Cardiomyocyte Troponin T Immunoreactivity Is Modified by Cross-linking Resulting From Intracellular Calcium Overload
Background During myocardial ischemia, the increase in cytosolic Ca2+ promotes the activation of neutral proteases such as calpains. Since the troponin T subunit is a substrate for calpains, we investigated the effects of irreversible myocyte damage on troponin T immunoreactivity.
Methods and Results Hearts from adult guinea pigs (n=32) were perfused under conditions of normoxia, ischemia, postischemic reperfusion, or Ca2+ paradox. Hearts were frozen and processed for immunohistochemistry and Western blot with three anti–troponin T monoclonal antibodies. Two of these antibodies are unreactive on cryosections of freshly isolated and normoxic hearts and of hearts exposed to 30 minutes of no-flow ischemia. In contrast, reactivity is detected in rare myocytes after 60 minutes of ischemia, in a large population of myocytes after 60 minutes of ischemia followed by 30 minutes of reperfusion, and in every myocyte exposed to Ca2+ paradox. In Western blots, samples from ischemia-reperfusion and Ca2+overloaded hearts show reactive polypeptides of about 240 to 260 kD and 65 to 66 kD in addition to troponin T. A similar pattern of immunoreactivity is observed with an anti–troponin I antibody. Histochemical troponin T immunoreactivity and reactivity on high-molecular-weight polypeptides are detectable in normal heart samples after preincubation with 10 mmol/L Ca2+ or with transglutaminase, whereas they are not if either transglutaminase or calpain is inhibited.
Conclusions The evolution of the ischemic injury is accompanied by changes in troponin T immunoreactivity as a consequence of the calcium-dependent activation of both calpain proteolysis and transglutaminase cross-linking.
In the myocardium, the evolution of ischemic damage is characterized by a wide array of functional and structural alterations.1 Cardiac contraction is blocked immediately at the onset of ischemia, and a prolonged restriction of the coronary flow is associated with a progressive rise of the diastolic pressure that probably reflects ATP depletion. When the coronary flow is reestablished after prolonged ischemia (ie, postischemic reperfusion), mechanical activity is not recovered and the degree of contracture is increased. Reversible alterations of the contractility, such as myocardial stunning,2 are produced by reperfusion after a short ischemic period.
Myofibrillar protein degradation is likely to be involved in both the reversible and irreversible changes in cardiac function induced by ischemia. Intracellular proteolysis could be stimulated by the increase in [Ca2+]i, which promotes the activation of calpains. These neutral proteinases are activated in vitro by either micromolar (calpain I) or millimolar (calpain II) Ca2+ concentrations.3 Indeed, evidence has been provided that calpain activity is increased during myocardial ischemia.4 Since cardiac TnI and TnT subunits are both substrates for calpain I,5 troponin degradation could be involved in the contractile derangements caused by myocardial ischemia.
In the present study, we investigated TnT immunoreactivity of guinea pig cardiac myocytes after exposure of isolated hearts to global ischemia, followed or not followed by reperfusion, and to the Ca2+ paradox phenomenon. We used a panel of anti-TnT monoclonal antibodies that react in Western blotting with cardiac TnT but differ in the pattern of immunohistochemical staining. We show that two anti-TnT antibodies are unreactive on sections of normal heart but become reactive after 90 minutes of global ischemia or after a shorter period of ischemia when followed by reperfusion or in the presence of the Ca2+ paradox phenomenon. Comparable changes in TnT immunoreactivity are observed in normal heart sections exposed to buffers with high Ca2+ concentration, to transglutaminase, or to paraformaldehyde. Changes in immunohistochemical staining are accompanied by the appearance of immunoreactive high-molecular-mass polypeptides in Western blots that we interpret as resulting from cross-linking between troponin subunits and between troponin and other cardiac proteins.
Tissue Source and In Vivo Treatments
Six-month-old male albino guinea pigs (n=38) were used for this study. After they were killed by decapitation according to authorized procedures, hearts were excised and either frozen in liquid nitrogen for immunochemical studies (n=6) or perfused with a Langendorff apparatus for the ischemia-reperfusion experiments (n=32). Soleus and tibialis anterior muscles were also excised from 2 animals.
