Skip Navigation

Cardiovascular Research 2003 57(3):715-726; doi:10.1016/S0008-6363(02)00738-1
© 2003 by European Society of Cardiology
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Ashton, K. J
Right arrow Articles by Headrick, J. P
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ashton, K. J
Right arrow Articles by Headrick, J. P
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Copyright © 2003, European Society of Cardiology

Effects of A1 adenosine receptor overexpression on normoxic and post-ischemic gene expression

Kevin J Ashtona, Kirsty Holmgrena, Jason Pearta, Amy R Lankfordb, G Paul Matherneb, Sean Grimmondc and John P Headricka,*

aHeart Foundation Research Centre, Griffith University Gold Coast Campus, Southport, Queensland, QLD 4217, Australia
bDepartment of Pediatrics and the Cardiovascular Research Center, University of Virginia Health Sciences Center, Charlottesville, VA, USA
cInstitute for Molecular Bioscience, University of Queensland, St. Lucia, Queensland, Australia

* Corresponding author. Tel.: +61-7-5552-8292; fax: +61-7-5552-8802. j.headrick{at}mailbox.gu.edu.au

Received 15 August 2002; accepted 14 October 2002


   Abstract
Top
 Abstract
1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 
Objectives: To identify potential molecular genetic determinants of cardiovascular ischemic tolerance in wild-type and transgenic hearts overexpressing A1 adenosine receptors (A1ARs). Methods: cDNA microarrays were used to explore expression of 1824 genes in wild-type hearts and ischemia-tolerant mouse hearts overexpressing A1ARs. Results: Overexpression of A1ARs reduced post-ischemic contractile dysfunction, limited arrhythmogenesis, and reduced necrosis by ~80% in hearts subjected to 30 min global ischemia 60 min reperfusion. Cardioprotection was abrogated by acute A1AR antagonism, and only a small number (19) of genes were modified by A1AR overexpression in normoxic hearts. Ischemia-reperfusion significantly altered expression of 75 genes in wild-type hearts (14 induced, 61 down-regulated), including genes for metabolic enzymes, structural/motility proteins, cell signaling proteins, defense/growth proteins, and regulators of transcription and translation. A1AR overexpression reversed the majority of gene down-regulation whereas gene induction was generally unaltered. Additionally, genes involved in cell defence, signaling and gene expression were selectively modified by ischemia in transgenic hearts (33 induced, 10 down-regulated), possibly contributing to the protected phenotype. Real-time PCR verified changes in nine selected genes, revealing concordance with array data. Transcription of the A1AR gene was also modestly reduced post-ischemia, consistent with impaired functional sensitivity to A1AR stimulation Conclusions: Data are presented regarding the early post-ischemic gene profile of intact heart. Reduced A1AR transcription is observed which may contribute to poor outcome from ischemia. A1AR overexpression selectively modifies post-ischemic gene expression, potentially contributing to ischemic-tolerance.

KEYWORDS Adenosine; Gene expression; Ischemia; Receptors; Reperfusion


   1. Introduction
Top
 Abstract
 1. Introduction
2. Methods
 3. Results
 4. Discussion
 References
 
Molecular mechanisms underlying cell death and dysfunction following ischemia-reperfusion are undefined, and molecular genetic determinants of ischemic tolerance poorly characterized. Research has focussed on post-translational modification of phenotype. However, pre- and post-ischemic phenotypes will be determined by gene expression prior to and following insult. Recent evidence indicates acute responses, previously considered post-translational, involve protein expression elements [1] and may be sensitive to gene transcription. Investigators have commenced analysis of gene expression in ischemic hearts [2–4] in the hope of identifying genes involved in genesis of ischemic injury. Such investigations are essential if the therapeutic potential of ‘gene therapy’ is to be realized in ischemic tissue. Few if any studies have addressed the genetic basis of ischemia-tolerant phenotypes.

We recently characterized marked protection from ischemia with cardiac A1AR overexpression, leading to reduced contractile dysfunction, de-energization, necrosis and infarction [5–8]. Degree of protection exceeds that for other interventions, including preconditioning [8]. Molecular mechanisms remain to be determined, but may involve KATP channels [7], are common to preconditioning [8], and differ from those for A3AR activation [9]. We have not ascertained whether tolerance involves transcriptional changes prior to and following ischemia. The aims of the present study were to identify effects of A1AR overexpression and ischemia on myocardial gene expression. A cDNA microarray containing ~2000 genes, or ~10% of the cardiovascular genome [10], was employed to assess expression in wild-type hearts and hearts overexpressing A1ARs. Differential expression was verified by quantitative real-time PCR analysis in a subset of genes, and expression of the A1AR gene was studied.


   2. Methods
Top
 Abstract
 1. Introduction
 2. Methods
3. Results
 4. Discussion
 References
 
2.1 Langendorff model
Investigations conformed with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996). Hearts were obtained from mature 20–28-week-old male and female wild-type mice (n=44, 29.1±3.6 g body weight, 133±7 mg wet heart weight) and transgenic mice overexpressing cardiac A1AR (n=41, 27.8±2.4 g body weight, 126±8 mg wet heart weight) anesthetized with 50 mg/kg sodium pentobarbital administered intraperitoneally. Details regarding generation and characterization of transgenic mice have been published previously [5,7,8]. Hearts were removed and perfused in a Langendorff mode as described in detail previously [6,7,9].

