Am J Physiol Heart Circ Physiol 291: H1959 –H1971, 2006.
First published April 28, 2006; doi:10.1152/ajpheart.00956.2005.
L-type calcium channel ␣-subunit and protein kinase inhibitors modulate
Rem-mediated regulation of current
Shawn M. Crump,1 Robert N. Correll,2 Elizabeth A. Schroder,1
William C. Lester,1 Brian S. Finlin,2 Douglas A. Andres,2 and Jonathan Satin1
Departments of 1Physiology and 2Biochemistry, University of Kentucky College of Medicine, Lexington, Kentucky
Submitted 7 September 2005; accepted in final form 20 April 2006
CARDIAC L-type Ca channel (CaV1.2) current is a major determinant of excitable cell function, and CaV1.2 function is tightly
regulated by intracellular signaling systems. CaV1.2 is both an
initiator and integrator of diverse signaling pathways. As an
initiator of cell signaling, CaV1.2 regulates cytosolic Ca entry
and cell excitability. As an integrator of cell signaling, a
diverse array of intracellular pathways, including but not limited to PKA, CaM, CaMK, and protein phosphatase 2B, regulate Ca channel function. For example, PKA mediates phos-
phorylation of CaV1.2 on the COOH terminus, leading to a
marked increase of Ca channel current (18).
The cardiac L-type Ca channel complex consists of multiple
subunits, CaV1.2 being the main pore-forming ␣-subunit (8). In
addition, auxiliary subunits in the heart include CaV␤2 and
␣2-␦ subunits. All of the subunits contribute to shaping the
kinetics and voltage dependence of channel gating (8, 16, 28,
29, 32). However, CaV␤2 has the important role of enhancing
Ca channel expression levels in myocytes (12, 51) and in
heterologous expression systems (6, 10, 24, 52). Within the
CaV␤2 family, CaV␤2a most effectively retards voltage-dependent inactivation of current (49). Thus the time course of
current decay can serve as index of the degree of CaV␤
interaction with functioning channels. We recently described
the members of the Rad and Gem/Kir-related (RGK) family of
small G-binding proteins as a novel class of Ca channel
regulators. RGK proteins include Rem, Rem2, Kir/Gem, and
Rad. RGK GTPases bind to CaV␤ (1, 20) and inhibit CaV1
current expression in muscle cells (20) and in heterologous
expression systems (1, 20).
RGK protein overexpression can be used as an extrinsic
mechanism for native cardiac Ca channel block (34). The RGK
protein Rem (19) is expressed in cardiac tissue, yet Rem
coexpressed with CaV␤2 prevents expression of L-type Ca
current (ICa,L) (1, 20). This raises the puzzling question of how
excitable cells maintain ICa,L in the presence of endogenous
RGK proteins such as Rem. One possibility is that Rem-CaV␤
regulation of ICa,L is regulated by intracellular signaling systems. Initial work on RGK inhibition of ICa,L investigated only
CaV␤ interaction (1, 20, 21), and initial studies focused on
RGK-CaV␤ interactions—including possible regulation by
binding to calmodulin (1) or 14-3-3 proteins (3). All of these
studies indicated that RGK regulation of ICa,L involves CaV␤,
but the contribution of CaV1.2 remains to be explored.
PKA modulation is a physiologically important mechanism
of Ca channel function in diverse excitable tissues such as heart
muscle and pancreatic beta cells. PKA increases Ca current via
phosphorylation of a serine residue on the COOH terminus
distal to the calmodulin interaction sites of CaV1.2 (8). Using
pharmacological and site-directed mutagenesis approaches, we
tested the hypothesis that Rem block of Ca channel expression
is not an all-or-none effect but rather a regulated process. Our
results show that the CaV1.2 COOH terminus, in particular,
serine 1928 (S1928) on the COOH terminus, contributes to
Rem reduction of Ca current. This property is recapitulated in
two divergent native cell systems, suggesting a novel and
Address for reprint requests and other correspondence: J. Satin, Dept. of
Physiology, MS-508, Univ. of Kentucky College of Medicine, 800 Rose St.
Lexington, KY 40536-0298 (e-mail: [email protected]).
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked “advertisement”
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
cardiac myocyte; monomeric GTPase; Rad and Gem/Kir-related family; insulin-secreting cell; calcium imaging; heart development
http://www.ajpheart.org
0363-6135/06 $8.00 Copyright © 2006 the American Physiological Society
H1959
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.4 on June 14, 2017
Crump, Shawn M., Robert N. Correll, Elizabeth A. Schroder,
William C. Lester, Brian S. Finlin, Douglas A. Andres, and
Jonathan Satin. L-type calcium channel ␣-subunit and protein kinase
inhibitors modulate Rem-mediated regulation of current. Am J Physiol
Heart Circ Physiol 291: H1959 –H1971, 2006. First published April
28, 2006; doi:10.1152/ajpheart.00956.2005.—Cardiac voltage-gated
L-type Ca channels (CaV) are multiprotein complexes, including
accessory subunits such as CaV␤2 that increase current expression.
Recently, members of the Rad and Gem/Kir-related family of small
GTPases have been shown to decrease current, although the mechanism remains poorly defined. In this study, we evaluated the contribution of the L-type Ca channel ␣-subunit (CaV1.2) to CaV␤2-Rem
inhibition of Ca channel current. Specifically, we addressed whether
protein kinase A (PKA) modulation of the Ca channel modifies
CaV␤2-Rem inhibition of Ca channel current. We first tested the effect
of Rem on CaV1.2 in human embryonic kidney 293 (HEK-293) cells
using the whole cell patch-clamp configuration. Rem coexpression
with CaV1.2 reduces Ba current expression under basal conditions,
and CaV␤2a coexpression enhances Rem block of CaV1.2 current.
Surprisingly, PKA inhibition by 133 nM H-89 or 50 ␮M Rp-cAMP-S
partially relieved the Rem-mediated inhibition of current activity both
with and without CaV␤2a. To test whether the H-89 action was a
consequence of the phosphorylation status of CaV1.2, we examined
Rem regulation of the PKA-insensitive CaV1.2 serine 1928 (S1928) to
alanine mutation (CaV1.2-S1928A). CaV1.2-S1928A current was not
inhibited by Rem and when coexpression with CaV␤2a was not
completely blocked by Rem coexpression, suggesting that the phosphorylation of S1928 contributes to Rem-mediated Ca channel modulation. As a model for native Ca channel complexes, we tested the
ability of Rem overexpression in HIT-T15 cells and embryonic
ventricular myocytes to interfere with native current. We find that
native current is also sensitive to Rem block and that H-89 pretreatment relieves the ability of Rem to regulate Ca current. We conclude
that Rem is capable of regulating L-type current, that release of Rem
block is modulated by cellular kinase pathways, and that the CaV1.2
COOH terminus contributes to Rem-dependent channel inhibition.
H1960
REM GTPASE AND CALCIUM CHANNELS
perhaps universal mechanism for long-term regulation of Ca
current levels by the coordinated actions of RGK GTPases and
the phosphorylation status of excitable cells.
MATERIALS AND METHODS
AJP-Heart Circ Physiol • VOL
291 • OCTOBER 2006 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.4 on June 14, 2017
Adenoviral Rem production. Replication-deficient recombinant adenoviruses coexpressing green fluorescent protein (GFP) and wildtype Rem were constructed as described (20). HIT-T15 cells were
infected at 1 ⫻ 107 placque-forming units/ml and analyzed 24 h
postinfection, a time point that allowed maximal Rem protein expression (21).
Plasmids and mutagenesis. CaV1.2 plasmid (kindly provided by
Dr. T. Kamp, University of Wisconsin) was identical to the originally
cloned full-length rabbit cardiac ␣1C-subunit (33) except for alternative splicing in domain IV S3 (47). CaV␤2a plasmid (kindly provided
by Dr. E. Perez-Reyes, University of Virginia) was identical to rat
(38) and contains the two cysteines residing in the D1 domain
associated with palmitoylation. The CaV1.2 serine 1928 to alanine
(S1928A) mutation was created by using the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) according to the
manufacturer’s instructions, using full-length CaV1.2 cloned into
pCC1GW1H as template. Mutation of S1928 was verified by DNA
sequence analysis. GFP-CaV1.2 was kindly provided by K. Beam
(Colorado State University, Fort Collins, CO) and is fully described
by Grabner et al. (27). For confocal microscopy, cells were fixed 2–3
days after transfection. Cells were washed three times with PBS
supplemented with 1 mM CaCl2 and 1 mM MgCl2 and then fixed in
4% paraformaldehyde on ice for 20 min. Cells were then washed three
times with PBS and mounted on slides with Vectashield. For examination of GFP-CaV1.2 distribution, GFP-CaV1.2 was expressed
alone, coexpressed with either hemagglutinin (HA)-tagged Rem or
Flag-tagged CaV␤2a, or coexpressed with both HA-tagged Rem and
Flag-tagged CaV␤2a. GFP-CaV1.2 was visualized by using a Leica
laser scanning confocal inverted microscope equipped with argon
laser to excite the cells at 488 nm and emission long-pass filtering
⬎505 nm. The cellular distribution of GFP-Rem was examined by
using a similar experimental protocol, except that GFP-Rem was
expressed alone, cotransfected with a vector expressing untagged
CaV1.2 or Flag-tagged CaV␤2a, or coexpressed with both untagged
CaV1.2 and Flag-tagged CaV␤2a.
