Potassium Transport in Normal and Transformed Mouse 3T3 Cells SANTA SPAGGIARE,' MICHAEL J. WALLACH AND JOSEPH T. TUPPER Department of Biology, Syracuse University, Syracuse, New York 13210

ABSTRACT The components of unidirectional K influx and efflux have been investigated in the 3T3 cell and the SV40 transformed 3T3 cell in exponential and stationary growth phase. Over the cell densities used for transport experiments the 3T3 cell goes from exponential growth to density dependent inhibition of growth ( 4 X 104 to 4 X lo5 cell cm -2) whereas the SV40 3T3 maintains exponential or near exponential growth (4 X 104 to 1 X lo6 cell cm-2). In agreement with previous observations, volume per cell and mg protein per cell decrease with increasing cell density. Thus, transport measurements have been expressed on a per volume basis. Total unidirectional K influx and efflux in the 3T3 cell is approximately double that of the SV40 3T3 cell at all cell densities investigated. Both cell types have similar volumes initially and show similar decreases with increasing cell density. Thus, in this clone of the 3T3 cell SV40 transformation specifically decreases unidirectional K flux. The magnitude of the total K flux does not change substantially for either cell line during transition from sparse to dense cultures. However, the components of the K transport undergo distinct changes. Both cell lines possess a ouabain sensitive component of K influx, presumably representing the active inward K pump. Both also possess components of K influx and efflux sensitive to furosemide. The data suggest this component represents a one-for-one K exchange mechanism. The fraction of K influx mediated by the ouabain sensitive component is reduced to one half its value when exponential versus density inhibited 3T3 cells are compared (63% versus 31% of total influx). No comparable drop occurs in the SV40 3T3 cell at equivalent cell densities (64% versus 56% of total influx). Thus, the pump mediated component of K influx would appear to be correlated with growth. In contrast, the furosemide sensitive component represents approximately 20 % of the total unidirectional K influx and efflux in both cell lines in sparse culture. At high cell densities, where growth inhibition occurs in the 3T3 cell but not the SV40 3T3, the furosemide sensitive component doubles in both cell lines. Thus, the apparent K-K exchange mechanism is density dependent rather than growth dependent. Alterations in membrane transport properties have been shown to occur during transition from exponential to density inhibited growth and upon cell transformation. Such changes have been implicated in the growth control exhibited by normal cells and the loss of such control in transformed cells (e.g., Pardee, '71; Holley, '72). Amino acid transport (e.g., Foster and Pardee, '69; Cecchini et al., '75; Otsuka and Moskowitz, '75) and sugar transport (e.g., Hatanaka and Hanafusa, '70; Weber, '73; Kletzien and Perdue, '74) have been investigated during cell growth and upon cell transformation. However, only a limited amount of information exists on the nature of cation transport. J . CELL.PHYSIOL., 89: 403-416.

Na and K movements across the cell membrane represent a large fraction of transport activity. The net result of the active and passive movements of these cations is the establishment of a n asymmetric distribution of Na and K, intracellular Na being low and intracellular K being high as related to the external medium. This, in turn, establishes a cell membrane potential, provides a n environment necessary for macromolecular synthesis and provides a favorable gradient for the uphill transport of certain Na-linked nutrients. Perturbation of any of these factors has been Received Dec 17, '75 Accepted Mar 29, '76 I Visiting research associate, Institute of General Physiology. Unlverslty of Parma. Parma, Italy

403

404

S . SPAGGIARE, M. J . WALLACH A N D J. T. TUPPER

shown to alter the growth characteristics of various mammalian cells. In this study we have investigated the nature of active and passive K movements in 3T3 cells and SV40 transformed 3T3 cells in sparse and dense cultures. One previous study has dealt with the nature of K uptake during exponential and density inhibited growth in a non-transformed cell (Raab and Humphreys, '74). In this study only total unidirectional K flux was investigated without reference to the components comprising this flux. One other study has dealt with the active, ouabain sensitive component of K influx during growth i n normal and transformed cells. Kimelberg and Mayhew ('75) have used the 3T3 and SV40 3T3 cell at various growth stages. However, the experimental approach used in this study suggests that the active component of K flux was not being measured under conditions which reflect normal growth conditions (DISCUSSION). DucourentPrigent et al. ('75) have investigated total and active K flux in normal versus transformed MOPC 173 cells. In this study the components were measured in the normal cell during exponential and density inhibited growth. However, studies on the transformed counterpart were not undertaken at similar cell densities for technical reasons. Thus, the data available to this point do not allow a clear assessment of the components of K transport in a normal versus transformed cell and how, if at all, these components are altered in response to cell growth and cell density. The data presented here approach this question. Our results indicate that the components of K flux fall into at least three categories in the normal and transformed 3T3 cell. These are an active component of K influx, a component of K-K exchange and a residual component, most likely representing the diffusional K flux. All three components are altered by transformation of the 3T3 cell with SV40. Furthermore, the magnitude of the active component of K flux appears to be related to the growth characteristics of the 3T3 cell whereas the magnitude of the exchange component of K flux appears to be related to the density of the culture and to be independent of the growth stage. MATERIALS A N D METHODS

