Inflammatory Cell Findings in the Female Rabbit Heart and Stress-associated Exacerbation with Handling and Procedures Used in Nonclinical Studies

Abstract

Despite the use of rabbits in biomedical research, including regulatory toxicology and cardiovascular studies, little data exist on heart findings in this species. This study was designed to document myocardial findings in female rabbits and the impact of study-related procedures typical for vaccine toxicology studies. One hundred and forty 6- to 8-month-old female New Zealand White rabbits were divided equally into 2 groups, high and low study procedure groups (group 1 and group 2, respectively). All animals received intramuscular (IM) injections of sterile saline every 2 weeks for 5 times and were necropsied 2 days after the final IM injection. Clinical chemistry, hematology, and urinalysis were evaluated. Blood for stress biomarkers (norepinephrine, epinephrine, cortisol, and corticosterone), C-reactive protein, cardiac troponin I, and creatine kinase were collected at time 0 (just before dose administration) and then at 4, 24, and 48 hr after dose administration in group 1 only. Hearts were assessed histologically. Focal to multifocal minimal inflammatory cell infiltrates were common (∼80%), particularly in the left ventricle and interventricular septum, and were similar to the types of infiltrates identified in other laboratory animal species. Additionally, study-related procedures elevated serum stress biomarkers and exacerbated the frequency and severity of myocardial inflammatory cell infiltrates.

Keywords

animal models catecholamines cardiomyopathy cardiovascular system New Zealand White rabbit risk assessment stress

In order to assess human risk from animal toxicity studies, it is essential to have an understanding of the range of morphological findings that occur spontaneously or are associated with study procedures within the species under examination. Rabbits have been used extensively in biomedical research, most frequently as animal models of disease, but also in nonclinical safety studies, particularly for developmental toxicity or local irritation studies. Rabbits have been a typical species for regulatory toxicology studies for vaccines (World Health Organization 2013). Despite the use of rabbits in various nonclinical studies, including as models of cardiovascular disease and physiology (Fan et al. 2015; Kang et al. 2016; Lang, Koren, and Odening 2016), there is a paucity of literature on background microscopic heart findings in rabbits. Inflammatory cell findings in rabbit hearts are well recognized within the comparative pathology community. The earliest publication on inflammatory heart lesions in otherwise healthy rabbits was in 1924 (Miller 1924), which described multifocal myocardial infiltrates of lymphocytes, macrophages, or a combination of both cell types with or without fibrosis, heterophils, plasma cells, mast cells, or degenerating cardiomyocytes (loss of striations). The locations of these findings, from greatest to least, were papillary muscle, interventricular septum, left ventricular free wall, right ventricular free wall, and atria. The lesions were described as being most common within the myocardium but were occasionally evident subendocardially or subepicardially. While the lesions were less frequent in younger rabbits and occurred in approximately 60% of the 34 rabbits evaluated, the ages were not specified. Myocardial inflammatory cell infiltrates have been reported in the rabbit heart as an incidental finding (Bradley 2012; Van Vleet and Ferrans 1986) and within Pfizer’s internal historical database.

In an effort to characterize the nature and range of microscopic findings in rabbit hearts, Pfizer designed a large study in female New Zealand White (NZW) rabbits patterned after standard vaccine studies but with either more or fewer procedures than in a typical study. Inflammatory findings were the most common diagnosis in the rabbit heart, and this article details these inflammatory findings and demonstrates that increases in the number of routine study procedures in rabbits can increase the distribution and severity of these heart findings.

Materials and Methods

Animals and Housing

One hundred and forty (140) 6- to 8-month-old female NZW rabbits (Covance Research Products, Denver, PA) were housed single in stainless steel/plastic cages (Allentown Euro Rack, Allentown, Allentown, NJ) at the Pfizer, Groton, CT animal facility. Only female rabbits were evaluated in order to optimize the power of the study in the face of limitations in study size capacity at the Pfizer, Groton, CT facility. Rabbits were fed a standard diet of pelleted food (Certified Rabbit Diet 5322; PMI Feeds, Richmond, IN) and municipal drinking water, further purified by reverse osmosis, was provided ad libidum. Rabbits were randomly divided into 2 groups of 70 animals each. Animals in group 1 were housed in rooms separate from animals assigned to group 2. This study was conducted in an Association for Assessment and Accreditation of Laboratory Animal Care–accredited institution and in accordance with the current guidelines for animal welfare (National Research Council Guide for the Care and Use of Laboratory Animals, 2011).

Study Design to Evaluate the Effect of Handling and Procedural Stress in Rabbits

This study was carried out under an approved Pfizer Institutional Animal Care and Use Protocol and designed to mimic a 5-dose vaccine toxicology study. All animals received intramuscular (IM) injections of sterile saline (1 ml per dosing day administered as bilateral 0.5 ml injections into each quadriceps muscle) on 5 occasions (once every 2 weeks). Animals were necropsied on day 59, 2 days following the last injection.

All rabbits were subjected to the same types of standard procedures performed in a typical repeat-dose toxicology study; however, the number/frequency of procedures conducted was different between animals in groups 1 and 2 (Table 1). Animals in group 1 (increased procedures) were subjected to the typical procedures performed in a standard repeat-dose toxicology study, except that additional blood samples were taken (i.e., 21 blood collections in this study vs. ∼6 in a standard study). Animals in group 2 (minimal procedures) had few procedures and only 2 blood collections in order to minimize the restraint and handling (less than in a standard study) of the animals.

Table 1.

