| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
-adrenergic signaling in transgenic mice that express the amino terminus of -ARK1
1Department of Surgery and 2Division of Cardiology, Department of Medicine, Duke University Medical Center, Durham, North Carolina 27710
Submitted 4 February 2003 ; accepted in final form 15 July 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-adrenergic receptor (-AR) kinase-1 (-ARK1) is elevated during heart failure; however, its role is not fully understood. -ARK1 contains several domains that are capable of protein-protein interactions that may play critical roles in the regulation of GPCR signaling. In this study, we developed a novel line of transgenic mice that express an amino-terminal peptide of -ARK1 that is comprised of amino acid residues 50–145 (-ARKnt) in the heart to determine whether this domain has any functional significance in vivo. Surprisingly, the -ARKnt transgenic mice presented with cardiac hypertrophy. Our data suggest that the phenotype was driven via an enhanced -AR system, as -ARKnt mice had elevated cardiac -AR density. Moreover, administration of a -AR antagonist reversed hypertrophy in these mice. Interestingly, signaling through the -AR in response to agonist stimulation was not enhanced in these mice. Thus the amino terminus of -ARK1 appears to be critical for normal -AR regulation in vivo, which further supports the hypothesis that -ARK1 plays a key role in normal and compromised cardiac GPCR signaling.
G protein-coupled receptors; isoproterenol; -adrenergic receptor kinase-1
-ADRENERGIC RECEPTOR KINASE-1 (-ARK1), which is a G protein-coupled receptor (GPCR) kinase (GRK), is a critical regulator of myocardial signaling and function. Myocardial expression of this member of the GRK family is elevated in several cardiovascular disorders including ventricular hypertrophy and heart failure (reviewed in Ref. 15). Moreover, genetic manipulation of -ARK1 in the heart can profoundly affect in vivo cardiac performance (22, 29). These in vivo studies include transgenic mice with cardiac-specific expression of the carboxyl terminal (CT) domain (-ARKct) of -ARK1 (22). The -ARKct inhibits -ARK1 by blocking the binding of -ARK1 to the 
-subunits of dissociated heterotrimeric G proteins (G), a process that is required for -ARK1 membrane translocation and GRK activation toward activated GPCRs (21). This significant influence on cardiac function by the manipulation of myocardial -ARK1 activity appears to be due at least in part to altered desensitization of -adrenergic receptors (-ARs), which are the primary GPCRs that regulate cardiac contractility; however, signaling through other myocardial GPCRs may also be altered (31).
As is common to all GRKs, -ARK1 has a three-domain structure with the central catalytic domain flanked by amino terminal (NT) and CT domains of 
200 amino acids (17). The -ARKct is involved in membrane targeting, whereas the role of the NT domain (-ARKnt) is less well characterized. However, the -ARKnt may also be critical for the in vivo regulation of this GRK, as it contains several identified domains that may take part in novel protein-protein interactions. Previous studies have shown that the -ARKnt may be involved in GPCR recognition and binding (27) and receptor phosphorylation (39) as well as membrane targeting (18) and anchoring (19, 21, 42). In addition, a fusion protein that contains a peptide from the -ARKnt (amino acids 50–145) was found to strongly inhibit its association with microsomal membranes (25).
Known to play a role in membrane binding, the -ARKnt is the subject of the current study. Interestingly, it has also been shown to contain domains that can interact with calmodulin (24), the membrane phospholipid phosphatidylinositol 4,5-bisphosphate (28), tubulin (7), and caveolin (5). Carman and co-workers (5) have recently suggested that -ARK1 might be subject to some level of tonic inhibition in cells by the binding of calveolin to a domain within the -ARKnt. Other recent data have suggested that the -ARKnt contributes to the phosphorylation-independent regulation of metabotropic glutamate receptors (9). Finally, there is an apparent conserved regulator of G protein signaling (RGS) domain (amino acids 51–173) within the -ARKnt (6, 37). This putative RGS domain of -ARK1 is somewhat less conserved than among other RGS proteins, which suggests that the -ARKnt RGS domain may have atypical functions (13). Consistent with this are the recent in vitro studies that show that the -ARK1 RGS domain can directly interact with G
q to regulate its signaling and that it does so without alteration of GTPase activity (6, 37).
