The Pivotal Role of Electron Microscopic Evaluation in Investigation of the Cardiotoxicity of bis(2-chloroethoxy)methane in Rats and Mice

Heart disease is the leading cause of mortality (National Center for Health Statistics 2005a) and health care costs in the United States (National Center for Health Statistics 2005b). It is important to develop sensitive techniques for identifying environmental exposures that might contribute to the burden of heart disease, and this paper describes our continuing work to develop these techniques using a bis-2-(chloroethoxy)-methane (CEM) rodent model that develops cardiotoxicity after only a few days of dosing.

The metabolism of CEM to thiodiglycolic acid is shared with a number of other chemicals including ifosfamide, monochloroacetic acid, chloroacetaldehyde, trichloroethane, trichloroethylene, 1-dichloroethylene, cyclophosphamide, vinylidene chloride, and vinyl chloride that produce cardiotoxicity in rodents and/or humans (Dunnick, et al. 2006; Dunnick, Johnson, et al. 2004; Dunnick, Lieuallen, et al. 2004).

Bis-2-(chloroethoxy)methane is a synthetic compound used in the production of polysulfide elastomers. It is widely used in sealant applications because of its resistance to solvents and high temperature degradation. It has been identified in industrial waste from metal finishing, plastics, rubber and chemical manufacturing, and steam electric power industries, and in water from inland highways (Dunnick, Johnson, et al. 2004; Dunnick, Lieuallen, et al. 2004).

The comparative cardiotoxicity after oral administration of CEM in F344/N rats and B6C3F1 mice is reported in this paper. Solutions of CEM were administered to male F344 rats and male B6C3F1 mice (Taconic Laboratories, Germantown, NY, USA), for up to sixteen days. Dosing started when the mice were thirteen to fifteen weeks of age, whereas dosing of the rats started at seven or eight weeks of age. Mice were housed one per cage in polycarbonate cages in rooms maintained at 72°F + 2°F with relative humidity at 50% + 10% and a twelve-hour light/dark cycle. Control and treated groups received irradiated NTP-2000 diet (Zeigler Brothers, Gardners, PA, USA) ad libitum.

Bis-2-(chloroethoxy)methane (Pfaltz and Bauer, Waterbury, CT, USA, stock number B12370, purity 97%, lot number 111005-5) was administered by oral gavage in corn oil at doses of 0, 25, 50, 75, or 100 mg/kg/day to mice and to rats at doses of 0 or 50 mg/kg/day. The mouse study consisted of two-day treated animals (two each at 25, 50, 75, or 100 mg/kg/day), five-day treated animals (two each at 25, 50, 75, or 100 mg/kg), and sixteen-day treated animals (two controls, nine at 25 mg/kg/day, six at 50 mg/kg/day, six at 75 mg/kg/day, and six at 100 mg/kg/day), for a total of forty-five mice. The rat study consisted of two-day treated animals (four at 50 mg/kg/day), five-day treated animals (four at 50 mg/kg/day), seven-day treated animals (four at 50 mg/kg/day), and sixteen-day treated animals (four controls and four at 50 mg/kg/day), for a total of twenty rats. The volume of the dosing solution delivered to rats was 5 mL/kg body weight and to mice 10 mL/kg body weight. Animals were dosed between 8 a.m. and 9 a.m. and they were sacrificed after two, five, seven, or sixteen days of dosing (see Table 1). All sacrifices occurred at 10 a.m. twenty-four hours after the last dose.

The care of animals on this study was according to NIH procedures, as described in the United States Public Health Service Policy on Humane Care and Use of Laboratory Animals, available from the Office of Laboratory Animal Welfare, national Institutes of Health, Department of Health and Human Services, RKLI, Suite 360, MSC 7982, 6705 Rockledge Drive, Bethesda, MD 20892-7982 or online at http://grants.nih.gov/grants/olaw/olaw.htm#pol. The NIEHS animal care committee reviewed the protocol for this study and monitored the study through its in-life phase.

