Branched-chain amino acids promote endothelial dysfunction through increased reactive oxygen species generation and inflammation

Abstract

Branched-chain amino acids (BCAA: leucine, isoleucine and valine) are essential amino acids implicated in glucose metabolism and maintenance of correct brain function. Elevated BCAA levels can promote an inflammatory response in peripheral blood mononuclear cells. However, there are no studies analysing the direct effects of BCAA on endothelial cells (ECs) and its possible modulation of vascular function. In vitro and ex vivo studies were performed in human ECs and aorta from male C57BL/6J mice, respectively. In ECs, BCAA (6 mmol/L) increased eNOS expression, reactive oxygen species production by mitochondria and NADPH oxidases, peroxynitrite formation and nitrotyrosine expression. Moreover, BCAA induced pro-inflammatory responses through the transcription factor NF-κB that resulted in the release of intracellular adhesion molecule-1 and E-selectin conferring endothelial activation and adhesion capacity to inflammatory cells. Pharmacological inhibition of mTORC1 intracellular signalling pathway decreased BCAA-induced pro-oxidant and pro-inflammatory effects in ECs. In isolated murine aorta, BCAA elicited vasoconstrictor responses, particularly in pre-contracted vessels and after NO synthase blockade, and triggered endothelial dysfunction, effects that were inhibited by different antioxidants, further demonstrating the potential of BCAA to induce oxidative stress with functional impact. In summary, we demonstrate that elevated BCAA levels generate inflammation and oxidative stress in ECs, thereby facilitating inflammatory cells adhesion and endothelial dysfunction. This might contribute to the increased cardiovascular risk observed in patients with elevated BCAA blood levels.

Keywords: BCAA; aorta; endothelial cells; endothelial dysfunction; inflammation; oxidative stress.

Figure 1

BCAA activate the PI3K/Akt‐mTORC1 and AMPK axis in endothelial cells. Effects of increasing BCAA concentrations (0.2‐12 mmol/L, 1 h) on A, mTORC1 and B, AMPK activation on human endothelial cells (ECs). Effect of BCAA (6 mmol/L, 1 h) on human ECs pre‐incubated 30 min with or without rapamycin (100 nmol/L) or AICAR (0.5 mmol/L) on C, AMPK, D, Akt and E, mTOR activation. F, eNOS mRNA and protein expression in ECs. mTOR, AMPK and Akt activation and eNOS expression were calculated as ratios of phosphorylated proteins vs corresponding total mTOR, GAPDH or α‐tubulin values and expressed as fold increase over control. For each panel, representative blots are shown above. Data are expressed as mean ± SEM. *P < .05; **P < .01 vs Control vs Control (C); ^≠ P < .05 vs BCAA. n = 6

Figure 2

BCAA induce ROS production via mTORC1. Effect of BCAA (6 mmol/L, 1 h) on human ECs pre‐incubated 30 min with or without rapamycin (RAPA, 100 nmol/L), AICAR (0.5 mmol/L), DPI (10 μmol/L), ML171 (ML, 0.5 μmol/L), gp91dstat (dstat, 5 μmol/L), mito‐TEMPO (MITO, 0.5 μmol/L) or BAY‐11‐7082 (BAY, 1 mmol/L) on A, NADPH oxidase activity and B, mitochondrial O₂ ^·−. C, Confocal microscopy images showing mitochondrial O₂ ^·− production using Mitosox (red) and DAPI for nuclei (blue). Effect of BCAA with or without rapamycin or AICAR on gene expression of the NADPH oxidase subunits NOX‐1 (D) and NOX‐2 (E). Data are expressed as mean ± SEM. *P < .05; **P < .01 vs Control vs Control (C). ^≠ P < .05 vs BCAA. n = 5‐7

Figure 3

BCAA trigger NF‐κB pathway and pro‐inflammatory genes expression. Effect of BCAA (6 mmol/L, 1 h) on human ECs pre‐incubated 30 min with or without rapamycin (RAPA), AICAR, ML171 (ML), gp91dstat (dstat), mito‐TEMPO (MITO) or BAY‐11‐7082 (BAY) on A, p65 phosphorylation, B, DNA‐binding activity of p65 and on gene expression of C, E‐selectin and D, ICAM‐1. E, Effect of BCAA on iNOS mRNA expression in ECs. Data are expressed as mean ± SEM. *P < .05; **P < .01 vs Control (C). ^≠ P < .05 vs BCAA. n = 6‐7

