Characterization of ANIT-Induced Toxicity using Precision-Cut Rat and Dog Liver Slices Cultured in a Dynamic Organ Roller System

Abstract

This article describes the toxicity of α-naphthylisothiocyanate (ANIT), a compound known to induce dose-dependent hepatobiliary toxicity in vivo, using the slice model. Liver slices (200 μm thick) from male Sprague–Dawley rats and male beagle dogs were cultured for 7 days while exposed to a range of ANIT concentrations (1– 100 μM for rat and 4–320 μM for dog). Tissues (and medium for dog) were evaluated using a panel of clinically relevant biomarkers for liver and histological endpoints to assess viability and proliferation. ANIT increased slice levels of enzyme biomarkers corresponding to biliary markers. At high concentrations (80–100 μM for rat, 320 μM for dog) a diminution of tissue enzyme levels was observed, corresponding to severe hepatobiliary injury. By days 5 and 7, biochemical markers in the medium of dog slices indicated an elevation of hepatocellular and biliary markers. Histologically for both species, minimal hepatocellular injury was noted, but proliferation of biliary epithelial cells (BEC) was observed using 5-bromo-2-deoxyuridine (BrdU) immunostaining. In rat slices, ANIT increased the expression of inducible nitrous oxide synthase (iNOS) within 12 hrs of exposure. In summary, additional experimentation using slice culture may further demonstrate its value in screening compounds that cause hepatobiliary toxicity.

Introduction

α-Naphthylisothiocyanate (ANIT) toxicity has been well documented (Lopez and Mazzanti, 1955; Goldfarb et al., 1962; Ungar et al., 1962; Roberts and Plaa, 1965; Desmet et al., 1968; Plaa, 1969; Plaa and Priestly, 1976; Kossor et al., 1993). Treatment of rats with ANIT (75–150 mg/kg) has resulted in biliary epithelial cell (BEC) necrosis followed by bile duct obstruction, cholestasis, and finally Type I BEC hyperplasia (Kossor et al., 1995). More recent publications have indicated that ANIT-induced hyperplasia may not be due to bile duct obstruction (Kossor et al., 1998). ANIT-induced toxicity to BEC and hepatocytes (HPC) has been shown to involve glutathione (Carpenter-Deyo et al., 1991; Jean and Roth, 1995), the accumulation of ANIT in the bile (Jean et al., 1995), and the release of toxic elements by neutrophils (Dahm et al., 1991a, 1991b; Hill and Roth, 1998; Jean et al., 1998; Hill et al., 1999). It has been suggested that the ANIT-induced toxicity can be attenuated by supplementation with Vitamin A or decreasing hepatic nonprotein sulfhydryl content (Dahm and Roth, 1991; Bailie et al., 1995).

Several studies have been conducted where the species sensitivity of ANIT exposure has been examined. It was found that rats, mice, and guinea pigs are more sensitive to ANIT toxicity than other species such as hamsters, rabbits and dogs (Capizzo and Roberts, 1971; Indacochea-Redmond and Plaa, 1971). Although one study found no hyperbilirubemia in dogs, even at high doses (Indacochea-Redmond and Plaa, 1971), another study contradicted these findings (Shibata et al., 1989b). It has been suggested that the differences of species’ sensitivities could be based on different rates and/or pathways of transformation (Capizzo and Roberts, 1971; Indacochea-Redmond and Plaa, 1971; Plaa and El-Hawari, 1976). Additional recent work indicates that the degree of increase of serum parameters was much lower in dogs than rats (when both species given the same dose per body weight) and the onset of hyperbilirubemia in dogs was later (Shibata et al., 1989a, 1989b).

