- Email: [email protected]
An improved in vitro model for testing the pulmonary toxicity of complex mixtures such as cigarette smoke
Exp Toxic Pathol 2003; 55: 51–57 URBAN & FISCHER http://www.urbanfischer.de/journals/exptoxpath
Fraunhofer Institute of Toxicology and Experimental Medicine, Hannover, Germany
An improved in vitro model for testing the pulmonary toxicity of complex mixtures such as cigarette smoke* MICHAELA AUFDERHEIDE, JAN W. KNEBEL, and DETLEF RITTER With 8 figures Received: December 2, 2002; Revised: January 21, 2003; Accepted: February 25, 2003 Address for correspondence: Prof. Dr. MICHAELA AUFDERHEIDE, Fraunhofer Institute of Toxicology and Experimental Medicine, Department of In Vitro Toxicology, Nikolai-Fuchs-Strasse 1, 30625 Hannover, Germany; Tel.: +49 511 5350 252, Fax: +49 511 5350 444, e-mail: [email protected] Key words: In vitro exposure; air/liquid interface; cigarette mainstream smoke; human lung cells.
Summary
Introduction
Numerous approaches have been employed for testing the biological activity of cigarette smoke in vitro. None of them has managed to expose cultured lung cells in a realistic manner to the complex gaseous and particulate mixture that constitutes cigarette smoke. We have devised a system that makes this possible. The system presented here enables the direct exposure of human lung cells to native, unmodified cigarette mainstream smoke. It consists of a smoking machine, a dilution device for the smoke, analytical devices for online monitoring and a specially adapted exposure module based on the Cultex** cell cultivation system that is equipped with a gas-exposure top. Due to the special design of the exposure device and the optimised exposure conditions, this equipment allows cultured human lung cells to be exposed to freshly generated cigarette mainstream smoke. Exploratory experiments revealed that the smoke could be diluted over a wide concentration range in a reproducible way with respect to gas and particulate phases, and also demonstrated reproducible particle deposition depending on smoke concentration. Furthermore, it was shown that the exposed cells maintained their viability. Native cigarette mainstream smoke induced dose-dependent cellular effects in exposed cells with respect to cellular viability (viable cell number monitored by tetrazolium salt cleavage) and intracellular parameters (ATP and glutathione content). Therefore, fresh, physically and chemically unmodified cigarette mainstream smoke can be tested using this novel system.
Testing the biological effects of smoke aerosols in vitro has been carried out using a variety of strategies. Nevertheless, all approaches involve three common fundamental objectives: (1) the use of a specified biological indicator system, (2) the exposure of the biological test system in a way that ensures contact with the cigarette smoke without altering its properties, and (3) the analysis of biological/biochemical alterations induced by the smoke in the indicator system. The choice of the biological indicator system is determined by many factors, including the possibility of comparison with historical data determined using the same system, reproducibility, availability, and applicability to the in vitro situation of the biological endpoints determined after exposure in vitro. Because of its obvious relevance, a human lung cell system is favoured. The choice of the endpoints for the analysis depends to a large extent on the aim of a particular study and is also subject to the considerations relevant for the indicator system. In contrast, the exposure conditions dictate a number of clear conditions because of the physical and chemical properties of the smoke and the fundamental requirements for maintaining the viability of the cells during exposure. Typically, cells are exposed to smoke extracts. Cigarettes are smoked in a smoking machine according to interna-
* Reported in part at the CORESTA Congress, September 22–27, 2002. ** (Pat. No. DE 19801763/PCT/EP99/00295)
Abbreviations: DMSO: dimethyl sulphoxide; PBS: phosphate-buffered saline; TPM: total particulate matter. 0940-2993/03/55/01-051 $ 15.00/0
51
tional smoking standards and total particulate matter (TPM) is collected on a Cambridge filter or by other trapping methods. TPM is extracted with an organic solvent, usually dimethyl sulphoxide (DMSO), and the organic extract is then added to the cell culture medium for the submerged exposure of cultivated cells (LAFI et al. 1991; MCKARNS et al. 2000; WILLEY et al. 1987). Since this strategy takes into account only the effects of the particulate phase of the smoke, efforts have been made to test the gaseous phase of the smoke in a similar way. While TPM is sampled on a filter, the remaining gas phase is bubbled through an aqueous solution, usually phosphate-buffered saline (PBS). This extract is added to the culture medium of a cell culture for testing purposes (ROEMER et a. 2002). There have also been attempts to test gas and particulate phases together. The whole smoke aerosol can be passed through culture flasks where cells are submerged in culture medium. The flask is steadily moved during exposure in such a way that the culture medium runs to one side of the flask and sets the cells free for exposure (BOMBICK et al. 1997). Such scenarios result in intermittent exposures that change periodically from a situation where cells are covered by a thick layer of medium to a situation allowing a more direct exposure with a thinner overlay of culture medium on the cells. Hence, common exposure techniques used to date all involve more or less drastic modifications of the smoke aerosol. When using extracts of smoke, the individual constituents of the crude smoke condensate mixture will be effective to a different extent according to their physico/chemical properties, mainly their solubility. This is the case, for example, for compounds of the gaseous phase when extracted using an aqueous salt solution. Organic gaseous components not soluble in the aqueous phase are extracted to a lesser extent and, thus, their effect may be underestimated in the subsequent toxicological tests. Additionally, components are aged by the procedures and the particulate character is lost completely. Furthermore, any liquid between the biological indicator system and the whole smoke aerosol or its extracted components acts as a barrier between the cells and the test substance. Firstly, this is due to physical reasons. Gas and particulate constituents of the smoke aerosol have to diffuse through the liquid or have to be dissolved to come into contact with the cells. Consequently, the effective exposure concentration of each constituent is determined by solubility and diffusion properties. Secondly, this is due to possible chemical reactions. Culture media are composed of many highly reactive substances like glutathione, amino acids and others, and will readily change the smoke components by way of chemical reactions. As a result of these factors, the qualitative and quantitative composition and dosage of cigarette mainstream smoke in the toxicological test are defined only partly by the composition and doses of the smoke aerosol generated. They are determined to a great extent by the physical and chemical changes induced in the smoke by processes that take place during extraction or exposure. 52
Exp Toxic Pathol 55 (2003) 1
The aim of this study was to develop a system for direct exposure of human lung cells to native cigarette mainstream smoke with as little change as possible in the smoke properties. To achieve this, an exposure strategy was used that rendered possible the exposure of human lung cells to the smoke aerosol directly at the air-liquid interface. Adherent cells were grown on microporous membranes and cultured biphasically, i.e. nourished by culture medium through the membrane and exposed directly, without medium coverage, to the atmosphere from above (fig. 3). Using this mode of exposure, cells have successfully been exposed to native gaseous compounds (RITTER et al. 2001), native cigarette sidestream smoke (AUFDERHEIDE et al. 2001) and native diesel exhaust fumes (KNEBEL et al. 2002). An adaptation of this technique was developed to allow exposure to diluted cigarette mainstream smoke using an automatic smoking machine. This paper presents a description and physical characterisation of the system as well as validation by means of measuring selected indices of activity in cells exposed to various concentrations of native cigarette mainstream smoke.
