J Mol
Cell
Cardiol
Propranolol
23,
1231-1244
(1991)
Reduces Anoxia/Reoxygenation-mediated MyocytesThroughanAnti-radicalMechanism
Jay H. Kramer,
I. Tong
Mak,
Anthony
M. Freedman
Injury
and William
ofAdult
B. Weglicki
Department ofMedicine, Division of Ex@rirnentalMedicine, The George Washington University Medical Center, 2300 EyeStreet, N. W., WnFhington, D. C. 20037, USA (Received 8 January 1991, accepted in revisedform 12 June 1991) J. H. KRAMER, I.T. MAK, A.M. FREEDMAN, AND W.B. WEGLICKI. Propmnolol Reduces Anoxia/Reoxygenation-mediated Injury of Adult Myocytes Through an Anti-radical Mechanism. Joumul o/ Molmhr and Cellular Cardiolqy (1991) 23, 1231-1244. The effects of propranolol (PRO) and atenolol (ATE) on adult canine myocytes exposed to 30 min anoxia (A:95% N,/5 % CO,) and subsequent reoxygenation (R:95% 0,/5% CO,) for up to 20 min was investigated. In some studies, comparison of effects were made with that of superoxide dismutase (SOD). Although anoxia alone produced only minimal injury, reoxygenation in the absence of the j3-blockers or SOD was associated with significant losses of cellular viability, elevated release of cellular lactate dehydrogenase, and increased formation of lipid peroxidation products. Myocytes exposed to A/R in the presence of d,l-PRO (20, 200~~) were afforded substantial, concentration-dependent protection during 20 min reoxygenation. Significant protection was also observed in the presence of 2 pM d-PRO (non-active flblocker), but only after a longer preincubation period (2 h). SOD (lOpg/ml) provided equi-potent protection to that of 2OOpM d,l-PRO. By contrast, the more water-soluble b-blocker, ATE (2OOcM), offered only minor protection. Electron Spin Resonance spin trapping studies using wphenyl-tert-butylnitrone (PBN) were also performed with A/R myocytes in the presence or absence of drug treatment. Short-term (10min) exposure to d,l-PRO (200~~) prior to A/R, or to SOD, resulted in a 71-84% reduction in total PBN lipid radical adduct formation (alkoxyl; csH = 2.0-2.5G, csN = 13.5-13.75G); long-term exposure (2h) to 2p~ d-PRO resulted in a 51% reduction. These data suggest that the superoxide anion was an initiator ofevents leading to subsequent lipid radical formation and that the anti-peroxidative properties of PRO appear to be independent of &receptor blockade.
KEY WORDS: Resonance
Isolated adult Spin Trapping;
canine myocytes; cr-Phenyl-tert-butylnitrone;
Free
radicals; d,l-
Lipid and
peroxidation; d-Propranolol.
Cellular
viability;
Electron
Spin
Introduction bursts of production, is the formation of conEvidence is accumulating which suggests an jugated dienes, lipid peroxides and lipid association between elevated levels of free aldehydes, suggesting damage to cellular radical production and the subsequent tissue membrane components [42]. Thus, the relatinjury resulting from post-ischemic reperfusively high phospholipid and polyunsaturated ion [19, 28, 29, 491. Although the cellular fatty acid content of sarcolemma from ventricorigin of free radical production remains a ular myocytes [30] may enhance the vulnermatter of conjecture, various studies using ability of this membrane system to free radical electron spin resonance spin trapping techattack [19, 301. In previous studies, we deniques have clearly demonstrated bursts of scribed both the potentiation of free radicalproduction of primary oxygen (superoxide mediated peroxidative injury to sarcolemma anion, hydroxyl) and secondary oxygen- and by surface-active lipid amphiphiles, as well as, carbon-centered (alkoxyl, alkyl, etc.) free protection against peroxidative events by the radicals during early minutes of tissue reperamphiphilic beta-blocker, d,l-propranolol fusion [I, 7, 20, 281. Coincidental with these PI. Grant
409
support:
NIH
grants
POl-HL38079
Please address all correspondence Ross Hall, The George Washington
0022-2828/91/111231
+14$03,00/O
and
ROl-HL36418.
to: Jay H. Kramer, University Med.
