Mechanisms involved in the desflurane-induced post-conditioning of isolated human right atria from patients with type 2 diabetes

British Journal of Anaesthesia 107 (4): 510 18 (2011) Advance Access publication 25 July 2011. doi:10.1093/bja/aer201 Mechanisms involved in the desflurane-induced post-conditioning of isolated human right atria from patients with type 2 diabetes S. Lemoine 1, L. Zhu 2, C. Buléon 2, M. Massetti 4, J-L. Gérard 1,2, P. Galera 3 and J-L. Hanouz 1,2 1 Laboratory of Experimental Anaesthesiology and Cellular Physiology EA3212, Institut Fédératif de Recherche ICORE146, Université de Caen Basse Normandie, Caen, France 2 Department of Anaesthesiology and Intensive Care and 3 Laboratory of Extracellular Matrix and Pathology EA 3214 and Pathology, CHU Caen, Caen, France 4 Department of Cardiac and Surgery, CHU de Caen, Caen, France Corresponding author. E-mail: sand.lemoine2@wanadoo.fr Editor s key points Desflurane is known to protect the diabetic heart against subsequent reperfusion injury. Using an ex vivo model, mechanisms of desflurane post-conditioning on the diabetic heart were assessed. Key mechanisms were identified which were associated with cardioprotection. The authors suggest that post-conditioning may be beneficial in diabetic patients. Background. Desflurane triggers post-conditioning in the diabetic human myocardium. We determined whether protein kinase C (PKC), mitochondrial adenosine triphosphatesensitive potassium (mitok ATP ) channels, Akt, and glycogen synthase kinase-3b (GSK-3b) were involved in the in vitro desflurane-induced post-conditioning of human myocardium from patients with type 2 diabetes. Methods. The isometric force of contraction (FoC) of human right atrial trabeculae obtained from patients with type 2 diabetes was recorded during 30 min of hypoxia followed by 60 min of reoxygenation. Desflurane (6%) was administered during the first 5 min of reoxygenation either alone or in the presence of calphostin C (PKC inhibitor) or 5-hydroxydecanoate (5-HD) (mitok ATP channel antagonist). Phorbol 12-myristate 13-acetate (PKC activator) and diazoxide (a mitok ATP channel opener) were superfused during early reoxygenation. The FoC at the end of the 60 min reoxygenation period was compared among treatment groups (FoC 60 ; mean and SD). The phosphorylation of Akt and GSK-3b was studied using western blotting. Results. Desflurane enhanced the recovery of force [FoC 60 : 79 (3)% of baseline] after 60 min of reoxygenation when compared with the control group (P.0.0001). Calphostin C and 5- HD abolished the beneficial effect of desflurane-induced post-conditioning (both P,0.0001). Phorbol 12-myristate 13-acetate and diazoxide enhanced the FoC 60 when compared with the control group (both P,0.0001). Desflurane increased the level of phosphorylation of Akt and GSK-3b (P,0.0001). Conclusions. Desflurane-induced post-conditioning in human myocardium from patients with type 2 diabetes was mediated by the activation of PKC, the opening of the mitok ATP channels, and the phosphorylation of Akt and GSK-3b. Keywords: desflurane; diabetes; myocardium; post-conditioning Accepted for publication: 29 April 2011 Diabetes increases mortality and morbidity after coronary surgery and acute myocardial infarction. 1 3 In addition, poor intraoperative blood glucose control has been shown to increase morbidity in diabetic patients. 4 Experimental studies have suggested that the cardioprotective effect of ischaemic and anaesthetic preconditioning and postconditioning is impaired or abolished in diabetic and hyperglycaemic animals. 5 8 We have shown that brief administration of desflurane during early reoxygenation in vitro is able to induce post-conditioning in human myocardium from patients with type 2 diabetes. 9 At the present time, experimental studies on the mechanisms and cellular signalling pathways that specifically mediate the effects of desflurane-induced post-conditioning in the diabetic myocardium are rare. In the isolated non-diabetic human myocardium, it has been shown that protein kinase C (PKC) activation, the opening of the mitochondrial adenosine triphosphate-sensitive potassium (mitok ATP ) channels, and some key components of the Reperfusion Injury Salvage Kinase pathway were involved in desflurane-induced post-conditioning. 10 The goal of the present study was to examine the involvement of PKC and mitok ATP channels in desflurane-induced post-conditioning using isolated human right atrial muscle obtained from patients with type 2 diabetes. Furthermore, we examined the & The Author [2011]. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved. For Permissions, please email: journals.permissions@oup.com