Material and methods Smoke generation, dilution and cell exposure system: An experimental system was developed as shown in figure 1. It consisted of: (1) an automatic smoking machine (Borgwaldt RM 1/G), working with smoking parameters according to FTC/ISO standards (35 ml puff volume/2 s, one puff per minute), (2) a dilution system for the fresh cigarette mainstream smoke, driven by synthetic air, (3) analytical instruments for online monitoring of aerosol concentrations (particles and gas), and (4) the exposure device for cultured cells growing adherently on microporous membranes (AUFDERHEIDE and MOHR 1999; RITTER et al. 2001). Figure 2 illustrates the status of the cells during exposure. For physico-chemical characterisation of the exposure, 1R4F cigarettes (University of Kentucky, conditioned according to ISO) were smoked with a constant number of 9 puffs. The mainstream smoke generated was fed into a constant flow of synthetic air (20.5% O2 in N2, flow control by mass-flow controller, Analyt, Müllheim, Germany) according to figure 1. In this way the discontinuously generated smoke was incorporated into a continuous flow. The smooth constant flow was passed through a light-scattering photometer and through the exposure device. The periodical changes in relative smoke and synthetic air concentrations were recorded (fig. 3). Cells were located in the exposure device on microporous membranes (0.4 µm pore size, 1 cm2, Falcon). The exposure device was heated by a water-bath. The cell exposure did not require the use of a cell culture incubator. For the investigation of the biological effects of different cigarette types, two research cigarettes were conditioned and smoked in a blind study according to ISO: Cigarette A had 9.0 mg of tar/cigarette, 14.3 mg of carbon monoxide (CO)/cigarette, 0.6 mg of nicotine/cigarette: Cigarette B had 1.4 mg of tar/cigarette, 0.6 mg of CO/ cigarette, 0.09 mg of nicotine/cigarette. British American Tobacco (BAT), Germany, kindly provided the experimental cigarettes.
Fig. 1. Experimental system for the generation, dilution and online monitoring of cigarette mainstream smoke and exposure of human lung cells adhering to microporous membranes. For details see text.
Physico-chemical measurements: Particle analysis was carried out using light-scattering photometry. The light-scattering photometers (from the Fraunhofer Department of Aerosol Research) were positioned in front of and behind the exposure device. They were calibrated for linearity, and performance in the desired aerosol concentration range was checked by filter analysis. Gas monitoring was carried out online by non-dispersive infrared spectroscopy (NDIR) with CO and CO2 monitors (Maihak Unor 6N, Hamburg, Germany) calibrated for the desired concentration ranges. Quantification of particle deposition was carried out by organic extraction of exposed culture membranes and fluorescence measurement of the extract in a microtiter plate fluorescence reader (Spectramax Gemini XS, Molecular Devices, Ismaning, Germany) Acquisition and processing of data was carried out online using data acquisition equipment (National Instruments, Munich, Germany) and software (DASYLab32, Datalog, Mönchengladbach, Germany) on a standard personal computer.
Fig. 2. Cell exposure in detail. Cells were grown on a microporous membrane (2) with culture medium beneath the membrane (3) and aerosol flow (1) above them without an overlying liquid layer. Right: View of one of three inserts being exposed in parallel in one exposure device.
Cell culture: A549 cells and human foetal bronchial epithelial cells (HFBE cells) (EMURA et al. 1990) were routinely thawed from a stock pool for each experiment and cultured in 75 cm2 flasks. Subconfluent cultures were trypsinised and cell viability was determined using an electronic cell counter (CASY, Schärfe Systems, Reutlingen, Germany). Cells were seeded onto polyethylene terephthalate (PET) track-etched membranes with a pore size of 0.4 µm, 1.6 × 106 pores/cm2 and a growth area of approximately 1 cm2 (Becton Dickinson, Germany), and cultured in an incubator at 95% RH (37 °C) and 5% CO2. Culture membranes were transferred into the exposure device before exposure. Culture medium below the membrane during the exposure contained no foetal calf serum (FCS). Analysis of cells: Tetrazolium salt cleavage was measured using the WST-1 dye (Boehringer, Mannheim, Ger-
Fig. 3. Time-dependent particle concentrations at the inlet (particles in) and outlet (particles out) of the cell exposure device. The time count started with the beginning of the puff. Exp Toxic Pathol 55 (2003) 1
53
Fig. 4. Results of particle analysis during smoke exposure using different dilution flow rates. Each point represents the smoking of one 1R4F cigarette with 9 puffs. Means represent smoke concentrations during the smoking time of 9 minutes.
many). Cells were incubated with a dilution of the dye according to the protocol developed by the manufacturer. After incubation, formazan production was quantified by absorption reading using a microtiter plate reader (Spectramax 340PC, Molecular Devices, Ismaning, Germany) at wavelengths of 450/630 nm. The number of viable cells was determined directly after tetrazolium salt analysis of the same cells using an electronic cell counter (CASY, Schärfe Systems, Reutlingen, Germany) after trypsinisation and dilution of the cell suspension (KNEBEL et al. 1998). This method is appropriate for an automated, quantitative discrimination between living and dead cells (WINKELMEIER et al. 1993). The protein content of the cells was quantified using a commercial kit according to the protocol of the manufacturer (DC Protein Assay, Biorad, Munich, Germany). The ATP/ADP ratio of exposed cells was analysed using reversed phase HPLC with UV detection at 259 nm following extraction with trichloroacetic acid (RITTER et al. 1999). Intracellular glutathione in cells was determined using a modified Tietze recycling assay after extraction with metaphosphoric acid (RITTER et al. 1999). The intracellular redox ratio of oxidised and reduced glutathione GSSG/GSH was quantified using extraction with meta-phosphoric acid, derivatisation using 2,4-dinitrofluorobenzene and detection of the N-2,4-dinitrophenyl derivatives by UV at 355 nm after HPLC (RITTER et al. 2001).
