Bicarbonate-induced activation of taurocholate transport across the basal plasma membrane of human term trophoblast MOHAMAD Y. A. EL-MIR, NELIDA ELENO, MARIA A. SERRANO, PILAR BRAVO, AND JOSE J. G. MARIN Departments of Physiology and Pharmacology and of Biochemistry and Molecular Biology, University of Salamanca, 37007-Salamanca, Spain

Y. A., NELIDA ELENO, MARIA A. SERJOSE J. G. MARIN. Bicarbonateinduced activation of taurocholate transport across the basal plasma membrane of human term trophoblast. Am. J. Physiol. 260 (Gastrointest. Liver Physiol. 23): G887-G894, 1991.-The efflux of [“Cl taurocholate from previously loaded vesicles, obtained from basal plasma membrane of human trophoblast, was studied. Apparent & (620 PM) and V,,,;lx (1.79 nmol min-’ . mg protein-‘) values were similar to those found in influx experiments (Marin et al., Gastroenterology 99: 1431-1438, 1990) . Transmembrane gradients of both bicarbonate (100 mM) and unlabeled taurocholate (0.5 mM) accelerated [“Cl taurocholate efflux. The bicarbonate-induced effect was not abolished by carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP) and K’-valinomycin voltage clamp. Neither was it mimicked by 5,5’-dimethyloxazolidine 2,4-dione (DMO) or by other organic (taurine, glycine, lactate, or acetate) or inorganic (Cl-, SCN-, HPO,f-, or SO:-) anions, and it was not sensitive to carbonic anhydrase inhibitors. No effect of bicarbonate was observed either in the absence of gradient or in the presence of a c&directed gradient. Bicarbonate-induced transstimulation was related to an increase in the value for the apparent V (+30%). Study of the stoichiometry suggests that the most probable coupling ratio is one, bicarbonate: taurocholate. In summary, these results provide evidence for the existence of a bicarbonate-driven anion exchange in the basal plasma membrane of the human term placental trophoblast. EL-MIR, MOHAMAD HANO, PILAR BRAVO,

AND

l

1lliX

bile acids; human

placenta

are restricted to the enterohepatic circulation in the adult. However, the concentrations of these compounds in the newborn serum are relatively high (8), decreasing progressively after birth (24). The fetal liver synthesizes and conjugates bile acids, but the enterohepatic circulation, although probably present, is much less important than during the extrauterine life. Evidence supporting the existence of this immaturity includes the absence of efficient fetal intestinal (17) and hepatic (3, 27) bile acid transport. Thus the bile acid concentrations in the fetal bile are extremely low (24). The main fate for fetal bile acid molecules, therefore, is probably to be transferred to the mother via the placenta. As under physiological conditions of plasma pH values, the major portion of these compounds in the fetal bile acid pool is in the ionized state, and the permeability of lipid bilayers for anionic bile acid molecules is very low BILE ACID MOLECULES

0193-l&57/91

$1.50 Copyright

CO 1991

(7). It is reasonable to suppose that more efficient transport systems than simple diffusion must underlie the bile acid transfer across the trophoblast. In addition, in all cell types in which bile acid transfer has been described, i.e., hepatocytes (1, 4, 13, 20), ileal cells (10, 31, 35) and kidney tubular cells (34), secondary active or potential driven carrier-dependent transport systems have been identified. In fact, several pieces of evidence indicate that also in human trophoblast bile acid transfer is mainly via a carrier-mediated process (9, 12, 18). The nature of the driving forces underlying the transport across the human trophoblastic brush-border plasma membrane is still a subject of controversy; both the facilitated diffusion via an electrical potential-sensitive pathway (12) and the hydroxyl-bile acid exchange (9) have been claimed as the responsible mechanism for bile acid transport across this membrane. As far as the other region of the plasma membrane is concerned, i.e., the basal area, we have recently provided evidence for the fact that the gradient of sodium, protons, or the electrical potential are not the driving forces underlying taurocholate transport across this membrane (18). In addition, as observed in the ileal basolateral membrane (3l), an inversely directed bicarbonate gradient stimulates taurocholate uptake by vesicles derived from the trophoblastic basal membrane (18). Thus a persuasive hypothesis is that a transport system similar to that described in the ileal cells, from the energetic standpoint, i.e., a bicarbonate-taurocholate exchanger, might be involved in the transfer of bile acids from the fetus to the trophoblast. This is even more interesting from the physiological point of view if one takes into consideration that both bile acid and bicarbonate gradients are inversely directed in the in vivo situation. Hence, the aim of the present work was to gain insight into the mechanism by which bicarbonate induces activation of taurocholate transport across the basal plasma membrane of human term trophoblast. MATERIALS

AND

METHODS

Materials. 14C-labeled taurocholate and L-alanine were obtained from New England Nuclear (Boston, MA). Unlabeled taurocholate, taurine, glycine, lactate, acetate, 5,5’-dimethyloxazolidine 2,4dione (DMO), bovine serum albumin (fraction V), valinomycin, tris( hydroxymethyl)aminomethane (Tris), acetazolamide [N-(5-sulthe American Physiological Society GM47

Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (129.237.035.237) on January 19, 2019.