Division of Experimental Ctr., 2300 Eye Street,
Medicine, Department N.W., Washington, D.C. @ 1991 Academic
of Medicine, 20037, USA. Press
LImited
J. H. Kramer
1232
In a more recent study, we were able to show that the anti-radical membrane protective effect of d,l-propranolol could be translated into cytoprotective effects in freshly isolated adult canine myocytes subjected to exogenouslygenerated oxygen free radicals [35]. Propranolol treatment provided significant preservation of and ultrastructure, improved viability, reduced both the release of cellular lactate dehydrogenase (LDH) and formation of lipid peroxidation products (malondialdehyde), compared to untreated myocytes. In the present study, we present evidence for the occurrence and participation of free radicals during injury of myocytes exposed to anoxia/reoxygenation. The current investigation, like the previous one, demonstrates a dose-dependent (2.0-200~~) protective effect of d,l-propranolol. This study also suggests the unlikely involvement of P-blockade as the mechanism of protection based on our comparative studies: (1) with the more water soluble P-blocker, atenolol; and (2)with a clinically-relevant concentration of the nonactive d-form of propranolol. Finally, comparisons are made between the effects of propranolol and the classic anti-radical agent, superoxide dismutase, on secondary production of lipid free radicals in electron spin resonance spin trapping studies using the trap (Yphenyl-tert-butylnitrone. This represents the first report of anoxialreoxygenation-mediated secondary free radical production in freshly isolated adult ventricular myocytes using ESR spin trapping.
Materials
and Methods
Chemicals D,Land d-Propranolol, atenolol, 2-thiobarbituric acid (TBA), FeSO.+’ 7H,O, cumene hydroperoxide and superoxide dismutase (Bovine erythrocyte, 3500 U/mg) were obtained from Sigma Chemicals. The traps OLphenyl-N-tert-butylnitrone and PW 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) were purchased from Aldrich Chemicals. All other reagents were purchased from either Sigma Chemicals or Fisher Scientific. Buffers and reagents were prepared in double deionized. distilled water.
et al. Isolation of calcium-tolerant
myocytes
Adult canine ventricular myocytes were isolated using a modification of the enzymatic digestion procedure reported previously [ 481. The cells obtained following collagenase digestion of tissue were resuspended in 10 mM potassium phosphate buffer (pH 7.4) containing (mM): glucose, 11; pyruvate, 5; succinate, 5; MgS04’ 7H20, 1.2; taurine, 20; KCl, 120; ATP, 1 .O; and 0.2% BSA. The cells were pelleted by centrifugation (50 x g, 1 min) and were then resuspended and exposed to several 30 min incubations (gassing, gently stirred, 24OC) with progressively higher calcium concentrations (0.05-1.25mM). The calcium tolerant cells were washed 3 to 5 times and the final pellet was resuspended in a 10 mM potassium phosphate incubation buffer (pH 7.4) containing (mM): NaCl, 120; MgS04.7H20, 1.2; and glucose, pyruvate and acetate (5mM each). The resulting myocyte preparations were usually 90-95% free of non-myocytic cells, with average yields of 2 x lo8 cells. Viabilities of 65 to 85 % rod-shaped myocytes, based on morphology and trypan blue exclusion, were typically obtained. Incubation procedure The isolated myocytes were preincubated for 10 min at 37’C with intermittent gassing (95 % 0$5% CO*) in the presence or absence of 2OOpM atenolol or d,l-propranolol (2, 20 or 2OOpM) prior to anoxic exposure. Although propranolol is lipophilic, it is also reasonably soluble in aqueous solution. Consequently, the preincubation period reflects the time needed to permit partitioning of the drug between aqueous and hydrophobic (membrane bilayer) phases. Previous studies have indicated that a was minimum of 10 min preincubation required to display significant anti-oxidant effects of propranolol at the higher concentrations [32, 351. Cell suspensions were diluted (1 .O x lo6 cells/ml or 10.13 +- 1.1 mg protein/ml final) into nitrogen-gassed incubation buffer and were then gassed again (less than 1 min) with 95% N,/5% CO2 to achieve an anoxic state (monitored with a Yellow Springs Instrument oxygen sensor and Clark oxygen electrode). Samples were incubated (37W) for 30 min in anaerobically-sealed test-tubes. At the end of 30min, cell suspensions were re-
