JOURNAL OF CELLULAR PHYSIOLOGY 148:17-23 (1991)

Characterization of Calcium Transport by Basal Plasma Membranes From Human Placental Syncytiotrophoblast J. LAFOND, M. LECLERC, AND M.G. BRUNETTE* Maisonneuve-Rosemont Hospital, Montreal, Quebec, H1 f 2 M4 Canada We have studied the mechanisms involved in calcium (Ca2+)transport through the basal plasma membranes (BPM) of the syncytiotrophoblastcells from full-term human placenta. These purified membranes were enriched 25-fold in Na+/ K+-adenosinetriphosphatase (ATPase), 37-fold in [3H]dihydroalprenoloIbinding sites, and fivefold in alkaline phosphatase activity compared with the placenta homogenates. In the absence of ATP and Mg'+, a basal Ca2+ uptake was observed, which followed Michaelis-Menten kinetics, with a K, Ca2+ of 0.18 k 0.05 pM and V ,,, of 0.93 2 0.1 1 nmol/mgimin. The addition of Mg2+ to the incubation medium significantly decreased this uptake in a concentrationdependent manner, with a maximal inhibition at 3 mM Mg2+ and above. The Lineweaver-Burk plots of Ca2+uptake in the absence and in the presence of 1 mM Mg2+ suggest a noncompetitive type of inhibition. Preloading the BPM vesicles with 5 mM Mg2+ had no significant effect on Ca2+ uptake, eliminating the hypothesis of a Ca2+iMg2+exchange mechanism. This ATP-independent Ca2+ uptake was not sensitive to M nitrendipine nor to M verapamil. An ATP-dependent Ca2+ transport was also detected in these BPM, whose K, Ca2+ ,, 3.4 +. 0.2 nmolesimgi3 min. This Ca2+ transport was 0.09 t- 0.02 pM and V requires Mg2+, the optimal concentration of MgL+ being approximately 1 mM. Preincubation of the membrane with lop6 M calmodulin strongly enhanced the initial ATP-dependent Ca2+ uptake. Finally, no Na+/Ca2+ exchange process could be demonstrated.

The continuous layer of the syncytiotrophoblast cell constitutes the main barrier to maternal-fetal transport. Fetal mineralization requires lar e quantities of calcium (Ca2+),particularly during the ast 3 months of gestation, and it is precisely during late part of gestation that Ca2+ transport through placenta occurs against the steepest concentration gradient (Schauberger and Pitkin, 1978; Pitkin, 1985). Because Ca2+is probably at concentrations below the micromolar range in the syncytiotro hoblast cell, it is likely that the basal plasma mem rane (BPM) plays the major role in the process of Ca2+ translocation through placenta. Shami et al. (1975)measured Ca2+transport through guinea pig placental mixed plasma membranes using a flow dialysis technique. They re orted a slow type of adenosine triphosphate (ATP)- ependent transport, not yet saturated after 2 hours of incubation with 5 mM Ca2+ as the substrate. A few years later, Whitsett and Tsang (1980)described ATP-dependent Ca2+uptake by microsomal fractions containing BPM and some intracellular organelles. Ca2+ uptake by this fraction showed a high affinity for Ca2+ (K, 74 nM) and required the presence of Mg2+. According to these authors, a similar ATP-dependent Ca2+ uptake was also present in brush border membrane preparations, though much less active. More recently, Fisher et al. (1987) further purified syncytiotrophoblast BPM and

Y

r&

B

0 1991 WILEY-LISS, INC.

demonstrated an ATP-dependent Ca2+ uptake (K, Ca2+ 119 nM) requiring Mg2+ and stimulated by calmodulin. This high-affinity Ca2+transport system was similar to that described for kidney and intestinal cell basal plasma membranes. The purpose of the present stud was to determine whether the ATP-dependent CazY transport system constitutes the only mechanism utilized by the BPM to assume the Ca2+ flux to the fetus. In a first step, we confirmed the presence of this system in our purified BPM. In a second step, we studied Ca2+transport in the absence of ATP. Finally, we investigated whether lacenta, as with some other Ca2+transporting epithe[a, is the site of a Ca2+/Na+exchanger. MATERIALS AND METHODS Purification of placental membranes Fresh human placentas were obtained from full-term normal vaginal deliveries. After perfusion with saline solution through the umbilical artery ramifications, 120-150 g placental tissue was excised from the central part of the placenta, after the amnion and decidua were removed. The tissue was cut into 2-5 mm3 fragments Received January 10, 1990; accepted March 18, 1991. *To whom reprint requestsicorrespondence should be addressed.

