Journal Logo


EDHF-Mediated Relaxation in Rat Gastric Small Arteries: Influence of Ouabain/Ba2+ and Relation to Potassium Ions

Van de Voorde, Johan; Vanheel, Bert

Author Information
Journal of Cardiovascular Pharmacology: April 2000 - Volume 35 - Issue 4 - p 543-548
  • Free


The endothelium plays a key role in modulating the responses of the underlying vascular smooth muscle cells by releasing endothelium-derived relaxing factors (EDRFs) under basal conditions and in response to agents such as acetylcholine. Nitric oxide (NO) and in some blood vessels also prostacyclin are well-characterized EDRFs. However, endothelium-dependent vasorelaxation is still observed in vessels in which NO and prostanoid synthesis is blocked, and this residual relaxation is attributed to an endothelium-derived hyperpolarizing factor (EDHF) (1,2). The chemical identity of this EDHF is as yet not established.

Controversial evidence has been provided for the suggestions that EDHF might be anandamide (3-5) or an epoxyeicosatrienoic acid formed through a cytochrome P450-dependent mechanism (6-10). Very recently strong evidence was reported suggesting that the EDHF could be identified as the K+ ion, at least in rat hepatic arteries (11). From the results, it was postulated that the increase in cytosolic calcium seen after stimulation of endothelial cells with acetylcholine might activate endothelial KCa channels, leading to efflux of K+ into the myoendothelial space. This increase in K+ then would hyperpolarize and relax the adjacent smooth muscle cells by activating the Na+/K+ pump and the inwardly rectifying K+ channels.

In this study we investigated whether EDHF-mediated relaxation in rat gastric small arteries is blocked after inhibition of the Na+/K+-pump and the inward rectifying K+ channel with respectively ouabain and Ba2+ and tried to link our observations with the hypothesis that K+ is the EDHF.


Tension measurements

Experiments were performed on small gastric arteries (normalized diameter ranging from 180 to 504 μm) from young female Wistar rats (200-260 g). The animals were killed by cervical dislocation. The arteries were dissected free and mounted in an automated dual small-vessel myograph (model 500 A; J.P. Trading, Aarhus, Denmark). After isolation, the preparations were transferred to a tissue chamber filled with 10 ml of Krebs-Ringer bicarbonate solution. Two stainless steel wires (40 μm diameter) were guided through the lumen of segments, having a length of 1.5-2 mm. One wire was fixed to a force-displacement transducer, and the other was connected to a micrometer. After mounting, the preparations were allowed to equilibrate for 30 min in the Krebs-Ringer bicarbonate solution bubbled with 95% O2 and 5% CO2 at 37°C. The arteries were then normalized as described by Mulvany and Halpern (12) to obtain optimal conditions for active force development. In brief, the passive wall tension-internal circumference characteristics were determined. On the basis of this relation, the circumference was set to a normalized internal circumference (from which normalized diameter can be calculated) corresponding to 90% of the internal circumference that the vessels would have under a passive transmural pressure of 100 mm Hg. After normalization, the preparations were allowed to equilibrate again for ≥30 min. Thereafter the preparations were 3 times challenged with a mixture of K+ 120 mM and norepinephrine 10 μM. When reproducible contractions were obtained, cumulative concentration-relaxation curves to agonists were made after precontraction of the preparations with norepinephrine, 10 μM (eliciting 98% of maximal contraction). All experiments were performed in the presence of the NO-synthase inhibitor nitro-L-arginine (L-NA, 0.1 mM) and the cyclooxygenase inhibitor indomethacin (50 μM) to exclude the potential involvement of NO and prostanoids in the observations.

Membrane potential measurements

Transmembrane potentials were measured as described previously (10). In brief, conventional microelectrodes were pulled with a vertical pipette puller (model 750; David Kopf, Tujunga, CA, U.S.A.) from 1 mm O.D. filamented glass tubings (Hilgenberg, Germany). Microelectrodes were filled with 1 M KCl. Their electrical resistance, measured in the normal Krebs-Ringer solution, ranged from 40 to 80 MΩ. The measured potential was followed on an oscilloscope and traced with a pen recorder at low speed. Absolute values of membrane potential were taken as the difference of the stabilized potential after cell impalement and the zero potential on withdrawal of the microelectrode from the cell. Changes in membrane potential produced by applications of acetylcholine or K+ were measured in the same cell during continuous recordings.