Electrophysiology. HEK-293 cells were transiently transfected with
plasmids 12– 48 h before recordings were performed, as previously
described (55). Transfected cells were identified by the expression of
GFP. The whole cell configuration of the patch-clamp technique was
used to measure ionic current. Patch electrodes with resistances of 1–2
M⍀ contained (in mM) 110 K-gluconate, 40 CsCl, 3 EGTA, 1 MgCl2,
5 Mg-ATP, and 5 HEPES, pH 7.36. The bath solution for HEK cells
consisted of (in mM) 140 (or 102.5) CsCl, 2.5 (or 40) BaCl2 or CaCl2,
1 MgCl2, 10 tetraethylammonium-Cl, and 5 HEPES, pH 7.4. For
embryonic ventricular myocytes, the normal Tyrode solution contained 140 NaCl, 1.8 CaCl2, 1 MgCl2, 5.4 KCl, 10 glucose, and 10
HEPES, pH 7.4. The Na-free bath was 150 N-methyl-D-glucamine,
2.5 BaCl2, 1 MgCl2, 10 HEPES, 10 glucose, and 5 4-aminopyridine,
pH 7.4. Signals were amplified with an Axopatch 200B amplifier and
333 kHz A/D system (Axon Instruments, Union City, CA). Data were
analyzed with Clampfit 8 (Axon Instruments) and Origin statistical
software (OriginLab, Northampton, MA). There was no difference in
the Rem inhibition of Ca2⫹ channel current amplitude for either Ca2⫹
or Ba2⫹ as the charge carrier. All recordings were performed at room
temperature (20 –22°C). We used post hoc Student’s t-test to test for
significance between control and experimental groups.
Chronic H-89 is defined as 133 nM H-89 added at the time of
plasmid transfection in HEK-293 cells. This concentration was maintained 18 –24 h. Immediately before recording was performed, H-89
was washed out. The washout period ranged from 11 to 30 min.
Current-voltage (I-V) relationships were evaluated by voltageclamp protocols consisting of 5-s duration holding potential at ⫺80
mV followed by test depolarizations ranging from ⫺90 to ⫹60 in
5-mV increments with a 5-s interpulse interval. Test depolarization
duration was either 225 ms (Figs. 1, 3, 4, and 5) or 800 ms in later
experiments (Fig. 6). This protocol was initiated ⬎2 min after patch
break and was repeated at least twice to confirm stability of the
amplitude. Each protocol required ⬃3 min to complete.
Embryonic ventricular myocyte tissue harvest or cell culture. All
animals and animal procedures used in this study were approved by
the University of Kentucky Institutional Animal Care and Use Committee. E10 mouse (ICR outbred strain, Harlan) hearts were dissected
free of connective tissues, and ventricles were separated from
conotruncus and sinus venosus. For cell electrophysiology, cells were
enzymatically dispersed and cultured as previously described (15).
Briefly, 10 – 40 embryos were minced and quickly transferred to
nominally Ca-free digestion buffer containing 0.5 mg/ml collagenase
(type II, Worthington) and 1 mg/ml pancreatin for two 15-min cycles.
Digested tissue yielded a large fraction of single spontaneously
beating cells in culture media consisting of DMEM ⫹ 10% FBS. Cells
were infected with adenovirus at 100 multiplicity of infection within
24 h of dispersal. Adenoviral exposure was limited to 16 h. Adenovirus-infected cells were identified by GFP fluorescence and recorded
48 –72 h after initial viral exposure. To positively identify a cell as a
cardiac myocyte, we relied on an embryonic ventricular myocyte ionic
current signature primarily consisting of a large sodium current (INa)
in normal Tyrode solution as our selection criterion. To isolate ICa,L,
the bath was then changed to Na-free bath solution (see Embryonic
ventricular myocyte tissue harvest).
HIT-T15 cells were dispersed and infected similarly to embryonic
ventricular myocytes, with the exception that HIT-T15 cell media
consisted of Ham’s F-12 ⫹ 10% dialyzed horse serum ⫹ 2.5% FBS.
Cytosolic calcium imaging. Cardiac myocytes were loaded with 2
␮M fura-2 AM for 8 min in a 10% CO2 incubator and then deesterified in Tyrode solution for ⬃20 min. Spontaneous Ca transients were
recorded from the annulus of the photometry tube focused on a single
cell or a cell cluster. All recordings were performed at 37°C. The cells
were excited with light of 340-nm and 380-nm wavelengths. The
images obtained at 340 nm and 380 nm were divided pixel by pixel,
and the ratio data were reported. Data were collected and analyzed
with IonOptix (Milton, MA) hardware and software. Additional offline transient analysis was performed with custom routines written in
MatLab (Northampton, MA).
Real-time RT-PCR. Total RNA was isolated by using the RNAqueous ⫺4PCR kit (Ambion) and quantitated spectrophotometrically at
260 nm. Contaminating genomic DNA was eliminated by DNase
treatment (Ambion). A portion of the resulting RNA (1 ␮g) was
immediately used as a template for cDNA synthesis. Reverse transcription was performed with the use of the SuperScript First-Strand
Synthesis System for RT-PCR (Invitrogen). Removal of genomic
DNA was confirmed by preparing a no reverse transcription control
for each sample. cDNA was then stored at ⫺20°C. Purified Rem-GFP
plasmid (19) was quantified by using the PicoGreen (Molecular
Probes) quantification method. With the molecular weight of the
plasmid and insert known, it was possible to calculate the copy
number. Knowing the copy number and concentration of plasmid
DNA, the precise number of molecules added to subsequent real-time
PCR runs was calculated. Real-time PCR was performed in 96-well
optical plates in triplicate using an ABI 7700 Sequence Detector.
Samples (0.2 ng cDNA) were prepared using the TaqMan PCR core
reagent kit (Applied Biosystems), with the following primer and probe
sets: GAPDH: forward 5⬘-ATGTTCCAGTATGACTCCACTCACG3⬘, reverse 5⬘-GAAGACACCAGTAGACTCCACGACA, probe 5⬘FAM-AAGCCCATCACCATCTTCCAGGAGCGAGA-TAMRA-3⬘;
Ca V 1.2: forward 5⬘-CACCGTCTCCACCTTCGAA-3⬘, reverse
5⬘-CTTGTCTTCTGTGTGGGAGTCAAT-3⬘, probe 5⬘-FAMTGGCCAGAGCTGCTGTACCGCTC-TAMRA-3⬘; CaV␤2: forward
H1961
REM GTPASE AND CALCIUM CHANNELS
RESULTS
Regulation of Ca channel current by Rem. It is now established that Rem inhibits ICa,L and that Rem interacts with CaV␤
(20, 21), but the question of whether L-type channel ␣-subunit
contributes to Rem regulation of ICa,L is unexplored. Therefore,
we initially tested the effect of Rem on ICa in the absence of
exogenous CaV␤ expression. CaV1.2 expressed alone traffick
to the plasma membrane (26) and allows measurement of Ba
current (Fig. 1A). Rem coexpression significantly decreases
peak inward current density by 64% compared with CaV1.2
expressed alone [from 10.3 ⫾ 1.9 (n ⫽ 15) to 3.7 ⫾ 0.5; P ⫽
0.003; n ⫽ 14; Table 1]. The I-V relationship shows no
Rem-induced-shift of peak current or obvious voltage dependence (Fig. 1A). Coexpression of CaV␤2a with CaV1.2 increases peak current density approximately 10-fold and shifts
the peak voltage approximately ⫺10 mV (compare Fig. 1, A
and B). In contrast to the partial block of ICa by Rem alone,
coexpression of Rem, CaV␤2a, and CaV1.2 results in the
complete absence of detectable current (20). The time course
of current decay is an index of CaV␤2a regulation of CaV1.2
function. Figure 1C shows the current time course for CaV1.2
alone superimposed on that coexpressed with Rem; Fig. 1D
shows the well-established inactivation slowing induced by
CaV␤2a. At 200 ms after the onset of the depolarization, ⬃75%
of the peak current remains for CaV1.2 alone, and Rem coexpression had no effect on inactivation kinetics (Fig. 1, C and
E). In contrast, coexpression of CaV1.2 with CaV␤2a slows the
time course of macroscopic current inactivation (Fig. 1, D and
Fig. 1. Rem partially blocks L-type voltage-gated Ca
(CaV) channel CaV1.2 but completely eliminates
CaV1.2 ⫹ CaV␤2a current expression. A and B: currentvoltage (I-V) curve for CaV1.2 ⫹ Rem- (A) or CaV1.2 ⫹
CaV␤2a ⫹ Rem- (B) expressing cells. Number of cells
tested is listed in parenthesis; 40 mM bath Ba2⫹. C and
D: time course of inactivation is slower with coexpression of CaV␤2a. Average of current traces for Vtest ⫹ 20
mV from cells expressing CaV1.2 ⫾ Rem (C) or
CaV1.2 ⫹ CaV␤2a ⫹ Rem (D) normalized and superimposed. E: pooled remaining current at 200 (r200);
r200 is defined as fraction of current at 200 ms/peak
current. Coexpression of CaV␤2a significantly retards
inactivation; *P ⬍ 0.001.
