Icariin protects cardiomyocytes against ischaemia/reperfusion injury by attenuating sirtuin 1‐dependent mitochondrial oxidative damage

Abstract

Background and Purpose

Icariin, a major active ingredient in traditional Chinese medicines, is attracting increasing attention because of its unique pharmacological effects against ischaemic heart disease. The histone deacetylase, sirtuin‐1, plays a protective role in ischaemia/reperfusion (I/R) injury, and this study was designed to investigate the protective role of icariin in models of cardiac I/R injury and to elucidate the potential involvement of sirtuin‐1.

Experimental Approach

I/R injury was simulated in vivo (mouse hearts), ex vivo (isolated rat hearts) and in vitro (neonatal rat cardiomyocytes and H9c2 cells). Prior to I/R injury, animals or cells were exposed to icariin, with or without inhibitors of sirtuin‐1 (sirtinol and SIRT1 siRNA).

Key Results

In vivo and in vitro, icariin given before I/R significantly improved post‐I/R heart contraction and limited the infarct size and leakage of creatine kinase‐MB and LDH from the damaged myocardium. Icariin also attenuated I/R‐induced mitochondrial oxidative damage, decreasing malondialdehyde content and increasing superoxide dismutase activity and expression of Mn‐superoxide dismutase. Icariin significantly improved mitochondrial membrane homeostasis by increasing mitochondrial membrane potential and cytochrome C stabilization, which further inhibited cell apoptosis. Sirtuin‐1 was significantly up‐regulated in hearts treated with icariin, whereas Ac‐FOXO1 was simultaneously down‐regulated. Importantly, sirtinol and SIRT1 siRNA either blocked icariin‐induced cardioprotection or disrupted icariin‐mediated mitochondrial homeostasis.

Conclusions and Implications

Pretreatment with icariin protected cardiomyocytes from I/R‐induced oxidative stress through activation of sirtuin‐1 /FOXO1 signalling.

Abbreviations

AAR

area at risk

Ac‐FOXO1

acetylated FOXO1

Bax

Bcl‐2‐associated X protein

C‐caspase‐3

cleaved caspase‐3

CK‐MB

MB isoenzyme of creatine kinase

Cyto CC

cytoplasmic cytochrome C

EF

ejection fraction

FOXO

forkhead box O

FS

fractional shortening

HR

heart rate

I/R

ischaemia reperfusion

IF

infarct size

KHB

Krebs–Henseleit buffer

LVDP

left ventricular developed pressure

MDA

malondialdehyde

Mito CC

mitochondrial cytochrome C

MnSOD

manganese SOD

NRVMs

neonatal rat ventricular myocytes

PI

propidium iodide

SI/R

simulated ischaemia/reperfusion

TTC

triphenyl tetrazolium chloride

Introduction

Heart ischaemia/reperfusion (I/R) injury is a pathological condition characterized by an initial restriction in blood supply that is followed by restored perfusion and concomitant reoxygenation, which eventually leads to acute myocardial infarction or heart failure (Eltzschig and Eckle, 2011). Previous evidence has indicated that a period of ischaemia primes the tissue for the subsequent reperfusion damage upon reperfusion and that a certain duration of ischaemia is required for reperfusion injury in the heart (Bernardi et al., 2015). Under normal conditions, mitochondria are the powerhouse of the living cell, producing most of the cell's energy by oxidative phosphorylation (Chen and Zweier, 2014). However, in ischaemic cardiomyopathy, mitochondria trigger apoptosis by releasing cytochrome C and activating caspases that mediate apoptosis and by altering cellular redox potential via ROS production (Wallace, 2005; Luo et al., 2016). For this reason, mitochondria appear to be a key factor in cardiomyocyte death and protecting mitochondria from oxidative stress and maintaining mitochondrial function may provide a means of ameliorating myocardial I/R injury (Murphy and Steenbergen, 2008). Despite major progress in the treatment of myocardial I/R injury during the last few decades, the therapeutic effect of these treatments is unsatisfactory; therefore, discovering new drugs and therapeutic targets remains an urgent priority.

Traditional Chinese medicine has been practised in Chinese clinics for approximately 2000 years, providing an extensive source of major pharmacological ingredients, such as Ren Shen, Huang Qi, Dan Shen and Epimedium (Han et al., 2008; Wu et al., 2011; Wei et al., 2014). Earlier work has shown that extracts of plant species of Epimedium, alone or in combination with other herbs, have a wide range of beneficial biological effects, including sexual enhancement, anti‐inflammatory, antioxidant, antidepressant, immunity improving and neuroprotective activities (Meng et al., 2016; Zhang et al., 2005; Zhang et al., 2007; Liang et al., 2012). Icariin, a https://en.wikipedia.org/wiki/Prenylflavonoid https://en.wikipedia.org/wiki/Flavonol https://en.wikipedia.org/wiki/Glycoside related to kaempferol (Figure S1A), is the major active ingredient in extracts of Epimedium, forming up to 6.5% of the dry weight (Xu et al., 2010; Schluesener and Schluesener, 2014). Icariin was neuroprotective in ischaemic stroke in rats by inhibiting the inflammatory responses mediated by NF‐κB and PPARs (Xiong et al., 2016). Furthermore, icariin acts as a potential agent for the prevention of cardiac I/R injury in both isolated and in vivo rat hearts via the http://www.guidetopharmacology.org/GRAC/FamilyDisplayForward?familyId=673‐http://www.guidetopharmacology.org/GRAC/FamilyDisplayForward?familyId=285 pathway (Ke et al., 2015; Meng et al., 2015; Zhai et al., 2015). These protective effects of icariin are ascribed to its strong activity against oxidative stress and mitochondrial dysfunction (Ke et al., 2015; Meng et al., 2015; Zhai et al., 2015; Xiong et al., 2016). However, considering the extensive pharmacological effects of icariin and the complex pathogenesis of ischaemic cardiomyopathy, the cardioprotective effect of icariin still needs further study.

The sirtuins are a highly conserved family of histone/protein deacetylases that can prolong lifespan in organisms (Nogueiras et al., 2012). Among the seven sirtuins expressed in mammals, http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=2707 is localized predominantly in the nuclei or cytoplasm, depending on the cell type, and this protein regulates a wide array of cellular processes especially in cardiomyocytes (Sundaresan et al., 2011). In ischaemic cardiomyopathy, these pathophysiological functions of SIRT1 are mainly mediated by the deacetylation of histones, transcription factors and coactivators, such as p53, forkhead box O (FOXO) family members, NF‐κB, http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=595 coactivator 1‐α and liver X receptor (Li et al., 2007; Hariharan et al., 2010; Hsu et al., 2010; Li et al., 2011; Planavila et al., 2011; Zhang et al., 2011). As it possesses a polyphenolic structure that is similar to that of http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=8741 (Figure S1B), icariin may activate SIRT1 through similar mechanisms (Park et al., 2012; Verdin, 2015). Recent studies have suggested that the benefits of icariin may be due to up‐regulation of SIRT1, which contributes to FOXO3 deacetylation and modulation of downstream antioxidative and anti‐apoptotic factors against lung and intestinal I/R injury (Zhang et al., 2015a; Zhang et al., 2015b). Moreover, by regulating SIRT1 expression, icariin activates the MAPK pathway to promote neuroprotection in ischaemia‐related brain injury (Wang et al., 2009). However, the ability of SIRT1 to mediate the icariin‐induced mitigation of myocardial and cardiac mitochondrial injury following I/R injury has not been explored.

The objective of the current study was to investigate the effect of icariin on I/R‐induced myocardial injury in vitro and in isolated and in vivo hearts. We hypothesized that icariin treatment would prevent I/R‐induced heart injury through its influence on mitochondrial oxidative damage and cell apoptosis via the SIRT1 pathway. To test this hypothesis, low and high concentrations of icariin were administered to isolated and in vivo hearts and to cardiomyocytes. The infarct size (IF), echocardiography indexes, myocardial necrosis, apoptosis indicators and mitochondrial oxidative damage were assessed. In addition, the expression levels of SIRT1 and apoptosis pathway proteins were evaluated by Western blot analysis.

