Abstract
1,3,5-Trinitrobenzene (TNB) is a munitions chemical that causes gliovascular lesions in the brain stem of rats similar to those produced by thiamine deficiency and nitroaromatic compounds, including m-dinitrobenzene. To identify neuropathic indices of toxicity, the effects of varying concentrations (0 to 2 mM) of TNB on cytotoxicity and cellular metabolic activity were examined using cultured astrocytes from Fischer-344 rats. The cytotoxicity was assessed by lactate dehydrogenase (LDH) leakage into the culture medium. Astrocyte metabolic activity was assessed by measuring the conversion of a tetrazolium salt to a formazan product. Additionally, the effects of oxidative stress on cellular metabolic activity - were determined by varying oxygen tension via alteration of culture media depth. In vitro, the toxic concentration 50% (TC50) of TNB, which induced cell death, was 16 μM following a 24-h exposure. The concentration of TNB that reduced cellular metabolic activity by 50% was 29 μM following a 24-h exposure. Varying the depth of the culture media did not influence the cellular metabolic activity in control or TNB-treated astrocytes. These results support the hypothesis that TNB induced neurotoxicity could partially be mediated via injury to astrocytes, a major component of the blood-brain barrier.
1,3,5-Trinitrobenzene (TNB) is a munitions chemical produced following the oxidation of trinitrotoluene (TNT) and the decarboxylation of the resulting trinitrobenzoic acid. It is used primarly in explosives and munitions and has had limited use in the vulcanization of rubber (Barnhart 1981). However, most TNB contamination results from photodegradation of TNT. TNB is a soil and water contaminant at installations that are currently or have been involved in the manufacture of explosives. Previous studies in Fischer-344 rats receiving single daily oral doses of trinitrobenzene over 6 to 10 days have resulted in methemoglobinemia, hemolytic anemia, testicular atrophy, and neurologic pathology (Chandra 1995; Chandra, Qualls, and Reddy 1995a; Chandra et al. 1997a; Chandra, Qualls, and Reddy 1997b; Chandra et al. 1999). Ataxia progressing to recumbence is a manifestation of the neurologic abnormalities. In experimental intoxications with TNB, symmetrical morphological changes in the brain stem associated with the ataxia have been reported. These changes involve petechial hemorrhages and necrosis (malacia) with reactive gliosis in the cerebellar peduncles. The presence of vacuolation and associated extravasated serum proteins is an indication of vasogenic brain edema, which appears to be a critical event in TNB toxicity. The lesions appear primarily glial and vascular in origin, with secondary neuronal involvement (Chandra et al. 1995b, 1999).
The mechanisms by which TNB exerts its toxic action on the central nervous system (CNS) are unclear, but previous investigators have shown that TNB-mediated tissue damage is accompanied by breakdown of the blood-brain barrier (BBB) with extravasation of albumin (Chandra et al. 1999). Astrocytes contribute to the maintenance and induction of the BBB characteristics in the brain endothelium, including tightness of tight junctions (Stewart and Wiley 1981; Butt, Jones, and Abbott 1990; Rubin et al. 1991). Changes in the permeability of membranes are a common sequela of toxic insult at either the subcellular or cellular level.
The neurotoxic effects of dinitrobenzene (DNB) have been extensively studied both in vitro and in vivo. Dinitrobenzene has been shown to induce vacuolated lesions together with petechial hemorrhages and edematous swelling in the brain stem of rats (Philbert et al. 1987; Romero et al. 1991, 1995; Ray et al. 1994). Increases in blood flow and glucose consumption in areas of the brain susceptible to toxic insult have been shown to precede the onset of morphologic changes, suggesting an initial involvement of the vascular bed and subsequently the BBB in the development of the lesions (Romero et al. 1991, 1995).
Lactate dehydrogenase (LDH) release in vitro has been used as an index of cytotoxicity in astroctyes with DNB (Romero et al. 1995, 1996). The basal metabolic rate has been suggested as being a possible critical factor in the neurotoxicity of DNB, cells with increased metabolic rates being more sensitive to damage. In single cultures, astrocytes have been shown to be more metabolically active than endothelial cells, which may contribute to the increased sensitivity of astrocytes to DNB. The highly toxic hydroxyl radical generated from H2O2 by the Fenton reaction has been theorized to play a partial role in DNB-induced toxicity in both astrocytes and endothelial cells. Excessive production of reactive oxygen species and other free radicals is also thought to significantly contribute to futile redox cycling in the toxicity of DNB (Romero et al. 1996).
