Endothelial inflammation induced by excess glucose is associated with cytosolic glucose 6-phosphate but not increased mitochondrial respiration

Abstract

Aims/hypothesis: Exposure of endothelial cells to high glucose levels suppresses responses to insulin, including induction of endothelial nitric oxide synthase activity, through pro-inflammatory signalling via the inhibitor of nuclear factor kappaB (IkappaB)alpha-nuclear factor kappaB (NF-kappaB) pathway. In the current study, we aimed to identify metabolic responses to glucose excess that mediate endothelial cell inflammation and insulin resistance. Since endothelial cells decrease their oxygen consumption rate (OCR) in response to glucose, we hypothesised that increased mitochondrial function would not mediate these cells' response to excess substrate.

Methods: The effects of glycolytic and mitochondrial fuels on metabolic intermediates and end-products of glycolytic and oxidative metabolism, including glucose 6-phosphate (G6P), lactate, CO(2), NAD(P)H and OCR, were measured in cultured human microvascular endothelial cells and correlated with IkappaBalpha phosphorylation.

Results: In response to increases in glucose concentration from low to physiological levels (0-5 mmol/l), production of G6P, lactate, NAD(P)H and CO(2) each increased as expected, while OCR was sharply reduced. IkappaBalpha activation was detected at glucose concentrations >5 mmol/l, which was associated with parallel increases of G6P levels, whereas downstream metabolic pathways were insensitive to excess substrate.

Conclusions/interpretation: Phosphorylation of IkappaBalpha by excess glucose correlates with increased levels of the glycolytic intermediate G6P, but not with lactate generation or OCR, which are inhibited well below saturation levels at physiological glucose concentrations. These findings suggest that oxidative stress due to increased mitochondrial respiration is unlikely to mediate endothelial inflammation induced by excess glucose and suggests instead the involvement of G6P accumulation in the adverse effects of hyperglycaemia on endothelial cells.

Fig. 1. The temporal effect of glucose on activation of NF-kB signaling

HMEC were cultured for the denoted time in the presence of various concentrations of glucose. Densitometry was performed, and after normalizing the data for total IKKβ, the fold increase over the 5 mM glucose control was calculated (N=3) A. Levels of phospho-IκBα determined by Western blot in lysates of HMEC treated with 25 mM glucose for time period between 1–18 h. B. IL-6 levels in response to 25 mM glucose as measured by ELISA. C. Representative Western blot (N=3) of ICAM and actin protein levels in response to 25 mM glucose D. NF-κB activity in response to 25 mM glucose. Comparisons between groups of treatments were made by ANOVA, followed by Bonferroni post hoc test when significance was indicated. Values of p<0.05 were considered significant (*).

Fig. 2. The effect of NF-kB on insulin signaling in endothelial cells exposed to 25 mM of glucose

HMEC transduced with either GFP or phosphorylation resistant IκBα (NF-kB super) construct were treated with vehicle or insulin (100 nM) for 15 min in 5 mM (control) or 25 mM glucose. A. Insulin-mediated pAkt as measured by Western blot were quantified and normalized to total Akt levels for 3 independent experiments. Fold increase over the low glucose (5 mM) vehicle condition were calculated for the GFP transduced HMEC (open square) and NF-kB super transduced HMEC (gray square) (*p<0.05) B. Insulin-mediated peNOS. as measured by Western blot were quantified and normalized to total eNOS.

Fig. 3. Temporal changes in OCR in response to the addition and removal of 20 mM glucose

HMEC were loaded into a perifusion system -two chambers were run in parallel with 4 × 10⁶ cells loaded into each- and allowed to equilibrate for 90-minute in the presence of 20 mM glucose. The glucose was then washed out or added to the inflow media as indicated. Due to mixing within the chamber, the temporal resolution of the measurements is limited to about 5 minutes. Data points are the average of two data sets collected in parallel.

