Ceramide-induced preconditioning involves reactive oxygen species

Life Sciences 78 (2006) 1702 – 1706 www.elsevier.com/locate/lifescie

Ceramide-induced preconditioning involves reactive oxygen species Sandrine Lecour *, Peter Owira, Lionel H. Opie Hatter Institute for Cardiology Research, Department of Medicine, University of Cape Town Medical School, Cape Town, South Africa Received 28 January 2005; accepted 9 August 2005

Abstract Introduction: Ceramide induces programmed cell death and it is thought to contribute to cardiac ischemia/reperfusion (I/R) injury. In contrast, we have demonstrated that administration of low doses of ceramide engenders cardiac preconditioning (PC). Ceramide is known to generate reactive oxygen species (ROS) in cells. Since mechanisms triggering the ceramide-induced cardioprotection remain unknown, we investigated the role of ROS in the genesis of this protective mechanism. Methods: Using an isolated Langendorff-perfused rat heart model, four groups (n  6 in each group) were considered: Control hearts underwent 30 min index regional ischemia and 120 min of reperfusion. In the ceramide group, hearts were preconditioned with c2-ceramide 1 AM for 7 min followed by 10 min washout prior to the I/R insult. In additional groups, MPG (1 mM), a synthetic antioxidant was given for 15 min alone or bracketing the ceramide perfusion. In each group, infarct size was determined at the end of the reperfusion period and superoxide dismutases (CuZnSOD and MnSOD) and catalase activities were evaluated. Results: Ceramide preconditioning reduced the infarct/area at risk (I/AAR) ratio (8.3 T 1.1% for ceramide vs. 36.4 T 1.2% for control, p < 0.001). Perfusion with MPG abolished the preconditioning effect of ceramide (I/AAR ratio = 36.7 T 4.9%). Ceramide was also associated with a 29% and 38% increase in catalase and CuZnSOD activities, respectively, compared with control group. Conclusion: Production of reactive oxygen species following ceramide preconditioning of the ischemic – reperfused heart appears to play a role in the cardioprotective effect of ceramide. D 2005 Elsevier Inc. All rights reserved. Keywords: Ceramide; Ischemia; Reactive oxygen species; Preconditioning; Reperfusion

Introduction Sphingolipids are a diverse family of phospholipids based on a long chain sphingoid backbone, generally sphingosine. Over the last few years, many studies provided evidence implicating the sphingolipid ceramide as a second messenger in multiple pathways initiated on binding of TNFa to its TNFR1/ p55 receptor (Adam et al., 1996; Schutze et al., 1994; Wiegmann et al., 1994). This hydrolyses sphingomyelin to form ceramide after activation of the sphingomyelinase. Ceramide is well known as a trigger for apoptosis in many different cell systems as well as an inhibitor of cell growth (Lecour et al., 2003). * Corresponding author. Hatter Institute for Cardiology Research, Cape Heart Centre and Medical Research Council Interuniversity Group, University of Cape Town Medical School, Observatory 7925, Cape Town, South Africa. Tel.: +27 21 406 63 58; fax: +27 21 447 87 89. E-mail address: [email protected] (S. Lecour). 0024-3205/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.lfs.2005.08.013

Preconditioning is the phenomenon whereby a transient non-lethal ischemic insult enables the cells to become more resistant to subsequent ischemic injury (Murry et al., 1986). The biological programs involved still remain unknown but the endogenous immune cardiac system has been implicated. Thus, we have previously demonstrated that exogenous low doses of TNFa or ceramide could mimic ischemic preconditioning (Lecour et al., 2002). Furthermore, Hallenbeck’s group reported that hypoxic preconditioning protects cultured neurons against hypoxic stress via TNFa and ceramide (Liu et al., 2000). More recently, Karliner’s group elegantly demonstrated that sphingosine kinase, an enzyme involved in the sphingolipid cascade, mediates ischemic preconditioning-induced cardioprotection (Jin et al., 2004). Interestingly, another biological program that has been suggested in ischemic preconditioning is the generation of reactive oxygen species (ROS) (see review in Yellon and Downey, 2003). ROS have generally been described as being harmful for cells but now there is growing evidence that ROS

