The data presented here have shown that stress induces changes in gene expression and that β2-AR modulates those changes in the heart. The vast majority of genes with expression disregulated by stress were different when β2-AR was upregulated and when it was blocked by the ICI treatment; only three genes were disregulated in both cases. This indicates that, given the same stressful situation, the profile of gene expression in the heart is substantially different when β2-AR is active or when it is blocked. The three genes with expression independent of β2-AR were CREM, related to the cAMP signaling pathway, HSP90AA1, which codifies HSP90, the glucocorticoid receptor chaperone, and ERRFI1, which codifies the feedback inhibitor of the epidermal growth factor receptor, which attenuates the PI3K-Akt signaling pathway.
β-AR activation by catecholamines is the mainly signal for cAMP generation in cardiac myocytes. The cAMP-dependent transcription factor, CREM, expressed in the myocardium, is involved in the regulation of the expression of various components of the cAMP signaling pathway [25]. It has been described as essential for normal cardiac function [26, 27] and required for the β1-AR response to overstimulation [27]. Transgenic mice expressing a CREM repressor isoform in cardiomyocytes developed myocardial dysfunction [27] and atrial fibrillation [28, 29]. Mice with complete ablation of the CREM gene presented contractile impairment, with a reduction in systolic and diastolic functions due to sarcoplasmic reticulum Ca2+-ATPase (SERCA) downregulation [26]. The data presented here show upregulation of CREM in the heart of stressed rats in the context of overstimulation of β1-AR (with or without the participation of β2-AR) due to stress-induced high levels of catecholamines.
Moreover, many other genes involved in that same signaling pathway had their expression disregulated under stress, whether or not treated with ICI118,551. RGS1 and RGS16 were upregulated in the untreated groups comparison. The RGS family acts as a negative regulator of G protein signaling. In the heart, RGS2 has been involved in the control of Gs signaling and cAMP production after β-AR stimulation [30]. RGS1 has been related to the control of inflammation and the immune response [31, 32]. The stress-induced alteration in the expression of these genes depends on an active β2-AR, since it is unaltered in the group treated with ICI118,551. On the other hand, CREB1 and GPER1 belong to the group of molecules predicted to be upstream regulators activated by stress, independent of β2-AR activation. CREB1, a transcription factor responsive to cAMP, leads to effects similar to CREM [26], while GPER1 belongs to the G-protein coupled receptor family that binds estrogen. This receptor activates both the adenylate cyclase-cAMP/PKA signaling pathway [33, 34] and the PI3K-Akt-mTOR signaling pathway [35].
The APLNR gene, also called the APJ receptor gene, encodes the G-protein coupled receptor for apelin, a protein expressed in the cardiovascular system which promotes a positive inotropic effect on the heart as well as angiogenesis and blood vessel relaxation [36, 37]. The inotropic action of apelin is the result of an increase in the availability of intracellular calcium [38]. Apelin not only increases inotropy, but also decreases left ventricular pre- and afterload due to its pronounced vasodilation effect [39]. It induces cAMP synthesis, as well as activating the PI3K/Akt signaling pathway [37]. The concentration of apelin is reduced in the failing heart [40] and the contractile function is impaired in cardiomyocytes with knockout for APLNR [41]. Apelin expression was downregulated under stress when no ICI118,551 treatment was given, although the expression of its receptor (APLNR) was reduced when the β2-AR was blocked.
Therefore, in the heart of untreated stressed rats, the downregulation of β1-AR and apelin seems to be counterbalanced by the upregulation of β2-AR and the action of CREM and RGS2. Therefore, although the proportion of β-AR subtypes is altered, there is no difference in cAMP formation by left atrial membranes of control and foot shock stressed rats stimulated by β1/β2-AR non-selective agonists [13]. However, when the membranes are stimulated in the presence of the β2-AR antagonist (ICI118,551), the amount of cAMP synthetized by the atrial membranes of stressed rats is lower than that of unstressed ones [13]. This is probably because the increase in the expression of CREM is not sufficient to sustain the synthesis of cAMP, since β1-AR is downregulated, as has been reported elsewhere [13].
The present data thus confirm that when stress is applied under conditions of β2-AR blockade, the canonical pathway of cAMP signaling is negatively regulated (see Table 2), due largely to APLNR downregulation. Many positive and negative influence thus work together to influence the cAMP level and the stress response.
The increasing cAMP level due to β1-AR and β2-AR activity culminates in an increased rate of beating and force developed by cardiomyocytes. The calcium transient plays a central role in this process [42]. Indeed, calcium was predicted to be an upstream regulator in the presence of higher β2-AR expression. Although extremely important for the proper functioning of cardiac cells, excessive calcium signaling leads to disorders such as arrhythmia, hypertrophy, and cell death [42]. Therefore, the stress induced modulation of positive and negative influences on the expression of molecules related to cAMP and the calcium signaling pathways in the heart adds complexity, as well as more possibility for control of the cardiac function and structure, with β2-AR apparently playing an essential role in the process.