The development of improved in vitro models for studying systemic toxic effects has become of great interest (Spielmann et al., 1998; ICCVAM/NTP, 2001; Pfaller et al., 2004; Tomaszewski, 2004). Liver toxicity targeting BEC cannot be detected using hepatocyte cultures alone. For example, dose-limiting injuries are sometimes due to injury to the biliary duct and the surrounding tissue (Grever and Grieshaber, 1993; Demetris, 1997). Thus, model systems that retain multiple cell types are necessary to detect toxicities associated with heterogenous cell populations.

Advances in slice culture techniques have extended the longevity of cultures, making it possible to examine slices as a predictive model for toxicology (Saulnier and Vickers, 2002; Behrsing et al., 2003; Fisher and Vickers, 2003; Vickers et al., 2004). Recently, dog and rat liver slices have been used to detect differential and cell-specific toxicities associated with the anticancer compounds geldanamycin and 17-allyaminogeldanamycin (Behrsing et al., 2005; Amin et al. 2005). Using plasma concentrations cited in in vivo experiments, slice experiments were conducted, and in vitro results were compared to published in vivo results. It was determined that the findings were remarkably similar with regard to biochemical and histological assessments. Current experimentation is further demonstrating the utility of the slice model with regard to predictive toxicology.

Using the rat and dog liver slice models, ANIT toxicity was examined using clinically relevant biochemical markers (aspartate amino transferase (AST), alanine amino transferase (ALT), Akaline phosphatase (ALP), and gamma glutamyl transferase (GGT) coupled with histological analysis using Hematoxylin and eosin (H&E) and the 5-bromo-2-deoxyuridine (BrdU) immunostain. Since ANIT toxicity involves an inflammatory component, induction of iNOS, a pleiotropic mediator of inflammation was studied in the rat slices. Results were compared to published in vivo data using these two species.

Materials and Methods

All work was conducted at SRI International. All animal work was done in accordance with institutional and governmental guiding principles in the use and care of animals. The following materials were used: a total of 11 (5 for iNOS experiments and the remaining for all other data) young adult male (250–350 g) Sprague–Dawley rats (Charles River, Hollister, California); glutathione (EM Science, Gibbstown, New Jersey); bovine serum albumin (BSA) and sodium pyruvate (ICN Pharmaceuticals, Aurora, Ohio); Waymouth’s MB 752/1 basal medium (#078-5105EL), Ca⁺-and Mg⁺ free phosphate-buffered saline (PBS), Glutamax I, and antibiotic-antimycotic solution (Invitrogen/Gibco, Carlsbad, California); Viaspan [Belzer—University of Wisconsin (UW) cold storage solution] (Fisher Scientific, Pittsburgh, Pennsylvania); other chemical supplies, including ANIT (Sigma, St Louis, Missouri). Equipment included the following: a tissue coring press and titanium inserts (Vitron, Tucson, Arizona); a 0.45-μm surfactant-free hydroanalysis mixed cellulose ester triton-free (HATF) filter paper used for slice placement (Millipore, Bedford, Massachusetts); PTFE membrane TF-200 and 0.2 μm filters (Pall Life Sciences, VWR International, West Chester, Pennsylvania); low-background, glass scintillation vials (Research Products International, Mt. Prospect, Illinois); TC-8 roller drum unit (New Brunswick Scientific, Edison, New Jersey).

Liver Slice Preparation

Rats were anesthetized with sodium pentobarbital (50 mg/kg), injected iv through the tail vein with 200 U heparin, and placed in a sterile biosafety hood, where all further work was done aseptically. After exposure of the abdominal cavity, the portal vein was perfused with ice-cold UW solution supplemented with 3 mM glutathione, 2 mM Glutamax I, 1x antibiotic-antimycotic solution, 30 μM L-ascorbic acid, 2 mM sodium pyruvate, 1 μM corticosterone, and 100 nM insulin (human). After the blood was cleared from the liver, it was quickly removed, and the lobes were separated with a scalpel and cored into cylinders 8 mm in diameter.

