Electrogenic Na+/K+-transport in human endothelial cells

Pfltigers Arch (1993) 424:301-307

Et [ hfin Journal of Physiology 9 Springer-Verlag 1993

Electrogenic Na+/K+-transport in human endothelial cells Masahiro Oike 1, Guy Droogmans 1, Rik Casteels 1, Bernd Nilius 2 1 K. U. Leuven, Department of Physiology, Campus Gasthuisberg, B-3000 Leuven, Belgium 2 K. U. Leuven, Max Planck Group Molecular and Cellular Physiology, Campus Gasthuisberg, B-3000 Leuven, Belgium Received December 8, 1992/Received after revision February 17, 1993/Accepted February 23, 1993

Abstract. Na+/K + pump currents were measured in endothelial cells from human umbilical cord vein using the whole-cell or nystatin-perforated-patch-clamp technique combined with intracellular calcium concentration ([Ca2+]i) measurements with Fura-2/AM. Loading endothelial cells through the patch pipette with 40 mmol/1 [Na +] did not induce significant changes of [Ca2+]i. Superfusing the cells with K+-free solutions also did not significantly affect [Ca2+]i. Reapplication of K + after superfusion of the cells with K+-free solution induced an outward current at a holding potential of 0 mV. This current was nearly completely blocked by 100 ~tmol/1 dihydroouabain (DHO) and was therefore identified as a Na+/K + pump current. During block and reactivation of the Na+/K + pump no changes in [Ca2+]i could be observed. Pump currents were blocked concentration dependently by DHO. The concentration for half-maximal inhibition was 21 ~tmol/1. This value is larger than that reported for other tissues and the block was practically irreversible. Insulin (10-1000U/1) did not affect the pump currents. An increase of the intracellular Na + concentration ([Na+]i) enhanced the amplitude of the pump current. Half-maximal activation of the pump current by [Na+]i occurred at about 60 mmol/1. The concentration for half-maximal activation by extracellular K + was 2.4 _+ 1.2 mmol/1, and 0.4 _+ 0.1 and 8.7 _+ 0.7 mmol]l for T1+ and NH4+ respectively. The voltage dependence of the DHO-sensitive current was obtained by applying linear voltage ramps. Its reversal potential was more negative than - 1 5 0 mV. Pump currents measured with the conventional whole-cell technique were about four times smaller than pump currents recorded with the nystatin-perforated-patch method. If however 100 gmol/1 guanosine 5"-O-(3-thiotriphosphate) (GTPTS) were added to the pipette solution, the currents measured in the ruptured-whole-cell-mode were not significantly different from the currents measured with the perforated-patch technique. We suppose that the use of the perforatedpatch technique prevents wash out of a guanine nucleotiCorrespondence to: B. Nilius

de-binding protein (G-protein)-connected intracellular regulator that is necessary for pump activation.

Key words: Endothelium - Perforated-patch-clamp technique - Electrogenic Na+/K + pump - Ca2+-signal ling

Introduction The maintenance of the electrochemical gradients for Na + and K + ions is essential for almost all cells and is mainly achieved by the action of the Na+/K + adenosine triphosphatase (ATPase). This pump transports three Na + ions outwards and two K + ions inwards for each hydrolysed ATP and thus generates a current. These pump currents have been measured in many cell types and can be used as a sensitive indicator of the pump function (see [7] for a review). In coronary endothelial cells readmission of K + after an exposure to K+-free solution induces a hyperpolarization that can be blocked by dihydroouabain (DHO) [6]. This report was the first hint of the existence of an electrogenic Na+/K + ATPase in endothelial cells, that contributes approximately 8 mV to the resting potential of these cells. Agonists such as histamine induce large inward currents through non-selective cation channels, which are mainly carried by Na + ions [24]. Restoration of the electrochemical gradients for Na + and K + after agonist stimulation would certainly require activation of a Na § extrusion mechanism. The ensuing rise in intracellular sodium concentration ([Na+]i) during agonist stimulation could also activate the Na+/K + ATPase and modulate the membrane potential and hence the driving force for Ca 2+ entry. Pump currents have not been measured directly in endothelial cells to date. In this report we show for the first time that the Na+/K + ATPase in endothelial cells induces a pump current. It has been reported that ouabain completely changes the pattern of intracellular Ca2+

302 transients in e n d o t h e l i a l cells i n d u c e d b y v a s o a c t i v e agonists, such as b r a d y k i n i n [17]. T h e s e o b s e r v a t i o n s m a y i n d i c a t e that the N a + / K + A T P a s e a c t i v i t y c o u l d p l a y an i m p o r t a n t role, n o t o n l y for the m a i n t e n a n c e o f the e l e c t r o c h e m i c a l g r a d i e n t s o f N a § a n d K + ions, b u t also for C a 2+ signalling. H o w e v e r , in our e x p e r i m e n t s n e i t h e r c h a n g e s in N a + / K + A T P a s e a c t i v i t y n o r c h a n g e s in [Na+]i w e r e a s s o c i a t e d w i t h s i g n i f i c a n t c h a n g e s in the i n t r a c e l l u l a r C s + c o n c e n t r a t i o n ([Ca2+]0.