Desflurane-induced post-conditioning phosphorylation of Akt and glycogen synthase kinase-3b (GSK-3b), which are important proteins involved in the prosurvival Reperfusion Injury Salvage Kinase pathway. Methods After the approval of the local medical ethics committee and after receiving written informed consent, right atrial appendages were obtained during cannulation for cardiopulmonary bypass from non-diabetic patients and patients with type 2 diabetes, who were undergoing routine coronary artery bypass surgery or aortic valve replacement. All patients received total i.v. anaesthesia with propofol, sufentanil, and pancuronium. Preoperative glycosylated haemoglobin (HbA1c) was measured during the operative period; in all patients (with and without diabetes), continuous infusion of fast-acting insulin (Actrapid H; Novo Nordisk Pharmaceutique, Puteaux, France) was initiated as soon as arterial blood glucose concentration exceeded 180 mg dl 21. Subsequently, the infusion rate was titrated according to a local protocol (,180 mg dl 21, 0 units h 21 ; 180 220 mg dl 21, 1 units h 21 ; 221 249 mg dl 21, 2 units h 21 ;.249 mg dl 21, 3 units h 21 ). Arterial blood glucose concentration was measured every 60 and 30 min after change in the rate of infusion. The rate of infusion was titrated according to the following protocol:,140 mg dl 21, infusion stopped until 180 mg dl 21 ; 140 179 mg dl 21, the rate of infusion was decreased by 0.5 units h 21 ; 180 220 mg dl 21, no change in the rate of infusion; 221 249 mg dl 21, the rate of infusion was increased or decreased by 0.5 units h 21 according to the previous blood glucose concentration;.249 mg dl 21, the rate of infusion was increased by 1 units h 21. Patients with chronic atrial arrhythmia were excluded from the study. The data for the patient characteristics, preoperative drug treatment, preoperative left ventricular ejection fraction, and HbA1c are reported in Table 1. Human atrial trabeculae model of hypoxia/ reoxygenation injury Experimental conditions Right atrial trabeculae (one to two per appendage) were dissected and suspended vertically between an isometric force transducer (MLT0202, ADInstruments, Sydney, Australia) and a stationary stainless clip in a 200 ml jacketed reservoir filled with daily prepared modified Tyrode s solution containing Table 1 Patient characteristic data, preoperative drug treatments, and preoperative left ventricular ejection fraction. The number in parentheses in the columns Groups and heart disease and Preoperative drug treatments indicates the number of patients. Age is expressed as mean (range), and LVEF, and HBA1c are expressed as mean (SD). At atrial appendage dissection. ACE, angiotensin-converting enzyme inhibitors; AVR, aortic valve replacement; BA, b-adrenergic-blocking drugs; BZD, benzodiazepine; CA, calcium channel antagonists; CABG, coronary artery bypass graft; COR, amiodarone; Des, desflurane; FUR, furosemide; K + A, mitochondrial adenosine triphosphate-sensitive potassium channel agonists; K + AT, mitochondrial adenosine triphosphate-sensitive potassium channel antagonists; STA, statins; Cal, calphostin C; Diab, diabetic; HBA1c, glycosylated haemoglobin; 5-HD, 5-hydroxydecanoate; PMA, phorbol 12-myristate 13-acetate Groups and heart disease Age (yr) Preoperative drug treatments LVEF (%) Diab-Control: AVR (n¼4); CABG (n¼6) Diab-Des: AVR (n¼2); CABG (n¼4) Diab-Des+Cal: AVR (n¼4); CABG (n¼2) Diab-Des+5-HD: AVR (n¼0); CABG (n¼6) Diab-Cal: AVR (n¼4); CABG (n¼2) Diab-5-HD: AVR (n¼2); CABG (n¼4) Diab-PMA: AVR (n¼4); CABG (n¼2) Diab-Diazoxide: AVR (n¼3); CABG (n¼3) Western: Control: AVR (n¼3); CABG (n¼1) Western: Des: AVR (n¼2); CABG (n¼2) Western: Diab-Control: AVR (n¼0); CABG (n¼4) Western: Diab-Des: AVR (n¼3); CABG (n¼1) 61 (47 80) ACE (6), BA (6), BZD (2), CA (2), COR (0), FUR (1), K + A (0), K + AT (1), MET (3), INS (0), STA (7) 71 (65 81) ACE (3), BA (4), BZD (1), CA (2), COR (0), FUR (1), K + A (0), K + AT (0), MET (0), INS (1), STA (3) 68 (61 73) ACE (2), BA (4), BZD (2), CA (1), COR (0), FUR (0), K + A (2), K + AT (0), MET (0), INS (0), STA (6) 72 (68 74) ACE (2), BA (4), BZD (0), CA (0), COR (0), FUR (O), K + A (2), K + AT (1), MET (0), INS (0), STA (4) 70 (60 77) ACE (2), BA (2), BZD (0), CA (3), COR (0), FUR (0), K + A (2), K + AT (0), MET (0), INS (0), STA (3) 68 (57-0) ACE (3), BA (2), BZD (3), CA (2), COR (2), FUR (0), K + A (2), K + AT (0), MET (0), INS (0), STA (8) 68 (55 79) ACE (2), BA (4), BZD (0), CA (0), COR (0), FUR (0), K + A (0), K + AT (0), MET (0), INS (0), STA (4) 71 (65 79) ACE (5), BA (4), BZD (0), CA (1), COR (0), FUR (1), K + A (2), K + AT (0), MET (0), INS (1), STA (8) 58 (50 72) ACE (1), BA (3), BZD (0), CA (0), COR (0), FUR (0), K + A (0), K + AT (0), MET (0), INS (0), STA (2) 65 (58 74) ACE (2), BA (3), BZD (0), CA (1), COR (0), FUR (0), K + A (0), K + AT (0), MET (0), INS (0), STA (4) 66 (57 76) ACE (3), BA (3), BZD (2), CA (0), COR (0), FUR (0), K + A (1), K + AT (0), MET (0), INS (0), STA (3) 68 (63 71) ACE (1), BA (0), BZD (0), CA (0), COR (0), FUR (0), K + A (1), K + AT (0), MET (0), INS (0), STA (1) HbA1c (%) Blood glucose (mg dl 21 ) 63 (17) 7.4 (1.4) 6.8 (1.2) 64 (10) 7.4 (0.8) 5.8 (1.8) 54 (3) 6.3 (0.9) 6.1 (0.9) 77 (2) 6.0 (0.5) 6.0 (0.3) 73 (7) 6.5 (0.6) 6.7 (2.0) 66 (10) 6.7 (0.7) 6.8 (1.1) 70 (9) 6.6 (1.2) 6.1 (0.2) 47 (19) 7.7 (1.7) 7.4 (1.2) 70 (8) 5.9 (0.3) 5.8 (0.3) 57 (18) 5.8 (0.5) 5.7 (0.4) 66 (15) 7.1 (0.7) 6.7 (1.5) 70 (9) 6.8 (0.9) 6.2 (0.7) 511

