Hypoxia-inducible factor 2-alpha-dependent induction of IL-6 protects the heart from ischemia/reperfusion injury

HIF2α has cardioprotective effects in cardiomyocytes during simulated hypoxia/reperfusion

Studies have shown that HIF2α plays a protective role during MIRI, and therefore, we investigated the underlying mechanisms. To assess the effect of HIF2α in H9c2 cardiomyocytes during hypoxia/reoxygenation (H/R), we generated a lentiviral vector to knock down the endogenous HIF2α gene in H9c2 cardiomyocytes. qPCR and Western blot results showed that the downregulation efficiency of shRNA 1 and 2 reached more than 60% (Figure 1A).

Figure 1. Effects of HIF2α on cell apoptosis and cell survival of cardiomyocytes during hypoxia. (A) qPCR and Western blot analysis of the efficiency of HIF2α gene- specific downregulation of lentiviral vectors. (B) CCK-8 detection of the effect of downregulation of the HIF2α gene on early cell apoptosis and cell survival. (C) Application of FACS-Annexin V/PI double staining to detect the effect of downregulation of the HIF2α gene on cell apoptosis. n = 3 per group. Data represent the mean ± SEM. **P<0.01 versus the indicated group.

After knocking down the endogenous HIF2α gene, we subjected H9c2 cardiomyocytes to H/R treatment. Our results showed that the apoptotic rate of cardiomyocytes increased significantly after H/R. Moreover, after knockdown of the HIF2α gene in H9c2 cells, the proportion of apoptotic cells increased further (Figure 1C). In addition, we found that knocking down the endogenous HIF2α gene reduced cell survival after H/R (Figure 1B), suggesting that HIF2α confers cardioprotective effects on H9c2 cardiomyocytes during I/R.

HIF2α regulates the IL-6 level in cardiomyocytes during simulated hypoxia/reperfusion

Our results showed that compared with the cells in a normoxic environment, those under H/R conditions expressed significantly increased levels of HIF2α and IL-6 proteins. This result suggests that H/R stimulates the expression of HIF2α and IL-6. After knockdown of endogenous HIF2α, the IL-6 protein level was also low in the mice (Figure 2A).

Figure 2. Expression of IL-6 during the H/R process in cardiomyocytes after knocking down endogenous HIF2α. (A) H9c2 myocardial cell line: Western blot to detect the expression level of IL-6 after knocking down endogenous HIF2α. (B) AC16 human cardiomyocyte line: Western blot to detect the expression level of IL-6 after knocking down endogenous HIF2α. (C) H9c2 cardiomyocyte cell line: qPCR to detect the expression level of IL-6 mRNA after knocking down endogenous HIF2α. (D) The results of ChIP: Diagram of the IL-6 promoter. Quantitative PCR analysis of anti- HIF2α antibody-precipitated and control IgG antibody-precipitated chromatin from cardiomyocytes cultured in 21 or 1% O₂. HRE: hypoxia response element. n = 3 per group. Data represent the mean ± SEM. *P<0.05, **P<0.01, ***P<0.001 versus the indicated group.

Based on the finding that HIF2α regulates IL-6 protein levels in the murine in vitro model of H/R, we performed studies in AC16 human cardiomyocytes to verify whether the observed changes in murine IL-6 levels also occur in AC16 human cardiomyocytes. Our results also showed that the H/R process stimulated the expression of IL-6 in AC16 human cardiomyocytes. After knocking down endogenous HIF2α, we observed a reduction in the protein expression levels of HIF2α and IL-6 in these AC16 human cardiomyocytes during H/R (Figure 2B). Additionally, we used qPCR to detect the expression of IL-6 mRNA during H/R. Our results were consistent with those of the Western blots. The IL-6 mRNA level in H9c2 cardiomyocytes increased during H/R. However, after knockdown of the HIF2α gene, the IL-6 mRNA levels decreased significantly (Figure 2C). Together, these data suggest that HIF2α regulates IL-6 levels in cardiomyocytes during simulated H/R conditions.

