Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Amiodarone (AM), [2-butyl-3-(3',5'-diiodo-4'-diethylaminoethoxybenzoyl)-benzofuran], is a potent and efficacious class III (Vaughan Williams' classification) antidysrhythmic agent (Heger et al., 1981). It is used for treating ventricular and supraventricular arrhythmias and may have a role in preventing mortality postmyocardial infarction (Singh, 1996; Julian et al., 1997) and treating sudden cardiac arrest (Kudenchuk et al., 1999). However, despite the common use of AM (Connolly, 1999), there exists a potential for the development of life-threatening AM-induced pulmonary fibrosis (Mason, 1987). AM-induced pulmonary toxicity (AIPT) has been clinically diagnosed in 1 to 13% of individuals receiving high doses of AM and 1.6% of patients receiving doses of AM equal to and less than 400 mg/day (Sunderji et al., 2000). Because of its high potential for mortality, AIPT is the adverse effect of greatest concern for patients receiving AM therapy. Although the etiology of AIPT has not been fully elucidated, it is generally accepted that, during the development of chemically induced pulmonary fibrosis, a prolonged cytotoxic insult to the airway epithelium initiates the release of inflammatory mediators, influx of inflammatory cells, fibroblast proliferation, and collagen deposition (Sheppard and Harrison, 1992). Furthermore, in a rat model of AIPT, cytotoxicity occurs shortly after exposure to AM (Taylor et al., 2001), and in the hamster, early alveolar type II cell proliferation consistent with epithelial cell loss has been demonstrated (Cantor et al., 1987).

Previously, we demonstrated that hamster alveolar macrophages, alveolar type II cells and nonciliated bronchiolar epithelial (Clara) cells were all susceptible to cytotoxicity induced by various concentrations of AM (Bolt et al., 1998). Furthermore, these cell types require relatively high levels of energy to maintain normal function (Crystal, 1991; Plopper et al., 1991). However, the mechanism of AM-induced cytotoxicity that initiates pulmonary fibrosis has not been determined. In nonpulmonary tissues, AM causes both structural (Yasuda et al., 1996) and functional (Fromenty et al., 1990, 1993) perturbations to mitochondria. In addition, we have demonstrated that AM disrupts oxygen consumption in isolated hamster lung mitochondria (Card et al., 1998). However, the relationship between mitochondrial disruption and lung cell death has not yet been established.

The N-dealkylated AM metabolite N-desethylamiodarone (DEA) possesses antidysrhythmic properties (Abdollah et al., 1989). However, following AM treatment, DEA accumulates extensively in the lungs of animals and humans (Adams et al., 1985; Daniels et al., 1989), often to a greater extent than AM itself (Daniels et al., 1989). Furthermore, DEA is more cytotoxic than AM in rat alveolar macrophages (Ogle and Reasor, 1990) and in nonpulmonary cell types (Ruch et al., 1991), and is a more potent fibrogen in hamsters (Daniels et al., 1989). However, the relative susceptibilities of different lung cell types to DEA-induced cytotoxicity have not been reported. Identification of a particularly susceptible cell type may provide valuable information regarding the etiology of AM- and DEA-induced pulmonary toxicity.

The purpose of this study was to determine the effects of AM and DEA on mitochondrial membrane potential and intracellular ATP levels, and the temporal relationship between these effects and loss of viability in different isolated hamster lung cell types. Agents that potentially could attenuate AM- and DEA-induced mitochondrial disruption or cellular energy loss were also investigated. Furthermore, experiments in isolated whole lung mitochondria compared accumulation of AM and DEA as well as effects on mitochondrial function.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals. Chemicals were obtained from the following suppliers: 5,5',6,6'-tetrachloro-1,1',3,3'-tetraethylbenzimidazolylcarbocyanine iodide (JC-1) from Molecular Probes (Eugene, OR); trichlorotrifluoroethane (Freon 113) from Ladd Research Chemicals Inc. (Burlington, VT); amiodarone HCl, protease type I, carbonyl cyanide p-(trifluoromethoxy)phenylhydrazone (FCCP), [³H]2,8-adenosine, RPMI 1640 media with and without phenol red, heparin, tri-n-octylamine, tetrabutylammonium dihydrogen phosphate, rotenone (95-98%), ADP (free acid), L-glutamate (monosodium salt), L-()-malate (monosodium salt), EDTA (disodium salt, dihydrate), succinate (disodium salt, hexahydrate), D-mannitol, 3-[N-morpholino]propanesulfonic acid, and fatty acid-free bovine serum albumin from Sigma Chemical Co. (St. Louis, MO); Percoll from Amersham Pharmacia Biotech AB (Uppsala, Sweden); HEPES from Roche Molecular Biochemicals (Laval, PQ, Canada); and sodium pentobarbital from M.T.C. Pharmaceuticals (Mississauga, ON, Canada). N-Desethylamiodarone hydrochloride was generously donated by Wyeth-Ayerst Research (Princeton, NJ). All other reagents were of analytical grade and purchased from standard chemical suppliers.

