A-type currents are voltage-gated,
calcium-independent potassium (Kv) currents that undergo rapid
activation and inactivation. Commonly associated with neuronal and
cardiac cell-types, A-type currents have also been identified and
characterized in vascular, genitourinary, and gastrointestinal smooth
muscle cells. This review examines the molecular identity, biophysical
properties, pharmacology, regulation, and physiological function of
smooth muscle A-type currents. In general, this review is intended to facilitate the comparison of A-type currents present in different smooth muscles by providing a comprehensive report of the literature to
date. This approach should also aid in the identification of areas of
research requiring further attention.
 |
INTRODUCTION |
SMOOTH MUSCLE CONTRACTILE ACTIVITY
correlates well with the level of free intracellular Ca2+
available to the contractile apparatus (Ref. 24; see,
e.g., Refs. 13 and 52). Ca2+ may enter the
cytosol from the extracellular space or may be released from
intracellular stores (i.e., the sarcoplasmic reticulum; e.g., Ref.
99). Ca2+ influx through voltage-gated
Ca2+ channels is required for physiological patterns of
smooth muscle contraction (17, 21, 50, 64). Although
release of intracellular Ca2+ is necessary for effective
contraction, it is likely that voltage-dependent entry of
Ca2+ across the plasma membrane is responsible for
initiating contractile events (81, 115). Because
of this voltage dependence, membrane potential is the primary
determinant of smooth muscle tone.
K+ channels are important regulators of membrane potential.
Under physiological conditions, membrane potentials of smooth muscles lie positive to the reversal potential for K+
(EK 
80 to 90 mV). Consequently,
increasing membrane permeability to K+ results in outward
current that opposes depolarization and cellular excitability (i.e.,
contraction). K+ channels exhibit remarkable molecular and
functional diversity, and the electrical behavior of a given tissue is
highly influenced by the complement K+ currents present.
The broad diversity of electrical activities of smooth muscles largely
results from the great diversity of K+ channels expressed
in these tissues. Other reviews have examined the general topic of
K+ currents in smooth muscles (e.g., Refs. 33,
47, 73, 102, 116).
This review is intended to complement and expand on these publications
by focusing specifically on rapidly inactivating A-type K+
conductances that are present in various smooth muscles.
| |
GENERAL PROPERTIES OF A-TYPE CURRENTS |
Voltage-gated, Ca2+-independent K+ (Kv)
currents are present in all smooth muscles. By using time dependence as
a basis for classification, Kv currents may be divided into two
archetypical categories: slow "delayed rectifier" currents and
rapid "A-type" currents. Hodgkin and Huxley
(44) first used the term delayed rectifier to describe the
voltage-dependent K+ current that developed after the
Na+ current in response to depolarization of Loligo
giant axons. The term delayed rectifier persists and is generally
used to describe Kv currents with kinetics similar to the original
axonal K+ currents (i.e., delayed onset of activation
followed by little or slow inactivation).
A-type currents were first observed in molluscan neurons by Hagiwara
and coworkers (41) and later characterized by Connor and
Stevens (22), Neher (72), and Thompson
(106). The general properties of A-type currents
in neurons have been the subject of previous reviews (85,
87). A-type currents are voltage-gated, Ca2+-independent K+ currents that are
distinguished from typical delayed rectifier currents by rapid rates of
inactivation. Kinetically, A-type currents bear closer resemblance to
voltage-gated Na+ currents than to delayed rectifier
K+ currents (72). A-type currents activate at
negative membrane potentials, with measurable thresholds typically
between 45 and 60 mV. Strong steady-state, voltage-dependent
inactivation is a feature typical of A-type currents, and
repolarization to potentials negative to 50 mV is typically required
for restoration of channel availability (Fig.
1). Inhibition by 4-aminopyridine (4-AP)
and insensitivity to extracellular tetraethylammonium (TEA) ions are considered to be pharmacological hallmarks of A-type currents (e.g.,
Ref. 106), although notable exceptions exist (e.g., Refs. 88 and 91).
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 1.
General properties of A-type K+ currents. A:
whole cell A-type currents from holding potentials of 80
(a) and 40 mV (b) recorded from mouse antral
myocytes. B: steady-state inactivation shown as a plot of
normalized peak current (I/Imax) as a
function of conditioning potential ( ) and fit with a
Boltzmann function. For voltage dependence of activation, peak
K+ currents at test potentials between 80 and +40 mV were
converted into permeabilities using the Goldman-Hodgkin-Katz current
equation. Permeabilities were normalized
(P/Pmax), plotted as a function of
test potential ( ), and fit with a Boltzmann function.
Dashed lines mark the voltages of half activation and inactivation.