G888

TAIJROCHOLATE

TRANSPORT

famoyl-1,3,4-thiadiazol-2-yl)acetamide] and 6-ethoxyzolamide (6-ethoxy-2-benzothiazolesulfonamide) were purchased from Sigma (St. Louis, MO). Carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP) was from Aldrich (Steinheim, FRG). N-2-hydroxyethylpiperazineN ‘-2-ethanesulfonic acid (HEPES) was purchased from Boehringer (Mannheim, FRG). All other reagents were from Merck (Darmstadt, FRG) or were of similar analytical grade. Plasma membrane vesicle orientation. Purification of basal plasma membrane vesicles (BPMV) was carried out from normal term human placentas kindly supplied by the Gynecology and Obstetrics Department of the Virgen de la Vega Hospital, Salamanca, Spain. The BPMV were prepared by the method of Kelley et al. (14) as described elsewhere (18). The purity and contamination of the preparations assayed, as indicated previously (18) by dihydroalprenolol binding (33), alkaline phosphatase activity (EC 3.1.3.1.) (6), L(+)-tartrate-sensitive acid phosphatase (EC 3.1.3.2.) (22), glucose-6-phosphatase (EC 3.1.3.9.) (2), and succinic dehydrogenase (EC 1.3.99.1.) (25) gave similar results to those previously reported (18). Protein was determined by the method of Lowry et al., as modified by Markwell et al. (19), with bovine serum albumin as a standard. The integrity of the vesicles was calculated from the latency of Na’-K’ATPase activity revealed by the addition of the detergent sodium dodecyl sulfate (SDS). The orientation of the BPMV was inferred by the determination of the ouabain binding to BPMV that were incubated in the presence or absence of SDS. To be loaded with taurocholate, frozen BPMV (-70°C) were first thawed and then diluted to -10 mg protein/ ml by using a buffer of similar composition to that in which the BPMV were resuspended before storage, i.e., 450 mM sucrose, 10 mM MgCla, 0.2 mM CaCIZ, 10 mM HEPES-Tris, pH 7.40. They were vesiculated by six passages through a 25gauge needle. The membrane preparation was diluted 1:l with a similar buffer containing 2 x C mM taurocholate plus +OOO dpm/pl [‘“Cltaurocholate (sp act 46.7 mCi/mmol), where C was the desired final concentration of taurocholate (usually 0.1 mM). The BPMV were incubated with the buffer containing labeled and unlabeled bile acid at 25°C for 2 h before using them. Efflux experiments. Taurocholate retention by the BPMV was measured by a rapid filtration technique (11). Experiments were initiated by adding 80 ~1 of incubation buffer to 20 ~1 of BPMV suspension (usually containing 0.1 mM taurocholate and -5 pg vesicles protein/pl). The compositions and conditions of different incubation and loading buffers are indicated in the table and figure legends. The incubation time was terminated by the addition of 4 ml of ice-cold stop solution (250 mM KCl, 25 mM MgS04, 10 mM HEPES-Tris, pH 7.4) and immediate filtration through 0.65-pm Millipore cellulose-nitrate filters (Millipore, Afora, Madrid, Spain). The incubation test tube and subsequently the filters were rinsed once again with the same stop solution and then three additional times with a similar stop solution that contained 0.1 mM unlabeled taurocholate. This procedure, selected on the basis of preliminary studies (18),

ACROSS

THE

PLACENTA

reduces the radioactivity retention by the filters (blank). Typically, 5 x lo4 dpm applied to the filters gave an average blank of ~50 dpm. Total loaded radioactivity or To was measured by rapid filtration after adding both 20 ~1 of the vesicle suspension and 80 ~1 of the incubation medium directly to 4 ml of stop solution. Net efflux was calculated by subtracting the actual value of radioactivity found at the considered incubation time from the To determined for this specific loaded vesicle preparation. Radioactivity on the filters was measured in a liquid scintillation counter (LS1800, Beckman, Madrid, Spain) using the Ready Safe Scintillation Cocktail, also from Beckman, as a scintillant. Statistical analysis. As indicated in the legends, all incubations were performed in duplicate or triplicate and all observations were confirmed on three or more separate BPMV preparations. Values are given as means t SE. The experimental protocol for efflux studies has the advantage of using aliquots from the same vesicle suspension in incubations with different buffers, which allows the use of a paired t test to compare the results obtained in these experiments. When the experimental protocol implied the use of different vesicle suspensions, comparisons were made by Student’s t test. Regression lines were calculated by the least-squares method. For the kinetic analysis, the values for initial efflux rate were fitted to an equation comprising the sum of saturable and diffusional components. The estimations were made by nonlinear regression analysis on a logarithmic scale as described by Van Melle and Robinson (29). Statistical analysis was made using a Macintosh SE computer (Apple, Cupertino, CA). RESULTS

BPMV orientation. Na’-K’-ATPase is known to be specifically located at the basal plasma membrane of the human trophoblast (32). We used the latency of this activity as a test to calculate the membrane vesicle integrity. For this purpose SDS was selected, considering the reported interference of other detergents such as Triton X-100 with Na’-K’-ATPase (26). Figure 1 shows that ouabain-sensitive ATPase activity was markedly increased by incubation of the BPMV with SDS. These results indicate that -56% of the membrane preparations formed sealed vesicles. Ouabain binding revealed that the great majority of these sealed membranes (-100%) were orientated in the inside-out direction. Taurocholate efflux from BPMV. Figure 2 shows the time course of taurocholate efflux from membrane vesicles that were previously loaded with labeled plus unlabeled taurocholate as described in MATERIALS AND METHODS. When the data are expressed as extruded taurocholate as percent of To, the shape of the curve is very similar to that previously reported for taurocholate uptake in a similar experimental system (18). As shown in Fig. 2, the efflux rate was lineal, at least up to 60 s. To confirm the previously suggested existence of a transport system for taurocholate in this membrane, we carried out the experiments whose results are shown in Table 1. Taurocholate-loaded BPMV were incubated in the presence or absence of 0.5 mM unlabeled extravesic-

Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (129.237.035.237) on January 19, 2019.

