (specific gravity=1.041). Uptake was measured as the rate of Ca incorporation from 30 s to 120 s. Results Identification and developmental expression of placental Co2*-ATPase The initial enzyme assay was carried out using total extract from day-18 mouse placental tissues. The extract exhibited Ca2+-activated ATP hydrolysis in a timeand substrate-dependent manner, with a mean specific activity of 14-18 nmol phosphate mg" 1 protein min"1. The membranous nature of the enzyme activity was indicated, as the specific activity was increased 5- to 10fold when microsomal membrane preparations were assayed. The total Ca2+-ATPase activity of the mouse placenta was measured as a function of embryonic development. As shown in Fig. 1, the specific activity level of the enzyme increased steadily during gestation, particularly during the late stages, perhaps reflecting the increased calcium need for skeletal mineralization of the rapidly growing rodent embryo (Bruns et al., 1978; Garel, 1983). The developmental profile of the mouse placental Ca2+-ATPase is thus consistent with a possible functional role for the enzyme in placental calcium transport. Characterization of placental Ca2+-ATPase The enzyme activity was next characterized with respect to its ion dependence and specificity, apparent MT, pi, kinetic parameters, substrate specificity, temperature and pH optima, and thermal stability.
Mouse placental Ca2+-ATPase 20 -.
18 h3
ivit :ein
B
o
o
nzyr
c'
16
1 0 0
, Cd
i^ 80
^
/
\ «
on
14
"o
1
/
1
g* 40
hi
| u
v 1
0.06 14 15 16 17 18 Embryonic age (days)
19 2+
Fig. 1. Developmental profile of mouse placental Ca ATPase activity. Placental tissues were dissected from mice at different stages of gestation and were extracted and assayed for Ca -ATPase activity as described in Materials and methods. The activities (closed circles) are the mean±s.D. from 2 experiments and are expressed as specific activities (nmol Pi mg" 1 protein min" 1 ). For comparison, the age profile of embryo wet weight (open circles) is also presented based on previously published values (Brans et al. 1978; Garel, 1983).
Ba
Sr
/
\
10 13
Mg \
60
1° 12 CL"
507
/ minus Ca
1
0.08 0.10 0.12 Ionic radius (nm)
1
0.14
Fig. 3. Ion specificity of mouse placental Ca 2+ -ATPase activity. Mouse placental extract was assayed for ATPase activity as described in Materials and methods in the absence of Ca 2 + (2mM EGTA; minus Ca 2 + ), in the presence of Ca 2 + (100 HM), or equimolar concentrations of Ca 2 + and the indicated divalent cations (100/iM total). All activities are the mean of triplicates, and are expressed as a percentage of the activity in the presence of Ca alone and plotted against the ionic radii of the respective ions.
To examine the ion specificity of the Ca2+ activation of the enzyme, a number of divalent cations were tested for their ability to compete with Ca 2+ . The data in Fig. 3 clearly demonstrated the Ca 2+ specificity of enzyme activation, and furthermore indicated that the ability of other cations to displace Ca2+showed some apparent relationship to the similarity of their ionic radii to that of Ca 2+ , suggesting some degree of steric specificity for Ca 2+ in the binding site of the enzyme.
10 100 Calcium (//M)
1000
Fig. 2. Ca 2 + and Mg2"1" dependence of mouse placental ATPase. Mouse placental extract was assayed in varying [Ca 2+ ] and [Mg2"1"] as indicated. All activities are the mean of triplicates and expressed as a percentage of the highest value.
(1) Ion dependence and specificity Ionic activation of the placental enzyme was examined in varying [Ca2+] and [Mg2"1"], and the results are shown in Fig. 2. The data clearly showed a Ca2+-activated, Mg2+-dependent ATPase activity. Thus, at all Mg2"1" concentrations, Ca 2+ served to activate the enzyme, with the activation being optimal at 1CT5 M [Mg2"1"]. It is noteworthy that although Mg2"1" at 1CT5 M was required for the Ca-activated ATPase activity, higher [Mg2"1"] actually appeared to inhibit the Ca2+-activated enzyme, particularly at low [Ca2+], except at [Ca2+] 3^500/iM, where some Mg2+-activated ATPase activity was observed. This latter finding suggested the presence in the total placental extract of another enzyme that operated at high [Mg24"] and [Ca 2+ ]. However, this activity was not observed when placental microsomal membranes were assayed.
