Content uploaded by Carol J Milligan
Author content
All content in this area was uploaded by Carol J Milligan on Jul 01, 2014
Content may be subject to copyright.
Available via license: CC BY 4.0
Content may be subject to copyright.
Regulation of Cloned Cardiac L-type Calcium Channels by
cGMP-dependent Protein Kinase*
(Received for publication, October 14, 1999)
L. H. Jiang, D. J. Gawler, N. Hodson, C. J. Milligan, H. A. Pearson, V. Porter, and D. Wray‡
From the School of Biomedical Sciences, University of Leeds, Leeds LS2 9JT, United Kingdom
We have studied the effect of 8-bromo-cyclic GMP (8-
Br-cGMP) on cloned cardiac L-type calcium channel cur-
rents to determine the site and mechanism of action
underlying the functional effect. Rabbit cardiac
␣
1C
sub-
unit, in the presence or absence of

1
subunit (rabbit
skeletal muscle) or

2
subunit (rat cardiac/brain), was
expressed in Xenopus oocytes, and two-electrode volt-
age-clamp recordings were made 2 or 3 days later. Ap-
plication of 8-Br-cGMP caused decreases in calcium
channel currents in cells expressing the
␣
1C
subunit,
whether or not a

subunit was co-expressed. No inhibi-
tion of currents by 8-Br-cGMP was observed in the pres-
ence of the protein kinase G inhibitor KT5823. Substitu-
tions of serine residues by alanine were made at
residues Ser
533
and Ser
1371
on the
␣
1C
subunit. As for
wild type, the mutant S1371A exhibited inhibition of
calcium channel currents by 8-Br-cGMP, whereas no
effect of 8-Br-cGMP was observed for mutant S533A.
Inhibition of calcium currents by 8-Br-cGMP was also
observed in the additional presence of the
␣
2
␦
subunit
for wild type channels but not for the mutant S533A.
These results indicate that cGMP causes inhibition of
L-type calcium channel currents by phosphorylation of
the
␣
1C
subunit at position Ser
533
via the action of pro-
tein kinase G.
Voltage-dependent calcium channels play important roles in
cell function, including for example the coupling of excitation to
release of neurotransmitters and hormones, and also play an
important role in the cardiac action potential. The dihydropy-
ridine-sensitive L-type Ca
2⫹
channels in cardiac cells provide a
major pathway for entry of extracellular Ca
2⫹
into the cyto-
plasm (1). The cardiac L-type Ca
2⫹
channel molecule consists of
three subunits, the channel pore-forming
␣
1
subunit, an intra-
cellular

subunit, and the largely extracellular
␣
2
/
␦
subunit
(2–7). Similar L-type Ca
2⫹
channels are also expressed in
smooth muscle and neuronal cells (8, 9). The
␣
1
subunit con-
tains four domains denoted I–IV, with N and C termini located
in the cytoplasm (see Fig. 4A), as is also the case for sodium and
potassium channels (the latter having four subunits) (10). Each
domain is composed of six hydrophobic segments, S1–S6, which
span the plasma membrane. Owing to the unique presence of
positively charged residues (arginine or lysine) at every three
or four residues, the S4 segment has been recognized as the
voltage sensor (10–13). The P region connecting the S5 and S6
segments forms the external mouth of the channel pore, where
a ring of four negatively charged glutamate residues contrib-
utes to the Ca
2⫹
selectivity of the channel (14–16). Functional
expression studies have demonstrated that the
␣
1
subunit has
the ability to constitute a voltage-dependent Ca
2⫹
channel on
its own. However, the ancillary subunits, in particular the

subunit, are essential to form fully functioning channels (2, 4,
17–19). The precise structural domains responsible for the
interaction between the
␣
1
subunit and the

subunit have
been defined; the interaction domain on the
␣
1
subunit has
been identified on the cytoplasmic loop connecting domains I
and II (20, 21) (see Fig. 4A), whereas the complementary in-
teraction region is located on a conserved domain of the

subunit (22).
Second messenger-activated protein kinase phosphorylation
has been well documented to be a crucial physiological regula-
tion mechanism of cardiac L-type Ca
2⫹
channels (6, 9, 23, 24).
For instance, the N-terminal region of the rabbit
␣
1C
subunit
has been identified as being important for the ability of protein
kinase C (PKC)
1
to regulate Ca
2⫹
channel currents (25),
whereas phosphorylation of the C-terminal region by cAMP-de-
pendent kinase (PKA) has been demonstrated to increase Ca
2⫹
channel currents (26–28). Cyclic GMP is recognized as being an
important second messenger in native cells (29), leading to
inhibition of calcium currents via activation of protein kinase
G. However, the mechanism of action of PKG on calcium chan-
nels has not previously been studied at the molecular/struc-
tural level.
Regulation by cGMP is an important mechanism, especially
because of the key role of nitric oxide in activating guanylate
cyclase (30). For instance, nitric oxide donors cause vasodila-
tation in vascular smooth muscle, and this may occur in part by
a decrease in L-type calcium channel currents, which is known
to occur via the activation of cGMP-dependent protein kinase
(9, 24). A similar mechanism is found in cardiac muscle, where
phosphorylation by PKG again decreases calcium channel cur-
rents (31, 32), and this could explain the negative inotropic
effects of nitric oxide on the heart. Kinetic analysis of the effect
of cGMP on single Ca
2⫹
channel activity in patch-clamped
cardiac cells indicates that cGMP-induced activation of PKG
prolongs the closed time with no effect on open time and single
channel conductance (33). Although it therefore seems clear
that PKG can inhibit calcium channel currents, the situation is
complicated because, in some (but not all) cell preparations, the
PKG-mediated effect may only be observable after PKA activa-
tion by cAMP; under basal conditions, activation of PKG may
have no detectable effect upon Ca
2⫹
channel currents (34, 35).
To complicate matters further, there is also evidence that the
inhibitory effects of cGMP on frog cardiac L-type Ca
2⫹
channel
currents may be mediated by cGMP-stimulated phosphodies-
* This work was supported in part by the British Heart Foundation.
The costs of publication of this article were defrayed in part by the
payment of page charges. This article must therefore be hereby marked
“advertisement” in accordance with 18 U.S.C. Section 1734 solely to
indicate this fact.
‡ To whom correspondence should be addressed. Tel.: 44-113-
2334320; d.wray@leeds.ac.uk.
1
The abbreviations used are: PKC, protein kinase C; PKA, cAMP-de-
pendent kinase; PKG, protein kinase G; 8-Br-cGMP, 8-bromo-cyclic
GMP; PCR, polymerase chain reaction; kb, kilobase(s).
THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 275, No. 9, Issue of March 3, pp. 6135–6143, 2000
© 2000 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.
This paper is available on line at http://www.jbc.org 6135
by guest, on February 19, 2013www.jbc.orgDownloaded from
terase activity rather than activation of PKG. This mechanism
may involve the breakdown of intracellular cAMP by the phos-
phodiesterase, which then results in the subsequent reduction
in the PKA-mediated phosphorylation of the Ca
2⫹
channels
and consequent reduction in currents (36, 37).
Although there is therefore much evidence for an important
role for cGMP/PKG in the regulation of cardiac L-type channels,
the site of phosphorylation remains to be determined. For in-
stance the effect may not be mediated by direct phosphoryla-
tion of the channel. Alternatively, if the channel itself is phos-
phorylated, it is not known which subunit, or indeed which
residue, is phosphorylated. There is some evidence that the
␣
1
and

subunits can be phosphorylated in vitro by protein kinase
G (8, 38). However, there is still no direct evidence that func-
tional effects accompany any such phosphorylation. Here, we
have attempted to gain a better understanding of the mecha-
nism involved in the cGMP-mediated inhibition of the cardiac
L-type Ca
2⫹
channel currents. By expressing cloned rabbit cal-
cium channels in Xenopus oocytes, we have investigated the
molecular/structural basis of the functional effect of cGMP; in
particular, direct effects on the channel, the subunit concerned,
and the residue phosphorylated have been studied.
EXPERIMENTAL PROCEDURES
Mutagenesis and cRNA Preparation—cDNA clones for rabbit cardiac
L-type Ca
2⫹
channel
␣
1C
subunit (2), rabbit skeletal muscle