All mice were euthanized with carbon dioxide one day after completion of dosing days 2, 3, 5, and 16, whereas rats were euthanized with carbon dioxide one day after completion of dosing days 2, 5, 7, and 16. At necropsy, all organs and tissues were examined for grossly visible lesions. For light microscopic examination, the heart was fixed in 10% neutral buffered formalin, processed, trimmed, embedded in paraffin, sectioned to a thickness of 4 to 6 mm, and stained with hematoxylin and eosin (H&E). Briefly, the heart was sectioned transversely into two halves, both placed cut-surface down per cassette and embedded. A semiquantitative grading scheme was used to evaluate the extent of the lesions such as myofiber vacuolation, myofiber fragmentation, inflammatory cell infiltration, and/or fibrosis, in the heart sections, as follows: minimal (grade 1) lesions involved less than 10% of the heart section; mild (grade 2), 11% to 40%; moderate (grade 3), 41% to 80%; marked (grade 4), 81% to 100%.

Two controls, two 2-day–treated mice at 100 mg/kg/day, and six 16-day–treated mice at 100 mg/kg/day were examined for cardiotoxicity by electron microscopic analysis. A sample from the right ventricle, left ventricle, and septum was taken from selected mouse hearts for electron microscopic evaluation. Following swift excision, the tissues were immersed immediately in fixative (3% glutaraldehyde [Ladd Research, Burlington, VT, USA] buffered in 0.1M sodium cacodylate [Electron Microscopy Sciences, Fort Washington, PA, USA], pH 7.2), cut into 1-mm cubes. Following two to three days of storage in the fixative, specimens were rinsed in buffer, postfixed in cacodylate-buffered 1% osmium tetroxide (Electron Microscopy Sciences), en bloc–stained in 2% aqueous uranyl acetate (Ted Pella, Inc., Redding, CA, USA), dehydrated through a series of graded alcohols and propylene oxide, and embedded in Polybed 812 (Polysciences, Warrington, PA, USA). At least three blocks from each region of each animal were analyzed. Semithin (0.5 mm) sections stained with 1% toluidine blue + 1% sodium borate were scanned light-microscopically to locate regions containing longitudinal fibers. Ultrathin (90-nm) sections were cut from these regions, placed on 150-mesh copper grids, stained with 5% uranyl acetate followed by Reynolds lead citrate, and examined in a Philips electron microscope (TECNA 12).

Although there were no clinical signs in treated rats or mice at the dose levels used in this study, histopathologic and/or electron microscopic evaluation revealed treatment-related cardiotoxicity.

Our results indicate that rats were more sensitive than mice to the degree of functional impairment of CEM. Light microscopy revealed damage throughout the rat hearts, including in the ventricular myocardium and interventricular septum (Table 1). This damage included treatment-related myocardial degeneration, a diagnosis that includes myocardial inflammation, myofiber vacuolation, and myofiber necrosis. At study day 16, the severity of the lesions was not greater than that seen at day 2, after two CEM doses in rats. The spectrum of lesions observed in these studies in rats is similar to what was seen after dermal administration (Dunnick, Lieuallen, et al. 2004). Since evaluation by light microscopy revealed CEM-induced cardiotoxic lesions in rats, electron microscopic evaluation was not deemed necessary. In addition, we have previously reported on the CEM evaluation of early changes in the heart of rats treated with CEM (Dunnick, Johnson, et al. 2004).

In mice, there was no evidence for light microscopic treatment-related cardiac lesions, However, ultrastructural evaluation of two mice, one mouse treated with two doses of 100 mg/kg CEM and another receiving sixteen doses of 100 mg/kg CEM, showed cardiotoxic lesions. The cardiotoxic lesions were located in the interventricular septum and consisted of clusters of mitochondria with fragmented cristae, mitochondrial electronlucency owing to loss of cristae, as well as inclusions consisting of concentric or nonconcentric whorls and/or other unusual membranous profiles in one animal treated with 100 mg/kg for sixteen days (Figure 1A–H). In addition, in one of these samples, multiple intrasarcoplasmic vacuoles were noted, consisting of electronlucent areas containing disintegrating debris or that were devoid of any contents (Figure 1G and H). Such irregular empty spaces may reflect previously disintegrating mitochondria.