Lemoine et al. (mm) 120 NaCl, 3.5 KCl, 1.1 MgCl 2, 1.8 NaH 2 PO 4, 25.7 NaHCO 3, 2.0 CaCl 2, and 5.5 glucose. The jacketed reservoir was maintained at 348C using a thermostatic water circulator (Polystat micropros, Bioblock, Illkirch, France). The bathing solution was oxygenated with 95% O 2 and 5% CO 2, resulting in a ph of 7.40 and PO 2 of 600 mm Hg. Isolated muscles were field-stimulated at 1 Hz by two platinum electrodes with rectangular wave pulses with a 5 ms duration, 20% above threshold (CMS 95107, Bionic Instruments, Paris, France). Trabeculae were equilibrated for 90 min to allow optimal mechanical recovery and performance at the apex of the length-active isometric tension curve. At the end of the stabilization period, the trabeculae were randomized to one of the experimental groups detailed below. Trabeculae obtained from the same appendage were included in different experimental groups. The force of contraction (FoC) was measured continuously, digitized at a sampling frequency of 400 Hz, and stored on a writable compact disc for analysis (MacLab, ADInstruments). At the end of each experiment, the length and weight of the muscles were measured. The cross-sectional area of the muscle was calculated, assuming a cylindrical shape and a density of 1. Trabeculae with a cross-sectional area more than 1.0 mm 2 were excluded. Accordingly, the preparations included had a FoC normalized per cross-sectional area.5.0 mn mm 22, and the ratio of resting to total force was,0.45. The endpoint of the study was the recovery of the FoC at 60 min of reoxygenation (FoC 60, expressed as a percentage of baseline). Experimental protocol At the end of the stabilization period, the trabeculae were randomly assigned (sealed envelopes) to one of the experimental groups. In all groups, hypoxia reoxygenation was performed by replacing 95% O 2 5% CO 2 with 95% N 2 5% CO 2 in the buffer for 30 min, followed by a 60 min oxygenated recovery period. In the diabetic control group (Diab-Control; n¼10), the trabeculae were treated according to the hypoxia reoxygenation protocol with no drug treatment (Fig. 1). In the desflurane treatment groups, desflurane was delivered to the organ bath using a specific calibrated vaporizer. A Diab-Control group (n =10) 30 min hypoxia 5 min desflurane 6% 60 min reoxygenation Diab-Des group (n =6) Diab-Des + Cal group (n =6) Diab-Des + 5-HD group (n =6) 5 min desflurane 6% Calphostin C, 5-HD Diab-Cal group (n =6) Diab-5-HD group (n =6) Calphostin C, 5-HD Diab-PMA group (n =6) Diab-Diazoxide group (n =6) B Control group (n = 4) Diab-Control group (n = 4) PMA, Diazoxide 30 min hypoxia Reoxygenation Tissue sampling 5 min desflurane 6% Des group (n = 4) Diab-Des group (n = 4) Tissue sampling Fig 1 Schematic diagram depicting the experimental protocol. (A) Contracting muscle experimental protocols. In the Diab-Des+Cal and Diab-Cal groups, calphostin C was administered at 1 mm. In the Diab-Des+5-HD and Diab-5-HD groups, 5-HD was administered at 800 mm. In the Diab-PMA and Diab-Diazoxide groups, PMA was administered at 1 mm and diazoxide was administered at 100 mm, respectively. (B) Western blot experimental protocols. Cal, calphostin C; Des, desflurane; Diab, diabetic; 5-HD, 5-hydroxydecanoate; PMA, phorbol 12-myristate 13-acetate. 512

Desflurane-induced post-conditioning The desflurane concentration in the carrier gas phase was measured with a calibrated infrared analyser (Capnomac; Datex, Helsinki, Finland). Desflurane-induced post-conditioning was triggered with 6% desflurane (Diab-Des; n¼6) during the first 5 min of reoxygenation; this concentration corresponds to a one minimum alveolar concentration of desflurane in adult humans at 378C. The mechanisms involved in desflurane-induced post-conditioning were studied in the presence of 1 mm calphostin C, a PKC inhibitor (Diab-Des+Cal; n¼6), and 800 mm 5-hydroxydecanoate (5-HD), a mitok ATP channel antagonist (Diab-Des+5-HD; n¼6). Pharmacological agents were administered 5 min before, throughout, and 10 min after desflurane exposure. Desflurane at 6% was chosen because we have previously shown that this is the optimal concentration for in vitro induction of post-conditioning in human myocardium. 