Impairment of Diazoxide-Induced Formation of Reactive Oxygen Species and Loss of Cardioprotection in Connexin 43 Deficient Mice
Protection by ischemic preconditioning is lost in cardiomyocytes and hearts of heterozygous connexin 43 deficient (Cx43+/−) mice. Because connexin 43 (Cx43) is localized in cardiomyocyte mitochondria and mitochondrial Cx43 content is increased with ischemic preconditioning, we now tried to identify a functional defect at the level of the mitochondria in Cx43+/− mice by use of diazoxide and menadione. Diazoxide stimulates the mitochondrial formation of reactive oxygen species (ROS) and menadione generates superoxide at multiple intracellular sites; both substances elicit cardioprotection through increased ROS formation. ROS formation in response to the potassium ionophore valinomycin was also measured for comparison. Menadione (2 μmol/L) and valinomycin (10 nmol/L) induced similar ROS formation in wild-type (WT) and Cx43+/− cardiomyocytes. In contrast, diazoxide (200 μmol/L) increased ROS formation by 43±10% versus vehicle in WT, but only by 18±4% in Cx43+/− cardiomyoctes (P<0.05). Two hour–simulated ischemia and oxygenated, hypo-osmolar reperfusion reduced viability as compared with normoxia (WT: 7±1% versus 39±2%, Cx43+/−: 8±1% versus 40±3%, P<0.01). Although menadione protected WT and Cx43+/− cardiomyocytes, diazoxide increased viability (17±2%, P<0.01) in WT, but not in Cx43+/− (9±1%). Menadione (37 μg/kg i.v.) before 30 minutes coronary occlusion and 2 hour reperfusion reduced infarct size in WT and Cx43+/− mice (24±4% versus 24±5%). In contrast, diazoxide (5 mg/kg i.v.) reduced infarct size in WT (35±4% versus 55±3% of area at risk, P<0.01), but not in Cx43+/− mice (56±2% versus 54±3%). Cardiomyocytes of Cx43+/− mice have a specific functional deficit in ROS formation in response to diazoxide and accordingly less protection.
Hearts from heterozygous connexin 43 (Cx43) deficient (Cx43+/−) mice cannot be protected by ischemic preconditioning.1,2 Isolated cardiomyocytes are also not protected by ischemic preconditioning, and loss of protection is therefore independent of gap junctions.3 Most of the signaling pathways of ischemic preconditioning converge at the mitochondria4–6 and mitochondrial free radical formation mediates signal transduction through posttranslational modification of redox-sensitive proteins.7,8 We have recently demonstrated the presence of Cx43 in the mitochondria and the increase of its mitochondrial content with ischemic preconditioning.9
Now we attempted to define the functional role of Cx43 in the mitochondria by use of diazoxide, which stimulates the formation of reactive oxygen species (ROS). Diazoxide was compared with menadione, which generates superoxide at multiple intracellular sites caused by redox cycling reactions catalyzed by several flavoenzymes.10 The cardioprotective effects elicited by diazoxide and menadione have been attributed to ROS formation.7,8,11,12 ROS formation by diazoxide was also compared with that by valinomycin, which is a potassium ionophore13 and assumed to induce ROS formation secondary to potassium influx into mitochondria and matrix swelling.14
Materials and Methods
All experiments were performed according to the guidelines of the American Physiological Society and approved by the local bioethical committee on animal experimentation.
Total cardiac Cx43 content was determined as described,1 and the localization of Cx43 at the mitochondria was measured as the ratio of Cx43 to the adenine nucleotide transporter in isolated mitochondria of 6 WT and Cx43+/− each.9
In Vitro Experiments
Cardiomyocytes were isolated from Cx43+/− and WT mice, as previously described.3 Simulated ischemia was induced by pelleting cardiomyocytes in hypoxic solution (pH=6.5), sealed with a layer of mineral oil.3 Cardiomyocytes were incubated with diazoxide (200 or 500 μmol/L) or menadione (2 μmol/L) (Sigma Aldrich, Germany) for 30 minutes followed by 15 minutes wash-out before simulated ischemia. At 0, 60, and 120 minutes of simulated ischemia, aliquots from each group were subjected to simulated reperfusion, ie, resuspended in oxygenated, iso-osmolar (at 0 minutes: 310 mOsm/L, pH=7.4) or hypo-osmolar solution (at 60 and 120 minutes: 250 mOsm/L, pH=7.4). Images were taken after 3 to 5 minutes, comprising ≥200 cells per sample for offline quantification of cell viability (trypan blue exclusion).3
For measurement of ROS formation, cardiomyocytes from Cx43+/− and WT mice were incubated with diazoxide (200 μmol/L), menadione (2 μmol/L), valinomycin (10nmol/L; a concentration which did not cause significant cellular contraction but ROS formation), or vehicle (0.1% DMSO, Sigma Aldrich) and 2 μmol/L reduced MitoTracker Red (MTR, Invitrogen, Germany) for 30 minutes. MTR, when oxidized by H2O2, is fluorescent and covalently bound to mitochondrial structures.15 Cardiomyocytes were then washed twice and transferred to an inverted microscope (Axiovert 100 TV, Zeiss) equipped with a fluorescence-sensitive CCD camera (T.I.L.L. Photonics). ROS formation was quantified from the mean fluorescence of 70 to 100 randomly selected rod-shaped cells per sample. The investigator was blinded for the genotype in all experiments.
