- Email: [email protected]
Effect of Treatment on Ventricular Function and Troponin I Proteolysis in Reperfused Myocardium
J Mol Cell Cardiol 34, 401–411 (2002) doi:10.1006/jmcc.2002.1522, available online at http://www.idealibrary.com on
Effect of Treatment on Ventricular Function and Troponin I Proteolysis in Reperfused Myocardium Ananth M. Prasan1, Hugh C. K. McCarron1, Brett D. Hambly2, Gary G. Fermanis3, David R. Sullivan4 and Richmond W. Jeremy1 Departments of 1Medicine and 2Pathology, University of Sydney, Sydney, 2006, NSW, Australia, 3 Department of Cardiothoracic Surgery, St. George Hospital, Kogarah, 2217 and 4Department of Clinical Biochemistry, Royal Prince Alfred Hospital, Camperdown 2050, Australia (Received 6 August 2001, accepted for publication 4 January 2002) A. M. P, H. C. K. MC, B. D. H, G. G. F, D. R. S R. W. J. Effect of Treatment on Ventricular Function and Troponin I Proteolysis in Reperfused Myocardium. Journal of Molecular and Cellular Cardiology (2002) 34, 401–411. Effects of ischemia time and treatment interventions upon troponin I (TnI) proteolysis and function of reperfused myocardium were examined in isolated, perfused rabbit hearts. Hearts were randomized to 90 min aerobic perfusion, 15 min low-flow (1 ml/min) ischemia (I) and 60 min reperfusion (R) or 60 min low-flow I and 60 min R. Hearts subject to 60 min I and 60 min R received either no treatment, -arginine treatment, or treatment with oxygen free radical (OFR) scavengers (mercapto-proponylglycine, catalase and superoxide dismutase). Hearts from cholesterol-fed rabbits were also studied after 60 min I and R. Isovolumic LV pressure and heart rate were recorded throughout and Western analysis of ventricular myocardium, using 3 specific antibodies, detected intact TnI (29 kDa) and TnI fragment (25 kDa). Hearts subject to 15 min I had minimal irreversible injury (TTC negative region=0.6±0.4% LV) but hearts subject to 60 min I had more extensive injury (TTC negative=40.7±5.8% LV). Recovery of rate–pressure product after 15 min I and 60 min R (56±9% of baseline) was better than after 60 min I and 60 min R (23±9%, P<0.01). Both arginine and OFR scavengers were associated with better recovery of function after 60 min I, (66±7% and 72±3% of baseline respectively, P<0.01 v no treatment) but cholesterol hearts had poor recovery after 60 min I (37±8%). The 25 kDa TnI (% total TnI immunoreactivity) was 8.7±0.9% in controls, 10.0±1.6% after 15 min I and 60 min R, and 17.4±2.4% after 60 min I and 60 min R (P<0.01 v controls and 15 min I). The proportion of 25 kDa TnI was increased in all hearts after 60 min I and did not change with treatment (-arginine 16.8±1.8%, OFR scavengers 16.0±3.2%, cholesterol 14.0±1.9%). There was no relation between proportion of 25 kDa TnI and recovery of function. Samples from freshly excised rabbit hearts and human right atria also had 25 kDa TnI (relative intensities 8.5±2.3% and 5.1±2.6% respectively). Although TnI fragmentation increases after prolonged ischemia and reperfusion, the functional recovery of stunned myocardium is independent of degree of TnI fragmentation. 2002 Elsevier Science Ltd. All rights reserved. K W: Myocardium, ischemia and reperfusion; Troponin I; Humans; Rabbits; Proteolysis.
Introduction The pathogenesis of persistent contractile dysfunction (stunning) of post-ischemic myocardium has been the subject of intense investigation. We know that intracellular calcium overload and oxygen free radicals contribute and that stunned
myocardium exhibits reduced Ca2+ sensitivity.1,2 Delayed recovery of function over several days and abnormal responses to Ca2+ suggest that injury to critical myofilament proteins occurs during ischemia and reperfusion.3 Evidence in support of this hypothesis has been presented by Gao et al.4 and by McDonough et al.,5 who observed apparent
Please address all correspondence to: Prof. Richmond W. Jeremy, Room 383, Blackburn Building, Department of Medicine, University of Sydney 2006, New South Wales, Australia. Tel: (612) 95157608. Fax: (612) 95506262. E-mail: [email protected]
0022–2828/02/040401+11 $35.00/0
2002 Elsevier Science Ltd. All rights reserved.
402
A. M. Prasan et al.
degradation of troponin I (TnI) in post-ischemic rat hearts. It has been hypothesized that Ca2+dependent activation of the cysteine protease calpain causes proteolytic cleavage of TnI near the C terminus, leaving a dysfunctional fragment (TnI1-193). The data about TnI fragmentation in stunned myocardium are not, however, uniform. Other groups have failed to find evidence of TnI fragmentation6–9 or have only observed a fragment in association with irreversible myocardial injury.8 There are also conflicting data about the functional consequences of C-terminus truncation of TnI. Theoretical considerations suggest that such truncation would increase Ca2+ sensitivity of the myofilaments. Indeed, studies of cTnI mutants, with C terminus truncations similar to those proposed for stunned myocardium, have shown that inhibition of actin-myosin interaction by the troponin complex is lessened, so that Ca2+ sensitivity of the myofilaments is actually increased.10 This is the opposite of observations in stunned myocardium. In contrast, transgenic mice with overexpression of TnI1-193 have been reported to have impairment of myocardial contractility and reduced Ca2+ sensitivity.11,12 Importantly, if TnI fragmentation is a fundamental pathogenic feature of stunned myocardium, then therapeutic interventions, which ameliorate the severity of post-ischemic dysfunction, should be associated with less TnI fragmentation. This study therefore compared the degree of TnI fragmentation with contractile function after short and long ischemic periods, and examined the effects of treatment interventions upon TnI fragmentation and myocardial function.
Materials and Methods Experimental studies were performed using male adult New Zealand White rabbits (2.5 to 3.5 kg), according to protocols approved by the Animal Research Ethics Committee of the University of Sydney.
Ischemia–reperfusion studies in isolated hearts After administration of heparin (2500 units i.v.), rabbits were killed with sodium pentobarbital (150 mg/kg i.v.) and hearts rapidly excised for Langendorff perfusion at a pressure of 80 mmHg with modified Krebs–Henseleit buffer containing 16 m glucose (37°C, pH 7.4, bubbled with 95% O2 and 5% CO2). Hearts were perfused within a
temperature-controlled organ-bath. A latex balloon was introduced into the left ventricle (LV) across the mitral valve for pressure measurement (P23Db, Gould Inc, CA, USA). The balloon was inflated with physiological saline solution (37°C) to obtain a LV end-diastolic pressure of 8–10 mmHg under baseline conditions. Heart rate and isovolumic ventricular pressures were recorded continuously with a digital acquisition system (MP100WSW, Biopac Systems, CA, USA). In hearts subject to low-flow ischemia, perfusion was by peristaltic pump (Harvard Instruments Model 66, Harvard Apparatus Inc., Holliston, MO, USA) at a rate of 1 ml/min. During low-flow ischemia, the LV balloon was deflated and during reperfusion the balloon was refilled with the same volume of saline as used during baseline measurements. Before each study, hearts were randomly assigned by ballot to one of the experimental protocols. If hearts developed ventricular fibrillation or rigor at the time of reperfusion, they were excluded from the subsequent data analysis. In the first series of experiments, the relation between duration of ischemia, myocardial necrosis and TnI fragmentation was examined. After an initial 15 min stabilization period, hearts were subject to either normal perfusion for 90 min without an ischemic period (n=8), or to 15 min low-flow ischemia (as a model of moderate post-ischemic dysfunction without necrosis) followed by 60 min free reperfusion (n=14) or to 60 min low-flow ischemia (as a model of severe post-ischemic dysfunction with necrosis) followed by 60 min reperfusion (n=14). The presence and extent of myocardial necrosis occurring after low-flow ischemia was quantified by horseradish peroxidase (HPR) staining14 and triphenyl-tetrazolium chloride (TTC) staining15 and by measurement of creatine kinase (CK) and cardiac troponin T (cTnT) in coronary effluent. Horseradish peroxidase (Type II, Sigma, St. Louis, MO, USA, 10 mg) was added to perfusate for the last 5 min of perfusion for each protocol. Two mid-left ventricular samples (2 mm thick) were then cut before the remainder of the ventricle was frozen in liquid nitrogen for protein studies. After fixation (4% paraformaldehyde in 0.1 sodium cacodylate buffer, pH 7.4, for 4 h at 4°C before washing in 0.1 sodium cacodylate buffer and incubation in buffer solution overnight), frozen sections (60 m thick) were cut from the samples. Sections were washed with 0.1 cacodylate buffer and incubated for 10 min at room temperature in 100 ml of 0.1 Tris-HCl buffer (pH 7.6), containing 100 mg 3,3′diaminobenzidine tetrahydrochloride (Sigma) and
Troponin I in Stunned Myocardium
33.3 l of 30% hydrogen peroxide. Sections were examined by light microscopy to detect abnormal staining, indicative of irreversible injury. Twenty medium power fields (×20) for each heart were examined by two blinded observers and the percentage of positively stained cells per field was determined. For TTC staining, four hearts from each group were immediately chilled to 4°C and cut into five 2-mm-thick transverse slices. Slices were then incubated at 37°C for 20 min in 1% TTC (0.1 phosphate buffer, pH 7.4) prior to fixing in 15% formalin. Each slice was digitally photographed and the area of necrosis determined by computer planimetry (Image-Pro Plus, Media Cybernetics, Silver Spring, Maryland, USA). For the measurement of CK and cTnT the pulmonary artery was cannulated and coronary effluent collected into chilled vials at timed intervals (pre-ischemia, 1, 5 and 30 min reperfusion). The CK was measured by enzymatic assay (Roche/Hitachi 917 clinical chemistry analyser) and cTnT was quantified by immunoassay (Roche Elecsys 2010 immunoassay analyser). In the second series of experiments, the effects of interventions associated with altered post-ischemic function2,13 were examined in hearts exposed to 60 min ischemia followed by 60 min reperfusion. In seven hearts, -arginine (200 ) was included in the perfusate throughout the experiment; in another six hearts OFR scavengers [mercapto-proponyl-glycine (3 m), superoxide dismutase (6 U/ ml) and catalase (6 U/ml)] were included in the perfusate throughout the experiment. The effect of hypercholesterolemia (which is associated with endothelial dysfunction and abnormal vascular NO synthesis) was examined in another ten hearts exposed to 60 min ischemia followed by 60 min reperfusion. These rabbits had been fed 0.5% dietary cholesterol for 16 weeks, before the ischemia and reperfusion studies, and had a mean total blood cholesterol level of 30±3 mmol/l at the time of study.
