Surgery was followed by a 20-minute stabilisation period. Thereafter, haemorrhagic shock was induced by the graded withdrawal of blood from the femoral artery to a reservoir until MAP decreased to 40 mmHg and maintained during 60 minutes by further blood withdrawal or reinfusion of shed blood. At 60 minutes, shed blood was retransfused via the venous line within 10 minutes. Animals were continuously monitored for HR, MAP and CBF during 300 minutes. Hydration was performed with a perfusion of 0.9% NaCl at a rate of 1.2 ml/hour.

At the end of the experiment, rats were sacrificed and blood samples were collected for arterial lactate measurement, centrifuged (4,000 rpm, 15 minutes, 4°C) and plasma aliquoted and stored at -80°C until biochemical analysis. Organs (aorta, heart and liver) were also collected and stored at -80°C until biochemical analyses.

The dehydrated NaHS powder (anhydrous, 2 g; Alpha Aesar GmbH & Co, Ward Hill, MA , USA) was dissolved in isotonic saline under argon gas bubbling until a concentration of 40 mM was obtained and intravenously administered as a single bolus (0.2 mg/kg body weight) 10 minutes before retransfusion or 10 minutes after the end of retransfusion (late NaHS). PNU-37883A (guanidine; 4-morpholinecarboximidine-N-1-adamantyl-N'-cyclohexyl hydrochloride) (Sigma Aldrich, St Quentin Fallavier, France) was dissolved in a 1:1 mixture of dimethyl sulphoxide and intravenously administered as a bolus (1.5 mg/kg) followed by 1 mg/kg/hour. The inducible NO synthase inhibitor 1400W (Sigma Aldrich) was administered intraperitoneally (20 mg/kg) at T0.

Eight groups of eight rats were studied, namely: sham rats, haemorrhagic shocked rats, shock + PAG (CSE inhibitor)-treated rats (50 mg/kg), shock + NaHS-treated rats, shock + late NaHS-treated rats, shock + PNU-37883A-treated rats, shock + PNU + NaHS-treated rats, and shock + 1400W-treated rats.

Arterial blood gases were controlled after the stabilisation period, in order to establish mechanical ventilation. Measurements of blood gas and blood glucose were recorded at baseline (t = 0 minutes at the beginning of haemorrhagic shock) and at two critical periods, namely at the end of reperfusion (t = 70 minutes) and at the end of the experiment (t = 300 minutes). MAP, HR, CBF and rectal temperature were recorded at baseline and every 10 minutes thereafter during the observation period.

Lactate concentrations were determined using an automated blood gas analyser (ABL5 Radiometer; Neuilly-Plaisance, France).

Plasma levels of IL-6 and TNFα were measured in duplicate with the use of rat IL-6 and TNFα ELISA kits (Quantikine ELISA; R&D Systems Europe, LILLE, France) according to the manufacturer's instructions. Results were expressed as picograms of the measured cytokine per millitre of plasma.

NO₂^- and NO₃^- are the primary oxidised products of NO reacting with water, and therefore the total concentration of NO₂^-/NO₃^- in plasma was used as an indicator of NO production in vivo. Briefly, the nitrate in the supernatant was first reduced to nitrite by incubation with nitrate reductase (10 U/ml) and NADPH (629.2 µg/ml) at room temperature for 30 minutes. Thereafter, total nitrite concentration in the samples was measured by Griess reaction following the addition of 100 µl Griess reagent to 100 µl sample in a 96-well plate with a flat transparent bottom. The optical density at 550 nm was measured by an ELISA microplate reader and normalised with the optical density at 550 nm of standard saline solutions.

Primers for quantitative RT-PCR were obtained from Eurogentec (Angers, France). Total RNA extraction was carried out with the RNA Plus mini kit (Qiagen, Courtaboeuf Cedex, France) according to the manufacturer's instructions. Total RNA was reverse-transcribed to cDNA using the iScript One-Step RT-PCR Kit for Probes (Biorad, Marnes-la-Coquette, France). cDNA obtained from the RT reaction was subjected to quantitative PCR using iTaq Fast SYBR Green Supermix with ROX (Biorad, Marnes-la-Coquette, France). The primer and concentrations were optimised according to the manufacturer's guidelines. Expression of, Kir6.1 mRNA and SUR2B mRNA were measured using iTaq Fast SYBR Green Supermix (Biorad).

