CpG oligodeoxynucleotide preconditioning improves cardiac function after myocardial infarction via modulation of energy metabolism and angiogenesis

2.1 Vector construction and lentivirus production

To construct the lentiviral vector expressing an shRNA targeting mouse TLR9, an shRNA expression cassette (7SK-shRNA-TLR9) was isolated from the psiRNA-TLR9 plasmid (InvivoGen, San Diego, CA) using ClaI and XbaI double digestion, which was inserted into the pLOXCMV-E/P lentiviral vector (Qi et al., 2015) to replace the CMV promoter by ClaI and SpeI (SpeI and XbaI has the same cohesive terminus) double digestion. The pLOXshRNA-TLR9 lentiviral vector was then constructed. For TLR9 overexpression in the lentiviral vector, the ORF of the mouse TLR9 gene was amplified by PCR primer pairs containing the EcoRI and XhoI restriction sites from the template of the pUNO1-mTLR9 plasmid (InvivoGen). The amplified fragments were then digested and ligated into the pLOXCMV-E/P vector linearized by the EcoRI and XhoI enzymes, thereby constructing the overexpression lentiviral vector pLOXCMV-TLR9 (pLOX-TLR9).

Lentiviruses were prepared in HEK293T cells by co-transfection of pLOX-TLR9 or pLOXshRNA-TLR9 together with the packaging plasmids pCMVR8.74 (#22036, Addgene, Cambridge, MA) and pMD2.G (#12259, Addgene) using the LipoFiter™ Liposomal Transfection Reagent (Hanbio Biotechnology, Shanghai, China). The viral supernatant was concentrated with the Lenti Virus Concentration Reagent (Biomiga, San Diego, CA) and titrated in HEK293T cells. High titer viruses (1 × 10⁹ PFU/ml) were re-suspended in PBS. A large number of small stock aliquots (10 µl) were made and frozen at −80°C for myocardium injection in mice. For animal experiments, a total of 30 µl of lentivirus was injected into three sites in the infarct zone of MI mouse models through an insulin syringe with an incorporated 29 G needle (BD).

2.2 Animal protocols

Male C57BL/6 mice at 8–10 weeks of age were randomized to receive an intraperitoneal (i.p.) injection of CpG-ODNs (1 mg/kg body weight) 2 hr prior to left anterior descending artery (LAD) ligation surgery as previously described (Lu et al., 2014). For certain experiments confirming the efficiency of lentiviral infection in vivo, lentiviruses were injected into the myocardium 3 days prior to MI induction to first upregulate or downregulate TLR9 expression. CpG-ODNs were provided by MWG-Biotech AG (Ebersberg, Germany) and completely phosphorothioate modified as previously described (Qi, Zheng, et al., 2013). The sequences and structures are as follows: KSK-CpG (5′-TCG TCG TTT TCG TCG TCG TTT T-3′), 1826-CpG (5′-TCC ATG ACG TTC CTG ACG TT-3′), and non-CpG (5′-TCC AGG ACT TCT CTC AGG TT-3′). The CG dinucleotides are indicated by italics and underlines. All mouse surgical and experimental procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of Jinan University. Sham controls were the same as the MI operation but without the LAD ligation. Animals were anaesthetized with ketamine (100 mg/kg, i.p.).

2.3 Echocardiography

Two weeks after the ligation of LAD, transthoracic echocardiography was performed using the Vevo^® 2100 ultrasound system (Visualsonics, Toronto, Canada) equipped with a high-frequency (30 MHz) linear array transducer. Parasternal long-axis, short-axis, and two apical four-chamber views were used to obtain two-dimensional and M-mode images. The left ventricular (LV) ejection fraction (LVEF) and LV fractional shortening (LVFS) were automatically calculated by the Vevo^® 2100 ultrasound system.

2.4 Histological analysis

Following serum collection, hearts were rapidly excised and transected into two segments through the infarct zone. One segment was fixed in 4% paraformaldehyde for 48 hr and embedded in paraffin. The other segment was embedded in Tissue-Tek optimal cutting temperature compound (OCT) (Sakura) for frozen sections. Paraffin sections (5 µm) were prepared and stained with hematoxylin and eosin (H&E) or with Masson trichrome for morphological and fibrosis analysis, respectively. The fibrotic area in each section was calculated by Image-Pro Plus version 6.0 software.

2.5 Serum VEGF levels

After echocardiography analysis, serum was collected and measured with a mouse VEGF ELISA Kit (R&D Systems, Minneapolis, MN) to evaluate the level of VEGF in the serum.