Number of Scheduled Procedures per Group.^a

Procedures	Group 1 (increased procedures)	Group 2 (minimal procedures)
Dosing	5	5
Cage-side observations on the day of dosing	15	10
Detailed clinical observations^b	10	1
Body weight	19	3
Injection site and body temperature	15	0
Ophthalmology exam	2	0
Blood collection	21	2

^aThe number of procedures differed between groups 1 and 2; group 1 had more procedures than that of a typical regulatory toxicology study, while group 2 had fewer procedures than that of a typical regulatory toxicology study.

^bAnimal removed from cage for more detailed examination.

Blood and Urine Collection

Blood was collected primarily from the auricular artery from rabbits restrained in metal restrainers (rabbit restrainer cage, model no. RR1; Draper Metal Fabrication Co., Holbrook, MA). Blood was collected from all animals for coagulation, hematology, and clinical chemistry analysis prior to the start of the study in sodium citrate anticoagulant tubes, ethylenediaminetetraacetic acid anticoagulant tubes, and serum separator tubes, respectively (total of 8.5 ml blood collected); blood for clinical chemistry was also collected at necropsy. Serum C-reactive protein (CRP), creatine kinase (CK), cardiac troponin I (cTnI), and stress-associated serum biomarkers (epinephrine, norepinephrine, cortisol, and corticosterone) were obtained prior to the start of the study and at necropsy in both groups. In addition, 1 ml of blood was collected from group 1 animals for cTnI, CK, CRP, epinephrine, norepinephrine, cortisol, and corticosterone prior to IM administration of saline (time 0) and at 4, 24, and 48 hr after administration on each of the 5 dosing days (with the exception of the 48-hr time point on day 57 which coincided with necropsy; a total of 20 ml blood/rabbit over 8 weeks). Serum for corticosterone, cortisol, epinephrine, and norepinephrine and urine for epinephrine and norepinephrine were protected from light and stored at −70°C until analyzed. Urine was collected by cystocentesis at necropsy.

Serum Clinical Chemistry, cTnI, CRP, Creatine Kinase, and Stress Biomarkers

Routine clinical chemistry parameters were evaluated using a Siemens Advia 1800 clinical chemistry analyzer (Siemens Corp., Tarrytown, NY). CRP was evaluated using the Life Diagnostic rabbit-specific solid phase sandwich immunoassay (Life Diagnostics, Inc., West Chester, PA). Hematology was evaluated using a Siemens Advia 120 analyzer (Siemens Corp.). Fibrinogen, activated partial thromboplastin time, and prothrombin time were evaluated on the Diagnostic Stago STA Compact coagulation analyzer (Stago, Parsippany, NJ). cTnI was measured using the human TnI-Ultra® assay on the Siemens Advia Centaur immunoanalyzer (Siemens Corp.). An internal assay validation was performed as well as a preliminary evaluation of cTnI responses in normal minimally restrained rabbits. As part of the internal validation, the limit of quantitation was determined by running (in duplicate) daily for 10 days a series of rabbit serum samples close to the limit of blank and calculating the mean, standard deviation, and percentage coefficient of variation. The limit of quantitation (the lowest concentration of cTnI within the acceptable precision limit of 20%) was 0.03 ng/ml. When normal rabbits were assessed with this assay, many had results <0.03 ng/ml but some had cTnI results up to 0.05 ng/ml and thus a threshold for background finding of 0.06 ng/ml was established. Preliminary studies also included collection of serum from a rabbit administered 3 mg/kg isoproterenol (peak mean concentration of cTnI of 133.5 ng/ml at 6 hr postdosing) as positive a control (data not shown). Epinephrine and norepinephrine were quantitated using an ultrahigh-pressure liquid chromatography–tandem mass spectrometry (LC-MS/MS) assay. Samples were prepared for analysis by the addition of an internal standard followed by solid phase extraction. Epinephrine and norepinephrine were separated from the other sample components by gradient elution with reverse phase chromatography and then detected by monitoring unique fragment ions in the mass spectrometer. Calibration curves for epinephrine and norepinephrine were constructed from known standards by plotting the ratio of the analyte peak area to internal standard peak area versus concentration. The concentration in the samples were calculated by interpolation using 1/x ² weighted linear regression analysis from the calibration curve for each analyte. Urine measurements were normalized to urinary creatinine. Methods deployed were developed in-house specifically for rabbit samples and are similar to those published in literature (Bicker et al. 2013). Assay performance was determined by quality control standards throughout the individual runs.

Serum corticosterone and cortisol were measured using an LC-MS/MS assay. Samples were prepared for analysis by the addition of an internal standard followed by supported liquid extraction. Prior to detection, the corticosterone and cortisol were separated from the other sample components by reverse phase chromatography coupled with gradient elution. The analytes and internal standard were detected by monitoring unique fragment ions. Calibration curves for corticosterone and cortisol were constructed from known standards by plotting the ratio of the analyte peak height to internal standard peak height versus concentration. The concentration in the samples was calculated by interpolation using 1/x ² weighted linear regression analysis from the calibration curve for each analyte. Assay performance was determined by quality control standards throughout the individual runs. Methods deployed were developed in-house specifically for rabbit samples and are similar to those published in literature (Guo et al. 2006).

Necropsy and Histology

Animals were euthanized by intravenous administration of barbiturate anesthesia (Nembutal® [sodium pentobarbital]; Oak Pharmaceutical, Lake Forest, IL) followed by exsanguination on day 59, 2 days after the last IM saline injection. All animals were fasted overnight prior to necropsy. The necropsy included macroscopic examination of the external surface of the body and the thoracic and abdominal cavities and their contents. Hearts were excised, gently flushed with 10% neutral buffered formalin (StatLab Medical Products, McKinney, TX) to remove the blood, and immersion fixed in 10% neutral buffered formalin for at least 48 hr. After fixation, the hearts were sectioned generating 2 longitudinal sections. In order to fit in cassettes, these longitudinal samples were then cut in cross section to generate 4 sections, 2 with the apical region of the heart and 2 with the base of the heart (Figure 1). All samples were processed with paraffin, and 5 µm sections were stained with hematoxylin and eosin (H&E). In addition to the routine H&E staining, special stains (periodic acid Schiff [PAS], Grocott’s methenamine silver [GMS], Gram, Giemsa, Masson’s trichrome, and/or phosphotungstic acid hematoxylin [PTAH]) were utilized on sections of heart from selected animals to further clarify and/or characterize findings.

