Cardiovascular Research 1992;26:449-455
449
Oxygen free radical injury and Gs mediated signal transduction in the stunned porcine myocardium Liang-Xiong Fu, Amfinn Ilebekk, Knut Arvid Kirkeben, Gunnar Aksnes, Finn Waagstein, Claes-HAkan Bergh, Johan Hoebeke, Qi-Ming Liang, Ake Hjalmarson Objective: The aim was to investigate involvement of oxygen free radicals and any changes in the Gs mediated p adrenergic signalling system of stunned porcine myocardium. Methods: Myocardial stunning was induced in eight pentobarbitone anaesthetised pigs by brief occlusions of the distal left anterior descending coronary artery for periods of up to 10 min. Segment length function was measured in the ischaemic region and in a control region supplied by the circumflex artery. Left ventricular biopsies were obtained from the two regions 1 h after the last occlusion for ultrastructural and biochemical studies. Timolol was been used to prevent arrhythmia during ischaemia. Results: At the time when biopsies were obtained, percent systolic shortening was reduced to 58% in the region subjected to ischaemia and was only minimally reduced in the control region. In the biopsies from the stunned region: ( 1 ) electron microscopy showed mild and reversible intracellular changes in the stunned myocardium; (2) the activities of superoxide dismutase and glutathione peroxidase were decreased by 66% and 52%. respectively; (3) the content of malondialdehyde was increased by 49%; (4) neither density nor affinity of p adrenoceptors showed any changes; (5) there were no alterations in messenger RNA encoding for the (Y subunit of the stimulatory guanine necleotide binding protein (Gs), demonstrated by northern and dot-blot hybridisations; (6) ELISA technique utilising a specific antipeptide antibody showed no quantitative change in Gs; (7) the activity of adenyl cyclase was unchanged. Conclusions: Even though the stunned porcine myocardium showed substantial evidence of free radical injury, the p adrenergic signalling system was intact.
M
yocardial stunning is defined as a transient reversible myocardial dysfunction induced by brief periods of ischaemia, which do not lead to necrosis. This phenomenon is likely to be an important 'component in the dysfunction observed in patients after periods with unstable angina or coronary spasm and after therapeutic thrombolysis. During ischaemia and reperfusion production of oxygen free radicals has been demonstrated.' Free radical injury may cause lipid peroxidation of cellular membranes and thereby alter the physical pro erties of membranes and P membrane bound ion pumps? The reversible myocardial injury caused by short lasting ischaemia is in part caused Fy free radical production during ischaemia and reperfusion.The generation of free radicals associated with ischaemia and reperfu$on sequences varies inversely with collateral blood flow.- Since myocardial collateral blood flow in the pig heart is minimal,' this species should be optimal for the detection of oxygen free radical injury. The purpose of the present study was to determine if production of oxygen free radicals occurred during development of myocardial stunning in our preparation. We therefore measured the activities of the oxygen free radical scavengers superoxide dismutase and glutathione peroxidase and the content of the lipid peroxidation product, malondialdehyde. The f3 blocker timolol was used to prevent arrhythmia in the early period of ischaemia. The use of timolol does not exert any significant influence on the I
'
receptor-(; protein-adenylyl cyclase system in the present protocol since it has been shown that there is no change in the p adrenoce tor density after timolol treatment in human myocardium. 1 2 The importance of the myocardial p adrenergic signalling system has been investigated in early' and in prolonged periods of ischaemia.'&l' After long lasting ischaemia, which induces myocardial infarction, an increased density of I adrenoceptors has been shown in the canine myocardium. Lipid peroxidation has been shown to decrease the density of p adrenoceptors.' Since it has been suggested that free radical injury is an important factor in the development of myocardial stunning, the function of the p adrenoceptor signalling system in the reversibly injured myocardium could be affected. To our knowledge, this has not previously been investigated. A second objective of this study was therefore to determine the density of p adrenoceptors, as well as any alterations in the Gs protein and the adenylyl cyclase system in the stunned myocardium.
I-k
Methods Animul preparution
Eight domestic pigs of either six weighing 19-27 kg (range) were anaesthetised with pentobarbitone sodium intraperitoneally (30 mg.kg-'), followed by a continuous intravenous sustaining dose of 5-20 mg.kg-'.h-l according to depth of anaesthesia. The pigs were tracheostomised and
Wallenberg Laboratory, Sahlgren's Hospital, University of Goteborg, 41 3 45 Goteborg, Sweden: L-X Fu, F Waagstein, C-H Bergh, Q-M Liang, p\ Hjalmarson; Institute for Experimental Medical Research, Ullevaal Hospital, Oslo, Norway: A Ilebekk, K A Kirkeben, G Aksnes; Laboratoire des Proteines des Liquides Biologiques, URA 1334, CNRS, Tours, France: J Hoebeke. Correspondence to Dr Fu.