Glucocorticoids (GC) released by the adrenal cortex are, in conjunction with the catecholamines, known as the stress hormones. The glucocorticoid receptor (GR) is located in the cytoplasm of target cells, bound to a chaperone complex and immunophilins. Upon the binding of the glucocorticoid, the GR undergoes a conformational change that causes the release of the chaperones and dimerization of the complexes GC-GR. Then, the GR-GC dimers translocate to the nucleus, where they start its genomic signaling [43, 44]. Recently, Severinova and colleagues (2019) have shown that a 1-hour treatment with a synthetic corticosteroid was enough to change the expression of genes that were GR targets [45]. Increased corticosterone plasma levels led to an increase in GR-mediated genomic signaling and the modulation of gene expression.
The heat shock proteins (HSP) 90 and 70, as well as Src, released from the GR complex, also influence cell signaling [44]. These proteins play several fundamental roles in cellular processes by regulating protein folding, proteostasis and intracellular signal transduction. HSP90 and HSP70 are inducible isoforms that present incremented expression after myocardium injury, oxidative stress, and hypoxia. In the GR complex, HSP 90 and HSP70 provide structural stability and function [46]. HSP90aa1, the mRNA for HSP90, was upregulated under stress, both with and without β2-AR participation, while HSPa1a, the mRNA for HSP70, was upregulated only under stress and β2-AR blockade. Several papers have linked the upregulation of HSP90 and HSP70 to cardioprotection against injury [46–48]. The in vitro overexpression of HSP90aa1 in cardiomyocytes attenuates apoptosis by increasing Bcl-2 expression [48]. HSP90 is also associated with the survival pathway of PI3K-Akt. The association of HSP90 with Akt leads to phosphorylation and the activation of endothelial nitric oxide synthase [46, 47]. Under stress, the reduction of fibrosis and apoptosis in the presence of β2-AR upregulation, and the increase in cellular homeostasis, viability, and survival mechanisms in the rats treated with ICI118,551 (β2-AR blockade), indicated by disease and function categorization, suggest that the modulation of the expression of genes is part of the adaptive mechanisms related to stress.
ERRFI1, also known as receptor-associated late transducer (RALT) or Gene 33, is a feedback inhibitor of the epidermal growth factor receptor (EGFR) [49]. The inhibition of EGFR by this gene in cardiac myocytes reduces cardiac hypertrophy [50]. It has been reported that the induction of the expression of the gene is related to the activation of GR signaling and coincides with a reduction in Akt phosphorylation [49]. The expression of EREFI1 has been found to increase in ischemic injuries and myocardial infarction, where it is associated with apoptosis of cardiomyocytes through a reduction of PI3K-Akt and extracellular signal-regulated kinase (ERK) signaling pathways [51]. In a recent paper, reduction of ERRFI1 activity in cardiomyocytes has been linked to miR-126, and found to protect them against apoptosis and oxidative stress [52]. The results presented here have demonstrated a greater concentration of circulating corticosterone, as well as upregulation of ERRFI1 under stress whether β2-AR is upregulated or blocked. The PI3K-Akt signaling pathway is an important pathway for survival which has been implicated in much of the microarray data presented here. We have also recently reported that Akt phosphorylation and PI3K expression are downregulated by foot shock stress [21], a result which has been compatible with ERRFI1 upregulation and APLN downregulation.
The present data have thus shown that two signaling pathways are the most affected by stress: the Gs-AC-cAMP and the PI3K-Akt signaling pathways. Moreover, the modulation of several of their components depends on the presence of the β2-AR. The changes induced by stress in the β-AR-Gs-AC, glucocorticoid receptor/glucocorticoid, and PI3K-Akt signaling pathways are summarized in Fig. 4. On the figure left side are some of the genes with expression altered by stress (upregulation of ERRFI1, CREM, RGS1, HSP90aa1; and downregulation of APLN). The right side of Fig. 4 shows the genes that are altered when stress is applied in conjunction with the β2-AR blockade. ERRFI1, CREM, HSP90aa1, and HSPA1A/A1B are upregulated whereas APLNR is downregulated.
The changes post stress in cardiac gene expression also include the regulation of the progression of the cell cycle, which can be impaired by downregulation of CDK1, and the presence of epigenetic factors, such as the upregulation of the class I histone deacetylase, HDAC8, and, in ICI-treated rats, stress induced upregulation of HSP40, and IRS2. The upregulation of IRS2 suggests that insulin may be an upstream regulator of the stress response, independent of β2-AR, since it appears in the comparisons of both non-ICI treated and ICI treated rats.