Figure 4

BCAA promote leucocytes adhesion to endothelium. Effect of BCAA (6 mmol/L, 1 h) on leucocytes adhesion to ECs pre‐incubated with or without rapamycin (RAPA), AICAR, wortmannin (W), mito‐TEMPO (MITO), gp91dstat (dstat), ML171 (ML) and BAY‐11‐7082 (BAY). Data are expressed as mean ± SEM. ***P < .001 vs Control (C). ^≠ P < .05; ^≠≠ P < .01 vs BCAA. n = 7

Figure 5

BCAA induce vasoconstrictor responses through ROS and COX‐2‐derived contractile prostanoids. Representative tracings showing the effect of BCAA (0.2‐10 mmol/L) on contraction of aortic segments in the absence (Control) (A) and in the presence of phenylphrine (Phe) pre‐contraction (B). Quantification of contractile responses (red arrows) is shown in C, together with the effect of gp91dstat, ML171 and mito‐TEMPO on concentration‐response curves to BCAA in aortic segments pre‐contracted with phenylephrine. D, Effect of gp91dstat, ML171, mito‐TEMPO and celecoxib on BCAA‐induced contractile responses in aortic segments incubated with L‐NAME (100 μmol/L, 30 min) without phenylephrine pre‐contraction. Concentration‐response curves to BCAA in aorta without endothelium (E‐) pre‐contracted with phenylephrine (E) or incubated with L‐NAME without phenylephrine pre‐contraction (F). Data represent mean ± SEM. ***P < .001 vs Control. ^≠ P < .05; ^≠≠ P < .01; ^≠≠≠ P < .001 vs Phe or L‐NAME. n = 7‐10

Figure 6

BCAA induce endothelial dysfunction through ROS. A, Contraction to depolarizing solution of high KCl (K+‐KHS) in arteries incubated with BCAA (6 mmol/L, 24 h) in the absence or presence of different antioxidants: gp91dstat, ML171, and mito‐TEMPO. Concentration‐response curves to phenylephrine (Phe) (B), acetylcholine (ACh) (C) and diethylamine NONOate (DEA‐NO) (D) in aortic segments incubated with BCAA in the absence or presence of gp91dstat, ML171 and mito‐TEMPO. Data represent mean ± SEM. *P < .05; ***P < .001 vs Control; ^≠ P < .05; ^≠≠ P < .01; ^≠≠≠ P < .001 vs BCAA.

Figure 7

BCAA induce nitrotyrosine expression and increase peroxynitrite levels. Effect of BCAA (6 mmol/L, 24 h) pre‐incubated 30 min with or without ML171 and gp91dstat (dstat) on A, protein nitrosylation levels in aortic sections and B, peroxinytrite levels measured in supernatants from aortic segments. Data are expressed as mean ± SEM. *P < .05 vs Control (C). ^≠ P < .05 vs BCAA. n = 3‐5

Figure 8

Scheme demonstrating the possible relationship between branched‐chain amino acids (BCAA), reactive oxygen species (ROS) and inflammation and its putative role in endothelial dysfunction and cardiovascular diseases. The influx of BCAA into endothelial cells is mediated by binding to specific nutrient transporters. In cytoplasm, BCAA activate PI3K‐Akt/mTORC1 and AMPK signalling pathways. The BCAA‐dependent activation of these pathways seems to induce NADPH oxidase activation, mitochondrial oxidative stress and nuclear transcription factor‐κB (NF‐κB) leading to increased production of ROS and pro‐inflammatory factors. BCAA‐induced oxidative and pro‐inflammatory status promote leucocytes migration and adhesion to the endothelium. In addition, ROS derived from mitochondria or through mechanisms involving NOX‐1 and NOX‐2 subunits of NADPH oxidase could reduce NO availability. These events, in turn, would induce endothelial dysfunction and vasoconstriction. All together, these vascular alterations might contribute to the development of cardiovascular diseases in clinical conditions associated with elevated levels of BCAA