(2) Apparent Mr To determine the apparent native MT of the placental Ca2+-ATPase, we carried out Ferguson analysis of nondenaturing gel electrophoresis (Chrambach, 1980), followed by enzyme histochemistry as described previously (Tuan and Knowles, 1984; Tuan and Kushner, 1987). The histochemical electrophoretogram revealed two dark brown bands of enzyme activity, bands I and II (Fig. 4A). The two activity bands, which were observed in both whole placental extract as well as placental microsomal membranes (total staining shown in Fig. 4B), behaved as distinct moieties with different patterns of electrophoretic mobility in acrylamide gels of varying %T (acrylamide+bis-acrylamide). It should be noted that band II was consistently the more prominent activity species. By plotting the R{ values versus varying %T, the retardation coefficients, KT, of the two activity bands were calculated (Chrambach, 1980) and compared to those of standard proteins of known Mx (Fig. 4C). This comparison yielded apparent Mr values of 118xlO3 and 150X103 for bands I and II, respectively. (3) Isoelectric point (pi) The isoelectric zymogram of the placental Ca2+ATPase revealed only one species of enzyme (Fig. 5), which corresponded to a pi of 6.8-7.0.
508
R. S. Tuan and N. Bigioni
P MM
M
P
A 1
i
B
3.77
5.42
6.78 7.39 7.82 8.20 pH
8.77 9.50 10.96
Fig. 5. Isoelectric focusing profile of mouse Ca2+-ATPase activity. Mouse placental extract was electrofocussed in polyacrylamide gel containing Triton X-100 as described in Materials and methods. Lanes A and B are two separate preparations of placental extract; the enzyme activity (arrow) corresponded to a pi of 6.8-7.0. •§24 S
23
2
21
-
20 00
!
|18o 30
40
50 60 70 - K r (xlOOO)
80
90
50
100
150
200
250
100
Fig. 4. Electrophoretic analysis of mouse Ca2+-ATPase. Total mouse placental extract and solubilized microsomes were fractionated by non-denaturing Triton X-100 polyacrylamide gel electrophoresis, and Ca2+-ATPase activity was visualized by enzyme histochemistry as described in Materials and methods. (A) Enzyme histogram of Ca2+-ATPase showing two activity bands (arrows, Bands I and II). Lane P, total placenta. Lanes M, microsomes. (B) Coomassie blue staining of proteins on the denaturing gel. Lanes P, total placenta; Lanes M, microsomes. (C) Size estimation of Ca2+-ATPase activity by means of Ferguson analysis of the enzyme histogram compared to standard proteins of known molecular weights. Bands I and II correspond to apparent Mv values of 118xlO3 and 150X103, respectively. (4) Kinetic parameters A Lineweaver-Burke reciprocal plot of l / v l/[Ca 2 + ] of the placental Ca 2+ -ATPase activity yielded an apparent Km of 0.97 ^M [Ca 2+ ] and a 1.05 nmol phosphate released m g " 1 placental -i mm
versus (Fig. 6) Vmax of protein
(5) Substrate specificity The Ca 2+ -activated enzyme activity was tested using various substrates. The relative activities (%) based on phosphate released were: ATP, 100; ADP, 52; AMP, 5; /5-glycerophosphate, 3; and GTP, 100. These data
