507
Calcium-Dependent Fluxes of Potassium-42 and
Chloride-36 during Norepinephrine Activation of Rat
Aorta
Jacquelyn M. Smith and Allan W. Jones
From the Department of Physiology, University of Missouri, Columbia, Missouri
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
SUMMARY. This study was designed to determine whether a-receptor-stimulated monovalent
ionic fluxes in rat aorta required calcium, and, if so, whether both extracellular calcium and
cellularly stored calcium are active. Calcium removal in the presence of 10 mM magnesium (to
maintain membrane stability) inhibited the norepinephrine-stimulated increase in potassium-42
and chloride-36 efflux. However, the norepinephrine-stimulated increase in sodium-24 influx
was relatively resistant to calcium depletion. Protocols were designed to measure the time course
for the changes in potassium-42 efflux and contraction when calcium was removed or replaced
in the presence of norepinephrine. The dose-dependent effect of a calcium antagonist (diltiazem)
was also measured. A close correlation (r = 0.94) was found between inhibition of contraction
and potassium-42 effluxes which followed the regression: % potassium-42 response = 1.0 X (%
contraction) + 1.8%). The slope of 1.0 and intercept near zero suggests the hypothesis that
norepinephrine-stimulated potassium-42 efflux and contraction are codependent on cellular
calcium concentration. This co-dependence held for short phasic responses (~1 minute), as well
as longer tonic responses (>5 minutes). It appears that calcium-dependent potassium-42 effluxes
can be supported by both the influx of extracellular calcium and release of cellular stores. It is
concluded that calcium-dependent potassium channels (and possibly chloride channels) are
operative in rat aorta and are an important component of the graded membrane response to
norepinephrine. The sodium channels, however, do not appear to share this same calcium
dependency. (CircRes 56: 507-516, 1985)
EXCITATION of arterial smooth muscle by catecholamines is associated with increases in fluxes of
ions down their electrochemical gradients (Jones,
1980). Much attention has been given to measures
of Ca++ fluxes in a variety of arteries and species
during excitation-contraction coupling (Cauvin et
al., 1983; Daniel et al., 1983), and the concept has
been developed that a-receptor activation of arterial
smooth muscle increases both Ca++ influx and release from cellular sites (Van Breemen et al., 1980).
These preparations generally respond to such stimuli
with graded depolarization (Somlyo et al., 1969;
Hermsmeyer, 1976); hence, they have been classified as a tonic type of smooth muscle. The timing
and magnitude of the changes in K+ conductance
which accompany a-receptor activation appear to
be important determinants of the tonic response.
Large dose-dependent increases in 42K efflux accompanied a-receptor activation in arteries from rats
(Jones, 1973) and rabbits (Droogmans, et al., 1977;
Martin and Gordon, 1983). The changes reflected
alterations in K+ permeability since they exceeded
the changes in efflux predicted for a simple reduction of the electrochemical driving force (Jones,
1980). Inhibition of K+ conductance by tetraethylammonium ion (TEA) resulted in the generation of
action potentials in previously tonic arterial smooth
muscle (Droogmans et al., 1977; Harder, 1982).
These observations indicate that the coupling between conductances which tend to depolarize
smooth muscle (Ca++, Na+, Cl~) and that which
hyperpolarizes (K+), is a critical determinant of the
type of response.
The coupling between a-receptor occupancy and
subsequent changes in ionic events is not well
understood. Bolton (1979) suggested that receptors
could operate ionic channels directly. Separation of
events modulated by receptors from those secondary
to changes in membrane potential is not complete.
In general, the a-receptor-controlled influx of Ca++
was additive to that induced by depolarization, and
was not as sensitive to blockade by Ca++ antagonists
(Meisheri et al., 1981). It is unclear to what extent
a-receptor-related changes in the movements of
Na+, K+, and Cl~ fall directly under the control of
receptors or are secondary events. We reported that
cAMP-dependent inhibition of norepinephrine
(NE)-induced contraction in rat aorta was closely
related to parallel inhibition of a-receptor-induced
increases in 42K and 36C1 efflux (Jones et al., 1984).
One explanation for this parallel inhibition is that
elevated cAMP results in maintenance of cytoplasmic Ca++ at levels below those needed for contraction and to open Ca++-dependent ionic channels.
Circulation Research/Vol. 56, No. 4, April 1985
508
++
+
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
The Ca dependence of K movements has been
observed in a wide variety of tissues. Gardos (1958)
reported that the net efflux of K+ from red blood
cells was increased by cellular Ca++. This observation has been confirmed in red cells with analyses
of unidirectional fluxes of isotopes (Yingst and Hoffman, 1984) and by patch clamp methods (Grygorczyk et al., 1984). Ca ++ -dependent activation of K+channels has also been observed in excitable cells
such as nerve (Meech, 1978) and skeletal muscle
(Barrett et al., 1982). Relatively few studies of Ca++dependent K+-conductance have been published for
smooth muscle. Recently, Singer and Walsh (1984)
have used patch clamp methods to identify a Ca++dependent K+-conductance in membranes from the
stomach muscle of the Bufo marinus. Den Hertog
(1981) observed that the sustained hyperpolarization induced in guinea pig Taeni caeci by a-receptor
activation required extracellular Ca++, as did the
increase in 42K efflux. These smooth muscles represent the intestinal, phasic type in which a-receptor
activation tends to inhibit contraction. One study is
available on arterial smooth muscle in which a possible Ca++ dependency was observed during norepinephrine (NE) stimulation of 86Rb efflux and
contraction (Martin and Gordon, 1983).
It was the purpose of this investigation to determine whether a-receptor-stimulated 42K efflux required Ca++. If such a dependence existed, it was of
further interest to determine whether extracellular
Ca ++ and/or cellularly sequestered Ca++ are able to
activate the 42K efflux. Correlations were made between the Ca++-dependent changes in 42K efflux
and in contractile response to test the hypothesis
that a Ca++-dependent K+ channel is operative in
arterial smooth muscle of the tonic type.