Mouse BALBtc 3T3 and SV40 3T3 cells

were generously provided by Dr. Eric Mayhew and Dr. George Poste, Roswell Park Institute, Buffalo, New York. Cells were grown in Dulbecco's modified Eagle's medium (DM) supplemented with 10% fetal calf serum, penicillin and streptomycin, all obtained from GIBCO. Cells were maintained at 37°C in 5% COz. Stock cultures were grown in 75 cm2 T-flasks or 100 mm tissue culture plates and subcultured every four to five days. Isotope uptake and washout studies were carried out on cells in 60 mm tissue culture plates grown under the conditions described above. 86Rb or 42K were added i n trace amounts. All experiments were conducted i n DM plus 10% fetal calf serum. For influx studies, the appropriate tracer or tracers were added at zero time and dishes were maintained at 37°C with gentle shaking in a water bath. At appropriate intervals, dishes were removed and washed three times with approximately 5 ml of Hanks' balanced salt solution or isotonic choline chloride, depending on the nature of the experiment. Distilled H 2 0 was then added to lyse the cells. After one hour the lysate was added to counting vials and isotope content was determined by detection of Cerenkov radiation i n a liquid scintillation counter. In experiments in which internal K content of the cells was determined, lysis was carried out in distilled HzO containing 15 mM LiC1. Samples were first counted for isotope content and subsequently analyzed on a flame photometer using Li as an internal standard. Preliminary experiments indicated that three washes removed virtually all external label, as determined from the isotope content of the washes. This was verified by the observation that the linear portion of the isotope uptake curves intercepted zero time at a value of counts representing less than 1 % of the uptake. For unidirectional efflux measurements cells were pre-loaded with the appropriate tracer for approximately two hours. This time period allows the bulk of intracellular K to exchange with extracellular label, as determined from uptake experiments. Cells were then washed three times i n DM and placed i n a known volume of DM. This was designated as zero time. Cells were maintained in a 37" C bath with gentle shaking. At intervals, dishes were removed and isotope content of the cells or

CATION TRANSPORT I N NORMAL A N D TRANSFORMED CELLS

the medium was determined as described above. The unidirectional efflux rate coefficient (time- I ) was determined from plots of the fraction of counts remaining i n the cells against time. The plots were linear over the time intervals used. Thus, no correction for backflux was required. In any experiment, cell volume and cell number were determined from replicate plates washed i n a manner identical to that used for flux measurements. Over the cell densities used in this study, maximum cell loss was 8% of the initial cell count prior to washing. At greater than 2 X lo6 cells c m - 2 detachment was pronounced for SV40 3T3 cells and transport experiments were not carried out at these densities. Cells were lifted using 0.1 % trypsin, 0.1 % EDTA in Hanks' balanced salt solution. As soon as the monolayer was dispersed, the cells were passed several times through a small bore pipet to remove cell clumps and a n aliquot was placed in warm DM. Cell number and median cell volume were determined using a Coulter counter with a 100 Fm aperture. For protein measurements, cells were washed as described above, solubilized in 0.1 N NaOH and processed using the Biuret procedure. In a series of experiments, unidirectional fluxes, cell volume, cell number, protein content and K content were measured on the same cell population at increasing cell densities. In these experiments, K flux, K content and protein were measured on the same cell plate. Cell volume and number were determined on replicate plates. RESULTS

Expression of transport measurements In agreement with the findings of Foster and Pardee ('69), a decrease in volume per cell and protein per cell was observed with increasing cell density for both 3T3 and SV40 3T3 (fig. 1). Thus, expression of transport data on a per cell basis, when comparing different cell densities, would not take into account these changes i n cell size. To compensate for these changes the present data are expressed on a per volume basis, a parameter which we have found to be highly reproducible from one culture to another. This does not appear to be the case for protein content in certain cell lines (e.g., Cook et al., '75). It should be noted that the volumes determined are the vol-

405

umes of cells lifted from the monolayer condition. Thus, they may not reflect precisely monolayer cell volume. However, Lamb and McCall ('72) have shown that monolayer versus trypsin lifted cells maintain similar volumes. Furthermore, the drop i n protein per cell would indicate that the volume decrease is not a n artifact of the lifting procedure and intracellular Na and K concentrations based on these volumes yield values i n reasonable agreement with numerous mammalian cell lines (see below). Unidirectional K f l u x measurements Unidirectional K influx and efflux were determined from the kinetics of uptake or washout of 4 P K or H6Rb, the latter offering the advantage of longer half-life. In order to assess the reliability of Rb as an analog for K movements in these cells, experiments were undertaken involving simultaneous measurement of 86Rb and 42K movement. Both isotopes were added in trace amounts to the same cell populations and influx and efflux were monitored in the presence or absence of the two transport inhibitors used (ouabain and furosemide). Samples were analyzed as described in MATERIALS AND METHODS. 86Rb versus 42K movements were assessed from recounting of samples after known decay of 42K. In four experiments on 3T3 and SV40 3T3, no significant difference was observed in the uptake or washout kinetics of H6Rb versus 42K. Furthermore, the inhibition of unidirectional K influx by ouabain or furosemide was not significantly different using s6Rbor 42K as tracer. However, a consistent difference was observed when the inhibition of unidirectional K efflux by furosemide was compared using 42K or R6Rb(table 1). Inhibition of 42K efflux in two experiments on 3T3 cells represented 21 % of total efflux; a value i n agreement with the inhibition by furosemide on unidirectional K influx using either 42K or R6Rbas tracer. In contrast, inhibition of R6Rbefflux by furosemide was 46% of the total efflux in three experiments (table 1). Similar results were obtained using SV40 3T3 (table 1). No net gain of K occurs i n the presence of furosemide despite the apparent larger inhibition of efflux versus influx using R6Rb as tracer (fig. 3). Based on the equality of the inhibition by furosemide of 42K uptake and loss and the absence of

S. SPAGGIARE, M. J. WALLACH AND J . T. TUPPER

406

1o6

L

A

/

S V 4 0 373

5

\

2

1

2

h

'F

*

3

5

4

DAYS

LA

0 3T3

A S V 4 0 3T3

2.5

0

8

I

1

2

CELLS

3 cm2x

I

I

I

4

5

6

B

I

10-5

Fig. 1 Cell volume a n d protein content as a function of cell density in 3T3 a n d S V 4 0 3T3 cells. (A) Growth characteristics of a typical 3T3 and S V 4 0 3T3 population. Cells were plated at the density indicated i n 60 mm plates plus 5 ml of medium. (B) Cell volume a s a function of cell density. Median cell volume was determined using a Coulter counter. Graph represents pooled data from four 3T3 and three S V 4 0 31'3 populations. (C) Protein content a s a function of cell density. Each point represents the mean of duplicate dishes taken from the same 3T3 or S V 4 0 3T3 population at increasing densities following plating at approximately 3 X 104 cells cm-2.