Fig. 6. Lineweaver-Burke kinetics plot of mouse placental Ca^-ATPase activity. The plot yielded a Km of 0.97 UM [Ca2+] and a Vmax of 1.05 nmol Pi mg"1 protein min . indicated that the enzyme activity was highly specific for nucleotide triphosphates, and did not result from nonspecific phosphatases. (6) Thermal properties, pH optimum, and stability The mouse placental Ca 2+ -ATPase was thermolabile; more than 60% of activity was lost by a 15 min incubation at temperatures >65°C, and boiling for 5 min completely destroyed enzyme activity. The temperature optimum of the enzyme activity itself was 45 °C (Fig. 7A). The pH profile of the placental Ca 2+ -ATPase activity was also measured. As shown in Fig. 7B, the enzyme activity was maximal at pH values (8.0-8.5) slightly above physiological, and was thus unlikely to be related to placental alkaline phosphatase (also see data on inhibitors below). With respect to stability, enzyme activity was preserved in whole placental tissues or pelleted microsomal membranes kept at —80°C for at least 6 months; however, detergent-solubilized enzyme lost activity after >24h upon storage at 4°C or -20°C. Inhibitors of mouse placental Ca2+-ATPase Various pharmacochemical reagents that were reported
Mouse placental Ca2+-ATPase 100r (A)
509
Control
•a 80
Phenothiazin
60 40 Erythrosin B
20 40 60 Temperature (°C)
80
Quercetin
Ouabain
100 r (B)
1 mM
0.01 mM
Levamisole
1 mM 0 01 mM 1 mM 20
I"
40
60
80
100
Relativ tnzyme actMty (%)
« 60 40 20 10
Fig. 7. Thermal (A) and pH (B) optima of mouse placental Qr + -ATPase. The enzyme assays were carried out at the indicated temperatures (A), or in Hepes-Pipes buffers at the indicated pH (B). The activities are the mean of triplicates and expressed as a percentage of the highest value.
to affect ATPase or phosphatase activities were tested for their ability to interfere with the mouse placental Ca2+-ATPase activity. These compounds included anticalmodulin compounds (phenothiazin: Gietzen et al. 1980; Vincenzi, 1982), erythrosin B (Morris etal. 1982), ouabain, vanadate (Caroni and Carafoli, 1981), quercetin (Barzilari and Rahamimoff, 1983; Bambauer et al. 1984), and levamisole (van Belle, 1972), varying doses of which were used to treat the placental extract prior to and during the Ca2+-ATPase assay. The results in Fig. 8 showed that whereas ouabain and levamisole did not affect the activity at all doses, the other compounds were effective inhibitors of the placental Ca2+-ATPase activity. The relative inhibitory strength based on effective dose appeared to be: phenothiazin>sodium vanadate, erythrosin B^quercetin (Fig. 8). The high sensitivity to phenothiazin suggested a possible functional association between calmodulin and the enzyme. Cytohistochemical localization of placental Ca2+ATPase The cellular location of the Ca2+-ATPase in the mouse placenta was detected histochemically as described previously for the human term placenta (Tuan and Kushner, 1987) and the chick embryonic chorioallantoic membrane (Tuan and Knowles, 1984). As shown in Fig. 9(A-D), Ca2+-ATPase histochemistry on cryosections of day-16 mouse placenta revealed the presence of enzyme activity in the inner placenta region; further-
Fig. 8. Inhibition of mouse placental Ca2+-ATPase activity. Mouse placental extract was incubated with the respective agents at the indicated concentrations for 15 min and then assayed for enzyme activity under the same conditions. All activities are the mean of 2-3 experiments and are expressed as a percentage of the activity in the absence of any pharmacochemical agents (control).