Low oxygen tension favors the reduction of nitroaromatic and nitroheterocyclic compounds in vitro by nitroreductases, thus forming the nitro radical anion by a single electron transfer from an electron donor, which is usually NADPH (Mason and Josephy 1985; Biaglow et al. 1986). Higher oxygen tension favors the reaction of the nitro radical with molecular oxygen, which is usually faster than the ability of the enzyme to add a second electron, leading to production of superoxide radicals and the regeneration of the parent nitro compound (Mason 1990). Previous studies have demonstrated that decreasing medium depth in culture wells does not result in increased toxicity to type II pneumocytes from the superoxide generating compound paraquat (Hoet et al. 1997). However, others have observed increased toxicity to Hep G2 cells from paraquat by decreasing the culture well medium depth (Qualls et al. 2000).
The major objectives of this study were to determine the levels of TNB cytotoxic to astrocytes in culture by measuring release of LDH, and monitoring changes in cellular metabolic activity. Additional experiments were conducted to examine the influence of media depth changes on metabolic activity.
MATERIALS AND METHODS
Materials
TNB (99.83% purity) was obtained from Naval Surface Warfare Center, MD, and the purity of the compound was confirmed by high-performance liquid chromatography (HPLC). DNB (99.8 % purity), Hank’s balanced salt solution (HBSS), glucose, sucrose, sodium ascorbate, and sodium bicarbonate were purchased from Sigma Chemical (St. Louis, MO). Monoclonal glial fibrillary acidic protein (GFAP) antibody was purchased from BioGenex (San Ramon, CA). CytoTox 96 Non-Radioactive Cytotoxicity Assay kits and the CellTiter 96 Aqueous One Solution Cell Proliferation Assay kits were obtained from Promega (Madison, WI). Dulbecco’s modified Eagle’s medium (DMEM), N-2-hydroxyethyl piperazine-N′-2-ethane-sulfonic acid (HEPES), and sodium pyruvate were purchased from JRH Biosciences (Lenexa, KS). Trypsin, fetal bovine serum (FBS), penicillin G, streptomycin, and
Cell Culture
Primary cultures of type I astrocytes were prepared from the hippocampi of 2-day-old Fischer 344 rat pups using a modification of a previously established procedure (Booher and Sensenbrenner 1972). The hippocampi were dissected in ice-cold Ca2+- and Mg2+-free HBSS and freed of meninges and chorid plexus. HBSS contained 17 mM glucose, 22 mM sucrose, 10 mM HEPES, 100 U/ml penicillin G, and 100 mg/ml streptomycin at pH 7.3 to 7.4. Tissue desegregations was done by digestion in 0.125% trypsin at 37°C for 30 min. Digested tissue was filtered through sterile cheesecloth and trypsinization halted by addition of 10 ml FBS. The suspension was spun at 750 rpm for 10 min and then the supernatant was removed. Cells were resuspended in culture medium and then seeded on poly-
Determination of Cytotoxicity
Single cultures of astrocytes were seeded on Costar 96-well cell culture plates and allowed to reach confluence. Cultures were treated with different concentrations (0 μM, 0.002 μM, 0.02 μM, 0.2 μM, 2 μM, 20 μM, 30 μM, 40 μM, 50 μM, 60 μM, 70 μM, 80 μM, 1 mM, and 2 mM) of TNB in 100 μl of the medium (using 0.5% dimethyl sulfoxide [DMSO] as a vehicle) and allowed to incubate for 24 h. Medium was not changed during the experiments. Astrocyte cytotoxicity was assessed with the CytoTox 96 Non-Radioactive Cytotoxicity Assay kit. Cell death was assessed by the percentage of LDH released into the culture medium upon cell lysis. Released LDH was measured with a 30-min coupled enzymatic assay that converts NAD+ and lactate to pyruvate and NADH. 2-(4-Iodophenyl)-3-(4-nitrophenyl)-5-phenyl tetrazolium chloride (INT) was converted into a red formazan product in the presence of NADH that was detected colorimetrically at 490 nm with a 96-well plate reader. The amount of color formed was proportional to quantity of LDH released from lysed cells.