Beagle dog liver specimens were obtained during necropsy procedures from control animals in other ongoing studies. Dogs (Marshall Farms, Rose, NY) were quarantined for 14 days after arrival before the study began. On study completion, the dogs were necropsied following euthanasia with sodium pentobarbital (150 mg/kg iv). The liver tissue for liver slice studies (consisting of one or two lobes) was removed immediately after the abdominal cavity was opened. The lobes were immediately placed in ice-cold UW solution, supplemented with 3 mM glutathione, 2 mM Glutamax I, 1 × antibiotic-antimycotic solution, 30 μM L-ascorbic acid, 2 mM sodium pyruvate, 1 μM corticosterone, and 100 nM insulin (human). The vessel containing the lobes was placed in a biosafety hood where all further work was done aseptically.

The cores of either rat or dog liver were then sliced (using the Krumdieck slicer) in supplemented UW solution into ~200-μm-thick discs using a Krumdieck slicer (Alabama Research and Development, Munford, Alabama). Only the most uniform-shaped slices were selected for experiments; each contained approximately 3 mg protein. Using a sterile cotton swab, the slices were placed on sterile HATF paper inside the titanium inserts. All slice manipulations were done in ice-cold supplemented UW solution.

Slice Pre-incubation/Culture

The titanium inserts with the slices were placed in sterile scintillation vials, each containing 1.7 ml culture medium consisting of Waymouth’s MB 752/1 basal medium containing, per ml, 2 mg BSA, 0.084 mg gentamicin sulfate, 5 μg oleic acid, 5 μg linoleic acid, 0.5 μg DL-α-tocopherol, 7.9 μg D-thyroxine, 5 μg insulin, 5 μg transferrin, 5 ng selenium (ITS), 288 ng testosterone, 272 ng β-17estradiol, 39.3 ng dexamethasone, 30 ng glucagon, 0.02 U insulin, 0.2 μmol L-2-phosphate ascorbic acid, and 2 mM sodium pyruvate. The vials were then capped with sterilized open-end caps containing PTFE membrane filters, each held in place by a hole-punched Teflon liner to allow gas exchange with the external atmosphere. Vials were placed in the roller drum inside a humidified incubator at 37°C under 70% O₂/25% N₂/5% CO₂ or 75% O₂/20%N₂/5% air atmosphere (the latter environment proved slightly better than the former after trigas incubators were purchased). The roller drum containing the vials was rotated at 6–7 rpm. Following equilibration for 2–3 hrs, the zero time-point group was harvested. In a biosafety hood, the medium in the remaining vials was aspirated and replaced with designated warmed media with or without ANIT at designated concentrations. Medium containing ANIT was spiked using 1:1000 dilutions from a stock solution made up in ethanol. Control groups had ethanol at the same concentration. In the case of the highest concentration used (320 μM), ANIT was weighed and dissolved directly into the medium. The slice medium in each vial was replaced with fresh medium daily until the slices were harvested; for examination of the medium biomarker content, the collected medium (1 ml/vial) was pooled for respective slices and stored at 4°C until later biochemical analysis. After medium change, the vials were repositioned on the roller system in the incubator, and the incubation continued under the conditions described above until all groups (3–4 replicates/group) were harvested. Each group was comprised of 3–4 replicates (slices cultured in separate tubes). Experiments assessing the toxicity of ANIT were run for a total of 7 days, with cultures harvested on study days 0, 1, 3, 5, and 7.

Slice Harvest

At the indicated time points, slices and filter paper were removed from their titanium inserts and rinsed in PBS by briefly submerging the slice (still on its HATF paper) into a vessel containing ~20 ml PBS. The slices were bisected so that one-half could be used for biomarker analysis and the other for histology and morphological examination. For BrdU incorporation experiments, 20 μM BrdU was added to the medium and allowed to incubate with the slices for 18 hrs before their harvest.