[Reprinted from Amberg et al. (2).]
|
|
| |
A-TYPE CURRENTS IN NON-SMOOTH MUSCLE TISSUES |
A-type currents have been examined extensively in neurons, and
these currents are thought to regulate firing frequency (14, 22,
63, 109). In many neurons, A-type channels are unavailable at
resting membrane potentials due to pronounced steady-state inactivation. A-type channels become available in these cells during
afterhyperpolarizations following action potentials, when the membrane
potential becomes sufficiently negative to remove inactivation
(22). A-type channels activate during the decay of the
afterhyperpolarization and tend to delay depolarization. In this
manner, A-type currents prolong the period between action potentials.
A-type currents are also present in atrial and ventricular myocytes and
in these cells are referred to as "transient" outward current
(Ito). In contrast to neuronal A-type currents,
Ito channels are available at resting membrane
potentials (Ref. 7; e.g., Refs. 9 and 74).
Ito is predominantly responsible for the initial
repolarization (phase 1) of the cardiac action potential. Pharmacological blockade of Ito with 4-AP causes
an increase in action potential amplitude and duration
(34) and increases force generation (94). The
distribution of Ito within the myocardium is
nonuniform and contributes toward regional variations in action potential waveforms (6, 15, 121).
| |
A-TYPE CURRENTS IN SMOOTH MUSCLE |
Voltage-dependent, transient outward K+ currents
have also been identified in smooth muscle cells (see Table
1). The smooth muscle literature is
confusing in that the term transient outward current has been used to
designate Ca2+-activated K+ currents (elicited
by step depolarizations that activate large, initial Ca2+
entry events), slowly inactivating, voltage-dependent K+
currents, and true A-type currents. In this review we will use the term
A-type current to designate rapidly inactivating, voltage-dependent K+ currents.
A-type currents have been identified in vascular smooth muscle cells of
the rabbit (portal vein, pulmonary artery, aorta), rat (pulmonary
artery, renal resistance artery), and human (mesenteric artery). See
Table 2 for appropriate references. In
contrast to the A-type currents of the myocardium (discussed elsewhere; e.g., Refs. 74 and 80), the physiological function of
A-type currents in vascular myocytes has not been fully clarified.
A-type currents have been identified in genitourinary (GU) smooth
muscle cells of the guinea pig (ureter, seminal vesicles, and vas
deferens), rabbit (vas deferens), rat (myometrium), and human
(myometrium). A-type currents are also present in gastrointestinal (GI)
smooth muscle cells of the mouse (fundus, antrum, jejunum, and colon), rat (ileum), guinea pig (colon), and opossum (esophagus), and a
"transient delayed rectifier" has also been characterized in the
human esophagus. As this review describes, the physiological function
of A-type currents in smooth muscles may be related to the maintenance
of membrane potential and regulation of excitability.
| |
MOLECULAR IDENTITY OF SMOOTH MUSCLE A-TYPE CONDUCTANCES |
K+ channel 
-subunits with A-type properties are
found in several K+ channel families including
Shaker (Kv1.4), Shaw (Kv3.4), and Shal
(Kv4.1, Kv4.2, and Kv4.3). Kv1.4 and Kv4 channels are responsible for
the typical A-type conductances susceptible to 4-AP in a wide range of
tissues, whereas Kv3.4 channels are mainly expressed in skeletal muscle
but also in the central nervous system. In central and peripheral
neurons, Kv1.4 and Kv4 channels distribute primarily to presynaptic and
postsynaptic membranes, respectively (93). Differences in
the local distribution of specific A-type channel subtypes has been
reported to be one of the major determinants of distinct functional
characteristics observed in various regions of the heart and central
nervous system.
Transcripts of a variety of pore-forming -subunits with A-type
properties (i.e., Kv1.4, Kv3.3, Kv3.4, Kv4.1, Kv4.2, and Kv4.3) have
been detected in rat mesenteric and tail arteries (122, 123). Kv 
-subunits, which confer A-type properties to some
delayed rectifier pore-forming subunits (e.g., Ref. 84),
have been found in these two tissues as well. Kv1.4, Kv4.1-4.3,
and Kv transcripts have been detected in rat pulmonary artery smooth
muscle (25, 104, 127), and transcripts encoding the long
isoform of Kv4.3 have been detected in rat aortic smooth muscle
(77). The distribution of A-type channel transcripts in
various smooth muscles is included in Table
3. Evidence regarding the molecular
composition of A-type currents in smooth muscles is, however,
incomplete, and experiments using antisense techniques or production of
A-type channel knockout animals have yet to be performed.