18

LAFOND ET AL

and stirred in 270 mM mannitol and 10 mM TrisHepes, H 7.4, for 30 min at 4°C with a magnetic barrel. Then, t e suspension was filtered through two layers of cotton gauze, and the filtrate was centrifuged at 100,OOOgfor 60 min. The whitish pellet was separated from the red blood cells, suspended in 25 ml Trismannitol medium, supplemented with 10 mM magnesium chloride (final concentration), stirred 20 rnin at 4"C, and finally centrifuged at 3,OOOg for 15 min. The brush border membranes (BBM) were collected from the supernatant. The BPM, although not yet purified, were contained in the pellet, which was resuspended in 10 mM Tris HC1, pH 7.4, and frozen at -70°C. On the day of experiment, the latter suspensions were washed twice in 10 mM Tris HC1, pH 7.4, and loaded on top of a discontinuous gradient of Ficoll (4% and 10%)as proposed by Kelley et al. (1983).Following a 60 min centrifugation (55,000g1, the BPM were collected at the 6 1 0 % interface, washed again, and suspended in 150 mM KC1 (or NaCl), 20 mM TrisHepes, pH 7.0, to a final protein concentration of 15-20 pgiul . The purity of the membranes was monitored by measuring the alkaline phosphatase (BBM marker) activity using the technique of Kelly and Hamilton (19701, the Na+/K+-ATPase (BPM marker enzyme) activity according to the technique of Post and Sen (19671, the succinate dehydrogenase (reticuloendothelium marker) with the technique of Pennington (19611, the glucose-6-phosphatase (mitochondria marker) by the method of Aronson and Toutser (1974), and the [3H]dihydroalprenolo1 binding (BPM marker; Kelley et al., 1983) according to the technique of Williams et al. (1976). Protein concentrations were determined by the method of Lowry et al. (1951). Ca2+transport The transport of Ca2+ through BPM was measured by the standard Millipore filtration assay using Millipore filters (type HAWP, 0.45 pm). Uptake was initiated by mixing 5 pl of the BPM vesicles suspension (10-20 pg) with 255 pl of incubation medium containing 150 mM KC1, 20 mM Tris-Hepes, H 7.0, 0.5 mM EGTA, 1 mM CDTA, 30 pCi 45Ca&2/ml, and the appropriate amount of CaClz to yield the desired concentration of free Ca2+.In the ATP-dependent experiments, 3 mM MgClz and 0 or 3 mM ATP was added to the medium. The total amount of calcium and magnesium to be added t o the media was calculated using a computerized ro ram, with the following association constants: CaB+-EGTA, g 4.37 x M; Mg2+-EGTA, 2.57 x lop5 M; Ca2+-CDTA,1.10 x 10-1 M; Mg2+Mi Ca2+-ATP,5.37 x M; and CDTA, 1.51 X Mg2+-ATP,1.55 x 10- M as described by Ghijsen et al. (1982) and Van Heeswijk et al. (1984). The incubation was allowed to proceed for the indicated times at 37°C. At the end of each incubation period, 40 p1 of the mixture was filtered through soaked filters (250 mM KCl for 30-60 min) and washed with 6.0 ml 250 mM sucrose, 150 mM KC1, 20 mM TrisHepes,fH 7.0, and 4 mM EGTA. The radioactivity retaine on the filter was measured in scintillation liquid Formula P-963 and counted in a beta LKB Mini-1211 scintillation counter (LKB, Rockville, MD).

R

ATP-dependent Ca2+ transport was calculated as the difference between the Ca2+uptake in the presence and in the absence of 3 mM ATP. In calmodulin experiments, membranes were first depleted in calmodulin by preincubation in a medium containing 10 mM 3-[N-morpholinolpropane sulfonic acid (MOPS), 1 mM dithiothreital (DTT),5 mM ethylenediamine-tetraacetic acid (EDTA), 20 mM TrisHepes, pH 7.4, during 20 min at 4°C. Following preincubation, 1 M KC1 (final concentration) was added to the suspension for another 20 min at 4°C. Then, these membranes were washed, resuspended in KC1 TrisHepes, and separated into two groups, the control and M calmodulin was the experimental, to which added. Ca2+ uptake was measured in the usual incubating medium containing (ex erimental group) or not containing (control group) 10- M calmodulin. Finally, the presence of Ca2+ channels in our BPM was investigated b adding M nitrendipine or M verapami or the carrier to the incubation medium.