407

CATION TRANSPORT IN NORMAL AND TRANSFORMED CELLS TABLE 1

Ufeect offurosemide o n unidirectional K + flux Cell type

Unidirectional K+ influx X 1 0 - 4 mole hr-lml-1

3T3

3.5 f.0.4 (10)

21 f 2 (4) 42K 20 f 2 (6) 86Rb

3.4 rt 0.3 (5)

21 f 5 (2) 42K 46 f 3 (3) 86Rb

SV40 3T3

1.4 f 0.2 (8)

27 ? 3 (2) 42K 2 3 '.3 (5) 86Rb

1.6 rt 0.2 (9)

24 f 6 (2) 42K 52 & 5 (7) SsRb

7i inhibition by furosemide

Unidirectional K+ efflux X 1 0 - 4 mole h r - ' m l - I

c/o

inhibition by furosemide

All experiments were done using sparse cell populations in exponential growth (less t h a n 1.5 X lo5 cells cm-2). Values for unidirectional K influx and efflux were determined from the kinetics of isotope uptake or washout (e.g., fig. 2 ) . Data are expressed a s mean f S.E.M. where the number of separate cell populations used i s indicated i n parentheses. 4*K and SKRb designate the % inhibition observed with each tracer. Furosemide was present at a 1 mM concentration.

TABLE 2

Components of unidirectional K+ influx in 3T3 and SV40 3T3 cells 3T3

X lo-"ole hr-k-' Total influx Active (ouabain sensitive) Exchange (furosemide sensitive) Residual (ouabain and furosemide insensitive) K, mmolefliter cells

SV40 3T3

Sparse

Confluent

Sparse

Confluent

3.5

3.1

1.4

1.8

2.2

0.9

0.9

1 .o

0.7

1.3

0.3

0.7

0.6 154 f 13 (6)

0.9 139 f 16 (6)

0.2 145 -t 18 (4)

0.1 132 f 20 (4)

The components of the total flux were determined from the mean value of that fraction of the total flux sensitive to ouabain or furosemide a t a 1 mM concentration. The residual flux i s that fraction of the total flux insensitive to these transport inhibitors. The d a t a are not meant to imply that the residual fraction represents a single component of K flux. K content is expressed as mean f S.E.M. where the number in parentheses i s the number of separate cell populations used.

changes i n net K flux in the presence of furosemide, it would appear that a specific interaction between Rb efflux and furosemide exists. At present, the basis of this is unknown. Thus, s6Rb appears to be a reliable analog to monitor K movements in both 3T3 and SV40 3T3 cells with the exception of the kinetics determined from efflux experiments i n the presence of furosemide. The data presented here are based on 42K movements under these conditions. Values for all unidirectional influxes were determined from the linear portions of the isotope uptake curves. In both cell lines isotope uptake was linear for at least 15 minutes (e.g., fig. 2 ) . Flux values were generally determined over a ten minute interval and computed based on external specific activity (cpmlmole) and cell volume, as determined from replicate cell plates. During this time interval, control plates

and those i n the presence of furosemide showed no significant alteration in K content of the cells. Thus, the criteria of steadystate or near steady-state is met (see below). In the presence of ouabain, net K loss and Na gain occurs. However, over this time interval the loss of K was maximally 1 5 % of total K. Thus, unidirectional K fluxes in the presence of ouabain over this time interval are reasonable estimates of unidirectional flux under steady-state conditions.

The components of Kflux as related to cell growth A . Exponential cell populations Cells were routinely plated at 3 to 5 X lo4 cells cm - 2 . Transport measurements were made one to two days after plating at this density. At this time both cell types exhibited exponential growth (fig. 1). Figure 4 and table 2 summarize the flux data

408

S. SPAGGIARE, M. J. WALLACH AND J. T. TUPPER

30

0 CONTROL

CONTROL 0 4 * K

A

A FUROSEMIDE 0 OUAEAIN

0.1 -

p

20-

B

2

e 0.2-

F

8 15-

\

Q 4 I

d

5-

A

I

0

0.3 -

5

I

1

I

I

I

10 15 20 25 30 35

MINUTES

0.4 -

MINUTES

Fig. 2 42K and SSRb uptake or washout kinetics i n the 3T3 cell. (A) Uptake of K a s monitored from *6Rb accumulation i n exponentially growing 3T3 cells (approximately 1 X 1 0 5 cells cm-2). Ouabain and furosemide were added at zero time at 1 mM concentrations. Unidirectional fluxes were determined from the linear portions of the uptake curves (RESULTS). Summary of 44 such experiments on 3T3 and SV40 3T3 at various densities is given in figure 4 and table 2. (B) Washout kinetics of K as monitored from S6Rb or 4*K loss. In this experiment SV40 3T3 cells (2.3 X lOJ cells cm-2) were preloaded with 86Rb or 42K for approximately two hours. Isotope loss was then monitored for each in the presence and absence of 1 mM furosemide. Slope of the line represents the unidirectional efflux rate coefficient (time-'), from which the unidirectional efflux was determined (RESULTS). Twenty-two such experiments dealing with 3T3 or SV40 3T3 at various densities are summarized in table 1 or the text.

