| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Basic Medical Sciences, School of Veterinary Medicine;
Department of Medicinal Chemistry and Molecular Pharmacology, College of Pharmacy;
Department of Veterinary Clinical Sciences, School of Veterinary Medicine;
Department of Veterinary Pathobiology and Cancer Center; and
¶ Department of Animal Sciences, College of Agriculture, Purdue University, West Lafayette, Indiana, USA; and
|| Cardiovascular Institute, University of Pittsburgh, Pittsburgh, Pennsylvania, USA
2Correspondence: School of Veterinary Medicine, Purdue University, 625 Harrison St., West Lafayette, IN 47907, USA. E-mail: pond{at}purdue.edu
ABSTRACT
Skeletal muscle atrophy results from an imbalance in protein degradation and protein synthesis and occurs in response to injury, various disease states, disuse, and normal aging. Current treatments for this debilitating condition are inadequate. More information about mechanisms involved in the onset and progression of muscle atrophy is necessary for development of more effective therapies. Here we show that expression of the mouse ether-a-go-go related gene (Merg1a) K+ channel is up-regulated in skeletal muscle of mice experiencing atrophy as a result of both malignant tumor expression and disuse. Further, ectopic expression of Merg1a in vivo induces atrophy in healthy wt-bearing mice, while expression of a dysfunctional Merg1a mutant suppresses atrophy in hindlimb-suspended mice. Treatment of hindlimb-suspended mice with astemizole, a known Merg1a channel blocker, inhibits atrophy in these animals. Importantly, in vivo expression of Merg1a in mouse skeletal muscle activates the ubiquitin proteasome pathway that is responsible for the majority of protein degradation that causes muscle atrophy, yet expression of a dysfunctional Merg1a mutant decreases levels of ubiquitin-proteasome proteolysis. Thus, expression of Merg1a likely initiates atrophy by activating ubiquitin-proteasome proteolysis. This gene and its product are potential targets for prevention and treatment of muscle atrophy.—Wang, X., Hockerman, G. H., Green, H. W. III, Babbs, C. F., Mohammad, S. I., Gerrard, D., Latour, M. A., London, B., Hannon, K. M., Pond, A. L. Merg1a K+ channel induces skeletal muscle atrophy by activating the ubiquitin proteasome pathway.
Key Words: ectopic gene expression • hindlimb suspension • astemizole
SKELETAL MUSCLE ATROPHY, a reduction of contractile protein content and muscle strength, can result from muscle damage, disease, disuse or aging (1
2
3
4)
; the resultant debilitation can seriously compromise quality of life. Physical therapy and administration of growth factors (5)
, inhibitors of proteolysis (6)
, or stimulators of protein synthesis (7)
are methods being studied to treat atrophy; however, more effective treatments are needed. The ubiquitin proteasome pathway (UPP), by which protein substrates are degraded (8)
, is prominent among mechanisms known to modulate skeletal muscle atrophy; however, little is known about the factors activating this pathway.
The erg1 (Kv11.1, KCNH2) K+ channel conducts IKr current, which is partially responsible for repolarization of the cardiac action potential in many mammalian species, including mice and humans. Mutations in human erg1 (HERG1) can produce Long QT Syndrome (LQT2), an inherited cardiac disorder characterized by delay of cardiac repolarization, syncope, ventricular arrhythmias, and risk of sudden death (9)
. Erg1 K+ channels are composed of four separate 
-subunit proteins. Two alternative splice variants of the erg1 -subunit have been cloned from mouse (Merg1a and 1b; 10
) and human (HERG1A and 1B; 11
) cDNA libraries. The Merg1b alternative splice variant lacks the first 380 amino acids found in the 1a subunit and, instead, begins with 40 unique amino acids (10)
. Electrophysiological studies show that distinct functional K+ channels are produced by expression of each splice variant alone and by coexpression of both (10
, 11)
. A recent report indicates that in vivo cardiac IKr current is produced by a heteromultimeric erg1 channel, composed of two 1a and two 1b subunits (12)
. High levels of erg1a protein have been detected in adult heart and brain tissues of various mammals (10
11
12
13
14)
. Merg1 gene expression is detected in numerous tissues of embryonic mice, including skeletal muscle (13)
. To our knowledge, there are no reports of erg1a gene expression in adult mammalian skeletal muscle or of a role for erg1 function in modulation of proteolysis. Here we show that erg1, a voltage-gated K+ channel, is up-regulated in atrophic skeletal muscle of mice. Further, ectopic expression of Merg1a in skeletal muscle induces UPP proteolysis and atrophy, while ectopic expression of a dysfunctional dominant negative Merg1a (DN Merg1a, G628S; 15
) mutant decreases UPP proteolysis and inhibits atrophy in hindlimb-suspended mice. The data suggest that Merg1a participates in the onset of skeletal muscle atrophy by signaling increased UPP proteolysis.