These findings using peptides from the -ARKnt raise the possibility that the -ARKnt may play a unique and key role in the regulation of -ARK1 function. Moreover, because several of the proteins shown to associate with the -ARKnt (including Gq, calmodulin, tubulin, and caveolin) are involved in different aspects of cardiac signaling and regulation, targeting the NT of this GRK may provide critical insight into novel in vivo actions of -ARK1 in the heart. In this study, we tested whether the -ARKnt has any cardiac physiological significance by characterizing transgenic mice with myocardial-targeted expression of the -ARKnt peptide.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-ARK1 was ligated into a previously described plasmid that contains the -myosin heavy chain (-MHC) gene promoter and the SV40 intro poly(A)+ signal (22). The -MHC--ARKnt transgene underwent pronuclear injection at the Duke Comprehensive Cancer Center Transgenic Facility. Second-generation adult animals (2–6 mo of age) were used for all studies, and results were compared with those found in nontransgenic littermate control (NLC) mice. Institutional review board approval for all mouse experiments was obtained from Duke University. Positive transgene expression was confirmed by SV40 Northern and -ARKnt Western blotting assays. Morphological and genomic characterization of myocardial hypertrophy. Heart weight-to-body weight ratios (HW/BW; measured in mg/g) were calculated after mice were weighed, killed, and dissected as described (14). In addition, the hearts of mice were perfusion fixed, and histological staining for collagen expression was done with Masson's trichrome using standard procedures (14). For gene expression studies, total RNA was extracted from frozen hearts with RNAzol (Biotecx Laboratories; Houston, TX), and this was used for Northern blot analysis for SV40, atrial naturetic factor (ANF), or GAPDH with cDNA probes using standard methods (2).
Protein immunoblotting. Frozen hearts were homogenized in RIPA buffer [150 mM NaCl, 50 mM Tris·HCl (pH 8.0) 5 mM EDTA, 1% (vol/vol) Nonidet P-40, 0.5% (wt/vol) deoxycholate, 10 mM NaF, 10 mM sodium pyrophosphate, and 1 mM each PMSF, aprotinin, and leupeptin], and proteins were transferred to nitrocellulose membranes as previously described (16). -ARKnt was immunoblotted using rabbit serum obtained from rabbits immunized with a glutathione-S-transferase--ARKnt peptide (Strategic Biosciences; Ramona, CA). Protein immunoblotting for -ARK1 was carried out using polyclonal antibodies specific for the carboxyl terminus (Santa Cruz Biotechnologies; La Jolla, CA). For GRK fractionation studies, frozen hearts were homogenized in buffer A (in mM: 25 Tris·HCl, 5 EDTA, 5 EGTA, pH 7.4, 20 NaF, 5 dTT, and 1 each PMSF, aprotinin, and leupeptin) and centrifuged at 39,000 g for 30 min at 4°C, and the supernatant was collected for the cytosolic fraction. The pellet was resuspended in buffer A-1% Triton X-100 (Mallinckrodt Chemical; Paris, KY) and was Dounce homogenized before being placed on ice for 30 min and centrifuged at 39,000 g for 30 min at 4°C. The supernatant was then collected for the particulate fraction. Equal amounts of cytosolic and particulate proteins were Western immunoblotted using -ARK1 antibodies. The ratio of protein is expressed as the amount in the particulate fraction divided by the sum of the cytoslic and particulate fractions. Other proteins were immunoblotted using specific antibodies and protein extracted using RIPA buffer or buffer A.
Echocardiography. Two-dimensional guided M-mode echocardiography (ATL; Bothell, WA) was carried out as previously described (11). Mice were studied while in the conscious state using gentle manual restraint after a period of acclimatization. Fractional shortening was calculated as (LVDd–LVDs)/LVDd, and left ventricular (LV) mass (LVM) was calculated as 1.055 x [(LVDd + IVSW + PW)3–LVDd3], where LVDd is the LV diastolic dimension, LVDs is the LV systolic dimension, IVSW is the intraventricular septum wall width, and PW is the posterior wall width.
Myocyte isolation and contractility. Adult mouse myocytes were isolated as previously described (30). Briefly, hearts were excised and perfused using the Langendorff technique with Joklik's modified MEM containing 150 U/ml collagenase II. Cells were maintained on laminin-coated plates. Myocytes were visualized with a Nikon inverted microscope with a solid-state charge-coupled device camera attached and were displayed on a video monitor. Two platinum electrodes placed in the bathing fluid were connected to a stimulator to field stimulate the myocytes at a pulse duration of 5 ms and a frequency of 0.5 Hz. Myocyte cell edges were enhanced and processed using a video edge detection system after basal stimulation and also 1–2 min after isoproterenol (Iso; 1 µM) administration. Calibrated myocyte length was converted from analog to digital online (PowerLab, ADInstruments) and stored on a computer. We studied all myocytes 2 h after isolation. Data from five to eight consecutive contractions were averaged. The contractile parameter was percent cell shortening, which was calculated as the percent change in myocyte length from rest to minimum length. We studied 10–15 myocytes from each heart for each group. Each individual heart was considered as n = 1.
