CABG was performed under normothermic (36-37°C) cardiopulmonary bypass (CPB). All components of the circuits were coated with phosphorylcholine inert surface (PHISIO, Sorin®). The pump manufacturer is Maquet® for the roller pumps.1.5 cm of ITA distality was sampled before blood flow interruption into the graft and before starting CPB (Time 1) and another segment (1.5 cm) before the last coronary anastomosis during aortic cross clamping (Time 2) (Figure 1). Each arterial segment was cut into three parts: a fresh part for arterial myography bathed and stored in a 50 ml organ bath containing a physiological salt solution (PSS). The other two parts were cooled in liquid nitrogen and stored at -80°C for immunohistochemistry and RT-PCR analysis.

CABG was performed under normothermic (36-37°C) cardiopulmonary bypass (CPB). All components of the circuits were coated with phosphorylcholine inert surface (PHISIO, Sorin®). The pump manufacturer is Sorin® for the centrifugal pumps.1.5 cm of ITA distality was sampled before blood flow interruption into the graft and before starting CPB (Time 1) and another segment (1.5 cm) before the last coronary anastomosis during aortic cross clamping (Time 2) (Figure 1). Each arterial segment was cut into three parts: a fresh part for arterial myography bathed and stored in a 50 ml organ bath containing a physiological salt solution (PSS). The other two parts were cooled in liquid nitrogen and stored at -80°C for immunohistochemistry and RT-PCR analysis.

1. Internal thoracic arteries analysis Myography Superoxide detection and confocal microscopy Immunochemistry Quantitative real time transcription-polymerase chain reaction (RT-PCR) analysis
2. Blood sampling and biochemical analysis

for each patient, 2 fresh segments of ITA (Time 1 and Time 2) stored in PSS were analyzed. On day+1, these segments were mounted on a wire-myograph (DMT, Aarhens, DK) . Two tungsten wires (25 μm diameter) were inserted into the lumen of the arteries and connected to a force transducer and a micrometer, respectively. The arteries were bathed in the PSS solution. Wall tension, equivalent to intra-arterial pressure (90 mmHg), was applied and the blood vessels were allowed to stabilize for thirty minutes. Arterial contractility was assessed with phenylephrine (PE, 10 μmol/L). Acetylcholine-induced (Ach 10 μmol/L) relaxation was then obtained after phenylephrine-induced preconstruction (50% of maximal contraction) in the presence or in the absence of the NO synthesis blocker L-NMMA (3.10-4 mol/L) and in the presence or in the absence of the COX synthesis blocker Indomethacin (10-5 mol/L).

: dihydroethidium staining (DHE, Sigma-Aldrich) was used to evaluate the in-situ levels of superoxide anions (O2-). DHE is permeable to cells and is oxidized by superoxide (O2-) to fluorescent products that are trapped by intercalation into the DNA. Sections (10 μm thickness) were incubated with DHE (1 μmol/L) in phosphate-buffered solution (PBS) and DAPI (4',6'-diamidino-2-phénulindole-Molecular probes, Invitrogen) for nuclear cells Fluorescent images of ethidium bromide were obtained using a confocal microscope (Nikon Eclipse TE2000S).