Pfliigers Arch (1992) 421:394--396

Journal of Physiology 9 Springer-Verlag 1992

Short communication Shear stress induced membrane currents and calcium transients in human vascular endothelial cells Gero Sehwarz 1, Guy Droogmans 2, and Bernd Nilius 1 * Max Planck Group "Molecular and Cellular Physiology", 0-6900 Jena, Federal Republic of Germany 2 Catholic University Leuven, Department of Physiology, B-3000 Leuven, Belgium Received May 18, 1992/Accepted June 11, 1992

Abstract. We have measured membrane currents induced by shear stress together with intracellular calcium signals in endothelial cells from human umbilical cord veins. In the presence o f extracellular calcium (Ca2+]o), shear stress induced an inward current at a holding potential of 0 mV which is accompanied by a rise in intracellular Ca 2+ ([Ca2+]i). In the absence of extracellular calcium shear stress was unable to evoke a calcium signal but still induced a membrane current. The voltage dependence of the shear stress induced current was obtained from difference currents evoked by linear voltage ramps before and duri.~g application of shear stress. Its reversa! potential Erev shifted from -2.3 * 0.8 mV (n=4) in a nominally Ca 2+ free solution to +1.5 -+ 1.6 mV at 1.5 mM [Ca2+]o (n=4) and to +21.9 _+ 4.4 mV (n=7) at 10 mM [Ca2+]o. From our data we conclude that shear stress opens an ion channel that is 12.5 -+ 2.9 (n=7) times more permeable for calcium than for sodium or cesium. Key words: endothelium - shear stress - Ca2+-permeable ion channels - intracellular calcium

Introduction. Shear stress mediates several biological effects in human endothelial cells, such as EDRF release, prostacydin synthesis, expression of tissue plasminogen activator (for a review see [1]. In a recent paper, we have described shear stress mediated Ca 2+ transients in vascular endothelial cells from human umbilical veins [2]. In these cells shear stress induced quenching of Fura II fluorescence by divalents, such as cobalt and nickel, indicating the activation of a pathway preferentially permeable for divalents. In this report we describe a membrane current that is activated by shear stress and carries calcium more than 10 times better than monovalent cations. This current could mediate intracellular Ca 2+ transients stimulated by shear stress and could therefore couple mechanical events to biological effects in endothelial cells.

Offprint requests to B. Nilius, Laboratorium voor Fysiologie, Campus Gasthuisberg, B-3000 Leuven (Belgium).

Methods. Endothelial cells were prepared from human umbilical cord veins by a collagenase digestion procedure as described previously [3]. Under our culture conditions, cells were not confluent between day 1 and 4 and single endothelial cells could be used for our experiments. We only used cells from primary culture up to the second passage. For [Ca2+]i measurements, cells were incubated with 2 laM of the acetoxymethylester Fura-2/AM (Molecular Probes, Eugene, OR, USA ). The bath perfusion solution (Krebs' solution) used in all experiments had the following composition '(in retool/l): 140 NaC1, 1.5 or 10 CaCI 2, 5.9 KC1, 1.2 MgC12, 11.5 Hepes-NaOH, 10 glucose, titrated to pH 7.3 with NaOH. The Ca 2+ measurement was coupled with simultaneous current measurements by the patch clamp technique in the whole cell mode. The technique used was described in detail elsewhere [2,4,5]. The following pipette solution was used (in mmol/l): 100 Cs aspartate, 40 mM CsCI, 5 NaCI, 5.5 MgC12, 10 Hepes, 5 Na2ATP, 0.1 EGTA buffered at pH 7.2 with CsOH. Under resting conditions the cells were superfused by a slow bath perfusion. Shear stress was induced by directing a stream of Krebs' solution along the surface of a single endothelial cell through a multibarrelled pipette with a common opening of about 150 lam. The flow system and the generation of shear stress are described in detail elsewhere [2]. Mean +_standard error of the mean were calculated from pooled data. For tests of significance, we used the unpaired t-test. Permeation ratios were calculated from the measured reversal potentials by using a method described in detail elsewhere [6]. Results and discussion. Figure 1A shows the first simultaneous measurement of shear stress induced membrane currents and intracellular Ca 2+ signals in a single endothelial cell. At a holding potential of 0 mV, shear stress (approximately 10 dyne/cm 2 as calculated from "r =4pQ/z~r3 where p represents the fluid viscosity, Q the rate of flow, r the radius of the cylinder of fluid that streams out of the pipette) evoked an inward current together with a slowly

