Lin28a protects against cardiac ischaemia/reperfusion injury in diabetic mice through the insulin-PI3K-mTOR pathway

The insulin-PI3K-mTOR pathway exhibits a variety of cardiovascular activities including protection against I/R injury. Lin28a enhanced glucose uptake and insulin-sensitivity via insulin-PI3K-mTOR signalling pathway. However, the role of lin28a on experimental cardiac I/R injury in diabetic mice are not well understood. Diabetic mice underwent 30 min. of ischaemia followed by 3 hrs of reperfusion. Animals were randomized to be treated with lentivirus carrying lin28a siRNA (siLin28a) or lin28a cDNA (Lin28a) 72 hrs before coronary artery ligation. Myocardial infarct size (IS), cardiac function, cardiomyocyte apoptosis and mitochondria morphology in diabetic mice who underwent cardiac I/R injury were compared between groups. The target proteins of lin28a were examined by western blot analysis. Lin28a overexpression significantly reduced myocardial IS, improved LV ejection fraction (LVEF), decreased myocardial apoptotic index and alleviated mitochondria cristae destruction in diabetic mice underwent cardiac I/R injury. Lin28a knockdown exacerbated cardiac I/R injury as demonstrated by increased IS, decreased LVEF, increased apoptotic index and aggravated mitochondria cristae destruction. Interestingly, pre-treatment with rapamycin abolished the beneficial effects of lin28a overexpression. Lin28a overexpression increased, while Lin28a knockdown decreased the expression of IGF1R, p-Akt, p-mTOR and p-p70s6k after cardiac I/R injury in diabetic mice. Rapamycin pre-treatment abolished the effects of increased p-mTOR and p-p70s6k expression exerted by lin28a overexpression. This study indicates that lin28a overexpression reduces IS, improves cardiac function, decreases cardiomyocyte apoptosis index and alleviates cardiomyocyte mitochondria impairment after cardiac I/R injury in diabetic mice. The mechanism responsible for the effects of lin28a is associated with the insulin-PI3K-mTOR dependent pathway.


Introduction
The rising prevalence and incidence of diabetes mellitus (DM) have been reported in response to increasing rates of overweight, obesity and more sedentary lifestyle [1][2][3]. To our knowledge, the risk of coronary heart disease (CHD) is about three to fourfold higher in diabetic patients, while diabetes is associated with two to four times increased risk of CHD mortality compared with patients without diabetes [4,5]. These findings emphasize the need for increasing efforts to prevent DM and to aggressively treat and control cardiovascular disease (CVD) risk factors among diabetic patients [6].
MicroRNAs and RNA-binding proteins are emerging as pivotal modulators of mammalian cardiovascular development and disease [7][8][9][10][11][12]. Lin28, a developmentally regulated RNA-binding protein, selectively blocks the production of let-7 miRNAs [8]. When overexpressed in mice, Lin28a promoted an insulin-sensitized state that resisted high-fat diet-induced diabetes. On the contrary, muscle-specific loss of lin28a and overexpression of let-7a resulted in insulin resistance and impaired glucose tolerance [13]. These data establish the Lin28/let-7 pathway as a central regulator of mammalian glucose metabolism.
The insulin-PI3K-mTOR pathway exhibits a variety of cardiovascular activities including protection against I/R injury [14]. Zhu et al. reported that the mTOR inhibitor rapamycin abrogated the enhanced glucose uptake and insulin-sensitivity conferred by lin28a in vitro and in vivo, indicating that lin28a enhanced glucose uptake and insulin-sensitivity via insulin-PI3K-mTOR signalling pathway [13,15,16]. However, the role of lin28a on experimental cardiac I/R injury in diabetic mice is not well understood. The aims of the present study were to (i) determine whether lin28a protects diabetic mice from cardiac I/R injury and (ii) identify whether the underlying mechanisms is associated with the insulin-PI3K-mTOR dependent pathway.