A preliminary communication of this study has
been made (Jones et al., 1984).
Methods
Animals and Tissues
Male Sprague-Dawley rats (200-250 g wt) were decapitated, and the thoracic aorta was prepared according to a
standardized method described previously (Jones, 1973;
Jones et al., 1984). We removed the endothelial cells from
the strips (flux studies) and rings (contractile studies) by
lightly stroking the intimal surface with filter paper while
the tissues were in the dissection solution. The tissues
were mounted on stainless steel holders, followed by a 3hour incubation at 37°C before initiation of the protocols.
Solutions
The normal physiological solution had the following
composition (in IHM): Na+, 146.2; K+, 5.0; Mg++, 1.2; Ca++,
2.5; Cl", 143.9; HCCV, 13.5; H2PCV, 1.2; and glucose,
11.4. Solutions were gassed with a 97% O2-3% CO2
mixture to obtain a pH of 7.4. Propranolol (3 /IM, Sigma),
ethylenediaminetetracetic acid (0.025 ITIM), and ascorbic
acid (1 mM, Sigma) were added to all solutions for /3receptor blockade and to inhibit oxidation of catecholamines. Aqueous stock solutions were prepared for NE
(Sigma) and diltiazem, (gift from Marion Laboratores).
The 0 Ca++ solutions were prepared with no Ca++, 10 mM
Mg++ (as Cl~ salt), and 2 mM ethyleneglycol-bis-(/3-aminoethyl ether)N,N'-tetraacetic acid (EGTA). The corresponding control solutions contained elevated Mg++ (10 mM)
and Ca++ = 2.5. In one series, 4.5 mM Ca++ (Cl" salt) was
added to the 0 Ca++ solution to compensate for the 2 mM
EGTA.
Isotope Fluxes
Routine equilibration and washout methods were used
(Jones, 1973, 1981). For efflux analyses, open strips were
incubated for 3 hours at 37°C in solution containing 42K
(20 /iCi/ml, University of Missouri Research Reactor) or
42
K plus 36C1 (2 /tCi/ml, ICN). After a 2-second rinse, each
tissue was passed through a series of vigorously gassed
tubes containing nonradioactive solution. Solution
changes, e.g., drugs, 0 Ca++, temperature, can be made
quickly with this method, and diffusional delays can be
maintained to a minimum (Jones, 1980). Washout curves
were computed (IBM) by sequentially adding the tissue
and tube radioactivity (measured by standard gamma and
liquid scintillation methods) in reverse order and by normalization in terms of initial activity. Corrections for isotopic decay were also made. The fraction of the isotope
exchanged per minute which represents the rate constant,
k (per min), was calculated for each washout period and
was used to compare the effects of drugs and ionic
changes.
The influx of 24Na (University of Missouri Research
Reactor) was done on tissues that had been equilibrated 3
hours at 37°C in the normal physiological solution. The
strips then were transferred to either 0 Ca++ solution or a
high Mg++ (10 mM) physiological solution for either 5 or
20 minutes. Half the strips were exposed to NE for an
additional half-minute, or for 5 minutes. The strips then
were transferred into the same solution (±Ca++, ±NE)
with 24Na added (10 /tCi/ml) for a 1-minute period. The
influx was terminated by quickly transferring the tissues
to the physiological solution chilled to 1°C. After several
rinses, the tissues were removed at 10 minutes, blotted,
dried for 2 hours at 110°C, then weighed. Previous studies
have shown that the 10-minute wash at 1°C is sufficient
to remove 24Na from the extracellular space without reducing greatly the slowly exchanging (cellular) component
(Jones, 1981). The 24Na was released by adding 0.5 ml
H2O2 (30% wt/vol) to each tissue, followed by 10-minute
treatment in a microwave oven. The sample was extracted
into a 0.1 M solution of HNO3 for 10 minutes, then
neutralized, and 10 ml of liquid scintillation cocktail were
added. Standards were prepared similarly from the 24Na
incubation media, with Na+ concentration verified by
flame photometry. The counts were corrected for background and isotopic decay, and the data are presented as
jiEq Na+/g dry w * P e r min. The relatively short exposure
to Na (1 minute) resulted in the exchange of less than
20% of the cellular contents. Because this exchange represents only a 10% average over the period for the influx,
back flux corrections were not applied.
Contraction
The rings were prepared and mounted according to a
standardized procedure (Jones et al., 1984). The rings were
stretched to 1.3 times their resting diameter and isometric
force was measured in response to agonists added to the
bath. The responses during the various protocols were
Smith and /ones/Ca++-Dependent Fluxes
509
normalized in terms of the force developed during an
initial control response. The response at 1 minute was
used to measure the relatively fast, phasic component
which was reported to depend mostly on the release of
cellular Ca++ (Deth and Van Breemen, 1977). The response
after 5 minutes was used to measure the tonic component
which required extracellular Ca++ for its maintenance
(Deth and Van Breemen, 1977).
Statistics
Results are presented as means ±SE, and significance
was determined by Student's f-test, with P < 0.05 deemed
to be significant. Standard linear regression methods were
used in one series.
Results
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
Ca++ Removal and Antagonism
Removal of Ca++ (0 Ca++ solution), or addition of
diltiazem, a Ca++ antagonist (Triggle and Swamy,
1980) inhibited the contractile response to NE in rat
aorta, as shown in Figure 1A. Both the initial fast
(phasic) response, and the slow (tonic) response were
inhibited by prolonged exposure to 0 Ca++ and
diltiazem (DZ). Similar treatment also inhibited the
NE-induced increase in 42K efflux (Fig. IB). As in
the case of the contractile response (Fig. 1A), inhibition was more complete by 0 Ca++ than by DZ (P
++
< 0.001). Because removal of Ca from physiological solution (Mg++ = 1.2 mM) was associated with
a major increase in 42K efflux 0ones, 1974), raising
the Mg++ was an important adjustment to maintain
membrane stability in 0 Ca++. After a preliminary
study, Mg++ = 10 mM was chosen because it maintained the resting 42K efflux within 30% of the basal
level and had little effect on the maximal 42K response to NE in the presence of Ca++ (Ak = 0.012
± 0.001/min, n = 12 for both Mg ++ = 1.2 and 10
mM). Higher concentrations of Mg++ maintained the
resting K effluxes closer to basal levels in 0 Ca++,
but significantly inhibited contractile and ionic flux
responses to NE (data not shown).
