Molecular and Cellular Biochemistry 100: 39-50, 1991. © 1991 Kluwer Academic Publishers. Printed in the Netherlands.

Original Article

Altered coupling states between calcium transport and (Ca 2+, Mg2+)-ATPase in the AS-30D Ascites hepatocarcinoma plasma membrane

Jaime Mas-Oliva, 1 Ruy P6rez-Montfort, 2 Maura Cfirdenas-Garcfa 1 and Miguel Rivas-Duro 1 1 Departamento de BioenergOtica y 2 Microbiologfa, Instituto de Fisiologia Celular. Universidad Nacional Aut6noma de MOxico, MOxico

Received 18 December 1989, accepted 10 April 1990

Key words: (Ca 2+, Mg2+)-ATPase, calcium transport, ATP ~ Pi exchange, plasma membrane, AS-30D hepatocarcinoma

Abstract

Plasma membrane fractions from normal, regenerating liver and the AS-30D ascites hepatocarcinoma exhibited a high degree of enrichment when a set of plasma membrane enzyme markers were studied in comparison to the ones associated to the mitochondrial and cytosolic compartments. While the (Ca 2+, MgZ+)-ATPase observed for the plasma membrane fraction isolated from normal liver showed an activity of 1.2/xmoles/mg/min, the regenerating liver and the AS-30D plasma membrane fractions presented a much lower ATPase activity (0.3 and 0.22/~moles/mg/min respectively). Despite the differences in ATPase activity observed between models, the plasma membrane fraction from the AS-30D hepatocarcinoma presented a calcium transport activity similar to the value observed for the normal system (5.9 and 5.5 nmoles Ca2+/ mg/10 min, respectively). Interestingly, the ATP ~ Pi exchange experiments carried out with the different plasma membrane fractions revealed that the (Ca 2+, Mg2+)-ATPase contained in the plasma membrane from the AS-30D cells shows an exchange activity of 26 nmoles ATP ~ Pi/mg/min, similar to the one observed for the enzyme from normal liver (30 nmoles ATP-~ Pi/mg/min). Our results suggest that the plasma membrane from the transformed model presents a more efficient mechanism to regulate the movement of calcium through the calcium pump, with an optimum expenditure of energy.

Introduction

During the last decades, the generated knowledge and better understanding of the effect of calcium ions upon the cell metabolism has stimulated the study of the regulatory function of calcium in mul- tiple systems [1, 2]. Among these systems, the study of the homeostasis of calcium in neoplastic transformation has attracted important attention. Nowadays there are many reports in the literature that suggest abnormal interactions between the

plasma membrane of neoplastic cells and the calci- um ion [3]. These interactions have been mainly studied in relation to cellular proliferation, inva- siveness and cell growth [4]. Interestingly, the ac- celerated growth of neoplastic cells has not only been associated with an elevated calcium concen- tration in the cytoplasm [5-7], but also with elevat- ed concentrations of regulatory proteins closely dependent on calcium, like calmodulin [8, 9] and oncomodulin [10]. Among the better known mech- anisms proposed as a partial explanation for the

Dedicated to the memory of Catalina Mas Oliva and Valent/n Mas Morera.

40

accelerated cell growth in transformation, the pro- posal that alterations at the level of the plasma membrane of these cells cause an increased perme- ability of calcium, has been considered an impor- tant one [6, 9].

In order to maintain the necessary conditions for normal cellular metabolism, most cells keep a low intracellular calcium concentration [11, 12] through the activity of an ATP-dependent calcium pump supported by the (Ca 2+, MgZ+)-ATPase and the Na+/Ca 2+ exchange system, both present in the plasma membrane [13-17]. The plasma membrane (Ca 2+, Mg2+)-ATPase, considered as the catalytic expression of the calcium pump, has been demon- strated to be an important molecule by which non- excitable [18, 19], as well as excitable cells [20-24] maintain a low concentration of calcium in the cytoplasm of normal cells; yet, its role in the plasma membrane of transformed cells is still obscure.

Although the ratio ATPase activity/calcium transport is known for most normal cell plasma membranes where the calcium pump has been studied; the comparison between these ratios with the ones obtained from their transformed counter- part is still lacking. Moreover, the way these sys- tems might balance their energy consumption through the participation of the ATP--~Pi exchange reaction catalyzed by the ATPase [25, 26] in the transformed cells, is still an open matter for dis- cussion. We believe the regulatory sites that con- trol this exchange reaction, considered a partial reaction in the synthesis of ATP carried out by the plasma membrane (Ca 2+, Mg2+)-ATPase, have an important role in the control of the ATPase activ- ity/calcium transport ratios in the transformed cell system.

The present study describes the isolation of an enriched plasma membrane fraction from normal and regenerated liver in parallel with the procedure for the isolation of a plasma membrane fraction from the AS-30 ascites hepatocarcinoma cell line. The characterization of the plasma membrane (Ca 2+, Mg2+)-ATPase and the ATP-dependent cal- cium transport contained in the three plasma mem- brane fractions is also presented. This character- ization indicated that the plasma membrane (Ca 2÷,

Mg2+)-ATPase from the AS-30D hepatoma has modulatory mechanisms that differ from the plas- ma membrane ATPase found in normal liver.

Materials and methods

Materials

All chemicals employed were of the highest quality available. [45Ca]CaC12, and [32p]Pi were purchased from New England Nuclear (Boston Mass, USA). Calmodulin was purified from calf testis [27].