Fig. 4. Effect of glucose on OCR and LPR (A) and NAD(P)H (B) by HMEC

(A) Experiments were carried out as described in Fig. 3, except that the protocol involved testing the effect of 0, 5 and 25 mM glucose, and data represents the average ± SEM of separate perifusions. OCR decreased and LPR increased in response to 5 mM glucose, but switching between 5 and 25 mM glucose appeared to have little effect. (B) NAD(P)H was measured in response to sequential increases in glucose to 5 and 25 mM. Data shown is average of two separate perifusions, where each kinetic profile was an average time course of 3 cells. NAD(P)H was normalized relative to 100% (in the presence of KCN) and 0% (in the presence of FCCP) as described in the Methods section. Statistical analysis was carried out on the steady state values at 5 and 25 mM glucose (calculated as the average of the final 15 minutes of data obtained during each concentration of glucose) using a Student’s t-test with paired data calculated with Kaleidagraph (Synergy Software, Reading PA). The differences in OCR, LPR and NAD(P)H between the two conditions were not significant.

Fig. 4. Effect of glucose on OCR and LPR (A) and NAD(P)H (B) by HMEC

(A) Experiments were carried out as described in Fig. 3, except that the protocol involved testing the effect of 0, 5 and 25 mM glucose, and data represents the average ± SEM of separate perifusions. OCR decreased and LPR increased in response to 5 mM glucose, but switching between 5 and 25 mM glucose appeared to have little effect. (B) NAD(P)H was measured in response to sequential increases in glucose to 5 and 25 mM. Data shown is average of two separate perifusions, where each kinetic profile was an average time course of 3 cells. NAD(P)H was normalized relative to 100% (in the presence of KCN) and 0% (in the presence of FCCP) as described in the Methods section. Statistical analysis was carried out on the steady state values at 5 and 25 mM glucose (calculated as the average of the final 15 minutes of data obtained during each concentration of glucose) using a Student’s t-test with paired data calculated with Kaleidagraph (Synergy Software, Reading PA). The differences in OCR, LPR and NAD(P)H between the two conditions were not significant.

Fig. 5. Glucose dose responses of G6P, LPR, CO₂ production, NAD(P)H and OCR

LPR, OCR and NAD(P)H values were measured using flow systems and are steady state values averaged from 15 to 30 min after the change in glucose concentration. CO₂ production was measured under static conditions and is an integrated response over 45 min. G6P values were single time point measurements obtained 30 min after the change in glucose. Each data point is an average of 4 separate experiments, except for the NAD(P)H data (n = 3), and the CO₂ data (n = 5) and error bars are ± SEM. Statistical significance of differences between the values of each parameter at 5 and 25 mM glucose was carried out using a paired Student’s t-test calculated with Kaleidagraph (Synergy Software, Reading PA). Only the differences at 5 and 25 mM glucose in G6P and CO₂ production were significant, as shown by the bar graphs in the inset (*p<0.05 and **p< 0.01).

Fig. 6. Effect of pyruvate on OCR by HMEC in the absence of glucose

As shown, the protocol entailed washing out the glucose, and after 30 min, assessing the effect of 10 mM pyruvate on OCR. LPR was also measured, but lactate levels were below the detection limit of the assay. In order to calculate error bars relative to the change in OCR, ΔOCR was calculated as the average of 3 separate perifusions, where individual time points were normalized by subtracting the baseline OCR at 5 mM glucose (averaged from −15 to 0 minutes).

Fig. 7. Effect of pyruvate on phospho-IκBα, a reflection of IKKβ activity, in the absence and presence of 5 mM glucose

Western blots demonstrating the effect of pyruvate and glucose on activation of IKKβ. HMEC were cultured for the denoted time in the presence of various concentrations of pyruvate and glucose. A two-tailed t-test was used to determine statistical significance of differences in mean values using Stata 8. Values of p<0.05 were considered significant.

Fig. 8. Effect of pyruvate on OCR (A) and G6P (B) in the presence of 5 mM glucose

Measurements are obtained as described in Fig. 5 except that each data point is an average of 5 separate experiments. Error bars are ± SEM; p-value was calculated by paired t-test using Kaleidagraph.

Fig. 8. Effect of pyruvate on OCR (A) and G6P (B) in the presence of 5 mM glucose

Measurements are obtained as described in Fig. 5 except that each data point is an average of 5 separate experiments. Error bars are ± SEM; p-value was calculated by paired t-test using Kaleidagraph.