Increases in 36C1 effluxes (Fig. 2B) which accompanied 42K efflux during NE stimulation (Fig. 2A)
also exhibited Ca++ dependency. About 50% of the
initial phasic response (time <1 minute) was inhibited by 10 minutes of exposure to 0 Ca++ (Table 1).
However, the tonic response at 4.5 minutes was
completely inhibited by 0 Ca++ (Table 1). These
differential effects may reflect an ability for stored
Ca++ released by NE (Deth and Casteels, 1977; Deth
and Van Breemen, 1977) to stimulate 42K and 36C1
effluxes for a brief period.
0.025
0.02
0.015
0.02
0.01
0.015
0.0
-NE
z
a.
5
0.01
0.25
Mg = 10mM
O -Ca
• DZ, 3OuM
0.005
•t
0.0
20
10
MINUTES
0.2
1
0
20
40
60
80
100
MINUTES
FIGURE 1. Panel A, effects of 0 Ca*+ solution and 30 MM diltiazem
(DZ) on a typical contractile response of rat aorta to a supermaximal
concentration of 3 HM NE. Control responses are from the same ring.
The DZ and 0 Ca+* were applied 20 minutes before NE. W indicates
washout of NE from controls. The dashed line represents the basal
tension before NE. All responses were measured in Mg** = 10 mM.
Panel B, effects of 0 Ca+* and DZ on "K efflux responses to NE.
Means plus SE(representative values) of the rate constants are plotted.
After 40 minutes, tissues were placed in either a control solution (•)
(Ca++ = 2.5 mM, Mg++ = 30 mM, n = 11 rats), 0 Co** solution (O)
(ECTA = 2 mM, Mg" = 10 mM, n = 5 rats), or DZ (U) (Ca*+ = 2.5
mM, Mg++ = 10 mM, n = 6 rats). Time is in minutes. Horizontal
lines indicate the period of application for the treatments.
0.15
0.1
0.0 I
0
5
10
MINUTES
15
20
FIGURE 2. Effect of 0 Ca++ (O) on NE stimulation of "K and 36C1
efflux. Panel A, 42K efflux plotted as in Figure W (n = 6 rats). The 0
Ca++ solution was applied at time = 0, and the composition of the 0
Cfl++ and control solution (9) are the same as Figure IB. Panel B,
M
C/ efflux measured simultaneously with the "K efflux in panel A.
Circulation Research/Vol. 56, No. 4, April 1985
510
TABLE 1
Effects of 0 Ca++ Solution (10 minutes) on NE (3 jiM)-Stimulated Increase in the Rate (Ak) of "K
and *C1 Efflux Measured Simultaneously from Rat Aorta (n = 6) and on Contraction (n = 6)
2
,6 C ,
K
Minutes in NE
Ca
++
(Ak per min)
0 Ca++ (Ak per min)
Contraction (%)
0.5
4.5
0.5
4.5
0.5
4.5
0.012
±0.001
0.010
±0.001
0.071
±0.008
0.020
±0.009
100
100
0.007
±0.001
0.001
±0.001
0.032
±0.008
-0.002
±0.003
33
±6
17
±1
<0.02
O.001
<0.01
<0.05
<0.001
<0.001
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
If the NE-induced increases in ion fluxes are dependent on free intracellular Ca++, then a close
relation would be expected between alterations in
the efflux response and contraction. Systematic investigations were made to test the closeness of this
relation. Protocols were designed to measure (1) the
concentration-dependent effect of DZ on NE-stimulated 42K efflux and contraction, (2) the time course
for changes in 42K efflux and contraction when Ca++
was replaced in 0 Ca ++ + NE, and (3) the time
course for changes in NE-stimulated 42K efflux and
contraction when Ca ++ was removed.
traction and on 42K efflux was relatively small compared to the inhibition of KCl-induced changes in
K efflux and contraction (Fig. 3). This contractile
effect is consistent with the report of Triggle and
Swamy (1980) that Ca++ antagonists were more
potent inhibitors of KCl-induced contractions than
of NE contractions in arterial smooth muscle. Similar
to the results of Saida and Van Breemen (1983),
extremely high concentrations of DZ could produce
almost complete inhibition of NE contractions (Fig.
3). This inhibition also extended to NE-induced
changes in 42K efflux (Fig. 3).
Diltiazem Concentration
Diltiazem inhibited the NE-induced changes in
both 42K efflux and contraction in a similar manner
(Fig. 3). The inhibitory responses were evaluated at
1-minute exposure to NE for the phasic response
and at 10 minutes for the tonic response. The percent
inhibition at 1 and at 10 minutes was similar for
each concentration of DZ. At a DZ concentration of
30 fiM, the inhibitory effect on NE-stimulated con-
Ca ++ Replacement
100
o
80
to
60
o
o
o
1.0
20
0
-6
LOG
DOSE
-5
-l.
DILTIAZEM
-3
FIGURE 3. Effects of various concentrations of DZ (log molar scale)
on the NE stimulation of rat aorta. All tissues were exposed to DZ
for 20 minutes before stimulation. Effects on "K efflux (n = 6 rats) at
1 minute (•) and after 10 minutes (O) in NE are plotted with
equivalent analyses of effects on contraction (n = 5 rats) at 1 (M) and
10 minutes (O). KCl (85 msi) was used as a stimulus at DZ = 30 tiM.