Propranolol
protects
Anoxic/Reoxygenated
oxygenated with 95% 0,/5% CO?. After 20 min reoxygenation, aliquots of suspension were retrieved and were used for assessment of lipid peroxidation, free and cell bound activities of lactate dehydrogenase (LDH), or loss of cellular viability. ESR spin trapping studies were conducted on A/R-exposed myocytes in the presence or absence of 200pM d,l-propranolol as described below. In some studies, we evaluated the influence of exogenous calcium on cellular viability and PBN adduct formation in myocytes exposed to AIR with or without 200~~ d,l-propranolol pretreatment. Paired experiments (n = 4) were conducted in the presence or absence of 1.25 mM calciumsupplemented incubation medium. Selected studies (viability and ESR spin trapping) were also conducted in which myocytes were subjected to: (i) 2 h preincubation with 2pM d-propranolol prior to AIR exposure; or (ii) SOD (iO~g/ml final) introduction 1 min prior to reoxygenation, to simulate an in vivo approach which considers the relatively short half-life of the enzyme. Since SOD will generally be confined to the extracellular aqueous space, preincubation was not considered necessary. In all instances, myocyte appropriately-incubated control samples were taken for comparisons with the experimentally-treated cells. Pharmacological treatments had no significant effects on control myocyte viability, nor did they induce LDH release or promote lipid peroxidation.
Assays and measurements Cellular viability was monitored by trypan blue exclusion (0.1% final) and was found to be consistent with morphological changes; with few exceptions, rounded myocytes took up the dye and rod-shaped cells were unstained. Free (incubation media) and cell pellet (cell pellet homogenized in 0.15% Triton-X100) LDH activities were spectrophotometrically determined from 1 ml suspension aliquots using Sigma kits. Lipid peroxidation measurements were assessed (0.5 ml samples) as the formation of thiobarbituric acid (TBA) reactive materials by a previously described calorimetric procedure [35] and results were expressed as malondialdehyde (MDA) equivalents. The antioxidant, BHT (0.01 %), was added during the heating step (80%) to pre-
Myocytes
1233
vent non-specific color development. The pharmacological agents (up to 200 PM) used in this study caused no interference in color development. The simplicity and sensitivity of this assay makes it the most widely used procedure to assess lipid peroxidation in vitro. Nevertheless, the potential non-specificity of the TBA reaction permits this assay to be used only as a qualitative index of lipid peroxidation.
ESR spin trapping and sample extraction ESR spin trapping studies were performed in room darkness on control and A/R-exposed myocyte samples in the presence or absence of interventions as described above. Since some spin trapping agents may possess concentration-dependend toxic [ 7, II, 131 or non-toxic [ 7, 22, 231 cardioactive properties which could confound interpretation of results, a method was devised which circumvents direct spin trap exposure to the cells. Following 30 min anoxia and at various times of reoxygenation up to 15 min, 1.5 ml aliquots of cell suspension were removed and rapidly filtered through a Millipore vacuum filtration manifold (Whatman GF/C filters) into glass tubes containing 1.5 m! 120 mM PBN (saline). Filtration of the cell suspension negates any chance of cellular metabolism of spin traps or adducts [43, 441. The filtrate/PBN mixtures were immediately frozen in liquid nitrogen and subsequently extracted (l-v01 sample: 1.5-vol toluene; 20%) with nitrogen-gassed, HPLC-grade toluene; centrifuged at 170 x g, 5min; extracted volumes were reduced under nitrogen gas; and samples were reconstituted in 0.3 ml toluene prior to ESR spectroscopy. Extraction efficiency for the spin trap was 85-95s as determined by UV spectroscopy [II]. This procedure provides reasonable stability to the spin adduct and has not been associated with significant ESR signal artifact.