2.6 Cell culture

Primary cardiomyocytes were isolated from 2- to 3-day old mice (C57BL/6) using enzymatic tissue digestion as previously described (Ehler, Moore-Morris, & Lange, 2013; Shintani et al., 2013). Briefly, hearts were incubated with 0.05% trypsin/EDTA at 4°C overnight followed by digestion with collagenase type II. Cardiomyocytes were then collected based on a differential plating method, and they were cultured with Dulbecco's-Modified Eagle Medium (DMEM) supplemented with 5% fetal bovine serum (FBS) at 37°C. Adult primary cardiomyocytes were isolated as described previously (Ackers-Johnson et al., 2016).

H9C2 cells were cultured in DMEM supplemented with 10% FBS at 37°C. The culture media was changed to DMEM containing 2% FBS for CpG-ODN administration. For H/Iexperiments, both primary cardiomyocytes and H9C2 cells were cultured in DMEM medium containing 2% FBS and incubated in a hypoxic incubator adjusted at 1% O₂ and 5% CO₂. For CpG-ODN preconditioning, primary cells and the cell line were pretreated with CpG-ODNs (1 μg/ml) for 1 hr under normoxic conditions. Fresh media was then changed and followed by continuous hypoxic or normoxic incubation. Mouse cardiac microvascular endothelial cells (CMECs) were isolated and cultured as previously described (Qi et al., 2015). NIH-3T3 cells were cultured in DMEM supplemented with 10% FBS at 37°C and used to examine mouse TLR9 knockdown or overexpression mediated by lentiviral infection.

2.7 Cell viability assay

H9C2 cell viability was analyzed using the CCK-8 Cell Counting Kit (Beyotime Biotechnology, China) according to the manufacturer's instructions as previously described (Qi, Li, et al., 2013). For primary cardiomyocytes, a semi-quantitative method of immunofluorescence staining for anti-cTnT antibody was used. Briefly, surviving cardiomyocytes were specifically identified by cTnT antibody staining. Cell viability was evaluated by the relative ratio of cTnT-positive cells to all nuclei.

2.8 Intracellular ATP measurement

Intracellular ATP levels were measured using the Enhanced ATP Assay Kit (Beyotime Biotechnology, China) according to the manufacturer's instructions. Values were normalized by protein content.

2.9 Immunofluorescence staining

Immunocytochemistry was performed as previously described (Qi et al., 2015). Briefly, cells were fixed, permeabilized, washed, blocked, and incubated with anti-TLR9, anti-cTnT and anti-SERCA2 antibodies (Cell Signaling) overnight at 4°C. The cells were then washed and incubated with goat anti-rabbit IgG-FITC or goat anti-rabbit IgG-Cy5 (Santa Cruz, CA) for 1 hr at room temperature, and nuclei were stained for 10 min with DAPI (10 mg/ml; Sigma–Aldrich Co., St. Louis, MO). The stained cells were observed by fluorescence microscopy (Olympus, Tokyo, Japan). Image-Pro Plus version 6.0 software was used to quantify the fluorescent intensity of the cells.

For tissues samples, frozen sections (4 µm) were incubated with the primary antibodies directed against alpha-smooth muscle actin (α-SMA) (1:100 dilution, Abcam, CA), cleaved caspase 3 (1:400 dilution, Cell Signaling), or TLR9 (1:400 dilution, Cell Signaling) at 4°C overnight. After washing, the sections were incubated with the goat anti-rabbit IgG-FITC or goat anti-rabbit IgG-Cy5 (Santa Cruz) at room temperature for 2 hr. Nuclei were counterstained with DAPI. α-SMA signaling was used to evaluate the density of the capillary vessels. In addition, EGFP expression was directly examined using frozen sections to monitor the lentiviral infection in vivo.

2.10 In vitro capillary-like tube formation

Prechilled 48-well microtiter plates were coated with 100 µl growth factor-reduced Matrigel (BD Biosciences, San Jose, CA) and incubated for 2 hr at 37°C. Where indicated, CMECs (5 × 10⁴) were loaded into each well and incubated in DMEM containing 2% FBS for 16 hr. Tube formation was analyzed with an inverted light microscope.

2.11 Western blotting

Whole cell extract preparations were obtained using Cell Lysis Buffer (Beyotime Biotechnology, China) as previously described (Qi et al., 2015). Briefly, after separation by SDS-PAGE and transference, PVDF membranes were then blocked with 5% nonfat milk, washed briefly, incubated with primary antibodies (against TLR9 and β-actin) at 4°C overnight, and then incubated with corresponding HRP-conjugated secondary antibodies for 1 hr at room temperature. Protein bands were visualized by incubating the membranes with chemiluminescence reagents prior to exposure to a gel imaging system.