I?

r

Drugs and materials l-[4,6-Propyl-3Hldihydroalprenolol (30 Cilmmol) was purchased from Amersham Co. (Oakville, Ontario, Canada); scintillation liquid Formula P-963 and 45CaC12(23 mCi/mol) were purchased from NEN (Boston, MA); and calmodulin (from bovine brain 40,000 Uimg), ATP, and all other products were obtained from Sigma (St. Louis, MO). Unless otherwise specified, statistical analysis was performed using two tailed unpaired t tests. Results are expressed as the mean i SEM.

RESULTS Purity of membrane preparations The membrane marker-specific activities in BPM and in whole tissue homogenates are presented in Table 1. For comparison, the corresponding values in BBM are also presented. The enrichments in alkaline phosphatase activity were 24- and fivefold in BBM and BPM, respectively, compared with placental homogenates. The enrichments of Na+/K+-ATPaseand specific binding of [3H]dihydroalprenolo1were five- and 25-fold and two- and 37-fold in BBM and BPM, respectively. Succinate dehydrogenase activity was very low in both BBM and BPM, whereas the two membrane preparations were slightly enriched in glucose-6-phosphatase (2.6- and twofold, respectively). ATP-dependent Ca2+uptake The time course of Ca2+transport by placental BPM vesicles is shown in Figure 1.In the absence of ATP, but in the presence of 3 mM Mg2+,there was a ro ressive but modest accumulation of radioactive Ca!+ ?n these BPM vesicles. When 3 mM ATP was added to the incubation medium, 45Ca2+uptake was significantly increased and reached 4.0 nmoleimg after 5 min. The inset in Figure 1 shows the initial time course of the ATP-dependent Ca2+transport by purified BPM vesicles. This transport was linear for the first 5 min of incubation. Addition of 10 pM A23187 (final concentration)

CALCIUM TRANSPORT BY SYNCYTIOTROPHOBLASTBASAL MEMBRANE

19

TABLE 1. Membrane marker activities of basal plasma membrane (BPM) and brush border membrane (BBM) preparations

Membrane HOMO' BBM BPM

Alkaline phosphatase (pmole Pi/mg/l5 min) (n = 20) 14.1 k 2.2* 332.6 zk 41.4** (24)' 80.3 k 15.6 (5)

[3H]dihydroalprenolol

Na+ K'-ATPase (pmole Pi/mg/l5 min) (n = 20) 0.27 1.9 6.4

+ 0.05* * 0.1** (5) + 1.59 (25)

binding (pmole/mg) (n = 5)

Succinate dehydrogenase (nmole/fig/l5 min) (n = 3)

3.10 k 0.37* 7.05 zk 0.76** (2) 115.2 0.8 (37)

21.8 2.0* 5.45b k 0.5 (0.25) 5.09 zk 0.4 (0.23)

*

+

Glucose-6-phosphatase (pmole/mg/min) (n = 4) 0.55 k 0.001* 0.142 -t 0.003 0.111 ?c 0.001

'HOMO, whole tissue homogenate. "he

magnitude of enrichment is in parenthesis.

*P < 0.01 vs. BPM and BBM. **P< 0.01 vs. BPM.