Normal myocardium in vivo Further studies of TnI fragmentation were performed on freshly harvested normal rabbit left ventricular myocardium and human right atrial myocardium. After anesthesia, as described above, the rabbit hearts were rapidly excised whole. These hearts (n=3) were immediately rinsed in cold physiological saline solution (4°C), frozen in liquid nitrogen and stored at −80°C for subsequent protein studies. Human samples were obtained from 3
403
patients, who were undergoing elective coronary artery bypass grafting, and who had given informed consent for tissue sampling. None of the patients had a history of atrial fibrillation. Samples from the right atrial appendage (150–300 mg wet weight) were excised at the time of atrial cannulation, prior to commencement of bypass and in the absence of ischemic preconditioning. These samples were immediately rinsed in cold saline solution and frozen in liquid nitrogen for subsequent analysis.
Protein studies The frozen myocardial samples were homogenized (4°C) in buffer solution containing imidazole-HCl 50 m, EGTA 5 m, dithiothreitol 1 m, pH 7.8. The buffer contained a cocktail of eight protease inhibitors: leupeptin 20 , phenylmethylsulphonyl fluoride 250 , Calpain I inhibitor 20 , AEBSF [4-(2-aminoethyl)-benzenesulfonylfluoride] 1 m, aprotinin 1 , bestatin 50 , E-64 [trans-epoxysuccinyl-leucylamido-(4-guanidino] butane] 15 , pepstatin 10 . Samples were centrifuged at 10 000 g for 10 min at 4°C and the pellet resuspended in the above solution, with addition of 0.5% Triton X-100 and -mercaptoethanol 10 m. Subsequently, 50 l of sample was diluted 2:1 in Tricine sample buffer and boiled for 5 min before electrophoresis. Protein determination in the homogenate was performed by bicinchoninic acid protein assay (Pierce Inc., Rockford, IL, USA). Total protein (40 g/lane) was separated by SDS-PAGE 16 (16.5% Tris-Tricine Ready Gels, Bio-Rad, Hercules, CA, USA). After separation, proteins were visualized by Coomassie blue or were transferred onto nitrocellulose membrane sheets (Bio-Rad) at 40 V for 45 min and 80 V for a further 30 min. Following transfer, the membrane was incubated with 4% bovine serum albumin in Tris buffered saline (TBS, 50 m Tris-HCl, 150 m NaCl; pH 7.5) at 4°C overnight. Three monoclonal antibodies to TnI were used to detect TnI fragmentation; clones 4E5 and 1H8 (Chiron Diagnostics Corporation, East Walpole, MA, USA, epitopes aa 70-87) and clone 19C7 (Biodesign International, Kennebunk, Maine, USA, epitope aa 41-49). The membrane was incubated with TnI antibody at a dilution of 1:2000 for 60 min. Subsequently, membranes were rinsed in TBS containing 0.1% Tween 20 and incubated with alkaline phosphatase-conjugated anti-mouse IgG (Sigma) at a dilution of 1:5000 for 60 min. Immunoreactive bands were visualized with Western Blue Substrate
404
A. M. Prasan et al.
Table 1 Ischemia time and hemodynamics of perfused hearts Controls No ischemia
15 min Ischemia
60 min Ischemia
Baseline 163±9 109±5 7±1 102±5 16591±1303
Baseline 166±8 101±6 8±1 94±6 15626±1270
Heart rate (beats/min) LVs (mmHg) LVd (mmHg) LVdev (mmHg) RPP (mmHg.beats/min)
Baseline 159±4 102±3 9±1 94±3 14877±689
Heart rate (beats/min) LVs (mmHg) LVd (mmHg) LVdev (mmHg) RPP (mmHg.beats/min)
30 min 156±5 103±5 8±1 95±4 14840±702
5 min reflow 152±7 56±6∗∗ 6±1 50±6∗∗ 7791±984∗∗
5 min reflow 131±8 31±6¶¶ 9±2 22±6¶¶ 2802±810¶¶
Heart rate (beats/min) LVs (mmHg) LVd (mmHg) LVdev (mmHg) RPP (mmHg.beats/min)
90 min 146±6 104±6 10±1 94±5 13592±592
60 min reflow 166±9 67±5∗∗ 9±1 58±5∗∗ 9719±1189∗
60 min reflow 132±17 38±8¶ 14±5 23±7¶¶ 3365±1076¶¶
∗P<0.05, ∗∗P<0.01 v no ischemia, ¶P<0.05, ¶¶P<0.01 v 15 min ischemia. LVs, LVd, LVdev=left ventricular systolic, diastolic and developed pressures. Rate Pressure Product (RPP)=LVdev×heart rate. Baseline data were recorded after 10 min initial perfusion in each group.
(Promega, Madison, WI, USA) and intensity measured by video gel densitometry (GDS8000, UltraViolet Products, UVP Ltd, Cambridge, UK) and (Phoretix 1D Advanced software, Phoretix International, Newcastle upon Tyne, UK). Quantitation of TnI fragmentation (antibody 19C7) was performed on duplicate samples for each heart. Immunoblotting identified two TnI bands, at 29 kDa (intact TnI) and at 25 kDa (fragment TnI), similar to those previously reported. The relation between protein load per lane and intensity of immunoblot stain for TnI was linear between 15 and 75 g per lane and a standard protein loading of 40 g per lane was used for experimental protocols. Each band was measured as the proportion of total immunoreactivity per lane. The mean variance between duplicate measurements of TnI band intensity in samples from the same heart was 3.0±0.6%.
Statistics Group data are described as mean±standard error. The means of duplicate measurements for each heart were used for statistical analysis. Hemodynamic and densitometry data were compared between groups by analysis of variance with posthoc Newman–Keuls comparison of group means when significance was indicated (SPSS 10.0 SPSS Inc. Chicago, IL, USA). A P value <0.05 is described as statistically significant.
Results Ischemia duration and myocardial recovery Hemodynamic data are shown in Table 1 for hearts assigned to control perfusion with no ischemia, or 15 min or 60 min ischemia with reperfusion. Of the eight hearts assigned to control perfusion without any ischemia, all completed the protocol. Thirteen of the 14 hearts assigned to 15 min ischemia and 12 of the 14 hearts assigned to 60 min ischemia completed the protocol. Hemodynamics at the start of the experiment did not differ significantly between groups. At the end of 90 min normal perfusion, LVdev was maintained in non-ischemic controls (94±5 mmHg, NS v baseline). After 60 min reperfusion, LVdev was reduced to 58±5 mmHg in the 15 min ischemia group and to 23±7 mmHg in the 60 min ischemia group (both P<0.01 v controls). Left ventricular systolic function, indexed by ratepressure product was better preserved after 15 min ischemia (59±9% of baseline) than after 60 min ischemia (22±9% of baseline, P<0.01). The LV end-diastolic pressure was mildly increased after 60 min ischemia and reperfusion, although the difference from controls was not significant. In the non-ischemic controls, both cTnT and CK in coronary effluent were within normal limits throughout the perfusion period. In the 15 min ischemia hearts, the cTnT was at the limit of sensitivity of the assay (0.01 ng/ml) before ischemia,
Troponin I in Stunned Myocardium
405
Figure 1 Examples of Western blots for TnI from control, 15 min ischemia and 60 min reperfusion, and 60 min ischemia and 60 min reperfusion hearts. In each heart a band of approximately 29 kDa (intact TnI) and a band of lower molecular weight (approximately 25 kDa) are evident. Blots are shown for antibody 19C7 and similar results were obtained with antibodies 4E5 and 1H8. No other bands of higher or lower molecular weight were detected.
with a small increase to 0.07±0.01 ng/ml after 5 min reperfusion and CK did not increase above pre-ischemic levels during reperfusion in these hearts. In contrast, the 60 min ischemia hearts had a marked increase in cTnT to 0.75±0.27 ng/ml and in CK to 331±237 U/ml after 5 min reperfusion (both P<0.01 v controls and v 15 min ischemia). In non-ischemic controls, no infarct TTC staining was observed. In the 15 min ischemia hearts, a tiny amount of TTC negative myocardium was observed in only two hearts (0.6±0.4% of LV mass). Among the 60 min ischemia hearts, 40.7±5.8% of the LV mass was TTC negative, consistent with irreversible injury. The data obtained from horseradish peroxidase staining were similar. In non-ischemic control hearts only isolated cells were observed with weak positive staining. Similarly, in the 15 min ischemia hearts, less than 2% of cells examined showed positive staining. In the 60 min ischemia group, all hearts exhibited positive peroxidase staining, with a mean of 32±4% of cells examined showing staining indicative of irreversible injury.
Ischemia duration and TnI fragmentation The distribution of contractile proteins on SDS PAGE gels stained with Coomassie blue was indistinguishable between samples from control, 15 min and 60 min ischemia hearts. Western blotting identified both intact TnI (29 kDa) and an additional, less intense band of 25 kDa in hearts from each of the three groups (Fig. 1). Similar immunoreactive
Figure 2 Group data for intensity of 25 kDa TnI band as proportion of total TnI immunoreactivity in control perfused hearts (no ischemia), and hearts exposed to 15 min ischemia and 60 min reperfusion or 60 min ischemia and 60 min reperfusion.
bands were observed using each of the antibody clones 4E5, 1H8 and 19C7. In hearts from nonischemic controls and 15 min ischemia groups, the 25 kDa bands were of similar intensity but the intensity was significantly increased in hearts from the 60 min ischemia group (Fig. 2). The mean intensities of the 25 kDa TnI bands were 8.7±0.9% and 10.0±1.6% of total TnI immunoreactivity in control and 15 min ischemia groups respectively, but intensity was increased to 17.4±2.4% in 60 min ischemia hearts (P<0.01 v both controls and 15 min ischemia ). It was assumed in calculations that the affinity of the 19C7 monoclonal antibody for intact TnI and the 25 kDa fragment were the same, as the antibody epitope is far removed from the putative cleavage site at residue 193. No correlation was observed between the
406
A. M. Prasan et al.
Table 2 Treatment intervention and hemodynamics of perfused hearts after 60 min ischemia Controls
-arginine
OFR scavengers
Baseline Heart rate (beats/min) LVs (mmHg) LVd (mmHg) LVdev (mmHg) RPP (mmHg.beats/min)
166±8 101±6 8±1 94±6 15626±1270
195±16∗ 103±5 8±1 95±6 18630±1921
170±21 109±6 7±1 102±6 18109±1050
178±8 85±5∗ 12±1∗ 73±6∗ 11745±989∗
Reflow Heart rate (beats/min) LVs (mmHg) LVd (mmHg) LVdev (mmHg) RPP (mmHg.beats/min)
131±8 31±6 9±2 22±6 2802±810
146±13 64±13∗ 8±1 56±13∗ 8607±2302∗∗
170±8∗ 75±85∗∗ 2±1∗ 73±5∗∗ 12329±610∗∗
148±17 29±6 15±6 14±13 2095±460
60 min Heart rate (beats/min) LVs (mmHg) LVd (mmHg) LVdev (mmHg) RPP (mmHg.beats/min)
132±17 38±8 14±5 23±7 3365±1076
170±14∗ 82±7∗∗ 10±3 71±9∗∗ 12407±2180∗∗
154±7 92±5∗∗ 8±1 84±5∗∗ 12918±542∗∗
150±14 46±4 18±3 27±6 4052±924
Group
Cholesterol
∗P<0.05, ∗∗P<0.01 v no treatment controls. Abbreviations as in Table 1. Data at reflow were recorded after 5 min reperfusion. Data at 60 min were recorded at 60 min after reperfusion.
intensity of the 25 kDa TnI band and LV developed pressure (R2=0.04) or rate pressure product (R2= 0.18), when data from all hearts in each group were pooled.