450
Fu, Ilebekk, Kirkeben, Aksnes, Waagstein, Bergh, Hoebeke, Liang, Hjalmarsnn
artificially ventilated with air and oxygen by a volume regulated ventilator (Princeton Medical Instruments, Natick, MA, USA). Arterial blood gases were regularly checked by an automatic blood gas analyser (AVL Biomedical Instruments, Model 945, Graz, Austria). POZwas kept above 30 kPa in all animals, Pcoz at 6.4(SEM 1.1) and pH at 7.42(0.05). Body temperature was kept at 37.6(0.7)”C by wrappings and electric heating pads. Urine was continuously drained through a cystostoma. The heart was exposed through a midsternal split and suspended in a pericardial cradle. A segment 5-7 cm distal on the left anterior descending coronary artery was dissected free for later intermittent occlusions with a Mayfield clip. Segment lengths were continuously recorded by ultrasonic technique in the ischaemic region and in a control region supplied by the circumflex coronary artery. A peripheral artery was cannulated for blood pressure recordings and for sampling arterial blood. A femoral vein was cannulated and saline infused to compensate for fluid losses. All animals received heparin 15 000 IU intravenously to prevent blood clotting. The surface of the heart was rinsed with sterile saline before left ventricular biopsies were obtained from the postischaemic region and from a control region supplied by the circumflex artery. The “through wall” biopsies (weighing approximately 1 g) were obtained by a sharpened stainless steel cylinder (12 mm internal diameter) attached to a mechanical drill and put into liquid nitrogen within 30 s. The myocardial biopsies were stored at -70°C until use. Animals were maintained and housed throughout under the conditions set by the Norwegian Council for Animal Research. Haemodynamic measurements Left ventricular pressure was measured by a microtip pressure transducer catheter (Model PC-470, Millar Instruments, Houston, TX, USA) introduced into the left ventricle via the right common carotid artery. The maximum positive value of left ventricular pressure (LVdP/dt) was used as the contractility index. Arterial blood pressure was measured by a Statham pressure transducer (Model P23 Gb, Gould Instruments, Hato Rey, Puerto Rico) connected to a polyethylene catheter inserted into a femoral or carotid artery. Aortic flow was recorded by an electromagnetic flowprobe snugly fitting the ascending aorta and connected to a square wave electromagnetic flowmeter (Nycotron, Mode 376, Drammen, Norway). Stroke volume was calculated by electronic integration of the aortic flow signal. Segment lengths were recorded by ultrasonic technique (Bugge-Asperheim-69). Two pairs of piezoelectric crystals were sewn into the mid-myocardium in each region, one element emitting and the other receiving ultrasound signals. Regional contractile function was described by the variable “percent systolic shortening”, which was defined as: percent systolic shortening = (end diastolic segment length - end systolic segment length) X 1OO/end diastolic segment length. Haemodynamic variables were continuously recorded on an eight channel galvanometric recorder (Mode 7758B, Hewlett-Packard). All haemodynamic measurements were made at end expiration and mean values for haemodynamic variables from four to six consecutive beats were calculated by the computer. The output of the recorder was sampled at 100 Hz and the signals transformed by an analogue to digital converter and stored on floppy discs for further processing.
The details of the computer sampling system have been described previo~sly.’~ Experimental design Myocardial stunning was induced by two different protocols: in four pigs, the distal left anterior descending coronary artery was occluded for 2, 2, 5 , 10, and 2 min; and in the remaining four pigs the artery was occluded for 2, 10, and 2 min, with a reperfusion period of 30 min between each ischaemic period. As a similar degree of contractile dysfunction was induced by the two protocols, data were pooled in the statistical analysis. Biopsies from the two regions of the left ventricles were obtained 1 h after the last coronary artery occlusion. The non-specific p blocker timolol (Blocadrenm, 0.1 mgakg-’) was given intravenously to all animals 20 min before the first coronary occlusion, in order to reduce the incidence of arrhythmias during the ischaemia-reperfusion procedure. Morphological identification In order to achieve reasonable fixation and prevent fixation artefact, a part (0.1 cm block) of each biopsy from both regions was fixed overnight in 100 mM cacodylate buffer (pH 7.4) containing 2% glutaraldehyde and 2% paraformaldehyde. Postfixation was done in 2% osmium tetroxide for 2 h. The samples were dehydrated with graded ethanol and infiltrated in epoxy resin with DMP-30. The embedded samples were sectioned and examined by electron microscopy (Transmissionmicroscopy, Philips 400, The Netherlands). Superoxide dismutase, glutathione peroxidase, and malondialdehyde Another part of each biopsy from both regions without macroscopic adipose or connective tissues was minced and homogenised in buffer (pH 7.5): 20 mM Tris.HC1, 250 mM sucrose, 5 mM MgCh, to determine the activities of superoxide dismutase and glutathione peroxidase, as well as the content of malondialdehyde. Superoxide dismutase activity was measured according to the method of Winterbourn et al. IS Glutathione peroxidase activity was determined by measuring oxidised glutathione as previously described.16The content of malondialdehy? was measured according to the method of Ohkawa et al. The p adrenoceptor-Gs-adenylyl cyclase system Membrane