AJP-Heart Circ Physiol • VOL
291 • OCTOBER 2006 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.4 on June 14, 2017
5⬘-CATTTGCAGTTCGGACCAATG-3⬘, reverse 5⬘-GCACCGGAACGTCATCCT-3⬘, probe 5⬘-FAM-CAGATACAGCGCAGCCTAMRA-3⬘; Rem: forward 5⬘-GCGCCTACGTCATCGTGTACT-3⬘,
reverse 5⬘-TCCGAGGCGCTCTCAAAG-3⬘, probe 5⬘-FAMCATAGCGGATCGCAGCA-TAMRA-3⬘. Samples were cycled for
50 cycles using an ABI 7700 Sequence Detector (Applied Biosystems). Cycle conditions were as follows: 2 min at 50°C, 10 min at
95°C followed by 50 cycles of 15 s at 95°C, and 1 min at 60°C. A
standard curve was generated from dilutions of the Rem plasmid by
plotting the natural log of the threshold cycle (CT) against the natural
log of the number of molecules. The CT was defined as the cycle at
which a statistically significant increase in the magnitude of the signal
generated by the PCR reaction was first detected. CT was calculated
under default settings for the real-time sequence detection software
(Applied Biosystems). To maximize accuracy, dilutions were made
over the range of copy numbers that included the amount of target
cDNA expected in the experimental cDNA samples. Specific calcium
channel component cDNA molecules present in the experimental
cDNA were calculated from the standard curve.
H1962
REM GTPASE AND CALCIUM CHANNELS
Table 1. Current densities of HEK-293 and HIT-T15 cells
HEK-293 Cells
40 mM Ba
CaV1.2 (n)
CaV1.2 ⫹ Rem (n)
CaV1.2 ⫹ ␤2a (n)
CaV1.2 ⫹ ␤2a ⫹ Rem (n)
CaV1.2 (wild type)
CaV1.2 ⫹ ␤2a (S1928A)
CaV1.2 ⫹ ␤2a (S1928A) ⫹ H-89
⫺10.3⫾1.9 (15)
⫺7.7⫾2.2 (5)
⫺4.5⫾2.5 (6)
⫺3.7⫾0.5 (14)
⫺7.0⫾1.9 (13)
⫺8.3⫾1.8 (9)
⫺102⫾21 (7)
⫺59.0⫾11.2 (6)
⫺61.4⫾20.9 (6)
0.0⫾0 (42)
⫺3.2⫾1.1 (16)
⫺3.4⫾(11)
HIT-T15 Cells
40 mM Ba
HIT-T15 (n)
HIT-T15 ⫹ H-89 (n)
Ad.control
Ad.Rem
⫺29.5⫾6.8 (7)
0.0⫾0 (8)
⫺13.7⫾4.0 (7)*
⫺5.6⫾1.0 (11)†
E). Thus macroscopic inactivation kinetics is an index of
functional association of CaV␤2a with CaV1.2.
CaV␤2a has predominant plasma membrane localization
(49). It has been suggested that CaV␤s, in general, function by
masking an endoplasmic reticulum retention domain on
CaV1.2 (5), although recent studies of CaV␤ suggest that the
original model must be refined to account for both Src homology 3 and GK domains of CaV␤ that are required to reconstitute ␣-subunit trafficking (50). Given our recent unpublished
finding that Rem interacts with the GK domain of CaV␤2 (B. S.
Finlin, R. Correll, C. Pang, S. M. Crump, J. Satin, and D. A.
Andres, unpublished observation), it then follows that Rem
may decrease CaV1.2 plasma membrane localization by interfering with the CaV1.2-CaV␤ interaction. There are, however,
three mitigating arguments against this model: 1) CaV␤2aCaV1.2-␣-interacting domain (AID) interaction is in the lownanometer range (41), and Rem is incapable of competing with
CaV1.2-AID for CaV␤2a (B. S. Finlin, R. Correll, C. Pang,
S. M. Crump, J. Satin, and D. A. Andres, unpublished observation); 2) surface biotinylation studies show CaV1.2 at the
plasma membrane even in the presence of overexpressed Rem2
(21); and 3) labeled peptide toxin studies showed cell surface
expression of CaV2.2 in the presence of exogenous Rem2
expression (9). We tested this further by measuring the fluorescence distribution of transiently transfected cells expressing
GFP-CaV1.2 fusion protein alone, coexpressed with either
Flag-tagged CaV␤2a or HA-tagged Rem, or coexpressing both
Flag-CaV␤2a and HA-Rem. Figure 2 (top), shows the distribution of GFP-CaV1.2. Fluorescence is punctate and excluded
from the nucleus. GFP-CaV1.2 ⫹ CaV␤2a coexpression, a
condition that yields ionic current, does not show enriched
plasma membrane localization. Most importantly, the fluorescence pattern of GFP-CaV1.2 is unchanged regardless of coexpression of CaV␤2a, Rem, or the combination of CaV␤2a
Fig. 2. Visualization of localization of green fluorescent protein (GFP)-CaV1.2 and GFP-Rem. Top: central optical slice through cells transfected with
GFP-CaV1.2 and coexpressing the indicated recombinant proteins. Leftmost panel demonstrates localization of GFP-CaV1.2 cotransfected with empty expression
vectors, and remaining panels display a representative micrograph of GFP-CaV1.2 coexpressed with either Flag-tagged CaV␤2a, hemagglutinin (HA)-tagged
Rem, or expressing both Flag-CaV␤2a and HA-Rem. GFP-CaV1.2 fluorescence is punctuate, excluded from the nucleus, and only minor amounts of the protein
are located at cell periphery. Coexpressed proteins have little effect on GFP-CaV1.2 distribution. Bottom: central optical slice through cells transfected with
GFP-Rem and coexpressing indicated recombinant proteins. GFP-Rem is enriched at cell periphery, although fluorescence is located throughout cell interior and
is both punctate and diffuse. Coexpression of either CaV1.2 or CaV␤2a has no effect on fluorescence distribution.
AJP-Heart Circ Physiol • VOL
291 • OCTOBER 2006 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.4 on June 14, 2017
Values are means ⫾ SE (in pA/pF); (n), no. of cells tested. Peak inward current density is for Vtest ⫹ 20 mV. HEK, human embryonic kidney; CaV, L-type
voltage-gated Ca; S1928A, serine 1928 to alanine mutation; Ad.control, cells infected with a control adenovirus expressing green fluorescent protein alone;
Ad.Rem, cells infected with an adenovirus coexpressing wild-type Rem and green fluorescent protein. *P ⫽ 0.0003, †P ⫽ 0.0004, for HIT-T15 with H-89
treatment vs. HIT-T15 without H-89 treatment.
H1963
REM GTPASE AND CALCIUM CHANNELS
and Rem. GFP-Rem fluorescence is enriched in the plasma
membrane, although a mixture of diffuse and punctate fluorescence is noted in the cytosol (Fig. 2, bottom). As with GFPCaV1.2, GFP-Rem fluorescence is not influenced by coexpression of Flag-CaV␤2a, untagged CaV1.2, or the combined coexpression of CaV␤2a and CaV1.2. Thus a portion of Rem is
expressed at the cell surface regardless of CaV␤2a expression.
We conclude that Rem does not necessarily block trafficking to
the cell surface, and we note that HEK-293 cells may be
inappropriate for quantitative cell-sorting studies with these
channels because of the ectopic localization of overexpressed
protein. With the finding that CaV1.2, CaV␤2a, and Rem all
may populate surface membrane to some extent, we therefore
focused this study on determining possible modulatory mechanisms of Rem effects on Ca channel current.
CaV1.2 COOH terminus is required for Rem block of Ca2⫹
channel current. The carboxyl terminus of CaV1.2 contains
several important modulatory domains, most notably PKA/
PKC substrate at position S1928 (18, 53). Deletion of the
CaV1.2 COOH terminus distal to residue 1733 (CaV1.2⌬1733)
yields a channel with ⬃50% plasma membrane localization
compared with full length (25) but with CaM regulation intact.