Materials and methods Isolation and culture of endothelial cells. Endothelial cells were prepared from human umbilical cord veins as described previously [15, 29]. In short, cells were grown in Medium 199 containing 10% human serum, 2 mmol/I L-glutanfine, 100 U/ml penicillin and 100 rag/1 streptomycin. The culture medium was exchanged every 48 h. Cells were detached by exposure to 500 rag/1 trypsin in a C s +- and Mg2+-free solution for about 3 rain, reseeded on gelatincoated cover slips and kept in culture for 2 - 4 days before use. We used only non-confluent cells in our experiments.

nected to a computer system (ATARI Mega 4). Fluorescence signals and transmembrane current were sampled at a slow frequency. With this technique we were able to measure simultaneously ionic currents and C s + transients induced in these cells by agonist stimulation, application of thapsigargin and thimerosal and by shear stress [10, 11, 24, 29]. The absence of Ca 2+ transients during the various procedures used in the present report can therefore not be due to a deficiency of the recording technique.

Electrophysiology. The patch-clamp technique was applied in the standard whole-cell mode or in the perforated-patch configuration using a patch-clamp amplifier EPC-9 (List, Darmstadt, Germany) connected to a second Atari Mega 4 computer. Membrane currents were filtered at 1 kHz with an 8-pole Bessel filter and digitized on-line at sample intervals of 250 gs. Current/voltage (I/V) relationships were determined from linear voltage ramps from - 2 0 0 to + 50 mV with a duration of 500 ms. Control experiments showed that there was no significant time-dependent delay of the change in pump current during application of a voltage step. It is therefore likely that the pump currents measured during the voltage ramps represent steady-state currents.

Statistics. Pooled data are given as mean + SEM. Significance was tested at the 5% level by means of Student's t-test for unpaired data.

Materials. For measurements of [Ca2+]i, cells were incubated for 20 min at room temperature with 2 gmol/1 Fura-2-acetoxymethylester (Fura-2/AM, Molecular Probes, Eugene, Ore., USA) dissolved in normal Krebs' solution and thereafter for another 20 rain at 37~ In all experiments we used as extracellular solution (in mmol/ 1): NaC1 131, KC1 5.9, MgC12 1.2, CaClz 2.5, glucose 11.5, 4-(2hydroxyethyl)-l-piperazineethanesulphonic acid (HEPES)-NaOH 11.5, pH 7.3. In K+-free solutions we used 136.9 mmol/1 NaC1. For pump stimulation we applied various concentrations of the monovalent cations K +, T1+ or NIL+ and reduced the [Na +] correspondingly. All experiments were performed at room temperature (20-22~ The pipette solution for normal whole-cell recordings without nystatin contained in mmol/l: 110 CsC1, 40 NaC1, 5 MgC12, 0.1 ethylene-bis(oxonitrito)tetraacetic acid (EGTA), 5 Na2ATR 11.5 HEPES. If the pipette Na + concentration ([Na+]p) was changed (between 10 and 100 mmol/1), CsC1 was substituted by NaC1 and vice versa. The same pipette solution, but without Na2ATP and MgCI2, was used when the membrane patch was perforated with nystatin [16, 26]. A stock solution of 50 g/1 nystatin dissolved in dimethylsulphoxide was used. Immediately prior to use the stock solution was diluted 1000 times to a finn nystatin concentration of 50 mg/1 in the pipette solution and ultrasonicated. The tip of the pipette was filled by capillarity with nystatin-free pipette solution and back-filled with the nystatin-containing solution. Access to the cell interior occurred within 5 - 1 0 min after seal formation; within 20 to 30 min series resistances dropped to 8 - 3 0 Mr2 (mean 15 + 2 Mr2, n = 23). Access to the cell interior was monitored by recording the membrane potential. Because Cs + was used in the pipette solution the values of the measured potentials are irrelevant and do not reflect the normal resting potential of these cells. The membrane capacitance was 2 5 - 2 1 6 pF (mean 110 _+ 12 pK n = 23 cells). The osmolarity of all bath and pipette solutions was adjusted to 300 mOsm/kg.