preparation - The above mentioned homogenate was also used to obtain a membrane preparation. After the nuclei had been removed by centrifugation at 900 g for 10 min at 4°C the supernatant fraction was spun at 43 000 g for 10 min at 4°C. The resulting pellet (membrane particulate fraction) was resuspended in 600 mM KCl in the same buffer as mentioned above. It was placed on ice for 10 min, recentrifuged at 43 000 g for 20 min and washed twice. The protein content was measured according to Lowry.” Aliquots of the membrane fractions were stored at -70°C before binding studies and adenylyl cyclase assays were performed. p Adrenoceptor radioligand bindint - p Adrenoceptor density was determined by use of (I2 I) iodocyanopindolol (ICYP). Binding assays were carried out in a buffer (pH 7.4): 25 mM Tris-75mM MgClz. An aliquot of the membrane preparation was incubated for 2 h in a total volume of 200 pl at 30°C with 100 pM ICYP and different concentrations of isoprenaline (100 pM-100 FM, 17 points). The reaction was terminated by dilution with 5 ml of ice cold buffer. The
Free radicals and /3 adrenergic system in myocardial stunning
samples were then poured over GF/C 2.5 cm glass microfibre filters (Whatman International Ltd, England) under reduced pressure followed by a wash with 15 ml of buffer. Filters were counted in a 1282 Compugamma universal counter (LKB, Sweden). Saturation binding isotherms were obtained by incubating membranes for 2 h with varying concentrations of ICYP (12.5 pM-400 pM), in the same buffer as above. Specific binding was defined as that displaced by 2 p,M 1-propanolol.The binding parameters Bmax and KD were determined by non-linear least squares fitting." Adenylyl cyclase assay - Adenylyl cyclase activity was measured according to the method of Salomon et al." Maximal adenylyl cyclase activity was determined in the presence of 50 p,M isoprenaline, 100 p,M guanosine 5 ' 4 3 , y-imino) triphosphate [Gpp(NH)p], and 100 p,M forskolin. Cyclic AMP was determined by the use of a cyclic AMP (3H) assay kit (Amersham, UK). Peptide synthesis and antibody purijication - The carboxyl terminal decapeptide RMHLRQYELL of Gs-a peptide was synthestised by the solid phase method of Merrifield" using an applied Biosystems 430 A automated peptide synthesiser (Foster City, CA, USA). Conjugation of peptide to bovine serum albumin was performed by using N-succinimidyl bromoacetate.22After immunisation of New Zealand white rabbits, the immunoglobin fractions were prepared from sera by precipitation in 50% (NH4)zS04 and then loaded on a Sepharose 4B CNBr activated substrate (Pharmacia, Sweden) to which the Gs-a peptide was covalently linked.23After washing the immunosorbent with phosphate buffered saline (PBS: 10 mM phosphate, 140 mM NaCl, pH 7.4), the specific anti-Gs-ar peptide antibody was eluted with 200 mM glycine (pH 2.8), neutralised in 1 M Tris (pH 8.0), and extensively dialysed against PBS. As a positive control, rabbit affinity-purified antibody of Gs-a was used.24 ELISA immunoassay - The membrane fraction was extracted with 1% cholate, while stirred on ice for 1 h, followed by centrifugation at 100 000 g for 1 h at 4°C. The clear supernatant (extract) was collected between a turbid floating layer and a loose pellet. Gs-a peptide (5 p,g.ml-l) in 100 mM NazCO3 solution (pH=l 1) was coated for 1 h on NUNC (Kanstrup, Denmark) microtitre plates. The wells were then saturated with PBS supplemented with 3% (wdvol) of skimmed milk, 0.1% (vol/ vol) of Tween 20 (E Merck, Darmstadt, Germany), and 0.01% (wdvol) of Thimerosal (Sigma, St Louis, MO, USA) (PMT). In the meantime, specific Gs antipeptide antibody was allowed to react with the extracts of the membrane fraction (50 p,g) for 1 h. These reaction mixtures were then transferred to the saturated microtitre plates for 1 h. After washing the wells three times with PMT, an affinity-purified biotinylated donkey antirabbit IgG antibody solution diluted 1:lOOO in PMT was allowed to react for 1 h. After three more washings, the bound biotinylated antibody was detected by incubation of the plates for 1 h with streptavidin peroxidase (1 Fgaml-') (Sigma) solution in PMT. This was followed by three washings in PBS and addition of substrate HzOz (2.5 mM)-2,2'-azino-di-(ethyl-benzthiazoline) sulphonic acid (2 mM) (ABTS, Sigma). Optical densities were read immediately kinetically at 405 nm in a kinetic microplate reader (Molecular Devices, USA). Determination of mRNA encoding for Gs-a - To determine mRNA "encoding for Gs&, " total RNA was extracted from other parts of the biopsies by the uanidinum thiocyanate-caesium chloride (CsCl) method! Plasmids
45 1
(with insert lengths) containing rat cDNA encoding for the Gs-az6 as well as a plasmid containing a 911-b insert encoding frog rRNA were radiolabelled with (a-3P)dCTP (Dupont, 3000 ci-mmol-I). Northern blot analysis was performed as previously described." The relative amount of mRNA was determined by dot-blot analysis. An equal quantity of total RNA (40 p,) from the tissue of each biopsy was denatured and serially diluted before application to hybridisation membranes (Amersham, UK) mounted on a 96 well manifold (Schleicher and Schuell, USA). The membranes were hybridised with radiolabelled probes. Calculation of the relative level of specific mRNA was based upon the assumption that each hybridised probe molecule was specifically bound to the appropriate mRNA. The specific radioactivity of the probe was calculated from scintillation counting of radioactivity and spectrophometry of its molarity. All probes used were stable as radioactivity decay followed the expected half life of (w3'P)dCTP.
