| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1Department of Pediatrics and 2Department of Medicine, Duke University, Durham; and 3Department of Pathology, University of North Carolina, Chapel Hill, North Carolina
Submitted 10 February 2004 ; accepted in final form 25 October 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
thin filaments; transgenic mice; ventricular function; myocardium
Previous studies of the functional effects of the cTnT isoforms have used proteins in solution, reconstituted myofilaments, and cells in culture (17, 43, 49, 52, 54). Calcium-activated ATPase activity did not differ between myofilaments reconstituted with human TnT (hcTnT) cTnT3 or hcTnT4 (52) and hcTnT1, hcTnT2, hcTnT3, or hcTnT4 (17), but calcium sensitivity of S1-ATPase activity was less in the presence of the longer bovine cTnT isoform than the shorter isoform (50). In a motility assay, velocity of myofilaments constituted with bovine cTnT3 or cTnT4 did not differ in calcium sensitivity, cooperativity, maximal velocity or in maximal force (54). In Tn extracted and reconstituted fibers, cooperativity was greater in the presence of hcTnT1-containing Tn, whereas calcium sensitivity decreased progressively from hcTnT1 to hcTnT4 (17). Expression of the embryonic rat cTnT isoform (similar to hcTnT1, see Fig. 1) in adult cardiac myocytes in culture did not affect calcium sensitivity of force development (43).
|
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Transgene Construct
hcTnT1 cDNA, containing the complete coding region, was excised from a Bluescript vector (1) by digestion with XbaI and KpnI and cloned at these sites in pGEM-3Zf(–), allowing SalI restriction sites proximal and distal to the cTnT1 sequence. The hcTnT1 transcript was then cloned into the single SalI site in pGEM-5Z(–) containing an 
-myosin heavy chain (MHC) promoter-simian virus 40 (SV40) poly(A) signal [kindly provided by Dr. J. Robbins (47) and modified by Dr. W. Koch (41)]. The cDNA and its flanking sequences were sequenced to ensure proper orientation and fidelity of the sequence (1). The DNA transcript (MHC-hTnT1-SV40) was linearized with SfiI and NotI, purified, and microinjected into the pronucleus of single egg cells from Bl6SJLF1/J mice (Jackson Laboratories) and implanted into pseudopregnant foster mothers (Duke Comprehensive Cancer Center and Transgenic Facility).
Screening for Transgenic Mice
Screening mice by PCR was done using the primers 5'-TCCACATTCTTCAGGATTCTCT-3' and 5'-ATCGGGGATCTTGGGAGGC-3', which amplify the 3' portion of the -MHC promoter and the proximal portion of the hcTnT1 coding region, yielding a 560-bp amplicon. Genomic DNA was prepared from tail DNA. Southern blots were prepared by electrophoresing the DNA in 0.7% agarose, transferring it to a nitrocellulose filter, probing the filter with a 32P-labeled SV40 fragment, and exposing X-ray film to the filter.
Generation of Transgenic Mice
The founder mice were bred with C57BL/J6 mice. Three founder lines, F19, F27, and F35, were generated. F1 transgenic (TG) mice from a given founder line were crossbred. The resulting F2s were used in this study. Wild-type [non-TG (NTG)] mice were mice from the three founder lines that did not express hcTnT1.
In Vitro Myocardial Properties
Ventricular bundles [diameter 70–400 (median 180) µm, length 0.5–2.1 (median 0.93) mm] were obtained and detergent skinned as described previously (35). Briefly, the mice were anesthetized (100 mg/kg ketamine-10 mg/kg xylazine ip), and the heart was washed in physiological solution (in some preparations, the initial washing solution contained 25 mmol/l 2,3-butanedione monoxime) (13, 22, 34). The aorta was cannulated, and the heart was perfused for 3 min with physiological solution containing 2 mmol/l EGTA, followed by 3 min of skinning solution (relaxing solution containing 0.5% Triton X-100, see below). The heart was kept on ice, in skinning solution for 30 min, and then in relaxing solution, and a preparation was obtained for study.