Drugs and statistics

The experiments were performed using a Krebs-Ringer bicarbonate solution of the following composition (in mM): NaCl, 135; KCl, 5; NaHCO3, 20; glucose, 10; CaCl2, 2.5; MgSO4·7H2O, 1.3; KH2PO4, 1.2; and EDTA, 0.026. High-K+ (120 mM) Krebs-Ringer bicarbonate solution was prepared by equimolar replacement of NaCl with KCl.

Norepinephrine bitartrate, indomethacin, L-NA, acetylcholine chloride, and ouabain were all obtained from Sigma (St. Louis, MO, U.S.A.). Sodium nitroprusside and barium chloride were obtained from Merck (Darmstadt, Germany), NS 1619 from RBI (Natick, MA, U.S.A.), and 1-EBIO (1-ethyl-2-benzimidazolinone) was obtained from Tocris (Bristol, U.K.).

All concentrations are expressed as final molar concentrations in the organ bath. Concentration-response curves were made by cumulative additions of small volumes (maximal, 1%) into the experimental chamber. All solutions were freshly prepared by dilution from appropriate stock solutions. Stock solutions were made in water except for 1-EBIO and NS 1619 (dissolved in dimethylsulfoxide), acetylcholine (dissolved in phthalate buffer, pH 4.0), and indomethacin (dissolved in ethanol).

Statistical significance was evaluated using Student's t test for paired observations; n indicates the number of preparations tested.


Influence of Ba2+ and ouabain

In these experiments we investigated the influence of the presence of Ba2+ and/or ouabain on the L-NA/indomethacin-resistant relaxation to acetylcholine on precontracted rat gastric arteries. The simultaneous addition of Ba2+ and ouabain elicited in most preparations (16 of 26) no effect, and in the other preparations, a mostly transient contractile effect (mean = 0.92 ± 0.37 mN; n = 26). The contraction elicted with norepinephrine (10 μM) was not substantially influenced (12.98 ± 0.85 mN in the absence vs. 12.49 ± 0.84 mN in the presence of Ba2+ and ouabain; n = 26). In the combined presence of Ba2+ (30 μM) and ouabain (0.5 mM), both added 15 min before precontraction of the preparations, EDHF-mediated relaxations are strongly impaired. In the same experimental conditions, the combined presence of Ba2+ and ouabain had no influence on relaxations induced by sodium nitroprusside (Fig. 1). In the presence of only Ba2+ (30 μM), the L-NA/indomethacin-resistant relaxation to acetylcholine is somewhat shifted to the right, whereas the response is more severely impaired in the presence of only ouabain (0.5 mM) (Fig. 1). Addition of Ba2+ as such had no influence on basal tone and also not on the contraction elicited by norepinephrine (15.45 ± 2.1 mN in the absence and 15.46 ± 1.9 in the presence of Ba2+; n = 4).

FIG. 1
FIG. 1:
Relaxation effects (in percentage of active tension with norepinephrine, 10 μM) of increasing molar concentrations of acetylcholine and sodium nitroprusside in control conditions (C) and in the combined presence of Ba2+ (30 μM) and ouabain (0.5 mM) or in the presence of Ba2+ (30 μM) or ouabain (0.5 mM) separately. All experiments performed in the presence of indomethacin and L-NA. (n = 4-9; *p < 0.05; ****p < 0.001).