Results
Fig. 5. Results of particle deposition analysis on culture membranes using different dilution flow rates and smoking one or two cigarettes with 9 puffs.
In exploratory experiments without cells the experimental system was examined with regard to the physicochemical characteristics of the smoke aerosol generated. The age of the smoke when it came into contact with the exposed cells could be estimated by the online analysis of the particle concentration at the inlet of the exposure device. Measuring the time between the beginning of the smoke puff and the arrival of the corresponding particle peak in the photometer gave the age of the smoke. This time was approximately 6 seconds; it did not depend on
Fig. 6. Exposure of A549 cells to synthetic air for 60 minutes under conditions of cigarette mainstream smoke exposure. Cells were post-incubated for 2 hours after treatment and then analysed for the parameters indicated. Control cells were left unexposed in the incubator for the same time and analysed in the same way. 54
Exp Toxic Pathol 55 (2003) 1
the dilution used. Figure 3 shows the time-dependent particle concentrations measured at the inlet and outlet of the exposure device. The diagram illustrates the detection of the particles of a puff about 6 seconds after the start of the puff. Furthermore, it shows that the particles were detected at the outlet of the exposure device before the particles of the following puff entered. There was obviously no mixing of sequential puffs . The dilution of particles and gas phase of mainstream smoke was validated using light-scattering photometry and NDIR detection of carbon monoxide and carbon dioxide in the aerosol. Mean particle concentrations during the smoking of one cigarette with 9 puffs using dilution flows in the range of 1 to 30 l/min are illustrated in figure 4. The figure shows that the particulate phase was effectively diluted in a reproducible manner. The same characteristics were found for CO and CO2 concentrations (data not shown), resulting in a stable CO/CO2 ratio and a stable CO/particle ratio independent of the dilution. The particle deposition on the cell culture membrane was analysed by exposing membranes without cells under conditions of cell exposure and quantification of the deposited particle mass by a fluorescence method. Figure 5 shows the results for the deposition of smoke aerosol particles at different dilution flow rates when one or two cigarettes were smoked (9 or 18 puffs). The reproducible particle deposition, depending on smoke aerosol concentration (dilution) and dosage (number of cigarettes), was documented. In preliminary experiments human lung cells were exposed using the system. A549 cells and human foetal bronchial epithelial cells (HFBE cells) were exposed to synthetic air to investigate the effects of exposure on the viability of the cells. Furthermore, HFBE cells were exposed to the native, diluted cigarette mainstream smoke
of two different research cigarettes to investigate the smoke effects on the cellular viability in relationship to the dose. Additional exposures to low smoke concentrations using short exposure times of 6 minutes were carried out to examine the sensitivity of the system. Figure 6 shows the results of A549 cell exposures to synthetic air for 60 minutes under the conditions used subsequently for the exposure to cigarette mainstream smoke. Cells were post-incubated for 2 hours after exposure and analysed for the parameters indicated above. Protein content, tetrazolium salt cleavage, ATP content, ATP/ADP ratio and intracellular content of glutathione were decreased very slightly by exposure to synthetic air, whereas the glutathione redox ratio was increased very slightly in comparison to controls. None of these changes was statistically significant and no loss of cells could be detected after determining viable cell number. These results show that the exposure to synthetic air under the conditions of cigarette mainstream smoke exposure did not reduce the viability of the cells. The results of cell exposures to cigarette mainstream smoke using HFBE cells and varying numbers of cigarette puffs generated from research cigarettes A and B are presented in figure 7. Cells were exposed for a total period of 28 minutes. Within this time, different numbers of cigarettes (up to 4, max. 7 puffs per cigarette) were smoked with 1 puff per minute (lowest dose of cigarette A was 4 puffs) and constant smoke dilution. After smoke exposure, cells were exposed to synthetic air for the rest of the 28-minute exposure period. Controls were exposed to synthetic air only. Figure 7 shows the results of the tetrazolium salt assay, which represents a sensitive marker for cellular viability and activity of metabolic processes. Using cigarette A, viability was clearly decreased compared to controls at doses of 18 or more
Fig. 7. Tetrazolium salt cleavage per cell in HFBE cells exposed to the diluted native cigarette mainstream smoke of cigarettes A and B using increasing numbers of puffs during a total exposure time of 28 minutes at a constant dilution flow rate of 1 l/min. Cells were exposed to the indicated number of cigarette puffs and to synthetic air for the rest of the exposure time, and post-incubated for 2 hours. Controls were exposed to synthetic air only (lowest dosage of cigarette A was 4 puffs). T-test cigarette A versus cigarette B, *** P < 0.001, t-test cigarette B versus control, ++ P < 0.01, +++ P < 0.001. Exp Toxic Pathol 55 (2003) 1
55
puffs. Mainstream smoke of cigarette B did not induce any decrease in viability through the whole concentration range, but increased the cellular metabolic activity using 14, 21 or 28 cigarette puffs compared to controls. This increase was statistically significant in comparison to controls and to results after exposure to the smoke of cigarette A at the same doses. HFBE cells were also exposed to 6 puffs of cigarettes A or B or synthetic air during a shortened exposure period of 6 minutes. After exposure, cells were post-incubated for varying times up to 24 hours or were not post-incubated at all (time 0 hours). Tetrazolium salt cleavage, viable cell number, ATP/ADP ratio and intracellular glutathione content were analysed. Figure 8 shows the results of the determinations of the ATP/ADP ratio and the intracellular glutathione content at times of 0 hours and 24 hours post-incubation compared to controls (synthetic air exposure). ATP/ADP ratios were decreased in HFBE cells after exposure to smoke from cigarette A to about 50% at time 0 and to about 75% at 24 hours. In contrast, smoke from cigarette B did not induce a significant decrease. The difference in the ATP/ADP ratios at 24 hours post-incubation time in cigarettes A and B was statistically significant. Glutathione content of cells exposed to mainstream smoke from cigarette A was decreased to 10% of control at 0 hours and increased to about 175% of control at 24 hours post-incubation time. After exposure to smoke from cigarette B the glutathione content of cells was decreased only very slightly at 0 hours and was comparable to controls at 24 hours post-incubation. The difference in the glutathione content at 24 hours post-incubation in cigarette A and B was statistically significant.
