|
|
||||||||
Articles |
1
Human Safety Department, 500 Miami Valley Laboratories, The Procter and Gamble Co., P.O. Box 538707, Cincinnati, OH 45253.
2
Division of Laboratory Medicine, Department of
Pathology, Washington University School of Medicine, St. Louis, MO
63110.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Cardiac troponin I (cTnI) is a strong candidate as a cardiac biomarker in animals. cTnI is a sensitive and persistent indicator of cardiac injury in humans (8)(12)(13), with high tissue specificity in the presence of marked skeletal muscle injury (6)(7)(8) and chronic renal failure (6)(8)(9)(14). Additionally, cTnI expression has not been detected in fetal or diseased skeletal muscle (15), although cTnT and CK-MB may be upregulated with chronic muscle injury (16)(17), and CK-MB is upregulated with endurance training (18)(19). Increases in serum cTnI after myocardial injury are persistent; the diagnostic window after myocardial infarction in humans is at least 9 days for cTnI (20)(21), which is longer than increases of CK-MB (20)(22) and comparable with cTnT, although some evidence suggests that cTnT remains increased longer (22)(23). Accordingly, we tested a cTnI immunoassay developed for diagnosis of cardiac injury in humans for its usefulness across a wide range of species.
| Materials and Methods |
|---|
|
|
|---|
Biopsy specimens of ~1 g were collected from heart and skeletal muscle within 5 min after animals were killed by barbiturates. The left ventricle was used for cardiac studies. Biopsies were obtained from muscles of the upper limbs for mammals, from pectoral muscles for birds, and from trunk muscles for fish for skeletal muscle studies. Heart and skeletal muscle biopsy specimens were obtained from six dogs of mixed breed, three Holstein calves, three horses (muscle biopsy obtained from two), three sheep, three Yorkshire pigs (heart biopsy obtained from two), three New Zealand rabbits (heart biopsy obtained from two), three SpragueDawley rats, three mice, three turkey poults (heart biopsy obtained from two), three chickens (used for egg production), and three rainbow trout. Tissue was immediately blotted free of blood, trimmed of obvious connective tissue and fat, and ultrafrozen in liquid nitrogen. Samples were stored temporarily in polypropylene vials at <-70 °C until analyzed. This temperature preserves cTnI in myocardium for at least 1 year (data not shown). Samples for analysis were shipped on solid CO2 from Guelph in Ontario to the Division of Laboratory Medicine, Washington University School of Medicine, St. Louis, MO.
Human tissue samples were also analyzed with the use of a protocol approved by the Institutional Review Board at Washington University. Informed consent was obtained from each patient before enrollment in the study. Five heart samples were obtained (three from the Washington University Autopsy Service and two from explanted failing hearts) and three skeletal muscle samples were obtained from patients undergoing back surgery. The cardiac explants and skeletal muscle biopsy specimens were obtained fresh; however, at least several hours elapsed between the time of death and attainment of the samples from the autopsy service. Accordingly, reactivity in these latter samples may underestimate cTnI content, although this underestimation is assumed to be similar for failing and nonfailing hearts because of similar delay times and processing procedures. These samples were also stored at <-70 °C until analysis.
assays
Tissue (~100300 mg) was weighed and minced. Ten volumes of
homogenization buffer containing 75 mmol/L Tris, pH 8.0, 8 mol/L urea,
and 1.0 mmol/L CaCl2 were added. Homogenization was
accomplished with the use of a hand-held rechargeable homogenizer with
a generator bore diameter of 5 mm (Omni International, Warrenton, VA)
for two 30-s bursts and an additional burst if necessary. After the
homogenates were incubated for 30 min at 4 °C to extract the cTnI,
they were centrifuged at 20 000g for 1 h.
