Hyperglycemia was induced in male Sprague-Dawley rats (200–250 g) by a single intraperitoneal injection of streptozotocin (STZ, 65 mg/kg body weight; Sigma-Aldrich) which was dissolved in citrate buffer (pH 4.5). The control group received an equivalent volume of citrate buffer injection alone. Seven days later, MI was conducted by ligation of the left anterior descending coronary artery (LAD) while the control group underwent a sham-operation. In the inhibition experiments, 0.5 mg/kg PD98059 (19), a specific inhibitor of MEK (upstream of ERK1/2), or 1.0 mg/kg SB203580 (20), a selective inhibitor of p38 MAPK, or an equivalent volume of vehicle was intraperitoneally administered 30 min prior to operation and once every day on Days 1–4 after MI. All procedures were performed in accordance with the Guidelines of the Hubei Council of Animal Care and approved by the Animal Use Subcommittee at the Huazhong University of Science and Technology, China.

Blood samples were obtained from tail-tip bleedings, and blood glucose concentration was measured with a glucometer (Roche).

The CSCs were isolated from the hearts of male Sprague-Dawley rats by a magnet-activated cell sorting (MACS) system (Dynal Biotech) as described previously (6,8). Briefly, the heart was excised and the aorta was cannulated rapidly. The heart was then perfused with Ca²⁺-free Tyrode’s solution for 10 min, and digested by collagenase (0.5 mg/ml; Roche) and trypsin (0.05 mg/ml; Sigma-Aldrich) at 37°C for 30 min. Next, the heart tissue was chopped, and the cell suspension was filtered with a strainer (Becton Dickson). Afterwards, cells were incubated with a rabbit anti-c-kit antibody (1:150; Santa Cruz Biotechnology, Inc.) and separated by using sheep anti-rabbit immunomagnetic microbeads (Dynal Biotech). Small round cells, containing most of the c-kit⁺ population, were separated. These c-kit+ CSCs were cultured for 3–5 days with Dulbecco’s modified Eagle’s medium/Ham’s Nutrient Mixture F12 (1:1) (DMEM/F12; Sigma-Aldrich) containing 15% fetal bovine serum (FBS; Gibco), basic fibroblast growth factor (bFGF, 10 ng/ml; Sigma-Aldrich), epidermal growth factor (EGF, 20 ng/ml; Sigma-Aldrich) and leucocyte inhibitory factor (LIF, 10 ng/ml; Sigma-Aldrich) at 37°C. After recovery, these CSCs were used for subsequent experiments.

Ventricular cardiomyocytes were isolated from neonatal Sprague-Dawley rats and submitted to primary culture as reported previously (21). After recovery, these cells were incubated in FBS-free DMEM with glucose (5.5 or 25 mM), glucose (5.5 or 25 mM) + IL-1β (1 ng/ml; Sigma-Aldrich), glucose (5.5 mM) + IL-1β + PD98059 (10 μM), or glucose (5.5 mM) + IL-1β + SB203580 (10 μM) for 12 h. The supernatants of cultured cardiomyocytes were used as conditioned media (CM). SCF mRNA in cells was detected by RT-PCR, and SCF protein was analyzed by ELISA.

Total RNA was extracted from peri-infarcted myocardium of the left ventricle and cultured cardiomyocytes with TRIzol reagent (Invitrogen Life Technologies). RT-PCR was carried out using the pairs of primers as following for semi-quantitative assessment. SCF: sense, 5′-TGTTTTG CCTAGTCATTGTTG-3′ and anti-sense, 5′-TGTCATTCCTA AGGGAACTG-3′, yielding a 404 bp product; β-actin: sense, 5′-CGTTGACATCCGTAAAGA-3′ and anti-sense, 5′-AGCC ACCAATCCACACAG-3′, yielding a 173 bp product. The products of PCR were separated by 1.5% agarose gel electrophoresis and visualized under UV using a gel documentation system (Bio-Rad). β-actin was used as an internal standard to verify equal PCR product loading for each experiment.