HIF2α transcriptionally regulates IL-6 expression in cardiomyocytes

Next, a chromatin immunoprecipitation experiment was performed to further verify that HIF2α has a direct regulatory effect on IL-6 at the transcriptional level. First, by sequence analysis, we found two original HIF transcription factor response elements (HRE) in the IL-6 gene promoter regions (Figure 2D). The results from our chromatin immunoprecipitation experiments showed that the DNA from the promoter region of the IL-6 gene recruited by the HIF2α antibody showed a greater enhancement under hypoxic conditions than under normoxic conditions. However, we found no significant differences in terms of the DNA of the promoter region pulled down by the IgG antibody (Figure 2D). Taken together, our results indicate that HIF2α directly regulates the transcription of IL-6 in cardiomyocytes.

HIF2α confers cardioprotective effects under myocardial I/R injury

Then, we conducted in vivo studies in mice to investigate whether HIF2α confers protective effects on MIRI. We constructed a type 9 adeno-associated virus vector (AAV9- shHIF2α) that specifically downregulates HIF2α. Four weeks after injection of the AAV (5*10^10 PFU) vector via the mouse tail vein, we ligated the LAD coronary artery in each mouse to establish an in vivo myocardial ischemia/reperfusion (I/R) model (Figure 3A). Our results showed that the expression levels of HIF2α and IL-6 in the myocardium increased significantly after myocardial I/R, and this change was reversed with AAV9-shHIF2α delivery (Figure 3B). Additionally, we isolated the hearts of mice 24 h after reperfusion and extracted the left ventricular blood for LDH and TnI determination, while we used TTC and Evans blue staining to evaluate the myocardial infarction areas. Our results showed no significant differences in terms of the myocardial risk areas between the sham groups. However, the myocardial infarction area in the shHIF2α knockdown group was significantly increased compared to that in the control group (Figure 3C). We also found that the myocardial injury markers LDH and TnI were significantly increased in the I/R group and were further enhanced with HIF2α knockdown (Figure 3D, 3E). These results suggest that HIF2α deficiency in vivo exacerbates MIRI.

Figure 3. The effect of in vivo HIF2α knockdown on myocardial infarction during acute myocardial ischemia/reperfusion. (A) Myocardial I/R model was established 4 weeks after AAV injection. (B) Western blots were performed to examine the HIF2α and IL-6 protein expression after HIF2α knockdown in vivo. n = 3 per group. (C) TTC and Evans blue staining after mouse heart I/R. n = 8 per group. (D) Effect of HIF2α knockdown on myocardial LDH release subsequent to I/R. n = 5 per group. (E) Effect of HIF2α knockdown on myocardial TnI release subsequent to I/R. n = 5 per group. Data represent the mean ± SEM. **P<0.01 versus the indicated group.

IL-6 knockdown aggravates myocardial I/R injury

Next, we sought to determine whether IL-6 is involved in the protective effects of HIF2α during myocardial I/R. First, we constructed a type 9 adeno-associated virus vector (AAV9-shIL-6) that specifically downregulates the IL-6 gene. Our results showed that both the mice in the AAV-scramble and AAV-shIL-6 groups had similar cardiac risk areas (Figure 4A, 4B). However, the mice in the AAV-shIL-6 group had larger myocardial infarction ranges than the mice in the AAV-scramble group (Figure 4A, 4C), with higher LDH and TnI values (Figure 4D, 4E). These findings indicate that in vivo knockdown of IL-6 increases the range of myocardial infarction in mice. That is, IL-6 knockdown in vivo exacerbates MIRI. Our observations confirmed that IL-6 is important for the protective effects of HIF2α during myocardial I/R.

