Telmisartan cardioprotects from the ischaemic/hypoxic damage through a miR‐1‐dependent pathway

Abstract The aim of this study was to investigate whether telmisartan protects the heart from the ischaemia/reperfusion damage through a local microRNA‐1 modulation. Studies on the myocardial ischaemia/reperfusion injury in vivo and on the cardiomyocyte hypoxia/reoxygenation damage in vitro were done. In vivo, male Sprague‐Dawley rats administered for 3 weeks with telmisartan 12 mg/kg/d by gastric gavage underwent ischaemia/reperfusion of the left descending coronary artery. In these rats, infarct size measurement, ELISA, immunohistochemistry (IHC) and reverse transcriptase real‐time polymerase chain reaction showed that expressions of connexin 43, potassium voltage‐gated channel subfamily Q member 1 and the protein Bcl‐2 were significantly increased by telmisartan in the reperfused myocardium, paralleled by microRNA‐1 down‐regulation. In vitro, the transfection of cardiomyocytes with microRNA‐1 reduced the expressions of connexin 43, potassium voltage‐gated channel subfamily Q member 1 and Bcl‐2 in the cells. Telmisartan (50 µmol/L) 60 minutes before hypoxia/reoxygenation, while not affecting the levels of miR‐1 in transfected cells in normoxic condition, almost abolished the increment of miR‐1 induced by the hypoxia/reoxygenation to transfected cells. All together, telmisartan cardioprotected against the myocardial damage through the microRNA‐1 modulation, and consequent modifications of its downstream target connexin 43, potassium voltage‐gated channel subfamily Q member 1 and Bcl‐2.

Telmisartan, an angiotensin II type 1 (AT1) receptor blocker and partial peroxisome proliferator-activated receptor gamma (PPAR-γ) agonist, has recently shown protective effects in the treatment of hypoxic/ischaemic cardiac damage [15][16][17] related to increased levels of cardiac Cnx43 and reduced apoptosis. 18,19 However, it has not yet been studied whether the effects of telmisartan on Cnx43 have been exerted through modifications of the expression of the cardiac miR-1. Therefore, the first objective of the present study was to study the effects of telmisartan treatment on the cardiac expression of miR-1, Cnx43 and KCNQ1 in a rat model of myocardial ischaemia/reperfusion (I/R) injury. Then, to better define the molecular mechanisms determined by telmisartan, the cardiac expression of miR-1, Cnx43 and KCNQ1 was monitored in hypoxic/reoxygenated embryonic rat ventricle H9c2 cells. This in vitro setting was further used to evaluate the putative effects of telmisartan on the transfected miR-1 levels and thus on both Cnx43 and KCNQ1 expression. Furthermore, the anti-apoptotic protein Bcl-2, a further target of the silencing activity of the miR-1, was monitored. 20  (a) untreated rats subjected to thoracotomy only and used as control (SHAM group); (b) untreated rats subjected to myocardial ischaemia/reperfusion (I/R) injury (I/R group); (c) rats administered for 3 weeks by gastric gavage with vehicle (1% methylcellulose, 4 mL/kg/d) and then subjected to I/R injury (I/R veh group); and (d) rats administered for 3 weeks by gastric gavage with telmisartan 12 mg/kg/d (Boehringer Ingelheim) and then subjected to I/reperfusion injury (I/R Tel group). According to previous evidence, this dose was effective in experimental myocardial ischaemia/reperfusion (I/R) injury. 15,16 At the end of the 3 weeks, rats were anaesthetized with intraperitoneal urethane (1.2 g/kg) and underwent thoracotomy or myocardial injury (I/R), as previously described with modifications. 15 Briefly, in order to permit artificial ventilation when required, the rats were subjected to tracheotomy by using a polythene cannula. Then, the left thoracotomy was performed between the fourth and the fifth ribs. After the pericardium removal, the heart was exteriorized and then a fine silk ligature was placed around the left anterior descending coronary artery (LADCA), close to its origin. After 25 minutes of ischaemia, a 2-hour reperfusion was performed. Rats were kept under artificial ventilation with room air at a rate of 56 strokes/min, a stroke volume of 1.0-1.5 mL/100 g and a positive-end expiratory pressure of 0.5-1 cm H 2 O. Overall mortality was <5% throughout the entire study, in line with the usual procedures in our laboratory. 21