Relaxing (pCa 9) and activating (pCa 4.5) solutions contained 10 mmol/l EGTA, 30 mmol/l BES, 30 mmol/l Na+, 12 mmol/l phosphocreatine, 15 U/ml creatine phosphokinase, 2.5 mmol/l pMgATP (3.0 pMg), 2% polyvinylpyrrolidone, and 100 µg/l pepstatin A, with pH 7.10 adjusted with KOH and an ionic strength of 190 mmol/l (adjusted with KCl). The compositions of these solutions were calculated using a computer program based on the study of Fabiato and Fabiato (15). The apparent Ca-EGTA stability constant (log10K = 6.568) included corrections for temperature and ionic strength (21). Solutions of various pCa were prepared by mixing appropriate volumes of pCa 9 and pCa 4.5 solutions (15).
The preparations were placed in the 30-µl cuvette (temperature controlled at 22°C) of a Güth OPT1M system (20). The force transducer had a resonance frequency of 715 Hz. Data were recorded using Digidata 1200 analog-to-digital board and pCLAMP 6 software (Axon Instruments) using a sampling frequency of 100 Hz for the force versus pCa curve and 2 kHz for the rate constant of tension redevelopment (ktr) (see below). One preparation was studied from each mouse. The preparation was stretched to a sarcomere length of 2.2–2.3 µm measured using light diffraction (preparations that did not produce a diffraction pattern were excluded from the analysis). The dimensions of the muscle (free length and diameter) were recorded.
Calcium sensitivity of tension.
Test solutions (see above) were applied in steps of increasing [Ca2+], and the developed force was recorded. pCa 9 was reapplied, usually at every third step, to assess resting tension and allow later adjustment for any baseline drift. We used a release-restretch protocol (7) during the acquisition of the force versus pCa curve. Nonlinear regression was used to fit the force (F)-pCa data with the Hill equation, 
, where nH is the Hill coefficient, Fmax is the maximum force obtained at saturating calcium, and Ca50 is the [Ca2+] required for half Fmax (Fig. 2).
|
|
Cardiac catheterization.
Cardiac catheterization was performed as reported previously (32, 41, 42). Briefly, mice were anesthetized (100 mg/kg ketamine-2.5 mg/kg xylazine ip) and ventilated (room air, 0.3–0.5 ml; rate 104 breaths/min), and their body temperature maintained at 
37°C. After a bilateral vagotomy, a 1.4-Fr micromanometer catheter (Millar Instruments; Houston, TX) was inserted into the aorta and left ventricle (LV) via the right carotid artery, and isoproterenol was infused through a polyethylene-50 catheter into the jugular vein. LV pressure (LVP) was recorded at baseline and after isoproterenol infusions of 50, 500, and 1,000 pg (41). The data were recorded using data-acquisition software (Sonometrics) and analyzed using pCLAMP software (Axon Instruments). The maximum rates of the rise and fall of LVP (dP/dtmax and dP/dtmin; mean of 10 consecutive beats) and the half-time of isovolumic relaxation [T1/2; the time required for the pressure at dP/dtmin to drop to half its value (33), mean of 4 consecutive beats] were determined. The heart was excised and weighed. Part of the LV was fixed in 2% glutaraldehyde or 4% paraformaldehyde, and the remainder was frozen in liquid nitrogen or placed in sample buffer.
Echocardiography.
Mouse hearts were imaged using an Acuson Sequoia C256 Imaging System (Mountainview, CA) and a 15L8Tx linear transducer at 13 MHz. The mice were anesthetized (45 mg/kg etomadate-50 µg/kg fentanyl ip), and body temperature was maintained at 37°C. The LV was imaged from the parasternal short-axis view at the level of the papillary muscles (sweep speed 100–200 mm/s). M mode was used to measure LV end-systolic and end-diastolic dimensions (EDD and ESD) and LV posterior wall and ventricular septal thickness (45). Pulsed-wave Doppler spectral tracings of the mitral inflow and LV outflow were obtained from an apical four-chamber view with the sample volume placed at the papillary muscle tips.
