In the present study, we compared the effects of the sGC stimulator BAY 60-4552 with those of the sGC activator GSK2181236A on cardiovascular physiology in acute and chronic models of cardiovascular disease associated with oxidative stress. The main findings are: (1) in acute coronary artery I/R rats, pre-treatment with either compound did not offer cardioprotection, (2) in SHR-SP, both compounds improved survival and provided partial protection against chronic HSFD-induced end-organ damage, (3) GSK2181236A, but not BAY 60-4552, attenuated cardiac hypertrophy in a blood pressure-independent manner, and (4) the ex vivo vasodilatory effects of the NO-dependent vasodilators carbachol and SNP, but not the NO-independent vasodilators BAY 60-4552 or GSK2181236A, were attenuated by chronic HSFD in SHR-SP, indicating that impaired NO-bioavailability and not the oxidative state of sGC is responsible for the vascular dysfunction observed in this disease model. Together, these results indicate that the oxidation state of sGC in HSFD SHR-SP might be differentially altered between tissues. If sGC is oxidized to a greater extent in the heart compared to the vasculature, then sGC activation might be advantageous over sGC stimulation for mitigating cardiac hypertrophy associated with cardiovascular disease.
GSK2181236A was originally identified following a high throughput screen for compounds which increase the enzymatic activity of sGC. Subsequent mechanistic studies using ODQ, a sGC heme-oxidant, indicated that GSK2181236A functions as a sGC “activator” (i.e., it increases the enzyme activity of the oxidized, NO-insensitive form of sGC in a heme-independent manner). Specifically, consistent with the effects of ODQ on other sGC activators (Stasch et al., 2002b
; Schindler et al., 2006
; Zhou et al., 2008
), ODQ pretreatment of rat vascular smooth muscle cells or vascular tissue augmented the functional effects (P-VASP formation and vasodilation, respectively) of GSK2181236A. Additional characterization of GSK2181236A in cell-based assays revealed that the interaction with sGC was at least 23-fold more potent than any other non-sGC protein tested, further supporting the utility of GSK2181236A as a pharmacological tool. BAY 60-4552, a sGC “stimulator” (i.e., it increases the enzyme activity of only the reduced, NO-sensitive form of sGC in a heme-dependent manner) currently being investigated in patients with biventricular heart failure (Mitrovic et al., 2009
), was used for comparative purposes. As expected, in contrast to GSK2181236A, ODQ attenuated the functional effects of BAY 60-4552 in vitro
One of the main adverse events associated with sGC stimulation/activation is hypotension (Mitrovic et al., 2009
; Erdmann et al., 2010
). This problem was mitigated in the current study by selecting non-hypotensive and modestly hypotensive doses of GSK2181236A and BAY 60-4552. Both doses are likely to be physiologically relevant since sGC stimulators/activators have been previously shown to elicit blood pressure-independent beneficial effects during chronic renal disease in subtotally nephrectomized rats (Benz et al., 2007
) and in aged spontaneously hypertensive rats (Jones et al., 2009
). Overall, both compounds elicited similar dose-dependent effects on heart rate and blood pressure, with GSK2181236A being ~3-fold more potent. Equi-efficacious doses of both compounds were used for the I/R and SHR-SP studies.