The PCR reaction parameters were as follows: incubation at 50°C for 2 minutes, incubation at 95°C for 10 minutes, and thereafter 40 denaturation cycles at 95°C for 15 seconds and annealing and extension at 60°C for 1 minute. Each sample was determined in duplicate. To determine the relative mRNA levels, a standard curve for each gene was created using RNA isolated from the haemorrhagic shock group. Isolated RNA was reverse-transcribed, and dilution series of cDNA ranging from 1 pg to 10 ng were subjected to real-time PCR. The obtained threshold cycle values were plotted against the dilution factor to create a standard curve. Relative mRNA levels in test samples were then calculated from the standard curve.

For in vivo determination, basal and maximal MAP values obtained after administration of 1 µg/kg bolus of norepinephrine were recorded in the sham, haemorrhagic shock, haemorrhagic shock + PNU and haemorrhagic shock + 1400W groups.

For ex vivo determination, aortic rings and small mesenteric arteries were carefully dissected and mounted on a wire myograph (Danish Myo Technology, Arhus, Denmark). The experiments were performed at 37°C in a physiological salt solution with the following composition: NaCl 119 mM; KCl 4.7 mM; NaHCO₃ 14.9 mM; MgSO₄·7H₂O 1.2 mM; CaCl₂ 2.5 mM; KH₂PO₄ 1.18 mM; glucose 5.5 mM, continuously bubbled with 95% O₂ and 5% CO₂.

After an equilibration period (at least 20 minutes) under optimal passive tension, two successive contractions in response to the combination of KCl depolarisation (100 mM) and phenylephrine (PE) (10 µM) (Sigma-Aldrich) were used in order to test the maximal contractile capacity of the vessels. After a 20-minute washout period, concentration-response curves to PE were elicited by cumulative administration of this vasoconstrictor agonist (1 nM to 100 µM) in order to determine the same concentration producing an equal level of contraction in the different groups. To study endothelium-dependent relaxation, aortic rings with functional endothelium were precontracted with PE (1 µM) and then exposed to increasing incremental concentrations of acetylcholine (1 nM to 100 µM; Sigma, St Louis, MO, USA). The presence of functional endothelium was confirmed with acetylcholine (1 µM), which elicited a relaxation superior to 50%.

Aorta and small mesenteric arteries (200 to 230 μm) were homogenised and lysed. Proteins (20 µg) were separated on 10% SDS-PAGE. Blots were probed with the following antibodies: anti-iNOS (BD Biosciences, San Jose, CA, USA), phosphorylated endothelial nitric oxide synthase (p-eNOS) (rabbit anti-rat eNOS, phosphorylated (ser1177); Cell Signaling Technology Saint Quentin Yvelines, France), phosphorylated-Akt (p-Akt) (rabbit anti-rat Akt, phosphorylated (ser473); Cell Signaling Technology), phospho-SAPK/JNK (mouse, anti-rat SAPK/JNK, phosphorylated (Thr183/Tyr185); Cell Signaling Technology), phospho-p38 mitogen-activated protein kinase (mouse, anti-rat p38 MAPK, phosphorylated (Thr180/Tyr182); Cell Signaling Technology), and phosphor-p44/42 MAPK (Erk1/2) (rabbit anti-rat p44/p42 MAPK, phosphorylated (Thr1202/Tyr204); Cell Signaling Technology). Proteins were transferred onto nitrocellulose membranes and probed with a monoclonal mouse anti-α-Tubulin antibody (Sigma-Aldrich).

Bound antibodies were detected with a secondary peroxidase-conjugated anti-mouse IgG (Promega, Madison, WI, USA). The blots were visualised using an enhanced chemiluminescence system (ECL Plus; Amersham, GE Healthcare Europe, Velizy-Villacoublay, France).

Results are expressed as the median and interquartile range for n experiments (n representing the number of animals). Difference between groups was tested using a Kruskal-Wallis test. When the relevant F values were significant at the 5% level, further pairwise comparisons were performed using a Dunn's multiple comparison test. All statistics were performed with the Statview software (version 5.0 software; SAS Institute, Cary, NC, USA). P <0.05 was considered statistically significant.