Experimental models

The experimental animals employed consisted of male Wistar rats weighing between 250-300 g, fed on a commercial diet. Partial hepatectomy refers to the surgical removal of the two thirds of the liver, unless otherwise indicated. Regenerating liver samples were surgically removed at the indicated times for each experiment set. The AS-30D rat ascites hepatocarcinoma cell line [28] was a gift from Dr. Antonio Villalobo (Faculty of Pharma- ceutical Sciences, University of British Columbia (Vancouver, Can.). The tumor cell line was main- tained in the peritoneal cavity of young adult male Wistar rats weighing between 250-300 g as previ- ously described [29].

Plasma membrane preparations

The plasma membrane fractions were isolated from the different cell types with a method devel- oped in our laboratory based on the procedures reported by Neville [30] and Brown et al. [31]. The tumor cell suspension was obtained from the peri- toneal cavity of 3 rats inoculated with the tumor 7 days before their sacrifice by decapitation [29]. The ascites fluid was collected by abdominal incision (50 ml aprox), diluted in 150 mM NaC1, 5 mM KCI and 20 mM Tris-HC1 pH 7.4, and washed 4 times at 4 ° C in a total volume of l i t of the medium men-

tioned above. The cells were recovered by centrifu- gation at 90 x g for 4 min in a swinging bucket rotor. The tumor cells were homogenized in ap- proximately 100 ml of ice cold 150 mM KC1 and 5 mM Tris-HC1 pH 8.0 with 4 gentle strokes (15 sec total time), in a polytron homogenizer operating at top speed (setting 10).

Normal and hepatectomized rats were sacrificed by decapitation and their livers excised, cut into small pieces with scissors and placed in approxi- mately 100 ml of ice cold 150 mM KC1 and 5 mM Tris-HC1 at pH 8.0. The tissue was first homoge- nized with 4 gentle strokes (15 sec) in a polytron homogenizer operating as described above, and further homogenized at 4 ° C by means of 6 strokes in a glass-teflon homogenizer.

After homogenization, all subsequent steps were performed at 4 ° C and were the same for the tumor, the regenerated and the normal tissues. The crude homogenates were centrifuged for 15min at 2 000 x g using a SS-34 Sorvall rotor and the super- natants discarded. The pellets were resuspended in 5 mM Tris-HC1 pH 8.0 and solid sucrose added to a final concentration of 47% (w/w). These solutions were placed in the bottom of four to six centrifuge tubes, and covered by two successives layers of decreasing concentrations of sucrose (44% and 42.3% (w/w)) in 5mM Tris-HC1, pH8.0. These tubes were centrifuged in a Beckman SW-27 swing- ing bucket rotor for l h at 150 000 x g. The crude plasma membrane fractions migrated to the top of the 42.3% (w/w) sucrose layer and were collected with a Pasteur pipette. These fractions were diluted with 1 volume of 5 mM Tris-HC1 pH 8.0 and cen- trifuged at 40 000 x g for 15 min in a Beckman 60 Ti fixed angle rotor. The resulting pellets were re- suspended and further homogenized for 7 sec with a polytron homogenizer in 10-20 ml of 5 mM Tris- HC1 pH 8.0. The homogenates were then carefully layered on the top of a discontinuous three layer gradient made up of 32%, 34% and 40% (w/w) sucrose in 5 mM Tris-HC1 pH 8.0, and centrifuged in a Beckman SW 40 Ti rotor at 300 000 x g for 2 h. The so called light plasma membrane fractions, remaining on top of the 32% (w/w) sucrose layer, were collected with a Pasteur pipette and diluted

41

with 1 volume of 5mM Tris-HC1 pH8.0. These fractions were centrifuged in a Beckman 60 Ti fixed angle rotor at 100000 x g for 30min, and unless otherwise indicated, the resulting pellets resus- pended in 25 mM Na-Hepes, 10 mM MgC12 pH 7.4 at a protein concentration of approximately 1 rag/ ml. The purified plasma membrane fractions were aliquoted, quickly frozen in liquid nitrogen ( - 190 ° C), and stored until needed.

Enzyme assays

The purity of the plasma membrane fractions was studied measuring the following enzyme markers: (Na +, K+)-ATPase [32], 5' nucleotidase [33], succi- nate dehydrogenase [34], lactate dehydrogenase [35] cytochrome oxidase [36], and glucose-6-phos- phatase [37]. Protein determinations were carried out following the method of Lowry et al. [38] em- ploying bovine serum albumin as a standard.

The (Ca 2+, Mg2+)-ATPase activity was mea- sured spectrophotometrically following the release of inorganic phosphate to the medium as previous- ly described [39]. The activity was assayed incubat- ing 50/xg of plasma membrane protein for 10 min- utes at 37°C in a final volume of 400/xl of assay medium containing 50mM Tris-Malate pH7.4, 10 mM MgC12, 100 ~M CaC12 and 2 mM ATP (final concentrations). The stimulation of the (Ca2+,Mg2+)-ATPase activity by calmodulin was determined using the same assay conditions as de- scribed above, except that the experiments were carried out in the presence of different concentra- tions of calmodulin.

The calcium stimulated ATPase activity was as- sayed in the absence of MgC12 using the same assay conditions previously described, except that differ- ent calcium concentrations were added to the reac- tion medium using a Ca2+/EGTA buffer system. The free concentrations were calculated using stan- dard stability complexes [40] and a computer pro- gram. The Mg2+-ATPase activity was assayed as previously described, except that the incubation medium contained 50mM Tris-Malate pH7.4, 2 mM ATP and different concentrations of MgC12

42

using a Mg2+/EGTA buffer system. The free mag- nesium concentrations were calculated as de- scribed above for calcium.

