Introduction

Myocardial infarction is the result of coronary artery occlusion which contributes to the high mortality around the world. The loss of blood flow may lead to the death of cardiomyocytes in the ischemia region (Murphy and Steenbergen 2008; Rafiq et al. 2014). With the early recovery of the arterial blood flow, a controversial injury caused by reperfusion has been arisen as a novel topic during the treatment of ischemic heart diseases, which limits the efficiency of vascular recanalization such as percutaneous coronary intervention and thrombolysis medicine applications (Vander Heide and Steenbergen 2013; Matsushima et al. 2014). MIRI triggers a complex reaction of cardiac tissue, which results in cardiomyocyte death (Whelan et al. 2010). It is of great significance that the underlying mechanism of MIRI is a systematic network with a various number of pathophysiological process and numerous molecules (Gottlieb et al. 2009). To abolish the negative effect of this pathophysiological process, researchers found heat shock protein 70 (HSP70) was protective during MIRI, which was a promising target for clinical treatments (Garrido et al. 2006; Feng et al. 2014; Vicencio et al. 2015).

HSP70 is regarded as cardiac protective molecules and could be upregulated to inhibit cardiomyocyte apoptosis after MIR (Peng et al. 2010). In addition, overexpression of HSP70 could inhibit caspase-dependent cardiomyocyte apoptosis which significantly improves the cardiac function (Rani et al. 2013). Besides inhibition of apoptosis (Sun et al. 2015), HSP70 was recently proven to attenuate cell death and autophagy via reducing LC3B expression during heat shock response (Hsu et al. 2013). Autophagy, which can induce necroptosis, plays a detrimental role during MIRI (Huang et al. 2015; Jiang et al. 2015). Necroptosis, a programmed necrosis, triggers a variety of reactions enhancing cellular death during stress exposure (Dannappel et al. 2014; Koshinuma et al. 2014). Meanwhile, animal experiments showed that the necroptosis inhibitor, Necrostatin-1, could exert efficient protection function against MIRI (Dmitriev et al. 2013). However, whether HSP70 inhibits necroptosis via suppressing autophagy during MIRI is unknown.

In this study, we aim to profile the relationship between HSP70, autophagy, and necroptosis during MIRI, demonstrating that HSP70 inhibits necroptosis through suppression of autophagy to promote cardiomyocyte survival. Hereon, we found a novel scenario on what exact functions HSP70 acts as a protective molecule during MIRI.

Materials and Methods

Animal experimental protocols

Sprague–Dawley (SD) rats, weight 230–280 g, were obtained from the Experimental Animal Center, Nantong University. All procedures were in accordance with the Guide for the Care of Use of Laboratory Animals published by the US National Institute of Health (NIH Publication No. 85–23, revised in 1996). The rats were randomly divided into eight groups (n = 6 in each group): sham group in which rats underwent operation without suture tie-down of the left anterior descending (LAD) coronary artery; experimental groups in which rats underwent ischemia for 30 min with 0, 2, 4, 6, 8, 12, or 24 h reperfusion, respectively.

Surgery of animals

All animals were operated under anesthesia with chloral hydrate. After endotracheal intubation, the heart was rapidly exposed via a left thoracotomy. Then, a 6-0 polyproline ligature was placed under the left coronary artery (LCA). The ends of the tie were threaded through a small plastic tube to form a snare for reversible LCA occlusion. Myocardial ischemia was confirmed by a pale area below the suture and ST-T elevation shown in electrocardiogram (ECG). After ischemia, reperfusion was achieved by loosening the snare and characterized by rapid disappearance of cyanosis. Following the operation, the incisions were closed in layers. The chest and endotracheal tubes were removed. Room temperature was maintained at 37°C.

Western blot analysis

Western blot was prepared with myocardial tissue after operation (n = 6 at each time point). Samples were isolated from approximately 0.1 g of myocardial tissue at different time points and minced with eye scissors on ice. Total myocardial tissue protein was then homogenized in lysis buffer and clarified by centrifugation at 13,000 rpm for 15 min in a microcentrifuge at 4°C. The protein was separated by SDS-PAGE, transferred to PVDF membranes (Millipore; Billerica, MA) at 300 mA for appropriate timespan. Membranes were exposed for HSP70 (mouse, 1:800; Santa Cruz; Dallas, TX); glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (rabbit, 1:800; Santa Cruz); receptor-interacting protein 1 (RIP1) (goat, 1:800; Santa Cruz); and LC3A/B (mouse, 1:1000; Cell signaling) antibodies overnight at 4°C. Then, HRP-conjugated secondary antibodies (1:1000; Southern-Biotech, Birmingham, AL) were added and incubated for 2 h at 37°C. Finally, membranes were analyzed by enhanced chemiluminescence system (ECL, Cell Signaling).

