Long-lasting partnership between insulin resistance and endothelial dysfunction: role of metabolic memory

Abstract

Background and purpose: The persistence of deleterious effects of hyperglycaemia even after glucose normalization is referred to as 'metabolic memory'. However, similar persistent effects of the metabolic consequences of a high fat diet (HFD) have not been described.

Experimental approach: Rats were given a normal pellet diet (NPD) or a HFD for 3 months. The animals from the HFD group were then returned to the NPD to observe the long-term effects of insulin resistance. Endothelial dysfunction was assessed by carbachol-mediated vasorelaxation and eNOS phosphorylation.

Key results: As expected, HFD consumption resulted in insulin resistance and endothelial dysfunction. Phosphorylation of eNOS at S1177 was decreased in HFD rats, compared with that in the NPD group. Rats on 3 months of HFD showed glucose intolerance and impaired insulin sensitivity and were then switched back to NPD (REV group) . Levels of cholesterol and triglyceride, and adiposity returned to normal in REV rats. However, endothelium-dependent vascular responses to carbachol which were impaired in HFD rats, continued to be impaired in REV rats. Similarly, decreased eNOS phosphorylation after HFD was not improved after 1 or 6 months of REV.

Conclusions and implications: Our data indicate that returning to NPD did not improve the insulin sensitivity or the endothelial dysfunction induced by HFD. Although some biochemical parameters responsible for insulin resistance and endothelial dysfunction were normalized, molecular and vascular abnormalities, involving NO, persisted for several months, highlighting the long-lasting effects of metabolic memory.

Figure 1

Characterization of HFD-induced insulin resistance. (A) Growth curve demonstrating body weight gain of rats, assessed at 30 day intervals in each group. (B and C) Bar graphs of the increased fasting plasma glucose and insulin levels, respectively, of HFD-fed animals. (D and E) glucose tolerance test (IPGTT), with representative bar graph showing AUC values, confirms decreased insulin sensitivity with HFD intake. (F) HOMA-IR scores indicate compensatory hyperinsulinaemia and insulin-resistant state of HFD-fed animals. Data shown are means ± SEM, n = 6–8 rats per group. **P < 0.01, ***P < 0.001; significantly different as indicated.

Figure 2

HFD consumption alters endothelium-dependent vasorelaxant responses in thoracic aorta and causes endothelial dysfunction. (A) Cumulative concentration–response curve to carbachol after pre-contraction with phenylephrine suggests impaired endothelium-dependent vasorelaxation in thoracic aortic rings is due to decreased bioavailability of NO. (B) Phosphorylation of eNOS (S1177), shown by IHC, in the EC of the aortas was decreased by HFD consumption. Phospho-eNOS S1177 was stained in the aortas of NPD (a and b) and HFD-fed (c and d) rats (magnification: 100×). b and d indicate inverted images of the respective photomicrographs. Arrows point to positively staining ECs. 10–12 random fields, including those in a–d, were analysed from the aortas of NPD- and HFD-fed rats for eNOS phosphorylation. C, Quantitation of eNOS phosphorylation at S1177. D-E, shows Western blot of S1177 phosphorylation of eNOS in NPD- and HFD-fed animals. Data shown are means ± SEM, n = 6–8 rats per group. *P < 0.05, ***P < 0.001; significantly different as indicated.

Figure 3

Transient HFD consumption causes persistent insulin resistance. (A) Growth curve demonstrating body weight gain of rats, taken at 60 day intervals in each group, following return to NPD after 3 months of HFD (diet reversal, REV.). (B and C) Plasma glucose and insulin levels indicate sustained hyperglycaemia and hyperinsulinaemia following REV. (D and E) Glucose tolerance test (IPGTT) performed after different durations of REV (1, 2, 4 and 6 months) indicates relatively poor improvement in glucose disposal rate in animals on REV. (F) HOMA-IR scores indicate persistent compensatory hyperinsulinaemia and insulin-resistant state, even after 6 months of REV. Data shown are means ± SEM, n = 6–8 rats per group. **P < 0.01; ***P < 0.001; #P < 0.05, ##P < 0.01, ###P < 0.001; †P < 0.05, ††P < 0.01, †††P < 0.001; significantly different as indicated.

Figure 4

REV normalizes lipid profile and adiposity index. (A and B) Plasma triglyceride and plasma cholesterol levels, after REV for 1, 2, 4 and 6 months indicating normalized lipid profile. (C) Adiposity index (fat pad weights) after REV for 1 and 6 months. Data shown are means ± SEM, n = 6–8 rats per group. ***P < 0.001; ###P < 0.001; †††P < 0.001; significantly different as indicated.

Figure 5

Transient HFD consumption persistently alters endothelium-dependent vasorelaxant responses in rings of thoracic aorta and induces long-lasting endothelial dysfunction. (A and B) Cumulative concentration–response curve to carbachol after pre-contraction with phenylephrine in thoracic aortic rings suggesting decreased bioavailability of NO even after 1 and 6 months of REV respectively. (C) eNOS phosphorylation at S1177 was determined by IHC in the EC of the aortas and was reduced after 1 month of REV. Phospho-eNOS S1177 was stained in the aortas of NPD (a and b), HFD (c and d) and REV-1 (e and f) rats (magnification: 100×). b and d indicate inverted images of the respective photomicrographs. (D) S1177 eNOS phosphorylation levels were determined by IHC in the EC of the aortas and were reduced even after 6 months of REV. Phospho-eNOS S1177 was stained in the aortas of NPD (a and b), HFD (c and d) and REV-6 (e and f) rats (magnification: 100x). b, d and f indicate inverted images of the respective photomicrographs. Arrows point to positively staining ECs. In both the figures C and D, 10–12 random fields, including those in a–f, were analysed from the aortas of NPD- and HFD-fed rats for eNOS phosphorylation. (E) Quantitation of eNOS phosphorylation at S1177 after different durations of REV. Data shown are means ± SEM, n = 6–8 rats per group. ***P < 0.001; ###P < 0.001; †††P < 0.001; significantly different as indicated.

Figure 6

REV does not improve eNOS phosphorylation. (A and B) Western blot of S1177 phosphorylation of eNOS in aorta in REV-1 and REV-6 groups. The return to NPD did not increase eNOS phosphorylation indicating persistent endothelial dysfunction. Data shown are means ± SEM, n = at least three sets of independent experiments. *P < 0.05, **P < 0.01; significantly different as indicated.