S. Lecour et al. / Life Sciences 78 (2006) 1702 – 1706

may have a dual effect, becoming cytoprotective in certain physiopathological conditions. Our recent findings have provided firm evidence for the production and role of free radicals in TNFa-induced cardioprotection (Lecour et al., 2005). In the present study, we hypothesized that pre-ischemic ROS activation mediates ceramide-induced cardioprotection. Therefore, we investigated the effect of the antioxidant, N-2mercaptopropionyl glycine (MPG) on the resistance to myocardial infarction conferred by ceramide in the isolated rat heart system. In physiological conditions, a production of free radicals is associated with an increased activity of the endogenous antioxidant enzymes. To give indirect evidence of free radical production after ceramide stimulation, we measured the activities of the endogenous antioxidant enzymes catalase and superoxide dismutase at different time points.

1703

Stabilisation

Ischemia

Reperfusion

Control

Control + DMS/MPG Ceramide

Ceramide + DMS/MPG

Cer

Cer

Infarct assessment

DMS 1µM or MPG 1 mM

Enzyme activities

Ischemia

Fig. 1. Treatment protocols for isolated perfused rat heart studies: ceramide (1 AM); N-2-mercaptopropionyl glycine (MPG, 1 mM).

Methods Animal groups All the experiments were conducted on adult male LongEvans rats weighing 250 –300 g in accordance with the Guide for the Care and Use of Laboratory Animals published by the U.S. National Institutes of Health (NIH Publication No. 85 (23), revised 1996) and all procedures were approved by the Faculty of Health Sciences Animal Ethics Committee, University of Cape Town.

min followed by 5 min washout prior to the ischemia – reperfusion insult. For infarct size measurement, the coronary artery was reoccluded at the end of the reperfusion period and a solution of 2.5% Evans Blue was perfused to delineate the area at risk. Hearts were then frozen and cut into slices, incubated in sodium phosphate buffer containing 1% w/v triphenyltetrazolium chloride (TTC) for 15 min to visualize the unstained infarcted region. Infarct and risk zone areas were determined with planimetry and infarct size was expressed as a percentage of the risk zone.

Experimental protocols Measurements of antioxidant enzymes activities Rats were anesthetized with 50 mg/kg intraperitoneal sodium pentobarbitone and were given an intravenous injection of 200 international units (IU) heparin. Hearts were excised rapidly and perfused retrogradely using the Langendorff technique at a constant pressure (100 cm H2O) with oxygenated Krebs –Henseleit buffer. A balloon was inserted through the left atrium into the left ventricle and the left-ventricular end diastolic pressure (LVEDP) was adjusted between 4 and 8 mmHg. Cardiac parameters were monitored continuously and included heart rate (HR), left ventricular developed pressure (LVDP: difference between left ventricular endsystolic pressure and end-diastolic pressure) and the coronary flow (CF). The perfusion protocol is shown in Fig. 1. All hearts were allowed to equilibrate for 15 min and were consequently subjected to a standard 30 min regional ischemia (induced by tightening a snare around the left coronary artery) followed by 120 min of reperfusion. A low dose of the cell-permeable c2ceramide (1 AM, corresponding to a lower concentration than physiological plasma ceramide concentration in rats (Zimmermann et al., 2001)) was given for 7 min followed by a 10 min washout period before the standard ischemia. Additional groups were perfused with the free radical scavenger N-2mercaptopropionyl glycine (MPG, 1 mM) or with a sphingosine kinase inhibitor dimethyl sphingosine (DMS, 1 AM) for 15

To evaluate catalase and superoxide dismutase (SOD) activities, the heart was removed from the cannula and the atria excised. The remainder of the heart was instantaneously frozen and kept at 80 -C until use. Afterwards, hearts were homogenized in 5 vol. of a 0.25 M sucrose solution and centrifuged at 6500g at 4 -C. The supernatant was then used for catalase determination with a modified method derived from Aebi and Clairbone (Dalloz et al., 1999). CuZnSOD and MnSOD activities analyses were based on the method developed by Mc Cord and Fridovich, as previously described (Dalloz et al., 1999). The enzymes activities were expressed in international units per mg of protein (IU/mg prot). Pharmacologic agents All chemicals were obtained from Sigma. Statistical analysis Data are presented as mean T S.E.M (n  6). Comparisons between multiple groups were performed by one-way ANOVA followed by the Student – Newman – Keuls post hoc test (GraphPad Instat). A value of p < 0.05 was considered statistically significant.