zp
Drugs and materials Plasmids containing rat cDNA encoding a subunit of Gs was kindly provided by R R Reed, Johns Hopkins University School of Medicine, Baltimore, USA. The rabbit affinity purified Gs-a antibody was a generous gift from Dr K Sficher, Institute of Pharmacology, Free University, Berlin. ( I '1)ICYP (100 pcisp,l-'), (3H) cyclic AMP (1.0 mCi.ml-l), and ( c Y - ~ ~ P ) ~(3000 C T P Ciaml-l) were obtained from New England Nuclear. All other materials were of reagent grade and were obtained from Sigma Chemical Co, St Louis, MO. USA. Statistics Student's t test was used for statistics. A probability value of pe0.05 was regarded as statistically significant when data were normally distributed. Data are presented as mean(SEM).
Results Segment length measurements and haemodynamic data Percent systolic shortening in the region subjected to ischaemia and reperfusion was reduced to 58(SEM 9)% at the time the biopsies were obtained. Systolic shortening in the control region supplied by the circumflex artery was moderately reduced, to 92(7)% of preocclusion value. Left ventricular pressure, left ventricular dP/dt, arterial pressure, stroke volume, and heart rate remained unchanged by the occlusion and reperfusion procedures (table I). Ultrastructural studies Electron microscopy revealed potentially reversible ischaemic changes such as wide I bands, slight intracellular Table I Haemodynamic measurements before development of stunning. Values are means(SEM).
and
after
~~
Before stunning
LVSP (mm Hg) LVdP/dt (mm.s-') AP (mm Hg) HR (beatsemin-I) SV (ml) SS control region (%) SS stunning region (%)
105( 13) 1250(212) 86( 12) 91(8) 17.3(6.0) 100 100
After stunning
104(17)
1013( 169) 88( 11)
96( 12) 16.0(4.3) 92(7)* 58(9)*
LVSP=left ventricular systolic pressure; LVdP/dt=maximum positive first derivative of LVP; AP=mean aortic pressure; HR=heart rate: SV=stroke volume; SS=percent systolic *pco.o1
45 2
Fu, llebekk, Kirkeben, Aksnes, Waagstein, Bergh, Hoebeke, Liung, Hjalmarson
oedema, and mild mitochondrial swelling in biopsies from stunned myocardium (fig 1). Signs of irreversible ischaemic injury such as amorphous densities in the matrix space of mitochondria, subsarcolemmal blebs, disrupted sarcolemmal membranes, or myocardial contraction bands2*2y were not found. Free radical scavengers and lipid peroxidation
In the biopsies from stunned myocardium, the activities of
superoxide dismutase and glutathione peroxidase were reduced to 66% and 52% of control values, respectively, whereas the content of malondialdehyde was increased above the control value by about 49% (table 11).
p Adrenergic binding
characteristics
Scatchard analysis revealed a single population of binding sites in stunned and control myocardium. No significant differences were observed in the receptor densities (Bmax)
Figure I Electron microscopic pictures of myocardial cells in control and stunned myocardium. In control myocardium the ultrastructure of the rnyocytes was essentially normal (panel A: magnijcation X 10 000).Myocytes .from the stunned region showed mild and reversible myocardial cell damage, such as wide I band, variable degree of mitochondria1 swelling, and intracellular-oedema (panel B: magn@cation
x 10 000)
Free radicals and
p udrenergic system in myocardial stunning
453
Table I1 Activities of superoxide dis1nuta.w (SOD) and glutathione peroxiduse (GSH-px) und content of malondialdehyde (MDA) in control and stunned myocardium. Values ure meuriJ(SEM).