Calcium uptake experiments

pSCa]CaC12 uptake was assayed using the Millipore filtration technique (0.45/xm pore diameter fil- ters). 100/xg of plasma membrane vesicles resus- pended in 0.25 M sucrose and 50 mM Hepes-HC1 pH 6.6, were added to the assay medium contain- ing 20 mM Hepes-HC1 pH 6.6, 80 mM KC1, 5 mM M g C l 2 , 1 mM EGTA, 0.82 mM total CaC12 (10/zM free C a 2+) and 100/zl [45Ca]CaC12 (500/xCi) in a final volume of 500/xl. The mixture was pre-in- cubated for 5 rain at 37 ° C and the reaction started with the addition of ATP to give a final concentra- tion of 3 mM, and incubated at 37 ° C in a shaking waterbath. At given intervals, 450/zl samples were rapidly vacuum filtered, and the filters washed with 16ml of 50mM Hepes-HC1 pH6.6 containing 0.25 mM sucrose. The filters were dried and placed in scintillation vials with 5 ml of tritosol, and the radioactivity counted using a scintillation counter.

ATP ~ Pi exchange reaction

The ATP --~ Pi exchange reaction as a partial reac- tion in the synthesis of ATP, was assayed mea- suring the formation of [y-32p]ATP from [32p]Pi [41, 26]. The excess [32p]Pi not utilized in the reaction was extracted as a phosphomolybdate complex pri- or to precipitation and centrifugation of the protein at 4000 rpm for 5 rain. For the measurement of the [y -32p]ATP formed, the organic or butylacetate phase was discarded and the aqueous samples reex- tracted after the addition of 20/xl of 20 mM Pi as a carrier and 20/zl acetone for colour enhancement. After this procedure had been repeated six times, aliquots from the water phase of the different sam- pies, were placed in scintillation vials and the radio- activity measured.

Results

Purity of plasma membrane fractions

Table 1 shows the specific activities of the different enzyme markers tested, both in the crude homog-

Table 1. Enzyme activities contained in homogenates and plasma membrane fractions from normal liver, regenerating liver, and AS-30D Hepatoma. The activity of the different enzymes was assayed as described in Materials and Methods. (n) Number of preparations, mean values (+ S.E.)

Tissue Specific activities: (nmol/mg Protein/min)

Ouabain-sensitive 5'Nucleotidase Succinate Lactate Cytochrome Glucose-6- (Na ÷, K+)-ATPase dehydrogenase dehydrogenase Oxidase phosphatase

Normal liver Homogenate (3) 41.2+ 0.86 (3) 221.9+ 1.2 (3) 12.05+ 0.3 Purified (3) 181.6 + 0.1 (3) 3078 + 69.7 (3) 1.09 + 0.005 membrane fraction Regenerating liver Homogenate (3) 46.7+ 0.3 (3) 101.2+ 4.5 (3) 7.6+ 1.04 Purified (3) 153.3_+ 0.6 (3) 1216.4_+ 10.1 (3) 6.65+ 0.3 membrane fraction AS-30D hepatoma Homogenate (3) 11.3+0.06 (3) 10.1+ 0.1 (3) 5.05+0.34 Purified (3) 144.7+ 0.5 (3) 271.3+ 8.5 (3) 3.42+ 0.14 membrane fraction

(3) 4777.5 + 637.0 (3) 3.72 + 0.01 (2) 6.31 + 0.4 (3) 767.2+ 25.0 (3) 1.32_+ 0.02 (2) 0.93+ 0.02

(3) 9296.7± 4100.6 (3) 0.84± 0.005 - - (3) 123.06± 9.3 (3) 0.60± 0.005 - -

(3) 276.0+ 121.5 (3) 4.2+ 0.02 (3) 231.2+ 29.0 (3) 0.48+ 0.005

(2) 1.14 + 0.02 (2) o.oo

enates and in the purified plasma membrane frac- tions. The ouabain-sensitive (Na +, K+)-ATPase was found substantially enriched in the plasma membrane fractions of the three different sources, whereas the 5' nucleotidase was observed mainly enriched in the purified plasma membrane frac- tions of normal and regenerating liver cells. In- terestingly, in accordance with the activities previ- ously reported for the AH-130 ascites hepatoma [42], the AS-30D hepatocarcinoma cells showed a low 5'nucleotidase activity, both in the crude ho- mogenate and in the purified plasma membrane fraction. Succinate dehydrogenase, lactate dehy- drogenase, cytochrome oxidase and glucose-6- phosphatase were assayed in order to find out the degree of contamination with mitochondrial mem- branes cytosol and endoplasmic reticulum. From these results, that are in good agreement with the results published by Church et al. [29], we can conclude that the plasma membrane preparations obtained from the different sources can be consid- ered as enriched plasma membrane fractions.