Administration of GSK2181236A and BAY 60-4552 2
h prior to cardiac I/R failed to reduce infarct size. This was an unexpected finding considering that administration of NO-donors before cardiac I/R consistently diminishes infarct size (Bolli, 2001
; Schulz et al., 2004
). Although NO could exert cardioprotective effects via non-sGC/cGMP mechanisms such as direct nitrosylation of proteins (Stamler et al., 1992
; Choi et al., 2002
), it is widely accepted that the cGMP/PKG pathway is the pivotal mechanism by which NO mediates cardioprotection during I/R. Indeed, it has been shown that natriuretic peptides and phosphodiesterase inhibitors (which also modulate cGMP levels) reduce infarct size following I/R (Burley et al., 2007
). Whereas several decades of intense investigations support a protective role for cGMP in I/R injury, the role of direct NO-independent sGC stimulation/activation during cardiac I/R is only beginning to be established. The sGC activator BAY 58-2667 improved cardiac function when administered 5
min prior to reperfusion following 60
min global cardiac ischemia in dogs (Korkmaz et al., 2009
). Similarly, when administered 5
min prior to reperfusion following 30
min regional ischemia, BAY 58-2667 decreased infarct size in rat (Krieg et al., 2009
) and rabbit hearts (Krieg et al., 2009
; Cohen et al., 2010
). Interestingly, inhibition of NO production with the eNOS inhibitor Nω
-arginine methyl ester (l
-NAME) exerted differential effects in these studies, blocking the cardioprotective effects BAY 58-2667 in rabbit (Cohen et al., 2010
) but not in rats (Krieg et al., 2009
). These results indicate a PKG-independent role for endogenous NO during I/R which could be species-dependent. In comparison to these studies where drug treatment was initiated 5
min prior to reperfusion, we administered drug 2
h prior to the ischemic insult. Additional studies are required to determine whether or not the timing of drug administration or another factor is responsible for the lack of cardioprotection observed in the present study.
Similar to previous studies, HSFD in SHR-SP induced a progressive, malignant hypertension with subsequent development of renal dysfunction (microalbuminuria), mild cardiac hypertrophy, and increased morbidity/mortality (Kerr et al., 1999
; Barone et al., 2001
; Behr et al., 2001
; Ma et al., 2001
). The gene expression profile observed in this study (increased ANF, α-skeletal actin, and TGF-β1 and decreased α-MHC expression) was consistent with maladaptive cardiac remodeling. Endothelial dysfunction, resulting from increased reactive oxygen species (ROS; e.g., superoxide and peroxynitrite, the reactive nitrogen intermediate), reduced bioavailable NO, or impaired NO-induced activation of oxidized sGC, contributes to the development and progression of this chronic hypertension and is likely a responsible mechanism for the hypertension-induced end-organ damage in this model (McIntyre et al., 1997
; Ma et al., 2001
; Ju et al., 2003
; Stasch et al., 2006
Considering that the vasorelaxation and cardiac and renal protective effects of the sGC activator BAY 58-2667 were potentiated under pathophysiological and oxidative stress conditions (Stasch et al., 2006
) and that removal of the sGC heme moiety or its inactivation by oxidation strongly diminished the ability of sGC stimulators to modulate sGC (Evgenov et al., 2006
), it has been hypothesized that pharmacological enhancement of the NO/sGC pathway via sGC activators, not stimulators, would be the more effective therapy for prevention and treatment of cardiovascular disease in the presence of excess ROS. This hypothesis has not previously been tested comprehensively in vivo
. In contrast to expectations, however, improvement of endothelial function was not potentiated by sGC activation as compared to sGC stimulation. Instead, the vasorelaxation, blood pressure lowering, and renal protective effects of the sGC stimulator BAY 60-4552 were similar to or slighter greater than that of the sGC activator, GSK2181236A, at the doses examined in this study. Overall, these results suggest that the oxidation state of sGC capable of modulating blood pressure and renal function is not altered to an extent necessary to exploit the differing mechanisms of actions of these pharmacological tools.
Interestingly, however, GSK2181236A, but not BAY 60-4552, attenuated cardiac hypertrophy in a blood pressure-independent fashion. These results suggest that sGC in cardiac tissue might be oxidized to a level sufficient to distinguish between sGC stimulation and activation. Growing evidence suggests that ROS play a pathophysiological role in the development of cardiac hypertrophy (see Seddon et al., 2007
for review), and indicates that sGC activation might be advantageous over sGC stimulation for mitigating maladaptive cardiac hypertrophy associated oxidative stress. However, additional studies are required to elucidate regional differences in sGC oxidation state.