After 20 min stabilization hearts were switched to pacing at 400 beats/min [5–9]. After a further 10 min baseline measurements were made and hearts subjected to 30 min global ischemia and 60 min reperfusion or 90 min aerobic perfusion (non-ischemic controls). This was performed in wild-type (n=9 ischemia-reperfusion, n=9 non-ischemic) and transgenic hearts overexpressing A1AR (n=8 ischemia-reperfusion, n=9 non-ischemic). Function was monitored throughout, and coronary effluent throughout reperfusion collected for purine analysis by HPLC [11] and lactate dehydrogenase analysis using a commercial spectrophotometric assay kit (Sigma Chemicals, St. Louis, MO, USA) [8,9]. Ectopy during the initial 10 min reperfusion was assessed: total number of abnormal or premature beats were determined and ectopy calculated as: %Ectopy=abnormal beats/total beatsx100%. Time for hearts to commence beating following ischemia was also determined. On completion of experiments hearts were freeze-clamped in aluminium tongs cooled in liquid N2 and stored at –80 °C until RNA extraction.

To assess the role of A1AR activation, wild-type (n=9) and transgenic hearts (n=8) were subjected to ischemia-reperfusion with infusion of the A1AR antagonist DPCPX (200 nM) initiated 10 min prior to ischemia. The agent was infused during and following ischemia since we previously identified protective functions of adenosine at these times [11]. Finally, A1AR-mediated negative chronotropic responses to the non-selective agonist 2-chloroadenosine were assessed as described previously [12] in normoxic wild-type (n=8) and transgenic hearts (n=8), and post-ischemic wild-type (n=9) and transgenic hearts (n=8) after 45 min reperfusion.

2.2 RNA isolation and DNA microarray analysis
Ventricular myocardium was homogenized and total RNA extracted using standard TRIzol and DNase treatment with subsequent spin column purification (Qiagen, Hilden, Germany). Ventricular RNA from a subset of hearts investigated (n=8 for non-ischemic wild-types, n=6 for all other groups) was pooled per group to reduce inter-experiment and inter-animal variability.

Preparation of differentially labeled fluorescent cDNA, hybridization and washing of microarray slides were performed as described previously [13]. Microarrays consisted of 1632 sequence verified cDNAs from the Research Genetics mouse UniGene set, 192 other known genes (primarily involved in development and differentiation) and positive and negative controls, all immobilized in duplicate on glass slides. Each hybridization was repeated with experimental and control samples reciprocally labeled.

2.3 Digital image analysis and data processing
Scanning of arrays was performed using a GMS 418 scanner (Affymetrix, Santa Clara, CA), image analysis was performed using ImaGene 4.1 (BioDiscovery, Los Angeles, CA). Briefly, this involved spot segmentation, gridding and flagging of poor and empty spots (exclusion of spots with signal:noise ratio <3, spots <40% of average area, irregular spots). GeneSight-Lite (BioDiscovery, Los Angeles, CA) was used for local background subtraction and global normalization of fluorescent intensities before exporting as a spreadsheet. Since ≥2 replicates are essential in any microarray study [14] experiments were repeated on 2–3 replicate slides for each group, with each slide containing duplicate spots for every gene. The cDNA labeling with Cy3 and Cy5 was reversed on one slide from each group and expression patterns compared to validate consistency of expression profiles and eliminate possible non-uniform labeling.

2.4 Quantitative real-time PCR
Differential expression was confirmed in nine transcripts via real-time PCR using an iCycler iQ real-time PCR detection system (Bio-Rad Laboratories, Hercules, CA). We selected nine genes identified in arrays as being unchanged, induced or down-regulated in one or more experimental groups studied (providing a broad range of expression to validate with real-time PCR): Pdha1, Idh3a, Ddit3, Rgs5, Hspe1, Ywhag, Atp5F1, Fhl2 and Csf1. Additionally we investigated expression of the A1AR gene (Adora1) and cardiac myosin heavy chain alpha (Myhca). The 18S ribosomal RNA was used as endogenous control to correct for minor experimental variations. Expression Changes were calculated using the 2–{Delta}{Delta}CT method [15]. Expression is expressed relative to non-ischemic wild-type hearts.

2.5 Statistical analyses
Data are shown as means±S.E.M. Functional data, purine and LDH efflux, and real-time PCR data were analyzed by multi-way ANOVA, with a Tukeys post-hoc test when significance was detected. Significance was accepted for P<0.05. For microarray data, significant changes were identified for genes whose log2 transformed expression ratios were shown to differ by t-test analysis of replicated array data at the P<0.05 level. A Bonferroni's form of correction was applied in which P-values determined in t-tests were multiplied by the number of observations, reducing possible type I errors in large data sets.


   3. Results
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
4. Discussion
 References
 
3.1 Functional responses to ischemia-reperfusion
Baseline functional parameters for all wild-type and transgenic hearts perfused under normoxic conditions for 30 min are shown in Table 1. Function was similar in wild-type and transgenic hearts, although intrinsic heart rate was reduced with A1AR overexpression. Ischemia abolished contractile function in wild-type and transgenic hearts, as documented previously [5–9]. Diastolic contracture was rapid, with time to contracture (rise in diastolic pressure of 20 mmHg) shorter in wild-types (258±14 s) vs. transgenics (397±17 s) (P<0.05). Peak contracture pressure during ischemia was not different (84±7 mmHg in wild-types, 89±8 mmHg in transgenics). Coronary flow recovered to similar levels in both groups, and did not differ significantly from non-ischemic hearts (data not shown). Diastolic pressure remained elevated after reperfusion at ~20 mmHg in wild-type hearts but recovered to control in transgenic hearts (Fig. 1A). Ventricular developed pressure, +dP/dt and –dP/dt all remained depressed by ~50% in wild-type hearts (Fig. 1). Contractile recoveries were enhanced by A1AR overexpression. Tissue necrosis (LDH loss) and de-energization (purine efflux) were reduced by A1AR overexpression (Fig. 1C). Significant ectopy during the initial 10 min of reperfusion (38±5%) was reduced by A1AR overexpression (27±4%, P<0.05). Time for hearts to recommence beating on reperfusion was reduced in transgenic (22±3 s) vs. wild-type hearts (33±3 s, P<0.05). Treatment with A1AR antagonist (DPCPX) during ischemia-reperfusion reduced recoveries in wild-type and transgenic hearts (Fig. 1). The antagonist largely normalized tolerance between wild-types and transgenic.