under these conditions. Total unidirectional K influx in the 3T3 line was approximately two times that of the SV40 3T3 (3.5 X 10 - 4 versus 1.4 X 10 - 4 mole h r - ml - I). Percent of the total influx sensitive to inhibition by ouabain was similar in both cell types (63 % in 3T3 vs 64% in SV40 3T3). Pre-incubation of both cell types with ouabain for five minutes prior to isotope addition did not alter the results. This indicates that the affinity of the inhibitor for the pump site is similar in both conditions. Both cell lines were found to possess components of unidirectional K influx and efflux which were sensitive to furosemide. This component represented 20 % and 25 % of the total

unidirectional K influx and 2 1 % and 24 % of the efflux in 3T3 or SV40 3T3 respectively (table 1). In previous studies furosemide has been shown to inhibit components of cation flux which are defined as one-for-one exchange components (for references see Tupper, '75). Such components, as originally envisioned by Ussing ('49), are electrically silent, require the presence of the element in question on both sides of the membrane and do not contribute to net flux. The present data are consistent with the conclusion that the furosemide sensitive components of K influx and efflux i n the 3T3 and the SV40 3T3 cell represent one-for-one exchange of K. Prolonged in-

409

CATION TRANSPORT IN NORMAL AND TRANSFORMED CELLS

POI

$-

I-".:

4 :

t

-I

90 -

-

-

-

60-

CONTROL A FUROSEMIDE 0 OUABAIN

-A

[email protected]@-A I

TIME ( m i n ) Fig. 3 Na and K content of exponentially growing 3T3 cells (7.3 X 104 cells cm -2) in the presence or absence of 1 mM ouabain or 1 mM furosemide. At zero time the transport inhibitors were added. Cell number and volume were determined from replicate plates using the Coulter counter.

cubation of either cell line i n the presence of furosemide does not result i n alterations in net K flux (e.g., fig. 3). Unidirectional K flux is reduced but the reduction is similar for both unidirectional influx and efflux using 42K as tracer (table 1). Furthermore, the furosemide sensitive component of K influx is distinct from the ouabain sensitive component. The analysis used in the present study regarding the kinetics of isotope movements and the determination of K flux values assumes steady-state or near steadystate conditions. In order to assess the validity of our experimental procedures, estimates of steady state kinetics were made for the exponential 3T3 and SV40 3T3 populations. The kinetics of S6Rb or 4 2 K influx yielded a total unidirectional K in-

flux of 3.5 X and 1.4 X mole hr - 'ml- for 3T3 and SV40 3T3 respectively. From the kinetics of 42Kor SfiRb washout, the unidirectional efflux rate coeficient (ko, hr-I) was estimated for both cell types (e.g., fig. 2). This yielded 2.0 f 0.1 (n = 5) and 1.2 & 0.1 (n = 7) h r - l for 3T3 and SV40 3T3, respectively. Under steady state conditions Ji = Jo where Ji and Jo equal the unidirectional influx and efflux in mole hr -lml-'. Unidirectional efflux may be determined from : (1)

J, = k , . (Ki)/ml

where (Ki)/ml is the mole of intracellular K per ml of cell volume. In three experiments this yielded 154 and 145 pmole ml - I for 3T3 and SV40 3T3 respectively (table 2). Placing these values in equation (1)

410

S. SPAGGIARE, M. J . WALLACH AND J . T. TUPPER

1.5 lo5 CELL cm-2

~ S - ~ ~ I O ~ 74x10 CELL cm-2

CELL cm-*

Fig. 4 Total unidirectional K flux and the fraction of this flux inhibited by ouabain and furosemide at increasing cell density in 3T3 and SV40 3T3 cells. Unidirectional fluxes were determined from points taken on the linear portion of the isotope uptake curve (e.g., fig. 2A) under each set of conditions. Results are expressed as mean f S.E.M. The number adjacent to each point is the number of values determined o n separate cell populations. Closed symbols refer to 3T3 cells and open symbols to SV40 3T3 cells.

yields a unidirectional K efflux of 3.1 and 1.7 X mole hr-’rn1-l as compared to 3.5 and 1.4 X 1 0 - 4 mole hr-lm1-I determined from unidirectional influx studies for 3T3 and SV40 3T3 respectively. Thus, the experimental conditions employed appear to maintain steady-state or near steadystate conditions and reflect reasonably well the movements of K in both cell lines. B. Near confluent a n d confluent cell populations Near confluency of the 3T3 line was characterized by a decrease i n growth rate three days after plating at approximately

3 to 5 X lo4 cell cm-2. The SV40 3T3 line continued to grow at a n exponential or near exponential rate (e.g., fig. 1). In the 3T3 cell slight decreases in total unidirectional Kinflux (3.1 X mole hr-lm1-l) and ouabain sensitive K influx (52% of total) were observed as compared to exponential populations (fig. 4, table 2). Unidirectional K influx was inhibited 23% with furosemide, a value similar to the exponential cells. It was observed for the SV40 3T3 that total unidirectional K influx (1.5 X 10-4 mole hr-lm1-l) remained at approximately one-half that of the normal 3T3 (e.g., fig. 5) and the components of the

CATION TRANSPORT I N NORMAL AND TRANSFORMED CELLS

V

0

20

40

60 80 TIME (min.)

