Discussion

The primary finding of this study is that expression of RGS6 protects the heart from ischemic injury. In light of the expression of RGS6 in the mouse ventricle (Yang et al., 2013), the ability of RGS6 to suppress Gα_i signaling (Hooks et al., 2003), and previous work demonstrating that disruption of interactions between Gα_i2 and endogenous RGS proteins enhances Gα_i2 signaling and protects the heart from ischemic injury (Waterson et al., 2011; Parra et al., 2014), we anticipated that genetic deletion of RGS6 would protect the heart from ischemia. Thus, the finding that RGS6 deletion potentiates ischemic injury was unexpected. mRNA transcripts encoding at least 16 different RGS proteins have been identified in cardiac myocytes and fibroblasts (Zhang and Mende, 2011), and several have been found in the heart at the protein level (Jean-Baptiste et al., 2005; Chakir et al., 2011; Stewart et al., 2012; Miao et al., 2016). However, to our knowledge, this is the first study to provide evidence that a specific RGS protein (RGS6) plays a role in the myocardial response to an ischemic insult.

The observation that GRK2 expression (Fig. 5A) and GRK2 phosphorylation (Fig. 5B) are increased in RGS6^−/− hearts is consistent with increased infarct sizes (Fig. 1) in RGS6^−/− hearts and with previous work demonstrating that increased GRK2 signaling worsens ischemic injury (Brinks et al., 2010). Increased GRK2 expression and phosphorylation are also consistent with our findings that β2-AR phosphorylation was increased (Fig. 3A) and that the inotropic effect of β2-AR stimulation was diminished in RGS6^−/− hearts (Fig. 3B), suggesting that β2-ARs were modestly desensitized. GRK2-dependent phosphorylation of β₂-ARs causes these receptors to shift from coupling to Gαs to Gαi (Zhu et al., 2012). Gα_i2 and Gα_i3 are both expressed in the heart and are activated by cardiac β₂-ARs (Xiao et al., 1999). Importantly, these G_αi isoforms have divergent effects on the ischemic heart. Gα_i2 signaling is cardioprotective (Waterson et al., 2011; Kohler et al., 2014; Parra et al., 2014), whereas Gα_i3 signaling worsens ischemic injury (Kohler et al., 2014). In vitro assays indicate that RGS6 catalyzes the hydrolysis of Gα_i3-bound GTP at nearly twice (1.9-fold) the rate that it catalyzes the hydrolysis of Gα_i2-bound GTP (Hooks et al., 2003). If this selectivity toward Gα_i3 also occurs in the intact heart, then RGS6 deletion would be expected to enhance signaling through Gα_i3 to a greater extent than it enhances cardioprotective signaling through Gα_i2. Thus, it is possible that RGS6 protects the heart from ischemia by suppressing β2-AR–induced Gα_i3 signaling. Alternatively, the increased myocardial sensitivity to ischemic injury in RGS6^−/− hearts might result from increased GRK2 activity in the mitochondria. Previous work has demonstrated that under conditions of ischemia or oxidative stress, ERK-dependent phosphorylation of GRK2 causes GRK2 to be translocated to the mitochondria where it promotes opening of the mitochondrial permeability transition pore, leading to apoptosis and myocyte death (Chen et al., 2013). Our findings that RGS6^−/− hearts exhibit overexpression of GRK2, increased basal phosphorylation of GRK2, and increased sensitivity to ischemic injury suggest that RGS6 expression protects the heart from ischemic injury by suppressing signaling through this pathway. The observations that RGS6 deletion increases the presence of phospho-GRK2 and total GRK2 expression in mitochondrial lysates and increases caspase-3 cleavage are also consistent with this mechanism.

It is well established that myocardial injury can occur as a result of both ischemia and reperfusion (Fröhlich et al., 2013; Jivraj et al., 2015). Both male and female RGS6^−/− hearts demonstrated significantly elevated EDP relative to male and female RGS6^+/+ hearts during the 30-minute ischemic insult (Fig. 2). This state of contracture progressed more rapidly and to a greater extent in RGS6^−/− hearts compared with hearts from RGS6^+/+ mice. Importantly, this difference between RGS6^−/− hearts and RGS6^+/+ hearts was evident prior to the onset of reperfusion. The fact that RGS6^−/− hearts developed significantly greater contracture than RGS6^+/+ hearts during ischemia (prior to reperfusion) suggests that the increased myocardial injury observed in RGS6^−/− hearts was likely the result of increased sensitivity to ischemia-induced injury rather than increased sensitivity to reperfusion-induced injury.

