Growth factor-mediated K1 channel activity associated
with human myeloblastic ML-1 cell proliferation
LING WANG, BO XU, RICHARD E. WHITE, AND LUO LU
Department of Physiology and Biophysics, School of Medicine,
Wright State University, Dayton, Ohio 45435
patch clamp; adenosine 38,58-cyclic monophosphate; protein
kinase A; epidermal growth factor; DNA synthesis
HUMAN MYELOBLASTIC leukemia (ML-1) cells proliferate
in tissue culture as immature myeloblasts, and this
proliferation is stimulated by various growth factors
present in the culture serum. ML-1 cells can be programmed to differentiate into granulocytes or macrophages when specifically stimulated (4), and these
differentiated cells play significant roles in the immune
defense system and require membrane-mediated transduction of cell-cell and cell-environment signals. Our
previous studies demonstrated that a voltage-gated K1
current in ML-1 cells was altered during the entire
process of differentiation induced by 12-O-tetradecanoylphorbol 13-acetate (TPA) and revealed that K1
channel activity varied depending on the stage of ML-1
cell proliferation and differentiation (18). Channel activity was dramatically diminished in the early stages of
TPA-induced ML-1 cell differentiation and was completely suppressed in differentiated cells. However, it is
still unclear how voltage-gated K1 channel activity is
regulated and what is the precise role of this channel in
ML-1 cell proliferation and differentiation.
Ion channels located at the cell membrane sense
chemical and physical changes in the cell growth
environment and mediate functional adaptation of the
cell to environmental changes. In excitable tissues,
such as nerves and muscle and some hormone-releasing cells, voltage-dependent K1 channels play important roles in regulation of cell electrical activities in
response to various stimulations. Voltage-dependent
K1 channels also play crucial roles in cell development,
volume regulation, membrane potential stabilization,
and proliferation (5, 6, 8, 13, 20). A variety of studies
have suggested that the voltage-gated K1 channel
plays a functional role in the onset of cellular events
associated with both T and B lymphocyte activation (1,
17, 21). It has been found that enhanced K1 channel
gene expression or increased K1 channel activity is
associated with mitogenesis in several cell types (21).
Application of different K1 channel blockers to cultured
cells significantly inhibits various types of cell proliferation (1, 7, 10). Recently, the important role of the
voltage-gated K1 channel in mitogenesis has been
suggested to be a key determinant for cell progression
through G1 phase before the G1 checkpoint in ML-1 and
other cells (10, 16, 32). In K1 channel activitysuppressed ML-1 cells, retinoblastoma protein (pRB) is
dephosphorylated and effectively inhibits the cell from
progressing through the G1/S transition (32).
To investigate the precise role the voltage-gated K1
channel plays in growth factor-mediated ML-1 cell
proliferation, we have designed a series of experiments
to study mechanisms that underlie the correlation of
channel activity to ML-1 cell growth control and the
effect of growth factors on K1 channel activity during
ML-1 cell proliferation. We found that the growthrelated K1 channel activity was markedly diminished
in serum-deprived cells, and channel activity could be
restored to its full activity within 30 min after physiological concentrations of serum or epidermal growth
factor (EGF) were applied to the patch chamber. EGFstimulated K1 channel activity was mediated through
elevation of intracellular adenosine 38,58-cyclic monophosphate (cAMP) levels and activation of cAMPdependent protein kinase (PKA)-induced phosphorylation. Our results suggest that the K1 channel in ML-1
cells is regulated by growth factor-mediated intracellular signaling pathways and plays an important role in
controlling cell proliferation, specifically in the G1/S
transition of the cell cycle.
MATERIALS AND METHODS
Cell culture. ML-1 cells were originally isolated from an
acute myeloblastic leukemia patient and were received as a
generous gift from Dr. R. W. Craig, Dartmouth Medical School
(Hanover, NH). Cells were maintained in suspension culture
as described previously (18). Briefly, culture medium RPMI
0363-6143/97 $5.00 Copyright r 1997 the American Physiological Society
C1657
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on June 16, 2017
Wang, Ling, Bo Xu, Richard E. White, and Luo Lu.
Growth factor-mediated K1 channel activity associated with
human myeloblastic ML-1 cell proliferation. Am. J. Physiol.
273 (Cell Physiol. 42): C1657–C1665, 1997.—ML-1 cell proliferation is dependent on the presence of serum growth factors.
Removing serum from the culture medium results in growth
arrest and promotes differentiation. In this study, we found
that a 4-aminopyridine-sensitive K1 channel was highly
expressed in proliferating ML-1 cells and significantly diminished in G1-arrested ML-1 cells induced by serum deprivation
but was restored within 30 min in these cells with addition of
10% fetal bovine serum (FBS) or 5 ng/ml epidermal growth
factor (EGF). Intracellular adenosine 38,58-cyclic monophosphate (cAMP) levels, but not guanosine 38,58-cyclic monophosphate, were significantly increased in serum-deprived cells
stimulated by FBS or EGF, and the effects of FBS and EGF on
the channel activation were mimicked by exogenous cAMP. In
inside-out patches, K1 channel activity was significantly
increased by the cAMP-dependent protein kinase catalytic
subunit, whereas the effect of EGF on K1 channel activation
was blocked by Rp-8-(4-chlorophenylthio)adenosine 38,58cyclic monophosphothioate. Together, our results demonstrate that serum growth factors stimulate K1 channel
activity in proliferation of ML-1 cells through protein kinaseinduced phosphorylation and suggest an important molecular
mechanism for serum growth factor-stimulated mitogenesis
in ML-1 cells.
