Partial inhibition of the ubiquitin–proteasome system ameliorates cardiac dysfunction following ischemia–reperfusion in the presence of high glucose

The link between hyperglycemia and increased morbidity/mortality with heart diseases such as acute myocardial infarction (AMI) is receiving growing attention [1‐3]. Hyperglycemia can manifest as a more chronic condition (typically found with diabetes) or as a stress-induced state, e.g. in diabetic and non-diabetic persons challenged by an AMI. Chronic hyperglycemia may sometimes be linked to the development of macrovascular complications [4], with AMI the chief contributor to cardiovascular mortality with diabetes [5‐7]. Acute, stress-induced hyperglycemia is associated with increased in-hospital deaths, congestive heart failure, and cardiogenic shock [8‐10]. The presentation of hyperglycemia in non-diabetic individuals likely represents a combination of undiagnosed diabetes, impaired glucose tolerance [11, 12], and a robust response to acute stress [8, 9].

Hyperglycemia-induced oxidative stress can lead to the formation of misfolded or damaged proteins that are usually eliminated by the ubiquitin–proteasome system (UPS), the major role player responsible for non-lysosomal protein degradation [13, 14]. However, a dysfunctional UPS is linked with a pro-inflammatory milieu and impaired cardiac function in diabetic mice [15] and diabetic patients presenting with unstable angina [16]. Although altered UPS functioning is implicated with diminished cardiac function, it remains unclear whether greater or lesser UPS activation is detrimental with ischemia–reperfusion under such conditions. This conundrum also applies for UPS regulation in response to ischemia–reperfusion per se, i.e. under normoglycemic conditions. For example, some found that oxidative stress, inflammation and myocardial reperfusion injury can be blunted by proteasomal inhibitor treatment [15, 16]. Furthermore, UPS inhibition leads to a reduction in ventricular tachyarrhythmias [17] and preserved cardiac function following myocardial ischemia under baseline glucose conditions. Conversely, others established that UPS inactivation (due to oxidative damage occurring with ischemia–reperfusion) is a robust contributor to the damaging effects elicited by an AMI [18]. Such discordant findings may be due to different experimental protocols/models utilized and due to variations in UPS inhibitor class, dosages and length of administration [19, 20]. With the paucity of studies investigating UPS regulation with ischemia–reperfusion under acute hyperglycemic conditions, efforts to gain greater insight into this intriguing question are essential as it will contribute to (a) the delineation of underlying mechanisms, and (b) the development of novel cardioprotective agents.

As our earlier findings implicated increased UPS activation in hyperglycemia-mediated cardiac damage [21] we hypothesized that its inhibition during early reperfusion provides cardioprotection following ischemia–reperfusion under high glucose conditions. We employed an established rat model of ex vivo isolated heart perfusions subjected to ischemia–reperfusion and exposed to simulated acute hyperglycemia as before [21, 22]. To further strengthen these data two different UPS inhibitor classes were used, i.e. lactacystin (2-(acetylamino)-3-[({3-hydroxy-2-[1-hydroxy-2-methylpropyl]-4-methyl-5-oxopyrrolidin-2-yl}carbonyl)sulfanyl]propanoic acid) and MG-132 (N-(benzyloxycarbonyl) leucinylleucinylleucinal Z-Leu-Leu-Leu-al). The current study demonstrates that proteasome inhibition elicits cardioprotection in hearts exposed to ischemia–reperfusion with acute simulated hyperglycemia by upregulating anti-oxidant defenses, enhancing anti-inflammatory effects and increasing autophagy.

To rule out the possibility of attributing the effects of hyperglycemia on heart function to changes in osmolarity, separate perfusion experiments were completed with 11 mM glucose plus 22 mM mannitol (total molarity 33 mM). These data demonstrate no differences in LVDP recovery and dP/dt_max for the osmotic control group vs. hearts perfused under baseline glucose (11 mM) conditions (Fig. 1a, b). By contrast, hearts exposed to high glucose displayed lower LVDP recovery compared to the osmotic control group (p < 0.05) with no differences for dP/dt_max (Fig. 1c, d). High glucose perfusions also resulted in greater infarct sizes compared to both control groups (p < 0.01) (Fig. 1e).

We next determined the effects of high glucose exposure on post-ischemic UPS activities. Here trypsin-, chymotrypsin- and caspase-like activities were significantly upregulated (Fig. 2). The effects of lactacystin and MG-132 were determined and both compounds—as expected—significantly lowered myocardial chymotrypsin-like activity (Fig. 2). To further validate the degree of UPS inhibition we also assessed peptide levels of E3 ligases (MURF1 and MAFbx) involved in the final step of the ubiquitination process. Here MAFbx expression was significantly increased under high glucose conditions, while both lactacystin and MG-132 administration decreased MURF1 and MAFbx expression following ischemia–reperfusion under high glucose conditions (Fig. 3).