TAIJROCHOLATE

TRANSPORT

ACROSS

THE

G889

PLACENTA

1. Effect of extravesicular taurocholate (0.5 mM) on taurocholate efflux from basal plasma membrane vesicles obtained from human trophoblast TABLE

Incubation Time

20 s

90 s 40 min 0

0.05

0.10

SDS Concent

0.20 ration

0.30

0.40

(mg SDS/mg

0.50

protein)

FIG. 1. Latency of ouabain-sensitive ATPase activit,y as revealed in the presence of different, concentrations of SDS. Membrane vesicles were incubated with SDS for :30 min at 25°C before carrying out the determination of ATPase activity. Enzymatic measurements were performed in the presence or absence of 4 mM ouabain to calculate ouabain-sensitive fraction (t,ermed as Na’-K’-ATPase) of total ATPase activity. Values are means k SE. They represent relative activit,y compared with value found without SDS (100%) in each series of determinations. Number of series was 6.

Cl

g

I$ 250

;!.,;

V loo0 0

I

I

600

1800

I

-

0

3000 Incubation

Time

(s)

FIG. 2. Time course of’ taurocholate ef’flux from vesicles previously loaded with this bile acid. Data are expressed as net, efflux (= vesicle content at time 0 s - actual content at indicated time) and as percent of’ total loaded taurocholate (vesicle content at time 0 s). Loading media contained (in mM) 250 sucrose, 0.2 CaCl.,, 10 M&l,, 100 KNO:{, 10 HEPES-Tris, pH 7.40, and 0.1 [‘C Jtaurocholate. Aliyuots of 20 ~1 of membrane suspension were incubated at, :37”C with 80 ,ul of a buffer whose composition was similar to that of t,he loading medium except that it did not contain t,aurocholate. Values are means k SE from data ( n = 12) obtained in duplicate or t,riplicate measurements on 4 different placenta preparations.

ular taurocholate. The efflux of labeled taurocholate was accelerated by a transstimulating effect due to the presence of unlabeled bile acid. This was not accounted for by a reduction in taurocholate retention related to vesicle disruption as indicated by similar experiments with loaded L- [‘“Clalanine instead of [“Cl taurocholate. In these experiments, 1 mM extravesicular taurocholate did not modify L-alanine retention (data not shown). Kinetic studies were performed by incubating the vesicles with different concentrations of taurocholate (O800 PM) for 2 h and then measuring the initial rate (60 s) of the bile acid efflux as a function of taurocholate concentrations (Fig. 3). The Eadie-Hofstee plot (initial efflux rate/substrate concentration, as a function of initial efflux rate) of these results is shown in Fig. 3, right. That this plot is linear (r = 0.999; P < 0.001; n = 20) suggests that the taurocholate efflux is saturable and that it obeys Michaelis-

Taurocholate pmol/mg

Efflux, protein

Without t aurocholate

With taurocholate

271t29 t528t:30 806t41

4X3*30* 659t36* 895k46

Values are means k SE from data (n = 1:3) obtained in duplicat,e or triplicate measurements on 4 different placenta preparations. Membrane vesicles were loaded with (in mM) 250 sucrose, 0.2 CaCL, 10 M&l,, 100 KNO:{, 10 HEPES-Tris, pH 7.4, and 0.1 [‘“Cltaurocholate. They were incubated with a medium of similar composition except that it contained 0.5 mM of unlabeled taurocholate. Twenty-microliter aliquots from the same vesicle suspension were incubated with 80 ~1 buffer without or with 0.5 mM extravesicular taurocholate. Results were compared by a paired t test. * P < 0.05.

Menten kinetics for a single transport system. The results of this study indicate that efflux of taurocholate from BPMV follows saturation kinetics. The best fit for the experimental data was a Michaelis-Menten equation plus a very small diffusional term. The values for apparent K,, (affinity constant), Vmax(maximal velocity of transport), and & (diffusional constant) were 620 PM (&lo5 SD), 1.79 nmolmin-‘*mg protein-’ (kO.20 SD), and 1 X lo-" nl.60 s-‘=rng protein-’ (+3 X lo-"' SD), respectively. Bicarbonate-induced transstimulation. As previously described in experiments on taurocholate uptake by BPMV, 100 mM bicarbonate increased the efflux of taurocholate from loaded BPMV (Table 2). This was due to a modification of the velocity of the process because at the equilibrium (40 min of incubation time) no difference was found in the total taurocholate retained by BPMV. This effect was not observed if other inorganic anions such as Cl-, SCN-, HPOi-, or SO,‘- were used instead of bicarbonate (Table 3). Organic molecules such as taurine, glycine, lactate, acetate, or DMO were also observed to have no transstimulating effect on taurocholate efflux from BPMV (Table 4). Figure 4 shows that bicarbonate-induced acceleration of taurocholate efflux was not reduced in the presence of two different inhibitors of the carbonic anhydrase activity, namely, 6-ethoxyzolamide and acetazolamide (Fig. 4, top). The ability of bicarbonate to stimulate taurocholate efflux was not affected either by the incubation of the BPMV in the presence of a protonophore (FCCP) and 100 mM K+ plus an ionophore for K+ (valinomycin) (Fig. 4, bottom). When the existence and position of the bicarbonate gradient were studied (Fig. 5) we observed that when the BPMV were loaded with 100 mM bicarbonate (“cisdirected” taurocholate and bicarbonate gradients) no stimulation of taurocholate efflux was found (Fig. 5, top). Moreover, the presence of 100 mM bicarbonate at both sides of the membrane in vesicles loaded with 100 mM bicarbonate and incubated with 100 mM bicarbonate (absence of gradient) was observed to have no transstimulating effect (Fig. 5, bottom).

Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (129.237.035.237) on January 19, 2019.

G890

TAIJROCHOLATE

TRANSPORT

THE

A CROSS

1200 z 5Q, Q) h z ml 2 2E 600

5 E W

PLACENTA

1

r =l.OO;

E-j 2 3 bk I

0 0

I

300 Taurocholate

600 Concentration

I

0

900 (PM)

0

Pd.05

1

2

V/S (pmollmirdmg

3 ptiein)/@b!

