Hypoxia reduces potassium currents in cultured rat
pulmonary
but not mesenteric arterial myocytes
XIAO-JIAN
YUAN, WILLIAM
F. GOLDMAN,
MARY
LEWIS J. RUBIN, AND MORDECAI
P. BLAUSTEIN
L. TOD,
Department
of Physiology, Division of Pulmonary and Critical Care Medicine, Department of Medicine,
and Hypertension Center, University of Maryland School of Medicine, Baltimore, Maryland 21201
DECREASED
alveolar partial pressure of oxygen causes
pulmonary
arterial constriction
(pulmonary
hypertension), but systemic arterial relaxation
(systemic hypotension). This hypoxia-induced
pulmonary vasoconstriction (HPV) is not only an important mechanism in
the matching of regional blood flow and ventilation
in
the lung, but also a major cause of pulmonary hypertension in patients with mountain sickness and pulmonary
heart disease. There are, however, no adequate explanations for the mechanisms responsible for HPV and hypoxia-induced
systemic vasodilation.
We previously reported (35) that the reduction of O2
tension (POT) from 135 Torr to ~40 Torr gradually and
L116
1040-0605/93
$2.00 Copyright
0
reversibly increased the resting tension in endotheliumdenuded as well as endothelium-intact
rat pulmonary
arteries (PA) but did not affect resting tension in mesenteric arteries (MA). This HPV was inhibited by the
K+ channel activator, cromakalim,
by Ca2+-free solution and by the Ca 2+ channel blocker, verapamil. Moreover, glibenclamide,
a K+ channel blocker, reversed the
cromakalim-induced
relaxation during HPV. Based on
these data, we hypothesized that HPV is an intrinsic
mechanism of PA myocytes that is related to hypoxiainduced depolarization
resulting from either a decrease
in K+ conductance or an increase in Ca2+ conductance.
In contrast, the lack of a vasoconstrictor
response to
hypoxia in MA might be due to 1) little or no membrane
depolarization
during hypoxia, 2) voltage-gated
Ca2+
channels that are less sensitive to small changes in membrane potential, and/or 3) contractile machinery that
has a lower sensitivity to Ca2+. Thus the divergent responses between MA and PA to hypoxia might be attributed to differences in mechanisms inherent within
these respective smooth muscle cells.
Bergofsky and Holtzman (4) first suggested that HPV
may be due to depolarization
of the VSM. It was subsequently demonstrated that hypoxia depolarized isolated
cat pulmonary resistance arteries and induced Ca2+-dependent spontaneous electrical activity (16). Furthermore, Suzuki and Twarog (31) showed that chronic hypoxia (7 days) depolarized rat main PA smooth muscle
from -52 to -42 mV. Recently, Post et al. (23) reported
that hypoxia (Po2 = 40 Torr) inhibited Ca2+-activated
K+ currents in canine PA smooth muscle.
In a different type of preparation, rabbit carotid body
type I cells, reduced O2 tension inhibited the whole cell
K+ currents without altering the Ca2+ currents, thereby
decreasing the frequency of spontaneous action potentials (19). Ganfornina
and Lopez-Barneo
(14) used
membrane patches of these arterial chemoreceptor cells
to identify an 02-sensitive K+ channel that was not
regulated by intracellular
Ca2+ or adenos ine triphosphate (ATP). On exposure to hypox ia, the open probability of this 02-sensitive K+ channel was reversibly
decreased by at least 50% without altering the singlethan nel conductance .
In contrast to the aforementioned
observations, hypoxia (Po2 reduced to -15 Torr) increased K+ channel
activity in cat cerebral arterial smooth muscle cells recorded in the cell-attached
mode (5). This raises the
possibility that the increased K+ channel activity during
hypoxia may have been due to a decrease in intracellular
ATP and may have been mediated by ATP-sensitive
K+
channels. Also, in cardiac myocytes, anoxia (Po2 5 0.5
1993 the American
Physiological
Society
Downloaded from http://ajplung.physiology.org/ by 10.220.32.247 on June 14, 2017
Yuan, Xiao-Jian,
William
F. Goldman,
Mary L. Tod,
Lewis J. Rubin, and Mordecai
P. Blaustein.
Hypoxia reduces potassium currents in cultured rat pulmonary
but not
mesenteric arterial myocytes. Am. J. Physiol. 264 (Lung Cell.
Mol. PhysioZ. 8): Lll6-L123,
1993.-To
explore possible mechanisms underlying
hypoxia-induced
pulmonary
vasoconstriction, the effect of hypoxia on outward K+ current (I&
was
evaluated in primary cultured rat pulmonary (PA) and mesenteric (MA) arterial smooth muscle cells using the whole cell
patch-clamp
technique. When the cells were bathed in standard
physiological
salt solution and the patch pipettes contained
Ca2+-free media with 10 mM ethylene glycol-his@-aminoethyl
ether)-N,N,N’,N’-tetraacetic
acid (EGTA), virtually all of the
Iout, including both the rapidly inactivating component (I,J and
the steady-state (noninactivating)
component
(I& was mediated by voltage-gated
I(+ channels. Reduction of 0, tension in
the bath solution from 155 Torr to ~74 Torr with sodium
dithionite
reversibly inhibited both Irt and I,, in PA myocytes,
but not in MA myocytes. The hypoxia-sensitive
I,, was activated at about -50 mV; thus, some of the channels responsible
for this current may be open at the resting membrane potential
(-40 t 1 mV) of PA cells used in this study. Hypoxia also
significantly depolarized PA cells bathed in PSS (1.8 mM Ca2+)
from -40.7 t 1.3 to -24.0 t 2.4 mV, and PA cells bathed in
Ca2+-free PSS (0.1 mM EGTA) from -38.4 t 1.3 to -26.1 of:3.9
mV. The hypoxia-induced
inhibition
of Iout in PA cells was
accompanied by an apparent increase in inward Ca2+ current.
