Spontaneous Cardiomyopathy in Cynomolgus Monkeys ( Macaca Fascicularis )

Introduction

In preclinical toxicology studies, purpose-bred nonhuman primates are used for safety assessment of drugs intended for human use, of which one major area of concern is the cardiovascular system. Thus, it is important to characterize, by lesion morphology and distribution, the spontaneous lesions that can occur in clinically healthy monkeys in order to distinguish them from drug-induced toxicity. The literature describing such heart lesions in nonhuman primates that are purpose-bred (Chamanza et al. 2006; Drevon-Gaillot et al. 2006; Ito et al. 1992; Qureshi 1979) or wild-caught (Bennet et al. 1998; Lowenstine 2003) is sparse. Consistently, the most common spontaneous heart finding in purpose-bred cynomolgus (Macaca fascicularis), rhesus (Macaca mulatta), assam (Macaca assamensis), and squirrel (Saimiri sciureus) monkeys as well as common marmosets (Callithrix jacchus) was focal mononuclear leukocyte infiltrates in the interstitium or perivascular region often of the subendocardium, usually without concurrent myocyte damage and less often with minimal myocardial degeneration or necrosis (Chamanza et al. 2006; Drevon-Gaillot et al. 2006; Ito et al. 1992; Qureshi 1979). Other less common findings include mineralization of the myocardium and heart vessel wall, endocarditis, pericarditis, eosinophilic infiltrates or granulomas in the epicardial adipose tissue, squamous (epithelial) cysts, myocardial fibrosis, myocardial mucin deposition, extramedullary hematopoiesis, blood-filled valvular cysts, and ectopic thyroid tissue (Chamanza et al. 2006; Drevon-Gaillot et al. 2006).

This report characterizes a previously undescribed spontaneous heart lesion that was identified by microscopic examination in four clinically healthy purpose-bred cynomolgus monkeys. Special stains and immunohistochemistry were used to facilitate lesion characterization and assess cardiomyocyte viability.

Materials And Methods

Six cynomolgus monkeys were maintained and humanely euthanized in accordance with regulatory compliance for animal care and use. All animals were clinically healthy. Animal no. 1 was a 9-year-old male that was not on study but was euthanized due to a suspected chronic tuberculosis lesion in the vertebrae that after complete necropsy, was attributed to previous trauma. Animal no. 2 was a 7-year-old female that was part of the vehicle control group in a 9-month toxicology study and survived to scheduled necropsy. Animal no. 3 was a 6-year-old male and no. 4 was a 5-year-old female that were part of another toxicology study in which there were no adverse treatment-related findings and survived to scheduled necropsy on day 8 and day 101, respectively. Two animals from the control group of this toxicology study, a 5-year-old male terminated on day 18 (animal no. 5) and a 6-year-old female terminated on day 101 (animal no. 6), were used as a positive control for immunohistochemistry and special stains. As part of the in-life data for the second toxicology study, a 5-minute electrocardiogram was performed twice prior to study onset and immediately prior to and within 15 minutes of compound administration.

Immediately after euthanasia, a necropsy was performed, during which the heart was collected whole and preserved in 10% neutral buffered formalin. After fixation, a longitudinal incision was made to produce two sections. One section included the interventricular septum, right ventricular free wall and atrium, aortic valve and right atrioventricular valve; and the second section included the left ventricular free wall and atrium, dorsal papillary muscle, and left atrioventricular valve. Each section was paraffin embedded and processed routinely into 5 μm-thick hematoxylin and eosin-stained slides for light microscopic examination. Serial sections of heart were examined using the following special stains: Mallory’s phosphotungstic acid hematoxylin (PTAH) (animals nos. 1 and 3–6) (Luna 1968) for muscle bands; Masson’s trichrome (Luna 1968) (animals nos. 1–6) and Sirius red (animal no. 1) (Kiernan 2007) for collagen fibers; and Lie’s hematoxylin basic fuschin picric acid (HBFP) (animal nos. 3–6) (Sheehan and Hrapchak 1980) for myocyte viability. Serial sections from animals nos. 1 and 3–6 also were examined by immunohistochemistry using mouse antihuman CD68 antibody (dilution 1:1000; clone KP1; DAKO, Glostrup, Denmark) for macrophages, mouse antihuman desmin antibody (dilution 1:50; clone D33; DAKO) for muscle, and mouse antihuman troponin-I antibody (dilution 1:200; clone 2D5; Labvision, Fremont, CA) for heart muscle. Immunohistochemistry was performed according to the manufacturer’s instructions using the Ventana discovery XT stainer (Jackson Laboratories Inc., West Grove, PA) and a biotin-secondary antibody detection system (Jackson Laboratories Inc.) consisting of biotinylated goat antimouse antibody (dilution 1:250) for the secondary antibody and HRP-DAB MAP kit for the chromagen. Some special stains and immunohistochemistry could not be performed on all animals in toxicology studies due to study protocol limitations.