Stunning region Control region
11
SOD"
GSH-~X'
MDA'
8 8
0.4Y(0.02)* 0.74(0.07)
0. I2(0.0 I )* 0.23(0.03)
I I20(34.75)* 752( 16.33)
Values are expressed as unit.mg-l protein. One unit is that amount of enzyme activity causing 50% inhibition of the superoxide dependent photolyt!c reduction of nitrobluetetrazolium Activities are expressed as absorbance 340min?.mg ' protein ' Content of MDA is expressed as prno1.rng-l protein which is calculated from the absorbance at S32 nM using tetramethoxypropanc as an external standard. *p
449
Oxygen free radical injury and Gs mediated signal transduction in the stunned porcine myocardium Liang-Xiong Fu, Amfinn Ilebekk, Knut Arvid Kirkeben, Gunnar Aksnes, Finn Waagstein, Claes-HAkan Bergh, Johan Hoebeke, Qi-Ming Liang, Ake Hjalmarson Objective: The aim was to investigate involvement of oxygen free radicals and any changes in the Gs mediated p adrenergic signalling system of stunned porcine myocardium. Methods: Myocardial stunning was induced in eight pentobarbitone anaesthetised pigs by brief occlusions of the distal left anterior descending coronary artery for periods of up to 10 min. Segment length function was measured in the ischaemic region and in a control region supplied by the circumflex artery. Left ventricular biopsies were obtained from the two regions 1 h after the last occlusion for ultrastructural and biochemical studies. Timolol was been used to prevent arrhythmia during ischaemia. Results: At the time when biopsies were obtained, percent systolic shortening was reduced to 58% in the region subjected to ischaemia and was only minimally reduced in the control region. In the biopsies from the stunned region: ( 1 ) electron microscopy showed mild and reversible intracellular changes in the stunned myocardium; (2) the activities of superoxide dismutase and glutathione peroxidase were decreased by 66% and 52%. respectively; (3) the content of malondialdehyde was increased by 49%; (4) neither density nor affinity of p adrenoceptors showed any changes; (5) there were no alterations in messenger RNA encoding for the (Y subunit of the stimulatory guanine necleotide binding protein (Gs), demonstrated by northern and dot-blot hybridisations; (6) ELISA technique utilising a specific antipeptide antibody showed no quantitative change in Gs; (7) the activity of adenyl cyclase was unchanged. Conclusions: Even though the stunned porcine myocardium showed substantial evidence of free radical injury, the p adrenergic signalling system was intact.
M
yocardial stunning is defined as a transient reversible myocardial dysfunction induced by brief periods of ischaemia, which do not lead to necrosis. This phenomenon is likely to be an important 'component in the dysfunction observed in patients after periods with unstable angina or coronary spasm and after therapeutic thrombolysis. During ischaemia and reperfusion production of oxygen free radicals has been demonstrated.' Free radical injury may cause lipid peroxidation of cellular membranes and thereby alter the physical pro erties of membranes and P membrane bound ion pumps? The reversible myocardial injury caused by short lasting ischaemia is in part caused Fy free radical production during ischaemia and reperfusion.The generation of free radicals associated with ischaemia and reperfu$on sequences varies inversely with collateral blood flow.- Since myocardial collateral blood flow in the pig heart is minimal,' this species should be optimal for the detection of oxygen free radical injury. The purpose of the present study was to determine if production of oxygen free radicals occurred during development of myocardial stunning in our preparation. We therefore measured the activities of the oxygen free radical scavengers superoxide dismutase and glutathione peroxidase and the content of the lipid peroxidation product, malondialdehyde. The f3 blocker timolol was used to prevent arrhythmia in the early period of ischaemia. The use of timolol does not exert any significant influence on the I
'
receptor-(; protein-adenylyl cyclase system in the present protocol since it has been shown that there is no change in the p adrenoce tor density after timolol treatment in human myocardium. 1 2 The importance of the myocardial p adrenergic signalling system has been investigated in early' and in prolonged periods of ischaemia.'&l' After long lasting ischaemia, which induces myocardial infarction, an increased density of I adrenoceptors has been shown in the canine myocardium. Lipid peroxidation has been shown to decrease the density of p adrenoceptors.' Since it has been suggested that free radical injury is an important factor in the development of myocardial stunning, the function of the p adrenoceptor signalling system in the reversibly injured myocardium could be affected. To our knowledge, this has not previously been investigated. A second objective of this study was therefore to determine the density of p adrenoceptors, as well as any alterations in the Gs protein and the adenylyl cyclase system in the stunned myocardium.
I-k
Methods Animul preparution
Eight domestic pigs of either six weighing 19-27 kg (range) were anaesthetised with pentobarbitone sodium intraperitoneally (30 mg.kg-'), followed by a continuous intravenous sustaining dose of 5-20 mg.kg-'.h-l according to depth of anaesthesia. The pigs were tracheostomised and
Wallenberg Laboratory, Sahlgren's Hospital, University of Goteborg, 41 3 45 Goteborg, Sweden: L-X Fu, F Waagstein, C-H Bergh, Q-M Liang, p\ Hjalmarson; Institute for Experimental Medical Research, Ullevaal Hospital, Oslo, Norway: A Ilebekk, K A Kirkeben, G Aksnes; Laboratoire des Proteines des Liquides Biologiques, URA 1334, CNRS, Tours, France: J Hoebeke. Correspondence to Dr Fu.