Methods

Animals

All animal care and experimental procedures in the present study were conducted in accordance with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (National Institutes of Health Publication No. 85‐23, revised in 1996), and approval was obtained from the Ethics Committee of the Fourth Military Medical University. All efforts were made to minimize the animals' suffering and to reduce the number of animals used. Animal studies are reported in compliance with the ARRIVE guidelines (Kilkenny et al., 2010; McGrath and Lilley, 2015). All animal models are commonly used in cardiovascular disease research (Rakhit et al., 2001; Bendale et al., 2013; Matsuda et al., 2016).

Neonatal (1–2 day old) and adult (220–250 g) Sprague–Dawley (SD) rats and adult male C57BL/6 mice (25–30 g) were obtained from the animal centre of the Fourth Military Medical University. The adult rats and mice were housed in an Association for Assessment and Accreditation of Laboratory Animal Care‐accredited facility in an environmentally controlled room (12 h light–dark cycle, 60% humidity) at 20–23°C and were allowed ad libitum access to standard chow and sterile water. Body weight was monitored throughout the experiment.

Culture of immortalized rat myoblast H9c2 cells and preparation of primary neonatal rat ventricular myocytes (NRVMs)

In a standard CO₂ incubator (95% air, 5% CO₂), immortalized rat myoblast H9c2 cells (ATCC, Manassas, VA, USA) were cultured in growth medium, which included DMEM, 10% FBS and antibiotics (100 units·mL⁻¹ of penicillin and 100 μg·mL⁻¹ of streptomycin) as previously described (Yu et al., 2017).

NRVMs were isolated from 1‐ to 2‐day‐old SD rats by serial enzymic digestion as described previously (Qin et al., 2017). After cervical dislocation of the pups, the hearts were removed and were placed in ice‐cold Hank's balanced salt solution. After rinsing, the ventricular tissue was chopped and digested with digestion solution containing trypsin and collagenase II. To reduce fibroblast contamination to <5% of the total cell population, pellets were suspended in growth medium and were plated onto six‐well plates (1 × 10⁶ cells/well) at 37°C for 25 min (pre‐plating step). Then, the cells in the supernatant were re‐plated into fresh 24‐ or 6‐well plates and were incubated in growth medium at 37°C in a standard CO₂ incubator. Under these conditions, more than 80% of the cells spontaneously beat for the duration of the remaining experiments, which were performed between days 2 and 3 after cell isolation.

Experimental settings

First, icariin or resveratrol mediated regulation of SIRT1 was evaluated in primary NRVMs for 12 h (Figure S1C). The icariin (2, 4 and 8 μM) and resveratrol (8 μM) concentrations were selected according to previous studies (Zhu et al., 2010; Becatti et al., 2012). Second, as shown in Figure S1D, the following five experimental groups were created. (i) For the control group, the NRVMs were treated with growth medium for 12 h and then cultured in serum‐free (SF) medium until the end of the testing. (ii) In the sirtinol group, cells were preconditioned with sirtinol (a widely used SIRT1 inhibitor, 25 μM) for 2 h followed by SF medium until the end of the test. (iii) For the simulated I/R (SI/R) group, the cardiomyocytes were stimulated with hypoxia for 2 h and then with reoxygenation for 4 h. (iv) In the SI/R + ICA group, cells were preconditioned with icariin (4 μM) for 2 h before the SI/R injury. (v) The SI/R + ICA + sirtinol group was preconditioned with icariin (4 μM) and sirtinol (25 μM) for 2 h prior to SI/R injury. In Figure S1E, an additional four groups are detailed. (i) For the SI/R + Con siRNA group, control siRNA transfection was conducted in H9c2 cells 48 h prior to SI/R injury. (ii) To silence SIRT1 in the SI/R + SIRT1 siRNA group, siRNA specifically targeted to rat SIRT1 was transfected into the cells 48 h prior to SI/R injury. (iii) For the third supplemental group, the SI/R + Con siRNA + ICA group, cells were transfected with control siRNA and then were preconditioned with icariin (4 μM) for 2 h before the SI/R injury. (iv) In the SI/R + SIRT1 siRNA + ICA group, after the cells were transfected with SIRT1 siRNA, they were preconditioned with icariin (4 μM) for 2 h before the SI/R injury.

To determine whether icariin induces SIRT1 activation, hearts were isolated from normal rats and randomly assigned to one of two groups (Figure S1F): the control group or the icariin group. The Control hearts were treated with modified Krebs–Henseleit buffer (KHB). For the icariin hearts, icariin was added to the modified KHB for 10 min, and the final icariin concentrations, as determined by HPLC analysis, were 10 μM. The contribution of icariin‐induced SIRT1 activation and the involvement of mitochondrial oxidative stress injuries in functional cardiac recovery and reduced IF were studied in the following experimental groups (Figure S1F): the I/R, I/R + ICA and I/R + ICA + sirtinol groups. To establish the I/R injury model in the I/R group, the isolated hearts were subjected to 45 min of ischaemia, followed by 60 min of reperfusion. Icariin and sirtinol (4 μM) preconditioning for 10 min was used in the I/R + ICA + sirtinol group.

Furthermore, the cardioprotective effect of SIRT1 activation induced by icariin was investigated in vivo (Figure S1G). The control group was treated with a 0.05% DMSO/PBS solution for 14 days following the sham operation. Icariin (60 mg·kg⁻¹) was dissolved in a 0.05% DMSO solution and administered to the icariin group by gavage twice a day for 2 weeks following the sham operation. In the I/R group, the left coronary artery was ligated for 0.5 h and was reperfused for 24 h. For the I/R + ICA group, icariin was orally administered following the I/R operation. For the I/R + ICA + sirtinol group, the icariin treatment lasted 2 weeks, and then sirtinol (1 mg·kg⁻¹) was dissolved in a 0.05% DMSO solution and administered i.p. for 1 week.

I/R surgery in vivo

I/R surgery was performed according to a previously described new method for inducing myocardial I/R injury without a ventilator (Gao et al., 2010). During this procedure of I/R, the C57BL/6 mice were maintained under anaesthesia with inhalation of 2% isoflurane using a delivery system (Viking Medical, Medford, NJ, USA) but not ventilation to ensure that the experimental subjects suffered minimal discomfort. The model includes a small skin incision made over the left front chest wall, and a purse string suture, reserved for later use. After this, a small hole was made with a mosquito clamp through the fourth intercostal space, and the heart was smoothly and gently extracted. Then, the left main descending coronary artery was ligated with a 6–0 silk slipknot, followed by the heart being placed back into the body and the skin being sutured. After 0.5 h of ischaemia, the slipknot was released to initiate myocardial reperfusion, which lasted for an additional 24 h. During the period of reperfusion, the mice were allowed to breathe room air and monitored until the recovery period, which was generally completed within 3–5 min. None of the mice exhibited dyspnoea, struggling and other symptoms of hypoxia. After I/R injury, the mice were immediately re‐anesthetized with 10 mg·mL⁻¹ of pentobarbital sodium (0.1 mL/20 mg) to ensure minimal discomfort. Then, 1% Evans Blue was injected apically with the heart beating to determine the area at risk (AAR). The entire procedure took less than 40 min. At this time, the mice were killed with minimal discomfort because of maintenance of anaesthesia. Next, the hearts were collected and 1 mm sections of the hearts were stained with 1% triphenyl tetrazolium chloride (TTC) to measure the IF, as previously described. The AAR, IF and left ventricle area (LV) of the digital heart section images were analysed using ImageJ software (National Institutes of Health, Bethesda, MD, USA) (Jin et al., 2017). The percentages of the IF and AAR areas of each section were multiplied by the weight of the section and then totalled from all sections. The AAR/LV and IF/AAR were expressed as percentages (Matsuda et al., 2016).