Animals. Male golden Syrian hamsters (140-160 g) were obtained from Charles River Canada (St. Constant, QC, Canada). Animals were housed at room temperature on a 12-h light/dark schedule (lights on at 7:00 AM) for a minimum of 1 week prior to experimentation. Hamsters were fed Purina rodent chow and water ad libitum and were cared for in accordance with the principles and guidelines of the Canadian Council on Animal Care.

Enriched Lung Cell Preparations. The procedures for isolating alveolar macrophages and obtaining enriched preparations of alveolar type II cells, and nonciliated bronchiolar epithelial (Clara) cells from unseparated cells (cell digest) are described in detail elsewhere (Bolt et al., 1998). Briefly, hamsters were injected with heparin sodium (2000 U in 0.5 ml of saline) 45 min prior to anesthetization with sodium pentobarbital (300 mg/kg i.p.). Alveolar macrophages were obtained by bronchoalveolar lavage. Other lung cells were released by treatment with protease type I, and enriched preparations of alveolar type II and Clara cells were obtained by centrifugal elutriation. The alveolar type II cell preparation was further enriched by Percoll density gradient centrifugation [28% (v/v) in RPMI 1640 media]. Viabilities and cell numbers were assessed on a hemacytometer by light microscopy using 0.5% trypan blue dye exclusion. Alveolar type II cells and Clara cells were identified using modified Papanicolaou (Kikkawa and Yoneda, 1974) and nitro blue tetrazolium (Devereux and Fouts, 1980) staining, respectively. For each experiment, cells isolated from 12 hamsters were pooled.

Incubations and Viability Assessments. Cells (4 × 10⁵ cells/well) were incubated in flat-bottomed, irradiated, nonadhesive, polystyrene Nunc 96 microwell plates (Canadian Life Technologies, Burlington, ON, Canada) in RPMI 1640 media (with phenol red and buffered with 50 mM NaHCO₃) containing vehicle (0.1% ethanol), 100 µM AM, or 100 µM FCCP under 95% air, 5% CO₂ at 37°C, pH 7.4. This concentration of AM was selected because we previously found it to cause appreciable AM-induced cell death at 24 and 36 h (Bolt et al., 1998) and it is within the range of concentrations of AM present in human lungs following clinical treatment with AM (Brien et al., 1987). To minimize clumping, cells were agitated by drawing the cells gently in a 10-ml pipette every 6 to 12 h. In separate experiments, cells were incubated with vehicle or DEA (100, 50, 25, and 10 µM). Because of their relatively high yield and demonstrated susceptibility to AM cytotoxicity (Bolt et al., 1998), alveolar macrophages alone were used in experiments to assess effects of preincubation with 5.0 mM glucose or 2 mM niacin (potential sources for ATP) for 1 h prior to addition of AM, on loss of viability. Viability and cell numbers were measured by 0.5% trypan blue dye exclusion because it represents irreversible loss of plasma membrane integrity, a late event in processes leading to cell death (Tyson and Green, 1987). It also requires a relatively small number of cells and facilitates microscopic identification of cell targets for cytotoxicity in fractions containing more than one cell type.