View this table:
[in this window]
[in a new window]

  		Table 1 Baseline functional parameters in perfused hearts from wild-type and transgenic mice

 

Figure 1
View larger version (33K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Recovery from 30 min ischemia and 60 min reperfusion in wild-type (n=9) and transgenic hearts overexpressing A1ARs (n=8). Effects of A1AR antagonism with 200 nM DPCPX are shown in wild-type (n=9) and transgenic (n=8) hearts. (A) left ventricular end-diastolic (EDP) and developed pressure (LVDP); (B) +dP/dt and –dP/dt; (C) LDH and purine efflux. Values are means±S.E.M. *, P<0.05 vs. wild-type; {dagger}, P<0.05 vs. untreated. Note no statistical differences between functional recoveries and effluxes in DPCPX treated transgenic vs. wild-type hearts.

 
3.2 Effects of A1AR overexpression on gene expression in normoxic hearts
A total of 989 arrayed genes (~55%) were detectable in hearts. Transgenic hearts displayed differential expression of only 19 genes (Table 2). Macrophage colony stimulating factor 1 (Csf1) and MARCKS-like protein were the only genes induced. Of 17 genes down-regulated, four were involved in metabolism, two for structural/motility proteins, six involved in cell signaling, two involved in gene transcription, two in protein synthesis, and one was unclassified.


View this table:
[in this window]
[in a new window]

  		Table 2 Genes significantly altered by ischemia-reperfusion in both wild-type and transgenic hearts

 
3.3 Effects of ischemia-reperfusion on gene expression
Analysis of expression revealed 75 genes altered by ischemia-reperfusion in wild-type hearts, and 67 in transgenic hearts (Table 2). We grouped genes into those encoding proteins involved in: (i) energy/metabolism; (ii) structure/motility (matrix, cytoskeletal, contractile, microtubule); (iii) cell signaling (receptors, effectors, intracellular transducers, protein modifiers); (iv) cell defence/growth (stress response, apoptotic, DNA synthesis/repair, immune/inflammatory); (v) gene expression (RNA synthesis, transcription factors, chromatin-related); (vi) modifying protein synthesis (processing, turnover, post-translational modification, trafficking, ribosomal); and (vii) unclassified functions.

Transcriptional patterns for post-ischemic hearts are depicted in Fig. 2. Similar total numbers of transcripts were altered in wild-type (75) and transgenic (67) hearts (Fig. 2, Table 2). However, the general pattern was quite different in the groups—less were induced vs. down-regulated in wild-type hearts whereas the reverse occurred in transgenics (Fig. 2). Of genes modified, 24 were common to wild-types and transgenic hearts (indicated in Table 2). Thus, 43 genes were modified in transgenic but not wild-type hearts.


Figure 2
View larger version (34K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Transcriptional responses to ischemia-reperfusion in wild-type and transgenic hearts. Shown are numbers of genes belonging to different functional groupings which are (A) induced, or (B) repressed.

 
3.4 Cellular metabolism genes
Repression of metabolic genes was characteristic of post-ischemic wild-type hearts (Table 2). Enzymes of carbohydrate metabolism (e.g. pyruvate dehydrogenase, isocitrate dehydrogenase, GAPDH), fatty acid metabolism (e.g. malonyl-CoA-decarboxylase, dodecenoyl-coenzyme A delta isomerase), and mitochondrial oxidative phosphorylation (adenine nucleotide translocase, mitochondrial ATP synthase) were repressed post-ischemia. PEP carboxykinase 1 and adenylosuccinate lyase were induced in wild-types (Table 2). A1AR overexpression completely or partially reversed repression of metabolic genes (Table 2, Fig. 3A), selectively reduced mitochondrial-uncoupling protein 2, and induced glycogen phosphorylase, Slc3a2, and NADH dehydrogenase (ubiquinone) (Table 2).


Figure 3
View larger version (36K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Real-time PCR and microarray analysis of genes involved in: (A) energy or substrate metabolism and cell signaling; and (B) cell defence/growth and gene transcription. Expression levels (log2) in transgenic non-ischemic (TG Non-Isc, n=6), wild-type post-ischemic (WT Isc, n=6) and transgenic post-ischemic (Tg Isc, n=6) hearts are shown relative to wild-type non-ischemic hearts (n=8). Open bars show real-time PCR data, shaded bars microarray data. Values are means±S.E.M. *, P<0.05 vs. wild-type normoxic hearts; {dagger}, P<0.05 vs. transgenic hearts. §, P<0.05 vs. post-ischemic hearts; {ddagger}, P<0.05 vs. post-ischemic transgenic hearts.