One of the distinguishing features of various A-type currents is the
time course of recovery from inactivation. For example, A-type currents
mediated by the Kv4 family of channels recover rapidly from
inactivation (cf. Refs. 92 and 124). The rapid recoveries
from inactivation of many GI A-type currents (time constants <100 ms;
Refs. 3, 57, 97, and 114)
are comparable to rates reported for cardiac A-type currents
(Ito), which are due to Kv4 channels in most
mammals (Refs. 7, 10, and 28; e.g.,
Ref. 74). For comparison, kinetic profiles of Kv4.2 and Kv4.3 are presented in Table 4. In human
mesenteric artery the reported recovery from inactivation is relatively
fast (time constant 250 ms; Ref. 95), suggesting that the
conductance responsible may be Kv4-derived. The rapid recovery from
inactivation of the A-type currents in guinea pig ureter (time constant
100 ms; Ref. 49) is also suggestive of Kv4 channel
involvement. In contrast to the currents described above, the slower
recovery from inactivation reported in rabbit portal vein
(12) and pulmonary artery (79) are
considerably slower (time constants 0.8s and >10s, respectively). These rates of recovery are more consistent with currents mediated by
Kv1 channels. It is important to note that kinetic features of A-type
K+ currents depend not only on the corresponding
-subunits but also on the presence of and interaction with accessory
-subunits (e.g., Ref. 84). Different heteromeric
subunit combinations likely underlie the diversity of A-type currents
present in smooth muscles (see Table 1).
View this table:
[in this window]
[in a new window]
|
Table 4.
Kinetic profile of Kv4.2 and Kv4.3 -subunits heterologously
expressed in Xenopus oocytes, CHO cells, and HEK cells
|
|
| |
KV4 CHANNELS AS A MOLECULAR COMPONENT OF A-TYPE CONDUCTANCES IN
SMOOTH MUSCLE |
A substantial line of evidence has accumulated suggesting that the
Kv4 family of K+ channels, particularly Kv4.2 and Kv4.3,
may be one of the major components of A-type conductances in many
smooth muscles. In GU smooth muscles, Kv4.3L (long isoform) transcripts
show greater abundance than those encoding Kv1.4, Kv3.3, Kv3.4, or
Kv4.2. A full-length clone of Kv4.3L, obtained by RT-PCR from rat vas
deferens smooth muscle, produced a 4-AP-sensitive A-type current when
heterologously expressed in HEK-293 cells (77). This study
also identified Kv4.3L transcripts in rat urinary bladder smooth
muscle. The most complete study addressing the molecular underpinnings
of a GU smooth muscle A-type current focused on rat myometrium
(100). Kv4 transcripts were detected in uterine smooth
muscle tissue. Real-time PCR showed that Kv4.3 transcripts were more
than twice as abundant as those encoding Kv4.1 and Kv4.2; Western blot
analysis confirmed Kv4.3 expression. Short Kv4.3 isoforms were not
detected in myometrial smooth muscle.
Smooth muscles from the rodent GI tract express the three known Kv4
isoforms. Transcripts encoding Kv4 channels have been detected in
isolated myocytes from the murine antrum, jejunum, and colon (2,
3, 57). Kv1.4 transcripts were not detected in colonic myocytes
(57). At the tissue level, real-time PCR demonstrated a
relative predominance of Kv4.3 transcripts over Kv4.1 and Kv4.2 in the
mouse colon (3). Although Kv4.3 transcripts may be
alternatively spliced (e.g., Ref. 78), only the long form
was detected in mouse GI smooth muscles. Kv4.3 transcripts were also
detected in rat stomach and colon (77). Myocyte-specific expression of Kv4 isoforms in the mouse GI tract has been confirmed with assays of protein expression, using commercially available Kv4.2-
and Kv4.3-specific antibodies (2, 3). The intensity of
Kv4-like immunoreactivity in myocytes in different regions of the mouse
GI tract mirrored that of respective A-type current densities (i.e.,
greater Kv4-like immunoreactivity in antrum and colon than in jejunum).
Immunohistochemical findings also corresponded well with the results
from quantitative PCR in that Kv4.3-like immunoreactivity was more
intense than Kv4.2-like immunoreactivity in mouse colonic myocytes.
The discovery of KChIPs (K+ channel-interacting proteins)
as -subunits specific for Kv4 channels (5) was a
pivotal step in understanding the mechanisms responsible for the
heterogeneity of Kv4-mediated A-type currents. KChIPs, which interact
with the NH2 terminus of Kv4 proteins (8),
enhance surface expression, thereby increasing A-type current densities
when coexpressed in heterologous systems and in native tissues
(5, 26, 43, 63). It has been shown that a defect in the
KChIP2 gene results in complete loss of Ito in
cardiac myocytes (58). In contrast to KChAP
(K+ channel-associated protein), a chaperone for Kv1.3,
Kv2.1, and Kv4.3 channels (61), KChIPs are specific for
the Kv4 family of K+ channels. KChIPs have high sequence
homology to neuronal Ca2+ sensor-1 (NCS-1; frequenin) and
calsenilin/DREAM, which belong to the superfamily of EF-hand-containing
proteins (16, 101). It has been demonstrated that
frequenin (NCS-1) is responsible for Ca2+-dependent
enhancement of Kv4 expression and modulation of kinetic behavior
(70). Quantitative analyses of transcriptional expression of neuronal Ca2+-binding proteins (NCBPs) indicated that
KChIP1 and KChIP3, as well as other NCBPs, are extensively
expressed in mouse GI smooth muscles (76).