Lung Cell Mitochondrial Membrane Potential. Mitochondrial membrane potential in intact lung cells was assessed using the JC-1 probe, a cationic chemical that exists as a green-fluorescing monomer at low membrane potentials (<120 mV) and as a red-fluorescing dimer (referred to as J-aggregates) at membrane potentials greater than 180 mV (Reers et al., 1995). Following excitation at 488 nm, the ratio of red (595-nm emission) to green (525-nm emission) fluorescence measures the ratio of high-to-low mitochondrial membrane potential (Reers et al., 1995). Lung cells (5.0 × 10⁵ cells/0.5 ml of RPMI 1640 media without phenol red) were incubated in irradiated, Nunclon Delta 24 multiwell plates (Canadian Life Technologies) with vehicle (0.1% ethanol), AM, DEA or FCCP, under 95% air, 5% CO₂ at 37°C for 2 h. JC-1 (5.0 µM) was added to each well, and cells were incubated for 30 min at room temperature in the dark, with gentle agitation. Cells were then washed twice with fresh media and resuspended in 0.5 ml RPMI 1640 media without phenol red. To measure basal fluorescence of each cell preparation, an aliquot was removed prior to addition of JC-1 and analyzed cytofluorometrically. Due to the relatively large number of red blood cells as well as debris present, effects of AM and DEA on mitochondrial membrane potential were not measured in cell digest.

Assessment of Cellular ATP Levels. To label adenine nucleotide pools, cells (1.2 × 10⁶) were loaded with 10.0 µCi of [³H]2,8-adenosine for 1 h at 37°C, under 95% air, 5% CO₂. Cells were centrifuged (500g for 5.0 min), washed twice with media (containing phenol red), and incubated with vehicle or 100 µM AM for 2, 4, or 6 h or 50 µM DEA for 2 h. In some experiments, alveolar macrophages were incubated with 5.0 mM glucose or 2.0 mM niacin for 1 h prior to exposure to 100 µM AM for 6 h to assess the ability of these agents to prevent AM-induced depletion of ATP. ATP measurement was conducted by a previously published method (Tekkanat and Fox, 1988) with modifications. Cells were washed twice with HEPES-buffered salt solution (Devereux and Fouts, 1981) and nucleotides extracted by adding 100 µl of 0.5 M ice-cold trichloroacetic acid, followed by neutralization with 100 µl of 0.25 M tri-n-octylamine in Freon 113. The organic layer was filtered through a 0.45-µm low protein binding Durapore (polyvinylidene difluoride) filter, snap frozen in liquid nitrogen, and stored at 80°C for no more than 3 days before analyzed.

Involvement of Mitochondrial Permeability Transition. To determine whether mitochondrial permeability transition occurred during exposure to AM, enriched fractions were loaded with 1.0 µM cyclosporin A 20 min prior to incubation with 100 µM AM (Bernardi, 1996). Cell viability was measured by 0.5% trypan blue exclusion.

Isolation of Whole Lung Mitochondria. Hamsters were deeply anesthetized with sodium pentobarbital (300 mg/kg i.p.), and lungs were perfused in situ with ice-cold 0.9% saline solution, removed, blotted dry, and weighed. Lung mitochondria (pooled from four to eight hamsters for each experiment) were isolated by differential centrifugation as described previously (Card et al., 1998). Aliquots of mitochondrial suspensions were removed for determination of protein content by the method of Lowry et al. (1951), with bovine serum albumin as the standard.

Polarographic Measurement of Oxygen Consumption. Oxygen consumption of isolated lung mitochondria was measured at 30°C as described previously (Card et al., 1998). Respiration supported by complex I of the respiratory chain was assessed using glutamate (5.0 mM) and malate (5.0 mM), and respiration at complex II was measured using succinate (10 mM, in the presence of 3.0 µM rotenone). Effects of AM or DEA on state 4 respiration at complexes I and II were examined by adding them at least 2 min following the total expenditure of 0.2 mM ADP (i.e., following transition from state 3 to state 4 respiration). Respiratory control ratios (RCRs) and ADP:O ratios were calculated as indicators of integrity of mitochondrial respiratory function (Estabrook, 1967).