Echocardiography

All echocardiography measurements in mice with myocardial ischaemia were performed using the Vevo®2100 high‐resolution in vivo imaging system (Visual Sonics, Toronto, Canada) equipped with a high frequency transducer (MS400, 18–38 MHz). All the procedures followed the official open access user manual (FUJIFILM VisualSonics, Inc., 2014). The mice were anesthetized with 2% isoflurane until they were sedated, and then, anaesthesia was maintained using 1% isoflurane during the M‐mode echocardiographic measurement. The fractional shortening (FS) and the ejection fraction (EF) were surveyed during five consecutive cardiac cycles.

Preparation of isolated perfused hearts as an ex vivo model

Male SD rats were anesthetized with sodium pentobarbital (45 mg·kg⁻¹, i.p.). The heart was rapidly excised and retrogradely perfused through the aorta by a Langendorff perfusion system with oxygenated KHB (Ma et al., 2015), containing 120 mM NaCl, 4.7 mM KCl, 1.25 mM CaCl₂, 25 mM NaHCO₃, 1.2 mM KH₂PO₄, 1.2 mM MgSO₄ and 11 mM glucose (pH 7.4), with a constant pressure of 80 mm Hg at pH 7.4 bubbled with 95% O₂/5% CO₂ at 37°C (Wang et al., 2014a). To establish the I/R injury model, isolated, perfused rat hearts were subjected to 40 min of ischaemia followed by 60 min of reperfusion. LV pressure was monitored using a water‐filled latex balloon connected to a pressure transducer (Model 100 BP‐Biopac System Inc., Goleta, CA, USA) and inserted into the LV cavity. The heart rate (HR), the left ventricular developed pressure (LVDP), the ventricular pressure generation (+dP/dt max) and the ventricular pressure relaxation (−dP/dt min) were collected. At the end of the reperfusion, the coronary effluent was rapidly frozen at −80°C for further analysis. Rat hearts were immediately stored at −20°C for 1 h. Then, the hearts were incubated in 1% TTC solution at 37°C for 15 min, followed by immersion in 4% paraformaldehyde solution for 30 min. Five cross‐sectional slices were imaged with a digital camera, and the percentage of the infarct area (white area) to the total ventricular area was calculated using ImageJ software (National Institutes of Health) (Li et al., 2013).

Simulated I/R (SI/R) injury in vitro

Two‐day‐old NRVMs and H9c2 cells were subjected to the SI/R injury protocol as described earlier (Luo et al., 2016). During the ischaemic simulation, the cells were covered with a modified hypoxic solution (119 mM NaCl, 5.4 mM KCl, 0.5 mM MgCl₂, 0.9 mM CaCl·2H₂O, 4 mM HEPES, 10 mM deoxyglucose and 20 mM sodium lactate; pH 6.2) and incubated in a hypoxic chamber (STEMCELL Technologies Inc., Vancouver, BC, Canada) with a humidified atmosphere of 5% CO₂ and 95% nitrogen at 37°C for 2 h. Then, the cells were transferred into a normoxic incubator and were covered with SF medium for 4 h to simulate reperfusion. At the end of reperfusion, cell viability was assessed using the CCK‐8 assay at 450 nm with a microplate reader (SpectraMax 190, Molecular Device, Sunnyvale, CA, USA) (Obis et al., 2014).

SIRT1 knockdown using siRNA in H9C2 cells

The SIRT1 and negative control siRNA sequences were as follows: SIRT1, sense 5′‐CACCUGAGUUGGAUGAUAUTT‐3′ and antisense 5′‐AUAUCAUCCAACUCAGGUGTT‐3′; and negative control, sense 5′‐UUCUCCGAACGUGUCACGUTT‐3′ and antisense 5′‐ACGUGACACGUUCGGAGAATT‐3′ (Wang et al., 2014a, 2014b). After H9c2 cells had grown to 80% confluency on six‐well plates, transfection was accomplished by using Lipofectamine RNAiMAX according to the manufacturer's instructions. Forty‐eight hours later, the cells were harvested for further experiments.

Determination of cardiomyocyte apoptosis

A transferase dUTP nick‐end labelling (TUNEL) staining kit was used to detect the rate of apoptosis after different treatments. Briefly, NRVMs and H9c2 cells were embedded in 4% paraformaldehyde and permeabilized with 0.1% Triton X‐100. Together with formalin‐fixed heart tissue sections from the peri‐infarct area, the cells were incubated with 50 μL of TUNEL reagent for 1.5 h at 37°C. DAPI staining was used to assess the number of nuclei. The TUNEL signal was observed with an FV1000 Olympus confocal microscope (Olympus, Tokyo, Japan). Cell apoptosis was determined as the ratio of the number of TUNEL‐positive nuclei to that of DAPI‐positive nuclei.

Furthermore, annexin V (Ann V)/propidium iodide (PI) staining was performed using a commercial kit (Kempf et al., 2006). The membrane phospholipid phosphatidylserine (PS) is translocated from the inner leaflet of the plasma membrane to the outer leaflet in apoptotic cells, thereby exposing PS to the external environment. Ann V has a high affinity for PS, which is useful for the identification of apoptotic cells with exposed PS. PI is a standard flow cytometric viability probe that is used to distinguish viable from nonviable cells (Cserepes et al., 2007). NRVMs (1.5 × 10⁶) were seeded into six‐well plates for 12 h. Following different treatments, cells of each group were collected, washed twice with PBS and resuspended at a density of 1 × 10⁶ cells·mL⁻¹. Next, a 500 μL cell suspension was incubated with 5 μL of annexin V‐FITC for 15 min in the dark at room temperature, and 10 μL of PI was added 5 min before the analysis. The fluorescence intensities were analysed within 1 h by flow cytometry using the CellQuest software (BD Biosciences, San Jose, CA, USA). Moreover, cell fluorography was performed on slides with an FV1000 Olympus confocal microscope (Olympus). This assay discriminates between intact (FITC⁻/PI⁻), early apoptotic (FITC⁺/PI⁻) and late apoptotic (FITC⁺/PI⁺) cells (Guo et al., 2011).

Myocardial necrosis analysis

The cell death markers LDH and the MB isoenzyme of creatine kinase (CK‐MB) were assessed spectrophotometrically in the collected serum, isolated heart perfusates and cell culture supernatants using commercial assay kits according to the provided instructions (Ma et al., 2015).

Moreover, H9c2 cells were stained with Trypan blue to assess the severity of membrane integrity impairment. Briefly, Trypan blue was dissolved in PBS and stored at 4°C. After different treatments, the cells were suspended in 0.25% trypsin for 3 to 5 min and centrifuged at 4°C for 15 min. Then, a 45 μL cell suspension and 5 μL of Trypan blue were mixed and incubated for 5 min (the final Trypan blue solution concentration was 0.04%). The total number of cells and the number of cells stained with Trypan blue (presumed to be nonviable) were counted in a 1 mm² area in a haemocytometer (Qiujing, Shanghai, China), and cell death was determined as the percentage of Trypan blue‐positive cells. Additionally, the same concentration of Trypan blue was used to evaluate dead cells in six‐well plates without trypsin treatment. The morphologies were obtained under an inverted/phase contrast microscope, and images were taken using a 600D camera (Canon Company, Japan).

Malondialdehyde (MDA) content and SOD activity assays

After treatment, heart tissues or cardiomyocytes were harvested and lysed by using a homogenizer. The MDA content in the lysates was detected by measuring levels of thiobarbital‐reactive substances, as previously reported, with a lipid peroxidation assay kit, and the results were expressed as nmol (mg protein)^‐1 (Wang et al., 2016). SOD enzyme activity was evaluated following the reagent manufacturer's instructions (Hu et al., 2018).