Myocardial Structure
LV longitudinal and cross sections were stained with hematoxylin-eosin and Masson’s trichrome. Morphometric data were captured using a Nikon Microphot FXA microscope/Optronics DEI-750 charge-coupled device video camera system and Apple Power Macintosh G3 computer using a Scion CG-7 capture card and analyzed using NIH Image. LV myocardium was processed for electron microscopy (31), and myofibril organization and sarcomere structure were examined.
Identification and Quantitation of Sarcomere Proteins
Myocardial proteins were resolved with SDS-PAGE using a modified Laemmli protocol (3). hcTnT1 expression was examined by loading two to five lanes with different amounts of myocardium. Staining of the gels and Western blots were performed as described previously (2, 26). Two-dimensional (2-D) gels were run using Invitrogen Zoom Strips (pH 4–7) for the first dimension. The cTnT isoforms were identified on Western blots using monoclonal antibody (MAb) 13-11, which recognizes a cTnT-specific epitope conserved across species (26). The relative amounts of the cTnT isoforms were determined, using densitometric analysis, for each mouse as a fraction of total cTnT. The expression of total TnT was compared with the expression of TnI and tropomyosin in TG and NTG mice using Western blots stained with MAb 3350 and Sigma T9283 for TnI and tropomyosin, respectively.
The isoforms of TnI, tropomyosin, myosin light chain (MLC)1, and MLC2 were resolved in 8.5%, 9.5%, and 12.5% acrylamide gels using SDS-PAGE as described previously (3). For cTnI and MLC2, 2-D gels were run using Invitrogen Zoom Strips (pH 3–10 and pH 4–7, respectively) for the first dimension. The myosin heavy chain (MHC) isoforms were resolved after the method of Warren and Greaser (55) and stained with Invitrogen "SimplyBlue Safestain." The Western blots were probed with cTnI-specific MAb 3350 (4); MAb 3309, which recognizes cTnI and ssTnI (4); MAbs 3A8 and 10C6, which are specific for MLC1 and MLC2 isoforms (12); and a MAb (Sigma CH1, T9283) specific for striated muscle tropomyosin.
Statistical Analysis
Except where indicated, values are given as means ± SE. The measures of LV and myofilament function were evaluated as a function of the percent hcTnT1 expression using linear regression, and the results are reported as the change in the variable (slopes ± SE) for an increase in hcTnT1 expression of 10%. Force and stiffness were corrected for cross-sectional area of the preparations by regression. Except as noted, we found no significant effects of age or sex. Statistical analysis was carried out using R software (38) and package NLME (36) for repeated measurements.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
We examined cTnT isoform expression in C57BL/6 mouse hearts at 15 and 17 days of gestation and in neonatal and adult mice. Western blots probed with MAb 13-11 demonstrated four bands (Fig. 4, A and B), numbered mouse cTnT1–4 (mcTnT1–4) to correspond to the four hcTnT isoforms (2). In the fetal mouse heart, mcTnT1 and mcTnT2 were the most highly expressed isoforms. In the neonate, mcTnT1 and mcTnT2 expression decreased and mcTnT3 and mcTnT4 increased. In the adult mouse heart, only mcTnT3 and mcTnT4 were expressed (Fig. 4A). This pattern of cTnT isoform expression in the mouse heart is consistent with combinatorial alternative splicing of two exons of cTnT cDNA (25) yielding four isoforms, cTnT1, cTnT2, cTnT3, and cTnT4 (Fig. 1); the charges of the peptides encoded by these two exons are –2 and –8 in the mouse and –2 and –7 in the human.
|
The ranges of hcTnT1 expression in the three founder lines, F19, F35, and F27, were 1–7%, 4–16%, and 14–34% of total cTnT expression (see also Table 1 and Fig. 5). There were no significant difference between NTG and TG mice in total cTnT-to-TnI (P = 0.7) and cTnT-to-tropomyosin (P = 0.4) ratios. In the TG mice, the expression of hcTnT1 correlated negatively with the expression of mcTnT4 but appeared to be independent of the expression of mcTnT3 (Fig. 5, A and B). hcTnT1 was more negatively charged than mcTnT3 or mcTnT4 (Fig. 6). The 2-D electrophoretic profiles were similar, suggesting that there were no differences in posttranslational modifications of mcTnT3 and mcTnT4 (Fig. 7).