(Ca 2+, Mg>)-ATPase activity of plasma membrane vesicles

In order to characterize the (Ca 2+, Mg2+)-ATPase activity present in the enriched plasma membrane fractions of normal, regenerated and transformed hepatocytes; optimal assay conditions for the mea- surement of the (Ca 2+, Mg2+)-ATPase activity at different pH values were employed as described in Materials and Methods. Figure 1 shows that inde- pendently of the pH employed in the reaction mix- ture, the ATPase, contained in the plasma mem- brane fraction isolated from control liver tissue, had a higher activity than plasma membrane from regenerating liver and AS-30D hepatoma. Never- theless, the optimal pH for the expression of the ATPase corresponded to 7.4. With respect to the calmodulin sensivity of the ATPase contained in the different plasma membrane fractions, it was observed that only the (Ca 2+, MgZ*)-ATPase activ- ity contained in the plasma membranes from nor- mal liver tissue was slightly stimulated by calmodu-

43

-g t,0

- , I . . - , - ,

O

13_

E . w

o_ 0 ,5 09

0 E

I I I 6 7 8

pH

Fig. 1. pH dependence of (Ca 2+, MgZ+)-ATPase activity (Ca 2+, Mg2+)-ATPase of normal liver (O), 48 hr regenerating liver (O) and AS-30D hepatocarcinoma (•) plasma membrane was mea- sured using optimal assay conditions as described in Materials and Methods. Each point represents the mean + S.E. of four experiments.

lin at concentrations ranging from 5-50/zg/mg pro- tein (Fig. 2).

In the absence of magnesium ions, normal liver (Ca 2+, Mg>)-ATPase was found to be stimulated by calcium, while the (Ca > , Mg2+)-ATPase of re- generating and transformed liver plasma mem- branes were shown to be less responsive to this cation (Fig. 3).

The magnesium sensitivity of the (Ca 2+, Mg2+) - ATPase was tested using assay conditions for opti- mal (Ca 2+, Mg2+)-ATPase activity in the absence of CaCI2 and in the presence of different free MgCI2 concentrations adjusted with EGTA as described in the Materials and Methods section. As shown in Fig. 4, the ATPase of the three different sources showed an important magnesium stimulation that followed a sigmoid-like response. However, the absolute activity values were found to be much

1,5

¢..-

bS_ 0.75

0 5 15 25 50 calmodulin

( j i g / rag p ro t )

44

Fig. 2. Dependence of the (Ca 2+, Mg2+)-ATPase activity on calmodulin. (Ca 2÷, Mg2+)-ATPase activity of normal liver (O), 48 hr regenerating liver (O) and AS-30D hepatocarcinoma (A) plasma membrane vesicles were assayed as described in Materi- als and Methods, with different calmodulin concentrations. Each point represents the mean + S.E. of two experiments.

lower in the regenerated and the hepatocarcinoma cells plasma membranes than the one observed in the plasma membrane from normal cells.

Since the diminished hydrolytic activities observ- ed in the regenerating liver could be associated with the degree of proliferation of this cell system, we compared the ATPase hydrolytic activity of the plasma membrane vesicles obtained from regener- ating liver at 2, 3, 4, 8, 18 and 35 days, using optimal assay conditions for the expression of the (Ca 2+, Mg2+)-ATPase activity. As shown in Fig. 5, the hydrolytic activity of 48hr regenerating liver showed a marked decrease in the ATPase activity, that was recovered progressively, reaching the con- trol value on the 20 th day after surgical removal of the left lobule of the liver.

1,5

c-

LT_ 0,75

:1,

, , , ,

9 7 5 .5

pCa

Fig. 3. Dependence of (Ca 2+, Mg2+)-ATPase activity on free calcium concentration in the absence of magnesium. (Ca 2+, Mg2+)-ATPase activity of normal liver (O), 48 hr regenerating liver ((3) and AS-30D hepatocarcinoma (A) plasma membrane vesicles were assayed as described in Materials and Methods, in the absence of magnesium ions. Free calcium concentrations were adjusted using CaZ+/EGTA buffers. Each point represents the mean + S.E. of four experiments.

Effect of Triton X-IO0 on (Ca 2+, Mg2+)-ATPase activity

In order to exclude the possibility that the plasma membrane vesicles from the three different sources had different vesicle orientations; optimal assay conditions for the measurement of the (Ca 2+, MgZ+)-ATPase activity at two different Triton X-100 concentrations, 0.1 and 0.5%, were employ- ed. Table 2 shows that the plasma membrane (Ca 2÷, Mg2+)-ATPase activity from the three dif- ferent sources, was diminished in the presence of Triton X-100. However, the degree of inhibition observed is the same for the three different vesicle types used here. Based on these data, we conclude that there are no important differences in orien- tation between the vesicles originated from the

45

r-

E

E

a_

o E

1,0

0,5

I

6 5 4 .5

pMg

Fig. 4. Dependence of (Ca 2+, Mg2+)-ATPase activity on free magnesium concentrations in the absence of calcium ions. (Ca 2÷, Mg2+)-ATPase activity of normal liver (O), 48 hr regen- erating liver (O) and AS-30D hepatocarcinoma (A) plasma membrane vesicles were assayed as described in Materials and Methods, in the absence of calcium ions. Magnesium free con- centrations were adjusted using Mg2+/EGTA buffers. Each point represent the mean + S.E. of two experiments.

r -

E

E

8..

03

0 E

1,0

0,5

i l • i I I I

0 4 12 20 28 36

Days Posthepatectomy

Fig. 5. (Ca 2+, Mg2+)-ATPase activity of plasma membrane ves- icles isolated from regenerating liver after different time peri- ods. (Ca 2+, Mg2+)-ATPase activity was assayed as described in Materials and Methods. Each point represents the mean + S.E. of two experiments.

different cell types. The consistent ATPase inhib- ition observed in this set of experiments, could be due to a direct effect of the detergent upon the enzyme, as has been suggested for other plasma membrane ATPase [43].