Animals and surgical protocols
Male C57BL/6 mice received humane care in adherence with the National Institutes of Health Guidelines on the Use of Laboratory Animals and were approved by the Fourth Military Medical University Committee on Animal Care. Diabetic mice were induced in adult male C57BL/6 mice (6-8 weeks), weight 20-25 g by intraperitoneal (i.p.) injections of Streptozocin (STZ; 50 mg/kg, STZ was dissolved in 0.1 M citrate buffer, pH 4.5) for 5 days and high-fat diets for 2 months. A one-drop blood sample was obtained at 8 weeks from all mice through the tip of the tail for the determination of blood glucose concentration by using a reflectance meter (Accu-Chek, Roche Diagnostics GmbH, Mannheim, Germany). Animals with glucose levels not less than 16.6 mmol/l were classified as diabetes. Mice were randomly divided into the following groups with n = 20 each: (i) Non-DM; (ii) DM + sham (Sham); (iii) DM + I/R (I/R); (iv) DM + siControl + I/R (I/R + siControl); (v) DM + Lin28a siRNA + I/R (I/R +siLin28a); (vi) DM + Control vector+ I/ R (I/R + Control vector); (vii) DM + Lin28a Overexpression + I/R (I/R + Lin28a) and (viii) DM + Lin28a Overexpression + Rapamycin + I/R (I/R + Lin28a + RAP).
Ten weeks after high-fat diets was given, diabetic mice were anesthetized with 2% isoflurane, and the heart was exposed via left fifth intercostal space thoracotomy. Lentivirus (30 ll, 1 9 10 9 TU/ml) was delivered via three separate intramyocardial injections, temporarily blanching the LV free wall. Hearts were subjected to I/R injury 72 hrs after lentivirus injection [17]. Rapamycin (a specific mTOR inhibitor, 5 mg/kg) was injected via the tail vein 10 min. before cardiac I/R in the I/R + Lin28a + RAP group.

Cardiac I/R injury model construction
Cardiac I/R injury model was constructed as previously described [18]. A left thoracic incision was used to open the chest. A 6-0 silk suture slipknot was placed at the proximal one-third of the left anterior descending artery. After 30 min. of ischaemia, the slipknot was released, and the myocardium was reperfused for 3 hrs. Sham group underwent the same surgical protocols except that the suture placed under the left coronary artery was not tied.

Measurement of myocardial infarct size
After 3 hrs reperfusion, the ligature around the coronary artery was retied, and 1 ml of 2% Evans Blue dye was injected into the side arm of the aortic cannula. When the dye was well-distributed, the heart was quickly excised, frozen at À80°C and sliced transversally into 1 mm thick sections. The slices were incubated in 1%2,3,5-triphenyltetrazoliumchloride (TTC; Sigma-Aldrich, St Louis, MO, USA) for 30 min. at 37°C as previously described [18]. Blue areas which were stained by Evans Blue indicated area not at risk (ANAR). TTC stained areas which were red parts in the heart represented ischaemic but viable tissue. Staining negative areas indicated infracted myocardium. Areas of infarct size (IS) and area at risk (AAR) were measured digitally by using IMAGE PRO PLUS software (Media Cybernetics, Bethesda, MD, USA). IS and AAR were expressed as percentages of the LV area (IS/LV and AAR/LV respectively).

Determination of myocardial apoptosis
Myocardial apoptosis was determined by terminal deoxyribonucleotidyl transferase-mediated dUTP-biotin nick end labelling (TUNEL) staining as previously described [19]. TUNEL staining was performed with fluorescein-dUTP (In Situ Cell Death Detection Kit; Roche Diagnostics) for apoptotic cell nuclei and 4,6-diamidino-2-phenylindole (DAPI; Sigma-Aldrich) stained all cell nuclei. AI is the number of TUNEL-positive myocytes divided by the total number of myocytes stained with DAPI from a total of 40 fields per heart (n = 5). Cleaved Caspase-3, Caspase-3, Bcl-2, Bax were detected by Western Blot evaluation. All of these assays were performed in a blinded manner.

Determination of cardiac function
Echocardiography performed at 24 hrs after reperfusion as previously described [18]. Sedated mice (2% isoflurane) were studied on an echocardiography system (Sequoia Acuson, 15-MHz linear transducer; Siemens, Erlangen, Germany). Cardiac dimensions and function were assessed by M-mode echocardiography. LV end-diastolic diameter and LV end-systolic diameter were measured on the parasternal LV long axis view. All measurements represent the mean of 5 consecutive cardiac cycles. LV end-systolic volume (LVESV), LV end-diastolic volume (LVEDV) and LV ejection fraction (LVEF) were calculated by the use of computer algorithms. All of these measurements were performed in a blinded manner. The LV pressure was measured via a Millar Mikro-tip catheter transducer that was inserted into the LV cavity through the left carotid artery. The LV systolic pressure, LV end-diastolic pressure, first derivative of the LV pressure (AELV dp/dt max) and heart rate were obtained by use of computer algorithms and an interactive videographics programme (Po-Ne-Mah Physiology Platform P3 Plus; Gould Instrument Systems, Valley View, OH, USA).

Determination of myocardium IL-6, TNF-a and MPO activity
The concentrations of interleukin-6 (IL-6) and tumour necrosis factoralpha (TNF-a) were measured by ELISA kits according to the manufacturer's instructions. Values were expressed as pictograms per milligram of total protein. Following the 3 hrs reperfusion period, tissue samples were taken from the AAR zones for myeloperoxidase (MPO) activity analysis. The activity of MPO was measured spectrophotometrically at 460 nm and expressed as units per 100 mg of tissue.