Measurement of [Ca2+]i. After loading with Fura-2/AM cells on cover slips were washed three times in the experimental chamber with Krebs' solution to remove the extracellular Fura-2/AM. The system for [Ca2+]i measurements and calibration are based on a method previously described [21]. In short, a single cell was excited alternately with light of 360 and 390 nm wave lengths via a rotating filter wheel (speed 2 - 3 rev/s) and fluorescence measured at 510 nm. Apparent free [Caa+]~ was calculated from the fluorescence ratio R of the background con'ected fluorescence signals [12]. For data acquisition we used an 8-channel A/D converter (Max Planck Institute of Biophysical Chemistry, G6ttingen) con-

Results Voltage-dependent pump currents and [Ca2+]i in endothelial cells P u m p currents w e r e a c t i v a t e d b y a p p l i c a t i o n o f 11 m m o l f l K + to cells w h i c h w e r e e x p o s e d p r e v i o u s l y for 10 m i n to a K + - f r e e solution. A s w i l l b e s h o w n b e low, this e x t r a c e l l u l a r K + c o n c e n t r a t i o n ([K+]o) i n d u c e s a n e a r - m a x i m a l a c t i v a t i o n o f the N a + / K + ATPase. F i g ure 1 A s h o w s a t y p i c a l e x p e r i m e n t using a n y s t a t i n - p e r f o r a t e d p a t c h in w h i c h r e a p p l i c a t i o n o f 11 mmol/1 [K+]o i n d u c e d an o u t w a r d current at the h o l d i n g p o t e n t i a l o f 0 m V a n d with [Na+]p o f 40 mmol/1. This current w a s l a r g e l y a b o l i s h e d after a p p l i c a t i o n o f 100 pmol/1 D H O . In this e x p e r i m e n t the current d i d n o t r e c o v e r after w a s h i n g out D H O , b u t a partial r e c o v e r y o f the current was observed in most experiments. Readmission o f 11 mmol/1 [K+]o in the p r e s e n c e o f 100 gmol/1 D H O d i d n o t i n d u c e a n y current n o r d i d this D H O c o n c e n t r a t i o n affect the h o l d i n g current in K + - f r e e m e d i u m . W e tried to m e a s u r e the D H O - b l o c k e d p u m p current at different m e m b r a n e p o t e n t i a l s b y a p p l y i n g linear voltage r a m p s f r o m - 2 0 0 to + 50 m V w i t h a d u r a t i o n o f 500 ms. In Fig. 1 A these a p p e a r as fast transients in the v o l t a g e p r o t o c o l and the spikes in the current trace corres p o n d to the c o n c o m i t a n t currents. T h e currents i n d u c e d b y these r a m p s in K + - f r e e solution (trace 1), at the p e a k o f the p u m p c u r r e n t (trace 2) a n d after n e a r m a x i m a l b l o c k o f the current b y 100 g m o l B D H O (trace 5), are r e p r e s e n t e d as a f u n c t i o n o f the i n s t a n t a n e o u s p o t e n t i a l in Fig. 1 B. T h e I/V c u r v e in K + - f r e e s o l u t i o n (trace 1) is r e l a t i v e l y linear. T h e current r e v e r s e s n e a r 0 m V a n d is t h e r e f o r e m a i n l y due to p a s s i v e leaks. T h e s m a l l diff e r e n c e b e t w e e n the current r e c o r d e d in K + - f r e e s o l u t i o n and that o b s e r v e d in 11 mmol/1 [K+]o in the p r e s e n c e o f D H O (trace 5) is m o s t l i k e l y due to an i n c o m p l e t e b l o c k

303

K+,J

A 11

11 K " 100 pM DHO AOK §

OK+

10

100#M OHO

1

B

11 K +

1 .uM DHO

----N. . . . . . . . . . . . . . . . . . . . . . . . . . . . '

BlOO

I

I

60 s

'

, [pAl

pA

20

2

.........

50

: .......................

60 s

17

C % fnhibition

0

---

-50

100.