CaV1.2⌬1733 preserves the EF hand and the A, C, and IQ
motif required for CaM interaction but omits the PKA phosphorylation site of CaV1.2 (18). Coexpression of CaV1.2⌬1733
with CaV␤2a resulted in current that was ⬃25% peak ICa
compared with wild-type CaV1.2. In stark contrast to wild type,
however, coexpression of CaV␤2a ⫹ Rem with CaV1.2⌬1733
failed to completely inhibit current. Although mean current
Fig. 4. H-89 rapidly runs down current but does
not alter the inactivation time course. A: peak
CaV1.2 ⫹ CaV␤2a current plotted as a function
of H-89 exposure time. B: representative current
traces superimposed from 6-, 10-, 14-, and 18min exposure times. Note the indistinguishable
time course of inactivation.
AJP-Heart Circ Physiol • VOL
291 • OCTOBER 2006 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.4 on June 14, 2017
Fig. 3. Distal carboxyl terminus is required for Rem elimination of Ca channel
current amplitude. Current-voltage (I-V) relations for deletion of CaV1.2
COOH terminus distal to residue 1733 (CaV1.2⌬1733)coexpressed with
CaV␤2a (solid squares) or with CaV␤2a ⫹ Rem (open circles) are not
significantly different.
was reduced, it was not statistically different from that in the
absence of Rem coexpression (Fig. 3). The key important
finding is that Rem-CaV␤2a fails to block current of the
COOH-terminal truncated CaV1.2 channel. This suggests that
the COOH terminus of CaV1.2 distal to position 1733 contributes to Rem regulation of Ca channel current.
PKA inhibition relieves Rem inhibition of Ca current expression. To explore the contribution of the CaV1.2 COOH terminus to Rem-dependent regulation of Ca channel current, we
next examined the involvement of PKA signaling to this
process. In HEK-293 cells, PKA may be constitutively active
(39). Thus we next tested the effect of acute PKA inhibition on
Ca channel current. Figure 4A shows that H-89 progressively
decreases Ba current, without altering the inactivation kinetics
(Fig. 4B). This finding is consistent with the notion that CaV1.2
is constitutively modulated by active PKA in HEK-293 cells.
We next performed a series of experiments to test whether the
phosphorylation state of CaV1.2 modulates Rem regulation of Ca
channels. Although H-89 modulation of Ca channel current occurs within seconds, the Rem obliteration of detectable current
may occur over the hours of de novo channel complex formation.
Therefore, we first treated transfected HEK-293 cells for ⬃18 h
with H-89 (133 nM) to test whether steps leading to reduced
channel function were governed by H-89-sensitive and Remsensitive processes. Given that acute H-89 exposure severely
reduces current [Fig. 4 (39)], we then washed out H-89 20 – 45
min before initiating patch-clamp recordings. Chronic H-89 exposure reduced CaV1.2 ionic current. In contrast, chronic H-89
pretreatment prevented Rem inhibition of peak ICa (Fig. 5A).
Although H-89 tended to increase peak current density in Rem ⫹
CaV1.2-expressing cells compared with cells expressing CaV1.2
alone, this difference was not significant. Neither the peak I-V
relationship nor current kinetics was altered by chronic H-89 in
Rem ⫹ CaV1.2-expressing cells (Fig. 5, A, D, and G). These data
suggest that Rem-mediated channel blockade of CaV1.2 current is
enhanced by an H-89-sensitive modulation of CaV1.2. Thus, in
the presence of Rem GTPase, PKA activity can decrease, rather
than increase, current.
Rem block of current expression is complete when coexpressed with CaV␤2a. We therefore extended our studies to
evaluate whether H-89 can override this profound block of
CaV1.2 current in HEK cells coexpressing Rem ⫹ CaV␤2a.
Overnight incubation of cells with H-89 reduces CaV1.2 ⫹
CaV␤2a current (measured after 20 – 45 min washout) 67%
(from 25.7 ⫾ 2.5 to 8.5 ⫾ 5.3 pA/pF; data for 2.5-mM Ba
H1964
REM GTPASE AND CALCIUM CHANNELS
control current not shown). This fractional decrease is in
parallel with experiments in the absence of CaV␤2a (Figs. 1A
and 5A, solid squares), and these data suggest that the H-89
fractional current reduction is independent of CaV␤2a coexpression. Rem ⫹ CaV␤2a coexpression completely blocks
current, and H-89 reduces current in the absence of Rem. Thus
it is surprising that H-89 incubation partially relieved the
Rem-CaV␤2a block of current expression (Fig. 5B). As higher
concentrations of H-89 have been reported to inhibit kinases in
addition to PKA, we repeated these experiments with RpcAMP-S, a second relatively selective competitive inhibitor of
PKA. H-89 and Rp-cAMP-S (applied for ⬎18 h) had similar
actions, namely, restoration of detectable current with Rem ⫹
CaV␤2a coexpression (Fig. 5, B and C). The current restored
by H-89 had kinetics that inactivated with the same time course
as that observed in the absence of CaV␤2a expression (Fig. 5,
AJP-Heart Circ Physiol • VOL
E and H). Given that H-89 has no effect on inactivation
kinetics, but CaV␤2a coexpression does, the data suggest that
H-89 restores CaV1.2 current with CaV␤2a functionally dissociated with respect to channel kinetics modification. This
kinetic distinction differs from the Rp-cAMP-S restored current (Fig. 5, F and I); however, H-89 may have targets in
addition to PKA.
Status of PKA phosphorylation site on CaV1.2 modulates
Rem regulation of Ca channels. A well-documented substrate
for PKA phosphorylation of CaV1.2 is serine 1928, located on
the COOH terminus of the channel (18, 23). To evaluate the
specificity of our pharmacological experiments, we next tested
Rem regulation of the CaV1.2 PKA phosphorylation-deficient
mutant (CaV1.2-S1928A). Peak current density of CaV1.2S1928A was ⬃75% of that for wild-type ␣-subunit (Table 1).
Rem coexpression had no effect on current density of CaV1.2-
291 • OCTOBER 2006 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.4 on June 14, 2017
Fig. 5. PKA inhibition regulates Rem block of CaV1.2 current. Chronic (18 h) H-89 pretreatment restores CaV1.2 ionic current expression even in the presence
of Rem coexpression. H-89 was present for ⬎18 h but was removed for 20 – 45 min before recording was performed and was absent during current recordings.
A: peak current density for cells expressing CaV1.2 (solid squares) or CaV1.2 ⫹ Rem (open circles). B: peak current density of CaV1.2 ⫹ CaV␤2a (solid squares)
or CaV1.2 ⫹ CaV␤2a ⫹ Rem (open circles). C: same as B except 50 ␮M Rp-cAMP-S (RPc) was used instead of H-89. D–F: current traces are normalized and
superimposed for Vtest ⫽ ⫹30 mV corresponding to conditions in A, and Vtest⫽ 0 mV corresponding to conditions in B and C. G–I: pooled r200 for conditions
corresponding to those in A–C.
REM GTPASE AND CALCIUM CHANNELS
H1965
S1928A (Fig. 6A), and H-89 incubation had no further effect
(Fig. 6B). Figure 6C illustrates the difference in response to
Rem coexpression for wild-type versus S1928A mutant ␣-subunit. As seen with the wild-type channels, CaV␤2a coexpression slows macroscopic current decay of the S1928A mutant
channel (Fig. 6, D and G). However, Rem coexpressed with
CaV␤2a resulted in inactivation kinetics that was intermediate
between CaV␤2a slowing and that for CaV1.2 alone (Fig. 6E).
In summary, these data show that Rem-dependent regulation of
the S1928A mutant channel is significantly different from the
wild-type channel (Fig. 6C), and these differences are consisAJP-Heart Circ Physiol • VOL
tent with the notion that the phosphorylation of CaV1.2 enhances Rem block of ICa,L.
Modulation in native cells. Ca channels are a complex of
several proteins. Therefore, we next sought native cell systems
to confirm results from the simplified heterologous expression
of Ca channels in HEK-293 cells. HIT-T15 cells are insulinsecreting cells that express predominantly CaV1.2 (30, 44).
These cells express consistent levels of Ca current, are efficiently infected with recombinant adenoviruses, and are easily
cultured for several days. First, to test the ability of Rem
protein to inhibit CaV1.2 current, we infected cells with either
291 • OCTOBER 2006 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.4 on June 14, 2017
Fig. 6. CaV1.2 phosphorylation site serine 1928 (S1928) contributes to Rem regulation of CaV1.2 current. In all panels, PKA phosphorylation-deficient CaV1.2
mutant S1928 to alanine (S1928A) is expressed. A: Rem has no effect on CaV1.2-S1928A current. Peak current density-voltage plot of CaV1.2-S1928A (solid
squares) and CaV1.2-S1928A ⫹ Rem (open circles). B: same as A, except cells were pretreated 18 h with H-89. C: peak current density-voltage plot of
CaV1.2-S1928A ⫹ CaV␤2a ⫹ Rem with H-89 pretreatment (open squares) and without H-89 (open circles) compared with wild-type CaV1.2 ⫹ CaV␤2a ⫹ Rem
(closed squares). Rem block of wild-type channel coexpressed with CaV␤2a is complete, and thus Rem-CaV␤2a block of CaV1.2-wild-type channel current is
significantly greater than that of CaV1.2-S1928A. D–F: current traces normalized and superimposed for Vtest ⫽ ⫹30 mV. Conditions are as labeled in legends.