Control experiments to assess the spin trapping methodology Due to the relatively high concentrations of cells required to conduct our spin trapping studies, some experiments aimed at characterizing the spin trapping methodology were performed in vitro using spin traps and exogenous
free radical generating systems. We were able to generate PBN adducts from analogs of lipid hydro~roxides (LOOH) which had spectral parameters virtually identical to that observed in A/R-exposed myocytes. Spin trapping was performed at 37% in a 3 mi aqueous reaction mixture containing (final) 60 rnM PBN, 8.3 ,ni~ FeSO+ 7H20 and 0.67 % (v/v> cumene hydroperoxide (CuQQH). An ahquot (0.5 ml) of the aqueous sample was analyzed by ESR spectroscopy and the remaining reaction mixture was extracted with toluene prior to ESR analysis, as described above. The hyper~~e splitting constants (ct~ f 3.5G; oN = 15.OG) measured for the aqueous sofution was essentially = 15.08G) identical to values (a~ = 3.52G; @$4J reported for a PBN-aikoxyl adduct in the aqueous phase [8]. Similar studies were conducted in the presence of 2~~~M d,~-proprano101, to determine if this agent directly interfered with the i~teractian between PBN and free radicals. Propranoloi was found to have no effect on ESR spectral parameters and only a slight effect on signal intensity (5-8s reduction) of the toiuene-extracted samples (data not shown); this effect on signal intensity can represent only a minor contribution to the drug’s anti-radical effect observed in the myocyte studies. Recent evidence suggests that the hydroxyl adduct of PBN (PBN-OH . ) may be the duminant adduct of spectra obtained fo~owing toluene extraction of ~st-ischemic corona eflluent samples [4]. Zn order to test for its possible contribution in our studies, ESR spectroscopy was performed on toluene extracts of an aqueous Fenton system mixture (HI@* + Fe2 + and 60 mM PBN); the hyper~ne splitting constants (cry = 2.56; ffN = 14.5 G) of the PBN-OH * adduct obtained under these conditions were not consistent with the spectral parameters observed in the myocyte studies. We also considered the possibility that contaminating iron in our incubation medium substantially contr~bnted to the free radicals detected during AIR. However, as determined by ESR spin trapping, contaminating iron does not appear to have been a significant participant in the generation of either primary oxygen-centered (x120, + 50 rnM DMPO in buffer, no added iron) or secondary lipid (CuOOH -t 60m~ PBN in buffer, no added iron) free radicals. No significant ESR signais
were detected unless l-2 CM (or greater) Fe” + was introduced. This suggests that most of the trace iron participating in free radical production had an endogenous origin, presumabIy from myocytes injured during exposure to A/R.
ESR analysis [I, 281 was performed at room temperature with a Bruker ER 100 series, Xband spectrometer using a magnetic field modulation frequency of 1OOKHz. The ESR spectra was obtained using a quartz flat ceff (60 x 10 x 0.25 mm). The microwave power was maintained at 10 mW to avoid saturation, and scans were performed with a modulation amplitude equal to or lower than 1 G. Unless otherwise stated, gain was set at 1.25 x 106, and sweep time was 200 s. Hyperfine coupling constants were measured directly from the field scan using 10G marker signal for c~ibration and measurements of signal intensity were performed as previously reported [28].