100

411

120

Fig. 5 Comparison of K uptake in near confluent 3T3 cells versus SV40 3T3 cells. Cell density was 2.8 X 1 0 5 cell cm - 2 and 2.2 X 105 cell cm - 2 for 3T3 and SV40 3T3, respectively. SSRb was used as tracer for K movements. Exchanged K is computed from external specific activity (cpm/mole K) and isotope uptake. Total K is determined from flame photometric analysis of the same cell plates used for uptake studies. Cell number and volume were determined on replicate plates. Estimates of unidirectional K influx, based on K exchange over the first 15-minute interval yield 1.2 X 10-4 mole hr-1ml-1 and 0.7 X 10-4 mole hr-1ml-1 for 3T3 and SV40 3T3, respectively.

flux were similar to the exponential SV40 period (fig. 1) with densities of approxi3T3 populations (fig. 4, table 2). mately 1.7 X 1 O6 cells cm - being achieved. Four to five days following plating of the Refeeding on the third day following platcells, the 3T3 line exhibited inhibition of ing did not significantly increase the dengrowth with saturating densities of approx- sities achieved (approximately 1.8 X 1 0 6 imately 4.0 X lo5 cells c m - 2 (fig. 1). Re- cells cm-*). At these densities the number feeding of the cells on the third day fol- of detached cells was insignificant, as lowing plating resulted in a n increase in judged by counting of decanted medium in the saturation density to approximately the Coulter counter. However, with con5.3 X lo5 cells cm-2. The SV40 3T3 cell tinued growth the SV40 3T3 cells became line continued at a n exponential or near spontaneously detached and their growth exponential growth rate during this time rate declined.

412

S. SPAGGIARE, M. J . WALLACH AND J . T. TUPPER

Stationary populations of the 3T3 cell cells. In these conditions the growth of the had a total unidirectional K influx of 3.1 cells was limited by the available surface x 1 0 - 4 mole hr-Im1-l (fig. 4, table 2). Of area and not the medium since the medium this total influx, 29% was ouabain sensi- supported exponential growth when surface tive, as compared to 6 3 % ouabain inhibi- area was not limiting (also Thrash and tion in the exponential cells. Paralleling Cunningham, '75). Thus, the reduction i n this decrease i n ouabain sensitive K influx ouabain sensitive K influx would appear to was a n increase in the furosemide sensi- be attributable to the density dependent tive component of K influx from 20% in growth inhibition and not to medium dethe exponential cell to 42% i n the station- pletion. SV40 3T3 cells continued exponential ary 3T3 cell. In 3T3 populations refed on the third day following plating, saturation or near exponential growth four to five densities were elevated (see above). How- days after plating. Total unidirectional K mole hr-*rnlP1.In ever, the decrease i n the ouabain sensitive influx was 1.8 X component of K influx still occurred, indi- contrast to the stationary normal 3T3, only cating the changes in transport properties a slight drop in the fraction of total influx are not due to medium depletion. It has sensitive to ouabain was observed. The innot been determined i n our confluent cells hibition by ouabain was 56% of total inwhether there exists a n increase in oua- flux in high density cultures versus 64% bain sensitive K influx shortly after addi- in the non-confluent populations (fig. 4, table 2). However, the fraction of the total tion of fresh medium. FJligsen et al. ('74) have observed a drop in Na-K ATPase ac- influx sensitive to furosemide also increased tivity upon reaching confluence in 3T3 cells significantly in the transformed cells as it and this drop could be reversed by fresh did i n the normal 3T3 cells at high denmedium or addition of fresh serum. The sity. Inhibition by furosemide represented addition of fresh medium or serum was 39% of total influx as compared to 21 % in accompanied by a stimulation of growth in the sparse SV40 3T3 cells. In table 2 are summarized the compoconfluent cultures and these authors concluded that the Na-K ATPase levels are nents of K influx in 3T3 and SV40 3T3 correlated with the growth characteristics cells in sparse and dense cultures. These of the cell (DISCUSSION). In order to further data indicate that confluency and subseassess any medium effects on the reduction quent growth arrest of the 3T3 cells are in ouabain sensitive K influx in the 3T3 characterized by an absolute decrease of cell, the following experiment was under- approximately 60% (2.2 to 0.9 X 10-4 taken. Approximately 5 X lo5 cells were mole hr-'ml-') in the amount of K influx plated under two conditions. In one case the mediated by the ouabain sensitive compocells were plated in 100 mm tissue culture nent. In the transformed 3T3 line we obplates. In the other condition, the cells served no significant change in the amount were plated i n 35 mm tissue culture plates. of ouabain sensitive K influx in sparse These plates were in turn placed in 100 versus dense cells (0.9 versus 1.0 X 1 0 - 4 mm bacteriological culture plates. The two mole hr - Iml - I ) . Dense populations of both conditions contained the same amount of 3T3 and SV40 3T3 are characterized by an medium (47 ml). The bacteriological plates approximate 2-fold increase in the amount do not support the growth of 3T3 cells due of K influx mediated by the furosemide to poor attachment of the cells to the sub- sensitive component. Based on these obserstrate. Thus, the growth was limited to the vations, it would appear that the pump 35 mm plate within the 100 mm dish. Cell mediated K influx may be related to the growth was monitored and four days after growth characteristics of the 3T3 cell in plating the cells in the 100 mm tissue cul- that it is significantly reduced upon cesture dish were in exponential growth phase sation of exponential growth. In contrast, (1.9 X lo5 cells cm-2). At this time cells the K-K exchange component would appear in the 35 mm plate were at a density of to be density dependent i n that it increases 4.9 X lo5 cells Ouabain sensitive K significantly at high cell density, indeuptake was determined at this time. It was pendent of the growth characteristics of 74% of the total uptake in the sparse cells the cell. In table 2 are also listed the comand 32% of the total in the high density ponents of K influx which are both ouabain

