The influence of high glucose on the aerobic metabolism of endothelial EA.hy926 cells

Abstract

The endothelium is considered to be relatively independent of the mitochondrial energy supply. The goals of this study were to examine mitochondrial respiratory functions in endothelial cells and isolated mitochondria and to assess the influence of chronic high glucose exposure on the aerobic metabolism of these cells. A procedure to isolate of bioenergetically active endothelial mitochondria was elaborated. Human umbilical vein endothelial cells (EA.hy926 line) were grown in medium containing either 5.5 or 25 mM glucose. The respiratory response to elevated glucose was observed in cells grown in 25 mM glucose for at least 6 days or longer. In EA.hy926 cells, growth in high glucose induced considerably lower mitochondrial respiration with glycolytic fuels, less pronounced with glutamine, and higher respiration with palmitate. The Crabtree effect was observed in both types of cells. High glucose conditions produced elevated levels of cellular Q10, increased ROS generation, increased hexokinase I, lactate dehydrogenase, acyl-CoA dehydrogenase, uncoupling protein 2 (UCP2), and superoxide dismutase 2 expression, and decreased E3-binding protein of pyruvate dehydrogenase expression. In isolated mitochondria, hyperglycaemia induced an increase in the oxidation of palmitoylcarnitine and glycerol-3-phosphate (lipid-derived fuels) and a decrease in the oxidation of pyruvate (a mitochondrial fuel); in addition, increased UCP2 activity was observed. Our results demonstrate that primarily glycolytic endothelial cells possess highly active mitochondria with a functioning energy-dissipating pathway (UCP2). High-glucose exposure induces a shift of the endothelial aerobic metabolism towards the oxidation of lipids and amino acids.

Fig. 1

Time-course of cell respiratory response induced by excess glucose. EA.hy926 cells were cultured for 3, 6, and 9 days in normal-glucose concentration (nG cells) or high-glucose concentration (hG cells). The basal oxygen consumption rate (OCR) was measured with 5 mM pyruvate

Fig. 2

Mitochondrial function of EA.hy926 cells grown in normal-glucose concentration (nG cells) or high-glucose concentration (hG cells). Substrate-dependent changes in the basal oxygen consumption rate (OCR) (a), maximal OCR (b), proton leak (c), and ATP-dependent OCR (d). Substrates: 5 mM pyruvate, 5.5 mM glucose (nG), 25 mM glucose (hG), pyruvate and nG, pyruvate and hG, 3 mM l-glutamine, or 0.3 mM palmitate. The vertical lines illustrate the values in the presence of pyruvate alone in nG cells (dashed lines) and hG cells (solid lines). e The representative oxygen uptake measurement with EA.hy926 cells (using nG cells as an example), illustrating the Crabtree effect. The numbers on the traces refer to the OCR in nmol O/min/mg protein. f The COX activity, CS activity, and cellular concentration of Q ₁₀ (expressed as the percentage of Q10 concentration in nG cells). g NBT reduction in the absence or presence of 10 μM DPI

Fig. 3

Determination of protein levels in EA.hy926 cells grown in normal glucose (nG cells) or high glucose (hG cells) (a) and in mitochondria isolated from these cells (nG mito and hG mito, respectively) (b). a Representative Western blots and analyses of the protein expression of HKI, LDH, mitochondrial marker (Mito marker), CS, and COXII. b Representative Western blots and analyses of the protein expression of UCP2, SOD2, ACADS, E3BP, CS, and particular subunits of ATP synthase, Complex III (CIII), Complex II (CII) and Complex I (CI). Expression levels normalised for β actin (a), mito marker (b , upper panel), or COXII (b , lower panel) protein abundance are shown

Fig. 4

Functional characteristics of endothelial mitochondria isolated from normal-glucose and high-glucose cells (nG mito and hG mito, respectively). a Representative oxygen uptake measurements (using nG mitochondria as an example) of non-phosphorylating respiration, phosphorylating respiration, and uncoupled respiration, as well as the obtained coupling parameters, ADP/O and the respiratory control ratio (RCR). Malate (10 mM) and 10 mM succinate (plus 1 μM rotenone) were used as respiratory substrates. The numbers on the traces refer to OCR in nmol O/min/mg protein. b The maximal respiration (phosphorylating respiration or uncoupled respiration) with different respiratory substrates: 10 mM malate, 10 mM pyruvate, 10 mM glutamate, 10 mM succinate, 3 mM glycerol-3-phosphate, 0.3 mM palmitoyl- dl -carnitine. c The COX activity, concentration of Q ₁₀ in mitochondria, and mitochondrial Q reduction level (QH₂/Q _tot) and the membrane potential (ΔΨ) during non-phosphorylating oxidation of succinate

Fig. 5

UCP2 activity in endothelial mitochondria from normal-glucose and high-glucose cells. a Effects of UCP2 activation by palmitic acid (PA) (7 μM per addition), and UCP2 inhibition by 2 mM GTP on the oxygen uptake, ΔΨ, and Q reduction level. The normal-glucose mitochondria (nG mito) are represented by broken lines and numbers in italics, and the high-glucose mitochondria (hG mito) are represented by solid lines and bold numbers. b Relationship between the respiratory rate and ΔΨ (proton leak kinetics) during non-phosphorylating succinate oxidation titrated with malonate. The PA-induced, GTP-inhibited, UCP2-mediated proton leak at the same ΔΨ (155 mV) is indicated as vertical lines (a broken line indicates nG mitochondria, and a solid line indicates hG mitochondria). c The influence of PA and GTP on the ADP/O ratio during the phosphorylating respiration. Relative changes compared with the control values (in the absence of PA and GTP) are shown. b, c Succinate (with 1 μM rotenone) was used as and oxidisable substrate. The mitochondria were incubated in the absence or presence of 14 μM PA and/or 2 mM GTP