Effects on "K efflux (n = 6 rats) at 1 (A) and after 10 minutes (A; in
KCl are plotted with equivalent effects on contraction (n = 5 rats) at
1 (•) and 10 minutes (0). The standard errors were of the order
±5%.
The addition of Ca++ to 0 Ca++ solution containing
NE caused a contraction which approached that
developed when NE was added to a Ca++-containing solution (Fig. 4A). The 42K efflux response to
replacement of Ca++ in the presence of NE (Fig. 4B)
was similar to the contractile response. NE increased
42
K efflux transiently in 0 Ca++, followed by a return
toward basal rates, in sharp contrast to the response
in the presence of Ca++. Ca++ replacement after 15
minutes in 0 Ca++ + NE resulted in an increase in
42
K efflux (P < 0.001) to the levels achieved by
tissues that had been exposed to NE and Ca++
throughout (Fig. 4B). The response to Ca++ replacement would be expected to be in the opposite direction if its major effect were to stabilize the membrane.
The time courses for the return of the responses
after Ca++ replacement appear in Figure 5. The effect
on 42K efflux reached a maximum before contraction,
but good agreement was noted overall. The return
of both 42K efflux and contraction when Ca++ was
replaced in the presence of NE indicates that the
availability of cellular Ca++, rather than a-receptor
activation alone, determined the 42K efflux response.
Ca ++ Removal
The transient stimulation of 42K efflux by NE in 0
Ca++ is consistent with the hypothesis that [Ca++]ceU
is elevated by the release of Ca++ from a cellular
pool of limited capacity (Deth and Casteels, 1977;
Karaki and Weiss, 1979). This hypothesis was tested
further by varying the time in 0 Ca ++ before expo-
Smith and Jones/Ca++-Dependent Fluxes
511
0.02
FIGURE 4. Panel A, effects of 0 Ca** solution (20
minutes before NE) and replacement of Ca** on a
typical contractile response in rat aorta. The data
are plotted as in Figure 1A. Panel B, effects on 0
Ca** solution and Ca** replacement (O) on NE
stimulation of i2K efflux (n — 7 rats for each set).
Controls (•) were exposed to Ca** throughout.
Data are plotted as in Figure IB.
Y 0.015
g
NEMg=10mM
O -Ca
•
+Ca
0.005
0.0
20
OCa = 4-5mM
80
40
60
MINUTES
sure to NE. The phasic contractile response (1 minute in NE) was reduced as the time in 0 Ca++ was
increased (Fig. 6). Only a small tonic contraction
remained after 12.5 minutes in 0 Ca++ + NE (Fig.
6). The initial response of 42K efflux to NE (k at 1
minute) also showed a reduction which depended
on the time the tissue was preincubated in 0 Ca++
100
RESPOI
UJ
LO
7
80
60
CONTRACTION
L0
T
Axir
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
a. 001
20
oL
0
10
20
30
MINUTES
FIGURE 5. Percent of maximum response after Ca** replacement in
the presence of NE for "K efflux (•) (n = 7 rats) and contraction (M)
(n = 5 rats). Arrows indicate the representative rise times required to
reach 63% maximum response, (0) 3.3 ± 0.6 and (U) 6.6 ± 2.0
minutes (not significant). The SE (representative values) are shown by
vertical bars. The only significant difference (P < 0.01) occurred at
7.5 minutes.
100
(Fig. 7). For instance, when the tissue was simultaneously placed in 0 Ca++ plus NE, approximately
90% of the control response occurred, while 40
minutes in 0 Ca++ removed any vestiges of a phasic
response to NE (Fig. 7). The 42K efflux returned
toward basal levels with continued exposure to 0
Ca++ + NE (Fig. 7). The decrement of the contractile
and 42K responses was closely related over a range
of times which led to almost complete inhibition
(Fig. 8). However, there was a tendency for contraction to decrease more rapidly over the initial period
and to retain some long-term residual response in
comparison to 42K efflux (Fig. 8).
Correlation of 42K Efflux and Contraction
The hypothesis that both NE-induced changes in
42
K efflux and in contraction were dependent on the
same factor, [Ca++]ceii, was tested by correlating the
average responses derived from the three protocols
(Figs. 3, 5 and 8). The correlation was close to 1 and
highly significant (P < 0.001). Moreover, the slope
of 1.0 and intercept near zero indicated a co-dependence which may reside in a common dependency on [Ca++]Ceii- This observation suggests the
hypothesis that the established role of [CJr + ]ceu as a
controller of contraction extends to the activation of
42
K efflux during NE stimulation.
Ca++ Dependency of 24Na Influx
The 24Na influx was measured under basal conditions and during NE treatment in the presence
Circulation Research/Vol. 56, No. 4, April 1985
512
0 MIN - C a
100
LU
<5
£
o
o
in
o
o
5 MIN -Ca
L2
NE 1MIN
80
60
40
20
10
20
-Ca.
t
K.
CONTRACTION, NE 1MIN
30
40
MINUTES
FIGURE 8. Percent of control response to NE at various times after
removal of Ca**. Data for "K efflux (0) (n = 6 rats) and contraction
(M) (n= 6 rats) are plotted + SE. Significant differences (P < 0.05)
occurred at 1, 21, and 41 minutes.
NE
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
COMIN -Ca
t
NF
FIGURE 6. Effects of the time in 0 Ca** solution on typical contractile
responses to NE. Data are plotted as in Figure IA.
and absence of Ca++ (Table 2). The basal 24Na influx
was increased in the 0 Ca++ solution (P < 0.001).