Effects of interventions on myocardial recovery and TnI fragmentation Hemodynamic data for hearts exposed to 60 min ischemia and 60 min reperfusion, with different treatment interventions, are compared with the group receiving no treatment in Table 2. All seven hearts treated with -arginine, five of six treated with oxygen radical scavengers and seven of ten hearts from cholesterol-fed rabbits completed the protocol. Before ischemia, there were no differences between controls, -arginine and OFR scavenger groups, although the -arginine treated group tended to a higher heart rate. The cholesterol group had lower LVdev and RPP and slightly higher LVd than did the other groups. After ischemia– reperfusion, hearts treated with -arginine had better LVdev (71±9 mmHg, P<0.01) and recovery of RPP (66±7% of pre-ischemia, P<0.01) than did hearts receiving no treatment (LVdev= 23±7 mmHg, RPP=22±9% of pre-ischemia). Similarly, hearts treated with oxygen free radical scavengers had better recovery of LVdev (84±5 mmHg, P<0.01) and RPP (72±3%, P<0.01) after 60 min reperfusion than did the controls. In contrast, hearts from the cholesterol group
Figure 3 Examples of Western blots for TnI (antibody 19C7) from hearts exposed to 60 min ischemia and reperfusion. Blots are shown for hearts without additional treatment (controls), or with addition of -arginine or oxygen free radical scavengers to perfusate. Also shown is a blot for a heart from a hypercholesterolemic rabbit. In each heart both 29 kDa and 25 kDa immunoreactive bands are evident. No other bands are seen.
had poor functional recovery after 60 min ischemia and reperfusion (RPP=37±8% baseline). The cholesterol hearts also tended to have a higher LV enddiastolic pressure than did the other groups. Western blots for representative hearts from each treatment group are shown in Figure 3, and group data for intensities of the intact TnI 29 kDa and fragment TnI 25 kDa bands are shown in Figure 4. In hearts from each group, both 29 kDa and 25 kDa bands were detected. Despite the marked differences in functional recovery, there were no differences between groups in the relative intensity of the 25 kDa band after 60 min ischemia and reperfusion (control 17.4±2.4%, -arginine
Troponin I in Stunned Myocardium
Figure 4 Group data for intensity of the TnI 25 kDa band in hearts exposed to 60 min ischemia and 60 min reperfusion, without treatment (controls), or with addition of -arginine, or with addition of oxygen free radical scavengers to the perfusate. Also shown are data for hearts from hypercholesterolemic rabbits. Data shown were obtained by Western blotting with antibody 19C7. There was no significant difference between groups.
Figure 5 Representative Western blots of myocardium from a control heart, a freshly harvested rabbit heart and human atrial appendage tissue. Each sample shows the presence of a 25 kDa TnI band in addition to intact TnI. Data shown are for blots with antibody 19C7.
16.8±1.8%, OFR scavengers 16.0±3.2% cholesterol 14.0±1.9%). Furthermore, there was no relation between post-ischemic function and intensity of the 25 kDa band when all data were pooled.
Myocardium in vivo The same 29 kDa and 25 kDa bands were identified by Western blotting in freshly excised rabbit and human myocardium (Fig. 5). Both bands were consistently identified in all normal myocardial samples. The relative intensity of the 25 kDa band in the freshly excised rabbit hearts (8.5±2.3%) was similar to that of both control and 15 min ischemia hearts, indicating that the 25 kDa band is unlikely to be an artefact resulting from crystalloid perfusion. The same results were obtained using
407
the antibodies with epitopes near the centre of the TnI sequence (4E5 and 1H8) and the antibody binding nearer the N-terminus (19C7). The same 25 kDa band was also observed in each of the human myocardial samples, with a mean intensity of 5.1±2.6% of the total TnI immunoreactivity. No other TnI bands were seen in the human samples. All samples were prepared under stringent conditions to exclude artefact due to proteolysis during preparation of the tissue. These conditions included a cocktail of eight protease inhibitors, immediate and rapid processing, and handling at 4°C. Omitting the cocktail of protease inhibitors increased TnI proteolysis if tissue samples were processed at room temperature, but not at 4°C (data not shown).
Discussion The exact pathogenesis of post-ischemic myocardial stunning has eluded researchers for over a decade. The most recent hypothesis for the mechanism of stunning is centered upon proteolytic fragmentation of TnI. It has been proposed that increased [Ca2+]I during ischemia–reperfusion results in activation of proteases, such as -calpain, with cleavage of TnI at the C terminus (TnI1-193) and subsequent impairment of myofilament contractility. We therefore examined the relationship between TnI fragmentation and myocardial contractile function after different ischemic events and after treatment interventions which can attenuate myocardial stunning. Our data show that although TnI fragmentation is increased after prolonged ischemia, the functional recovery of post-ischemic myocardium is independent of the degree of TnI fragmentation.
Troponin I and myocardial stunning Considerable evidence indicates that myocardial stunning is a disorder of the myofilaments,17,18 possibly due to a functional defect of the troponin complex.19 Reconstitution of rabbit skeletal myofilaments with troponin complex isolated from stunned hearts results in reduced contractile function.20 Reports from Gao et al.4 and McDonough et al.5 have identified a TnI degradation fragment (TnI1-193) in rat hearts after global ischemia. In contrast, other groups have not found any evidence of increased TnI fragmentation in post-ischemic myocardium,6,7,21 although it may be argued that these negative findings reflect lack of sensitivity of the detection methods used. As independent
408
A. M. Prasan et al.
measures of myocardial viability after prolonged global ischemia were not available for the earlier studies,4,5 the possibility that TnI fragmentation is a feature of irreversibly injured myocardium could not be excluded. Indeed, the data of Thomas et al. suggest that increased TnI fragmentation is a feature of irreversible injury, rather than reversible stunning.8 Resolution of this question requires quantitation of TnI fragmentation after different degrees of myocardial injuries, and correlation with contractile state and histopathology. In perfused rabbit hearts, we have observed fragmentation of TnI similar to that reported for rat4,5 and porcine8 hearts. Our data show no increase in TnI fragmentation after 15 min ischemia and subsequent reperfusion, despite the presence of stunning in the absence of significant myocardial necrosis. After 60 min ischemia and reperfusion however, we observed an increase in TnI fragmentation and histological and biochemical evidence of irreversible injury, in addition to poor functional recovery. The extent of irreversible injury in our model is comparable to that previously documented after prolonged ischemia in both isolated15 and in vivo22 rabbit hearts. To this point our data are consistent with the findings of Thomas et al.,8 suggesting that increased TnI fragmentation only occurs after more severe ischemia–reperfusion insults, associated with irreversibly injured cells. This is the first study to report TnI fragmentation after treatment interventions, which alter post-ischemic myocardial function. Our findings suggest that the potential role of TnI in myocardial stunning may be more complex than simple fragmentation. Both -arginine and oxygen free radical scavenger treatments were associated with markedly improved myocardial function after 60 min ischemia and reperfusion. As neither oxygen radical scavengers, such as mercapto-proponylglycine, nor -arginine treatment reduce infarct size after prolonged ischemia in rabbits,22,23 the functional improvement would appear to result from attenuation of stunning of residual viable myocardium. Despite this, there was no reduction in the degree of TnI fragmentation in the treated hearts. Conversely, hypercholesterolemia was associated with poor recovery of function, but the TnI fragmentation in these hearts was similar to the -arginine and oxygen free radical scavenger groups. Although increased TnI fragmentation was observed with a longer duration of ischemia, it did not correlate with functional recovery of the myocardium and was unaffected by treatment interventions which altered functional recovery of the hearts. In contrast to previous reports from rat hearts,
we did not observe any other TnI fragmentation, or higher molecular weight immunoreactive complexes in rabbit hearts. Subsequent cleavage of TnI1-193 to TnI63-193 and TnI73-193 has been described in rats.5 All our antibodies would detect TnI1-193 and the 4E5 and 1H8 antibodies would detect TnI63-193 but not necessarily TnI73-193. Our observations suggest that such cleavage does not occur in rabbits after even 60 min low-flow ischemia. Our observations of some TnI fragmentation in normal human and rabbit myocardium (both in vivo and ex vivo) cast further doubt upon the role of the TnI fragment in the pathogenesis of myocardial stunning. Some TnI fragmentation has been observed in normal porcine myocardium8 and was also reported by Gao et al.4 in non-ischemic rat hearts. Previously, McDonough et al.24 have described TnI fragmentation in left and right ventricular biopsies after cardioplegia and reperfusion, concluding that TnI fragmentation may be a cause of contractile dysfunction in humans. Both this study and another from the same group11 also showed TnI fragments in samples obtained before bypass. Although the finding of fragmentation may be ascribed to previous recurrent ischemic injury, an equally plausible explanation is that the fragmentation is a feature of normal myocardium, for other reasons. As the right atrial samples we examined are unlikely to have suffered ischemiareperfusion injury, it is more likely that limited TnI fragmentation is indeed a feature of normal human myocardium.