| Measurement of area at risk and infarct size
After the 2-hour reperfusion period, the ratios between the weights of area at risk (AR) and left ventricle (LV) (AR/LV), infarct size (IS) and area at risk (IS/AR) and infarct size and left ventricle (IS/LV) were measured with Evans Blue dye as previously described. 15 Selected experiments (n = 5 for each experimental group) were repeated monitoring AR without the staining procedures to measure IS. After the AR collection, half of each specimen was immediately frozen in liquid nitrogen for biochemical analyses and the other half was fixed by immersion in 10% buffered formalin and paraffin-embedded for IHC. No. BA-1000). An expert pathologist, blinded to the experimental protocol, analysed the specimens (intra-observer variability 6%). The antigen expression was automatically calculated by using Image program LEICA IM500 and LEICA QWIN statistic program. Five distinct preparations for each group of animals were carried out, by analysing in each of them 20 microscope fields, for a total area of 2.4755 e +004 at 400× magnifications. Cat. No. 15862) 400 µmol/L for 6 hours (H/R group) in DMEM 10%. [22][23][24][25] Hypoxia was confirmed by Western blot analysis of HIF-1α protein expression. In order to investigate the effects of telmisartan treatment on hypoxic conditions alone, and not in combination with the serum and glucose deprivation (SGD), following the 6-hour hypoxia the medium with CoCl 2 was replaced by DMEM without CoCl 2 for 2-hour reoxygenation. 26,27 Sixty minutes before the hypoxia/reoxygenation, 23 cells were exposed to a dose of telmisartan (50 µmol/L; H/R Tel group) already reported as effective on H9c2 cells, and to its vehicle dimethyl sulfoxide (DMSO 1% in cell medium; H/R DMSO group). 18 At the end of the hypoxia/reoxygenation period, cell morphology was observed with optic microscopy (Leica DMi1, Germany). According to previous experience, cells were seeded at 5 × 10 3 cells/cm 2 in 96-well plates for the viability assay; at 1 × 10 6 cells/cm 2 in 10 cm cell culture dishes for total RNA and protein detection; and at 1 × 10 4 cells/cm 2 in 24-well plates for immunofluorescence assay. 28 Three independent experiments were performed. In a single experiment, each treatment was repeated three times.

| Cell viability assay
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay was used to measure cell viability. 14 Briefly, the addition of MTT solution (1:10 in culture medium) to each well was followed by a 3hour incubation at 37°C. Once MTT solution was removed, each well was washed for 20 minutes at room temperature with isopropanol-HCl 0.2 N. A 96-well plate reader (iMark; Bio-Rad Laboratories) was used to detect the optical density (OD) values at 570 nm.

| Protein isolation and quantization
Heart tissues were homogenized in RIPA lysis buffer (Sigma; Cat. No. R0278) including a protease inhibitor cocktail (Roche; Cat. No. 1187358000) and centrifuged at 13 800 g for 10 minutes at 4°C, to F I G U R E 1 Effects of telmisartan on myocardial damage after I/R injury. A, AR/LV, IS/AR and IS/LV ratios and (B) cTnI levels in the different experimental groups. SHAM, thoracotomy only; I/R, ischaemia/reperfusion; veh, 1% methylcellulose 4 mL/kg/d; Tel, telmisartan 12 mg/kg; AR, area at risk; LV, left ventricle; IS, infarct size. Results are expressed as mean ± SEM of five observations. *P < .05 vs SHAM; °P < .05 and °°P < .01 vs I/R veh purify the protein supernatants from nucleic acids. 14 Total protein concentration was determined following the Bio-Rad protein assay protocol (Bio-Rad Laboratories; Cat. No. 500-0006) and used for Western blotting and ELISAs.

| Western blotting assay
Thirty microgram of protein sample was separated on a 12% separation gel, electrotransferred onto a polyvinylidenedifluoride

| ELISAs
Troponin I as marker of myocardial infarction and Bcl-2 protein levels were quantified by using the Rat Cardiac Troponin I (cTnI) ELISA Kit (MyBioSource; Cat. No. MBS727624) and the Rat B-cell CLL/ lymphoma 2, Bcl-2 ELISA Kit (MyBioSource; Cat. No. MBS704498), according to the manufacturer's protocols.

| Statistical analysis
The results are presented as mean ± standard error of the mean (SEM) of five observations per group in vivo, and as mean ± SEM of nine observations in vitro. Statistical significance was determined using ANOVA followed by Bonferroni's test. A P-value <.05 was considered significant to reject the null hypothesis.