These results showed that electron microscopic analysis was able to characterize the morphological manifestations of the heart toxicity in mice hearts beyond the level possible with conventional histological methods. This analysis of left ventricle and interventricular septum revealed vacuolations arising predominantly from disintegrating mitochondria and, to a lesser extent, expanding sacroplasma reticulum (SR). Our previous studies (Dunnick, Johnson, et al. 2004; Dunnick, Lieuallen, et al. 2004) indicated that CEM-induced cardiac damage can be widespread through the myocardium.

It is speculated that mitochondrial disintegration and vacuolation apparently led to subsequent myofibrillary degeneration and destruction and eventual myocytic rupture. Since we did not observe proliferation or reduplication of the SR membrane, typified by bundled tubules (Cheville 1994), a hallmark of an adaptive response, we inferred that the CEM-induced damage was indeed degenerative. In contrast to the present study in mice, in which changes were limited to the mitochondria, our EM studies in rats (Dunnick, Johnson, et al. 2004) indicated that alterations in mitochondria constituted the most prominent feature of this lesion, but distention of SR, myofibrillary degeneration, and occasional Z-banding misalignments were also apparent.

Mitochondrial impairment may be initiated even before morphological manifestations are detectable. Damage to mitochondria can impair energy production, release triggers for apoptosis such as cytochrome c, and/or decrease the ability to eliminate oxygen radicals through the manganese-containing mitochondrial superoxide dismutase. Loss of mitochondrial function may also result in an accumulation of toxic compounds in the myocyte, such as those produced from acidosis or a buildup of lactate (Lesnefsky, Moghaddas, et al. 2001). A decrease in mitochondrial function of at least 30% to 50% is ordinarily required before lower rates of energy production occur (Lesnefsky, Moghaddas, et al. 2001). Most likely, this degree of damage was not attained in our study, because treated mice had no clinical signs of cardiotoxicity, suggesting that there was a sufficient reserve of energy-generational capacity to survive asymptomatically.

Although cardiac failure arises from many general, diverse causes (Jessup and Brozena 2003), mitochondrial damage and malfunction are widely associated with human heart disease, including dilated and hypertrophic cardiomyopathy, ischemic and alcoholic cardiomyopathy, and other syndromes (Marin-Garcia, Ananthakrishnan, and Goldenthal 1995). Mitochondrial defects may arise from mutations or deletions in the mitochondrial genome (Marin-Garcia and Goldenthal 2002) or through free-radical damage (Childs et al. 2002; Cummings et al. 1992). Megamitochondria, as observed in CEM cardiotoxicity, have been closely associated with the generation of free radicals (Wakabayashi et al. 1997). Mitochondria damage may also lead to free radical formation and oxidative damage (Fariss et al. 2005).

One study suggests that thiodiglycolic acid, the proposed active metabolite of CEM, can interfere with oxidation of palmitic acid, a long-chain fatty acid, but not the metabolism of succinic acid that would be metabolized by the Krebs cycle (Visarius et al. 1998). The dicarboxylic acid structure of thiodiglycolic acid may enable it to compete with the fatty-acid metabolic pathways that involve activation of fatty acids by long-chain fatty acid CoA synthetase, followed by formation of acylcarnitines by carnitine palmitoyltransferases (Lesnefsky, Slabe, et al. 2001). This proposed mechanism might in part explain the CEM-induced heart toxicity and that of other chemicals metabolized to thiodiglycolic acid.

Sudden cardiac death is seen in humans without known preexisting conditions (Lopshire and Zipes 2006; Zipes et al. 2006). Occult cardiotoxicity may occur in rodents where there are no clinical signs or histopathologic evidence for cardiotoxicity, but upon subsequent cardiac toxic events (e.g., ischemia, exposure to a cardiotoxin), the first exposure magnifies the severity of the second event (Golomb et al. 2007). Thus, even without CEM-induced clinical signs of cardiac toxicity, cardiac damage may occur and lead to more severe cardiac toxicity from subsequent exposures.

This study showed that the F344/N rat is a more sensitive model than the B6C3F1 mouse for detecting cardiotoxicity. The results of this study showed that use of electron microscopy is an important tool in identifying the cardiotoxic potential of chemical/drug exposures.