-Adrenergic receptor binding. Receptor binding on myocardial membranes was performed as previously described using the nonselective -AR ligand 125I-labeled cyanopindolol (16). Saturation binding assays were performed on 15 µg of membrane protein using increasing amounts of ligand (4.3–412.5 pM). Nonspecific binding values were determined in the presence of 10 µM alprenolol. Reactions were conducted in 500 µl of binding buffer at 37°C for 1 h and then terminated by vacuum filtration through glass-fiber filters. All assay points were performed in triplicate, receptor density (Bmax) was determined (using GraphPad software), and receptor density was normalized to milligrams of membrane protein.
Hemodynamic evaluation of intact animals. Mice were anesthetized with ketamine (100 mg/kg of body wt) and xylazine (5 mg/kg of body wt) (16) and were analyzed as previously described (22). Briefly, after endotracheal intubation was complete, mice were connected to a rodent ventilator. After bilateral vagotomy, the chest was opened and a 1.8-Fr high-fidelity micromanometer catheter (Millar Instruments) was inserted into the left atrium, advanced through the mitral valve, and secured in the LV. Hemodynamic measurements were recorded at baseline and 45–60 s after injection of incremental doses of Iso. Doses of Iso were specifically chosen to maximize the contractile response but limit the increase in heart rate. Ten sequential beats were averaged for each measurement.
Adenylyl cyclase activity. Membrane fractions were prepared from hearts as described (16). Adenylyl cyclase activity was assessed at baseline and in response to the -AR agonist Iso using methodology previously described (16). Briefly, myocardial membranes (20 µg) were incubated for 10 min at 37°C, [32-P]ATP was isolated by anion exchange chromatography, and then cAMP was quantitated.
Lipid kinase assays. Myocardial membrane fractions were prepared, and -ARK1 was immunoprecipitated from protein using a -ARK1 monoclonal antibody before assay for phosphoinositide 3-kinase (PI3-K) activity (26). Briefly, after immunoprecipitation, the organic phase was spotted on TLC plates and resolved by chromatography. TLC plates were exposed to autoradiography.
Pump implantation and chronic drug treatment. Mini osmotic pumps (model 2002, Alzet) were implanted in mice anesthetized with ketamine (100 mg/kg of body wt) and xylazine (5 mg/kg of body wt) as described (16). Pumps were filled with Iso, nadolol, or vehicle (0.002% ascorbic acid in PBS) and were set to deliver Iso or nadolol at 30 mg·kg–1·day–1 for 7 days. Cyclosporine A (CsA, Sandimmune; Novartis; East Hanover, NJ) was administered to the mice by daily injection (30 mg·kg–1·day–1 ip) for 7 days.
Transverse aortic constriction. Mice were anesthetized with ketamine (100 mg/kg of body wt) and xylazine (5 mg/kg of body wt) as described (16), and microsurgical procedures were performed using a dissecting microscope (32). Animals were intubated, and transverse aortic constriction (TAC) was performed as described to yield a reproducible constriction. Sham-operated animals underwent the same procedure except for the aortic constriction. After 7 days of TAC, mice were anesthetized, and right and left carotid artery pressures were recorded (32). Hearts were excised, weighed, and frozen. Echocardiography was performed immediately before TAC and at 7 days post-TAC.
Generation of dual transgenic mice. Transgenic mice that express the -MHC-Gq transgene have been previously described (8) and were obtained with much appreciation from Dr. Gerald Dorn (University of Cincinnati). These mice were mated with the -MHC--ARKnt-18 line of mice to generate dual transgenic mice that expressed the -ARKnt and overexpressed Gq in the heart.
Statistical analysis. Data are expressed as means ± SE. Data were analyzed using two-way ANOVA or unpaired Student's t-test as indicated; P < 0.05 was considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-ARKnt in heart. To generate mice with targeted cardiomyocyte expression of the -ARKnt, a 285-bp cDNA that codes for bovine -ARK1 residues 45–150 was cloned into a vector that contains the -MHC promoter and SV40 poly(A)+ signal as previously described (22). Two independent founder lines of -ARKnt transgenic mice were established: -ARKnt-18 and -ARKnt-33. No gross phenotypic changes or unusual neonatal mortality levels were observed in these transgenic mice compared with the corresponding NLC mice. To assess transgene expression in these lines of mice, Northern and Western blotting assays were carried out. Positive expression of transgene mRNA was observed in both lines of -ARKnt mice (Fig. 1A), and this expression was cardiac specific (data not shown). Myocardial protein extracts prepared from the hearts of these two different lines of mice had identifiable -ARKnt peptide expression upon Western blotting with custom-made polyclonal antibodies raised against the -ARKnt (Fig. 1B). As shown in Fig. 1, cardiac expression of the -ARKnt is variable between lines with -ARKnt-18 mice showing higher expression. Unless indicated, subsequent assays were done with these mice.