In another series of experiments, the influence of different concentrations (500, 50, and 5 μM) of ouabain was investigated on the L-NA/indomethacin-resistant relaxation to acetylcholine. In these experiments, ouabain was added after precontraction of the preparations with 10 μM norepinephrine and 10 min before starting the additions of acetylcholine. In the experiments with ouabain, 500 μM, an additional concentration-response curve with acetylcholine was established 30 min after washout of ouabain. Addition of ouabain to the precontracted preparations, in all concentrations used, elicited on itself a transient contraction followed by a more slowly developing relaxation. An original recording of such an experiment is shown in Fig. 2. The influences of the different concentrations of ouabain on the concentration-response curve of acetylcholine are shown in Fig. 3. They illustrate that the responses to acetylcholine are still inhibited in the presence of 50 μM ouabain, but no more in the presence of 5 μM ouabain. The inhibitory influence of ouabain, 0.5 mM, is a reversible phenomenon, because the response to acetylcholine is significantly restored after washout of ouabain.

FIG. 2
FIG. 2:
Original tracing of an experiment showing the influence of the presence of ouabain (0.5 mM) on acetylcholine-induced relaxations (concentrations expressed in log molar) in a gastric artery. Ouabain is added after precontraction of the preparation with norepinephrine (10 μM) and elicits a transient contraction. Concentration-responses are assessed in the presence of indomethacin (50 μM) and L-NA (0.1 mM).
FIG. 3
FIG. 3:
Relaxation effects (in percentage of active tension with norepinephrine, 10 μM) of increasing molar concentrations of acetylcholine in control conditions (C) and in the presence of ouabain at a concentration of 0.5 mM, 50 μM, or 5 μM. The triangles represent the effect of acetylcholine after washing out ouabain, 0.5 mM, for 30 min. All experiments performed in the presence of indomethacin and L-NA. (n = 4-5; **p < 0.02 vs. control; ***p < 0.01 vs. control; †††p < 0.01 vs. ouabain, 0.5 mM).

Effect of K+

On many preparations, the influence of increasing extracellular concentrations of K+ was tested by cumulative additions of KCl, 1-15 mM. This was done only on preparations showing previously substantial L-NA/indomethacin-resistant relaxations in response to acetylcholine. Two original recordings of such experiments are shown in Fig. 4. In some of the preparations (14 of 29), very transient relaxations (>10% of tone) were observed with the lower concentrations of K+ (upper tracing). Other preparations (15 of 29), however, showed no or only a small relaxation (<10% of tone) or even contraction in response to K+ (bottom tracing) despite strong relaxing responses to additions of acetylcholine. The summarized data of these experiments are presented in Fig. 5.

FIG. 4
FIG. 4:
Original tracings of two experiments showing the effects of increasing millimolar concentrations of K+ on isolated small rat gastric arteries. Preparations are contracted with norepinephrine (10 μM) in the presence of indomethacin (50 μM) and L-NA (0.1 mM).
FIG. 5
FIG. 5:
Relaxation effects (in percentage of active tension with norepinephrine, 10 μM) of increasing molar concentrations of acetylcholine and of increasing millimolar concentrations of K+ measured on the same preparations (n = 14-29). All experiments performed in the presence of indomethacin and L-NA.

Experiments with NS 1619 and 1-EBIO

In another series of experiments, the influence of removal of the endothelium was investigated on relaxations elicited by addition of the KCa-channel openers NS 1619 and 1-EBIO. After removal of the endothelium, resulting in a complete loss of the relaxing influence of acetylcholine, the relaxation response to NS 1619 (n = 4) and 1-EBIO (n = 6) was not blocked (Fig. 6A).

FIG. 6
FIG. 6:
Relaxation effects (in percentage of active tension with norepinephrine, 10 μM) of increasing molar concentrations of NS 1619 and 1-EBIO (30 μM) in preparations with and without endothelium (A) and in preparations in the absence and presence of Ba2+ (30 μM) and ouabain (0.5 mM) (B). All experiments performed in the presence of indomethacin and L-NA (n = 4-5).

In other series of experiments, the influence of ouabain and Ba2+ was investigated on relaxations elicited by addition of the KCa-channel openers NS 1619 and 1-EBIO. Whereas ouabain/Ba2+ strongly impaired the L-NA/indomethacin-resistant response to acetylcholine in these preparations (not shown), the responses to NS 1619 and 1-EBIO were not diminished (Fig. 6B).

