Activation of κ-opioid receptor by U50,488H improves vascular dysfunction in streptozotocin-induced diabetic rats

In order to establish diabetic model, streptozotocin (STZ, Sigma, St. Louis, MO, USA) injection was applied. Male Sprague–Dawley (SD) rats (weight 200–240 g) received intraperitoneal injection of STZ (35 mg/kg) for 3 days [31]. Blood glucose levels were tested by blood glucose meter (Lifescan, Inc., Milpitas, CA, USA) one week following STZ injection. Animals with glucose levels ≥16.6 mmol/L were considered as diabetic. Diabetic rats received standard food for 4 weeks and then randomized into the following 4 groups (n = 15): (1) DM group: diabetic rats without any treatment; (2) DM + U50,488H group: diabetic rats that received daily treatment of U50,488H (Biomol, Plymouth Meeting, PA, USA) at 1 mg/kg for 4 weeks; (3) DM + vehicle group: diabetic rats that received only saline daily for 4 weeks; (4) DM + nor-BNI group: diabetic rats that received daily treatment of nor-BNI (Sigma, St. Louis, MO, USA) at 0.5 mg/kg for 4 weeks. Age-matched normal rats that received only saline daily (n = 15) were used as non-diabetic controls (CON). All experiments were conducted under the National Institutes of Health Guidelines on the Use of Laboratory Animal and were approved by the Fourth Military Medical University Committee on Animal Care.

Blood pressure was monitored with a tail cuff system (Non-invasive Blood Pressure System, PanLab) as described previously [32]. Briefly, rats were placed in a warm chamber (37°C) for 10 minutes to rest, and then occluding cuffs and pneumatic pulse transducers were placed on the tails. Five readings were obtained from each rat. Commercially available ELISA kit was used to determine plasma insulin levels (Cusabio Biotech Co., China).

To detect the KOR expression in thoracic aorta, immunohistochemical staining was performed. Briefly, the thoracic aortas were isolated, fixed by 4% paraformaldehyde and paraffin embedded. Serial tissue sections (5 μm) were mounted on slides, then xylene and ethanol were used to deparaffinize and dehydrate the tissue respectively. Non-specific antigens were blocked by 20% normal bovine serum for 30 minutes, which was followed by treatment with 0.3% hydrogen peroxide. Then the sections were incubated overnight at 4°C with KOR antibody (1:100, Santa Cruz Biotechnology, Santa Cruz, CA, USA). After three washes in PBS, the sections were incubated with a biotinylated secondary antibody (1:2000, Santa Cruz Biotechnology, Santa Cruz, CA, USA) for 1 hour at 37°C and washed again in PBS. Then horseradish peroxidase was developed with 3,3-diaminobenzadine (DAB) (Roche diagnostics, Mannheim, Germany) as the chromogen substrate. After rinsing, the slides were mounted with cover slips, and the positive samples displayed a brown color under a light microscope.

Transmission electron microscope (TEM) was performed to identify the ultrastructural changes of thoracic aortas. After anesthetized, the thoracic aortas from the rats were immediately isolated and cut into 1–2 mm wide, and then kept in glutaraldehyde and osmium tetroxide in order. Subsequently, the TEM specimens were dehydrated using graded ethanol and embedded in resin. The ultrastructure of thoracic aortas were visualized by ultra-thin sections under transmission electron microscope (JEOL JEM-2000EX, Tokyo, Japan).

Enzyme-linked immunosorbent assay (ELISA, Bionewtrans Pharmaciutical Biotechnology Co.Ltd, USA) was performed to determine the serum levels of ANG II, sICAM-1, IL-6 and IL-8, as previously described [33]. Standard curves were drawn by manufacturers’ instructions. Absorbance of standards and samples was measured by microplate reader (Model 550, Bio-Rad, Japan) at 450 nm and the serum levels of ANG II, sICAM-1, IL-6 and IL-8 were calculated according to the standard curves.

The arterial functions were assessed with endothelium-intact isolated rats vessels mounted for isometric tension recordings in organ chambers, as described previously [34]. Rats were anesthetized with pentobarbital sodium (40 mg/kg, i.p.), followed by the rapid removal of thoracic aortas. The vessels were cleared of excess connective tissues under a stereomicroscope and cut into rings (diameter ≈ 3 mm).