Figure 4. The effect of in vivo IL-6 knockdown on the myocardial infarction area and markers of myocardial injury. (A) TTC and Evans blue staining after mouse heart I/R. (B) Mice in the AAV-scramble group and AAV-shIL-6 group experienced similar cardiac risk areas. n = 8 per group. (C) The range of myocardial infarction in the AAV- shIL-6 group was larger than that in the AAV-scramble group. n = 8 per group. (D) Effect of IL-6 knockdown on the LDH release subsequent to I/R. n = 5 per group. (E) Effect of IL-6 knockdown on the troponin I release subsequent to I/R. n = 5 per group. Data represent the mean ± SEM. *P<0.05, **P<0.01 versus the indicated group.

Recombinant IL-6 treatment diminishes I/R injury in IL-6 knockdown mice

Based on our studies in mice with the IL-6 gene downregulated in cardiomyocytes, we hypothesized that treatment with recombinant IL-6 could provide cardioprotection from I/R injury. Mice were exposed to I/R injury with intravenous injection of IL-6 at a dose of 10 μg/kg or vehicle 15 min before the onset of myocardial ischemia. After a 30-min exposure to myocardial ischemia followed by 24 h of reperfusion, we found significantly reduced myocardial infarct sizes and troponin I serum levels in the mice treated with recombinant IL-6 compared with those treated with vehicle (Figure 5A, 5B). These findings suggest that supplementation with recombinant IL-6 reduced the extent of myocardial infarction.

Figure 5. Effect of exogenous IL-6 supplementation on myocardial infarct size and troponin I in endogenous IL-6 knockdown mice or HIF2α deficiency mice. (A) Representative image and quantitative analysis of TTC and Evans blue staining after mouse heart I/R with IL-6 supplementation in IL-6 deficiency mice. n = 8 per group. (B) The results of troponin I release subsequent to I/R. n = 5 per group. (C) Representative image and quantitative analysis of TTC and Evans blue staining after mouse heart I/R with IL-6 supplementation in HIF2α deficiency mice. n = 8 per group. (D) The results of troponin I analysis. n = 5 per group. Data represent the mean± SEM. *P<0.05, **P<0.01 versus the indicated group.

IL-6 treatment rescues the HIF2α-knockdown mice

Next, we investigated whether treatment with recombinant IL-6 rescues mice with downregulated HIF2α. After knocking down the HIF2α gene, we treated mice with 10 μg/kg recombinant IL-6. The results of Evans blue staining showed similar cardiac risk areas after subsequent exposure to 30 min of ischemia and 24 h of reperfusion between the IL-6-treated and control mice (Figure 5C). However, the area of myocardial infarction and the level of troponin I were both significantly lower in the IL-6-treated mice than in the control mice (Figure 5C, 5D). Thus, recombinant IL-6 treatment is associated with attenuated MIRIs in the HIF2α-knockdown mice.

The PI3K/Akt and STAT3 signaling pathways are involved in HIF2α/IL-6- induced cardioprotection

After showing that treatment with recombinant IL-6 reduced the extent of myocardial injury following I/R, we investigated whether administration of recombinant IL-6 activates cardioprotective pathways, e.g., via activation of survival kinases such as Akt and the STAT3 pathway. As described above, we treated the IL-6 knockdown mice and the HIF2α deficient mice with recombinant IL-6 or vehicle and examined the phosphorylation of Akt and STAT3. Our results showed that phosphorylated STAT3 (Tyr705) and AKT (Ser473) in the IL-6-treated hearts were significantly increased compared with the control hearts (Figure 6A, 6B).

Figure 6. The roles of the PI3K/Akt and STAT3 pathways in the protective effects of HIF2α/IL-6. (A, B) Western blotting was performed to determine the protein expression of AKT, Phos-AKT, STAT3 and Phos-STAT3 in myocardium. n = 3 per group. (C) Representative image and quantitative analysis of infarct staining with different treatments. n = 8 per group. Data represent the mean ± SEM. *P<0.05, **P<0.01, ***P<0.001 versus the indicated group.