450
Fu, Ilebekk, Kirkeben, Aksnes, Waagstein, Bergh, Hoebeke, Liang, Hjalmarsnn
artificially ventilated with air and oxygen by a volume regulated ventilator (Princeton Medical Instruments, Natick, MA, USA). Arterial blood gases were regularly checked by an automatic blood gas analyser (AVL Biomedical Instruments, Model 945, Graz, Austria). POZwas kept above 30 kPa in all animals, Pcoz at 6.4(SEM 1.1) and pH at 7.42(0.05). Body temperature was kept at 37.6(0.7)”C by wrappings and electric heating pads. Urine was continuously drained through a cystostoma. The heart was exposed through a midsternal split and suspended in a pericardial cradle. A segment 5-7 cm distal on the left anterior descending coronary artery was dissected free for later intermittent occlusions with a Mayfield clip. Segment lengths were continuously recorded by ultrasonic technique in the ischaemic region and in a control region supplied by the circumflex coronary artery. A peripheral artery was cannulated for blood pressure recordings and for sampling arterial blood. A femoral vein was cannulated and saline infused to compensate for fluid losses. All animals received heparin 15 000 IU intravenously to prevent blood clotting. The surface of the heart was rinsed with sterile saline before left ventricular biopsies were obtained from the postischaemic region and from a control region supplied by the circumflex artery. The “through wall” biopsies (weighing approximately 1 g) were obtained by a sharpened stainless steel cylinder (12 mm internal diameter) attached to a mechanical drill and put into liquid nitrogen within 30 s. The myocardial biopsies were stored at -70°C until use. Animals were maintained and housed throughout under the conditions set by the Norwegian Council for Animal Research. Haemodynamic measurements Left ventricular pressure was measured by a microtip pressure transducer catheter (Model PC-470, Millar Instruments, Houston, TX, USA) introduced into the left ventricle via the right common carotid artery. The maximum positive value of left ventricular pressure (LVdP/dt) was used as the contractility index. Arterial blood pressure was measured by a Statham pressure transducer (Model P23 Gb, Gould Instruments, Hato Rey, Puerto Rico) connected to a polyethylene catheter inserted into a femoral or carotid artery. Aortic flow was recorded by an electromagnetic flowprobe snugly fitting the ascending aorta and connected to a square wave electromagnetic flowmeter (Nycotron, Mode 376, Drammen, Norway). Stroke volume was calculated by electronic integration of the aortic flow signal. Segment lengths were recorded by ultrasonic technique (Bugge-Asperheim-69). Two pairs of piezoelectric crystals were sewn into the mid-myocardium in each region, one element emitting and the other receiving ultrasound signals. Regional contractile function was described by the variable “percent systolic shortening”, which was defined as: percent systolic shortening = (end diastolic segment length - end systolic segment length) X 1OO/end diastolic segment length. Haemodynamic variables were continuously recorded on an eight channel galvanometric recorder (Mode 7758B, Hewlett-Packard). All haemodynamic measurements were made at end expiration and mean values for haemodynamic variables from four to six consecutive beats were calculated by the computer. The output of the recorder was sampled at 100 Hz and the signals transformed by an analogue to digital converter and stored on floppy discs for further processing.
The details of the computer sampling system have been described previo~sly.’~ Experimental design Myocardial stunning was induced by two different protocols: in four pigs, the distal left anterior descending coronary artery was occluded for 2, 2, 5 , 10, and 2 min; and in the remaining four pigs the artery was occluded for 2, 10, and 2 min, with a reperfusion period of 30 min between each ischaemic period. As a similar degree of contractile dysfunction was induced by the two protocols, data were pooled in the statistical analysis. Biopsies from the two regions of the left ventricles were obtained 1 h after the last coronary artery occlusion. The non-specific p blocker timolol (Blocadrenm, 0.1 mgakg-’) was given intravenously to all animals 20 min before the first coronary occlusion, in order to reduce the incidence of arrhythmias during the ischaemia-reperfusion procedure. Morphological identification In order to achieve reasonable fixation and prevent fixation artefact, a part (0.1 cm block) of each biopsy from both regions was fixed overnight in 100 mM cacodylate buffer (pH 7.4) containing 2% glutaraldehyde and 2% paraformaldehyde. Postfixation was done in 2% osmium tetroxide for 2 h. The samples were dehydrated with graded ethanol and infiltrated in epoxy resin with DMP-30. The embedded samples were sectioned and examined by electron microscopy (Transmissionmicroscopy, Philips 400, The Netherlands). Superoxide dismutase, glutathione peroxidase, and malondialdehyde Another part of each biopsy from both regions without macroscopic adipose or connective tissues was minced and homogenised in buffer (pH 7.5): 20 mM Tris.HC1, 250 mM sucrose, 5 mM MgCh, to determine the activities of superoxide dismutase and glutathione peroxidase, as well as the content of malondialdehyde. Superoxide dismutase activity was measured according to the method of Winterbourn et al. IS Glutathione peroxidase activity was determined by measuring oxidised glutathione as previously described.16The content of malondialdehy? was measured according to the method of Ohkawa et al. The p adrenoceptor-Gs-adenylyl cyclase system Membrane