Significance of TnI fragment An obvious issue is whether our findings of TnI fragmentation in normal myocardium could be an artefact of tissue processing. In order to minimize this, all tissue samples were immediately frozen in liquid nitrogen and protein preparation was performed at 4°C in the presence of eight different protease inhibitors (a wider range than used in previous studies). The cocktail of protease inhibitors includes two calpain inhibitors, calpain inhibitor I and E-64, which penetrate the cell membrane and rapidly inhibit intracellular calpain activity. Other groups, using a variety of tissue preparation techniques, have also observed similar TnI fragmentation in non-ischemic myocardium in different species.5,8,25 Not all studies have found TnI fragmentation. Some of the apparent differences between previous reports may relate to methodological differences, such as protein loading. Thus
Troponin I in Stunned Myocardium
Thomas et al.,8 using a protein loading of 200 g, found up to 10% basal TnI fragmentation, but Papp et al.,25 using a loading of only 3 g per lane, failed to demonstrate any TnI fragmentation. It is possible that the TnI fragment is a feature of normal protein turnover in the heart. All myofilament proteins are subject to continuous synthesis, incorporation and degradation. TnI has a relatively short half-life of 3.2 days.27 Current evidence indicates that TnI is exchanged stochastically along the length of the thin filament28 and there appears to be a pool of free TnI continuously available for incorporation into the myofilaments.27 This free pool has been estimated to be at least 3% of total TnI in human hearts.28 The 25 kDa TnI fragment may therefore be a degenerate protein, from either the free TnI pool or the myofilaments. Increased TnI fragmentation after prolonged ischemia is consistent with increased proteolytic activity4,29 and clinical studies show that both intact TnI and the truncated TnI can be detected in the circulation after myocardial infarction in humans.30 Although myocardial stunning may well be associated with dysfunction of the troponin regulatory complex,11 we find no correlation between degree of TnI fragmentation and severity of stunning. Furthermore, no other proteolytic degradation of the regulatory proteins has yet been described. This evidence suggests that the aetiology of myocardial stunning may involve a more subtle alteration of the regulatory proteins than gross structural change. One possible mechanism is via phosphorylation of TnI by protein kinase C (PKC), which can result in reduced myofilament Ca2+ sensitivity and impaired contractile performance.31–34 Indeed, there is evidence that such phosphorylation occurs in post-ischemic myocardium.35,36 Furthermore, phosphorylation of TnI by PKC, whilst reducing myofilament Ca2+ sensitivity, also increases the susceptibility of this protein to proteolytic cleavage by -calpain in vitro.37 This raises the intriguing possibility that postischemic myocardial dysfunction is a two-stage process, with initial functional deficit due to phosphorylation of TnI (and possibly other proteins) with subsequent structural damage due to proteolysis. This two-stage process would be consistent with previous studies reporting increased TnI fragmentation after ischemia–reperfusion and with our observations of dissociation between function and TnI degradation in stunned myocardium. The first stage may not inevitably lead to the second stage, which would require activation of proteolytic enzymes, possibly partly mediated by an increase in preload.38 Thus our observations after 15 min
409
ischemia and reperfusion would be consistent with an ischemic insult sufficient to result in initial functional deficit of the regulatory complex, but not severe enough to initiate subsequent proteolytic damage.
Conclusions Our results indicate that proteolysis of TnI is not the primary cause of stunned myocardium. Some TnI fragmentation is evident in normal myocardium of different species, possibly resulting from the normal turnover of TnI. Although an increase in TnI fragmentation is observed after prolonged ischemia and reperfusion, this does not correlate with myocardial function. We propose that initial functional alteration of TnI (or another regulatory protein) is the cause of myocardial stunning and precedes structural damage to the protein.
Acknowledgements This study was supported by a National Health and Medical Research Council of Australia Project Grant No. 2000: 107272 to R.W.J. and B.D.H. The authors wish to thank Virginia Turner and Jane Radford for assistance with the preparation of histological specimens and Dorothy Kouzios for assistance with CK and cTnT measurement and the Chiron Corporation for the generous gift of two monoclonal antibodies to TnI.
References 1. B R. Mechanism of myocardial stunning. Circulation 1990; 82: 723–738. 2. B R, M E. Molecular and cellular mechanisms of myocardial stunning. Physiol Rev 1999; 79: 609–634. 3. G WD, A D, B PH, M E. Relationship between intracellular calcium and contractile force in stunned myocardium: direct evidence for decreased myofilament Ca-responsiveness and altered diastolic function in intact ventricular muscle. Circ Res 1995; 76: 1036–1048. 4. G WD, A D, L Y, P NG, M AM, M E. Role of Troponin I proteolysis in the pathogenesis of stunned myocardium. Circ Res 1997; 80: 393–399. 5. MD JL, A DK, V E JE. Troponin I degradation and covalent complex formation accompanies myocardial ischaemia/reperfusion injury. Circ Res 1999; 84: 9–20. 6. B R, M R, D P, C E, S S, D L F. Binding of cytosolic
410
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18. 19. 20.
21.
A. M. Prasan et al.
proteins to myofibrils in ischemic rat hearts. Circ Res 1996; 78: 821–828. G L, M R, V M, B CM, DL F. Cardiomyocyte Troponin T immunoreactivity is modified by cross-linking resulting from intracellular calcium overload. Circulation 1996; 93: 1896–1904. T SA, F JA, L T-C, F J, C JM. Absence of Troponin I degradation or altered sarcoplasmic reticulum uptake protein expression after reversible ischaemia in swine. Circ Res 1999; 85: 446–456. S AJ, K FJ, D RS. Myofibrillar disruption in hypocontractile myocardium showing perfusion–contraction matches and mismatches. Am J Physiol 2000; 278: H1320–H1334. R HM, T X-H, S RJ, M AF. The C terminus is essential for full inhibitory activity and Ca2+ sensitivity of rat myofibrils. J Biol Chem 1997; 272: 26887–26892. M AM, K H, G D, MD JL, K DA, V E JE, M E. Transgenic mouse model of stunned myocardium. Science 2000; 287: 488–491. K H, S DG, M AM, M E. Maintained contractile reserve in a transgenic mouse model of myocardial stunning. Am J Physiol (Heart Circ Physiol) 2001; 280: H2623–H2630. E DT, W M, M N, C GA, E RM, C GA, E RM, R JA, F JE, D DW, D DK. L-arginine reduces endothelial inflammation and myocardial stunning during ischemia. Ann Thorac Surg 1995; 60: 1275–1281. T E, A M. Pathologic assessment of myocardial cell necrosis and apoptosis after ischemia and reperfusion with molecular and morphological markers. J Mol Cell Cardiol 2000; 32: 209–224. Y K, L Y, T A, M T, L GS, Y X, H D, C MV, D JM. Rat and rabbit heart infarction: effects of anesthesia, perfusate, risk zone, and method of infarct sizing. Am J Physiol 1994; 267: H2383–H2390. S H, J G. Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa. Anal Biochem 1987; 166: 368–379. G WD, L Y, M R, M E. Intrinsic myofilament alterations underlying the decreased contractility of stunned myocardium: a consequence of Ca2+-dependent proteolysis? Circ Res 1996; 78: 455–465. M E, G WD. Stunned myocardium: a disease of the myofilaments? Basic Res Cardiol 1995; 90: 269–272. F CS, R FC. The troponin complex and regulation of muscle contraction. FASEB J 1996; 9: 755–767. MD KS, M RL, M WP. Incorporation of the Troponin regulatory complex of post-ischemic stunned porcine myocardium reduces myofilament calcium sensitivity in rabbit psoas skeletal muscle fibers. J Mol Cell Cardiol 1998; 30: 285–296. L H, M A, R N, V A H, B P, K U, K J, L S, L B, L I, M FU, N J, S W. Biochemical
22.
23.
24.
25.
26. 27.
28.
29. 30.
31.
32.
33.
34.
35.
mechanism(s) of stunning in conscious dogs. Am J Physiol (Heart Circ Physiol) 2000; 279: H176–H184. M T, C MV, D JM. Failure of N-2mercaptoproponyl glycine to reduce myocardial infarction after 3 days of reperfusion in rabbits. Basic Res Cardiol 1999; 94: 180–187. W MW, T CS, R S, Z Z-Q, V-J J. Endogenous nitric oxide (NO) protects against ischaemia-reperfusion injury in the rabbit. Cardiovasc Res 1995; 30: 79–86. MD JL, L R, P W, T MY, M S, P SC, A D, R G, V E JE. Cardiac Troponin I is modified in the myocardium of bypass patients. Circulation 2001; 103: 58–64. P Z, V J, S GJM. Calpain-I induced alterations in the cytoskeletal structure and impaired mechanical properties of single myocytes of rat heart. Cardiovasc Res 2000; 45: 981–993. M AF. Turnover of cardiac troponin subunits: kinetic evidence for a precursor pool of Troponin I. J Biol Chem 1981; 256: 964–968. M DE, A FP, M JM. Thin filament protein dynamics in fully differentiated adult cardiac myocytes: toward a model of sarcomere maintenance. J Cell Biol 1999; 145: 1483–1495. A JE, S KB, L Y, L JH, J AS. Comparable detection of acute myocardial infarction by CKMB isoenzyme and cardiac troponin I. Clin Chem 1994; 40: 1291–1295. Y K, S Y, F M, H K. Calpain is implicated in rat myocardial injury after ischemia or reperfusion. Jpn Circ J 1995; 59: 40–48. S Q, L M, Z X, Z M, K L, L S, L JP. Degradation of cardiac troponin I in serum complicates comparisons of cardiac troponin I assays. Clin Chem 1999; 45: 1018–1025. J NM, N JTA, R RL, B GC, F D, K MG, B PM, H YA, K JF. Phosphorylation specificities of protein kinase C isozymes for bovine cardiac troponin I and troponin T and sites within these proteins and regulation of myofilament properties. J Biol Chem 1996; 271: 23277–23283. V RC, K JF. Protein kinase C-mediated phosphorylation of troponin I and C-protein in isolated myocardial cells is associated with inhibition of myofibrillar actomyosin MgATPase. J Biol Chem 1993; 268: 2705–2711. T Y, C G, K DM, L Z, W H, K EG, K GL, W RA. In vivo phosphorylation of cardiac troponin I by protein kinase C decreases cardiomyocyte calcium responsiveness and contractility in transgenic mouse hearts. J Clin Invest 1998; 102: 72–78. W YG, B WJ, H J, S AM, B LA, L SL. Brief rapid pacing depresses contractile function via Ca2+/PKC-dependent signaling in cat ventricular myocytes. Am J Physiol (Heart Circ Physiol) 2000; 280: H90–H98. P P, Z J, Q Y, T X-L, M S, C X, B R. Ischemic preconditioning induces selective translocation of protein kinase C isoforms epsilon and eta in the heart of conscious rabbits without subcellular redistribution of total protein kinase C activity. Circ Res 1997; 81: 404–414.
Troponin I in Stunned Myocardium 36. U F, W PE, D SB, H KD, W AA. MDL-28170, a membrane-permeant calpain inhibitor, attenuates stunning and PKC proteolysis in reperfused ferret hearts. Cardiovasc Res 1997; 35: 60–67. 37. D L F, D T R, S F, B R, M E, S N, S S, P
411
S. Specific degradation of troponin T and I by mucalpain and its modulation by substrate phosphorylation. Biochemical J 1995; 308: 57–61. 38. F J, S BJ, F JA, L TC, C JM. Preload induces troponin I degradation independently of myocardial ischemia. Circulation 2001; 103: 2035–2037.