more, the reaction product was associated with aggregates of polyploid trophoblastic cells. Occasionally, reaction product could also be seen associated with cellular components of the fetal vasculature (not shown), perhaps a result of the Ca2+-ATPase activity of the vascular smooth muscle (Githens, 1983; GodfraindDe Becker and Godfraind, 1980). The specificity of the enzyme histochemistry was demonstrated, as in previous studies (Tuan and Knowles, 1984; Tuan and Kushner, 1987), by the lack of reaction product in the absence of Ca2+ (Fig. 9E and F) or ATP (not shown). Levamisole (1 mM) was routinely included in the incubation mixture to eliminate interference from alkaline phosphatase (van Belle, 1972). The developmental profile of Ca2+-ATPase expression was also examined by cytohistochemistry. Fig. 10 illustrates the results obtained using day-11 and -14 placenta samples. In agreement with the biochemical activity data shown in Fig. 1, day-11 placenta did not exhibit any enzyme staining (Fig. 10A and B), whereas day-14 placenta was clearly reactive (Fig. 10C-F). Interestingly, the first appearance of Ca2+-ATPase was seen in the inner placenta (Fig. 10C-E), also associated with aggregates of polyploid trophoblastic cells (Fig. 10F). The outer placenta remained negative for Ca -ATPase histochemical activity, at least up to day 16. Calcium uptake by cell-free mouse placental membranes and functional involvement of Ca2+ATPase Cell-free membranes isolated by differential centrifugation fractionation of mouse placenta exhibited ATPdependent 45Ca uptake (Fig. 11) in a manner analogous to similar membranes isolated from the human term placenta (Tuan, 1985), although the apparent specific
510
R. S. Tuan and N. Bigioni
Fig. 9. Histochemical localization of Ca2+-ATPase activity in mouse placenta. Cryosections (8/an) of mouse placenta were stained for Ca2+-ATPase as described in Materials and methods. (A and B) Positive reaction for Ca2+-ATPase activity observed using phase contrast (A) and bright-field (B) optics. Note association of reaction product with large polyploid cells (arrows). (C and D) Similar to A and B, respectively. Some reactivity was also observed in the erythrocytes located in a fetal capillary (open arrow). (E and F) Similar to A and B, respectively, except that Ca2+ was omitted from the incubation medium. Bar=10/an.
activity was lower in the mouse placenta. Functional involvement of the Ca2+-ATPase was also suggested since pre-incubation with phenothiazin, a potent inhibitor of its activity (Fig. 8), also significantly inhibited membrane calcium uptake (Fig. 11). Discussion
We have reported here the identification, characterization and cellular localization of a specific, Ca 2+ activated, Mg2+-dependent ATPase in mouse chorioallantoic placenta. Enzyme activity is expressed as a function of fetal gestation, and is localized principally in the large, polyploid fetal trophoblasts of the inner placenta. The enzyme has a pi of 6.8-7.0, and corresponds to two Mr species of 118 and 150X103. The
activity is highly sensitive to phenothiazin, suggesting a calmodulin dependence, and is also inhibited by vanadate, erythrosin B and quercetin. Phenothiazin also strongly inhibits calcium uptake by placental membranes in vitro. The biochemical properties of the mouse placental Ca2+-ATPase suggest that it is a membrane-bound enzyme that functions at physiological pH. Although two electrophoretic forms of the enzyme are observed on non-denaturing gels (Fig. 4A), they are isoelectric (see Fig. 5) and probably differ only in apparent MT. By comparison with the enzyme of the human placenta (Treinin and Kulkarni, 1986; Tuan and Kushner, 1987), and the known properties of other Ca2+-ATPases (see reviews by Sarkadi; 1980; Penniston, 1982; Schatzmann, 1983; Carafoli, 1987), we suggest that the 118X103 form is derived from the vasculature, whereas
Mouse placental Ca2* -ATPase
511
Fig. 10. Developmental profile of Ca2+-ATPase activity in mouse placenta revealed by enzyme histochemistry. Cryosections of mouse placentae (day-11, A and B; day-14, C-F) were stained as described in the legend to Fig. 9. (A,C,F) Phase contrast optics; (B,D,E) bright-field optics; op, outer placenta; ip, inner placenta. (A and B) Day-11 placenta showing no detectable enzyme activity; opposing arrowheads demarcate the apparent border between the inner (ip) and outer (op) placenta. (C and D) Similar regions in a day-14 placenta, showing distinct activity staining in the inner placenta, whereas the outer placenta remained negative for enzyme activity. (E) Another region of the placenta with a staining pattern similar to D. (F) Higher magnification of stained regions, showing association of enzyme activity (arrowheads) with the peripheral area of large, polyploid cells. Bar=40^m in A-E, and 10/an in F.