Removal of extracellular
Ca2+ and addition of 2 mM EGTA to
the bath solution while maintaining
a Ca2+-free intracellular
solution with 10 mM EGTA in the pipette (to prevent Ca2+activated K+ channels from opening) did not preclude the hypoxia-induced
inhibition
of Iout in PA cells. These data indicate
that hypoxia attenuates voltage-gated
K+ channel activity in
PA cells but not in MA cells. The mechanism by which hypoxia
inhibits IoUt is not known, but might be related to inhibition
of
oxidative metabolism. This inhibition
of Iout depolarizes the PA
cells. By secondarily opening voltage-gated Ca2+ channels and
promoting
Ca2+ entry, the block of these K+ channels might
be responsible for initiating
hypoxia-induced
pulmonary vasoconstriction.
pulmonary arterial smooth muscle cells; potassium channel; hypoxia; patch-clamp
technique
HYPOXIA
INHIBITS
K+ CURRENTS
Torr) gradually induced time-independent
outward currents without affecting Ca2+ currents (13). Hypoxia
(lowering PO, from 650-700 to 9-10 Torr) caused coronary vasodilation, presumably by inhibiting
ATP-sensitive K+ channels as a result of decreased ATP production (10, 32).
In the studies described in this report, we directly
investigated
and compared the electrical responses of
outward K+ currents during normoxia and hypoxia in
primary cultured PA and MA smooth muscle cells. Preliminary findings have been communicated
(33).
MATERIALS
AND
METHODS
Cell Preparation
Patch Clamp
Membrane currents were recorded with an Axopatch-ID
(Axon Instruments, Foster City, CA) amplifier using the patchclamptechnique (15) in the whole cell configuration (seeRef. 34
for details). Step-pulseprotocols and data acquisition were performed by a digital interface (TL-1 DMA interface of Axon
Instruments) coupled to an IBM compatible computer (Cintronix AT 286, Cintronix, Annapolis, MD). Whole cell currents
were filtered at 2 kHz. Data analysis was routinely performed
using the pClamp program (Axon Instruments). All the experiments were performed at room temperature (22-24°C).
Reagents
and Solutions
A cover slip containing the cellswaspositioned in the recording chamber (~0.75 ml) and superfused at a rate of 0.61.2 ml/min. The standard extracellular (bath) physiological
salt solution (PSS) for recording outward potassium currents (lout) contained (in mM) 141 NaCl, 4.7 KCl, 1.8 CaCl,,
1.2 MgCl,, IO N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic
acid (HEPES), and 10 glucose,buffered to pH 7.4 with 5 M
NaOH. When the extracellular Ca2+ was varied between 0
(Ca2+-freePSS) and 1.8 mM, equimolarconcentrations of Mg2+
were adjustedto compensatefor surface chargedifferencesand
osmolarity. In Ca2+-free PSS, 0.1-2 mM ethyleneglycol-his-(Paminoethyl ether)-N,N,N’,N’-tetraacetic acid (EGTA) was
added. The internal (pipette) solution used for recording I;,ut
contained (mM) 125 KCl, 4 MgCl,, 10 HEPES, IO EGTA, 5
Na,ATP, buffered to pH 7.2-7.3 with 1 M KOH. In someexperiments (seeRESULTS)
the EGTA concentration wasreduced
to 0.1 mM.
Oxygen Tension
Normoxic conditions were establishedby bubbling the superfusion PSS with room air to achieve a PO, = 155Torr (at 24°C).
Hypoxic conditions were establishedby dissolving 0.8-l mM
sodiumdithionite (Na,S,O,, Sigma), an oxygen scavengerthat
combines with oxygen (28), directly in extracellular PSS to
achievea PO, 5 74 Torr. An oxygen electrode(Microelectrodes,
Londonderry, NH) was positioned in the cell chamber on the
microscopestageto monitor the PO, of the extracellular PSS
continuously. Hypoxia was defined as PO, 5 74 Torr. Sodium
dithionite had no effect unlessaccompaniedby a reduction in
0, tension (25; alsoseeRESULTS).
Statis tics
The compositedata are expressedasmeanszf~SE. Statistical
analysiswasperformed using the paired and unpaired Student’s
PULMONARY
ARTERY
CELLS
Ll17
t test and one-way analysis of variance (ANOVA). Differences
were consideredto be significant when P < 0.05.
RESULTS
Effects of Hypoxia on Iout in Primary Cultured PA
and MA Smooth Muscle Cells
As described in the accompanying article (34), patchclamp studies of primary cultured PA and MA cells
bathed in standard PSS revealed that these cells possess
both voltage-gated inward Ca2+ current (I;n) and outward
K+ current (&). The Iout consists of a rapidly inactivating (transient) component
(&J, a slowly inactivating
component &), and a steady-state (noninactivating)
component (I,,). Little of the Iout could be attributed to
the Ca2+-activated
K+ (Kc,) component,
because 1)
patch pipette solutions were Ca2+-free and contained 10
mM EGTA and 2) Iout was virtually unaffected by removal of extracellular Ca2+ (34; and see below). When the
EGTA concentration in the pipette solution was reduced
to 0.1 mM, however, an external Ca2+-dependent component of Iout was uncovered.
A reduction of PO, from 140 to 52 Torr reversibly
decreased Iout in PA cells (Fig. lA), but a similar reduction, from 155 to 15 Torr, did not affect Iout in MA cells
(Fig. 1B). Both the inactivating and steady-state components of Iout in the PA cells were affected. The fact that
the MA cells were not affected by the Na+ dithionite
in
the hypoxic superfusion solution indicates that this O2
scavenger did not, itself, have a deleterious effect on cultured arterial cells. Furthermore, when PA cells were superfused with normoxic solution (vigorously bubbled with
room air) containing 1 mM Na+ dithionite,
no reduction
was
observed
(Fig.
1C).
These
findings
demonin Lt
strate that the hypoxia, and not the Na+ dithionite, per
se, caused Iout to decline in PA but not MA cells.