Figure 1.

Histological sections of the rabbit heart. Formalin-fixed hearts were incised to generate two longitudinal sections. These samples were then bisected to fit in cassettes for processing, generating four hematoxylin and eosin slides; two slides included the atria (right atrium [RA] and left atrium [LA]) and an upper part of the ventricles and two slides contained the ventricles, including the right ventricle (RV), left ventricle (LV), papillary muscle (P), and the interventricular septum (IVS). Hematoxylin and eosin. Original Objective 0.4×.

Histopathologic Evaluation

Microscopic findings were graded on a scale of 1–5 as minimal, mild, moderate, marked, or severe, respectively. Each region of the heart was evaluated and in order to understand the distribution of findings within the heart, these findings were recorded separately for the left ventricle, right ventricle, interventricular septum, right atrium, and left atrium. All findings were recorded, regardless of the size or extent. Following completion of the tissue evaluation by the study pathologist, an independent peer-review evaluation of all heart sections was performed by a second pathologist.

Results

In life

All rabbits survived to scheduled euthanasia and were in good health. There were no differences in body weight, food consumption, or procedure-related clinical effects.

Necropsy

Complete necropsies with gross evaluation of the external surfaces, thoracic, and abdominal cavities and contents were performed. Abnormal color at the injection site was observed at necropsy in 5 animals in group 1 and 13 animals in group 2. No findings were identified grossly in the heart or in any other tissues.

Inflammatory and Other Findings in the Rabbit Heart

A range of inflammatory findings were evident in the heart microscopically. Inflammatory findings were identified in 80% (56 out of 70) of the rabbits in the limited handling group (group 2) and 86% (60 out of 70) in the increased handling group (group 1). Very rarely, there was individual cardiomyocyte necrosis unassociated with inflammatory cells (Figure 2A). More commonly, however, the necrotic cardiomyocytes were associated with macrophages (Figure 2B) and occasionally with low numbers of heterophils (Figure 2C; termed mixed inflammatory cell infiltrates) and rarely with interstitial edema/basophilic interstitial fibrillar matrix. Aggregates of macrophages without evidence of necrotic cardiomyocytes were also evident (Figure 2D), in which the necrotic cardiomyocyte(s) was either no longer present or obscured by inflammatory cells. Small lymphocytic infiltrates with or without plasma cells were rarely evident (Figure 2E); these may be comparable to the lymphocytic infiltrates common in many organs unassociated with tissue damage and will not be the subject of further discussion here (Drevon-Gaillot et al. 2006; Helke et al. 2016; Sato, Doi, Wako, et al. 2012). Occasionally fibrosis was present (Figure 2F). These findings were minimal to moderate in severity grade, focal to multifocal in distribution, and located within the myocardium. While minimal severity grades were found in both groups at comparable (similar) incidences (60 of the 70 and 56 of the 70 animals, respectively; Table 2), mild and moderate inflammatory findings were only identified in group 1. Individual cardiomyocyte necrosis unassociated with inflammatory cells was present in two of the group 1 hearts and none of the group 2 hearts. Fibrosis was evident in a single group 2 rabbit (with minimal inflammatory cell infiltrates) and a single group 1 rabbit (unassociated with inflammatory cell infiltrates). The foci were usually comprised of macrophages and occasionally heterophils and/or individual necrotic cardiomyocytes. These findings were widely distributed but were most commonly localized to the left ventricular free wall, less commonly in the interventricular septum and the right ventricular free wall, and least frequently in the atria (Table 3). Many of the minimal foci recorded in this study had such low numbers of inflammatory cells that they would most likely fall beneath the routine recording thresholds by most pathologists because of their small size and/or incidence across all study groups. Compared to the Pfizer historical database (of the 1,254 rabbits where an approximately 13% incidence of inflammatory cell foci were recorded in the heart), much higher incidences were recorded in this study. Possible reasons for such large differences in the recorded incidences include the judicious application of thresholds by individual pathologists during evaluation of regulatory toxicology studies, the high sensitivity for recording heart findings in this study regardless of extent, and the increase in surface area evaluated in this investigation (approximately 200% more).

Figure 2.

Photomicrographs of the heart from minimally handled rabbits (group 2), showing the progression of lesions. (A) Focal cardiomyocyte necrosis with no apparent inflammation (arrowhead). (B) Focal cardiomyocyte necrosis (arrowhead) with few inflammatory cells adjacent to focal small macrophage aggregate (black arrow). (C) Focal cardiomyocyte necrosis (arrowhead) associated with a slightly greater number of inflammatory cells (black arrow). (D) Inflammatory cell infiltrate (black arrow) with macrophages and heterophils without cardiomyocyte necrosis within the lesion. (E) Chronic inflammatory cell infiltrate comprised of lymphocytes (black arrow). (F) Myocardial fibrosis (blue arrows) associated with low numbers of mixed inflammatory cell infiltrates (black arrows). Hematoxylin and eosin. (A–E) Original objective 20×.

Table 2.