Membrane potential measurements

Smooth muscle cells of small gastric arteries had a mean resting membrane potential of −63.1 ± 2.7 mV (n = 4). In all experiments, cumulative increases of the extracellular concentration of K+ by 2, 4, and 8 mM depolarized the cells by 1.8 ± 0.8, 4.8 ± 1.4, and 9.4 ± 2.1 mV, respectively, whereas the application of acetylcholine (0.3 μM) to the preparations hyperpolarized the same cells by 5.9 ± 1.8 mV. A representative recording is shown in Fig. 7.

FIG. 7
FIG. 7:
Continuous recording of the membrane potential (Em) of a smooth muscle cell in a gastric artery during stepwise increases of K+ concentrations by 2, 4, and 8 mM and during application of acetylcholine (0.5 μM).


The experiments described here were performed in an effort to find out whether the hypothesis that EDHF might be K+, as recently suggested from experiments on rat hepatic artery (11), can be extended to gastric small arteries of this species. That the hypothesis is unlikely to be universally valid is suggested by the recently published observations of Guignard et al. (13) on guinea pig carotid and porcine coronary arteries. Also our recent membrane potential measurements on rat mesenteric arteries do not support the hypothesis that EDHF and K+ are the same entity (14). In preliminary experiments, we could establish that L-NA/indomethacin-resistant relaxation response to acetylcholine in rat gastric small arteries is likely mediated by EDHF because the response completely disappears after removal of the endothelium and in the presence of high concentrations of K+.

Our experiments with ouabain and Ba2+ confirm the observations of Edwards et al. (11). The inhibitory influence of these substances relies mainly on blockade of the Na+/K+-pump because the inhibitory influence of ouabain is much more pronounced than that of Ba2+. It should be noted that the sensitivity of EDHF-mediated responses to ouabain has already been demonstrated in other studies (15,16). The inhibitory influence of ouabain is reversible and specific because the non-EDHF-mediated relaxation elicited by sodium nitroprusside is not influenced in the presence of Ba2+ and ouabain.

In the gastric artery, the high concentration of ouabain (1 mM) as used by Edwards et al. (11) on hepatic artery is not required for inhibition of EDHF-mediated relaxations. Lower concentrations are also active. An influence is no more observed with ouabain in a concentration of 5 μM. Nevertheless, on addition of this lower concentration of ouabain, a transient contraction is still elicited on precontracted preparations, illustrating that this concentration is somehow active. It should be noted that there exist different isoenzymes with different sensitivities to ouabain. For the α1β1 Na,K-ATPase isoenzyme in rat, the inhibition constant (Ki) is in the order of 0.04-0.1 mM, whereas the other isoenzymes are more sensitive to ouabain (Ki in the order of 0.03-0.2 μM) (17). Our data thus suggest the potential involvement of the α1β1 Na,K-ATPase isoenzyme in the inhibitory influence of ouabain on EDHF-mediated relaxation.

Although our observations with Ba2+ and ouabain are in line with the suggestion of K+ being the EDHF, our observations with additions of K+ are not so convincing. Some of the preparations relaxed in a transient way in response to additions of small concentrations of K+. However, many other preparations showed no or only small relaxations, although these vessels showed a strong EDHF-mediated relaxation. If endogenous K+ were the EDHF, one would expect that addition of exogenous K+ would induce relaxation in all EDHF-sensitive preparations.

In an attempt to further evaluate whether EDHF could be identified as K+, experiments were performed using the KCa channel openers NS 1619 (18,19) and 1-EBIO (20,21). Considering the hypothesis that K+ is released as EDHF by endothelial cells through activation of KCa channels, it could be expected that activation of these channels might elicit a relaxation at least in part endothelium dependent. However, we found that preparations without endothelial cells (as evidenced by the lack of relaxation to acetylcholine) showed no diminished response to NS 1619 or 1-EBIO compared with preparations with endothelium. It should be noted that as yet no data are available that NS 1619 activates endothelial KCa channels, in contrast to 1-EBIO, which has been reported to act more specifically on endothelial KCa channels (20,21).