Measurement of mitochondrial membrane potential

The mitochondrial membrane potential of NRVMs and H9c2 cells was analysed by staining with the cationic JC‐1 dye according to the method described previously (Yang et al., 2017). In cells with depolarized mitochondria, JC‐1 predominantly exists in a monomeric form and displays green fluorescence; but in polarized mitochondria, these monomers aggregate and appear as red fluorescence. The fluorescence intensity was detected with an FV1000 Olympus confocal microscope (Olympus). After staining, red fluorescence was excited at 550 nm and detected at 600 nm, and green fluorescence was excited at 485 nm and detected at 535 nm. Mitochondrial membrane potential was represented by the ratio of red to green fluorescence intensity.

Detection of mitochondrial superoxide production

Mitochondrial ROS (Mito‐ROS) in the NRVMs were detected with fluorescent MitoSOX Red, which is a cell‐permeable probe that accumulates specifically in the inner mitochondrial compartment (Becatti et al., 2012). MitoSOX was dissolved in DMSO immediately before use and then applied to NRVMs on glass cover slips at a final concentration of 5 μM. After 20 min of staining, the slips were washed twice in PBS. DAPI staining was used to assess the positions of the nuclei after staining for 10 min at 37°C. Finally, the red fluorescence intensity was measured under an FV1000 Olympus confocal microscope (Olympus) at excitation/emission wavelengths of 510 nm/580 nm (Song et al., 2018).

Western blotting analysis

Protein lysates were prepared from heart tissues and cardiomyocytes after different treatments. The nuclear and cytoplasmic fractions were then isolated using a cytoplasm extraction kit (Beyotime Institute of Biotechnology, Haimen, China) according to the manufacturer's instructions. The primary antibody concentrations used in this analysis are as follows (details in Table S1): SIRT1 (1:500), acetylated FOXO1 (Ac‐FOXO1, 1:500), manganese SOD (MnSOD, 1:500), http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=2844 (1:500), http://www.guidetopharmacology.org/GRAC/FamilyDisplayForward?familyId=910 (1:500), cytochrome C (1:500), cytochrome C oxidase IV (1:1000), http://www.guidetopharmacology.org/GRAC/FamilyDisplayForward?familyId=848#2712 (1:1000), http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=1619 (1:1000), cleaved caspase‐3 (C‐caspase‐3, 1:1000), and β‐actin (1:1000). Secondary antibodies were purchased from the Zhongshan Company (Beijing, China). Finally, the immunoreactive signals were visualized using an electrochemiluminescence system (Bio‐Rad, West Berkeley, CA, USA), and the relative intensity of the bands was quantified using Image Lab software (Bio‐Rad).

Data and statistical analysis

The data and statistical analysis comply with the recommendations on experimental design and analysis in pharmacology (Curtis et al., 2018). A minimums of five or eight independent samples were used in the cell or animal experiments respectively, and every analysis was repeated three times, unless otherwise stated. The distribution normality was assessed with the GraphPad Prism 6.01 software (Paragraph Software, Inc., La Jolla, CA, USA). The results are expressed as the means ± SD. The data were analysed by operators blinded to the treatment groups. Multi‐group comparisons were performed using one‐way ANOVA, followed by Bonferroni's post hoc test. A P value < 0.05 was used as an indicator of statistically significant differences.

Materials

Icariin (>94% purity), resveratrol (>99% purity), sirtinol, trypsin, TTC, Evans Blue, Trypan blue, DMSO, DAPI, and collagenase were obtained from Sigma‐Aldrich (St. Louis, MO, USA). DMEM and FBS were purchased from Gibco Laboratories (Life Technologies, Inc., Burlington, ON, Canada). Lipofectamine RNAiMAX, penicillin, and streptomycin were purchased from Invitrogen (Carlsbad, CA, USA). Cell Counting Kit‐8 (CCK‐8) was purchased from Dojindo Laboratories (Tokyo, Japan). SIRT1 and the negative control siRNAs were from Shanghai GenePharma Co., Ltd. (Shanghai, China). The TUNEL staining kit used for apoptosis detection was purchased from Roche Molecular Biochemicals (Mannheim, Germany). A FITC annexin V apoptosis detection kit was obtained from BD Biosciences (San Jose, CA, USA). 5,5′,6,6′‐tetraethyl‐benzimidazolylcarbocyanine iodide (JC‐1) and MitoSOX Red were purchased from Yisheng Biotechnology Co., LTD. (Shanghai, China). The commercial assay kits for the LDH, CK‐MB, MDA, and SOD analyses were from Jiancheng Bioengineering (Nanjing, Jiangsu, China). The antibodies against SIRT1, Ac‐FOXO1, MnSOD, Bcl2, Bax, cytochrome C, and β‐actin were from Santa Cruz Biotechnology (Santa Cruz, CA, USA), whereas the antibodies against cytochrome C oxidase IV, SIRT6, caspase‐3 and cleaved caspase‐3 were from Cell Signaling Technology (Beverly, MA, USA). Other chemicals and reagents were of analytical grade. Icariin and resveratrol were dissolved in DMSO at the indicated concentrations and stored at −20°C before application.

Nomenclature of targets and ligands

Key protein targets and ligands in this article are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY (Harding et al., 2018), and are permanently archived in the Concise Guide to PHARMACOLOGY 2017/18 (Alexander et al., 2017a,b,c).

Results

Icariin attenuates I/R‐induced cell death and apoptosis in NRVMs

As shown in Figure 1A, gradually increasing the icariin concentration for 12 h did not significantly affect the viability of the NRVMs. However, icariin treatment clearly increased SIRT1 expression, which was weaker in the icariin group than in the resveratrol group (8 μM) at same time point (Figure 1B). SI/R induced cardiomyocyte death, which was ameliorated in a dose‐dependent manner by icariin treatment, and icariin‐treated cells displayed typical morphological features (Figure 1C). In addition, the apoptosis of NRVMs showed that the SI/R injury increased the apoptotic response, and icariin treatment reduced SI/R‐induced apoptosis (Figures 1D and S2B). As expected, SI/R injury strongly disturbed mitochondrial function and the oxidative stress level, but these indicators were restored by icariin, as shown by decreased mitochondrial depolarization (Figures 1E and S2A), increased SOD activity (Figure 1G) and lower MDA content (Figure 1H). Necrotic cell death in NRVMs was further assessed by measuring LDH levels. Our results showed that icariin triggered a significant decrease in the release of LDH (Figure 1F). SIRT1 and SIRT6 expression decreased after the SI/R but was restored by icariin treatment. In contrast, inverse changes were observed with FOXO1 acetylation (Figures 1I and S2C). Furthermore, after SI/R injury, icariin increased Bcl2, decreased Bax and cleaved caspase‐3, and promoted translocation of cytochrome C from the cytoplasm into the mitochondria (Figure S2C).

Figure 1.

Icariin protects NRVMs from SI/R‐induced apoptosis, decreases mitochondrial oxidative stress and activates the SIRT1 pathway. (A) NRVMs were treated with different concentrations of icariin (ICA; 2, 4 or 8 μM), and cell viability was analysed. Scale bar: 100 μm. (B) SIRT1 expression was measured after icariin and resveratrol (RSV) treatment. (C) NRVMs were pretreated with icariin (2, 4 or 8 μM) and then subjected to SI/R injury, and cell viability was analysed. Scale bar: 100 μm. (D) The NRVM apoptotic index was increased after SI/R, but icariin treatment (2, 4 and 8 μM) prevented cell apoptosis. (E) Data show the protective effect of icariin on SI/R‐induced cardiomyocyte mitochondrial depolarization, as evidenced by JC‐1 staining. (F) Data show the protective effect of icariin on SI/R‐induced LDH release in cell culture supernatants. (G) Data show the protective effect of icariin on SI/R‐induced SOD activity in cell lysates. (H) Data show the protective effect of icariin on SI/R‐induced MDA content in cell lysate. (I) SIRT1 activation and FOXO1 deacetylation were detected by Western blotting analysis. SI/R injury decreased SIRT1 expression and markedly increased FOXO1 acetylation in the NRVMs. Icariin treatment (2, 4 and 8 μM) significantly promoted SIRT1 activation, which caused Ac‐FOXO1 deacetylation. Values are expressed as the means ± SD; n = 5. ^# P < 0.05, significantly different from the control group (each test was repeated three times).

