Myocardial Insulin Resistance After Myocardial Ischemia Contributes To Post-ischemic Heart Failure And Its Underlying Mechanism

Backgroud Accumulating evidence from clinical and experimental studies has documented a strong association between systemic insulin resistance and heart failure(HF). Systemic insulin resistance has been demonstrated to be an independent risk factor for the development of HF. On the other hand, cinical studies have indicated that there is a high prevalence of systemic insulin resistance and impaired glucose homeostasis in nondiabetic HF patients, and such abnormalities are associated with decreased cardiac function. In these studies, only systemic insulin resistance is evaluated without considering myocardialinsulin resistance. In contrast to systemic insulin resistance, myocardial insulin resistance has little influence on systemic insulin resistance because the heart comprises less than 0.5 percent of the total body weight. Given the importance of myocardial insulin signaling in protection against ischemia-induced injury and subsequent cardiac dysfunction/remodelling, local myocardial insulin resistance per se and its association with post-ischemic HF are inadequately investigated. Therefore, rigorous investigation of ischemic HF in the context of impaired myocardial insulin signaling/action is warranted.Aims The present study was aimed to determine the role of myocardial insulin resistance in ischemic HF and its underlying mechanisms.Methods Adult male rats were subjected to permanent ligation of the left anterior descending coronary artery. Serial Doppler echocardiography was performed before the operation and 1, 2, 4 and 8 wk after the operation. LV end-systolic and end-diastolic diameters(LVESD and LVEDD respectively) were measured. Ejection fraction(EF) and fractional shortening(FS) were calculated. Fasting plasma glucose(FPG) and fasting plasma insulin(FIN) were measured. Rats were also given an oral glucose(2 g/kg) challenge or an i.p. injection of insulin(0.5 U/kg) to assess whole body glucose tolerance and insulin sensitivity. Insulin-stimulated myocardial [18F]-fluorodeoxyglucose(FDG) uptake was measured by PET/CT. Insulin signaling pathway with or without insulin stimulation was analyzed by western blot. Myocardial and serum TNF-α levels were measured using a rat ELISA kit.Adenoviral vectors for overexpression of TNF-α(Ad TNF-α) were used. Gene transfer with adenovirus encoding GFP was used as a control. Rats that had surgically induced MI(or control rats) were treated with the TNF-α inhibitor etanercept by i.p. injection(300 μg/250 g body weight) 2 d prior to coronary artery ligation and every 2 d thereafter in the first week after MI. Insulin(0.5 U/ml, 1ml/kg/d) was administrated daily in the first week after MI to investigate the effect of insulin on myocardial insulin sensitivity and the development of ischemic HF. To obtain additional direct evidence that myocardial insulin resistance contributes to post-MI HF, we obtained the mice that were homozygous for lox P-flanked insulin receptor exon, and positive for tamoxifen-inducible Cre recombinase driven by the cardiomyocyte-specific α-myosin heavy chain(MHC) promoter(MHC-Mer Cre Mer/IRfl/fl). Male mice aged 8-10 wk were injected intraperitoneally with tamoxifen(50 mg/kg) for 3 d to induce insulin receptor gene excision in cardiomyocytes. Age-matched male littermates(MHC-Mer Cre Mer/IR+/+) were used as control animals. Overnight-fasted(16?h) mice were injected i.p. with glucose(2?g?/kg) for glucose tolerance tests(GTT) or insulin(0.75?U/?kg) for insulin tolerance tests(ITT). Serial Doppler echocardiography was performed on the knockout mice and the controls before the operation and 4 wk after the operation.Results 1. Progressive LV dilation and heart dysfunction were observed over the course of the experiment. When compared with sham-operated rats, substantial reduction in EF and increased LV diameter were observed 1 wk post-MI. In contrast with 1 wk after MI, worsened LV dysfunction and dilation were observed at 4 wk post-MI. In comparison with MI 4 wk, there was no significant progression at 8 wk post-MI.2. At baseline, FDG uptake in the heart of rats assigned to MI or sham groups was comparable. Insulin injection(10 U/kg) produced robust increases in cardiac FDG uptake in sham rats 30 min after surgery. By contrast, insulin-stimulated myocardial FDG uptake was even higher at the 30 min in rats post-MI. However, after 30 min, over time, progressive decrease in insulin-stimulated myocardial FDG uptake consistent with progressive increase in myocardial insulin resistance was observed 1 d, 1wk, and 2 wk post-MI. There were no significant differences in FPG, FIN,QUICKI, OGTT, or ITT when sham-operated and post-MI rats were evaluated and compared at 1 or 2 wk post-MI. Thus, surgically-induced MI caused local myocardial insulin resistance without systemic insulin resistance or glucose intolerance. 