1 Genomic Physiology Group, Department of Physiology, University of Massachusetts Medical School, Worcester 01655
2 Laboratory of Genetic Physiology, Department of Medicine, Brigham and Womens Hospital, Boston, Massachusetts 02115
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
gene expression; aging; anti-adrenergic; G protein-coupled receptors; physiological genomics
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The ß-adrenergic signaling cascade is composed of membrane receptors, signal transducers, and downstream effector molecules. The receptor-signaling cascade is dynamically regulated by the phosphorylation state catalyzed by common and pathway-specific enzymes. Although the individual candidate approach has provided some clues as to the changes that occur with aging, the basic premise guiding these studies is that the determinants of ß-adrenergic responsiveness will be few and the regulatory circuits will be simple. However, it is clear that the signaling pathways for membrane receptors are not simple but are highly integrated and exhibit heterologous regulation. Thus we hypothesized that the regulation of ß-adrenergic responsiveness would be more complex than can be examined in a simple candidate approach and that a more complete understanding of the mechanisms responsible for this overall loss of responsiveness requires a much broader approach. Therefore, to evaluate the contributions of the individual elements of these complex, interacting systems to the deficit in ß-adrenergic signaling, we used a genomic approach to quantitate the levels of expression of 8,799 transcripts coupled with direct analysis of the candidate proteins identified as being differentially expressed. Consistent with our notion, we identified 19 genes whose differential expressions may contribute to the modulation of adrenergic responsiveness known to occur in the aged heart.
![]() |
MATERIALS AND METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Tissue harvesting.
Young (4 mo of age) and aged (22 mo of age) adult rats were anesthetized with 50 mg/kg ip pentobarbital and hearts were excised. The hearts were perfused and rinsed to remove blood and immediately frozen with clamps prechilled in liquid nitrogen. The atria and any extraneous tissue were removed from the heart tissue wafers, and the remaining ventricular tissue was stored in liquid nitrogen until used.
RNA extraction and analysis.
Between 400 and 500 mg of each frozen tissue sample (left and right ventricle combined) was rapidly thawed and the total RNA was extracted using TRIzol (Invitrogen, Carlsbad, CA). Transcript profiling with Affymetrix GeneChips (Affymetrix, Santa Clara, CA) was performed using U34A rat expression arrays (containing 8,799 sequences) according to Affymetrix protocols, and the data were collected using MicroArray Suite 5. We maintained strict adherence to indices of quality control. Total RNA yields for young and aged hearts were similar at 1.19 ± 0.32 and 1.25 ± 0.15 µg/mg wet weight of ventricular myocardium. All RNA samples were analyzed by denaturing agarose gel electrophoresis prior to analysis on the arrays, and all RNA samples exhibited intact 28S and 18S ribosomal RNA. Yields of in vitro transcribed RNA were high ranging from 70110 µg and were not different between the two groups; in vitro transcription products that were less than 45 µg were discarded and the labeling repeated. All samples were analyzed on Test2 chips prior to analysis on the U34A expression arrays. Sample quality as assessed by 3'/5' ratios of endogenous transcripts (e.g., actin, GAPDH), and exogenous added RNA (Bio B, C, and D) was high (<2), and these values were not different between groups. Analysis on the expression levels using the rat U34A chips revealed no gross differences between the groups. Regarding the overall patterns of expression, there was an insignificant reduction in the number of transcripts called present in the hearts from the old rats because 26.4 and 21.8% of the transcripts were called present in the hearts from young and aged rats, respectively (P = 0.353).
The data has been deposited at the Gene Expression Omnibus web site (GEO, http://www.ncbi.nlm.nih.gov/geo/); the set designation is GSE421.
Immunoblot analysis.
Approximately 100 mg of ventricular myocardium from an additional group of animals (4 young, 4 aged) was homogenized in 10 vol of 25 mM Tris, 25 mM NaCl, 1 mM sodium vanadate, 10 mM NaF, 10 mM pyrophosphate, 0.5 mM EGTA, and 1 mM phenylmethylsulfonyl fluoride. Protein content of the homogenates was determined using a bicinchoninic acid technique (Pierce) and solubilized in 5% glycerol, 1% SDS, 32 mM Tris, 2.5% ß-mercaptoethanol, and 0.05% bromophenol blue.