In addition to enhancing surface expression, KChIPs also modulate Kv4
channel kinetic behavior (5, 8, 11). A variety of
modulation patterns have been observed depending on the specific combination of Kv4 and KChIP proteins. For Kv4.2, coexpression with
KChIP1 slowed the rate of inactivation, accelerated the time course of
recovery from inactivation, and shifted the voltage of
half-activation to more negative potentials (5, 8, 69). The effects of KChIP1 on Kv4.3L are similar to that of Kv4.2, although
there are quantitative differences between the heteromeric subunit
combinations (43). On the other hand, coexpression of KChIP1 with Kv4.1 accelerated inactivation and shifted the voltage of
half-activation in the positive direction (69). More
strikingly, coexpression of a novel KChIP4 isoform nearly abolished Kv4
channel inactivation (46). Thus the assembly of different
KChIP and Kv4 combinations likely contributes to the diversity and
complexity of native A-type currents in neuronal, cardiac, and smooth
muscle cells.
Observed levels of Kv4 transcripts were found to be the same in murine
colon and jejunum; however, the density of A-type current in colonic
myocytes was approximately twice that of jejunum, and Kv4-like
immunoreactivity was substantially greater in the colon (3). Studies were performed to determine whether
differences in A-type current density could be due to differential
expression of KChIPs (3). KChIP2 expression has recently
been shown to mirror the transmural gradient of
Ito in canine and human ventricles (86). In analogous fashion, KChIPs appear to mediate the
density of A-type current in mouse colonic and jejunal myocytes
(3). KChIP transcripts, which were detected in isolated
colonic and jejunal myocytes (3, 76), were approximately
twice as abundant in colon as in jejunum. Thus, in mouse colon and
jejunum, A-type current density parallels KChIP expression.
| |
PLASTICITY OF A-TYPE CURRENT EXPRESSION |
Electrical remodeling of myometrial smooth muscle results from
pregnancy-associated changing hormone levels (e.g., Refs.
107 and 111). In most species, including humans, plasma
levels of 17-estradiol and progesterone increase during pregnancy
(110). These changes are associated with a decline in
A-type current densities in myometrial smooth muscle cells during the
transition from a nonpregnant to a late-pregnant state
(119). The changes in myometrial A-type currents have been
mimicked by administration of 17-estradiol to immature female
rats. However, progesterone was without effect
(30). In cultured human myometrial cells, chronic exposure
to estrogen or progesterone induced a hyperpolarizing shift in the
A-type current steady-state inactivation curve, whereas simultaneous
exposure of estrogen and progesterone was without effect
(54). Acute effects of estrogen and progesterone were not
evident. As noted above, Kv4.3 channels are thought to underlie the
A-type current in rat myometrium. In this species, at the end of
pregnancy, Kv4.3 mRNA and Kv4.3 protein levels were reduced (100), with the Kv4.3 protein being predominantly confined
to the perinuclear region. Administering 17-estradiol, but not
progesterone, to ovariectomized rats reduced Kv4.3 transcript and
protein levels (100). As with pregnant rats (e.g., Ref.
119), ovariectomized rats treated with 17-estradiol had
drastically diminished A-type currents. Taken together, these data
suggest that estrogen is a negative regulator of A-type channel
expression in the rat myometrium.
| |
BIOPHYSICAL PROPERTIES OF SMOOTH MUSCLE A-TYPE CURRENTS |
A-type currents recorded from vascular smooth muscle cells tend to
be relatively small in amplitude with rather variable biophysical properties (see Table 1). Activation thresholds for vascular A-type
currents have ranged between 65 (rabbit portal vein) and +10 mV
(human mesenteric artery). With the exception of rabbit portal vein,
vascular A-type current activation thresholds and voltages of
half-inactivation were positive to resting membrane potentials (Tables
1 and 2). Cells of the rabbit portal vein are an exception to this
generalization, however, because the A-type current in these cells
appears to be completely inactivated at the resting membrane potential
(12). In contrast to other vascular smooth muscles in
which A-type currents have been recorded, portal vein smooth muscle
displays spontaneous phasic electrical activity (see Table 2); however,
the role of A-type current in the spontaneous activity of this tissue
has not been clarified.
Vascular smooth muscle A-type currents activate and inactivate (e.g.,
20 to 220 ms) rapidly. Inactivation of A-type currents in rat pulmonary
and renal arteries was best described by the sum of two exponentials,
and inactivation of similar currents in rabbit portal vein was fit with
a single component. Recovery from inactivation was slow (0.8 to 10 s) in rabbit portal vein and pulmonary artery. Recovery from
inactivation was considerably faster in human mesenteric artery (time
constant 250 ms). The single-channel conductance(s) responsible for
A-type currents in vascular smooth muscle cells have not been identified.