Effect of Treatment on Ventricular Function and Troponin I Proteolysis in Reperfused Myocardium Ananth M. Prasan1, Hugh C. K. McCarron1, Brett D. Hambly2, Gary G. Fermanis3, David R. Sullivan4 and Richmond W. Jeremy1 Departments of 1Medicine and 2Pathology, University of Sydney, Sydney, 2006, NSW, Australia, 3 Department of Cardiothoracic Surgery, St. George Hospital, Kogarah, 2217 and 4Department of Clinical Biochemistry, Royal Prince Alfred Hospital, Camperdown 2050, Australia (Received 6 August 2001, accepted for publication 4 January 2002) A. M. P, H. C. K. MC, B. D. H, G. G. F, D. R. S R. W. J. Effect of Treatment on Ventricular Function and Troponin I Proteolysis in Reperfused Myocardium. Journal of Molecular and Cellular Cardiology (2002) 34, 401–411. Effects of ischemia time and treatment interventions upon troponin I (TnI) proteolysis and function of reperfused myocardium were examined in isolated, perfused rabbit hearts. Hearts were randomized to 90 min aerobic perfusion, 15 min low-flow (1 ml/min) ischemia (I) and 60 min reperfusion (R) or 60 min low-flow I and 60 min R. Hearts subject to 60 min I and 60 min R received either no treatment, -arginine treatment, or treatment with oxygen free radical (OFR) scavengers (mercapto-proponylglycine, catalase and superoxide dismutase). Hearts from cholesterol-fed rabbits were also studied after 60 min I and R. Isovolumic LV pressure and heart rate were recorded throughout and Western analysis of ventricular myocardium, using 3 specific antibodies, detected intact TnI (29 kDa) and TnI fragment (25 kDa). Hearts subject to 15 min I had minimal irreversible injury (TTC negative region=0.6±0.4% LV) but hearts subject to 60 min I had more extensive injury (TTC negative=40.7±5.8% LV). Recovery of rate–pressure product after 15 min I and 60 min R (56±9% of baseline) was better than after 60 min I and 60 min R (23±9%, P<0.01). Both arginine and OFR scavengers were associated with better recovery of function after 60 min I, (66±7% and 72±3% of baseline respectively, P<0.01 v no treatment) but cholesterol hearts had poor recovery after 60 min I (37±8%). The 25 kDa TnI (% total TnI immunoreactivity) was 8.7±0.9% in controls, 10.0±1.6% after 15 min I and 60 min R, and 17.4±2.4% after 60 min I and 60 min R (P<0.01 v controls and 15 min I). The proportion of 25 kDa TnI was increased in all hearts after 60 min I and did not change with treatment (-arginine 16.8±1.8%, OFR scavengers 16.0±3.2%, cholesterol 14.0±1.9%). There was no relation between proportion of 25 kDa TnI and recovery of function. Samples from freshly excised rabbit hearts and human right atria also had 25 kDa TnI (relative intensities 8.5±2.3% and 5.1±2.6% respectively). Although TnI fragmentation increases after prolonged ischemia and reperfusion, the functional recovery of stunned myocardium is independent of degree of TnI fragmentation. 2002 Elsevier Science Ltd. All rights reserved. K W: Myocardium, ischemia and reperfusion; Troponin I; Humans; Rabbits; Proteolysis.
Introduction The pathogenesis of persistent contractile dysfunction (stunning) of post-ischemic myocardium has been the subject of intense investigation. We know that intracellular calcium overload and oxygen free radicals contribute and that stunned
myocardium exhibits reduced Ca2+ sensitivity.1,2 Delayed recovery of function over several days and abnormal responses to Ca2+ suggest that injury to critical myofilament proteins occurs during ischemia and reperfusion.3 Evidence in support of this hypothesis has been presented by Gao et al.4 and by McDonough et al.,5 who observed apparent
Please address all correspondence to: Prof. Richmond W. Jeremy, Room 383, Blackburn Building, Department of Medicine, University of Sydney 2006, New South Wales, Australia. Tel: (612) 95157608. Fax: (612) 95506262. E-mail: [email protected]
0022–2828/02/040401+11 $35.00/0
2002 Elsevier Science Ltd. All rights reserved.
402
A. M. Prasan et al.
degradation of troponin I (TnI) in post-ischemic rat hearts. It has been hypothesized that Ca2+dependent activation of the cysteine protease calpain causes proteolytic cleavage of TnI near the C terminus, leaving a dysfunctional fragment (TnI1-193). The data about TnI fragmentation in stunned myocardium are not, however, uniform. Other groups have failed to find evidence of TnI fragmentation6–9 or have only observed a fragment in association with irreversible myocardial injury.8 There are also conflicting data about the functional consequences of C-terminus truncation of TnI. Theoretical considerations suggest that such truncation would increase Ca2+ sensitivity of the myofilaments. Indeed, studies of cTnI mutants, with C terminus truncations similar to those proposed for stunned myocardium, have shown that inhibition of actin-myosin interaction by the troponin complex is lessened, so that Ca2+ sensitivity of the myofilaments is actually increased.10 This is the opposite of observations in stunned myocardium. In contrast, transgenic mice with overexpression of TnI1-193 have been reported to have impairment of myocardial contractility and reduced Ca2+ sensitivity.11,12 Importantly, if TnI fragmentation is a fundamental pathogenic feature of stunned myocardium, then therapeutic interventions, which ameliorate the severity of post-ischemic dysfunction, should be associated with less TnI fragmentation. This study therefore compared the degree of TnI fragmentation with contractile function after short and long ischemic periods, and examined the effects of treatment interventions upon TnI fragmentation and myocardial function.
Materials and Methods Experimental studies were performed using male adult New Zealand White rabbits (2.5 to 3.5 kg), according to protocols approved by the Animal Research Ethics Committee of the University of Sydney.
Ischemia–reperfusion studies in isolated hearts After administration of heparin (2500 units i.v.), rabbits were killed with sodium pentobarbital (150 mg/kg i.v.) and hearts rapidly excised for Langendorff perfusion at a pressure of 80 mmHg with modified Krebs–Henseleit buffer containing 16 m glucose (37°C, pH 7.4, bubbled with 95% O2 and 5% CO2). Hearts were perfused within a
temperature-controlled organ-bath. A latex balloon was introduced into the left ventricle (LV) across the mitral valve for pressure measurement (P23Db, Gould Inc, CA, USA). The balloon was inflated with physiological saline solution (37°C) to obtain a LV end-diastolic pressure of 8–10 mmHg under baseline conditions. Heart rate and isovolumic ventricular pressures were recorded continuously with a digital acquisition system (MP100WSW, Biopac Systems, CA, USA). In hearts subject to low-flow ischemia, perfusion was by peristaltic pump (Harvard Instruments Model 66, Harvard Apparatus Inc., Holliston, MO, USA) at a rate of 1 ml/min. During low-flow ischemia, the LV balloon was deflated and during reperfusion the balloon was refilled with the same volume of saline as used during baseline measurements. Before each study, hearts were randomly assigned by ballot to one of the experimental protocols. If hearts developed ventricular fibrillation or rigor at the time of reperfusion, they were excluded from the subsequent data analysis. In the first series of experiments, the relation between duration of ischemia, myocardial necrosis and TnI fragmentation was examined. After an initial 15 min stabilization period, hearts were subject to either normal perfusion for 90 min without an ischemic period (n=8), or to 15 min low-flow ischemia (as a model of moderate post-ischemic dysfunction without necrosis) followed by 60 min free reperfusion (n=14) or to 60 min low-flow ischemia (as a model of severe post-ischemic dysfunction with necrosis) followed by 60 min reperfusion (n=14). The presence and extent of myocardial necrosis occurring after low-flow ischemia was quantified by horseradish peroxidase (HPR) staining14 and triphenyl-tetrazolium chloride (TTC) staining15 and by measurement of creatine kinase (CK) and cardiac troponin T (cTnT) in coronary effluent. Horseradish peroxidase (Type II, Sigma, St. Louis, MO, USA, 10 mg) was added to perfusate for the last 5 min of perfusion for each protocol. Two mid-left ventricular samples (2 mm thick) were then cut before the remainder of the ventricle was frozen in liquid nitrogen for protein studies. After fixation (4% paraformaldehyde in 0.1 sodium cacodylate buffer, pH 7.4, for 4 h at 4°C before washing in 0.1 sodium cacodylate buffer and incubation in buffer solution overnight), frozen sections (60 m thick) were cut from the samples. Sections were washed with 0.1 cacodylate buffer and incubated for 10 min at room temperature in 100 ml of 0.1 Tris-HCl buffer (pH 7.6), containing 100 mg 3,3′diaminobenzidine tetrahydrochloride (Sigma) and
Troponin I in Stunned Myocardium
33.3 l of 30% hydrogen peroxide. Sections were examined by light microscopy to detect abnormal staining, indicative of irreversible injury. Twenty medium power fields (×20) for each heart were examined by two blinded observers and the percentage of positively stained cells per field was determined. For TTC staining, four hearts from each group were immediately chilled to 4°C and cut into five 2-mm-thick transverse slices. Slices were then incubated at 37°C for 20 min in 1% TTC (0.1 phosphate buffer, pH 7.4) prior to fixing in 15% formalin. Each slice was digitally photographed and the area of necrosis determined by computer planimetry (Image-Pro Plus, Media Cybernetics, Silver Spring, Maryland, USA). For the measurement of CK and cTnT the pulmonary artery was cannulated and coronary effluent collected into chilled vials at timed intervals (pre-ischemia, 1, 5 and 30 min reperfusion). The CK was measured by enzymatic assay (Roche/Hitachi 917 clinical chemistry analyser) and cTnT was quantified by immunoassay (Roche Elecsys 2010 immunoassay analyser). In the second series of experiments, the effects of interventions associated with altered post-ischemic function2,13 were examined in hearts exposed to 60 min ischemia followed by 60 min reperfusion. In seven hearts, -arginine (200 ) was included in the perfusate throughout the experiment; in another six hearts OFR scavengers [mercapto-proponyl-glycine (3 m), superoxide dismutase (6 U/ ml) and catalase (6 U/ml)] were included in the perfusate throughout the experiment. The effect of hypercholesterolemia (which is associated with endothelial dysfunction and abnormal vascular NO synthesis) was examined in another ten hearts exposed to 60 min ischemia followed by 60 min reperfusion. These rabbits had been fed 0.5% dietary cholesterol for 16 weeks, before the ischemia and reperfusion studies, and had a mean total blood cholesterol level of 30±3 mmol/l at the time of study.