Vascular rings were mounted on a myograph (Multi Wire Myograph System-610 M, Danish Myo Technology A/S, Denmark) under their optimal resting tension (1.0 g) in 15 ml organ baths containing warmed (37°C), oxygenated (95% O₂ + 5% CO₂) Krebs solution (118.3 mmol/L NaCl, 4.7 mmol/L KCl, 25 mmol/L NaHCO3, 1.2 mmol/L MgSO4, 1.2 mmol/L KH2PO4, 2.5 mmol/L CaCl2, 11.1 mmol/L glucose and 0.026 mmol/L EDTA, pH 7.4). The vessels were allowed to equilibrate for 60 minutes before the onset of experiments, with Krebs solution changed every 15 minutes. After a washout period, vascular rings were challenged with a cumulative concentration of KCl (10–90 mmol/L) or noradrenaline (NE, 10⁻⁹-10⁻⁴ mol/L) to test vasoconstrictive activity. In addition, arterial contractions were induced by phenylephrine (PE, 80 μmol/L), and then were relaxed by acetylcholine (ACh, 10⁻⁹-10⁻⁴ mol/L) or sodium nitroprusside (SNP, 10⁻⁹-10⁻⁴ mol/L) to test vasodilative activity. Myograph data were recorded by PowerLab and Chart software (AD Instruments, Australia), and cumulative concentration response curves were then generated.

Western blot was performed as previously described [31]. Briefly, the thoracic aortic tissues from all groups were quickly isolated and snap-frozen. Total proteins (for KOR and eNOS) were extracted with lysate and nuclear proteins (for NF-κB p65) were isolated using a nuclear NE-PER extract kit (Thermo Scientific, IL, USA). Equal weight of proteins (40 μg) were loaded into a SDS-PAGE gel. After separated, the proteins were transferred onto nitrocellulose membranes electrophoretically. Then the membranes were incubated with 5% skim milk for 1 hour, appropriate primary antibody at 4°C overnight, and secondary antibody at room temperature for 1 hour in order. Enhanced chemiluminescence reagent kit (Millipore) was used for development of the blots, which were detected with UVP Bio-Imaging Systems. Protein levels were evaluated by Vision Works LS Acquisition and Analysis Software.

The following primary antibodies were used: KOR (1:1 000), endothelial nitric oxide synthase (eNOS, 1:1 000), phospho-eNOS (at Ser-1177) (1:1 000), NF-κB p65 (1:1 000), histone 3 (1:1 000) and β-actin (1:1 000). Secondary antibodies were horseradish peroxidase-conjugated goat anti-rabbit IgG and rabbit anti-goat IgG at 1:5,000 dilution. All above antibodies were purchased from Santa Cruz Biotechnology.

All statistical analysis were performed by GraphPad Prism software version 5.0 (GraphPad Software, CA, USA). Values were presented as mean ± SD and analyzed using ANOVA with Bonferroni corrected t-test. Western blot results were analyzed using the Kruskal–Wallis test followed by Dunn’s post hoc test. Differences were considered statistically significant if P < 0.05.

Administration of U50,488H or nor-BNI was initiated 4 weeks following the onset of experimental DM. The metabolic characteristics of rats in different groups are shown in Table 1. Diabetic rats exhibited hyperglycemia (22.3 ± 2.19 mmol/L vs. 8.2 ± 0.94 mmol/L, P < 0.05), decreased body weight gain (286.6 ± 14.61 g vs. 419.7 ± 12.25 g, P < 0.05) and reduced insulin levels (0.51 ± 0.093 ng/mL vs. 2.42 ± 0.466 ng/mL, P < 0.05) compared with CON group. There was no significant difference in blood pressure (103.6 ± 6.40 mmHg vs. 104.9 ± 6.14 mmHg, P > 0.05) between CON and DM groups. Neither U50,488H nor nor-BNI administration influenced body weight, blood glucose levels, blood pressure and insulin levels in diabetic rats.