 
3.5 Structural/motility genes
Structural, cytoskeletal and myofibrillar genes was altered by ischemia. β-actin, cadherin-2, β-catenin, tropomodulin 2, {alpha}-tubulin 4, and ankyrin-1 were all down-regulated (Table 2). Only elastin was significantly induced in wild-types. These changes, excluding down-regulation of β-catenin, were reversed by A1AR overexpression.

3.6 Cell signaling genes
In wild-type hearts ischemia-reperfusion down-regulated HLA-B associated transcript 2, a Ca+2 signal transducer, protein phosphatase-2 regulatory subunit A, PCTAIRE protein kinase and Rgs5 (Table 2). A1AR overexpression reversed changes in all genes down-regulated in wild-types, excluding PCTAIRE protein kinase 1 and Rgs5 (the latter was down-regulated to a greater extent in transgenic hearts). Insulin-like growth factor binding protein 1, a DNA activated protein kinase, and midkine were induced in wild-type and transgenics.

3.7 Cell defence and growth-related genes
Transcription of 14-3-3 gamma, referred to as Ywhag, and a phosphoprotein (enriched in astrocytes), were down-regulated by ischemia in wild-type and transgenic hearts. Heat shock proteins Hsp84, Hsp47 and chaperonin 10 (Hspe1) were induced by ischemia-reperfusion in both groups. Induction of DNA-damage inducible transcript 3 (Ddit3 or Gadd153/Chop-10) was also observed in both groups. A1AR overexpression selectively induced an early growth response gene, interferon-related developmental regulator 1, clusterin, and Hsp40.

3.8 Transcription/chromatin, and protein synthesis genes
Transcripts encoding DEAD box polypeptides were induced in transgenic hearts, together with other transcription factors (Msx1, Spop and Sox9). Significant down-regulation of the four and a half LIM domains 2 gene (Fhl2) was observed in all groups (Table 2). There was a general trend of post-ischemic gene induction in transgenic hearts compared to down-regulation in wild-type hearts (Table 2). Deoxyhypusine synthase was the only gene induced by ischemia in both groups.

3.9 Quantitative real-time PCR analysis and A1AR function
Real-time PCR confirmed differential expression of Pdha1, Idh3a, Atp5F1, Csf1, Rgs5, Ywhag, Hspe1, Ddit3 and Fhl2 (Fig. 3). Expression data from the alternate methodologies were comparable (Fig. 4). Expression of A1AR gene (Adora1) was predictably higher in transgenic vs. wild-type hearts. Interestingly, Adora1 was induced by ischemia-reperfusion in transgenic hearts but down-regulated in wild-types (Fig. 5). This pattern was consistent with reduced sensitivity of A1AR mediated bradycardia in wild-type but not transgenic hearts. The pEC50 for 2-chloroadenosine-mediated bradycardia in normoxic hearts was greater in transgenic (7.9±0.2) vs. wild-type hearts (6.5±0.2, P<0.05). Ischemia-reperfusion reduced sensitivity to 2-chloroadenosine in wild-type (pEC50=5.1±0.3, P<0.05 vs. normoxic), equivalent to a 10-fold decline in sensitivity. This did not occur in post-ischemic transgenic hearts (pEC50=7.4±0.3).


Figure 4
View larger version (20K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 Relationship between gene expression determined via cDNA microarray analysis and quantitative real-time PCR. Expression data (from Fig. 3) is shown for the genes Pdha1, Idh3a, Atp5F1, Csf1, Rgs5, Ywhag, Hspe1, Ddit3 and Fhl2. Data were acquired from transgenic non-ischemic (n=6) and post-ischemic hearts (n=6), and wild-type post-ischemic hearts (n=6). The mean log2 expression ratios are shown relative to wild-type non-ischemic hearts (n=8) for the purposes of comparing real-time PCR data with microarray data.

 

Figure 5
View larger version (20K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 Expression of the A1AR gene (Adora1) and the cardiac myosin heavy chain alpha gene (Myhca) measured by quantitative real-time PCR in non-ischemic and ischemic transgenic hearts, and ischemic wild-type hearts (n=6 in all cases). Expression ratios are shown relative to non-ischemic wild-type hearts (n=8) for purposes of comparison. Values are means±S.E.M. *, P<0.0001 vs. wild-type normoxic; {dagger}, P<0.0001 vs. transgenic.

 

   4. Discussion
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
References
 
Identifying transcriptional patterns conferring resistance (or sensitivity) to ischemic insult is an important step in understanding ischemic and reperfusion injures, and in developing genetic approaches to ischemic disorders. As indicated in the recent review by Henriksen and Kotelevtsev, although it is not technically feasible to simultaneously profile expression of large numbers of proteins it is possible to profile transcription of large numbers of genes [16]. This facilitates identification of potential proteins and pathways involved in generating particular phenotypes. In the current study we identify changes in expression of novel and characterized genes in ischemia-tolerant hearts, and also observe a number of novel gene changes associated with ischemic insult.