~

)

0 mM K"

50 84 -100

-200 ClOO

,

,

,

,

,

-150

-100

-50

0

50

0 (3)=~j// (4)

IpgMp [pAl

75

1

lO

lOO

lOOO

C,HO [IJmol/I]

50 25 0 -25

V [mVl -50 -200

, -150

, -100

, -50

, 0

50

Fig. 1 A - C , Voltage-dependent Na+/K + pump currents and intracellular calcium concentration [Ca2+]i. A A current trace induced by readmission of 11 mmol K + extracellularly ([K+]o) and with 40 mmol/1 Na + in the pipette ([Na+]p) and the effect of 100 lamol/l dihych'oouabain (DHO) on this current. The holding potential is 0 mV. The dashed line indicates the zero current level. The artefacts in the current trace correspond to currents induced by applying linear voltage ramps between - 2 0 0 and + 50 mV. B Current/ voltage ([/V) relationships obtained from linear voltage ramps in 0 retool/1 K + (1) and after readmission of l l mmol K + before (2) and after (5) application of 100 Ixmol/1 DHO. The numbers refer to the corresponding numbers in A. C I/V relationship of the DHOblocked current (2-5), obtained by subtracting trace 5 from 2

of the pump current. It cannot be ruled out however that readmission of extracellular K + activates a membrane conductance although we never observed activation of an inwardly rectifying K + current under our experimental conditions with high [Na+]i and [Cs+]i. Although it cannot be excluded that DHO affects some membrane

Fig. 2 A - C . Concentration dependent inhibition of Na+/K+ pump currents by DHO. A and B The effects of 100 and I gmol/1 DHO on the current activated by readmission of 11 mmol/1 K + to ceils which have been superfused during 10 min with a K+-free solution. [Na+]p was 40 mmol/1, The three artefacts in A are from slowly sampled ramps. C The concentration dependence of the DHO inhibition of the pump current. The inhibition was measured after the current trace had reached a new stable level. The curve was drawn using the equation percentage INHIBITION = 100/ (1 + KJ[DHO]), where Kd represents the concentration for half maximal inhibition (Ka = 21 pmol/1)

conductances, we have defined the pump current as the difference current in the absence and presence of DHO. The difference between both currents (Fig. 1 C) shows an apparent reversal potential around - 1 4 0 mV. From three cells in which the I/V curves in the absence and presence of DHO clearly intersected we calculate a mean reversal potential of - 1 5 6 _+ 11 mV. In three other cells the I/V curves in the absence and presence of DHO did not intersect, but converged to an apparent reversal potential that was even more negative than - 1 5 0 mV. In a few other cells the pump current showed only little voltage dependence in the voltage range from - 2 0 0 to + 50 mV and it was not possible to estimate a reversal potential from these I/V curves. Incubation of the cells in K+-free solution and loading them with a high [Na +] (40 mmol/1) through the pipette did not significantly affect the [Ca2+]i (Fig. 1 A, lower trace). Moreover, during activation of the outward

304 ]p,[el

{4~

•

"" {5

,y

//1" + 9

A

40 pA

100Na+p

///-o

.

.

.

.

.

.

r

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

t

60 s

B

20 pA (4) O0

~} i

0.01

. . . . . . . .

I

0.1

. . . . . . . .

I

. . . . . . . .

1

i

,

I

lOmmo[/I

Fig. 3. Effect of various extracellular cations on the pump current. The Na+/K+ adenosine triphosphatase (ATPase) of cells which were exposed for 10 rain to K+-free solution was activated with various concentrations of the monovalent cations K + (e), T1+ ( I ) or NH/ (O). The resultant pump currents, expressed relative to the pump current induced by 5.9 mmol/1 K +, were plotted as a function of the corresponding cation concentration. The number beside each symbol indicates the number of cells. These data were fitted with a Hill equation as described in the text

C

20 pA

o

,o72

_.................. i_:.__

.

.

.

.

.

r

60 s

D

200

current or application of D H O no significant changes in [Caa+]i occurred. These unexpected findings were confirmed in all tested cells (n = 29). This lack of change in [Ca2+]i was not due to a failure of our Ca 2+ analysis because we could record Ca 2+ transients induced by various agonists and by shear stress in these and other cells [24, 29] using either perforated or ruptured patches.

B

60 s

l~,uw [pA]

(s)

150

100

50

Concentration dependence of the block of endothelial pump currents by DHO Figure 2 (A and B) shows outward currents that are activated by reapplication of 11 mmol/1 [K+]o and their inhibition by 100 and I pmol/l D H O applied at the plateau of the currents. The block at concentrations exceeding 10 ~tmol/1 developed very quickly and was never completely reversible at these concentrations. Figure 2 C shows the concentration dependence of the inhibition of the p u m p current by DHO. The concentration required for half-maximal inhibition of the current (Ks) was 21 pmol/1. This rather high value is similar to that observed in cardiac muscle, where a Ks for D H O binding at 10.8 retool/1 [K+]o of 24 pmol/1 has been reported [3]. This high value might suggest that the a-subunit of the p u m p isoform has a low affinity for ouabain [14]. A characteristic property of this isoform type is that it is not sensitive to insulin [33], although up till now the possibility that the reported effects of insulin are indirect and not related to isoform subtypes has not been excluded. We therefore investigated the effect of insulin on the p u m p current in these endothelial cells but could not observed any effect of 10 (n = 2 cells), 100 (n = 4 cells) or 1000 (n = 2 cells) U/1 insulin on the p u m p currents activated by readmission of 11 retool/1 [K+]o (data not

i

i

i

i

I

i

0

20

40

60

80

100

[ Na+] I' [mmol/I]