preparation - The above mentioned homogenate was also used to obtain a membrane preparation. After the nuclei had been removed by centrifugation at 900 g for 10 min at 4°C the supernatant fraction was spun at 43 000 g for 10 min at 4°C. The resulting pellet (membrane particulate fraction) was resuspended in 600 mM KCl in the same buffer as mentioned above. It was placed on ice for 10 min, recentrifuged at 43 000 g for 20 min and washed twice. The protein content was measured according to Lowry.” Aliquots of the membrane fractions were stored at -70°C before binding studies and adenylyl cyclase assays were performed. p Adrenoceptor radioligand bindint - p Adrenoceptor density was determined by use of (I2 I) iodocyanopindolol (ICYP). Binding assays were carried out in a buffer (pH 7.4): 25 mM Tris-75mM MgClz. An aliquot of the membrane preparation was incubated for 2 h in a total volume of 200 pl at 30°C with 100 pM ICYP and different concentrations of isoprenaline (100 pM-100 FM, 17 points). The reaction was terminated by dilution with 5 ml of ice cold buffer. The
Free radicals and /3 adrenergic system in myocardial stunning
samples were then poured over GF/C 2.5 cm glass microfibre filters (Whatman International Ltd, England) under reduced pressure followed by a wash with 15 ml of buffer. Filters were counted in a 1282 Compugamma universal counter (LKB, Sweden). Saturation binding isotherms were obtained by incubating membranes for 2 h with varying concentrations of ICYP (12.5 pM-400 pM), in the same buffer as above. Specific binding was defined as that displaced by 2 p,M 1-propanolol.The binding parameters Bmax and KD were determined by non-linear least squares fitting." Adenylyl cyclase assay - Adenylyl cyclase activity was measured according to the method of Salomon et al." Maximal adenylyl cyclase activity was determined in the presence of 50 p,M isoprenaline, 100 p,M guanosine 5 ' 4 3 , y-imino) triphosphate [Gpp(NH)p], and 100 p,M forskolin. Cyclic AMP was determined by the use of a cyclic AMP (3H) assay kit (Amersham, UK). Peptide synthesis and antibody purijication - The carboxyl terminal decapeptide RMHLRQYELL of Gs-a peptide was synthestised by the solid phase method of Merrifield" using an applied Biosystems 430 A automated peptide synthesiser (Foster City, CA, USA). Conjugation of peptide to bovine serum albumin was performed by using N-succinimidyl bromoacetate.22After immunisation of New Zealand white rabbits, the immunoglobin fractions were prepared from sera by precipitation in 50% (NH4)zS04 and then loaded on a Sepharose 4B CNBr activated substrate (Pharmacia, Sweden) to which the Gs-a peptide was covalently linked.23After washing the immunosorbent with phosphate buffered saline (PBS: 10 mM phosphate, 140 mM NaCl, pH 7.4), the specific anti-Gs-ar peptide antibody was eluted with 200 mM glycine (pH 2.8), neutralised in 1 M Tris (pH 8.0), and extensively dialysed against PBS. As a positive control, rabbit affinity-purified antibody of Gs-a was used.24 ELISA immunoassay - The membrane fraction was extracted with 1% cholate, while stirred on ice for 1 h, followed by centrifugation at 100 000 g for 1 h at 4°C. The clear supernatant (extract) was collected between a turbid floating layer and a loose pellet. Gs-a peptide (5 p,g.ml-l) in 100 mM NazCO3 solution (pH=l 1) was coated for 1 h on NUNC (Kanstrup, Denmark) microtitre plates. The wells were then saturated with PBS supplemented with 3% (wdvol) of skimmed milk, 0.1% (vol/ vol) of Tween 20 (E Merck, Darmstadt, Germany), and 0.01% (wdvol) of Thimerosal (Sigma, St Louis, MO, USA) (PMT). In the meantime, specific Gs antipeptide antibody was allowed to react with the extracts of the membrane fraction (50 p,g) for 1 h. These reaction mixtures were then transferred to the saturated microtitre plates for 1 h. After washing the wells three times with PMT, an affinity-purified biotinylated donkey antirabbit IgG antibody solution diluted 1:lOOO in PMT was allowed to react for 1 h. After three more washings, the bound biotinylated antibody was detected by incubation of the plates for 1 h with streptavidin peroxidase (1 Fgaml-') (Sigma) solution in PMT. This was followed by three washings in PBS and addition of substrate HzOz (2.5 mM)-2,2'-azino-di-(ethyl-benzthiazoline) sulphonic acid (2 mM) (ABTS, Sigma). Optical densities were read immediately kinetically at 405 nm in a kinetic microplate reader (Molecular Devices, USA). Determination of mRNA encoding for Gs-a - To determine mRNA "encoding for Gs&, " total RNA was extracted from other parts of the biopsies by the uanidinum thiocyanate-caesium chloride (CsCl) method! Plasmids
45 1
(with insert lengths) containing rat cDNA encoding for the Gs-az6 as well as a plasmid containing a 911-b insert encoding frog rRNA were radiolabelled with (a-3P)dCTP (Dupont, 3000 ci-mmol-I). Northern blot analysis was performed as previously described." The relative amount of mRNA was determined by dot-blot analysis. An equal quantity of total RNA (40 p,) from the tissue of each biopsy was denatured and serially diluted before application to hybridisation membranes (Amersham, UK) mounted on a 96 well manifold (Schleicher and Schuell, USA). The membranes were hybridised with radiolabelled probes. Calculation of the relative level of specific mRNA was based upon the assumption that each hybridised probe molecule was specifically bound to the appropriate mRNA. The specific radioactivity of the probe was calculated from scintillation counting of radioactivity and spectrophometry of its molarity. All probes used were stable as radioactivity decay followed the expected half life of (w3'P)dCTP.