MATERIALS AND METHODS
Animals
All procedures were approved by the Purdue Animal Care and Use Committee. ND4-Swiss Webster mice (Harlan-Sprague, Indianapolis, IN) were used in all procedures except for tumor studies which used athymic mice (NCr-nu; Harlan-Sprague). Animals were housed in Purdue University facilities, monitored by lab animal veterinarians and provided food and water ad libitum.
Hindlimb suspension
Custom suspension cages were constructed as described previously (16)
. Mice were placed in these cages resting in approximately a 30° head down tilt with their hindlimbs elevated so that they were unable to place any load on their hindlimbs. Control mice were kept in commercial mouse cages in a normal wt-bearing state.
Tumor induction
Kb human esophageal cancer cells (7000 cells/100 µl of RPMI medium with 10% FBS and 1% glutamine; 17
) were implanted subcutaneously (s.c.) in the right axilla of athymic mice (n=8; NCr-nu; Harlan-Sprague). Control mice (n=8) were injected with vehicle. Mice were weighed weekly for 6 wk. Tumors were measured by a digital caliper biweekly.
Western blot
For immunoblots, membrane proteins were extracted from gastrocnemius muscles and brain (14)
. Samples were immunoblotted by using erg1 antibody (Ab) (14)
. After blotting, membranes were stained with 0.1% Coomassie R-250 to confirm that membrane samples contained equal protein.
Tissue sections and staining
Gastrocnemius muscles were prepared and cryo-sectioned (14 µm) as described earlier (16)
. Sections were stained for ß-galactosidase (lacZ) activity (16)
or immunostained (16)
using erg1 Ab (14)
, except where noted. Images of sections were captured with a Leaf Micro–Lumina digital camera (Scitex; Tel-Aviv, Israel). The pixel number of each muscle fiber cross section was determined (Adobe Photoshop 6) and converted to µm2. Two sections (50 fibers each) from each muscle mid-section were analyzed.
Plasmids
The Merg1a (10)
, Merg1b (10)
and DN-Merg1a (G628S; 15) clones were in pBK/cytomeglovirus (CMV). The ubiquinated firefly luciferase (Ub-FL) in pGL-3/CMV was a gift from Dr. David Piwnica-Worms (18
; Washington University, St. Louis, MO). The CMV-nlacZ in pNL vector was purchased from the Center Commercial de Gros (Toulouse, France). The phRL synthetic Renilla luciferase (RL) reporter vector was purchased from ProMega (Madison, WI).
Electroporation
Mice were anesthetized with 0.01 µl/mg body wt of xylazine (1 mg/ml) and ketamine (9 mg/ml) in sterile saline. Gastrocnemius muscles of shaved hindlimbs were injected with plasmids and electroporated with 8 pulses at 200V/cm for 20 ms at 1 Hertz (19)
with an extracellular matrix (ECM) 830 ElectroSquare Porator (BTX; Hawthorne, NY).
Reverse-transcription polymerase chain reaction (RT-PCR)
RT-PCR was performed as described earlier (16)
, except that contaminating DNA was removed from total RNA extract by two 10-min treatments with DNase I (ProMega). Duplicate RNA samples were exposed either to reverse-transcriptase or to vehicle only. Template control samples did not receive RNA. The reverse-transcription product was amplified with PCR, combined with ethidium bromide, and electrophoresed on a 2% agarose gel. Primers for Merg1a were: 5' – CGC AGA ACA CCT TCC TCG ACA C - 3' (forward) and 5' – GCA GAA GCC GTC GTT GCA GTA G - 3' (reverse). Primers for Merg1b were: 5'- AGT CCT CCA TGG CGA TTC - 3' (forward) and 5' - GGC CTG CAG CTT ATA CTC - 3' (reverse).
Astemizole (Sigma, St. Louis, MO)
The antihistamine astemizole (20)
was suspended in water.