Biomarker Analyses

Each slice section designated for biochemical analysis was transferred into a 1.5 ml Eppendorf tube containing 0.5 ml PBS + 0.5% Triton X-100 on ice. The sections were then homogenized and briefly sonicated at ice-cold temperatures. The resulting lysates were centrifuged at 9000 × g for 5 min to remove particulate matter. The resulting supernatants were stored at 4°C until the experiment was completed and then sent on ice with the corresponding media by courier to a local laboratory (Quality Clinical Labs, Mountain View, California) for analysis with a Hitachi 911 clinical analyzer. Biomarker (AST, ALT, ALP, and GGT) levels measured from slice lysates were taken within the first 24 hrs and analyzed periodically thereafter (tests indicated they were stable at 4°C for a minimum of 18 days). A small fraction of the lysate was retained for protein analysis using a Pierce BCA protein assay kit (VWR International, West Chester, Pennsylvania) and BSA standards in PBS + 0.5% Triton X-100. A chemiluminescent ALP-specific substrate (Michigan Diagnostics, Troy, Michigan) was used to measure ALP activity in the medium. For these experiments, medium ALP activity was extrapolated from an ALP standard curve. All medium ALP assays were performed in 96-well microtiter plates.

Histology

The faces of slice sections designated for morphological examination were covered by lens paper (prewetted in 10% buffered formalin) and placed between 2 foam inserts. This “sandwich” was first placed in a histological cassette, fixed in 10% buffered formalin for 18–24 hrs, then transferred to 70% alcohol, and finally submitted to a local laboratory (Biopathology Lab, South San Francisco, California) for paraffin embedding. H&E, Periodic Acid Schiff (PAS), Masson’s Trichrome, and BrdU immunostaining were done on 4-μm-thick tissue sections cut from the paraffin blocks. BrdU immunostaining was done using a staining kit, according to the protocol provided by the manufacturer (Zymed Labs, South San Francisco, California).

Semi-Quantititative Histological Analysis

H&E-stained slice sections were thoroughly examined under low (25 × and 50 ×) and medium (100 × and 200×) magnification and assessed for HPC and BEC viability. Following examination, a viability percentage score was assigned to both HPC and BEC. In addition, glycogen deposition in HPCs was estimated on a scale of 0–4, where 0 was no glycogen, 1 was minimal, 2 was low, 3 was moderate, and 4 was high. BEC proliferation was evaluated on sections immunostained with BrdU. From each slice section, the five portal tracts with the best morphology containing small to medium-size bile ducts were selected and counted for BrdU-positive BEC using a Zeiss Axioskop 2 equipped with Axiocam digital camera and KS 300 interactive imaging software. Positive cells present in the stroma or lining the portal vessels were excluded. On the basis of these numbers, a mean BEC proliferation count for each slice was deduced.

Western Blot

Protein concentrations for the liver slice lysates were determined using BCA reagent (Pierce Biotechnology, Rockford, IL). A positive control for iNOS was prepared with RAW 264.7 cells (mouse macrophage) treated with 10 μg/ml LPS. Samples were prepared for Western analysis according to the Invitrogen (Carlsbad, CA) protocol: 20 μg protein per lane were prepared in 4X loading buffer, heated at 70°C for 10 min, then run in 4–12% NuPage gels (Invitrogen, Carlsbad, CA) at 150 volts for 2 hours 45 minutes. After protein transfer onto Immobilon membrane (Millipore, Bedford, MA) and blocking in 5% dry milk in PBS + 0.1% Tween-20, membranes were probed for iNOS using BD Transduction Laboratories antibody (#610332) and horseradish peroxidase-conjugated anti-rabbit IgG secondary antibody (Santa Cruz Biotechnology, SC2062, Santa Cruz, CA). ECL Plus (Amersham Biosciences, Piscataway, NJ) was used for chemiluminescent detection. Liver slices from a total of five animals were used for Western blotting experiments.

Statistics

Replicates from control and treatment groups among experiments were consolidated and compared using t-test analysis. Values were considered significantly different if p < 0.05. Treatment group means and standard deviations were calculated. Values were converted to percent of control, and standard deviations were determined as a percent of the ratio between the treated groups and the control.