Ca ++ depletion for 5 minutes did not greatly inhibit
the NE-stimulated 24Na influx. At 1.0 minutes in
NE, the stimulated influxes were similar in 0 Ca++
and controls (+Ca ++ ), whereas 4.5 minutes of additional exposure to 0 Ca ++ led to only a 30%
reduction in NE-stimulated influx, compared to the
response in the presence of Ca++. This reduction is
in sharp contrast to the attenuation (>80%) of contraction (Fig. 6) and 42K efflux (Fig. 7) in 0 Ca++ 5
minutes, plus 5 minutes in 0 Ca ++ + NE. Prolonged
exposure to 0 Ca++ (20 minute) before NE did not
reduce the initial 24Na influx (1.0 minutes in NE),
but did inhibit the response after an additional 4.5
minutes in 0 Ca++ plus NE. Under this latter condition, the contractile response (Fig. 4A) and 42K efflux
(Fig. 4B) were more than 90% inhibited. The NEstimulated influx of 24Na appears to be resistant to
depletion of Ca++, except under conditions of almost
complete removal of extracellular and cellular
sources for contractile responses.
Discussion
++
Ca -Dependent K+ Channels
The results of the three protocols, which were
designed to alter cellular Ca , showed that a close
relation exists between force development and 42K
efflux from rat aorta during a-receptor stimulation.
Since little evidence exists for a cause-and-effect
relation between these two events, it is reasonable
to conclude that both responses are controlled by a
common factor, e.g., [Ca++]cen. Force development
in skinned arterial smooth muscle was dependent
on [Ca++], with half maximal stimulation achieved
0.025
0.02
FIGURE 7. Effects of the time in 0 Ca** solution on "K
efflux and response to NE. Data are plotted as in Figure
IB (n = 6 rats for each protocol). Note that, in three
series, 0 Ca** solution was applied at the same time
(40 minutes), and that NE was applied at 0 (V, 5 W,
and 40 minutes (M) later. Controls (9) were incubated
in Ca**-containing solution throughout.
0.015
0.01
I
0.0
20
60
MINUTES
80
100
120
Smith and /ones/Ca++-Dependent Fluxes
513
TABLE 2
Influx of "Na in Rat Aorta and the Effects of NE (3 MM) and 0 Ca** Solution Applied at Various
Times
Basal (-NE)
+ NE
A 24Na influx (NE)
2
*Na influx
Time in NE "Na influx
(/iEq/g dry wt per
(^Eq/g d.w./min) P*
(min) (^Eq/g d.w./min)
min)
n
Time in
0 Ca++
(min)
12
0
1.91
±0.14
1.0
2.40
±0.13
0.49
±0.19
0.03
7
0
1.96
±0.11
5.5
2.89
±0.12
0.93
±0.09
0.001
P|
PJ
3.04
±0.21
0.001
1.0
3.56
±0.24
0.52
±0.25
0.07 >0.5
3.10
±0.18
0.001
5.5
3.70
±0.21
0.63
±0.11
0.001 0.0
20
3.04
±0.26
0.005
1.0
4.26
±0.45
1.22
±0.42
0.03 >0.1
20
3.47
±0.03
0.001
5.5
3.69
±0.41
0.22
±0.37
11
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
>0.5
0.1
* Comparison between +Ca++ and 0 Ca++ conditions —NE.
•f Comparison between —NE and +NE conditions with no change in Ca++.
X Comparison between +Ca++ and 0 Ca++ on A 24Na influx at equivalent time in NE.
at 0.8 to 6 X 10 7 M and maximal responses at 1-3
/iM (Ruegg and Paul, 1982; Chatterjee and Murphy,
1983). Studies that employed patch clamp techniques have demonstrated Ca++ dependency of K+
channels (as well as voltage dependency) in a variety
of types of cells. (Miller, 1983; Schwarz and Passow,
1983). At zero or slightly negative membrane potentials, however, [Ca ] of 1 fiM or greater was required
to increase the time that K+ channels were open in
red blood cells (Grygorcyzk et al., 1984; Yingst and
Hoffman, 1984) and in skeletal muscle (Barrett et
al., 1982; Latorre et al., 1982). A unique study on
membrane patches from smooth muscle also used
high [Ca++] to increase the open time of K+ channels
(Singer and Walsh, 1984). If our hypothesis is correct, then we would predict that the open time of
Ca++-dependent K+ channels in arterial smooth
muscle would increase with similar [Ca++] (0.1-3
(IM) as contraction in the presence of physiological
membrane potentials [Em = —25 to —45 mV in NE
(Jones, 1980; Haeusler, 1983). K+ channels from
skeletal muscle required 5 HM Ca++ (or higher) to
open at Em = -20 to -40 mV (Barrett et al., 1982;
Latorre et al., 1982). These comparisons are based
on skeletal muscle, because the conductances of
Ca++-activated K+ channels in smooth muscle
(Singer and Walsh, 1984) were similar to those in
skeletal muscle [200-300 pS (Barrett et al., 1982;
Schwarz and Passow, 1983)], in contrast to red blood
cells which were 10-fold lower (Schwarz and Passow, 1983; Grygorsczyk et al., 1984). The difference
in Ca++ dependence between isolated patches and
intact cells may reside in the availability of a cofactor
such as calmodulin which would be lost when cellular structures are exposed to experimental solu-
tions (Ruegg and Paul, 1982). This explanation
needs experimental verification.
The Ca++ activation and K+ effluxes during NE
stimulation in this study is consistent with observations that increased membrane conductance occurred at constant membrane potential during NE
stimulation (Haeusler, 1982, 1983). Treatment with
TEA reduced the conductances resulting in depolarization and ability to generate action potentials (Casteels, et al., 1977; Haeusler, 1983). Ca++ activation
of K+ channels (and contractile response) with increasing levels of NE could account for much of the
voltage-independent change in conductance. It is
also possible that the Ca++ dependency of K+ channels in the tonic type of smooth muscle requires
lower [Ca++] than the phasic type; hence, regenerative action potentials would be more readily inhibited. A shift to higher [Ca++] for activation of K+
channels (at a given membrane potential) would
allow a larger inward current to be unopposed by
an increase in K+ conductance, and hence a regenerative response (action potential) would be more
likely to occur. Action potentials tend to be conducted in smooth muscle and result in relatively
large contractile responses (Golenhofen, 1981) consistent with propulsive motility. Closer coupling of
inward Na+ and Ca++ currents with K+ efflux would
result in graded changes in membrane potential and
in contraction (Somlyo et al., 1969; Harder, 1982;
Haeusler, 1983). Such graded responses can sustain
a continuous opposition to distending forces, e.g.,
blood pressure, with less expenditure of energy than
phasic contractile events (Chatterjee and Murphy,
1983).