zp
Drugs and materials Plasmids containing rat cDNA encoding a subunit of Gs was kindly provided by R R Reed, Johns Hopkins University School of Medicine, Baltimore, USA. The rabbit affinity purified Gs-a antibody was a generous gift from Dr K Sficher, Institute of Pharmacology, Free University, Berlin. ( I '1)ICYP (100 pcisp,l-'), (3H) cyclic AMP (1.0 mCi.ml-l), and ( c Y - ~ ~ P ) ~(3000 C T P Ciaml-l) were obtained from New England Nuclear. All other materials were of reagent grade and were obtained from Sigma Chemical Co, St Louis, MO. USA. Statistics Student's t test was used for statistics. A probability value of pe0.05 was regarded as statistically significant when data were normally distributed. Data are presented as mean(SEM).
Results Segment length measurements and haemodynamic data Percent systolic shortening in the region subjected to ischaemia and reperfusion was reduced to 58(SEM 9)% at the time the biopsies were obtained. Systolic shortening in the control region supplied by the circumflex artery was moderately reduced, to 92(7)% of preocclusion value. Left ventricular pressure, left ventricular dP/dt, arterial pressure, stroke volume, and heart rate remained unchanged by the occlusion and reperfusion procedures (table I). Ultrastructural studies Electron microscopy revealed potentially reversible ischaemic changes such as wide I bands, slight intracellular Table I Haemodynamic measurements before development of stunning. Values are means(SEM).
and
after
~~
Before stunning
LVSP (mm Hg) LVdP/dt (mm.s-') AP (mm Hg) HR (beatsemin-I) SV (ml) SS control region (%) SS stunning region (%)
105( 13) 1250(212) 86( 12) 91(8) 17.3(6.0) 100 100
After stunning
104(17)
1013( 169) 88( 11)
96( 12) 16.0(4.3) 92(7)* 58(9)*
LVSP=left ventricular systolic pressure; LVdP/dt=maximum positive first derivative of LVP; AP=mean aortic pressure; HR=heart rate: SV=stroke volume; SS=percent systolic *pco.o1
45 2
Fu, llebekk, Kirkeben, Aksnes, Waagstein, Bergh, Hoebeke, Liung, Hjalmarson
oedema, and mild mitochondrial swelling in biopsies from stunned myocardium (fig 1). Signs of irreversible ischaemic injury such as amorphous densities in the matrix space of mitochondria, subsarcolemmal blebs, disrupted sarcolemmal membranes, or myocardial contraction bands2*2y were not found. Free radical scavengers and lipid peroxidation
In the biopsies from stunned myocardium, the activities of
superoxide dismutase and glutathione peroxidase were reduced to 66% and 52% of control values, respectively, whereas the content of malondialdehyde was increased above the control value by about 49% (table 11).
p Adrenergic binding
characteristics
Scatchard analysis revealed a single population of binding sites in stunned and control myocardium. No significant differences were observed in the receptor densities (Bmax)
Figure I Electron microscopic pictures of myocardial cells in control and stunned myocardium. In control myocardium the ultrastructure of the rnyocytes was essentially normal (panel A: magnijcation X 10 000).Myocytes .from the stunned region showed mild and reversible myocardial cell damage, such as wide I band, variable degree of mitochondria1 swelling, and intracellular-oedema (panel B: magn@cation
x 10 000)
Free radicals and
p udrenergic system in myocardial stunning
453
Table I1 Activities of superoxide dis1nuta.w (SOD) and glutathione peroxiduse (GSH-px) und content of malondialdehyde (MDA) in control and stunned myocardium. Values ure meuriJ(SEM).
Stunning region Control region
11
SOD"
GSH-~X'
MDA'
8 8
0.4Y(0.02)* 0.74(0.07)
0. I2(0.0 I )* 0.23(0.03)
I I20(34.75)* 752( 16.33)
Values are expressed as unit.mg-l protein. One unit is that amount of enzyme activity causing 50% inhibition of the superoxide dependent photolyt!c reduction of nitrobluetetrazolium Activities are expressed as absorbance 340min?.mg ' protein ' Content of MDA is expressed as prno1.rng-l protein which is calculated from the absorbance at S32 nM using tetramethoxypropanc as an external standard. *p