Statistical analysis
All values and figures were expressed as mean AE SD. Comparison between groups was subjected to ANOVA followed by Bonferroni correc-tion for post hoc t-test. Two-sided tests have been used and P < 0.05 were considered statistically significant. SPSS software package version 14.0 (SPSS, Chicago, IL, USA) was performed for data analysis.

Basic parameters
There was no major difference between groups in terms of heart rate, blood glucose, body mass before the cardiac I/R injury except that blood glucose and body mass were significantly lower in the non-diabetic mice compared with the diabetic mice (Table S1).
Lin28a overexpression inhibits, while lin28a knockdown promotes let7a expression after cardiac I/R injury in mice with diabetes Lin28a overexpression and lin28a knockdown model were successfully constructed and confirmed by western blot and real-time PCR analysis ( Fig. 1A and B). Lin28a overexpression inhibited, while lin28a siRNA administration promoted let7a expression as demonstrated by real-time PCR analysis (Fig. 1C). CTs values were presented in Table S2.

Lin28a overexpression alleviates cardiomyocytes ultrastructure impairment induced by cardiac I/R injury in diabetic mice
Electron microscope showed that the number of cardiomyocytes mitochondria was increased in the diabetic group. Cardiomyocytes mitochondria were found greater in size with the destruction of cristae after cardiac I/R (30 min./3 hrs) injury in diabetic mice. Lin28a overexpression alleviated mitochondria ultrastructure impairment as demonstrated by intact membrane, normalized cristae density and architecture. On the contrary, Lin28a siRNA administration aggravated mitochondria ultrastructure impairment as compared with the I/R group. The cristae and matrix were cleared out resulting in vacuoles in most of the swelling mitochondria. Cristae disorientation and breakage were also found in most of the mitochondria. As expected, rapamycin pre-treatment abolished the effects of Lin28a protecting against cardiomyocytes mitochondria ultrastructure impairment (Fig. 2D).

Lin28a reduces cytokine levels and alleviates leucocyte infiltration after cardiac I/R injury in diabetic mice
The levels of LV IL-6, TNF-a and MPO were measured (Fig. 3G-I). Cardiac I/R injury resulted in a noticeable increase in IL-6 (31.7 AE 4.6 versus 16.5 AE 2.7 pg/mg protein, P < 0.05), TNF-a  0.211 AE 0.013, P < 0.05) compared with the I/R group. Rapamycin pre-treatment abolished the beneficial effects of lin28a overexpression on cardiomyocytes apoptosis (I/R + Lin28a + RAP: 0.204 AE 0.025, I/R + Lin28a: 0.125 AE 0.026, P < 0.05) compared with the I/R + Lin28a group (Fig. 4A and B). Cleaved Caspase-3, Caspase-3 and Bax protein levels evaluated by western blot were down-regulated by lin28a overexpression and up-regulated by lin28a siRNA administration. The pro-to anti-apoptotic protein (Bax/Bcl-2) ratio were significantly decreased in the I/R+ Lin28a group, while increased in the I/R+ siLin28a group. Furthermore, I/R + Lin28a +RAP administration increased Bax/Bcl-2 ratio, Cleaved Caspase-3, Caspase-3, Bax protein levels and decreased Bcl-2 protein level in ischaemic cardiac tissue compared with the I/R+ Lin28a group (Fig. 4C-H).

Lin28a overexpression inhibits cardiomyocytes apoptosis after cardiac I/R injury in diabetic mice
Lin28a overexpression increases IGF1R, p-Akt, p-mTOR, p-p70s6k expression in myocardium exposed to I/R injury in diabetic mice After 3 hrs of reperfusion, western blot analysis revealed that lin28a overexpression was associated with a significant increase in IGF1R and phosphorylation of Akt, mTOR and p70s6k protein in ischaemic cardiac tissue that was exposed to I/R injury in diabetic mice.