The current-voltage
(I-V) curves in Fig. 2 show the
marked inhibitory effect of hypoxia on Iss in PA myocytes
but not in MA myocytes. Hypoxia significantly reduced
the average slope conductance calculated from the I-V
curves in PA cells from 12.2 t 2.6 nS to 6.2 t 1.8 nS (P
< 0.001, n = 22), but did not significantly affect the slope
conductance in MA cells (from 6.3 t 1.3 nS to 5.8 t 1.2
nS, P = 0.82, n = 5).
Hypoxia also reduced & in cultured PA cells. This is
apparent in Fig. lA, where the initial peak of Iout (34) can
be seen to decline markedly during the period of hypoxia.
In contrast, although the early peak of Iout was more
prominent in the MA cell record illustrated in Fig. lB,
neither Irt nor I,, were affected by the reduction of PO, to
15 Torr.
As shown in Fig. 3A, a maximal effect of hypoxia on I,,,
about a 65% reduction, was observed at a PO, of 74 Torr
(the highest PO, of the hypoxic challenges examined) in
PA cells; further reduction of PO, to 15 Torr had no
additional effect. In contrast, there was no effect of hypoxia on I,, in MA cells in this range of PO,. The implication is that PA cells, but not MA cells, are sensitive to
very modest reductions in PO,.
The effects of hypoxia on Iss, in 30 primary cultured PA
cells and 18 primary cultured MA cells, are summarized in
Fig. 3B. When the PO, in the extracellular superfusate
Downloaded from http://ajplung.physiology.org/ by 10.220.32.247 on June 14, 2017
The method used for isolation and primary culture of rat
pulmonary and mesentericarterial smoothmusclecellshasbeen
described(34).
IN
L118
A
HYPOXIA
Control
INHIBITS
K+ CURRENTS
Hypoxia
IN PULMONARY
Torr)
(PO,=52
Torr)
(PO,=155
Torr)
(PO,=15
Torr)
(PO,=140
B
Torr)
500
pA
ms
Na Dithionite
Torr)
(PO,=162
-70
Torr)
100
ms
Fig. 1. Effects of hypoxia
on inward and outward
currents
in primary
cultured
rat pulmonary
artery
(PA) and mesenteric artery
(MA)
cells. Representative
families
of superimposed
current
records were elicited
by depolarizing
cells to a series of test potentials
between -20 and +80 mV
for a PA cell (A) and a MA cell (B) during normoxia
and
hypoxia.
C: effects of Na+ dithionite
without
accompanying hypoxia
on outward
K+ currents
in a PA cell. Superimposed current
traces in C were elicited by depolarizing
the cell to 20, 40, and 60 mV. The holding potential
in all
cases was -70 mV. PO, during
normoxic
and hypoxic
conditions
are shown in parentheses.
Leakage
currents
were subtracted.
Recovery
(PO,=162
mVRY” mV
Torr)
200
was reduced from 140 to 44 Torr, on the average, I,, in PA
cells was decreased by -60%, but I,, in MA cells was not
significantly affected.
The time course of the responses to hypoxia in single
PA and MA cells further illustrates the difference between these two types of cells (Fig. 4). In the PA cell (Fig.
4A), changes in PO, preceded concomitant changes in I,,.
In contrast, comparable changes in PO, in the MA cell
had no significant effect on Iss (Fig. 4B).
Hypoxia-induced
attenuation
of IOut in PA cells was
accompanied by an apparent increase in lin (Figs. 1A and
24). It seems likely, however, that Iin and IOut flow concurrently during the depolarizing pulses. Therefore the
peak amplitudes of both Iin and Irt underestimate, respectively, the voltage-gated Ca2+ and early K+ currents.
Thus much (if not all) of the apparent hypoxia-induced
increase in Iin may simply be due to the inhibition
of Irt.
Effects of Hypoxia on the Membrane Potential
in PA cells
Figure 5 shows that hypoxia also significantly and reversibly depolarized primary cultured PA cells when the
cells were superfused with either standard (Ca2+-containing) PSS (Fig. 5, A, B, and D) or with Ca2+-free PSS
containing 0.1 mM EGTA (Fig. 5, C and 0). Hypoxia also
increased spontaneous electrical activity in some cells
pA
(Fig. 5A shows an example). Hypoxia did not significantly affect these parameters in MA cells (data not
shown).
This depolarization
likely results from a decrease in
resting K+ conductance that shifts the membrane potential away from the potassium equilibrium
potential. The
augmented rate of spontaneous electrical activity associated with the hypoxia-induced
depolarization
(Fig. 5A) is
probably due to the markedly increased opening of voltage-gated Ca2+ channels.
Hypoxia-Induced Inhibition of Outward K+ Currents
in PA Cells is Independent of External Ca2+
The effects of removing extracellular Ca2+ and of buffering intracellular
Ca2+ on the hypoxia-induced
inhibition of IOut were tested on PA cells bathed in Ca2+-free
PSS containing 2 mM EGTA (Fig. 6). Note that Iin,
which appears to be carried by Ca2+ (34), was completely
eliminated under these conditions (Fig. 6A), whereas IOUt
at large depolarizations
was comparable to that in cells
bathed in standard PSS (compare Fig. 6 with Figs. lA,
lC, and 3B). Furthermore,
the patch pipette solution
contained 10 mM EGTA; this should have been sufficient
to buffer most of the intracellular [ Ca2+], especially in the
absence of Ca2+ entry. Nevertheless, reduction of PO,
from 155 to 15 Torr still significantly and reversibly de-
Downloaded from http://ajplung.physiology.org/ by 10.220.32.247 on June 14, 2017
100
C
Control
CELLS
Recovery
(PO,=140
(PO,=162
ARTERY
HYPOXIA
INHIBITS
K+ CURRENTS
IN
PULMONARY
ARTERY
Lll9
CELLS
(PA)
1200
-(T
822
Control
25
I
t
:
I
160
-40
-60
600
I
I
120
100
’
I
80
’
I
60
’
I
’
40
I
20
’
I
0
I
0
0
Normoxia
Hypoxia
T
T
f
I
f
-40
20
-200
i
I
40
I
I
hv)
60
-L
Fig. 2. Composite
steady-state
current-voltage
relationships
(I- V
curves) from 22 PA cells (A) and 5 MA cells (B) during normoxia
and
hypoxia.