S. Lecour et al. / Life Sciences 78 (2006) 1702 – 1706

Results Haemodynamic data and myocardial infarct size Table 1 summarises haemodynamic data recorded in the four different experimental groups before the index ischemia, and during the ischemia –reperfusion insult. Left ventricular developed pressure, heart rate and coronary flow did not significantly differ between various groups. As shown in Fig. 2, control hearts subjected to 30 min regional ischemia and 120 min reperfusion had an infarct of 36.4 T 1.2% (expressed as a percentage of the area of risk). Pretreatment of the hearts with ceramide (1 AM) resulted in a reduction of the infarct size to 8.3 T 1.1% ( p < 0.05 vs. control). Addition of the sphingosine kinase inhibitor DMS during perfusion with ceramide blocked the protective effect of ceramide (31.0 T 2.8%, p = n.s. vs. control). When ceramide was perfused in the presence of the free radical scavenger MPG, the infarct-sparing effect was lost, resulting in a similar infarct size to the control hearts (36.7 T 4.9 vs. 36.4 T 1.2, p = n.s.).

Infarct size (% of the ischemic area)

1704

60

50

40

30

20

*

10

0 Control

Reperfusion 120 min

93 T 3 102 T 4 101 T 5 102 T 4 102 T 5 108 T 6

57 T 8 53 T 6 62 T 5 72 T 4 53 T 5 74 T 11

67 T 3 65 T 4 71 T 7 88 T 12 59 T 6 70 T 7

304 T 10 330 T 19 284 T 10 300 T 20 306 T 7 295 T 10

276 T 24 320 T 16 248 T 15 248 T 29 257 T 23 232 T 15

236 T 20 268 T 11 272 T 15 272 T 19 251 T 24 248 T 20

11.2 T 1.1 12.0 T 1.4 12.0 T 0.6 13.2 T 0.8 13.4 T 0.8 10.6 T 1.1

6.4 T 2.2 6.3 T 0.3 7.0 T 1.0 6.8 T 0.7 7.1 T 0.3 5.8 T 0.5

8.6 T 1.9 6.5 T 0.5 9.8 T 1.1 11.0 T 1.1 6.7 T 0.4 8.2 T 0.2

Parameters measured prior to ischemia (pre-ischemic), after 30 min of ischemia and after 120 min of reperfusion. Values are mean T S.E.M. (n  6). Ceramide = 1 AM, DMS = 1 AM, and MPG = 1 mM. LVDP= left ventricular developed pressure.

CuZnSOD (IU/mg of protein) Catalase (IU/mg of protein)

LVDP (mmHg) Control Control + DMS Control + MPG Ceramide Ceramide + DMS Ceramide + MPG Heart rate (beats/min) Control Control + DMS Control + MPG Ceramide Ceramide + DMS Ceramide + MPG Flow (ml/min) Control Control + DMS Control + MPG Ceramide Ceramide + DMS Ceramide + MPG

Ischemia 30 min

MPG

Ischemia

Preischemic

MnSOD (IU/mg of protein)

Table 1 Hemodynamic parameters of isolated perfused rat hearts

Ceramide + MPG

Ischemia Cer

Pre-ischemic biochemical measurements (Fig. 3) in hearts pretreated for 7 min with ceramide demonstrated that catalase and CuZnSOD activities were both increased compared with the control group (21.6 T 0.4 vs. 16.8 T 0.5 IU/mg of protein for catalase and 2.42 T 0.22 vs. 1.76 T 0.15 IU/mg of protein for CuZnSOD, p  0.05) while MnSOD activity remained unchanged. Ceramide pre-treatment did not affect the enzymes activities at the end of the 30 min ischemia.

DMS

Fig. 2. Effect of the sphingosine kinase inhibitor dimethyl sphingosine (DMS, 1 AM) or the antioxidant, N-2-mercaptopropionyl glycine (MPG, 1 mM) on the resistance to myocardial infarction conferred by ceramide (1 AM). Infarct size was measured following 30 min occlusion of the left coronary artery and 120 min of reperfusion. *p < 0.05 versus control.