ATP-Dependent calcium uptake by plasma membrane vesicles

Since it has been reported that the (Ca 2+, Mg2+) - ATPase activity of plasma membrane from normal liver might not necessarily be the biochemical ex- pression of the calcium pump [44, 45], we studied the uptake of labelled calcium carried out by the plasma membrane vesicles of normal, regenerating liver and the hepatocarcinoma AS-30D cells. Fig- ure 6 shows a time-course for the ATP-dependent calcium transport. The plasma membrane vesicles of normal liver showed a greater pumping activity

compared with the low transport activity found for the regenerating liver. Surprisingly, the ATPase from transformed cell plasma membrane showed a similar pumping activity when compared the AT- Pase of normal liver plasma membranes. As shown in Fig. 6, the accumulated calcium was released with the addition of the calcium ionophore A23187.

These results show that despite the important differences observed with the ATPase hydrolytic activities between transformed and normal mem- branes, calcium uptake is maintained at similar levels in both vesicle types. It is important to men- tion that despite the low ratios observed between Ca 2+ transport and ATP hydrolysis, that could be due to the different sensitivities of the methods employed for their measurements, the 'coupling efficiency' is higher in the tumor plasma mem- brane.

46

Effect of vanadate, NaCl and nucleotides on (Ca 2+, Mg2+)-ATPase and ATP-dependent calcium transport

Three criteria were used in order to further charac- terize the (Ca 2+, Mg2+)-ATPase and its associated calcium pump: the effect of sodium, the effect of vanadate and the capability of the ATPase to hy- drolyze CTP, GTP and ITP. Table 3 shows the effect of NaC1 (80 mM) and vanadate (100/xM) on both (Ca 2+, Mg2+)-ATPase and ATP-dependent calcium uptake. The ATP-dependent calcium up- take, present in the three different vesicles types was inhibited by vanadate; whereas the (Ca 2+,

MgZ+)-ATPase activity remained unaltered. On the other hand, sodium ions slightly enhanced the (Ca 2+, MgZ+)-ATPase activity of the normal liver and AS-30D hepatocarcinoma plasma membrane vesicles, and diminished the ATP-dependent calci- um accumulative capacity of these vesicles. These results are in accordance with the existence of the Na+/Ca 2+ antiporter as suggested for other liver plasma membrane fraction [46]. Interestingly, if ATP is changed for CTP, GTP or ITP in the reac- tion mixture for the hydrolytic reaction, these nu- cleotides are capable to support the ATPase reac- tion in the three plasma membrane models, rela- tive to their control value observed when ATP is employed (Table 4). Based on these results, we conclude that the (Ca 2+, Mg2+)-ATPase and the ATP-dependent calcium pump, contained in the three plasma membrane models, seem to represent the expression of the same molecule.

Table 2. Effect of Triton X-100 on plasma membrane (Ca 2+, MgZ+)-ATPase activity

Plasma membrane Triton X-100 (Ca 2+, Mg2+)-ATPase isolated from % % remnant activity

Normal liver 0.05 85 Regenerating liver 0.05 80 AS-30D hepatoma 0.05 80

Normal liver 0.1 60 Regenerating liver 0.1 50 AS-30D hepatoma 0. i 76

6

% (D

3 i I 5 I0 15

t (min)

Fig. 6. Time course of ATP-dependent calcium uptake into plasma membrane vesicles. ATP-dependent calcium uptake of normal liver (O), 48hr regenerating liver (O) and AS-30D hepatocarcinoma (A) was assayed as described in Materials and Methods at a free calcium concentrations of 2 x 10 -s M. The arrow indicates the addition of 0.2/xg A-23187. The results represent the mean _+ S.E. of four experiments.

Discussion

In agreement with the results shown by Church et al. [29], the adaptation of the normal liver plasma membrane purification procedure [30, 31], to the tumor cell line, proved to be successful when the plasma membrane markers were compared with those for mitochondria and cytoplasm. From these results, it is concluded that the plasma membrane fractions used in this study are highly enriched with the important characteristic that they have a sever- al-fold increase in (Ca 2+, MgZ+)-ATPase activity and ATP-dependent calcium transport, when com- pared with the preparations obtained by Neville [30]. Moreover, the plasma membrane prepara- tions employed in our experiments proved useful for both, ATP-hydrolysis as well as for calcium uptake experiments.

Based on the following characteristics, it is pro-

47

Table 3. Effect of vanadate and NaC1 on (Ca 2+, Mf+)-ATPase and Ca 2+ uptake. ATPase activity and calcium uptake were assayed as described in Materials and Methods. Vanadate and NaC1 were added to the reaction medium at the indicated concentrations. NaCI was used in the reaction medium instead of KC1. The results represent the mean of triplicate determinations