That the responses to NS 1619 and 1-EBIO are not diminished in endothelium-denuded preparations might be due to an overwhelming influence of the KCa channel openers on KCa channels on the smooth muscle. Opening of KCa channels, either in smooth muscle or endothelial cells is, according to the hypothesis of Edwards et al. (11), expected to increase the K+ concentration in the intercellular fluid surrounding the smooth muscle cells. This increase in intercellular K+ might activate the Na+/K+ pump and the inward rectifier. Therefore, one could speculate that the responses to NS 1619 and 1-EBIO would be at least in part inhibited by ouabain and Ba2+. However, our results clearly demonstrate no inhibitory influence. These data thus give no support to the hypothesis of K+ being the EDHF in the gastric artery. It should be noted, however, that other mechanisms than the opening of KCa channels might eventually contribute to the relaxant effect of NS 1619 (22).

That EDHF in rat small gastric arteries is unlikely to be K+ is further supported by membrane potential measurements. Smooth muscle cells that hyperpolarize to acetylcholine fail to do so in response to increasing concentrations of K+. In contrast, the addition of K+ consistently depolarized the same cells.

What might be the mechanism responsible for the inhibitory influence of ouabain/Ba2+ on EDHF-mediated relaxation responses besides activation of inward rectifier and the sodium pump by K+? It can be hypothesized that blockade of the Na+/K+ pump by ouabain eventually might lower the concentration gradient for K+ over the smooth muscle cell membrane and in that way diminish the driving force for K+ efflux from the smooth muscle in response to EDHF, which often has been associated with opening of K+-channels on the smooth muscle cells (1,2). Interference also might occur at the level of the endothelial cells. Inhibition of the Na+/K+ pump and the inward rectifier, which are also present on endothelial cells (23), may depolarize the endothelial cells and in that way diminish the driving force on Ca2+ influx, an essential step in EDHF synthesis.

It should also be mentioned that ouabain is able to block gap junctional communication (24). This feature of ouabain also must be considered in relation to EDHF-mediated effects because recent observations provide evidence that heterocellular gap junctional communication may be involved in EDHF-mediated effects (25,26). Blockade of these gap junctions by ouabain thus might also explain its inhibitory influence on EDHF-mediated relaxation.

In summary, it is concluded that the combination ouabain/Ba2+ has a pronounced inhibitory influence on acetylcholine-induced EDHF-mediated relaxation of rat gastric small arteries. The relation of this inhibitory influence to an action of K+ as EDHF is, however, uncertain.

Acknowledgment: This work was supported by the Fund for Scientific Research-Flanders. B.V. is a senior research associate of the FWO-Vlaanderen. Thanks to E. Tack for the expert technical assistance.