|
Expression of -ARKnt induces cardiac hypertrophy. Because there was no overt gross phenotype evident in these mice, upon death of adult animals, we examined cardiac size and structure. First, we measured HW/BW (in mg/g) as a measure of cardiac hypertrophy. Hearts from -ARKnt-18 mice presented with statistically significant hypertrophy compared with NLC mice as heart mass was 17% higher (Fig. 2A). This hypertrophy was also reflected in the -ARKnt-33 line of mice (HW/BW: NLC, 4.50 ± 0.1 mg/g; -ARKnt-33, 5.05 ± 0.2 mg/g, or 13% higher; P < 0.05), which suggests specificity to -ARKnt expression and not a nonspecific transgene insertional process. In addition, there were no differences in hypertrophic levels in young vs. old mice (HW/BW of -ARKnt-18: 8 wk old, 5.02 ± 0.2 mg/g; >7 mo old, 4.90 ± 0.3 mg/g), which demonstrates that the increase in heart mass occurs within the first 2 mo of life. Hearts were also perfusion fixed, sectioned, and stained for collagen expression to determine whether the hypertrophy associated with -ARKnt expression promotes fibrosis. Histological sections contained no differences in collagen or fibrosis between -ARKnt-18 mice and the corresponding NLC mice at either 2 mo or 1 yr of age (data not shown).
|
To investigate whether the morphological alterations induced by -ARKnt expression were accompanied by molecular markers of hypertrophy, we measured ventricular expression of ANF, which is a cardinal marker of hypertrophy (32). Northern blots revealed a significant increase in ANF mRNA levels in ventricles from -ARKnt-18 mice (Fig. 2B), which is consistent with the increased heart mass measurements. To assess whether the hypertrophy observed was cardiomyocyte specific, we isolated individual myocytes from NLC and -ARKnt-18 mouse hearts. Using a video edge detection system, we determined the lengths of the individual cells and found that there was a significant increase in the length of myocytes isolated from -ARKnt-18 hearts compared with cells isolated from NLC hearts (Fig. 2C).
An initial assessment of in vivo physiology in these mice by echocardiography revealed no significant alterations in cardiac dimensions, wall thickness, or function between -ARKnt-18 and NLC mice (Table 1). However, when LVM was calculated from these measurements, there was a significant 23% increase found in the -ARKnt mice (Table 1), which is consistent with the determined heart mass increases that were identified (Fig. 2A). As an initial mechanistic assessment into this hypertrophy, we used phospho-antibodies to analyze the activity of MAPKs, specifically, ERK and JNK, in -ARKnt mouse hearts and found no difference compared with hearts of NLC mice (data not shown).
|
Effects of -ARKnt expression on -ARs in heart. Because previous transgenic mice that expressed the -ARKct region had altered -AR signaling (22), we measured myocardial -AR density and function. Saturation binding experiments revealed a moderate but statistically significant increase (1.5-fold) in cardiac -AR density in -ARKnt-18 mice (Fig. 3A). Protein immunoblotting of cardiac extracts for other members of the -AR signaling cascade, including the stimulatory and inhibitory G proteins (Gs and Gi, respectively) as well as adenylyl cyclase revealed no differences in myocardial levels of these proteins between -ARKnt and NLC mice (data not shown).
|
To determine whether the increase in -ARs in the -ARKnt transgenic mice had any functional consequences, we isolated individual myocytes from the hearts of the mice and used a field stimulator to determine the contractile responses. Cell shortening, or contractility, after electrical stimulation did not differ between NLC and -ARKnt cells (Fig. 3B). In addition, we treated the myocytes with the -AR agonist Iso and again observed no differences in the single-cell contractile response between NLC cells and cells that expressed the -ARKnt (Fig. 3B). We also used an intact mouse model to examine the hemodynamic responses of these mice. LV contractility and relaxation as measured by changes in maximum and minimum LV dP/dt, respectively, at baseline and after Iso stimulation did not differ between NLC and -ARKnt mice (Fig. 3C). These studies suggest that the increase in -AR number in the -ARKnt mice had no effect on acute responses to -AR stimulation in either the myocytes or the intact heart. To further confirm these findings, we examined the activity of adenylyl cyclase. We observed no differences in the baseline levels of adenylyl cyclase activity between NLC hearts and hearts that expressed the -ARKnt (Fig. 3D). Furthermore, there were no differences after Iso treatment (Fig. 3D).