Perinatal hearts were obtained from 30-day-old fetuses and from 10-day-old neonates.
Perfusion of Isolated Hearts
Hearts were perfused by the nonrecirculating Langendorff technique as previously described.6 Glucose (11 mmol/L) was used as a substrate in a modified Krebs-Henseleit solution containing (in mmol/L): NaCl 115, KCl 4.0, CaCl2 1.5, MgSO4 1.1, NaHCO3 25.0, and KH2PO4 0.9. The buffer was gassed with 95% O2/5% CO2 to give a Po2 >600 mm Hg and a final pH of 7.4. This medium was delivered to the aortic cannula by means of a peristaltic pump. During normoxic perfusions, the flow was maintained at 12 mL/min for 20 minutes before ischemia and for 30 minutes after. Ischemia was induced by the complete abolition of the coronary flow. Left ventricular wall temperature was maintained at 36°C to 37°C irrespective of coronary flow by suspension of the heart in a water-jacketed chamber. For the Ca2+ paradox protocol, after 20 minutes of equilibration with the Ca2+-containing solution, the hearts were perfused for 10 minutes with a solution in which Ca2+ had been substituted with 0.25 mmol/L EGTA. This Ca2+-free solution was then replaced with the Ca2+-containing buffer for a further 30-minute period. All the perfusions were terminated by immersion of the hearts in liquid nitrogen.
Antigens and Antibodies
Monoclonal antibodies RV-C2, BN-59, and TT-2 have previously been shown to react specifically with rat and bovine cardiac TnT isoforms.7 8 9 Monoclonal antibody TI-1 reacts specifically with rat and bovine cardiac TnI subunit.10 Tropomyosin was isolated from bovine and rat hearts as previously described10 ; anti-tropomyosin antibody was purchased from Sigma Chemical Co. Anti-myosin antibody BA-D5 reacts with the slow skeletal/cardiac myosin heavy chain subunit.11 Fluorescein-conjugated and peroxidase-conjugated anti-mouse immunoglobulins were purchased from Dakopatts.
Whole homogenates of tissue samples were processed for Western blotting as described,12 with slight modifications. In brief, cryostat sections 20 μm thick were collected in Eppendorf microtubes, and electrophoresis loading buffer was added, together with antiprotease inhibitors (0.01 mol/L PMSF, 1 μg/mL pepstatin). When necessary, sections were preincubated under some of the conditions described in the following paragraph; supernatants were removed after centrifugation at 10 000g for 10 minutes at 4°C, and pellets were resuspended in electrophoresis loading buffer. Samples were boiled for 3 minutes and centrifuged for 15 minutes at 4°C. Samples were run either in an 8% or in a 4% to 10% gradient polyacrylamide gel with the acrylamide-to-bisacrylamide ratio used by Anderson and Oakeley.13 High-molecular-weight standards included commercially available preparations (BioRad) and sarcolemmal preparations of skeletal muscle.14 Comparable amounts of each sample, as determined by densitometric analysis of the actin band in Coomassie blue–stained gels, were transferred to nitrocellulose paper for 2 hours at 400 mA in the absence of methanol, and the efficiency of transfer was checked by Ponceau red staining. Blots were then saturated with BSA and incubated with appropriate antibodies, as previously described.12
Cryosections were incubated with undiluted supernatants from hybridomas for 20 minutes at 37°C. After several rinses with PBS, sections were incubated with fluorescein- or peroxidase-conjugated secondary antibodies. Peroxidase activity of bound antibodies was revealed by incubation with 3,3′-diaminobenzidine in 0.05 mol/L Tris-HCl, pH 7.6, containing 0.03% hydrogen peroxide.