Fig. 4 A-D. Dependence of the Na+/K+-pump currents on intracellular Na +. Pump currents, measured with the nystatin-perforated-patch method, were obtained at various [Na+]p (A 100 retool/l, B 40 mmol/1, C t0 retool/l). D The amplitude of the pump current as a function of [Na+]p. Current amplitudes were measured as the maximal current activated by readmission of K + at a holding potential of 0 mV. Each point represents the mean + SEM of 5-11 ceils

shown). This finding is therefore compatible with our previous contention about the isoform type of the pump.

Concentration dependence of the pump current activated by extracellular cations Pump currents were activated by various [K+]o after prior exposure to K+-free solution for 10 min. The data, as presented in Fig. 3, show a clear dependence of the p u m p current on the [K+]o used to activate the pump. In this figure each data point represents the p u m p current

305 11 K I~

A

10 p A

whole cell D 5o r

n=5 IpuMP [pAl n=l 1

i 60 s

40 mM Na~p 40

1 1 K .F

T

30 10 p A

i: ot 60 s

'

perforated

20

i

n=9 lO 11 K +

ii whole cell

perforated

whole celI+GTP~S

20 p A

50 s

whole celI+GTP3,S

Fig. 5 A-D. Na+/K+ pump currents obtained using the conventional whole-cell technique (whole cell) and the nystatin-perforated-patch method (perforated). In both experiments, [Na+]v was 40 mmol/1. In whole-cell experiments 5 mmol/1 ATP and 5 mmol/1 MgClz were added to the pipette solution. The holding potential was 0 mV. Cells were superfused with K+-free solution for t0 rain before pump currents were activated by reappfication of 11 mmol/1 K +. In the presence of 100 ~tmol guanosine-5'-O-(3-thiotriphosphate) (GTP?,S) in the patch pipette, nearly the same current size was

measured in the whole cell configuration as in the perforated patch mode. A, B and C show typical current traces for each method. Dashed lines indicate the zero current level. D shows the mean _+ SEM' of the current amplitude obtained from 9, 11, and 5 cells, using the whole cell, the perforated patch method, and the whole cell method in presence of 100 ~tmol/1GTP?S, respectively. The differences between "whole cell" and "perforated" are significant, but not between "perforated" and "whole cell + GTPTS"

(Ip,roi) as a fraction of the current activated by 5.9 mmol/1 [K+]o. The Hill equation

Dependence of pump currents on [Na+]i

Iv,re1 = Ip.... [1 + (KoJ[K+]o) n] 1 was fitted to the data, where Ip,max represents the current amplitude at maximal p u m p stimulation, Ko.5 the concentration required for half-maximal activation, and n a parameter which describes the steepness of the concentration dependence. The values obtained for the best fit w e r e Ip,max = 1.42 _+ 0.26, Ko.~ = 2.4 +_ 1.2 mmol/1 and n = 0.85 _+ 0.15. Figure 3 also shows the concentration dependence of the p u m p current activated with either T1 + or NH~+. It is obvious that the activation curve in the presence of T1 + is shifted to lower concentrations and that NH4+ is less efficient than K + for activation of the Na+/K + ATPase. The data for T1 + were also fitted to the Hill equation, with Ip,max ~- 1.40 + 0.21, Ko5 = 0.4 _+ 0.1 mmol/1 and n = 1.1 _+ 0.19. The data obtained with NH4+ as the pump-activating cation were also fitted to the Hill equation, but Ip.... was fixed at 1.41, the mean of the values obtained for the other two cations, because these data did not show a clear-cut plateau level. For the two remaining parameters we obtained Ko.5 = 8.7 + 0.7 mmol/1 and n= 1.0+0.1.

