4-HNE and 4-HHE were synthesised as previously described . Insulin (Actrapid, 100 U/ml) was from Novo Nordisk (La Défense, France) and ECL reagent and 2-deoxy-d-2,6-[3H]glucose were from GE Healthcare (Orsay, France). 1,2-Dithiole-3-thione (D3T) was from Interchim (Montlucon, France). All other reagents were from Sigma-Aldrich (Saint Quentin Fallavier, France).
Antibodies to Ser473-Akt, total Akt, p85 and total IRS1 (dilution 1:1000) were from Cell Signaling (Leiden, the Netherlands). Secondary antibodies (dilution 1:10,000) were from Sigma-Aldrich (Heidelberg, Germany). Anti-HHE-Michael adducts antibodies were from Cosmobio (Tokyo, Japan). Antibodies for insulin signalling were validated using insulin as a positive control. Anti-HHE antibody specificity was tested against protein lysates incubated with 4-HHE or 4-HNE.
Fifteen individuals with type 2 diabetes mellitus and 17 healthy volunteers were recruited from an ongoing study at Hospices Civils de Lyon. The characteristics of the participants are shown in Table 1. The study was approved by the local ethics committee (reference D-09-17) and written informed consent was obtained from all volunteers. After an overnight fast, blood samples were collected, centrifuged at 1500 g for 10 min to isolate plasma and stored at −80°C.
All animal experiments were performed under authorisation no. 69-266-0501 (INSA-Lyon, DDPP-SV, Direction Départementale de la Protection des Populations - Services Vétérinaires du Rhône) according to the guidelines laid down by the French Ministry of Agriculture (no. 2013-118) and the European Union Council Directive for the protection of animals (2010/63UE).
The rats used in all experiments were raised in an air-conditioned room with a controlled environment of 21 ± 0.5°C and 60–70% humidity, under a 12 h light–dark cycle (light on from 07:00 to 19:00 hours) with free access to food (2016C, 12.6 kJ/g; Harlan, Gannat, France) and water. Rats were housed together and randomised into groups on the day of the experiment using random numbers.
Zucker diabetic fatty rats
Five-week-old male lean (ZDF-Leprfa/+/?, n = 5) and obese Zucker diabetic fatty (ZDF) rats (ZDF-Leprfa/Crl, n = 5), a rat model of diabetes, were purchased from Charles River Laboratories (L’Arbresle, France). The characteristics of the rats are shown in Table 2. At 15 weeks of age, rats were deeply anaesthetised with sodium pentobarbital (120 mg/kg i.p.). Terminal cardiac blood puncture was collected into a heparinised syringe and blood was centrifuged for 2 min at 3500 g to prepare plasma, snap-frozen in liquid nitrogen and stored at −80°C. Blood glucose was measured with a glucometer (Accu-Check Performa; Roche, Meylan, France) and insulin was determined using an enzyme immunoassay (EIA) according to the manufacturer’s recommendations (A05005; Bertin Pharma, Montigny le Bretonneux, France). HbA1c was determined on whole blood using a rat glycated haemoglobin assay kit (80300; Crystal Chem, Zaandam, the Netherlands).
Clamps were performed in 3-month-old anaesthetised male Wistar rats  purchased from Janvier (Le-Genest-Saint-Isle, France). Rats were fasted overnight, anaesthetised with sodium pentobarbital (35 mg/kg i.p.; Sanofi Santé Animale, Centravet, Lapalisse, France) and chlorpromazine (5 mg/kg, i.p., Largactil; Sanofi-Aventis, Paris, France) and then implanted with indwelling catheters (PE-20; Phymep, Paris, France) in the left carotid artery and left and right jugular veins. After catheterisation, four rats were infused with 4-HHE (10 mg/kg, 0.1 ml) and four rats were infused with the vehicle (DMSO, 0.1 ml) in the left jugular vein. A standard 2 h clamp  was conducted using a primed and continuous infusion of human recombinant insulin (Actrapid) at a rate of 6 mU kg−1 min−1 coupled with a variable infusion of 25% (wt/vol.) glucose to maintain blood glucose concentration at approximately 6 mmol/l. Blood glucose was measured every 5 min using a glucometer (Accu-Check Performa). Plasma insulin concentration was determined at the end of the clamp by EIA assay (SpiBio, Montigny le Bretonneux, France).