Immunohistochemical analysis

After MIR, rat hearts were perfused with sterile saline and 4% formalin successively (Fisher Scientific, Fairlawn, NJ). Then, hearts were rapidly collected and immersion fixed in 10% formalin for 24 h. The sections were cut serially (5 μm). For immunohistochemical staining, the sections were boiled at 121°C for 20 min in 10 mM citrate buffer solution (pH 6.0) for antigen retrieval. Endogenous peroxidase activity was blocked by soaking in 0.3% hydrogen peroxide. After rinsing in PBS (pH 7.2), the sections were incubated with HSP70, RIP1, and LC3A/B antibodies for 2 h at 37°C. After incubation, the sections were washed with PBS for three times and incubated with secondary antibody for 30 and 20 min at 37°C separately. After three times washing by PBS, the sections were incubated with DAB in 0.05 mol/L Tris buffer (pH 7.6) containing 0.03% H2O2 for signal development. Finally, the slides were dehydrated, cleared, and cover slipped (Wang et al. 2012). All of analyses were conducted by Image-Pro plus 6.0 software.

Isolation and culture of rat ventricular cardiomyocytes

Neonatal rat ventricular myocytes were isolated from 2-d-old SD rats. The hearts were collected, minced, and digested in trypsin. The primary neonatal rat cardiomyocytes were cultured in Dulbecco’s modified Eagle’s medium (DMEM) containing 15% fetal bovine serum (FBS). The medium was replaced with DMEM lacking FBS before the cells were placed into a hypoxic incubator (95% N2, 5% CO2, 37°C). After 6 h in the hypoxic incubator, cells were transferred to a normal incubator (21% O2, 5% CO2, 74%N2, 37°C) for 6 h to reoxygenation (Ji et al. 2013).

Double immunofluorescent staining

The primary cardiomyocytes were cultured in 24-wells plate (Corning Inc., Corning, NY). After H/R stimuli, the cells were fixed in 4% paraformaldehyde. All samples were incubated overnight at 4°C with primary antibody for HSP70 (1:100; Santa Cruz) and different markers as follows: α-actinin (cardiomyocyte marker, 1:100; Santa Cruz), RIP1 (1:100; Santa Cruz), and LC3A/B (1:100; cell signaling). After washing in PBS for three times, each time 15 min, a mixture of 4, 6-diamidino-2-phenylindole (DAPI) with FITC- and Cy3-conjugated secondary antibodies were added in a dark room and incubated overnight at 4°C. The stained sections were analyzed with a Leica fluorescence microscope (Leica DM 5000B, Germany).

Small interfering RNA-based experiment

Three small interfering RNAs (siRNAs) targeting HSP70 gene were designed and synthesized by Shanghai Genepharma (Shanghai, China), and the most effective siRNA (siHSP70) identified by western blot was applied for sequent experiments. The sequences of three siHSP70s are as follows: 5′-GCUGAGAAAGAGGAGUUCGTT-3′, 5′-GAAUGCGCUCGAGUCCUAUTT-3′, 5′-GCUGAGAAAGAGGAGUUCGTT-3′. Scrambled RNA oligonucleotides were used as control. Twenty-four hours prior to transfection, cardiomyocytes were plated onto a 6-well plate (Corning Inc., Corning, NY) at 40–60% confluence. For each well, 33.3 nM siHSP70 or scrambled siRNA was transfected using Lipofectamine 2000 (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions. The medium was replaced with DMEM containing 10% FBS after 6 h (Wan et al. 2015).

Overexpression plasmid-based experiment

Overexpression plasmid of HSP70 was purchased from PPL plasmid library (Nanjing, Jiangsu, China). The transfection procedure was identical to siRNA transfection described above.

LDH assay

The LDH assay kit was purchased from Jiancheng bioengineering institute (Nanjing, Jiangsu, China). All measurements were conducted following the instructions.