An unexpected finding was the recruitment of the immune system to the cardiac tissue of stressed rats, with the profile clearly different for the two groups, as a function of β2-AR. Immunological activation was identified in microarray data in expressed genes, canonical pathways and upstream regulators. The correlation of immunological response with cardiac injury and repair is widely recognized. A cardiac injury initiates an inflammatory process by first stimulating resident immune cells, mainly macrophages, dendritic cells and mast cells, to release pro-inflammatory mediators such as IL-12, IFN-γ and TNF-α. This will amplify the inflammatory response through the recruitment of lymphocytic cells [53, 54]. In the failing heart, mechanical stress promotes immune activation due to the enhancement of immune cells and the release of inflammatory cytokines, which clearly contribute to myocardial dysfunction [54].
Among the differently expressed genes in the comparison of non-ICI treated rats, IL22RA2 (interleukin 22 receptor subunit alpha 2), IL2RA (interleukin 2 receptor antagonist) and IFRD1 (interferon related developmental regulator 1) were upregulated and are described as regulators of the inflammatory response [55, 56]. IL-6, a pro-inflammatory cytokine produced by myeloid cells and cardiomyocytes in an autocrine mechanism [53], was also upregulated in this group. The expression of IL6 and IL6R mRNA has been detected in the myocardium of patients with severe heart failure [57]. The expression of IL6 mRNA in the myocardium has been suggested to be related to cardiac disease and the impairment of cardiac performance [57, 58]. The IL-6R signaling pathway is also positively related to human coronary heart disease [59–61]. It has been predicted that the pro-inflammatory cytokines IL-1β and TNF act as upstream regulators of the cardiac stress response in the presence of β2-AR. Both are related to acute inflammation in cardiac disease, such as,heart failure and myocardial infarction [53, 54]. IL-1β, TNF-α, and IL-12 are cytokines released by T helper 1 lymphocytes (Th1), which control the cellular immune response. Interestingly, immune cells, including Th1 lymphocytes, express β2-AR and are, therefore, susceptible to catecholamines action [62].
In immune cells, cAMP/PKA signaling inhibits the transcription of nuclear factor kappa B (NF-κB), another important factor in the immune response; through CREB, it activates the transcription of IL-10. This mechanism promotes the differentiation of the Th2 response, which is markedly anti-inflammatory, and also inhibits the development of the Th1 response, which is markedly pro-inflammatory [62, 63]. In vitro, norepinephrine suppresses the release of TNF-α by bone marrow cells through β2-AR activation, an effect that is canceled by ICI188,551 [63]. β2-AR agonists cause a similar reduction in TNF-α production in mononuclear cells of the peripheral blood from diabetic rats [64]. Furthermore, the chronic use of β2-AR–selective and nonselective blockers in mice impairs the recruitment of leukocytes to the injured heart and reduces survival [65].
The data reported here thus show that in the presence of greater expression of β2-AR, as previously reported by Moura et al. (2017), a pro-inflammatory signaling is triggered in the heart of stressed rats. If the upregulation of β2-AR occurs only in the cardiomyocytes or in the resident immune cells as well is, however, unknown at the present time. Indeed, data for the comparison of stressed ICI-treated rats with non-stressed ICI-treated rats did not involve pro-inflammatory genes, which suggests that the treatment with ICI188,551 reduced the development of the Th1 response by blocking β2-AR in the immune cells as well. Several canonical pathways implicated in that analysis are related to the immune response, such as IL-10, IL-17, CD40, CD27 and IL-6 signaling, the acute phase response, and GR signaling. One interesting characteristic of these canonical pathways is the expression of the same gene, NFKBIA, at times in conjunction with MAP2K3 and/or JUN. NFKBIA, also known as IκBα, encodes a potent inhibitor of the NF-κB [66], which is a transcription factor that mediates activation of the inflammatory response with important consequences in heart disease. The overexpression of IκBα in cardiomyocytes inhibits NF-κB activity, reduces hypertrophy and improves cardiac performance and survival signaling via Akt [67]. The presence of IκBα in the left ventricle tissue suppresses the expression of NF-κB-inducible inflammatory molecules and attenuates myocardial fibrosis [68]. Therefore, the expression of IκBα mRNA in ICI-treated stressed rats could indicate the presence of a regulatory or anti-inflammatory mechanism in the ventricle of rats submitted to stress when faced with β2-AR blockage.
The stress induced alterations in the expression of such a large number of genes seems to be an adaptive mechanism attempting to sustain the function of the heart and cellular homeostasis in order to protect the cardiomyocytes from apoptosis. β2-AR clearly plays a role in this process, since the alterations in gene expression that occurred in the presence of this adrenoceptor subtype are completely different from those seen when it is blocked. The microarray data presented here provide an overview of stress-mediated regulation in cardiac tissue. However, further exploration of the more relevant signaling pathways is necessary to clarify how and if those alterations in gene expression would affect the heart function in vivo.