Effects of -ARKnt expression on -ARK1 in heart. Because these data indicate that the increase in -AR density in -ARKnt hearts surprisingly does not lead to enhanced acute cardiac -AR signaling and function, we investigated potential reasons why this may occur. Because -ARK1 is a critical regulator of -ARs in the heart via densensitization (22), we first assessed whether the myocardial activity of this GRK was altered in -ARKnt mice. To determine whether -ARKnt expression altered endogenous -ARK1 in these mice, we measured protein levels, and, although there were no differences in overall protein levels between -ARKnt-18 and NLC mice (Fig. 4A), -ARKnt mice had a significantly higher proportion of -ARK1 in the membrane fraction vs. the soluble cytosolic fraction (Fig. 4B). Because -ARK1 must first translocate to membrane before it is activated to phosphorylate agonist-occupied GPCRs (17), this result indicates that the activation state of -ARK1 is enhanced in -ARKnt compared with NLC mice. In addition, we measured -ARK1-associated PI3-K activity levels, because PI3-K has been found to translocate with activated -ARK (26). We found that PI3-K activity was upregulated in the hearts of -ARKnt mice compared with NLC mice (Fig. 4C). Overall, these results (Fig. 4) suggest that -ARK1 activity is enhanced in the -ARKnt mice and may serve to negate acute signaling through the enhanced -AR density.
|
Cardiac hypertrophic responses in -ARKnt transgenic mice. To further investigate the cardiac -AR system in these animals, we chronically treated -ARKnt and NLC mice with Iso for 7 days via implanted mini osmotic pumps. This treatment has previously been shown to induce myocardial hypertrophy (23). We found that the HW/BW ratios of both -ARKnt and NLC mice were significantly increased after 7 days of Iso, and once again the hearts of -ARKnt-18 mice were larger than those of NLC mice (Fig. 5A). However, this increased cardiac hypertrophy in transgenic mice was apparently not due to enhanced -AR responsiveness but instead due to the increased baseline hypertrophy, as the HW/BW data plotted as the percent change compared with vehicle-treated control mice showed no differences in the hypertrophic responses between NLC and -ARKnt-18 mice (Fig. 5B).
|
To determine whether the -ARKnt would influence other forms of hypertrophy, mice were subjected to 7 days of TAC (1). This experimental model of pressure overload has previously been shown to induce an increase in heart size that is dependent on Gq-mediated signaling (1). Thus TAC introduces a hypertrophic mechanism that is independent of -AR signaling, which, interestingly, may also test whether the -ARKnt has any RGS-like function in vivo. Functional analysis by echocardiography established that there was no difference in the cardiac fractional shortening of NLC and -ARKnt mice in response to 7-day TAC (Fig. 5C), which is consistent with our other indexes of cardiac function in these mice (see Fig. 3). We found the hypertrophic response to TAC was similar to that induced by Iso, that is, larger hearts were observed in the -ARKnt mice (Fig. 5D), but there was no difference in the percent change in HW/BW compared with baseline (Fig. 5E). These results indicate that the enhanced cardiac Gq-mediated signaling induced by TAC (1) is not attenuated by -ARKnt expression and the hypertrophy mediated by the transgene appears to be independent and additive of any pressure-overload mechanism.
Determination of signaling pathways responsible for cardiac hypertrophy in -ARKnt-18 mice. Several mouse models of hypertrophy have been shown to be due to increased cardiac signaling of the Ca2+/calmodulin-activated phosphatase calcineurin (43). To investigate whether this could be a contributing mechanism for the hypertrophy observed with cardiac -ARKnt expression, we treated NLC and -ARKnt-18 mice for 7 days with CsA, a calcineurin inhibitor that has been shown to reverse hypertrophy in several animal models (10). This is an intriguing hypothesis, because Ca2+/calmodulin has been shown to bind and interact with the NT of -ARK1 (24). As shown in Fig. 6A, CsA treatment had no significant effects on the cardiac hypertrophy in -ARKnt mice, which indicates that the calcineurin signaling pathway is apparently not involved. Protein immunoblotting also revealed no changes in expression of calmodulin or calcineurin in -ARKnt mice (data not shown).
|
We also directly tested whether the increased -AR density and potentially chronically activated signaling contributed to the cardiac hypertrophy in these transgenic mice. To test this, we treated -ARKnt and NLC mice for 7 days with the nonselective -AR antagonist nadolol again using subcutaneously implanted mini osmotic pumps. After nadolol treatment, there was a significant decrease in the HW/BW ratio in -ARKnt mice (Fig. 6B). NLC mice were not affected by the treatment (Fig. 6B). Thus blocking chronic -AR signaling reversed the hypertrophy, which demonstrates the contribution of the increased -AR density in -ARKnt mice to the phenotypical characterization. In addition, enhanced -ARK1 activation as determined by the percentage of this GRK that was present in the membrane fraction (see Fig. 4B) was returned to normal levels by nadolol treatment (Fig. 6C), which suggests that chronic -AR activation induces -ARK1 activation.