Group Incidences and Severities of the Inflammatory Microscopic Heart Findings.^a

	Group 1 (increased procedures)^b	Group 2 (minimal procedures)^b
Heart	N = 70	N = 70
Infiltrate, mononuclear, or mixed
Minimal	60	56
Mild	2	0
Moderate	2	0
Fibrosis	0	1
Minimal	1	1
Necrosis, cardiomyocyte	37	23
No infiltrate identified	6	14

^aThe data represent the number of rabbits that had at least one inflammatory cell focus in the heart; necrotic cardiomyocytes were colocalized with inflammatory cell infiltrates except on rare occasions in which a single necrotic cardiomyocyte was identified unassociated with inflammatory cell infiltrates.

^bNumber of animals with finding.

Table 3.

Incidence and Severity of Inflammatory Findings According to Location in the Various Heart Subregions.^a

Heart	Group 1 (increased procedures)					Group 2 (decreased procedures)
	Number of animals with finding (N = 70)					Number of animals with finding (N = 70)
	LV	RV	IVS	LA	RA	LV	RV	IVS	LA	RA
Infiltration, mononuclear cell
Minimal	46	11	29	7	6	35	8	11	1	5
Infiltration, mixed cell
Minimal	28	9	9	8	9	7	9	9	8	6
Mild	1	0	1	0	1	0	0	0	0	0
Moderate	1	0	2	0	0	0	0	0	0	0
Fibrosis
Minimal	1	0	0	0	0	1	0	0	0	0
Necrosis, cardiomyocyte
Minimal	31	10	16	3	4	15	4	7	2	2
Mild	2	0	2	0	0	0	0	0	0	0

Note: LV = left ventricle; RV = right ventricle; IVS = interventricular septum; LA = left atrium; RA = right atrium.

^aFindings were recorded once for each region of the heart (i.e., a single score for each finding based on multiple slides); however, one animal might have two types of infiltrates recorded for the same region of heart (mixed cell or mononuclear cell). Cardiomyocyte necrosis was associated with inflammatory cell infiltrates in all but two instances.

In addition to these findings, aortic mineralization was evident in 9% of the rabbits (ranging from minimal to moderate and focal to multifocal) and was similarly distributed between the two groups. Mineralization of the pulmonary artery was identified in a single animal, but as this vessel was infrequently present in the heart sections, the true incidence of mineralization at this site is uncertain.

Influence of Procedures on Myocardial Inflammatory Findings

While the incidence of inflammatory findings was similar between groups 1 and 2, the distribution of foci within the heart was increased in group 1 as compared to group 2. Specifically, more subregions (i.e., left ventricle, right ventricle, left atrium, right atrium, and interventricular septum) in individual hearts were identified as having at least one inflammatory focus (Table 3). The areas of the heart that demonstrated the greatest increase in inflammatory findings were the left ventricle and interventricular septum (Table 3). In addition, mild and moderate inflammatory findings were present only in group 1 (4 of the 70 hearts) and, similar to most of the minimal findings, were comprised of macrophages with scattered heterophils and necrotic cardiomyocytes (Figures 3 and 4); these were differentiated from minimal findings in that they were more extensive and/or more widely distributed. Mild and moderate findings were localized to the interventricular septum and left ventricle/papillary muscle (Table 3; Figure 3). These more severe inflammatory infiltrates were sometimes organized in a linear pattern, suggesting a response along the length of individual cardiomyocytes (Figures 3C and 4D). Mild and moderate lesions were not attributable to infectious disease (negative PAS, GMS, Gram, and Geimsa histochemical stains) or vascular fibrin thrombi (negative PTAH staining in vasculature; data not shown).

Figure 3.

Photomicrographs of the heart from an increased procedures rabbit (group 1). (A) Low magnification of cardiac ventricles and interventricular septum (IVS). The IVS and left ventricle had multifocal widespread inflammatory cell infiltrates (arrows). (B) Higher magnification of the IVS shows the infiltrate distribution. (C) High magnification of the inflammatory cell infiltrates; infiltrates included macrophages (arrows) with occasional necrotic cardiomyocytes (arrow head). Infiltrates were often in a linear pattern, presumably along the length of necrotic individual cardiomyocytes. Hematoxylin and eosin. (A) Original objective 0.4× (B) original objective 4× and (C) original objective 20×.

Figure 4.

Photomicrographs of the heart from an increased procedures rabbit (group 1; different from those in Figure 3). (A) Necrotic cardiomyocyte (arrow head) associated with few inflammatory cells. (B) Inflammatory cell infiltrates (black arrow) associated with focal cardiomyocyte necrosis and interstitial edema (clear spaces). (C) Inflammatory cell infiltrates comprising macrophages and heterophils (black arrows) associated necrotic cardiomyocytes (arrow head) with basophilic matrix and early fibrosis. (D) Distribution of inflammatory cell infiltrates (black arrow) was occasionally closely associated with small vessels (blue arrows). Hematoxylin and eosin. (A–C) Original objective 20× and (D) original objective 10×.

Stress Biomarkers

Serum stress biomarkers were evaluated before the study start to obtain baseline values and at necropsy. Changes in serum stress biomarkers in response to handling and procedures (IM injection and 4 blood collections over 48 hr) were evaluated only in group 1. Changes were most evident with norepinephrine (NE) and cortisol (Figure 5A and C). NE concentrations gradually increased after time 0 (immediately prior to IM saline administration), with peak values typically occurring at 48 hr postadministration. Mean NE concentrations increased at each dosing interval compared to baseline, peaking at day 29, which coincided with the 3rd IM saline administration (peak at 1.7× baseline; Figure 5A). Serum cortisol concentrations mimicked the findings in NE (Figure 5C). However, elevations in cortisol often peaked 4 hr after IM saline administration. Serum cortisol concentrations typically decreased or returned to near baseline values just prior to the next dosing interval. There were similar changes in mean corticosterone values as those seen with cortisol but based on higher corticosterone baseline values, increases in mean values were slight and only identified on days 15 and 29 at 4 hr postdose (Figure 5D). The values were typically lower than baseline values at time 0 at each dosing interval as well as on day 59. Epinephrine also showed changes over time, but based on high baseline values, no conclusions could be made (Figure 5B). Variations in NE and epinephrine were high, possibly in part due to the nature of the stress response in the individual animals and/or to the very short half-life of NE and epinephrine (a few minutes) in the blood (Friedgen, Halbrugge, and Graefe 1993). To minimize handling and procedures in group 2 animals, blood was collected and stress biomarkers evaluated only prior to the initiation of dosing and prior to necropsy (Figure 5). Serum stress biomarker values were similar between groups 1 and 2 at comparable time points (i.e., prior to initiation of dosing and at necropsy).