PO, during normoxia
and hypoxia
were, respectively,
143 k 3
Torr and 41 t 6 Torr for PA cells and 145 t 6 Torr and 35 t 11 Torr
for MA cells. Duration
of test pulse was 300 ms; steady-state
(noninactivating)
current
(&J was measured
at 250-290
ms. Data are means t
SE. I- V curve for PA cells during hypoxia
is significantly
different
from
I-V curve during normoxia
(P < 0.001; ANOVA).
creased both the Irt and I,, components of IOUt (Fig. 6).
The composite I- V curves obtained from seven PA cells
bathed in Ca 2+- free PSS (with 2 mM EGTA present)
during normoxia and hypoxia are shown in Fig. 6B. The
hypoxia-sensitive
component of -I,,, indicated by the solid
triangles, was activated at about -52 mV (as shown by
the extrapolated broken line). This potent ial is more negative than the resting membrane potential (- ~41 mV) that
we measured in these PA cells. Thus these data support
the hypothesis that the hypoxia-induced
depolarization
is due to the inhibition
of these K+ channels.
Data from 12 PA cells incubated in Ca2+-free PSS are
summarized in Fig. 6C. The fact that hypoxia significantly inhibited I,, in the absence of extracellular (and
intracellular)
Ca2+ suggests that this hypoxia-sensitive
IOut was not dependent on Ca2+ entry or mediated by Kc,
channels. The average fractional inhibition
of I,, by hypoxia in the absence of external Ca2+ (-30%; Fig. 6C)
appeared to be only half as great as in the presence of
external Ca2+ (-60%; Fig. 3B), some of this difference in
the hypoxia-induced
inhibition
of IOut may have been due
to a Ca2+-dependent component. In the presence of external Ca2+ however selective inhibition
of outward K+
current will enhance the relative contribution
to the net
current that may be carried by influx of Ca2+ (Figs. IA
and Fig. 2A, middle). Thus the observed fractional reduction of Ca2+ -independent
IOut by hypoxia should be
greater under conditions in which some inward Ca2+
MA Cells
(n=18)
PA Cells
(n=30)
Fig. 3. Effects of hypoxia
on steady-state
outward
K+ current
(I,,) of PA
and MA cells. A: effects of hypoxia
on I,, in PA (0) and MA (0) are
plotted
as function
of 0, tension
(POT = 15-74 Torr).
Data points
correspond
to means f: SE of I,, during hypoxia,
relative
to I,, during
normoxia;
number above or below each symbol indicates
number of cells
tested. I,, was elicited by depolarizing
cells from a holding potential
of
-70 mV to a test potential
of +60 mV for 300 ms (see Fig. 2 legend). B:
data in A show that effects of hypoxia
were constant for PO, between 74
and 15 Torr;
therefore,
data for 30 PA cells and 18 MA cells were
averaged.
Bars in B indicate
means t SE of I,, for PA and MA cells
during normoxia
(PO, was 142 t 2 Torr for PA cells and 145 k 2 Torr
for MA cells) and during hypoxia
(POT was 43 k 5 Torr for PA cells and
47 t 5 Torr for MA cells). *** P < 0.001, hypoxia
vs. normoxia.
current is flowing (as in Fig. 3) than when it is not (as in
Fig. 6B).
DISCUSSION
The present study was based on three key earlier observations concerning hypoxia-induced
vasoconstriction.
1) Contraction of arterial smooth muscle, including PA
and MA, is normally triggered by an increase in intracellular free [Ca2+] (12, 18). The hypoxia-induced
increase
in resting PA tension is dependent on extracellular Ca2+
and can be prevented by Ca2+ channel blockers (16, 35).
2) Hypoxia-induced
pulmonary vasoconstriction
is not
dependent on the endothelium.
At least part of the response is mediated directly by the VSM cells (3, 25, 35).
3) Hypoxia-induced
contractions can be demonstrated in
cultured PA cells (19). Thus we hypothesized (35) that
HPV is an intrinsic mechanism of PA myocytes that is
related to hypoxia-induced
depolarization
(16,3 1) resulting from either a decrease in K+ conductance or an increase in Ca2+ conductance. We sought to test this hypothesis directly by investigating the inward and outward
Downloaded from http://ajplung.physiology.org/ by 10.220.32.247 on June 14, 2017
(PA)
800
t
I
140
PO, (Torr)
B
-60
1
:
L120
HYPOXIA
INHIBITS
K+ CURRENTS
IN PULMONARY
ARTERY
CELLS
A
180
Hypoxia
-
r
-lo
1
-25
-
1.8
mM
Ca2+
500
c
vE -30
E
w -35
dW
/(
-
4 min
-
100
o’,
,
0
,
12
w,
,
,
,
,
,
3
4
5
6
7
8
1
1 min
t
LO
Normoxia
B
-25
1
mM
Ca
2+
i
1 min
30 0 j
Hypoxia
,
,
0
12
,
‘b----d
,
3
Time
,
,
,
,
,
,
4
5
(min)
6
7
8
9
LO
Normoxia
C -30 -
Fig. 4. Time course of hypoxia-induced responses of steady-state outward K+ current (I,,, open circles) elicited by repeated test pulses of +60
mV from a holding potential of -70 mV in a PA cell (A) and a MA cell
(B). Simultaneously measured PO, is indicated by closed circles, time 0
corresponds to normoxic condition. Hypoxia bars indicate period during
which cells were superfused with physiological salt solution containing
1 mM sodium dithionite.
ionic currents and membrane potential during normoxia
and hypoxia in primary cultured PA and MA smooth
muscle cells.