Catalase and SOD activities analyses

Pre-ischemic

Ceramide Ceramide + DMS

24 20

*

30 min ischemia Ischemic area Non ischemic area

16 12 8 4 0

2.5

*

CTL Ceramide

2.0 1.5 1.0 0.5 0 20 16 12 8 4 0

Fig. 3. Determination of catalase and superoxide dismutase (CuZnSOD and MnSOD) activities in the isolated rat hearts. The activities were determined at the end of ceramide (1 AM) perfusion and at the end of the 30 min regional ischemia. Results are expressed in international units (IU)/mg of protein. Each bar represents the mean T S.E.M. (n = 10). *p < 0.05 versus control.

S. Lecour et al. / Life Sciences 78 (2006) 1702 – 1706

Discussion Using an isolated Langendorff rat heart model, exposure to a brief perfusion with ceramide 1 AM prior to an ischemia – reperfusion insult significantly reduced infarct size, in accordance with our previous study (Lecour et al., 2002). The mechanism for this preconditioning-like effect is currently unknown. The pertinent findings of this work are the implication of ROS as triggers to this cardioprotective effect, since MPG, administered simultaneously with ceramide, abolished the protective effect of ceramide. Sphingolipids and cardioprotection The sphingolipids signaling pathway is one of the pathways activated in response to ischemia – reperfusion. We propose that accumulation of ceramide during ischemia and/or reperfusion mediates, at least in part, the associated contractile dysfunction, apoptotic cell death and associated myocardial structural abnormalities. We have previously reported that ceramide and sphingosine1 phosphate could mimic ischemic preconditioning(Lecour et al., 2002). Moreover, the cardioprotective effect of ceramide could be abolished by (1) N-oleoyl ethanolamine (NOE), an inhibitor of the enzyme ceramidase that converts ceramide into sphingosine; or (2) by the sphingosine kinase inhibitor DMS (enzyme converting sphingosine into sphingosine-1 phosphate). These data strongly suggest that downstream products of ceramide in the sphingolipid pathway, i.e. sphingosine and sphingosine-1 phosphate, are required for this preconditioninglike effect. There are several studies now suggesting that the sphingolipid pathway is also involved in ischemic preconditioning. NOE or other inhibitors of the sphingolipid cascade can abolish ischemic preconditioning in the isolated rat heart (Cui et al., 2004; Lecour et al, 2002) and sphingosine kinase activation mediates ischemic preconditioning in the murine heart (Jin et al, 2004). ROS and cardioprotection It is now well established that ceramide can either promote cell survival or cell death in response to ischemic injury. Interestingly, similar properties have been reported for free radicals. Low doses of oxygen radicals can mimic ischemic preconditioning (Tritto et al., 1997) and MPG can abrogate a single episode of preconditioning (Baines et al., 1997). Free radicals may protect via activation of different kinases such as protein kinase C (Baines et al., 1997) or by acting on the mitochondrial permeability transition pore (Hausenloy et al., 2004). Recently, we have reported that TNFa-induced preconditioning was a free radical-mediated event. Using an isolated rat heart perfusion system and electron spin resonance spectroscopy, we have shown that TNFa perfusion is associated with release of free radicals into the coronary effluent (Lecour et al., 2005). In addition, the protective effect of TNFa was blocked in the presence of MPG, providing firm evidence for the