4.1 Transcriptional effects of A1AR overexpression in non-ischemic tissue
Consistent with previous data [7,8], A1AR overexpression generates an ischemia-tolerant phenotype characterized by improved functional recovery and impaired necrosis (Fig. 1). Additionally, we show arrhythmogenesis is reduced. While data support roles for mitochondrial KATP channel activation [7] and energy metabolism [6], molecular determinants of ischemia tolerance are poorly defined. The ischemia-tolerant phenotype is sensitive to A1AR antagonism, as is tolerance in wild-type hearts (Fig. 1). These data support a protective role for endogenous adenosine in wild-type hearts, and demonstrate ischemia tolerance in transgenic hearts results primarily from activation of A1ARs during ischemia-reperfusion rather than adaptive changes over the life of the animal, in agreement with prior observations [8]. Consistent with this conclusion, A1AR overexpression only modified a small number of genes in normoxic hearts (Table 2). Of two genes induced one is involved in inflammation (Csf1) while the other (MARCKS-like protein) is of unknown function. A number of genes down-regulated with A1AR overexpression are involved in cell signaling (G-protein coupled receptor kinase, insulin-like growth factor binding protein, regulator of G-protein signaling), and it is tempting to speculate enhanced activation of G-coupled A1AR leads to modulation of proteins involved in G-protein signaling. One novel observation is down-regulation of mitochondrial uncoupling protein 2, which is an inhibitor of myocardial efficiency [17]. We have previously shown adenosine enhances efficiency of O2 use in myocardium [18]. Since the protected phenotype is sensitive to A1AR antagonism, the importance of these varied gene changes in ischemic-tolerance is questionable.

4.2 Gene expression in ischemic-reperfused myocardium
There is evidence acute myocardial responses, thought to be post-translational in nature (e.g. preconditioning), may involve translational or transcriptional components [1]. Moreover, there is indirect evidence adenosine modifies myocardial gene expression [19,20]. Microarrays are powerful tools in identifying transcriptional effects of ischemia and modification by A1AR overexpression. The approach has been employed in post-ischemic [3,4], cardiomyopathic [21], hypertrophic [22], and failing hearts [23]. Of the almost 1000 arrayed genes expressed, 75 were consistently modified by ischemia-reperfusion in wild-type myocardium. The major proportion (~80%) are repressed as opposed to induced (Table 2, Fig. 2). The post-ischemic profile involves down-regulation of cellular metabolic genes and cell/tissue structure and motility genes, with lesser numbers of genes involved in signaling, defence/growth, and gene and protein expression being modified. A1AR overexpression switches the post-ischemic profile from primarily gene repression to gene induction (Fig. 2).

While data regarding gene expression is insufficient to confirm functional relationships between and roles for different proteins, transcriptional profiling provides clues to mechanisms contributing to ischemic tolerance, injury and remodeling, and permits development of testable hypotheses. For example, a ‘cluster’ of glycolytic genes were down-regulated, including pyruvate dehydrogenase, GAPDH, isocitrate dehydrogenase, and aldolase. These changes are predicted to reduce flux through glycolysis, consistent with impaired post-ischemic glucose oxidation and pyruvate dehydrogenase and GAPDH activities [24–26]. Changes were reduced or abolished by A1AR overexpression (Table 2, Fig. 3A), and A1ARs can modify post-ischemic glucose metabolism which is considered a primary protective mechanism [27]. Enzymes of oxidative phosphorylation were also repressed including ATP synthase, mitochondrial adenine nucleotide translocator, and cytochrome c oxidase. Thus, there is evidence to suggest impaired oxidative phosphorylation post-ischemia. A number of lipid metabolism genes were also down-regulated (malonyl-CoA-decarboxylase, dodecenoyl-coenzyme A delta isomerase, and L-3-hydroxyacyl-Coenzyme A dehydrogenase), suggestive of a shift from fatty acid catabolism. Importantly, A1AR overexpression reversed the majority of changes in metabolic genes, and induced several including phosphoenolpyruvate carboxykinase 1 and glycogen phosphorylase, implicating improved energy metabolism in the protected phenotype.

Another notable observation was repression of structural/motility genes (Table 2, Fig. 2), consistent with cytoskeletal and myofibrillar abnormalities post-ischemia [28]. It is reasonable to hypothesize some may play a role in contractile dysfunction and subsequent remodeling. Although proteolysis may play a role [28], reduced expression could contribute to dysfunction. Greatest changes were observed for β-catenin (cell adhesion), peroxisomal farnesylated protein (peroxisomal biogenesis), cadherin 2 (intercalated disc protein), {alpha}-tubulin (cytoskeletal protein), and β-actin. The majority of changes were abolished with A1AR overexpression (Table 2), consistent with roles in the post-ischemic phenotype.

A small number of genes involved in cell signaling were altered by ischemia, and some were selectively modified by A1AR overexpression. Ischemia down-regulated protein phosphatase-2 subunits and PCTAIRE protein kinase 1, and induced DNA-activated protein kinase. Since contractility and myocardial apoptosis are sensitive to protein phosphorylation, modification of phosphatases and kinases could have important effects in post-ischemic tissue (which is characterized by enhanced apoptosis and contractile dysfunction).

Genes involved in cell defence and modulation of apoptosis were selectively modified by ischemia and A1AR overexpression. In addition to Hsp84, Hsp47, Hspe1 and Ddit3 (all induced in both groups), clusterin, Hsp40 and Egr1 were selectively induced in transgenic hearts. Induction of Hspe1 and Egr1 in transgenic hearts is consistent with cardioprotection since Hspe1 protects against ischemia [29], and Egr1 is up-regulated after injury, inducing growth related genes [30]. The glycoprotein clusterin (or apolipoprotein J) also inhibits apoptosis [31], and enhanced expression in transgenic hearts could protect viable cells from apoptosis. We demonstrate reduced necrosis/infarction in this model [7,8], and others demonstrate A1AR-mediated inhibition of apoptosis [32].

