In this study, we show that the 10-week mild exercise aerobic training (~60% VO2max) induces cardioprotection to I/R and alters the gene expression profile in rat left ventricles at 48 hours after the last training session. The differentially expressed genes were: Cav3; Eno3; cytochrome P450, family 27, subfamily a, polypeptide 1 (Cyp27a1); Egln1; cystatin C (Cst3) and tumor necrosis factor, alpha-induced protein 1 (endothelial) (Tnfaip1) genes, along with the two RGD1306959 and RGD1307773 predicted genes and two unidentified clones. Present data indicate that the applied mild chronic training protocol is effective, as demonstrated by the higher running capacity and the soleus to body weight ratio in TRA rats with respect to CTR rats, without inducing cardiac hypertrophy, as demonstrated by the absence of differences in heart-to-body weight ratio and in the thickness of the left ventricular wall between TRA and CTR rats.
The described mild exercise training model determined development of cardioprotection, as evaluated by I/R experiments, similarly to what reported in the case of intense training procedures [29
], as demonstrated by the fact that the infarct size was larger in CTR than in TRA. These data are coherent with the previous observation obtained by our group with a similar training protocol [14
The selected experimental design satisfies several purposes. First, the applied mild training protocol resembles that practiced by adult humans for cardiovascular fitness. Ten-week training duration enabled the onset of cardiac adaptations and allowed for the estimation of the possible occurrence of cardiac hypertrophy [9
]. Second, rats were sacrificed 48 hours after the last training session in order to rule out acute effects of exercise and to evaluate gene expression changes due to chronic adaptation. It might be hypothesized that the cardioprotection resulting from the adaptation to mild chronic exercise training is the result of a summation of several second windows of protection similar to what observed during ischemic preconditioning. In fact, transient ischemic events may occur during exercise at high heart rate, when the time available for diastolic coronary flow is reduced and myocardial contractility as well as mechanical compression is likely to impede coronary flow [30
]. Moreover, the experimental design allowed us to compare and extend on a larger number of genes data reported in a previous paper [14
] employing a similar chronic mild training protocol.
Gene expression profile changes induced by exercise training have been investigated in previous studies that employed training procedures of a more intense nature than those used in this study and that, in contrast to the protocol of this study, induced cardiac hypertrophy [9
]. Differentially expressed genes reported in those studies differ highly in number and type, not only when compared to those reported in the present work, but also when a comparison is drawn among them. Conflicting results may stem from the intensity, duration of protocols, and distance of the sacrifice from the last exercise bout. However, it is interesting to note that in the studies cited above the number of differentially expressed genes decreased according to the time that elapsed from the last exercise bout to the sacrifice, suggesting a restoring of gene expression to levels comparable to those of sedentary control animals: 24 hours (305 genes) [9
], 48 hours (75 genes) [17
], 72 hours (27 genes) [18
Recently, Freimann and coworkers [11
] evaluated gene expression profiles of heart from rats which received a 7-week swimming exercise training and underwent surgically-induced myocardial infarction. The gene expression profiles were evaluated before, 4 hours, 2 days and 4 weeks after surgery. Differentially expressed genes did not overlap, at any time point, with those reported here. However, the experimental design of the Freimann's group differed from ours as we evaluated the gene expression profiles that characterized animals after the mild chronic exercise and, in a different set of rats trained in parallel using an identical training protocol, we determined the cardioprotection by measuring the extent of the injury caused by ischemia/reperfusion experiments. Concerning the previous paper of our group [14
], none of the 14 differentially expressed genes identified (96 candidate genes involved in stress and toxicity) was included within the genes found differentially expressed in the present study. On the other hand, two different technologies – a macroarray focused on 96 genes in the previous paper and Affymetrix GeneChips representing 31.000 transcripts in the present paper – were used.
Among the differentially expressed transcripts, we pointed our attention on 3 genes: similar to C11orf17 protein (RGD1306959), Eno3 and Cav3.