following a 5 min incubation did not result in a significant release of Ca2+.This release was obtained only with a rapid dilution of the suspension in 20 volumes of an isotonic solution containing both 10 pM A23187 and 4 mM EGTA (Fig. 1).The lack of Ca2+ release with A23187 suggests that the sequestered Ca2+was not free in the intravesicular space. Extravesicular binding was prevented by careful washing of the vesicuIes with 4 mM EGTA, following transport. To confirm the absence of extravesicular binding, a new series of Ca2+ transport experiments were performed with the stepwise addition of sucrose in the incubation medium, in order to produce progressive shrinkage of the vesicular volume. Ca2+ transport decreased as a direct function of osmolarity of the incubation medium, indicating that the amount of retained Ca2+depended on the intravesicular space. In accordance with the data reported by Fisher et al. (1987), the ATP-dependent Ca2+ uptake was strongly stimulated by calmodulin. Figure 2 represents the M calmodulin on the time course of influence of 1 pM Ca2+ uptake. The effect was particularly strong on the very initial uptake of the substrate. As for any ATP-dependent transport, Mg2+ was an absolute requirement. In the absence of Mg2+, 3 mM ATP did not si nificantly enhance Ca2+ uptake. The addition of Mg' at a concentration as low as 0.1 mM induced a significant ATP-dependent transport, which was further enhanced by increasing Mg2 until the maximal effect was reached at Mg2+ 1 mM. At higher Mg2+concentrations, Ca2+uptake showed a slight although not significant tendency to decrease (Fig. 3). Finally, Figure 4 shows the dependence of ATPdependent Ca2+ uptake on free Ca2+concentration, in the presence 3 mM free Mg2+.This transport followed Michaelis-Menten kinetics. The inset in Figure 4 represents the Lineweaver-Burk plot of these data. The com uterized regression line revealed a value for K, Ca2 of 0.09 k 0.02 pM and for V,, of 3.4 2 0.2 nmolimgi3 min. These values are quite close to those reported by Whitsett and Tsang (1980) and Fisher et al. (1987). Not shown in Figure 4, addition of 100 pM vanadate to the membrane vesicles and in the incubation medium completely abolished the ATP-dependent Ca2' uptake, Therefore, according to the characteristics of the ATP-dependent Ca2+ uptake, our membranes were very similar to those reported in the two studies mentioned above. We investigated whether this type of transport was exclusive in our BPM vesicles.

P

ATP-independent Ca2+transport: Effect of Mg2+ In the absence of ATP and Mg2+,a Ca2+u take by BPM vesicles was still observed, which reac ed 3.65 nmolimg after 5 rnin of incubation when 1pM Ca2+was used as the substrate. As was mentioned above, rapid M dilution in an isotonic solution containing A23187, 4 mM EGTA, and no Ca2+ entirely released this transported Ca2+ (Fig. 1). In contrast to the ATP-dependent Ca2+ uptake, the time course of this transport was linear for only 1 min, reached equilibrium after 5 min, and was strongly inhibited by the addition of mM range of Mg2+(Fig. 5).The effect of free Mg2+ concentration on 1 pM Ca2+uytake is shown in Figure 6. Upon addition of Mg2+,Ca2 uptake abruptly decreased to reach a minimal value at 2.5 mM Mg2+. This ATP-independent Ca" uptake was saturable with respect to free Ca2+ concentration. Figure 7 represents Ca2+transport at various substrate concentrations in the absence and in the presence of 1 mM Mg2+. The Lineweaver-Burk plots of these data are resented in the inset of Figure 7. In the absence of hg2+,the apparent K, Ca2+was 0.18 0.05 pM and V,, 0.93 2 0.11 nmolimgimin. In the presence of 1mM Mg2+ kinetic parameters became 0.28 i 0.07 pM and 0.565 2 0.05 nmol/mg/min. Only the change in V,, was significant (P < 0.02, n = 5 ) . Therefore, the influence of Mg2+ does not result from competition with Ca2+. Hypothetically, Mg2+ may decrease Ca2+ uptake according to an exchange mechanism: Once into the vesicles, Ca2+ could be expelled from the vesicle in exchange with Mg2+ ions. To investigate this possible mechanism, we preincubated the vesicles 30 rnin at 25'C in a medium containing 0 or 5 mM free Mg2+. Ca2+uptake was then measured by adding 5 p1 of the vesicle suspension to 255 pl of incubation medium containing no Mg2+.The time course of Ca2+uptake by the two series of vesicle preparations is presented in Figure 8. The resence of Mg2+into the vesicles did not influence Ca" uptake, eliminating the possibility of a Ca2+iMg2+exchange.

R

*

Effect of Ca2+channel-blockers We have previously demonstrated (Lafond et al., 1988) that 62% of placental BPM were oriented inside out. Therefore, it is not known whether the ATPindependent Ca2+uptake measured in our experiments reflects an influx or an efflux process of the syncy-

20

LAFOND ET AL

1. 5 h

F E

L 1.5 0

v

I-'

1.0

Y

s5 0.5

N

z 1

0

1

2

3

4

5

6

0

10

120

60

a

k

+ N

m

-

U

n

I ,

V .