| Myocardial tissue damage after telmisartan administration
Untreated Sprague-Dawley rats subjected to I/R procedure (I/R group) exhibited an AR equal to the 63% of the LV. The IS was 65% of the AR and 40% of the LV ( Figure 1A). The IS/AR values were not affected by the administration of vehicle (I/R veh) but were significantly reduced following treatment with telmisartan 12 mg/kg (I/R Tel; −38%, P < .01 vs I/R veh), as well as IS/LV ratios (−25%, P < .01 vs I/R veh; Figure 1A). Cardiac troponin I (cTnI) levels were significantly increased by I/R procedure (+67.9%, P < .05 vs SHAM) and significantly reduced in rats receiving telmisartan 12 mg/kg only (−31%, P < .05 vs I/R veh; Figure 1B).

| Effects of telmisartan on Cnx 43, KCNQ1 and Bcl-2 expression in I/R hearts
Following the I/R procedure, the IHC showed a significant reduction in Cnx43 and KCNQ1 expressions in I/R rats compared with SHAM group (respectively, −61.3% and −50%, P < .01 vs SHAM; Figure 2A,B). These were not significantly modified in I/R veh rats compared with I/R group (respectively, +6.9% and −5%), whereas I/R Tel group exhibited a consistent increase in both Cnx 43 and KCNQ1 levels compared with I/R group (respectively, +89.7% and +70%, P < .01 vs I/R veh; Figure 2A,B). The same trend was shown by Bcl-2 levels, significantly decreased in I/R group (−70.6%, P < .01 vs SHAM; Figure 2C). These were markedly increased by telmisartan treatment +140% P < .05 vs I/R veh) and not by its vehicle (−4% vs I/R veh; Figure 2C).

| miR-1 modulation by telmisartan in I/R hearts
qRT-PCR analysis showed a significant increase in miR-1 levels in rat hearts subjected to I/R procedure compared with SHAM rats (+100%, P < .05 vs SHAM; Figure 3A). Infarcted hearts exhibited an evident reduction in miR-1 levels when administered with telmisartan 12 mg/kg (−36.4%, P < .05 vs I/R) and not with 1% methylcellulose vehicle (−2.3% vs I/R; Figure 3A). This miR-1 down-regulation was paralleled by an increased expression of Cnx43, KCNQ1 and Bcl-2 in I/R Tel group ( Figure 3B).

| Cnx 43, KCNQ1 and Bcl-2 expression in H/R H9c2 cardiomyocytes
In contrast, both the vehicle DMSO and telmisartan at the doses used did not affect the factors monitored in normoxic H9c2 cells (data not shown).