0 mM
Ca
2
-50
2+
Hypoxia
I
1
1 min
Normoxia
Control
(11)
D
0
-
(18)
HYPox
(11)(18)
Recovery
l-
sE
v-10
..-I
Effects of Hypoxia on Iout and E,
in Primary Cultured PA and MA Smooth Muscle Cells
The resting membrane potential of most cells including
smooth muscle is dominated by the relative K+ permeability and the transmembrane
K+ distribution;
other
ions (e.g., Na+ and Cl-) generally make a much smaller
contribution
to the resting membrane potential (9, 21).
Thus hypoxia-induced
inhibition
of voltage-gated K+
channels would lead to a gradual enhancement of the
resting tension as we observed in isolated and endothelium-denuded
PA rings (35). In other studies (3, 26)
where the PA was partially depolarized or precontracted
by vasoconstrictors or KCl, hypoxia could be expected to
augment the rate and strength of PA contraction as a
consequence of 1) the inhibition
of voltage-gated K+
channels that were relatively more activated, and 2) the
further increase in the availability
of Ca2+ secondary to
depolarization. Our hypothesis is also consistent with the
report that some K+ channel blockers (tetraethylammonium and 4-aminopyridine,
but not glibenclamide)
cause
vasoconstriction
in the normoxic lung (17); this finding
suggests that a background K+ conductance helps maintain the low vascular resistance of the normoxic lung.
In our experiments, hypoxia (Po2 5 74 Torr) significantly attenuated both of the voltage-gated components
& and Iss) of the outward K+ current in PA but not MA
cells (Figs. l-4 and 6). The selective effect on the PA cells
implies that the underlying mechanisms that mediate
this hypoxic response are prevalent in PA but not MA
cells; these differences are maintained when the cells are
cultured. The implication
is that smooth muscle cells in
-20
3
4
2
e,
-30
2
2
E
-40
Gz
2
-50
~
Fig. 5. Effects of hypoxia on membrane potential (Em) measured with
current-clamp (I=O) in primary cultured PA cells. A: Em determined in
cell bathed in standard PSS (with 1.8 M Ca2+ present) during normoxia
was about -38 mV. Cell depolarized reversibly during hypoxia (Po2 =
15 Torr). This was accompanied by an increase in spontaneous electrical activity. B: another example of depolarization in response to hypoxia
(PO 2 = 15 Torr) in a PA cell bathed in standard PSS. Em during
normoxia was -45 mV initially and returned to about -43 mV following
period of hypoxia. C: Em was initially -42.5 mV during normoxia in a
cell bathed in Ca 2+-free PSS (with 0.1 mM EGTA present); cell depolarized, reversibly, to about -36 mV during hypoxia (Po2 = 8 Torr). D:
means t SE membrane potentials for cells bathed in standard PSS
(open bars) or in Ca-free PSS (hatched bars). Data were obtained from
11 PA cells bathed in standard PSS during normoxia (Po2 = 155 t 1
Torr), hypoxia (PO, = 17 t 3 Torr), and normoxic recovery (Po2 = 151
t 2 Torr), and from I8 PA cells bathed in Ca2+-free PSS during normoxia (PO, = 145 t 2 Torr), hypoxia (PO, = 16 ,t 4 Torr), and normoxie recovery (Po2
= 142 t 2 Torr). *** P < 0.001, ** P < 0.01,
hypoxia vs. control and vs. recovery.
different vascular beds may have some fundamentally different properties consistent with their specific functions.
The membrane input resistance of resting VSM cells is
high, on the order of l-10 GQ (8, 21, 34). Thus even a
small decrease of outward current through hypoxia-sensitive K+ channels would be expected to depolarize the
cells, as observed (Fig. 5; and see Refs. 15, 29). These
hypoxia-sensitive
K+ channels activate at a potential of
about -52 mV (Fig. 6B), which is more negative than the
Downloaded from http://ajplung.physiology.org/ by 10.220.32.247 on June 14, 2017
2 -35
.5
wg-40
J
1.8
-i
-30
60
I
9
B
’
,
Hypoxia
HYPOXIA
Control
(PO,=155
INHIBITS
K+ CURRENTS
Hypoxia
Torr)
(PO,=
15
IN
PULMONARY
ARTERY
Recovery
Torr)
(PO,=155
Torr)
1000 pA
-70
B
mV
(PA)
Hypoxia
Difference
__--I
8
-40
i
20
40
60
80
-500
Control
resting potential of our PA cells (about -40 mV). This
hypoxia-induced depolarization would open voltage-gated
Ca2+ channels and increase Ca2+ influx. The concomitant
increase in spontaneous electrical activity that we sometimes observed (Fig. 5A) was probably due to the opening
of increasing numbers of voltage-gated Ca2+ channels
that were activated when the PA cells were depolarized.
Is Hypoxia-Induced Inhibition
of Iout Dependent on Ca2+ ?
Recently, Post et al. (23) demonstrated that hypoxia
(PO = 40 Torr) inhibits K+ currents in freshly dispersed
can&e PA cells. In their study, the hypoxia-induced
attenuation of K+ currents was substantially
enhanced
when the pipette solution contained 0.1 mM EGTA instead of 5 mM EGTA and was completely prevented by
either addition of nisoldipine (a Ca2+ channel blocker) to
the bath solution, or replacement of EGTA by 10 mM
1,2-bis(2-aminophenoxy)ethane-N,N,N’,N’-tetraacetic
acid (BAPTA,
another Ca2+ chelator) in the pipette
solution. They therefore concluded that hypoxia-induced
inhibition
of Iout was due mainly to block of Kc,
channels.
In contrast, in our primary cultured rat PA cells, IOut
was still markedly and reversibly inhibited by a reduced
PO, when the PA cells were bathed in Ca2+-free PSS
containing 2 mM EGTA, whereas the Ca2+-free pipette
solution contained 10 mM EGTA (Fig. 6). Clearly, a large
fraction of this hypoxia-induced
decrease of IOut was not
dependent on extracellular or intracellular Ca2+ and could
not be attributed to block of Kc, channels. Furthermore,
the fact that hypoxia caused PA cells to depolarize even
in the absence of external Ca2+, and with 10 mM EGTA
in the pipette solution, is further evidence that these
Hypoxia
ms
L121
Fig. 6. Effect of hypoxia
on outward
current (&)
in PA cells in absence of extracellular Ca 2+. Ca2+-free
PSS in all of these
experiments
contained
2 mM EGTA.