Figure 5.

Plasma concentrations of stress biomarkers associated with dosing in groups 1 and 2. (A) Norepinephrine (NE), (B) epinephrine (EP), (C) cortisol, and (D) corticosterone. In group 1 (•), dosing days (1, 15, 29, 43, and 57) included 4 blood collections: 0 hr (predose, vertical dotted blue lines), 4-hr postdose, 24-hr postdose, and 48-hr postdose. Stress biomarker concentrations typically increased over 48 hr after saline administration as compared to the time 0 (predose) blood sampling. Maximal NE and EP levels over the course of the study occurred at day 29. Measurements were taken prior to the study initiation and at necropsy in group 2 (▪).

There were no notable changes in urine chemistry biomarker parameters (NE or epinephrine) at the end of study (day 59). The timing of the urine collections (day 59 only) allowed for the assessment of changes in the stress biomarkers only at the end of the study (data not shown).

cTnI

Based on internal Pfizer studies conducted to determine the suitability and sensitivity of using the human TnI-Ultra assay to measure rabbit cTnI, concentrations of cTnI up to 0.06 ng/ml were considered to be within the range of background. Individual cTnI concentrations >0.06 ng/ml were present in 23 (33%) of the 70 rabbits in group 1 on one or more of the study days. These elevations typically were present at 4 and/or 24 hr postdose and were indicative of cardiomyocyte injury with release of cTnI into the circulation. Elevations in cTnI did not correlate well with histological findings, as >90% of the animals in group 1 had inflammatory infiltrates in the heart, but elevations in cTnI were only evident in 33% of rabbits. The 2 animals with the highest elevations in cTnI (peak values of 2.04 and 1.69 ng/ml), which occurred on day 57, had the most severe inflammatory findings in the heart (moderate). Interestingly, the 2 rabbits with mild inflammatory cell infiltrates did not have cTnI > 0.06 ng/ml at any time point during the study, indicating that these lesions may have initiated during the intervals between sampling. More animals had cTnI > 0.06 ng/ml on day 57 (6 animals) as compared with day 1 (4 animals). There were no animals with cTnI > 0.06 ng/ml in group 2, the minimally handled animals (only evaluated prior to the initiation of dosing and day 59 [48 hr after the last dose]).

CRP and CK

Study procedure–related increases in CRP and CK, as compared to the baseline values, were evident in group 1 (increased procedures) and increases in CRP were also observed compared with baseline in group 2 (minimal procedures). In group 1, CRP increased up to 4.3× baseline and CK increased up to 2.9× baseline at multiple time points over the study (after the 1st, 3rd, 4th, and 5th doses). Both analytes typically started to increase 4 hr postdose, peaked at 24 hr postdose, decreased at 48 hr postdose, and returned to baseline or near baseline at the subsequent predose time point. The increases in CRP and CK were attributed to an acute phase response and muscle injury (possibly as a result of restraint, blood collection vascular trauma, and/or IM injection), respectively. Group 2 rabbits also had slight increases in CRP 1.8× baseline on day 59 (48 hr postdose) but did not have changes in CK.

Discussion

This report details the nature, incidence, and distribution of microscopic inflammatory findings in hearts from female NZW rabbits. Minimal scattered aggregates of macrophages, with or without heterophils or individual necrotic cardiac myocytes, represented the most common background finding in the rabbit heart. Rabbits subjected to more handling and procedures had more widespread myocardial inflammatory foci and, in a few animals, increased severity of these findings.

The type of infiltrate described in detail here is similar to that described previously in the heart of rabbits (Miller 1924; Osheroff et al. 2013; Weber and Van Der Walt 1975). These findings appear to fall along a continuum representing the chronicity of the lesion. The initiating cause may be individual cardiomyocyte necrosis which is rapidly followed by infiltration of macrophages and heterophils. Ultimately, if lesions are severe enough, the inflammation resolves with fibrosis. Presumably small lesions resolve with limited or no scarring. The presence of inflammatory cells tracking along the length of cardiomyocytes is consistent with underlying individual cardiomyocyte necrosis. Small and transient elevations in serum cTnI during the study may have been the result of cardiomyocyte perturbations or death underlying some inflammatory foci. cTnI had limited correlation with microscopic lesions except in the most severe cases; this may reflect timing of blood collection relative to tissue sampling as well as the region evaluated microscopically. The active inflammatory lesions described in this article may be distinguished from effects of direct cardiotoxins, as toxic insult tends to have more widespread cardiomyocyte degeneration or necrosis, evidence of vascular injury, or have toxicity in other organ systems (Dunnick et al. 2004; Hanton et al. 2008; Jokinen et al. 2011, 2005; Lamberti et al. 2014; Merrick, Robinson, and Condie 1991; Nyska et al. 2005; Pettersen et al. 2012; Varga et al. 2015).