2.12 RT-PCR analysis

Total RNA was isolated using TRI Reagent (Molecular Research Center Inc., Cincinnati, OH). The quantity and purity of RNA was verified by measuring A260 and A280. cDNA was synthesized from total RNA (2 µg) with oligo (dT)₁₈ primers (0.5 µg) using the ReverTra Ace^® qPCR RT Kit (Toyobo, Japan). The primers used in this study are as follows: TLR9: 5′-GAC TTA CTG TTG GAG GTG CAG ACC-3′ (forward) and 5′-GAA CAC CAC GAA GGC ATC ATA GG-3′ (reverse) and β-actin: 5′-GCA CCA CAC CTT CTA CAA TGA G-3′ (forward) and 5′-TTG GCA TAG AGG TCT TTA CGG A-3′ (reverse). β-actin was used as an internal control. All samples were pre-denatured for 5 min at 94 °C according to the PCR Master Mix instructions (Thermo Fisher Scientific, Hudson, NH). The conditions for PCR amplification were as follows: 28 cycles of 94 °C for 30 s, 55 °C for 30 s, and 72 °C for 30 s. Amplified DNAs were further extended by an additional extension at 72 °C for 7 min. The PCR products were subjected to electrophoresis in 2% agarose gels and visualized by staining with ethidium bromide.

2.13 Statistical analysis

All data are presented as the mean ± SEM. Statistical analyses were performed using the unpaired Student's t-test. A p value less than 0.05 was considered statistically significant.

3.1 CpG-ODN preconditioning improves cardiac function after MI

To examine the role of CpG-ODN preconditioning in MI, we administered CpG-ODN and non-CpG-ODN (non-CpG, control) to mice 2 hr prior to MI induction. For some experiments, lentiviruses containing TLR9 shRNA were injected into the heart 3 days prior to MI induction. Cardiac function was evaluated by echocardiography 2 weeks after MI. As shown in Figure 1a, MI-induced decreases in LV systolic function were improved to various degrees by the preconditioning of CpG-ODNs but not by non-CpG. The LVEF was significantly increased by the preconditioning of 1826-CpG compared with non-CpG controls. Although the difference was not statistically significant, KSK-CpG also increased the LVEF values compared with non-CpG. Moreover, LVFS was significantly higher in mice preconditioned by CpG-ODNs compared with the non-CpG control (Figure 1b). However, TLR9 knockdown by lentivirus significantly blocked the cardiac protection of 1826-CpG against MI (Fiure. 1).

To evaluate the myocardium morphology and fibrosis, H&E and Masson trichrome staining were performed. As shown in Figure 2a, almost no integral cardiomyocytes were observed in the infarct zone in the non-CpG preconditioning group. However, integral cardiomyocytes were observed in CpG-ODN, particularly 1826-CpG, preconditioned mice. Masson trichrome staining showed that CpG-ODN preconditioning significantly attenuated the development of cardiac fibrosis in mice with MI (Figures 2b and 2c). Moreover, CpG-ODN preconditioning could reduce the apoptosis level in the infarcted myocardium, which was confirmed by cleaved caspase three staining (Figure 2d). Taken together, these results demonstrate that CpG-ODN preconditioning improves cardiac function after MI.

3.2 CPG-ODN reconditioning protects cardiomyocytes from hypoxia

Given that cardiomyocyte cellular injury is mainly induced by hypoxia and ischemia during MI, the potential protective effects of CpG-ODN preconditioning on cardiomyocytes were further examined in vitro. Both the H9C2 cell line and primary cardiomyocytes were pretreated with 1826-CpG (1 μg/ml) for 1 hr prior to H/I induction. As shown in Figure 3a, H/I induced marked injury of H9C2 cells from 24 to 48 hr, but the decrease in cell survival was significantly abolished by 1826-CpG preconditioning. For primary cardiomyocytes, both neonatal and adult cardiomyocyte viability were markedly suppressed by H/I treatment for 48 hr, which is consistent with results in H9C2 cells. However, preconditioning of 1826-CpG significantly increased the cell survival under H/I (Figures 3b and 3c). These findings suggest that CpG-ODN preconditioning protects cardiomyocyte survival against hypoxic and ischemic conditions.

3.3 CpG-ODN preconditioning increases TLR9 expression in cardiomyocytes accompanied by metabolism regulation

We next examined the expression of TLR9 in cardiomyocytes treated with CpG-ODN preconditioning by immunocytochemistry. As shown in Figure 4a, 1826-CpG preconditioning promoted TLR9 expression in H9C2 cells even under normoxic conditions. Moreover, marked TLR9 expression was observed in H9C2 cells preconditioned by 1826-CpG 6 hr after H/I induction. Consistent with the cardiomyocyte cell line, 1826-CpG preconditioning also increased the expression of TLR9 in neonatal primary mouse cardiomyocytes under both normoxic and H/I conditions (Figures 4b and 4c). These results indicated that TLR9 may be involved in CpG-ODN-mediated cardiomyocyte protection against H/I treatment.