Statistical assay

The SPSS19.0 software was used for statistical analysis. The results were presented as means ± SD. Differences between groups were calculated using the Student t test or one-way analysis of variance. P < 0.05 was considered statistically significant. All the experiments were repeated for at least three times with the consistent results.

Results

HSP70 expression is opposite to autophagy and necroptosis markers after MIRI

The rat myocardial IR model was constructed by temporarily occluding the LAD coronary artery, which was verified by ECG. Normal ECG was used as the negative control Fig. 1A . The changes of ST segment elevation and wide QRS waves were observed after myocardial ischemia immediately. With myocardial reperfusion, the electrophysiological changes indicated the recovery of cardiac function. However, reperfusion could not reverse the changes of ECG completely compared to the ischemia group, which showed the establishment of MIR Fig. 1B, C .

Figure 1
figure 1

HSP70 expression is opposite to autophagy and necroptosis markers after MIR in vivo. Adult SD rats were treated with ischemia for 30 min and different time points of reperfusion. A Normal ECG was taken as control. B ECG of acute myocardial ischemia. Marked ST-segment elevation was recorded, which indicated proximal occlusion of LAD artery. C ECG showed recovery of ST elevation and decrease in the T wave amplitude. D HSP70, LC3 A/B, and RIP1 expressions were detected by western blot after MIR. E Statistical graphs (relative optical density) show HSP70 and RIP1 expressions to GAPDH. F Statistical graph (relative optical density) shows the ratio of LC3B to LC3A. G Immunohistochemistry detected HSP70, LC3A/B, and RIP1. Scale bars = 50 μm in first image (applied to all). H Statistical graphs for the positive areas of tissues. Asterisk, dollar sign, ampersand, and commercial at represent P < 0.05 compared with the sham group.

Full size image

With the elongation of reperfusion, HSP70 expression increased immediately after MIR and peaked during 6 to 8 h. Afterwards, HSP70 expression decreased, while the ratio of LC3B to LC3A was increased rapidly during 12 h of reperfusion, and RIP1 expression showed a gradually increase from 4 h of reperfusion Fig. 1D . Interestingly, after 2 h reperfusion following ischemia, HSP70 expression was postponed after the blooming of autophagy, followed by the increase of RIP1 expression which indicated the initiation of necroptosis Fig.1E, F . These results suggested that HSP70 might be related to autophagy and necroptosis during MIR.

To identify the cellular location of HSP70, LC3A/B, and RIP1, we chose the notable time point, 6 h reperfusion after ischemia to perform the immunohistochemistry. It is shown that after MIR, HSP70 was located in the nucleus and cytoplasm of cardiomyocytes while RIP1 and LC3A/B were located predominately in the cytoplasm of cardiomyocytes Fig. 1G . Meanwhile, the analysis of positive cells showed the increase of HSP70, LC3A/B, and RIP1 expression following MIR Fig. 1H .

HSP70 is associated with cardiomyocyte autophagy and necroptosis after MIRI

To mimic MIRI in vitro, we used the neonatal rat ventricular myocytes with H/R treatment. LDH release burst after hypoxia during 8 h reoxygenation. However, LDH release was blunt after 8 h reoxygenation, but reoxygenation could not completely reverse LDH release to normal level Fig. 2A . The HSP70, LC3A/B, and RIP1 expression showed similar change after H/R treatment compared with that of MIRI in vivo Fig. 2B, C, D . These results further suggested that HSP70 was associated with cardiomyocyte autophagy and necroptosis. In the consequent experiments, 6 h of reoxygenation after hypoxia and 6 h was set for H/R stimuli.

Figure 2
figure 2

HSP70 expression is opposite to autophagy and necroptosis markers in neonatal ventricular cardiomyocytes following H/R treatment in vitro. Neonatal ventricular cardiomyocytes were treated with 6 h hypoxia and different times of reoxygenation. A LDH assay was conducted to detect cell death. B HSP70, LC3 A/B, and RIP1 expressions were detected by western blot after H/R. C Statistical graph (relative optical density) shows HSP70 and RIP1 expressions to GAPDH. D Statistical graph (relative optical density) shows the ratio of LC3B to LC3A. E Distribution of HSP70 with LC3A/B and RIP1 with or without H/R. Neonatal ventricular cardiomyocytes fixed on the small plates were detected for HSP70, LC3A/B, and RIP1. DAPI was used to label nuclei (H/R: hypoxia for 6 h and reoxygenation for 6 h). Scale bars = 25 μm in first image (applied to all).