Discussion An experimental system was devised to expose human lung cells to freshly generated, diluted cigarette main-
stream smoke aerosol. The aim of the study was to induce only minimal changes in the smoke, which was generated according to international smoking standards, and to achieve effective contact between exposed cells and the exposure atmosphere both for the gaseous and particulate constituents of the smoke. The system works by exposing cells directly at the airliquid interface on microporous membranes. Experiments using such a system for investigations of native gaseous compounds and other airborne materials (RITTER et al. 2001; AUFDERHEIDE et al. 2001; KNEBEL et al. 2002) had previously shown that the exposure of human lung cells could be accomplished without any physical modification of the test atmosphere conditions (e.g. by humidification, heating or addition of CO2) that are routine procedures in cell culture. Such an in vitro system is essential for testing cigarette mainstream smoke under conditions as close as feasible to those in vivo. Physico-chemical characterisation of the smoke aerosol generated in the system demonstrated that the main particle peak entered the exposure device about 6 seconds after the start of the puff. Both the particulate and gaseous phases of the aerosol were diluted effectively and in a reproducible manner. The stable ratios of CO to particle concentration and CO to CO2 concentration irrespective of the dilution demonstrated that this method of smoke aerosol dilution induced no changes in the particulate phase or in the gaseous phase. The effective contact of the gas phase and the cells using this cell culture system had been shown in a study on gaseous substances (RITTER et al. 2001). Reproducible and dose-dependent particle deposition on the culture membrane was demonstrated in this study. Moreover, the physical mechanisms of sedimentation and diffusion used here for particle deposition follow the same principles as in human distal lung airspaces in vivo. Experiments using two kinds of human lung cells, A549 cells and human foetal bronchial cells, showed that
Fig. 8. Exposure of HFBE cells to the diluted native cigarette mainstream smoke of cigarettes A and B. Cells were exposed to 6 puffs, corresponding to an exposure time of 6 minutes, and analysed directly after exposure or after a post-incubation time of 24 hours for intracellular ATP/ADP ratio and glutathione content. Control cells were exposed to synthetic air only. T-test cigarette A versus cigarette B, * P < 0.05. 56
Exp Toxic Pathol 55 (2003) 1
the cells could be exposed in the system without loss of viability. Moreover, exposure to two different types of research cigarettes in preliminary experiments resulted in different dose/effect relationships. Smoke from cigarette A, which has more tar, CO and nicotine than cigarette B, clearly decreased the viability of exposed cells, whereas smoke from cigarette B induced metabolic activation under comparable conditions. Smoke from cigarette A induced a typical kinetic profile for the glutathione content of exposed cells, i.e. the depletion directly after exposure and an “overshoot” reaction after a longer post-incubation time. Such kinetics have also been found in other studies with tobacco-burning products (RAHMAN et al. 1995, 1996). The differences between the two types of cigarette were statistically significant. In summary we conclude that this system allows the direct exposure of human lung cells to fresh, diluted native cigarette smoke and the analysis of specific biological effects of different cigarettes. The system offers many advantages over previous model systems and should have wide applicability in future studies. Acknowledgements: The authors would like to thank K. Hoffmann for his excellent technical assistance. This work was sponsored by a grant from the the FR+G (Forschungsgesellschaft Rauchen und Gesundheit mbH, Berlin, Germany).
References AUFDERHEIDE M, RITTER D, KNEBEL JW, et al.: A method for in vitro analysis of the biological activity of complex mixtures such as sidestream cigarette smoke. Exp Toxicol Pathol 2001; 53: 141–152. AUFDERHEIDE M, MOHR U: CULTEX – a new system and technique for the cultivation and exposure of cells at the air/liquid interface. Exp Toxicol Pathol 1999; 51: 489–490. BOMBICK DW, AYRES PH, DOOLITTLE DJ: Cytotoxicity assessment of whole smoke and vapor phase of mainstream and sidestream cigarette smoke from three Kentucky reference cigarettes. Toxicology Methods 1997; 7: 179–192.
EMURA M, RIEBE M, OCHIAI A, et al.: New functional cellculture approach to pulmonary carcinogenesis and toxicology. J Cancer Res Clin Oncol 1990; 116: 557–562. KNEBEL JW, RITTER D, AUFDERHEIDE M: Development of an in vitro system for studying effects of native and photochemically transformed gaseous compounds using an air/liquid culture technique. Toxicol Lett 1998; 96–97: 1–11. KNEBEL JW, RITTER D, AUFDERHEIDE M: Exposure of human lung cells to native diesel motor exhaust – development of an optimized in vitro test strategy. Toxicol In Vitro 2002; 16: 185–192. LAFI A, PARRY JM: The effects of ventilation and the dilution of smoke upon the clastogenic and aneugenic activity of tobacco particulate matter in cultured mammalian cells. Mutat Res 1991; 264: 51–57. MCKARNS SC, BOMBICK DW, MORTON MJ, et al.: Gap junction intercellular communication and cytotoxicity in normal human cells after exposure to smoke condensates from cigarettes that burn or primarily heat tobacco. Toxicol In Vitro 2000; 14: 41–51. RAHMAN I, LI XY, DONALDSON K, et al.: Glutathione homeostasis in alveolar epithelial cells in vitro and lung in vivo under oxidative stress. Am J Physiol 1995; 269: L285–292. RAHMAN I, SMITH CA, LAWSON MF, et al.: Induction of gamma-glutamylcysteine synthetase by cigarette smoke is associated with AP-1 in human alveolar epithelial cells. FEBS Lett 1996; 396: 21–25. RITTER D, KNEBEL JW, AUFDERHEIDE M, et al.: Development of a cell culture model system for routine testing of substances inducing oxidative stress. In Vitro Toxicology 1999; 13: 745–751. RITTER D, KNEBEL JW, AUFDERHEIDE M: In vitro exposure of isolated cells to native gaseous compounds – development and validation of an optimised system for human lung cells. Exp. Toxic. Pathol. 2001; 53: 373–386. ROEMER E, TEWES FJ, MEISGEN TJ, et al.: Evaluation of the potential effects of ingredients added to cigarettes. Part 3: in vitro genotoxicity and cytotoxicity. Food Chem Toxicol 2002; 40: 105–111. WILLEY JC, GRAFSTROM RC, MOSER JR CE, et al. : Biochemical and morphological effects of cigarette smoke condensate and its fractions on normal human bronchial epithelial cells in vitro. Cancer Res 1987; 47: 2045–2049. WINKELMEIER P, GLAUNER B, LINDL T: Quantification of cytotoxicity by cell volume and cell proliferation. ATLA 1993; 21: 269–280.