Results

Slice Culture

Rat liver slice cultures were well-maintained during experimentation for 7 days (Table 1). As seen previously with rat (Behrsing et al., 2005) and dog (Amin et al., 2005) liver slices, tissue AST and ALT levels diminish while ALP and GGT markers increase over time in culture. Histological results indicate good viability, and active glycogen accumulation is evident over the culture period. The viability of rat liver slices between days 3 to 7 ranged from 69% to 78% for HPC and from 87% to 90% for BEC.

Rat Slice ANIT Toxicity

Initial screening experiments were conducted using rat slices to determine the responsiveness of the slice model to ANIT exposure. Due to the negligible proliferation response in control tissue, day 1 BrdU was difficult to assess. However, by day 3, BrdU staining showed an increased number of proliferating BECs in ANIT-treated vs. control groups. Further experimentation also examined both slice histology and biochemical biomarkers corresponding to HPC and BEC, following exposure to a range of ANIT concentrations from day 1 until day 7 (Figures 1, and 2). Interestingly, ALP values measured in the ANIT treated tissue were found elevated at the earliest time point (Figure 1).

Initially, ranges of concentrations (1–8, 20, and 80–100 μM) were used to determine the ANIT concentrations that produced the best effects on tissue. Results from treated rat slices indicate that application of ANIT at the lowest concentration range (1–8 μM) caused a statistically significant increase in slice content of AST and ALT at all time points with the exception of day 1. HPC appeared normal except after prolonged application of ANIT; in that case, 20 and 80–100 μM ANIT treatments caused a slight decrease in viability by day 7 (87 ± 3% and 86 ± 6% of control values, respectively). Lower AST, ALT, and glycogen values are seen at day 7 only with the application of 80–100 μM ANIT (Figure 1). BEC viability was not affected at any concentration or time point. Biochemical markers for BEC were also found to be elevated at days 1, 3, and 5, but the increase diminished by day 7. An increase in the number of BrdU positive BEC cells was noted at days 3 (largest increase) and 5 with the application of 20 and 80–100 μM ANIT (Figure 1). Western blot analysis to determine iNOS levels in rat slice lysates showed that ANIT exposure increased the production of iNOS at all concentrations within 12 hrs (Figure 3). However by day 3, iNOS dropped to undetectable levels at all concentrations. The highest level of iNOS expression was seen at 24 hr at a concentration of 20 μM ANIT and this was followed by a small decrease in expression at the highest concentration (80 μM).

Dog Slice ANIT Toxicity

Experiments were conducted using dog tissue to compare species response to ANIT toxicity. Dog liver slices were cultured for 7 days, and average control cultures indicate slightly lower baseline (day 0) AST values, but higher ALT values (Table 2) compared to rat liver slices. ALP and GGT baseline values were approximately the same as rat values, but slice ALP was found to increase more and GGT less than rat slice values. Although a similar trend in biomarker level fluctuation was found when comparing 7 day cultures of rat liver slices, dog slices appeared to retain a higher overall viability of HPC and BEC (Table 1 and Table 2). Glycogen and BrdU staining of control slices appeared comparable for both rat and dog slices.

When exposed to a range of concentrations (4, 20, 80, and 320 μM), dog slices were less responsive to the toxic effects of ANIT. Although lower concentrations of ANIT (1–8 and 20 μM) caused an elevation of AST and/or ALT in rat slices at days 3, 5, and 7, this increase was not observed with dog slices at any concentration or time point (Figure 4). The highest concentration of ANIT applied (320 μM) tended to decrease ALT and glycogen values at several time points, suggesting HPC toxicity. Histological assessment indicated that HPC and BEC viability is unchanged except at the highest ANIT concentration of 320 μM (87%, 74%, and 74% of control for HPC, and 95%, 87%, and 84% of control for BEC, at days 3, 5, and 7, respectively). Although elevations of BEC biochemical parameters are noted with rat tissue on all days assayed (1, 3, 5, and 7), only slight elevations are noted with dog slices, and neither ALP nor GGT elevations were found to be statistically significant (p < 0.05). However, BrdU staining indicated increased BEC proliferation at days 5 and 7 (Figure 5), although it was less pronounced than that seen with rat slices at day 3 (Figure 1, Figure 4). Since biochemical biomarkers of slices did not reflect the histological observations of increased proliferation, slice medium was examined in an attempt to explain histological findings.