Determination of Mitochondrial Drug Accumulation. Accumulation of AM and DEA in isolated lung mitochondria was determined following incubation with 400 µM AM or DEA at 30°C. Isolated mitochondria (1.0-2.0 mg of protein) were incubated in 1.5 ml of respiratory buffer 5 min prior to addition of AM or DEA. Following incubation with AM or DEA for 0.5 or 3 min, samples were centrifuged for 1 min (4000g) to pellet any precipitated drug. Aliquots (250 µl) of supernatant were used for analysis of drug levels, by HPLC (Brien et al., 1987; Bolt et al., 1998), and protein. Because of limited solubility of AM and DEA in the respiratory buffer, concurrent incubations with AM or DEA in the absence of mitochondrial protein controlled for drug precipitation into the buffer (background drug). To measure drug levels, 250-µl samples were centrifuged (5 min at 13,000g) and the resulting mitochondrial pellets were washed, mixed with 250 µl of respiration buffer, and centrifuged again. The supernatants were discarded, and the pellets stored at 20°C for up to 1 month. On the day of analysis, 500 µl of HPLC mobile phase [5% (v/v) acetic acid: acetonitrile, 20:80 (v/v)] was added to the individual pellets, which were mixed vigorously and centrifuged at 13,000g for 5 min. Supernatants were removed and filtered using 0.45-µm syringe filters (Millex-HV syringe-driven filter units; Millipore, Bedford, MA). Fifty microliters of these filtrates was analyzed by reverse phase HPLC with UV detection (Brien et al., 1987). Background drug levels in the buffer were subtracted from the drug concentrations measured in the presence of mitochondrial protein.

Data Analysis. Unless indicated otherwise, all data are expressed as sample group means ± S.D. Differences between treatment groups were determined by one- or two-way repeated measures analysis of variance (ANOVA) followed by Newman-Keuls post hoc test or by paired t tests, as indicated. Some percentage data underwent arcsine transformation prior to statistical analysis (Sokal and Rohlf, 1973). In all cases, p < 0.05 was considered statistically significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Lung Cell Isolations. Bronchoalveolar lavage fluid contained 98% alveolar macrophages, and cell digest contained approximately 21% alveolar type II cells and 4% Clara cells. Centrifugal elutriation of cell digest resulted in preparations containing 50 to 60% alveolar type II cells, and 35 to 50% Clara cells. Percoll density gradient centrifugation of the alveolar type II cell preparation enriched it to 75 to 85%. As reported previously, the alveolar type II cell preparation also contained lymphocytes (5-10%) and a few Clara cells (2-5%). Ciliated cells (10-15%), clumped red blood cells (10-15%), and a relatively small percentage (2-10%) each of fibroblasts and polymorphonuclear leukocytes were present in the Clara cell preparations.

AM Effects in Isolated Lung Cells. Initial viabilities in all fractions were greater than 90%, and control viabilities remained greater than 75% at 36 h. Total cell numbers in all fractions did not decrease over the course of the incubation period. Incubation with 100 µM AM for 2 h caused a 33 to 45% decrease in red/green fluorescence ratio with JC-1, indicative of a substantial loss of mitochondrial membrane potential in hamster macrophages, as well as in preparations enriched in type II cells and Clara cells (Fig. 1). This effect was similar to that seen with the classical mitochondrial uncoupler FCCP (Fig. 1). However, neither cellular ATP levels (Fig. 2) nor viability was significantly affected by AM at this time point. By 4 h of incubation, 100 µM AM caused an approximately 50% decrease in ATP content in cell digest and macrophages (Fig. 2), although viability was still at control values. By 6 h, ATP levels were significantly decreased in all cell fractions exposed to 100 µM AM, with the type II cell fraction showing a lesser degree of depletion than the other cell fractions (Fig. 2). Despite the pronounced loss of ATP relative to control, viability remained high in all the cell fractions at 6 h (viability in all fractions ranged from 94-98% in control fractions and 90-97% in AM-treated fractions). Consistent with our previous observation (Bolt et al., 1998), 100 µM AM caused substantial cytotoxicity in all the cell preparations at 24 and 36 h (Fig. 4). FCCP (100 µM) caused time-dependent loss of viability in all cell preparations, especially during the first 12 h, with cell digest, macrophages, and Clara cells showing particular sensitivity (Fig. 3).