In a series of sections, incubation with anti-troponin antibodies was preceded by the following pretreatments: (1) The effect of fixatives was tested by preincubation with 4% paraformaldehyde in PBS for 10 minutes at RT or acetone 95% in water for 5 minutes at −20°C. (2) The effect of detergents was tested by preincubation with Triton 0.1% in PBS for 30 minutes at RT or SDS 0.03% in PBS for 2 minutes at RT. (3) The effect of buffers at different ionic strengths was tested by preincubation with Tris-HCl buffers ranging from 0.4 to 1.5 mol/L at pH 7.5 to 8.0 for 30 minutes at 37°C; pyrophosphate 0.1 mol/L+EGTA 5 mmol/L+DTT 5 mmol/L, pH 8.6, for 15 minutes at RT; and sodium phosphate 3 mmol/L+NaCl 40 mmol/L, pH 7.0, for 15 minutes at RT15 . (4) In phosphorylation experiments, sections were incubated with 0.1 mg/mL protein kinase type I or type II for 5 minutes at 37°C in 20 mmol/L histidine+2.5 mmol/L EGTA+2.5 mmol/L magnesium sulfate+1 mmol/L DTT, pH 7.0, and in the presence of 20 mmol/L NaF+50 mmol/L ATP+2.5 mmol/L cAMP, as described.16 Parallel experiments were performed in which purified anti-TnT immunoglobulins were coincubated in the same medium. Control treatments were performed without ATP and/or protein kinase. (5) The effect of Ca2+ concentration was tested by use of propionate buffer containing 170 mmol/L potassium+2.5 mmol/L Mg+5 mmol/L ATP+5 mmol/L EGTA+10 mmol/L imidazole, pH 7.0. Free Ca2+ concentration at 10−5 and 10−7 mol/L was determined by calculation after addition of different Ca2+ amounts, as described.17 Ca2+ concentration at 10−2 and 10−3 mol/L was obtained by addition of CaCl2. Sections were incubated with these media for 30 minutes at 37°C and subsequently treated with antibodies. In parallel experiments, 1 to 4 μg/mL RV-C2 and 10 μg/mL BN-59 or TT-2 purified immunoglobulins were added to the media. Sections also were treated with media without ATP or with incubation with EGTA 5 mmol/L. (6) A propionate buffer containing 170 mmol/L potassium+2.5 mmol/L magnesium+10 mmol/L imidazole, pH 7.0, with 10 mmol/L CaCl2 added, was used for incubation of sections with 40 and 160 μg/mL human erythrocyte transglutaminase for 30 minutes at 37°C.18 Transglutaminase activity was checked by incubation of sections with 0.5 mmol/L dansylcadaverine and observation of fluorescence incorporation under a microscope. Transglutaminase is inhibited by acrylamide at very low concentrations (0.1 to 5 mmol/L), as determined by 3[H]spermidine incorporation. Endogenous and exogenous transglutaminase were inhibited by addition of 0.1 to 5 mmol/L acrylamide to the same buffer.19 Reactions were performed at 37°C for 30 minutes. Controls were incubated in the presence of the inhibitor in Ca2+-free buffer. (7) The same buffer as described in (6) was used to test the effects of the calpain inhibitor calpeptin (Z-Leu-Nle-OH, Novabiochem). Since calpeptin has to be bound to calpain to be effective, sections were incubated with 10 μmol/L calpeptin in Ca2+-free propionate buffer at 37°C for 30 minutes; 10 mmol/L CaCl2 was then added to the medium, and incubation lasted an additional 30 minutes. Controls were incubated in Ca2+-free buffer. Neither calpain I nor calpain II isolated from rat skeletal muscle is inhibited by 5 mmol/L acrylamide (E. Melloni, PhD, and F. Salamino, PhD, unpublished observations). (8) The effect of TnC extraction was achieved by incubation for 30 minutes at 37°C with 5 mmol/L EDTA/10 mmol/L imidazole, pH 7.2, as described.19
Statistical analysis was performed by the unpaired Student’s t test.
Immunoreactivity of Anti-TnT Antibodies in the Guinea Pig Heart
Western blot analysis with anti-TnT antibodies RV-C2, BN-59, and TT-2 identifies several TnT isoforms in guinea pig heart (Fig 1⇓). Four major polypeptides of 42, 41.5, 41, and 40 kD correspond to those described in rabbit heart.13 Reactivity was also observed on two faster-migrating polypeptides of about 35 kD, similar to those seen in rat heart.8 These high-mobility bands are unlikely to correspond to degradation products, since they are seen after quick freezing of the guinea pig hearts and in the presence of protease inhibitors. We have previously shown that these anti-TnT antibodies did not cross-react with GAPDH,7 9 and we have also found that they do not cross-react with purified bovine and rat cardiac tropomyosin (not shown). Anti-TnT antibodies differ in their reactivity with skeletal muscle: as described for other mammalian species,8 BN-59 and TT-2 react with cardiac and slow and fast skeletal TnT isoforms (Fig 1a⇓), whereas RV-C2 reacts only with cardiac TnT isoforms (Fig 1b⇓).