Dog Medium

Measurement of medium from ANIT-treated dog slices indicated that although no histological indication of HPC viability loss was noted, an increase of AST and ALT activity was found at days 5 and 7 (Figure 6). Previous work indicated that a direct correlation exists between the levels of ALP in the medium and the level of observed proliferation (Behrsing et al., 2005). Results from ALP measurements indicate an initial reduction of ALP release by day 3 (in a concentration-dependent manner), but increasing values over time. By day 7, average medium ALP values from all treatment concentrations are well above control values.

Discussion

ANIT has been widely used as a chemical inducer of cholestasis and biliary hyperplasia for models of toxicity (Rodriguez-Garay, 2003). In an attempt to further evaluate the value of liver slice model to predict in vivo toxicity, rat and dog liver slices were exposed to ANIT and the results assessed using accepted biochemical (AST, ALT, ALP, and GGT) and histological (H&E and BrdU staining) endpoints. The findings indicate relatively similar patterns of toxicity to those previously determined by in vivo studies using rats and dogs. In short review, some species such as rats have been found more sensitive to ANIT toxicity than others such as dogs (Capizzo and Roberts, 1971; Indacochea-Redmond and Plaa, 1971; Shibata et al., 1989a, 1989b). It has been speculated that the rate of ANIT transformation may be responsible for these differences (Capizzo and Roberts, 1971; Indacochea-Redmond and Plaa, 1971; Plaa and El-Hawari, 1976).

In both slices and in vivo studies, ANIT caused an increase in transaminase levels, although with different lag times following ANIT exposure. It was important to note that in most in vivo studies, ANIT is given as a single dose, but in the current in vitro studies ANIT was applied for the duration of the experimental time points. In vivo, serum transaminase measurements were found to increase within 24 h of ANIT treatment (Shibata et al., 1989a, 1989b; Jean and Roth, 1995). The application of 80 μM ANIT to rat slices resulted in a decrease of ALT and AST in tissue content as early as 48 hr in one of three experiments. However, an apparent decrease in tissue content of these enzymes does not necessarily translate into a detectable increase in the medium, which would better correlate to serum elevations of biomarkers.

When dog liver slices were used for ANIT experimentation, the incubation medium was also collected and examined for AST and ALT following ANIT treatment; the clinical analyzer used was not sufficiently sensitive to measure ALP or GGT in the medium. To examine a BEC marker in the medium, a highly sensitive luminescence assay was developed to measure ALP. Medium biochemistry found that although only ANIT concentrations of 80 μM or higher resulted in statistically lower ALT in tissue, the elevation of AST and ALT in medium was observed by day 5 at all concentrations (Figure 5). Work performed concurrently with this study indicated that the proliferation of BEC in slice cultures is reflected in levels of ALP in the medium (Behrsing et al., 2005). For example, control cultures, which normally display increased proliferation over time, show corresponding increased levels of ALP in the medium; conversely, increasing biliary toxicity that manifests as inhibited proliferation results in lowered ALP in the medium.