Icariin‐induced SIRT1 overexpression protects cardiomyocytes from I/R‐induced necrosis and apoptosis

To investigate the role of SIRT1 in icariin‐induced cardioprotection in SI/R injury, we inhibited the expression of SIRT1 with sirtinol, a specific NAD‐dependent class III histone deacetylase sirtuin inhibitor, and SIRT1 siRNA in NRVMs and H9c2 cells respectively. The results showed that sirtinol did not affect cell viability (Figure 2A), apoptosis (Figures 2B, S3A and S4A) or the baseline LDH concentration (Figure 2C) or CK‐MB release (Figure 2D). However, pretreating with sirtinol for 2 h significantly attenuated the cardioprotective effects of icariin on cell viability (Figure 2A), cell apoptosis (Figures 2B, S3A and S4A), LDH concentration (Figure 2C) and CK‐MB release (Figure 2D). Meanwhile, SIRT1 siRNA exacerbated the SI/R‐induced injury, which also blunted the myocardial protective effect of icariin (Figures 2E–2H, S3B and S4B). In addition, sirtinol treatment inhibited SIRT1 expression and promoted FOXO1 acetylation. Compared with the cells in the SI/R treatment group, NRVMs and H9c2 cells treated with icariin presented significantly down‐regulated Bax and cleaved caspase‐3 and up‐regulated Bcl2 protein levels, whereas the sirtinol and SIRT1 siRNAs markedly reversed the ICA‐induced regulation of the expression levels of these proteins (Figure 2I and 2J).

Figure 2.

Inhibition of SIRT1 impairs the anti‐apoptotic effects of icariin in the NRVM and H9c2 cell response to SI/R injury. NRVMs were subjected to SI/R injury and then changes in (A) cell viability, (B) apoptotic index, (C) LDH concentration and (D) CK‐MB release were measured after treatment with icariin (ICA) and sirtinol (Sir). The control or SIRT1 siRNA‐transfected H9c2 cells were subjected to SI/R injury; then, the changes in (E) cell viability, (F) apoptotic index, (G) LDH concentration and (H) CK‐MB release were measured after treatment with ICA. After SI/R injury, changes in the protein expression levels (I) of SIRT1, Ac‐FOXO1, Bcl2, Bax and C‐caspase‐3 were analysed by Western blotting in NRVMs treated with icariin and sirtinol. After SI/R injury, changes in the protein expression levels (J) of SIRT1, Ac‐FOXO1, Bcl2, Bax and C‐caspase‐3 were analysed by Western blotting in the control or SIRT1 siRNA‐transfected H9c2 cells treated with ICA. Values are expressed as the means ± SD; n = 5. ^# P < 0.05, significantly different from the control group (each test was repeated three times).

SIRT1 is necessary and sufficient for icariin to restore mitochondrial function and suppress intracellular oxidative stress during SI/R

Mitochondrial dysfunction induced by SI/R increases oxidative stress and plays an important role in I/R‐induced myocardial damage by increasing membrane permeability and cell apoptosis (Murphy and Steenbergen, 2008; Eltzschig and Eckle, 2011). Pretreating with sirtinol for 2 h did not affect the baseline SOD activity values, MDA content or mitochondrial membrane potential in NRVMs. However, compared with the SI/R + ICA cells, the cells pretreated with sirtinol presented with alleviated SOD levels and mitochondrial membrane potential and increased MDA stress (Figure 3A–3C). As expected, excessive Mito‐ROS production was detected together with mitochondrial dysfunction in the NRVMs. SI/R treatment significantly increased the MitoSOX Red intensity, indicating an increase in mitochondrial superoxide production (Figure S3C). Icariin markedly inhibited the SI/R‐induced Mito‐ROS production, and this inhibition was attenuated by sirtinol preconditioning (Figure S3C). In addition, decreasing SIRT1 expression by specific siRNA further reduced SOD activity and mitochondrial membrane potential and increased MDA content in H9c2 cells compared with SI/R‐treated control cells (Figure 3D–3F). SIRT1‐silencing conditions also reversed the ICA‐induced effects on SOD, MDA and mitochondrial membrane potential in H9c2 cells, similar to the results from sirtinol and icariin pretreatment on SI/R injury in NRVMs. Icariin protected against SI/R‐induced mitochondrial dysfunction and dysregulation of the endogenous antioxidant enzyme system, as shown by the decreased mitochondrial translocation of cytochrome C to the cytoplasm and the increased MnSOD expression (Figure 3G and 3H). Sirtinol and SIRT1 siRNA treatments in NRVMs and H9c2 cells, respectively, decreased the effects of icariin on these proteins. These results confirm that SIRT1‐mediated protection by icariin plays an important role in maintaining mitochondrial function and redox balance in I/R‐challenged cardiomyocytes.

Figure 3.

Inhibition of SIRT1 impairs ICA‐mediated mitochondrial homeostasis in NRVMs and H9c2 cells responding to SI/R insults. NRVMs were subjected to SI/R injury; then, the changes in (A) SOD activity, (B) MDA content and (C) mitochondrial membrane potential were measured after treatment with icariin (ICA) and sirtinol (Sir). The control or SIRT1 siRNA‐transfected H9c2 cells were subjected to SI/R injury; then, the changes in (D) SOD activity, (E) MDA content and (F) mitochondrial membrane potential were measured after treatment with ICA. After SI/R injury, changes in the protein expression levels (I) of Mito CC, Cyto CC and MnSOD were analysed by Western blotting in NRVMs treated with icariin and sirtinol. After SI/R injury, changes in the protein expression levels (J) of Mito CC, Cyto CC and MnSOD were analysed by Western blot in the control or SIRT1 siRNA‐transfected H9c2 cells treated with ICA. Values are expressed as the means ± SD; n = 5. ^# P < 0.05, significantly different from the control group (each test was repeated three times).

Icariin improves cardiac function and alleviates mitochondrial oxidative stress and apoptosis following I/R in rat isolated hearts via a SIRT1‐dependent mechanism

We further investigated whether icariin ameliorates myocardial lesions in rat isolated hearts with I/R injury. In general, compared with the functional parameters of the control group, IF, tissue cell apoptosis, LV function (+dP/dt max, −dP/dt min, LVDP and HR) and biochemical indexes in the perfusate (LDH and CK‐MB) and tissues (MDA and SOD) were not significantly affected in the icariin‐treated group (Figure 4A–4F). After 40 min of ischaemia and 60 min of reperfusion, the +dP/dt max, LVDP and HR values were decreased (Figure 4A, 4C and 4D), and the −dP/dt min (Figure 4B) value was increased significantly. Pretreating with icariin for 10 min alleviated myocardial damage and restored LV function, as demonstrated by the substantial improvement in the +dP/dt max, LVDP and HR values (Figure 4A, 4C and 4D) and the reduced −dP/dt min value (Figure 4B) after the reperfusion period. Sirtinol pretreatment for 10 min attenuated the effects of icariin on cardiac function (Figure 4A–4D).

Figure 4.