We observe two novel transcriptional changes with A1AR overexpression, confirmed by real-time PCR, inconsistent with a protected phenotype. First, Ywhag was down-regulated post-ischemia, and this was exaggerated by A1AR overexpression. The family of 14-3-3 proteins suppress apoptosis [33], and this is the first report the Ywhag gene may be modulated by ischemia. Secondly, Ddit3 was induced by ischemia in transgenic and wild-type hearts (Fig. 3B). Regulation of this gene, which induces growth arrest and cell death, has not been reported in heart, although another group recently reported ischemic induction of closely related Gadd45 [34]. Comparable induction of this gene in wild-type and transgenic hearts suggests activation of a cell death program which is insensitive to A1AR activation.

Induction of several transcription factors was observed in transgenic hearts (DEAD box polypeptides, Msx1, Spop, Sox9). It is tempting to speculate these factors may coordinate expression of groups of genes in transgenic hearts, contributing to ischemic tolerance. Curiously, Fhl2 was repressed in wild-type and, to a lesser extent, transgenic hearts (Fig. 3B). Due to its structure and expression in heart, Fhl2 is implicated as a regulator of cardiac differentiation and phenotype [35]. In terms of protein synthesis and modification, we observe repression of mitochondrial and ribosomal genes in wild-type hearts. This is reversed in transgenics, with additional ribosomal genes induced. Again, one might speculate induction of ribosomal proteins facilitates recovery in transgenic tissue.

4.3 Effects of ischemia-reperfusion on A1AR expression and function
Since we were interested in effects of A1ARs, expression of Adora1 was quantified. Prior analysis of transgenic hearts shows 100-fold overexpression of coupled A1AR [5], consistent with expression of Adora1 in normoxic transgenic hearts (Fig. 5). Paradoxically, ischemia-reperfusion induced Adora1 in transgenic hearts whereas it was moderately down-regulated in wild-type hearts. The A1AR transgene is controlled by the MEF-2 mutated myosin heavy chain (Myhca) promoter. Thus, the differences could reflect enhanced Myhca expression in post-ischemic transgenic hearts. Since Myhca levels are orders of magnitude greater than Adora1, small relative changes in Myhca expression could result in substantial changes in Adora1 expression. However, there is no detectable difference in Myhca expression (Fig. 5). Importantly, down-regulation of Adora1 in wild-types may adversely effect post-ischemic outcome, and receptor density may additionally be reduced by agonist triggered down-regulation. This, coupled with repressed transcription, could limit A1AR-mediated protection. This is supported by a post-ischemic reduction in sensitivity of A1AR-mediated bradycardia (see Section 3), an effect not observed in transgenic hearts. While this functional response is atrial in origin, data are collectively consistent with post-ischemic changes in A1AR expression which may limit A1AR protection.

4.4 Study limitations
Limitations regarding microarray analysis revolve around variance, methods of data handling and analysis, and tissue specificity. Variability in expression data is best eliminated by replication [14,36,37], and 2–3 replicates are essential in array studies [14]. Many recent cardiac studies incorporate limited replicates or report data from single arrays [3,22,23]. Investigators have also utilized 1.8- or 2-fold cut-offs for identifying expression changes [3,4,23]. These are not statistically valid, even in duplicated experiments [37], and it is critical to have sufficient replicate microarray assays to reach reliable conclusions [14]. Similar statistical standards applying to other biological experiments also apply to microarray experiments [36,37]. We examine expression for genes duplicated on each slide, and report expression data from multiple microarray experiments (i.e. a minimum of four replicates) and only genes whose expression changes significantly at the P<0.05 level are reported. Importantly, a subset of genes underwent secondary analysis via quantitative real-time PCR, validating differential expression (induction and repression) determined via microarray analysis of pooled RNA samples (Fig. 4).

A limitation relevant to any transcriptional analysis of intact myocardium [2–4,19,20,22] is that it is not possible to discern between alterations in specific cell populations, including cardiomyocytes, fibroblasts, resident inflammatory cells, and vascular tissue. Information is acquired regarding the genotype of the entire ventricular myocardium and associated vasculature. A drawback, therefore, is that where transcriptional responses are restricted to specific cell types, changes measured in intact heart will underestimate these responses. Although cell-culture might permit analysis of homogenous cell populations, the drawback with this alternate approach is that genotype and phenotype may be substantially altered from the in situ state, metabolic rate is low, and interactions between cell types important in inducing injurious or protective responses are absent. Analysis of the intact heart does provide insight into transcriptional control in the intact organ. Subsequent analysis may be necessary to identify the cellular sites of such responses.

4.5 Conclusions
The ischemia-tolerant phenotype with A1AR overexpression appears to result from acute activation of A1AR receptors during and following ischemia rather than adaptive changes in phenotype and genotype over the life-span of the animal. We identify a number of rapid changes in gene expression which may contribute to the post-ischemic phenotype, and are selectively modified by A1AR overexpression. The post-ischemic transcriptional profile is characterized by repression of metabolic and structural/motility genes, and changes in lesser numbers of signaling and cell defence genes. A1AR overexpression reverses ~80% of gene repression but less than 20% of gene induction observed in wild-type hearts. Notably, A1AR overexpression reversed repression of metabolic and structural/motility genes, enhanced induction of genes controlling gene expression, and altered pro- and anti-apoptotic genes, potentially contributing to an anti-apoptotic phenotype. While the ~2000 arrayed genes analysed may represent only 10% of the cardiovascular genome [10], these responses provide clues regarding molecular determinants of ischemic tolerance and the post-ischemic phenotype.

Time for primary review 20 days.


   Acknowledgements
 
This work was supported by grants from the National Institutes of Health RO1 grant (HL 59419) and the National Heart Foundation of Australia (G 98B 0080 and G 99B 0246).