395 SHEAR

A

$I-IgAR

i

A

1

7

200 p A ~

VH =

?,- .:r..:

12,

0 mV

V H = 0 mV

I

i

0.2 #M Cai -1

J

i

i 60s

i

I

300 s

I L~] 500 v [~v]

-80

-80

-40

-20

9

500 11 [pA]

c -1000

aoo[1 I~] -1500

5

v [mvl -80

-flO

-4O

-2,O

,

C 1 [pAl v [mv] '

-60 i

'

-80 i

Figure 2. In Ca-~ee solution (nominal Ca2+ flee in the presence

-80r

Figure 1. Shear stress induces Ca2+ transients and inward cur-

rents in human vascular endothelial cells at 10 m M extracellular calcium. A: Simultaneous measurement of intracellular calcium and membrane current. The bar indicates the application of shear stress on the surface of a single endothelial cell (approximately I0 dyne/cm2). A holding potential of O m V was applied throughout the experiment (slow sampling rate of 3 Hz). The peaks in the current trace are due to application of voltage ramps, and are represented in B at an expanded time scale. B: i-v curves obtained by application of linear voltage ramps 3%m -100 to 50 m V (500 ms, sampling rate 500 Hz). C: i-v curves of the shear stress induced current were obtained by subtracting an averaged ramp current before application of shear stressjgom the currents 2 to 6 during shear stress. The reversal potential obtained is close to 30 inV.

of O.1 m M EGTA) shear stress does not induce Ca2+ transients but still activates membrane currents. A: The same protocol as in figure 1 was applied but now in the absence of calcium. During shear stress a small decrease of intracellular calcium could be observed. B: Ramp currents 1, 2, 6, and 7 are plotted against membrane potential. C: Shear stress induced currents were obtained jgom difference currents ( here 2-1, 6-1, 7-1 ). The currents reversed near 0 mV.

developing Ca 2§ transient. Shear stress was applied in the presence of 10 m M extracellular calcium. Figure 1A shows records at a slow sampling rate (3Hz). During the experiment, 500 ms linear voltage ramps from -100 to +50 mV were applied before and during application of shear stress. The currents induced by these voltage ramps were also sampled at a much higher sampling rate (500 Hz), and converted to i-v curves, as shown in Fig. lB. The numbers along the traces refer to the corresponding traces in Fig. 1A. Control i-v curves, obtained from traces before application of shear stress, were subtracted from the i-v curves measured during shear stress. The i-v curves of the shear

396 stress activated current are represented in Fig. 1C, from which it is obvious that this current develops slowly during application of shear stress. Its reversal potential was near to +30 mV. From 7 cells a reversal potential of +21.9 • 4.4 mV was calculated at 10 mM [Ca2+]o. These experiments were repeated at different extracellular Ca 2+ concentrations. At 1.5 mM [Ca2+]o we measured a reversal potential of the shear stress induced currents of + 1.5 + 1.6 mV ( n=4 ). Figure 2A shows an experiment in which a Ca 2+ free solution is blown onto the cell to induce shear stress. Under these conditions no Ca 2+ transient could be evoked. The intracellular calcium concentration even decreased during the superfusion with Ca 2+ free solution. This could be explained by a decreased leak Ca 2+ flux in the presence of an active Ca 2+ extrusion mechanism, or by a reversed driving force for calcium. Under Ca 2+ free conditions, no inward current could be recorded at a holding potential of 0 inV. However, from the applied voltage ramps, as shown in figure 2B, it is obvious that shear stress still activates a membrane current. This current reverses near 0 mV, as can be seen from the difference currents, represented in Fig. 2C. From four cells we measured a reversal potential of-2.3 + 0.8 inV. To estimate the contribution of calcium influx to the shear stress induced current, we have calculated the respective permeation ratios from the reversal potentials measured under the different experimental conditions. The permeation ratios were calculated at each of the measured reversal potentials as the roots of the implicit equations RT PNa'[Na]o E r e v - " - In -0 F PCs'[Cs]i + PNa' [Na]i

