Cardiac Troponin T Isoform Expression Correlates With Pathophysiological Descriptors in Patients Who Underwent Corrective Surgery for Congenital Heart Disease
Background This study examined cardiac troponin T (cTnT) isoform expression in patients who had undergone surgery at Duke University Medical Center (Durham, NC) between December 1, 1993, and January 31, 1995, to correct congenital heart defects. The human heart expresses four cTnT isoforms (cTnT1 through cTnT4) whose sequence differences result from combinatorial alternative splicing of two exons. We have previously shown that cTnT4 is expressed at higher levels in severely failing hearts from transplant patients. In this study, we tested the hypothesis that congenital heart defects that have a more negative effect on myocardial function increase cTnT4 expression. We used the presence or absence of drug treatment for heart failure or congested circulation before surgery and the duration of inotropic support after corrective surgery as indicators of the pathophysiological state of the heart just before surgery.
Methods and Results Right atrial appendage tissue was collected from 34 patients, 6 days to 35 years old (median age, 3.4 months). The amounts of the cTnT1 through cTnT4 isoforms, measured as a percentage of total cTnT, were determined from Western blots probed with MAb13-11, a cTnT-specific monoclonal antibody. We found that cTnT4 expression correlated positively with the duration of inotropic support and was higher in patients who received drug treatment before surgery than in those who did not. Furthermore, we found that the percent of cTnT4 was significantly higher in hearts with congenital defects that caused congestive failure than in hearts with tetralogy of Fallot.
Conclusions These findings suggest that in patients with congenital cardiac defects, cTnT4 expression is modulated by heart failure and is increased in hearts that are more hemodynamically stressed.
Cardiac troponin T is an essential component of the system that regulates cardiac contraction and the sensitivity of cardiac myofilaments to calcium. Mammalian hearts express multiple cTnT isoforms.1 2 3 4 5 6 These isoforms are products of combinatorial alternative splicing of the primary transcript of the cTnT gene.2 3 That these isoforms have functional significance has been suggested by findings that correlate their expression with myofibrillar ATPase activity,4 the sensitivity of myofilaments to calcium,7 the sarcomere length dependence of myofilament sensitivity to calcium,8 and myofilament binding of calcium.9
The human heart expresses four cardiac TnT isoforms (cTnT1 through cTnT4) whose sequence differences result from the splicing of a 30-nt and a 15-nt exon, both of which encode highly acidic peptides.4 10 (In our initial studies of human expression of cTnT, we examined adult ventricular tissue and identified only two isoforms.4 These isoforms were named cTnT1 and cTnT2. Subsequently, we identified two additional isoforms in fetal human tissue, and cTnT1 and cTnT2 were then renamed cTnT3 and cTnT4, respectively.) cTnT1 contains both the 10- and 5-residue peptides; cTnT2, the 10-residue peptide; cTnT3, the 5-residue peptide; and cTnT4, neither.10 As in other mammals, in the human cTnT isoform expression is developmentally regulated: cTnT1 is expressed at a high level in the fetal heart and at a very low level in the adult heart.4 8 Severe heart failure also has been shown to affect human cTnT isoform expression: failing hearts from transplant patients showed increased cTnT4 expression.4 This change in cTnT isoform expression with heart disease is supported by the recent findings that cTnT isoform expression is altered in the hypertrophied failing guinea pig heart11 and the diabetic rat heart.8
This study is the first to examine the effect of congenital cardiac defects on cTnT isoform expression. Based on our previous observation that cTnT4 expression is increased in the severely failing human heart4 and the subsequent observation in the guinea pig by Gulati et al11 that heart failure secondary to a surgically produced coarctation of the aorta resulted in increased expression of two cTnT isoforms with electrophoretic mobilities similar to cTnT4, we proposed the hypothesis that cTnT4 expression is increased in hearts in which pathophysiological states have had a negative effect on myocardial function. Although atrial tissue was used in this study (ventricular tissue was available in only a few patients), we expected to see changes in atrial cTnT isoform expression because the hemodynamic loading found in many congenital defects affects the atrium. Moreover, the negative effects of hemodynamic loading on ventricular function also could have global effects on cardiac cTnT isoform expression. Because there were no noninvasive measures of function that are independent of the various ventricular loading conditions found in congenital cardiac defects, we used the postoperative ability of the heart to recover from surgery as indicated by the duration of IS and the presence or absence of drug treatment before surgery as measures of the preoperative condition of the heart.