In Vivo Experiments
Myocardial infarction was induced in WT or heterozygous Cx43+/− mice, as previously described.1 The left anterior descending coronary artery was occluded for 30 minutes, followed by 120 minutes of reperfusion. Diazoxide (5 mg/kg), menadione(37 μg/kg) or saline was injected into the jugular vein 30 minutes before ischemia. After removal of the heart, infarct size and area at risk were measured by TTC and Evans blue staining.2
Data are expressed as mean±SEM. For comparison of ROS formation and cell viability between untreated and treated groups, two-way ANOVA was used. For the analysis of in vivo infarct size data, two-way ANOVA was used. Fisher’s least-significance test was performed when significant overall effects were detected. P<0.05 was considered significant.
Total cardiac Cx43 content in Cx43+/− mice was 56±9% of that in WT;1 the Cx43/adenine nucleotide transporter ratio in Cx43+/− isolated mitochondria was 21±7% of that in WT, indicating that the mitochondrial Cx43-deficiency was at least as pronounced as total cardiac Cx43 deficiency.
Mitochondrial ROS Formation in Response to Diazoxide, Menadione, and Valinomycin in Cx43 Deficient Cardiomyocytes
Diazoxide (200 μmol/L) increased ROS formation by 43±10% versus vehicle control in WT, but only by 18±4% in Cx43+/− cardiomyocytes; increase of the diazoxide concentration to 500 μmol/L did not further increase ROS formation in Cx43+/− (13±3%, n=4 mice). Menadione induced similar ROS formation in WT and in Cx43+/− cardiomyocytes, by 28±9% and 34±15%, respectively (Figure 1). Also, valinomycin induced similar ROS formation in WT and Cx43+/− cardiomyocytes by 46±11% and 43±12%, respectively (Figure 1).
No Protection by Diazoxide in Isolated Cardiomyocytes and in Hearts In Situ From Cx43-Deficient Mice
In isolated cardiomyocytes, viability remained relatively stable during normoxia: WT (51±1% at baseline versus 40±2% at 60 minutes and 39±2% at 120 minutes) and Cx43+/− (52±1% at baseline versus 40±1% at 60 minutes and 40±3% at 120 minutes). Simulated ischemia/reperfusion reduced viability (at 60 minutes: WT: 11±1%, Cx43+/−: 12±2%; at 120 minutes: WT: 7±1%, Cx43+/−: 8±1%, all P<0.01 versus baseline and normoxia). Diazoxide preserved viability in cardiomyocytes of WT (at 60 minutes: 22±2%; at 120 minutes: 17±2%, P<0.01 versus simulated ischemia/reperfusion), but not of Cx43+/− mice (at 60 minutes: 15±3%; at 120 minutes: 9±1%) (Figure 2). With menadione, viability was increased both in WT (at 60 minutes: 23±4% versus 16±1%, at 120 minutes: 19±1% versus 14±2%) and in Cx43+/− cardiomyocytes (at 60 minutes: 23±2% versus 17±1%, at 120 minutes: 20±2% versus 15±1%, all P<0.05) (Figure 3).
Infarct size following ischemia/reperfusion (I/R) was not different between WT and Cx43+/− (Figure 4). Infarct size was reduced with diazoxide in WT, but not in Cx43+/− mice. In contrast, menadione reduced infarct size both in WT and in Cx43+/− mice (Figure 4).
Apart from the role of Cx43 in the intercellular communication and propagation of ischemic cell death,16,17 Cx43 is important for protection by ischemic preconditioning.1–3 In the rat heart, a decrease of Cx43 at gap junctions during myocardial ischemia was observed, whereas total Cx43 expression was unchanged, suggesting alternative subcellular localizations.18 Indeed, during oxidative stress translocation of Cx43 to the mitochondria has been reported in endothelial cells.19 Recently, we have demonstrated that Cx43 is present in the mitochondria of cardiomyocytes and that its mitochondrial content is increased following ischemic preconditioning.9 In the present study, diazoxide before ischemia/reperfusion reduced infarct size, confirming previous reports in other species.6 In line with our previous findings on ischemic preconditioning,3 diazoxide was also effective in isolated WT cardiomyocytes and thus in the absence of gap junctions. Cx43 deficiency abolished the protection afforded by ischemic preconditioning and diazoxide in isolated and in situ cardiomyocytes, suggesting that Cx43 is part of the preconditioning signal cascade triggered at the level or upstream of mitochondria.