We used three different values of [Na+]p to adjust the [Na+]i. A clear dependence of the p u m p current on [Na+]p could be observed in all experiments. Figure 4 ( A - C ) shows representative traces of p u m p currents activated by 11 mmol/1 [K+]o at various [Na+]p. At 100 mmol/1 [Na+]p the p u m p current was 155 -+ 47 pA (n=5 cells); 3 8 . 9 _ + 5 . 3 p A (n=11 cells) and 18.5 -+ 3.4 pA (n = 8 cells) were measured at 40 and 10 mmol/1 [Na+]p respectively. I f we assume that the p u m p is maximally activated at 100 mmol/1 [Na+]p, the value for half-maximal activation would be around 60 mmol/l. This rather high value could be partially due to a large discrepancy between the [Na+]p and that at the p u m p sites. Pump activation could locally deplete intracellular Na + if the diffusion of Na + from the pipette were much slower than the pumping rate. No significant changes in [Ca2+]i could be observed in any of these experiments. These results are at variance with the findings that showed an increase in [Ca2+]i by reducing the extracellular Na + concentration ([Na+]o) in Na+-overloaded endothelial cells that were explained by an action on the endothelial Na+/Ca 2+ exchanger ([27], B. Nilius, G. Droogmans, unpublished observations). I f

306 this were the case then we should also expect an increase of [Ca2+]i by raising [Na+]i to 100 mmol/1 through the pipette. It is not clear why the endothelial cells respond differently to a decrease in the electrochemical gradient of Na + ions by manipulating either [Na+]o or [Na+]i.

Perforated-patch versus whole-cell configuration To ensure that the endothelial cells had a constant supply of intracellular ATP we initially performed our experiments in the normal whole-cell configuration [13]. However, in these experiments we recorded much smaller pump currents than with the perforated-patch-clamp method. We subsequently studied this effect systematically (Fig. 5). Cells were loaded with 40 mmol/l [Na+]i either by rupturing the cell membrane by suction or by using the pore-forming agent nystatin. In nine cells studied with the ruptured-membrane technique we recorded an mean pump current of 9.9 -- 1.1 pA at a holding potential of 0 mV. In nystatin-perforated patches we recorded a pump current of 38.9 _+ 5.3 pA (n -- 11 cells) under the same conditions. This latter value is significantly different from that obtained in whole-cell mode. The onset and relaxation of the pump current were similar for both techniques. If we added 100 ~tmol/1 guanosine 5'-0-(3thiotriphosphate (GTPTS) to the pipette solution, we obtained pump currents with an amplitude similar to that in the perforated-patch-clamp mode. The mean maximum pump current under these conditions was 39.8 _+ 10.2 pA (n = 5 cells). This value is not significantly different from that obtained with the perforated-patch technique. This finding suggests that pump activation in endothelial cells may be regulated by a guanine nucleotide-binding protein (G-protein)-controlled mechanism, some component of which is washed out in ruptured patches.

Discussion

The findings of the present study demonstrate that endothelial cells contain an electrogenic Na+/K + ATPase. The existence of [3H]-ouabain binding sites has been demonstrated [8] and the Na+/K + ATPase has been localized ultracytochemically [20] in endothelial cells. A DHO-sensitive, K+-activated contribution to the endothelial resting potential has also been reported [6], which could be explained by activation of an electrogenic pump. Electrogenic Na+/K + pump currents, which are a more direct parameter of pump activity, have not been measured in these cells so far. We present here the first recordings of such currents in human endothelial ceils. These currents could be blocked with DHO and their amplitude was modulated by [Na+]i and various external cations. With the pump fully activated rather large currents could be recorded in these small endothelial cells at room temperature. Taking into account the high temperature sensitivity of the pump this may point to a very high Na+/K + ATPase activity in these ceils.