FIG. 3. Kinetics of taurocholate efflux from basal plasma membrane vesicles of human trophoblast. Vesicles were loaded with (in mM) 250 sucrose, 0.2 CaCIZ, 10 M&l,, 100 KNOtr, 10 HEPES-Tris, pH 7.40, and taurocholate at indicated concentrations. They were incubated at 37°C for 60 s with incubation buffers similar to loading medium except that they did not contain taurocholate. Left: initial rates (60 s) of taurocholate efflux (V) as a function of substrate concentrations. Right: Eadie-Hofstee plot of data given on the left. Values are means t SE from data (n = 20) obtained in duplicate or triplicate measurements on 7 different placenta preparations. Apparent K,,,, 620 t 105 1.79 t 0.20 nmol min. mg protein-‘; &, 1 X lo-” t 3 X lo-“’ nlmin-’ mg protein? PM; Kw l

2. Effect of extravesicular bicarbonate (100 mM) on taurocholate efflux from basal plasma membrane vesicles obtained from human trophoblast TABLE

Taurocholate pmol/mg Incubation Time

100 mM extravesicular NO,

Inorganic Anion

Efflux, protein

Control ClSCNHPO:SOi-

100 mM extravesicular HCO,

20s

275t31

406t39”

90 s 40 min

515t28

647&38* 81Ok64

830t44

3. Absence of transstimulating effect in the presence of several inorganic anions TABLE

Values are means k SE from data (n = 15) obtained in duplicate or triplicate measurements on 4 different placenta preparations. Membrane vesicles were loaded with (in mM) 250 sucrose, 0.2 CaCl,, 10 MgCIY, 100 KNO:{, 10 HEPES-Tris, pH 7.4, and 0.1 [14C]taurocholate. They were incubated with a medium of similar composition where, in some experiments, 100 mM KNO:, was replaced by 100 mM KHCO:+ Twenty-microliter aliquots from the same vesicle suspension were incubated with 80 ~1 buffer containing 100 mM NO; or HCO,. Results were compared by a paired t test. * P < 0.05.

Quantitative aspects of bicarbonate-induced stimulation of taurocholate efflux. The results obtained in the study of the initial rate of taurocholate efflux from BPMV as a function of taurocholate concentrations, in the presence of 100 mM bicarbonate, are shown in Fig. 6. The analysis of the best fit for these values indicates that the bicarbonate-modified Michaelis-Menten equation was similar to that obtained in the absence of bicarbonate, as far as the value for the apparent K, was concerned. These values were 704 PM (+81 SD) and 653 PM (k97 SD) in the presence or absence of 100 mM bicarbonate, respectively. However, the value for apparent Vmaxwas markedly affected by the presence of 100 mM extravesicular bicarbonate. This value was 2.27 nmol. min. mg protein-‘, i.e., -30% higher than that obtained in the absence of extravesicular bicarbonate. The stoichiometry of the interaction between bicarbonate and the taurocholate transport system was investigated by the activation method (Fig. 7, bottom). The result was a hyperbolic-shaped curve (not a sigmoidal curve as would be expected for a stoichiometry higher than one). These data may be fitted to the Hill equation by the method of least-squares for n = 1 and n = 2 as shown in Fig. 7, top left and right; the best fit was for

Incubation 20 s 205t30 216k37

221t39 225233 184-+37

Time 40 min

90 s 384t60

597288

385t69 401261 394t61 336t49

632k103

633t92 664k105 628t104

Values are means t SE from data (n = 8) obtained in duplicate measurements on 4 different placenta preparations, and represent amount of taurocholate efflux in picomoles per milligram protein. Membrane vesicles were loaded with (in mM) 250 sucrose, 0.2 CaCIT, 10 MgC12, 100 KNOs, 10 HEPES-Tris, pH 7.4, and 0.1 [ 14C]taurocholate. They were incubated with a medium of similar composition except that 100 mM KNOZj (control) was replaced by 100 mM KCl, KSCN or 50 mM K,HPO, or KgSOA, and the osmolarity was balanced with sucrose. Before being used, vesicles were incubated for 15 min at room temperature with valinomycin (20 pg/mg protein) and FCCP (5 PM). Twenty-microliter aliquots from the same vesicle suspension were incubated with 80 ~1 buffer containing the selected organic anion. None of the results were different from control experiments as compared by a paired t test.

4. Absence of transstimulating effect in the presence of several organic compounds TABLE

Organic Compound

Control Taurine Glycine Lactate Acetate DMO

Incubation 20 s 245k22 224k13

Time 90 s

41Ok24

396t28 389t27

23Ok14 236kl3 221t12

4Olk21

243t16

402t25

384k25

40 min

659k44 649t50 657t47 661t63 637t48 647t52

Values are means t SE from data (n = 9) obtained in duplicate or triplicate measurements on 4 different placenta preparations, and represent amount of taurocholate efflux in picomoles per milligram protein. Membrane vesicles were loaded with (in mM) 250 sucrose, 0.2 CaCl,, 10 MgC12, 100 KNOn, 10 HEPES-Tris, pH 7.4, and 0.1 [‘“Cl t aurocholate. They were incubated with a medium of similar composition except that 10 mM sucrose (control) was replaced by 10 mM taurine, glycine, L-lactate, acetate, or 5,5’-dimethyloxazolidine 2,4dione (DMO). Twenty-microliter aliquots from the same vesicle suspension were incubated with 80 ~1 of the medium containing the selected organic compound. None of the results were different from control experiments as compared by a paired t test.

Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (129.237.035.237) on January 19, 2019.