The vasodilatory responses to compounds which modulate the NO/sGC pathway at different points were investigated in HSFD SHR-SP to help elucidate the mechanism(s) responsible for the vascular dysfunction. The vasodilatory responses to the endothelium-dependent vasodilator carbachol and the endothelium-independent vasodilator SNP were attenuated by HSFD, consistent with previous observations (Ju et al., 2003
; Willette et al., 2009
). These results alone could be explained by multiple impairments in the NO/sGC pathway, including reduced NO production, excessive NO degradation/neutralization, sGC oxidation, altered phosphodiesterase activity, etc., and thus do not clarify the mechanism for the impaired vasodilation. In contrast to NO-dependent vasodilation, however, the vasodilatory responses to the NO-independent compounds BAY 60-4552 and GSK2181236A were unaltered by HSFD. Taken together, three important conclusions can be made. First, the mechanism responsible for the impaired vasodilation is upstream of sGC and is likely reduced NO-bioavailability due to the high levels of oxidative stress and subsequent peroxynitrite formation associated with HSFD in SHR-SP (Ma et al., 2001
). Second, the level of oxidative stress in this model is high enough to inactivate NO, but not high enough to alter the oxidation state of sGC to a point necessary to impact the vasodilatory effects of direct (NO-independent) sGC stimulators or activators. Rather, a level of sGC oxidation seen with 10
μM ODQ is likely necessary to impact NO-independent sGC stimulator/activator-mediated vasodilation. As such, it is questionable whether or not the ODQ results represent a biochemical artifact or a pathophysiological state. Interestingly, in contrast to our results with GSK2181236A, Stasch et al. (2006
) have shown that the vasodilatory effects of a different sGC activator, BAY 58-2667, are potentiated under pathophysiological conditions in animal models [aged SHR, hyperlipidemic (WHHL) rabbits and atherosclerotic (ApoEl−/−
) mice] and humans (mesocolon arteries taken from patients with type 2 diabetes). The precise reason for this discrepancy is unknown, but could be the result of different animal models or compounds. Indeed, although both GSK2181236A and BAY 58-2667 are sGC activators with structural similarities, their precise mechanism(s) of action might differ. Additional studies are necessary to elucidate these discrepancies. Third, consistent with the in vivo
hemodynamic findings, the hypothesis that sGC activation offers a therapeutic advantage over sGC stimulation is not necessarily correct. Although this hypothesis appears accurate when comparing NO-mediated stimulation of sGC (e.g., via muscarinic receptor activation or NO-donors) with sGC activators (e.g., BAY 58-2667; Stasch et al., 2006
, GSK2181236A), other than differential effects on cardiac hypertrophy, there was no clear differentiation between NO-independent sGC stimulation and activation in the present study. These findings emphasize that additional studies which directly compare NO-independent sGC stimulators with activators are necessary in order to better predict which class of agents might offer a therapeutic advantage in a particular patient population.
In summary, the present study is the first to compare a NO-independent sGC stimulator (BAY 60-4552) with a sGC activator (GSK2181236A) directly and comprehensively both in vitro and in vivo. Whereas neither compound attenuated acute cardiac I/R injury, both provided dose-dependent levels of protection against HSFD-induced end-organ damage in SHR-SP, a model associated with profound oxidative stress, hypertension, and renal dysfunction. Isolated artery dilation studies indicate that reduced NO-bioavailability and not the oxidation state of sGC is responsible for the HSFD-induced impaired vasodilation, and thus provide a rationale as to why the hypotensive effects of sGC activation was not potentiated. In contrast, the differential effects of sGC stimulation and activation on HSFD-induced cardiac hypertrophy suggest that the oxidative state of sGC is altered to a greater degree in cardiac than vascular tissue. Overall, these results suggest that the vasorelaxation, blood pressure, and end-organ protective effects of sGC activators can be clearly differentiated from those of NO-donors, but not from those of sGC stimulators in rodent models of oxidative stress. However, sGC activation might be advantageous over sGC stimulation for mitigating cardiac hypertrophy associated with cardiovascular disease. Additional studies designed to further define the pharmacological differences between sGC stimulators and sGC activators will be necessary to understand the potential clinical differentiation of these two classes of agents.