G: pooled r200 data. CaV␤2a significantly slows inactivation of CaV1.2-S1928A. With Rem coexpression, inactivation kinetics is intermediate between that
which is characteristic for ␣ alone versus ␣ ⫹ CaV␤2a. Number of cells tested indicated in bars.
H1966
REM GTPASE AND CALCIUM CHANNELS
Vtest 0 mV ICa,L ⫽ ⫺3.1 ⫾ 0.6 pA/pF, or ⫺0.5 ⫾ 0.2 pA/pF for
Ad.control or Ad.Rem, respectively, P ⫽ 0.02; Fig. 9D). We
next tested for the ability of long-term H-89 exposure to
reverse Rem-induced ICa,L inhibition. For these studies, H-89
was applied for 6 h to Ad.Rem-infected cells, media containing
H-89 were removed 30 – 60 min before recording was performed, and recordings were initiated in the absence of H-89.
After H-89 pretreatment, Ad.Rem-infected cells compared
with Ad.control cells exhibited ICa,L peak amplitudes that
tended to be larger but were statistically indistinguishable from
those for Ad.control cells (Fig. 9E). One point of complexity is
that mouse embryonic ventricular myocytes express multiple
Ca channel types, principally, ICa,L carried by CaV1.2 (45), and
T-type Ca current (ICa,T) carried by CaV3.1d (15). Ba2⫹ as the
charge carrier and EGTA in the pipette dramatically slows
macroscopic ICa,L decay but does not appreciably alter voltagedependent inactivation of ICa,T (14). In embryonic ventricular
myocytes, ICa,T inactivates with a time constant of ⬃14 ms for
Vtest positive to ⫺30 mV (7, 15, 42). To eliminate possible ICa,T
contamination, we therefore replotted the I-V relationship with
isochronal current 80 ms after onset of Vtest to allow for ⬎95%
inactivation of ICa,T. The isochronal I-V shows larger ICa,L in
Ad.Rem cells pretreated with H-89 compared with those that
were not treated. Taken together with the heterologous expression studies, we therefore conclude that the phosphorylation
status of CaV1.2 is an important contributor to Rem block of
Ca channel current in native cells.
DISCUSSION
This study demonstrates a novel mechanism for L-type Ca
channel functional regulation. Rem ⫹ CaV␤ coexpression
completely blocks Ca channel current in HEK-293 cells (1,
20), but Rem-CaV␤2a-inhibited current can be restored by
either pharmacological inhibition of PKA signaling (H-89 or
Rp-cAMP-S) or site-directed mutagenesis of the PKA phosphorylation site within the CaV1.2 subunit (S1928A). Treatment with H-89 and Rp-cAMP-S yields slightly different basal
current, suggesting the possibility of additional cellular targets
beyond PKA. Nonetheless, two pharmacological manipulations and the results with the S1928A mutant share PKA as a
common effector. Thus we suggest that the phosphorylation
status of CaV1.2 contributes to Rem regulation of the Ca
channel complex. The second principal finding is that Rem, in
the absence of exogenous CaV␤ expression, can inhibit CaV1.2
current. The third major finding is that, in a pancreatic ␤-cell
line and in primary cultures of isolated cardiac embryonic
Fig. 7. Rem partially inhibits native Ca channel current, and H-89
pretreatment inhibits RGK regulation of Ca channel current. A:
current density-voltage relationship of HIT-T15 cells infected with
adenovirus coexpressing wild-type Rem and GFP (Ad.Rem; 0 pA/pF,
n ⫽ 8; open circles) or control adenovirus expressing GFP alone
(Ad.control; 29.5 ⫾ 6.8 pA/pF, n ⫽ 7, solid squares). B: treatment of
infected cells with H-89 inhibits Rem attenuation of current. (Ad.
Rem ⫹ H-89, 5.6 ⫾ 1.0 pA/pF; n ⫽ 11, open circles; or Ad.control ⫹
H-89, 13.7 ⫾ 4.0 pA/pF, n ⫽ 7, solid squares; P ⫽ 0.0004.)
AJP-Heart Circ Physiol • VOL
291 • OCTOBER 2006 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.4 on June 14, 2017
a control adenovirus expressing GFP alone (Ad.control) or an
adenovirus coexpressing wild-type Rem and GFP (Ad.Rem).
Figure 7A shows that Rem overexpression completely blocks
Ca channel current in HIT-T15 cells. Our heterologous expression studies suggest that PKA inhibition should decrease the
control current but should act to reverse Rem-mediated current
inhibition. Figure 7B shows current-voltage curves demonstrating that, in the presence of H-89 pretreatment, there is Ca
channel current despite exogenous Rem expression. The decrease of current by H-89 under control conditions (compare
closed squares of Fig. 7, A and B) is anticipated on the basis of
a reduced phosphorylation of CaV1.2 by inhibition of PKA.
Thus an H-89-sensitive pathway plays a crucial role in Remdependent Ca channel regulation in this cell line.
Embryonic ventricular myocytes express mRNA for Rem
and CaV␤2 (Fig. 8A) and thus are a desirable native model
system to address the complex interplay between Rem and
ionic current. Adenovirus-mediated Rem overexpression results in ⬎3 order of magnitude increase of Rem mRNA but
caused no significant changes in CaV␤2, CaV1.2, or NaV1.5
mRNA. Rem overexpression inhibits spontaneous beating and
Ca transients (Fig. 8, B and C). In Ad.control-infected embryonic ventricular myocytes, 17 of 21 cells exhibited spontaneous global Ca transients (Table 2). In contrast, Ad.Reminfected embryonic ventricular myocytes failed to display
spontaneous Ca transients in all cells tested (n ⫽ 21; Table 2).
Nifedipine (10 ␮M) also inhibits spontaneous Ca transients in
control cell clusters (not shown). This confirms the essential
role of CaV1.2 and/or 1.3 channels for spontaneous activity. To
determine the minimum H-89 pretreatment duration required to
release Rem block of Ca channel current, we incubated cells
for 1 h, followed with a washout period of at least 45 min, and
observed spontaneous Ca transients in only 1 of 6 cell clusters
tested (Table 2). However, a 2-h H-89 pretreatment, followed
by 20- to 25-min washout period, restored spontaneous Ca
transients in embryonic ventricular myocytes infected with
Ad.Rem (Fig. 8D and Table 2). These results are consistent
with that observed in heterologous expression (Figs. 1– 6) and
HIT cell studies (Fig. 7) and show that an H-89-sensitive
pathway modulates Rem regulation of Ca channels on a rather
slow timescale.
To directly measure ICa,L, we voltage clamped single cells in
the whole cell configuration. Representative Ad.control-infected cell currents are shown in Fig. 9A. The current noticeably activates at ⬃⫺25 mV and decays slowly with sustained
depolarization. Infection of cells with Ad.Rem significantly
reduces ICa,L compared with that from Ad.control cells (for
H1967
REM GTPASE AND CALCIUM CHANNELS
ventricular myocytes, exogenous Rem blocks ICa,L, and longterm treatment with H-89 can reverse this effect. A key
principle is that PKA modulation of Rem block of ICa,L is
occurring over a relatively long time frame compared with
well-studied acute channel modulation by PKA. Interestingly,
our data suggest that, in the long term, PKA enhances Rem
block of ICa,L, and this is opposite the well-characterized effect
of acute PKA stimulation of ICa,L.
Bimodal modulation of CaV1.2 by PKA. Acute activation of
PKA increases CaV1.2 current and is reversible by acute
addition of H-89 (37, 39). The rapid rundown in the presence
of H-89 has previously been interpreted as a consequence of
Table 2. Ca2⫹ transient kinetic parameters
Treatment
Tested
Calcium
Transient
Present
Ad.control
Ad.control ⫹ H89 (2-h pretreatment)
Ad.Rem
Ad.Rem ⫹ H-89 (2-h pretreatment)
Ad.Rem ⫹ H-89 (1-h pretreatment)
21
17
21
14
6
17
15
0
10
1
t50, s
0.207⫾0.013
0.187⫾0.016
0.448⫾0.035
0.401⫾0.044
0.318⫾0.021
0.424⫾0.016*
0.211⫾0.010
0.225
0.382⫾0.018
0.673
0.396⫾0.048
0.333
Values are means ⫾ SE. TTP, time to peak of an individual Ca2⫹ transient; t50, time for 50% decay. *P ⫽ 0.02.