To determine the roles of PI3K/Akt activation and STAT3 activation in the cardioprotection of HIF2α /IL-6 against myocardial I/R, we administered the PI3K inhibitor wortmannin (1 mg/kg) and the STAT3 inhibitor Stattic (2 mg/kg) to the mice 60 min prior to ischemic insult. Our data indicate that blockade of PI3K/Akt as well as the STAT3 pathway abolished the HIF2α/IL-6-induced cardioprotective effects (Figure 6C). These findings suggest that HIF2α/IL-6 is dependent on PI3K/Akt activation and STAT3 activation to mediate its effects.

Animals

Adult male C57BL/6 mice (8 to 10 weeks old) from the Shanghai Slac Animal Laboratory were maintained and handled in accordance with the Guidelines for Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication, 8^th Edition, 2011). The Institutional Review Board of The First Affiliated Hospital of USTC approved all animal procedures (Hefei, China).

Myocardial I/R model and IL-6 treatment

We performed conventional surgical ligations of the left anterior descending (LAD) coronary arteries in mice as previous studied [38, 39]. Briefly, we anesthetized the mice with intraperitoneal injections of ketamine (50 mg/kg) and pentobarbital sodium (50 mg/kg), orally intubated them, and ventilated them. We maintained their core body temperature constantly at 37° C. Medial sternotomies were then performed using an electrocautery pen to visualize the proximal LAD in each mouse and ligate it. The coronary artery remained occluded for 30 min, after which we cut the sutures and allowed the vessel to reperfuse. We closed the sternum and skin separately and allowed the animals to recover. In the PI3K inhibition and STAT3 inhibition experiments, mice were treated with the PI3K inhibitor wortmannin (1 mg/kg, Selleck Chemicals, No. S2758, USA) and the STAT3 inhibitor Stattic (2 mg/kg, Selleck Chemicals, No. S7024, USA) by intraperitoneal injection 60 min prior to ischemic insult.

Recombinant IL-6 treatment

Recombinant mouse IL-6 was purchased from R&D Systems (No. 406-ML-025, USA) and dissolved IL-6 in 0.9% NaCl solution. In the IL-6 group, 10 μg/kg recombinant mouse IL-6 was intravenously administered 15 min prior to the onset of myocardial ischemia, whereas the control group received the same volume of 0.9% NaCl over the same period as described previously [34]. There was no difference in mortality between the groups. When the injection was complete, we began the in situ myocardial ischemia induction.

Myocardial area-at-risk and infarct size determination

We measured the area at risk and infarct size. In brief, 24 h after reperfusion, we anesthetized the mice, ventilated them, and catheterized them through the common carotid artery. We performed median sternotomies and religated the LAD coronary arteries in the same location as before. We injected Evans blue dye (1.2 mL of a 4.0% solution; Sigma, No. E2129, USA) into the carotid artery catheter to differentiate the cardiac ischemic zone from the nonischemic zone. We then rapidly excised the hearts and sectioned them serially and incubated the sections in 1.0% 2,3,5-triphenyltetrazolium chloride (Sigma, No. T8877, USA). We weighed each of the five 1-mm thick myocardial slices, and a blinded observer assessed the areas of infarction, risk, and nonischemic left ventricle using computer-assisted planimetry (NIH ImageJ 1.37).