|
|
|
|
-Tropomyosin was the only tropomyosin isoform recognized in NTG and TG hearts (n = 12; Fig. 8). -MHC was the only MHC isoform expressed in NTG and TG ventricular myocardium (Fig. 8). cTnI, but not ssTnI (n = 45), and only the ventricular isoforms of MLC1 and MLC2 (n = 4) were expressed in NTG and TG ventricular myocardium (Fig. 8). mcTnT, cTnI, and MLC2 were resolved into similar patterns for NTG and TG myocardium using 2-D gel electrophoresis (Figs. 7 and 8).
|
Table 1 gives a summary of the characteristics of mice from the three founder lines. TG mice appeared normal and healthy. LV myocardium of NTG and TG mice had normal morphology (Fig. 9), and examination under the electron microscope demonstrated normal sarcomere and myofibril structure. Morphometric examination of LV myocardium (3 sections/mouse) from NTG and TG mice showed no significant difference in the percentage of myocardium made up of myocytes: 98.3 ± 1.3% for NTG (mean ± SD, n = 5 mice) and 96 ± 5.4% for TG (n = 7). Myocyte cross-sectional area (100 cells/mouse) was also similar: 363 ± 30 µm2 for NTG (mean ± SD, n = 5) and 374 ± 73 µm2 for TG (n = 6).
|
Force. At 0% hcTnT1 expression (NTG), Fmax was 25.9 ± 2.4 (n = 10) vs. 16.5 ± 1.2 mN/mm2 (n = 15) for mice expressing >20% hcTnT1 (P = 0.011, Welch test). With the use of all the data, regression analysis gave a similar result: Fmax decreased significantly with increasing hcTnT1 expression (R2 = 0.65, n = 30, P = 0.019; see Fig. 10A).
|
Maximum rate of redevelopment of tension. ktr was sensitive to calcium (57) (Fig. 2), but the ktr versus pCa relation was not significantly affected by hcTnT1 expression: mean pCa for one-half ktr,max for NTG mice was 5.63 ± 0.03 and did not change significantly with hcTnT1 expression (slope –0.01 ± 0.02/10% increase in hcTnT1, R2 = 0.014, n = 15, P = 0.7). At saturating calcium, mean ktr,max was 10.0 ± 0.8 s–1 at 0% hcTnT1 expression. Its increase, by 0.57 ± 0.38 s–1/10% increase in hcTnT1, was not significantly different from 0 (R2 = 0.086, n = 26, P = 0.15; Fig. 10D).
In Vivo Myocardial Function
Echocardiography. Fractional shortening (FS) was calculated from the measured dimensions as FS = (EDD – ESD)/EDD. Mean FS at 0% hcTnT1 expression (NTG) was 27.6 ± 2.2%. FS increased by 2.0 ± 1.5%/10% increase in hcTnT1 expression (Fig. 11); the increase was not significantly different from 0 (R2 = 0.10, n = 19, P = 0.19). The E wave-to-A wave ratio (E/A; Fig. 12) was also not significantly affected by hcTnT1 expression: at 0% hcTnT1, E/A was 1.09 ± 0.12, and it increased by 0.028 ± 0.076 for a 10% increase in hcTnT1 expression (R2 = 0.014, n = 12, P = 0.72). Heart rate (mean ± SD) under etomadate-fentanyl anesthesia was 470 ± 67 beats/min.
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
cTnT Isoforms
A developmental shift in expression of TnT isoforms to those with shorter and less negatively charged amino-terminal regions occurs in cTnT, fsTnT, and ssTnT and in indirect flight muscle TnT (1, 5, 19, 23, 27). The similarity in this shift across muscle types strongly suggests that alterations in TnT isoform expression contribute to the plasticity of muscle function.