AJP-Heart Circ Physiol • VOL
Frequency, Hz
TTP, s
291 • OCTOBER 2006 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.4 on June 14, 2017
Fig. 8. Ad.Rem-infected embryonic ventricular
myocytes effects on mRNA levels and spontaneous
Ca transients. A: Ad.Rem increases Rem expression
⬎103-fold but has no significant effect on CaV␤2,
NaV1.5, or CaV1.2 mRNA levels as measured by
RT-PCR. B–D: spontaneous Ca transients manifested as ratio of fluorescence at 340 nm to fluorescence at 380 nm (340/380) in Ad.control-infected
embryonic ventricular myocytes (B), Ad.Rem-infected embryonic ventricular myocytes (C), or
Ad.Rem-infected embryonic ventricular myocytes
that were pretreated with 133 nM H-89 (D). Note
that H-89 restores transients in Ad.Rem-infected
embryonic ventricular myocytes that are indistinguishable from those in control conditions. Sweeps
in B and C are 30 s. Kinetic parameters of individual calcium transients are listed in Table 2. D:
individual 10-s duration sweeps are concatenated.
Solid vertical lines denote end and beginning of
individual 10-s duration sweeps. Note that resting
Ca2⫹ increases at 25 min post-H-89 washout, and
spontaneous transients commence beginning with
the 30-min recording.
H1968
REM GTPASE AND CALCIUM CHANNELS
dephosphorylation of a protein in the Ca channel complex (39).
However, our data support a model in which PKA phosphorylation of CaV1.2 promotes Rem-mediated block of Ca channel
current. These findings suggest that PKA has bimodal effects
on Ca channel current. In the face of relatively high Rem
expression, S1928 phosphorylation promotes a paradoxical
decrease of current. In the absence of Rem, S1928 phosphorylation causes a well-described increase of current (Fig. 5).
Thus PKA signaling serves to enhance Ca currents via direct
phosphorylation of the channel complex but also plays an
essential role in RGK GTPase family-dependent negative regulation of L-type Ca channel activity. We speculate that longterm Rem downregulation of ICa,L is a homeostatic regulator of
current that is increased via PKA stimulation. Studies are
ongoing to determine whether a similar mechanism serves to
regulate other RGK family proteins.
Mechanism of Rem block of Ca channel expression. CaV␤2
subunits promote trafficking of CaV ␣-subunits to the cell
AJP-Heart Circ Physiol • VOL
surface; in this regard, the palmitylated CaV␤2a is highly
efficient (11). CaV␤2a also adds the distinctive functional
feature of retarding voltage-dependent inactivation in heterologously expressed channels (26) and in native systems (12). In
the absence of Ca2⫹, CaV␤2a-modified channels inactivate less
than either CaV1.2 alone or CaV1.2 coexpressed with other
CaV␤ proteins. We therefore can use current decay kinetics as
an index of CaV␤2a-␣-subunit functional interaction. For the
wild-type channel coexpressed with Rem and CaV␤2a, the
current that was restored by H-89 pretreatment tended to
display decay kinetics reflective of ␣-subunit expression alone
(Fig. 5). This would be consistent with models invoking either
Rem blocking CaV␤2a-directed trafficking of ␣-subunit to the
surface membrane or Rem preventing CaV␤2a modulation of a
preformed ␣-␤ channel complex at the surface membrane. In
the former model, ␣-subunit-only Ca channels populate surface
membrane. Support for the latter mechanism is the observation
that the CaV1.2-S1928 mutant channel current has intermediate
291 • OCTOBER 2006 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.4 on June 14, 2017
Fig. 9. Rem inhibits ICa,L in embryonic ventricular myocytes, and Rem inhibition is antagonized
by H-89 pretreatment. A: representative ionic current from Ad.control-infected embryonic ventricular myocyte cells for Vtest as shown in left column. B: representative current from Ad.Rem
cells. C: representative current from Ad.Reminfected cells pretreated with 133 nM H-89. H-89
was removed before patch-clamp recording was
performed and was absent for ⬎10 min. D: pooled
peak inward I-V relationship for Ad.control superimposed on Ad.Rem. Peak currents at Vtest 0 mV
are significantly decreased in Ad.Rem-infected
cells (open circles; n ⫽ 4) compared with Ad.control (solid squares; n ⫽ 9; P ⫽ 0.01). E: pooled
peak inward I-V relationship for Ad.Rem pretreated with 133 nM H-89 (upright triangles)
compared with Ad.Rem (smooth line reproduced
from D). To eliminate any possible contribution
of T-type Ca current, current was also measured
and plotted for isochrone of 80 ms after onset of
Vtest (inverted triangles; n ⫽ 9). Note restoration
of isochronal current with H-89 pretreatment (inverted triangles) despite Ad.Rem infection.
REM GTPASE AND CALCIUM CHANNELS
AJP-Heart Circ Physiol • VOL
tions are shared by ICa,T, INa(13, 22), and hyperpolarizationactivated current If (40). Thus, in embryonic ventricular
myocytes, ICa,L cannot be separated on the basis of voltage
dependence but can be partially separated on the basis of
kinetics. The key point is that kinetic-based separation
requires the replacement of bath Ca2⫹ with Ba2⫹. With
Ca2⫹ as the charge carrier, ICa,L fast inactivation overlaps
with that of ICa,T, because of the relatively rapid Cadependent inactivation properties of the L-type Ca channel
complex. However, ICa,L with Ba2⫹ as the charge carrier and
with internal Ca buffered exhibits voltage-dependent decay
that is ⬎10-fold slower than ICa,T. Finally, Ni2⫹ is known to
block ICa,T but also has subunit composition-specific effects
on ICa,L (54). Thus we did not use Ni2⫹ to discriminate
between ICa,T and ICa,L in embryonic ventricular myocytes
because of uncertainty of the effects of Ni2⫹ might have on
the ICa,L of embryonic ventricular myocytes. Given the
ambiguity of voltage dependence and potential nonselective
effects of pharmacological reagents, we view Rem as an
excellent discriminator between ICa,L and ICa,T. In heterologous expression systems, Rem blocks CaV1.2 channel
current but has no effect on either CaV3.1 or CaV3.2 (20).
Thus our finding that Rem blocks spontaneous Ca transients
suggests that CaV1.2 channels contribute trigger Ca to
Ca-induced Ca release in embryonic ventricular myocytes.
Moreover, in the presence of CaV1.2, channel block by Rem
ICa,T cannot compensate for L-type trigger Ca.
Physiological consequences of PKA enhancement of RGKchannel effects. A recent study (34) showed that focal application of RGK can serve as an effective therapy for local Ca
channel block in the heart. Our present work shows that PKA
phosphorylation of the ␣-subunit would be expected to enhance such block. Murata et al. (34) allude to a potential benefit
of RGK block of Ca current in ailments such as hypertrophic
cardiomyopathy. However, a common first-line treatment of
hypertrophic cardiomyopathy and other related cardiac disorders often includes administration of ␤-adrenergic receptor
blockade (48). Given our present results, ␤-blockade would
reduce the ability of Rem, and perhaps all RGK proteins, to
inhibit Ca channel. Thus RGK as a Ca-entry blocking therapy
may, in fact, be quite desirable, and our findings suggest that
RGK therapy would have more effect for a sympathetic nervous system-stimulated heart.
The original observation that linked RGK to CaV␤ was in
insulin-secreting cells (1). Although controversial, the most
convincing evidence shows that CaV1.2 is the essential secretagogue in insulin-secreting cells (43, 46). Nutrient stimulation
of insulin-secreting cells results in a well-established increase
of cytosolic cAMP. This, in turn, results in cascades that,
among the downstream targets, include increased PKA activity
that increases Ca channel opening (see Ref. 35 for a review).
Our results show that RGK in insulin-secreting cells add a
negative feedback pathway to PKA-induced increases in Ca
current. Indeed, our recent analysis of Rem2 function in ␤-islet
cell function supports such a model (21).
In summary, we show evidence for a novel mechanism by
which PKA stimulation can paradoxically decrease L-type Ca
current. We conclude that the phosphorylation status of CaV1.2
is a contributor to RGK inhibition of Ca channel current.
291 • OCTOBER 2006 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.4 on June 14, 2017
inactivation kinetics when compared with either ␣-subunit
alone or ␣ ⫹ CaV␤2a-expressed ICa. The inclusive explanation
is that both general mechanisms of current block may exist.