We next evaluated functional parameters in our experimental system and found that high glucose exposure blunted cardiac contractile function vs. matched controls following ischemia–reperfusion (Figs. 4, 5). Acute lactacystin administration during the reperfusion phase significantly improved LVDP recovery at baseline (Fig. 4a) and also under high glucose conditions (from 10 ± 2.3% to 36 ± 4%; p < 0.01 vs. untreated high glucose) (Fig. 4c). RPP data exhibited a similar trend although lactacystin had no effect under baseline glucose conditions compared to the untreated group (Fig. 4b). MG-132 elicited robust cardioprotection (LVDP, RPP) at baseline and in response to high glucose exposure (Fig. 5). Moreover, both lactacystin and MG-132 decreased end-diastolic pressure in hearts exposed to high glucose conditions, although there were no differences for the control groups (Table 1). Proteasome inhibitor treatment also enhanced coronary flow although this was only statistically significant for MG-132 (p < 0.05 vs. untreated high glucose) (Table 1). Our comparative analyses show that MG-132 administration resulted in greater cardioprotection compared to lactacystin (Fig. 6). The regional ischemia experiments further support these findings by demonstrating that lactacystin blunted the high glucose-induced increase in the infarct size/area-at-risk from 66.4 ± 3.4% to 48 ± 3.9% (p < 0.05 vs. high glucose untreated) (Fig. 7). Likewise, MG-132 treatment decreased infarct sizes under baseline and high glucose perfusion conditions. Here MG-132 significantly attenuated the infarct size/area-at-risk to 40.3 ± 2.5% and 49.7 ± 7.4% vs. 56.3 ± 2.8% and 73.5 ± 4.9% at baseline and high glucose perfusions, respectively. There were no differences in the area-at-risk for any of the experimental groups investigated (data not shown).

For the current study, high glucose exposure diminished myocardial SOD1, but not SOD2 expression levels following ischemia–reperfusion (Fig. 8). Myocardial TNF-α expression was also significantly elevated while IkBα levels remained unaltered (vs. baseline glucose conditions) following ischemia–reperfusion in the presence of high glucose (Fig. 9). Due to the interplay between the UPS and autophagy we also investigated the effects of lactacystin (more specific inhibitor vs. MG-132) in this context. LC3-II protein levels were not altered following ischemia–reperfusion under high glucose conditions (Fig. 10). However, lactacystin treatment resulted in a robust elevation in cardiac LC3-II expression (p < 0.001 vs. high glucose untreated). The expression of p62 was not significantly altered in response to high glucose but was significantly increased after lactacystin administration (Fig. 10).

This study demonstrates that both lactacystin and MG-132 administration elicited cardioprotection during ischemia–reperfusion performed under acute high glucose conditions. Here initial experiments demonstrated a robust increase in post-ischemic myocardial UPS activities together with elevated protein expression of an E3 ligase (MAFbx). Moreover, the proteasome inhibitor doses employed and the treatment duration resulted in partial UPS impairment together with cardioprotective effects.

Previous studies found decreased proteasome function following ischemia–reperfusion, although the underlying mechanisms remain unclear (reviewed in [31]). This likely occurs due to higher free radical generation with ischemia–reperfusion, thereby leading to direct oxidative modification and subsequent attenuation of proteasome function [18, 31]. In agreement, we detected increased proteasome activities in rat hearts subjected to ischemia–reperfusion performed under baseline glucose conditions vs. pre-ischemic hearts (data not shown). However, with hyperglycemia excess oxidative stress triggers higher UPS activity to help degrade and remove damaged proteins. For example, we recently reported increased cardiac superoxide levels together with decreased SOD activity following ischemia–reperfusion performed under high glucose perfusion conditions [21, 22]. These findings are consistent with other studies demonstrating higher UPS activity in diabetic mouse hearts [15], streptozotocin-infarcted rat hearts and diabetic patients suffering from unstable angina [16]. What about the functional effects of proteasome inhibition with ischemia–reperfusion under hyperglycemic conditions? This study shows—for the first time as far as we are aware—that both lactacystin and MG-132 administration triggered cardioprotective effects with ischemia–reperfusion performed under such conditions. This is in agreement with previous findings showing that high glucose levels trigger myocardial oxidative stress and apoptosis in parallel with impaired contractile function [22].

The effects observed under baseline glucose conditions are consistent with earlier research establishing that proteasome inhibition results in cardioprotection with ischemia–reperfusion [17, 32‐34]. Although both beneficial and detrimental effects were observed with proteasome inhibition, the collective data indicate that cardioprotection with ischemia–reperfusion occurs only with acute drug administration [31]. Moreover, the outcomes of UPS inhibition are strongly dose- and time-dependent with lower, non-toxic doses likely upregulating anti-oxidant defenses and exerting anti-inflammatory effects, while higher doses and chronic treatment may result in damaging consequences on protein turnover and cell death [31]. This may also depend on distinct intracellular sub-compartments, with some suggesting that beneficial effects may be due to the inhibition of proteasomal activity within the non-cardiomyocyte compartment where it may exert anti-inflammatory effects [35]. In support, others found that moderate cardiomyocyte-restricted proteasome inhibition exacerbated ischemia–reperfusion injury in mouse hearts [36]. Further studies are, however, required to investigate these intriguing possibilities.