(1)

in Ca 2§ free solutions, and from RT PN~.[Na]o + P'.[Ca]o Erev-"*ln = 0 (2) F PC~'[Cs]i + PNa "[Na]i + P"' [Ca]i

in Ca 2§ containing solutions, where p,

PCa 1+ e x p ( ~ )

and P " - P c a ' e x p ( ~ ) l+exp(~)

For the latter equation we used the permeation ratio PNa/PCs obtained from equation (1). R, T and F have their usual meaning. PCs, PNa, PCa represent the permeability coefficients for Cs+, Na + and Ca 2+. Erev is the measured reversal potential under the different conditions. From the experiments in the absence of extracellular calcium, we obtained from 4 cells: PNa/PCs = 0.94 + 0.03 indicating that the channel is not discriminating between these two cations. Using this ratio for the permeation of the monovalent cations in (2), we obtained the

permeation ratio PCa/PCs = 8.3 + 3.4 (n = 4 cells) in 1.5 mM [Ca2+]o . From 7 cells in 10 mM [Ca2+]o we calculated a PCa/PCs ratio of 12.5 + 2.9. The permeation ratios between 1.5 and 10 mM [Ca2+]o were not significantly different. These data indicate that shear stress activates a membrane channel that is permeable for calcium. Calcium permeates the channel approximately 10 times better than monovalent cations. At 10 mM extracellular calcium, which is the most reliable situation, shear stress activates an inward current of approximately 100 pA at a holding potential of 0 mV. This current would be composed of a sodium inward current, a potassium (or here Cs § outward current, and a calcium inward current. Taking into account the calculated permeation ratios, the contribution of Ca 2+ to this current would be approximately 80 pA, which could explain the increase in intracellular calcium induced by shear stress. We conclude that shear stress activates ion channels in endothelial cells by a still unknown mechanism, thereby inducing Ca 2§ influx and an increase in intracellular calcium. From the calculated permeation properties the recorded current shows some similarities with the single channel currents measured in endothelial cells from porcine aorta [7] but not with those from bovine pulmonary artery cells [8], References. [1] Nollert M.U., Diamond S.L. and Mclntire L.V. (1991) Hydrodynamic shear stress and mass transport modulation of endothelial cell metabolism. Biotechnology and Bioengineering 38:588-602 [2] Schwarz G, Droogmans G, Callewaert G, Nilius B (1992) Shear stress induced calcium transients in human endothelial cells from umbilical cord veins. Journal of Physiology ( London ) in the press [3] Jaffe E.A., Nachman R.L., Becker C.G. and Minick C.R. (1973). Culture of human endothelial cells derived from umbilical veins. Journal of Clinical lnvestigation 52:2745-2756 [4] Neher E. (1989). Combined Fura-2 and patch clamp measurements in rat peritoneal mast cells. In: Neuromuscular Junction, eds. LC Sellin, R Libelius, S Thesleff, Elsevier Science Publishers, Amsterdam, p. 65-76 [5] Grynkiewicz G., Poenie M. and Tsien R.Y. (1985). A new generation of Ca 2+ indicators with greatly improved fluorescence properties. Journal of Biological Chemistry 260: 34403450 [6] Nilius B. (1990). Permeation properties of a non-selective cation channel in human vascular endothelial cells. Pfliigers Archiv EuropeanJournal of Physiology416:609-611 [7] Lansman J.B., Hallam T.J. and Rink T.J. (1987). Single stretch-activated ion channels in vascular endothelial cells as mechanotransducers? Nature 235:811-813 [8] Oleson S.-O., Clapham D.E. and Davies P.F. (1988). Haemodynamic shear stress activates a K + current in vascular endothelial cells. Nature 331: 168-170