The data regarding a trafficking-related mechanism remain
controversial. Recent biochemical studies with another RGK
family member, Rem2, demonstrate that Rem2 does not prevent cell surface accumulation of CaV1.2 (21) and CaV2.2 (9),
but contradictory results obtained by surface immunofluorescence were reported by others for Rad and Rem (4), Rem2 (2),
and Kir/Gem (3). We suggest that this immunostaining technique may be less sensitive to low levels of cell surface
expression than the other techniques. Our justification for this
speculation is that CaV1.2 expressed alone in HEK-293 cells
shows ionic current, but CaV1.2 cell surface expression is not
detectable in the HA-immunostaining assay used by Beguin
and colleagues (3). Thus RGK proteins do not necessarily
block Ca2⫹ channel current by preventing transport to the
plasma membrane. CaV ␣-␤ subunit interaction stoichiometry
is 1:1 (17), and recent studies show CaV␤2a enhances Ca
channel modulation via secondary binding sites (31). This
notion is consistent with our data suggesting that CaV1.2,
particularly the phosphorylation of the COOH-terminal S1928,
may contribute to the combined Rem-CaV␤ modulation of
ICa,L. In our working model, we suggest that a putative
CaV␤2a-Rem-CaV1.2 complex is enhanced by CaV1.2 phosphorylation. In such a dynamic model, there are theoretically
four possible channel combinations: a complex of ␣-␤-Rem,
␣-␤, ␣- alone, or ␣-Rem. ␣-␤-Rem should yield no current and
␣-␤ will yield large-amplitude current with slow inactivation,
whereas ␣-Rem or ␣- alone would yield small current with
faster inactivation. Why then does ␣(S1928A)-␤-Rem yield
current with intermediate kinetics? The simplest explanation is
that S1928 phosphorylation status enhances ␤-Rem-mediated
current blockade, and in the absence of phosphorylation of
␣-S1928, ␤-Rem is less likely to complex with ␣- to completely block current. Thus we speculate that the functional
impact of Rem in complex with the channel is to inhibit
channel gating and channel kinetics, in part, via modulation of
accessory subunit interactions. A second noncompeting idea is
that Rem may function to promote retrograde transport, thus
destabilizing plasma membrane localization of the Ca channel
complex. Impetus for this suggestion is the ⬎1-h time course
of H-89 required for restoration of current (Table 2). This
suggests the possible presence of a cycling pool of Ca channel
complexes that exhibit H-89-sensitive regulation. The data
from our present study motivate future studies to delineate the
mechanisms revealed by RGK block, and H-89 release of Rem
block of Ca channel current.
Contribution of ICa,L to embryonic ventricular myocyte function. In mature, nonpathological heart cells, CaV1.2 apparently is the only voltage-gated Ca channel underlying ICa,L,
whereas ICa,T is reexpressed in adult pathological states
(36). In adult myocytes, as in heterologous expression
systems, ICa,L and ICa,T can be separated by either the
relatively high- versus low-voltage activation range or by
altering holding potential. However, in embryonic ventricular myocytes, ICa,T activates at higher potential and thus
overlaps with the activation range of ICa,L (15). ICa,L also
shows inactivation dependence that overlaps with ICa,T.
Such depolarizing shifts of voltage dependence in embryonic ventricular myocytes compared with mature prepara-
H1969
H1970
REM GTPASE AND CALCIUM CHANNELS
ACKNOWLEDGMENTS
We thank Tim Kamp (University of Wisconsin) for CaV1.2 and CaV␤2a,
and Kurt Beam (University of Connecticut) for GFP-CaV1.2.
GRANTS
This study was supported by National Institutes of Health Grants HL074091 (to J. Satin), HL-072936 (to D. A. Andres), and P-20RR20171-Centers
of Biomedical Research Excellence program of National Center for Research
Resources (to D. A. Andres); ADA award (to B. Finlin); and American Heart
Association predoctoral fellowship (to R. N. Correll). J. Satin is an Established
Investigator of the American Heart Association.
REFERENCES
AJP-Heart Circ Physiol • VOL
291 • OCTOBER 2006 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.4 on June 14, 2017
1. Beguin P, Nagashima K, Gonoi T, Shibasaki T, Takahashi K,
Kashima Y, Ozaki N, Geering K, Iwanaga T, and Seino S. Regulation
of Ca2⫹ channel expression at the cell surface by the small G-protein
kir/Gem. Nature 411: 701–706, 2001.
2. Beguin P, Mahalakshmi RN, Nagashima K, Cher DH, Kuwamura N,
Yamada Y, Seino Y, and Hunziker W. Roles of 14-3-3 and calmodulin
binding in subcellular localization and function of the small G protein
Rem2. Biochem J 390: 67–75, 2005.
3. Beguin P, Mahalakshmi RN, Nagashima K, Cher DH, Takahashi A,
Yamada Y, Seino Y, and Hunziker W. 14-3-3 And calmodulin control
subcellular distribution of Kir/Gem and its regulation of cell shape and
calcium channel activity. J Cell Sci 118: 1923–1934, 2005.
4. Beguin P, Mahalakshmi RN, Nagashima K, Cher DH, Ikeda H,
Yamada Y, Seino Y, and Hunziker W. Nuclear sequestration of betasubunits by Rad and Rem is controlled by 14-3-3 and calmodulin and
reveals a novel mechanism for Ca2⫹ channel regulation. J Mol Biol 355:
34 – 46, 2006.
5. Bichet D, Cornet V, Geib S, Carlier E, Volsen S, Hoshi T, Mori Y, and
De Waard M. The I-II loop of the Ca2⫹ channel alpha1 subunit contains
an endoplasmic reticulum retention signal antagonized by the beta subunit.
Neuron 25: 177–190, 2000.
6. Brice NL, Berrow NS, Campbell V, Page KM, Brickley K, Tedder I,
and Dolphin AC. Importance of the different beta subunits in the
membrane expression of the alpha1A and alpha2 calcium channel subunits: studies using a depolarization-sensitive alpha1A antibody. Eur
J Neurosci 9: 749 –759, 1997.
7. Burgess DE, Crawford O, Delisle BP, and Satin J. Mechanism of
inactivation gating of human T-type (low-voltage activated) calcium
channels. Biophys J 82: 1894 –1906, 2002.
8. Catterall WA. Structure and regulation of voltage-gated Ca2⫹ channels.
Annu Rev Cell Dev Biol 16: 521–555, 2000.
9. Chen H, Puhl HL, 3rd Niu SL, Mitchell DC, and Ikeda SR. Expression
of Rem2, an RGK family small GTPase, reduces N-type calcium current
without affecting channel surface density. J Neurosci 25: 9762–9772,
2005.
10. Chien AJ, Zhao X, Shirokov RE, Puri TS, Chang CF, Sun D, Rios E,
and Hosey MM. Roles of a membrane-localized beta subunit in the
formation and targeting of functional L-type Ca[IMAGE] channels. J Biol
Chem 270: 30036 –30044, 1995.
11. Chien AJ, Carr KM, Shirokov RE, Rios E, and Hosey MM. Identification of palmitoylation sites within the L-type calcium channel beta 2a
subunit and effects on channel function. J Biol Chem 271: 26465–26468,
1996.
12. Colecraft HM, Alseikhan B, Takahashi SX, Chaudhuri D, Mittman S,
Yegnasubramanian V, Alvania RS, Johns DC, Marban E, and Yue
DT. Novel functional properties of Ca2⫹ channel beta subunits revealed
by their expression in adult rat heart cells. J Physiol 541: 435– 452, 2002.
13. Conforti L, Tohse N, and Sperelakis N. Tetrodotoxin-sensitive sodium
current in rat fetal ventricular myocytes– contribution to the plateau phase
of action potential. J Mol Cell Cardiol 25: 159 –173, 1993.
14. Cribbs LL, Lee JH, Yang J, Satin J, Zhang Y, Daud A, Barclay J,
Williamson MP, Fox M, Rees M, and Perez-Reyes E. Cloning and
characterization of ␣1H from human heart, a member of the T-type
calcium channel gene family. Circ Res 83: 103–109, 1998.
15. Cribbs LL, Martin BL, Schroder EA, Keller BB, Delisle BP, and Satin
J. Identification of the T-type calcium channel (CaV3.1d) in developing
mouse heart. Circ Res 88: 403– 407, 2001.
16. Dai S, Klugbauer N, Zong X, Seisenberger C, and Hofmann F. The
role of subunit composition on prepulse facilitation of the cardiac L-type
calcium channel. FEBS Lett 442: 70 –74, 1999.
17. Dalton S, Takahashi SX, Miriyala J, and Colecraft HM. A single CaV␤
can reconstitute both trafficking and macroscopic conductance of voltagedependent calcium channels. J Physiol 567: 757–769, 2005.
18. De Jongh KS, Murphy BJ, Colvin AA, Hell JW, Takahashi M, and
Catterall WA. Specific phosphorylation of a site in the full-length form of
the alpha 1 subunit of the cardiac L-type calcium channel by adenosine
3⬘,5⬘-cyclic monophosphate-dependent protein kinase. Biochemistry 35:
10392–10402, 1996.
19. Finlin BS and Andres DA. Rem is a new member of the Rad- and
Gem/Kir Ras-related GTP-binding protein family repressed by lipopolysaccharide stimulation. J Biol Chem 272: 21982–21988, 1997.
20. Finlin BS, Crump SM, Satin J, and Andres DA. Regulation of voltagegated calcium channel activity by the Rem and Rad GTPases. Proc Natl
Acad Sci USA 100: 14469 –14474, 2003.
21. Finlin BS, Mosley AL, Crump SM, Correll RN, Ozcan S, Satin J, and
Andres DA. Regulation of L-type Ca2⫹ channel activity and insulin
secretion by the Rem2 GTPase. J Biol Chem 280: 41864 – 41871, 2005.
22. Fujii S, Ayer RK Jr, and DeHaan RL. Development of the fast sodium
current in early embryonic chick heart cells. J Membr Biol 101: 209 –223,
1988.