1a
subunit
(39), rat cardiac/brain

2a
subunit (3), and rabbit skeletal muscle
␣
2
/
␦
-1
subunit (40) were used in this study. The cDNA encoding the rabbit
␣
1C
subunit was subcloned into pcDNA3 (Invitrogen) between the HindIII
and KpnI restriction sites in the vector multiple cloning sequence (the
first nonessential 171 bases of the 5⬘-untranslated region of the
␣
1C
cDNA were deleted, as was a short fragment containing an unwanted
EcoRI site between Asp718 and XbaI within the multiple cloning region
of the vector). This construct was then used as the template for PCR
mutagenesis of S533A and S1371A using the method of Sarkar and
Sommer (41). PCR was carried out using Deep Vent polymerase (New
England Biolab) for 30 cycles (40 s at 95 °C, 1 min at 60–64 °C, and at
a 1 kb/min extension rate at 72 °C).
To make mutant S533A, first round PCR was carried out using a
forward mutagenic primer (5⬘-AGT CGA AAT TCGCCC GCT ACT
GGC) and a reverse primer (5⬘-CAC GGC CTT CTC CTT AAG GTG) to
produce a 1.1-kb product. Second round PCR was then undertaken
using a forward primer (5⬘-TCT ACA TCT CTC CTG GAG GTT C) and
the 1.1-kb first round PCR product as the reverse “megaprimer” to
produce a 2.6-kb product. This 2.6-kb product was then digested with
restriction endonucleases ClaI and EcoRI to produce a ⬃2.0-kb frag-
ment. The wild type
␣
1C
cDNA was similarly digested with ClaI and
EcoRI, and the mutagenic 2-kb fragment was then ligated into the
␣
1C
cDNA.
For mutant S1371A, PCR mutagenesis was carried out similarly.
First round PCR was undertaken using a forward mutagenic primer
(5⬘-CAA GCT GCT GGC CCG CGG GGA G-3⬘) and reverse primer
(5⬘-GCC GGA GGA GGG TCA CCA CA-3⬘), again using the wild type
␣
1C
-pcDNA3 template to synthesize a first round product of ⬃0.6 kb.
Second round PCR was undertaken with a forward primer (5⬘-TCC ATT
ACA GCT GAT GGA GAG T) and the 0.6-kb product as the reverse
megaprimer. The 2.1-kb PCR product synthesized was digested with
AflII and BstEII to produce a 1.9-kb fragment. The wild type
␣
1C
cDNA
was similarly digested with AflII and BstEII, and the mutagenic frag-
ment was then ligated into the
␣
1C
cDNA. The introduced mutations
were both sequenced and verified; the S1371A mutant also had an
additional mutation G1375A outside the consensus phosphorylation
sequence (see Fig. 4).
For making RNA, the cDNAs for
␣
1C
(in pcDNA3) and

1
(in
pSPORTS2) were first linearized by Asp718 and SalI, respectively, and

2
(in pBluescript or pcDNA3) and
␣
2
␦
-1 (in pcDNA3) were linearized by
NotI. Capped cRNAs were synthesized in vitro under T7 polymerase
using MEGAscript (Ambion). The RNA concentration was estimated
using a formaldehyde gel, comparing with standard markers.
Xenopus Oocyte Culture and cRNA Microinjection—The preparation
of oocytes from Xenopus laevis frogs, microinjection of RNA, and main-
tenance of injected oocytes have been detailed elsewhere (42). In brief,
each oocyte was injected with 50 nl containing cRNA for
␣
1C
alone (⬃25
ng) or together with 8 ng of RNA for

1
or

2
. The
␣
2
␦
-1 subunit was not
normally co-injected, except where specifically stated in the text
(amounts of RNA injected:
␣
1C
,15ng;

2
, 5.1 ng; and
␣
2
␦
, 7.5 ng).
Oocytes were incubated at 19.8 °C for 2–3 days in modified Barth’s
solution (88 mMNaCl, 1 mMKCl, 2.4 mMNaHCO
3
, 0.82 mMMgSO
4
,
0.33 mMCa(NO
3
)
2
, 0.4 mMCaCl
2
, 7.5 mMTris-HCl, pH 7.6, 10
4
units/
liter penicillin, and 10 mg/liter streptomycin) before examining the
expression of Ca
2⫹
channels.
Electrophysiological Recording—The oocytes were held in a 50-
l
recording chamber and perfused by a high barium solution containing
40 mMBa(OH)
2
,50mMNaOH, 2 mMKOH, and 5 mMHEPES, pH 7.4,
with methanesulfonic acid (22–25 °C). The calcium channel currents
using Ba
2⫹
as charge carrier were measured with the two-electrode
voltage clamp using the Geneclamp500 amplifier (Axon Instruments)
as described previously (42). Currents were filtered at 2 kHz and
sampled at 4 kHz by using a CED1401Plus interface with CED data
acquisition software. The membrane potential of oocytes was held at
⫺80 mV. To construct the current-voltage relationships, Ba
2⫹
currents
were elicited by a series of 200-ms depolarizing pulses every 10 s, from
⫺70 mV to ⫹60 mV in 10-mV increments, followed by 20 200-ms
hyperpolarizing pulses every 2 s stepping to ⫺90 mV for subsequent
leak and capacitative current subtractions. In experiments investigat-
ing the effects of 8-bromo-cyclic GMP (Sigma) or KT5823 (Calbiochem),
200-ms depolarizing pulses to ⫹10 mV were applied every 30 s (or
exceptionally where stated every 10 s) from the holding potential of ⫺80
mV; the agents were then applied by continuous superfusion while
repetitively stimulating. In some experiments, the oocytes were per-
fused first with KT5823 for 15–20 min before application of 8-Br-cGMP
as above. 8-Br-cGMP was directly dissolved in the bathing solution
prior to use, whereas KT5823 was diluted from 1 mMstock in Me
2
SO to
the specified final concentration (final concentration of Me
2
SO was
always less than 0.1% v/v). Experiments using 8-Br-cGMP were carried
out in a darkened room to prevent breakdown. All data were presented
as the means ⫾S.E. unless otherwise stated. Comparisons between
means were made using the Student’s ttest, with a significance level at
p⬍0.05.
COS Cell Culture and Transfection with Calcium Channel Sub-
units—COS-1 cells were grown in Dulbecco’s modified Eagle’s medium
containing 10% (v/v) fetal bovine serum, 100 units/ml penicillin, 100
g/ml streptomycin, and 4 mMglutamine and maintained in a 5% CO
2
atmosphere at 37 °C prior to their use. COS cells were transfected using
the Lipofectin (Life Technologies, Inc.) transfection procedure. For this,
cells were seeded to a confluency of ⬃20% 24 h prior to transfection. For
each plate of cells, cDNA (5
gof
␣
1C
and 5
gof