To qualitatively confirm the tissue specificity of the cTnI assay in different species, Western blot analysis was performed with the use of the two cTnI-specific monoclonal antibodies of the assay (2B1.9 and 2F6.6) and a third monoclonal antibody (3C5.10) that recognizes both cTnI and skeletal muscle troponin I (sTnI) from human and dog (20). Homogenates were prepared from tissue samples as described above and diluted 50-fold in gel buffer containing sodium dodecyl sulfate. Of this, 10 µL was applied to 16.5 g/L polyacrylamide gels containing sodium dodecyl sulfate (26). After electrophoretic separation, proteins were transferred to nitrocellulose and incubated first with the anti-TnI monoclonal antibodies and then with goat anti-mouse antibodies conjugated with alkaline phosphatase. Color was developed with 5-bromo-4-chloro-3-indolylphosphate and nitroblue tetrazolium (BCIP/NBT Phosphate Substrate System, Kirkegaard and Perry) as substrates. Purified canine cTnI and sTnI was prepared from canine heart and skeletal muscle (27) and used as controls.
For measurements of tissue reactivity in the cTnI assay, supernatant of centrifuged homogenates was removed and diluted with pooled normal human sera, 100-fold for homogenates of muscle and fish hearts and 20 000-fold for homogenates of mammalian and avian hearts. The concentration of cTnI was determined with a research-application, sandwich enzyme immunoassay (Stratus I, Dade International) in which two cTnI-specific murine monoclonal antibodies (20) are used, 2F6.6 as the capture antibody and 2B1.9 as the labeling antibody. Human cTnI was used in the calibrators, its concentration determined with use of a Bradford protein assay (Bio-Rad). Because this method overestimates cTnI concentrations fivefold compared with the Lowry assay (28) (data not shown), the tissue reactivities are fivefold greater than would be obtained by the current, commercially available cTnI immunoassay (Stratus II, Dade International) (29).
calculations and statistical analysis
Tissue reactivity in the cTnI assay was defined as the
concentration of human cTnI that gave the equivalent reactivity. Data
were analyzed with the use of Instat 2.0 software and graphed with
Prism (GraphPad, San Diego, CA) and are reported as mean ± SE.
For comparisons of 3 or more mean values, one-way ANOVA was used. Data
were log10-transformed before statistical analysis when
standard deviations were >10-fold different according to Bartlett's
test for homogeneity of variances. When the F statistic
indicated between-group differences in a comparison, the
StudentNewmanKeuls multiple comparisons test was performed. To test
for differences when there were only two groups, the two-sided,
unpaired Student's t-test was used, or if Bartlett's test
indicated different standard deviations, Welch's alternative
t-test was used. A decrease in tissue reactivity with heart
failure was tested with the use of a one-sided, unpaired Student's
t-test.
For comparative purposes, data were grouped into large mammals (dogs, calves, horses, sheep, and pigs), small mammals (rabbits, rats, and mice), birds, and fish. Differences between groups were considered significant for P <0.05. For the species in which the TnI sequence was known, tissue reactivity in the cTnI immunoassay was compared with the degree of homology of that TnI with human cTnI (30).
| Results |
|---|
|
|
|---|
|
The human cTnI concentrations equivalent to the reactivities of heart
and skeletal muscle tissue from each species studied are listed in
Table 1
. They are ordered according to decreasing reactivities in
hearts. Two outlying values, one from a rat heart (10.9 mg/g) and the
other from a dog with heart failure (3.3 mg/g), were excluded. These
values were >3 SD from the mean of the remainder of their group. The
identity of these samples was thought to have been interchanged because
of a transcription error. In hearts, the mean reactivity varied across
species by >100-fold; skeletal muscle showed at most a fourfold
difference between species.
|
The reactivities of hearts in the cTnI assay were ~10-fold greater for each species of mammals than for each species of birds and 6- to 10-fold greater for each species of birds than for fish. Mammals could be divided into two subgroups that differed widely in body size and had similar cTnI reactivities. One subgroup comprised dogs, calves, horses, and sheep; the other subgroup comprised laboratory animals (rabbits, rats, and mice). Values for pigs were intermediate to those for the two other groups of mammals but, for statistical analyses and figures, were grouped with values for other large mammals because of their size.
Figure 2
indicates the reactivities of cardiac (top panel) and skeletal
muscle (middle panel) in the cTnI immunoassay for the four groups of
animals, large and small mammals, birds, and fish. Cardiac reactivity
was significantly different between groups (P <0.001).