0.2

1

~

1'-

1

V

~~

3

300

TIME (s)

TIME (min)

5

0.3

0.1

0.5

0.8

Mg2+(mM)

h

F4

1 0

E 3 w v

Y

a 2

I-

n

3

1

Ca2+ (pr4)

c5 -

5

I/-

w 2

Mg2+= 1 mM

Y

a ! i 3

+

1

N

m

0

0

1

2

3

4

5 TIME (min)

0

1

2

3

5

6

Mg2+ (mM)

Figures 1-6.

tiotrophoblast cell. If this transport does not require These two Ca2+ blockers were simply added to the any energy, and if it does not result from an exchange incubation medium. Neither l o p 6 M nifedipine nor mechanism either, it may correspond t o a passive entry lop4 M verapamil had any influence on 1 FM Ca2+ of Ca2+ along an electrochemical gradient, through uptake (Fig. 9). some type of Ca'+ channels. Does syncytiotrophoblast BPM contain a We have studied the effect of two Ca2+ channel Ca2+/Na+exchanger? blockers on this tranmort. nitrendbine. a dihvdropyridine, i.e., an L-type, voltage-gated Ca2+ channel Because Na+/Ca2+exchange systems have been deinhibitor, and verapamil, which is a phenylalkylamine. scribed in the membranes of a variety of cell types, in

CALCIUM TRANSPORT BY SYNCYTIOTROPHOBLASTBASAL MEMBRANE

particular in kidney (Gmaj et al., 1979; Khalifa et al., 1983; Van Heeswijk et al., 1984; Van Os, 1987; Ramachandran and Brunette, 1989; Bouhtiauy et al., 1991), we have also investigated whether the syncytiotrophoblast BPM also possesses such a system. The vesicles were preincubated at 25°C in a medium containing 150 mM of either NaCl or KC1 and 20 mM Tris-Hepes, pH 7.0. Then Ca2+uptake was measured as described in Materials and Methods. Figure 10a shows the time course of the 1 pM Ca2+ uptake by the two series of vesicles. No significant increase in Ca2+ uptake could be observed in the Na+ loaded vesicles compared with the control at any of the transport times. Similarly, when the vesicles were preloaded with 1pM free 45Ca2+ and incubated in a medium containing 150 mM of either KC1 or NaC1, the efflux of Ca2+from the vesicles was not significantly influenced by the presence of Na+ in the transposition (Fig. lob).