Using this same paradigm, our experiments confirmed histological indications from BrdU staining that BEC proliferation was indeed occurring following ANIT application. BrdU staining was not assessed at day 1 since there were only occasional BrdU-positive BEC slices, and no proliferation at all was observed in many control slices. Because of this very low BrdU-positive cell count, a meaningful interpretation of the data was not feasible. The delay (by comparison with in vivo data) of hepatocyte biomarkers in the medium of slices appears longer than that observed in vivo. Reviewing the sequence of events leading to in vivo toxicity suggests one possible explanation. In vivo, the infiltration of neutrophils has been associated with ANIT-induced toxicity. The in vitro slice system lacks a functioning vascular system that allows for the recruitment of neutrophils into the liver. Any neutrophil reaction in the slice is limited to the neutrophils present in the tissue at the time of slice preparation, and therefore neutrophil recruitment cannot occur. However an interesting finding in this study that points to an inflammatory process as a mediator of ANIT toxicity was the detection of increased induction of iNOS, a key enzyme involved in nitrous oxide (NO) production. NO is a potent biological mediator in a myriad of physiological and pathological events, particularly inflammation (Chen et al., 2003). It has been shown that hepatocytes and Kupffer cells are the major source of iNOS in the liver and its levels are increased following acute insult by hepatotoxicants (Sass et al., 2005). The increase in iNOS levels following ANIT exposure is indicative of hepatocellular toxicity and activation of Kupffer cells. The concentration of ANIT that caused maximal induction of iNOS in rat liver slices was 20 μM. Interestingly, the highest concentration of 80 μM did not result in further increase of iNOS. A possible explanation of this observation could be an ANIT-induced acute lethal insult to the cells before the inflammatory pathways are activated and detected. Moreover, iNOS induction was seen only in the first 24 hr of ANIT exposure signifying its possible role as a transient early injury mediator.

Histologically, rat and dog slice tissues maintained good viability throughout the 7 days of experimentation. The changes in HPC and BEC biochemical and histological biomarkers have previously been described and are considered to be a normal part of slice culture using the Waymouths-based medium. It was noted that more BEC stained in dog slice control cultures, and this was attributed to a higher complement of biliary cells due to the larger size of the liver. An interesting histological finding seen in rat slices, and more frequently in ANIT-treated slices, was the presence of secretions plugging mostly the small to medium-size biliary channels, possibly indicating cholestasis and/or biliary obstruction caused by ANIT. Although the composition of these obstructions was not investigated, their presence may warrant further experimentation, especially since this phenomenon was not observed in dog liver slices.

One drawback of analyzing slice tissue data is that all biomarker values are normalized between slices by dividing out the protein content to eliminate subtle differences. Thus, even with significant induced toxicity, the lower tissue or medium biomarker values are elevated by normalizing with lower slice protein content. This was the case when 320 μM ANIT was applied to dog slices for 7 days. Although the biochemical endpoints suggest moderate injury, the damage is considerably underestimated because the average protein content of these slices was only 77% of control. In other words, the values represented in the biochemical data only correspond to the remaining tissue after dead cells have sloughed off. It was not feasible to include protein levels as endpoints because slices were bisected to maximize utility for biochemical and histological assessment. The slices are bisected manually, and it is assumed that this procedure is not 100% accurate. However, histological examination provides an opportunity to directly visualize the extent of tissue damage. Therefore, the use of histology enables a better assessment of severe toxicity in slices that show greater loss of protein.

The in vivo toxicity of ANIT, as determined by the degree of elevation of serum transaminases and bilirubin, was found to be more severe in rats than dogs (Shibata et al., 1989a, 1989b). The in vitro studies conducted using rat and dog tissue slices concurred with these findings. Our tissue slice results using biochemical and histological endpoints also indicate that the rat is more sensitive to ANIT toxicity than the dog. BEC biomarkers (ALP, GGT, and BrdU incorporation) are more highly elevated in rat than in dog, and rat HPC are also more susceptible to damage at the same concentration.

This study further indicates the utility of the slice model for detecting in vivo toxicities in a concentration-dependent and cell-specific manner. It also confirms the ability of the slice model to discriminate between species’ susceptibility to toxicity.