Determination of Cellular Metabolic Activity
Astrocyte cultures that had reached confluence on Costar 96-well culture plates were treated with different concentrations (0 μM, 20 μM, 30 μM, 40 μM, 50 μM, 60 μM, 70 μM, 80 μM, 90 μM, 100 μM, and 125 μM) of TNB in 100 μl of the medium (using 0.5% DMSO as a vehicle) and allowed to incubate for 24 h. Medium was not changed during the experiments. Cellular metabolic activity was then compared in astrocytes exposed to varying concentrations of either TNB or m-DNB to assess differences in toxicity to these two nitroaromatic compounds. Cellular metabolic activity was assessed in astrocytes with the CellTiter 96 Aqueous One Solution Cell Proliferation Assay kit. Metabolic activity of astrocytes was measured via 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymetho-xyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS), a water-soluble tetrazolium salt, in the presence of phenazine ethosul-fate (PES), an intermediate electron acceptor. PES amplifies the MTS signal when it is converted to a formazan product that can be determined colorimetrically. The conversion of MTS to the formazan product is accomplished via NADPH or NADH by dehydrogenase enzymes in metabolically active cells. The reduction of MTS to the MTS formazan product was recorded at 490 nm with a 96-well plate reader. Reduction of the tetrazolium salt occurs in active mitochondria residing in living cells (Sun et al. 1997). This assay was also repeated on single cultures of astrocytes exposed to similar concentrations of DNB (0 mM, 0.5 mM, 1.0 mM, and 2.0 mM) used by Romero et al. (1996).
Effects of Media Depth on Cellular Metabolic Activity
Costar 96-well culture plates containing astrocyte monolayers were treated with different concentrations (0 μM, 10 μM, 20 μM, 30 μM, 40 μM, 50 μM, 60 μM, 70 μM, 80 μM, and 90 μM) of TNB in the medium (using 0.5% DMSO as a vehicle) and allowed to incubate for 2 h. Different volumes (50, 100, and 200 μl) of medium were used in each well to determine if increasing oxygen tension would increase cellular toxicity. Decreasing medium volumes in individual culture chamber wells should allow for greater oxygen saturation of the medium and oxygen availability for radical formation (Hoet, Demedts, and Nemery 1997) Medium was not changed during the experiments. Cellular metabolic activity was assessed in astrocytes with the CellTiter 96 Aqueous One Solution Cell Proliferation Assay kit as described previously.
Data Analysis
All data were initially examined by analysis of variance (ANOVA) with a general linear model (GLM) using SAS (SAS Institute, Carey, NC). Mean ± standard deviation of astrocytic LDH release and MTS formazan formation for each group were calculated and compared to appropriate controls using Dunnett’s test. Cytotoxicity, measured by LDH release from astrocytes, and metabolic activity, measured by MTS formazan formation, were both transformed using a probit by logarithmic dose scale. The toxic concentration 50% (TC50) and TNB quantity resulting in a 50% decrease in cellular metabolic activity were determined using linear regression of transformed data.
RESULTS
Determination of Cytotoxicity
TNB at concentrations of 70 μM or higher produced a loss of continuity of the cell monolayer when cultured astrocytes were observed microscopically after 24 h of exposure. Closely apposed contacts between cells were lost and the monolayer appeared to have gaps where cells had lysed in comparison to controls not exposed to TNB (Figures 1 and 2). Concentrations of 60 μM or less did not induce noticeable morphologic changes in astrocytes over 24 h.
The threshold TNB concentration that induced an increase in the percentage of dead astrocytes, measured by LDH release into the culture medium, was 20 μM after 24 h exposure (Figure 3). Transformation of the percentage cell death versus linear dose data to a probit response versus log dose produced a linear, dose-response curve. These lines allow for more precise predictions using linear regression analysis. The concentration of TNB eliciting 50% cell death in rat astrocytes after 24 h incubation is 16 μM (Figure 4).
Determination of Cellular Metabolic Activity
Cellular metabolic activity was assessed in cultured astrocytes after exposure to both TNB and DNB. Cellular metabolic activity was increased at 1.0 mM DNB and significantly decreased in astrocytes exposed to 2.0 mM (Figure 5).
Cultured astrocytes displayed increased sensitivity to the toxic effects of TNB (micromolar range) in comparison with astrocytes treated with DNB (millimolar range) (Figure 6). Astrocytes displayed a 50% decrease in metabolic activity when exposed to 29 μM TNB (Figure 7).
Effects of Medium Depth on Cellular Metabolic Activity
Changing medium depth to allow greater oxygen penetration of the culture medium layer did not significantly change cellular metabolic activity from the three volumes (50, 100, and 200 μl) used in this experiment (Figure 8). Different incubation times also did not change the results with different media depths (results not shown).
DISCUSSION
This study represents a new approach towards discovery of the neurotoxic mechanisms of TNB. All previous investigations with TNB neurotoxicity have focused on in vivo exposure in the rat. Neuropathology induced by TNB has been shown to be accompanied by the breakdown of the blood-brain barrier (Chandra et al. 1999). In vivo work has helped to identify areas of brain pathology but has not examined the response of individual cells of these brain regions, specifically astrocytes and endothelial cells of the BBB. This investigation was the first attempt to begin looking at individual cells, specifically astrocytes and their response to toxic concentrations of TNB. Investigations were also aimed at comparing the response of astrocytes exposed to TNB with DNB, another nitroaromatic compound producing lesions in similar regions of the rat brain.