Tissue samples were homogenized and separated by SDS polyacrylamide gel (12.5%) electrophoresis followed by electrophoretic transfer of proteins from the gel to nitrocellulose membranes (Bio-Rad). The membranes were probed with the following primary antibodies: rabbit anti-SCF, mouse anti-β-actin, mouse anti-P-p38, rabbit anti-p38, mouse anti-P-ERK, or rabbit anti-ERK1/2 (1:600; all from Santa Cruz Bioetchnology, Inc.) overnight at 4°C. Bands were visualized by using corresponding horseradish peroxidase (HRP)-conjugated anti-biotin antibody and enhanced chemiluminescence reagents (Pierce Biotechnology, Inc.).

On Day 5 after MI, heart slices from the peri-injured myocardium were prepared for immunohistochemical staining. Slices were incubated with rabbit polyclonal antibodies (1:100; PeproTech) against rat SCF overnight at 4°C (for negative control studies, the antibodies were substituted by phosphate-buffered saline). Endogenous peroxidase was blocked by 0.3% H₂O₂ for 20 min at room temperature and the biotin-conjugated anti-rabbit immunoglobulin (1:200; Dako) was used as the secondary antibody. After incubation with streptavidin peroxidase, visualization of peroxidase localization was performed using diaminobenzidine-hydrogen peroxide (DAB-H₂O₂) substrate to give a brown color.

To assess the migration of CSCs in vivo after MI, cultured CSCs were labeled with 5-bromo-2′-deoxyuridine (BrdU; Sigma-Aldrich) as described previously (22). The BrdU-labeled CSCs (1×10⁶) were injected into the AV-groove of the rat heart followed by LAD ligation or sham operation. On Day 5 post-MI, histological examination was performed to visualize the localization of the BrdU-labeled CSCs using mouse anti-BrdU antibody (1:150; Zymed) and TRITC-conjugated goat anti-mouse IgG antibody. The numbers of BrdU⁺ cells homing to the peri-infarcted area were counted in 5 randomly selected sections per heart, and for each section 10 fields were randomly chosen by high-power fields.

Three weeks after LAD ligation or sham operation, rats were anesthetized again. The right carotid artery was cannulated with a Millar Micro-Tip pressure transducer catheter. After obtaining the arterial blood pressure and heart rate, the catheter was advanced into the left ventricle (LV) to measure the systolic and end-diastolic pressures as well as the rate of pressure development (+dP/dt_max) and rate of relaxation (−dP/dt_min) of the LV.

A sensitive ELISA procedure was used to quantify immunoreactive SCF released into the supernatant of the cultured cardiomyocytes. The ELISA was performed according to the instructions provided by the manufacturer (R&D systems). A polystyrene microplate (96 wells) was coated with a rabbit polyclonal anti-SCF antibody, and recombinant rat SCF was used as the standard.

Chemotaxis experiments were performed using a 48-well chemotaxis chamber technique (Neuro Probe) as previously described (23). Briefly, 25 μl of DMEM + glucose (5.5 mM), virus conditioned media (CM), or CM + SCF-Ab (rabbit anti-rat SCF, 1:600; Santa Cruz Biotechnology, Inc.) was placed in the lower chamber. A polycarbonate membrane with a 5-μm pore size separated the upper and lower chamber. CSCs resuspended in DMEM (50 μl) were placed in each well of the upper chamber. The chamber was then incubated for 3 h at 37°C in a humidified atmosphere with 5% CO₂ and then disassembled. The membrane was removed and scraped to remove non-migrating CSCs from the upper surface. Then the membrane was fixed and stained. The numbers of CSCs that had migrated to the lower surface of the membrane were counted in ten random high-power fields (HPFs) by light microscopy, and a chemotactic index (CI) was calculated to express stimulated migration. Each assay was performed in triplicate wells. CI = Stimulated   migration   ( Number   of   CSCs / HPF ) Random   migration   ( Number   of   CSCs / HPF )

All data are expressed as means ± SEM. For analysis of differences between 2 groups, the Student’s t-test was performed. For multiple groups, ANOVA was carried out followed by the Student-Newman-Keuls test. The level of statistical significance was set at P<0.05.

Blood glucose concentration was measured to validate whether STZ induced hyperglycemia as expected. Seven days after STZ administration, the blood glucose concentration reached 20.33±1.84 mM, significantly higher than that of 6.72±0.35 mM in the control group (P<0.01).