Inhibition of SIRT1 aggravates both cardiac dysfunction and myocardial impairment induced by I/R, despite the protective effect of ICA, in rat isolated hearts. (A–D) Haemodynamics of the Langendorff perfusion system in I/R‐induced isolated rat hearts after icariin (ICA) or sirtinol (Sir) treatment. The (A) +dP/dt max, (B) −dP/dt min, (C) LVDP and (D) HR data are presented. (E) Representative TTC‐stained ventricular sections from the I/R‐induced rat isolated hearts after icariin or sirtinol treatment. Red‐stained areas indicate viable tissue, and unstained pale areas indicate infarcted tissue (left panel). Scale bar: 1 cm. Morphometric analysis of the IF is shown in the right panel. (F) Representative TUNEL‐stained ventricular sections in the peri‐infarct region of the I/R‐induced isolated rat hearts after icariin or sirtinol treatment. Myocardial nuclei appear light blue; TUNEL‐positive nuclei appear green (left panel). Scale bar: 200 μm. Morphometric analysis of the apoptotic index is shown in the right panel. (G) Data show the I/R‐induced LDH concentration in perfusates after icariin or sirtinol treatment in isolated rat hearts. (H) Data show the I/R‐induced CK‐MB release in perfusates after icariin or sirtinol treatment in isolated rat hearts. (I) Data show the I/R‐induced SOD activity in tissue after icariin or sirtinol treatment in isolated rat hearts. (J) Data show the I/R‐induced MDA content in tissue after icariin or sirtinol treatment in isolated rat hearts. Values are expressed as the means ± SD. n = 8. ^# P < 0.05, significantly different from the control group (each test was repeated three times).

The gross histological analysis of TTC‐stained sections in the I/R injury group had a larger necrotic area than the tissues from the control and icariin‐treated groups (Figure 4E). Icariin reduced the cross‐sectional cardiac IF in the I/R insult group (Figure 4E). The results of the apoptosis assays revealed an increased percentage of TUNEL‐positive cells in the I/R rat hearts, but this percentage was reduced by icariin pretreatment (Figure 4F). The biochemical analysis results showed that more LDH and CK‐MB were released in the coronary perfusates from I/R rat hearts than those from control rat hearts (Figure 4G and 4H). However, in the I/R group, the oxidative stress tests revealed that antioxidase activity (i.e. SOD) was depressed (Figure 4I) in the tissues and that the lipid peroxide level (i.e. MDA) was raised (Figure 4J). icariin pretreatment consistently increased SOD activity and decreased LDH, CK‐MB and MDA levels (Figure 4G–4J). Importantly, the inhibition of SIRT1 activity with sirtinol abolished the protective effects of icariin on the variables measured (Figure 4G–4J), which further demonstrated that after I/R injury in isolated hearts, icariin plays a regulatory role on cell apoptosis and oxidative stress via SIRT1.

As shown in Figure S5, SIRT1 protein expression was significantly increased in the icariin pretreatment group and decreased in the I/R group and was accompanied by an opposing trend in the expression of its substrate (Ac‐FOXO1). Icariin up‐regulated SIRT1 expression and down‐regulated the Ac‐FOXO1 levels in the I/R‐induced isolated heart tissue. Moreover, icariin pretreatment increased the level of cytochrome C in the mitochondria and decreased its levels in the cytoplasm after I/R injury. In the I/R group, the MnSOD level was decreased, but this effect was reversed by icariin pretreatment. The expression of Bcl2 was also increased when I/R hearts were pretreated with ICA; in contrast, the Bax and cleaved caspase‐3 expression levels were reduced when the I/R hearts were subjected to icariin pretreatment. However, sirtinol treatment abolished the icariin‐induced alterations on all the above proteins.

SIRT1 is essential for icariin to ameliorate I/R‐induced myocardial damage in vivo

To monitor heart function after myocardial infarction and reperfusion, echocardiography was carried out in vivo. The EF and FS values from the group treated with icariin for 2 weeks were not significantly different from the control group values (Figure 5A–5C). However, icariin markedly prevented the decrease in EF and FS observed in the I/R‐treated mouse hearts (Figure 5A–5C). Similarly, the ligation and release of the coronary artery created a larger IF, but this effect was markedly reduced by icariin (Figure 5D). Notably, the AAR was not significantly different between these groups (Figure 5E).

Figure 5.

Despite the protective effect of icariin, Inhibition of SIRT1 further promotes I/R‐induced left ventricular dysfunction and myocardial impairment in vivo. (A) Representative M‐mode images demonstrating the preservation of LV contractility in I/R‐induced mouse hearts after icariin (ICA) or sirtinol (Sir) treatment. The (B) EF and (C) FS data are presented. (D) Representative photographs of Evans blue and TTC double‐stained heart sections from each group. Data show the I/R‐induced increase in IF in mouse hearts after icariin or sirtinol treatment. Scale bar: 1 mm. (E) Data show the I/R‐induced changes in the AAR in mouse hearts after icariin or sirtinol treatment. (F) Representative TUNEL‐stained ventricular sections in the peri‐infarct region from the I/R‐induced mouse hearts after icariin or sirtinol treatment. Myocardial nuclei appear light blue; TUNEL‐positive nuclei appear green (left panel). Scale bar: 200 μm. Morphometric analysis of the apoptotic index is shown in the right panel. (G) Data show the I/R‐induced LDH concentration in mouse serum after icariin or sirtinol treatment. (H) Data show the I/R‐induced CK‐MB release in mouse serum after icariin or sirtinol treatment. (I) Data show the I/R‐induced SOD activity in mouse heart tissue after icariin or sirtinol treatment. (J) Data show the I/R‐induced MDA content in mouse heart tissue after icariin or sirtinol treatment. Values are expressed as the means ± SD. n = 8. ^# P < 0.05, significantly different from the control group (each test was repeated three times).

In the mouse myocardial I/R model, icariin treatment effectively attenuated the increased apoptotic index (Figure 5F) and inhibited the increase in LDH (Figure 5G) and CK‐MB (Figure 5H) serum levels, which demonstrated that icariin inhibited myocardial necrosis in vivo (Figure 5F–5H).

Assays of oxidative stress‐related enzymes showed that, although pretreatment with icariin did not affect the baseline values of SOD and MDA in the mouse hearts, it did improve SOD activity and reduced the MDA level, compared with those in the I/R‐treated group (Figure 5I, J).

We also investigated the role of icariin in regulating SIRT1, Ac‐FOXO1, mitochondrial oxidative stress and apoptotic proteins after the mouse hearts were subjected to I/R injury (Figure S6). The Western blotting results showed that icariin up‐regulated SIRT1 expression, which further reduced the FOXO1 acetylation level. Compared with the I/R group, icariin treatment triggered the SIRT1‐FOXO1 pathway, restored cytochrome C stabilization by increasing its translocation into the mitochondria, activated an essential intracellular antioxidant enzyme (i.e. up‐regulated MnSOD) and inhibited cell apoptosis (i.e. up‐regulated Bcl2 and down‐regulated Bax and cleaved caspase‐3).

Importantly, this cardioprotective effect of icariin in vivo was blocked by SIRT1 inhibition with sirtinol, as shown by the decreased heart function (Figure 5A–5C), increased cell apoptosis and necrosis (Figures 5D–5H and S6), and increased oxidative stress response (Figures 5I,J and 6) in the I/R + ICA + sirtinol‐treated mouse hearts, compared to the I/R + ICA‐treated hearts.

Figure 6.

The SIRT1‐dependent mechanism involved in the cardioprotective effects of icariin against the mitochondrial oxidative stress induced by I/R injury. Icariin treatment reduces cardiac injury and improves myocardial function during I/R injury. Icariin attenuates ROS overproduction and maintains mitochondrial homeostasis by regulating the SIRT1/FOXO1 signalling pathway, which subsequently relieves myocardial cell apoptosis after I/R injury.

Discussion

Myocardial cell death resulting from I/R injury is a major cause of morbidity and mortality worldwide (Ferdinandy et al., 2014). In the past few decades, it has become clear that the myocardial response to reperfusion may lead to further irreversible myocardial cell death (Murphy and Steenbergen, 2008), which has motivated extensive study of the cardioprotective mechanisms against I/R injury. In this study, we systematically evaluated the protective effects of icariin against I/R injury, which might be mediated by the up‐regulation of myocardial SIRT1 expression in vitro and in vivo.