2
) was diluted in 2 ml
of OPTI-MEM medium and then mixed by agitation for 15 min with 25
g of Lipofectin also suspended in 2 ml of OPTI-MEM medium. Cells
were washed twice with OPTI-MEM medium and then incubated at
37 °C in 5% CO
2
with the DNA-Lipofectin mix for 12 h. Culture medium
was then replaced with Dulbecco’s modified Eagle’s medium containing
10% fetal bovine serum, and cells were allowed to recover for a further
48 h prior to their use.
Preparation of Antibody against the
␣
1C
Subunit—Rabbit polyclonal
antibody against an
␣
1C
subunit fragment (comprising IIS6 and the
II/III linker) was prepared as follows. A GST-
␣
1C
fusion protein was
made by subcloning the cDNA encoding amino acids Asn
739
–His
896
(between EcoRI and AflII sites) of the rabbit
␣
1C
subunit into the
pGEX-3X vector at the SmaI site of the vector multiple cloning se-
quence. The construct was expressed in E. coli and purified by batch
adsorption to glutathione-agarose beads, and the eluted fusion protein
was used to immunize rabbits. Anti-GST antibody was removed from
the rabbit serum by adsorption to GST on nitrocellulose and checked by
Western blotting.
Immunoprecipitation and Western Blot Detection of the
␣
1C
Sub-
unit—Transfected COS cells expressing the
␣
1C
and

2
subunits were
lysed by homogenization at 4 °C in 1 ml of RIPA lysis buffer containing
10 mMTris, 1 mMEDTA, 1% (v/v) Nonidet P-40, 0.5% (w/v) sodium
deoxycholate, 1 mMphenylmethylsulfonyl fluoride, and 0.1% (w/v) SDS,
pH 7.5. This was then incubated for1hat4°Cwith anti-
␣
1c
serum (10
l) conjugated to protein A-Sepharose. The antibody-protein complexes
were then collected by centrifugation at 16,000 ⫻gand washed three
times with RIPA buffer. Immunoprecipitated proteins were resolved
using SDS-polyacrylamide gel electrophoresis and Western transferred
onto nitrocellulose filters. Filters were probed with anti-
␣
1c
antibody
(1:1000 dilution) and detected with anti-rabbit horseradish peroxidase
antibody followed by ECL.
32
P Metabolic Labeling of Oocytes—Oocytes were injected with cRNA,
and the expression of Ca
2⫹
channels was confirmed in samples taken
from each batch using electrophysiological recording prior to their use
in labeling experiments. Batches of 50 oocytes were pooled and incu-
Calcium Channel Regulation by PKG6136
by guest, on February 19, 2013www.jbc.orgDownloaded from
bated for3hat25°Cwith [
32
P]inorganic phosphate (1 mCi/5 ml) in
modified Barth’s solution. After incubation, oocytes were washed twice
and resuspended in fresh Barth’s solution and then reincubated in the
dark in the presence or absence of 8-Br-cGMP (1 mM) for 20 min.
Oocytes were then washed twice with fresh Barth’s solution and then
resuspended in RIPA lysis buffer. The lysed oocytes were subjected to
preclearing by incubation for1hat4°Cwith preimmune rabbit serum
that had been preconjugated to protein A-Sepharose, followed by sepa-
ration by centrifugation. Lysates were precleared three times prior to
immunoprecipitation with the anti-
␣
1C
subunit antibody as described
above for COS cells.
RESULTS
Characterization of Wild Type Ca
2⫹
Channels Expressed in
Xenopus Oocytes—Before studying the effects of cGMP, the
characteristics of the wild type channels expressed in oocytes
were determined. Inward currents were observed for oocytes
injected with
␣
1C
subunit alone or in combination with either

1
or

2
subunit. It can be seen (Fig. 1) that currents were small
for the
␣
1C
subunit, whereas the additional presence of either

1
or

2
subunit produced larger inward currents, which
shifted the I-V curve to the left. The presence of the

subunits
led to faster activation and to some inactivation in the case of

1
(but hardly any inactivation for

2
) (Fig. 1). The sensitivity
of the expressed calcium channel currents to dihydropyridines
was checked using the agonist BAY K8644 for oocytes injected
with
␣
1C

2
subunits. It can be seen from Fig. 1Ethat BAY
K8644 gave the expected increase in current and shift to the
left of the I-V curve. In uninjected oocytes there were no ap-
preciable inward currents. In summary, the data reported in
this section indicate that the injected
␣
1C
and

subunits be-
haved in our hands in a similar manner to previous reports (3,
4, 17), constituting an L-type calcium channel with expected
voltage dependence and kinetics.
Effects of 8-Br-cGMP on Wild type Ca
2⫹
Channel Cur-
rents—To study the effects of phosphorylation by protein ki-
nase G, the membrane-permeable cGMP analogue, 8-Br-cGMP,
was applied by perfusion in the bath. The roles of the different
subunits in phosphorylation were examined by applying this
reagent to oocytes expressing either
␣
1C
subunit alone or in
combination with

1
or

2
subunits.
During repetitive depolarizations to ⫹10 mV from a holding
potential of ⫺80 mV, oocytes were continuously perfused with
8-Br-cGMP. For oocytes expressing the
␣
1C
subunit alone,
there was a reduction in calcium channel currents by 8-Br-
cGMP (Fig. 2A). Reductions in calcium channel currents were
also observed when 8-Br-cGMP was applied in the same man-
ner to oocytes expressing either
␣
1C

1
subunits or
␣
1C

2
subunits (Fig. 2, Band C). The extent of reduction in currents
produced by 8-Br-cGMP was similar for both
␣
1C

1
and
␣
1C

2
combinations. The effect seemed to be greater in the case of
␣
1C
subunit alone, but this did not attain statistical significance.
Example currents are also displayed in Fig. 2 (insets), which
show that there were no obvious effects of 8-Br-cGMP on the
activation and inactivation kinetics of the Ca
2⫹
channel
currents.
Further experiments were carried out to investigate whether
the above effects are indeed caused via the activation of protein
kinase G or by some other mechanism (such as direct 8-Br-
cGMP action). For this, oocytes expressing
␣
1C

2
subunits
FIG.1. Properties of wild type Ca
2ⴙ
channel subunits expressed in oocytes. A—C, example currents are shown for recordings from
oocytes expressing
␣
1C
subunit alone,
␣
1C

1
subunits, and
␣
1C

2
subunits, respectively. Currents were elicited by the depolarizing steps to the
indicated potentials from a holding potential of ⫺80 mV. Oocytes were injected with cRNA for
␣
1C
subunit alone or in combination with

1
or

2
subunit, and the recordings were made 2–3 days post injection using 40 mMBa
2⫹
as charge carrier. D, the I-V curves for peak current amplitudes
are shown for calcium channel currents for the
␣
1C
subunit alone (f,n⫽8),
␣
1C

1
subunits (,n⫽6), and
␣
1C

2
subunit (Œ,n⫽17). The I-V
curve for the endogenous Ca
2⫹
channels in uninjected oocytes (●,n⫽18) is also shown. E, the I-V curves for peak current amplitudes are shown
for calcium channel currents for
␣
1C