Results

A spontaneous cardiomyopathy was identified by microscopic examination in four clinically healthy purpose-bred cynomolgus monkeys that ranged from 4 to 9 years of age and included 2 males and 2 females. Animal no. 1 was not part of a toxicology study, animal no. 2 was in the vehicle control group of a toxicology study, and animal nos. 3 and 4 were enrolled in a toxicology study in which there were no adverse treatment-related findings. The electrocardiograms performed in-life on animals with (nos. 3 and 4) and without (nos. 5 and 6) the cardiomyopathy were not remarkable. At necropsy, gross heart changes or evidence of heart failure were not present.

On routine microscopic examination of the heart, each monkey had a chronic cardiomyopathy, ranging from mild (animal no. 3) to moderate (animal no. 1) to severe (animal nos. 2 and 4). The cardiomyopathy had a consistent distribution (Figure 1A–E). It affected the myocardium to subendocardium of the apical to mid-left ventricular free wall and interventricular septum and, in more severe cases, extended to the ventricular base and right ventricular free wall. The atria were not affected. Morphologically, the cardiomyopathy was characterized by foci of cardiomyocyte disarray, vacuolization of the perimyseal connective tissue, and a meshwork of fibrous tissue (Figure 2A). The fibrous tissue tended to concentrate around medium-sized blood vessels and dissected into the interstitium between and isolating disrupted cardiomyocytes and occasionally replacing lost cardiomyocytes. Disrupted cardiomyocytes often had cytoplasmic pallor or stippling and karyomegaly. Foci had a minimal attending mononuclear leukocyte infiltrate comprised predominantly of cells immunoreactive for CD68 antibody, a macrophage marker (Figure 2B).

Masson’s trichrome and Sirius stains (Figure 2C) indicated that the fibrous tissue was collagen. It also highlighted the meshwork pattern and the perimyseal localization of vacuoles associated with foci of disrupted cardiomyocytes. Cardiomyocytes in affected regions were immunoreactive for anti-desmin and troponin-I antibodies (Figure 2D) in a similar manner to those in unaffected regions and in control hearts (animal nos. 5 and 6). These antibodies also highlighted the misalignment (i.e., disarray) of cardiac fibers, which often manifested as a slight variation in staining intensity (Figure 2D inset). PTAH stain indicated that disrupted cardiomyocytes had a normal cross-striation pattern. Technical difficulties with the HBFP stain precluded its consistency among sections (Schrerer and Massi 1975; Rufaie, Florio, and Olsen 1983), and thus it was not used for interpretation.

In addition to the characteristic findings of the primary cardiomyopathy, animal no. 2 had one distinct focus in the left basal ventricle (Figure 1E) that was removed from the other affected foci in the left apical ventricle and interventricular septum. This focus extended from the endocardium to the epicardium. It shared morphological features with primary cardiomyopathy; however, it also contained hemosiderophages, an increased number of lymphocytes and plasma cells in the attending infiltrate, and more dense thick bands of fibrous tissue. The distribution and morphology of this focus were suggestive of a chronic reparative response from a previous infarct.

The previously described spontaneous finding of scattered mononuclear infiltrates (Chamanza et al. 2006; Drevon-Gaillot et al. 2006) also was identified in the heart from animal nos. 2, 4, and 5. This finding was considered separate from the cardiomyopathy, as it usually was removed from regions affected by the described cardiomyopathy and also occurred in several other animals from that toxicology study that did not have the concurrent cardiomyopathy.