A:
representative
current
recordings
elicited
by depolarizing
PA cell to a family of test
potentials
from -20 to +80 mV during
normoxia,
hypoxia,
and normoxic
recovery. Note that inward
current
(lin) was
eliminated
in absence of external
Ca2+.
Linear leakage currents
were subtracted
as
described
(34). B: composite
I- V curves of
steady-state
&
(Is& obtained
from 7 PA
cells, bathed in Ca 2+-free PSS during normoxia
and hypoxia.
Hypoxia-sensitive
component
of I,, (difference
= I,, during
normoxia
minus Iss during hypoxia)
is also
shown. Broken
line shows linear extrapolation of hypoxia-sensitive
Iss from 0 mV
to abscissa intercept
(-52 mV). I- V curve
during
hypoxia
is significantly
different
from I- V curve during normoxia
(P < 0.01,
ANOVA).
C: mean I,, evoked by repeated
test pulses of +60 mV from a holding potential of -70 mV during normoxia
(Po2 =
147 t 2 Torr),
hypoxia
(POT = 42 t 9
Torr),
and normoxic
recovery
(Po2 = 145
& 3 Torr).
Data are means t SE of I,,
values from 12 primary
cultured
PA cells.
** P < 0.01, hypoxia
vs. normoxia;
# P <
0.05, recovery
vs. hypoxia.
Recovery
hypoxic responses are Ca2+ independent. Therefore, our
findings suggest that voltage-gated K+ channels play an
important role in the hypoxia-induced
inhibition
of IOut
and membrane depolarization
in PA cells. Moreover,
using the same primary cultured rat PA cells, Salvaterra
and Goldman
(27) report that hypoxia
(Po2 =
22-29 Torr) causes marked elevation in cytosolic Ca2+
( [Ca2+],&
within 30 s; during maintained
hypoxia,
increases
uniformly
throughout
responsive
cells
[Ca2+lcyt
until maximal levels are reached at 2-3 min. Such a rise
in [Ca2+1,,t,as a result of membrane depolarization and
the opening of voltage-gated Ca2+ channels (and perhaps
release of Ca 2+ from internal stores), could explain the
HPV response. A rise in [Ca”+],, should, however, activate Kc, channels and hyperpolarize
the membrane.
Blocking these Kc, channels (23) might then augment
and prolong the contractions by inhibiting repolarization
(6), but we would not expect block of the Kc, channels to
initiate the HPV response.
Do ATP-Sensitive K+ Channels Play a Role
in Hypoxia-Inhibited
IOUt?
K+ channels that are inhibited by intracellular
ATP
(KATP channels) have been identified in VSM cells (30).
Activation of K ATp channels in certain systemic arterial
smooth muscle cells may be important in the vasodilator
response to hypoxia or ischemia (10, 11, 22). Daut and
colleagues (10, 32) found that hypoxia-induced
coronary
vasodilation in guinea pig hearts was mimicked by cromakalim, a K ATp channel activator and inhibited by glibenclamide a K ATp channel blocker. They proposed that
activation of K ATp channels, as a result of decreased ATP
production during hypoxia, hyperpolarizes the VSM cells
and thereby relaxes coronary arteries.
Downloaded from http://ajplung.physiology.org/ by 10.220.32.247 on June 14, 2017
Normoxia
-60
L
50
CELLS
L122
HYPOXIA
INHIBITS
K+ CURRENTS
Summary and Conclusions: Implications
for
the Mechanism of Hypoxic Pulmonary Vasoconstriction
Our observations confirm that aspects of the hypoxiainduced responses in the pulmonary vasculature can profitably be studied in cultured pulmonary arterial smooth
muscle cells (20). Moreover, our results demonstrate that
the electrophysiological
responses of PA and MA cells to
hypoxia are different and are consistent with the different contractile responses of pulmonary and mesenteric
arteries to hypoxia (35).
As discussed in the preceding sections, initiation
of
HPV cannot be explained by the block of either Kc, or
KATP channels in PA cells. Instead, our results provide
ARTERY
CELLS
direct evidence that hypoxia selectively blocks voltagegated K+ channels in PA but not MA cells. These voltagegated K+ channels are activated at potentials more negative than the resting membrane potential, and it appears
likely that some of these channels may be open at rest and
may therefore contribute to the resting membrane conductance and resting potential. Block of these hypoxiasensitive, voltage-gated K+ channels would be expected to
depolarize the PA cells, as we observed, and secondarily
open depolarization-activated
Ca+ channels, thereby
raising [ Ca2+] cyt and promoting
contraction.
Whether
these hypoxia-sensitive,
voltage-gated K+ channels are
affected directly by the reduced 0, level (14) or by free
radicals or another second messenger that is generated as
a result of the hypoxia (1) remains to be determined.
We gratefully
acknowledge
Drs. M. T. Nelson, D. R. Matteson
and B.
K. Krueger
for their assistance
with patch-clamp
techniques.
We are
also indebted
to Drs. E.M. Santiago
and Huang Yu for their advice on
cell isolation
and culture.
This research was supported
by a Research
Fellowship
from American Heart Association-Maryland
Affiliate
(X.-J. Yuan),
by National
Heart, Lung, and Blood Institute
Grants HL-32276
(M. P. Blaustein),
HL-43091
(W. F. Goldman),
and HL-43304
(M. L. Tod), and by a
research
grant from the Veterans
Affairs
Research
Services
(L. J.
Rubin).
M. L. Tod is an Established
Investigator
of the American
Heart
Association.
Address for reprint
request: M. P. Blaustein,
Dept. of Physiology,
University
of Maryland
School of Medicine,
655 W. Baltimore
St.,
Baltimore,
MD 21201
Received
9 July
1992; accepted
in final
form
6 October
1992.