2
subunits before (Œ) and 5 min after application of BayK 8644 (1
M)(⽧,n⫽7). Currents elicited at each
test potential were normalized to the value obtained at ⫹10 mV before application of BayK8644 for each cell. The mean ⫾S.E. averaged over the
cells is shown throughout, in this and subsequent figures.
Calcium Channel Regulation by PKG 6137
by guest, on February 19, 2013www.jbc.orgDownloaded from
were pretreated with the protein kinase G inhibitor, KT5823
(43). Fig. 3Ashows that the inhibitor alone had no effect on
calcium channel currents, but, as can be seen from Fig. 3B, the
reduction in current by 8-Br-cGMP was essentially abolished
by previous incubation with the inhibitor, KT5823. The I-V
curves shown in Fig. 3Chave been normalized at ⫹10 mV and
show that neither 8-Br-cGMP nor the inhibitor had an effect on
the voltage dependence of the calcium channel currents. In
summary, the results in this section indicate that 8-Br-cGMP
caused reductions in wild type calcium channel currents
whether or not a

subunit was present and that the effect is
indeed due to modulation by protein kinase G.
Effect of Point Mutations at Consensus Phosphorylation Sites
of the
␣
1C
Subunit—Because the data so far described suggest
a role for the
␣
1C
subunit in phosphorylation mediated by
protein kinase G, we have therefore analyzed the amino acid
sequence of the
␣
1C
subunit for potential PKG consensus se-
quences. Based on previously recognized PKG consensus se-
quences (44) together with a basic arginine residue down-
stream (45), two potential PKG phosphorylation sites were
identified at serine 533 and serine 1371 of the rabbit
␣
1C
subunit used in the present studies (Fig. 4). Each of these
serines was mutated to alanine (which introduces only a sub-
stitution of -OH by -H), and the effect of these mutations on the
regulation of the Ca
2⫹
channel by 8-Br-cGMP was investigated.
The S1371A/G1375A mutant
␣
1C
subunit was first investi-
gated and coexpressed with the

2
subunit in oocytes. Channel
currents for this mutant were similar in magnitude (data not
shown) and exhibited similar voltage dependence to wild type
currents (normalized I-V curves shown in Fig. 5B). Perfusion of
the cells with 8-Br-cGMP resulted in a significant reduction
(p⬍0.05) of Ca
2⫹
channel currents (Fig. 5A), as for wild type.
Normalized I-V curves showed that this effect was voltage-
independent (Fig. 5C), also as for wild type. Again, pretreat-
ment of cells with the PKG inhibitor KT5823 reduced the effect
of 8-Br-cGMP (Fig. 5B). Clearly these experiments show that
mutation of serine 1371 (and glycine 1375) are not concerned
with functional effects of phosphorylation by protein kinase G.
The mutant S533A
␣
1C
subunit was also co-expressed with
the

2
subunit in oocytes. Again the calcium channel currents
were similar in magnitude (data not shown) and voltage de-
pendence (normalized I-V curve, Fig. 6B). However, as can be
seen from Fig. 6A, 8-Br-cGMP perfusion on oocytes expressing
this mutant did not inhibit the Ca
2⫹
channel currents (Fig. 6A).
Furthermore, the I-V curves for mutant currents before and
after 8-Br-cGMP treatment were not significantly different
(Fig. 6B). The results for this mutant indicate that (in contrast
to residue Ser
1371
) residue Ser
533
on the
␣
1C
subunit is essen-
tial for cGMP-induced regulation of calcium channel currents
via phosphorylation by protein kinase G.
Effects of Other Subunits on the 8-Br-cGMP-dependent Chan-
nel Activity—All the previous recordings were made in the
absence of the
␣
2
␦
subunit to eliminate any complications from
the latter subunit. However, to investigate a more “native”
subunit combination, experiments were repeated in the pres-
ence of the
␣
2
␦
subunit. Fig. 7 (Aand B) shows the effect of
application of 8-Br-cGMP to oocytes expressing
␣
1C
,

2
, and
␣
2
␦
subunits. For wild type currents, inhibition by 8-Br-cGMP was
observed, whereas there was no inhibition for the S533A mu-
tant (Fig. 7A). Normalized I-V curves for wild type currents
showed that there was no effect of 8-Br-cGMP on the voltage
dependence of the inhibition (Fig. 7B). These results are simi-
Br-cGMP for the
␣
1C

1
subunit shown here were not significantly
different from those shown in Afor the
␣
1C
subunit alone or from those
shown in Bfor the
␣
1C

2
subunits.
FIG.2.The inhibitory effect of 8-Br-cGMP on wild type Ca
2ⴙ
channel currents. A, calcium channel currents were recorded during
repetitive stimulation by depolarizing pulses (every 10 s) to ⫹10 mV
from a holding potential of ⫺80 mV for oocytes injected with
␣
1C
subunit
alone and perfused in the presence (f, initial current ⫽⫺0.026 ⫾0.014
A, n⫽5) or absence (䡺,n⫽5) of 8-Br-cGMP (1 mM). For each oocyte,
the peak current amplitudes were averaged over 1-min periods and
then normalized to the value obtained for the first minute of the re-
cording. Currents were significantly reduced (p⬍0.05, t⬎6 min) by
8-Br-cGMP as compared with untreated control, but the reduction was
not statistically significant when compared with the initial size of the
currents. Example currents before and 15 min after application of
8-Br-cGMP are shown in the inset.B, this figure shows calcium channel
currents obtained as in Aabove using
␣
1C

2
subunits, in the presence
(, initial current -0.144 ⫾0.031
A, n⫽5) or absence (ƒ,n⫽4) of
8-Br-cGMP (1 mM). Currents were significantly decreased by 8-Br-
cGMP (p⬍0.05, t⬎7 min)) when compared with untreated controls or
with initial current values. Currents in the presence of 8-Br-cGMP for
the
␣
1C

2
subunit shown here were not significantly different from
those shown in Afor the
␣
1C
subunit alone. C, this figure shows calcium
channel currents obtained as in Aabove using
␣
1C

1
subunits, in the
presence (Œ, initial current -0.293 ⫾0.084
A, n⫽4) or absence (‚,n⫽
7) of 8-Br-cGMP (1 mM). Currents were significantly decreased by
8-Br-cGMP (p⬍0.05, t⬎7 min) when compared with untreated
controls or with initial current values. Currents in the presence of 8-
Calcium Channel Regulation by PKG6138
by guest, on February 19, 2013www.jbc.orgDownloaded from
lar to those obtained in the absence of the
␣
2
␦
subunit, except
that initial currents were several times larger in the presence
of the
␣
2
␦
subunit (see figure legends for detailed current
magnitude measurements), even though less
␣
1C
subunit was
injected into the oocytes (see “Experimental Procedures”).
Another subunit that might complicate the interpretation of
our results is the endogenous
␣
1
subunit found in oocytes. It is
known that expressed

subunit can combine with endogenous
␣
1
subunits in oocytes to produce calcium channel currents
(46), and this could in principle contribute to the inhibition
observed with 8-Br-cGMP. When the

2
subunit alone was
injected into oocytes, there was indeed a small measurable
calcium current (⫺0.084 ⫾0.012
A, n⫽3). However, this
current was not sensitive to 8-Br-cGMP (Fig. 7C), showing that
the effects of 8-Br-cGMP observed in this study for injected
cells are not due to effects on such endogenous currents.
Studies Using
32
P Labeling to Assess Phosphorylation of the
␣
1C
Subunit—The above studies, which examine effects on
function, have strongly indicated that phosphorylation of the
␣
1C
subunit by PKG is the mechanism of action of 8-Br-cGMP.
Here we have attempted to evaluate directly the effect of 8-Br-
cGMP on the phosphorylation state using
32
P labeling. For this,
we raised a polyclonal antibody against the
␣
1C
subunit and
used it as a tool to detect
32
P-labeled subunit before and after
treatment with 8-Br-cGMP.
The specificity of the antibody was tested by transiently
expressing the
␣
1C
subunit in COS cells, followed by immuno-
FIG.3.Effect of the PKG inhibitor KT5823 on wild type Ca
2ⴙ
channel currents either alone or with 8-Br-cGMP. A, calcium
channel currents were recorded during repetitive stimulation by depo-
larizing pulses (every 30 s) to ⫹10 mV from a holding potential of ⫺80
mV for oocytes injected with
␣
1C