21

ment of x5 vs. X1.3) and less contaminated with endoplasmic reticulum (glucose-6-phosphatase x 0.2 vs. X 0.7) and mitochondria (succinate dehydrogenase X0.23 vs. ~ 3 )Finally, . enrichment in Na+/K+-ATPase was lower ( x 25) than that in dihydroalprenolol ( x 381, suggesting that in the original homogenates, the Na+ pump was also present in other cell membranes. ATP-dependent Ca2+transport Except for few details, our data concerning the ATP-dependent Ca2+ transport are very close to those reported by Whitsett and Tsang (1980) and Fisher et al. (1987). In particular, the kinetic parameters are similar in the three studies. However, in contrast to what was reported in the Fisher et al. studies, in our experiments intravesicular Ca2+ was not released by the simple addition of 10 pM A23187 into the medium. This release was obtained only in the presence of both A23187 and 4 mM EGTA. It is therefore believed that, DISCUSSION in the presence of ATP, Ca2+ is actively transported Membrane purity into the vesicles, where high concentrations are This study describes two types of Ca2+transport by reached. However, a large proportion of this Ca2+binds isolated placental BPM vesicles: 1) a basal, saturable to the internal surface of the vesicles. Confirming the transport, which is inhibited by Mg2+, and 2) an hypothesis of Ca2+ binding inside the vesicles is the ATP-dependent transport. The validity of these data calculated vesicular volume (1 ml/mg) using the sedepends on the membrane purity. Although at the end questered Ca2+ value at equilibrium, in the absence of of gestation the syncytiotrophoblast cells constitute the ATP. This volume is far above the usual size of vesicles. predominant epithelial cell population se arating ma- The intravesicular binding of Ca2+, however, should ternal and fetal compartments, our initial omogenates not influence the initial uptake and, therefore, the were probably also rich in capillary cells. Therefore, it kinetics parameters. Also slightly different from the was mandatory to assess the purity of these membranes Fisher et al. data, the ATP-dependent transport in our with enzyme markers specific for BPM of the syncy- experimental conditions was extremely sensitive to tiotrophoblast cell and those specific for the undesired Mg2+:0.2 mM free Mg2+ sufficed to permit an almost membranes. maximal ATP-dependent transport. Finally, the strong Until now, the purest BPM reported in the literature enhancement of Ca2+ uptake by calmodulin that we are those from Fisher et al. (1987), with which our also observed was probably due to the complete deplemembranes are comparable. In particular, the mem- tion of our membranes in endogenous calmodulin folbranes from both laboratories have the same enrich- lowing their treatment with 1 M KC1. Therefore, BPM ment in dihydroalprenolol binding sites ( x 38 vs. x 36). from syncytiotrophoblast cells is the site of an ATPOur membranes, however, are more contaminated with dependent Ca2+ transport, which 1) reaches equilibbrush border membranes (alkaline phosphatase enrich- rium relatively slowly compared with the homologous transport through kidney membranes, 21 necessitates the presence of low concentrations of Mg2+,and 3) is Fig. 1. Time course of 1 pM Ca2+uptake by isolated BPM vesicles in very sensitive to calmodulin. the presence (open symbols) or absence (solid symbols) of 3 mM ATP. In a recent study, we failed to detect any ATPAt 5 min of incubation (arrow), the suspension was diluted in 20 dependent Ca2+transport through the other pole memvolumes of incubation medium containing 10 pM A23987 and 4 mM EGTA. Inset: Time course of the ATP-dependent uptake (n = 3). Data brane of the syncytiotrophoblast, i.e., through the BBM of this cell (Brunette et al., 1991). Therefore, this are mean i SEM. Fig. 2. Effect of 10+ M calmodulin on the ATP-dependent Ca2' transport system, which requires energy, has a high transDort through BPM vesicles. The membranes were first deoleted affinity for Ca2+, and is exclusively localized in the in en'dogenous -calmodulin by a n incubation with 10 mM '3-[Nmorpholimolpropane sulfonic acid, 1 mM dithiothreital, 5 mM EDTA, BPM, probably constitutes the main mechanism for 20 mM Tris-Hepes, pH 7.4, over 20 min at 4°C followed by the addition expulsion of Ca2+from the cytosol to the fetal compartof 1 M KCl. Then the membranes were washed and separated in two ment, against a steep potential and concentration M was added to the experimental sample gradient. Since only 62% of our BPM are inside out samples. Calmodulin at of membranes and to the corresponding incubating medium. reported in our experiFig. 3. Dose-response curve of the effect of Mg'+ on the ATP- (Lafond et al., 1988), the V, dependent Ca2+uptake by the BPM vesicles. Ca2+ = 1 pM (n = 3). ments is underestimated. Data are mean ? SEM. As for any ATP-dependent Ca2+ trans Fig. 4. Effect of Ca2' concentration on the ATP-dependent Ca2' cells this activity should be related to a uptake by the BPM vesicles. Inset: Lineweaver-Burk plot of these Ca2'/Mg2+-ATPase. This presumption = 3.43 ? 0.2 nmolimgi3 min data. K, Ca2' = 0.09 i- 0.02 pM, V,, tioned by Whitsett (1980) and Whitsett and Tsang (n = 3). Data are mean i- SEM. Fig. 5 . Time course of 1 pM Ca'+ uptake (in the absence of ATP) by (1980), who failed to demonstrate any topographical BPM vesicles in the presence and in the absence of 1 mM Mg2+ association between the ATP-dependent transport and (n = 3). Data are mean t SEM. Fig. 6. Dose-response curve of the effect of Mg2+ on ATP-indepen- the enzyme activity. We hypothetize that chemical +-ATPase dent Ca2+uptake by BPM vesicles (n varies from 3 to 15 depending on measurement of the high-affinity Ca2+/Mg2 in placenta might be complicated by the presence of Mg2+ concentration).

ph

LAFOND ET AL.