| D ISCUSS I ON
Here, we report that telmisartan protected the infarcted rat heart by locally increasing two important players in hypoxia-induced cell survival, 29 Cnx43 and KCNQ1 potassium channel. In contrast, telmisartan reduced miR-1 expression within the infarcted heart along with increased expression of the known apoptotic marker 20,30 B-cell lymphoma 2 (Bcl-2). This action resulted in a lower degree of tissue damage. Telmisartan also copied this effect on embryonic rat ventricle H9c2 cells exposed to a hypoxia/reoxygenation procedure.
The cardioprotection afforded by telmisartan has been widely reported in myocardial I/R injury [15][16][17][18][19]31 ; however, the present study for the first time sheds light on the involvement of the potassium channel KCNQ1 and Cnx43 with respect to telmisartan action. As it modulates cardiac cell susceptibility to hypoxia 32,33 and regulates the impulse propagation and electrical synchronization between cardiomyocytes, 34,35 Cnx43 seems to have a key role in the setting of functional I/R damage. Similarly, recent evidence reported that Cnx43 together with KCNQ1 is part of the same signalling pathway that provides cell protection after myocardial infarction by acting in a functionally dependent manner. 3,29,36 Indeed, mutations in both genes encoding for KCNQ1 and Cnx43 have been associated with clinical sudden infant death syndrome (SIDS), for which hypoxia is a major risk factor. 37,38 The data presented, therefore, identified telmisartan as a regulatory candidate of these two important factors in the infarcted myocardium, Cnx43 and KCNQ1. Moreover, they highlighted that telmisartan cardioprotection is exerted by reducing the apoptosis within the myocardium relating to tissue protection. Indeed, apoptosis of cardiomyocytes, associated with quantitative disorders of proteins, aggravates myocardial I/R injury. 39 It should also be noted that telmisartan decreased the resident miR-1, which controls both the cardiac structure and the functionality. miR-1 predominantly regulates the electrical and contractile activity of the heart, by modulating atrioventricular and ventricular conduction at multiple levels. [40][41][42] Alterations of its expression levels result in atrioventricular block, impaired contractile function and increase in ROS levels following cardiac I/R injury. 9,[43][44][45] This microRNA targets both Cnx43 and KCNQ1, 11,13,14,40,41 and, when down-regulated, improves KCNQ1 expression in H9c2 cells, 14 in line with the results achieved here. Again, this is the first time that a study reports miR-1 involvement in a telmisartan-induced cardioprotection. One F I G U R E 7 Telmisartan effects on hypoxic H9c2 cardiomyocytes transfected with miR-1 mimic. A, Representative immunofluorescence images, showing Cnx3 and KCNQ1 levels in miR-1-transfected cells exposed to hypoxia/reperfusion. Cell nuclei are labelled in blue with Hoechst, whereas cells positive to Cnx43 or KCNQ1 antibodies are labelled in green. B, Bar graph showing the percentage of Cnx or KCNQ1-positive cells/total counted cells. C, Bcl-2 protein levels in miR-1-transfected H/R cardiomyocytes, detected by ELISA. mim1-, normoxic cells transfected with negative control mimic; mim1, normoxic cells transfected with miR-1 mimic 5 nmol/L; H/R mim1, cells transfected with miR-1 mimic 5 nmol/L and exposed to hypoxia/reoxygenation; H/R mim1 DMSO, H/R mim1 cells exposed to DMSO 1%; H/R mim1 Tel, H/R mim1 cells exposed to telmisartan 50 µmol/L. Results are expressed as mean ± SEM of nine observations. # P < .05 vs mim1-; ## P < 0.01 vs mim1-;^P < .05 vs mim1; § § P < .01 vs H/R mim1 DMSO. Scale bar = 10 µmol/L; 20× magnification hypothesis that could be formulated for this phenomenon is that the miR-1 down-regulation could be linked to the blocking activity of telmisartan on the AT1 receptors, which notoriously stimulate the expression of cellular miRNAs under hypoxia conditions. 46,47 Induced by hypoxia, the AT1 receptors promote miRNA expression through mechanisms depending on Gaq/11 and Erk1/2 activation and by blocking AT1 receptors, telmisartan could impair the activation of Gaq/11 and Erk1/2 G-protein-dependent signalling, leading to a reduced miR-1 expression. 46 and useful in vitro model for exploring the mechanisms driven by hypoxia/reoxygenation. [23][24][25]54,55 These lead to the hydroxylase enzyme inactivation, hypoxia-inducible factor HIF-1α stabilization, ROS generation, apoptosis and increased Bax/Bcl-2 ratio followed by activation of caspase-3 cleavage. 55,56 The data obtained in the present study on H9c2 cells confirmed the reduction in Cnx43, KCNQ1 and Bcl-2 expression following exposure to the hypoxia-mimetic agent, while treatment with telmisartan restored them. This in line with other experiences showing that angiotensin II (ATII) exerts an inhibitory effect on Cnx43, KCNQ1 and Bcl-2 expression, and PPAR-γ agonism is effective in preventing the reduction in Cnx43 induced by ATII. [57][58][59] Interestingly, the telmisartan-activated response on Cnx43, KCNQ1 and Bcl-2 exerted through a down-regulation of miR-1 was confirmed by mimic transfection of miR-1. Thus suggesting in vivo crosstalk between Cnx43, KCNQ1, Bcl-2 and miR-1, 3 one would argue that the properties exerted by telmisartan in hypoxic/ischaemic cardiac damage are exerted through miR-1 and consequent modifications of Cnx43, potassium channel KCNQ1 and Bcl-2 expression in the cardiomyocytes.

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

AUTH O R CO NTR I B UTI O N
MCT performed the research, FB performed immunofluorescence, AM analysed the data, IP performed the immunohistochemical analysis, EG and GP performed the surgical procedure, and GFN and MD designed the research and wrote the study.

DATA ACCE SS I B I LIT Y
The authors confirm that the data supporting the findings of this study are available within the article.