SDS-polyacrylamide gel electrophoresis was performed as previously described (13) except that mini-gels 0.75-mm thick and 10% polyacrylamide were used. Separated proteins were transferred to nitrocellulose and exposed to primary antibodies from Calbiochem (San Diego, CA), J-QUE Biologics (Worcester, MA) and Santa Cruz Biotechnology (Santa Cruz, CA). Secondary antibodies were conjugated to horseradish peroxidase, and the chemiluminescence was monitored using X-ray film. For some of the proteins (ß1-adrenergic receptor, adenylyl cyclase 5/6, and adenosine A1 receptor), blocking peptides were used. In all three cases, excess blocking peptide added to the primary antibody abolished the signal.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
We next examined the identity of the differentially expressed transcripts (Fig. 1, Supplemental Table 1). Of the 263 genes, 163 were annotated, and of these annotated genes, 34% (56 transcripts) could be classified as signaling/cell communication (Supplemental Table 2). Of these, approximately half encoded proteins directly or indirectly involved in G protein-coupled receptor signaling pathways. There were also a large number of genes associated with metabolism (24%) and defense processes (12%) that exhibited both up- and downregulation of expression. Since a major phenotypic alteration in the aged heart is a reduced signaling in response to adrenergic stimulation, we focused our further analysis on the profiles of transcripts involved in the regulation of this pathway. We identified 19 potentially important gene products (Table 1); of these, 16 were upregulated while only 3 were downregulated with aging. Importantly, we observed a decrease in ß1-adrenergic receptor and adenylyl cyclase mRNAs, whereas the mRNA encoding ß-arrestin increased. The results also demonstrate an increase in mRNAs encoding the adenosine A1 receptor and phospholipase D, which could increase anti-adrenergic effects. Moreover, the mRNAs encoding the muscarinic M3 receptor, nicotinic acetylcholine receptor ß3-subunit, and nicotinic acetylcholine receptor-related protein were increased, as was the mRNA encoding atrial natriuretic peptide (ANP), S-adenosylmethionine synthetase, and guanylate kinase-associated protein. Taken together, these changes may contribute to a reduced sympathetic drive.
|
|
Age-associated change in protein expression.
Immunoblot analysis was utilized to examine whether the changes in gene expression observed with aging were accompanied by changes in abundance of some of the key proteins (Fig. 2). The analysis was performed on cardiac extracts from an additional group of animals (4 young, 4 aged). The results indicate that the expression of ß1-adrenergic receptor and adenylyl cyclase are decreased in the aged heart (Fig. 3). [It should be noted that adenylyl cyclase 2 was found to be reduced in the microarray analysis; however, antibodies specific for adenylyl cyclase 2 are not available. The antibody used in this analysis recognizes adenylyl cyclases 17. The predominant from of adenylyl cyclase in rat heart is 5/6 (6, 20).] Immunoblot analysis also indicated that the expression of the adenosine A1 receptor, ß-arrestin-2, nicotinic acetylcholine receptor ß3-subunit, and phospholipase D are increased in the aged heart. The expression of these six proteins is parallel with the observed changes in the mRNA levels responsible for these proteins. As a control, we examined the expression of protein kinase C-, a protein whose mRNA did not exhibit a change. Consistent with this, the levels of protein kinase C-
protein did not change with aging.
|
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Reduced ß-adrenergic responsiveness in the heart is a key determinant of myocardial function in aged hearts as well as hypertrophic and failing hearts. In this study we employed transcript profiling to examine genes differentially expressed between young and aged hearts. We have identified 263 transcripts that are differentially expressed between young and aged hearts, and among these are 19 whose patterns of expression may contribute to ß-adrenergic responsiveness. Although we did not demonstrate causality, differential expression of 11 of these genes may result in a decrease in adrenergic responsiveness. However, we also identified eight genes whose differential expression might contribute to an enhanced adrenergic responsiveness, and these changes might be compensatory to the anti-adrenergic effects of the former set of genes. Interestingly, of these 19 genes identified, 16 were upregulated in the aged rat heart while only 3 were downregulated. Thus the regulation of adrenergic responsiveness during aging is not due to a generalized loss of transcriptional activity but is due to an active upregulation of a series of genes. Moreover, measurements of the overall pattern of gene expression revealed a similar occurrence, that aging in the rat heart is not due to generalized "shut-off" of gene expression. Indeed, out of the 263 genes that exhibit differential gene expression, more than two-thirds were upregulated. Array analysis has been performed on aging mouse liver, brain, and skeletal muscle (33), as well as fibroblasts isolated from humans of different ages (29). The current study is in agreement with these previous reports in that aging is not associated with widespread alterations in gene expression. Estimates of the number of genes that exhibit differential gene expression range from 1 to 4% of the total number of genes queried (16, 21, 28, 29). With the exception of the analysis of cultured cells, aging is not associated with an overall decrease in gene expression, because liver, brain, and skeletal muscle studies reveal that a near equal numbers of transcripts are up- or downregulated with aging (16, 21, 23, 28). However, the present study differs from these reports in that the number of genes whose expression increases with age outnumbers those decreasing by greater than twofold.