Analysis of variance was used for comparison of several means and the Tukey test was used for aI1 paired comparison of means. Significance was considered at P
Cell
Cardiol
Propranolol
23,
1231-1244
(1991)
Reduces Anoxia/Reoxygenation-mediated MyocytesThroughanAnti-radicalMechanism
Jay H. Kramer,
I. Tong
Mak,
Anthony
M. Freedman
Injury
and William
ofAdult
B. Weglicki
Department ofMedicine, Division of Ex@rirnentalMedicine, The George Washington University Medical Center, 2300 EyeStreet, N. W., WnFhington, D. C. 20037, USA (Received 8 January 1991, accepted in revisedform 12 June 1991) J. H. KRAMER, I.T. MAK, A.M. FREEDMAN, AND W.B. WEGLICKI. Propmnolol Reduces Anoxia/Reoxygenation-mediated Injury of Adult Myocytes Through an Anti-radical Mechanism. Joumul o/ Molmhr and Cellular Cardiolqy (1991) 23, 1231-1244. The effects of propranolol (PRO) and atenolol (ATE) on adult canine myocytes exposed to 30 min anoxia (A:95% N,/5 % CO,) and subsequent reoxygenation (R:95% 0,/5% CO,) for up to 20 min was investigated. In some studies, comparison of effects were made with that of superoxide dismutase (SOD). Although anoxia alone produced only minimal injury, reoxygenation in the absence of the j3-blockers or SOD was associated with significant losses of cellular viability, elevated release of cellular lactate dehydrogenase, and increased formation of lipid peroxidation products. Myocytes exposed to A/R in the presence of d,l-PRO (20, 200~~) were afforded substantial, concentration-dependent protection during 20 min reoxygenation. Significant protection was also observed in the presence of 2 pM d-PRO (non-active flblocker), but only after a longer preincubation period (2 h). SOD (lOpg/ml) provided equi-potent protection to that of 2OOpM d,l-PRO. By contrast, the more water-soluble b-blocker, ATE (2OOcM), offered only minor protection. Electron Spin Resonance spin trapping studies using wphenyl-tert-butylnitrone (PBN) were also performed with A/R myocytes in the presence or absence of drug treatment. Short-term (10min) exposure to d,l-PRO (200~~) prior to A/R, or to SOD, resulted in a 71-84% reduction in total PBN lipid radical adduct formation (alkoxyl; csH = 2.0-2.5G, csN = 13.5-13.75G); long-term exposure (2h) to 2p~ d-PRO resulted in a 51% reduction. These data suggest that the superoxide anion was an initiator ofevents leading to subsequent lipid radical formation and that the anti-peroxidative properties of PRO appear to be independent of &receptor blockade.
KEY WORDS: Resonance
Isolated adult Spin Trapping;
canine myocytes; cr-Phenyl-tert-butylnitrone;
Free
radicals; d,l-
Lipid and
peroxidation; d-Propranolol.
Cellular
viability;
Electron
Spin
Introduction bursts of production, is the formation of conEvidence is accumulating which suggests an jugated dienes, lipid peroxides and lipid association between elevated levels of free aldehydes, suggesting damage to cellular radical production and the subsequent tissue membrane components [42]. Thus, the relatinjury resulting from post-ischemic reperfusively high phospholipid and polyunsaturated ion [19, 28, 29, 491. Although the cellular fatty acid content of sarcolemma from ventricorigin of free radical production remains a ular myocytes [30] may enhance the vulnermatter of conjecture, various studies using ability of this membrane system to free radical electron spin resonance spin trapping techattack [19, 301. In previous studies, we deniques have clearly demonstrated bursts of scribed both the potentiation of free radicalproduction of primary oxygen (superoxide mediated peroxidative injury to sarcolemma anion, hydroxyl) and secondary oxygen- and by surface-active lipid amphiphiles, as well as, carbon-centered (alkoxyl, alkyl, etc.) free protection against peroxidative events by the radicals during early minutes of tissue reperamphiphilic beta-blocker, d,l-propranolol fusion [I, 7, 20, 281. Coincidental with these PI. Grant
409
support:
NIH
grants
POl-HL38079
Please address all correspondence Ross Hall, The George Washington
0022-2828/91/111231
+14$03,00/O
and
ROl-HL36418.
to: Jay H. Kramer, University Med.