Large mammals had reactivity 1.7-fold greater than small mammals,
10-fold greater than birds, and 100-fold greater than fish. In
contrast, there was no significant difference between mammals and birds
in skeletal muscle reactivity (Fig. 2
, middle panel), although in fish
it was 50% lower than in other species (P <0.0004). No
other difference between groups was detected for muscle cTnI.
|
The reactivity of hearts relative to muscle varied substantially across
groups (Fig. 2
, lower panel). In large mammals, the reactivity was
2-fold greater than in small mammals, 8-fold greater than in birds, and
42-fold greater than in fish.
In Fig. 3
the relationship between homology of the TnI gene sequence with
human cTnI is indicated for those species for which this information
was available (30); log reactivity is plotted against the
percentage homology for a large overlapping segment of the protein
sequence comprised of ~200 amino acids (of a possible 209). Sequence
homologies with human cTnI are 96.4% for bovine heart, 92.8% for
mouse and rat heart, 91.4% for rabbit heart, 71.5% for chicken heart,
and 57.1% for human skeletal muscle (30). This sequence
homology and log tissue reactivity were sigmoidally related
(r2 = 1, P <0.05).
|
Compared with healthy control dogs, cardiac reactivity was 32%
decreased in the dogs with heart failure (P <0.002; Fig. 4
). Cardiac reactivity of dogs with heart failure was not
significantly different from that of small mammals.
|
Mean reactivity of the three nonfailing human hearts was 5.7 ± 0.6 mg/g, similar to that observed for small mammals. Reactivity in the two failing human hearts was 4.02 ± 1.0 mg/g, a decrease of 30% compared with the nonfailing hearts (P = 0.09). Reactivity of skeletal muscle was 8.4 ± 2.7 µg/g, not significantly different from that of other mammals. Heart/skeletal muscle ratio of reactivities was estimated as 680 ± 75 and was similar to that for small mammals.
| Discussion |
|---|
|
|
|---|
The differences across phyla for cardiac reactivity in the cTnI
immunoassay can be attributed primarily to species differences in the
primary structure of the protein (30) (Fig. 3
). In
addition, the differences among mammals likely also result, in part,
from differences in myofibril content. Myocardial cTnI content of large
mammal species (9.43 ± 1.71, n = 5) was 1.6-fold greater
(P <0.01) than that of small mammal species (5.83 ±
0.22, n = 3). Compared with large mammals, the volume fraction of
cardiac myocytes composed of myofibrils is less for small mammals,
although the volume fraction of mitochondria is greater, apparently
because of a greater metabolic rate (31).
In ventricular and myocyte remodeling during the development of heart failure, a moderate reduction in the content of myofibrils, of which cTnI is a component, occurs (32). In Doberman pinscher dogs with heart failure, the 32% lower (than healthy controls) cTnI concentration can be attributed to lower myofibril content of myocardium (25) and matches the 30% lower concentration of cTnT found in myocardium of the same breed of dogs with the same cardiomyopathy (25). Myocardial content of cTnI was also decreased by 30% in failing human hearts, compared with nonfailing hearts.
Estimation of the effectiveness of potential cardiac biomarkers in animals requires consideration of their assay reactivity and tissue selectivity, especially since the primary limitation of the use of existing cardiac biomarkers in animals is their relatively low activity and lack of selectivity for cardiac vs skeletal muscle (1)(2)(3). Release of skeletal muscle proteins in association with animal handling and stress can severely restrict the value of biomarkers with low cardiac selectivity (1)(2)(3).
Although high cardiac reactivity and specificity reveal cTnI to be an excellent candidate biomarker of cardiac injury in mammals, they indicate that cTnI would be ~100-fold and ~5000-fold less effective in birds and fish, respectively. In fish, as a consequence of this low cardiac reactivity and specificity and also because of small heart to muscle mass ratio, skeletal muscle injury should release more apparent cTnI into blood than should cardiac injury. However, in birds, the cardiac reactivity and specificity of cTnI may be sufficient for its effective use as a biomarker of cardiac injury if skeletal injury is only mild.