CATION TRANSPORT I N NORMAL AND TRANSFORMED CELLS

insensitive and furosemide insensitive. These have been designated as residual flux. The nature of these has not been characterized. Previous studies (e.g., Mills and Tupper, ’75; Tupper, ’75) have indicated that such fluxes represent, at least in part, the passive K diffusional component, the magnitude of which is dependent on membrane permeability (cm sec - l ) membrane potential and ionic distribution. Sjodin and Ortiz (‘75) have recently presented data suggesting that reduction of K permeability is paralleled by enhancement of active K influx i n frog miiscle. In the 3T3 cell the residual component of K influx increases by approximately 30% at confluency and this is paralleled by a drop in active K uptake. The change in residual flux may represent in whole or in part a change in K permeability, thus suggesting a n interrelationship between active and passive K fluxes in the 3T3 cell. DISCUSSION

Membrane transport phenomena are likely to be most meaningful if expressed on a surface area basis. However, realistic estimates of the surface area of monolayer cultured cells and a n assessment of a change in this parameter associated with growth are plagued by technical difficulties. In culture systems in which various growth stages are being compared or where comparisons between different cell types are being made, consideration must be given to inherent differences in cell morphology. Protein content and cell volume have been used as monitors of changes i n cell morphology. Foster and Pardee (‘69) observed a linear relationship between cell protein and cell volume in exponential and confluent 3T3 cells. They also observed a decrease in cell volume with increasing cell density, thus concluding that expression of data on a per cell basis was not meaningful. We have confirmed these observations on the 3T3 cell (fig. 1) and have chosen to express our results on a per volume basis since we have found this parameter and changes in this parameter to be highly reproducible. Since protein content and cell volume are related, results expressed on a per mg protein basis are comparable to the present data. Three previous studies have dealt with aspects of K transport, as determined from

413

the kinetics of isotope uptake or loss, in sparse versus confluent populations of cultured mammalian cells. Raab and Humphreys (‘74) concluded that no significant change occurred in total unidirectional K influx and efAux in sparse versus confluent chick embryo fibroblasts. Furthermore, they observed a decrease in cell volume in confluent cultures and no significant change in intracellular K concentration during the growth cycle. These results are all consistent with our present observations in the 3T3 cell. Raab and Humphreys (‘74) concluded that changes in K metabolism were not associated with growth control. However, these studies did not investigate the components of unidirectional K flux contributing to total flux. Thus, it is possible that changes occur which are similar to those observed in the present study. Ducouret-Prigent et al., (‘75) have investigated K transport in the MOPC cell line. Using a normal (ME,) and a transformed line (MF,), K transport was determined at different culture densities. In agreement with the observations of Raab and Humphreys (’74) and our present observations, no significant difference in total unidirectional K influx or in intracellular K concentration was observed between the growing and density inhibited ME2 cells. The fraction of K influx sensitive to ouabain was also evaluated. Ducouret-Prigent et al. (‘75) observed no significant reduction in the ouabain sensitive component at high density, i n contrast to our observations i n the 3T3 cell. One obvious possibility for the difference lies i n the different cell lines used. However, if this is the case, then a reduction in active K transport would not appear to be a general physiological event associated with growth inhibition. However, i n the study of Ducouret-Prigent et al. (‘75) the confluent cells were fed one day prior to the flux measurements. Elligsen et al. (‘74) have demonstrated that refeeding will raise the level of (Na-K)ATPase in confluent 3T3 cells and this elevation is maintained 24 hours after feeding. Thus, the constancy of ouabain inhibition in the ME, line may be a result of the medium change. Studies on the MF2 line were also conducted, but only at near confluency. This line does not exhibit density dependent growth inhibition. No significant difference in K concentration was

414

S. SPAGGIARE, M. J. WALLACH AND J. T. TUPPER

observed between the ME2 and MF2 lines, a n observation consistent with the present comparison between 3T3 and SV40 3T3. The fraction of total unidirectional K influx sensitive to ouabain was also similar in ME2 and MF2 at this stage of growth, which is also consistent with the present observations on 3T3 and SV40 3T3. Kimelberg and Mayhew ('75) have investigated ouabain sensitive K uptake i n 3T3 and SV40 3T3 i n sparse and confluent cultures. Comparison of this data to the present study is hampered by the experimental procedures used in the work of Kimelberg and Mayhew ('75). The fraction of total K uptake sensitive to ouabain was determined from a comparison of the isotope content of cells incubated 60 minutes i n the presence or absence of ouabain. These comparisons do not represent steady state conditions nor unidirectional fluxes. Figure 3 illustrates that after 60 minutes intracellular K levels are more than halved and intracellular Na is more than doubled. This is also evident from the data of Kimelberg and Mayhew ('75: fig. 1) in which it is seen that equilibrium or near equilibrium levels of intracellular isotope are far below control levels. Under these conditions, backflux of isotope is significant. Thus, isotope content of the cells i n the presence of ouabain after 60 minutes is not a measure of unidirectional K influx sensitive to ouabain and, in turn, is not comparable to the present data. It is also probable that such large changes in intracellular Na and K alter the activity of the ouabain sensitive pump (e.g., Post et al., '60; Ducouret-Prigent et al., '75). Several studies have dealt with the specific activity of (Na-K)-ATPase in normal versus transformed cell lines and in sparse versus confluent cultures. There does not appear to be a general agreement as to the specific activity of this enzyme in normal versus transformed cells when compared within the same cell line. Perdue et al. ('71), Graham ('72), and YoshikawaFukada and Nojima ('72), have all observed high specific activities of (Na-K)ATPase in normal rat liver, BHK and 3T3 cells respectively a s compared to a transformed counterpart. In contrast Sheinin and Onodera ('72) have observed levels of (Na-K)-ATPase in various transformed lines of 3T3 cells which are above or below