9 In the pharmacological control groups, the muscles were exposed to 1 mm calphostin C (Diab-Cal; n¼6) and 800 mm 5-HD (Diab-5-HD; n¼6) 5 min before and during the first 15 min of reoxygenation (Fig. 1). In the phorbol 12-myristate 13-acetate (PMA) and diazoxide treatment groups, atrial trabeculae were treated according to the hypoxia reoxygenation protocol and were exposed during the first 15 min of reoxygenation to 1 mm PMA, a PKC activator (Diab-PMA; n¼6), and 100 mm diazoxide, a mitok ATP channel opener (Diab-Diazoxide; n¼6) (Fig. 1). The concentrations of calphostin C, 5-HD, PMA, and diazoxide used have been validated in previous in vitro studies 10 11 using human myocardium. Chemicals Calphostin C, diazoxide, and 5-HD were purchased from Calbiochem/VWR International (Fontenay sous Bois, France), and PMA was obtained from Sigma Aldrich (Saint Quentin Fallavier, France). Desflurane was purchased from Glaxo Wellcome (Marly-le-Roi, France). Western blot analysis The right atrial appendage was pinned in a chamber (5 ml) containing modified Tyrode s solution, oxygenated with 95% O 2 5% CO 2 and maintained at 34 (0.5)8C. The preparation was stimulated at a frequency of 1 Hz. In all groups, hypoxia was induced after a 90 min equilibration period by replacing 95% O 2 5% CO 2 with 95% N 2 5% CO 2 in the buffer for 30 min, followed by a 5 min oxygenated recovery period (Control, n¼4, and Diab-Control, n¼4) or a 5 min exposure to 6% desflurane (Des, n¼4 and Diab-Des, n¼4) (Fig. 1). For the western blot, atrial samples were frozen in liquid nitrogen and stored at 2808C before protein extraction and western blot analysis. Frozen tissue samples were subjected to SDS polyacrylamide gel electrophoresis as previously described. 10 After protein transfer, polyvinylidene fluoride membranes (Immobilon-P) were probed with antibodies for phospho-akt (Ser473), total Akt, phospho-gsk-3b (Ser9), and total GSK-3b (1/1000 dilution; Cell Signaling Technology, Ozyme, Saint Quentin Yvelines, France). The western blots of each group were stripped and probed again with an antibody against b-tubulin (Santa Cruz Technology) to ensure equivalent loading. The developed films were scanned, and the band densities were quantified using NIH Image J software (Research Service Branch, National Institutes of Mental Health, Bethesda, MD, USA). The phospho-akt level (Ser473) was normalized to the total Akt level, and the phospho-gsk-3b level (Ser9) was normalized to the total GSK-3b level. Glycosylated haemoglobin measurement Arterial blood samples for HbA1c (n¼66) measurement by high-performance liquid chromatography (ADAMS-A1c HA 81-60, Menarini Diagnostics, Firenze, Italy) were obtained before the cardiopulmonary bypass. Statistical analysis The endpoint of the study was the recovery of the FoC at 60 min of reoxygenation (FoC 60, expressed as a percentage of baseline). The power analysis demonstrated that a group size of n¼5 was necessary to detect a difference of 40% in the FoC [control and inhibitor groups: FoC 60 ¼50 (9)% of baseline, and desflurane (6%) group: FoC 60 ¼90 (9)% of baseline] with a power of 0.8 at an a-level of 0.05. The number of experiments per group was calculated based on a one-way analysis of variance (ANOVA) with four control and inhibitor groups and one 6% desflurane-treated group. Data are expressed as the mean (SD). Baseline values of the primary mechanical parameters, patient age, preoperative left ventricular ejection fraction, and the FoC 60 were compared by univariate ANOVA with the group factor as the independent variable. If the P-value was,0.05, a Bonferroni post hoc analysis was performed. Within-group data were analysed over time using a two-way ANOVA for repeated measures and the Bonferroni post hoc analysis with group factor and time (baseline; hypoxia for 5, 10, 20, 30 min; and reoxygenation for 5, 10, 20, 30, 40, 50, and 60 min) as the independent variables. In western blotting, the band densities for the protein of interest were then normalized to that of the band for b-tubulin in the same sample, and values are expressed as the mean (SD). Statistical comparisons were made by the use of the ANOVA for repeated measures and the Bonferroni post hoc analysis. All P-values were two-tailed, and a P-value of,0.05 was required to reject the null hypothesis. Statistical analysis was performed using Statview 5 software (Deltasoft, Meylan, France). Results The patient characteristics, preoperative treatments, left ventricular ejection fraction, and HbA1c levels are shown in Table 1. Fifty-two human right atrial trabeculae and 16 right atrial appendages were studied. There were no significant differences between the groups in terms of the trabecular length at the apex of the length-active isometric tension 513

Lemoine et al. Table 2 Control values of the primary mechanical parameters of human right atrial trabeculae. Data are mean (SD). L max, maximal length at the apex of the length-active force curve; CSA, cross-sectional area; AF, acting isometric force normalized per cross-sectional area; RF/TF, ratio of resting force to total force; Cal, calphostin C; Des, desflurane; Diab, diabetic; 5-HD, 5-hydroxydecanoate; PMA, phorbol 12-myristate 13-acetate Experimental groups L max (mm) CSA (mm 2 ) AF (mn mm 22 ) RF/TF Diab-Control (n¼10) 6.2 (2.5) 0.60 (0.20) 27 (12) 0.38 (0.06) Diab-Des (n¼6) 6.4 (2.4) 0.51 (0.23) 24 (12) 0.28 (0.06) Diab-Des+Cal (n¼6) 7.3 (1.4) 0.50 (0.28) 25 (16) 0.28 (0.09) Diab-Des+5-HD (n¼6) 6.8 (1.6) 0.48 (0.20) 27 (16) 0.30 (0.09) Diab-Cal (n¼6) 6.7 (1.0) 0.45 (0.16) 21 (4) 0.37 (0.09) Diab-5-HD (n¼6) 6.0 (1.3) 0.44 (0.07) 31 (11) 0.27 (0.08) Diab-PMA (n¼6) 7.1 (1.9) 0.55 (0.20) 29 (10) 0.23 (0.04) Diab-Diazoxide (n¼6) 5.6 (1.4) 0.41 (0.23) 34 (17) 0.29 (0.06) curve, cross-sectional area, ratio of