   References
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 

  1. Matsuyama N., Leavens J.E., McKinnon D., Gaudette G.R., Aksehirli T.O., Krukenkamp I.B. Ischemic but not pharmacological preconditioning requires protein synthesis. Circulation (2000) 102(Suppl):S312–S318.
  2. Kitakaze M., Asakura M., Sakata Y., Asanuma H., Sanada S., Kuzuya T., Miyazaki J., Takashima S., Hori M. cDNA array hybridization reveals cardiac gene expression in acute ischemic murine hearts. Cardiovasc. Drugs Ther. (2001) 15:125–130.[CrossRef][ISI][Medline]
  3. Lyn D., Liu X.W., Bennett N.A., Emmett N.L. Gene expression profile in mouse myocardium after ischemia. Physiol. Genomics (2000) 2:93–100.[Abstract/Free Full Text]
  4. Stanton L.W., Garrard L.J., Damm D., Garrick B.L., Lam A., Kapoun A.M., Zheng Q., Protter A.A., Schreiner G.F., White R.T. Altered patterns of gene expression in response to myocardial infarction. Circ. Res. (2000) 86:939–945.[Abstract/Free Full Text]
  5. Gauthier N.S., Morrison R.R., Byford A.M., Jones R., Headrick J.P., Matherne G.P. Functional genomics of transgenic overexpression of A1 adenosine receptors in the heart. Drug Dev. Res. (1998) 45:402–409.[CrossRef][ISI]
  6. Headrick J.P., Gauthier N.S., Berr S.S., Morrison R.R., Matherne G.P. Transgenic A1 adenosine receptor overexpression improves myocardial energy state during ischemia-reperfusion. J. Mol. Cell Cardiol. (1998) 30:1059–1064.[CrossRef][ISI][Medline]
  7. Headrick J.P., Gauthier N.S., Morrison R.R., Matherne G.P. Cardioprotection by KATP channels in wild-type hearts and hearts overexpressing A1 adenosine receptors. Am. J. Physiol. Heart Circ. Physiol. (2000) 279:H1690–H1697.[Abstract/Free Full Text]
  8. Morrison R.R., Jones R., Byford A.M., Stell A.R., Peart J., Headrick J.P., Matherne G.P. Transgenic overexpression of cardiac A1 adenosine receptors mimics ischemic preconditioning. Am. J. Physiol. Heart Circ. Physiol. (2000) 279:H1071–H1078.[Abstract/Free Full Text]
  9. Peart J., Flood A., Linden J., Matherne G.P., Headrick J.P. Adenosine mediated cardioprotection in ischemic reperfused mouse heart. J. Cardiovasc. Pharm(acol). (2002) 39:117–129.[CrossRef][ISI][Medline]
  10. Dempsey A.A., Dzau V.J., Liew C.-C. Cardiovascular genomics: estimating the total number of genes expressed in the human cardiovascular system. J. Mol. Cell Cardiol. (2001) 33:1879–1886.[CrossRef][ISI][Medline]
  11. Peart J., Headrick J.P. Intrinsic activation of A1 adenosine receptors during ischemia and reperfusion improves ischemic tolerance. Am. J. Physiol. Heart Circ. Physiol. (2000) 279:H2166–H2175.[Abstract/Free Full Text]
  12. Headrick J.P., Gauthier N.S., Morrison R.R., Matherne G.P. Chronotropic and vasodilatory responses to adenosine and isoproterenol in mouse heart: effects of adenosine A1 receptor overexpression. Clin. Exp. Pharmacol. Physiol. (2000) 27:185–190.[CrossRef][ISI][Medline]
  13. Grimmond S., Van Hateren N., Siggers P., Arkell R., Larder R., Soares M.B., de Fatima Bonaldo M., Smith L., Tymowska-Lalanne Z., Wells C., Greenfield A. Sexual dimorphic expression of protease nexin-1 and vanin-1 in the developing mouse gonad prior to overt differentiation suggests a role in mammalian sexual development. Hum. Mol. Genet. (2000) 9:1553–1560.[Abstract/Free Full Text]
  14. Lee M.L., Kuo F.C., Whitmore G.A., Sklar J. Importance of replication in microarray gene expression studies: Statistical methods and evidence from repetitive cDNA hybridizations. Proc. Natl. Acad. Sci. USA (2000) 97:9834–9839.[Abstract/Free Full Text]
  15. Livak K.J., Schmittgen T.D. Analysis of relative gene expression data using real-time quantitative PCR and the 2–{Delta}{Delta}CT method. Methods (2001) 25:402–408.[CrossRef][ISI][Medline]
  16. Henriksen P.A., Kotelevtsev Y. Application of gene expression profiling to cardiovascular disease. Cardiovasc. Res. (2002) 54:16–24.[Abstract/Free Full Text]
  17. Boehm E.A., Jones B.E., Radda G.K., Veech R.L., Clarke K. Increased uncoupling proteins and decreased efficiency in palmitate-perfused hyperthyroid rat heart. Am. J. Physiol. Heart Circ. Physiol. (2001) 280:H977–H983.[Abstract/Free Full Text]
  18. Headrick J. Willis R.J. Endogenous adenosine improves work rate to oxygen consumption ratio in catecholamine stimulated isovolumic rat heart. Pflügers Arch. 1989;413:354-358.
  19. Heads R.J., Latchman D.S., Yellon D.M. Differential stress protein mRNA expression during early ischaemic preconditioning in the rabbit heart and its relationship to adenosine receptor function. J. Mol. Cell Cardiol. (1995) 27:2133–2148.[CrossRef][ISI][Medline]