C1658
GF-MEDIATED K1 CHANNEL ACTIVITY AND ML-1 CELL GROWTH
4-aminopyridine (4-AP) were purchased from Sigma Chemical (St. Louis, MO). Rp-8-(4-chlorophentlthio)adenosine 38,58monophosphothioate (Rp-CPT-cAMPS) and EGF were obtained from Biolog Life Science (La Jolla, CA) and Calbiochem
(La Jolla, CA), respectively. The catalytic subunit of PKA was
diluted in KCl bath solution, and 1 mg/ml DTT was added to
the mixture. The mixture was then allowed to stand for 10
min at room temperature (22°C) before use or before storage
at 280°C. K1 channel blockers were prepared as stock
solutions with concentrations of 500 mM to 1 M in sterile
water.
RESULTS
A voltage-gated K1 channel in proliferating ML-1
cells. A voltage-gated and 4-AP-sensitive K1 channel at
the whole cell level in ML-1 cells has been shown
previously (18). To investigate the role of this channel
in growth factor-mediated ML-1 cell proliferation, single
K1 channel currents were measured (Fig. 1A). The
microscopic current-voltage (I-V) relationship was linear with a single-channel conductance of 31 6 0.7 pS
(measured from the slope of I-V curves; n 5 6) in
symmetric 140/140 mM KCl solution (Fig. 1B). When
extracellular KCl was isotonically replaced with NaCl,
the inward current was abolished (Fig. 1B), suggesting
selectivity of this channel for K1. The K1-selective
channel was also confirmed pharmacologically by demonstrating sensitivity of the channel to extracellular
4-AP. Channel activity was inhibited 63 and 97% by
extracellular application of 50 or 100 µM 4-AP, respectively (Fig. 1C). These results indicate that the singlechannel current is carried by K1 through a 4-APsensitive K1 channel.
Serum growth factors activate the K1 channel. Proliferation of ML-1 cells is dependent on various serum
growth factors in the culture medium. If the K1 channel
plays a role in growth factor-mediated ML-1 cell proliferation, then channel activity should be regulated by
serum growth factors. To test this hypothesis, the
precise role of this channel in growth factor-mediated
ML-1 cell proliferation was determined by patch-clamp
experiments. Possible regulatory effects of serum growth
factors on K1 channel function were examined with the
cell-attached patch clamp. Channel activity was observed in ML-1 cells cultured with serum-rich medium
(with 7.5% FBS), and the channel was frequently open
with an average activity (NPo ) of 21 6 5.5% (at 260 mV,
n 5 9) in cell-attached patches (Table 1). In contrast,
channel activity was significantly diminished to an NPo
of 0.4 6 0.2% (at 260 mV, n 5 16, P , 0.001) in ML-1
cells that were cultured in serum-free medium (with
0.3% FBS) for at least 12 h (Fig. 2A). The specific effect
of growth factors on channel activity was studied by
addition of 10% FBS onto serum-starved cells. When
10% FBS was applied in the patch chamber, channel
activity was restored in cell-attached patches in a few
minutes and reached full activity within 30 min (Fig.
2B). Channel activity was significantly increased, from
0.4 6 0.2% to 40 6 15% within 30 min and to 57 6 5%
after 30 min (n 5 6, P , 0.001) (Fig. 2C). In some
patches lasting for 150 min or longer, high channel
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on June 16, 2017
1640 containing 25 mM N-2-hydroxyethylpiperazine-N8-2ethanesulfonic acid (HEPES) buffer was supplemented with
7.5% heat-inactivated fetal bovine serum (FBS; GIBCO,
Grand Island, NY). For the high-K1 culture medium, NaCl in
the RPMI 1640 medium was isotonically replaced by 135 mM
KCl. Cells were grown in a humidified incubator with 5% CO2
at 37°C and passed at a seeding density of 3 3 105 cells/ml.
Cells were washed twice with phosphate buffer solution (PBS)
before they were transferred for patch-clamp experiments.
Cell growth assays. Proliferation of ML-1 cells was determined by counting cell numbers and measuring [3H]thymidine incorporation into the host DNA. ML-1 cells from
suspension cultures were plated in triplicate into 35-mm
culture dishes at a density of 3 3 105 cells/ml. The effect of K1
channel blockers on cell growth was tested by dilution of
concentrated stock solution directly into the plating medium.
After incubation for 4 or 24 h, all cultures were pulsed with 1
µCi/ml [3H]thymidine for 2 h. Cells were then harvested and
washed twice with PBS. Nucleic acids were precipitated with
10% trichloroacetic acid, and radioactivity of samples was
quantified by liquid scintillation counting. Growth arrest
induced by serum deprivation was achieved by culturing cells
in RPMI 1640 medium containing 0.3% FBS at 37°C for 24 h.