Icariin belongs to the flavonoid family of polyphenols, the structure of which comprises 15 carbons with two aromatic rings connected by a three‐carbon bridge (Vauzour, 2012; Shen et al., 2017). Icariin has been found to exert many biological and pharmacological effects, including protection against I/R injury in different tissues. Because it possesses a polyphenolic structure similar to resveratrol, icariin may activate SIRT1 through similar mechanisms (Canto et al., 2009; Park et al., 2012; Verdin, 2015). Recent studies have shown that icariin and its metabolites protect against cerebral I/R injury in rats via NF‐κB inhibition and PPAR activation (Deng et al., 2016; Xiong et al., 2016). In addition, cardioprotective effects of icariin have also been described, such as attenuating the damage induced by I/R in rats through activating the PI3K/Akt pathway (Ke et al., 2015; Meng et al., 2015; Zhai et al., 2015). Importantly, Schluesener demonstrated that network analysis based on the Comparative Toxicogenomics Database, linked icariin via SIRT1 to prominent age‐associated pathologies (Schluesener and Schluesener, 2014). However, the role of icariin in cardiovascular protection remained unclear.

SIRT1, a class III histone deacetylase, regulates a wide array of cellular processes by deacetylating both histones and an increasing number of non‐histone proteins (Finkel et al., 2009). In heart tissue, SIRT1 plays a protective role against myocardial I/R injury, and the stimulation of SIRT1 may represent a novel method for protecting the heart from ischaemic heart disease (Murry et al., 1986; Hsu et al., 2010). Some studies have further investigated the SIRT1‐regulated protective effects of icariin on ischaemic disease, including stroke and intestinal I/R‐induced acute lung injury (Zhai et al., 2015; Zhang et al., 2015b). However, the role of ICA‐mediated SIRT1 expression in myocardial I/R injury has not been fully elucidated.

Consistent with the studies summarised above, we showed, in this study, that icariin significantly mitigated heart damage after I/R injury. Supplying rat isolated hearts with icariin in the KHB improved the changes in LVDP, HR and ±dP/dt induced by I/R in the Langendorff perfused system, which was similar with the increase of FS and EF in icariin‐treated mice with ligation of coronary artery. There is strong correlation between the increase in LDH and CK‐MB levels and that of the IF (Wang et al., 2016). When the isolated and in vivo hearts were challenged with I/R injury, the LDH and CK‐MB levels were increased in the perfusate and serum. The icariin‐mediated protection was shown by the 20 to 50% reduction in IF in the ex vivo and in vivo models. Preconditioning with icariin in the KHB and icariin treatment by oral gavage effectively decreased LDH and CK‐MB levels, suggesting that icariin enhanced the heart's ability to resist I/R‐induced damage. Incremental increases in the icariin concentration did not significantly affect NRVMs viability. However, in the NRVMs and H9c2 cells, icariin clearly preserved cellular activity following SI/R injury. Treatment with icariin repressed the release of LDH and CK‐MB into the culture supernatants. In addition, icariin dose‐dependently increased the expression of SIRT1 in NRVMs. Inhibiting SIRT1 with sirtinol in rat isolated hearts abolished the icariin‐mediated improvements in cardiac function, such as the decreased LVDP, HR and +dP/dt max values, as well as the alterations to EF and FS in vivo. Treatment with sirtinol increased LDH and CK‐MB levels, which had been reduced in the perfusate and serum by icariin treatment. The in vitro experiment showed similar results: SIRT1 siRNA and sirtinol reversed the protective effects of icariin on viability and on the icariin‐induced down‐regulation of LDH and CK‐MB in NRVMs and H9c2 cells. The above results imply that icariin exerted cardioprotective effects, against I/R injury via SIRT1 activation in vivo and in vitro.

Oxidative stress is widely recognized as a primary mechanism and as an important cause of cell death during myocardial infarction, which is defined as an imbalance between the generation of ROS and activity of the antioxidant defence system (Wang et al., 2015; Luo et al., 2016). Previous studies have demonstrated that I/R‐induced oxidative stress leads to the accumulation of the lipid peroxidation product MDA (Rochette et al., 2013) and a decline in the endogenous free radical‐scavenging enzyme SOD (Gao et al., 2009). In addition, Hsu et al. showed that the oxidative stress level after I/R was negatively regulated by SIRT1, which suggested that the up‐regulation of SIRT1 reduces oxidative stress and enhances antioxidant capacity to protect the heart from I/R injury (Hsu et al., 2010). In our study, the activity of SOD was dramatically reduced after I/R, whereas MDA content was significantly increased. Icariin pretreatment attenuated these changes in SOD and MDA both in vitro and in vivo. However, these trends were reversed by SIRT1 siRNA and sirtinol treatment, suggesting that icariin might protect the myocardium against I/R injury via promoting SIRT1, which inhibits oxidative stress.

Mitochondria are critical targets and a root cause of superoxide‐induced tissue injury, particularly during ischaemia and subsequent reperfusion (Murphy and Steenbergen, 2008). In I/R, cardiac mitochondrial damage leads to dysregulated mitochondrial metabolism and is thus a key contributor to myocardial cell death from excessive ROS production, which ultimately leads to mitochondrial membrane potential depolarization (Lesnefsky et al., 2016; Wu et al., 2017). Meanwhile, aberrant mitochondrial depolarization was associated with a significantly increased mitochondrial superoxide production (Becatti et al., 2012). SIRT1 is known to exert protective effects against I/R‐induced mitochondrial oxidative damage (Hsu et al., 2010). In our study, I/R injury decreased SIRT1 expression and caused rapid dissipation of the mitochondrial membrane potential and a dramatic increase in the mitochondrial superoxide production, which were altered in a dose‐dependent manner by icariin treatment. SIRT1 siRNA and sirtinol significantly reduced the inhibitory effect of icariin on mitochondrial stress, which indicated that icariin might rescue mitochondrial function by regulating SIRT1.

Moreover, mitochondrial MnSOD plays a major role in scavenging superoxide (Hoshida et al., 2002). I/R injury significantly impaired MnSOD enzyme activity (Suzuki et al., 2002; Turoczi et al., 2003; Shao et al., 2014). In cardiomyocytes, SIRT1 stimulates the deacetylation of FOXO1, which, in turn, plays an essential role in mediating the SIRT1‐induced up‐regulation of MnSOD and the suppression of oxidative stress (Hsu et al., 2010). Our results showed that icariin pretreatment promoted MnSOD enzyme activity both in vitro and in vivo accompanied by an increase in SIRT1 and a decrease in Ac‐FOXO1. However, sirtinol and SIRT1 siRNA abolished the icariin‐induced enhancement in MnSOD activity.

Furthermore, several studies have demonstrated that cytochrome C is released from dysfunctional mitochondria into the cytosol during heart ischaemia (Lesnefsky et al., 1997; Borutaite et al., 2001) and the release of cytochrome C then initiates programmed cell death (Borutaite et al., 2003). Cytochrome C mediates apoptosis via the proapoptotic members of the Bcl‐2 family, such as Bcl‐2‐associated X protein (Bax), whose actions are opposed by the anti‐apoptotic Bcl‐2 members, such as Bcl2 (Tait and Green, 2010). In contrast, the release of cytochrome C into the cytosol leads to the recruitment and activation of http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=1625&familyId=734&familyType=ENZYME. Once activated, caspase‐9 cleaves and activates downstream caspase‐3 (Riedl and Salvesen, 2007). In addition, in the mouse myocardial infarction model, SIRT1 mediated the up‐regulation of the FOXOs and prevented cellular injury by activating Bcl‐xL and suppressing Bax and cleaved caspase‐3 (Hsu et al., 2010). Icariin also inhibited markers of mitochondrial stress (i.e. polarized mitochondrial membrane potential and stabilized cytochrome C) and cell apoptosis (i.e. increased Bcl2 and decreased Bax and caspase‐3 levels), indicating that the protective effect of icariin on alleviating oxidative stress and anti‐apoptotic action occurred mainly through the attenuation of mitochondrial dysfunction by inhibiting mitochondria‐dependent apoptosis. After co‐treatment with SIRT1 siRNA or sirtinol, the inhibitory effects of icariin on I/R injury‐induced mitochondrial stress and cell apoptosis were markedly reduced. Together, these findings indicate that icariin attenuated myocardial I/R injury‐induced mitochondrial oxidative damage via the SIRT1/FOXO1 pathway.