2
subunits and perfused with KT5823
(3
M)(䉫,n⫽3) as indicated by the bar. Peak current amplitudes were
normalized to the value (⫺0.178 ⫾0.023
A) at time 0. The current was
not significantly affected during the application of KT5823 as compared
with initial values. B, calcium channel currents were recorded and
plotted as in Afor oocytes superfused with 8-Br-cGMP (1 mM)(Œ, initial
current ⫽⫺0.39 ⫾0.16
A, n⫽5) alone, as shown by the bar. To test
the action of KT5823 (1
M), this inhibitor was applied 15 min previous
to, as well as during, the application of 8-Br-cGMP (1 mM)(⽧, initial
current ⫽⫺0.32 ⫾0.10
A, n⫽6). The inhibitor blocked the effect of
8-Br-cGMP (significant differences between the points of the two curves
shown, t⬎10 min, p⬍0.05), and indeed in the presence of the
inhibitor, 8-Br-cGMP showed no significant reduction in currents. In
the absence of the inhibitor, 8-Br-cGMP caused a significant reduction
(p⬍0.05, t⬎10 min) in currents as compared with initial values. C,
this shows I-V curves for calcium channel currents for
␣
1C

2
subunits
before application of reagents (‚,n⫽12), 15 min after KT5823 (3
M)
alone (ƒ,n⫽3), 10 min after 8-Br-cGMP (1 mM) with (⽧,n⫽6) or
without (〫,n⫽5) pretreatment with KT5823 (1
M). Currents at each
test potential were normalized to the value obtained at ⫹10 mV. There
were no significant differences between any of the corresponding points
on the I-V curves.
FIG.4.Consensus sequences for protein kinase G on the rabbit
␣
1C
subunit. A, schematic diagram of the
␣
1C
subunit of the rabbit
cardiac L-type Ca
2⫹
channel. The interaction domain on the
␣
1C
subunit
for

subunit binding (located on the I-II cytoplasmic linker) is shown
(f). The location of the two consensus phosphorylation sequences for
PKG are also indicated at Ser
533
(●) and Ser
1371
(Œ); these were mu-
tated to alanines. B, alignment of the amino acid regions around the two
consensus phosphorylation sequences for PKG on the rabbit
␣
1C
sub-
unit (cardiac L-type, used in the present study) accession number
X15539, along with the corresponding counterparts from rabbit
␣
1A
(P/Q-type) accession number X57477, human
␣
1B
(N-type) accession
number M94172, rat
␣
1E
(R-type) accession number L15453, rat
␣
1G
(T-type) accession number AF 027984, human
␣
1H
(T-type) accession
number AF051946, rabbit
␣
1S
(skeletal muscle L-type) accession num-
ber P07293, carp
␣
1S
(skeletal muscle L-type) accession number
PIR103831, human
␣
1D
(neuronal L-type) accession number PIR87002,
rat
␣
1C
(neuronal L-type) accession number M67515, and mouse
␣
1C
subunits accession number U17869 (cardiac L-type). The consensus
sequences for PKG are shown in the last line, and the corresponding
residues in the above
␣
1
subunits are shaded.
Calcium Channel Regulation by PKG 6139
by guest, on February 19, 2013www.jbc.orgDownloaded from
precipitation and Western blot detection of this protein by the
antibody. COS cells were used rather than oocytes because of
the low levels of expression in oocytes. As can be seen (Fig. 8A),
the antibody immunoprecipitated and detected a specific band
corresponding to the size expected for the
␣
1C
subunit protein
(lane 1) that is absent for the vector control-transfected cells
(lane 2). Thus the antibody is a useful tool for the next stage of
the investigation, which involved the isolation of
32
P-labeled
␣
1C
subunit in oocytes.
Oocytes were injected with cRNA for
␣
1C
and

2
subunits (or
uninjected as controls), metabolically labeled with [
32
P]inor-
ganic phosphate, and washed, followed by incubation in the
presence or absence of 8-Br-cGMP (1 mM) for 20 min. The
anti-
␣
1C
subunit antibody was used to immunoprecipitate the
expressed protein, which was resolved by SDS-polyacrylamide
gel electrophoresis, and
32
P labeling detected using x-ray film.
Fig. 8Bshows labeled bands for injected cells (lanes 1 and 2)
that were absent for controls (lane 3) and of the expected size
for the
␣
1C
subunit. There was also an additional labeled band
of lower molecular weight that could be a proteolytic break-
down product or an
␣
1C
subunit co-complexing protein. There
was no significant difference in extent of labeling of either of
these bands in the presence (lane 2) and absence of 8-Br-cGMP
(lane 1). Even in the absence of 8-Br-cGMP, considerable phos-
phorylated protein was detected, which may underlie the in-
ability to detect changes in phosphorylation by this methodol-
ogy. These experiments were also repeated in the additional
presence of the
␣
2
␦
subunit, but qualitatively similar results
were observed (data not shown).
DISCUSSION
In this study we have examined the molecular mechanism of
the effect of 8-Br-cGMP on the calcium channel activity of the
rabbit
␣
1C
subunit alone or when co-expressed with

1
or

2
subunit in Xenopus oocytes. In summary, we have shown that
8-Br-cGMP inhibits this L-type channel whether or not a

subunit is present and that the abolition of this effect by a
protein kinase G inhibitor indicates an action via protein ki-
nase G. Mutation of serines on the
␣
1C
subunit at consensus
phosphorylation sites for PKG showed abolition of the 8-Br-
cGMP effect by mutation S533A (without other effects on chan-
nel function) but not by mutation S1371A, indicating that
phosphorylation of Ser
533
underlies the functional effect of
8-Br-cGMP. Co-injection of RNA for
␣
2
␦
subunit together with
the above subunits produced larger currents but showed the
same general features as in its absence: an inhibition by 8-Br-
cGMP for wild type channel and no effect for S533A mutant,
indicating that the
␣
2
␦
subunit plays no additional role in this
process.
Previous studies by others have suggested a role for cGMP
and PKG activation in the regulation of Ca
2⫹
channel currents
in cells from isolated tissues, such as heart, vascular smooth
muscle, skeletal muscle, and nerve (9, 24, 31, 33–35, 47–49).
The effects of cGMP are usually, but not always, inhibitory.
However, the target and molecular mechanisms have not pre-
viously been established, and the present study is the first to
investigate molecular mechanisms of the functional effect of
cGMP in cloned calcium channels. Biochemical studies have
already shown for example that neuronal
␣
1C
subunit can be
phosphorylated by PKG in vitro (8). We have demonstrated
here that 8-Br-cGMP has the ability to significantly reduce
calcium channel current via an action on the
␣
1C
subunit alone.
The PKG phosphorylation site, which we have found at Ser
533
is located on the cytoplasmic linker between domains I and II
(Fig. 4). Interestingly, this position is located very near to the
site of binding of the

subunit (around residues 458–475 of the
rabbit
␣
1C
subunit). Intriguingly, it is tempting to speculate
that phosphorylation of residue 533 may cause local conforma-
tional changes that may affect binding or function of the

subunit. This might be expected to lead to a shift in the I-V
curve by 8-Br-cGMP given the shift in the I-V curve that the

FIG.5. Effect of 8-Br-cGMP on calcium channel currents for
the S1371A
␣
1C
mutant. A, calcium channel currents are shown dur-
ing perfusion of oocytes expressing S1371A mutant
␣
1C
and