Discussion

Four purpose-bred cynomolgus monkeys had the spontaneous finding of a chronic cardiomyopathy identified by microscopic examination of the heart. The number of animals in this report was too small to suggest an age or sex predisposition. All animals were clinically healthy, and for two monkeys, the repeated in-life electrocardiograms performed within 4 months of termination were not remarkable. Thus, the cardiomyopathy was not associated with detectable clinical manifestations proximal to euthanasia. There was no gross evidence of heart disease. Microscopically, the cardiomyopathy had a consistent distribution and morphology. The apical to mid-ventricular myocardium to subendocardium had foci of cardiomyocyte disarray, vacuolization of the perimyseal connective tissue, and a meshwork of fibrous tissue surrounding medium-sized blood vessels and dissecting between or less often replacing cardiomyocytes. Foci had a minimal, predominantly macrophage infiltrate, which was interpreted as a residual clean-up response. The disrupted cardiomyocytes were immunoreactive to anti-desmin and troponin-I antibodies, which indicated that the integrity of the muscle fiber antigens was intact and, thus, that the cardiomyocytes were not degenerative or necrotic at a submicroscopic level (Ribeiro-Silva et al. 2002; Zhang and Riddick 1996). The PTAH stain did not indicate contraction band necrosis. Thus, despite the disrupted appearance of cardiomyocytes on HE-stained sections, the chronic cardiomyopathy was not associated with active cardiomyocyte damage.

The consistent morphology and distribution of the cardiomyopathy suggested a common etiology and pathogenesis; however, a definitive cause was not identified. Catecholamine-induced or stress cardiomyopathy, myocardial infarction, hypertension associated with alterations in the renin-angiotensin-aldosterone system, cardiac hypertrophy, and heart failure are conditions documented to cause cardiac fibrosis and cardiomyocyte disruption in animals and humans (Khullar et al. 1989; Lyon et al. 2008; Sun and Weber 2005). In these monkeys, conditions associated with or evidence for the latter four conditions were not present. The cardiomyopathy, however, had features similar to those described in the chronic phase (greater than 21 days) of catecholamine-induced experimental cardiomyopathy in adult male rhesus monkeys (Khullar et al. 1989). Microscopically, the ventricular subendocardium to myocardium had foci of disrupted cardiomyocytes separated by irregular to linear scar tissue that assumed a variegated appearance, vacuolization of perimyseal connective tissue that were fat droplets by Sudan stain, and a minimal attending leukocyte infiltrate. The affected cardiomyocytes showed restoration of succinic dehydrogenase enzyme activity and normal ultrastructural morphology, which were lost during the acute phase (3 days). In humans, stress (also termed Takotsubo) cardiomyopathy is a form of myocardial stunning associated with a transient substantial elevation in serum catecholamine (epinephrine) levels due to sudden emotional or physical stress (Lyon et al. 2008). It is characterized by acute reversible ventricular dysfunction with a predilection for the apical myocardium. Interestingly, females have a higher incidence, constituting more than 90% of reported cases. There also is an increasing incidence of an atypical phenotype that has a paradoxical predilection for the basal to mid-ventricle. The associated histopathology is not well defined, especially as it is limited to endocardial biopsies obtained close to clinical presentation, but has included descriptions of interstitial fibrosis and mildly atrophic myocardial fibers with fatty infiltration (Akashi et al. 2003). There also are descriptions of cardiomyocyte contraction band necrosis, leukocyte infiltrates, and interstitial edema (Lyon et al. 2008), which is similar to the acute phase of catecholamine-induced experimental cardiomyopathy in monkeys (Khullar et al. 1989) and rats (Lyon et al. 2008). These latter acute changes were not morphologic features of the chronic cardiomyopathy in the monkeys of this report; however, if it is a shared pathogenesis, then more acute lesions of similar distribution may occur.

Stress cardiomyopathy is hypothesized to result primarily from direct epinephrine-mediated effects on cardiomyocytes and subsequent negative inotropism (Lyon et al. 2008). In addition to this mechanism, norepinephrine-induced coronary vasospasm could result in a secondary ischemic insult, superimposed on the primary epinephrine-induced apical stunning. Thus, the additional focus in the left basal ventricle in animal no. 2, which was distinct from the apically oriented primary cardiomyopathy and suggestive of a chronic infarct, may represent such a superimposed insult from coronary vasospasm. Alternatively, the pathogeneses may not be related.

This report documents another spontaneous heart lesion in clinically healthy monkeys for consideration during interpretation of toxicology studies. The consistent morphology and distribution of the cardiomyopathy suggested a common etiology and pathogenesis. It had features reminiscent of chronic catecholamine-induced experimental cardiomyopathy and stress cardiomyopathy in monkeys and humans, respectively. The cardiomyopathy described was chronic and quiescent, but should it have a shared pathogenesis, then a more acute phase of similar distribution may be identified in the future.