ECG
Mice were anesthetized by a 240 mg/kg body wt i.p. injection of 20 mg/ml Avertin (2, 2, 2-tribromoethanol; Sigma) in a solution of 1.25% tert-amyl alcohol in saline (21)
. A standard lead II ECG was obtained by placing 18-gauge needle electrodes s.c. in the right foreleg (–), left hind leg (+), and right hind leg (ground). ECG’s were recorded (3 min) using a multichannel Biopac MP100 physiological recording system (Biopac Systems Inc; Goleta). QT intervals were measured and corrected for heart rate using the formula: QTC = QTO/(RRO/100)0.5 (22)
.
Dual luciferase reporter assay
The Dual-Luciferase Reporter Assay Kit (Promega) was used in accordance with manufacturer’s instructions. Firefly luciferase and Renilla luciferase (RL) activities were measured with a TD-20/20 Luminometer (Promega).
Statistics
Data were analyzed by ANOVA for a completely randomized design. When significant differences were found, means were separated by Fisher’s Protected Least-Significance Difference. All data were analyzed using the General Linear Model Procedure of SAS. Statements of significance were based on P-levels as noted.
RESULTS
Tumor expression induces Merg1a synthesis and atrophy in skeletal muscle
Six weeks after injection of tumor cells, the average wt of tumor-bearing mice (minus tumor wt) was 10.8% (P
0.005) less than control mice. Tumor-expressing mice had an 18% (P0.001) lower gastrocnemius wt-to-body wt ratio (1.10±0.019%, SEM) than control mice (1.34±0.018%, SEM), demonstrating that tumor bearing mice lost muscle mass. The mean muscle fiber cross-sectional area (csa) of left gastrocnemius muscles from tumor-bearing mice was 25% less (P0.05) than that of control mice (1445.8±103.8 µm2 SEM vs. 1925.0±161.6 µm2 SEM, respectively), demonstrating that tumor-bearing mice experienced skeletal muscle atrophy. Western blot analysis of extracted membrane proteins detected two Merg1a proteins in cachectic mouse skeletal muscle as previously reported in brain and heart (14; Fig. 1
A). No Merg1 channel protein was detected in control tissue (Fig. 1A
).
|
Hindlimb suspension results in atrophy and Merg1a synthesis in skeletal muscle
Fourteen mice were hindlimb-suspended (16)
for either 4 (n=7) or 7 (n=7) d, while control animals (n=14) remained wt bearing. Left gastrocnemius muscles were embedded and cross sectioned, while right muscles were analyzed with Western blot (14)
and RT-PCR analyses (16)
. After 7 d, relative to control animals, suspended mice experienced a mean 25% decrease (P0.005) in ratio of gastrocnemius muscle wt to body wt and a 45% decrease (P0.05) in muscle fiber csa. Data show that hindlimb suspension induced atrophy in gastrocnemius muscles. Western blot analysis (14)
detected two Merg1a protein isoforms in atrophying skeletal muscle, but not in controls (Fig. 1B
). Consistently, Merg1a mRNA was detected in atrophic muscle (Fig. 1C
). A substantially lower concentration of Merg1a message was also detected in control muscle (Fig. 1C
, lane 3), suggesting that Merg1a protein is synthesized in this tissue at levels not detectable by our Western blotting procedure. After concentrating (4x) a control gastrocnemius muscle sample, Merg1a proteins were evident by blotting (Fig. 1D
). We did not detect proteins in the skeletal muscle sample that correlated to the predicted mass of Merg1b protein (95 kDa; Fig. 1D
). Indeed, RT-PCR analysis failed to detect Merg1b message in either control or atrophied muscle (Fig. 1E
). Using immunohistochemistry (16)
, we demonstrated that Merg1 proteins were localized in atrophic skeletal muscle fibers; none was found in control muscle (Fig. 1F
).
Electroporation results in expression of plasmid DNA
Left gastrocnemius muscles of 14 mice were injected with expression plasmid encoding Merg1a (30 µg; 10
), while left gastrocnemius muscles of another 14 mice received expression plasmid encoding Merg1b (30 µg; 10
). Right gastrocnemiuses of all mice were injected with expression plasmid encoding lacZ (30 µg; 10
). Gastrocnemius muscles were electroporated to facilitate plasmid uptake (19)
. After 7 d, Merg1a proteins (14)
were detected in Merg1a injected muscles by Western blot analysis; these proteins were not detected in control muscle (Fig. 2
A). A Merg1b protein was detected just above 95 kDa in muscles injected with Merg1b plasmid; this protein was absent from control muscle (Fig. 2B
).