2
subunit
with (●, initial current ⫺0.223 ⫾0.072
A, n⫽5) or without 8-Br-
cGMP (1 mM)(䡺, initial current ⫺0.24 ⫾0.10
A, n⫽4). Currents were
produced by depolarizing pulses to ⫹10 mV from a holding potential of
⫺80 mV every 10 s. Peak current amplitudes were normalized to the
value at time 0. Currents were significantly reduced when compared
with the value before application of 8-Br-cGMP or with time-matched
control (p⬍0.05). B, mutant calcium channel currents are shown for
the action of 8-Br-cGMP (1 mM) applied as indicated by the bar in
oocytes treated with KT5823 (1
M) for at least 15 min before as well as
during the application of 8-Br-cGMP (⽧, initial current ⫺0.362 ⫾0.046
A, n⫽4). Depolarizing pulses to ⫹10 mV were applied from a holding
potential of ⫺80 mV every 30 s. There was no significant effect of
8-Br-cGMP in the presence of KT5823 as compared with initial values.
C, this shows I-V curves for mutant S1371A
␣
1
and

2
currents before
(E,n⫽5), and 10 min after (●,n⫽5) application of 8-Br-cGMP. The
I-V curve for the wild type
␣
1C

2
currents, reproduced from Fig. 3C(Œ,
n⫽12), is included here for comparison. Currents at each test potential
were normalized to the value obtained at ⫹10 mV for each oocyte. There
were no significant differences between any of the corresponding points
on the I-V curves.
Calcium Channel Regulation by PKG6140
by guest, on February 19, 2013www.jbc.orgDownloaded from
subunit produces; however, no such shifts in I-V curves were
obtained by the action of 8-Br-cGMP, possibly arguing against
such a mechanism. However, our data did hint that the effect of
8-Br-cGMP was reduced in the presence of the

subunit (as
compared with the
␣
1C
subunit alone), but experimental scatter
obtained for the small currents produced by
␣
1C
alone did not
allow us to substantiate this.
Our data therefore indicate that it is PKG-dependent phos-
phorylation of the
␣
1C
subunit that leads to functional effects
rather than phosphorylation of the

subunit. However, there
are consensus PKG phosphorylation sequences on the

sub-
units (39), and in vitro phosphorylation of the

1
subunit by
PKG has been demonstrated (38), but this seems not to involve
effects on calcium channel function. Furthermore, the effects
reported here of 8-Br-cGMP were similar whether the

1
or

2
subunit was present (as well as the
␣
1C
subunit); differences
might have been expected if phosphorylation of the

1
subunit
had been involved, given the different potential PKG (consen-
sus) sites on the two

subunits used in the present study.
Although the effects of 8-Br-cGMP in the present study were
not large, they were always reproducible, as could be most
clearly seen when currents had been increased severalfold by
␣
2
␦
subunit expression. Furthermore, our results were consist-
ent in that no run-down was observed in control experiments
without 8-Br-cGMP. We have also shown that in oocytes in-
jected with

2
subunit alone, the calcium channel current
(which arises by association with endogenous
␣
1
subunits) is
insensitive to 8-Br-cGMP. For oocytes injected with
␣
1C
and

2
subunits, currents were approximately three times larger than
for oocytes injected with

2
alone. This suggests that, although
appreciable endogenous current would be present in cells in-
jected with
␣
1C
and

2
subunits, this would not contribute to
the 8-Br-cGMP action but would in fact have the effect of
making the inhibition appear smaller than it actually is. Even
taking this into account, the extent of inhibition by cGMP in
oocytes is still somewhat smaller than that observed in native
cells, e.g. cardiac (31, 35, 48) and neuronal cells (49). One
possible reason for this could be the fact that the magnitude of
the effect of cGMP in some native cells is dependent on resting
phosphorylation levels by cAMP (35, 48), with decreased rest-
ing cAMP-phosphorylation leading to decreased effect of cGMP.
However, this mechanism may not entirely account for the
smaller effect because resting cAMP-dependent phosphoryla-
tion is high in oocytes (26, 42). Another possible explanation for
the smaller currents in oocytes may be the requirement of
additional proteins (absent in oocytes) for the full effect of
cGMP. Indeed, in the case of PKA-mediated phosphorylation of
calcium channels, it is known that protein kinase A anchoring
proteins are additionally required in heterologous expression
systems to observe the full regulatory effect of cAMP/PKA (28,
50). It is therefore conceivable that similar anchoring or regu-
latory proteins may be necessary for a full cGMP/PKG effect.
Other mechanisms are also possible as explanations of the
smaller effect in oocytes, such as the possibility that PKG levels
are low in oocytes (51).
Although our electrophysiological data strongly imply that
the molecular mechanism by which cGMP regulates the cardiac
L-type Ca
2⫹
channel is indeed phosphorylation, the low expres-
sion of calcium channels and the small extent of the cGMP
effect make biochemical studies of cGMP-dependent phospho-
rylation difficult. Here we have used [
32
P]inorganic phosphate
labeling of oocytes expressing the channel, together with im-
munoprecipitation using anti-
␣
1C
antibodies. We were able to
observe high levels of resting phosphorylation (in agreement
with other studies (26, 42) referred to above). However, this
high basal phosphorylation level (presumably including phos-
phorylation by PKA, PKC, and other kinases) made it impos-
sible for us to detect the small additional cGMP dependent
phosphorylation. Nevertheless, the electrophysiological studies
of channel function reported here are clearly more sensitive
and have led to strong support for a cGMP-dependent phospho-
rylation mechanism.
The inhibitory effects of cGMP via phosphorylation by PKG
of the cloned L-type cardiac
␣
1C
subunit are supported by sim-
ilar reports for isolated cardiac tissues (24, 31, 34, 35, 48) and
indeed also for L-type calcium channels in vascular tissues (9).
The sequence alignments shown in Fig. 4 indicate the presence
in the
␣
1D
subunit of a serine residue (at amino acid 502 in the
human
␣
1D
) correspondingly positioned to 533 in the rabbit
␣
1C
subunit. Therefore, we would speculate that this could also be
a PKG phosphorylation site in
␣
1D
and that this neuronal
L-type channel could therefore be under the regulatory control
of cGMP via protein kinase G. Indeed an inhibitory action has
already been reported on a neuronal (rat pinealocyte) L-type
channel via PKG (49). Although skeletal muscle
␣
1S
subunit is
L-type, it lacks the serine residue corresponding to 533 of
␣
1C
(Fig. 4), suggesting that inhibitory cGMP action via PKG does
not occur. Interestingly, for calcium channels in isolated skel-
etal muscle tissues, cGMP produces an increase rather than a
decrease in calcium channel currents, presumably via phospho-
rylation elsewhere or via an unrelated mechanism (52). The
remaining sequences shown in Fig. 4,
␣
1A
(P/Q-type),
␣
1B
(N-
type),
␣
1E
(R-type),
␣
1G
(T-type), and
␣
1H
(T-type) do not have
the corresponding consensus sequence around Ser
533
of
␣
1C
(indicating absence of PKG effects at this site), but other con-
sensus sites for PKG phosphorylation are present, so further
investigation of possible regulation by cGMP of non-L-type
channels would be interesting. Existing data from isolated
FIG.6.Effect of 8-Br-cGMP on calcium channel currents for the S533A
␣
1C
mutant. A, calcium channel currents are shown during
perfusion of oocytes expressing 533A mutant
␣
1C
and

2
subunits with 8-Br-cGMP (1 mM)(●, initial current -0.184 ⫾0.069
A, n⫽5). Currents
were produced by depolarizing pulses to ⫹10 mV from a holding potential of ⫺80 mV every 30 s. Peak current amplitudes were normalized to the
value at time 0. Currents were not significantly changed as compared with initial values. Where not shown, error bars are smaller than the data
point. B, this shows I-V curves for mutant S533A
␣
1
and

2
currents before (E,n⫽15) and 10 min after (●,n⫽3) application of Br-cGMP. The
I-V curve for the wild type
␣
1C