Intracellular cAMP and guanosine 38,58-cyclic monophosphate assays. ML-1 cells were synchronized in the G1 phase of
the cell cycle by serum deprivation for 24 h. Cells were then
aliquoted into 35-mm culture dishes at a final concentration
of 1 3 106 cells/ml and were stimulated with either 10% FBS
or 5 ng/ml EGF. At the times indicated, cells were collected
and washed twice with ice-cold PBS and then resuspended in
1 ml of 65% (vol/vol) ice-cold ethanol. After they had settled
for 60 min at 22°C, supernatants were drawn into new test
tubes and remaining precipitates were washed with ice-cold
65% ethanol. Washing solutions were added into the appropriate tubes. Cell extracts were centrifuged at 2,000 g for 15 min
at 4°C, and supernatants were transferred into fresh tubes.
The extracts were then dried overnight by a vacuum lyophilizer. Intracellular cAMP and guanosine 38,58-cyclic monophosphate (cGMP) levels were assayed with the use of the enzyme
immunoassay system (EIA, nonacetylation protocol) provided
by Amersham Life Sciences (Buckinghamshire, UK).
Patch-clamp studies. Both cell-attached and inside-out
patch-clamp techniques were used in the present study.
Detailed methods for the patch pipette preparation, data
acquisition, and single-channel analysis were described previously (31). Briefly, pipettes were manufactured with a twostage puller (PP-83, Narishige) with a resistance of 3–4 MV
when filled with 150 mM KCl solution. The solutions used in
these experiments were 1) KCl bath solution containing (in
mM) 140 KCl, 2 MgCl2, 0.5 CaCl2, 1 ethylene glycol-bis(baminoethyl ether)-N,N,N8,N8-tetraacetic acid, and 10 HEPES
(pH 7.4) and 2) pipette solution containing (in mM) 140 KCl, 2
MgCl2, 1 CaCl2, and 10 HEPES (pH 7.4). Single-channel
currents were recorded with an Axonpatch 200A amplifier
(Axon Instruments, Foster City, CA) and filtered with a
four-pole low-pass filter at 1 kHz and digitalized at 22 kHz by
a pulse-code modulator (A. R. Vetter, Rebersburg, PA). The
pCLAMP program (Axon Instruments) was used to analyze
the single-channel data. The channel activity was determined
as NPo, where N represents number of channel openings in
the patch and Po represents the channel open probability. All
experiments were performed at room temperature (21–23°C).
Data are presented as original values or as means 6 SE,
when indicated. Significant differences were determined by
using the paired t-test at the confidence interval indicated.
Reagents. The catalytic subunit of PKA, MgATP, dithiothreitol (DTT), 8-(4-chlorophenylthio)-cAMP (CPT-cAMP), and
GF-MEDIATED K1 CHANNEL ACTIVITY AND ML-1 CELL GROWTH
C1659
Table 1. Effect of cAMP-dependent protein kinase
on K 1 channel activity in ML-1 cells
Patch-Clamp Conditions
Channel Activity
(NPo ), %
n
Cell-attached patches
Cell-attached patch plus CPT-cAMP
Inside-out patches
Inside-out patches plus PKA-C subunit
21 6 5.5
65 6 12.4*
15 6 3.2
38 6 5*
9
8
4
4
Values are means 6 SE; n 5 no. of cells. Channel activity was
determined as NPo , where N represents no. of channel openings in
the patch and Po represents channel open probability. CPT-cAMP,
8-(4-chlorophenylthio)adenosine 38,58-cyclic monophosphate; PKA-C,
cAMP-dependent protein kinase catalytic subunit. * Significant difference (P , 0.05).
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on June 16, 2017
Fig. 1. Single-channel recordings of K1 channel in proliferating ML-1
cells. Single-channel currents were recorded from cell-attached patches at
various membrane potentials. A: outward current recorded as an upward
deflection and inward current as a downward deflection. Arrows, closed
states of channel. Extracellular solution in the pipette contained 140 mM
K1. B: current-voltage (I-V) relationships obtained from patches with
symmetrical 140 mM K1 in the extracellular solution (j) or with 0
mM K1 in the extracellular solution (r). Data were plotted as means
with SE bars and fit with a linear curve (n 5 6). C: blocking effect of
4-aminopyridine (4-AP) on the K1 channel activity (NPo, where N
represents number of channel openings in the patch and Po represents channel open probability). In the cell-attached patches, 50 or
100 µM 4-AP was added in the extracellular solution at a membrane
potential of 260 mV. Data were presented as means 6 SE (n 5 6).
* Significant difference (P , 0.05), determined by the paired t-test.
activity was observed continuously. Thus these findings
suggest that K1 channel function in proliferating ML-1
cells is regulated by serum growth factors.
Intracellular signaling pathway stimulated by serum
growth factors. To investigate the possible intracellular
signaling pathway involved in regulating K1 channel
activity, intracellular cAMP and cGMP concentrations
were measured before and after stimulation with 10%
FBS in growth-arrested ML-1 cells. The intracellular
cAMP level was significantly increased within 5 min
(P , 0.001) after FBS treatment and continued to rise
for 55 min (Fig. 3). Because it has been shown that EGF
stimulates cAMP production in cardiac myocytes (18,
19) and hepatoma cells (20), we examined the effect of
EGF on the cAMP levels in ML-1 cells. When 5 ng/ml
EGF was applied to growth-arrested ML-1 cells in
culture, the intracellular cAMP concentration was significantly increased within 5 min and reached a plateau level at 30 min (Fig. 3). On the other hand,
application of FBS or EGF did not significantly increase intracellular cGMP levels.