View larger version (48K): [in this window] [in a new window]   Fig. 1.   Isolated lung cells incubated with vehicle (0.1% ethanol, ), 100 µM AM (), or 100 µM FCCP () for 2 h and loaded with JC-1 probe for 30 min. Mean (red/green) fluorescence (±S.D., n = 4) (measured cytofluorometrically and expressed as percentage of control) indicates ratio of high/low mitochondrial membrane potential. *, indicates difference (p < 0.05, repeated measures ANOVA with Newman-Keuls post hoc test) from control. Cells pooled from 12 hamsters were used for each n value.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   [3H]ATP levels in isolated hamster lung cells following incubation with 100 µM AM (mean percentage of control ± S.D.). *, significant difference from control (p < 0.05, paired t test, n = 4). At each time point, cells were pooled from 12 hamsters for each n value.

View larger version (25K): [in this window] [in a new window]   Fig. 3.   Viability (mean ± S.D., n = 4) of cell digest (CD), macrophages (MAC), alveolar type II cells (type II), and Clara cells (Clara) incubated with vehicle (0.1% ethanol) or 100 µM FCCP. All FCCP-treated fractions were significantly different from respective controls. *, difference between FCCP-treated type II preparations and FCCP-treated Clara cell preparations and FCCP-treated type II cell preparations and FCCP-treated macrophage preparations; +, significant difference between 36 h FCCP-treated Clara, type II, and macrophage fractions from respective 12-h FCCP-treated fractions (p < 0.05, repeated measures two-way ANOVA with Newman-Keuls post hoc test). For each n value, cells from 12 hamsters were pooled and analyzed at 12, 24, and 36 h. , CD-control; , CD-FCCP; , MAC-control; , MAC-FCCP; , Type II-control; , Type II-FCCP; , Clara-control; , Clara-FCCP.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Effect of 1.0 µM cyclosporin A (cyclo) on AM cytotoxicity in cell digest (CD), macrophages (MACs), alveolar type II cells (type II), and Clara cells (Clara) (data points represent mean of two experiments) Cells from 12 hamsters were pooled for each n value.

DEA Effects in Isolated Lung Cells. Incubation of hamster lung cells with 100 µM DEA for 2 h resulted in 100% cell death in all cell preparations (data not shown). Fifty micromolar DEA caused a rapid decrease in cell viability relative to 100 µM AM (Figs. 4 and 5) and the viability loss was first observed in the Clara cell-enriched preparation, in which a significant loss (approximately 20%) was apparent by 2 h (Fig. 5). Because cell viability, measured by trypan blue exclusion, was determined by light microscopy, morphological characteristics (size, lack of cilia, lack of surfactant-containing inclusions) confirmed that Clara cells, rather than contaminating cell types, were the principal cells demonstrating DEA cytotoxicity in this cell fraction. Following incubation with 50 µM DEA for 6 h, substantial and similar degrees of viability loss had occurred in all the cell fractions, and the decreases in viability progressed to 24 h (Fig. 5). With 25 µM DEA, time-dependent decreases in viability were evident in the cell digest beginning at 6 h and in alveolar type II cells and Clara cells beginning at 12 h. Loss of viability in alveolar macrophages did not occur during exposure to 25 µM DEA. At 10 µM, DEA did not cause significant cytotoxicity in any enriched preparation at any time point (Fig. 5).

View larger version (20K): [in this window] [in a new window]   Fig. 5.   Viability (mean ± S.D., n = 4) of cell digest (A), macrophages (B), type II cells (C), and Clara cells (D) incubated with vehicle (0.1% ethanol, ) or 10 (), 25 (), or 50 µM () DEA. *, significant difference from control; +, significant difference between 36 h DEA-treated fraction from respective 6-h DEA-treated fraction (p < 0.05, repeated measures two-way ANOVA with Newman-Keuls post hoc test). For each n value, cells from 12 hamsters were pooled and analyzed at 2, 6, 12, and 24 h. Separate cell preparations were used for the time course of each concentration of DEA.