|
Merg1a expression in wt-bearing mice induces skeletal muscle atrophy
Left gatrocnemius muscles of 14 mice were injected with lacZ (20 µg) and control (40 µg) expression plasmids. Right gastrocnemius muscles of seven of these mice were coinjected with Merg1a (20 µg; 10), lacZ (20 µg) and control pBK/CMV DNA expression plasmids (20 µg). The right gastrocnemius muscles of the remaining mice (n=7) were coinjected with three expression plasmids, each encoding for either: Merg1a (20 µg; 10
), lacZ (20 µg), or a dominant-negative Merg1a mutant (DN Merg1a; 20 µg; 15
). The DN Merg1a gene product is a pore mutant that is transported to the cell membrane but prevents Merg1 current conduction (15)
. Mice remained wt-bearing. Seven days after electroporation, gastrocnemius fibers expressing Merg1a and lacZ (blue fibers) experienced a significant 16.3% decrease in muscle fiber csa relative to fibers injected with lacZ plasmid only (Fig. 3
A1, 2; B1, 2). Blockade of Merg1a function by DN Merg1a prevented all but 6.7% of the atrophy induced by Merg1a expression (Fig. 3A
1, 3; B1, 3). This incomplete rescue is not surprising, because it is not likely that all randomly assembled channels would contain mutant channel subunit.
|
Block of Merg1a function inhibits the skeletal muscle atrophy induced by hindlimb suspension
Left gastrocnemius muscles of 14 mice were injected with lacZ expression plasmid (20 µg) and control plasmid (30 µg). Right gastrocnemius muscles received lacZ expression plasmid (20 µg) plus expression plasmid encoding DN Merg1a (30 µg; 16
). Twelve hours after electroporation (19)
, mice were either hindlimb-suspended (n=7) or remained wt-bearing (n=7). After 7 d, there was a 45% decrease in csa of suspended mouse muscle fibers relative to wt-bearing controls, showing that suspension induced atrophy (Fig. 3C
1, 2; D1, 2). Importantly, the muscle fibers from suspended mice expressing lacZ and DN Merg1a mutants (15)
were 13.4% smaller than fibers from muscles of wt bearing mice expressing lacZ (Fig. 3C
2, 3; D). Data demonstrate that, although DN Merg1a expressing fibers do not completely retain control fiber size (as expected), block of Merg1a function does attenuate atrophy. Further, DN Merg1a (15)
and lacZ expressing fibers did not undergo atrophy to the extent of fibers from the same section not expressing plasmid DNA (Fig. 3D
, compare stained and nonstained fibers). Thus, synthesis of DN Merg1a mutant blocks the onset of disuse atrophy in suspended mice, showing that Merg1a current conduction plays a role in atrophy. Note: Expression of lacZ alone did not affect muscle fiber size in either wt-bearing or suspended mice (Fig. 3D
1, 2; compare stained and nonstained fibers within each section). This observation was confirmed by fiber csa (data not shown).
Ectopic expression of Merg1b in gastrocnemius muscle does not induce atrophy
Left gastrocnemius muscles of seven mice were injected with lacZ expression plasmid (20 µg) and control expression plasmid (30 µg), while right gastrocnemiuses received lacZ expression plasmid (20 µg) and Merg1b expression plasmid (30 µg; 10). Seven days after electroporation (19)
, there was no significant difference in fiber csa of control muscle expressing lacZ alone (2456±13 µm2, SEM) and those expressing lacZ and Merg1b (2449±16 µm2, SEM). This finding suggests that the Merg1-induced significant decrease in fiber size is specific to the 1a variant.
Astemizole dose-response curve
Three groups of three mice each were gavaged once every 12 h with a suspension of astemizole, a potent erg1 channel blocker (20)
; to yield 80, 160, or 320 mg/kg body wt. A control group was gavaged with an equal vol. of water. ECGs and animal weights were measured daily. By treatment day seven, the corrected QT intervals (QTc’s, 22) of all treated animals were significantly longer (P0.02) than those of the control group, demonstrating that the drug blocked Merg1 channel in heart. Because wt loss was minimal and QTc prolongation was maximal, 160 mg/kg was used in the suspension study.