resting-to-total force (Table 2). Effects of 5-HD and calphostin C on desflurane-induced post-conditioning Desflurane significantly increased the FoC 60 of the desflurane-treated group when compared with that of the control group [Diab-Des: 79 (3)% of baseline vs Diab-Control: 49 (7)% of baseline; P,0.0001]. The desflurane-induced enhanced recovery of the FoC 60 was abolished in the presence of calphostin C [Diab-Des+Cal: 54 (3)% of baseline vs Diab-Des: 79 (3)% of baseline; P,0.0001] and 5-HD [Diab-Des+5-HD: 57 (6)% of baseline vs Diab-Des: 79 (3)% of baseline; P,0.0001] (Fig. 2). When compared with the Diab-Control group, the groups treated with calphostin C [Diab-Cal: 59 (3)% of baseline; P¼0.18] and 5-HD [Diab-5-HD: 54 (6)% of baseline; P¼0.11] did not have significantly different FoC 60 values (Fig. 2). Post-conditioning effects of PMA and diazoxide When compared with the Diab-Control group, the groups treated with PMA and diazoxide showed enhanced recovery of the FoC 60 [Diab-PMA: 91 (4)% of baseline and Diab-Diazoxide: 87 (9)% of baseline; P,0.0001 vs Diab-Control] (Fig. 2). Phosphorylation of Akt and glycogen synthase kinase 3b in the diabetic and non-diabetic human myocardium exposed to desflurane As illustrated in Figure 3, desflurane (6%) administered in the first 5 min of reoxygenation significantly increased the ratio of phospho-akt (Ser473) to Akt in the non-diabetic group (+89% in Des vs Control; P,0.0001) and in the diabetic group (+84% in Diab-Des vs Diab-Control; P,0.0001). Similarly, desflurane (6%) administered in the first 5 min of reoxygenation significantly increased the ratio of phosphorylated GSK-3b (Ser9) to GSK-3b in the non-diabetic group (+110% in Des vs Control, P,0.0001) and in the diabetic group (+102% in Diab-Des vs Diab-Control, P,0.0001) (Fig. 3). FoC at the end of reoxygenation (% of baseline) 100 90 80 70 60 50 40 30 20 10 0 Discussion Diab-Control Diab-Des Diab-Des+Cal Diab-Cal Diab-Des+5-HD Diab-5-HD Diab-PMA Diab-Diazoxide Fig 2 Recovery of the FoC of isolated human right atrial trabeculae at the end of the 60 min reoxygenation period after the 30 min hypoxic challenge in groups exposed to desflurane (6%) (Diab-Des) alone or in the presence of calphostin C (Diab-Des+ Cal) and 5-HD (Diab-Des+5-HD). Effect of the administration of PMA (Diab-PMA) and diazoxide (Diab-Diazoxide) at the beginning of reoxygenation on the time course of the FoC (expressed as a percentage of baseline) of the isolated human right atrial trabeculae during a 30 min hypoxic challenge followed by a 60 min reoxygenation period. Data are mean (SD). P,0.001 vs the Diab-Control, Diab-Des+Cal, Diab-Des+5-HD, Diab-Cal, and Diab-5-HD groups. Cal, calphostin C; Des, desflurane; Diab, diabetic; 5-HD, 5-hydroxydecanoate; PMA, phorbol 12-myristate 13-acetate. The present study showed that the in vitro desflurane-induced post-conditioning of human myocardium obtained from patients with type 2 diabetes involved the activation of PKC, the opening of mitok ATP channels, and the phosphorylation of Akt and GSK-3b. Limited data are available on the mechanisms involved in the cardioprotection of diabetic myocardium, and the results 514

Desflurane-induced post-conditioning A Relative density phospho-akt (Ser473) / Akt B Relative density phospho-gsk-3β (Ser9) / GSK-3β ratio 1.4 1.2 1 0.8 0.6 0.4 0.2 0 1.4 1.2 1 0.8 0.6 0.4 0.2 0 Control Control Control Diab-Control Diab- Control Control Des Diab- Control Diab-Control Des Des Diab-Des of the experimental studies remain contradictory. 12 To our knowledge, only two studies have examined the role of PKC during ischaemia reperfusion in diabetic rat hearts. 13 14 It has been shown using isolated diabetic rat hearts that 4 week streptozotocin-induced diabetes increases the resistance to ischaemia reperfusion injury and that this effect was eliminated by chelerythrine, a non-specific PKC inhibitor. 13 Moreover, translocation of PKC-1 from the cytosolic fraction to the mitochondrial fraction was persistent, whereas it was transiently in the non-diabetic heart. 14 The present study shows that inhibition of PKC abolished the desflurane-induced post-conditioning in isolated human Diab-Des Des Diab-Des Phospho-GSK-3β (Ser9) GSK-3β Diab-Des Phospho-Akt (Ser473) Akt Fig 3 Western blot analysis showing levels of total Akt, phosphorylated Akt (A), total glycogen synthase kinase 3b (GSK-3b total), and phosphorylated GSK-3b (Ser9) (phospho-gsk-3b) (B) after 5 min of reoxygenation alone (Control) or in the presence of 5 min of 6% desflurane (Des). Equal loading was confirmed by western blot with an anti-b-tubulin antibody. The relative total Akt, phospho-akt, total GSK-3b, and phospho-gsk-3b protein levels were calculated by averaging the results obtained from four independent experiments. Example blot shown is a representative image of a single blot. Data are mean (SD). P,0.0001 vs the Control and Diab-Control groups. Des, desflurane; Diab, diabetic; GSK-3b, glycogen synthase kinase-3b. myocardium obtained from patients with type 2 diabetes. In isolated