I-'

0

T

2

4

6

8

1

0

0

1

2

3

4

+ 0

60

120

300

0

60

120

5 1IME (rnin)

l/Ca2+ (vM)

300

TIME (s)

0

60

120

180

300

+

0

60

120

180

TIME

300

(s)

Fig. 7. Effect of 1 mM Mg2+ on the kinetic parameters of ATPindependent Ca2+uptake by BPM (n = 5). Data are mean i SEM. Fig. 8. Effect of intravesicular Mg2+ on the ATP-independent Ca2+ uptake by the BPM vesicles. The vesicles were preloaded with either 150 mM KCl (solid symbols) or 143 mM KCl and 5 mM MgC1, (open symbols) n = 3. Data are mean t SEM. Fig, 9. Effect of M verapamil (a) and M nitrendipine (b)on the time course of 1JLMCa2+uptake by BPM vesicles (n = 3).Data are mean 5 SEM.

Fig. 10. Effect of trans-Na+ on Cazi uptake (a) and release (b) by BPM vesicles. In the experiments in a, the vesicles were preloaded with 20 mM Tris-Hepes, pH 7.0, and 150 mM of either KC1 (solid symbols) or NaCl (open symbols) and incubated in a medium containing 1 JLMCa2+ in 20 mM Tris-Hepes and 150 mM KC1 (n = 8). In the experiments in b, the vesicles were reloaded with 1 JLMCaZ+ in 20 mM Tris-Hepes, pH 7, and 150 m b KCl and incubated in a medium containing 20 mM Tris-Hepes, pH 7, 4 mM EGTA, and 150 mM of either KC1 (solid symbols) or NaCl (open symbols) (n = 3). Data are mean ? SEM.

other enzymes able to hydrolyze ATP under the activation by Ca2+. Further investigation using immunoassay techniques or enzyme purification and reconstitution should yield precision about the localization of the enzyme and its implication in the ATP dependent Ca2+transport.

port. Mg2+ inhibited Ca2+ transport in a dose-dependent manner, according to a noncompetitive mechanism. The identity of this new carrier or channel is unknown. Because no energy is likely required by this type of Ca2+ transport, we hypothetize that it could reflect the presence of a channel, permitting the entry of Ca2+into the cell, rather than the outflux as for the ATP-dependent transport. In our experimental conditions, neither nitrendipin nor verapamil had any influence on this type of Ca2+ transport. This lack of effect does not exclude the possibility of a Ca2+ channel. Indeed the action of these two Ca2+ inhibitors are mostly directed toward potential dependent channels, excluding the other types of channels. Moreover, there are examples in the literature of membranes on which Ca2+ inhibitors exhibit their action only followin a pretreatment with a hormone such as PTH (Bacs ai and Friedman, 1990). These Ca2+ channels are probably regulated by phosphorylation rather than voltage. Therefore, it is evident that further investigation is

The ATP-independent Ca2+transport by BPM With the absence of Mg2+ and ATP, we observed a relatively rapid Ca2+ uptake, whereas the presence of 0.25 mM Mg2+ abruptly decreased this uptake. The difference between Ca2+ uptake in the resence and absence of Mg2+ could not be accounte for by Ca2+ binding to the external surface of the membranes, since the vesicles were always rinsed with an EGTA-containing medium following the uptake and since the value of retained Ca2+ was strongly related to the intravesicular volume. Moreover, the influence of Mg2+ on this Ca2+uptake does not reflect a Ca2+iMg2+ exchange mechanism, since preloading the vesicles with Mg2+ did not affect the subsequent Ca2+ trans-

a

a

CALCIUM TRANSPORT BY SYNCYTIOTROPHOBLAST BASAL MEMBRANE

necessary to elucidate the identity of this new type of Ca2' transport in the syncytiotrophoblast BPM and to determine whether or not it corresponds to a Ca2+ channel. Absence of Ca2+/Na+exchange in placental BPM Finally, although in some experiments the presence of Na+ instead of K+ in the transposition resulted in a higher Ca2+ uptake, this effect was not re roducibly observed; therefore, the presence of a Ca8INa' exchanger cannot be confirmed. It is possible, however, that a higher purification of our membranes would permit us to unmask some exchange activity.