Full size image

To further investigate the relationship among HSP70, autophagy, and necroptosis, we performed double immunofluorescence staining of HSP70, LC3A/B, and RIP1 in neonatal rat ventricular myocytes after H/R. It was obvious that HSP70, LC3A/B, and RIP1 co-localized in the cytoplasm of myocytes after 6 h reoxygenation Fig. 2E . These results further indicated that HSP70 was related to autophagy and necroptosis of cardiomyotes.

The knockdown of HSP70 facilitates necroptosis and autophagy in cardiomyocytes after H/R treatment

To explore the effects of HSP70 on autophagy and necroptosis, we utilized siRNA to downregulate HSP70 expression. The HSP70 siRNA3 could significantly downregulate HSP70 expression compared to scrambled siRNA, and the rescue experiments confirmed that HSP70 siRNA3 specifically downregulated HSP70 expression Fig. 3A . After transfected with siHSP70, HSP70 expression was downregulated in normal group and H/R group, respectively Fig. 3B . However, the ratio of LC3B to LC3A and RIP1 expression were upregulated following H/R. Interestingly, the changes did not appear in the normal groups Fig. 3C, D . While LDH assay showed that HSP70 knockdown led to increasing cell death after H/R Fig. 3E .

Figure 3
figure 3

HSP70 knockdown upregulates neonatal ventricular cardiomyocyte autophagy and necroptosis. Knocking down of HSP70 in neonatal ventricular cardiomyocytes with or without H/R. A HSP70 protein expression after HSP70 knockdown in cardiomyocytes with overexpression of HSP70 rescued. B HSP70, LC3 A/B, and RIP1 expressions were detected by western blot. C Statistical graph (relative optical density) shows HSP70 and RIP1 expressions to GAPDH. D Statistical graph (relative optical density) shows the ratio of LC3B to LC3A. E LDH assay was conducted to detect cell death. F Co-location of HSP70 with LC3A/B and RIP1 with or without H/R in neonatal ventricular cardiomyocytes after HSP70 knockdown. Neonatal ventricular cardiomyocytes fixed on the small plates were detected for HSP70, LC3A/B, and RIP1. DAPI was used to label nuclei (H/R: hypoxia for 6 h and reoxygenation for 6 h). Scale bars = 25 μm in first image (applied to all).

Full size image

To clarify the changes of distribution and expression of LC3A/B and RIP1 after HSP70 knockdown, double immunofluorescence staining was carried out. After HSP70 knockdown, autophagy and necroptosis were upregulated in cadiomyocytes following H/R stimuli, while the cells without H/R stimuli had no obvious changes of autophagy and necroptosis Fig. 3F . These results indicated that HSP70 only had effects on cardiomyocytes after H/R stimuli.

HSP70 inhibits cardiomyocyte necroptosis via suppressing autophagy

After given z-VAD-FMK (an apoptosis inhibitor, 50 μM for 1 h), the cardiomyocytes were selectively treated with 3-MA (an autophagy inhibitor, 5 mM for 2 h), rapamycin (an autophagy activator, 0.5 μM for 4 h), and necrostatin-1 (a necroptosis inhibitor, 50 μM for 2 h) and transfected with siHSP70. When autophagy or necroptosis was inhibited, HSP70 expression did not change obviously. However, when autophagy was inhibited, RIP1 expression was downregulated. Additionally, the autophagy inhibitor could abolish the increased expression of RIP1 caused by siHSP70 transfection. Meanwhile, rapamycin caused further increase of RIP1 expression which was in relevance with siHSP70 transfection Fig. 4A-C .

Figure 4
figure 4

HSP70 inhibits cardiomyocyte necroptosis via suppressing autophagy. After given z-VAD-FMK, an apoptosis inhibitor, cardiomyocytes were selectively treated with 3-MA (an autophagy inhibitor), rapamycin (an autophagy activator), and necrostatin-1 (a necroptosis inhibitor) and transfected with siHSP70. A HSP70, LC3 A/B, and RIP1 expressions were detected by western blot. B Statistical graph (relative optical density) shows HSP70 and RIP1 expressions to GAPDH. C Statistical graph (relative optical density) shows the ratio of LC3B to LC3A. D LDH assay was conducted to detect cell death.