Of the four developmentally regulated mcTnT isoforms, the larger, more negatively charged isoforms are expressed early in development, whereas the smaller, less negatively charged isoforms are expressed in the adult. Combinatorial alternative splicing of two 5' cTnT exons provides the basis for four cTnT cDNAs in the human and other mammals (1, 19). Three previously cloned mouse cDNAs variably contain those two 5' exons (25). As in hcTnT1, they encode two negatively charged peptides with charges of –2 and –8. The successively less negative isoelectric points of mcTnT1, mcTnT2, mcTnT3, and mcTnT4 (Fig. 6) suggest that these isoforms are products of combinatorial alternative splicing of these two exons.
hcTnT1 is expressed in the fetal heart and in infants with congenital heart defects (2, 44). It is absent from the normal adult human heart and is reexpressed in the failing adult heart (2, 53). The proximity of the cTnT amino-terminal region to the overlap of adjoining tropomyosin molecules may enable cTnT isoform amino-terminal differences to modulate the Tn-tropomyosin calcium-regulated switch where activation with calcium results in movement of tropomyosin and exposure of myosin binding sites on actin leading to strong cross-bridge formation (18, 56, 58).
Functional Consequences of Transgenic Expression of hcTnT1
Correlations between the force-pCa and myofibrillar ATPase-pCa relations and cTnT isoform expression during development have been studied (29, 35, 46). For example, in vitro studies of rabbit neonatal myocardium demonstrated a positive correlation between cTnT2 expression and the sensitivity of the myofilaments to calcium (35). A review of the effects of development on cTnT isoform expression, which involves a change in expression from longer, more acidic cTnT isoforms (cTnT1 and cTnT2) to shorter, less acidic ones (cTnT3 and cTnT4), indicates that the calcium sensitivity of both the myofibrillar ATPase activity (29) and force (14, 40) decreased with development. Because the adult mouse expresses only cTnT isoforms that are smaller and less negatively charged than hcTnT1, we expected that hcTnT1 expression in the adult mouse would result in an increase in the sensitivity of the myofilaments in vitro.
In a recent study, Gomes et al. (17) described the results of in vitro experiments in which they replaced endogenous cTnT in pig myocardial fibers with each of the four recombinant human isoforms and show a nearly linear relationship between pCa50 and charge (hcTnT1 being the most negatively charged isoform). Using their values and assuming complete cTnT replacement in their preparations, we can calculate, based on charge, the expected difference in pCa50, pCa50(NTG – TG), between NTG and TG mice expressing 24.3% hcTnT1 (the mean expression level of founder line F27) as –0.042. This expected difference is well within the 95% confidence interval (–0.084 to +0.117) for the difference in pCa50 obtained from our mice.
Our in vitro measurement of force in the skinned preparation showed a decrease of Fmax with increased hcTnT1 expression. The three-state model of thin filament activation of McKillop and Geeves (30) provides a possible explanation for the apparent effect of hcTnT1 expression on Fmax. According to the model, the thin filaments are in a blocked, closed, or open activation state. In the blocked position, cross-bridge formation is prevented; in the closed state, myosin can weakly bind actin; and in the open state, myosin can strongly bind actin. In the presence of saturating calcium, 20% of the actin sites are in the open state and 80% in the closed state (30). Different cardiac and skeletal muscle TnT fragments have different effects on thin filament activation (28, 51). The sequence differences between these fragments, which are in the TnT amino-terminal region, and their relation to the overlap of adjoining tropomyosin molecules may modulate thin filament activation. If increased hcTnT1 expression causes a shift in the average position of tropomyosin towards the closed state, a decrease in Fmax could be expected.
The in vivo study demonstrated the effects of increased hcTnT1 expression on cardiac function: dP/dtmax and –dP/dtmin fell and T1/2 increased. When we omitted the NTG mice and analyzed only the data from the TG mice (n = 24), we reached the same conclusions: dP/dtmax decreased at a rate of 1,120 ± 490 mmHg/s per 10% increase in %hcTnT1 (P = 0.030), –dP/dtmin decreased by 980 ± 360 mmHg/s per 10% hcTnT1 (P = 0.013), and T1/2 increased by 0.54 ± 0.24 ms for a 10% increase in hcTnT1 (P = 0.034). These data are consistent with the effect of hcTnT1 expression on Fmax and the force-pCa relation. We found a decrease in Fmax with no change in pCa50; such a decrease would be expected to scale the force-pCa relation downward and result in a decrease in dP/dtmax and –dP/dtmin, assuming other properties, e.g., cytosolic calcium concentration as a function of time during the contraction-relaxation cycle, remains constant.