-ARKnt expression attenuates cardiac -AR down-regulation. The data in Fig. 6C demonstrate that enhanced -ARK1 translocation is a function of -AR signaling and is not necessarily due to a direct effect of -ARKnt on -ARK1 regulation. Thus the -ARKnt may directly alter -AR density in cardiomyocytes. To address how the expression of the -ARKnt may lead to enhanced myocardial -AR density, we examined receptor downregulation after chronic -AR agonist exposure. As in Fig. 5, -ARKnt and NLC mice were chronically treated for 7 days with Iso. Previous studies have shown that after chronic Iso stimulation, cardiac -AR is reduced due to receptor downregulation (12, 23). After treatment with Iso or vehicle, cardiac membranes were prepared from NLC and -ARKnt mouse hearts and saturation binding was carried out to determine -AR density. As expected, Iso-treated NLC mice demonstrated significantly decreased cardiac -AR density compared with vehicle-treated NLC mice (Fig. 7). However, in surprising contrast, -AR density was absolutely maintained in -ARKnt mice and stayed enhanced over NLC mice (Fig. 7). Thus the -ARKnt imparts abnormal regulation on cardiac -ARs in these mice that leads to increased endogenous -AR density apparently through attenuation of -AR downregulation. One intriguing possibility for a mechanism is that the -ARKnt interferes with a caveolin-ARK1 or caveolin--AR interaction. These interactions may potentially be involved in -AR internalization (40). To begin to address this, the cardiac expression of caveolin-3, the primary caveolin isoform expressed in heart, was examined, and there were no differences in the levels or distribution (data not shown), which suggests that levels of the protein were not altered by the -ARKnt.
|
Does -ARKnt act as an RGS molecule in vivo? Data in Fig. 5 suggest that the -ARKnt does not have RGS activity in the heart after TAC. To address this potential function of the -ARKnt peptide more directly, we crossed -ARKnt mice with a transgenic mouse model of cardiac Gq overexpression (8). The Gq mice present with cardiac hypertrophy and other abnormalities (8). Their phenotype has previously been shown to be rescued by a transgenic mouse that overexpresses RGS4 in the heart (34). Interestingly, the dual-transgenic mice we generated had a synergistic increase in heart weight compared with either the -ARKnt or the Gq single-transgenic animals (Fig. 8). ANF levels were also found to be higher in the dual-transgenic mice (data not shown). These data coupled with the TAC results (Fig. 5) clearly demonstrate that the -ARKnt peptide does not exert any RGS activity on myocardial Gq-mediated signaling and that the hypertrophic phenotype is mediated by independent mechanisms (i.e., -AR signaling).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-ARKnt to examine its potential role in cardiac GPCR signaling in vivo. Surprisingly, these mice presented with cardiac hypertrophy that was characterized by an increase in the size of individual myocytes with no ventricular dilation, dysfunction, or fibrosis. Moreover, expression of the -ARKnt peptide led to alterations in myocardial -AR properties including increased density. Interestingly, this increased number of -ARs did not alter acute cardiac signaling or function following -agonist stimulation. This may be a result of enhanced desensitization due to increased -ARK1 activation as we found in -ARKnt transgenic mice an increase in the amount of -ARK1 in the membrane component of cells, which would suggest heightened GRK activation. Included was enhanced -ARK1-associated PI3-K activity in cardiac membranes isolated from -ARKnt mice. Chronically, the increased -AR density in the hearts of these transgenic mice appears to induce minor enhanced signaling that with time is ultimately responsible for the cardiac hypertrophy, as -AR blockade reversed the enlargement of cardiac mass in -ARKnt transgenic mice. The -ARKnt did attenuate -AR downregulation in response to chronic Iso treatment. These data suggest a potential mechanism, as cardiac -ARKnt expression can alter normal regulatory processes that involve -AR internalization thereby preventing downregulation, and chronically this can lead to increased baseline -AR density and the cardiac hypertrophic phenotype that was seen.
Previously, in vitro studies have implicated the NT of -ARK1 in either receptor-associated roles such as receptor targeting or binding (18, 19, 27, 42) or in unique GRK protein-protein interactions such as with tubulin (7) and caveolin (5). Logically, this follows other research that clearly defines applications for the central catalytic domain of the protein in receptor phosphorylation and for the CT of the protein in G binding, membrane translocation, and GPCR desensitization (21). Indeed, the -ARKnt may function as an adapter domain by binding to other molecules; however, this interaction may be more important than previously thought. The finding that chronic expression of the -ARKnt induces cardiac hypertrophy and prevents -AR downregulation suggests a significant role for this domain in GPCR signaling in the heart. More specifically, this role appears to be through effects on myocardial -ARs. Interestingly, from a mechanistic standpoint, the -ARKnt may perturb normal -AR regulation directly or indirectly through alterations in GRK function.