TAUROCHOLATE Acetazolamide Acetazolamide

w

1000

TRANSPORT

(Nitrate)

G891

PLACENTA

q

(Nitrate) (Bicarbonate)

90 Incubation

THE

0

(Bicarbonate)

6-Ethoxyzolamide 6-Ethoxyzolamide

ACROSS

Bicarbonate

(100 mM

Control

0

“cis”)

q

cl

q

2400 Time (s) Incubation

Control Bicarbonate (100 mM “trans”) + Valinomycin + FCCP

Time

(s)

0

q

Bicarbonate Bicarbonate

(100 mM “trans”) (100 mM in=out)

q q

* * 20

90 Incubation

Time (s)

2400

IJIG. 4. Sensitivity of bicarbonate-induced activation of taurocholate efflux from basal plasma membrane vesicles to the presence of inhibitors (top) of carbonic anhydrase activity or a protonophore (bottom) in voltage clamp conditions. Vesicles were loaded with (in mM) 250 sucrose, 0.2 CaC12, 10 MgC$, 100 KNO:{, 10 HEPES-Tris, pH 7.40, 0.1 [ ‘“C]taurocholate, and 1 PM 6-ethoxyzolamide (n = 7) or acetazolamide (n = 7). Top: vesicles were incubated with a similar taurocholatefree buffer where 100 mM KNO,] was or was not replaced by 100 mM KHCO:,. Bottom: vesicles were treated as described above except that they were preincubated for 15 min with valinomycin (20 g/mg protein) and FCCP (5 FM) before carrying out efflux experiments (n = 6). Values are means & SE from data obtained in duplicate or triplicate measurements on 3 different placenta preparations. Results were compared by a paired t test. * P < 0.05.

n=

1, which suggests that the most probable stoichiometry is l:l, bicarbonate:taurocholate.

20

90 Incubation

Time

2400 (s)

FIG. 5. Importance of a trans-gradient in bicarbonate-induced activation of taurocholate efflux from basal plasma membrane vesicles. Vesicles were loaded with (in mM) 250 sucrose, 0.2 CaCl?, 10 M&l?, 100 KNOZr, 10 HEPES-Tris, pH 7.40, and 0.1 mM [“Cltaurocholate. Top: vesicles were loaded as described above except that an aliquot of them was loaded with a buffer containing 100 mM KHCO:, instead of 100 mM KNO:,. Vesicles were incubated with a similar buffer containing 100 mM KNO,] (c&gradient) (n = 10). Bottom: vesicles were loaded with 100 mM KNO:, or KHCO,,. They were incubated with a similar buffer containing 100 mM KHCO:$ (absence of gradient) (n = 7). Values are means t SE from data obtained in duplicate or triplicate measurement,s on 3 (top) or 4 (bottom) different placenta preparations. Results were compared by a Student t test (top) or a paired t test (bottom). * P < 0.05.

DISCUSSION Bicarbonate

The vectorial translocation of bile acids through the trophoblast requires their transport across distinct membrane areas: the basal membrane during the passagefrom the fetal blood to the trophoblast and the brush-border membrane during the exit from this tissue to the maternal blood, although inversely directed transfer for these compounds probably also occurs (for review, see Ref. 30). The existence of carrier-mediated transport systems in the brush-border region of the plasma membrane has been reported (9, 12). Previous studies in our laboratory have demonstrated that basal plasma membrane vesicles derived from human term trophoblast are able to take up taurocholate via a carrier-mediated transport system that is sensitive to the presence of a transmembrane bicarbonate gradient (18). The present studies were designed to extend these observations. The estimated orientation for the membrane vesicle preparations used in the present study was different from that usually reported for plasma membrane obtained

Control

300 Taurocholatc

600 Concentration

(PM)

FIG. 6. Effect of extravesicular bicarbonate on the initial rates (60 s) of taurocholate efflux as a function of substrate concentrations. Vesicles were loaded with (in mM) 250 sucrose, 0.2 CaCIZ, 10 MgC$, 100 KNO:(, 10 HEPES-Tris, pH 7.40, and taurocholate at indicated concentrations. They were incubated at 37°C for 60 s with similar incubation buffers to that used to load the vesicles except that they were free of taurocholate and that 100 mM KNO:{ was replaced by 100 mM KHCO:{. Values are means t SE from data (n = 8) obtained in duplicate or triplicate measurements on 3 different p1acent.a preparations. Calculat,ed kinetic parameters in the presence of bicarbonate are apparent K,,, 704 * 81 PM; Km,, 2.27 t 0.20 nmol . min-’ . mg protein-‘; and &, 1 x lo-‘” & 2 X lo-“’ nl. min. mg protein?

Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (129.237.035.237) on January 19, 2019.

G892

TAUROCHOLATE (V/(Bicarbonate)

(pmol/20

(V/[Bicarbonate)

) x 100

(pmol/20

s/mg protein)/mM

TRANSPORT

3000

2000

2000

THE

PLACENTA

BA = Bile

* ) x 1000

s/mg protein)/mM 3000-

ACROSS

Acid

*

OH-

BA-

Id &us~-Bor&r

1ooo

1000

n "0

200 V (pmol/20

400

-

0’ -0

600

”

BA 200 V (pmol/20

s/mg protein)

400

600

s/mg protein)

Mcmbranc

FIG. 8. Schematic representation of suggested bicarbonate-bile acid exchanger across basal plasma membrane of human term placental trophoblast. Previously reported carrier-mediated transfer for bile acids across brush-border membrane of this tissue has also been included.