The effects of EGF on K1 channel activity were
further characterized by applying EGF to the bath
solution of cell-attached patches in serum-deprived
ML-1 cells. Channel activity was markedly increased
by 5 ng/ml EGF in a few minutes and reached maximal
activity within 30 min (Fig. 4A). The channel activity
measured at 260 mV was significantly increased, from
0.4 6 0.2% to 47 6 15% within 30 min and to 48 6 16%
after 30 min (n 5 4, P , 0.001). The time course showed
that the increase of EGF-induced K1 channel activity
was parallel to the increase of intracellular cAMP level
(Fig. 4B). It was notable that there was a lag phase (a
few minutes) in the increase of the channel activity.
Increasing the concentration of EGF to 25 ng/ml stimulated the K1 channel activity in serum-deprived ML-1
cells, but there was no further increase after 30 min
(data not shown). Effects of FBS and EGF on K1
channel activity were not a voltage-dependent process
in hematopoietic ML-1 cells when the membrane potential was varied from 260 to 160 mV (Fig. 4C). To
further confirm the effect of cAMP on K1 channel
regulation, 100 µM CPT-cAMP, a membrane-permeable
cAMP analog, was added directly to serum-deprived
ML-1 cells in the patch chamber. Channel activity was
stimulated by CPT-cAMP within a few minutes (Fig.
C1660
GF-MEDIATED K1 CHANNEL ACTIVITY AND ML-1 CELL GROWTH
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on June 16, 2017
Fig. 2. Effect of serum growth factors on K1 channel
activity. Cell-attached patch-clamp experiments were
performed in a symmetrical 140 mM K1 condition at a
membrane potential of 260 mV. A: comparison of K1
channel activity in proliferating (top trace) and serumdeprived (bottom trace) ML-1 cells. Histogram bars
represent channel activities from traces shown at top.
B: activation of the K1 channel by 10% fetal bovine
serum (FBS) in serum-deprived ML-1 cells. Channel
activities were observed in the same cell before and
after addition of 10% FBS in the patch chamber. Histogram bars represent channel activities from traces
shown at top. C: statistics of K1 channel activity
stimulated by 10% FBS in serum-deprived ML-1 cells.
Bars (means with SE bars) represent K1 channel
activity stimulated by 10% FBS at 0, ,30, and .30 min.
Maximal activity (Maxi) was selected from the entire
patch period after the FBS stimulation. * Significant
difference (P , 0.001); data were collected from 6
independent experiments.
4D) and significantly increased to 65 6 12% (n 5 8)
within 30 min (Table 1). Together, these results suggest
that the intracellular signaling pathway for the growth
factor-mediated K1 channel activity involves production of the second messenger cAMP and that EGF may
be one of the growth factors that stimulates this
pathway.
Verification of PKA involvement. To further characterize the intracellular signaling pathway and the potential involvement of PKA-induced phosphorylation in
growth factor-mediated K1 channel activation, insideout patches from ML-1 cells cultured in normal medium were exposed to a phosphorylation mixture solution containing MgATP and the catalytic subunit of
GF-MEDIATED K1 CHANNEL ACTIVITY AND ML-1 CELL GROWTH
PKA. Excised patches were held at 260 mV for 2 min to
measure control channel activity, and then 1 mM
MgATP was added to the patch bath solution. K1
channel activity was not significantly changed during a
waiting period of 7–10 min. Then, 50 nM PKA catalytic
subunit was added in the bath, resulting in a significant increase in NPo from 15 6 3.2% to 38 6 5.1% (n 5
4, P , 0.05) in 5–10 min (Fig. 5A and Table 1). In
inside-out patches that were not exposed to the phosphorylation mixture solution, K1 activity faded away
within 10–30 min after patch excision. Results from
these experiments suggest that phosphorylation of the
channel protein induced by PKA is required to maintain the normal activity of the K1 channel in these cells.
This conclusion was further supported by experiments
that demonstrated that Rp-CPT-cAMPS, an antagonist
of PKA, blocked channel activity induced by EGF in
serum-deprived ML-1 cells (Fig. 5B). Cell-attached
patches were held at 260 mV, and 5 ng/ml EGF was
then added to stimulate channel activity. Rp-CPTcAMPS (100 µM) was added to the patched cell after a
dramatic increase of channel activity was observed (5
min). The K1 channel activity induced by EGF was
significantly diminished, from 47 6 15% to 10 6 3%
(n 5 4, P , 0.01), within 30–50 min after application of
100 µM Rp-CPT-cAMPS (Fig. 5C).
Effect of suppressed K1 channel activity on ML-1 cell
proliferation. The effect of inhibition of K1 channel on
ML-1 cell proliferation was evaluated by adding K1
channel blockers in the culture medium individually or
by replacing Na1 with K1 in the medium and by
monitoring cell numbers and DNA synthesis measured
by [3H]thymidine incorporation. ML-1 cells were growth
arrested in the G1 phase by culturing cells in serumdeprived medium for 24 h. Growth-arrested cells were
then released into normal culture medium with 7.5%
FBS (controls), in 135 mM K1 medium with 7.5% FBS
(high K1 medium), or in normal culture medium containing 7.5% FBS plus 2 mM 4-AP, 2.5 mM BaCl2, 10
mM tetraethylammonium, or 30 µM quinine. After
applications of different K1 channel blockers at the
indicated concentrations or in the high-K1 concentration culture condition for 24 h, there were no significant
changes in cell viability measured with the trypan blue
exclusion method. Viability measurements of ML-1
cells in the absence and presence of different K1
channel blockers and in high-K1 concentration culture
condition are summarized in Table 2. The fractional
inhibition of DNA synthesis was measured at 4 and 24
h after release of growth-arrested cells. Rates of [3H]thymidine incorporation after exposure to different channel blockers or the high-K1 concentration culture condition were significantly inhibited as early as 4 h and
reached a much higher level at 24 h (Fig. 6). These
results suggest that blockade of the K1 channel by K1
channel blockers and growing cells in the high-K1
concentration medium inhibited DNA synthesis, preventing ML-1 cells from progressing through the G1
phase to S phase of the cell cycle.