Plasma membrane isolated from control vanadate NaC1

(100/.tM) (80mM)

nmoles Pi/mg/min

Normal liver 1041.0 + 182.0 979.0 + 49.0 1158.0 + 33.0 Regeneratingliver 177.0+ 9.2 140.0+_ 3.0 160.0+ 6.0 AS-30D hepatoma 200.0_+ 4.4 151.0 + 7.5 263.0 _+ 21.0

nmoles Ca2+/mg/min

Normalliver 1.69_+ 0.77 0.17 + 0.09 0.39_+ 0.09 Regeneratingliver 0.33+ 0.09 0.08+ 0.01 0.69_+ 0.11 AS-30Dhepatoma 1.12+ 0.62 0.22+ 0.06 0.09+ 0.01

posed that the ( C a 2+, Mg2+)-ATPase and ATP- dependent calcium transport present in the three different plasma membrane preparations, are in- deed an expression of the same molecule: 1) In- sensitiveness to calmodulin. Although there seems to be a small response to calmodulin when the plasma membrane from normal liver cells were assayed, we believe this could be due to a small contamination by a non-parenchymal (Ca 2+, MgZ+)-ATPase that has been reported to be sensi- tive to calmodulin [47]. 2) Insensitiveness to 100/xM vanadate. When compared to plasma membrane ( C a 2+, MgZ+)-ATPase from other sources, the liver ATPase reaction is not inhibited by vanadate [48] in contrast to the inhibition ob- served for the ATP dependent Ca 2+ uptake [48]. 3) Lack of specificity in the reaction with nucleotides.

The ATPase hydrolysis values supported by the different nucleotides employed are in accordance with previous reports by other investigators [50]. 4) Diminished calcium accumulative capacity in the presence of NaC1. This result suggests the presence of the plasma membrane Na+/Ca 2+ exchanger in the different preparations.

Analysing the data shown in Figs 3, 6 and 7, it can be suggested that the vesicle systems used in this study are not leaky. This is an important issue, since we did not measure initial rates of calcium uptake. In the presence of an important back flow of calcium, the regenerated vesicle system would have presented a stimulated ATPase activity, which in fact stays very low (Fig. 3). Moreover, since the accumulation of calcium in the intravesic- ular medium is a requisite for the expression of the

Table 4. Effect of nucleotides on the (Ca 2+, Mg2+)-ATPase activity. ATPase activity was assayed as described under Materials and Methods. ATP, GTP, ITP and CTP were added at to the reaction medium at the indicated concentrations. The results represent the mean of triplicate determinations

(Ca 2+, Mg2+)-ATPase

nmoles Pi/mg/min.

Plasma membrane ATP CTP GTP ITP isolated from 3 mM 3 mM 3 mM 3 mM

Normal liver 1041.0 + 182.0 919.0 _+ 72.0 731.0_+ 65.0 808 + 5.3 Regenerating liver 177.0 + 9.2 40.0_+ 2.0 80.0_+ 4.0 100 + 12.0 AS-30D hepatoma 200.0 + 4.4 144.0_+ 25.0 167.0 + 6.6 167 + 6.6

48

o { " { { E r-

I I I I I

0 I 2 3 4 5 6

Ca +] mM

Fig. 7. ATP ~ Pi exchange reaction catalized by the (Ca 2+, Mg2+)-ATPase. Plasma membranes isolated from normal liver (O), 48 hr regenerating liver (O) and AS-30D hepatocarcinoma (ZX). The assay procedure was carried out as described in Mate- rials and Methods. Each point represents the mean + S.E. of three experiments.

ATP ~ Pi exchange reaction, neither the normal nor the neoplastic vesicle systems would have pre- sented a high ATP ~ Pi exchange activity as ob- served in Fig. 7. Based on these arguments, the calcium uptake data presented in this study (Fig. 6), support the hypothesis of the presence of an optimum coupling efficiency between the Ca 2+- ATPase and the calcium pump in the three vesicle systems studied in this investigation.

Although several conditions have been reported to cause uncoupling between of the Ca 2÷ pump and the Ca2+-ATPase [51-54], at present, the factors that determine the expression and the control of the different coupling states of the enzyme, are still not fully understood. Our observations could give a new perspective to this phenomenon, since we clearly demonstrate a higher coupling state AT- Pase/calcium transport in the plasma membranes of the tumor cells in comparison to the regenerated and normal cell models. The observation of the presence of an ATP --~ Pi exchange reaction in the transformed ceils similar to the one found in the

plasma membrane of normal cells, opens the possi- bility of different energy handling capabilities in the two systems. Although more experiments are needed for the complete elucidation of these phe- nomena, these initial observations suggest that the regulatory sites controling the expression of the ATPase exchange reaction in the transformed model might be altered or modulated in a spacial way. These possible alterations might confer the system with an optimum calcium transport activity associated to the lowest possible expenditure of energy.

In conclusion, the results presented here indicate that the plasma membrane of transformed cells had a more efficient mechanism to regulate the move- ment of calcium through the calcium pump with an optimum use of their energy. The nature of the mechanisms that control the expression of the ATP

Pi exchange reaction in the plasma membrane of these cells, that results in a high coupling state between the hydrolytic activity of the ATPase and the transport of calcium, could be related to chang- es in the calcium binding sites of the enzyme. This possibility is at present under investigation in our laboratory.

Acknowledgements

The authors thank Dr. Antonio Villalobo for the generous gift of the AS-A30D hepatoma cells. This study has been supported by Consejo Nacional de Ciencia y Tecnologia, M6xico, D.F. (Grant P219CCOL880870), The Third World Academy of Sciences, Trieste, Italy (Grant 97-Mex 2) and the Aida Weiss Prize (Programa Universitario en Sa- lud, Universidad Nacional Aut6noma de M6xico, M6xico D.F.). M.R.D. has been supported by a Fellowship from Consejo Nacional de Ciencia y Tecnologia, M6xico, D.F.