Normal myocardium in vivo Further studies of TnI fragmentation were performed on freshly harvested normal rabbit left ventricular myocardium and human right atrial myocardium. After anesthesia, as described above, the rabbit hearts were rapidly excised whole. These hearts (n=3) were immediately rinsed in cold physiological saline solution (4°C), frozen in liquid nitrogen and stored at −80°C for subsequent protein studies. Human samples were obtained from 3
403
patients, who were undergoing elective coronary artery bypass grafting, and who had given informed consent for tissue sampling. None of the patients had a history of atrial fibrillation. Samples from the right atrial appendage (150–300 mg wet weight) were excised at the time of atrial cannulation, prior to commencement of bypass and in the absence of ischemic preconditioning. These samples were immediately rinsed in cold saline solution and frozen in liquid nitrogen for subsequent analysis.
Protein studies The frozen myocardial samples were homogenized (4°C) in buffer solution containing imidazole-HCl 50 m, EGTA 5 m, dithiothreitol 1 m, pH 7.8. The buffer contained a cocktail of eight protease inhibitors: leupeptin 20 , phenylmethylsulphonyl fluoride 250 , Calpain I inhibitor 20 , AEBSF [4-(2-aminoethyl)-benzenesulfonylfluoride] 1 m, aprotinin 1 , bestatin 50 , E-64 [trans-epoxysuccinyl-leucylamido-(4-guanidino] butane] 15 , pepstatin 10 . Samples were centrifuged at 10 000 g for 10 min at 4°C and the pellet resuspended in the above solution, with addition of 0.5% Triton X-100 and -mercaptoethanol 10 m. Subsequently, 50 l of sample was diluted 2:1 in Tricine sample buffer and boiled for 5 min before electrophoresis. Protein determination in the homogenate was performed by bicinchoninic acid protein assay (Pierce Inc., Rockford, IL, USA). Total protein (40 g/lane) was separated by SDS-PAGE 16 (16.5% Tris-Tricine Ready Gels, Bio-Rad, Hercules, CA, USA). After separation, proteins were visualized by Coomassie blue or were transferred onto nitrocellulose membrane sheets (Bio-Rad) at 40 V for 45 min and 80 V for a further 30 min. Following transfer, the membrane was incubated with 4% bovine serum albumin in Tris buffered saline (TBS, 50 m Tris-HCl, 150 m NaCl; pH 7.5) at 4°C overnight. Three monoclonal antibodies to TnI were used to detect TnI fragmentation; clones 4E5 and 1H8 (Chiron Diagnostics Corporation, East Walpole, MA, USA, epitopes aa 70-87) and clone 19C7 (Biodesign International, Kennebunk, Maine, USA, epitope aa 41-49). The membrane was incubated with TnI antibody at a dilution of 1:2000 for 60 min. Subsequently, membranes were rinsed in TBS containing 0.1% Tween 20 and incubated with alkaline phosphatase-conjugated anti-mouse IgG (Sigma) at a dilution of 1:5000 for 60 min. Immunoreactive bands were visualized with Western Blue Substrate
404
A. M. Prasan et al.
Table 1 Ischemia time and hemodynamics of perfused hearts Controls No ischemia
15 min Ischemia
60 min Ischemia
Baseline 163±9 109±5 7±1 102±5 16591±1303
Baseline 166±8 101±6 8±1 94±6 15626±1270
Heart rate (beats/min) LVs (mmHg) LVd (mmHg) LVdev (mmHg) RPP (mmHg.beats/min)
Baseline 159±4 102±3 9±1 94±3 14877±689
Heart rate (beats/min) LVs (mmHg) LVd (mmHg) LVdev (mmHg) RPP (mmHg.beats/min)
30 min 156±5 103±5 8±1 95±4 14840±702
5 min reflow 152±7 56±6∗∗ 6±1 50±6∗∗ 7791±984∗∗
5 min reflow 131±8 31±6¶¶ 9±2 22±6¶¶ 2802±810¶¶
Heart rate (beats/min) LVs (mmHg) LVd (mmHg) LVdev (mmHg) RPP (mmHg.beats/min)
90 min 146±6 104±6 10±1 94±5 13592±592
60 min reflow 166±9 67±5∗∗ 9±1 58±5∗∗ 9719±1189∗
60 min reflow 132±17 38±8¶ 14±5 23±7¶¶ 3365±1076¶¶
∗P<0.05, ∗∗P<0.01 v no ischemia, ¶P<0.05, ¶¶P<0.01 v 15 min ischemia. LVs, LVd, LVdev=left ventricular systolic, diastolic and developed pressures. Rate Pressure Product (RPP)=LVdev×heart rate. Baseline data were recorded after 10 min initial perfusion in each group.
(Promega, Madison, WI, USA) and intensity measured by video gel densitometry (GDS8000, UltraViolet Products, UVP Ltd, Cambridge, UK) and (Phoretix 1D Advanced software, Phoretix International, Newcastle upon Tyne, UK). Quantitation of TnI fragmentation (antibody 19C7) was performed on duplicate samples for each heart. Immunoblotting identified two TnI bands, at 29 kDa (intact TnI) and at 25 kDa (fragment TnI), similar to those previously reported. The relation between protein load per lane and intensity of immunoblot stain for TnI was linear between 15 and 75 g per lane and a standard protein loading of 40 g per lane was used for experimental protocols. Each band was measured as the proportion of total immunoreactivity per lane. The mean variance between duplicate measurements of TnI band intensity in samples from the same heart was 3.0±0.6%.
Statistics Group data are described as mean±standard error. The means of duplicate measurements for each heart were used for statistical analysis. Hemodynamic and densitometry data were compared between groups by analysis of variance with posthoc Newman–Keuls comparison of group means when significance was indicated (SPSS 10.0 SPSS Inc. Chicago, IL, USA). A P value <0.05 is described as statistically significant.
Results Ischemia duration and myocardial recovery Hemodynamic data are shown in Table 1 for hearts assigned to control perfusion with no ischemia, or 15 min or 60 min ischemia with reperfusion. Of the eight hearts assigned to control perfusion without any ischemia, all completed the protocol. Thirteen of the 14 hearts assigned to 15 min ischemia and 12 of the 14 hearts assigned to 60 min ischemia completed the protocol. Hemodynamics at the start of the experiment did not differ significantly between groups. At the end of 90 min normal perfusion, LVdev was maintained in non-ischemic controls (94±5 mmHg, NS v baseline). After 60 min reperfusion, LVdev was reduced to 58±5 mmHg in the 15 min ischemia group and to 23±7 mmHg in the 60 min ischemia group (both P<0.01 v controls). Left ventricular systolic function, indexed by ratepressure product was better preserved after 15 min ischemia (59±9% of baseline) than after 60 min ischemia (22±9% of baseline, P<0.01). The LV end-diastolic pressure was mildly increased after 60 min ischemia and reperfusion, although the difference from controls was not significant. In the non-ischemic controls, both cTnT and CK in coronary effluent were within normal limits throughout the perfusion period. In the 15 min ischemia hearts, the cTnT was at the limit of sensitivity of the assay (0.01 ng/ml) before ischemia,
Troponin I in Stunned Myocardium
405
Figure 1 Examples of Western blots for TnI from control, 15 min ischemia and 60 min reperfusion, and 60 min ischemia and 60 min reperfusion hearts. In each heart a band of approximately 29 kDa (intact TnI) and a band of lower molecular weight (approximately 25 kDa) are evident. Blots are shown for antibody 19C7 and similar results were obtained with antibodies 4E5 and 1H8. No other bands of higher or lower molecular weight were detected.
with a small increase to 0.07±0.01 ng/ml after 5 min reperfusion and CK did not increase above pre-ischemic levels during reperfusion in these hearts. In contrast, the 60 min ischemia hearts had a marked increase in cTnT to 0.75±0.27 ng/ml and in CK to 331±237 U/ml after 5 min reperfusion (both P<0.01 v controls and v 15 min ischemia). In non-ischemic controls, no infarct TTC staining was observed. In the 15 min ischemia hearts, a tiny amount of TTC negative myocardium was observed in only two hearts (0.6±0.4% of LV mass). Among the 60 min ischemia hearts, 40.7±5.8% of the LV mass was TTC negative, consistent with irreversible injury. The data obtained from horseradish peroxidase staining were similar. In non-ischemic control hearts only isolated cells were observed with weak positive staining. Similarly, in the 15 min ischemia hearts, less than 2% of cells examined showed positive staining. In the 60 min ischemia group, all hearts exhibited positive peroxidase staining, with a mean of 32±4% of cells examined showing staining indicative of irreversible injury.
Ischemia duration and TnI fragmentation The distribution of contractile proteins on SDS PAGE gels stained with Coomassie blue was indistinguishable between samples from control, 15 min and 60 min ischemia hearts. Western blotting identified both intact TnI (29 kDa) and an additional, less intense band of 25 kDa in hearts from each of the three groups (Fig. 1). Similar immunoreactive
Figure 2 Group data for intensity of 25 kDa TnI band as proportion of total TnI immunoreactivity in control perfused hearts (no ischemia), and hearts exposed to 15 min ischemia and 60 min reperfusion or 60 min ischemia and 60 min reperfusion.
bands were observed using each of the antibody clones 4E5, 1H8 and 19C7. In hearts from nonischemic controls and 15 min ischemia groups, the 25 kDa bands were of similar intensity but the intensity was significantly increased in hearts from the 60 min ischemia group (Fig. 2). The mean intensities of the 25 kDa TnI bands were 8.7±0.9% and 10.0±1.6% of total TnI immunoreactivity in control and 15 min ischemia groups respectively, but intensity was increased to 17.4±2.4% in 60 min ischemia hearts (P<0.01 v both controls and 15 min ischemia ). It was assumed in calculations that the affinity of the 19C7 monoclonal antibody for intact TnI and the 25 kDa fragment were the same, as the antibody epitope is far removed from the putative cleavage site at residue 193. No correlation was observed between the
406
A. M. Prasan et al.
Table 2 Treatment intervention and hemodynamics of perfused hearts after 60 min ischemia Controls
-arginine
OFR scavengers
Baseline Heart rate (beats/min) LVs (mmHg) LVd (mmHg) LVdev (mmHg) RPP (mmHg.beats/min)
166±8 101±6 8±1 94±6 15626±1270
195±16∗ 103±5 8±1 95±6 18630±1921
170±21 109±6 7±1 102±6 18109±1050
178±8 85±5∗ 12±1∗ 73±6∗ 11745±989∗
Reflow Heart rate (beats/min) LVs (mmHg) LVd (mmHg) LVdev (mmHg) RPP (mmHg.beats/min)
131±8 31±6 9±2 22±6 2802±810
146±13 64±13∗ 8±1 56±13∗ 8607±2302∗∗
170±8∗ 75±85∗∗ 2±1∗ 73±5∗∗ 12329±610∗∗
148±17 29±6 15±6 14±13 2095±460
60 min Heart rate (beats/min) LVs (mmHg) LVd (mmHg) LVdev (mmHg) RPP (mmHg.beats/min)
132±17 38±8 14±5 23±7 3365±1076
170±14∗ 82±7∗∗ 10±3 71±9∗∗ 12407±2180∗∗
154±7 92±5∗∗ 8±1 84±5∗∗ 12918±542∗∗
150±14 46±4 18±3 27±6 4052±924
Group
Cholesterol
∗P<0.05, ∗∗P<0.01 v no treatment controls. Abbreviations as in Table 1. Data at reflow were recorded after 5 min reperfusion. Data at 60 min were recorded at 60 min after reperfusion.
intensity of the 25 kDa TnI band and LV developed pressure (R2=0.04) or rate pressure product (R2= 0.18), when data from all hearts in each group were pooled.