As shown in Figure 1A, KOR was mainly detectable in the endothelium and smooth muscle of the thoracic aortas. Compared with the CON group, the KOR protein expression increased significantly in the DM group, which was also confirmed by Western blot (Figure 1B). We further tested KOR expression in the DM + U50,488H group and found no statistically significant difference compared with the DM group (Figure 1C).

To determine the effects of KOR activation on vasoconstrictive function of thoracic aortas in DM rats, isometric tension recordings were performed. The concentration-response curves for KCl (0–100 mmol/L) and NE (10⁻⁹-10⁻⁴ mol/L) are shown in Figure 2A-B. The vasoconstrictive function of thoracic aortas was significantly greater for the DM group compared with CON group. U50,488H administration induced a marked rightward shift in concentration-response curves, while nor-BNI administration induced a marked leftward shift in concentration-response curves, indicating that KOR activation was involved in attenuation of the vasoconstrictive abnormalities in DM rats. However, we also found that U50,488H administration could not restore the vasoconstrictive function to the normal level.

Vasodilative function was determined by isometric tension recordings with addition of ACh (10⁻⁹-10⁻⁴ mol/L) or SNP (10⁻⁹-10⁻⁴). Concentration-response curves (Figure 2C) for ACh revealed that DM led to lower vasodilatory potency, which was significantly improved with U50,488H treatment. In contrast, nor-BNI administration further aggravated the vasodilative abnormalities of diabetic rats. However, we also found that U50,488H administration could not restore the vasodilative function to the normal level. There were no statistically significant differences in relaxation induced by SNP administration among all five groups (Figure 2D). Taken together, all these data indicated that KOR activation by U50,488H could improve the vasodilative function in DM endothelium-dependently.

Hematoxylin and eosin (HE) staining showed intact endothelium in the CON group, that the endothelium was falling in the DM group, and only part of the endothelium was falling in the DM + U50,488H group (Figure 3A). Transmission electron microscope analysis was performed to further evaluate the effects of KOR activation on the thoracic aortal ultrastructure in DM rats. In the CON group, the surface of the endothelium was smooth, well-integrated and tightly affixed to the underlying smooth muscle. However, the endothelium of aortas from the DM group showed signs of degeneration, such as swelling and necrosis, and was noticeably detached from the smooth muscle. The endothelial cell junctions were not integrated, and smooth muscle cells migrated into the endothelium. Additionally, more collagen fibrils and irregular thickening of elastic fibers were observed in the pericellular spaces of enlarged smooth muscle cells. U50,488H treatment effectively attenuated the changes, preserving the integrity of the endothelium and decreasing smooth muscle cells migration, while treatment with vehicle or nor-BNI did not show these positive effects (Figure 3B).

Serum NO concentrations in the DM group were significantly decreased compared with the CON group (Figure 4A). Treatment with U50,488H enhanced NO secretion, while treatment with vehicle did not have this positive effect. Additionally, nor-BNI treatment resulted in less NO secretion. ANG II levels in the DM group were higher compared with the CON group, while U50,488H treatment attenuated ANG II production, and nor-BNI treatment further aggravated ANG II production (Figure 4B). We also investigated the effect of U50,488H on eNOS phosphorylation in thoracic aortas. The DM group showed decreased eNOS phosphorylation, while U50,488H treatment increased eNOS phosphorylation (Figure 4C). Furthermore, in the DM + U50,488H group, NO levels and eNOS phosphorylation were lower and ANG II levels were higher compared with the CON group.

To determine whether U50,488H inhibited the DM-induced inflammatory response in rats, pro-inflammatory cytokines and adhesion molecule levels were measured. As shown in Figure 5A-C, concentrations of IL-6, IL-8 and sICAM-1 increased markedly in the DM group compared with the CON group. Treatment with U50,488H inhibited DM-induced IL-6, IL-8 and sICAM-1 production but did not restore them to the normal levels, while nor-BNI treatment further increased IL-6, IL-8 and sICAM-1 production.

It is reported that inhibition of the transcription factor NF-кB is important in mediating anti-inflammatory activity. Therefore, Western blot was performed to investigate NF-кB nuclear translocation. The results showed that nuclear translocation of p65 protein induced by DM was markedly attenuated by U50,488H treatment, while nor-BNI treatment increased nuclear translocation of p65 protein (Figure 5D).