Tissue was collected from 34 patients (17 women, 17 men) undergoing open heart surgery to correct congenital heart defects between December 1, 1993, and January 31, 1995, at Duke (NC) University Medical Center. The patients and/or their families gave informed consent as part of the surgical consent in accordance with institutional guidelines. Tissue was obtained from the right atrial appendage when an atriotomy was performed for insertion of the venous cannula for cardiopulmonary bypass. The tissue was immediately placed in liquid nitrogen and stored at −170°C for future analysis. Age at the time of surgery ranged from 6 days to 35 years (median, 3.4 months). All patients underwent complete repair of their congenital cardiac defects and had no significant residual defects as assessed by postbypass transesophageal and epicardial echocardiography.12 13
The durations of cardiopulmonary bypass, cross-clamp, and circulatory arrest times for the patients were within expected ranges: 33 to 227 minutes (median, 105 minutes) and 13 to 93 minutes (median, 47 minutes) for cardiopulmonary bypass and cross-clamp time, respectively. Of the 34 patients, 16 underwent circulatory arrest for 5 to 45 minutes (median, 26 minutes). The postoperative durations of IS ranged from 0 to 8 days (median, 3 days); ventilator support, 0 to 20 days (median, 3 days); pediatric intensive care unit stay, 1 to 29 days (median, 5 days); and hospital stay, 4 to 44 days (median, 10 days).
Quantification of Troponin T Isoforms
The tissue was placed in sample buffer,1 and the proteins were resolved with SDS-PAGE with the method of Laemmli14 with these modifications: 30% acrylamide and 1.1% bis-acrylamide stock solutions were used in 7.5% running gels and a 3.3% stacking gel. cTnT isoforms were identified by use of Western blots probed with the cTnT-specific monoclonal antibody MAB 13-1115 as primary antibody and an alkaline phosphatase-labeled secondary antibody. Nitro blue tetrazolium and 5-bromo-4-chloro-3-indolyl-phosphate (Promega Corp) were used as color reagents. Western blots were scanned with an LKB laser densitometer (Pharmacia LKB), and the area under the cTnT densitometric waveform was measured (Fig 1⇓). The relative amount of each cTnT isoform was calculated as a percentage of total cTnT.4
The preoperative status of drug therapy; the postoperative duration of intravenous IS, mechanical ventilation, and intensive care unit and hospital stay; age at time of surgery and diagnosis; cross-clamp, cardiopulmonary bypass, and circulatory arrest times; and postbypass transesophageal and epicardial echocardiography results were obtained from a review of the medical records. Proteins from each tissue specimen were run on two separate gels, and the percent isoform content was determined as detailed above. The average of the two values thus obtained was used in the data analysis. Parameter estimates are given, followed by ±SE. We used nondirectional significance tests.