Exp Toxic Pathol 55 (2003) 1
57
Fraunhofer Institute of Toxicology and Experimental Medicine, Hannover, Germany
An improved in vitro model for testing the pulmonary toxicity of complex mixtures such as cigarette smoke* MICHAELA AUFDERHEIDE, JAN W. KNEBEL, and DETLEF RITTER With 8 figures Received: December 2, 2002; Revised: January 21, 2003; Accepted: February 25, 2003 Address for correspondence: Prof. Dr. MICHAELA AUFDERHEIDE, Fraunhofer Institute of Toxicology and Experimental Medicine, Department of In Vitro Toxicology, Nikolai-Fuchs-Strasse 1, 30625 Hannover, Germany; Tel.: +49 511 5350 252, Fax: +49 511 5350 444, e-mail: [email protected] Key words: In vitro exposure; air/liquid interface; cigarette mainstream smoke; human lung cells.
Summary
Introduction
Numerous approaches have been employed for testing the biological activity of cigarette smoke in vitro. None of them has managed to expose cultured lung cells in a realistic manner to the complex gaseous and particulate mixture that constitutes cigarette smoke. We have devised a system that makes this possible. The system presented here enables the direct exposure of human lung cells to native, unmodified cigarette mainstream smoke. It consists of a smoking machine, a dilution device for the smoke, analytical devices for online monitoring and a specially adapted exposure module based on the Cultex** cell cultivation system that is equipped with a gas-exposure top. Due to the special design of the exposure device and the optimised exposure conditions, this equipment allows cultured human lung cells to be exposed to freshly generated cigarette mainstream smoke. Exploratory experiments revealed that the smoke could be diluted over a wide concentration range in a reproducible way with respect to gas and particulate phases, and also demonstrated reproducible particle deposition depending on smoke concentration. Furthermore, it was shown that the exposed cells maintained their viability. Native cigarette mainstream smoke induced dose-dependent cellular effects in exposed cells with respect to cellular viability (viable cell number monitored by tetrazolium salt cleavage) and intracellular parameters (ATP and glutathione content). Therefore, fresh, physically and chemically unmodified cigarette mainstream smoke can be tested using this novel system.
Testing the biological effects of smoke aerosols in vitro has been carried out using a variety of strategies. Nevertheless, all approaches involve three common fundamental objectives: (1) the use of a specified biological indicator system, (2) the exposure of the biological test system in a way that ensures contact with the cigarette smoke without altering its properties, and (3) the analysis of biological/biochemical alterations induced by the smoke in the indicator system. The choice of the biological indicator system is determined by many factors, including the possibility of comparison with historical data determined using the same system, reproducibility, availability, and applicability to the in vitro situation of the biological endpoints determined after exposure in vitro. Because of its obvious relevance, a human lung cell system is favoured. The choice of the endpoints for the analysis depends to a large extent on the aim of a particular study and is also subject to the considerations relevant for the indicator system. In contrast, the exposure conditions dictate a number of clear conditions because of the physical and chemical properties of the smoke and the fundamental requirements for maintaining the viability of the cells during exposure. Typically, cells are exposed to smoke extracts. Cigarettes are smoked in a smoking machine according to interna-
* Reported in part at the CORESTA Congress, September 22–27, 2002. ** (Pat. No. DE 19801763/PCT/EP99/00295)
Abbreviations: DMSO: dimethyl sulphoxide; PBS: phosphate-buffered saline; TPM: total particulate matter. 0940-2993/03/55/01-051 $ 15.00/0
51
tional smoking standards and total particulate matter (TPM) is collected on a Cambridge filter or by other trapping methods. TPM is extracted with an organic solvent, usually dimethyl sulphoxide (DMSO), and the organic extract is then added to the cell culture medium for the submerged exposure of cultivated cells (LAFI et al. 1991; MCKARNS et al. 2000; WILLEY et al. 1987). Since this strategy takes into account only the effects of the particulate phase of the smoke, efforts have been made to test the gaseous phase of the smoke in a similar way. While TPM is sampled on a filter, the remaining gas phase is bubbled through an aqueous solution, usually phosphate-buffered saline (PBS). This extract is added to the culture medium of a cell culture for testing purposes (ROEMER et a. 2002). There have also been attempts to test gas and particulate phases together. The whole smoke aerosol can be passed through culture flasks where cells are submerged in culture medium. The flask is steadily moved during exposure in such a way that the culture medium runs to one side of the flask and sets the cells free for exposure (BOMBICK et al. 1997). Such scenarios result in intermittent exposures that change periodically from a situation where cells are covered by a thick layer of medium to a situation allowing a more direct exposure with a thinner overlay of culture medium on the cells. Hence, common exposure techniques used to date all involve more or less drastic modifications of the smoke aerosol. When using extracts of smoke, the individual constituents of the crude smoke condensate mixture will be effective to a different extent according to their physico/chemical properties, mainly their solubility. This is the case, for example, for compounds of the gaseous phase when extracted using an aqueous salt solution. Organic gaseous components not soluble in the aqueous phase are extracted to a lesser extent and, thus, their effect may be underestimated in the subsequent toxicological tests. Additionally, components are aged by the procedures and the particulate character is lost completely. Furthermore, any liquid between the biological indicator system and the whole smoke aerosol or its extracted components acts as a barrier between the cells and the test substance. Firstly, this is due to physical reasons. Gas and particulate constituents of the smoke aerosol have to diffuse through the liquid or have to be dissolved to come into contact with the cells. Consequently, the effective exposure concentration of each constituent is determined by solubility and diffusion properties. Secondly, this is due to possible chemical reactions. Culture media are composed of many highly reactive substances like glutathione, amino acids and others, and will readily change the smoke components by way of chemical reactions. As a result of these factors, the qualitative and quantitative composition and dosage of cigarette mainstream smoke in the toxicological test are defined only partly by the composition and doses of the smoke aerosol generated. They are determined to a great extent by the physical and chemical changes induced in the smoke by processes that take place during extraction or exposure. 52
Exp Toxic Pathol 55 (2003) 1
The aim of this study was to develop a system for direct exposure of human lung cells to native cigarette mainstream smoke with as little change as possible in the smoke properties. To achieve this, an exposure strategy was used that rendered possible the exposure of human lung cells to the smoke aerosol directly at the air-liquid interface. Adherent cells were grown on microporous membranes and cultured biphasically, i.e. nourished by culture medium through the membrane and exposed directly, without medium coverage, to the atmosphere from above (fig. 3). Using this mode of exposure, cells have successfully been exposed to native gaseous compounds (RITTER et al. 2001), native cigarette sidestream smoke (AUFDERHEIDE et al. 2001) and native diesel exhaust fumes (KNEBEL et al. 2002). An adaptation of this technique was developed to allow exposure to diluted cigarette mainstream smoke using an automatic smoking machine. This paper presents a description and physical characterisation of the system as well as validation by means of measuring selected indices of activity in cells exposed to various concentrations of native cigarette mainstream smoke.