non-contractile human diabetic myocardium, Hassouna and colleagues 7 showed that PKC activation by PMA before the prolonged ischaemia provided cardioprotective effects. In the present study, the activation of PKC during the first few minutes of reoxygenation was shown to trigger post-conditioning. Thus, preconditioning and post-conditioning may be effective in the diabetic myocardium and may share a common signalling pathway, including the activation of PKC. 15 However, further studies should be performed to identify the role of the different PKC isoforms. At the present time, the role and function of mitok ATP channels in the diabetic myocardium remain unclear. Our results showed that, in the myocardium from patients with type 2 diabetes, desflurane-induced myocardial post-conditioning was abolished in the presence of 5-HD. It has been shown that 5-HD inhibits myocardial post-conditioning in the non-diabetic myocardium; 10 11 however, there were no data for the diabetic myocardium. Although protective signalling pathways are shared by preconditioning and postconditioning in the non-diabetic myocardium, 15 this result must be examined in the diabetic myocardium. Most experimental studies have shown that diazoxide is not able to trigger preconditioning in the diabetic myocardium or during acute hyperglycaemia. 16 18 The present results show that diazoxide that is administered at the beginning of reoxygenation significantly increased the recovery of the FoC 60. Several reasons may explain this discrepancy. First, diazoxide was administered as a preconditioning stimulus 14 16 in previous studies, whereas in the present study, it was administered as a post-conditioning stimulus. Secondly, in Ghosh and colleagues study, 16 all patients with type 2 diabetes were treated with K ATP channel antagonists, whereas in our study, only a few patients with type 2 diabetes were treated with K ATP channel antagonists (Table 1). Thirdly, HbA1c measurements (Table 1) suggested that diabetic patients had well-controlled glycaemia in the present study. Fourthly, the species and experimental model differences may be of importance. Isolated human myocardium obtained from patients with diabetes should definitely not be compared with the experimental diabetes obtained previously using animal models of type 2 diabetes, such as young dogs with diabetes induced by alloxan and streptozotocin 18 and the Zucker obese rat. 17 Additionally, it has been suggested that a higher threshold may be required to activate the cardioprotective signalling pathways in the diabetic myocardium. 919 Finally, it has been argued that the effects of diazoxide and 5-HD in inducing and blocking the preconditioning, respectively, may reflect mitochondrial bioenergetic function actions that do not necessarily involve mitok ATP 20 22 channels. Western blot experiments showed a significant increase in the phosphorylation of Akt and GSK-3b after a brief exposure to desflurane (6%) during early reoxygenation in the nondiabetic and diabetic myocardium. This result suggests that the Akt/GSK-3b protective signalling pathway can be 515

Lemoine et al. recruited by desflurane in both the diabetic and non-diabetic myocardium. It has been shown that the inactivation (phosphorylation) of GSK-3b plays a key role in myocardial postconditioning, both ischaemic and pharmacological, through the inhibition of the mitochondrial permeability the inhibition of transition pore opening. 23 25 In addition, pharmacological inhibition of GSK-3b at the time of reperfusion reduces the myocardial infarct size in streptozotocin-induced diabetic rats. 25 Tsang and colleagues reported that three cycles of ischaemic preconditioning are necessary to reduce the myocardial infarct size in diabetic rats and that the effect was commensurate with significant Akt phosphorylation. They suggested that repeated ischaemic preconditioning stimuli are needed to achieve the threshold for cardioprotection and that a critical level of Akt phosphorylation is necessary to protect the diabetic myocardium. 26 Taken together, these results suggest that the Reperfusion Injury Salvage Kinase signalling pathway can be activated by desflurane-induced post-conditioning in human myocardium from patients with type 2 diabetes. The current findings should be interpreted within the constraints of potential limitations. First, the effects of the anaesthetic drugs; physiological conditions, including other diseases; or medical treatments received by the patients before obtaining the atrial appendages cannot be ruled out. However, in clinical practice, patients for whom cardioprotective strategies may be proposed are likely to have similar conditions, treatments, or both, and it is of interest to examine the effects of cardioprotective strategies under conditions closely resembling clinical situations. Secondly, although our experimental groups were comparable demographically (Table 1), age has been shown to impair the protective effects of ischaemic post-conditioning in senescent mouse hearts. 