The comparative analyses performed establish that MG-132 administration lead to greater cardioprotection compared to lactacystin. We are unclear regarding the underlying mechanism(s) whereby MG-132 resulted in such a striking improvement in cardiac function. As indicated before, proteasome inhibitor dose and treatment duration are crucial parameters that may affect functional outcomes and we therefore cannot exclude the possibility that lower lactacystin dosages could potentially result in enhanced cardioprotection. Lactacystin is also considered a more specific inhibitor than MG-132 and it is therefore possible that the superiority of MG-132 may be due to non-UPS-related effects. For example, MG-132 can exert additional effects on calpains and cathepsins [37] and also protect cardiomyocytes from injury through the inhibition of NF-kB and downregulation of transforming growth factor beta 1 (TGFβ1) [38]. As MG-132 significantly augmented coronary flow, it is possible that it may also exert advantageous effects on cardiac endothelial cells. In support, an in vitro study found that low-dose proteasome inhibition upregulated endothelial nitric oxide synthase expression and activity together with improved endothelial function [39]. This may therefore lead to increased oxygen supply to the myocardium resulting in improved post-ischemic function [40].

To gain a better understanding of the mechanisms whereby proteasome inhibitors elicit cardioprotection, we assessed whether it exhibits anti-oxidant and/or anti-inflammatory properties in our experimental system. It is well established that myocardial ischemia–reperfusion is associated with increased oxidative stress resulting in cell death [41, 42]. For example, we recently reported that several markers of myocardial oxidative stress were significantly increased in our experimental model [21, 22]. Of note, high glucose availability leads to oxidative stress due to an imbalance in terms of free radical production and intracellular anti-oxidant defenses [43‐45]. For the current study high glucose exposure diminished myocardial SOD1, but not SOD2 expression levels following ischemia–reperfusion. Both lactacystin and MG-132 administration resulted in higher cardiac SOD1 and SOD2 protein levels and this is likely a key mechanism to detoxify increased free radical levels. Although we are unclear regarding the transcriptional modulators implicated in this process, the basic leucine zipper transcription factor, nuclear factor-erythroid 2-related factor 2 (Nrf2), is a strong candidate as it is a known UPS target [46, 47]. It is likely that UPS inhibition in our experimental system attenuates Nrf2 ubiquitination and subsequent degradation, thereby increasing its stability and translocation to the nucleus for activation of target anti-oxidant genes [48‐50].

The effects of proteasome inhibition on myocardial inflammation were investigated by evaluating TNF-α peptide levels and NFkβ activation. For the latter, we determined myocardial IkBα levels—a UPS target—that when bound to NFκβ in the cytosol prevents its translocation to the nucleus and the subsequent activation of target genes such as TNF-α. Here TNF-α expression was significantly elevated while IkBα levels remained unaltered following ischemia–reperfusion in the presence of high glucose. Lactacystin administration blunted the TNF-α spike under baseline and high glucose conditions, while this effect was not observed with MG-132. However, IkBα levels were significantly higher following lactacystin and MG-132 administration. As lactacystin is a more potent proteasome inhibitor than MG-132, this suggests that its cardioprotective effects are more strongly linked to the attenuation of UPS-mediated induction of inflammatory pathways. In support, a disproportionately elevated UPS can lead to NFκβ activation, thereby resulting in downstream transcriptional effects and the establishment of a pro-inflammatory milieu with hyperglycemia and/or ischemia [16]. Thus these data establish that the strong pro-inflammatory effect triggered in response to ischemia–reperfusion and hyperglycemia can be blunted by partial proteasome inhibition.

Due to the interplay between the UPS and autophagy we also investigated the effects of lactacystin (more specific inhibitor vs. MG-132) in this context. LC3-II levels increased while there was also a significant decrease in p62 expression, indicating greater autophagic flux. Here p62 can shuttle ubiquitinated proteins not removed by the UPS to the autophagic clearance system [51], thus serving as an adapter that is able to interact with both the UPS and autophagic machinery to eventually help with the degradation of misfolded and damaged proteins [52]. These data are consistent with others demonstrating that low dose proteasome inhibitor treatment (reducing chymotrypsin-like activity to ~46%) increased myocardial autophagy [53]. Lower p62 levels would indicate greater clearance and uptake of damaged proteins by cardiac autophagosomes. Thus with UPS inhibition there is an upregulation of autophagy following ischemia–reperfusion under high glucose conditions, likely an adaptive mechanism that assists with the removal of misfolded and damaged proteins. This is in agreement with others reporting that experimentally-induced upregulation of autophagy was linked to improved heart function in diabetic mice [54].

This study established that partial proteasome inhibition elicits cardioprotection in hearts exposed to ischemia–reperfusion with acute simulated hyperglycemia. Our findings also show three major protective effects, i.e. (a) enhancing myocardial anti-oxidant defences, (b) attenuating inflammation, and (c) increasing the autophagic response. These data are encouraging as it could lead to the development of novel therapeutic interventions that may help treat individuals burdened with hyperglycemia and suffering associated cardiovascular complications such as an AMI.