Material and methods Smoke generation, dilution and cell exposure system: An experimental system was developed as shown in figure 1. It consisted of: (1) an automatic smoking machine (Borgwaldt RM 1/G), working with smoking parameters according to FTC/ISO standards (35 ml puff volume/2 s, one puff per minute), (2) a dilution system for the fresh cigarette mainstream smoke, driven by synthetic air, (3) analytical instruments for online monitoring of aerosol concentrations (particles and gas), and (4) the exposure device for cultured cells growing adherently on microporous membranes (AUFDERHEIDE and MOHR 1999; RITTER et al. 2001). Figure 2 illustrates the status of the cells during exposure. For physico-chemical characterisation of the exposure, 1R4F cigarettes (University of Kentucky, conditioned according to ISO) were smoked with a constant number of 9 puffs. The mainstream smoke generated was fed into a constant flow of synthetic air (20.5% O2 in N2, flow control by mass-flow controller, Analyt, Müllheim, Germany) according to figure 1. In this way the discontinuously generated smoke was incorporated into a continuous flow. The smooth constant flow was passed through a light-scattering photometer and through the exposure device. The periodical changes in relative smoke and synthetic air concentrations were recorded (fig. 3). Cells were located in the exposure device on microporous membranes (0.4 µm pore size, 1 cm2, Falcon). The exposure device was heated by a water-bath. The cell exposure did not require the use of a cell culture incubator. For the investigation of the biological effects of different cigarette types, two research cigarettes were conditioned and smoked in a blind study according to ISO: Cigarette A had 9.0 mg of tar/cigarette, 14.3 mg of carbon monoxide (CO)/cigarette, 0.6 mg of nicotine/cigarette: Cigarette B had 1.4 mg of tar/cigarette, 0.6 mg of CO/ cigarette, 0.09 mg of nicotine/cigarette. British American Tobacco (BAT), Germany, kindly provided the experimental cigarettes.
Fig. 1. Experimental system for the generation, dilution and online monitoring of cigarette mainstream smoke and exposure of human lung cells adhering to microporous membranes. For details see text.
Physico-chemical measurements: Particle analysis was carried out using light-scattering photometry. The light-scattering photometers (from the Fraunhofer Department of Aerosol Research) were positioned in front of and behind the exposure device. They were calibrated for linearity, and performance in the desired aerosol concentration range was checked by filter analysis. Gas monitoring was carried out online by non-dispersive infrared spectroscopy (NDIR) with CO and CO2 monitors (Maihak Unor 6N, Hamburg, Germany) calibrated for the desired concentration ranges. Quantification of particle deposition was carried out by organic extraction of exposed culture membranes and fluorescence measurement of the extract in a microtiter plate fluorescence reader (Spectramax Gemini XS, Molecular Devices, Ismaning, Germany) Acquisition and processing of data was carried out online using data acquisition equipment (National Instruments, Munich, Germany) and software (DASYLab32, Datalog, Mönchengladbach, Germany) on a standard personal computer.
Fig. 2. Cell exposure in detail. Cells were grown on a microporous membrane (2) with culture medium beneath the membrane (3) and aerosol flow (1) above them without an overlying liquid layer. Right: View of one of three inserts being exposed in parallel in one exposure device.
Cell culture: A549 cells and human foetal bronchial epithelial cells (HFBE cells) (EMURA et al. 1990) were routinely thawed from a stock pool for each experiment and cultured in 75 cm2 flasks. Subconfluent cultures were trypsinised and cell viability was determined using an electronic cell counter (CASY, Schärfe Systems, Reutlingen, Germany). Cells were seeded onto polyethylene terephthalate (PET) track-etched membranes with a pore size of 0.4 µm, 1.6 × 106 pores/cm2 and a growth area of approximately 1 cm2 (Becton Dickinson, Germany), and cultured in an incubator at 95% RH (37 °C) and 5% CO2. Culture membranes were transferred into the exposure device before exposure. Culture medium below the membrane during the exposure contained no foetal calf serum (FCS). Analysis of cells: Tetrazolium salt cleavage was measured using the WST-1 dye (Boehringer, Mannheim, Ger-
Fig. 3. Time-dependent particle concentrations at the inlet (particles in) and outlet (particles out) of the cell exposure device. The time count started with the beginning of the puff. Exp Toxic Pathol 55 (2003) 1
53
Fig. 4. Results of particle analysis during smoke exposure using different dilution flow rates. Each point represents the smoking of one 1R4F cigarette with 9 puffs. Means represent smoke concentrations during the smoking time of 9 minutes.