I

1

0

50 Bicarbonate

Concentration

I

1

100

150 (mM)

7. Bottom: relationship between extravesicular bicarbonate concentration and initial rate (20 s) of taurocholate efflux. Vesicles were loaded with (in mM) 250 sucrose, 0.2 CaCl?, 10 MgCl,, 10 HEPESTris, pH 7.40, 0.1 [“‘Cltaurocholate, and 100 KNO:,. They were incubated at 37°C with similar incubation buffers to those used to load the vesicles except that they were free of taurocholate and that extravesicular 100 mM KNO:, was replaced by lo-130 mM KHCO,(. Osmolarity was corrected with sucrose. Values are means k SE from data (n = 8) obtained in duplicate or triplicate measurements on 3 different placenta preparations. Hill plot (V vs. V/A”, where V is initial rate of efflux; A is activator, i.e., bicarbonate, concentrations; and n is number of molecules of activator/molecule of substrate interacting with carrier) and n = 1 (top left) and n = 2 (top right). FIG.

from other cellular types. This a priori surprising data might be due to the particular microanatomy of this tissue. The basal side of the trophoblast lacks microvilli; i.e., it is more or less flat, except that it is curved at the place where it surrounds the chorionic vessels. That means that the orientation that was discovered is just that expected if the membrane forms vesicles keeping its in vivo curvature. In terms of transport studies, these data are of great relevance, because it must be kept in mind that when using these membrane vesicle preparations for uptake studies we are simulating trophoblastto-fetal blood transfer, whereas in efflux experiments we simulate the transfer from the fetal blood into the trophoblast. The similarity in the results obtained in the time course, the kinetic parameters, and the bicarbonate sensitivity between those shown here from efflux studies and those reported elsewhere (18) from uptake studies suggests that the transport system for taurocholate across the basal plasma membrane of the trophoblast has symmetry properties. At this point, it is worth noting that efflux experiments present two advantages. First, efflux represents the physiological direction of net trans-

fer in intact trophoblast, i.e., from the fetal blood into this tissue. Second, in this experimental model the same vesicle suspension can be used with different incubation media. This permits clearer results and makes easier the comparisons between different experimental conditions. This has been useful, for instance, to ascertain the transstimulation of taurocholate efflux by unlabeled extravesicular taurocholate, provided that this was not due to vesicle disruption (Table 1). This phenomenon of countertransport is usually considered a strong evidence for the involvement of a carrier-mediated process. In addition, the efflux experiments using isolated hepatocytes (15) or canalicular plasma membrane vesicles (21) have been proved useful by other groups interested in bile acid transport. Therefore, efflux experiments were performed in the present study. The absence of an inhibitory effect on bicarbonateinduced taurocholate transport by two different inhibitors of the carbonic anhydrase activity, namely 6-ethoxyzolamide and acetazolamide, indicates that the conversion between COZ and HCO:y does not play an important role in the overall mechanism of bicarbonate-induced stimulation. Thus bicarbonate as an anion is directly involved in the process. This is in agreement with the lack of transstimulating effect on taurocholate transport of DMO, an organic compound that can behave as CO2 carrying H+ across the membrane in the protonated form but that poorly permeates the membrane in anionic form. Moreover, its pK, is close to that for the couple COzHCO;. Other evidence supporting the idea that the anion HCO: but not the buffer system C02-HCO: is responsible for the activation of taurocholate transport is the insensitivity of the bicarbonate effect to the presence of a protonophore (FCCP). The bicarbonate-induced activation is not sensitive either to the membrane potential as indicated by the absence of modification in the ability of bicarbonate to stimulate taurocholate efflux in voltage clamp conditions (Fig. 4, bottom).

Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (129.237.035.237) on January 19, 2019.

TAUROCHOLATE

TRANSPORT

To activate taurocholate efflux, a gradient of bicarbonate is required. The presence of bicarbonate at the same concentration at both sides of the membrane is not enough to accelerate taurocholate transport. Moreover, the gradient must be located in a trans-side, i.e., inversely directed. A &-directed gradient has no effect on taurocholate efflux (Fig. 5, top). The studies on the kinetics of the process indicate that the increase in the efflux of taurocholate from the vesicles is due to a velocity effect. Bicarbonate appears to modify the translocation rate of the complex taurocholate carrier rather than the affinity of the carrier to bind taurocholate. Although the actual bicarbonate flux has not been measured in this study, to produce energetic activation it is usually accepted that coupling of activator and substrate fluxes via a transport protein occurs. Taken together, these results strongly suggest, but do not prove, that bicarbonate-induced activation of taurocholate transport is accounted for by the exchange between bicarbonate and bile acid. The involvement of an anion exchanger system is also in agreement with the previously reported sensitivity of taurocholate uptake to the inhibition by 4,4’-diisothiocyanestilbene-2,2’-disulfonic acid (DIDS) and bromosulfophthalein (18). Although these experimental data are usually accepted as indirect evidence for the existence of an exchange, that bicarbonate anions actually cross the membrane in exchange with taurocholate must be demonstrated in further research. However, at present, and bearing that in mind, several approaches can be employed to gain insight into the interaction among bicarbonate-carrier-bile acid. In this sense, the activator-substrate stoichiometry is of particular interest, since it figures heavily in the understanding of the transport event and the concentration capacity of the transporter. Thus we studied the apparent stoichiometry for the hypothesized exchange by the activation method (for review, see Ref. 28). The results indicate that the most probable stoichiometry is l:l, bicarbonate:taurocholate. As far as the binding order of the substrate and the activator to the carrier is concerned, a somewhat straightforward but not definitive approach is the following: the existence of transstimulation by unlabeled taurocholate in the absence of an activator allows models considering mandatory the binding of the activator to the carrier to be previous to that of the substrate (e.g., the mirror model) to be reasonably ruled out. Moreover, the similar values for the apparent K, in uptake and efflux studies suggest that models considering different affinity and hence different binding order for taurocholate and bicarbonate at both sides of the membrane (e.g., the glide model) are not probable. Although this must still remain speculative, at present, the model of random binding order is the one that best fits with our experimental data (28). To delineate the activator specificity of this transporter, several anions other than bicarbonate were assayed. They were selected on the basis of their role in fetal metabolism (lactate, acetate) or bile acid conjugation (glycine, taurine). Inorganic anions were also used because carrier-mediated transport for them has been reported to exist in this tissue (5). However, not one of them was found to have a transstimulating effect on