Much of the above analyses of Ca++-dependent
Circulation Research/Vol. 56, No. 4, April 1985
514
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
K+ efflux was based on the assumption that force
development quantitatively reflects changes in
[Ca++]Ceii, and that Ca++ activation of the K+ channel
(increased probability of being open) occurred
quickly with little inactivation (alteration in the
probability of being open during constant stimulation). The first assumption is reasonable, although
not entirely correct. Hysteresis has been observed
between the [Ca++] dependence of contraction induced by increasing [Ca++] and relaxation resulting
from reduction of [Ca++] in skinned arterial smooth
muscle (Chatterjee and Murphy, 1983). Approximately 10 times lower [Ca++] was required for force
maintenance than for force development. This difference may underlie some of the differences in the
time courses for changes in 42K efflux and contraction (Figs. 5 and 8). These discrepancies are relatively
small, however, in comparison to the correlation
derived over the whole range of responses (Fig. 9).
Opening of Ca++-activated channels occurs much
more rapidly than the resolution of the 42K efflux
method (Barrett et al., 1982). Likewise, inactivation
of these channels during constant stimulation was
not a dominant feature (Barrett et al., 1982; Latorre
et al., 1982). The assumptions that the opening is
rapid and stable appear to be reasonable approximations. Some fade occurred in the response of 42K
efflux to NE (Jones 1973; Fig. 2a). It is not clear
whether this fade reflected slow inactivation of
Ca+"%activated K+ channel or represented a decay
of [Ca++]cen during force maintenance (Morgan and
Morgan, 1982; Kato et al., 1984). Such transients
could result from the acute influx and release of
Ca ++ from a pool of limited capacity, followed by
a-receptor stimulated influx of Ca ++ operating alone
in the presence of increased Ca ++ efflux. If it can be
confirmed by direct measures than the Ca++-activated K+ channel has a stable dependence on voltage and [Ca++], once the membrane has reached a
stable voltage, then the time course for the a-recep:1OO
LU
80
LO
O
0_
m 60
a:
CO
LU
cc
20
o
• Dz NE 1 MIN
• Dz NE10MIN
• -Ca NE 1 MIN
O - C a NE 12.5MIN
A Ca REPLACEMENT
0
0
20 40 60 80 100
% CONTROL RESPONSE.CONTRACTION
FIGURE 9. Correlation between % control responses for t2K efflux and
for contraction when Ca*+ was altered by DZ (U, Q), Ca++ removal
(9, O), or replacement (A). The line represents the linear regression,
% *2K = 1.0 x (% contraction) + 1.8%, with a correlation coefficient,
r = 0.94(P< 0.001).
tor-stimulated 42K efflux may reflect that for
[Ca++]«n.
Our finding of Ca++-dependent increases in 42K
efflux during NE stimulation are in general agreement with those of Martin and Gordon (1983), with
one notable exception. They showed a close relation
between NE-induced increases in 86Rb efflux and
tension development by rabbit aorta. As in our
study, these two events were partially inhibited by
a Ca++ antagonist. It was also noted that the increase
in 86Rb efflux during Ca++ removal could be reversed
by elevating Mg ++ . On the other hand, La+++ had a
differential action on the NE-stimulated contracture
and 86Rb efflux. A transient contraction occurred,
consistent with the release of stored Ca++, but no
related change occurred in 86Rb efflux. The authors'
interpretation, that increased [Ca++] from release
had no effect on 86Rb effluxes, has an alternate
explanation. Various types of K+ channels can be
blocked by divalent and trivalent ions, such as Ba++,
Cu++, Cd++, and La+++ (Takata et al., 1967; Miller,
1983). Treatment with La+++ may have directly
blocked K+ channels and, hence, the 86Rb response
to NE. Despite this point, we feel that it is likely that
Ca++-dependent K channels are operative in both
rat and rabbit aorta, and are an important factor for
the modulation of the membrane potential during
NE stimulation.
Ca++-Dependent Cl" Efflux
NE also increased 36C1 efflux from vascular
smooth muscles with a computed increase in permeability that exceeded that for K+ (Table 8, in Jones,
1980). As in the case of 42K, removal of Ca++ greatly
inhibited the phasic response and abolished the later
tonic response (Table 1). Based on the parallelism
with the Ca++ dependency of 42K efflux, it is speculated that a Ca++-dependent Cl~ channel is operative in rat aortic smooth muscle. There is additional
basis for this speculation, in that a Ca++-dependent
Cl~ current has been observed in at least one cell,
the Xenopus oocyte (Barish, 1983). The Cl" equilibrium potential, EQ, in the oocyte was significantly
less than the membrane potential (E a = —24 mV vs.
Em = -60 to -90 mV), a characteristic it shares with
arterial smooth muscle which also exhibited a significant difference between ECi and Em (—30 mV vs.
-55 mM, Jones, 1980).
It is unlikely that Cl~ and K+ share the same
channel. The peak Cl~ efflux exceeded that for K+
by 2-fold (Jones, 1980), and single channel analyses
indicated that Ca++-dependent K+ channels from
skeletal muscle were impermeant to Cl~ (Latorre et
al., 1982). Although little evidence is available concerning Ca++ dependency of Cl~ fluxes, more detailed investigations of, e.g., single channel behavior, may be a productive endeavor.