Astemizole treatment inhibits atrophy in hindlimb-suspended mice and increases muscle size in wt-bearing controls
Twenty-eight mice were randomly assigned to four treatment groups. All animals received an oral gavage of water twice daily for 4 d. All mice were then gavaged once every 12 h for 7 d with either: water (groups 1 and 3) or 160 mg/kg astemizole in water (groups 2 and 4). During the 7 d, groups 1 and 2 remained wt-bearing, while groups 3 and 4 were subjected to hindlimb-suspension. Data (Table 1
) show that astemizole treatment alone had no significant effect on body wt (note percent body wt change in groups 1 and 2). However, suspension caused a 7% decrease in body wt that was significantly alleviated by drug treatment (groups 3 and 4). Most importantly, the 15% decrease in muscle fiber csa that was experienced by hindlimb-suspended animals (groups 1 and 3) was blocked by astemizole treatment, bringing fiber csa values to control levels (compare groups 1 and 4). Interestingly, astemizole treatment produced significant increases in fiber csa, gastrocnemius muscle wt-to-body wt ratio and absolute gastrocnemius muscle wt in wt-bearing mice (groups 1 and 2). Data show that pharmacological block of Merg1 alleviates disuse atrophy and suggests that block of Merg1a function results in muscle hypertrophy. These effects likely result from decreased proteolysis produced by block of endogenous Merg1 channel. This is a reasonable supposition because we detect very low levels of Merg1a mRNA (Fig. 1C
) and protein (Fig. 1D
) in skeletal muscle of wt-bearing mice.
|
Merg1a expression induces UPP activity
Mouse gastrocnemius muscles (n=26) were injected with expression plasmids encoding Ub-FL (40 µg; 18
) and the RL reporter (10 µg). Activity of Ub-FL was normalized to RL activity to control for differences in the transfection efficiency from muscle to muscle. Ub-FL expression produces a measurable protein degraded by the UPP (18)
; the RL protein will not be degraded. Therefore, a reduction in the Ub-FL-to-RL activities ratio will represent an increase in UPP activity. Ub-FL and RL-injected mice (n=6) also received a plasmid encoding Merg1a (30 µg; 10
), while others (n=7) received the DN Merg1a mutant (30 µg; 15). Remaining mice (n=13) received control plasmids (30 µg). Twelve hours after electroporation (19)
, mice injected with DN Merg1a (n=7), and those injected with control plasmids (n=7) were hindlimb-suspended, while other mice (n=12) remained wt-bearing. After 7 d, ratios of Ub-FL-to-RL activity in gastrocnemius muscles were determined as measures of UPP activity; the lower ratios indicated higher UPP activity. Hindlimb suspension induced UPP activity (Fig. 4
A1, 3). Ectopic expression of Merg1a in gastrocnemius muscles of wt-bearing mice induced proteolysis of Ub-FL, decreasing the Ub-FL-to-RL activity ratio by 48% relative to control (Fig. 4A
3, 4). Also, block of endogenous Merg1a channel function by DN Merg1a expression significantly reduced UPP activity in suspended mice (Fig. 4A
1, 2). Data show that Merg1a K+ channel function modulates UPP activity.
|
Induction of UPP activity is Merg1a-specific
Both gastrocnemius muscles of 12 mice were coinjected with Ub-FL (40 µg; 18
) and RL (10 µg) expression plasmids. The right gastrocnemius muscles of six of these mice were coinjected with Merg1a expression plasmid (30 µg; 10), while right gastrocnemius muscles of the remaining six mice were coinjected with Merg1b expression plasmid (30 µg; 10
). Left gastrocnemius muscles received appropriate control plasmid (30 µg). Seven days after electroporation (19)
, ectopic expression of Merg1a significantly increased UPP activity as evidenced by a decrease (80%) in the relative activity of Ub-FL (Fig. 4B
1, 2); however, ectopic expression of Merg1b did not increase UPP activity as demonstrated by an insignificant (9.4%) decrease in the Ub-FL-to-RL ratio (Fig. 4A
1, 3), an amount that could result from coassembly of Merg1b subunit with low levels of endogenous Merg1a. Data show that Merg1a, and not Merg1b, specifically increases UPP activity.