many). Cells were incubated with a dilution of the dye according to the protocol developed by the manufacturer. After incubation, formazan production was quantified by absorption reading using a microtiter plate reader (Spectramax 340PC, Molecular Devices, Ismaning, Germany) at wavelengths of 450/630 nm. The number of viable cells was determined directly after tetrazolium salt analysis of the same cells using an electronic cell counter (CASY, Schärfe Systems, Reutlingen, Germany) after trypsinisation and dilution of the cell suspension (KNEBEL et al. 1998). This method is appropriate for an automated, quantitative discrimination between living and dead cells (WINKELMEIER et al. 1993). The protein content of the cells was quantified using a commercial kit according to the protocol of the manufacturer (DC Protein Assay, Biorad, Munich, Germany). The ATP/ADP ratio of exposed cells was analysed using reversed phase HPLC with UV detection at 259 nm following extraction with trichloroacetic acid (RITTER et al. 1999). Intracellular glutathione in cells was determined using a modified Tietze recycling assay after extraction with metaphosphoric acid (RITTER et al. 1999). The intracellular redox ratio of oxidised and reduced glutathione GSSG/GSH was quantified using extraction with meta-phosphoric acid, derivatisation using 2,4-dinitrofluorobenzene and detection of the N-2,4-dinitrophenyl derivatives by UV at 355 nm after HPLC (RITTER et al. 2001).
Results
Fig. 5. Results of particle deposition analysis on culture membranes using different dilution flow rates and smoking one or two cigarettes with 9 puffs.
In exploratory experiments without cells the experimental system was examined with regard to the physicochemical characteristics of the smoke aerosol generated. The age of the smoke when it came into contact with the exposed cells could be estimated by the online analysis of the particle concentration at the inlet of the exposure device. Measuring the time between the beginning of the smoke puff and the arrival of the corresponding particle peak in the photometer gave the age of the smoke. This time was approximately 6 seconds; it did not depend on
Fig. 6. Exposure of A549 cells to synthetic air for 60 minutes under conditions of cigarette mainstream smoke exposure. Cells were post-incubated for 2 hours after treatment and then analysed for the parameters indicated. Control cells were left unexposed in the incubator for the same time and analysed in the same way. 54
Exp Toxic Pathol 55 (2003) 1
the dilution used. Figure 3 shows the time-dependent particle concentrations measured at the inlet and outlet of the exposure device. The diagram illustrates the detection of the particles of a puff about 6 seconds after the start of the puff. Furthermore, it shows that the particles were detected at the outlet of the exposure device before the particles of the following puff entered. There was obviously no mixing of sequential puffs . The dilution of particles and gas phase of mainstream smoke was validated using light-scattering photometry and NDIR detection of carbon monoxide and carbon dioxide in the aerosol. Mean particle concentrations during the smoking of one cigarette with 9 puffs using dilution flows in the range of 1 to 30 l/min are illustrated in figure 4. The figure shows that the particulate phase was effectively diluted in a reproducible manner. The same characteristics were found for CO and CO2 concentrations (data not shown), resulting in a stable CO/CO2 ratio and a stable CO/particle ratio independent of the dilution. The particle deposition on the cell culture membrane was analysed by exposing membranes without cells under conditions of cell exposure and quantification of the deposited particle mass by a fluorescence method. Figure 5 shows the results for the deposition of smoke aerosol particles at different dilution flow rates when one or two cigarettes were smoked (9 or 18 puffs). The reproducible particle deposition, depending on smoke aerosol concentration (dilution) and dosage (number of cigarettes), was documented. In preliminary experiments human lung cells were exposed using the system. A549 cells and human foetal bronchial epithelial cells (HFBE cells) were exposed to synthetic air to investigate the effects of exposure on the viability of the cells. Furthermore, HFBE cells were exposed to the native, diluted cigarette mainstream smoke
of two different research cigarettes to investigate the smoke effects on the cellular viability in relationship to the dose. Additional exposures to low smoke concentrations using short exposure times of 6 minutes were carried out to examine the sensitivity of the system. Figure 6 shows the results of A549 cell exposures to synthetic air for 60 minutes under the conditions used subsequently for the exposure to cigarette mainstream smoke. Cells were post-incubated for 2 hours after exposure and analysed for the parameters indicated above. Protein content, tetrazolium salt cleavage, ATP content, ATP/ADP ratio and intracellular content of glutathione were decreased very slightly by exposure to synthetic air, whereas the glutathione redox ratio was increased very slightly in comparison to controls. None of these changes was statistically significant and no loss of cells could be detected after determining viable cell number. These results show that the exposure to synthetic air under the conditions of cigarette mainstream smoke exposure did not reduce the viability of the cells. The results of cell exposures to cigarette mainstream smoke using HFBE cells and varying numbers of cigarette puffs generated from research cigarettes A and B are presented in figure 7. Cells were exposed for a total period of 28 minutes. Within this time, different numbers of cigarettes (up to 4, max. 7 puffs per cigarette) were smoked with 1 puff per minute (lowest dose of cigarette A was 4 puffs) and constant smoke dilution. After smoke exposure, cells were exposed to synthetic air for the rest of the 28-minute exposure period. Controls were exposed to synthetic air only. Figure 7 shows the results of the tetrazolium salt assay, which represents a sensitive marker for cellular viability and activity of metabolic processes. Using cigarette A, viability was clearly decreased compared to controls at doses of 18 or more
Fig. 7. Tetrazolium salt cleavage per cell in HFBE cells exposed to the diluted native cigarette mainstream smoke of cigarettes A and B using increasing numbers of puffs during a total exposure time of 28 minutes at a constant dilution flow rate of 1 l/min. Cells were exposed to the indicated number of cigarette puffs and to synthetic air for the rest of the exposure time, and post-incubated for 2 hours. Controls were exposed to synthetic air only (lowest dosage of cigarette A was 4 puffs). T-test cigarette A versus cigarette B, *** P < 0.001, t-test cigarette B versus control, ++ P < 0.01, +++ P < 0.001. Exp Toxic Pathol 55 (2003) 1
55
puffs. Mainstream smoke of cigarette B did not induce any decrease in viability through the whole concentration range, but increased the cellular metabolic activity using 14, 21 or 28 cigarette puffs compared to controls. This increase was statistically significant in comparison to controls and to results after exposure to the smoke of cigarette A at the same doses. HFBE cells were also exposed to 6 puffs of cigarettes A or B or synthetic air during a shortened exposure period of 6 minutes. After exposure, cells were post-incubated for varying times up to 24 hours or were not post-incubated at all (time 0 hours). Tetrazolium salt cleavage, viable cell number, ATP/ADP ratio and intracellular glutathione content were analysed. Figure 8 shows the results of the determinations of the ATP/ADP ratio and the intracellular glutathione content at times of 0 hours and 24 hours post-incubation compared to controls (synthetic air exposure). ATP/ADP ratios were decreased in HFBE cells after exposure to smoke from cigarette A to about 50% at time 0 and to about 75% at 24 hours. In contrast, smoke from cigarette B did not induce a significant decrease. The difference in the ATP/ADP ratios at 24 hours post-incubation time in cigarettes A and B was statistically significant. Glutathione content of cells exposed to mainstream smoke from cigarette A was decreased to 10% of control at 0 hours and increased to about 175% of control at 24 hours post-incubation time. After exposure to smoke from cigarette B the glutathione content of cells was decreased only very slightly at 0 hours and was comparable to controls at 24 hours post-incubation. The difference in the glutathione content at 24 hours post-incubation in cigarette A and B was statistically significant.