the normal 3T3. Kimelberg and Mayhew ('75) have observed elevated levels of (NaK)-ATPase in SV40 3T3 cells as compared to normal 3T3 at cell densities of approximately 1 X l o 5 cm-2. This difference subsequently diminished at higher cell densities. In comparisons of sparse versus confluent cultures of normal and transformed cells, Lelievre et al. ('72), Lelievre and Parof ('73) and Elligsen et al. ('74) all conclude that the specific activity of (Na-K)ATPase drops at confluency for cells exhibiting density dependent growth inhibition whereas it is unaffected i n the transformed counterpart at confluency. These observations are consistent with the present study on ouabain sensitive K uptake in sparse versus confluent 3T3 and SV40 3T3, if it is assumed that ouabain sensitive K uptake corresponds to the activity of Na-K activated, ouabain sensitive ATPase. However, differences in assay conditions, unmasking of latent enzyme and differential sensitivity of enzyme i n normal versus transformed andlor sparse versus confluent cultures must be assessed before direct comparisons between transport studies and enzyme levels are meaningful. It would appear, however, that transport measurements provide information under conditions in which such variables are non-existent or minimized. The present study is consistent with the view of Elligsen et al. ('74) that the degree of active K transport is a function of the growth behavior of the cell. Density dependent inhibition of growth in the 3T3 cell is paralleled by a decrease i n active K transport. At equivalent cell densities no significant reduction i n active transport occurs i n the SV40 3T3. The data also indicate that transformation of the 3T3 cell by SV40 results in a n absolute decrease in total unidirectional K flux and the components of the flux. The present study has identified components of K influx and efflux in the 3T3 cell which fall into the category of one-forone exchange (Ussing, '49). Such exchange mechanisms have been identified for Na, K, C1 and Ca in a variety of cells (see Tupper, '75, for references). Furosemide has been shown to selectively inhibit Na-Na exchange in the red cell at physiological levels of extracellular Na (Dunn, '70; Sachs,

CATION TRANSPORT IN NORMAL AND TRANSFORMED CELLS

'71). It has also been shown to selectively inhibit Na-Na and K-K exchange in the Ehrlich ascites cell (Mills and Tupper, '75; Tupper, '75). Exchange components are electrically silent and thus their inhibition should not alter cell membrane potential. Using fluorescent dye techniques, Laris et al. ('75) have monitored membrane potentials in the Ehrlich ascites cell. Furosemide does not alter the Ehrlich cell membrane potential (personal communication, P. C. Laris), thus substantiating the conclusion that it selectively inhibits exchange components. The physiological significance of exchange components has not been firmly established. Certain observations suggest they may represent an alternate mode for the active Na-K transport system. For example, in the red cell (Garrahan and Glynn, '67) and the Girardi cell (Lamb and McCall, '72) increases in Na-K transport are paralleled by decreases in Na-Na exchange. In the Ehrlich ascites tumor cell the magnitude of the K-K exchange component is cell cycle dependent and reciprocally related to the magnitude of the active K transport mechanism in S and Gz phase (Mills and Tupper, '76). The present study is consistent with the hypothesis of a reciprocity between active K transport and K-K exchange in the normal 3T3 cell in that the absolute decrease in ouabain sensitive K influx is paralleled by an increase in furosemide sensitive K influx at confluency (table 2). However, at high density in the SV40 3T3, an equivalent increase in the furosemide sensitive component occurs with no concomitant decrease in the ouabain sensitive component. These observations lead us to suggest that the magnitude of the exchange component is density dependent and not growth dependent. Cecchini et al. ('75) also present evidence for an apparent density dependent change in membrane transport phenomena in 3T3 and SV40 3T3. In investigating the A and L amino acid transport systems in these cells they observe a decrease in the A system and an increase in the L system as both the 3T3 and SV40 3T3 cells increase in density. Cell membrane transport properties have been suggested as playing a primary rolc in the regulation of cell growth. Cation transport mechanisms are directly linked to cell cation content, cell membrane po-

415

tential, cation dependent accumulation of cell nutrients and maintenance of intracellular environments necessary for macromolecular synthesis. However, no direct evidence exists indicating the specific manner by which transport phenomena are integrated with growth control. The present study indicates that such changes occur in cation transport. They indicate further, however, that the changes fall into at least two categories, those unique to the process of transformation and those unique to the physical state of the culture. ACKNOWLEDGMENTS

Supported by grants to J. T. T. from the American Cancer Society (BD-181), the National Institutes of Health (CA 172 0301) and the Syracuse University Research fund. LITERATURE CITED Banerjee, S. P., and H. B. Bosmann 1976 Rubidium transport and ouabain binding in normal and virally transformed mouse fibroblasts. Exp. Cell Res., 100: 153-158. Cecchini, G., M. Lee and D. L. Oxender 1975 Amino acid transport systems in 3T3 and SV3T3 mouse cells. Fed. Proc., 34: 556. Cook, J. S., G. Vaughan, W. Proctor and E. Brake 1975 Interaction of two mechanisms reeulating alkali cations in HeLa cells. J. Cell. Phisiol., 86: 59-70. _. .~ Ducouret-Prigent, B., L. Lelievre, A. Parof and A. Kepes 1975 Relationships between intracelM a r K concentrations and K fluxes in growing and contact-inhibited cells. Biochim. Biophys. Acta, 401;119-127. Dunn, M. J. 1970 The effects of transport inhibitors on sodium outflux and influx in red blood cells: evidence for exchange diffusion. 49; 18041814. Elligsen, J., J. Thompson, H. Frey and J. Kruuv 1974 Correlation of (Na-K)-ATPase activity with growth of normal and transformed cells. Exp. Cell Res., 87: 233-240. Foster, D. U., and A. B. Pardee 1969 Transport of amino acids by confluent and nonconfluent 3T3 and polyoma virus-transformed 3T3 cells growing on glass cover slips. J. Biol. Chem., 244:2675-2681. Garrahan, P. J., and I. M. Glynn 1967 The behavior of the sodium pump in red cells in the absence of external potassium. J. Physiol., 197: 159-175. Graham, J. M. 1972 Isolation and characterization of membranes from normal and transformed tissue culture cells. Biochem. J., 130: 1113-1124. Hatanaka, M., and H. Hanafusa 1970 Analysis of a functional change in membrane i n the process of cell transformation by Rous sarcoma virus: Alteration in the characteristics of sugar transport. Virology, 41 ; 647-652. Hempling, H. G. 1972 The ascites tumor cell.