2
currents, reproduced from Fig. 3C(Œ,n⫽12), is included here for comparison. Currents at each test potential
were normalized to the value obtained at ⫹10 mV for each oocyte. There were no significant differences between any of the corresponding points
on the I-V curves.
Calcium Channel Regulation by PKG 6141
by guest, on February 19, 2013www.jbc.orgDownloaded from
tissues for non-L-type channels are limited, although it does
appear that T-type channels are not regulated by PKG (9).
Furthermore, as reported in the present paper, currents in
oocytes injected with

subunit alone showed no sensitivity to
8-Br-cGMP, consistent with the reported nature of such endog-
enous oocyte channels as N and T type (46).
Calcium channels can also be regulated directly by G pro-
teins (G
␥
subunits). The calcium channel

subunit binds to a
region on the
␣
1
subunit (53, 54), which overlaps with the
binding region for G proteins, and it might at first sight be
thought that phosphorylation at the nearby Ser
533
site could
affect G protein binding/modulation. However, this interaction
may not occur in practice; L-type channel
␣
1C
subunits do not
possess a G
␥
binding site (55), and the other types of channel
do not possess the PKG consensus sequence pattern shown in
Fig. 4 at the
␣
1C
Ser
533
site (although weak PKG consensus
patterns are present elsewhere).
Our conclusion that protein kinase G causes its functional
effect via phosphorylation of the
␣
1C
subunit rather than the

subunit complements other studies (25–28) that have shown
functional effects via the
␣
1
subunit of PKA and PKC. The
activation of PKA by cAMP leads to increases in calcium chan-
nel currents in cardiac muscle, specifically by phosphorylation
of residue Ser
1928
in the C terminus of rabbit
␣
1C
subunit, but
residue Ser
533
(found here as the site for PKG-dependent phos-
phorylation) was not a target for functional effects of PKA-de-
pendent phosphorylation (26, 28). The activation of PKC can
lead to an increase followed by a decrease in current (25, 56,
57); for rabbit
␣
1C
the N terminus is crucial for the increase
(but not for the decrease) in current. On the other hand for
␣
1A
or
␣
1B
subunits, PKC phosphorylation may occur at the I/II
linker region leading to increased currents; interestingly, in
addition PKC activation leads to decreased effect of G proteins,
which, as already mentioned, also bind to the I/II linker in
␣
1A
or
␣
1B
subunits (53, 54), reminiscent of a similar possible in-
on these I-V curves. C, this shows the effect of 8-Br-cGMP on calcium
channel currents in cells expressing the

2
subunit alone. The same
protocol described for Awas used. Currents were not significantly
altered by 8-Br-cGMP. The initial current before application of 8-Br-
cGMP was ⫺0.084 ⫾0.012
A(n⫽3).
FIG.7. Effect of 8-Br-cAMP on calcium channel currents for
oocytes expressing
␣
1C
/

2
with
␣
2
-
␦
subunits and for the

2
sub-
unit expressed alone. A, calcium channel currents are shown during
perfusion of oocytes expressing wild type (f, initial current ⫺0.86 ⫾
0.17
A, n⫽3) or the S533A mutant (䢇, initial current ⫺0.90 ⫾0.07
A, n⫽4) with 8-Br-cGMP (1 mM). Currents were produced by depo-
larizing pulses to ⫹10 mV from a holding potential of ⫺80 mV every
10 s. Peak current amplitudes were normalized to the value at time 0.
Currents for the wild type subunit were significantly reduced when
compared with the value before application of 8-Br-cGMP or with the
time-matched current recorded from S533A injected oocytes (p⬍0.05).
8-Br-cGMP had no significant effect on currents recorded from the
S533A injected oocytes. The inset shows typical current traces recorded
from an oocyte injected with wild type
␣
1C
,

2
, and
␣
2
-
␦
subunits before
and during application of 8-Br-cGMP. B, I-V curves for the wild type
␣
1C
,

2
, and
␣
2
-
␦
subunit currents taken from the same cells shown in
A, before (䡺) and during (E) application of 8-Br-cGMP. Currents at each
potential were normalized to the value obtained at ⫹10 mV for each
oocyte. There were no significant differences between any of the points
FIG.8.Detection of expression by antibody, and
32
P labeling of
the
␣
1C
subunit. A, this shows detection of
␣
1C
subunit expression in
COS cells. Either
␣
1C
-pcDNA3 (lane 1) or pcDNA3 vector control (lane
2) DNA was transfected into COS cells, and expressed protein was
immunoprecipitated from cell lysates using anti-
␣
1C
antibody. Follow-
ing SDS-polyacrylamide gel electrophoresis resolution and Western
transfer of cellular proteins, blots were probed with the same antibody
followed by incubation with anti-rabbit horseradish peroxidase antibod-
ies and ECL detection. The antibody immunoprecipitated and detected
only one protein of the expected size for the
␣
1C
subunit. B, effect of
8-Br-cGMP on
32
P metabolic labeling of the
␣
1C
subunit in Xenopus
oocytes. Oocytes were either uninjected (lane 3) or injected with cRNA
for
␣
1C
and

2
subunits (lanes 1 and 2). Two days after injection, cells
were incubated with
32
P
i
followed by incubation in the presence (lanes
2and 3) or absence (lane 1) of cGMP. Cells were lysed, and proteins
were immunoprecipitated using anti-
␣
1C
antibodies, and resolved by
SDS-polyacrylamide gel electrophoresis. Labeled proteins were visual-
ized by exposure to x-ray film. Two predominant labeled bands were
detected in the injected cells, and one band corresponds to the expected
␣
1C
subunit size.
Calcium Channel Regulation by PKG6142
by guest, on February 19, 2013www.jbc.orgDownloaded from
teraction between PKG and the

subunit at this region in
␣
1C
(see above).
In conclusion, we have demonstrated here inhibitory effects
of 8-Br-cGMP on cloned L-type cardiac channel currents via
activation of protein kinase G and the implicated phosphoryl-
ation of the
␣
1C
subunit at position Ser
533
. The widespread role
of cGMP in the regulation of cardiac and vascular smooth
muscle L-type calcium channels is apparent, but the possible
role of cGMP as a modulator of neuronal non-L-type calcium
channels is largely unexplored.
Acknowledgments—We thank the British Heart Foundation for sup-
port. We are also grateful to F. Hofmann for providing the
␣
1C
,