27 29 It is important to note that our study included a control group that was equally affected by these potentially modifying factors. Thirdly, in 12 patients, i.v. insulin has been used to control blood glucose concentration before atrial appendage (two in the Diab-Control, Diab-Des+5-HD, Diab-Cal, Diab-5-HD, Diab-PMA groups; one in Diab-Des and Diab-Diazoxide groups). Insulin has been shown to activate cell survival pathways, including the PI3-kinase/Akt-dependent pathway. 30 Fourthly, the treatments used to equilibrate glycaemia, such as sulfonylureas which increase insulin secretion by blocking the K ATP channel in the pancreatic b-cell membrane, can affect myocardial K ATP channel. 31 We have included a number of patients with type 2 diabetes treated with antagonists of K ATP channels (Table 1, preoperative treatments). Fifthly, our experiments were performed under moderate hypothermia (348C) to ensure the stability of the trabeculae over time. It has been shown that hypothermia may decrease the sensitivity of mitok ATP channels. 32 However, during surgical procedures, moderate hypothermia may occur. Then, the specificity of the activators and inhibitors used in the Administration of desflurane in early reoxygenation period Activation Phosphorylation Phosphorylation PMA + PKC Opening Calphostin C Akt Diazoxide 5 HD + mitok ATP channels GSK-3β Post-conditioning of human myocardium with type 2 diabetes Fig 4 Proposed schematic representation of the signalling pathways leading to desflurane induced post-conditioning of the human myocardium with type 2 diabetes, in vitro. Brief administration of desflurane in the early reoxygenation period induced PKC activation, the opening of mitok ATP channels, phosphorylation of Akt and GSK 3b, which mediate the cardioprotective effect. Moreover, in early reoxygenation, activation of PKC (via PMA administration) and the opening of mitok ATP channels (via diazoxide administration) induced myocardial postconditioning. GSK-3b, glycogen synthase kinase-3b; mitok ATP, mitochondrial adenosine triphosphate-sensitive potassium channel; PKC, protein kinase C; PMA, phorbol 12-myristate 13-acetate; 5-HD, 5-hydroxydecanoate. 516

Desflurane-induced post-conditioning present study should be considered. Diazoxide and 5-HD have also been shown to modify the mitochondrial respiratory chain complex. 20 22 Finally, we studied isolated contracting human atrial trabeculae but not the myocardial ventricular infarct size. Important functional and structural differences exist between the atrial and ventricular myocardium. However, for ethical reasons, it is not possible to obtain systematic ventricular biopsies from patients undergoing coronary artery bypass graft surgery or aortic valve replacement. Additionally, as described for the myocardial preconditioning, the beneficial effects of post-conditioning have also been described for reperfusion-induced arrhythmias 33 and myocardial conctractility. 34 Conclusion Using isolated human myocardium from patients with type 2 diabetes, we showed that the activation of PKC, the opening of mitok ATP channels and changes in mitochondrial bioenergetic function, the phosphorylation of Akt, and the phosphorylation of GSK-3b were involved in desflurane-induced post-conditioning (Fig. 4). Moreover, activation of PKC and treatment with diazoxide at the beginning of reoxygenation were cardioprotective in the diabetic myocardium (Fig. 4). These results raise the possibility that diabetic patients may benefit from pharmacological post-conditioning. Conflict of interest None declared. Funding This research was supported by a grant from the Société Française d Endocrinologie LIFESCAN (to S.L.) and by institutional, departmental, or both sources. References 1 Woods SE, Eppley C, Engel A. The influence of diabetes mellitus in patients undergoing coronary artery bypass graft surgery: a prospective cohort study. Am Surg 2008; 74: 839 44 2 Haffner SM, Lehto S, Rönnemaa T, Pyörälä K, Laakso M. Mortality from coronary heart disease in subjects with type 2 diabetes and in non diabetic subjects with and without prior myocardial infarction. N Engl J Med 1998; 339: 229 34 3 Canbaz S, Duran E. Ischaemia reperfusion studies and diabetes mellitus. Br J Anaesth 2003; 91: 158 9 4 Ouattara A, Lecomte P, Le Manach Y, et al. Poor intraoperative blood glucose control is associated with a worsened hospital outcome after cardiac surgery in diabetic patients. Anesthesiology 2005; 103: 687 94 5 Kersten JR, Toller WG, Gross ER, Pagel PS, Warltier DC. Diabetes abolishes ischemic preconditioning: role of glucose, insulin, and osmolality. Am J Physiol Heart Circ Physiol 2000; 278: H1218 24 6 Matsumoto S, Cho S, Tosaka S, et al. Pharmacological preconditioning in type 2 diabetic rat hearts: the roles of mitochondrial ATP-sensitive potassium channels and the phosphatidylinositol 3-kinase-Akt pathway. Cardiovasc Drugs Ther 2009; 23: 263 70 7 Hassouna A, Loubani M, Matata BM, Fowler A, Standen NB, Galinanes M. Mitochondrial dysfunction as the cause of the failure to precondition the diabetic human myocardium. Cardiovasc Res 2006; 69: 450 8 8 Huhn R, Heinen A, Weber NC, Hollmann MW, Schlack W, Preckel B. Hyperglycaemia blocks sevoflurane-induced postconditioning in the rat heart in vivo: cardioprotection can be restored by blocking the mitochondrial permeability transition pore. Br J Anaesth 2008; 100: 465 71 9 Lemoine S, Durand C, Zhu L, et al. Desflurane-induced postconditioning of diabetic human right atrial myocardium in vitro. Diabetes Metab 2010; 36: 21 8. 