Upregulation of inducible NO synthase by exogenous adenosine in vascular smooth muscle cells activated by inflammatory stimuli in experimental diabetes

Abstract

Background: Adenosine has been shown to induce nitric oxide (NO) production via inducible NO synthase (iNOS) activation in vascular smooth muscle cells (VSMCs). Although this is interpreted as a beneficial vasodilating pathway in vaso-occlusive disorders, iNOS is also involved in diabetic vascular dysfunction. Because the turnover of and the potential to modulate iNOS by adenosine in experimental diabetes have not been explored, we hypothesized that both the adenosine system and control of iNOS function are impaired in VSMCs from streptozotocin-diabetic rats.

Methods: Male Sprague-Dawley rats were injected with streptozotocin once to induce diabetes. Aortic VSMCs from diabetic and nondiabetic rats were isolated, cultured and exposed to lipopolysaccharide (LPS) plus a cytokine mix for 24 h in the presence or absence of (1) exogenous adenosine and related compounds, and/or (2) pharmacological agents affecting adenosine turnover. iNOS functional expression was determined by immunoblotting and NO metabolite assays. Concentrations of adenosine, related compounds and metabolites thereof were assayed by HPLC. Vasomotor responses to adenosine were determined in endothelium-deprived aortic rings.

Results: Treatment with adenosine-degrading enzymes or receptor antagonists increased iNOS formation in activated VSMCs from nondiabetic and diabetic rats. Following treatment with the adenosine transport inhibitor NBTI, iNOS levels increased in nondiabetic but decreased in diabetic VSMCs. The amount of secreted NO metabolites was uncoupled from iNOS levels in diabetic VSMCs. Addition of high concentrations of adenosine and its precursors or analogues enhanced iNOS formation solely in diabetic VSMCs. Exogenous adenosine and AMP were completely removed from the culture medium and converted into metabolites. A tendency towards elevated inosine generation was observed in diabetic VSMCs, which were also less sensitive to CD73 inhibition, but inosine supplementation did not affect iNOS levels. Pharmacological inhibition of NOS abolished adenosine-induced vasorelaxation in aortic tissues from diabetic but not nondiabetic animals.

Conclusions: Endogenous adenosine prevented cytokine- and LPS-induced iNOS activation in VSMCs. By contrast, supplementation with adenosine and its precursors or analogues enhanced iNOS levels in diabetic VSMCs. This effect was associated with alterations in exogenous adenosine turnover. Thus, overactivation of the adenosine system may foster iNOS-mediated diabetic vascular dysfunction.

Fig. 1

Immunoblots for iNOS in VSMCs from control (a) and STZ-diabetic rats (b) in the presence of endogenous adenosine modulators. VSMCs were incubated with cytomix comprising 10 ng/mL interleukin (IL)-1β, 10 ng/mL interferon (IFN)-γ and 25 ng/mL tumor necrosis factor (TNF)-α plus 1 µg/mL LPS for 24 h. Representative blots are shown. Densitometric analysis of iNOS protein levels are shown as % cytomix. The data are presented as mean ± SEM (n = 8). *P < 0.05, **P < 0.01 (two-tailed t-test). ADA, adenosine deaminase; 8-PT, 8-phenyltheophylline

Fig. 2

Immunoblots for iNOS in VSMCs from control (a) and STZ-diabetic rats (b) in the presence of adenosine turnover modulators. VSMCs were incubated as described in the legend to Fig. 1. Representative blots are shown. Densitometric analysis of iNOS protein levels are shown as  % cytomix. The data are presented as mean ± SEM (n = 8). *P < 0.05 (two-tailed t-test). NBTI, S-(4-Nitrobenzyl)-6-thioinosine; EHNA, erythro-9-(2-Hydroxy-3-nonyl)adenine; AOPCP, α,β-Methylene-ADP

Fig. 3

Correlations between iNOS levels and NO metabolite release into the culture medium of VSMCs from control (a) and STZ-diabetic rats (b). iNOS level data are from experiments shown in Figs. 1 and 2

Fig. 4

Residual adenosine concentration as measured by HPLC in the culture medium of VSMCs from control (a) and STZ-diabetic rats (b) following incubation with increasing adenosine concentrations (1–100 µM) for 24 h. The values represent the mean ± SEM (n = 10)

Fig. 5

Immunoblots for iNOS in VSMCs from control (a) and STZ-diabetic rats (b) in the presence of exogenous purines. VSMCs were incubated as described in the legend to Fig. 1. Representative blots are shown. Densitometric analysis of iNOS protein levels are shown as % cytomix. The values represent the mean ± SEM (n = 5). *P < 0.05 (two-tailed t-test). PIA, N-(2-phenylisopropyl)-adenosine

Fig. 6

Residual concentration of adenosine and its metabolites as measured by HPLC following incubation with adenosine, AMP or inosine (all 100 µM) for 24 h in cell-free culture medium (a) or the culture medium of VSMCs from control (b) and STZ-diabetic rats (c). The values represent the mean ± SEM (n = 10)

Fig. 7

Residual concentration of adenosine, AMP and metabolites thereof as measured by HPLC following incubation with adenosine or AMP (both 100 µM) for 24 h in the culture medium of VSMCs from control (a) and STZ-diabetic rats (b). Inhibitors of adenosine/AMP turnover were added as indicated. The values represent the mean ± SEM (n = 6). NBTI, S-(4-Nitrobenzyl)-6-thioinosine; EHNA, erythro-9-(2-Hydroxy-3-nonyl)adenine; AOPCP, α,β-methylene-ADP

Fig. 8

Concentration-response curves of endothelium-deprived aortic rings from control and STZ-diabetic rats precontracted with 0.1 µM noradrenaline and incubated with increasing concentrations of adenosine in the presence or absence of the NO synthase inhibitor L-NMA (10 µM). n = 5–7. Diabetes + NMA vs. diabetes, P < 0.001 (two-way ANOVA). Control vs. diabetes and control + NMA vs. control, ns not significant