REFERENCES
1. Archer,
S. L., J. A. Will,
and E. K. Weir.
Redox status in the
control
of pulmonary
vascular
tone. Herz 11: 127-141,
1986.
2. Ashford,
M. L. J. ATP-regulated
K+ channels
in rat hypothalamic neurones
(Abstract).
J. Physiol.
Lond. In press.
3. Bennie,
R. E., C. S. Packer,
D. R. Powell,
N. Jin, and R. A.
Rhoades.
Biphasic
contractile
response
of pulmonary
artery
to
hypoxia.
Am. J. Physiol.
261 (Lung Cell. Mol. Physiol.
5): Ll56L163, 1991.
4. Bergofsky,
E. H., and S. Holtzman.
A study of the mechanisms involved
in the pulmonary
arterial
pressor response to hypoxia. Circ. Res. 20: 506-519,
1967.
5. Bonnet,
P., D. Gebremedhin,
N. J. Rush,
and D. R.
Harder.
Effects of hypoxia
on a potassium
channel in cat cerebral
arterial
muscle cells. Kurdiol.
80 Suppl. 7: 25-27, 1991.
6. Brayden,
J. E., and M. T. Nelson.
Regulation
of arterial
tone
by activation
of calcium-dependent
potassium
channels.
Science
Wash. DC 256: 532-535,
1992.
7. Buescher,
P. C., D. B. Pearse,
R. P. Pillai,
M. C. Litt,
M. C.
Mitchell,
and J. T. Sylvester.
Energy state and vasomotor
tone
in hypoxic
pig lungs. J. Appl. PhysioZ. 70: 1874-1881,
1991.
8. Clapp,
L. H., and A. M. Gurney.
ATP-sensitive
K+ channels
regulate
resting potential
of pulmonary
arterial
smooth
muscle
cells. Am. J. Physiol.
262 (Heart
Circ. Physiol.
31): H916-H920,
1992.
9. Cox,
R. H. Potassium
channel
activators
in vascular
smooth
muscle. In: Cellular
and Molecular
Mechanisms
in Hypertension,
edited by R. H. Cox. New York: Plenum,
1991, p. 27-43.
10. Daut,
J.,
W. Maier-Rudolph,
N. von
Beckerath,
G.
Mehrke,
K. Guntherk,
and L. Goedel-Meinen.
Hypoxic
dilation
of coronary
arteries
is mediated
by ATP-sensitive
potassium channels.
Science Wash. DC 247: 1341-1344,
1990.
11. Davies,
N. W., N. B. Standen,
and P. R. Stanfield.
ATPdependent
potassium
channels
of muscle cells: their properties,
regulation,
and possible function.
J. Bioenerg.
Biomembr.
23: 509535, 1991.
12. Filo, R. S., D. F. Bohr,
and J. C. Ruegg.
Glycerinated
skeletal
and smooth muscle: calcium and magnesium
dependence.
Science
Wash. DC 147: 1581-1583,
1972.
Downloaded from http://ajplung.physiology.org/ by 10.220.32.247 on June 14, 2017
In the pulmonary circulation, however, hypoxia causes
vasoconstriction.
Furthermore,
we found that cromakalim has no effect on isolated rat PA rings during normoxia, but inhibits HPV in these rings; glibenclamide
also has no effect during normoxia, but reverses the inhibitory effect of cromakalim on HPV (35). Because cromakalim opens glibenclamide-sensitive
KATP channels,
we must assume either 1) that the VSM cell membranes
were sufficiently polarized during normoxia so that opening of these channels had no effect on membrane potential, and therefore did not influence [Ca”+].,, and contraction in the PA cells, or 2) that the ATP levels were
sufficiently low during normoxia so that the KATP channels were already opened (which seems unlikely). If the
latter possibility were true, to explain our observations
(35) we would then need to postulate that hypoxia paradoxically increases cytosolic ATP to a level sufficient to
close the K ATp channels in order to depolarize pulmonary
arterial smooth muscle and induce contraction. However,
preliminary data indicate that glibenclamide has no effect
on the resting membrane potential, nor does it prevent
hypoxia-induced
depolarization,
in primary cultured PA
cells (Yuan et al., unpublished observations).
The relationship
between hypoxia and intracellular
ATP levels in PA cells has been controversial. Buescher
and colleagues (7) have reported that the ATP concentration in isolated, degassed pig lung is unchanged or even
(paradoxically) increased during hypoxia and the induction of HPV. However, several other workers (1,22) have
shown that HPV is initiated or augmented when the ATP
level {and the phosphate potential, i.e., the ratio, [ATP/
(ADP+Pi)]}
is reduced as a consequence of hypoxia and,
in some studies, subsequent inhibition
of glycolysis (29).
The weight of evidence strongly suggests that hypoxia
does not raise ATP levels and, therefore, that block of
KATP channels is not involved in the initiation
of the
HPV response. However, it is clear that KATP channels
exist in rat PA smooth muscle and potentially could have
a modulatory effect on membrane potential and contraction in response to hypoxia. Recently, K+ channels that,
in contrast to the aforementioned
“classic” KATP channels, are activated by intracellular
ATP have been observed in rat hypothalamic
neurons (2) and PA smooth
muscle cells (24). Indeed, Robertson et al. (24) have suggested that hypoxia may depolarize PA ceils and promote
vasoconstriction
by reducing the activity of these ATPactivated K+ channels.
IN PULMONARY
HYPOXIA
13. Friedrich,
M.,
K. Benndorf,
INHIBITS
M.
K+ CURRENTS
Schwalb,
and
IN PULMONARY
H. Hirche.
0. P., A. Marty,
E. Neher,
B. Sakmann,
and
F. J.
Improved patch-clamp techniques for high-resolution
recording from cells and cell-free membrane patches.
current
Pfluegers
16. Harder,
Arch. 391: 85-100,
1981.