416

S. SPAGGIARE, M. J . WALLACH AND J . T. TUPPER

In: Transport and Accumulation in Biological Systems. E. J. Harris, ed. Butterworth and Co. Ltd., London, p. 271. Holley, R. W. 1972 A unifying hypothesis concerning the nature of malignant growth. Proc. Natl. Acad. Sci. (U.S.A.), 69:2840-2841. Kimelberg, H., and E. Mayhew 1975 Increased ouabain-sensitive Rb uptake and sodium and potassium ion-activated adenosine triphosphatase activity in transformed cell lines. J. Biol. Chem., 250: 100-104. Kletzien, R., and J. F. Perdue 1974 Sugar transport in chick embryo fibroblasts. J. Biol. Chem., 249: 3366-3374. Lamb, J. F., and D. McCall 1972 Effect of prolonged ouabain treatment on Na, K, C1 and Ca concentration and fluxes i n cultured human cells. J. Physiol., 225: 599-417. Laris, P. C., H. Pershadsingh and R. Johnstone 1975 Monitoring membrane potentials in Ehrlich ascites tumor cells by means of a fluorescent dye. J. Gen. Physiol., 66: 14a. Lelievre, L., and A. Parof 1973 Enzyme activities in membranes from three pleiotypes of the murine plasmocytoma MOPC 173,cultivated i n vitro. Biochem. Biophys. Acta, 291 : 671-679. Lelievre, L., B. Prigent and A. Parof 1971 Contact inhibition -plasma membrane enzymatic activities in cultured cell lines. Biochem. Biophys. Res. Commun., 45: 637443. Mills, B.,and J . T. Tupper 1975 Cation permeability and ouabain-insensitive cation flux in the Ehrlich ascites tumor cell. J . Membrane Biol., 20: 75-97. 1976 Cell cycle dependent changes in potassium transport. J. Cell. Physiol., 89: 123-132. Otsuka, H.,and M. Moskowitz 1975 Difference in transport of leucine in attached and suspended 3T3 cells. J. Cell. Physiol., 85: 665-674. Pardee, A. B. 1971 The surface membrane as

regulator of animal cell division. In Vitro, 7: 95-105. Perdue, J . F., R. Kletzien, K. Miller, G. Pridmore and V. Wray 1971 The isolation and characterization of plasma membranes from cultured cells. Biochim. Biophys. Acta, 249: 435-453. Post, R. L., C. Merritt and C. Albright 1960 Membrane adenosine triphosphates as a participant i n the active transport of sodium and potassium in the human erythrocyte. J. Biol. Chem., 235: 1796-1802. Rabb, J. L., and T. Humphreys 1974 Potassium ions and the regulation of cell growth i n culture. Exp. Cell Res., 89: 407-410. Sachs, J. R. 1971 Ouabain-insensitive sodium movements in the human red blood cell. J. Gen. Physiol., 57: 259-281. Sheinin, R., and K. Onodera 1972 Studies of the plasma membrane of normal and virustransformed 3T3 mouse cells. Biochim. Biophys. Acta, 274: 49-63. Sjodin, R. A., and 0. Ortiz 1975 Resolution of the potassium ion pump in muscle fibers using barium ions. J. Gen. Physiol., 66: 269-286. Thrash, C., and D. Cunningham 1975 Growth limitation of 3T3 mouse fibroblasts by available growth surface area and medium components. J. Cell. Physiol., 86: 301-310. Tupper, J. T. 1975 Cation flux in the Ehrlich ascites tumor cell; evidence for Na-Na and K-K exchange diffision. Biochim. Biophys. Acta, 394: 586-596. Ussing, H. H. 1949 Transport of ions across cellular membrane. Physiol. Rev., 29: 127-155. Weber, M. J. 1973 Hexose transport in normal and in Rous sarcoma virus-transformed cells. J. Biol. Chem., 248: 2978-2983. Yoshikawa-Fukada, M., and T. Nojima 1972 Biochemical characteristics of normal and virally transformed mouse cell lines. J. Cell. Physiol., 80: 421430.

Note added in proof: Banerjee a n d Bosmann (‘76) h a v e reported t h a t t h e fraction of *GRb uptake sensitive to ouabain is reduced in confluent 3T3 cells as compared to confluent SV40 3T3 cells. This is in qualitative agreement with t h e present study.

Potassium transport in normal and transformed mouse 3T3 cells.

Potassium Transport in Normal and Transformed Mouse 3T3 Cells SANTA SPAGGIARE,' MICHAEL J. WALLACH AND JOSEPH T. TUPPER Department of Biology, Syracus...
1019KB Sizes 0 Downloads 0 Views