3. Insulin stimulation of either sham or post-MI rats 1 d after operation caused robust increases in cardiac Akt phosphorylation. However, 1 wk post-MI, insulin-stimulated Akt phosphorylation was barely detectable. With respect to ERK 1/2 phosphorylation, both basal and insulin stimulated phosphorylation were a little higher in post-MI rats at 1 d when compared with sham animals. By 1 wk post-MI, basal ERK 1/2 phosphorylation was extremely elevated with no detectable insulin-stimulated increase. For p38 MAPK phosphorylation, there was no detectable insulin-stimulated increase in the heart of sham animals at baseline. At 1 wk-post-MI, insulin stimulated a robust increase in p38 MAPK phosphorylation. In sham animals, as expected, insulin stimulation caused increased GLUT4 translocation to the myocardial cell surface. In post-MI animals, insulin also stimulated GLUT4 translocation, but the magnitude of this effect was smaller 1 d post-MI, and almost absent 1 wk post-MI. 4. Overexpression of TNF-α resulted in significant increased TNF-α levels in the heart, which reached to the highest level at 4 d after adenovirus infection. An insulin injection(10 U/kg) produced a significant increase in FDG uptake and GLUT4 membrane translocation and Akt phosphorylation in the heart of ad GFP-treated rats, all of whichwere significantly lower in ad TNF-α-treated rats at 1 wk post-adenovirus infection. This was accompanied by increased p38 MAPK phosphorylation in ad TNF-α-treated hearts. 5. Echocardiographic examination at baseline revealed no significant differences in EF and LVESD when ad TNF-α-treated and ad GFP-treated rats(without MI) were evaluated. However, substantial LV dysfunction and dilation were observed in the hearts overexpressing TNF-α 1 wk post-MI when compared with ad GFP-treated controls 1 wk post-MI. Taken together, these results demonstrated that local cardiac overexpression of TNF-α mimicked myocardial insulin resistance, and exacerbated poor heart function observed with surgically-induced MI. 6. Pretreatment with etanercept during the first week after surgically-induced MI(as described in Methods) repressed local myocardial TNF-α production(circulating serum TNF-α levels were undetectable). This substantially restored insulin-stimulated Akt phosphorylation, and GLUT4 translocation as well as myocardial FDG uptake 1 wk post-MI. Moreover, etanercept treatment significantly suppressed p38 MAPK phosphorylation 1 wk post-MI. There were no significant differences in cardiac function or LV dilation when hearts from saline and etanercept-treated animals were evaluated 4 wk post-MI. 7. Early insulin treatment suppressed TNF-α production when compared with saline treatment. Insulin treatment also improved subsequent insulin-stimulated myocardial FDG uptake 1 wk post-MI, and alleviated cardiac dysfunction and dilation 4 wk post-MI as evidenced by increased EF, ±LV dp/dtmax, and decreased LVESD. 8. Immunoblotting confirmed reduction of insulin receptor protein by ~78% in cardiac muscle but not in skeletal muscle or liver. Thus, tamoxifen at 50 mg/kg body weight was sufficient to induce cardiomyocyte-specific insulin receptor deletion. Hereafter, we refer to the tamoxifen-treated MHC-Mer Cre Mer/IRfl/fl mice as tamoxifen-induced cardiomyocyte-specific insulin receptor knockout(TCIRKO). No significant differencesin heart weight/body weight were observed when control and TCIRKO mice were compared. No significant differences in GTT or ITT were observed among control and TCIRKO mice. Thus, systemic insulin sensitivity and glucose tolerance was not altered in TCIRKO mice. No significant differences in cardiac function or LV dimensions were observed when TCIRKO sham mice were compared with control mice(MHC-Mer Cre Mer/IR+/+). TCIRKO mice exhibited aggravated post-ischemic ventricular remodeling and dysfunction 4 wk after MI compared with littermate controls.Conclusion In conclusion, our results have shown that myocardial insulin resistance, independently of systemic insulin resistance, is an early event in the development of ischemic HF following MI. Impaired myocardial insulin action is at least partly induced by TNF-α overproduction, which precipitates the progression of post-ischemic HF. The results suggest that potential therapeutic strategies targeted at reversing post-ischemic myocardial insulin resistance may prevent or delay the progression of HF. Our findings are potentially translatable to the prevention and treatment of ischemic heart disease and HF.