NE-Stimulated Na+ Influx
The influx of 24Na was stimulated by NE, in
addition to the stimulation of 42K and 36C1 effluxes.
Previously, NE effects were measured on 24Na ef-
Smith and /ones/Ca++-Dependent Fluxes
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
fluxes from Na+-loaded arteries (Droogmans et al.,
1977; Jones, 1980). The percent increases (25-50%)
above basal levels were in agreement with the results of this study. The increase in 24Na influx was
maintained for at least 5 minutes in NE plus Ca++,
which indicated that a-receptor stimulation led to
simultaneous increases in 24Na, 42K, and 36C1 fluxes
down their electrochemical gradients. Given the limitations of measuring rapid fluxes in tissues strips
(Jones, 1980) it was not possible to determine
whether increases in 24Na influx occurred before
changes in 42K and 36C1 efflux during the initial
application of NE. The NE stimulation of 24Na influx
was more resistant to Ca++ removal than the 42K
and 36C1 effluxes. Only under the extreme condition
used (20 minutes in 0 Ca++ plus 5.5 minutes in NE)
was the NE-stimulated 24Na influx inhibited. It appears, however, that the increase in 24Na influx falls
more directly under the control of the a-receptor
than 42K or 36C1 effluxes, which are linked with
changes in cellular Ca++.
The basal 24Na influx increased in 0 Ca ++ in
contrast to the NE-stimulated influx. The basal and
NE-stimulated influx of 24Na exceeded the 45Ca
influx reported for rat aorta (Godfraind, 1976). The
basal 45Ca influx was only one-tenth (molar basis)
that for Na+; thus, the elevation in Na+ influx in 0
Ca++ solution was beyond any reduction in Ca++
influx. This elevation may result from increased Na+
permeability (despite the 10 urn Mg++), or perhaps
removal of adsorbed Ca++ which had occupied sites
for the membrane transport of Na+.
Control of Excitation
The observation that ionic events during a-receptor stimulation were controlled by [Ca++]ceii alters
the interpretation of some previous work. Haeusler
(1983) proposed that a combination of voltage- and
Ca++-dependent processes are operative in arteries
during NE stimulation. It is probably not realistic to
separate the responses into independent voltageand Ca++-regulated events (channels), since it is
established that the Ca++-dependent K+ channels in
several cell types exhibit a marked voltage dependency (Miller, 1983). The Ca++ sensitivity of the K+
channels (high-conductance type) increased as the
inside of the membrane was made more positive
(depolarization). Electromechanical coupling and
pharmacomechanical coupling (Somlyo and Somlyo, 1970) would therefore be an interactive process
during NE stimulation of polarized arterial smooth
muscle. It would be difficult to achieve an isolated
electromechanical event, since the Ca++ entering
during depolarization, e.g., KG treatment, would
itself alter membrane conductance. That is—in addition to controlling Em, it is necessary to control
[Ca++]ceii in order to define membrane channel characteristics. The development of a comprehensive
model for excitation of arterial smooth muscle will
therefore be even more complex than previously
thought. The application of recent improvements in
515
electrophysiological and biochemical techniques to
the study of vascular smooth muscle may provide
the required definitive approach.
We thank Nilam Dheri and Anthony Sanchez for capably conducting many of the protocols. We also thank Susanna Harwood for
technical assistance. Diltiazem was a gift from Marion Laboratories.
This work was supported in part by DHHS Grant HL 15852 and
postdoctoral support from DHHS HL 07094.
Address for reprints: Dr. Jacquelyn M. Smith, Department of
Physiology, University of Missouri, Columbia, Missouri 65212.
Received August 28, 1984; accepted for publication January 17,
1985.
References
Barish M (1983) A transient calcium-dependent chloride current
in the immature Xenopus oocyte. J Physiol (Lond) 342: 309-325
Barrett JN, Magleby KL, Pallotta BS (1982) Properties of single
calcium-activated potassium channels in cultured rat muscle. I
Physiol (Lond) 331: 211-230
Bolton TB (1979) Mechanisms of action of transmitters and other
substances on smooth muscle. Physiol Rev 59: 606-718
Casteels R, Kitamura K, Kuriyama H, Suzuki H (1977) The
membrane properties of the smooth muscle cells of the rabbit
main pulmonary artery. J Physiol (Lond) 271: 41-61
Cauvin C, Loutzenhiser R, Van Breemen C (1983) Mechanisms
of calcium antagonist-induced vasodilation. Annu Rev Pharmacol 23: 373-396
Chatterjee M, Murphy RA (1983) Calcium-dependent stress
maintenance without myosin phosphorylation in skinned
smooth muscle. Science 221: 464-466
Daniel EE, Grover AK, Kwan CY (1983) Calcium. In Biochemistry
of Smooth Muscle, edited by NL Stephens. Boca Raton, Florida,
CRC Press, pp 1-88
Den Hertog A (1981) Calcium and the a-action of catecholamines
on guinea pig Taenia caeci. J Physiol (Lond) 316: 109-125
Deth R, Casteels R (1977) A study of releasible Ca++ fractions in
smooth muscle cells of the rabbit aorta. J Gen Physiol 69: 401416
Deth R, Van Breemen C (1977) Agonist induced release of intracellular Ca++ in the rabbit aorta. J Membr Biol 30: 363-380
Droogmans G, Raeymaekers L, Casteels R (1977) Electro- and
pharmacomechanical coupling in the smooth muscle cells in
the rabbit ear artery. J Gen Physiol 70: 129-148
Gardos G (1958) The function of calcium in the potassium permeability of human erythrocytes. Biochim Biophys Acta 30: 653654
Godfraind T (1976) Calcium exchange in vascular smooth muscle,
action of noradrenaline and lanthanum. J Physiol (Lond) 260:
21-35
Golenhofen K (1981) Differentiation of calcium activation processes in smooth muscle using selective antagonists. In Smooth
Muscle: An Assessment of Current Knowledge, edited by E
Bulbring, AF Brading, AW Jones, T Tomita. London, Edward
Arnold, pp 157-170
Grygorcyzyk R, Schwarz W, Passow H (1984) Ca2+-activated K+
channels in human red cells: Comparison of single-channel
currents with ion fluxes. Biophys J 45: 693-698
Haeusler G (1982) a-Adrenoceptor mediated contractile and electrical responses of vascular smooth muscle. J Cardiovasc Pharmacol 4: S97-S100
Haeusler G (1983) Contraction, membrane potential, and calcium
fluxes in rabbit pulmonary arterial muscle. Fed Proc 42: 263268
Harder DR (1982) Membrane electrical activation of arterial
smooth muscle. In Vascular Smooth Muscle: Metabolism, Ionic
and Contractile Mechanisms, edited by MF Crass, III, CD
Barnes. New York, Academic Press, pp 71-97
Hermsmeyer K (1976) Electrogenesis of increased norepinephrine
sensitivity of arterial vascular smooth muscle in hypertension.