Division of Experimental Ctr., 2300 Eye Street,
Medicine, Department N.W., Washington, D.C. @ 1991 Academic
of Medicine, 20037, USA. Press
LImited
J. H. Kramer
1232
In a more recent study, we were able to show that the anti-radical membrane protective effect of d,l-propranolol could be translated into cytoprotective effects in freshly isolated adult canine myocytes subjected to exogenouslygenerated oxygen free radicals [35]. Propranolol treatment provided significant preservation of and ultrastructure, improved viability, reduced both the release of cellular lactate dehydrogenase (LDH) and formation of lipid peroxidation products (malondialdehyde), compared to untreated myocytes. In the present study, we present evidence for the occurrence and participation of free radicals during injury of myocytes exposed to anoxia/reoxygenation. The current investigation, like the previous one, demonstrates a dose-dependent (2.0-200~~) protective effect of d,l-propranolol. This study also suggests the unlikely involvement of P-blockade as the mechanism of protection based on our comparative studies: (1) with the more water soluble P-blocker, atenolol; and (2)with a clinically-relevant concentration of the nonactive d-form of propranolol. Finally, comparisons are made between the effects of propranolol and the classic anti-radical agent, superoxide dismutase, on secondary production of lipid free radicals in electron spin resonance spin trapping studies using the trap (Yphenyl-tert-butylnitrone. This represents the first report of anoxialreoxygenation-mediated secondary free radical production in freshly isolated adult ventricular myocytes using ESR spin trapping.
Materials
and Methods
Chemicals D,Land d-Propranolol, atenolol, 2-thiobarbituric acid (TBA), FeSO.+’ 7H,O, cumene hydroperoxide and superoxide dismutase (Bovine erythrocyte, 3500 U/mg) were obtained from Sigma Chemicals. The traps OLphenyl-N-tert-butylnitrone and PW 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) were purchased from Aldrich Chemicals. All other reagents were purchased from either Sigma Chemicals or Fisher Scientific. Buffers and reagents were prepared in double deionized. distilled water.
et al. Isolation of calcium-tolerant
myocytes
Adult canine ventricular myocytes were isolated using a modification of the enzymatic digestion procedure reported previously [ 481. The cells obtained following collagenase digestion of tissue were resuspended in 10 mM potassium phosphate buffer (pH 7.4) containing (mM): glucose, 11; pyruvate, 5; succinate, 5; MgS04’ 7H20, 1.2; taurine, 20; KCl, 120; ATP, 1 .O; and 0.2% BSA. The cells were pelleted by centrifugation (50 x g, 1 min) and were then resuspended and exposed to several 30 min incubations (gassing, gently stirred, 24OC) with progressively higher calcium concentrations (0.05-1.25mM). The calcium tolerant cells were washed 3 to 5 times and the final pellet was resuspended in a 10 mM potassium phosphate incubation buffer (pH 7.4) containing (mM): NaCl, 120; MgS04.7H20, 1.2; and glucose, pyruvate and acetate (5mM each). The resulting myocyte preparations were usually 90-95% free of non-myocytic cells, with average yields of 2 x lo8 cells. Viabilities of 65 to 85 % rod-shaped myocytes, based on morphology and trypan blue exclusion, were typically obtained. Incubation procedure The isolated myocytes were preincubated for 10 min at 37’C with intermittent gassing (95 % 0$5% CO*) in the presence or absence of 2OOpM atenolol or d,l-propranolol (2, 20 or 2OOpM) prior to anoxic exposure. Although propranolol is lipophilic, it is also reasonably soluble in aqueous solution. Consequently, the preincubation period reflects the time needed to permit partitioning of the drug between aqueous and hydrophobic (membrane bilayer) phases. Previous studies have indicated that a was minimum of 10 min preincubation required to display significant anti-oxidant effects of propranolol at the higher concentrations [32, 351. Cell suspensions were diluted (1 .O x lo6 cells/ml or 10.13 +- 1.1 mg protein/ml final) into nitrogen-gassed incubation buffer and were then gassed again (less than 1 min) with 95% N,/5% CO2 to achieve an anoxic state (monitored with a Yellow Springs Instrument oxygen sensor and Clark oxygen electrode). Samples were incubated (37W) for 30 min in anaerobically-sealed test-tubes. At the end of 30min, cell suspensions were re-