cTnT4 Expression Versus Recovery Time From Surgery
We studied the relationship between the level of expression of cTnT4 and the pathophysiological state of the heart at the time of surgery. We used two indicators of the state of the heart: the requirement for drug treatment for heart failure or a congested circulation (characterized by tachypnea and hepatomegaly) with a diuretic or digoxin before surgery and the length of recovery time after surgery. We used the presence or absence of drug therapy before surgery to indicate the extent of hemodynamic loading imposed by the pathophysiological state and the inability of the heart to sustain this load. The recovery time after surgery, as measured by the duration of IS, indicated how the myocardium was affected by the defect and therefore the ability of the heart to recover from the stress of surgery. We used linear regression to examine the relationships between IS and drug therapy and the level of expression of cTnT4, while the effects of sex and age of the patient at the time of surgery were controlled for: % cTnT_|<|4|>||<|=|>|B_|<|0|>||<|+|>|B_|<|1|>||<|\cdot|>|age|<|+|>|B_|<|2|>||<|\cdot|>|IS|<|+|>|B_|<|3|>||<|\cdot|>|drug|<|+|>|B_|<|4|>||<|\cdot|>|sex|<|+|>|B_|<|5|>||<|\cdot|>|age|<|\cdot|>|drug where B0 is a constant, age is the natural logarithm of the patient's age in days at the time of surgery, IS is the duration of inotropic support in days, drug is an indicator variable for drug therapy before surgery (drug=0 for absence of therapy, drug=1 for therapy), and sex is an indicator variable for the sex of the patient (sex=0 for male, sex=1 for female). B1 was −0.30±0.47; B2 was 1.13±0.47; B3 was 24.2±8.4; B4 was 3.61±1.5; and B5 was −4.19±1.7. Multiple R2 for the regression was .49. Our conclusions remained the same when we substituted either the duration of ventilator support or the length of intensive care unit stay for IS.
The results show that percent cTnT4 correlated positively with the duration of postsurgery IS (P=.023 for B2; see Fig 2⇓), and cTnT4 expression was higher in patients who required treatment with a diuretic or digoxin before the operation (P=.007 for B3). We interpret these results to indicate that cTnT4 increased with the severity of heart disease in these patients.
Effect of the Type of Congenital Defect on cTnT4
If cTnT4 increases as a result of a negative effect of the congenital defect on the myocardium as our data suggest, we reasoned that patients with little or no evidence of congestive heart failure would not express elevated levels of cTnT4. We therefore hypothesized that cTnT4 would be lower in TOF patients than in patients with defects that often cause congestive heart failure. We divided the data (see the Table⇓) into three roughly equal age intervals (based on a logarithmic scale): 0 to 3 weeks (n=7), 3 weeks to 18 months (n=21), and 1.5 to 36 years (n=6). We also divided them by diagnosis: TOF (n=7), transposition of the great arteries (n=6), VSD (n=8), AVSD (n=6). A fifth group, other (n=7), was made up of patients with a wide variety of diagnoses that were all associated with volume loading. The “other” group comprised patients with total anomalous pulmonary venous return (n=2), truncus arteriosus (n=1), double-outlet right ventricle (n=1), secundum atrial defect (n=1), anomalous origin of the right pulmonary artery from the ascending aorta (n=1), and transposition of the great arteries with a VSD (n=1).
The Table⇑ shows the mean percent cTnT4 for the various age and diagnostic groups. All the patients with TOF were in age group 2. We compared in age group 2 the mean percent cTnT4 in the TOF group with the average of the means of the VSD, AVSD, and other groups: cTnT4 in the TOF group was 4.6% compared with 10.4% for the other three groups (P=.018). This result is consistent with our hypothesis that cTnT4 expression increases as a result of the negative effects of increased hemodynamic loading, eg, secondary to left-to-right shunting found in the presence of a VSD or AVSD. Comparison of percent cTnT4 between the TOF and TGV groups reveals no apparent effect of cyanosis on cTnT4 expression.
The human heart expresses four cTnT isoforms that differ by the combinatorial alternative splicing of a 30-nt and a 15-nt exon, which encode highly acidic peptides.10 Expression is developmentally regulated1 2 3 4 and has been found to be affected in adult transplant patients with severe heart failure who demonstrated increased expression of the smallest isoform cTnT4.4 On the basis of this finding, we proposed that cTnT4 expression will be increased in patients whose myocardial function is negatively affected by the hemodynamic load imposed by a congenital cardiac defect.