References

1. Carafoli E: Intracellular calcium homeostasis. Ann Rev Biochem 56: 395--433, 1987

2. Rasmussen H, Waisman DM: Modulation of cell function

in the calcium messenger system. Rev Physiol Biochem Pharmacol 95: 111-142, 1982

3. Wallach DFH: Membrane anomalies of neoplastic cells. Med Hypoth 2: 241-246, 1976

4. Whitfield JF: The roles of calcium and magnesium in cell proliferation. An overview. In: Ions, Cell Proliferation, and Cancer. Academic Press, New York, 1982, pp 283-294

5. Baker PF, Schapira AHV: Anesthetics increase light emis- sion from aequorin at constant ionised calcium. Nature 284: 168-169, 1980

6. Jaffe LF: Eggs are activated by a calcium explosion, carci- nogenesis may involve calcium adaptation and habituation. In: AL Boynton, WL Mckeehan, JF Whitfield (eds.) Ion, Cell Proliferation, and Cancer. Academic Press, New York, 1982, pp 295-310

7. Van Rossumm GDV, Galeotti T, Morris HP: The mineral content and water compartments of liver and of Morris hepatoma 5123 tc and 3924A and the changes of composi- tion occurring during necrosis in Hepatoma 3924A. Cancer Res 33: 1078-1085, 1973

8. Means AR, Rasmussen CD: Calcium calmodulin and cell proliferation. Cell Calcium 9: 313-319, 1988

9. Criss WE, Kakiuchi S: Calcium, calmodulin and cancer. Fed Proc 41: 2289-2291, 1982

10. MacManus JP: Calmodulin and Oncomodulin content of tumours. In: AL Boynton, WL McKeehan, JF Whitfield (eds.) Ions, Cell proliferation, and Cancer. Academic Press, New York, 1982, pp 489-498

11. DiPolo R, Requena J, Brinley FJ, Mullins LS, Scarpa A, Tiffert T: Ionized calcium concentrations in squid axons. J Gen Physio167: 433--467, 1976

12. Debetto P, Catley L: Characterization of a CaZ+-stimulated Mg2+-dependent adenosine triphosphatase in friend mu- rine erythroleukemia cell plasma membranes. J Biol Chem 259: 13824-13831, 1984

13. Niggli V, Penniston JT, Carafoli E: Purification of the (Ca2+,Mg2+)-ATPase from human erythrocyte membrane using a calmodulin affinity column. J Biol Chem 254: 9955- 9958, 1979

14. Niggli V, Adunyah ES, Penniston JT, Carafoli E: Purified (Ca 2+,Mg2+)-ATPase of erythrocyte membrane. Reconsti- tution and effect of calmodulin and phospholipids. J Biol Chem 256: 395-401, 1981

15. Pitts B JR: Stoichiometry of sodium-calcium exchange in cardiac sarcolemmal vesicles. J Biol Chem 254: 6232-6235, 1979

16. Lamers JMJ, Stinis JT: An electrogenic Na+/Ca 2+ anti- porter in addition to the Ca 2+ pump in cardiac sarcolemma. Biochim Biophys Acta 640: 521-534, 1981

17. Busselen P: Effect of potassium depolarization on sodium- dependent calcium effux from goldfish heart ventricles and guinea-pig atria. J Physiol 327: 309-324, 1982

18. Pershadsingh HA, Landt M, McDonald JM: A high affinity calcium-stimulated magnesium dependent adenosine tri- phosphatase in rat adipocyte plasma membranes. J Biol Chem 255: 8983-8986, 1980

49

19. Nellans HN, Popovitch JE: Calmodulin-regulated ATP- driven calcium transport by basolateral membranes of rat small intestine. J Biol Chem 256: 9932-9936, 1981

20. Dipolo R: Ca pump driven by ATP in squid axons. Nature 274: 390-392, 1978

21. Dipolo R, Beaugu6 L: Physiological role of ATP-driven calcium pump in squid axons. Nature 278: 271-273, 1979

22. WuytackF, deShutter G, Casteels R: Partialpurification of (Ca2+-Mg2+)-dependent ATPase from pig smooth muscle and reconstitution of an ATP-dependent Ca2+-transport system. Biochem J 198: 265-271, 1981

23. Michaelis EK, Michaelis ML, Chang HA, Kitos TE: High affinity Ca2+-stimulated Mg2+-dependent ATPase in rat brain, synaptosomes, synaptic membranes and micro- somes. J Biol Chem 258: 6101-6108, 1983

24. Mas-Oliva J, Williams A, Nayler WG: ATP-induced stim- ulation of calcium binding to cardiac sarcolemma. Biochem Biophys Res Comm 87: 441--447, 1979

25. Mas-Oliva J, de Meis L, Inesi G: Calmodulin stimulates both adenosine 5'-triphosphate hydrolysis and synthesis catalyzed by a cardiac calcium ion dependent adenosine triphosphatase. Biochemistry 22: 5822-5825, 1983

26. Mas-Oliva J: Synthesis of ATP-catalyzed by the (Ca 2+- Mg2+)-ATPase from erythrocyte ghosts. Energy conserva- tion in plasma membranes. Biochim Biophys Acta 812: 163-167, 1985

27. Gopalakrishna R, Anderson WB: CaZ+-induced hydro- phobic site on calmodulin: Application for purification of calmodulin by phenyl-sepharose affinity chromatography. Biochem Biophys Res Commun 104: 830-836, 1982

28. Smith FD, Walburg FE, Chang PJ: Establishment of a transplantable ascites variant of a rat hepatoma induced by 3'-methyl-4-dimethyl-aminobenzene. Cancer Res 30: 2306-2309, 1970