512
R. S. Tuan and N. Bigioni 3
100
§ • _ 80
1 ^ - 60 S;E
40 Control
-ATP
+PHE
S o Fig. 11. Calcium uptake by cell-free placental membranes in vitro. The membrane preparation and calcium uptake procedures were as described in Materials and methods. Calcium uptake activity was expressed as pmolmg"1 1 protein min" . Samples were treated as follows: Control, standard assay conditions; —ATP, in the absence of ATP in the incubation mixture; and +PHE, membranes pre-treated with 100 HM phenothiazin and assayed in the presence of the inhibitor. Values represent the mean of triplicates, 1and are expressed as percentage of control (10.1pmolmg~ min~1). the more predominant, higher M r , 150X103 form is likely to be a plasma membrane component of the placental trophoblasts, which stain prominently for Ca2+-ATPase activity in the histochemical reaction. Indeed, Borke et al. (1989) have recently detected by immunohistochemistry epitopes of the human erythrocyte plasma membrane 140 xlO3 Ca2+-ATPase in the basal membranes of human and rat placental trophoblasts. It is highly likely that the Ca -ATPase activity described here is similar to that reported by these workers. Calcium transport is a major function of the placental structure, and serves to provide the necessary calcium to the developing embryo. In vivo and in vitro calcium uptake studies (e.g. Fisher et al., 1987; Miller and Berndt, 1975; Shami et al. 1975; Sweiry et al., 1986; Tuan, 1985; Twardock, 1967; van Kreel and van Dijk, 1983; Whitsett and Tsang, 1980), using whole animal or perfused placenta or membrane vesicles derived from human and other animal species, have clearly established a bioenergetic requirement for calcium transport. Since the active nature of placental calcium transport would suggest that it is coupled to ATP hydrolysis, the Ca2+-ATPase identified and characterized in this study may indeed be functionally involved in the transport process, as suggested by the ATP dependence of calcium uptake by mouse placental membranes as shown here (Fig. 11). Similar association of placental membrane calcium uptake transport with ATP hydrolysis has also been reported by other workers (Fisher et al., 1987; Shami et al., 1975; Treinen and Kulkarni, 1987). In support of this hypothesis, the mouse enzyme shares many common properties with the human placental Ca2+-ATPase, which has been clearly implicated in placental membrane calcium uptake in vitro (Tuan and Kushner, 1987). These common properties include Mr, substrate specificity and, most importantly, sensitivity to various pharmacochemical compounds, which also inhibit calcium uptake by placental membrane vesicles (Fig. 11). However, it should be pointed out that since these compounds are not specific inhibitors of
Ca -ATPase activities (as there are no 'ouabainequivalent' specific inhibitors of plasma membrane Ca2+-ATPases), the similarities between the two placental Ca2+-ATPases thus only serve as circumstantial correlations. Additional support for a transport role for the mouse placental Ca2+-ATPase is provided by its cellular distribution within the mouse placenta, i.e. association with the fetal trophoblast cells. It is thus of interest to note that the human (Tuan, 1985) and mouse (Tuan and Cavanaugh, 1986) CaBPs and their mRNA (Tuan et al. 1988), as well as the human Ca2+-ATPase (Tuan and Kushner, 1987), are also specifically localized within the trophoblastic cells, which are generally believed to the transporting cell type of the hemochorial placenta (Dearden and Ockleford, 1983; Sideri et al. 1983; Truman and Ford, 1984). At present, the mechanisms of trans-placental transport processes are poorly understood (Shennan and Boyd, 1987). With respect to calcium transport, the work in our laboratory has established the high Mr CaBPs as functional components, and much progress has been made on their biochemical and molecular characteristics (Tuan, 1982 and 1985). Specific reagents for the CaBPs, such as antibodies (Borke et al. 1989; Tuan and Cavanaugh, 1986) and cDNAs (Tuan and Kirwin, 1988), have also been prepared. Work is currently in progress to extend these studies to the placental Ca -ATPases to gain further understanding of the basic mechanism and regulation of placental calcium transport. This research was supported in part by grants from the National Institutes of Health (HD 15822, HD 21355), March of Dimes Birth Defects Foundation (1-1146), and U. S. Department of Agriculture (88-37200-3746). The authors also acknowledge the assistance of Robert Akins and James Kirwin in the microsomal membrane study. References BAMBAUER, H., UENO, S., UMAR, H. AND UECK, M. (1984).