View larger version (49K): [in this window] [in a new window]   Fig. 6.   Cytofluorometrically measured red/green fluorescence (mean percentage of control ± S.D., n = 4) representative of mitochondrial membrane potential in isolated cells incubated with vehicle (0.1% ethanol), 50 µM DEA, or 100 µM FCCP for 2 h and loaded with JC-1 for 30 min. *, indicates difference (p < 0.05, repeated measures ANOVA with Newman-Keuls post hoc test) from control. Cells from 12 hamsters were pooled for each n value.

Effects of AM and DEA on Isolated Whole Lung Mitochondria. In mitochondria isolated from whole hamster lungs, calculated RCRs were 2.90 ± 0.43 and 1.52 ± 0.10 for complex I and II, respectively. The ADP:O ratio values were 4.27 ± 0.42 and 2.20 ± 0.43 for complex I and II, respectively. As indicated by the higher RCR values (i.e., >2.50), tight coupling was observed at complex I but not at complex II, similar to our previous report (Card et al., 1998).

View larger version (20K): [in this window] [in a new window]   Fig. 7.   Effect of AM and DEA on complex 1-supported (A) and complex II-supported (B) state 4 (resting) oxygen consumption in isolated hamster lung mitochondria. *, significant difference (p < 0.05; mean ± S.D., n = 4, paired t tests) from equimolar concentrations of AM. Mitochondria were pooled from eight hamsters for each n value.

View larger version (11K): [in this window] [in a new window]   Fig. 8.   Representative lung mitochondrial oxygen consumption tracing depicting the effect of AM or DEA on state 4 respiration supported by complex I. Note the immediate inhibition of respiration as a result of DEA addition, compared with the initial stimulation followed by secondary inhibition as a result of AM. Similar effects were observed for respiration supported by complex II in the presence of AM or DEA.

View larger version (18K): [in this window] [in a new window]   Fig. 9.   Mitochondrial AM and DEA levels (mean ± S.D.) in isolated whole lung mitochondria following incubation with 400 µM AM or DEA. *, significant difference from AM at the same time point (p < 0.05, n = 3, repeated measured two-way ANOVA). Mitochondria were pooled from eight hamsters for each n value.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Although the etiology of AIPT still has not been fully elucidated, the results of the present investigation suggest a sequence of events involving disruption of mitochondrial function and loss of cellular ATP as pivotal events in initiation of AM cytotoxicity in the lung. Incubation with AM for 2 h resulted in decreased intracellular mitochondrial membrane potential in all cell preparations. However, no effects on ATP levels or viability were observed at this time. At 4 h in macrophages and 6 h in type II cell and Clara cell preparations, ATP levels decreased. Overt AM-induced cell death in alveolar macrophages, type II cells, and Clara cells did not occur until 12 h and was more pronounced at 24 and 36 h (Bolt et al., 1998) (Fig. 4). Therefore, during AM-induced cytotoxicity, mitochondrial function is disrupted prior to ATP depletion and ultimately cell death in freshly isolated target lung cell types. The early occurrence of AM-induced perturbation of mitochondrial function is suggestive of an initiating event in cytotoxicity.

AM-induced perturbations of mitochondrial structure (Yasuda et al., 1996) and function (Fromenty et al., 1990; Card et al., 1998) have been reported previously. However, those experiments were performed in isolated mitochondria or nonpulmonary cells. Our experiments confirmed disruptive effects of AM on mitochondrial function and examined, for the first time, these effects in freshly isolated, intact, target lung cell types. This was important because it allowed for the assessment of AM-induced effects in the presence of factors such as biotransformation enzymes, reducing agents, antioxidants, etc., which are greatly altered in isolated mitochondria and cultured cells. Although there are advantages and disadvantages to isolated organelles and cultured cells, freshly isolated cells more accurately reflect cellular properties as they occur in situ (Massey, 1989).