Concerning a -AR-mediated mechanism of the cardiac hypertrophy seen in -MHC--ARKnt transgenic mice, chronic stimulation of -ARs by agonist administration has been shown to induce cardiac hypertrophy (12, 23), which suggests the scenario that an increase in receptor number (with unchanged agonist concentration) may also lead to hypertrophy such as in the present -ARKnt mouse model. Importantly, a previous study has shown that transgenic overexpression of 1-ARs in the hearts of mice can induce myocyte hypertrophy accompanied by cardiac fibrosis and myofibrillar disarray (4), which suggests that upregulation of this receptor alone may be sufficient to induce the hypertrophic phenotype that we observe in the -ARKnt transgenic mice. However, because we did not see any negative physiology or pathology, either the increased density of receptors was not high enough for the toxic effects of 1-AR signaling, or the increased density was not exclusively of the 1-AR subtype. This will be explored in a future study.
Perhaps the most intriguing phenotype in these mice that implies a direct interaction between the -ARKnt and -AR regulation is the finding that -ARKnt expression in heart prevents downregulation of -ARs following chronic agonist stimulation. Thus the NT of -ARK1 may be involved in the receptor internalization processes, which precedes the loss of receptor density on the sarcolemmal membrane. Importantly, previous studies have suggested that -ARK1 may be involved in the sequestration process as it is known to colocalize in caveolae with the internalized 2-AR (36, 38). Furthermore, within the -ARKnt region is a caveolin-binding domain, and -ARK1 can associate with caveolin in vitro (5). Therefore, the -ARKnt in our mice may be acting as a caveolin-binding partner, which in the heart would be the caveolin-3 isoform. This -ARKnt-caveolin-3 interaction could potentially prevent caveolin from interacting with -ARK1 and/or internalizing -ARs. As a potential additional mechanism of this interaction, a decrease in caveolin-3 expression has recently been shown to be associated with cardiac hypertrophy and progressive cardiomyopathy in mice (44); thus in -ARKnt mice, although there was no overall decrease in caveolin-3 expression, there may be a functional decrease or alteration of the normal distribution of caveolin-3 that may lead to changes that could trigger cardiac hypertrophy. Additional studies are warranted to investigate this potential novel interaction of -ARK1 in the heart.
The rescue of the hypertrophic phenotype of the -ARKnt mice by chronic treatment with the neutral -antagonist nadolol is further evidence that signaling through the -AR is the principal component responsible for the phenotype of the model. Myocyte hypertrophy in a mouse model has previously been prevented by the administration of -AR blockers (3). In this study, administration of propranolol for 5 mo prevented cardiomyopathy in Gs transgenic mice (3). The -AR blocker treatment in the -ARKnt mice reversed the hypertrophy with only 1 wk of treatment. In addition, the increased membrane fractionation of endogenous -ARK1 in -ARKnt mice, which reflects heightened activity of this GRK, was normalized after nadolol treatment. These data suggest that the hypertrophy is not a developmental effect but rather is maintained by apparently low levels of constant signaling through the increased density of myocardial -ARs. Finally, in support of a -AR-mediated phenotype is the fact that inhibition of calcineurin activity via CsA treatment did not affect cardiac hypertrophy in -ARKnt mice.
-ARKnt could also regulate -AR signaling indirectly by affecting GRK function. Interestingly, although the -ARKnt transgenic mice had an increased number of -ARs, there was no alteration in acute -AR signal transduction as measured by adenylyl cyclase activity or -AR function as measured by in vivo hemodynamics and single myocyte contractility in response to Iso. In addition, one aspect of this study that is important to note is that -ARKnt expression does not appear to inhibit the actions of -ARK1 on GPCRs (i.e., -ARs), which is the result of -ARKct expression in the heart (22). Thus these two domains of -ARK1 have clearly distinct functions and interactions. However, -ARKnt targeting appears to alter other actions of this GRK that importantly also play a critical role in -AR regulation. The data suggest that the overall lack of increased -AR signaling may actually be due to increased membrane targeting of -ARK1, as this GRK is known to translocate from the cytoplasm to the plasma membrane in response to receptor occupancy and is thereby activated (36). This mechanism, however, appears to be the result of increased -AR density and is not due to the action of -ARKnt on -ARK1. The higher levels of receptor activation appear to be responsible for more -ARK1 membrane translocation and activation, because nadolol treatment also reversed this aspect of the -ARKnt phenotype. Regardless of the exact mediator (i.e., direct or indirect actions of the -ARKnt), this represents a potential mechanism to offset any increase in receptor number. Furthermore, increased -ARK1 translocation was associated with increased membrane PI3-K activity that also may play a role in altered -AR characteristics.