DISCUSSION
K1 channel activity has been found to influence cell
proliferation and differentiation in various systems (2,
13, 18). In the present study, we demonstrate that K1
channel activity is closely correlated to serum growth
factor-stimulated ML-1 cell proliferation. In proliferating ML-1 cells, the K1 channel activity was observed in
abundance (Table 1), but this activity was greatly
suppressed in serum-deprived ML-1 cells that were
synchronized in the G1 phase of the cell cycle (32). The
diminished K1 channel activity in serum-deprived cells
was then restored within 30 min by direct exposure of
ML-1 cells to physiological concentrations of FBS and
EGF (Figs. 2 and 4). Activation of K1 channels by serum
growth factors has been found in other cell types. For
example, in PC-12 cells, nerve growth factor regulates
the abundance and distribution of delayed rectifier K1
channels (24) and exposure of resting microglial cells to
interferon-g or granulocyte/macrophage colony-stimulating factor results in an inhibition of outward K1
current (11). These results suggest that a shift of the
resting membrane potential to more hyperpolarized
levels may be a prerequisite for intracellular mechanisms involved in macrophage and microglial cell activity. Generally, effects of growth factors and cytokines on
K1 channel activity can be divided into long-term and
short-term effects. The short-term effect is characterized by altering channel gating, most likely through
second messenger-mediated modulation of channel protein. However, the long-term effect corresponds to the
maximal K1 conductance affected by total channel
numbers resulting from altered K1 channel gene expression and insertion of channel proteins to the membrane. In human cultured oligodendrocytes, inward
rectifier K1 channels were modulated by tumor necrosis factor-a (TNF-a), a cytokine associated with activated macrophages (19, 25). Treatment of oligodendrocytes with TNF-a for 24–48 h significantly decreases
expression of the K1 channel gene and diminishes the
mean open time of the K1 channel relative to control
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on June 16, 2017
Fig. 3. Stimulation of intracellular cAMP levels by serum growth
factors in serum-deprived ML-1 cells. Intracellular cAMP level was
measured with the enzyme immunoassay system, following the time
course of 10% FBS (j) and 5 ng/ml epidermal growth factor (EGF; r)
treatments. Intracellular cGMP level (l) was also measured before
and after stimulation with 5 ng/ml EGF, following the indicated time
course. Data were plotted as means with SE bars and collected from 5
independent experiments.
C1661
C1662
GF-MEDIATED K1 CHANNEL ACTIVITY AND ML-1 CELL GROWTH
value. These data suggest that TNF-a possesses both
short- and long-term effects on the inward rectifier K1
channel in human oligodendrocytes. In the present
study, activation of K1 channels in ML-1 cells by 10%
FBS or 5 ng/ml EGF can be considered a short-term
effect of serum growth factors.
Growth factor/cytokine receptor-mediated second
messenger systems, such as the cAMP cascade, have
been suggested to modulate K1 channel function, and a
number of different experimental approaches have
been used to study the role of the cAMP cascade in
modulating K1 channel activity. Intracellular levels of
cAMP may be increased artificially by injection of
cAMP through microelectrodes, extracellular application of membrane-permeable cAMP analogs, such as
dibutyryl cAMP or 8-bromo-cAMP, use of phosphodies-
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on June 16, 2017
Fig. 4. EGF-stimulated K1 channel activity. Cell-attached patch-clamp experiments were performed in a symmetrical 140 mM K1 condition at a membrane potential of 260 mV. A: activation of the K1 channel by 5 ng/ml EGF
in serum-deprived ML-1 cells. Channel activities were observed from the same patch before and after addition of 5
ng/ml EGF in the patch chamber. Histogram bars demonstrate channel activities from traces shown at top. B: time
course (inset) and statistics of K1 channel activity stimulated by 5 ng/ml EGF in serum-deprived ML-1 cells. Bars
(mean values 6 SE) represent K1 channel activity stimulated by 5 ng/ml at 0, ,30, and .30 min. Maximal activity
was selected from the entire patch period after the EGF stimulation. * Significant difference (P , 0.001); data were
collected from 6 independent experiments. C: effect of membrane potential on FBS- and EGF-induced K1 channel
activity. Data were collected from 4–6 independent experiments and plotted as mean values with SE bars. D:
activation of K1 channel by 100 µM 8-(4-chlorophenylthio)-cAMP (8-CPT-cAMP) in serum-deprived ML-1 cells.
Channel activities were observed from the same patch before (top trace) and after (bottom trace) addition of 100 µM
8-CPT-cAMP in the patch chamber. Histogram bars demonstrate channel activities from traces shown at top.