Effects of interventions on myocardial recovery and TnI fragmentation Hemodynamic data for hearts exposed to 60 min ischemia and 60 min reperfusion, with different treatment interventions, are compared with the group receiving no treatment in Table 2. All seven hearts treated with -arginine, five of six treated with oxygen radical scavengers and seven of ten hearts from cholesterol-fed rabbits completed the protocol. Before ischemia, there were no differences between controls, -arginine and OFR scavenger groups, although the -arginine treated group tended to a higher heart rate. The cholesterol group had lower LVdev and RPP and slightly higher LVd than did the other groups. After ischemia– reperfusion, hearts treated with -arginine had better LVdev (71±9 mmHg, P<0.01) and recovery of RPP (66±7% of pre-ischemia, P<0.01) than did hearts receiving no treatment (LVdev= 23±7 mmHg, RPP=22±9% of pre-ischemia). Similarly, hearts treated with oxygen free radical scavengers had better recovery of LVdev (84±5 mmHg, P<0.01) and RPP (72±3%, P<0.01) after 60 min reperfusion than did the controls. In contrast, hearts from the cholesterol group
Figure 3 Examples of Western blots for TnI (antibody 19C7) from hearts exposed to 60 min ischemia and reperfusion. Blots are shown for hearts without additional treatment (controls), or with addition of -arginine or oxygen free radical scavengers to perfusate. Also shown is a blot for a heart from a hypercholesterolemic rabbit. In each heart both 29 kDa and 25 kDa immunoreactive bands are evident. No other bands are seen.
had poor functional recovery after 60 min ischemia and reperfusion (RPP=37±8% baseline). The cholesterol hearts also tended to have a higher LV enddiastolic pressure than did the other groups. Western blots for representative hearts from each treatment group are shown in Figure 3, and group data for intensities of the intact TnI 29 kDa and fragment TnI 25 kDa bands are shown in Figure 4. In hearts from each group, both 29 kDa and 25 kDa bands were detected. Despite the marked differences in functional recovery, there were no differences between groups in the relative intensity of the 25 kDa band after 60 min ischemia and reperfusion (control 17.4±2.4%, -arginine
Troponin I in Stunned Myocardium
Figure 4 Group data for intensity of the TnI 25 kDa band in hearts exposed to 60 min ischemia and 60 min reperfusion, without treatment (controls), or with addition of -arginine, or with addition of oxygen free radical scavengers to the perfusate. Also shown are data for hearts from hypercholesterolemic rabbits. Data shown were obtained by Western blotting with antibody 19C7. There was no significant difference between groups.
Figure 5 Representative Western blots of myocardium from a control heart, a freshly harvested rabbit heart and human atrial appendage tissue. Each sample shows the presence of a 25 kDa TnI band in addition to intact TnI. Data shown are for blots with antibody 19C7.
16.8±1.8%, OFR scavengers 16.0±3.2% cholesterol 14.0±1.9%). Furthermore, there was no relation between post-ischemic function and intensity of the 25 kDa band when all data were pooled.
Myocardium in vivo The same 29 kDa and 25 kDa bands were identified by Western blotting in freshly excised rabbit and human myocardium (Fig. 5). Both bands were consistently identified in all normal myocardial samples. The relative intensity of the 25 kDa band in the freshly excised rabbit hearts (8.5±2.3%) was similar to that of both control and 15 min ischemia hearts, indicating that the 25 kDa band is unlikely to be an artefact resulting from crystalloid perfusion. The same results were obtained using
407
the antibodies with epitopes near the centre of the TnI sequence (4E5 and 1H8) and the antibody binding nearer the N-terminus (19C7). The same 25 kDa band was also observed in each of the human myocardial samples, with a mean intensity of 5.1±2.6% of the total TnI immunoreactivity. No other TnI bands were seen in the human samples. All samples were prepared under stringent conditions to exclude artefact due to proteolysis during preparation of the tissue. These conditions included a cocktail of eight protease inhibitors, immediate and rapid processing, and handling at 4°C. Omitting the cocktail of protease inhibitors increased TnI proteolysis if tissue samples were processed at room temperature, but not at 4°C (data not shown).
Discussion The exact pathogenesis of post-ischemic myocardial stunning has eluded researchers for over a decade. The most recent hypothesis for the mechanism of stunning is centered upon proteolytic fragmentation of TnI. It has been proposed that increased [Ca2+]I during ischemia–reperfusion results in activation of proteases, such as -calpain, with cleavage of TnI at the C terminus (TnI1-193) and subsequent impairment of myofilament contractility. We therefore examined the relationship between TnI fragmentation and myocardial contractile function after different ischemic events and after treatment interventions which can attenuate myocardial stunning. Our data show that although TnI fragmentation is increased after prolonged ischemia, the functional recovery of post-ischemic myocardium is independent of the degree of TnI fragmentation.
Troponin I and myocardial stunning Considerable evidence indicates that myocardial stunning is a disorder of the myofilaments,17,18 possibly due to a functional defect of the troponin complex.19 Reconstitution of rabbit skeletal myofilaments with troponin complex isolated from stunned hearts results in reduced contractile function.20 Reports from Gao et al.4 and McDonough et al.5 have identified a TnI degradation fragment (TnI1-193) in rat hearts after global ischemia. In contrast, other groups have not found any evidence of increased TnI fragmentation in post-ischemic myocardium,6,7,21 although it may be argued that these negative findings reflect lack of sensitivity of the detection methods used. As independent
408
A. M. Prasan et al.
measures of myocardial viability after prolonged global ischemia were not available for the earlier studies,4,5 the possibility that TnI fragmentation is a feature of irreversibly injured myocardium could not be excluded. Indeed, the data of Thomas et al. suggest that increased TnI fragmentation is a feature of irreversible injury, rather than reversible stunning.8 Resolution of this question requires quantitation of TnI fragmentation after different degrees of myocardial injuries, and correlation with contractile state and histopathology. In perfused rabbit hearts, we have observed fragmentation of TnI similar to that reported for rat4,5 and porcine8 hearts. Our data show no increase in TnI fragmentation after 15 min ischemia and subsequent reperfusion, despite the presence of stunning in the absence of significant myocardial necrosis. After 60 min ischemia and reperfusion however, we observed an increase in TnI fragmentation and histological and biochemical evidence of irreversible injury, in addition to poor functional recovery. The extent of irreversible injury in our model is comparable to that previously documented after prolonged ischemia in both isolated15 and in vivo22 rabbit hearts. To this point our data are consistent with the findings of Thomas et al.,8 suggesting that increased TnI fragmentation only occurs after more severe ischemia–reperfusion insults, associated with irreversibly injured cells. This is the first study to report TnI fragmentation after treatment interventions, which alter post-ischemic myocardial function. Our findings suggest that the potential role of TnI in myocardial stunning may be more complex than simple fragmentation. Both -arginine and oxygen free radical scavenger treatments were associated with markedly improved myocardial function after 60 min ischemia and reperfusion. As neither oxygen radical scavengers, such as mercapto-proponylglycine, nor -arginine treatment reduce infarct size after prolonged ischemia in rabbits,22,23 the functional improvement would appear to result from attenuation of stunning of residual viable myocardium. Despite this, there was no reduction in the degree of TnI fragmentation in the treated hearts. Conversely, hypercholesterolemia was associated with poor recovery of function, but the TnI fragmentation in these hearts was similar to the -arginine and oxygen free radical scavenger groups. Although increased TnI fragmentation was observed with a longer duration of ischemia, it did not correlate with functional recovery of the myocardium and was unaffected by treatment interventions which altered functional recovery of the hearts. In contrast to previous reports from rat hearts,
we did not observe any other TnI fragmentation, or higher molecular weight immunoreactive complexes in rabbit hearts. Subsequent cleavage of TnI1-193 to TnI63-193 and TnI73-193 has been described in rats.5 All our antibodies would detect TnI1-193 and the 4E5 and 1H8 antibodies would detect TnI63-193 but not necessarily TnI73-193. Our observations suggest that such cleavage does not occur in rabbits after even 60 min low-flow ischemia. Our observations of some TnI fragmentation in normal human and rabbit myocardium (both in vivo and ex vivo) cast further doubt upon the role of the TnI fragment in the pathogenesis of myocardial stunning. Some TnI fragmentation has been observed in normal porcine myocardium8 and was also reported by Gao et al.4 in non-ischemic rat hearts. Previously, McDonough et al.24 have described TnI fragmentation in left and right ventricular biopsies after cardioplegia and reperfusion, concluding that TnI fragmentation may be a cause of contractile dysfunction in humans. Both this study and another from the same group11 also showed TnI fragments in samples obtained before bypass. Although the finding of fragmentation may be ascribed to previous recurrent ischemic injury, an equally plausible explanation is that the fragmentation is a feature of normal myocardium, for other reasons. As the right atrial samples we examined are unlikely to have suffered ischemiareperfusion injury, it is more likely that limited TnI fragmentation is indeed a feature of normal human myocardium.