23. Gao T, Yatani A, Dell’Acqua ML, Sako H, Green SA, Dascal N, Scott
JD, and Hosey MM. cAMP-dependent regulation of cardiac L-type Ca2⫹
channels requires membrane targeting of PKA and phosphorylation of
channel subunits. Neuron 19: 185–196, 1997.
24. Gao T, Chien AJ, and Hosey MM. Complexes of the alpha 1C and beta
subunits generate the necessary signal for membrane targeting of class C
L-type calcium channels. J Biol Chem 274: 2137–2144, 1999.
25. Gao T, Bunemann M, Gerhardstein BL, Ma H, and Hosey MM. Role
of the C terminus of the alpha 1C (CaV1.2) subunit in membrane targeting
of cardiac L-type calcium channels. J Biol Chem 275: 25436 –25444,
2000.
26. Gerster U, Neuhuber B, Groschner K, Striessnig J, and Flucher BE.
Current modulation and membrane targeting of the calcium channel
alpha1C subunit are independent functions of the beta subunit. J Physiol
517: 353–368, 1999.
27. Grabner M, Dirksen RT, and Beam KG. Tagging with green fluorescent
protein reveals a distinct subcellular distribution of L-type and non-L-type
Ca2⫹ channels expressed in dysgenic myotubes. Proc Natl Acad Sci USA
95: 1903–1908, 1998.
28. Gurnett CA, De Waard M, and Campbell KP. Dual function of the
voltage-dependent Ca2⫹ channel alpha 2 delta subunit in current stimulation and subunit interaction. Neuron 16: 431– 440, 1996.
29. Hofmann F, Biel M, and Flockerzi V. Molecular basis for Ca2⫹ channel
diversity. Annu Rev Neurosci 17: 399 – 418, 1994.
30. Ji J, Yang SN, Huang X, Li X, Sheu L, Diamant N, Berggren PO, and
Gaisano HY. Modulation of L-type Ca(2⫹) channels by distinct domains
within SNAP-25. Diabetes 51: 1425–1436, 2002.
31. Leroy J, Richards MS, Butcher AJ, Nieto-Rostro M, Pratt WS, Davies
A, and Dolphin AC. Interaction via a key tryptophan in the I-II linker of
N-type calcium channels is required for ␤1 but not for palmitoylated ␤2,
implicating an additional binding site in the regulation of channel voltagedependent properties. J Neurosci 25: 6984 – 6996, 2005.
32. Letts VA, Felix R, Biddlecome GH, Arikkath J, Mahaffey CL, Valenzuela A, Bartlett FS, 2nd Mori Y, Campbell KP, and Frankel WN. The
mouse stargazer gene encodes a neuronal Ca2⫹-channel gamma subunit.
Nat Genet 19: 340 –347, 1998.
33. Mikami A, Imoto K, Tanabe T, Niidome T, Mori Y, Takeshima H,
Narumiya S, and Numa S. Primary structure and function expression of
the cardiac dihydropyridine-sensitive calcium channel. Nature 340: 230 –
233, 1989.
34. Murata M, Cingolani E, McDonald AD, Donahue JK, and Marban E.
Creation of a genetic calcium channel blocker by targeted gem gene
transfer in the heart. Circ Res 95: 398 – 405, 2004.
35. Newgard CB and McGarry JD. Metabolic coupling factors in pancreatic
beta-cell signal transduction. Annu Rev Biochem 64: 689 –719, 1995.
36. Nuss HB and Houser SR. T-type Ca2⫹ current is expressed in hypertrophied adult feline left ventricular myocytes. Circ Res 73: 777–782, 1993.
37. Ono K and Fozzard HA. Phosphorylation restores activity of L-type
calcium channels after rundown in inside-out patches from rabbit cardiac
cells. J Physiol 454: 673– 688, 1992.
38. Perez-Reyes E, Castellano A, Kim HS, Bertrand P, Baggstrom E,
Lacerda AE, Wei XY, and Birnbaumer L. Cloning and expression of a
cardiac/brain beta subunit of the L-type calcium channel. J Biol Chem 267:
1792–1797, 1992.
REM GTPASE AND CALCIUM CHANNELS
AJP-Heart Circ Physiol • VOL
47. Slish DF, Engle DB, Varadi G, Lotan I, Singer D, Dascal N, and
Schwartz A. Evidence for the existence of a cardiac specific isoform of
the alpha 1 subunit of the voltage dependent calcium channel. FEBS Lett
250: 509 –514, 1989.
48. Spirito P, Seidman CE, McKenna WJ, and Maron BJ. The management of hypertrophic cardiomyopathy. N Engl J Med 336: 775–785, 1997.
49. Takahashi SX, Mittman S, and Colecraft HM. Distinctive modulatory
effects of five human auxiliary beta(2) subunit splice variants on L-type
calcium channel gating. Biophys J 84: 3007–3021, 2003.
50. Takahashi SX, Miriyala J, Tay LH, Yue DT, and Colecraft HM. A
CaVbeta SH3/guanylate kinase domain interaction regulates multiple
properties of voltage-gated Ca2⫹ channels. J Gen Physiol 126: 365–377,
2005.
51. Wei Sk Colecraft HM, DeMaria CD, Peterson BZ, Zhang R, Kohout
TA, Rogers TB, and Yue DT. Ca2⫹ channel modulation by recombinant
auxiliary ␤ subunits expressed in young adult heart cells. Circ Res 86:
175–184, 2000.
52. Yamaguchi H, Okuda M, Mikala G, Fukasawa K, and Varadi G.
Cloning of the beta(2a) subunit of the voltage-dependent calcium channel
from human heart: cooperative effect of alpha(2)/delta and beta(2a) on the
membrane expression of the alpha(1C) subunit. Biochem Biophys Res
Commun 267: 156 –163, 2000.
53. Yang L, Liu G, Zakharov SI, Morrow JP, Rybin VO, Steinberg SF,
and Marx SO. Ser1928 is a common site for Cav1.2 phosphorylation by
protein kinase C isoforms. J Biol Chem 280: 207–214, 2005.
54. Zamponi GW, Bourinet E, and Snutch TP. Nickel block of a family of
neuronal calcium channels: subtype and subunit-dependent action at multiple sites. J Membr Biol 151: 77–90, 1996.
55. Zhang Y, Cribbs LL, and Satin J. Arachidonic acid modulation of ␣1H,
a cloned human T-type calcium channel. Am J Physiol Heart Circ Physiol
278: H184 –H193, 2000.
291 • OCTOBER 2006 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.4 on June 14, 2017
39. Perez-Reyes E, Yuan W, Wei X, and Bers DM. Regulation of the cloned
L-type cardiac calcium channel by cyclic-AMP-dependent protein kinase.
FEBS Lett 342: 119 –123, 1994.
40. Qu J, Altomare C, Bucchi A, DiFrancesco D, and Robinson RB.
Functional comparison of HCN isoforms expressed in ventricular and
HEK 293 cells. Pflügers Arch 444: 597– 601, 2002.
41. Richards MW, Butcher AJ, and Dolphin AC. Ca2⫹ channel betasubunits: structural insights AID our understanding. Trends Pharmacol Sci
25: 626 – 632, 2004.
42. Satin J and Cribbs LL. Identification of a T-type calcium channel
isoform in murine atrial myocytes (AT-1). Circ Res 86: 636 – 642, 2000.
43. Schulla V, Renstrom E, Feil R, Feil S, Franklin I, Gjinovci A, Jing XJ,
Laux D, Lundquist I, Magnuson MA, Obermuller S, Olofsson CS,
Salehi A, Wendt A, Klugbauer N, Wollheim CB, Rorsman P, and
Hofmann F. Impaired insulin secretion and glucose tolerance in beta
cell-selective Cav1.2 Ca2⫹ channel null mice. EMBO J 22: 3844 –3854,
2003.
44. Seino S, Chen L, Seino M, Blondel O, Takeda J, Johnson JH, and Bell
GI. Cloning of the alpha 1 subunit of a voltage-dependent calcium channel
expressed in pancreatic beta cells. Proc Natl Acad Sci USA 89: 584 –588,
1992.
45. Seisenberger C, Specht V, Welling A, Platzer J, Pfeifer A, Kuhbandner S, Striessnig J, Klugbauer N, Feil R, and Hofmann F. Functional
embryonic cardiomyocytes after disruption of the L-type alpha 1C
(Cav1.2) calcium channel gene in the mouse. J Biol Chem 275: 39193–
39199, 2000.
46. Sinnegger-Brauns MJ, Hetzenauer A, Huber IG, Renstrom E, Wietzorrek G, Berjukov S, Cavalli M, Walter D, Koschak A, Waldschutz
R, Hering S, Bova S, Rorsman P, Pongs O, Singewald N, and
Striessnig J. Isoform-specific regulation of mood behavior and pancreatic
␤ cell and cardiovascular function by L-type Ca2⫹ channels. J Clin Invest
113: 1430 –1439, 2004.
H1971