Cell culture and treatment

The fetal cardiomyocyte-derived H9c2 (American Type Culture Collection) cells were cultured in Dulbecco’s modified Eagle’s medium (Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum, 100 U ml–1 penicillin, 100 μg ml–1 streptomycin and 110 mg ml–1 sodium pyruvate in a humidified atmosphere containing 5% CO₂ at 37° C. Human cardiomyocytes (HCMs) were purchased from ScienCell (Carlsbad, CA, USA). For the H/R experiments, we first perfused H9c2 cells in normal Hank's solution with a gas mixture of 21% O₂-5% CO₂ at 37° C, pH 7.4. For simulation of ischemia, Hank's solution was switched to pH 7.4 at 37° C without glucose or calcium, and then, the cells were aerated with a gas mixture of 95% N₂-5% CO₂ for 10 h. To simulate reperfusion, we treated the cells again with normal Hank's solution and a gas mixture of 21% O₂-5% CO₂ at 37° C, pH 7.4 for 2.5 h. We included cells under normoxic conditions throughout the experiments as controls.

Construction of lentivirus expressing shRNA for rat Hif2α

The rat HIF2α RNAi target sequence was 5'-CAGCACTGCTTCAGTGCCAT- GACAA-3'. We used a nonrelated, scrambled RNA sequence without a match in the rat genomic sequence as a control (5'-TCAGTCTTCATGGAACCTT-3'). We generated lentiviruses harboring these RNAi constructs using a lentivirus expression system (Invitrogen Corporation, MA, USA) as published. The constructs were verified by sequencing. We produced viral particles by third-generation packaging in 293FT cells and concentrated lentiviral stocks using ultracentrifugation.

Viability assays

We assessed cell viability using a cell counting kit-8 (CCK-8, No. C0038, Beyotime Institute of Biotechnology, Haimen, China). We seeded H9c2 cells in 48-well plates for 24 h. Then, we infected the cells with lentivirus for 48 h before subjecting them to H/R treatment. Cells were provided with fresh media, and 10 μl of CCK-8 solution was added to every well. We then incubated the plates under normal conditions for 2 h. We measured optical density values at 470 nm using a microplate reader (Multiskan MK33, Thermolab systems, Helsinki, Finland). We repeated each experiment at least three times.

Analysis of apoptosis

We used an Annexin V-FITC/PI apoptosis detection kit (Beyotime Institute of Biotechnology, No. C1062S, Haimen, China) to identify the presence of cell apoptosis according to the manufacturer's instructions. After treatments, cells were harvested and resuspended in 200 μl of binding buffer. Then, we incubated the cells with 10 μl of Annexin V-FITC and 5 μl of PI in the dark for 15 min. We assessed the apoptosis rate by flow cytometry analysis.

Lactate dehydrogenase (LDH) and cardiac troponin-I (TnI) release

We assessed necrotic cell death by measuring supernatant LDH activity. We used a spectrophotometric kit (Nanjing Jiancheng, No. A020-1, China) according to the manufacturer’s instructions. In brief, we collected 20 μl supernatant samples in a 96- well plate and then added 25 μl of matrix buffer and 5 μl of coenzyme I to each well. The mixture was incubated at 37° C for 15 min. Then, we added 25 μl of 2,4- dinitrophenylhydrazine to each well and incubated the cells again at 37° C for 15 min. We finally added 250 μl of 0.4 M NaOH to each well and incubated the plates at room temperature for 5 min. We measured the absorbance value at 450 nm with a spectrophotometer (BioTek, VT, USA) and calculated the LDH activity. We measured plasma TnI levels as an indicator of cardiomyocyte damage using a mouse TnI ELISA kit (Elabscience Biotechnology, No. E-EL-M1203c, China) according to the manufacturer’s instructions.

Quantitative real-time PCR

We isolated total RNA samples from heart tissues using TRIzol Reagent (Thermo Fisher Scientific, No. 15596026, USA). We performed real-time PCR using SYBR Green detection of PCR products in real time with the ABI Prism 7700 Sequence Detection System (Applied Biosystems, CA, USA) for relative quantification of HIF2α products. The primers used for rat HIF2 α were as follows: forward, 5'- GAGAAGGTGACTCAGCGACA-3'; and reverse, 5'-GTTCTTGCTCCCTCCAACTC-3'. The primers used for rat IL-6 were as follows: forward, 5'-CATTCTGTCTCGAGCCCACC-3'; and reverse, 5'- GCTGGAAGTCTCTTGCGGAG-3'. GAPDH RNA was amplified as a reference standard. The sequences of rat GAPDH primers were as follows: forward, 5'- TGGAGTCTACTGGCGTCTT-3'; reverse, 5'-TGTCATATTTCTCGTGGTTCA-3'. We conducted the reactions in triplicate by heating the reactants to 95° C for 5 min, followed by 40 cycles of 94° C for 30 s, 58° C for 30 s, and 72° C for 30 s.