The in vivo variables could have been affected by preload or afterload and subject to other potential compensatory changes, e.g., in cytosolic intracellular [Ca2+]. When we controlled statistically for end-diastolic pressure (as a surrogate for preload), the changes in dP/dtmax as a function of hcTnT1 expression remained the same. Although the Western blots of the 2-D gels showed similar patterns of resolution of mcTnT, mcTnI, and MLC2, differences in phosphorylation of contractile and calcium-regulating proteins could have contributed to the differences observed between NTG and TG hearts (9, 48). The effect of isoproterenol on the physiological variables, however, did not depend on the hcTnT1 expression level, leading us to speculate that the in vivo baseline level of cAMP-dependent phosphorylation is similar in NTG and TG myocardium. In our in vitro study, both sarcomere length and [Ca2+] were under our control, and therefore the differences in Fmax are unlikely to be due to differences in fiber length or [Ca2+]. The effects of hcTnT1 expression observed are unlikely to be due to abnormal structure because we saw no evidence of morphometric or structural differences between NTG and TG mice.
cTnT Isoform Charge and Function
To explore the potential effects of differences in charge resulting from the expression of hcTnT1, we refitted the regression models using charge in place of hcTnT1 expression and obtained similar results. The effect of charge brought about by hcTnT1 expression on Fmax was statistically significant, as were the slopes of dP/dtmax, –dP/dtmin, and T1/2 as a function of charge. The expression of hcTnT1 in TG mice appeared to be at the expense of mcTnT4 and not mcTnT3 (Fig. 5), leading to the potential ambiguity that the functional effects can be ascribed to either an increase in hcTnT1 or a decrease in mcTnT4 expression. However, ascribing these functional effects to differences in charge eliminates this ambiguity and at the same time invokes a more general concept.
In summary, our in vivo and in vitro results suggest that a change in cTnT isoform expression has a modulatory role in cardiac function and are consistent with the interpretation that cTnT amino-terminal charge and length may be the basis of this effect. Although human and mouse cTnT are highly homologous, they are not identical. Therefore, it is possible that some of the functional changes may have resulted from sequence differences other than the 15-residue peptide in the amino-terminal region of hcTnT1. We were interested in hcTnT1 because levels of hcTnT1 expression of 26–30% of total cTnT are present in the human fetal heart [scans of previously published Western blots (2)], and levels up to 15% are present in infants with severe congenital cardiac defects that require surgery (44). Our findings suggest that hcTnT1 expression at these levels affects myofilament function. This effect of cTnT1 could be potentially protective of the heart by decreasing its steady-state level of contractility whereby increasing its functional reserve. Future studies will assess whether hcTnT1 expression protects against or exacerbates the effects of pathological hemodynamic loading on survival and ventricular function.
Because the expression of hcTnT1 was almost exclusively at the expense of expression of mcTnT4 (Fig. 5), it is not possible to distinguish whether the effects were due to an increase in hcTnT1 or a decrease in mcTnT4. However, inasmuch as hcTnT1 and mcTnT4 differ in charge, the effect of the cTnT isoforms on ventricular and myofilament function can be stated in terms of a net increase in cTnT negative charge.
|
GRANTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
-myosin heavy chain promoter. We thank the Duke Comprehensive Cancer Center Transgenic Facility for the technical assistance in generating the transgenic founder lines. We are grateful to Dr. Scott Buck for many valuable suggestions. |
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||
; F35, 
; and F27, 
. The relative expression of hcTnT1 in TG mice appeared to have been at the expense of the dominant isoform, mcTnT4 (A), with no effect on the relative expression of mcTnT3 (B).
-tropomyosin, slow skeletal TnI (ssTnI), or atrial MLC1 expression. Western blots were probed with Sigma T-9283, which recognizes striated muscle tropomyosin, MAb 3309, which recognizes cTnI and ssTnI (
, NTG; 