Immunohistochemical analysis reveals two distinct labeling patterns in sections of guinea pig hearts: RV-C2 labels all myocytes from atrial, ventricular, and conduction system myocardium (Fig 2a⇓); in contrast, most myocytes are unreactive with BN-59 and TT-2 (Fig 2b⇓). Rare reactive myocytes are occasionally detected in the left or right myocardial wall; they are usually concentrated in small foci and, based on the pattern of distribution, do not appear to correspond to conduction myocytes. BN-59 and TT-2 anti-TnT antibodies are also unreactive in sections of fetal and neonatal hearts (not shown); therefore, their staining pattern cannot be due to differential reactivity with fetal and adult cardiac TnT isoforms.
TnT Immunoreactivity in the Isolated Heart: Effects of Ischemia, Reperfusion, and Ca2+ Overload
Isolated guinea pig hearts were subjected to different perfusion conditions, whose effect on TnT immunoreactivity is summarized in the Table⇓ and illustrated by Fig 3⇓. The staining pattern of BN-59 and TT-2 is unchanged (ie, no reactivity) in normally perfused hearts (Fig 3a⇓) as well as in hearts exposed to no-flow ischemia up to 30 minutes. Conversely, few labeled myocytes are observed after 60 minutes of ischemia (Fig 3b⇓), and their number increases significantly after 90 minutes of ischemia (Table⇓).
A larger increase in the proportion of the myocytes reactive with BN-59 and TT-2 is observed after reperfusion of ischemic hearts (Fig 3c⇑); reactive myocytes are concentrated primarily in the middle layer of the ventricular wall, the large majority being grouped in bundles. The percentage of labeled myocytes in reperfused hearts after either 30 minutes or 60 minutes of ischemia is significantly higher than the percentage of labeled myocytes observed in hearts exposed only to ischemia (Table⇑). The effects of reperfusion appear to be time dependent: no modification of immunoreactivity is detected after 5 minutes of reperfusion, whereas a slight increase in the proportion of reactive myocytes is observed after 15 minutes of reperfusion (not shown).
Ca2+ paradox experiments show even greater effects on BN-59 and TT-C2 immunoreactivity. All cardiac myocytes become reactive in hearts perfused for 10 minutes with a Ca2+-free medium followed by 30 minutes of perfusion with normal Ca2+-containing medium (Fig 3d⇑ and Table⇑). Strong immunoreactivity is seen in a large number of myocytes as early as 5 minutes after exposure to Ca2+-containing medium and in each myocyte after 15 minutes (not shown). No change in immunoreactivity is observed in any of these conditions with RV-C2 and anti-TnI antibodies.
Western blot analysis shows that anti-TnT antibodies react with TnT polypeptides in hearts perfused with oxygenated medium and under no-flow ischemia for 60 minutes (Fig 4⇓). In contrast, in ischemic-reperfused and Ca2+ paradox–treated hearts, the anti-TnT antibodies BN-59 and RV-C2, which recognize different epitopes in cardiac TnT (Fig 1⇑), react with additional polypeptides whose molecular weights appear to be identical and correspond to about 240 to 260 kD and 65 to 66 kD (Figs 4⇓ and 5⇓). Reactive polypeptides of molecular weight lower than cardiac TnT are also observed in ischemic-reperfused and Ca2+ paradox–treated hearts (Fig 4⇓).
Western blot analysis with an anti-TnI antibody shows labeling of TnI in control and ischemic heart homogenates, whereas additional reactive bands are detected in the ischemic-reperfused and Ca2+ paradox hearts (Figs 4⇑ and 5⇑). Some of these bands, including polypeptides of about 240 kD and 66 kD, appear to be identical to those labeled by anti-TnT antibodies, whereas a reactive polypeptide of about 55 kD is not labeled by anti-TnT antibodies.
