Impact of hyperglycemia on myocardial ischemia–reperfusion susceptibility and ischemic preconditioning in hearts from rats with type 2 diabetes

Circulating blood glucose levels influences the efficacy of endogenous cardioprotection. Here, we investigated specifically in hearts from animals with T2DM whether hyperglycemia had impact on endogenous cardioprotection. We find that infarct size in hearts from animals with T2DM is smaller than in hearts without T2DM during both normo- and hyperglycemia. This indicates that the endogenous cardioprotection observed in animals with T2DM during normoglycemia is preserved during hyperglycemia. We also investigated whether hyperglycemia per se influences myocardial susceptibility to ischemia–reperfusion injury. Myocardial infarct size was larger in animals with and without T2DM during hyperglycemia than during normoglycemia. This may indicate that myocardial susceptibility to ischemia–reperfusion injury is increased during hyperglycemia in T2DM to the same extend as demonstrated in animal models of type 1 diabetes [10, 11].

Circulating blood glucose levels also influences the efficacy of exogenous cardioprotection. We have recently reported that the effect of exogenous cardioprotection by IPC is preserved in animals with T2DM in contrast to animals without diabetes during hypoglycemia [9]. During hyperglycemia, a previous study failed to demonstrate an infarct sparring effect of preconditioning in hearts from animals with type 1 diabetes [11]. Here, we assessed the influence of hyperglycemia on the efficacy of exogenously induced cardioprotection in T2DM hearts and observed that IPC-induced infarct reduction was abrogated by a standard IPC protocol. However, using an intensified IPC protocol with four cycles of ischemia and reperfusion [7, 30], myocardial infarct size was reduced during hyperglycemia in hearts from animals with T2DM, while only to a minor and statistically insignificantly extend in hearts from animals without diabetes. These findings may reflect an additive effect of endogenous and exogenous cardioprotection in hearts from animals with T2DM and that hyperglycemia influences sensitivity to myocardial ischemia–reperfusion injury. The efficacy of IPC in hearts from animals with diabetes during normoglycemia remains controversial. The conflicting results may be related to the use of different models of diabetes. In models of T2DM, the cardioprotective effect of IPC during normoglycemia is abrogated [5, 31] or attenuated with an intensified preconditioning stimulus required to induce cardioprotection [7, 32, 33]. Here, we confirm that hearts from animals with T2DM are modifiable by exogenous cardioprotection by IPC during normoglycemia.

Evidence points towards alterations in myocardial glucose metabolism as a mechanistic link between preconditioning and cardioprotection because IPC is associated with increased glucose uptake during ischemia and reperfusion [12, 34‐36]. We validated and extended the findings from normoglycemia to hyperglycemia by demonstrating that IPC-induced cardioprotection is associated with increased myocardial glucose uptake during hyperglycemia. Notably, the cardioprotective effect of exogenous cardioprotection by IPC is associated with increased myocardial glucose uptake in T2DM as well as non-diabetic hearts, while the endogenous cardioprotection observed in T2DM hearts is not. Hence, the underlying mechanisms of exogenously induced and endogenous cardioprotection seem to differ at a metabolic level. The IPC mediated increment in glucose uptake during ischemia–reperfusion has largely been attributed to activation of AMP-activated protein kinase (AMPK) and GLUT4 translocation [36]. Protein O-GlcNAcylation is sensitive to extracellular glucose levels and may also influence glucose uptake [37‐39]. Exogenous cardioprotection by IPC and remote IPC involves modulation of myocardial O-GlcNAc levels and can be abrogated by azaserine, a commonly used inhibitor of glutamine:fructose-6-phosphate amidotransferase (GFAT), suggesting that O-GlcNAcylation is involved in the underlying mechanisms [15, 16, 28]. We extend these findings by demonstrating that exogenous cardioprotection by IPC during hyperglycemia is associated with increased myocardial levels of O-GlcNAc. In contrast, the endogenous cardioprotection observed in T2DM during both normo- and hyperglycemia was not associated with altered myocardial levels of O-GlcNAc suggesting different mechanisms of action between exogenous induced and endogenous cardioprotection at a post-translational level.

Alternative mechanisms of exogenously and endogenously induced cardioprotection may relate to the myocardial phosphatidylinositol 3-kinase/protein kinase B (PI3K/Akt) signaling system. The important role PI3/Akt signaling in ischemic preconditioning is well documented [40, 41], whereas it is reported to be defective in hearts from diabetic animals [7]. Furthermore, as hyperglycemia per-se suppresses Akt phosphorylation [42], our findings of reduced cardioprotection by IPC during hyperglycemia but preserved endogenous cardioprotection in diabetic animals are in support of a diverse role of PI3/Akt signaling in exogenously induced cardioprotection by IPC and the endogenous cardioprotection observed in hearts from diabetic animals. However, differences in PI3/Akt signaling do not seem to explain the disparate findings related to myocardial susceptibility to IR injury in different animal models of diabetes. In an animal models of type 1 diabetes mellitus induced by streptozotocin, decreased myocardial Sirt1 activity leading to decreased Akt phosphorylation and increased Drp1 activity increases myocardial susceptibility to IR injury [43]. In contrast, the Goto-Kakizaki rat, a lean model of T2DM, has lower myocardial levels of phosphorylated Akt and decreased myocardial susceptibility to IR injury. Future studies are needed to expand our understanding of the mechanisms related to differences in myocardial susceptibility to IR injury in different animal models of diabetes.

