Reduced NO-cGMP signaling contributes to vascular inflammation and insulin resistance induced by high-fat feeding

Abstract

Objective: Diet-induced obesity (DIO) in mice causes vascular inflammation and insulin resistance that are accompanied by decreased endothelial-derived NO production. We sought to determine whether reduced NO-cGMP signaling contributes to the deleterious effects of DIO on the vasculature and, if so, whether these effects can be blocked by increased vascular NO-cGMP signaling.

Methods and results: By using an established endothelial cell culture model of insulin resistance, exposure to palmitate, 100 micromol/L, for 3 hours induced both cellular inflammation (activation of IKK beta-nuclear factor-kappaB) and impaired insulin signaling via the insulin receptor substrate-phosphatidylinositol 3-kinase pathway. Sensitivity to palmitate-induced endothelial inflammation and insulin resistance was increased when NO signaling was reduced using an endothelial NO synthase inhibitor, whereas endothelial responses to palmitate were blocked by pretreatment with either an NO donor or a cGMP analogue. To investigate whether endogenous NO-cGMP signaling protects against vascular responses to nutrient excess in vivo, adult male mice lacking endothelial NO synthase were studied. As predicted, both vascular inflammation (phosphorylated I kappaB alpha and intercellular adhesion molecule levels) and insulin resistance (phosphorylated Akt [pAkt] and phosphorylated eNOS [peNOS] levels) were increased in endothelial NO synthase(-/-) (eNOS(-/-)) mice, reminiscent of the effect of DIO in wild-type controls. Next, we asked whether the vascular response to DIO in wild-type mice can be reversed by a pharmacological increase of cGMP signaling. C57BL6 mice were either fed a high-fat diet or remained on a low-fat diet for 8 weeks. During the final 2 weeks of the study, mice on each diet received either placebo or the phosphodiesterase-5 inhibitor sildenafil, 10 mg/kg per day orally. In high-fat diet-fed mice, vascular inflammation and insulin resistance were completely prevented by sildenafil administration at a dose that had no effect in mice fed the low-fat diet.

Conclusions: Reduced signaling via the NO-cGMP pathway is a mediator of vascular inflammation and insulin resistance during overnutrition induced by high-fat feeding. Therefore, phosphodiesterase-5, soluble guanylyl cyclase, and other molecules in the NO-cGMP pathway (eg, protein kinase G) constitute potential targets for the treatment of vascular dysfunction in the setting of obesity.

Figures

Figure 1. The effect of eNOS inhibitor, L-NAME, on palmitate-mediated endothelial inflammation and insulin signaling

HMEC were treated with vehicle or the eNOS inhibitor (L-NAME) 50 µM 3 h and then treated with BSA (C) or BSA-palmitate (F) (10 µM) for 3 h, followed by insulin stimulation (I) (100 nM, 15 min). Cell lysates were analyzed by Western blot and fold increase over vehicle, BSA control group was calculated for each experiment (n=3) A. Fold increase of phospho-IκBα protein normalized to GAPDH levels. B. Fold increase of ICAM protein levels normalized to GAPDH levels. C. Fold increase in IRS-1 tyrosine phosphorylation normalized to total IRS-1. D. Fold increase in pAkt normalized to total Akt E. Fold increase in peNOS normalized to total eNOS levels. *p<0.05

Figure 2. The effect of the NO donor, DETANO on palmitate-mediated endothelial inflammation and insulin resistance

HMEC were treated with vehicle or the NO donor-DETANO (50 µM) overnight and then treated with BSA (C) or BSA-palmitate (F) (100 µM) for 3 h, followed by insulin stimulation (I)(100 nM, 15 min). Cell lysates were analyzed by Western blot and fold increase over the vehicle, BSA control group was calculated for each experiment (n=3) A–C. Representative IRS-1 tyrosine, pAkt and peNOS Western blots are shown. IRS-1 tyrosine phosphorylation levels were normalized to total IRS-1 levels and pAkt and peNOS were normalized to total Akt and eNOS levels and the fold increase was calculated. * p<0.05. D. phospho-IκBα protein levels normalized to GAPDH E. Fold increase in IL-6 concentration in response to palmitate in the presence of DETANO (50 µM), sGC inhibitor ODQ (1 µM), or PKG inhibitor KT-5823 (1 µM) from HMEC supernatant as determined by ELISA. * p<0.05.

Figure 3. The effect of 8 Br-cGMP on endothelial inflammation and insulin signaling

HMEC were treated with 50 µM 8-Br-cGMP for 12 h in the presence or absence of ODQ (1 µM) and then stimulated with LPS (5 ng/ml) for 1 h or palmitate (100 µM) for 3 h. A.B. Phospho-IκBα levels were assessed by Western blot and normalized to GAPDH and fold increased over the control condition was calculated following densitometry. C. Following treatment with vehicle or 8-Br-cGMP, HMEC were stimulated with 100 nM insulin for 15 min and pAkt levels were assessed by Western blot. * p<0.05.

Figure 4. The effect of HF-feeding in eNos −/− mice on body weight, vascular insulin signaling and inflammation

WT and eNos −/− mice were fed a HF or LF diet for 4 wk. A. Weight gain for both WT and eNos −/− mice on the LF or HF diet. B. Fold increase in p-Akt/total Akt as measured by ELISA. C. Fold increase in phospho-IκBα normalized to GAPDH levels as measured by Western blot. D. ICAM protein levels normalized to total GAPDH levels. *p<0.05.

Figure 5. Effect of daily sildenafil during HF feeding on weight gain, vascular insulin signaling and inflammation

Male C57BL6 mice were maintained on either a low fat (LF; 10% saturated fat) or high fat (HF; 60% saturated fat) diet for 8 wk and for the last 2 wks of the diet study, mice received 0.1 mg/kg/day of sildenafil or placebo (control). A. Mean weekly weight during wk 6–8. B,C. Fold increase in pAkt and peNOS in response to insulin (0.06 U/g body weight IP) *P<0.05 vs. LF controls. D,E Fold increase in phospho-IκBα and ICAM in response to 8 wk of HF or LF feeding with sildenafil or placebo (control). *, p<0.05.