1705

production and the role of free radicals in TNFa-induced cardioprotection. In our present study, ceramide-induced cardioprotection was blocked in the presence of MPG and indirect evidence of ROS production was given by an increase of antioxidant enzymes activities following the perfusion of ceramide. Our findings concur with the work of others in which cells exposed to ceramide (5 to 10 AM) generated ROS production (Suematsu et al., 2003; Zhang et al., 2001). Mechanisms involved in ceramide-induced cardioprotection Little is known about molecular mechanisms involved in ceramide-induced cardioprotection. We have shown that the preconditioning effect of ceramide was abolished in the presence of a mitochondrial KATP channel blocker (Lecour et al., 2002). A role for nuclear factor nB (NF-nB) and calpain has also been suggested (Ginis et al., 1999, 2002). Our present study demonstrates that free radicals are likely to be part of the signaling cascade. However, the source of these free radicals still remains to be investigated. The mitochondrial electron chain transport, cyclooxygenase-2 and NADPH oxidase are sources of free radicals and can be activated by exposure to high doses of ceramide (Corda et al., 2001; Subbaramaiah et al., 1998; Zhang et al., 2003). A direct activation by ceramide of systems capable of generating free radicals such as xanthine oxidase or nitric oxide has never been reported to our knowledge. Although MnSOD activity was not increased in our model at the time points that we have considered, we cannot exclude a mitochondrial origin of the free radicals in our experiments as MnSOD and CuZnSOD activities may be increased at different time points after perfusion with ceramide. The exact role of free radicals in the cardioprotective effect of ceramide is also unknown. Sphingosine-1 phosphate, a downstream product of ceramide, induces cardioprotection via protein kinase C (Jin et al., 2002). In ischemic preconditioning, free radicals may protect via activation of protein kinase C (Yellon and Downey, 2003), therefore, it is appropriate to hypothesise that free radicals activated by ceramide may also protect via activation of protein kinase C. In conclusion, our data strongly support an essential role for free radicals in ceramide-induced preconditioning. However, further work will be required to delineate the source, the nature and the role of these free radials activated in this cardioprotective program. Acknowledgments This study was supported in part by the Hatter Institute Foundation, the Medical Research Council, and the Wellcome Trust. S.L. was supported by a Servier Senior lectureship for research in Heart Failure. We would like to thank Noel Markgraaf from the Animal Unit for his generous help. References Adam, D., Wiegmann, K., Adam-Klages, S., Ruff, A., Kronke, M., 1996. A novel cytoplasmic domain of the p55 tumor necrosis factor receptor initiates

1706

S. Lecour et al. / Life Sciences 78 (2006) 1702 – 1706

the neutral sphingomyelinase pathway. Journal of Biological Chemistry 271 (24), 14617 – 14622. Baines, C.P., Goto, M., Downey, J.M., 1997. Oxygen radicals released during ischemic preconditioning contribute to cardioprotection in the rabbit myocardium. Journal of Molecular and Cellular Cardiology 29 (1), 207 – 216. Corda, S., Laplace, C., Vicaut, E., Duranteau, J., 2001. Rapid reactive oxygen species production by mitochondria in endothelial cells exposed to tumor necrosis factor-alpha is mediated by ceramide. American Journal of Respiratory Cell and Molecular Biology 24 (6), 762 – 768. Cui, J., Engelman, R.M., Maulik, N., Das, D.K., 2004. Role of ceramide in ischemic preconditioning. Journal of the American College of Surgeons 198 (5), 770 – 777. Dalloz, F., Maingon, P., Cottin, Y., Briot, F., Horiot, J.C., Rochette, L., 1999. Effects of combined irradiation and doxorubicin treatment on cardiac function and antioxidant defenses in the rat. Free Radical Biology & Medicine 26 (7 – 8), 785 – 800. Ginis, I., Jaiswal, R., Klimanis, D., Liu, J., Greenspon, J., Hallenbeck, J.M., 2002. TNF-alpha-induced tolerance to ischemic injury involves differential control of NF-kappaB transactivation: the role of NF-kappaB association with p300 adaptor. Journal of Cerebral Blood Flow and Metabolism 22 (2), 142 – 152. Ginis, I., Schweizer, U., Brenner, M., Liu, J., Azzam, N., Spatz, M., Hallenbeck, J.M., 1999. TNF-alpha pretreatment prevents subsequent activation of cultured brain cells with TNF-alpha and hypoxia via ceramide. American Journal of Physiology 276 (5 Pt 1), C1171 – C1183. Hausenloy, D., Wynne, A., Duchen, M., Yellon, D., 2004. Transient mitochondrial permeability transition pore opening mediates preconditioning-induced protection. Circulation 109 (14), 1714 – 1717. Jin, Z.Q., Goetzl, E.J., Karliner, J.S., 2004. Sphingosine kinase activation mediates ischemic preconditioning in murine heart. Circulation 110 (14), 1980 – 1989. Jin, Z.Q., Zhou, H.Z., Zhu, P., Honbo, N., Mochly-Rosen, D., Messing, R.O., Goetzl, E.J., Karliner, J.S., Gray, M.O., 2002. Cardioprotection mediated by sphingosine-1-phosphate and ganglioside GM-1 in wild-type and PKC epsilon knockout mouse hearts. American Journal of Physiology. Heart and Circulatory Physiology 282 (6), H1970 – H1977. Lecour, S., Rochette, L., Opie, L., 2005. Free radicals trigger TNFalphainduced cardioprotection. Cardiovascular Research 65 (1), 239 – 243. Lecour, S., Smith, R.M., Sack, M.N., 2003. The role of immune and inflammatory cytokines in ischemic preconditioning of the heart—identification of novel cardiac cell survival signaling. In: Feuerstein, G., Libby, P., Mann, D. (Eds.), Progress in Inflammation Research: Inflammatory and Cardiac Diseases. Birkhauser, Basel, pp. 251 – 269. Lecour, S., Smith, R.M., Woodward, B., Opie, L.H., Rochette, L., Sack, M.N., 2002. Identification of a novel role for sphingolipid signaling in TNF alpha