In addition to anti-TnT and anti-TnI immunoreactivities, the 240- to 260-kD polypeptides detected in the ischemic-reperfused and Ca2+ paradox hearts display immunoreactivity for myosin heavy chain (Fig 5⇑).
Calpain and Transglutaminase Are Responsible for Changes in TnT Immunoreactivity
To identify the factors that may be implicated in making the TnT epitope accessible to BN-59 and TT-2 antibodies in sections of injured myocardium, we applied various pretreatments to sections of normal guinea pig hearts before incubation with the antibodies. The protocols used include the following treatments: (1) fixatives such as paraformaldehyde,20 which alters protein conformation with covalent cross-links; (2) detergents, which improve antibody accessibility; and (3) buffers at high ionic strength or containing pyrophosphate, which are known to affect the solubility of myofibrillar components.15 Other treatments that are known to affect the conformation of the troponin complex include phosphorylation by cAMP-dependent protein kinases, exposure to different Ca2+ concentrations, and selective extraction of the TnC subunit.19 21
Among all the protocols used, only two types of treatments, ie, paraformaldehyde fixation and high Ca2+ concentration, affected the immunohistochemical staining with anti-TnT antibodies BN-59 and TT-2. After paraformaldehyde, all myocytes become reactive with BN-59 and TT-2 antibodies (not shown). No change in anti-TnT immunoreactivity is observed when the sections are pretreated with buffers at pCa >5, which includes the range of cytosolic oscillations occurring during the contraction-relaxation phases. In contrast, a dramatic change in BN-59 and TT-2 antibody immunoreactivity is observed in sections exposed to 10 mmol/L Ca2+, either in the presence or in the absence of ATP (Fig 6⇓). All myocytes become reactive after pretreatment with 10 mmol/L Ca2+ buffer, whereas no significant difference in immunoreactivity is detected in the presence of lower Ca2+ concentration (not shown). High Ca2+ concentration at neutral pH is known to activate calpain and transglutaminase.3 22 Thus, we investigated whether incubation of sections from normal guinea pig heart in the presence of high Ca2+ and of transglutaminase or calpain inhibitors, such as acrylamide and calpeptin, respectively, would affect the observed change in TnT immunoreactivity. Results are illustrated by Fig 6⇓ and show that, in the presence of high Ca2+, calpeptin and acrylamide inhibit any change in reactivity, whereas all myocytes become reactive after incubation with transglutaminase.
Western blot analysis with anti-TnT antibody BN-59 was performed on cryostat sections that were preincubated with high Ca2+ and transglutaminase or calpain inhibitors before gel electrophoresis. As illustrated by Fig 7⇓, samples exposed to high Ca2+, in the presence of very low concentrations of acrylamide (10 μmol/L) or of transglutaminase, show other reactive bands in addition to TnT, which display the same apparent molecular weight as 240- to 260-kD and 65- to 66-kD polypeptides identified in the isolated hearts after ischemia-reperfusion and Ca2+ paradox. Conversely, no additional reactive band was observed when sections were incubated with higher concentrations of acrylamide in the presence or absence of transglutaminase or with calpeptin, except for a weak reactivity at the level of the 260-kD polypeptide in the latter case.
Effects of High-Ionic-Strength Buffers on TnT Immunoreactivity
We observed that the immunoreactivity of the anti-TnT antibody RV-C2 also may be changed. This antibody, at variance with BN-59 and TT-2, stains all myocytes in control heart sections, and this staining pattern is not modified by most pretreatments, including 10 mmol/L Ca2+ buffers. However, reactivity with RV-C2 is abolished in most myocytes after preincubation with buffers at high ionic strength or containing pyrophosphate16 that extract most myofibrillar protein, including troponins (Fig 8a⇓ and 8b⇓). These pretreatments do not change the reactivity of BN-59 and TT-2 antibodies (Fig 8c⇓ and 8d⇓). Interestingly, the rare myocytes that maintain RV-C2 labeling under these conditions correspond to the myocyte population labeled by BN-59 and TT-2 antibodies in untreated sections (Fig 8b⇓ and 8c⇓). The normal pattern of staining with RV-C2 antibody is not restored when paraformaldehyde fixation is performed immediately after the high-ionic-strength buffer pretreatment (not shown). A change in immunoreactivity similar to that seen with RV-C2 is observed with an antibody specific for the TnI subunit: whereas all guinea pig myocytes are labeled by anti-TnI in control sections, only a few myocytes retain the labeling after pretreatment with high-ionic-strength medium (Fig 8e⇓ and 8f⇓). These myocytes correspond to the myocyte population labeled by the anti-TnT antibodies BN-59 and TT-2 in control sections and by RV-C2 in sections pretreated by high-ionic-strength medium. If sections of ischemic-reperfused heart are pretreated with high-ionic-strength media, the majority of myocytes maintain strong immunoreactivity with all three anti-TnT antibodies (Fig 9⇓).