Because there is no direct noninvasive measure of myocardial function that is unaffected by the different loading conditions in congenitally malformed hearts, we tested the hypothesis by using indirect measures. One measure was how well the heart coped with the surgical intervention that corrected the abnormality, as exemplified by the duration of postsurgery IS. We assumed that the hearts that took longer to recover after an operation were sicker at the time of the operation. By our hypothesis, we expected cTnT4 from tissue removed at the time of surgery to correlate positively with the duration of postoperative recovery. Another measure was whether the patient was undergoing therapy with a diuretic or digoxin before surgery. We assumed that the sicker patients required drug therapy, and the expectation was that percent cTnT4 would be higher in patients who were on drug therapy. Our analysis included only those patients who underwent surgical correction of their defects and had an excellent surgical result as assessed by epicardial or transesophageal echocardiography in the operating room.12 13 These criteria were aimed at removing the effect of postoperative persistent abnormal hemodynamic states on recovery from surgery.
We used the duration of IS to test our hypothesis because it is relatively more specific as a measure of the cardiovascular system than the other variables. IS consisted of a continuous infusion of dopamine with other agents (eg, epinephrine and milrinone) as necessary. When we repeated the analysis but substituted the duration of ventilator support or the duration of the stay in the intensive care unit for the duration of IS, the conclusions remained the same. The durations of IS, ventilator support, and intensive care unit stay were determined by the staff physicians in the pediatric intensive care unit on the basis of their assessment of the patient and the adequacy of cardiac output. This determination and the hospital course were completed before the cTnT assays were performed and therefore were not subject to bias by knowledge of cTnT isoform expression. We showed that cTnT4 expression was correlated with the duration of IS or equivalently of ventilator support or intensive care unit stay after surgery. We also showed that percent cTnT4 was higher in those patients who were on drug therapy before surgery than those who were not.
These findings led to our testing whether cTnT4 expression is affected by the extent of the negative effects on ventricular function caused by the congenital defect. In particular, whether TOF patients, who underwent surgery because of hypercyanotic episodes or increasing levels of cyanosis, had lower cTnT4 expression than patients in the other groups with defects usually was associated with a greater hemodynamic load and signs of heart failure or a congested circulation as characterized by tachypnea and hepatomegaly. We found a lower expression of cTnT4 in TOF patients compared with the VSD, AVSD, and other groups (Fig 3⇓). In terms of the effect of cyanosis on cTnT4 expression, the lower expression of cTnT4 in TOF patients compared with the TGV patients suggests that cyanosis is not the cause of the lower level of cTnT4 expression in TOF patients. These results are consistent with the hypothesis that cTnT4 expression is increased in response to processes that affect myocardial function.
Effect of Age on cTnT Isoform Expression
The focus of this study was the effect of pathophysiological states on the expression of cTnT4 and not the effect of postnatal development on isoform expression. The study was not and could not be designed to examine the effect of age on isoform expression. In each patient, the timing of the age at which the tissue was obtained was a consequence of the natural history of the congenital defect and its pathophysiological effects. Consequently, the analysis of the effect of age on cTnT expression may be complicated by the effects of those preoperative variables that affected recovery from surgery.
These considerations aside, we found cTnT1 to be more prominent in the neonatal atrium and to decrease with age early during development (see Fig 4⇓). This decrease in cTnT1 appears to be at the expense of increasing cTnT3. Our description of cTnT1 isoform expression in the human right atrium in this group of patients is consistent with its higher expression in human fetal ventricle than in the adult heart, in which it either is not expressed at all or is expressed at a very low level.4 8 The cTnT2 isoform was undetectable in the atrium, as might be anticipated from its very low expression in the fetal and adult human hearts.4 10 The dominant isoform in the atria at all ages was cTnT3 as had been found in the human fetal and adult ventricles.4 10
The relationships we observed between cTnT4 and recovery from surgery and therapy before surgery and between cTnT1 and age suggest that cTnT isoform expression may modulate or be modulated by myocardial function: small changes in the relative amounts of other muscle proteins, β and α myosin heavy chains, have been shown to exert a very large effect on cross-bridge mechanical interactions.16 Correlations have been found among cTnT isoform expression, isoform sequence differences, and various measures of cardiac myofilament function in other studies.4 7 8 9 17 The dominantly expressed cTnT3 isoform and the cTnT4 isoform differ by only a negatively charged five-amino-acid peptide in the amino terminal region.10 Two bovine cTnT isoforms also have been shown to differ by a five-amino-acid peptide, which is negatively charged.18 In a contractile protein assay of myofibrillar ATPase activity, the bovine isoform lacking the peptide with a sequence that is similar to human cTnT4 was found to increase the sensitivity of ATPase activity to calcium.17 Although this finding suggests that an increase in cTnT4 is an adaptive response to an abnormal hemodynamic setting rather than the basis for abnormal myocardial function, the significance of human cTnT isoform expression in myofilament function remains to be established.