1a
, and
␣
2
␦
clones and to L. Birnbaumer for the

2a
clone.
REFERENCES
1. Hille, B. (1992) Ionic Channels of Excitable Membranes, 2nd Ed., pp. 83–114,
Sinauer Associates, Sunderland, MA
2. Mikama, A., Imoto, K., Tanabe, T., Niidome, T., Mori, Y., Takeshima, H.,
Narumiya, S., and Numa S. (1989) Nature 340, 230–233
3. Perez-Reyes, E., Castellano, A., Kim, H. S., Bertrand, P., Baggstrom, E.,
Lacerda, A. E., Wei, X. Y., and Birnbaumer, L. (1992) J. Biol. Chem. 267,
1792–1797
4. Hullin, R., Singer-Lahat, D., Freichel, M., Biel, M., Dascal, N., Hofmann, F.,
and Flockerzi, V. (1992) EMBO J. 11, 885–890
5. Brickley, K., Campbell, V., Berrow, N., Leach, R., Norman, R. I., Wray, D.,
Dolphin, A. C., and Baldwin, S. A. (1995) FEBS Lett. 364, 129–133
6. Hosey, M. M., Chien, A. J., and Puri, T. S. (1996) Trends Cardiovasc. Med. 6,
265–273
7. Gurnett, C. A., Felix, R., and Campbell, K. P. (1997) J. Biol. Chem. 272,
18508–18512
8. Hell, J. W., Yokoyama, C. T., Wong, S. T., Warner, C., Snutch, T. P., and
Catterall, W. A. (1993) J. Biol. Chem. 268, 19451–19457
9. Xiong, Z. L., and Sperelakis, N. (1995) J. Mol. Cell. Cardiol. 27, 75–91
10. Catterall, W. A. (1995) Annu. Rev Biochem. 64, 491–531
11. Larsson, H. P., Baker, O. S., Dhillon, D. S., and Isacoff, E. Y. (1996) Neuron 16,
387–397
12. Yusaf, S. P., Wray, D., and Sivaprasadarao, A. (1996) Pflugers Arch. Eur. J.
Physiol. 433, 91–97
13. Yang, A., George, A. L., and Horn, R. (1996) Neuron 16, 113–122
14. Heinemann, S. H., Terlau, H., Stuhmer, W., Imoto, K., and Numa, S. (1992)
Nature 356, 441–443
15. Yang, J., Ellinor, P. T., Sather, W. A., Zhang, J. F., and Tsien, R. W. (1993)
Nature 366, 158–161
16. Yatani, A., Bahinski, A., Mikala, G., Yamamoto, S., and Schwartz, A. (1994)
Circ. Res. 75, 315–323
17. Singer, D., Biel, M., Lotan, I., Flockerzi, V., Hofmann, F., and Dascal, N. (1991)
Science 253, 1553–1557
18. Neely, A., Wei, X. Y., Olcese, R., Birnbaumer, L., and Stefani E. (1993) Science
262, 575–578
19. Kamp, T. J., Perez-Garcia, M. T., and Marban, E. (1996) J. Physiol. 491, 89–96
20. Pragnell, M., De Waard, M., Mori, Y., Tanabe, T., Snutch, T. P., and Campbell,
K. P. (1994) Nature 368, 67–70
21. De Waard, Scott, V. E. S., Pragnell, M., and Campbell, K. P. (1996) FEBS Lett.
380, 272–276
22. De Waard, M., Pragnell, M., and Campbell, K. P. (1994) Neuron 13, 495–503
23. Trautwein, W., and Hescheler, J. (1990) Annu. Rev. Physiol. 52, 257–274
24. Sperelakis, N., Xiong, Z. L., Haddad, G., and Masuda, H. (1994) Mol. Cell.
Biochem. 140, 103–117
25. Shistik, E., Ivanina, T., Blumenstein, Y., and Dascal, N. (1998) J. Biol. Chem.
273, 17901–17909
26. Perets, T., Blumenstein, Y., Shistik, E., Lotan, I., and Dascal, N. (1996) FEBS
Lett. 384, 189–192
27. De Jongh, K. S., Murphy, B. J., Colin, A. A., Hell, J. W., Takahashi, M., and
Catterall, W. A. (1996) Biochemistry 35, 10392–10402
28. Gao, T. Y., Yatani, A., Dell’Acqua, M. L., Sako, H., Green, S. A., Dascal, N.,
Scott, J., and Hosey, M. M. (1997) Neuron 19, 185–196
29. Waldman, S. A., and Murad, F. (1987) Pharmacol. Rev. 39, 163–196
30. McDonald, L. J., and Murad F. (1996) Proc. Soc. Exp. Biol. Med. 211, 1–6
31. Haddad, G. E., Sperelakis, N., and Bkaily, G. (1995) Mol. Cell. Biochem. 148,
89–94
32. Schulz, R., and Triggle, C. R. (1994) Trends Pharmacol. Sci. 15, 255–259
33. Tohse, N., and Sperelakis, N. (1991) Circ. Res. 69, 325–331
34. Levi, R. C., Alloatti, G., and Fischmeister, R. (1989) Pflugers Arch. Eur. J.
Physiol. 413, 685–687
35. Wahler, G. M., and Dollinger, S. (1995) Am. J. Physiol. 268, C45–C54
36. Hartzell, H. C., and Fischmeister, R. (1986) Nature 323, 273–275
37. Mery, P., Pavoine, S. M., Water, U., and Fischmeister, R. (1993) J. Biol. Chem.
268, 26286–26295
38. Jahn, H., Nastainczyk, W., Rohrkasten, A., Schneider, T., and Hofmann, F.
(1987) Eur. J. Biochem. 178, 535–542
39. Ruth, P., Rohrkasten, A., Biel, M., Bosse, E., Regulia, S., Meyer, H. E.,
Flockerzi, V., and Hofmann, F. (1989) Science 245, 1115–1118
40. Ellis, S. B., Williams, M. E., Ways. N. R., Brenner, R., Sharp, A. H., Leung,
A. T., Campbell, K. P., McKenna, E., Kosh, W. J., Hui, A., Schwartz, A., and
Harpold, M. M. (1988) Science 241, 1661–1664
41. Sarkar, G., and Sommer, S. S. (1990) BioTechniques 8, 404– 407
42. Wilson, G., O’Neill, C. A., Sivaprasadarao, A., Findlay, J. B. C., and Wray, D.
(1994)Pflugers Arch. Eur. J. Physiol. 428, 186–193
43. Grider, J. R. (1993) Am. J. Physiol. 264, G334–G340
44. Pearson, R. B., and Kemp, B. E. (1991) Methods Enzymol. 200, 62–81
45. Edelman, A. M., Blumenthal, D. K., and Krebs, E. G. (1987) Annu. Rev.
Biochem. 56, 567–613
46. Lacerda, A. E., Perez-Reyes, E., Wei, X., Castellano, A., and Brown, A. M.
(1994) Biophys. J. 66, 1833–1843
47. Campbell, D., Stamler, J. S., and Strauss, H. C. (1996) J. Gen. Physiol. 108,
277–293
48. Mery, P., Lohmann, S., Walter, U., and Fischmeister, R. (1991) Proc. Natl.
Acad. Sci. U. S. A. 88, 1197–1201
49. Chik, C. L., Liu, Q. Y., Li, B., Karpinski, E., and Ho, A. K. (1995) J. Neurosci.
15, 3104–3109
50. Johnson, B. D., Brousal, J. P., Peterson, B. Z., Gallombardo, P. A., Hocherman,
G. H., Lai, Y., Scheuer, T., and Catterall, W. A. (1997) J. Neurosci. 17,
1243–1255
51. Alioua, A., Tanaka, Y., Wallner, M., Hofmann, F., Ruth, P., Meera, P., and
Toro, L. (1998) J. Biol. Chem. 273, 32950–32956
52. Kokate, T., Heiny, J. A., and Sperelakis, N. (1993) Am. J. Physiol. 265,
C47–C53
53. Zamponi, G., Bourinet, E., Nelson, D., Nargeot, J., and Snutch, T. (1997)
Nature 385, 442–446
54. De Waard, M., Liu, H. Y., Walker, D., Scott, V. E., Gurnett, C. A., and
Campbell, K. (1997) Nature 385, 446–450
55. Dolphin, A. C. (1998) J. Physiol. 506, 3–11
56. Singer-Lahat, D., Gershon, E., Lotan, I., Hullin, R., Biel, M., Flockerzi., V.,
Hofmann, F., and Dascal, N. (1992) FEBS Lett. 306, 113–118
57. Fournier, F., Charnet, P., Bourinet, E., Vilbert, C., Matifat, F., Charpentier,
G., Navarre, P., Brule, G., and Marlot, D. (1993) FEBS Lett. 317, 118–124
Calcium Channel Regulation by PKG 6143
by guest, on February 19, 2013www.jbc.orgDownloaded from