1. Cohen RA, Vanhoutte PM. Endothelium-dependent hyperpolarization: beyond nitric oxide and cyclic GMP. Circulation 1995;92:3337-49.
2. Garland CJ, Plane F, Kemp BK, Cocks TM. Endothelium-dependent hyperpolarization: a role in the control of vascular tone. Trends Pharmacol Sci 1995;16:23-30.
3. Chataigneau T, Félétou M, Thollon C, et al. Cannabinoid CB1 receptor and endothelium-dependent hyperpolarization in guinea-pig carotid, rat mesenteric and porcine coronary arteries. Br J Pharmacol 1998;123:968-4.
4. Randall MD, Alexander SPH, Bennett T, et al. An endogenous cannabinoid as an endothelium-derived vasorelaxant. Biochem Biophys Res Commun 1996;229:114-20.
5. Zygmunt PM, Högestätt ED, Waldeck K, Edwards G, Kirkup AJ, Weston AH. Studies on the effects of anandamide in rat hepatic artery. Br J Pharmacol 1997;122:1679-86.
6. Campbell WB, Gebremedhin D, Pratt PF, Harder DR. Identification of epoxyeicosatrienoic acids as endothelium-derived hyperpolarizing factors. Circ Res 1996;78:415-23.
7. Corriu C, Félétou M, Canet E, Vanhoutte PM. Inhibitors of the cytochrome P450-mono-oxygenase and endothelium-dependent hyperpolarization in the guinea-pig isolated coronary artery. Br J Pharmacol 1996;117:607-10.
8. Hecker M, Bara A, Bauersachs J, Busse R. Characterization of endothelium-dependent hyperpolarizing factor as a cytochrome P450-derived arachidonic acid metabolite in mammals. J Physiol (Lond) 1994;481:407-14.
9. Van de Voorde J, Vanheel B. Influence of cytochrome P-450 inhibitors on endothelium-dependent nitro-L-arginine-resistant relaxation and cromakalim-induced relaxation in rat mesenteric arteries. J Cardiovasc Pharmacol 1997;29:827-32.
10. Vanheel B, Van de Voorde J. Evidence against the involvement of cytochrome P450 metabolites in endothelium-dependent hyperpolarization in rat main mesenteric artery. J Physiol (Lond) 1997;501:331-41.
11. Edwards G, Dora KA, Gardener MJ, Garland CJ, Weston AH. K+ is an endothelium-derived hyperpolarizing factor in rat arteries. Nature 1998;396:269-72.
12. Mulvany MJ, Halpern W. Contractile properties of small arterial resistance vessels in spontaneously hypertensive and normotensive rats. Circ Res 1977;41:19-26.
13. Quignard J-F, Félétou M, Thollon C, Vilaine J-P, Duhault J, Vanhoutte PM. Potassium ions and endothelium-derived hyperpolarizing factor in guinea-pig carotid and porcine coronary arteries. Br J Pharmacol 1999;127:27-34.
14. Vanheel B, Van de Voorde J. Barium decreases endothelium-dependent smooth muscle responses to transient but not to more prolonged acetylcholine applications. Eur J Physiol 1999;439:123-9.
15. Félétou M, Vanhoutte PM. Endothelium-dependent hyperpolarization of canine coronary smooth muscle. Br J Pharmacol 1988;93:515-24.
16. Olanrewaju HA, Hargittai PT, Lieberman EM, Mustafa SJ. Effect of ouabain on adenosine receptor-mediated hyperpolarization in porcine coronary artery smooth muscle. Eur J Pharmacol 1997;322:185-90.
17. Blanco G, Mercer RW. Isoenzymes of the Na-K-ATPase: heterogeneity in structure, diversity in function. Am J Physiol 1998;275:F633-50.
18. Olesen SP, Munch E, Moldt P, Drejer J. Selective activation of Ca2+ dependent K+ channels by novel benzimidazolone. Eur J Pharmacol 1994;251:53-9.
19. Holland M, Langton PD, Standen NB, Boyle JP. Effects of the BKCa channel activator, NS 1619, on rat cerebral artery smooth muscle. Br J Pharmacol 1996;117:119-29.
20. Edwards G, Gardener MJ, Walker SD, Weston AH. Comparison of effects of 1-EBIO and NS1619 on K+ currents in vascular smooth muscle and endothelial cells. Br J Pharmacol 1998;125:106P.
21. Adeagbo ASO. 1-Ethyl-2-benzimidazolinone stimulates endothelial KCa channels and nitric oxide formation in rat mesenteric vessels. Eur J Pharmacol 1999;379:151-9.
22. Edwards G, Niederste-Hollenberg A, Schneider J, Noack T, Weston AH. Ion channel modulation by NS 1619, the putative BKCa channel opener, in vascular smooth muscle. Br J Pharmacol 1994;113:1538-47.
23. Nilius B, Viana F, Droogmans G. Ion channels in vascular endothelium. Annu Rev Physiol 1997;59:145-70.
24. Spray DC, Bennett MVL. Physiology and pharmacology of gap junctions. Annu Rev Physiol 1985;47:281-303.
25. Chaytor AT, Evans WH, Griffith TM. Central role of heterocellular gap junctional communication in endothelium-dependent relaxations of rabbit arteries. J Physiol (Lond) 1998;508:561-73.
26. Taylor HJ, Chaytor AT, Evans WH, Griffith TM. Inhibition of the gap junctional component of endothelium-dependent relaxations in rabbit iliac artery by 18-α glycyrrhetinic acid. Br J Pharmacol 1998;125:1-3.

Endothelium; EDHF; Ouabain; Sodium pump; Potassium ion; Gastric artery

© 2000 Lippincott Williams & Wilkins, Inc.