D. R., J. A. Madden,
and
C. Dawson.
Hypoxic induction of Ca2+-dependent action potentials in small pulmonary
arteries of the cat. J. Appl. Physiol. 59: 1389-1393,
1985.
17. Hasunuma,
K., D. M. Rodman,
and I. F. McMurtry.
Effects
of K+ channel blockers on vascular tone in the perfused rat lung.
Am. Rev. Respir.
18. Kamm,
K. E.,
Dis.
and
144: 884-887,
J. T. Stull.
1991.
B. E. Marshall,
and
Lopez-
M.
T.,
J.
B.
Patlak,
J.
F.
Worley,
and
N.
B.
Calcium channels, potassium channels, and voltage dependence of arterial smooth muscle tone. Am. J. Physiol. 259 (Cell
Physiol.
28): C3-C18,
22. Ohe, M., T. Mimata,
1990.
T. Haneda,
and T. Takishima.
Time
course of pulmonary vasoconstriction with repeated hypoxia and
glucose depletion. Respir. Physiol. 63: 177-186, 1986.
23. Post,
J.
M.,
J.
R.
Hume,
S. L.
Archer,
and
E. K.
Weir.
Direct role for potassium channel inhibition in hypoxic pulmonary vasoconstriction. Am. J. Physiol.
262 (CeZZ Physiol.
31):
C882-C890,
24. Robertson,
Kozloswski.
1992.
B. E.,
P. R.
Corry,
P. C. G. Nye,
D. M.,
T. Yamaguchi,
K.
and I. F. McMurtry.
Hypoxic
rat pulmonary artery. J. PharmacoZ.
Exp.
1988.
27. Salvaterra,
C. G. and W. F. Goldman.
and
R.
Z.
Ca2+ and Mg-ATP activated potassium channels
from rat pulmonary artery. Pfluegers Arch. 421: 94-96, 1992.
Hasunuma,
R.
F.
contraction of isolated
Ther.
248:
952-959,
Acute hypoxia increases cytosolic calcium in cultured pulmonary arterial myocytes.
Am. J. Physiol.
264 (Lung Cell. Mol. Physiol.)
In press.
28. See, K. L., I. J. Forbes,
and W. H. Betts.
Oxygen dependency
of phototoxicity with hematoporphyrin derivative. Biochem. Photobiol. 39: 631-634,
1984.
29. Stanbrook,
H. S., and I. F. McMurtry.
potentiates hypoxic vasoconstriction
Inhibition of glycolysis
in rat lung. J. Appl. Physiol.
55: 1467-1473,
1983.
30. Standen,
N. B., J. M. Quayle,
N. W. Davies,
J. E. Brayden,
Y. Huang,
and M. T. Nelson.
Hyperpolarizing
vasodilators
Science
31. Suzuki,
K+ channels in arterial smooth muscle.
Wash. DC 245: 177-190,
1989.
H., and B. M. Twarog.
Membrane properties of
smooth muscle cells in pulmonary hypertensive rats. Am. J. Phys-
iol. 242 (Heart
32. Von Beckerath,
E. J. Macarak.
Hypoxic contraction of cultured pulmonary vascular smooth muscle cells. Am. J. Respir. Cell MoZ. Biol. 3: 457-465, 1990.
21. Nelson,
Standen.
26. Rodman,
O’Brien,
activate ATP-sensitive
J.
activity in
chemoreceptor cells of the mammalian carotid body. J. Gen. Phys-
ioZ. 93: 1001-1015,
1989.
20. Murray,
T. R., L. Chen,
D. M., and N. F. Voelkel.
Regulation of vascular
tone. In: The Lung, Scientific Foundations,
edited by R. G. Crystal, J. B. West, P. J. Barnes, N. S. Cherniack, and E. R. Weibel.
New York: Raven, 1991, p. 1105-1119.
Circ.
Physiol.
11): H907-H915,
N., S. Cyrys,
A. Dischner,
1982.
and
J. Daut.
Hypoxic vasodilation in isolated, perfused guinea-pig heart: an
analysis of the underlying mechanisms. J. Physiol. Lond. 442:
297-319,
1991.
33. Yuan,
X.-J.,
W. F.
and M. P. Blaustein.
Goldman,
M.
L.
Tod,
L.
J.
Rubin,
Contrasting effects of hypoxia on potassium currents in cultured pulmonary (PA) and mesenteric
(MA) artery cells (Abstract). Am. Rev. Respir. Dis. 145: A227,
1992.
34. Yuan,
X.-J.,
W. F. Goldman,
M. P. Blaustein.
Ionic currents
M.
L. Tod,
L. J. Rubin,
and
in rat pulmonary and mesenteric
arterial myocytes in primary culture and subculture. Am. J. Phys-
ioZ. 264 (Lung
35. Yuan,
X.-J.,
CeZl. MoZ. Physiol.
8): LlO7-Ll15,
M. L. Tod, L. J. Rubin,
and
M.
1993.
P. Blaustein.
Contrasting effects of hypoxia on tension in rat pulmonary and
mesenteric arteries. Am. J. Physiol. 259 (Heart Circ. Physiol. 28):
H281-H289,
1990.
Downloaded from http://ajplung.physiology.org/ by 10.220.32.247 on June 14, 2017
The function of myosin and
myosin light chain kinase phosphorylation
in smooth muscle.
Annu. Rev. Pharmacol.
Toxicol.
25: 593-620,
1989.
19. Lopez-Lopez,
J., C. Gonzalez,
J. Urena,
and
Barneo.
Low PO, selectively inhibits K channel
L123
CELLS
25. Rodman,
Effects of anoxia on K and Ca currents in isolated guinea pig
cardiocytes. Pfluegers Arch. 416: 207-209,
1990.
14. Ganfornina,
M. D., and J. Lopez-Barneo.
Single K+ channels in membrane patches of arterial chemoreceptor cells are modulated by O2 tension. Proc. N&Z. Acad. Sci. USA 88: 2927-2930,
1991.
15. Hamill,
Sigworth.
ARTERY