The clinical implications of our findings are that myocardial infarct size in patients with or without T2DM may be influenced by changes in myocardial susceptibility to ischemia–reperfusion injury caused by hyperglycemia per-se similarly to the animals investigated in the present study. Furthermore, the effect of exogenous cardioprotection activated prior to an ACS may partly or completely be abrogated by hyperglycemia in patients with or without T2DM, which may result in increased myocardial infarct size. As myocardial infarct size is a prognostic factor after acute myocardial infarction, this finding may in part explain the observed increased mortality in patients with ACS and elevated circulating glucose levels at hospital admission. However, as infarct sizes at both normo- and hyperglycemia were smaller in animals with than without T2DM, infarct size alone does not seem to explain the overall higher mortality in patients with T2DM suffering acute myocardial infarction compared with patients without diabetes. The higher mortality in patients with diabetes may merely be related to arrhythmias after acute myocardial infarction, as suggested by findings of exogenously induced conditioning being ineffective against IR-induced atrioventricular block in animals with diabetes in contrast to animals without diabetes [44].

A strength of the present study is the use of a clinically relevant animal model of T2DM as patients with T2DM prevail in clinical patient cohorts of acute myocardial infarction. Yet, our T2DM rats were rather young with recent onset of diabetes and the concurrent effect of age was not investigated although T2DM is often diagnosed in elderly patients. A limitation of the study relates to the use of an isolated perfused heart model. However, induction of hyperglycemia in vivo influences insulin production and levels, which impacts myocardial ischemia–reperfusion injury [45] and may influence myocardial glucose uptake and O-GlcNAc levels differently in T2DM and non-diabetic hearts. Furthermore, the influence of the systemic response to hyperglycemia including activation of neurogenic and humoral components can be excluded in our model. Our diabetic animal were exposed to hyperinsulinemia until the time of IR, which was performed without insulin as this per se influences cardioprotection [46]. Performing IR during hyperinsulinemia in diabetic animals and during normoinsulinemia in non-diabetic animals in our model is not believed to have impacted the comparison of the results between animals with and without diabetes substantially as the mechanism whereby insulin leads to Akt activation is impaired in the diabetic state by upregulation of phosphatase and tensin homolog deleted on chromosome 10 (PTEN) [47, 48]. However, as hyperglycemia diminishes the cardioprotective effects of insulin [42], the presence of insulin during IR may have augmented the difference in IR injury between normo- and hyperglycemia in animals without diabetes. As the ZDF model of T2DM is also characterized by high circulating cholesterol, triglycerides and free fatty acids concentrations [5], our diabetic animals were also exposed to hyperlipidemia until the time of IR. Conflicting data regarding the impact of hyperlipidemia on cardioprotection have been reported [49, 50]. However, the potential influence of hyperlipidemia on cardioprotection does not require presence of the lipids at the time of IR as hyperlipidemia prior to IR has been demonstrated to modulate cardioprotection in isolated hearts perfused with Krebs–Henseleit buffer containing glucose as the only substrate in the absence of lipids [51, 52]. Such findings may be explained by altered myocardial signaling, gene and microRNA profiles caused by exposure to hyperlipidemia [50] and such effects would also present in our study. We specifically chose to investigate the effects of hyperglycemia on myocardial infarct size in a glucose-dependent model to investigate the impact on MGU without interference from other substrates including lipids and ketones. We cannot exclude that our findings may have been influenced by substrate competition in the presence of free fatty acids as well as other substrates. Variations in circulating glucose concentrations changes osmolarity, which may influence infarct size and an increase in plasma glucose from 11 to 22 mmol/L leads to a 3% increase in osmolarity of the perfusion buffer. The influence of osmolarity on infarct size has varied in previous studies. Increases in serum osmolarity by administration of raffinose did not modify infarct size or impact the capacity of IPC to protect against infarction as demonstrated by Kersten et al. [11]. In contrast, Zalesak et al. demonstrated that a hyperosmotic state reduced the cardioprotective capacity of IPC and decreased myocardial susceptibility to ischemia–reperfusion injury suggesting that increased osmolarity may play a role in lack of cardioprotection by IPC in the myocardium from animals with diabetes [23]. In the present study, glycemic and osmolarity levels were altered similarly by hyperglycemia and as infarct size was similarly in both animals with and without diabetes during hyper-compared with normoglycemia we conclude that the increased osmolarity during hyperglycemia in the present study does not influence myocardial ischemia–reperfusion injury. We cannot establish whether the blunted effect of IPC, although still present with an intensified stimulus in animals with T2DM, was caused by hyperglycemia or an increase in osmolarity.