JC-1 has been used to measure mitochondrial membrane potential in other isolated cell types (Di Lisa et al., 1995). However, to our knowledge, this is the first report to use JC-1 in freshly isolated lung cells. JC-1 has the advantage of being more specific for mitochondrial (as opposed to plasma) membrane potential, and more consistent in its response to depolarization, than other dyes such as rhodamine 123 and 3,3'-dihexyloxacarbocyanin iodide (Salvioli et al., 1997). Furthermore, flow cytometry allows measurement of fluorescence of each cell, a property particularly useful in the Clara cell preparations, where flow cytometer gates can be adjusted to measure fluorescence from only the larger cells (thereby excluding contributions of contaminating blood cells).

Previous studies have indicated the ability of AM to uncouple mitochondrial electron transport (Fromenty et al., 1990). Our observation that the time course of AM-induced disruption of lung mitochondrial membrane potential relative to cytotoxicity resembled that of FCCP, an established uncoupler of electron transport, suggests that the mechanism of mitochondrial disruption resulting in cytotoxicity for AM and FCCP may be the same.

Mitochondrial permeability transition (MPT), a phenomenon recognized to be important in both necrosis and apoptosis for some agents, is a sudden increase in permeability of the inner mitochondrial membrane, resulting in membrane depolarization, uncoupling of oxidative phosphorylation, mitochondrial swelling, and release of intramitochondrial ions (Lemasters et al., 1998). MPT is regulated by a voltage-dependent channel that is inhibited by nanomolar concentrations of cyclosporin A, and cyclosporin A protects against cell death involving MPT-activating substances or events (Seaton et al., 1998). The failure of cyclosporin A to reduce AM-induced cytotoxicity suggests that MPT is not involved in AM-induced disruption of mitochondrial function. Also, exposure of isolated hamster lung cells to cyclosporin A concentrations above 1.0 µM caused loss of viability (data not shown).

Despite occurrence of mitochondrial disruption in AM-treated cells at 2 h, [³H]ATP levels did not differ between control and AM-treated cells until approximately 4 to 6 h. Interestingly, we report, for the first time, that alveolar macrophages, the cell type most susceptible to 100 µM AM-induced cytotoxicity, demonstrated AM-induced ATP depletion earlier than other cell types. The reason for this phenomenon is not known but could be related to particularly high-energy requirements of macrophages. Although it is not possible to completely rule out potential effects of AM on ATP utilization, observed decreases in ATP levels are consistent with disrupted mitochondrial function and hence, ATP synthesis.

Since AM decreased ATP levels prior to cell death, we hypothesized that AM-induced cytotoxicity would be reduced by preventing ATP depletion. Theoretically, ATP depletion might be diminished by adding glucose or niacin to the incubation medium. Glucose, via either glycolysis or the pentose phosphate pathway, can generate ATP. Although glucose previously has been shown to decrease AM-induced lymphocyte cytotoxicity (Fromenty et al., 1993), its lack of effectiveness against lung cell cytotoxicity correlated with its inability to prevent AM-induced depletion of ATP.

Niacin has been found to prevent chemically induced cytotoxicity by maintaining cellular NAD levels (Weitberg and Corvese, 1990). However, similar to the situation for glucose, we found niacin to be ineffective at preventing AM-induced ATP depletion and cytotoxicity in hamster macrophages. Although the metabolic pathway for the synthesis of NAD may differ between cell types, this lack of effectiveness is consistent with that seen in cultured rat pulmonary macrophages (Nadeau and Lane, 1988). Thus, the partial attenuation of AM-induced pulmonary fibrosis previously reported for niacin (Wang et al., 1992) is likely attributable to some mechanism other than maintenance of ATP levels, such as decreased procollagen gene expression (Gurujeyalakshmi et al., 1996).

The cell isolation procedures resulted in considerable enrichment of both type II cells and Clara cells. Since Clara cells represented a very small percentage of cells in the cell digest (4%), enrichment was limited to 35 to 50% in the Clara cell fraction. This purity is within the range of Clara cell enrichments previously obtained by us (Bolt et al., 1998) and by other authors studying different species (Plopper et al., 1991). However, when making comparisons between the cell preparations, particularly involving the Clara cell fraction, one must carefully consider the presence of other cell types in the preparations before drawing conclusions based on methods (such as that used for measuring ATP) that do not differentiate between the cell types present.