A-type currents have also been recorded in GU smooth muscle cells. Like
vascular muscle cells, these currents are relatively small in amplitude
(see Table 1). Activation thresholds were between 60 and 30 mV and
occurred at potentials just positive to the resting membrane potential
(Table 2), near the firing threshold of GU smooth muscles. The A-type
current in nonpregnant rat myometrium appears to be an exception, and
significant activation occurred negative to the resting membrane
potential. Although the reported voltages of half-inactivation for
these currents are often negative to the resting membrane potential, GU
A-type currents are not fully inactivated at physiologically relevant membrane potentials.
Inactivation of rat myometrial and guinea pig seminal vesicle A-type
currents were best fit with a single exponential, whereas those of the
guinea pig ureter required two exponents. Recovery from inactivation
was relatively fast in guinea pig ureter with a time constant near 100 ms. In guinea pig vas deferens, recovery was slower and best described
by the sum of two exponents. In myocytes of the guinea pig ureter, a
14-pS (asymmetrical K+) channel with A-type properties has
been recorded in cell-attached patches (49).
A-type currents dominate macroscopic voltage-dependent outward currents
of GI smooth muscle cells isolated from rat ileum (97) and
murine colon (57), antrum (2), and fundus
(Amberg GC, Koh SD, and Sanders KM, unpublished observation) (see Table 1). Smaller A-type currents were reported in guinea pig colon (114), opossum esophagus (1), human esophagus
(117), and mouse jejunum (3). GI A-type
currents were resolved at potentials between 60 and 50 mV, which
corresponds with the resting potentials of many GI smooth muscles
(Table 2). These potentials are also near the voltages of
half-inactivation for the A-type currents. As a result, there is
sustained availability of A-type channels at physiologically relevant
membrane potentials throughout the GI tract.
Once activated, GI A-type currents inactivated rapidly with time
constants <100 ms. A-type currents of mouse (57) and
guinea pig (114) colonic myocytes were best fit by the sum
of two exponentials. At negative potentials, GI A-type currents
typically recovered from inactivation quickly, with time constants
<100 ms. However, opossum esophagus and mouse antral A-currents had
time constants for recovery from inactivation in excess of 125 ms. A
19-pS K+ (symmetrical K+) channel with A-type
properties has been identified in cell-attached patches of murine
colonic myocytes (see Fig. 2; Ref.
4). A 12-pS K+ (asymmetrical K+)
channel is thought to be responsible for the A-type current in the
guinea pig colon (113).
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 2.
Current-voltage relationship of 19-pS K+ channels in
colonic myocytes. A: single-channel currents recorded from a
colonic myocyte using the cell-attached patch-clamp technique. Membrane
potential was stepped for 4 s from 80 mV to potentials between
20 and +60 mV in 20-mV increments under asymmetric K+
gradients. Dashed lines indicate open channels; solid lines indicate
closed channels. B: plot of single-channel current as a
function of voltage under asymmetrical K+
() and symmetrical () K+.
SE bars are included at all points. The asymmetrical K+
data was fit with the Goldman-Hodgkin-Katz (GHK) equation; the
symmetrical K+ data was fit by linear regression.
[Reprinted from Amberg et al. (4).]
|
|
| |
PHARMACOLOGY OF SMOOTH MUSCLE A-TYPE CURRENTS |
All smooth muscle A-type currents showed sensitivity to 4-AP (
10
mM) and insensitivity to external TEA (>10 mM). 3,4-Diaminopyridine blocked seminal vesicle myocyte A-type currents (89). As
with other tissues [e.g., ventricle (18)], the effects
of 4-AP on some smooth muscle A-type currents [e.g., guinea pig ureter
(49), opossum esophagus (1), and guinea pig
(114) and mouse colon (57) were complex (see
Fig. 3)]. 4-AP slowed activation
and decreased the peak amplitudes of these A-type currents, but channel block by 4-AP was relieved during prolonged depolarization, resulting in partial recovery from block. These observations are consistent with
closed channel block by 4-AP that is relieved when the channels open
(see Ref. 57 for further discussion).
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 3.
Inhibition of colonic A-type currents by 4-aminopyridine (4-AP) and
flecainide. A and B: whole cell A-type currents
recorded from a colonic myocyte before and after 4-AP (5 mM;
top) and flecainide (10 µM; bottom). The membrane
potential was stepped for 500 ms from 80 mV to potentials between
80 mV and +20 mV. [Reprinted from Refs. 3 (bottom
traces) and 57 (top traces)].
|
|
The effects of inorganic cations (e.g., Cd2+ and
La3+) have been tested on A-type currents of smooth muscle
cells (1, 3, 12, 49, 95). From studies in neurons,
inorganic cations are known to shift A-type current voltage
dependencies of activation and inactivation to more depolarized
potentials (e.g., Ref. 66). In considering investigations
of A-type currents, it is important to note the presence of inorganic
cations, such as Cd2+ or Mn2+, which are often
included to minimize contamination by Ca2+-activated
K+ currents, since these ions will affect the
voltage-dependence of activation and inactivation.
A-type potassium currents in mouse GI smooth muscle were inhibited by
flecainide (Refs. 2 and 3; see Fig. 3). Flecainide, an
antiarrhythmic agent used in clinical practice, blocks A-type potassium
currents formed by the Kv4 family of K+ channels near
therapeutically relevant concentrations (i.e., low micromolar; Refs.