This study shows that in the guinea pig heart, conditions associated with irreversible cell damage and Ca2+ overload, such as postischemic reperfusion and the Ca2+ paradox phenomenon, are characterized by changes in the pattern of immunoreactivity of two anti-TnT monoclonal antibodies, whose epitope is masked in sections of the normal heart. The unmasking of the epitope might occur as a consequence of either conformational changes, degradation, or cross-linking of the troponin proteins.
The effect of section pretreatment on the immunoreactivity pattern of anti-TnT antibodies suggests that BN-59 and TT-2 recognize the same or two different epitopes that are not accessible in control heart. In fact, (1) comparison of paraformaldehyde-fixed and -unfixed sections shows that the epitope recognized by each of the two antibodies is present in every myocyte but is normally masked in the large majority of cardiac myocytes; (2) the epitope becomes accessible in the presence of high Ca2+ concentrations; and (3) the rare myocytes that are stained by the two antibodies are resistant to treatments, such as with high-ionic-strength medium, that increase the solubilization of TnT and TnI subunits. It has been shown that differences in epitope accessibility may be related to changes in protein conformation and different protein functional states.23 24 25 During contraction, the troponin complex undergoes a conformational change initiated by the binding of Ca2+ to the TnC subunit and by the dissociation of the TnI subunit from actin. However, exposure to a physiological range of free Ca2+ concentrations did not modify the immunoreactivity with BN-59 and TT-2 anti-TnT antibodies, whereas only extremely high free Ca2+ concentrations unmasked the epitope. It is thus possible that changes in [Ca2+]i, such as those occurring in the presence of irreversible cell damage, are essential for epitope unmasking. In the heart, high [Ca2+]i causes hypercontracture, with a 60% reduction of myocyte length.26 However, it seems unlikely that unmasking of the TnT epitope results from a conformational change due to the direct effect of high [Ca2+]i on myofibrils, since this mechanism would be active after short exposure to high [Ca2+]i. In contrast, the population of immunoreactive myocytes increases progressively with the duration of the perfusion with Ca2+-containing medium in the Ca2+ paradox experiments. Thus, other events are necessary to explain the effects of high [Ca2+]i on TnT immunoreactivity.