and ischemic preconditioning mediated cardioprotection. Journal of Molecular and Cellular Cardiology 34 (5), 509 – 518. Liu, J., Ginis, I., Spatz, M., Hallenbeck, J.M., 2000. Hypoxic preconditioning protects cultured neurons against hypoxic stress via TNF-alpha and ceramide. American Journal of Physiology. Cell Physiology 278 (1), C144 – C153. Murry, C.E., Jennings, R.B., Reimer, K.A., 1986. Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium. Circulation 74 (5), 1124 – 1136. Schutze, S., Machleidt, T., Kronke, M., 1994. The role of diacylglyceramideol and ceramide in tumor necrosis factor and interleukin-1 signal transduction. Journal of Leukocyte Biology 56 (5), 533 – 541. Subbaramaiah, K., Chung, W.J., Dannenberg, A.J., 1998. Ceramide regulates the transcription of cyclooxygenase-2. Evidence for involvement of extracellular signal-regulated kinase/c-Jun N-terminal kinase and p38 mitogen-activated protein kinase pathways. Journal of Biological Chemistry 273 (49), 32943 – 32949. Suematsu, N., Tsutsui, H., Wen, J., Kang, D., Ikeuchi, M., Ide, T., Hayashidani, S., Shiomi, T., Kubota, T., Hamasaki, N., Takeshita, A., 2003. Oxidative stress mediates tumor necrosis factor-alpha-induced mitochondrial DNA damage and dysfunction in cardiac myocytes. Circulation 107 (10), 1418 – 1423. Tritto, I., D’Andrea, D., Eramo, N., Scognamiglio, A., De Simone, C., Violante, A., Esposito, A., Chiariello, M., Ambrosio, G., 1997. Oxygen radicals can induce preconditioning in rabbit hearts. Circulation Research 80 (5), 743 – 748. Wiegmann, K., Schutze, S., Machleidt, T., Witte, D., Kronke, M., 1994. Functional dichotomy of neutral and acidic sphingomyelinases in tumor necrosis factor signaling. Cell 78 (6), 1005 – 1015. Yellon, D.M., Downey, J.M., 2003. Preconditioning the myocardium: from cellular physiology to clinical cardiology. Physiological Reviews 83 (4), 1113 – 1151. Zhang, D.X., Zou, A.P., Li, P.L., 2001. Ceramide reduces endotheliumdependent vasodilation by increasing superoxide production in small bovine coronary arteries. Circulation Research 88 (8), 824 – 831. Zhang, D.X., Zou, A.P., Li, P.L., 2003. Ceramide-induced activation of NADPH oxidase and endothelial dysfunction in small coronary arteries. American Journal of Physiology. Heart and Circulatory Physiology 284 (2), H605 – H612. Zimmermann, C., Ginis, I., Furuya, K., Klimanis, D., Ruetzler, C., Spatz, M., Hallenbeck, J.M., 2001. Lipopolysaccharide-induced ischemic tolerance is associated with increased levels of ceramide in brain and in plasma. Brain Research 895 (1 – 2), 59 – 65.