The predicted sequence of similar to C11orf17 protein (RGD1306959) has a high homology (mRNA nucleotide: 90%–92%; protein: 77% on positivity) with the human C11orf17 [alias: protein kinase-A interacting protein 1 (AKIP1) or breast cancer associated gene 3 (BCA3)]. C11orf17 has been recently shown to enhance NF-κB-mediated gene expression [32
], a redox-sensitive transcription factor, known to be involved in the control of a large number of cellular processes, such as immune and inflammatory responses, developmental processes, cellular growth, and apoptosis. The gene expression of similar to C11orf17 protein (RGD1306959) gene was found increased in LV of hypertensive rats and positively correlated with LV mass index, thus pointing to its potential implication in LV hypertrophy development [33
]. Interestingly, in these hypertensive rats, Eno3 showed a negative correlation with LV mass index [33
], whereas, in our model of mild physical exercise without cardiac hypertrophy, Eno3 gene showed an increased expression in trained rats. Our findings, together with data obtained in hypertensive rat models, strongly suggest a relevant role of similar to C11orf17 protein (RGD1306959) gene in the adaptive response of the cardiac muscle mass to the workload; this mechanism could be balanced by the up-regulation of the Eno3 gene. Eno3 gene encodes one of the three enolase isoenzymes found in mammals. This isoenzyme is found in skeletal and cardiac muscle cells. The functional role of this gene in heart muscle, as well as skeletal muscles, has not been elucidated yet. Several reports described a role for the hypoxia-inducible factor-1 (HIF-1) in the transcriptional activation of enolase-1 [34
], and enolase-2 genes [35
]. In the present study, Hif-1alpha showed a trend toward increased expression in trained rats, thus suggesting that Hif-1alpha could be regulated also by non-hypoxic stimuli. In a recent proteomics study, enolase 1 alpha and proteins similar to alpha enolase have been found to increase in exercised rats [36
]. Moreover, beta enolase was observed to increase in athletes after running a race [37
]. These data further indicate an important role of beta enolase in the preconditioning phenomena that induce cardioprotection.
Using the present protocol of mild exercise training an increased expression of caveolin 3 at transcriptomic and proteomic levels was demonstrated. In cardiomyocytes, caveolin 3 is a major constituent of caveolae, plasma membrane invaginations implicated in vesicular trafficking, signal transduction, and Ca2+
]. The influence of caveolin 3 on hypertrophy has been previously suggested by a number of evidences, including the development of cardiac hypertrophy in caveolin 3-null mice [39
], the inhibition of agonist-induced cardiomyocyte hypertrophy by overexpression of caveolin 3 [40
], and the observation that exercise training, preventing pathological hypertrophy in spontaneously hypertensive rats, is associated to the overexpression of caveolin 3 [41
]. Moreover, it has been found that caveolin 3 induces protection of cardiac myocytes from ischemic damage [42
]. All these results lend support to the hypothesis that exercise training-induced cardioprotection to I/R is not only accompanied by the overexpression of caveolin 3, but also strictly related to it by a cause-effect relationship.
In order to identify functional gene sets correlated with mild chronic exercise training, we applied GSA analysis. GSA analysis identified only 2 gene sets correlated with exercise training: the ARENRF2PATHWAY and the GABAPATHWAY. There are some evidences that physical exercise leads to the enhanced transient formation of reactive oxygen species (ROS) [14
]. Oxygen radicals are important intracellular second messengers that mediate cardioprotection [43
]. Increased levels of ROS generated in working muscles favor a change in the redox-balance towards a more pro-oxidant state counteracted by a complex network of antioxidant systems [44
], as suggested by the association of exercise with the "Oxidative stress induced gene expression via Nrf2" pathway. The functional consequences of exercise-induced oxidative stress are only partly understood; nevertheless, the present data demonstrated that our protocol is able to induce cardioprotection, as evaluated by functional I/R experiments. In addition, the association of the GABA pathway with training is consistent with the role of GABA receptor activity in the cardioprotection induced by ischemic preconditioning [43
]. The association of these two pathways is consistent with our data on Cav3 upregulation: in fact, several data showed that caveolin 3 protein plays an important role in ROS response through MAP kinases mediated intracellular signal transduction underlying cardioprotection [45
Even if we confirmed by other molecular biology approaches the coherence of our microarray data, this study have the limitations of all transcriptomics approaches concerning the post-transcriptional and post-transductional behavior of the differentially expressed genes and the possibility of type I (false positive genes) and type II (false negative genes) statistical error due to the high number of parameters investigated in each experiments.