Propranolol
protects
Anoxic/Reoxygenated
oxygenated with 95% 0,/5% CO?. After 20 min reoxygenation, aliquots of suspension were retrieved and were used for assessment of lipid peroxidation, free and cell bound activities of lactate dehydrogenase (LDH), or loss of cellular viability. ESR spin trapping studies were conducted on A/R-exposed myocytes in the presence or absence of 200pM d,l-propranolol as described below. In some studies, we evaluated the influence of exogenous calcium on cellular viability and PBN adduct formation in myocytes exposed to AIR with or without 200~~ d,l-propranolol pretreatment. Paired experiments (n = 4) were conducted in the presence or absence of 1.25 mM calciumsupplemented incubation medium. Selected studies (viability and ESR spin trapping) were also conducted in which myocytes were subjected to: (i) 2 h preincubation with 2pM d-propranolol prior to AIR exposure; or (ii) SOD (iO~g/ml final) introduction 1 min prior to reoxygenation, to simulate an in vivo approach which considers the relatively short half-life of the enzyme. Since SOD will generally be confined to the extracellular aqueous space, preincubation was not considered necessary. In all instances, myocyte appropriately-incubated control samples were taken for comparisons with the experimentally-treated cells. Pharmacological treatments had no significant effects on control myocyte viability, nor did they induce LDH release or promote lipid peroxidation.
Assays and measurements Cellular viability was monitored by trypan blue exclusion (0.1% final) and was found to be consistent with morphological changes; with few exceptions, rounded myocytes took up the dye and rod-shaped cells were unstained. Free (incubation media) and cell pellet (cell pellet homogenized in 0.15% Triton-X100) LDH activities were spectrophotometrically determined from 1 ml suspension aliquots using Sigma kits. Lipid peroxidation measurements were assessed (0.5 ml samples) as the formation of thiobarbituric acid (TBA) reactive materials by a previously described calorimetric procedure [35] and results were expressed as malondialdehyde (MDA) equivalents. The antioxidant, BHT (0.01 %), was added during the heating step (80%) to pre-
Myocytes
1233
vent non-specific color development. The pharmacological agents (up to 200 PM) used in this study caused no interference in color development. The simplicity and sensitivity of this assay makes it the most widely used procedure to assess lipid peroxidation in vitro. Nevertheless, the potential non-specificity of the TBA reaction permits this assay to be used only as a qualitative index of lipid peroxidation.
ESR spin trapping and sample extraction ESR spin trapping studies were performed in room darkness on control and A/R-exposed myocyte samples in the presence or absence of interventions as described above. Since some spin trapping agents may possess concentration-dependend toxic [ 7, II, 131 or non-toxic [ 7, 22, 231 cardioactive properties which could confound interpretation of results, a method was devised which circumvents direct spin trap exposure to the cells. Following 30 min anoxia and at various times of reoxygenation up to 15 min, 1.5 ml aliquots of cell suspension were removed and rapidly filtered through a Millipore vacuum filtration manifold (Whatman GF/C filters) into glass tubes containing 1.5 m! 120 mM PBN (saline). Filtration of the cell suspension negates any chance of cellular metabolism of spin traps or adducts [43, 441. The filtrate/PBN mixtures were immediately frozen in liquid nitrogen and subsequently extracted (l-v01 sample: 1.5-vol toluene; 20%) with nitrogen-gassed, HPLC-grade toluene; centrifuged at 170 x g, 5min; extracted volumes were reduced under nitrogen gas; and samples were reconstituted in 0.3 ml toluene prior to ESR spectroscopy. Extraction efficiency for the spin trap was 85-95s as determined by UV spectroscopy [II]. This procedure provides reasonable stability to the spin adduct and has not been associated with significant ESR signal artifact.