Western blot analyses

Left ventricular tissues were homogenized at 4° C with a homogenizer in lysis buffer containing 1% Triton X-100, 0.5% deoxycholate, and 5 mmol/L 2-mercaptoethanol. We scraped cell extracts into lysis buffer containing 20 mmol/L Tris-HCl (pH 7.4), 6 mM urea, and 200 mmol/L potassium chloride with a protease inhibitor cocktail, followed by vigorous vertexing and cooling on ice for 15 min before a 15-min centrifugation at 12,000 g. We then ran the samples on SDS-PAGE. We transferred the proteins on gels onto polyvinylidene fluoride microporous membranes (Bio-Rad, CA, USA) and probed them with primary antibodies against HIF2 α (No. ab109616, Abcam, USA; 1:1000), anti-GAPDH (internal control; No. KC-5G4, Kangcheng, Shanghai, China; 1:8000), anti-IL-6 (No. ab9324, Abcam, USA; 1;1000), anti-AKT (No. #4685, Cell Signaling Technology, USA; 1:2000) and anti-p-AKT (No. #4051, Cell Signaling Technology, USA; 1:2000). The secondary antibodies were coupled with horseradish peroxidase (Cell Signaling Technology, 1:6000). We visualized the immunoreactions using an enhanced ECL detection kit (No. GERPN2209, Amersham Pharmacia Biotech, USA) and then exposed them to film before quantifying them with a video documentation system (Gel Doc 2000, Bio-Rad).

Chromatin immunoprecipitation

We performed chromatin immunoprecipitation (ChIP) experiments using a Simple ChIPTM Enzymatic Chromatin IP Kit (Cell Signaling Technology) according to the manufacturer’s instructions. Cells were treated with phosphatidylethanolamine and then washed twice with phosphate-buffered saline before crosslinking with 1% formaldehyde at 37° C for 10 min. Next, we isolated nuclear fractions and enzymatically digested the DNA in them. We verified the quality and length of the digested fragments on 1.5% agarose gels. We immunoprecipitated protein-DNA complexes using antibodies against HIF2α (No. NB100-122, Novus Biologicals, USA) and normal rabbit IgG (No. #3900, Cell Signaling Technology, USA). The purified DNA was quantified by qRT-PCR with SYBR Green PCR reagents (Toyobo, Osaka, Japan) to detect the enrichment of the IL-6 promoter with the specific primers 5'- CAGGCATCACTACTGTCTGTG-3' (forward) and 5'-TTCTGACCTA-AGTTTCTTCAG-3' (reverse) and normalized to the total input control.

AAV9 vector generation and transfection

We prepared AAV9-shHIF2α, AAV9-shIL-6, and AAV9–mCherry vectors as described [40]. We injected 1 x 1012 vg of AAV9-mCherry, AAV9-shHIF2α or AAV9- shIL-6 intravenously into the tail veins of 4- to 5-week-old male C57 mice as described previously [41]. We conducted sham or myocardial ischemia operations 4 weeks after the AAV9 injections.

Statistical analysis

Data are shown as the mean ± SEM. We used GraphPad Prism® version 6.0 software to perform unpaired, 2-tailed t tests for comparisons between 2 groups and ANOVA or repeated ANOVA followed by the Bonferroni post hoc test for multiple comparisons. We considered P values < 0.05 as statistically significant.