Ultracytochemical localization of Ca 2+ -ATPase activity in pituicytes of the neurohypophysis of the guinea pig. Cell Tiss. Res. 237, 491-497. BARZILAI, A. AND RAHAMIMOFF, H. (1983). Inhibition of Ca 2 + transport ATPase from synaptosomal vesicles by flavonoids. Biochim. biophys. Acta 730, 245-254. BORKE, J., CARIDE, A., VERMA, A., KELLEY, L., SMITH, C ,
PENNISTON, J. AND KUMAR, R. (1989). Calcium pump epitopes in placental trophoblast basal plasma membranes. Am. J. Physiol. 257, C341-C346. BOYD, R. D. H. (1987). Placental transport: diversity and complexity. Arch. Dis. Child. 62, 1205-1206. BRUNS, M., FAIOTO, A. AND AVIOU, L. (1978). Placental calcium-
binding protein in rats. Apparent identity with vitamin Ddependent calcium-binding protein from rat intestine. /. biol. Chem. 253, 3186-3190. CARAFOLJ, E. (1987). Intracellular calcium homeostasis. A. Rev. Biochem. 56, 395-433. CAJIONI, P. AND CAJIAFOLI, E. (1981). The Ca 2+ -pumping ATPase of heart sarcolemma: characterization, calmodulin dependence and partial purification. J. biol. Chem. 256, 3263-3270. CHRAMBACH, A. (1980). Electrophoresis and electrofocusing on polyacrylamide gel in the study of native macromolecules. Molec. cell. Biochem. 29, 23-46. DEARDEN, L. AND OCKLEFORD, C. (1983). Structure of human
Mouse placental Ca2+-ATPase trophoblast: correlation with function. In Biology of Trophoblast (ed. Loke, Y. and Whyte, A.), pp. 69-110. Elsevier, Amsterdam
513
human erythrocyte Ca2+-transport ATPase by the phenothiazines and butyrophenones. Biochem. biophys. Res. Commun. 94, 674-481. GITHENS, S. (1983). Localization of alkaline phosphatase and adenosine triphosphatase in the mammalian pancreas. J. Histochem. Cytochem. 31, 697-705.
placenta by rapid paired-tracer dilution: Studies with calcium and choline. /. develop. Physiol. 8, 435-445. TREINEN, K. AND KULKARNI, A. (1987). ATP-dependent Ca2+ uptake in brush border membranes of human term placenta. Placenta 8, 477^86. TREINEN, K. AND KULKARNI, A. (1986). High-affinity, calciumstimulated ATPase in brush border membranes of the human term placenta. Placenta 7, 365-373. TRUMAN, P. AND FORD, H. (1984). The brush border of the human term placenta. Biochim. biophys. Acta 779, 139-160. TUAN, R. (1982). Identification and characterization of a calciumbinding protein from human placenta. Placenta 3, 145-158. TUAN, R. (1985). Ca2+-binding protein of the human placenta. Characterization, immunohistochemical localization, and functional involvement in Ca"+ transport. Biochem. J. 227, 317-326.
GODFRAIND-DE BECKER, A. AND GODFRAIND, T. (1980). Calcium
TUAN, R., CARSON, M., JOZEFIAK, J., KNOWLES, K. AND SHOTWELL,
FISHER, G., KELLEY, L. AND SMITH, C. (1987). ATP-dependent
calcium transport across basal plasma membranes of human placental trophoblast. Am. J. Physiol. 252, C38-C46. GAREL, J.-M. (1983). Parathyroid hormone, calcitonin and mineral metabolism in the mammalian fetus and neonate. In Perinatal Calcium and Phosphorus Metabolism (Holick M., Gray, T , Anast, C. eds.). Elsevier, Amsterdam, pp. 71-104. GIETZEN, K., MANSARD, A. AND BADER, H. (1980). Inhibition of
transport system: a comparative study in different cells. Int. Rev. Cytol. 67, 141-170. HOGAN, B., CONSTANTTNI, F. AND LACY, E. (1986). Manipulating the
Mouse Embryo. A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. LOWRY, O . , ROSEBROUGH, N . , FARR, A . AND RANDALL, R . (1951).
Protein measurement with the Folin phenol reagent. J. biol. Chem. 193, 265-275. MILLER, R. AND BERNDT, W. (1975). Mechanisms of transport across the placenta: an in vitro approach. Life Sci. 16, 7-30. MORRIS, S., SILBERGELD, E., BROWN, R. AND HAYNES, D. (1982).