Control experiments to assess the spin trapping methodology Due to the relatively high concentrations of cells required to conduct our spin trapping studies, some experiments aimed at characterizing the spin trapping methodology were performed in vitro using spin traps and exogenous
free radical generating systems. We were able to generate PBN adducts from analogs of lipid hydro~roxides (LOOH) which had spectral parameters virtually identical to that observed in A/R-exposed myocytes. Spin trapping was performed at 37% in a 3 mi aqueous reaction mixture containing (final) 60 rnM PBN, 8.3 ,ni~ FeSO+ 7H20 and 0.67 % (v/v> cumene hydroperoxide (CuQQH). An ahquot (0.5 ml) of the aqueous sample was analyzed by ESR spectroscopy and the remaining reaction mixture was extracted with toluene prior to ESR analysis, as described above. The hyper~~e splitting constants (ct~ f 3.5G; oN = 15.OG) measured for the aqueous sofution was essentially = 15.08G) identical to values (a~ = 3.52G; @$4J reported for a PBN-aikoxyl adduct in the aqueous phase [8]. Similar studies were conducted in the presence of 2~~~M d,~-proprano101, to determine if this agent directly interfered with the i~teractian between PBN and free radicals. Propranoloi was found to have no effect on ESR spectral parameters and only a slight effect on signal intensity (5-8s reduction) of the toiuene-extracted samples (data not shown); this effect on signal intensity can represent only a minor contribution to the drug’s anti-radical effect observed in the myocyte studies. Recent evidence suggests that the hydroxyl adduct of PBN (PBN-OH . ) may be the duminant adduct of spectra obtained fo~owing toluene extraction of ~st-ischemic corona eflluent samples [4]. Zn order to test for its possible contribution in our studies, ESR spectroscopy was performed on toluene extracts of an aqueous Fenton system mixture (HI@* + Fe2 + and 60 mM PBN); the hyper~ne splitting constants (cry = 2.56; ffN = 14.5 G) of the PBN-OH * adduct obtained under these conditions were not consistent with the spectral parameters observed in the myocyte studies. We also considered the possibility that contaminating iron in our incubation medium substantially contr~bnted to the free radicals detected during AIR. However, as determined by ESR spin trapping, contaminating iron does not appear to have been a significant participant in the generation of either primary oxygen-centered (x120, + 50 rnM DMPO in buffer, no added iron) or secondary lipid (CuOOH -t 60m~ PBN in buffer, no added iron) free radicals. No significant ESR signais
were detected unless l-2 CM (or greater) Fe” + was introduced. This suggests that most of the trace iron participating in free radical production had an endogenous origin, presumabIy from myocytes injured during exposure to A/R.
ESR analysis [I, 281 was performed at room temperature with a Bruker ER 100 series, Xband spectrometer using a magnetic field modulation frequency of 1OOKHz. The ESR spectra was obtained using a quartz flat ceff (60 x 10 x 0.25 mm). The microwave power was maintained at 10 mW to avoid saturation, and scans were performed with a modulation amplitude equal to or lower than 1 G. Unless otherwise stated, gain was set at 1.25 x 106, and sweep time was 200 s. Hyperfine coupling constants were measured directly from the field scan using 10G marker signal for c~ibration and measurements of signal intensity were performed as previously reported [28].
Analysis of variance was used for comparison of several means and the Tukey test was used for aI1 paired comparison of means. Significance was considered at P