Significance of TnI fragment An obvious issue is whether our findings of TnI fragmentation in normal myocardium could be an artefact of tissue processing. In order to minimize this, all tissue samples were immediately frozen in liquid nitrogen and protein preparation was performed at 4°C in the presence of eight different protease inhibitors (a wider range than used in previous studies). The cocktail of protease inhibitors includes two calpain inhibitors, calpain inhibitor I and E-64, which penetrate the cell membrane and rapidly inhibit intracellular calpain activity. Other groups, using a variety of tissue preparation techniques, have also observed similar TnI fragmentation in non-ischemic myocardium in different species.5,8,25 Not all studies have found TnI fragmentation. Some of the apparent differences between previous reports may relate to methodological differences, such as protein loading. Thus
Troponin I in Stunned Myocardium
Thomas et al.,8 using a protein loading of 200 g, found up to 10% basal TnI fragmentation, but Papp et al.,25 using a loading of only 3 g per lane, failed to demonstrate any TnI fragmentation. It is possible that the TnI fragment is a feature of normal protein turnover in the heart. All myofilament proteins are subject to continuous synthesis, incorporation and degradation. TnI has a relatively short half-life of 3.2 days.27 Current evidence indicates that TnI is exchanged stochastically along the length of the thin filament28 and there appears to be a pool of free TnI continuously available for incorporation into the myofilaments.27 This free pool has been estimated to be at least 3% of total TnI in human hearts.28 The 25 kDa TnI fragment may therefore be a degenerate protein, from either the free TnI pool or the myofilaments. Increased TnI fragmentation after prolonged ischemia is consistent with increased proteolytic activity4,29 and clinical studies show that both intact TnI and the truncated TnI can be detected in the circulation after myocardial infarction in humans.30 Although myocardial stunning may well be associated with dysfunction of the troponin regulatory complex,11 we find no correlation between degree of TnI fragmentation and severity of stunning. Furthermore, no other proteolytic degradation of the regulatory proteins has yet been described. This evidence suggests that the aetiology of myocardial stunning may involve a more subtle alteration of the regulatory proteins than gross structural change. One possible mechanism is via phosphorylation of TnI by protein kinase C (PKC), which can result in reduced myofilament Ca2+ sensitivity and impaired contractile performance.31–34 Indeed, there is evidence that such phosphorylation occurs in post-ischemic myocardium.35,36 Furthermore, phosphorylation of TnI by PKC, whilst reducing myofilament Ca2+ sensitivity, also increases the susceptibility of this protein to proteolytic cleavage by -calpain in vitro.37 This raises the intriguing possibility that postischemic myocardial dysfunction is a two-stage process, with initial functional deficit due to phosphorylation of TnI (and possibly other proteins) with subsequent structural damage due to proteolysis. This two-stage process would be consistent with previous studies reporting increased TnI fragmentation after ischemia–reperfusion and with our observations of dissociation between function and TnI degradation in stunned myocardium. The first stage may not inevitably lead to the second stage, which would require activation of proteolytic enzymes, possibly partly mediated by an increase in preload.38 Thus our observations after 15 min
409
ischemia and reperfusion would be consistent with an ischemic insult sufficient to result in initial functional deficit of the regulatory complex, but not severe enough to initiate subsequent proteolytic damage.
Conclusions Our results indicate that proteolysis of TnI is not the primary cause of stunned myocardium. Some TnI fragmentation is evident in normal myocardium of different species, possibly resulting from the normal turnover of TnI. Although an increase in TnI fragmentation is observed after prolonged ischemia and reperfusion, this does not correlate with myocardial function. We propose that initial functional alteration of TnI (or another regulatory protein) is the cause of myocardial stunning and precedes structural damage to the protein.
Acknowledgements This study was supported by a National Health and Medical Research Council of Australia Project Grant No. 2000: 107272 to R.W.J. and B.D.H. The authors wish to thank Virginia Turner and Jane Radford for assistance with the preparation of histological specimens and Dorothy Kouzios for assistance with CK and cTnT measurement and the Chiron Corporation for the generous gift of two monoclonal antibodies to TnI.
References 1. B R. Mechanism of myocardial stunning. Circulation 1990; 82: 723–738. 2. B R, M E. Molecular and cellular mechanisms of myocardial stunning. Physiol Rev 1999; 79: 609–634. 3. G WD, A D, B PH, M E. Relationship between intracellular calcium and contractile force in stunned myocardium: direct evidence for decreased myofilament Ca-responsiveness and altered diastolic function in intact ventricular muscle. Circ Res 1995; 76: 1036–1048. 4. G WD, A D, L Y, P NG, M AM, M E. Role of Troponin I proteolysis in the pathogenesis of stunned myocardium. Circ Res 1997; 80: 393–399. 5. MD JL, A DK, V E JE. Troponin I degradation and covalent complex formation accompanies myocardial ischaemia/reperfusion injury. Circ Res 1999; 84: 9–20. 6. B R, M R, D P, C E, S S, D L F. Binding of cytosolic
410
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18. 19. 20.
21.
A. M. Prasan et al.
proteins to myofibrils in ischemic rat hearts. Circ Res 1996; 78: 821–828. G L, M R, V M, B CM, DL F. Cardiomyocyte Troponin T immunoreactivity is modified by cross-linking resulting from intracellular calcium overload. Circulation 1996; 93: 1896–1904. T SA, F JA, L T-C, F J, C JM. Absence of Troponin I degradation or altered sarcoplasmic reticulum uptake protein expression after reversible ischaemia in swine. Circ Res 1999; 85: 446–456. S AJ, K FJ, D RS. Myofibrillar disruption in hypocontractile myocardium showing perfusion–contraction matches and mismatches. Am J Physiol 2000; 278: H1320–H1334. R HM, T X-H, S RJ, M AF. The C terminus is essential for full inhibitory activity and Ca2+ sensitivity of rat myofibrils. J Biol Chem 1997; 272: 26887–26892. M AM, K H, G D, MD JL, K DA, V E JE, M E. Transgenic mouse model of stunned myocardium. Science 2000; 287: 488–491. K H, S DG, M AM, M E. Maintained contractile reserve in a transgenic mouse model of myocardial stunning. Am J Physiol (Heart Circ Physiol) 2001; 280: H2623–H2630. E DT, W M, M N, C GA, E RM, C GA, E RM, R JA, F JE, D DW, D DK. L-arginine reduces endothelial inflammation and myocardial stunning during ischemia. Ann Thorac Surg 1995; 60: 1275–1281. T E, A M. Pathologic assessment of myocardial cell necrosis and apoptosis after ischemia and reperfusion with molecular and morphological markers. J Mol Cell Cardiol 2000; 32: 209–224. Y K, L Y, T A, M T, L GS, Y X, H D, C MV, D JM. Rat and rabbit heart infarction: effects of anesthesia, perfusate, risk zone, and method of infarct sizing. Am J Physiol 1994; 267: H2383–H2390. S H, J G. Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa. Anal Biochem 1987; 166: 368–379. G WD, L Y, M R, M E. Intrinsic myofilament alterations underlying the decreased contractility of stunned myocardium: a consequence of Ca2+-dependent proteolysis? Circ Res 1996; 78: 455–465. M E, G WD. Stunned myocardium: a disease of the myofilaments? Basic Res Cardiol 1995; 90: 269–272. F CS, R FC. The troponin complex and regulation of muscle contraction. FASEB J 1996; 9: 755–767. MD KS, M RL, M WP. Incorporation of the Troponin regulatory complex of post-ischemic stunned porcine myocardium reduces myofilament calcium sensitivity in rabbit psoas skeletal muscle fibers. J Mol Cell Cardiol 1998; 30: 285–296. L H, M A, R N, V A H, B P, K U, K J, L S, L B, L I, M FU, N J, S W. Biochemical
22.
23.
24.
25.
26. 27.
28.
29. 30.
31.
32.
33.
34.
35.
mechanism(s) of stunning in conscious dogs. Am J Physiol (Heart Circ Physiol) 2000; 279: H176–H184. M T, C MV, D JM. Failure of N-2mercaptoproponyl glycine to reduce myocardial infarction after 3 days of reperfusion in rabbits. Basic Res Cardiol 1999; 94: 180–187. W MW, T CS, R S, Z Z-Q, V-J J. Endogenous nitric oxide (NO) protects against ischaemia-reperfusion injury in the rabbit. Cardiovasc Res 1995; 30: 79–86. MD JL, L R, P W, T MY, M S, P SC, A D, R G, V E JE. Cardiac Troponin I is modified in the myocardium of bypass patients. Circulation 2001; 103: 58–64. P Z, V J, S GJM. Calpain-I induced alterations in the cytoskeletal structure and impaired mechanical properties of single myocytes of rat heart. Cardiovasc Res 2000; 45: 981–993. M AF. Turnover of cardiac troponin subunits: kinetic evidence for a precursor pool of Troponin I. J Biol Chem 1981; 256: 964–968. M DE, A FP, M JM. Thin filament protein dynamics in fully differentiated adult cardiac myocytes: toward a model of sarcomere maintenance. J Cell Biol 1999; 145: 1483–1495. A JE, S KB, L Y, L JH, J AS. Comparable detection of acute myocardial infarction by CKMB isoenzyme and cardiac troponin I. Clin Chem 1994; 40: 1291–1295. Y K, S Y, F M, H K. Calpain is implicated in rat myocardial injury after ischemia or reperfusion. Jpn Circ J 1995; 59: 40–48. S Q, L M, Z X, Z M, K L, L S, L JP. Degradation of cardiac troponin I in serum complicates comparisons of cardiac troponin I assays. Clin Chem 1999; 45: 1018–1025. J NM, N JTA, R RL, B GC, F D, K MG, B PM, H YA, K JF. Phosphorylation specificities of protein kinase C isozymes for bovine cardiac troponin I and troponin T and sites within these proteins and regulation of myofilament properties. J Biol Chem 1996; 271: 23277–23283. V RC, K JF. Protein kinase C-mediated phosphorylation of troponin I and C-protein in isolated myocardial cells is associated with inhibition of myofibrillar actomyosin MgATPase. J Biol Chem 1993; 268: 2705–2711. T Y, C G, K DM, L Z, W H, K EG, K GL, W RA. In vivo phosphorylation of cardiac troponin I by protein kinase C decreases cardiomyocyte calcium responsiveness and contractility in transgenic mouse hearts. J Clin Invest 1998; 102: 72–78. W YG, B WJ, H J, S AM, B LA, L SL. Brief rapid pacing depresses contractile function via Ca2+/PKC-dependent signaling in cat ventricular myocytes. Am J Physiol (Heart Circ Physiol) 2000; 280: H90–H98. P P, Z J, Q Y, T X-L, M S, C X, B R. Ischemic preconditioning induces selective translocation of protein kinase C isoforms epsilon and eta in the heart of conscious rabbits without subcellular redistribution of total protein kinase C activity. Circ Res 1997; 81: 404–414.
Troponin I in Stunned Myocardium 36. U F, W PE, D SB, H KD, W AA. MDL-28170, a membrane-permeant calpain inhibitor, attenuates stunning and PKC proteolysis in reperfused ferret hearts. Cardiovasc Res 1997; 35: 60–67. 37. D L F, D T R, S F, B R, M E, S N, S S, P
411
S. Specific degradation of troponin T and I by mucalpain and its modulation by substrate phosphorylation. Biochemical J 1995; 308: 57–61. 38. F J, S BJ, F JA, L TC, C JM. Preload induces troponin I degradation independently of myocardial ischemia. Circulation 2001; 103: 2035–2037.