Erythrosin B (USFD&C Red 3) inhibits calcium transport and ATPase of muscle sarcoplasmic reticulum. Biochem. biophys. Res. Commun. 104, 1306-1311. MUNRO, H., PILISTINE, S. AND FANT, M. (1983). The placenta in
nutrition. A. Rev. Nutr. 3, 97-124. ONO, T. AND TUAN, R. (1986). Effect of experimentally induced calcium deficiency on development, metabolism and liver morphogenesis of the chick embryo. /. Embryol. exp. Morph. 92, 207-222. PENNISTON, J. (1982). Plasma membrane Ca2+-pumping ATPases. Ann. N.Y. Acad. Sci. 402, 296-303. RAMSEY, E. (1975). The Placenta of Laboratory Animals and Man, pp. 126-141, Holt, Rinehart and Winston, New York. SARKADI, B. (1980). Active calcium transport in human red cells. Biochim. biophys. Acta 604, 159-190. SCHATZMANN, H. (1983). The red cell calcium pump. A. Rev. Physiol. 45, 303-312. SHAMI, Y., MESSER, H. AND COPP, D. (1975). Calcium uptake by
placental plasma membrane vesicles. Biochim. biophys. Acta 401, 256-264. SHENNAN, D. AND BOYD, C. (1987). Ion transport by the placenta: a review of membrane transport system. Biochim. biophys. Acta 906, 437^57. SIDERI, M., DEVIRGILIIS, G., RAINOLDI, R. AND REMOTTI, G.
(1983). The ultrastructural basis of nutritional transfer: Evidence of different patterns in the plasma membranes of the multilayered placental barrier. Trophoblast Res. 1, 15-26.
B. (1986). Calcium transport function of the chick embryonic chorioallantoic membrane. I. In vivo and in vitro characterization. J. Cell Sci. 82, 73-84. TUAN, R. AND CAVANAUGH, S. (1986). Identification and
characterization of a calcium-binding protein in the mouse chorioallantoic placenta. Biochem. J. 233, 41-49. TUAN, R. AND KTRWIN, J. (1988). Mouse placental 57-kDa calciumbinding protein: I. Cloning of cDNA and characterization of developmental expression. Differentiation 37, 98-103. TUAN, R. AND KNOWLES, K. (1984). Calcium-activated ATPase of the chick embryonic chorioallantoic membrane. Identification, developmental expression, and topographic relationship with calcium-binding protein. J. biol. Chem. 259, 2754-2763. TUAN, R. AND KUSHNER, T. (1987). Calcium-activated ATPase of the human placenta: Identification, characterization, and functional involvement in calcium transport. Placenta 8, 53-64. TUAN, R., LAMB, B. AND JESINKEY, C. (1988). Mouse placental 57-
kDa calcium-binding protein: II. Localization of mRNA in mouse and human placentae by in situ cDNA hybridization. Differentiation 37, 198-204. TWARDOCK, A. (1967). Placental transfer of calcium and strontium in guinea pig. Am. J. Physiol. 213, 837-842. VAN BELLE, H. (1972). Kinetics and inhibition of alkaline phosphatases from canine tissues. Biochim. biophys. Acta 289, 158-168. VAN KREEL, B. AND VAN DUK, J. (1983). Mechanisms involved in the transfer of calcium across the isolated guinea pig placenta. J. Devel. Physiol. 5, 155-165. VINCENZI, F. (1982). Pharmacological modification of the Ca2+pump ATPase activity of human erythrocyte. Ann N.Y. Acad. Sci. 402, 368-380. WHTTSETT, J. AND TSANG, R. (1980). Calcium uptake and binding by membrane fractions of human placenta: ATP-dependent calcium accumulation. Pediat. Res. 14, 769-775. WIDDOWSON, E. (1968). Growth and composition of the fetus and newborn. In The Biology of Gestation (ed. Assali, N.) Vol. 2, New York, Academic Press.
SWEIRY, J., PAGE, K., DACKE, C , ABRAMOVICH, D. AND
KUDILEVICH, D. (1980). Evidence of saturable uptake mechanisms at maternal and fetal sides of the perfused human
(Accepted 6 June 1990)