GF-MEDIATED K1 CHANNEL ACTIVITY AND ML-1 CELL GROWTH
C1663
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on June 16, 2017
Fig. 5. Increased K1 channel activity by
cAMP-dependent protein kinase (PKA)-induced phosphorylation. A: activation of the
K1 channel by PKA-induced phosphorylation.
Activities of the K1 channel in the excised
inside-out patch were compared before and
after a cocktail containing PKA catalytic subunit and MgATP was applied in the bath
solution at 260 mV. Bottom: channel activity
from the traces shown at top. B: blockade of
EGF effect on the K1 channel activity by
Rp-8-(4-chlorophenylthio)adenosine 38,58-cyclic monophosphothioate (Rp-CPT-cAMPS). In
the cell-attached patch from a serum-deprived cell, K1 channel was activated 15 min
after 5 ng/ml EGF was applied at 260 mV.
Channel activity was then inhibited by application of 100 µM Rp-CPT-cAMPS in the same
cell. Bottom: channel activities from traces
shown at top. C: statistics of Rp-CPT-cAMPS
blockade. Bars represent the blockade of EGFstimulated K1 channel activity by 100 µM
Rp-CPT-cAMPS. Data were plotted as means
with SE bars. * Significant difference (P ,
0.01, n 5 4).
C1664
GF-MEDIATED K1 CHANNEL ACTIVITY AND ML-1 CELL GROWTH
Table 2. Effect of altered culture conditions for 24 h
on viability of ML-1 cells
Culture Conditions
Viability, %
n
Normal culture
4-AP (2 mM)
Barium (2.5 mM)
TEA (10 mM)
Quinine (30 µM)
High-K1 culture
Serum starvation
98.6 6 0.5
98.2 6 0.6
98.8 6 0.7
97.9 6 1.1
97.8 6 0.5
97.8 6 1.2
97.0 6 0.4
12
6
6
6
6
6
6
Values are means 6 SE of n (no. of cells). 4-AP, 4-aminopyridine;
TEA, tetraethylammonium. There are no significant differences in
the viability among these groups.
Fig. 6. Effect of suppressed K1 channel activity on ML-1 cell proliferation. Fractional inhibition of [3H]thymidine incorporation was measured at 4 h and at 24 h in high-K1 culture condition (high K1 ); in the
presence of 2 mM 4-AP, 2.5 mM BaCl2, 10 mM tetraethylammonium
(TEA), or 30 µM quinine; or in the culture after serum-deprived ML-1
cells were supplemented with 7.5% FBS (starvation). Fractional
inhibition was calculated by the equation (1 2 T)/T0, where T
represents rate of [3H]thymidine incorporation in the altered culture
conditions or in the presence of individual K1 channel blockers and T0
represents rate of [3H]thymidine incorporation in the normal culture
condition. Data were plotted as means with SE bars and were
collected from 6 independent experiments.
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on June 16, 2017
terase inhibitors, or use of the diterpene compound
forskolin, a direct activator of adenylate cyclase (20).
Using EIA, we found that the effect of serum growth
factors or EGF on ML-1 cells was mediated through
regulation of intracellular cAMP levels but not intracellular cGMP levels (Fig. 3). The effect of increased
intracellular cAMP levels on K1 channel activity was
verified by direct application of CPT-cAMP (Fig. 4D).
Increased intracellular cAMP can further activate PKA,
leading to phosphorylation of serine/threonine residues
on a variety of substrate proteins, including ion channels.
The present study demonstrates that serum growth
factor-stimulated K1 channel activity is mediated via
cAMP-dependent phosphorylation. Using the insideout patch clamp, we confirmed that serum growth
factor-regulated K1 channels in ML-1 cells can be
activated by direct phosphorylation by the PKA catalytic subunit in vitro (Fig. 5A). Furthermore, we found
that the effect of EGF on K1 channel activity can be
blocked by the PKA inhibitor Rp-CPT-cAMPS (Fig. 5B).
These results raise the interesting possibility that PKA
mediates the effects of growth factors on ion channels
and other proteins. A large body of evidence has shown
that protein phosphorylation by PKA is an important
cellular mechanism modulating K1 channel function.
The phosphorylation state of the channel subunits or
associated proteins can influence the amplitude or the
time course of current initiated by a change of membrane potential or ligand binding (7). For example, the
delayed rectifier K1 channel in Aplysia bag cell neurons, the K1 channel in hippocampal neurons, and
Ca21-activated K1 channels in neuroendocrine cells are
also inhibited by cAMP analogs via activation of PKA
(26, 29). However, it is important to point out that
PKA-mediated phosphorylation can induce the opposite effect in different voltage-gated K1 channels. For
instance, the anomalous rectifier K1 channel functions
in Aplysia neurons and cardiac cells are upregulated by
activation of PKA (2, 12, 27). PKA-mediated phosphorylation affects the opening probability and the Ca21 or
voltage sensitivity of rat brain Ca21-activated K1 channels reconstituted into artificial lipid bilayers (23).