In conclusion, icariin treatment reduces cardiac injury and improves myocardial function during I/R injury. These cardioprotective effects seem to be mostly due to attenuation of excessive ROS production and maintenance of mitochondrial homeostasis; the SIRT1/FOXO1 signalling pathway is involved in both processes and subsequently inhibits cell apoptosis (Figure 6). These results indicate that icariin may be potentially beneficial in the treatment of myocardial I/R injury during cardiac surgery and ischaemic heart disease.

Author contributions

X.‐q.H. and C.‐x.F. designed the research and revised this manuscript. B.W., J.‐y.F. and L.‐m.Y. performed the in vivo and isolated heart experiments and drafted this manuscript. J.‐s.H., Y.‐c.W. and L.‐m.Y. performed the in vitro experiments. Y.‐q.C., Y.W. and Y.Z. analysed the data; and X.F., S.‐y.D. and Z.‐q.M. prepared the figures.

Conflict of interest

The authors declare no conflicts of interest.

Declaration of transparency and scientific rigour

This http://onlinelibrary.wiley.com/doi/10.1111/bph.13405/abstract acknowledges that this paper adheres to the principles for transparent reporting and scientific rigour of preclinical research recommended by funding agencies, publishers and other organisations engaged with supporting research.

Supporting information

Figure S1 Molecular structure and experimental protocols. (A) The chemical formula of icariin. (B) The chemical formula of resveratrol. (C) The adverse effects of icariin or Res treatment on·NRVMs·under normal culture condition (n = 5 per group). (D) The protective effects of icariin treatment on·NRVMs·through SIRT1 pathway after I/R injury (n = 5 per group). (E) The protective effects of icariin treatment on the H9c2 cell through SIRT1 pathway after I/R injury (n = 5 per group). (F) The contribution of SIRT1 induced by icariin and the involvement of mitochondrial oxidative stress to the improved functional cardiac recovery and reduced infarct size in isolated rat hearts (n = 8 per group). (G) The contribution of SIRT1 induced by icariin and the involvement of mitochondrial oxidative stress to the improved functional cardiac recovery and reduced infarct size in mice hearts (n = 8 per group). GM, growth medium; ICA, icariin; KHB, Krebs–Henseleit buffer; Res, resveratrol; SF, serum‐free medium; SI/R, simulated ischemia/reperfusion.

Figure S2 Icariin protects NRVMs from SI/R‐induced apoptosis which combined with decreasing mitochondrial oxidative stress. (A) Changes in mitochondrial membrane potential were indicated by JC‐1 staining. Red fluorescence represents the mitochondrial aggregate form of JC‐1, indicating intact mitochondrial membrane potential. Green fluorescence represents the monomeric form of JC‐1, indicating dissipation of potential. Scale bar: 100 μm. (B) Representative TUNEL‐stained in I/R‐induced NRVMs after icariin treatment. Myocardial nuclei appear light blue; TUNEL‐positive nuclei appear green (indicated with arrowhead). Scale bar: 100 μm. (C) Change of SIRT6, Bcl2, Bax, mitochondrial cytochrome C (Mito CC), cytoplasmic cytochrome C (Cyto CC), and cleaved caspase‐3 were detected by western blot analysis. n = 5. Values are expressed as mean ± SD. ^# P < 0.05 compared with control group (Each test was repeated for three times). Bax, Bcl‐2 associated X protein; CK‐MB, isoenzyme of creatine kinase; COX IV, cytochrome C oxidase IV; Cyto CC, cytoplasmic cytochrome C; C‐caspase‐3, cleaved caspase‐3; ICA, icariin; Mito CC, mitochondrial cytochrome C; SI/R, simulated ischemia/reperfusion.

Figure S3 Representative fluorescent images about TUNEL and Mito‐ROS production after icariin treatment combined with sirtinol or SIRT1 siRNA in SI/R‐induced cells. Apoptotic positive cells in (A) NRVMs and (B) H9c2 cells were showed. TUNEL‐positive nuclei appeared green. (C) Changes of mitochondrial ROS were indicated by MitoSOX Red intensity in NRVMs. Myocardial nuclei appear light blue. Scale bar: 100 μm. (Each test was repeated for three times). n = 5. ICA, icariin; SI/R, simulated ischemia/reperfusion; Sir, sirtinol.

Figure S4 Representative fluorescent images about Annexin V/PI and trypan blue staining after icariin treatment combined with sirtinol or SIRT1 siRNA in SI/R‐induced cells. Apoptotic positive cells were detected via Annexin V/PI staining in NRVMs by flow cytometry (A) and confocal microscope (B). Annexin V showed as green, PI displayed as red. (C) Evaluation of H9c2 cell viability and morphology by trypan blue exclusion. Scale bar: 100 μm. (Each test was repeated for three times). n = 5. Ann V, Annexin V; ICA, icariin; Ph C, phasecontrast; PI, propidium iodide; SI/R, simulated ischemia/reperfusion; Sir, sirtinol.

Figure S5 Changes of protein expressions in I/R‐induced isolated rat hearts after icariin treatment combined with sirtinol. The results of Western blots for SIRT1, Ac‐FOXO1, Mito CC, Cyto CC, Bcl2, Bax, and C‐caspase‐3 are provided. n = 8. Values are expressed as mean ± SD. ^# P < 0.05 compared with control group (Each test was repeated for three times). Ac‐FOXO1, acetylated FOXO1; Bax, Bcl‐2 associated X protein; COX IV, cytochrome C oxidase IV; Cyto CC, cytoplasmic cytochrome C; C‐caspase‐3, cleaved caspase‐3; FOXO, forkhead box O; ICA, icariin; Mito CC, mitochondrial cytochrome C; MnSOD, manganese SOD; I/R, ischemia/reperfusion; Sir, sirtinol.

Figure S6 Changes of protein expressions in I/R‐induced mice hearts after icariin treatment combined with sirtinol. The results of Western blots for SIRT1, Ac‐FOXO1, mitochondrial cytochrome C (Mito CC), cytoplasmic cytochrome C (Cyto CC), Bcl2, Bax, and cleaved caspase‐3 are provided. n = 8. Values are expressed as mean ± SD. ^# P < 0.05 compared with control group (Each test was repeated for three times). Ac‐FOXO1, acetylated FOXO1; Bax, Bcl‐2 associated X protein; COX IV, cytochrome C oxidase IV; Cyto CC, cytoplasmic cytochrome C; C‐caspase‐3, cleaved caspase‐3; FOXO, forkhead box O; ICA, icariin; Mito CC, mitochondrial cytochrome C; MnSOD, manganese SOD; I/R, ischemia/reperfusion; Sir, sirtinol.

Table S1 Antibodies for Western blotting analysis.

Supporting info item

Wu, B. , Feng, J. , Yu, L. , Wang, Y. , Chen, Y. , Wei, Y. , Han, J. , Feng, X. , Zhang, Y. , Di, S. , Ma, Z. , Fan, C. , and Ha, X. (2018) Icariin protects cardiomyocytes against ischaemia/reperfusion injury by attenuating sirtuin 1‐dependent mitochondrial oxidative damage. British Journal of Pharmacology, 175: 4137–4153. 10.1111/bph.14457.