39 and 124). In the mouse, the IC50 for inhibition of peak current was 11 µM in colon (3) and 35 µM in antrum (2). Data with flecainide are
highly suggestive of a role for Kv4 isoforms in GI A-type currents;
however, the most convincing pharmacological evidence for the
involvement of Kv4 channels is block by heteropodatoxin
(90) or phrixotoxin (27), two spider
venom-derived inhibitors that are highly specific for Kv4
channels. However, limited availability of these toxins has precluded their use in the pharmacological profiling of smooth muscle
A-type currents.
The A-type currents of rabbit vas deferens and guinea pig ureter, vas
deferens, and colon were inhibited by arachidonic acid (68). External application of arachidonic acid decreased
peak current amplitudes with an IC50 of ~1 µM.
Arachidonic acid also increased the apparent rate of inactivation and
decreased the slow component of recovery from inactivation. The effects
of arachidonic acid did not require superoxide generation or
transformation via cyclooxygenase or lipoxygenase pathways. Activation
of protein kinase A, protein kinase C, tyrosine kinases, and
GTP-binding proteins were not required as well. Thus it appears that
the inhibitory effect of arachidonic acid on these A-type currents was
a direct effect of arachidonic acid itself on the underlying channels. Cloned Kv4 channels are also inhibited by similar concentrations of
arachidonic acid (112). A recent report demonstrated that KChIPs are required for kinetic modulation of Kv4 currents by arachidonic acid (45).
Additional blockers of A-type currents include quinine, which inhibited
rabbit pulmonary artery A-type currents (10 µM; Ref. 79), phencyclidine (100 µM), which reportedly inhibited
the A-type current in rabbit portal vein (12), and
glibenclamide (>10 µM), which was reported to inhibit the A-type
current in guinea pig seminal vesicle (89).
1,4-Dihydropyridine Ca2+ channel antagonists (e.g.,
nicardipine), which potently block Ito in rat
cardiac ventricular myocytes (IC50 0.6 µM; Ref.
38), also blocked the A-type current in guinea-pig ureter
(Muraki K and Imaizumi Y, unpublished observation).
| |
REGULATION OF SMOOTH MUSCLE A-TYPE CURRENTS |
Relatively little is known about the regulation of A-type currents
in most of the smooth muscles in which these channels are expressed. In
mouse colonic myocytes, the A-type current is regulated by
Ca2+/calmodulin-dependent kinase II (CaMKII; Ref.
56) and by the Ca2+-dependent phosphatase
calcineurin (4). Dialysis of colonic myocytes with
exogenous CaMKII slowed the rate of inactivation of colonic A-type
currents (Fig. 4), whereas bath
application of the membrane-permeant CaMKII inhibitor KN-93 greatly
increased the rate of inactivation (56). Inhibitors of
calcineurin (cyclosporin A and FK506) and protein phosphatase 1 (okadaic acid) slowed the inactivation of 19-pS channels with A-type
current properties by blocking phosphatase activity and increasing the
phosphorylated state (4). Basal CaMKII activity in colonic
cells appears to influence A-type K+ channels, because
inhibition of CaMKII increased the rate of inactivation of whole cell
A-type currents in amphotericin-perforated patch experiments
(56). It is possible that a portion of the basal activity
results from Ca2+-independent activity of
autophosphorylated CaMKII, a well-characterized feature of this kinase
(e.g., Ref. 98).
View larger version (10K):
[in this window]
[in a new window]
|
Fig. 4.
Dialysis with autothiophosphorylated
Ca2+/calmodulin-dependent kinase II (CaMKII) slows the
inactivation of mouse colonic A-type currents. A: A-type
currents recorded from a myocyte dialyzed with autothiophosphorylated
CaMKII (10 nM) immediately upon establishing whole cell recording
(early) and after 200s of dialysis (late). B: A-type
currents recorded from a myocyte dialyzed with autothiophosphorylated
CaMKII (10 nM) alone, during exposure to TEA (10 mM), and during
exposure to TEA plus the CaMKII inhibitor KN-93 (5 µM). [Reprinted
from Koh et al. (56).]
|
|
In the presence of external TEA, A-type currents of myocytes dialyzed
with autophosphorylated CaMKII did not inactivate completely during
500-ms test potentials (Fig. 4). However, after inhibition of CaMKII
with KN-93, inactivation was complete within a few hundred milliseconds
in the presence of TEA. Slowing of inactivation due to endogenous
CaMKII activity resulted in incomplete inactivation of A-type current.