Correlation of PKA-mediated modulation of intrinsic
channel characteristics with the direct phosphorylation
of a K1 channel has been demonstrated for three
distinct voltage-gated K1 channels belonging to the
Shaker subfamily. Therefore, by integrating electrophysiological and molecular biology techniques in a
Xenopus oocyte expression system, the inactivation
gating of the Shaker K1 channels and the opening time
that a single Kv1.2 channel spends in different conductance states were shown to be regulated by PKAinduced phosphorylation at the COOH-terminal region
of the channels (9, 14). Direct phosphorylation of channel proteins by PKA has been demonstrated biochemically for the Shaker K1 channels purified from rat
(Kv1.1) or bovine brain (Kv1.2) (15). The opening
probabilities of these channel on reconstituted lipid
bilayers and the Kv1.3 channel residing in T lymphocyte membrane can be increased by cAMP-dependent
phosphorylation (3, 22).
We have shown that K1 channel activity was extremely low in growth-arrested ML-1 cells and that
channel activity can be restored by addition of 10% FBS
(Fig. 2). To study how the effect of growth factors on the
channel activity might influence cell proliferation,
[3H]thymidine incorporation was used to measure ML-1
cells entering the S phase and proliferation (Fig. 6).
Suppression of K1 channel activity by different K1
channel blockers effectively prevented growth-arrested
ML-1 cells from entering the S phase of the cell cycle. It
has been shown that pRB controls cell proliferation at
the G1 check point of the cell cycle, whereas pRB
dephosphorylation causes G1 arrest in many cell types
(28). We have shown that increases in the dephosphorylated form of pRB is an important event in the loss of
proliferation in K1 channel-suppressed ML-1 cells (32).
Our results have demonstrated that the effect of K1
channel inhibition on ML-1 cell proliferation is a phasespecific event.
In summary, the investigation of a functional role for
a growth-associated K1 channel in human myeloblastic
ML-1 cell proliferation has provided new evidence that
GF-MEDIATED K1 CHANNEL ACTIVITY AND ML-1 CELL GROWTH
K1 channel activity involves growth factor-mediated
G1/S transition of the cell cycle. Activity of this channel
is closely associated with this stage of cell growth and is
controlled by serum growth factors through the intracellular cAMP and cAMP-dependent kinase cascades.
PKA-mediated phosphorylation of this K1 channel resulted in increased activity that paralleled serum
growth factor-stimulated cell proliferation. Our findings also provide an additional molecular mechanism
supporting a role for K1 channel activity in the G1/S
transition of the cell cycle, and this mechanism constitutes a novel means of controlling ML-1 cell proliferation.
Received 6 January 1997; accepted in final form 17 June 1997.
15.
16.
17.
18.
19.
20.
REFERENCES
1. Amigorena, S., D. Choquet, J. L. Teillaud, H. Korn, and
W. H. Fridman. Ion channel blockers inhibit B cell activation at
a precise stage of the G1 phase of the cell cycle. Possible
involvement of K1 channels. J. Immunol. 144: 2038–2045, 1990.
2. Benson, J. A., and I. B. Levitan. Serotonin increases an
anomalously rectifying K1 current in the Aplysia neuron R15.
Proc. Natl. Acad. Sci. USA 80: 3522–3525, 1983.
3. Cai, Y. C., and J. Douglass. In vivo and in vitro phosphorylation of the T lymphocyte type n (Kv1.3) potassium channel. J.
Biol. Chem. 268: 23720–23727, 1993.
4. Craig, R. W., O. S. Frankfurt, H. Sakagami, K. Takeda, and
A. Bloch. Macromolecular and cell cycle effects of different
classes of agents including the maturation of human myeloblastic leukemia (ML-1) cells. Cancer Res. 44: 2421–2429, 1984.
5. Day, M. L., S. J. Pickering, M. H. Johnson, and D. I. Cook.
Cell-cycle control of a large-conductance K1 channel in mouse
early embryos. Nature 365: 560–562, 1993.
6. Decoursey, T. E., K. G. Chandy, S. Gupta, and M. D.
Cahalan. Mitogen induction of ion channels in murine T lymphocytes. J. Gen. Physiol. 89: 405–420, 1987.
7. Deterre, P., D. Paupardin-Tritsch, J. Bockaert, and H. M.
Gerschenfeld. Role of cyclic AMP in a serotonin-evoked slow
inward current in snail neurones. Nature 290: 783–785, 1981.
8. Deutsch, C., and L. Q. Chen. Heterologous expression of
specific K1 channels in T lymphocytes: functional consequences
for volume regulation. Proc. Natl. Acad. Sci. USA 90: 10036–
10040, 1993.
9. Drain, P., A. E. Dubin, and R. W. Aldrich. Regulation of
Shaker K1 channel inactivation gating by the cAMP-dependent
protein kinase. Neuron 12: 1097–1099, 1994.
10. Dubois, J. M., and B. Rouzaire-Dubois. Role of potassium
channels in mitogenesis. Prog. Biophys. Mol. Biol. 59: 1–21, 1993.
11. Fischer, H. G., C. Eder, U. Hadding, and U. Heinemann.
Cytokine-dependent K1 channel profile of microglia at immunologically defined functional states. Neuroscience 64: 183–191,
1995.
12. Frace, A. M., and H. C. Hartzell. Opposite effects of phosphatase inhibitors on L-type calcium and delayed rectifier currents
in frog cardiac myocytes. J. Physiol. (Lond.) 472: 305–326, 1993.