Endothelial cells possess non-selective cation channels through which considerable Na + influx occurs during agonist stimulation [4, 22, 23, 24]. The amplitude of these currents may reach 500 pA at - 5 0 mV. An efficient Na + extrusion mechanism is therefore necessary for the restoration of [Na+]i after agonist stimulation. However, the Na+/K + ATPase in non-excitable cells apparently has a larger functional impact than simply the maintenance of the electrochemical gradients for Na + and K + ions. It has been proposed that the endothelial Na+/K + ATPase is important for the control of angiotensin converting enzyme activity [5]. Inhibition of the Na+/K + ATPase in endothelial cells also drastically modified the pattern of [Ca2+]i during agonist stimulation [17]. It might also be involved in the organization of a vectorial transport: fine tuning of the electrochemical gradient of Ca 2+ ions through an electrogenic Na+/K + ATPase could modulate Ca ~+ influx in endothelial cells and thereby contribute to Ca 2+ signalling. So far, little is known about the characteristics of this transport mechanism in endothelia. The pump currents described in this study show some similarities to those observed in other tissues. Firstly, they are blocked concentration dependently by DHO. Second, the pump activity depends on [Na+]i with an apparent concentration for half-maximal activation around 60 retool/1. A simlar surprisingly high value has been reported in cardiac cells (around 70 mmol/1 [191), and could be due to a large discrepancy between [Na+]v and [Na +] at the pump sites. Third, the relative potency of external cations for activating pump currents was T1+ > K + > NH4+. This same order of potency was also observed for the activation of the isolated, Mg2+-dependent Na+/K + ATPase [28] and for pump currents in cardiac tissue [2]. Also the Ko.5 values for activation by these three external cations were similar to those reported previously (see e. g. [2]). Finally, the pump current shows a voltage dependence similar to that in other tissues [9, 25, 32]. On the other hand, some of our findings are somewhat at variance with those in other cells. Firstly, the inhibitory action of DHO was not completely reversible. Second, the Kd value of DHO was much higher than in other tissues, e. g. in oocytes for which a K~ value of 0.4 btmol/1 has been reported [30], but it is similar to that observed in cardiac tissue [3]. These divergent values may be attributed to a different subunit isoform [14], similar to that in cardiac tissue. Third, the current relaxation after K + washout is slower than in other preparations [2] and finally, the pump current measured after mechanical rupturing of the cell membrane was much smaller than that recorded using perforated patches. This run-down in the ruptured-whole-cell mode is presumably due to washout of some cytoplasmic components, and was completely prevented by adding 100 gmol/1 GTPTS to the patch pipette. Although possible contributions of protein kinase C and cyclic adenosine monophosphate on Na+/K + ATPase modulation have been reported in other cells (rat kidney [1], rat liver [18, 31]), we have only little knowledge of the mechanisms that modulate pump activity in endothelial cells. If insulin had had an effect on pump currents in endothelial cells, it might have provided

307 s o m e hint r e g a r d i n g i n t r a c e l l u l a r m e c h a n i s m s o f the p u m p . H o w e v e r , the e x p e r i m e n t s with GTPTS s h o w that a G - p r o t e i n - d e p e n d e n t m e c h a n i s m c o u l d b e i n v o l v e d in the a c t i v a t i o n o f the p u m p in e n d o t h e l i a l cells. In conclusion, w e p r e s e n t for the first t i m e m e a s u r e m e n t s o f N a + / K + p u m p currents in e n d o t h e l i a l cells. This tool c o u l d p r o v i d e p o s s i b i l i t i e s for further study o f this e l e c t r o g e n i c t r a n s p o r t in the to date e l e c t r i c a l l y less w e l l - c h a r a c t e r i z e d e n d o t h e l i a l cell.

References 1. Bertorello A, Aperio A (1989) Na+,K+-ATPase is an effector protein for protein kinase C in renal proximal tubule cells. Am J Physiol 256: F370-F373 2. Bielen FV, Glitsch HG, Verdonck F (1991) Dependence of Na + pump on external monovalent cations and membrane potential in rabbit cardiac Purldnje cells. J Physiol (Lond) 442:169-189 3. Bielen FV, Glitsch HG, Verdonck F (1992) The kinetics of the inhibition by dihydroouabaln of the sodium pump current in single rabbit Purkinje cells. Naunyn-Schmiedeberg's Arch Pharmacol 345 : 100-107 4. Bregestovski R Bakhramov A, Danilov S, Modobaeva A, Takeda K (1988) Histamine-induced inward currents in cultured endothelial cells from human umbilical vein. Br J Pharmacol 95 : 429-436 5. Dasarathy Y, Fanburg BL (1989) Elevation of bovine endothefial cell angiotensin converting enzyme by cationophores and inhibition by ouabain, Biochim Biophys Acta 1051 : 1 4 20 6. Daut J, Mehrke G, Nees S, Newman WH (1988) Passive electrical properties and electrogenic sodium transport of cultured guinea-pig coronay endothelial cells. J Physiol (Lond) 402 : 237-254 7. De Weer R Gadsby DC, Rakowski RF (1988) Voltage dependence of the Na-K pump. Annu Rev Physiol 50:225-241 8. Freissmuth M, Nees S, Btck M, Schtitz W (1987) Binding of [3H]ouabain to endothelial cells derived from various vascular beds. Basic Cardiol 82:544-550 9. Gadsby DC, Kimura J, Noma A (1985) Voltage dependence of Na/K pump current in isolated heart cells. Nature 315 : 6 3 65 10. Gericke M, Droogmans G, Nilius B (1993) Thimerosal induced changes of intracelhilar calcium in human endothelial cells. Cell Calcium 14:201-207 11. Gericke M, Droogmans G, Nilius B (1993) Thapsigargin discharges intracellular calcium stores and induces transmembrane currents in human endothelial cells. Pfltigers Arch 422 : 552-557 12. Grynkiewicz G, Poenie M, Tsien RY (1985) A new generation of Ca 2+ indicators with greatly improved fluorescence properties. J Biol Chem 260:3440-3450 13. Hamill OP, Marty A, Neher E, Sakmann B, Sigworth FJ (1981) Improved patch-clamp technique for high-resolution current recording from cells and cell-free membrane patches. Pfltigers Arch 391:85-100 14. Horisberger JD, Lemos V, Kraehenbtthl JR Rossier BC (1991) Structure-function relationship of Na +,K+-ATPase. Annu Rev Physiol 53 : 565-584 15. Jaffe EA, Nachman RL, Becker CG, Minick CR (1973) Culture of human endothelial cells derived from umbilical veins. J Clin Invest 52 : 2745-2756