A functional role of Cx43 in mitochondria is supported by our finding of a reduced ROS signal in response to diazoxide in Cx43+/− cardiomyocytes. In contrast, the potassium ionophore valinomycin and menadione induced similar ROS formation in WT and Cx43+/− cardiomyocytes, indicating a specific defect in the diazoxide response with Cx43 deficiency. Also, the signaling cascade downstream of mitochondria was not affected by Cx43 deficiency, as ROS formation induced by menadione still resulted in improved viability following simulated ischemia/reoxygenation in isolated cardiomyocytes and reduced infarct size in hearts from both WT and Cx43+/− mice.
In conclusion, cardiomyocytes of Cx43+/− mice have a specific functional deficit in ROS formation in response to diazoxide and accordingly less protection.
This study was funded from the institutional budget. Y.L. received a scholarship from the Deutscher Akademischer Austauschdienst (DAAD) and the China Scholarship Council. K.B. was awarded the Servier Fellowship of the European Section of the International Society for Heart Research for this project.
↵*Both authors contributed equally to this work.
Original received May 31, 2005; revision received July 29, 2005; accepted August 1, 2005.
Schulz R, Cohen MV, Behrends M, Downey JM, Heusch G. Signal transduction of ischemic preconditioning. Cardiovasc Res. 2001; 52: 181–198.
O’Rourke B. Evidence for mitochondrial K+ channels and their role in cardioprotection. Circ Res. 2004; 94: 420–432.
Pain T, Yang X-M, Critz SD, Yue Y, Nakano A, Liu GS, Heusch G, Cohen MV, Downey JM. Opening of mitochondrial KATP channels triggers the preconditioned state by generating free radicals. Circ Res. 2000; 87: 460–466.
Forbes RA, Steenbergen C, Murphy E. Diazoxide-induced cardioprotection requires signaling through a redox-sensitive mechanism. Circ Res. 2001; 88: 802–809.
Boengler K, Dodoni G, Ruiz-Meana M, Cabestrero A, Rodriguez-Sinovas A, Garcia-Dorado D, Gres P, Di Lisa F, Heusch G, Schulz R. Connexin 43 in cardiomyocyte mitochondria and its increase by ischemic preconditioning. Cardiovasc Res. 2005; 67: 234–244.
Gutierrez PL. The role of NAD(P)H oxireductase (DT-diaphorase) in the bioactivation of quinone-containing antitumor agents: a review. Free Radic Biol Med. 2005; 29: 263–275.
Yue Y, Krenz M, Cohen MV, Downey JM, Critz SD. Menadione mimics the infarct-limiting effect of preconditioning in isolated rat hearts. Am J Physiol Heart Circ Physiol. 2001; 281: H590–H595.
Yue Y, Qin Q, Cohen MV, Downey JM, Critz SD. The relative order of mK(ATP) channels, free radicals and p38 MAPK in preconditioning’s protective pathway in rat heart. Cardiovasc Res. 2005; 55: 681–689.
Bernardi P. Mitochondrial transport of cations: channels, exchangers, and permeability transition. Physiol Rev. 1999; 79: 1127–1155.
Oldenburg O, Qin Q, Sharma AR, Cohen MV, Downey JM, Benoit JN. Acetylcholine leads to free radical production dependent on KATP channels, Gi proteins, phosphatidylinositol 3-kinase and tyrosine kinase. Cardiovasc Res. 2002; 55: 544–552.
Oldenburg O, Qin Q, Krieg T, Yang X-M, Philipp S, Critz SD, Cohen MV, Downey JM. Bradykinin induces mitochondrial ROS generation via NO, cGMP, PKG, and mitoKATP channel opening and leads to cardioprotection. Am J Physiol Heart Circ Physiol. 2004; 286: H468–H476.
Garcia-Dorado D, Inserte J, Ruiz-Meana M, Gonzalez MA, Solares J, Julia M, Barrabés JA, Soler-Soler J. Gap junction uncoupler heptanol prevents cell-to-cell progression of hypercontracture and limits necrosis during myocardial reperfusion. Circulation. 1997; 96: 3579–3586.
Schulz R, Heusch G. Connexin 43 and ischemic preconditioning. Cardiovasc Res. 2004; 62: 335–344.
Beardslee MA, Lerner DL, Tadros PN, Laing JG, Beyer EC, Yamada KA, Kleber AG, Schuessler RB, Saffitz JE. Dephosphorylation and intracellular redistribution of ventricular connexin43 during electrical uncoupling induced by ischemia. Circ Res. 2000; 87: 656–662.
Li H, Brodsky S, Kumari S, Valiunas V, Brink P, Kaide J-I, Nasjletti A, Goligorsky MS. Paradoxical overexpression and translocation of connexin 43 in homocysteine-treated endothelial cells. Am J Physiol Heart Circ Physiol. 2001; 282: H2124–H2133.