10 Lemoine S, Beauchef G, Zhu L, et al. Signaling pathways involved in desflurane-induced postconditioning in human atrial myocardium in vitro. Anesthesiology 2008; 109: 1036 44 11 Lemoine S, Puddu PE, Durand C, et al. Signaling pathways involved in postconditioning-induced cardioprotection of human myocardium, in vitro. Exp Biol Med (Maywood) 2010; 235: 768 76 12 Monteiro P, Gonçalves L, Providência LA. Diabetes and cardiovascular disease: the road to cardioprotection. Heart 2005; 91: 1621 5 13 Moon CH, Jung YS, Lee SH, Baik EJ. Protein kinase C inhibitors abolish the increased resistance of diabetic rat heart to ischemia reperfusion injury. Jpn J Physiol 1999; 49: 409 15 14 Ooie T, Takahashi N, Nawata T, et al. Ischemia-induced translocation of protein kinase C-epsilon mediates cardioprotection in the streptozotocin-induced diabetic rat. Circ J 2003; 67: 955 61 15 Hausenloy DJ, Yellon DM. Preconditioning and postconditioning: united at reperfusion. Pharmacol Ther 2007; 116: 173 91. 16 Ghosh S, Standen NB, Galinanes M. Failure to precondition pathological human myocardium. J Am Coll Cardiol 2001; 37: 711 8. 17 Katakam PV, Jordan JE, Snipes JA, Tulbert CD, Miller AW, Busija DW. Myocardial preconditioning against ischemia reperfusion injury is abolished in Zucker obese rats with insulin resistance. Am J Physiol Regul Integr Comp Physiol 2007; 292: R920 6 18 Kersten JR, Montgomery MW, Ghassemi T, et al. Diabetes and hyperglycemia impair activation of mitochondrial K(ATP) channels. Am J Physiol Heart Circ Physiol 2001; 280: H1744 50 19 Tanaka K, Kehl F, Gu W, et al. Isoflurane-induced preconditioning is attenuated by diabetes. Am J Physiol Heart Circ Physiol 2002; 282: H2018 23 20 Hanley PJ, Mickel M, Löffler M, Brandt U, Daut J. K ATP channelindependent targets of diazoxide and 5-hydroxydecanoate in the heart. J Physiol 2002; 542: 735 41 21 Lim KH, Javadov SA, Das M, Clarke SJ, Suleiman MS, Halestrap AP. The effects of ischaemic preconditioning, diazoxide and 5- hydroxydecanoate on rat heart mitochondrial volume and respiration. J Physiol 2002; 545: 961 74 22 Schäfer G, Wegener C, Portenhauser R, Bojanovski D. Diazoxide, an inhibitor of succinate oxidation. Biochem Pharmacol 1969; 18: 2678 81 23 Tong H, Imahashi K, Steenbergen C, Murphy E. Phosphorylation of glycogen synthase kinase-3b during preconditioning through a phosphatidylinositol-3-kinase dependent pathway is cardioprotective. Circ Res 2002; 90: 377 9 24 Feng J, Lucchinetti E, Ahuja P, Pasch T, Perriard JC, Zaugg M. Isoflurane postconditioning prevents opening of the mitochondrial permeability transition pore through inhibition of glycogen synthase kinase 3b. Anesthesiology 2005; 103: 987 95 25 Gross ER, Hsu AK, Gross GJ. Diabetes abolishes morphine-induced cardioprotection via multiple pathways upstream of glycogen synthase kinase-3b. Diabetes 2007; 56: 127 36 517

Lemoine et al. 26 Tsang A, Hausenloy DJ, Mocanu MM, Carr RD, Yellon DM. Preconditioning the diabetic heart: the importance of Akt phosphorylation. Diabetes 2005; 54: 2360 4 27 Lauzier B, Delemasure S, Debin R, et al. Beneficial effects of myocardial postconditioning are associated with reduced oxidative stress in a senescent mouse model. Transplantation 2008; 85: 1802 8 28 Przyklenk K, Maynard M, Darling CE, Whittaker P. Aging mouse hearts are refractory to infarct size reduction with postconditioning. J Am Coll Cardiol 2008; 51: 1393 8 29 Boengler K, Buechert A, Heinen Y, et al. Cardioprotection by ischemic postconditioning is lost in aged and STAT3-deficient mice. Circ Res 2008; 102: 131 5 30 Gao F, Gao E, Yue TL, et al. Nitric oxide mediates the antiapoptotic effect of insulin in myocardial ischemia reperfusion: the roles of PI3-kinase, Akt, and endothelial nitric oxide synthase phosphorylation. Circulation 2002; 105: 1497 502 31 Gribble FM, Ashcroft FM. Sulfonylurea sensitivity of adenosine triphosphate-sensitive potassium channels from beta cells and extrapancreatic tissues. Metabolism 2000; 49: 3 6 32 Saito W, Noguchi K, Okazaki K, et al. Temperature-sensitive effects of potassium channel openers on isolated guinea pig myocardium and aorta. J Cardiovasc Pharmacol 1998; 31: 327 9 33 Dow J, Bhandari A, Kloner RA. Ischemic postconditioning s benefit on reperfusion ventricular arrhythmias is maintained in the senescent heart. J Cardiovasc Pharmacol Ther 2008; 13: 141 8 34 Sivaraman V, Mudalgiri NR, Di Salvo C, et al. Postconditioning protects human atrial muscle through the activation of the RISK pathway. Basic Res Cardiol 2007; 102: 453 9 518