Although AM, DEA, and FCCP were cytotoxic in all of the cell preparations examined, intercellular differences were observed. Notably, 50 µM DEA caused significant loss of ATP and viability in Clara cell-enriched preparations before effects were observed in macrophages and type II cells (Fig. 8; Table 1). Furthermore, Clara cells and type II cells were more susceptible to lower concentrations of DEA (25 µM) than macrophages. Similarly, in our earlier study, we found Clara cells to be more susceptible to a lower concentration of AM (50 µM) than were other hamster lung cell types (Bolt et al., 1998). Since both Clara cells and type II cells contain appreciable levels of cytochromes P450 (Massey, 1989; Plopper et al., 1991), one might suggest a role for P450-catalyzed biotransformation of AM and DEA in cytotoxicity. However, if P450-catalyzed bioactivation were central to the cytotoxicity of these agents then macrophages, which have very low expression of P450 unless treated with a P450 inducer (Crystal, 1991), should not be sensitive to AM or DEA toxicity. While this prediction holds true for macrophages exposed to 25 µM DEA (Fig. 5) and 50 µM AM (Bolt et al., 1998), loss of macrophage viability due to 50 µM DEA (Fig. 5) and 100 µM AM (Fig. 4; Bolt et al., 1998) was substantial. Furthermore, the macrophage and Clara cell-enriched preparations showed greater susceptibility to FCCP cytotoxicity than did the type II cell-enriched preparation (Fig. 3). Since FCCP is a direct-acting mitochondrial uncoupler, not requiring P450-catalyzed bioactivation, then the sensitivity of macrophages and Clara cells to FCCP may reflect a relative inability to overcome or survive disruption of cellular energy metabolism. This could also explain the early loss of Clara cell viability seen with DEA and a low concentration of AM (Bolt et al., 1998). It is also supported by the observation that cell preparations demonstrating the greatest depletion of ATP tended to show the greatest loss of viability later during incubations (Fig. 2; Table 1).

In all hamster cell types examined, DEA was much more cytotoxic than AM, causing cell death at a lower concentration. This suggests a potentially important contribution of DEA in clinical AIPT, particularly considering the fact that substantial accumulation of DEA occurs in lung tissues of patients undergoing chronic AM therapy (Brien et al., 1987). The explanation for the relatively high interexperiment variability in susceptibility of macrophages to 50 µM DEA-induced cytotoxicity (Fig. 5B) is not readily apparent.

The link between disruption of mitochondrial function and loss of cell viability was also supported by the inhibition by AM and DEA of state 4 respiration in whole lung mitochondria. The biphasic effects of AM may be associated with rapid accumulation of protonated AM within the mitochondrial matrix, resulting in release of protons, and hence initial stimulation of respiration due to an uncoupling effect; the subsequent inhibition is presumed to be due to further accumulation of AM (Fromenty et al., 1990; Card et al., 1998). The monophasic inhibition caused by DEA has been observed in other tissues (Fromenty et al., 1990). Of particular note were the more rapid inhibition of respiration by DEA, which was associated with greater accumulation into the mitochondria relative to AM, and greater inhibition of complex II by DEA. Although the relative accumulation of AM and DEA into mitochondria may differ in intact cells, the greater effects of DEA in isolated mitochondria coincided with the greater cytotoxicity of the N-dealkylated AM metabolite compared with AM itself. Also, the rapidity of onset of effects in isolated mitochondria relative to intact cells is consistent with a lack of barriers for diffusion in the former.

In conclusion, AM and DEA disrupt mitochondrial function prior to cellular ATP depletion and subsequent cell death in freshly isolated hamster lung cells, suggesting that perturbation of lung cell mitochondrial function initiates AM-induced cytotoxicity. Since cell death occurs early relative to the onset of fibrosis in animal models of AIPT (Taylor et al., 2001) then preventing collapse of mitochondrial membrane potential and/or subsequent ATP depletion might potentially prevent AM-induced cytotoxicity and ultimately pulmonary fibrosis. Furthermore, DEA is more cytotoxic and causes cytotoxicity faster than AM, suggesting an important contribution of DEA to the development of AIPT.