Comparison of the results of this study with a similar one conducted by one of us (P.J.O'B.) with the use of cTnT (second-generation assay, Boehringer Mannheim) (33) suggests that the cTnI immunoassay has ~10-fold greater heart to skeletal muscle ratio of reactivities than the cTnT immunoassay and 10-fold greater heart activity for mammals relative to other phyla than the cTnT immunoassay. Skeletal muscle had ~1% of the reactivity of cardiac muscle in the cTnT immunoassay (33), but in this study we find only 0.050.1% cardiac muscle reactivity in the cTnI immunoassay. Also, cardiac concentration of cTnT varied by <10-fold across phyla (33), whereas cTnI concentration varied by up to 100-fold. Although not specifically tested in this study, these differences in reactivity of the different species in the cTnI compared with the cTnT immunoassays support the proposal that the immunoassays recognize different proteins and that cross-reactivity of cTnT in the cTnI immunoassay is minimal and would not restrict its use.
limitations
Several limitations should be considered when the results of this
study are interpreted. Although high cardiac reactivity and heart to
skeletal muscle ratio of reactivities identify cTnI as a candidate
biomarker of cardiac injury, other factors may affect its
effectiveness. For example, species variations may exist in the
kinetics of cTnI release into the blood with various forms of cardiac
injury and in the kinetics of clearance from the blood that could
restrict its use (3). However, on the basis of the wide
diagnostic window of cTnI in humans, it seems likely that the duration
of increased cTnI following cardiac injury in other mammals, and in
birds, will be sufficient to provide valuable diagnostic and prognostic
information.
In conclusion, our data identify cTnI as the most specific cardiac biomarker described to date for mammals and birds and as a candidate for the specific and sensitive diagnosis and prognosis of cardiac injury in these phyla.
| Acknowledgments |
|---|
| Footnotes |
|---|
1 Nonstandard abbreviations: CK-MB, cardiac isoenzyme of creatine kinase; cTnI, cardiac troponin I; cTnT, cardiac troponin T; sTnI, skeletal muscle troponin I. ![]()
| References |
|---|
|
|
|---|
The following articles in journals at HighWire Press have cited this article:
![]() |
R. Tunca, M. Sozmen, H. Erdogan, M. Citil, E. Uzlu, H. Ozen, and E. Gokce Determination of cardiac troponin I in the blood and heart of calves with foot-and-mouth disease J Vet Diagn Invest, September 1, 2008; 20(5): 598 - 605. [Abstract] [Full Text] [PDF] |
||||
![]() |
V. Gunes, G. Atalan, M. Citil, and H. M. Erdogan Use of cardiac troponin kits for the qualitative determination of myocardial cell damage due to traumatic reticuloperitonitis in cattle Vet Rec., April 19, 2008; 162(16): 514 - 517. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Wang, J. Radhakrishnan, I. M. Ayoub, J. D. Kolarova, D. M. Taglieri, and R. J. Gazmuri Limiting sarcolemmal Na+ entry during resuscitation from ventricular fibrillation prevents excess mitochondrial Ca2+ accumulation and attenuates myocardial injury J Appl Physiol, July 1, 2007; 103(1): 55 - 65. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. S. Willis, J. A. Snyder, R. H. Poppenga, and D. G. Grenache Bovine cardiac troponin T is not accurately quantified with a common human clinical immunoassay J Vet Diagn Invest, January 1, 2007; 19(1): 106 - 108. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. B. Walker Serum Chemical Biomarkers of Cardiac Injury for Nonclinical Safety Testing Toxicol Pathol, January 1, 2006; 34(1): 94 - 104. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. M. Crandell and W. A. Ware Cardiac Toxicity From Phenylpropanolamine Overdose in a Dog J. Am. Anim. Hosp. Assoc., November 1, 2005; 41(6): 413 - 420. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. B. Wallace, E. Hausner, E. Herman, G. D. Holt, J. T. Macgregor, A. L. Metz, E. Murphy, I.Y. Rosenblum, F. D. Sistare, and M. J. York Serum Troponins as Biomarkers of Drug-Induced Cardiac Toxicity Toxicol Pathol, January 1, 2004; 32(1): 106 - 121. [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |