Abstract
The present study was undertaken to characterize myocardial lesions in the rat induced by low doses of isoproterenol (Iso) and to correlate lesion severity with release of cardiac troponin T (cTnT) and changes in myocyte iNOS expression. Two types of cardiac injury patterns were observed. A Type I response, noted 3 or 6 hours postdosing with 8, 16, 32, or 64 μg/kg Iso, included potential reversible myocardial alterations associated with slight increases in serum cTnT (< 0.3 ng/mL) and a slight reduction in myocyte cTnT immunoreactivity. The second type of response noted 3, 6, 12, 24 or 48 hours postdosing with 125, 250, or 500 μg/kg Iso consisted of irreversible myocyte alterations, together with significant increases in serum cTnT (3–14 ng/mL) and a marked reduction of cTnT immunoreactivity. By 48 hours the hearts of rats dosed with 125–500 μg/kg Iso had developed interstitial fibrosis, and serum cTnT had declined to near control levels (0.06–0.18 ng/mL). Increases in iNOS immunoreactivity correlated with the lesion severity. These findings suggest that low doses of Iso exert complex effects on the myocardium and that the generation of NO through increased expression of iNOS could be an important factor in the pathogenesis of myocyte injury.
Introduction
Isoproterenol (Iso) has been used as a model compound to induce infarct-like lesions in the rat and various other animal species (Rona 1985). The lesions are most prevalent in areas of the heart that are most susceptible to ischemia (Rona et al. 1963). The Iso-induced myocardial alterations are similar in certain respects to those occurring in human beings following a myocardial infarction (Wexler and Greenberg 1978). It is thought that the β adrenergic cardiostimulatory activity exerted by Iso increases cardiac oxidative metabolism to a level that exceeds the amount of oxygen available to the myocyte through the unobstructed coronary circulation. The energy imbalance, in conjunction with a number of complex biochemical (altered calcium flux, stimulation of the adenyl cyclase system, aggregation of platelets, and formation of reactive oxygen species) (Van Vleet et al. 2002) and structural changes (alterations in membrane permeability) (Boutet et al. 1976; Todd et al. 1980), appear to contribute to the pathogenesis of the myocyte damage (Rona 1985). The area of the heart most susceptible to hypoxia caused by tachycardia appears to be the left ventricular subendocardium (Balazs et al. 1986; Van Vleet et al. 2002). Recently, changes in iNOS expression have also been associated with Iso-induced cardiotoxicity (Sun et al. 2005; Liu et al. 2005). Myocyte damage observed following exposure to Iso includes both apoptosis and necrosis (Goldspink et al. 2004).
Cardiac troponin T (cTnT) and I (cTnI) have been shown to be specific and sensitive biomarkers of drug-induced myocardial cell injury in animals and humans (Bertsch et al. 1997; Wallace et al. 2004). We (Herman et al. 1998, 1999, 2001) and others have found that the clinical immunoassay for serum cTnT can be used in experimental situations. In rats, a number of studies have described a relationship between the serum levels of cTnT or cTnI and the severity of Iso-induced cardiotoxicity (Bleuel et al. 1995; Bertinchant et al. 2000; Herman et al. 2006). However, the use of anti-cTnT and/or anti-cTnI antibodies to examine the relationship between changes in cardiac tissue cTnT and/or cTnI immunoreactivity and the extent of microscopically detected myocardial lesions has not been reported.
Many morphological and biochemical features of the lesions observed following administration of Iso have been described and/or characterized. However, most of these evaluations were carried out in studies that used relatively high single or multiple doses of Iso and few time points (Rona 1985). The present study was initiated to examine in detail the characteristics of lesions induced by very low doses of Iso. Histopathological evaluation and immunohistochemical staining were used to monitor Iso-induced lesion progression and to explore the possible roles of apoptosis and iNOS in the pathogenesis of the cardiac alterations.
Materials and Methods
Animals
Male Sprague-Dawley rats (8 weeks old) were acclimated 1 week before the experiments, placed in transparent polycarbonate cages, housed in an environmentally controlled room, and given Certified Purina Rodent Chow and water ad libitum. The protocol was approved by the Institutional Animal Use Committee of the Center for Drug Evaluation and Research, FDA. Animal care procedures for the study conformed to the 1996 ILAR (Institute of Laboratory Animal Resources) Guide for the Care and Use of Laboratory Animals (National Research Council 1996).
Experimental Procedures
Rats received a single subcutaneous (sc) injection of saline or Iso (8–500 μg/kg). After dosing, rats were euthanized by exsanguination under isoflurane anesthesia at different time points: 3 hours (8, 16, 32, or 64 μg/kg); 6 hours (32 or 64 μg/kg); and 3, 6, 12, 24, or 48 hours (125, 250, or 500 μg/kg).
Immunoassay for cTnT
Blood samples were centrifuged immediately after collection, and the sera were frozen at −80° C until assayed. Serum concentration of cTnT was monitored in the laboratory of Dr. Nader Rifai (Children’s Hospital, Harvard Medical School, Boston, MA) by immunoassay (Elecsys 1010 system, Roche Diagnostics, Indianapolis, IN).
Pathologic Studies
At necropsy, entire hearts were collected from control and Iso-treated rats. Two portions of the bisected hearts were fixed in 10% neutral formalin, embedded in paraffin, and sectioned at a thickness of 5 μm. Individual sections were stained with H&E, Movat’s pentachrome, Masson trichrome, or Gomori’s hexamine silver (American HistoLabs, Inc., Gaithersburg, MD).
Grading System for Myocardial Necrosis and Apoptosis
Histopathological evaluation of acute cardiac lesions was performed in sections stained with H&E, Movat’s (for necrosis and fibrosis), Masson’s (for necrosis and fibrosis), and Gomori’s stains (for myocardial basement membranes [BM]), and Terminal deoxynucleotidyl transferase (TdT)-mediated modified dUTP Nick End Labeling (TUNEL) assay (for apoptosis). Myocardial degeneration, interstitial inflammation, and interstitial fibrosis were observed in the hearts of Iso-treated rats. In addition, myocyte necrosis, apoptosis, and cell membrane alterations were also observed. These latter 3 types of alterations are believed to be major factors associated with the Iso-induced lesions and thus were used as the basis for assessing lesion severity on a semiquantitative scoring scale of 0–5:
0 = normal myocardial cells;
1 = scant number of myocardial cells (< 5%) showing necrosis, partial absence of myocardial BM, a few apoptotic cells (one or two cells/high magnification field in an affected area);
2 = 5–15% of myocardial cells showing necrosis, partial absence of myocardial BM, no more than four apoptotic cells per field;
3 = multiple foci of necrotic myocardial cells (16–25%), partial absence of myocardial BM, up to five apoptotic cells/field;
4 = myocardial cells (26–35%) showing necrosis in confluent areas, marked loss of myocardial BM, up to six apoptotic cells/field;
5 = more than 35% of myocardial cells showing necrosis in multiple massive or coalescent areas (which also show hypercontraction bands and myofibrillar loss), complete absence of myocardial BM, seven or more apoptotic cells/field).
Immunohistochemical Studies
TUNEL assay and indirect immunoperoxidase staining for cTnT, cTnI, and iNOS were carried out in the sections of formalin-fixed, paraffin-embedded cardiac tissues.
For the detection of cardiac apoptosis, sections were mounted onto glass slides coated with amino-propyl-triethoxysilane and assayed with the CardioTACS in situ apoptosis detection kit (Trevigen, Inc., Gaithersburg, MD). The assay is based on DNA end labeling using TUNEL, which is subsequently detected using the TACS Blue Label detection system (Lovelace et al. 1996).
The procedures for immunoperoxidase staining for cTnT and iNOS were described previously (Zhang et al. 2006). For the detection of immunoreactivity of cTnT, cTnI, and iNOS, sections were mounted onto glass slides coated with poly-L-Lysine. Pretreatment of tissue sections for cTnT, cTnI, and iNOS immunostaining was performed with microwave irradiation in a pressure cooker with antigen retrieval Glyca solution (for cTnT and cTnI) or antigen retrieval Citra solution (for iNOS). The contents of Glyca solution are mainly glycine and hydrochloric acid (pH 3–4), and Citra solution is composed of 10 mM sodium citrate and 0.05% Tween 20 (pH 6.0) (BioGenex, San Ramon, CA). After microwave treatment, slides were cooled in the solution for 20 minutes and then rinsed with distilled water. To block endogenous peroxidase activity, sections were incubated with 0.3% hydrogen peroxide in methanol for 30 minutes and then with 5% normal horse serum for 30 minutes. Sections were incubated overnight at 4°C with the primary mAb against cTnT/MCA470 (clone: T1/16), cTnI/MCA1208 (clone: 110) (Serotec, Inc. Raleigh, NC), or iNOS (BD Biosciences, San Diego, CA) at a dilution of 1:100. After washing with phosphate buffered saline (PBS), the sections were incubated with a biotinylated second antibody (Vector Laboratories, Burlingame, CA) for 1 hour and then incubated with avidin-biotinylated horseradish peroxidase complex (Vector) for 30 minutes. The peroxidase reaction was carried out with 0.05% 3’3’-diaminobenzidine in 0.1 M Tris-HCl buffer and 0.01% hydrogen peroxide for 5 minutes. Finally, sections were counterstained with hematoxylin.
For negative control staining, the primary mAb (cTnT, cTnI, iNOS) was omitted from the incubation step or replaced by the isotype mouse IgG1 (Serotec, Inc., Raleigh, NC, Cat# MCA 1209) for cTnT mAb, mouse IgG2a (Serotec, Inc., Raleigh, NC, Cat# MCA 1210) for cTnI mAb, and mouse IgG2a (BD Biosciences, San Jose, CA, Cat# 550339) for iNOS mAb at a dilution of 1:100.
Statistics
Kruskal-Wallis rank test (nonparametric analysis of variance) was applied to determine the significance of differences among cardiac lesion scores. The Tukey-Kramer multiple comparison test was used to assess the significance of differences among the groups for cTnT serum data. A p value ≤ .05 was considered significant.
Results
Histopathological Findings
Control Rats (saline, 3, 6, 12, 24, or 48 hours)
All of the saline-treated rats exhibited normal histology (Figure 1A), a scant amount of inconspicuous interstitial fibrous materials (Figure 1B), and completely enclosed myocardial BM (Figure 1C).
Type I Response (8–64 μg/kg Iso for 3 or 6 hours)
Treatment with the lowest doses of Iso caused minimal to undetectable changes in cardiac tissues. No myocyte alterations were detected in the hearts 3 hours after dosing with 8 or 16 μg/kg Iso. Treatment with doses of 32 or 64 μg/kg Iso caused alterations consisting of minimal myofibrillar loss, cytoplasmic vesicles, minimal inflammatory cell infiltration (few leukocytes and lymphocytes), and a small amount of interstitial fluid. Hypercontraction bands and partial loss of myocardial BM were observed rarely in individual cells. At 3 hours, 3 of 19 rats given 32 or 64 μg/kg Iso had minimal lesions, whereas at 6 hours these doses induced minimal or mild lesions in 13 of 15 rats.
Type II Response (125–500 μg/kg Iso for 3–24 hours)
Alterations induced by 125–500 μg/kg Iso included severe and extensive myocardial and interstitial lesions that were located primarily in the left ventricle, septum, papillary muscle, and the apex of the heart. Subendocardial necrosis was more severe in left ventricular papillary muscles than in any other subendocardial locations. The cardiac lesions varied with treatment duration and doses. The early changes included myocyte necrosis (few myocytes with hypercontraction bands) (Figure 1D), numerous leukocytes, fluid accumulation in the interstitial spaces (Figure 1E), and a partial loss of myocardial BM in areas with necrotic myocytes (Figure 1F). As the lesions progressed, myocardial necrosis became more extensive (Figure 1G), as did interstitial fibrosis (delicate collagen fibers) (Figure 1H) and BM alterations (Figure 1I). Numerous macrophages were observed in the necrotic areas (Figures 1H and 1I).
Forty-eight hours after treatment, a large amount of dense connective tissue (thick groups of collagen fibrils) was present in the existent necrotic areas. At this time, a few necrotic myocytes (without myofibrils) remained detectable (Figures 1K and 1L). In contrast, myocytes from control rats had myofibrils interspersed with few interstitial collagen fibrils (Figure 1J).
Immunohistochemical Findings
Detection of Myocardial Apoptosis
In saline-treated rats, no myocyte nuclei were labeled by the TUNEL assay (Figure 2A).
Type I Response (8–64 μg/kg Iso for 3 or 6 hours)
Slightly stained nuclei could be detected in rare single cells, but morphological signs of apoptosis such as nucleus condensation, cell shrinkage, and hypereosinophilic cytoplasm were not found.
Type II Response (125–500 μg/kg Iso for 3–48 hours)
Slightly stained nuclei and cytoplasmic shrinkage in a few myocytes were barely observed in the hearts 3 hours after treatment with 125μg/kg Iso (Figure 2B). At the time of maximal cardiac alterations (250 or 500 μg/kg Iso, 3 hours), a large number of apoptotic myocytes were detected, and both necrotic and apoptotic myocytes were arranged side-by-side or end-to-end (Figures 2C–2E). Twelve hours after treatment, the number of apoptotic myocytes decreased (Figure 2F). At 24 hours post-treatment, apoptotic myocytes were rarely seen at the site of overwhelming necrosis (Figure 2G). By 48 hours, no apoptotic myocytes were detected; however, a few small blue spots, suggestive of apoptotic bodies, were noted in inflammatory areas (Figure 2H).
Immunoreactivity of cTnT and cTnI
The myocardial immunoreactivity of the anti-cTnT mAb was comparable to that of the anti-cTnI mAb in saline-treated rats. The mAbs against cTnT and cTnI stained the regularly arranged cross-striations (I bands) (Figures 3A, 4A, and 4a). The anti-cTnI mAb, but not the anti-cTnT mAb, also stained the intercalated disks, and staining intensity for cTnI was stronger than for cTnT (Figures 4A and 4B). Changes in the intensity of immunoperoxidase staining for cTnT or cTnI in the hearts of Iso-treated rats were noted earlier than observable morphologic alterations detected by more routine histological staining procedures.
Type I Response (8–64 μg/kg Iso for 3 or 6 hours)
In hearts from animals dosed with 8–16 μg/kg Iso at 3 hours, by visual inspection the staining intensity of cTnT and cTnI showed little or no change when compared with that observed in the hearts of saline-treated rats (Figures 4B, 4C, 4b, and 4c). A slightly more localized reduction in cTnT and cTnI staining intensity was detected in a few cardiac myocytes after a dose of 64 μg/kg Iso (Figures 3B, 4E, and 4e).
Type II Response (125–500 μg/kg Iso for 3–48 hours)
A marked reduction of cTnT and cTnI staining intensity increased with doses of 125–500 μg/kg Iso observed 3 hours after treatment (Figures 3C, 3D, 4F, 4G, and 4f–4h). As cardiac injury progressed in rats treated with 125–500 μg/kg Iso at 6, 12, or 24 hours, reduction of cTnT staining intensity and increase in size of less-stained areas varied with time (Figures 3E–3G). The reduction in cTnT immunoreactivity could still be detected in residual necrotic myocytes (Figure 3H) 48 hours after treatment with 125–500 μg/kg.
Immunoreactivity of iNOS
No cytoplasmic iNOS immunoreactivity was detected in the cardiac myocytes from saline-treated rats (Figure 5A).
Type I Response (8–64 μg/kg Iso for 3 or 6 hours)
No iNOS immunoreactivity was found in the hearts from these animals.
Type II Response (125–500 μg/kg Iso for 3–48 hours)
The most prevalent locations of increased iNOS immunoreactivity were observed in the tendinous cords (Figures 5B and 5D), papillary muscles (Figures 5C and 5E), and the subendocardium (Figure 5F). The high levels of iNOS immunoreactivity were widely distributed over the endocardium, myocardium, and pericardium of both the atria and the ventricles, particularly the left ventricular myocardium (Figures 5G and 5H), in the hearts from rats 24 hours after treatment with 500 μg/kg Iso.
Association among Pathological Findings with Serum Levels of cTnT
Pathological findings, cTnT immunoreactivity, and serum levels of cTnT are summarized in Table 1. The separation between a Type I response (8–64 μg/kg Iso for 3 or 6 hours) and a Type II response (125–500 μg/kg Iso for 3–48 hours) can be seen in Figure 6.
Discussion
The results of the present study indicate that on the basis of routine histologic evaluation, TUNEL assay for apoptosis, and immunoreactivity for cTnT and cTnI, the myocardial lesions induced by low doses of Iso were arbitrarily divided into two categories (Table 1). The coexistence of interstitial edema, inflammatory infiltration, BM damage, and myocardial degeneration was interpreted as indicating potential reversible lesions, because these changes are not necessarily the most important factors involved in the pathogenesis of cell death (Type 1 response). In contrast, the coexistence of apoptosis, necrosis with cell membrane rupture, and fibroblast proliferation was interpreted as indicating the presence of irreversible cell damage (Type II response). The two categories of cellular alterations correlated well with the reduction in cTnT and cTnI immunoreactivity and the corresponding increase in serum levels of these two cardiac proteins.
Dose- and Time-Related Cardiac Responses Induced by Iso
Type I Response
The present study shows that the serum level of cTnT was below the level of assay detection in most serum samples from saline-treated control animals. Serum cTnT levels were increased approximately ten-fold (0.13–0.26 ng/mL) in rats 3–6 hours after treatment with doses of 8–64 μg/kg Iso. Similar increases in serum cTnT levels have been reported after strenuous exercise. Rats subjected to 5 hours of intense exercise had serum cTnT levels of 0.06 to 0.26 ng/mL (Chen et al. 2000). The serum levels were the highest at the end of the exercise period. However, pathological lesions typical of ischemic myocardial damage were found in the hearts of only a small number of the rats 24 or 48 hours after the end of the exercise (Chen et al. 2000). Elevated cTnT levels (< 0.2 ng/mL) have also been observed at the end of a strenuous bicycle race (Neumayr et al. 2005). The appearance of low cTnT levels in the serum associated with strenuous exercise or low doses of Iso, together with the observations of a slight reduction of cTnT immunoreactivity in the absence of significant Iso-induced lesions, are consistent with the hypothesis that the source of these cardiac troponins could be the small pool of cTnT in the myocyte cytoplasm.
Previous studies have reported that Iso-induced acute myocardial injury appeared as early as 1–3 hours in rats (Dudnakova et al. 2003; Goldspink et al. 2004; Noronha-Dutra et al. 1984) and that low doses of Iso (0.01 or 0.02 mg/kg) could induce myocardial hypertrophy, necrosis, and apoptosis in rats (Alderman and Harrison 1971; Goldspink et al. 2004). The present study found similar patterns of Iso-induced cardiac injury. At the lowest Iso doses (8–64 μg/kg), only minimal changes were detectable on routine histopathology 3–6 hours after treatment. The myocardial alterations found in hearts of animals given low doses of Iso consisted mainly of early apoptosis and little or no necrosis, to a lesser extent edema, neutrophil infiltration, BM damage, or myocardial degeneration. These observations tend to identify Iso doses of 32–64 μg/kg as being threshold levels for the induction of myocyte apoptosis without necrosis. These same doses resulted in a decrease in cTnT immunostaining intensity in a few scattered myocytes. The regularly arranged cross-striations and normal configuration of myocytes were not altered.
Type II Response
Treatment of rats with Iso doses of 125 to 500 μg/kg was found to produce discernible cardiac lesions clearly seen by light microscopic examination 3–48 hours after treatment. Significant changes were also seen in these hearts using TUNEL assays for apoptosis and immunostaining for cTnT, cTnI, and iNOS. In some instances, alterations of interstitial edema, BM, and myocyte integrity were mild and thus could be considered reversible. In other instances, irreversible alterations such as myocyte necrosis and apoptosis and BM damage were prominent. At these doses of Iso, both reversible and irreversible lesions are seen as part of the spectrum of severe myocardial damage. Serum levels of cTnT showed highly significant changes that closely paralleled the severity of the lesion scores. Immunohistochemical staining for cTnT demonstrated that the characteristic striated pattern was greatly reduced or changed to a diffuse pattern. These changes in immunoreactivity show good association with the dramatic increases in serum cTnT levels. The serum levels of cTnT following Type II response were ten-fold above the levels detected in Type I response (8–64 μg/kg Iso) and were elevated ~100 times above those levels found in saline-control animals.
By 48 hours, cardiac tissue morphology was characterized by the predominance of dense connective tissue interspersed with the remnants of dying or dead cardiac myocytes. At this time, cardiac lesions were less severe, and the levels of serum cTnT were reduced to less than 1 ng/mL. The relatively rapid clearance of cTnT from serum following acute myocardial injury suggests that the timing of blood sample collection will be a critical factor in appropriate use of cTnT measurements to detect the onset and severity of acute cardiac injury.
Potential Contributions of Apoptosis to cTnT Release
Previous studies have suggested that cytoplasmic proteins are unlikely to be released into the blood from cells with an intact plasma membrane (Wallace et al. 2004). The present study observed slight changes in nuclei stained by TUNEL assay in rats treated with Iso doses of 8–64 μg/kg. Although weak positive reactions were detected in the nucleus of single cells, morphological evidence of cell shrinkage and hypereosinophilic cytoplasm was not observed. It is possible that these cells underwent a minimal amount of DNA damage, but the damage did not reach a level capable of inducing cell death. Thus, the small number of potentially apoptotic cells detected in this study is unlikely to be sufficient to contribute to the tenfold increase in serum cTnT observed with the lowest doses of Iso (8–64 μg/kg).
There is direct evidence that exposure to certain amounts of Iso can lead to myocyte apoptosis. Saito et al. (2000) reported that the incidence of apoptosis in cultured rat myocytes increased in relation to the concentration of Iso. Goldspink et al. (2004) have shown that, in rats, the induction of myocardial apoptosis and necrosis peaked at 3–6 hours and 18 hours, respectively, after a single sc injection of 5 mg/kg Iso. In this model, the incidence of myocardial necrosis was significantly greater than that of myocardial apoptosis (Goldspink et al. 2004). These findings are similar to our observations in the present study (125–500 μg/kg) of peak apoptosis at 3–6 hours, and peak necrosis from 3 hours to 24 hours. At present it is not known whether the cardiac troponins are released from apoptotic myocytes. If cTnT release did occur from these cells, conceivably the increased numbers of apoptotic myocytes seen after treatment with Iso doses of 125–500 μg/kg might contribute to the observed increase in serum levels of cTnT. However, because of considerably greater numbers, the release from necrotic myocytes appears to be the predominant source of elevated levels of serum cTnT in a Type II response.
Role of iNOS in Iso-induced Myocardial Injury
The present study showed that iNOS immunoreactivity significantly increased following treatment with Iso doses of 125–500 μg/kg. Recent studies have found that NO can play an important role in the response of the myocardium to injury. Excess levels of NO resulting from upregulation of iNOS contribute to myocyte apoptotic cell death, myocardial remodeling, and cardiac dysfunction (Liu et al. 2005; Schulz 2001). In contrast, inhibition of iNOS protects the aging rat heart against Iso-induced myocardial injury (Li et al. 2006). Lack of iNOS has also been reported to preserve cardiac function and to enhance cardiac contractile and relaxation responses following administration of Iso to iNOS knockout mice (Sun et al. 2005). There is evidence to suggest that iNOS can be induced in cardiomyocytes (Kleinert et al. 2005; Schulz 2001) and that the induced iNOS plays a role in initiating cardiac myocyte apoptosis (Arstall et al. 1999). The increased iNOS immumoreactivity detected in the present study also suggests a similar contribution by the NO system to the induction of cardiac myocyte apoptosis observed following treatment with the higher doses of Iso.
In summary, the data from this study show that treatment of rats with various doses of Iso results in two patterns of response. The first type, seen with doses of 8–64 μg/kg, is characterized by increases in serum cTnT to values of < 0.3 ng/mL with a slight reduction in cTnT immunoreactivity and minimal histological changes. The second type, which occurred with doses of 125–500 μg/kg, results in striking myocardial alterations (myocyte apoptosis and necrosis, loss of BM, inflammatory cell infiltration and fibrosis) that can easily be discerned by light microscopy and induction of iNOS. These changes are associated with a significant reduction in cTnT immunoreactivity and parallel increases in serum cTnT.
Footnotes
This article was written in a personal capacity and does not represent the opinion of the United States Food and Drug Administration. The authors wish to thank Dr. Elizabeth Hausner, CDER, FDA, and Dr. Stephen L. Hilbert, CDRH, FDA, for their critical review of the manuscript.