Full size image

Ultimately, we used LDH assay to detect cell damage, showing that HSP70 knockdown resulted in more cell damage compared to cardiomyocytes exposed to autophagy and necroptosis inhibitor. Interestingly, after siHSP70 transfection, autophagy inhibitor exerted a stronger effect of alleviating cell damage compared to cardiomyocytes given necroptosis inhibitor only Fig. 4D . Herein, these results suggested that HSP70 inhibited necroptosis and autophagy, while inhibition of autophagy had effects on necroptosis,

Discussion

Over the past decade, the treatment for acute myocardial ischemia diseases has enjoyed significant advances, especially the therapies for ST-segment elevation myocardial infarction (STEMI). However, STEMI still has high morbidity and mortality worldwide (Windecker et al. 2013). The main reason that blocks myocardial recovery from ischemia diseases is MIRI (Bagai et al. 2014). Recently, autophagy has been regarded as a novel pathophysiological process during MIR, which is of great significance besides apoptosis and creates a scenario of both pros and cons (Gottlieb and Mentzer 2010). However, more researches showed that inhibition of autophagy could reduce cell death after H/R stimuli in cardiomyocytes, while lysosomal degradative pathway activation which was characterized by the ratio of LC3B/LC3A after MIR was aggressive to promote cardiomyocyte death (Gurusamy et al. 2009; Wang et al. 2015). In our study, after 2 h reperfusion, autophagy was promoted and prolonged for the next 10 h, which was in line with the previous studies (Jian et al. 2015; Xiao et al. 2015).

Necroptosis, a programmed cell necrosis, leads to a reduction of cardiac function (Koshinuma et al. 2014). RIP1 is one of the representatives for the activation of necroptosis (Rosentreter et al. 2015). Autophagy could promote both necroptosis and apoptosis in human glioblastoma cells in vitro (Zhang et al. 2015). Herein, we used rat MIR and neonatal cardiomyocyte H/R models to detect autophagy and necroptosis, finding 6 h reoxygenation time point showed great changes and autophagy occurred before necroptosis initiated, suggesting that autophagy triggered necroptosis. Furthermore, in renal IR, HSP70 exerted a protective effect through suppressing autophagy (Yeh et al. 2010). In present study, after HSP70 knockdown, autophagy and necroptosis were activated following cardiomyocyte H/R stimuli, which was in line with the previous research (Huang et al. 2013).

Regulated autophagy could protect cells against death (Huang et al. 2015). However, MIR leads to excessive autophagy activation, which results in organelle damage, cellular dysfunction, and cell death (Levine and Yuan 2005). As excessive autophagy triggers apoptosis and necroptosis during MIR, we utilized z-VAD-FMK, a pan-caspase inhibitor to block apoptosis. Furthermore, autophagy activator rapamycin and autophagy inhibitor 3-MA, necroptosis inhibitor necrostatin-1 were applied. The autophagy activator could exert similar effect on cardiomyocytes as HSP70 knockdown, which could promote autophagy and cell death (Xiao et al. 2015). Moreover, with HSP70 knockdown in vitro, autophagy inhibitor could still block necroptosis, indicating that HSP70 was not the only regulator of autophagy, which was consistent as previously reported (Wei et al. 2015). In line with previous researches (Dmitriev et al. 2015), after HSP70 knockdown, necroptosis inhibitor had no effect on autophagy. Thereafter, necroptosis inhibitor was not capable to downregulate cell death as autophagy inhibitor, indicating besides caspase-dependent apoptosis, caspase-independent apoptosis also participated in cell death, which was in parallel with the previous study (Anitha et al. 2014). However, HSP70 knockdown promoted more cell death than autophagy activator, suggesting that HSP70 had other protective functions during MIR as previously reported (Peng et al. 2010; Vicencio et al. 2015). In summary, HSP70 attenuated necroptosis of cardiomyocytes via suppressing autophagy after MIR.

Our study has several limitations that need to be mentioned. Firstly, as the ratio of LC3B to LC3A represented macroautophagy, knockdown of HSP70 also affects chaperone-mediated autophagy during MIR, which needs more investigation. Secondly, whether HSP70 and autophagy affect other necroptosis-associated molecules such as RIP13, A20 needs further study.