16. Korn SJ, Horn R (1989) Influence of sodium-calcium exchange on calcium current rundown and the duration of calcium-dependent chloride currents in pituitary cells, studied with whole cell and perforated patch recording. J Gen Physiol 94 : 7 8 9 - 812 17. Laskey RE, Adams DJ, Johns A, Rubanyi GM, Breemen C van (1990) Membrane potential and Na+-K + pump activity modulate resting and bradykinin-stimulated changes in cytosolic free calcium in cultured endothelial cells from bovine atria. J Biol Chem 263 : 2613-2619 18. Lynch CJ, Wilson PB, Blackmore PF, Exton JH (1986) The hormone-sensitive hepatic Na+-pump. Evidence for regulation by diacylglycerol and tumor promoters. J Biol Chem 261 : 14 5 5 1 - 1 4 556 19. Mogul DJ, Singer DH, Eick RE ten (1990) Dependence of Na-K pump current on internal Na § in mammalian myocytes. Am J Physiol 259 :H488-H496 20. Nag S (1990) Ultracytochemical localization of Na+,K+-ATPase in cerebral endothefium in acute hypertension. Acta Neuropathol 80 : 7 - 1 1 21. Neher E (1989) Combined Fura-2 and patch clamp measurements in rat peritoneal mast cells. In: Sellin LC, Libelius R, Thesleff S (eds) Neuromuscular junction. Elsevier, Amsterdam, pp 6 5 - 7 6 22. Nilius B (1990) Permeation properties of a non-selective cation channel in human vascular endothelial cells. Pfltigers Arch 416: 609-611 23. Nilius B, Riemann D (1990) Ion channels in human endothelial cells. Gen Physiol Biophys 9 : 8 9 - 1 1 2 24. Nifius B, Schwarz G, Oike M, Droogmans G (1993) Histamine activated non-selective cation currents and Ca 2+ transients in endothelial cells from human umbilical vein. Pfltigers Arch (in press) 25. Omay HS, Schwarz W (1992) Voltage-dependent stimulation of Na+/K+-pump current by external cations: selectivity of different K + congeners. Biochim Biophys Acta 1104:167173 26. Rae JL, Fernandez J (1991) Perforated patch recordings in physiology. News Physiol Sci 6 : 2 7 3 - 2 7 7 27. Sage SO, Breemen C van, Cannel1 MB (1991) Sodium-calcium exchange in cultured bovine pulmonary artery endothelial cells. J Physiol (Lond) 440:569-580 28. Schuurmans Stekhoven F, Bonting SL (1981) Transport adenosine triphosphatases: properties and functions. Physiol Rev 61 : 1 - 7 6 29. Schwarz G, Droogmans G, Callewaert G, Nllius B (1992) Shear stress-induced calcium transients in human endothelial cells from umbilical cord veins. J Physiol (Lond) 458 : 5 2 7 538 30. Schweigert B, Lafaire AV, Schwarz W (1988) Voltage dependence of the Na-K ATPase: measurements of ouabain-dependent membrane currents and ouabain binding in oocytes of Xenopus laevis. Pfltigers Arch 412:579-588 31. Vasilets LA, Schwarz W (1992) Regulation of endogenous and expressed Na+/K + pumps in Xenopus oocytes by membrane potential and stimulation of protein kinases. J Membr Biol 125 : 119-132 32. Wu MM, Mortimer MM (1991) Voltage dependence of currents through the Na, K-exchange pump of Rana oocytes. J Membr Biol 121 : 2 3 - 3 6 33. Yingst DR (1988) Modulation of the Na-K-ATPase by Ca and intracellular proteins. Annu Rev Physiol 50 : 291-303