ACROSS

THE

G893

PLACENTA

taurocholate transport. Thus bicarbonate seems to be the specific activator of this transporter. At least as far as hepatocytes are concerned, bile acid transporters located both at the sinusoidal (36) and the canalicular (16) region of the plasma membrane extend their substrate specificity not only to different bile acids but also to a large variety of amphipathic compounds. Cholic acid is believed to be the major bile acid in the fetal bile acid pool at term, but this is not a constant characteristic throughout the gestational period. Thus concentrations of chenodeoxycholic acid in the amniotic fluid have been reported to exceed those of cholic acid in early gestation (23). This gives rise to two interesting questions. First, what is the specificity of the described transport system for different bile acid species, and second, is this a variable characteristic during intrauterinal life? Nevertheless, whether substrate specificity is restricted to taurocholate or whether this system has also multispecific anion-exchanger properties is a point that deserves further investigation. The fact that both taurocholate and bicarbonate gradients are inversely directed across the human placenta in vivo suggests that the transfer of these compounds between the maternal and fetal blood can occur by anion exchange. However, the actual physiological significance of these findings as well as the sensitivity of the overall process to alterations in the acid-base balance both in the mother and the fetus are interesting questions that remain to be elucidated. In summary, our results suggest that an anion exchanger with symmetry properties and a coupling ratio of bicarbonate to taurocholate of one is involved in the transfer of this bile acid across the basal plasma membrane of human term trophoblast. The bile acid molecules might then leave this tissue to enter the maternal blood via the carrier-mediated pathways reported to be located at the brush-border membrane of the trophoblast (9, 12). A schematic model of this process is proposed in Fig. 8. The authors thank Dr. C. A. R. Boyd for his useful comments and R. Picken for assistance in preparing the manuscript. This work was supported in part by the Fondo de Investigaciones Sanit,arias de la Seguridad Social (grant 90/430). M. Y. A. El-Mir was the recipient of a doctoral fellowship from the Ministerio de Asuntos Exteriores. P. Bravo was the recipient of a doctoral fellowship from the Ministerio de Education y Ciencia. This work was presented in part at the British Physiological Society, January 4-6, 1990, Liverpool, UK, and has appeared in abstract form in J. Physiol. Lond. 424: 17P, 1990. Address for reprint requests: J. J. Garcia Marin, Dept. de Fisiologia y Farmacologia, Facultad de Farmacia, Campo Charro s/n, 37007Salamanca, Spain. Received

30 (July 1990; accepted

in final

form

16 January

1991.

REFERENCES M. S., AND D. HEGNER. Effects of Na’ on bile acid uptake by isolated rat hepatocyte. Hoppe-Seyler’s 2. Ph.ysiol. Chum. 359: 181-192,1978. BAGINSKI, E. S., P. P. FOA, AND B. ZAK. Glucose-6-phosphatase. In: Methods of Enzymatic Analysis, edited by H. U. Bergmeyer. Weinheim, FRG: Verlag Chemie, 1974, vol. 2, p. 876-880. BATH, R., M. S. BERNSTEIN, R. J. ANDERSON, D. VIDYASAGAR, AND M. A. EVANS. Uptake of taurocholate by freshly isolated ANWER,

Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (129.237.035.237) on January 19, 2019.

G894

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

TAUROCHOLATE

TRANSPORT

hepat,ocytes from fetal and newborn rabbits. Riol. Neonate 47: 99106, 1985. BI,ITZER, B. L., C. TERZAKIS, AND K. A. SCOTT. Hydroxyl/bile acid exchange. A new mechanism for the uphill transport of cholate by basolateral liver plasma membrane vesicles. J. Biol. Chem. 261: 12042-12046, 1986. BOYD, C. A. R., AND D. B. SHENNAN. Sulphate transport into vesicles prepared from human placental brush border membranes: inhibition by trace element oxides. J. Physiol. Lond. 379: 367-376, 1986. BRETAUDIERE, J. P., AND T. SPILMANN. Alkaline phosphatases. Ortophosphoric monoester phosphohydrolase (alkaline optimum). Routine method. In: Methods of Enzymatic Analysis, edited by H. U. Bergmeyer. Weinheim, FRG: Verlag Chemie, 1984, vol. 4, p. 7582. CABRAL, D. J., D. M. SMALL, H. S. LILLY, AND J. A. HAMILTON. Transbilayer movement of bile acids in model membranes. Biochemistry 26: 1801-1804, 1987. COLOMBO, C., A. RODA, E. RODA, M. BUSCAGLIA, C. A. DELL’AGNOLA, P. FILIPPETTI, M. RONCHI, AND F. SERENI. Correlation between fetal and maternal serum bile acid concentrations. Pediatr. Res. 19: 227-231, 1985. DUMASWALA, R., M. ANANTHRAYANAN, AND F. J. SUCHY. Characterization of a specific transport mechanism for bile acids on the brush border membrane of human placenta (Abstract). Hepatology 8: 1260, 1988. GALLAGHER, K., J. MAUSKOPF, J. T. WALKER, AND L. LACK. Ionic requirements for the active bile salt transport system. J. Lipid Res. 17: 572-577, 1976. HOPFER, U., K. NEI,SON, J. PERROTO, AND K. J. ISSELBACHER. Glucose transport in isolated brush-border membranes from rat small intestine. J. Biol. Chem. 248: 25-32, 1973. IIOKA, H., I. MORIYAMA, K. HINO, AND M. A. ICHIJO. A study on human placental bile acid transport system of taurocholate across microvillous membrane. Acta Obstet. Gynaecol. Jpn. 38: 837-844, 1986. INOUE, M., R. KINNE, T. TRAN, ANI) I. M. ARIAS. Taurocholate transport by rat liver canalicular membrane vesicles. Evidence for the presence of a Na+/independent transport system. J. Clin. Inuest. 73: 659-66:3, 1984. KELI,EY, L. K., C. H. SMITH, AND B. F. KING. Isolation and characterization of the basal cell membrane of human placental trophoblast. Biochim. Biophys. Acta 734: 91-98, 1983. KUHN, W. F., AND D. A. GEWIRTZ. Stimulation of taurocholate and glycocholat,e efflux from the rat hepatocyte by arginine vasopressin. Am. J. Physiol. 254 (Gastrointest. Liver Physiol. 17): G732G740, 1988. KUIPERS, F., M. ENSERINK, R. HAVINGA, A. B. M. VAN DEERSTEEN, M. J. HARDONK, J. FEVERY, AND R. J. VONK. Separate transport systems for biliary secretion of sulphated and unsulphated bile acids in the rat. In: Trends in Bile Acid Research, edited by G. Paumgartner, A. Stiehl, and W. Gerok. London: Kluwer, 1988, p. 143-152. LITTI~E, J. M., R. A. SMALI,WOOD, R. LESTER, G. J. PIASECKI, AND B. T. JACKSON. Bile salt metabolism in primate fetus. Gastroenterology 69: 1315-1320, 1975. MARIN, J. J. G., M. A. SERRANO, M. Y. EL-MIR, N. ELENO, AND C. A. R. BOYD. Bile acid transport by basal plasma membrane vesicles of human term placental trophoblast. Gastroenterology 99: 1431-1438, 1990. MARKWEI,L. M. A. K.. S. M. HAAS. L. L. BIERER. AND N. E.

ACROSS

20.

21.

22.

23.

24. 25.

27.

28. 29.

30.

31.

32. 33.

34.

35.

36.

THE

PLACENTA

TOLBERT. A modification of the Lowry procedure to simplify protein determination in membrane and lipoprotein samples. Anal. Biochem. 87: 206-210, 1978. MEIER, P. J., A. S. MEIER-ABT, C. BARRETT, AND J. L. BOYER. Mechanisms of taurocholate transport in canalicular and basolatera1 rat liver plasma membrane vesicles. Evidence for an electrogenie canalicular organic anion carrier. J. Biol. Chem. 259: 1061410622, 1984. MEIER, P. J., A. S. MEIER-ABT, AND J. L. BOYER. Properties of the canalicular bile acid transport system in the rat. Biochem. J. 242: 465-469, 1987. MOSS, D. W. Acid phosphatases. Ortophosphoric monoester phosphohydrolase In: Methods of Enzymatic Analysis, edited by H. U. Bergmeyer. Weinheim, FRG: Verlag Chemie, 1984, vol. 4, p. 92101. NAKAGAWA, M., AND K. D. R. SETCHELL. Bile acid metabolism in early life: studies in amniotic fluid. J. Lipid Res. 31: 1089-1098, 1990. NIIJIMA, S. Studies on the conjugating activity of bile acids in children. Pediatr. Res. 19: 302-307, 1985. PENNINGTON, R. J. Biochemistry of dystrophic muscle. Mitochondrial succinate-tetrazolium reductase and adenosine triphosphatase. Biochem. J. 80: 649-654, 1961. SIPS, H. J., D. BROWN, R. OONK, AND L. ORCI. Orientation of ratliver plasma membrane vesicles. A biochemical and structural study. Biochim. Biophys. Acta 692: 447-454, 1982. SUCHY, F. J., J. C. BUCUVALAS, A. L. GOODRICH, M. S. MOYER, AND B. L. BLITZER. Taurocholate transport and Na-K-ATPase activity in fetal and neonatal rat liver plasma membrane vesicles. Am. J. Physiol. 251 (Gastrointest. Liver Physiol. 14): G665-G673, 1986. TURNER, R. J. Quantitative studies of cotransport system: models and vesicles. J. Membr. Biol. 76: l-15, 1983. VAN MELLE, G., AND J. W. L. ROBINSON. A systematic approach to the analysis of intestinal transport kinetics. J. Physiol. Paris 77: 1011-1016, 1981. WATKINS, J. B. Placental transport: bile acid conjugation and sulfation in the fetus. J. Pediatr. Gastroenterol. Nutr. 2: 365-373, 1983. WEINBERG, S. L., G. BURCKHARDT, AND F. A. WII,SON. Taurocholate transport by rat intestinal basolateral membrane vesicles. Evidence for the presence of an anion exchange transport system. J. Clin. Invest. 78: 44-50, 1986. WHITSETT, J. A. Specializations in plasma membranes of the human placenta. J. Pediatrics 96: 600-604, 1980. WILLIAMS, L. T., L. JARETT, AND R. J. LEFKOWITZ. Adipocyte @adrenergic receptors. Identification and subcellular localization by [“Hldihydroalprenolol binding. J. Biol. Chem. 251: 3096-3104, 1976. WILSON, F. A., G. BURCKHARDT, H. MURER, G. RUMRICH, ANI) K. J. ULLRICH. Sodium-coupled taurocholate transport in the proximal convolution of the rat kidney in vivo and in vitro. J. Clin. Invest. 67: 1141-1150, 1981. WILSON, F. A., AND L. L. TREANOR. Glycodeoxycholate transport in brush border membrane vesicles isolated from rat jejunum and ileum. Biochim. Biophys. Acta 554: 430-440, 1979. ZIMMERLI, B., J. VALANTINAS, AND P. J. MEIER. Multispecificity of Na’-dependent taurocholate uptake in basolateral (sinusoidal) rat liver plasma membrane vesicles. J. Pharmacol. Exp. Ther. 250: 301-308. 1989.

Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (129.237.035.237) on January 19, 2019.