The data presented here demonstrate that handling and procedures can increase both the distribution of inflammatory foci in the rabbit heart and, in a subset of animals, the severity of the inflammatory findings. The more severe (mild and moderate) inflammatory findings were morphologically similar to the minimal (background) findings except that they were more expansive or widespread. Thus, while background inflammatory findings are typically minimal, they could range in severity up to moderate in any given toxicology study. Mild and moderate inflammatory findings were also evident in of 3 out of 1,254 animals in the Pfizer historical database, indicating that lesions of this severity have also been seen in control animals historically.

In the published literature, heart lesions similar in nature and distribution (primarily left ventricle and papillary muscle) to those identified in this study have been associated with both stress and catecholamines in rabbits (Downing and Chen 1985; Weber and Van Der Walt 1975). Catecholamine toxicity in the heart has been attributed to effects on the vasculature (hypoxia/ischemia) as well as to direct cardiomyocyte damage from spillover of catecholamines released from local sympathetic nerve terminals (Fineschi et al. 2010; Samuels 2007; Y-Hassan 2014; Liaudet, Calderari, and Pacher 2014). Within hours of administration, catecholamines cause contraction band degeneration/necrosis; after 24 hr, the findings are primarily inflammatory, comprising macrophages with scattered polymorphonuclear cells (heterophils). The lesions caused by catecholamine administration in rabbits have been suggested to be α-adrenergic receptor driven (Downing and Chen 1985).

Stress-induced cardiac damage in humans is suggested to be due to direct injury by catecholamines (Liaudet, Calderari, and Pacher 2014). In this study, a one-to-one correlation between the serum stress biomarker concentrations and heart lesions was not identified. A relationship of the heart findings to local catecholamine release, however, cannot be ruled out; additionally, peak blood levels of these very short-lived stress hormones may have not have coincided with the blood collection schedule. It is also possible that heretofore unknown mechanisms may also play a role in the pathogenesis of background inflammatory cell findings in the rabbit heart.

As compared to other species, rabbits are known for being hyperexcitable and may be uniquely susceptible to the development of stress-associated heart lesions (Osheroff et al. 2013). Interestingly, catecholamine concentrations in the rabbit heart have been demonstrated to be notably higher (up to 3 times) than that of other species such as the rat or dog (Angelakos, Fuxe, and Torchiana 1963; Mukherjee et al. 1983). Additionally, there is well-documented interindividual variation in the ability to respond/adapt to stress in NZW rabbits which may account for the observed differences in the severity of myocardial inflammatory findings (Giannico et al. 2014; Hilton 1982; Oxley and Ellis 2015; Schadt and Hasser 1998; Trocino et al. 2013; Burhans, Smith-Bell, and Schreurs 2015; Smith-Bell, Burhans, and Schreurs 2012; Lidfors and Edström 2010; Verga, Luzi, and Carenzi 2007; Weber and Van Der Walt 1975). Variation in individual susceptibility and adaptability to stress could explain the presence of mild and moderate severity heart findings in only a small subset of rabbits after increased handling and procedures.

Myocardial inflammatory cell infiltrates have been well described in rats and to a lesser extent in other species, such as nonhuman primates and mini pigs (Jokinen et al. 2011, 2005; Kemi et al. 2000; Chanut et al. 2013; Helke et al. 2016; Sato, Doi, Kanno, et al. 2012). While the etiology of the myocardial lesions has not been clearly identified, vascular- or catecholamine-based mechanisms have been suggested (Berridge et al. 2016; Cross et al. 2015; Greaves 2007; Jokinen et al. 2011, 2005; Kemi et al. 2000; Yoshimura et al. 1989). A more detailed understanding of the progression of lesions with age as well as any sex-related differences in the rabbit is required to fully understand this background finding.

Conclusions

Female NZW rabbits have a high incidence of minimal myocardial inflammatory cell findings that can range in severity from minimal to marked, in which the predominant cell type is the macrophage. While some of these findings may represent incidental infiltrates common in tissues of animals (usually perivascular and lymphocytic), many appear to be the result of individual cardiomyocyte necrosis which is rapidly followed by inflammatory cell infiltration (typically macrophages with heterophils). In severe cases, these inflammatory cell foci may resolve as focal fibrosis. The chronicity of the lesion dictates the histological appearance: very early lesions have individual cardiomyocyte necrosis with little or no inflammatory infiltrate; intermediate lesions have mixed inflammatory cell infiltrates with or without necrotic cardiomyocytes; and late lesions consist of inflammatory cell infiltrates and fibrosis. The incidence and severity of these foci may be exacerbated by handling and procedures that occur in toxicology studies. Based on the data presented here and on the well-described catecholamine-related effects on rabbit hearts, we propose that the increase in distribution and severity of inflammatory cell foci in the rabbits subjected to increased procedures and handling could be mediated by a stress response. Additionally, a subset of NZW rabbits may be exceptionally vulnerable to stress responses and/or do not habituate to stressors, which may contribute to the subsequent development of myocardial inflammatory cell foci under stressful conditions. The development of heart lesions secondary to study procedure–related stress in routine repeat-dose toxicology studies may make differentiation between stress-associated heart lesions and test article–related findings in toxicology studies in rabbits challenging. Further efforts are needed to understand how age and sex (as only females were evaluated in this study) may impact the incidence and severity of the myocardial inflammatory cell infiltrates in rabbits, particularly in the face of study-related activities and stress.

Footnotes

Acknowledgments

Special thanks to Kimberly Ebersole for her careful review of the data included in this article and Alan Opshal for image analysis support.

Author Contribution

Authors contributed to conception or design (RSS, IP, GH, KK, RP, VM, CR, JC, WR, DC); data acquisition, analysis, or interpretation (RSS, IP, CR, JC, WR, DC); drafting the manuscript (RSS); and critically revising the manuscript (IP, GH, KK, RP, VM, CR, JC, WR, DC). All authors gave final approval and agreed to be accountable for all aspects of work in ensuring that questions relating to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Angelakos

E. T.

Fuxe

Torchiana

M. L.

(1963). Chemical and histochemical evaluation of the distribution of catecholamines in the rabbit and guinea pig hearts. Acta Physiol Scand 59, 184–92.