In the isolated heart, the Western blot profile of TnT polypeptides is essentially unchanged after 60 minutes of global ischemia. Conversely, additional bands of reactivity, corresponding to higher-molecular-weight polypeptides, are detected in ischemia-reperfusion and Ca2+ paradox hearts. Furthermore, reactive polypeptides with lower apparent molecular weight than TnT are observed in ischemic-reperfused and Ca2+ paradox hearts, suggesting the possibility of limited degradation of TnT. The most striking result is the appearance of immunoreactive bands with higher molecular weight than TnTs, which are apparently labeled by both BN-59 and RV-C2 anti-TnT antibodies, which recognize an epitope common to all TnT isoforms and an epitope specific for the cardiac TnT isoforms, respectively. A likely interpretation of this finding is that the intracellular Ca2+ overload accompanying cell death after both ischemia-reperfusion and Ca2+ paradox induces cross-linking between troponin subunits or their fragments and other cardiac proteins. Indeed, Western blots with anti-TnI and anti–myosin heavy chain antibodies (Figs 4⇑ and 5⇑) suggest that the same phenomenon may involve these proteins as well. Cross-linking of cytoplasmic proteins catalyzed by tissue transglutaminase has been described as part of the biochemical pathway that leads to programmed cell death.27 Transglutaminases (Ca2+-dependent protein-glutamine γ-glutamyltransferase) are activated in the presence of an increase in [Ca2+]i, such as occurs after ischemic cell injury.22 Skeletal muscle TnT polymerizes in vitro to high-molecular-weight aggregates in the presence of Ca2+ and transglutaminase28 ; in addition, the action of transglutaminase was demonstrated on purified skeletal myosin and actin.29 Exposure of heart sections to high Ca2+ and transglutaminase modifies TnT immunohistochemistry and shows in Western blot the presence of reactivity on the same high-molecular-weight polypeptides as observed in the isolated heart samples obtained after ischemia-reperfusion and Ca2+ paradox. The effects of exogenous transglutaminase on TnT immunohistochemistry and cross-linking are inhibited by acrylamide; the same result is obtained with incubation in the presence of acrylamide alone, suggesting inhibition of endogenous transglutaminase. The possibility that transglutaminase is activated in cardiac myocytes is suggested further by the evidence that reperfusion injury induces apoptosis in the rabbit heart.30 Interestingly, cross-linked myofibrillar proteins generated by transglutaminase activity are insoluble at high ionic strength.31 Such a property might explain the disappearance of myocyte immunoreactivity of normal heart sections stained with the anti-TnT antibody RV-C2 and the anti-TnI antibody after pretreatment with high-ionic-strength buffers or pyrophosphate. Conversely, the immunoreactivity of the large majority of myocytes of ischemic-reperfused and Ca2+ paradox hearts does not appear to be affected by treatment with solutions that solubilize myofibrillar proteins.
The possibility that partial degradation of TnT plays a major role in determining the observed change in TnT immunoreactivity is suggested by the results obtained in the presence of the calpain inhibitor calpeptin. Incubation of heart sections with calpeptin inhibits the change in TnT immunoreactivity on myocytes induced by exposure to high Ca2+. Western blot shows reactivity on TnTs and absence of reactivity with the 65- to 66-kD polypeptides, whereas the 260-kD polypeptide is barely detectable. Although we showed that undegraded skeletal TnT is a substrate for transglutaminase in vitro,28 it is possible that in the heart, calpain degradation of TnT may be necessary for the polymerization or the cross-linking to other myofibrillar proteins carried on by transglutaminase suggested by the weak reactivity observed on the 260-kD polypeptide. Conversely, degradation of TnT or of other myofibrillar proteins alone does not modify the pattern of TnT immunoreactivity that occurs when transglutaminase activity is inhibited. It must be pointed out that TnT was not detected in the coronary effluent in any of the experimental conditions used in the present study by Western blot analysis (R.M., F.D.L., unpublished observations). Nevertheless, it is tempting to speculate that the aggregates may represent a sort of storage from which TnT could be released during the late phases of reperfusion. This mechanism could explain the persistence of TnT immunoreactivity in patient serum for many days after a myocardial infarction.32
In summary, our results indicate that myocyte injury after ischemia-reperfusion and Ca2+ paradox is accompanied by a change in TnT immunoreactivity, which can be attributed to the formation of high-molecular-weight products generated through the combined activity of calpain and transglutaminase.
Selected Abbreviations and Acronyms
This work was supported by grants from the Ministero dell’Università e della Ricerca Scientifica e Tecnologica (quota 40% to 60%), by the Consiglio Nazionale delle Ricerche (grants 93.00301.CT04 and 94.2424.CT04), and by Telethon-Italy (project 202). The authors thank Drs Melloni and Salamino for sharing results before publication; Dr R. Betto for advice, for the use of pCa5 and pCa7 solutions, and for the gift of sarcolemmal preparations; and Dr S. Salvatori for providing protein kinase I and II.
- Received July 20, 1995.
- Revision received November 13, 1995.
- Accepted November 21, 1995.
- Copyright © 1996 by American Heart Association
Reimer KA, Jennings RB. Myocardial ischemia, hypoxia and infarction. In: Fozzard HA, Haber E, Jennings R, Katz A, Morgan H, eds. The Heart and Cardiovascular System. New York, NY: Raven Press; 1986:1133-1201.