  20. Sommerschild H.T., Lunde P.K., Deindl E., Jynge P., Ilebekk A., Kirkeboen K.A. Elevated levels of endogenous adenosine alter metabolism and enhance reduction in contractile function during low-flow ischemia: Associated changes in expression of Ca+2-ATPase and phospholamban. J. Mol. Cell Cardiol. (1999) 31:1897–1911.[CrossRef][ISI][Medline]
  21. Peng T.Q., Sadusky T., Li Y.W., Coulton G.R., Zhang H.Y., Archard L.C. Altered expression of Bag-1 in Coxsackievirus B3 infected mouse heart. Cardiovasc. Res. (2001) 50:46–55.[Abstract/Free Full Text]
  22. Johnatty S.E., Dyck J.R.B., Michael L.H., Olson E.N., Abdellatif M. Identification of genes regulated during mechanical load-induced cardiac hypertrophy. J. Mol. Cell Cardiol. (2000) 32:805–815.[CrossRef][ISI][Medline]
  23. Barrans J.D., Stamatiou D., Liew C.-C. Construction of a human cardiovascular cDNA microarray: Portrait of the failing heart. Biochem. Biophys. Res. Commun. (2001) 280:964–969.[CrossRef][ISI][Medline]
  24. Lopaschuk G.D., Spafford M.A., Davies N.J., Wall S.R. Glucose and palmitate oxidation in isolated working rat hearts reperfused after a period of transient global ischemia. Circ. Res. (1990) 66:546–553.[Abstract/Free Full Text]
  25. Knight R.J., Kofoed K.F., Schelbert H.R., Buxton D.B. Inhibition of glyceraldehyde-3-phosphate dehydrogenase in post-ischaemic myocardium. Cardiovasc. Res. (1996) 32:1016–1023.[Abstract/Free Full Text]
  26. Kofoed K.F., Schoeder H., Knight R.J., Buxton D.B. Glucose metabolism in reperfused myocardium measured by [2-18 F] 2-fluorodeoxyglucose and PET. Cardiovasc. Res. (2000) 45:321–329.[Abstract/Free Full Text]
  27. Finegan B.A., Lopaschuk G.D., Gandhi M., Clanachan A.S. Inhibition of glycolysis and enhanced mechanical function of working rat hearts as a result of adenosine A1 receptor stimulation during reperfusion following ischaemia. Br. J. Pharmacol. (1996) 118:355–363.[ISI][Medline]
  28. Gao W.D., Liu Y., Mellgren R., Marban E. Intrinsic myofilament alterations underlying the decreased contractility of stunned myocardium: consequence of Ca+2-dependent proteolysis? Circ. Res. (1996) 78:455–465.[Abstract/Free Full Text]
  29. Lin K.M., Lin B., Lian I.Y., Mestril R., Scheffler I.E., Dillmann W.H. Combined and individual mitochondrial HSP60 and HSP10 expression in cardiac myocytes protects mitochondrial function and prevents apoptotic cell deaths induced by simulated ischemia-reoxygenation. Circulation (2001) 103:1787–1792.[Abstract/Free Full Text]
  30. Brand T., Sharma H.S., Fleischmann K.E., Duncker D.J., McFalls E.O., Verdouw P.D., Schaper W. Proto-oncogene expression in porcine myocardium subjected to ischemia and reperfusion. Circ. Res. (1992) 71:1351–1360.[Abstract/Free Full Text]
  31. Zwain I.H., Amato P. cAMP-induced apoptosis in granulosa cells is associated with up-regulation of P53 and bax and down-regulation of clusterin. Endocr. Res. (2001) 27:233–249.[CrossRef][ISI][Medline]
  32. Vitolo O.V., Ciotti M.T., Galli C., Borsello T., Calissano P. Adenosine and ADP prevent apoptosis in cultured rat cerebellar granule cells. Brain Res. (1998) 809:297–301.[CrossRef][ISI][Medline]
  33. Zha J., Harada H., Osipov K., Jockel J., Waksman G., Korsmeyer S.J. BH3 domain of BAD is required for heterodimerization with BCL-XL and pro-apoptotic activity. J. Biol. Chem. (1997) 272:24101–24104.[Abstract/Free Full Text]
  34. Xue-Han N., Chen S.-H., Xu C.-S., Li L., Yao L.Y., Qian K., Krueger J.J., Hyyti O.M., Portman M.A. Hypothermic protection of the ischemic heart via alterations in apoptotic pathways as assessed by gene array analysis. J. Appl. Physiol. (2002) 92:2200–2207.[Abstract/Free Full Text]
  35. Ng E.K., Chan K.K., Wong C.H., Tsui S.K., Ngai S.M., Lee S.M., Kotaka M., Lee C.Y., Waye M.M., Fung K.P. Interaction of the heart-specific LIM domain protein, FHL2, with DNA-binding nuclear protein, hNP220. J. Cell Biochem. (2002) 84:556–566.[CrossRef][ISI][Medline]
  36. Baldi P., Long A.D. A Bayesian framework for the analysis of microarray expression data: regularized t-test and statistical inferences of gene changes. Bioinformatics (2001) 17:509–519.[Abstract/Free Full Text]
  37. Claverie J.M. Computational methods for the identification of differential and coordinated gene expression. Hum. Mol. Genet. (1999) 8:1821–1832.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Ashton, K. J
Right arrow Articles by Headrick, J. P
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ashton, K. J
Right arrow Articles by Headrick, J. P
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?