Berridge

B. R.

Mowat

Nagai

Nyska

Okazaki

Clements

P. J.

Rinke

. (2016). Non-proliferative and proliferative lesions of the cardiovascular system of the rat and mouse. J Toxicol Pathol 29, 1S–47S.

Bicker

Fortuna

Alves

Falcao

(2013). Liquid chromatographic methods for the quantification of catecholamines and their metabolites in several biological samples—A review. Anal Chim Acta 768, 12–34.

Bradley

A. E.

(2012). New Zealand White Rabbit. In Background Lesions in Laboratory Animals: A Color Atlas ( MacInnis

E. R.

, ed.), pp. 87–91. Elsevier Health Sciences, Philadelphia, PA.

Burhans

L. B.

Smith-Bell

C. A.

Schreurs

B. G.

(2015). Effects of extinction treatments on the reduction of conditioned responding and conditioned hyperarousal in a rabbit model of posttraumatic stress disorder (PTSD). Behav Neurosci 129, 611–20.

Chanut

Kimbrough

Hailey

Berridge

Hughes-Earle

Davies

Roland

. (2013). Spontaneous cardiomyopathy in young Sprague-Dawley rats: Evaluation of biological and environmental variability. Toxicol Pathol 41, 1126–36.

Cross

M. J.

Berridge

B. R.

Clements

P. J.

Cove-Smith

Force

T. L.

Hoffmann

Holbrook

. (2015). Physiological, pharmacological and toxicological considerations of drug-induced structural cardiac injury. Br J Pharmacol 172, 957–74.

Downing

S. E.

Chen

(1985). Myocardial injury following endogenous catecholamine release in rabbits. J Mol Cell Cardiol 17, 377–87.

Drevon-Gaillot

Perron-Lepage

M. F.

Clement

Burnett

(2006). A review of background findings in cynomolgus monkeys (Macaca fascicularis) from three different geographical origins. Exp Toxicol Pathol 58, 77–88.

10.

Dunnick

J. K.

Lieuallen

Moyer

Orzech

Nyska

(2004). Cardiac damage in rodents after exposure to bis(2-chloroethoxy)methane. Toxicol Pathol 32, 309–17.

11.

Fan

Kitajima

Watanabe

Zhang

Liu

Chen

Y. E.

(2015). Rabbit models for the study of human atherosclerosis: From pathophysiological mechanisms to translational medicine. Pharmacol Ther 146, 104–19.

12.

Fineschi

Michalodimitrakis

D’Errico

Neri

Pomara

Riezzo

Turillazzi

(2010). Insight into stress-induced cardiomyopathy and sudden cardiac death due to stress. A forensic cardio-pathologist point of view. Forensic Sci Int 194, 1–8.

13.

Friedgen

Halbrugge

Graefe

K. H.

(1993). Plasma clearances and extractions of four catecholamines in the anesthetized rabbit: The role of amine removal by blood cells. J Cardiovasc Pharmacol 21, 21–28.

14.

Giannico

A. T.

Lima

Lange

R. R.

Froes

T. R.

Montiani-Ferreira

(2014). Proven cardiac changes during death-feigning (tonic immobility) in rabbits (Oryctolagus cuniculus). J Comp Physiol A Neuroethol Sens Neural Behav Physiol 200, 305–10.

15.

Greaves

(2007). Cardiovascular System in Histopathology of Preclinical Toxicology Studies ( Greaves

, ed.). Academic Press, New York.

16.

Guo

Taylor

R. L.

Singh

R. J.

Soldin

S. J.

(2006). Simultaneous determination of 12 steroids by isotope dilution liquid chromatography-photospray ionization tandem mass spectrometry. Clin Chim Acta 372, 76–82.

17.

Hanton

Sobry

Dagues

Provost

J. P.

Le Net

J. L.

Comby

Chevalier

(2008). Characterisation of the vascular and inflammatory lesions induced by the PDE4 inhibitor CI-1044 in the dog. Toxicol Lett 179, 15–22.

18.

Helke

K. L.

Nelson

K. N.

Sargeant

A. M.

Jacob

McKeag

Haruna

Vemireddi

. (2016). Background pathological changes in minipigs: A comparison of the incidence and nature among different breeds and populations of minipigs. Toxicol Pathol 44, 325–37.

19.

Hilton

S. M.

(1982). The defence-arousal system and its relevance for circulatory and respiratory control. J Exp Biol 100, 159–74.

20.

Jokinen

M. P.

Lieuallen

W. G.

Boyle

M. C.

Johnson

C. L.

Malarkey

D. E.

Nyska

(2011). Morphologic aspects of rodent cardiotoxicity in a retrospective evaluation of national toxicology program studies. Toxicol Pathol 39, 850–60.

21.

Jokinen

M. P.

Lieuallen

W. G.

Johnson

C. L.

Dunnick

Nyska

(2005). Characterization of spontaneous and chemically induced cardiac lesions in rodent model systems: The national toxicology program experience. Cardiovasc Toxicol 5, 227–44.

22.

Kang

Brennan

J. A.

Kuzmiak-Glancy

Garrott

K. E.

Kay

M. W.

Efimov

I. R.

(2016). Technical advances in studying cardiac electrophysiology role of rabbit models. Prog Biophys Mol Biol 121, 97–109.

23.

Kemi

Keenan

K. P.

McCoy

Hoe

C. M.

Soper

K. A.

Ballam

G. C.

van Zwieten

M. J.

(2000). The relative protective effects of moderate dietary restriction versus dietary modification on spontaneous cardiomyopathy in male Sprague-Dawley rats. Toxicol Pathol 28, 285–96.