The release of LDH from single or mixed cultures of brain cells (astrocytes, endothelial cells, and neurons) is often used as a measure of cytotoxicity or loss of cellular viability (Romero et al. 1995, 1996; Sparapani et al. 1997; Deshpande and Nishino 1998; Robb and Connor 1998). The toxic concentration 50% (TC50) of TNB that induced cell death in cultured astrocytes was 16 μM following a 24-h exposure. Romero et al. (1996) reported dosage levels of 1.0 mM DNB resulting in approximately 10% loss of cell viability and concentrations of 2.0 mM causing greater than 90% cell death. The concentration of TNB required to elicit the same level of cytotoxicity was at least 50 times less than that of DNB. These results demonstrate that in single cultures of astrocytes, TNB has greater toxicity than DNB.
Cellular metabolic activity may play a critical role in the neurotoxicity caused by TNB, cells with higher metabolic rates being more sensitive to damage. Investigators working with DNB have found a strong relationship between brain areas susceptible to damage in experimental energy deprivation syndromes and the regions displaying the highest metabolic rates (Ray et al. 1992; Cavanagh 1993). Early increases in glucose consumption in these brain areas damaged by DNB have been shown to precede the morphologic lesion (Romero et al. 1995). The concentration of TNB that reduced cellular metabolic activity by 50% was 29 μM following 24 h exposure. In comparison with TNB, the concentration of m-DNB necessary to produce a 50% reduction in cellular metabolic activity was at least 50-fold greater than TNB. The dosage levels of TNB and DNB required to elicit a 50% change in metabolic activity were greater than the concentrations required for 50% cell death (as measured by LDH release). This might suggest that metabolic activity is a more accurate indicator of cell death than LDH release. To make sure that TNB alone was not reducing MTS to the MTS formazan product, culture wells were set up without astrocytes and medium with TNB at different concentrations (similar to the metabolic activity experiment) was run via the MTS assay. No significant reduction of MTS occurred when TNB-treated medium was run via the MTS assay without astrocytes (data not shown). Astrocytes exhibited increases in metabolic activity at concentrations of 20 μM TNB and 1.0 mM DNB. These increases in metabolic activity were attributed to stimulation of these cells, with subsequent increased metabolic output in response to toxic insults (TNB and DNB).
Different volumes (50, 100, and 200 μl) of medium were used in wells to determine if a mechanism requiring increased oxygen tension, such as redox cycling, would increase cellular toxicity. Decreasing medium volumes in individual culture chamber wells should allow for greater oxygen saturation of the medium and oxygen availability for radical formation. Varying the depth of the culture medium did not influence the cellular metabolic activity in control or TNB-treated astrocytes.
Another striking finding of this study is the difference between toxicity observed between TNB and DNB in vivo and in vitro. Three 10-mg/kg doses of m-DNB are ultimately neurotoxic to rats, with lesions developing 12 h after the third dose (Romero et al. 1991). Five to 10 71-mg/kg doses of TNB are neurotoxic to rats (either by observed clinical signs or histopathologic changes in the brain stem and cerebellum) (Chandra, Qualls, and Reddy 1995a). Therefore, DNB appears to exhibit greater toxicity in vivo versus TNB. The concentration of TNB resulting in 50% cell death of cultured astrocytes (measured by LDH release) was 16 μM after 24 h incubation. The concentration of DNB resulting in the same level of cell death of cultured astrocytes was between 1 and 2 mM after 24 h incubation (Romero et al. 1996). Therefore, although DNB appears to exhibit greater toxicity in vivo, TNB displays higher toxicity in vitro. Several possible explanations for this include differences in solubility, different metabolites produced after metabolism in organs other than the brain, or differences in absorption in the gastrointestinal tract.
In summary, the results of these studies demonstrate the ability of astrocytes to be utilized in culture and quantify changes when exposed to TNB or other nitroaromatic compounds. This experimental model begins to examine astrocytes, a major component of the BBB and their response to TNB. The determination of a range of TNB concentrations in vitro that can show astrocyte death (measured by LDH release or decreased metabolic activity) and should be a useful tool for cell culture investigations. Culture medium depth does not appear to play a significant role in the pathogenesis of in vitro toxicity.
Footnotes
Acknowledgements
The authors thank Howard Bausum for a critical review of the manuscript.
This work was presented in part at the 38th Annual Meeting of the Society of Toxicology, March 14–18, 1999, New Orleans, LA.
Disclaimer: This research was funded in part by the U.S Army. The views, opinions, and/or findings should not be construed as official Department of Army position, policy, or decision unless designated by other official documentation.