DISCUSSION
Skeletal muscle health involves maintenance of an intricate balance between protein synthesis and degradation. Atrophy results from a perturbation in this intricate modulation of the various synthetic and proteolytic pathways and very little is known about mechanisms involved in initiation of the imbalance (2
, 3
, 23)
. Our data demonstrate that Merg1a channel function participates in initiation of skeletal muscle atrophy in response to muscle disuse or cachexia by signaling an increase in UPP proteolysis. We show that Merg1 channel function is an initiating factor acting upstream of atrophy: 1) Merg1 proteins are detected [day 4 of suspension] before the onset of significant atrophy [day 7 of suspension]; 2) ectopic expression of Merg1a induces a decrease in fiber csa in wt-bearing limbs of mice; and 3) genetic and pharmacologic attenuation of Merg1 channel function prevents atrophy in hindlimb-suspended mice. Our studies also demonstrate that the Merg1a splice variant is expressed in skeletal muscle, while Merg1b is not detected, which strongly suggests that the Merg1 channel in this tissue is composed of Merg1a subunits only. Also, expression of Merg1a, and not Merg1b, results in decreased muscle fiber size and increased UPP activity in wt-bearing mice. Perhaps the more extensive Merg1a NH2 terminus is necessary for up-regulation of the UPP. This portion of the protein does contain a number of unique signaling sequences, including a PAS-PAC domain, numerous potential protein kinase C (PKC) phosphorylation sites, a protein kinase A (PKA) phosphorylation site, and a potential cAMP/cGMP-dependent protein kinase phosphorylation site (10)
. Further, our data strongly suggest that Merg1a channel function is necessary to the atrophic process because both expression of the dysfunctional DN Merg1a mutant and astemizole treatment (pharmacologic channel block) inhibit the decrease in fiber size induced by suspension. Interestingly, although UPP activity is known to function during atrophic remodeling of the heart (24)
, physiologically relevant levels of Merg1 current (IKr) are necessary for normal cardiac and are not likely to induce atrophy. Perhaps expression of the Merg1b splice variant in heart (12)
is involved in this regulation. Therefore, the functional consequence of Merg1 channel current conduction may be determined by Merg1 channel subunit composition.
The possible mechanism(s) by which Merg1a function signals UPP activity is intriguing. It is known that HERG function is regulated by both PKC and PKA (25)
and by PIP2 (26)
, serving as a link between adrenergic stimulation and regulation of the cardiac action potential. HERG channel function is also regulated by thyrotropin-releasing hormone (TRH) through G-proteins to mediate changes in the frequency of action potentials during the TRH response in GH3 cells (27)
. Further, Merg1 function affects spike-frequency adaptation in neuroblastoma cells (28)
and decreases the firing frequency of human ß-cells (29)
. Interestingly, high-frequency stimulation of myoblasts from 20 d old rat embryos potentiates L-type Ca2+ channels (by phosphorylation of the channel) and increases Ca2+ flux through the sarcolemmal membrane (30)
. Perhaps, Merg1 activity changes the response of skeletal myocytes to neural stimulation by modulating calcium channel activity. Merg1a activity may lower calcium flux across the sarcolemmal membrane and the concentration of depolarizing signal reaching the sarcoplasmic reticulum. Ultimately, the intracellular calcium concentration ([Ca2+]i) would be decreased. Diminished [Ca2+]i precedes many of the functional changes resulting from hind-limb suspension (31)
, including the switch in predominant fiber type that occurs in slow twitch muscle undergoing atrophy; however, the mechanism(s) responsible for lowered [Ca2+]i is not well understood (31
, 32)
. In summary, Merg1a channel function is an initiator of disuse- and cachexia-stimulated atrophy, acting upstream of UPP proteolysis. The mechanism involved in the modulation of UPP by Merg1a channel function obviously begs further study.
ACKNOWLEDGMENTS
The authors thank Dr. David Piwnica-Worms (Washington Univ.; St. Louis, MO) for his gift of plasmid encoding the ubiquinated firefly luciferase. We thank Drs. Gordon Coppoc and Eli Asem (Purdue Univ.) for support. A.L.P. would like to thank Drs. Jeanne M. Nerbonne (Washington University School of Medicine; St. Louis, MO) and David Van Wagoner (Cleveland Clinic Foiundation; Cleveland, OH) for helpful discussions about ion channel physiology. This project was funded in large part by the American Heart Association (Scientist Development Grant #0235363N to ALP) and in part by USDA-CSREES Hatch Project number INDO76045 and the Purdue University Department of Basic Medical Sciences.
FOOTNOTES
1 These authors contributed equally to this work. 
Received for publication November 18, 2005. Accepted for publication February 16, 2006.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  H. Shi, C. Zeng, A. Ricome, K. M. Hannon, A. L. Grant, and D. E. Gerrard Extracellular signal-regulated kinase pathway is differentially involved in beta-agonist-induced hypertrophy in slow and fast muscles Am J Physiol Cell Physiol, May 1, 2007; 292(5): C1681 - C1689. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