In vivo insulin stimulation and insulin signalling in skeletal muscle
Male Wistar rats, fasted for 7 h prior to the experiments, were anaesthetised with sodium pentobarbital (35 mg/kg i.p., Sanofi Santé Animale, Centravet) and chlorpromazine (5 mg/kg i.p., Largactil; Sanofi-Aventis). Body temperature was maintained at 37°C with a homeostatic blanket (Harvard apparatus, les Ullis, France). A tracheotomy (PE-240; Phymep, Paris, France) was performed to facilitate breathing. A catheter (PE-20) was inserted into the left jugular vein for 4-HHE infusion. Rats were either infused with 4-HHE (10 mg/kg in 0.1 ml DMSO, n = 5) or 0.1 ml of DMSO as a control (n = 4). One hour after infusion, rats received an intravenous injection of either saline (154 mmol/l NaCl) or insulin (0.75 U/kg body weight). Thirty minutes after insulin injection, rats were killed with an intravenous overdose of sodium pentobarbital (120 mg/kg), blood was collected by heart puncture into a heparinised syringe and gastrocnemius muscle was rapidly dissected out, blotted dry and snap-frozen in liquid nitrogen.
Plasma 4-HHE and 4-HNE measurement
Five hundred microlitres of plasma were used to measure 4-HHE and 4-HNE by GC–MS as previously described .
Rat L6 muscle cells were obtained from the American Type Culture Collection (LC/GC, Molsheim, France) and grown as described previously . Cells were starved of serum for 4 h before treatments, which were performed in serum-free Minimum Essential Medium Eagle Alpha Modification (Sigma-Aldrich, Saint Quentin Fallavier, France). Viability and cell death were determined by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) reduction (Cell Proliferation Kit I; Roche), lactate dehydrogenase release (In vitro toxicology assay kit; Sigma-Aldrich) and caspase 3 activity (Caspase 3 Assay Kit; Sigma-Aldrich). Mycoplasma contamination was regularly tested by PCR.
Cells were treated with 4-HHE and incubated for 20 min with 100 nmol/l insulin or 20 μmol/l cytochalasin B. Glucose uptake was initiated by the addition of 2-deoxy-d-[3H]glucose (747 GBq/mmol) to a final concentration of 0.1 mmol/l for 5 min at 37°C. Uptake was terminated by three washes in ice-cold PBS and solubilisation in 0.1% (wt/vol.) SDS. Tritium was detected by liquid scintillation counting and results normalised by protein concentration measured using the Bradford assay. Non-specific uptake measured in presence of cytochalasin B was subtracted from each determination.
Spectrophotometric 2,4-dinitrophenylhydrazine assay
Carbonyl groups on proteins were determined using 2,4-dinitrophenylhydrazine (DNPH) as previously described . Carbonyl content was determined from the absorbance at a wavelength of 370 nm using a molar absorption coefficient of 22,000 l mol−1 cm−1 and normalised to the protein concentration measured at 280 nm.
Protein extraction, immunoprecipitation and immunoblotting
After incubation with 4-HHE, cells were washed with PBS and proteins were extracted as described previously . For IRS1/p85 detection, whole-cell lysates were immunoprecipitated using antibodies against IRS1. Proteins were boiled in Laemmli buffer, separated by SDS-PAGE and transferred onto nitrocellulose membrane. For dot blotting, 20 μg of proteins from whole-cell lysate were loaded directly onto the nitrocellulose membrane. Following saturation with 5% (wt/vol.) BSA, membranes were probed with antibodies and processed for chemiluminescence (ECL plus; GE Healthcare, Coutaboeuf, France). Quantification was performed using the ImageJ software (National Institutes of Health, Bethesda, USA).
L6 cells grown on coverslips were incubated with 4-HHE, fixed for 30 min with 3% (wt/vol.) paraformaldehyde and quenched with 50 mmol/l NH4Cl for 5 min. After PBS washing, cells were permeabilised with 50 μg/ml digitonin for 10 min and blocked with 0.1% (wt/vol.) BSA for 30 min. The cells were incubated with the first antibodies and then washed and labelled with fluorescent secondary antibodies. The stained cells mounted in Mowiol (Sigma Aldrich, Saint Quentin Fallavier, France) were examined under a Zeiss LSM 510 confocal microscope (Zeiss, Marly le Roy, France) equipped with 63W oil objective and fluorescence was quantified using Image J software v1.56.
Reduced glutathione (GSH) was measured using a commercially available kit from BioVision (Clinisciences, Montrouge, France) and normalised to the protein concentration measured using the Bradford assay.
Experimenters were not blind to group assignment. Data were analysed using Graphpad Prism (Graphpad Software, La Jolla, CA, USA) and presented as means ± SEM. Results were compared by one-way or two-way ANOVA followed when appropriate by post hoc Fisher PLSD tests. Simple comparisons were performed with Student’s t test using Welch’s correction for variance in homogeneity whenever needed. Differences were considered significant at the p < 0.05 level.