RGS6^−/− hearts are hypersensitive to cholinergic stimulation (Posokhova et al., 2010, 2013; Yang et al., 2010), yet they have normal basal heart rates (Yang et al., 2010). This suggests that RGS6^−/− mice compensate for increased sensitivity to parasympathetic stimulation by increasing the activity of sympathetic neurons. The observations that β2-AR phosphorylation was increased and that β2-AR–induced responses were functionally desensitized in RGS6^−/− hearts are consistent with chronic β2-AR signaling. The finding that increased basal GRK2 phosphorylation and myocardial hypersensitivity to ischemic injury were reversed by chronic propranolol treatment provides evidence that RGS6^−/− hearts are hypersensitive to ischemic injury as a result of excessive β-AR signaling.

Our study used the Langendorff isolated heart model to investigate the effect of RGS6 deletion on the ischemic heart. However, our assessment of signaling pathways involved in RGS6-mediated cardioprotection was performed primarily in hearts that were isolated and immediately flash frozen in liquid nitrogen. Thus, our observations of increased β₂-AR phosphorylation (Fig. 3A), increased GRK2 expression (Fig. 5A), increased GRK2 phosphorylation (Fig. 5B), and increased caspase-3 cleavage (Fig. 6C) in RGS6^−/− hearts reflect their status under in vivo conditions. On the basis of these findings, we speculate that RGS6^−/− hearts would have also exhibited increased sensitivity to ischemic injury if myocardial ischemia had been induced in vivo.

Cardioprotective stimuli that decrease infarct size do not always enhance postischemic recovery of contractile function (Cohen et al., 1999; Sumeray et al., 2000; Schulz et al., 2001; Perrin et al., 2003). Some investigators have attributed this observation to “myocardial stunning,” which is characterized by contractile dysfunction without infarction. Myocardial stunning may result from the generation of reactive oxygen species, which damage contractile proteins (Bolli and Marban, 1999), or from calcium overloading in cardiac myocytes (Kumar et al., 2009), which induces degradation of contractile proteins by calcium-dependent proteases (Bolli and Marban, 1999). Importantly, these processes do not destroy cardiac myocytes, and the contractile function can recover with prolonged reperfusion time as new protein synthesis enables the damage to be repaired. In addition, postischemic coronary flow rates are typically much lower than preischemic coronary flow rates. This decrease in coronary flow (and subsequent decrease in the delivery of oxygen and nutrients) can suppress cardiac contractile function and mask the true contractile capacity of the myocytes. Importantly, our data indicate that RGS6 expression protects the heart from infarction but does not prevent postischemic contractile dysfunction within the 1-hour reperfusion period that was measured. One limitation of the Langendorff isolated heart model is that cardiac contractile function cannot be monitored over the long term. Thus, the extent to which RGS6 deletion would affect contractile function days or weeks after ischemia remains unknown.

RGS6^−/− mice are resistant to doxorubicin-induced production of reactive oxygen species and apoptosis, suggesting that RGS6 might mediate the cardiotoxic effect of doxorubicin that is commonly observed in patients with cancer who are treated with this drug (Yang et al., 2013). It should be noted that the proapoptotic role of RGS6 after doxorubicin treatment of cardiac myocytes, mouse embryonic fibroblasts, and MCF-7 breast cancer cells occurs through a mechanism that is independent of its interaction with G proteins (Huang et al., 2011; Yang et al., 2013). Thus, the involvement of RGS6 in both G protein–dependent and G protein–independent signaling mechanisms could account for its ability to either decrease or increase myocardial injury in the contexts of ischemia or doxorubicin treatment, respectively.

Single nucleotide polymorphisms have been identified in the human gene encoding RGS6 (Sibbel et al., 2011). Posokhova et al. (2013) identified a RGS6 single nucleotide polymorphism (Genbank accession number NM_004296.5) in which cytosine 37 is deleted, resulting in a frame shift and a premature stop codon. This mutation results in a loss of RGS6 function and is associated with increased heart rate variability in humans who are heterozygous for this polymorphism. The prevalence of this polymorphism is unknown. However, our data suggest that carriers of this polymorphism or other polymorphisms that result in a loss of RGS6 function may be at increased risk of cardiac ischemic injury.

RGS proteins represent a potential class of new therapeutic drug targets (Sjögren and Neubig, 2010). Our data demonstrate that a deficiency of RGS6 worsens ischemic injury, but it is unclear whether overexpression of RGS6 to quantities above endogenous levels would have a cardioprotective effect. Prior work has demonstrated that RGS2 and RGS6 expression can be enhanced in the ventricles by treatment with cardiotonic steroids and doxorubicin, respectively (Sjögren et al., 2012; Yang et al., 2013). Despite the cardiotoxic effects of these agents, these studies establish the proof of concept that pharmacological intervention can be used to increase the expression of RGS protein in the ventricles. The development of agents that increase RGS6 expression could potentially provide a new approach to treating ischemic heart disease.