Braunwald E, Kloner RA. The stunned myocardium: prolonged, postischemic ventricular dysfunction. Circulation. 1982;66:1146-1149.
Toyo-Oka T. Phosphorylation with cyclic adenosine 3′:5′ monophosphate-dependent protein kinase renders bovine cardiac troponin sensitive to the degradation by calcium-activated neutral protease. Biochem Biophys Res Commun. 1982;187:44-50.
Saggin L, Ausoni S, Gorza L, Sartore S, Schiaffino S. Troponin T switching in the developing rat heart. J Biol Chem. 1988;263:18488-18492.
Saggin L, Gorza L, Ausoni S, Schiaffino S. Cardiac troponin T in developing, regenerating and denervated rat skeletal muscle. Development. 1990;110:547-554.
Zanellato AMC, Borrione AC, Saggin L, Giuriato L, Schiaffino S, Sartore S. Troponin T– and troponin I–like proteins in bovine vascular smooth muscle. Circ Res. 1991;68:1349-1361.
Saggin L, Gorza L, Ausoni S, Schiaffino S. Troponin I switching in the developing heart. J Biol Chem. 1989;264:16299-16302.
Anderson PAW, Oakeley AE. Immunological identification of five troponin T isoforms reveals an elaborate maturational troponin T profile in rabbit myocardium. Circ Res. 1989;65:1087-1093.
Luise M, Presotto C, Senter L, Betto R, Ceoldo S, Furlan S, Salvatori S, Sabbadini RA, Salviati G. Dystrophin is phosphorylated by endogenous protein kinases. Biochem J. 1993;293:243-247.
Schiaffino S, Gorza L, Saggin L, Valfré C, Sartore S. Myosin changes in hypertrophied human atrial and ventricular myocardium: a correlated immunofluorescence and quantitative immunochemical study on serial cryosections. Eur Heart J. 1984;75(suppl F):95-102.
Salvatori S, Damiani E, Barhanin J, Furlan S, Salviati G, Margreth A. Co-localization of the dihydropyridine receptor and the cyclic AMP-binding subunit of an intrinsic kinase to the junctional membrane of the transverse tubules of skeletal muscle. Biochem J. 1990;267:679-687.
Bergamini CM, Signorini M. Calcium dependent reversible inactivation of erythrocyte transglutaminase by acrylamide. Biochem Int. 1988;17:865-872.
Babu A, Sonnenblick E, Gulati J. Molecular basis for the influence of muscle length on myocardial performance. Science. 1988;240:74-76.
Franke WW, Schid E, Wellsteed J, Grund C, Gigi O, Geiger B. Change of cytokeratin filament organization during the cell cycle: selective masking of an immunological determinant in interphase PtK2 cells. J Cell Biol. 1983;97:1255-1260.
Hegmann TE, Lin JLC, Lin JJC. Motility-dependence of the heterogeneous staining of culture cells by a monoclonal anti-tropomyosin antibody. J Cell Biol. 1988;106:385-393.
Larson DM, Fujiwara K, Wayne RA, Gimbrone A Jr. Myosin in cultured vascular smooth muscle cells: immunofluorescence and immunochemical studies of alterations in antigenic expression. J Cell Biol. 1984;99:1582-1589.
Haworth RA, Hunter DR, Berkoff HA. Contracture in isolated adult heart cells. Circ Res. 1981;49:1119-1128.
Huang YP, Seguro K, Motoki M, Tawada K. Cross-linking of contractile proteins from skeletal muscle by treatment with microbial transglutaminase. J Biochem. 1992;112:229-234.
Gottlieb RA, Burleson KO, Kloner RA, Babior BM, Engler RL. Reperfusion injury induces apoptosis in rabbit cardiomyocytes. J Clin Invest. 1994;94:1621-1628.
Gard DL, Lazarides E. Specific fluorescent labeling of chicken myofibril Z-line proteins catalyzed by guinea pig liver transglutaminase. J Cell Biol. 1979;81:336-347.
Katus AH, Remppis A, Neumann FJ, Sheffold T, Diederich KW, Vinar G, Noe A, Matern G, Kuebler W. Diagnostic efficiency of troponin T measurements in acute myocardial infarction. Circulation. 1991;83:902-912.