24.

Lamberti

Giovane

Garzillo

E. M.

Avino

Feola

Porto

Tombolini

Di Domenico

(2014). Animal models in studies of cardiotoxicity side effects from antiblastic drugs in patients and occupational exposed workers. BioMed Res Int 2014, 8.

25.

Lang

C. N.

Koren

Odening

K. E.

(2016). Transgenic rabbit models to investigate the cardiac ion channel disease long QT syndrome. Prog Biophys Mol Biol 121, 142–56.

26.

Liaudet

Calderari

Pacher

(2014). Pathophysiological mechanisms of catecholamine and cocaine-mediated cardiotoxicity. Heart Fail Rev 19, 815–24.

27.

Lidfors

Edström

(2010). The laboratory rabbit. In The UFAW Handbook on the Care and Management of Laboratory and Other Research Animals ( Hubrecht

Kirkwood

, eds.), 8th ed., pp. 179–187. Wiley-Blackwell, Oxford, UK.

28.

Merrick

B. A.

Robinson

Condie

L. W.

(1991). Cardiopathic effect of 1,2,3-trichloropropane after subacute and subchronic exposure in rats. J Appl Toxicol 11, 179–87.

29.

Miller

C. P.

(1924). Spontaneous interstitial myocarditis in rabbits. J Exp Med 40, 543–51.

30.

Mukherjee

Haghani

Brady

Bush

McBride

Buja

L. M.

Willerson

J. T.

(1983). Differences in myocardial alpha- and beta-adrenergic receptor numbers in different species. Am J Physiol 245, H957–61.

31.

National Research Council (2011). Guide for the Care and Use of Laboratory Animals. National Academy Press, Washington, DC.

32.

Nyska

Murphy

Foley

J. F.

Collins

B. J.

Petranka

Howden

Hanlon

Dunnick

J. K.

(2005). Acute hemorrhagic myocardial necrosis and sudden death of rats exposed to a combination of ephedrine and caffeine. Toxicol Sci 83, 388–96.

33.

Osheroff

M. R.

Kobs

D. J.

Buccellato

Croutch

C. R.

Elcock

L. E.

Burback

B. L.

Johnson

J. D.

(2013). Comparative toxicology studies in Sprague-Dawley rats, rhesus monkeys, and New Zealand White rabbits to determine a no observed adverse effect level for 1,1’-methylenebis[4-[(hydroxyimino)methyl]-pyridinium] dimethanesulfonate. Int J Toxicol 32, 59S–74S.

34.

Oxley

J. A.

Ellis

C. F.

(2015). Misconceptions regarding rabbit behaviour. Vet Rec 176, 339–40.

35.

Pettersen

J. C.

Pruimboom-Brees

Francone

O. L.

Amacher

D. E.

Boldt

S. E.

Kerlin

R. L.

Ballinger

W. E.

(2012). The PPARα agonists fenofibrate and CP-778875 cause increased beta-oxidation, leading to oxidative injury in skeletal and cardiac muscle in the rat. Toxicol Pathol 40, 435–47.

36.

Samuels

M. A.

(2007). The brain-heart connection. Circulation 116, 77–84.

37.

Sato

Doi

Kanno

Wako

Tsuchitani

Narama

(2012). Histopathology of incidental findings in cynomolgus monkeys (Macaca fascicularis) used in toxicity studies. J Toxicol Pathol 25, 63–101.

38.

Sato

Doi

Wako

Hamamura

Kanno

Tsuchitani

Narama

(2012). Histopathology of incidental findings in beagles used in toxicity studies. J Toxicol Pathol 25, 103–34.

39.

Schadt

J. C.

Hasser

E. M.

(1998). Hemodynamic effects of acute stressors in the conscious rabbit. Am J Physiol 274, R814–21.

40.

Smith-Bell

C. A.

Burhans

L. B.

Schreurs

B. G.

(2012). Predictors of susceptibility and resilience in an animal model of posttraumatic stress disorder. Behav Neurosci 126, 749–61.

41.

Trocino

Majolini

Tazzoli

Filiou

Xiccato

(2013). Housing of growing rabbits in individual, bicellular and collective cages: Fear level and behavioural patterns. Animal 7, 633–39.

42.

Van Vleet

J. F.

Ferrans

V. J.

(1986). Myocardial diseases of animals. Am J Pathol 124, 98–178.

43.

Varga

Z. V.

Ferdinandy

Liaudet

Pacher

(2015). Drug-induced mitochondrial dysfunction and cardiotoxicity. Am J Physiol Heart Circ Physiol 309, H1453–67.

44.

Verga

Luzi

Carenzi

(2007). Effects of husbandry and management systems on physiology and behaviour of farmed and laboratory rabbits. Horm Behav 52, 122–29.

45.

Weber

H. W.

Van Der Walt

J. J.

(1975). Cardiomyopathy in crowded rabbits. Recent Adv Stud Cardiac Struct Metab 6, 471–77.

46.

World Health Organization. (2013). Guidelines on the Nonclinical Evaluation of Vaccine Adjuvants and Adjuvanted Vaccines, 40CFR Part 261. US Government Printing Office, Washington, DC.

47.

Y-Hassan

(2014). Acute cardiac sympathetic disruption in the pathogenesis of the Takotsubo syndrome: A systematic review of the literature to date. Cardiovasc Revasc Med 15, 35–42.

48.

Yoshimura

Imai

Hayashi

Hashimoto

(1989). Role of renin-angiotensin system in the pathogenesis of spontaneous myocardial fibrosis in Sprague-Dawley rats: Effect of long-term administration of captopril. Tohoku J Exp Med 157, 241–49.