Circ Res 38: 362-367
516
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
Jones AW (1973) Altered ion transport in vascular smooth muscle
from spontaneously hypertensive rats and influence of aldosterone, norepinephrine and angiotensin. Circ Res 33: 563-572
Jones AW (1974) Altered ion transport in large and small arteries
from spontaneously hypertensive rats and the influence of
calcium. Circ Res 34: (suppl 1): 117-122
Jones AW (1980) Content and fluxes of electrolytes. In Handbook
of Physiology, The Cardiovascular System, vol II, Vascular
Smooth Muscle, edited by DF Bohr, AP Somlyo, HV Sparks.
Washington, D.C., Am Physiological Society, pp 253-299
Jones AW (1981) Kinetics of active sodium transport in aortae
from control and deoxycorticosterone hypertensive rats. Hypertension 3: 631-640
Jones AW, Smith JM, Dheri N (1984) Calcium activation of ionic
fluxes during norepinephrine stimulation of arterial smooth
muscle (abstr). Fed Proc 43: 168A
Karaki H, Weiss GB (1979) Alteration in high and low affinity
binding of 45Ca in rabbit aortic smooth muscle by norepinephrine and potassium after exposure to lanthanum and low
temperature. J Pharmacol Exp Ther 211: 86-92
Kato S, Osa T, Ogasawara T (1984) Kinetic model for isometric
contraction in smooth muscle on the basis of myosin phosphorylation hypothesis. Biophys J 46: 35-44
Latorre R, Vergara C, Hidalgo C (1982) Reconstitution in plasma
lipid bilayers of a Ca++-dependent K+ channel from transverse
tubule membranes isolated from rabbit skeletal muscle. Proc
Natl Acad Sri USA 79: 805-809
Martin W, Gordon JL (1983) Differential calcium dependence of
contractile response and 86Rb efflux from rabbit aorta induced
by vasoactive stimuli. J Cell Physiol 115: 46-52
Meech RW (1978) Calcium-dependent potassium activation in
nervous tissues. Annu Rev Biophys Bioeng 7: 1-18
Meisheri KD, Hwang O, Van Breemen C (1981) Evidence for two
separate Ca2+ pathways in smooth muscle plasmalemma. J
Circulation Research/Vol. 56, No. 4, April 1985
Membr Biol 59: 19-25
Miller C (1983) Integral membrane channels: Studies in model
membranes. Physiol Rev 63: 1209-1242.
Morgan JP, Morgan KG (1982) Vascular smooth muscle: The first
recorded Ca2+ transients. Pflugers Arch 395: 75-77
Ruegg JC, Paul RJ (1982) Vascular smooth muscle, calmodulin
and cyclic AMP-dependent protein kinase alter calcium sensitivity in porcine carotid skinned fibers. Circ Res 50: 394-399
Saida K, Van Breemen C (1983) Mechanisms of Ca++ antagonistinduced vasodilation, intracellular actions. Circ Res 52: 137142
Schwarz W, Passow H (1983) Ca2+-activated K+ channels in
erythrocytes and excitable cells. Annu Rev Physiol 45: 359-374
Singer JJ, Walsh JV (1984) Large conductance Ca++-activated K+
channels in smooth muscle cell membrane, reduction in unitary
currents due to internal Na+ ions. Biophys J 45: 68-70
Somlyo AP, Somlyo AV (1970) Vascular smooth muscle. II.
Pharmacology of normal and hypertensive vessels. Pharmacol
Rev 22: 249-353
Somlyo AV, Vinall P, Somlyo AP (1969) Excitation-contraction
coupling and electrical events in two types of vascular smooth
muscle. Microvasc Res 1: 354-373
Triggle DJ, Swamy VC (1980) Pharmacology of agents that affect
calcium, agonists and antagonists. Chest 78: 174s-179s
Van Breemen C, Aaronson P, Loutzenhiser R, Meisheri K (1980)
Ca2+ movements in smooth muscle. Chest 78: 157s-165s
Yingst DR, Hoffman JF (1984) Ca++-induced K+ transport in
human red blood cell ghosts containing arsenazo III. J Gen
Physiol 83: 19-45
INDEX TERMS: Rat aorta • 42K, 36C1, "Na fluxes • Contraction • Ca++ antagonist • Ca++-dependent K+ channel • Ca++dependent Cl" channel • Norepinephrine • Arterial smooth muscle.
Calcium-dependent fluxes of potassium-42 and chloride-36 during norepinephrine activation
of rat aorta.
J M Smith and A W Jones
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
Circ Res. 1985;56:507-516
doi: 10.1161/01.RES.56.4.507
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1985 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7330. Online ISSN: 1524-4571
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circres.ahajournals.org/content/56/4/507
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Circulation Research can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further information
about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation Research is online at:
http://circres.ahajournals.org//subscriptions/