29. Church JG, Ghosh SH, Roufogalis BD, Villalobo A: Endo- genous hyperphosphorylation in plasma membrane from an ascites hepatocarcinoma cell line. Biochem Cell Biol 66: 1-12, 1988

30. Neville DM: Isolation of an organ specific protein antigen from cell-surface membrane of rat liver. Biochem Biophys Acta 154: 540-552, 1968

31. Brown AE, Lok MP, Elovson J: Improved method for the isolation of rat liver plasma membrane. Biochem Biophys Acta 426: 418-432, 1976

32. Post RL, Sen AK: Sodium and potassium stimulated AT- Pase. Methods Enzymol X. 762-768, 1967

33. Evans WH: Properties of 5'nucleotidase purified from mouse liver plasma membranes. Clinical Sciences 58: 439- 444, 1980

34. Veeger C, Der Vastanian DV, Zeylemaker WP: Succinate dehydrogenase. Methods Enzymol XIII 81-90, 1969

35. Reeves WJ, Fimognari GM: Lactic dehydrogenase. Meth- ods Enzymol IX 288-294, 1967

36. Lazarow A, Cooperstein S J: A microspectrophotometric method for the determination of cytochrome oxidase. J Biol Chem 189: 665-670, 1951

50

37. Nordlie RC, Arion WJ: Evidence for the common identity of glucose 6-phosphatase, inorganic pyrophosphatase, and pyrophosphate-glucose phosphotransferase. J Biol Chem 239: 1680-1685, 1964

38. Lowry OH, Rosebrough NJ, Farr AL, Randall R: Protein measurement with the Folin phenol reagent. J Biol Chem 193: 265-275, 1951

39. Fiske CH, Subbarow Y: The colorimetric determination of phosphorous. J Biol Chem 66: 375-380, 1925

40. Sill6n LG, Martell AE: Stability constants on metal-ion complexes. Suppl i Spec publication 25 The Chemical So- ciety, London, 1971

41. de Meis L, Martins OB, Alves E" Role of water, hydrogen ion, and temperature on the synthesis of adenosine triphos- phate by the sarcoplasmic reticulum adenosine triphospha- tase in the abscence of a calcium ion gradient. Biochemistry 19: 4252-4261, 1980

42. Ikehara Y, Takahashi K, Mansho K, Eto S, Kato K: Con- trast manifestation of alkaline phosphatase and 5'nucleoti- dase in plasma membranes isolated from rat liver and as- cites hepatoma. Biochim Biophys Acta 470: 202-211, 1977

43. Knowles AF, Leng L: Purification of a low affinity Mg 2+ (Ca2+)-ATPase from the plasma membranes of a human oat cell carcinoma. J Biol Chem 259: 10919-10924, 1984

44. Lin SH: Novel ATP-dependent calcium transport compo- nent from rat liver plasma membranes. The transporter and the previously reported (Ca2+,Mg2+)-ATPase are different proteins. J Biol Chem 260: 7850-7856, 1985

45. Lin SH: The rat liver plasma membranes high affinity (Ca2+,Mg2+)-ATPase is not a calcium pump. J Biol Chem 260: 10976--10980, 1985

46. Schanne FAX, Moore L: Liver plasma membrane calcium transport. Evidence for a Na+-dependent Ca2+-Flux. J Biol Chem 261: 9886-9889, 1986

47. Schiitze S, Sling HD: Does a calmodulin-dependent Ca/+- regulated Mg2+-dependent ATPase contribute to hepatic microsomal calcium uptake. Biochem J 243: 729-737, 1987

48. Iwasa Y, Iwasa T, Higashi K, Matsui K, Miyamoto E: Demonstration of a high affinity CaZ+-ATPase in rat liver plasma membranes. Biochem Biophys Res Commun 105: 488-494, 1982

49. Bachs O, Famulski KS, Mirabelli F, Carafoli E: ATP-de- pendent Ca 2+ transport in vesicles isolated from the bile canalicular region of the hepatocyte plasma membrane. Eur J Biochem 147: 1-7, 1985

50. Chan K, Junger KD: Calcium transport and phosphorylat- ed intermediate of (CaZ+-MgZ+)-ATPase in plasma mem- branes of rat liver. J Biol Chem 258: 4404-4410, 1983

51. Lotersztajn S, Hanoune J, Pecker F: A high affinity calci- um-stimulated magnesium-dependent ATPase in rat liver plasma membranes. J Biol Chem 256: 11209-11215, 1981

52. Pecker F, Lotersztajn S: Fe z+ and other divalent metal ions uncouple Ca z+ transport from (Ca2+-Mg2+)-ATPase in rat liver plasma membranes. J Biol Chem 260: 731-735, 1985

53. Hidalgo C, Petrucci DA, Vergara C: Uncoupling of Ca 2+- transport in sarcoplasmic reticulum as a result of labeling lipid groups and inhibition of CaZ+-ATPase activity by mod- ification of lysine residues of the Ca2+-ATPase polypeptide. J Biol Chem 257: 208-216, 1982

54. Ochs DL, Reed PW: Ca2+-stimulated, Mg2+-dependent ATPase activity in neutrophil plasma membrane vesicles. Coupling to CaZ+-transport. J Biol Chem 259: 102-106, 1984

Address for offprints: J. Mas-Oliva, Departamento de Bioenerg6tica, Instituto de Fisiologia Celular, Universidad Nacional Aut6noma de M6xico, Apdo. Postal 70-600, 04510 M6xico, D.F. M6xico