To investigate whether hyperglycemia influences SCF expression which subsequently affects CSC homing, SCF mRNA and protein in the peri-infarcted myocardium were analyzed. RT-PCR revealed that the SCF mRNA level increased significantly in the peri-infarcted myocardium on Day 1, 3, 5 and 7 after MI in the normoglycemic and hyperglycemic rats (Fig. 1A). Immunohistochemical staining and western blot analysis also revealed that SCF protein expression dramatically increased in the peri-infarcted myocardium on Day 5 after MI, which was in line with the change in SCF mRNA (Figs. 1B and 3A). However, the increase in SCF mRNA and protein was significantly attenuated in the hyperglycemic group compared to the normoglycemic group.

To understand whether ERK1/2 and p38 MAPK are involved in the inhibition of SCF expression, western blot analysis was performed to detect the levels of phosphorylated or total ERK1/2 and p38 MAPK proteins in the peri-infarcted myocardium. Results showed that the phosphorylation of ERK1/2 (Fig. 2A) and p38 MAPK (Fig. 2B) in the periinfarcted regions were obviously upregulated on Day 1, 3, 5 and 7 after MI in the normoglycemic and hyperglycemic groups, but the levels of regulation in the hyperglycemic group were attenuated compared to the normoglycemic group. Total ERK1/2 and p38 MAPK protein expression showed no obvious differences in the normoglycemic and hyperglycemic groups. This suggests that the increase in SCF in the periinfarcted myocardium may be associated with the activation of ERK1/2 and p38 MAPK, which may be attenuated by hyperglycemia.

To elucidate the relationship between the increase in SCF expression and the upregulation of ERK1/2 and p38 MAPK activities, inhibitors of ERK1/2 (PD98059) and p38 MAPK (SB203580) were administered 30 min prior to operation and once every day on Days 1–4 after MI. Western blot analysis revealed that SCF protein significantly increased in the peri-infarcted myocardium on Day 5 after MI, but the increase was significantly attenuated by administration of PD98059 or SB203580, which was similar to the inhibiting effect of hyperglycemia (Fig. 3A). Meanwhile, in vitro experiments were performed in cultured cardiomyocytes. RT-PCR and ELISA showed that a high glucose level (25 mM) markedly decreased the SCF expression compared to a normal glucose level (5.5 mM) and osmotic control (5.5 mM glucose + 19.5 mM mannitol). After stimulation by IL-1β, SCF expression was greatly increased. However, the increase was significantly downregulated by PD98059 or SB203580, which was similar to the effect of high glucose (Fig. 3B and C). All of these results suggest that hyperglycemia inhibits SCF expression in the peri-infarcted myocardium via downregulation of ERK1/2 and p38 MAPK activities.

To investigate whether the upregulation of SCF leads to a higher accumulation of CSCs in the peri-infarcted region after MI, CSC migration in vivo was performed by injection of BrdU-labeled CSCs into the AV-groove followed by coronary ligation. As shown in Fig. 4A and B, more CSCs were attracted to the peri-infarcted myocardium on Day 5 after MI compared with the sham-operated group. However, in the hyperglycemic rats, CSC homing was markedly inhibited, which was similar to the effect of PD98059 or SB203580 administration in normoglycemic rats. In vitro chemotaxis experiments were also carried out to quantitatively evaluate CSC migration under different conditions. As shown in Fig. 4C, compared with the control group, the average number of migrated CSCs increased significantly in the conditioned medium groups. Adding high glucose, PD98059, SB203580 or an SCF antibody partially or totally inhibited the migration. These results imply that hyperglycemia suppresses CSC homing to the peri-infarcted myocardium through attenuation of SCF expression.

To ascertain whether CSC homing to the peri-infarct myocardium correlates with a change in cardiac function, cardiac function was evaluated on Day 21 after MI by hemodynamic measurement. As expected, the results of +dP/dt_max (Fig. 5A) and −dP/dt_min (Fig. 5B) indicated that cardiac function was notably aggravated after induction of MI. However, AV-groove injection of CSCs resulted in obvious improvement in cardiac function, and this improvement was partially blocked by administration of PD98059 or SB203580. In the hyperglycemic rats, cardiac function was much aggravated after MI compared with the normoglycemic rats. The recovery of cardiac function after AV-groove injection of CSCs was also observed in the hyperglycemic rats, but the levels of LV +dP/dt_max and −dP/dt_min were significantly lower than levels in the normoglycemic rats. These results imply that hyperglycemia inhibits the recovery of cardiac function through inhibition of CSC homing after MI.