13. Freedman, B. D., B. K. Fleischmann, J. A. Punt, G. Gaulton,
Y. Hashimoto, and M. I. Kotlikoff. Identification of Kv1.1
expression by murine CD4-CD8- thymocytes. A role for voltage-
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
dependent K1 channels in murine thymocyte development. J.
Biol. Chem. 270: 22406–22411, 1995.
Huang, X. Y., A. D. Morielli, and E. G. Peralta. Molecular
basis of cardiac potassium channel stimulation by protein kinase
A. Proc. Natl. Acad. Sci. USA 91: 624–628, 1994.
Ivanina, T., T. Perets, W. B. Thornhill, G. Levin, N. Dascal,
and I. Lotan. Phosphorylation by protein kinase A of RCK1 K1
channels expressed in Xenopus oocytes. Biochemistry 33: 8786–
8792, 1994.
Lewis, R. S., and M. D. Cahalan. Ion channels and signal
transduction in lymphocytes. Annu. Rev. Physiol. 52: 415–430,
1990.
Lin, C. S., R. C. Boltz, J. T. Blake, M. Nguyen, A. Talento,
P. A. Fischer, M. S. Springer, N. H. Sigal, R. S. Slaughter,
M. L. Garcia, G. J. Kaczorowski, and G. C. Koo. Voltagegated potassium channels regulate calcium-dependent pathways
involved in human T lymphocyte activation. J. Exp. Med. 177:
637–645, 1993.
Lu, L., T. Yang, D. Markakis, W. B. Guggino, and R. W.
Craig. Alterations in a voltage-gated K1 current during the
differentiation of ML-1 human myeloblastic leukemia cells. J.
Membr. Biol. 132: 267–274, 1993.
McLarnon, J. G., M. Michikawa, and S. U. Kim. Effects of
tumor necrosis factor on inward potassium current and cell
morphology in cultured human oligodendrocytes. Glia 9: 120–
126, 1993.
Pappone, P. A., and S. I. Ortiz-Miranda. Blockers of voltagegated K channels inhibit proliferation of cultured brown fat cells.
Am. J. Physiol. 264 (Cell Physiol. 33): C1014–C1019, 1993.
Price, M., S. C. Lee, and C. Deutsch. Charybdotoxin inhibits
proliferation and interleukin 2 production in human peripheral
blood lymphocytes. Proc. Natl. Acad. Sci. USA 86: 10171–10175,
1989.
Rehm, H., S. Pelzer, C. Cochet, E. Chambaz, B. L. Tempel,
W. Trautwein, D. Pelzer, and M. Lazdunski. Dendrotoxinbinding brain membrane protein displays a K1 channel activity
that is stimulated by both cAMP-dependent and endogenous
phosphorylations. Biochemistry 28: 6455–6460, 1989.
Reinhart, P. H., S. Chung, B. L. Martin, D. L. Brautigan,
and I. B. Levitan. Modulation of calcium-activated potassium
channels from rat brain by protein kinase A and phosphatase 2A.
J. Neurosci. 11: 1627–1635, 1991.
Seamon, K. B., and J. W. Daly. Forskolin: a unique diterpene
activator of cyclic AMP-generating systems. J. Cyclic Nucleotide
Res. 7: 201–224, 1981.
Soliven, B., S. Szuchet, and D. J. Nelson. Tumor necrosis
factor inhibits K1 current expression in cultured oligodendrocytes. J. Membr. Biol. 124: 127–137, 1991.
Strong, J. A., and L. K. Kaczmarek. Multiple components of
delayed potassium current in peptidergic neurons of Aplysia:
modulation by an activator of adenylate cyclase. J. Neurosci. 6:
814–822, 1986.
Walsh, K. B., and R. S. Kass. Distinct voltage-dependent
regulation of a heart-delayed IK by protein kinases A and C.
Am. J. Physiol. 261 (Cell Physiol. 30): C1081–C1090, 1991.
Weinberg, R. A. The retinoblastoma protein and cell cycle
control. Cell 81: 323–330, 1995.
White, R. E., A. Schonbrunn, and D. L. Armstrong. Somatostatin stimulates Ca21-activated K1 channels through protein
dephosphorylation. Nature 351: 570–573, 1991.
Woodfork, K. A., W. F. Wonderlin, V. A. Peterson, and J. S.
Strobl. Inhibition of ATP-sensitive potassium channels causes
reversible cell-cycle arrest of human breast cancer cells in tissue
culture. J. Cell. Physiol. 162: 163–171, 1995.
Xu, B., and L. Lu. Protein kinase A-regulated Cl2 channel in
ML-1 human hematopoietic myeloblasts. J. Membr. Biol. 142:
65–75, 1994.
Xu, B., B. A. Wilson, and L. Lu. Induction of human myeloblastic ML-1 cell G1 arrest by suppression of K1 channel activity.
Am. J. Physiol. 271 (Cell Physiol. 40): C2037–C2044, 1996.
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on June 16, 2017
We thank Dr. R. W. Craig for giving us ML-1 cells as a generous
gift.
This study was supported by National Institute of General Medical
Sciences Grant GM-46834 (to L. Lu) and was partially supported by
National Heart, Lung, and Blood Institute Grant HL-54844 (to R. E.
White) and by a grant from the American Foundation for Aging
Research (to R. E. White).
Address for reprint requests: L. Lu, Dept. of Physiology and
Biophysics, School of Medicine, Wright State Univ., Dayton, OH 45435.
14.
C1665