Human atrial cTnT isoform expression had not been examined previously. In this study, we have shown that atrial cTnT4 expression correlates with the duration of postoperative IS, an indicator of the pathophysiological state of the heart at the time of corrective surgery, and is higher in patients who require medical treatment before surgery. Our study suggests that congenital cardiac defects are associated with differences in the expression of the cTnT isoforms. Whether this is a global effect on cTnT gene expression in all the chambers of the heart and whether this improves or impairs heart function remain to be established.
The work was supported by NIH grants HL-20749 and HL-42250, and The Gustavus and Louise Pfeiffer Research Foundation.
Selected Abbreviations and Acronyms
|AVSD||=||atrioventricular septal defect|
|cTnT||=||cardiac troponin T|
|TOF||=||tetralogy of Fallot|
|TGV||=||transposition of great vessels|
|VSD||=||ventricular septal defect|
- Received October 4, 1995.
- Revision received January 4, 1996.
- Accepted January 15, 1996.
- Copyright © 1996 by American Heart Association
Anderson PAW, Moore GE, Nassar RN. Developmental changes in the expression of rabbit left ventricular troponin T. Circ Res. 1988;63:742-747.
Cooper TA, Ordahl CP. A single troponin T gene generates embryonic and adult isoforms via developmentally regulated alternative splicing. J Biol Chem. 1985;260:11140-11148.
Jin J-P, Lin JJ-C. Rapid purification of mammalian cardiac troponin T and its isoform switching in rat hearts during development. J Biol Chem. 1988;263:7309-7315.
Anderson PAW, Malouf NN, Oakeley AE, Pagani ED, Allen PD. Troponin T isoform expression in humans: a comparison among normal and failing adult heart, fetal heart, and adult and fetal skeletal muscle. Circ Res. 1991;69:1226-1233.
Spach MS, Dolber PC, Anderson PAW. Multiple regional differences in cellular properties that regulate repolarization and contraction in the right atrium of adult and newborn dogs. Circ Res. 1989;65:1594-1611.
Greig A, Hirschberg Y, Anderson PAW, Hainsworth C, Malouf NN, Oakeley AE, Kay BK. Molecular basis of cardiac troponin T isoform heterogeneity in rabbit heart. Circ Res. 1994;74:41-47.
Nassar R, Malouf NN, Kelly MB, Oakeley AE, Anderson PAW. Force-pCa relation and troponin T isoforms of rabbit myocardium. Circ Res. 1991;69:1470-1475.
Akella AB, Ding XL, Cheng R, Gulati J. Diminished Ca2+ sensitivity of skinned cardiac muscle contractility coincident with troponin T-band shifts in the diabetic rat. Circ Res. 1995;76:600-606.
McAuliffe JJ, Gao L, Solaro RJ. Changes in myofibrillar activation and troponin C Ca2+ binding associated with troponin T isoform switching in developing rabbit heart. Circ Res. 1990;66:1204-1212.
Anderson PAW, Greig A, Mark TA, Malouf NN, Oakeley AE, Ungerleider RM, Allen PD, Kay BK. Molecular basis of human cardiac troponin T isoforms expressed in the developing, adult, and failing heart. Circ Res. 1995;76:681-686.
Malouf NN, McMahon D, Oakeley AE, Anderson PAW. A cardiac troponin T epitope conserved across phyla. J Biol Chem. 1992;267:9269-9274.
Tobacman LS, Lee R. Isolation and functional comparison of bovine cardiac troponin T isoforms. J Biol Chem. 1987;262:4059-4064.