An alternate explanation of these findings is that the upregulation of -ARs is very modest at only 1.5-fold and may not be sufficient to induce any signaling alterations. However, this does not explain the hypertrophy that is developed, other than to speculate that -ARK1 activation does not completely compensate for the increased -AR number in the long term. Alternatively, -ARKnt expression may affect the cellular localization of -ARK1 independent of -AR signaling. These scenarios will be addressed in future studies.
Another important and novel finding of this study was that -ARKnt mice did not rescue the phenotype of the Gq transgenic mice. It was hypothesized that the -ARKnt may act as an RGS molecule since the -ARKnt does contain a partial RGS domain including residues found most importantly in -ARK-Gq binding (41). Thus the -ARKnt may be able to alter Gq signaling as previously demonstrated by cardiac RGS4 expression (34). The -ARKnt transgenic mice clearly cannot inhibit Gq-mediated signaling in the heart in vivo, as transgenic -ARKnt-Gq mice actually had an additional increase in hypertrophy (Fig. 8). This is further supported by the increased hypertrophy observed after TAC (see Fig. 5), which is also a Gq-mediated event (1). Thus the entire RGS-like domain of -ARK1 may be necessary for this activity. Importantly, the hypertrophic pathways induced by -ARKnt expression in the heart appear to be independent (and additive) to Gq-mediated pathways.
Herein, we have described a novel transgenic mouse model of cardiac hypertrophy caused by expression of an NT peptide of -ARK1 (GRK2). Although other cardiac GPCRs may be affected in these mice, altered -AR signaling in cardiomyocytes seems to be responsible for the hypertrophic phenotype. Our data demonstrate that the NT of -ARK1 is important in the interaction of the enzyme with the -AR and its signaling pathway perhaps by facilitating protein-protein interactions involved in receptor internalization and subsequent downregulation. These findings suggest a principal role for the -ARKnt in the -AR signal transduction pathway in heart and further demonstrate the importance of -ARK1 in overall cardiac regulation.
|
FOOTNOTES |
|---|
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.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1b-adrenergic receptors. In vivo 1-adrenergic receptor-mediated regulation of -adrenergic signaling. J Biol Chem 272: 21253–21259, 1997.-Adrenergic receptor blockade arrests myocyte damage and preserves cardiac function in the transgenic Gs mouse. J Clin Invest 104: 551–558, 1999.[ISI][Medline]
1-adrenergic receptor in transgenic mice. J Mol Cell Cardiol 32: 817–830, 2000.[ISI][Medline]
q/11 by an RGS domain in the G protein-coupled receptor kinase, GRK2. J Biol Chem 274: 34483–34492, 1999.q overexpression induces cardiac contractile failure in mice. Proc Natl Acad Sci USA 94: 8121–8126, 1997.-adrenergic signaling pathway alteration in isoproterenol-induced cardiac hypertrophy in male Sprague-Dawley rats. Jpn Heart J 38: 849–857, 1997.[Medline]
ARK1 expression in catecholamine-induced cardiac hypertrophy in transgenic mice over-expressing 1b adrenergic receptors. J Am Coll Cardiol 38: 534–540, 2001.-adrenergic receptor stimulation and blockade. Circulation 98: 1783–1789, 1998. interactions with PH domains and Ras-MAP kinase signaling pathways. Trends Biochem Sci 20: 151–156, 1995.[ISI][Medline]
ARK: an inhibitor of G-mediated stimulation of type II adenylyl cyclase. Proc Natl Acad Sci USA 91: 3637–3641, 1994. subunits of heterotrimeric G proteins on the -adrenergic receptor kinase. J Biol Chem 268: 8256–8260, 1993.-adrenergic receptor kinase or a ARK inhibitor. Science 268: 1350–1353, 1995.-adrenergic receptor stimulation in mice. J Mol Cell Cardiol 29: 2735–2746, 1997.[ISI][Medline]
-adrenergic receptor kinase to microsomal membranes. Modulation of the activity of bound kinase by heterotrimeric G protein activation. J Biol Chem 271: 985–994, 1996.-adrenergic receptor kinase 1. J Biol Chem 276: 18953–18959, 2001.-adrenergic receptor kinase 1 inhibitor prevents the development of myocardial failure in gene-targeted mice. Proc Natl Acad Sci USA 95: 7000–7005, 1998.-adrenergic receptor kinase in gene-targeted mice. J Biol Chem 273: 18180–18184, 1997.
q overexpressing mice. J Mol Cell Cardiol 33: 209–218, 2001.[ISI][Medline]
-adrenergic signaling and cardiac hypertrophy in transgenic mice overexpressing TGF-1. Am J Physiol Heart Circ Physiol 283: H1253–H1262, 2002.-Adrenergic receptor kinase (GRK2) colocalizes with -adrenergic receptors during agonist-induced receptor internalization. J Biol Chem 272: 9601–9604, 1997.