Indeed, it has been reported that TLR9 plays important roles in the cellular protection of cardiomyocytes by regulating SERCA2/AMPK pathway-mediated cellular metabolism (Shintani et al., 2014, 2013). We therefore examined the expression of SERCA2 in cardiomyocytes preconditioned with 1826-CpG. As shown in Figure 5a, SERCA2 expression was decreased in H9C2 cells treated with 1826-CpG under H/I although the expression of TLR9 was increased. However, there was no difference in SERCA2 expression in H9C2 cells directly subjected to H/I (Figures 5b and 5c). Given that SERCA2 is an Ca²⁺ ATPase that governs ATP production and energy metabolism (Shintani et al., 2014), ATP production was also analyzed in H9C2 cells treated with 1826-CpG. As shown in Figure 5d, 1826-CpG preconditioning significantly decreased ATP production in H9C2 cells under H/I from 6 to 24 hr compared with non-CpG. Previous studies have determined that decreases in ATP is a key trigger for the activation of AMP-activated kinase (AMPK) in cardiomyocytes, which thereby increases cellular tolerance against metabolic stress (Shintani et al., 2014, 2013). AMPK activation was subsequently analyzed by western blotting to further confirm whether the SERCA2/ATP/AMPK pathway was involved in CpG-ODN-mediated cardiomyocyte protection. We found that the preconditioning of 1826-CpG increased AMPK phosphorylation in H9C2 cells treated with H/I (Figure 5e). Importantly, a great increase in AMPK phosphorylation was also observed in infarcted myocardium tissues after preconditioning with CpG-ODN, which is in addition to the slight Akt phosphorylation (Figure 5f). These findings further reveal that the SERCA2/ATP/AMPK pathway may play an important role in the CpG-ODN preconditioning-mediated cellular protection of cardiomyocytes through TLR9 activation.

3.4 CpG-ODN preconditioning promotes angiogenesis after MI

It is known that angiogenesis plays a critical role in the repair of damaged hearts. However, a series of studies have suggested that continuous treatment with CpG-ODNs suppresses angiogenesis under different pathological or physiological statuses (Wu, Cui, Dick, & Liu, 2014; Wu et al., 2016; Yan et al., 2015). To further explore the potential involvement of CpG-ODN preconditioning in angiogenesis after MI, angiogenesis was analyzed by immunofluorescence staining for α-SMA. As shown in Figures 6a and 6b, CpG-ODN preconditioning significantly promoted angiogenesis compared with non-CpG although MI also induced angiogenesis to a certain extent compared with the sham group. Moreover, serum VEGF production was also increased by CpG-ODN preconditioning compared with non-CpG (Figure 6c). The in vitro angiogenesis of endothelial cells preconditioned by CpG-ODN was further examined by tube formation assays. We found that 1826-CpG preconditioning promoted the tube formation of CMECs as VEGF-treated endothelial cells (Figure 6d). However, the VEGF-mediated tube formation of endothelial cells was suppressed by continuous treatment with 1826-CpG (Figure 6e). These results indicate that different CpG-ODN-mediated effects on angiogenesis depend on the administration method, including preconditioning or continuous treatment.

3.5 Persistent expression of TLR9 fails to protect cardiac function from MI

Finally, we examined whether persistent TLR9 activation had the same protective effects on MI using lentiviral-mediated stable expression of the TLR9 gene. Surprisingly, overexpression of the TLR9 gene around the infarcted zone had no effect on improving cardiac function compared with the control group, which was injected only with PBS. Furthermore, TLR9 knockdown by lentiviral-mediated shRNA also had no influence on cardiac function compared with the control group. Both LVEF and LVFS were not significantly regulated by the persistent overexpression or knockdown of TLR9 compared with the control group (Figures 7a and 7b). The efficient overexpression and knockdown of TLR9 was verified both in vitro (NIH-3T3 cells) and in vivo (mice) (Figure 7c–f). Furthermore, the efficient infection of lentiviruses in vivo was further confirmed by expression of the EGFP reporter gene in the myocardium (Figure 7g). These findings further indicate that the persistent activation of TLR9, the target of CpG-ODNs, failed to improve cardiac function. However, TLR9 overexpression suppressed the tube formation of endothelial cells, indicating that persistent TLR9 activation inhibits angiogenesis (Figure 7h).