Discussion
The increased prevalence of DM is a major concern for cardiologists. DM is widely acknowledged to significantly increase the risk of developing cardiac death [20]. Treatment strategies for diabetic patients require both blood glucose control and adverse cardiovascular events control. These findings warrant the significance of aggressive primary prevention against ischaemia/reperfusion (I/R) injury in diabetic patients. In the present study, lin28a limited IS in diabetic mice after I/R injury. Hemodynamic measurements and echocardiography evaluation revealed preserved LV function by lin28a overexpression. The irreversible mitochondrial dysfunction is a crucial event in cardiomyocyte death following a myocardial I/R injury. After reperfusion, intracellular Ca 2+ elevation and increased ROS production trigger mPTP opening, leading to the depletion of ATP and loss of mitochondrial integrity [21,22]. Alleviating mitochondrial dysfunction may decrease cardiac I/R injury. Meanwhile, modulating mitochondrial function may improve insulin resistance and reduce subsequent cardiac mortality [23,24]. In the present study, lin28a overexpression alleviated mitochondria ultrastructure impairment as demonstrated by intact membrane and cristae although still with a greater size after cardiac I/R in diabetic mice. Lin28a protected against cardiac I/R injury in diabetic mice at least in part by alleviating mitochondrial dysfunction.
Previous studies have shown that a large accumulation of neutrophils was observed in both the permanently occluded and reperfused myocardium, suggesting that an inflammatory response may contributes to apoptosis in both settings [25]. In this study, we found that lin28a overexpression inhibited, while lin28a siRNA administration enhanced inflammatory responses by modulating inflammatory factors production including IL-6 and TNF-a after cardiac I/R injury in diabetic mice.
Cardiomyocytes apoptosis is one of the major pathogenic mechanisms underlying myocardial I/R injury. Blocking the apoptosis process could prevent the loss of contractile cells, minimize cardiac injury induced by I/R injury [26,27]. Thus, we performed TUNEL staining and measured Caspase-3 activity by western blot to explore the underlying mechanism responsible for the cardiac function improvement induced by lin28a overexpression in diabetic mice after cardiac I/R injury. The results indicated that lin28a overexpression inhibited cardiomyocytes apoptosis after I/R injury in diabetic mice as demonstrated by decreased TUNEL-positive cardiomyocytes and decreased caspase-3 expression levels. On the contrary, lin28a siRNA administration increased cardiomyocytes apoptosis, exacerbated cardiac I/R injury in diabetic mice after cardiac I/R injury.
One of the possible mechanisms which has been proposed for the protective effects of lin28a overexpression is blocking let7 production and inhibiting let-7-mediated repression of multiple components of the insulin-PI3K-mTOR pathway, including IGF1R, INSR and IRS2 [13,28]. The lin28a/let-7a axis regulated glucose metabolism in part through the insulin-PI3K-mTOR pathway and the insulin-PI3K-mTOR pathway was involved in the cardioprotection against myocardial ischaemia/reperfusion injury [13,14]. To explore whether the protective effects of lin28a overexpression are associated with insulin-PI3K-mTOR pathway, the specific mTOR inhibitor rapamycin was employed in the present study. Lin28a overexpression inhibited, while lin28a knockdown enhanced let7a expression. Furthermore, lin28a overexpression increased, while lin28a knockdown decreased IGF1R, p-Akt, p-mTOR and p-p70s6k expression after cardiac I/R injury in diabetic mice. Interestingly, pre-treatment with rapamycin abolished the effects of lin28a overexpression as demonstrated by increased IS, MPO activity, caspase-3 expression and increased production of IL-6 and TNF-a, worsened LV function, enhanced cardiomyocytes apoptosis and decreased p-mTOR and p-p70s6k expression. These results suggest that lin28a overexpression induces cardioprotective effects through the activation of insulin-PI3K-mTOR pathway.
In conclusion, the salient finding of the present study is that lin28a overexpression reduces IS and alleviates cardiac dysfunction after cardiac I/R injury in diabetic mice. This is accompanied by decreased cardiac apoptosis and inflammation. The mechanism responsible for the effects of lin28a overexpression is mediated, at least in part, by the insulin-PI3K-mTOR dependent pathway.

Limitations
Other members of RISK pathway may have participated in the link between Lin28a and its cardiac protection effects. The function of cardiomyocytes is closely related to mitochondria biogenesis and mitochondrial dysfunction is the primary cause of I/R-induced apoptosis of cardiomyocytes. To further demonstrate the efficacy of Lin28a administration on cardiomyocyte mitochondrial biogenesis and function, AMPK related pathway should be measured in future studies. The protection effects of Lin28a may link to the metabolic derangement. Further studies are needed to investigate whether Lin28a activates Akt phosphorylation by different mechanisms in diabetic mice or non-diabetic mice underwent cardiac ischaemia insult.

Acknowledgements
This work was supported by National Natural Science Foundation of China (no's. 81270263, 81100579, 81300149).

Conflicts of interest
None declared.

Supporting information
Additional Supporting Information may be found in the online version of this article: Figure S1 Representative gel bolts depicting respective protein expression using specific antibodies (A); Lin28a (B); phosphorylated AKT (p-AKT; C).