Thus kinetic modulation by CaMKII functionally transforms this current
from a rapidly inactivating A-type current into a conductance that
performs more like a delayed rectifier with a significant sustained
component. This transformation converts the A-type current into a
conductance that would tend to have sustained influence on excitability
at negative membrane potentials. KChIPs, which regulate A-type current
densities in mouse colon and jejunum, have been shown to produce
kinetic effects qualitatively similar to the effects of CaMKII
(5, 46).
| |
PHYSIOLOGICAL FUNCTION OF SMOOTH MUSCLE A-TYPE CURRENTS |
The availability of A-type channels to contribute to physiological
events in smooth muscle cells is a function of the intrinsic voltage-dependent properties of the channels and the voltage range of
the smooth muscle tissue. The physiological significance of A-type
currents in vascular smooth muscles, therefore, is controversial. In
rabbit portal vein, where the resting membrane potential is reported to
be 50 mV (e.g., Ref. 59), A-type currents should be
completely inactivated (12). In human mesenteric artery, application of 4-AP induced firing of action potentials and eliminated transient afterhyperpolarizations following the action potentials (95). Those authors suggested that a high-threshold A-type
current may suppress membrane excitability in these cells. A similar
argument may be made for the A-type current in rat renal resistance
arteries (37). In rabbit (79) and rat
(125) pulmonary arterial smooth muscle tissue, 4-AP caused
depolarization. However, this effect may not result from inhibition of
an A-type current. In rabbit pulmonary artery, quinine had little
effect on the resting membrane potential at concentrations expected to
inhibit A-type currents (79). Because 4-AP blocks other
species of Kv channels (e.g., Kv1.5), use of this compound does not
necessarily indicate a role for A-type currents, and other
pharmacological agents are needed to test the contribution of A-type currents.
In GU smooth muscle, the best described physiological function for an
A-type current is in the guinea pig ureter. Physiological assessment of
ureteral smooth muscle A-type current has been performed in
current-clamped myocytes (49, 103) and in intact tissue preparations (32). Application of 4-AP to intact tissues
decreased the maximum "diastolic" membrane potential by ~5 mV and
increased spiking frequency (32). The potential
immediately preceding action potential discharge (approximately 60
mV) was not altered by 4-AP. Thus it appears that the A-type current in
guinea pig ureter may regulate firing frequency by enhancing the
maximum diastolic potential between action potentials. These results
are consistent with the steady-state availability of the A-type current reported by Sui and Kao (103) but contrast with the
availability of channels reported by Imaizumi and coworkers
(49).
In ureteral myocytes, studied under current clamp, increasing the
availability of A-type current by changing the resting potential from
50 to 70 mV caused a slight delay in action potential onset (103). From a resting potential of 70 mV, but not 50
mV, 4-AP exposure increased the rate of action potential depolarization by a few milliseconds, whereas other action potential components were
unaffected (49, 103). However, in intact ureter, 4-AP decreased the maximum rate of rise of action potentials and reduced the
initial spike amplitude (32). In contrast to
current-clamped myocytes, where late components of the action potential
were unaffected by A-type current inhibition, the plateau phase of the
action potential in the intact ureter was greatly extended by 4-AP
(32). This effect was thought to result from either
inhibition of a sustained steady-state component of the A-type current
or an increase in initial Ca2+ influx during the action
potential (32). Further investigation is required to
adequately explain the observed differences between intact tissue and
isolated myocyte current-clamp experiments in the guinea pig ureter.
Very little is known about the physiological significance of the A-type
current in myometrial smooth muscle. However, it is clear that the
A-type current is predominantly a feature of the nonpregnant myometrium
(see PLASTICITY OF A-TYPE CURRENT EXPRESSION). Although
A-type currents have been reported in myocytes isolated from pregnant
rat and human uteri, they are largely overshadowed by noninactivating
K+ currents (e.g., Ref. 119). It is likely
that decreases in A-type currents, in concert with changes in other
conductances, contribute toward the increased excitability of the
myometrium at parturition (100).
In GI smooth muscles, reported A-type current activation thresholds and
voltages of half-inactivation are near the resting membrane potential
(Tables 1 and 2). Because activation occurs at potentials where
steady-state inactivation is incomplete, a small fraction of these
A-type currents is available (i.e., there is "window current") at
physiological relevant membrane potentials (Fig. 1). These small
steady-state currents can have significant influences on membrane
potential as a result of the high input resistances typical of GI
smooth muscle cells (i.e., gigaohms).
A-type currents contribute toward the resting membrane potential in
guinea pig colon (114), opossum esophagus
(1), and in mouse antrum (2) and colon
(57). These conclusions are supported by experimental
evidence in intact tissue preparations (mouse antrum and colon) and in
current-clamped isolated smooth muscle cells (opossum esophagus, guinea
pig colon, and mouse antrum). In this respect, the A-type currents
assume a role more typically associated with noninactivating delayed
rectifier K+ currents. At first consideration, this
function would seem unlikely for a rapidly inactivating K+
current; however, these findings are in agreement with steady-state activation and inactivation analyses that have revealed sustained currents (i.e., window currents) within the physiological range of
potentials. Small tonic currents would be expected to arise through
these channels within the window current region of the current-voltage relationship.