ACKNOWLEDGMENTS The authors thank Manon Arsenault for her secretarial help and the Department of Obstetrics at Maisonneuve-Rosemontand Notre-Dame Hospitals for the generous donation of placentas. We thank Dr. Daniel Lajeunesse for fruitful discussion during the preparation of the manuscript and also Mr. Raymond Seguin and Mr. Victor L'Heureux for graphic assistance. This study was supported by grant MA 9565 from the Medical Research Council of Canada. LITERATURE CITED Aronson, N., and Toutser, 0. (1974) Isolation of rat liver plasma membrane fragments in isotonic sucrose. Method Enzymol., 31: 90-102. Bacskai, B.J., and Friedman, P.A. (1990) Activation of latent Ca2 ' channels in renal epithelial cells by parathyroid hormone. Nature, 347:388-391. Bouhtiauy, I., Lajeunesse, D., and Brunette, M.G. (1991) The mechanism of PTH action on calcium reabsorption by the distal tubule. Endocrinology, 128:251-258. Brunette, M.G., Leclerc, M., Lafond, J., and Lajeunesse, D. (1991) Ca2+ transport through the brush border membrane of human placenta syncytiotrophoblast. Placenta, submitted. Fisher, G.J., Kelley, L.K., and Smith, C.H. (1987) ATP-dependent calcium transport across basal plasma membranes of human placental trophoblast. Am. J. Physiol., 252:C38-C46. Ghijsen, W.E.J.M., de Jong, M.D., and Van Os, C.H. (1982) ATP-

23

dependent calcium transport and its correlation with Ca2+-ATPase activity in basolateral plasma membrane of rat duodenum. Biochim. Biophys. Acta, 689:327-336. Gmaj, P., Murer, H., and Kinne, R. (1979) Calcium ion transport across plasma membranes isolated from rat kidney cortex. Biochem. J., 178:549-577. Kelley, L.K., Smith, C.H., and King, B.F. (1983) Isolation and partial characterization of the basal cell membrane of human placental trophoblast. Biochim. Biophys. Acta, 734.9-98. Kelly, M.H., and Hamilton, J.R. (1970)A microtechnique for the assay of intestinal alkaline phosphatase. Results in normal children and in children with celiac disease. Clin. Biochem., 3r33-43. Khalifa, S.. Milis. S.. and Hruska. K.A. (1983) Stimulation of calcium uptake by parathyroid hormone in renal brush border vesicles. J . Biol. Chem., 258:14400-14406. Lafond. J.. Auger. D., Fortier. J., and Brunette, M.G. (1988)Parathvroid hormone receptor in human placental syncytiotrophoblast brush border and basal plasma membrane. Endocrinology, 123: 2834-2840. Lowry, O.H., Rosebrough, N.J., Farr, A.L., and Randall, R.J. (1951) Protein measurement with folin phenol reagent. J . Biol. Chem., -1.9.1.%5-275 --- - . - . Pennington, R.J. 11961) Biochemistry of dystrophic muscle. Biochem. J., 80:649-654. Pitkin, R.M. (1985) Calcium metabolism in pregnancy and the perinatal period: a review. Am. J. Obstet. Gynecol., 151:99-107. Post, R.A., and Sen, A.K. (1967) Sodium and potassium stimulated ATPase. Methods Enzymol., 10:762-768. Ramachandran, C., and Brunette, M.G. (1989) The Na'iCa'' exchange system is localized exclusively in the distal tubule. Biochem. J., 257:259-264. Schauberger, C.W., and Pitkin, R.M. (1978) Maternal perinatal calcium relationships. Obstet. Gynecol., 57:74-76. Shami, Y., Messer, H.H., and Copp, D.H. (1975) Calcium uptake by placental plasma membrane vesicles. Biochim. Biophys. Acta, 401: 256-264. Van Heeswijk, M.P.E., Geertsen, J.A.M., and Van Os, C.H. (1984) Kinetic properties of the ATP-dependent Ca" pump and the Na+/Ca'+ exchange system in basolateral membrane from rat kidney cortex. J. Membrane Biol., 79:19-31. Van Os, C.H. (1987) High affinity Ca2 -ATPase in basolateral and kidney cortex. Biochim. Biophys. Acta, 906:195-222. Whitsett, J.A. (1980) Specialization in plasma membranes of the human placenta. J . Pediatr., 96:600-604. Whitsett, J.A., and Tsang, R.C. (1980)Calcium uptake and binding by membrane fractions of human olacenta: ATP-deoendent calcium concentration. Pediatr. Res., 14369-775. Williams, L.T., Jarett, L., and Lefiowitz, J . (1976) Adipocyte padrenergic receptors. J . Biol. Chem., 251:3016-3023. I