Studies on the molecular mechanisms of the reduced adrenergic signaling in aging are few. In our study, we were intrigued by the number of differentially expressed transcripts that could be classified as relating to signaling/cell communication. Approximately 34% of the annotated genes belonged to this category. Further categorization revealed that half of these transcripts were directly involved in G protein-coupled receptor signaling. Xiao and colleagues (34) reported a decrease in both ß1- and ß2-adrenergic receptors in aged rat myocytes with no change in the Gs or Gi subunits and no change in the ß-adrenergic receptor kinase (ßARK). Chin and colleagues (4) confirmed the lack of changes in
-subunits while Kilts et al. (24) demonstrated a decrease in Gi. In contrast to the paucity of studies in aged myocardium, numerous studies in left ventricular hypertrophy and failure have indicated that changes in components directly involved in ß-adrenergic signaling play a causal role. Studies have indicated decreases in ß1- and ß2-receptors, decreases in adenylyl cyclase, and increases in ßARK (1, 11, 25). A causal role for these changes has been shown in numerous transgenic, knockout, and gene transfer studies. Moderate overexpression of the ß-adrenergic receptors using transgenic or gene transfer approaches, decrease expression of ßARK in heterozygous knockout mice, or attenuation of ßARK activity by transgenic expression of the ßARK inhibitor (ßARKct) all lead to increases in ß-adrenergic signaling (10, 18, 30, 32). Our results, viewed in light of these studies, strongly suggest that the demonstrated decrease in ß1-adrenergic receptor and adenylyl cyclase and an increase in ß-arrestin play a causal role in the decreased adrenergic signaling in aged hearts.
In addition to gene products directly involved in ß-adrenergic signaling, several signaling pathways that can modulate ß-adrenergic signaling were also differentially regulated between the young and the aged hearts. Adenosinergic signaling pathways operating via adenosine A1 receptors are known inhibitors of ß-adrenergic signaling (7). The mRNA encoding the adenosine A1 receptor and a possible downstream adenosine receptor effector protein, phospholipase D (5, 31), are both increased with aging (2.9- and 4.8-fold increase, respectively). This, coupled with the previously reported increase in extracellular adenosine in the aging heart would further reduce ß-adrenergic signaling (8, 9). The mRNAs encoding the muscarinic receptor M3, nicotinic acetylcholine receptor-ß3, and the nicotinic acetylcholine receptor-related protein are increased in the aged heart (3.3-, 5.9-, and 4.6-fold increase, respectively). Because the parasympathetic nervous system opposes the adrenergic system, an increased expression of these elements might contribute to elevated anti-adrenergic activity in the aged heart. In addition, we also observed increases in guanylate kinase-associated protein and ANP, which would result in increased cGMP production/signaling and, subsequently, a decrease in adrenergic signaling (3, 22).
Given the profound decrease in adrenergic signaling in the aged heart, we were intrigued with the finding of differential expression that may constitute compensatory responses to the above described decreases in cAMP/adrenergic signaling. Several transcripts whose gene products couple with Gs and result in increased intracellular cAMP levels were increased with aging (dopamine D3 receptor, serotonin receptor; 5.1- and 3.1-fold increase, respectively). Similarly, the adenylyl cyclase-activating polypeptide was increased 2.3-fold, potentially enhancing cyclase activity, while AKAP 84, the cAMP-dependent protein kinase anchor protein, was increased 6.3-fold. Although the consequences of increased AKAP expression are unclear, previous reports suggest that increased AKAP expression results in enhanced protein kinase A activity (12). AMP deaminase 1 (6.4-fold increase) would be expected to decrease adenosine, thus modulating the effects of the increased adenosine receptor. Similarly, the increase in ANP would be modulated by the increased abundance of the ANP clearance receptor (4.1-fold increase). We also observed an increase (4.8-fold) in P2Y6 pyrimidinergic receptor. P2Y6 signaling exhibits positive inotropic effects, counteracting the effects of decrease inotropic effects of the loss of adrenergic signaling (19). Furthermore, we found a decrease (0.4-fold) in the expression of iNOS. iNOS expression is elevated in heart failure and has been shown to contribute to the decreased adrenergic stimulation in that pathophysiological state. Thus the decrease in iNOS with aging would have a compensatory effect on adrenergic signaling in that state (2, 3, 14, 15, 22).
In conclusion, these studies reveal several important points. First, consistent with several reports, aging affects the accumulation of a small fraction of the genes queried. Second, surprisingly in our study, two-thirds of the genes differentially expressed are upregulated with aging. Third, of the genes exhibiting differential gene expression, at least 19 genes may contribute to the altered adrenergic responsiveness known to occur in the aged heart.
![]() |
ACKNOWLEDGMENTS |
---|
Editor M. Stoll served as the review editor for this manuscript submitted by Editor R. E. Pratt.
GRANTS
This publication was made possible by National Institute on Aging Grant AG-11491 and National Heart, Lung, and Blood Institute Grant HL-66045 (both to J. G. Dobson, Jr.).
![]() |
FOOTNOTES |
---|
Address for reprint requests and other correspondence: J. G. Dobson, Jr., Dept. of Physiology, Univ. of Massachusetts Medical School, 55 Lake Ave. N, Worcester, MA 01655 (E-mail: James.Dobson{at}umassmed.edu).
10.1152/physiolgenomics.00076.2003.
1 The Supplementary Material for this article (Supplemental Tables 1 and 2) is available online at http://physiolgenomics.physiology.org/cgi/content/full/00076.2003/DC1 .
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|