With the exception of mouse fundus, the electrical activities of these
GI tissues are spontaneously phasic (Table 2). In mouse colon tissue,
4-AP exposure induced a slight depolarization and abolished the
quiescent periods between slow wave depolarizations (57).
These observations suggest that the contribution of the A-type current
to resting membrane potential is critical for maintaining the phasic
nature of this tissue. In contrast, after blockade of enteric
neurotransmission, flecainide inhibition of the A-type current in mouse
antrum caused a decrease in resting membrane potential without loss of
phasic activity (Fig. 5; Ref.
2). Similar results were obtained in the mouse antrum when
4-AP was used to block the A-type current (Amberg GC, Koh SD, and
Sanders KM, unpublished observation). Thus A-type currents are not a
requirement for phasic activity in the GI tract. Because flecainide
blocks Na+ channels, as well as other K+
channels, results obtained using this compound may result from inhibition of non-A-type channels, and other pharmacological agents are
needed to manipulate A-type currents. The use of 4-AP brings similar
uncertainties with regard to nonspecificity, thus highlighting the need
for greater availability of specific blockers and use of transgenic
technologies in assessing physiological functions of A-type currents.
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 5.
Effects of A-type current
inhibition on the electrical activity of intact murine antral smooth
muscle. Flecainide (30 µM) was applied after control in standard
oxygenated bathing solution supplemented with
NG-nitro-L-arginine, atropine,
nicardipine, glibenclamide, and TEA. A representative slow wave before
and after flecainide is shown at right with an expanded time
scale; for clarity, the flecainide slow wave is shifted to the right by
~2 s. Dashed line denotes the membrane potential between slow waves
under control. [Reprinted from Amberg et al. (2).]
|
|
In current-clamped myocytes isolated from guinea pig colon
(114) and opossum esophagus (1), evoked
action potentials peaked ~30 ms faster after 4-AP exposure, which may
have led to an increase in spiking frequency (116). In
mouse antral myocytes, inhibiting the A-type current did not accelerate
the upstroke of evoked action potentials (2). These
observations may reflect fundamental differences between GI smooth
muscles or may result from differences in experimental parameters. The
changes in electrical activity observed after inhibition of GI A-type
currents likely results from general membrane depolarization as opposed
to specific modulation of spike waveforms by A-type currents. In
contrast to the succinct depolarizations observed in isolated myocyte
current-clamp experiments, in tissue preparations at 37°C, action
potential spikes during slow wave complexes generally occur after
relatively slow depolarizations (see Fig.
6). Under these conditions, most of the
A-type current, which is small to begin with, will have accumulated in
an inactive state by the time action potential thresholds are reached.
Furthermore, at elevated temperatures (e.g., 37°C), current
availability at negative potentials is increased (35, 108)
and inactivation kinetics are more rapid (1).
Consequently, at physiological temperatures, A-type current
contributions toward resting membrane potentials will be increased,
whereas faster rates of inactivation will preclude availability at
depolarized potentials. Thus, in vivo, it is likely that A-type
currents modulate action potentials indirectly by regulating resting
membrane potential. This role for A-type currents is highly dependent
on the window current that results from overlapping of voltage
dependence of activation and inactivation.
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 6.
Relationship between A-type current availability and
membrane potential during slow wave activity in the mouse antrum. The
voltage recording is taken from a conventional microelectrode recording
of mouse antrum at 37°C. The steady-state availability is
extrapolated from the steady-state inactivation curve for mouse antral
A-type currents at room temperature. Steady-state availability is used
to approximate the A-type current availability because of the
relatively slow rate of depolarization (i.e., over 2 s). A-type
channel availability falls precipitously during the initial phase of
the slow wave, and little of this conductance is available further into
the slow wave.
|
|
| |
CONCLUSION |
Although not ubiquitous in smooth muscles, A-type currents clearly
are important regulators of membrane excitability in a variety of
smooth muscles. The physiological role of these channels in smooth
muscles deviates from the functions traditionally associated with
A-type currents. Considerably more information is needed about the
molecular species responsible for A-type currents in a variety of
smooth muscles, and studies to determine second messenger regulation of
these conductances still must be performed on most of the cells in
which these conductances have been identified. To better understand the
physiological significance of A-type current, improved blockers for
A-type currents are needed. Known specific antagonists are in short
supply and difficult to obtain in quantities necessary for
physiological experiments. Alternatively, employment of transgenic
technologies and other molecular techniques may allow a more rigorous
understanding of the physiological function of these currents in intact
smooth muscle organs. Finally, little is known about the presence and
function of A-type currents in human smooth muscles.
This project was supported by National Institute of Diabetes and
Digestive and Kidney Diseases Program Project Grant DK-41315.
Address for reprint requests and other correspondence:
K. M. Sanders, Dept. of Physiology and Cell Biology, Univ. of
Nevada School of Medicine, Reno, Nevada 89557 (E-mail:
kent{at}physio.unr.edu).
TOP
ABSTRACT
INTRODUCTION