Discussion An experimental system was devised to expose human lung cells to freshly generated, diluted cigarette main-
stream smoke aerosol. The aim of the study was to induce only minimal changes in the smoke, which was generated according to international smoking standards, and to achieve effective contact between exposed cells and the exposure atmosphere both for the gaseous and particulate constituents of the smoke. The system works by exposing cells directly at the airliquid interface on microporous membranes. Experiments using such a system for investigations of native gaseous compounds and other airborne materials (RITTER et al. 2001; AUFDERHEIDE et al. 2001; KNEBEL et al. 2002) had previously shown that the exposure of human lung cells could be accomplished without any physical modification of the test atmosphere conditions (e.g. by humidification, heating or addition of CO2) that are routine procedures in cell culture. Such an in vitro system is essential for testing cigarette mainstream smoke under conditions as close as feasible to those in vivo. Physico-chemical characterisation of the smoke aerosol generated in the system demonstrated that the main particle peak entered the exposure device about 6 seconds after the start of the puff. Both the particulate and gaseous phases of the aerosol were diluted effectively and in a reproducible manner. The stable ratios of CO to particle concentration and CO to CO2 concentration irrespective of the dilution demonstrated that this method of smoke aerosol dilution induced no changes in the particulate phase or in the gaseous phase. The effective contact of the gas phase and the cells using this cell culture system had been shown in a study on gaseous substances (RITTER et al. 2001). Reproducible and dose-dependent particle deposition on the culture membrane was demonstrated in this study. Moreover, the physical mechanisms of sedimentation and diffusion used here for particle deposition follow the same principles as in human distal lung airspaces in vivo. Experiments using two kinds of human lung cells, A549 cells and human foetal bronchial cells, showed that
Fig. 8. Exposure of HFBE cells to the diluted native cigarette mainstream smoke of cigarettes A and B. Cells were exposed to 6 puffs, corresponding to an exposure time of 6 minutes, and analysed directly after exposure or after a post-incubation time of 24 hours for intracellular ATP/ADP ratio and glutathione content. Control cells were exposed to synthetic air only. T-test cigarette A versus cigarette B, * P < 0.05. 56
Exp Toxic Pathol 55 (2003) 1
the cells could be exposed in the system without loss of viability. Moreover, exposure to two different types of research cigarettes in preliminary experiments resulted in different dose/effect relationships. Smoke from cigarette A, which has more tar, CO and nicotine than cigarette B, clearly decreased the viability of exposed cells, whereas smoke from cigarette B induced metabolic activation under comparable conditions. Smoke from cigarette A induced a typical kinetic profile for the glutathione content of exposed cells, i.e. the depletion directly after exposure and an “overshoot” reaction after a longer post-incubation time. Such kinetics have also been found in other studies with tobacco-burning products (RAHMAN et al. 1995, 1996). The differences between the two types of cigarette were statistically significant. In summary we conclude that this system allows the direct exposure of human lung cells to fresh, diluted native cigarette smoke and the analysis of specific biological effects of different cigarettes. The system offers many advantages over previous model systems and should have wide applicability in future studies. Acknowledgements: The authors would like to thank K. Hoffmann for his excellent technical assistance. This work was sponsored by a grant from the the FR+G (Forschungsgesellschaft Rauchen und Gesundheit mbH, Berlin, Germany).
References AUFDERHEIDE M, RITTER D, KNEBEL JW, et al.: A method for in vitro analysis of the biological activity of complex mixtures such as sidestream cigarette smoke. Exp Toxicol Pathol 2001; 53: 141–152. AUFDERHEIDE M, MOHR U: CULTEX – a new system and technique for the cultivation and exposure of cells at the air/liquid interface. Exp Toxicol Pathol 1999; 51: 489–490. BOMBICK DW, AYRES PH, DOOLITTLE DJ: Cytotoxicity assessment of whole smoke and vapor phase of mainstream and sidestream cigarette smoke from three Kentucky reference cigarettes. Toxicology Methods 1997; 7: 179–192.
EMURA M, RIEBE M, OCHIAI A, et al.: New functional cellculture approach to pulmonary carcinogenesis and toxicology. J Cancer Res Clin Oncol 1990; 116: 557–562. KNEBEL JW, RITTER D, AUFDERHEIDE M: Development of an in vitro system for studying effects of native and photochemically transformed gaseous compounds using an air/liquid culture technique. Toxicol Lett 1998; 96–97: 1–11. KNEBEL JW, RITTER D, AUFDERHEIDE M: Exposure of human lung cells to native diesel motor exhaust – development of an optimized in vitro test strategy. Toxicol In Vitro 2002; 16: 185–192. LAFI A, PARRY JM: The effects of ventilation and the dilution of smoke upon the clastogenic and aneugenic activity of tobacco particulate matter in cultured mammalian cells. Mutat Res 1991; 264: 51–57. MCKARNS SC, BOMBICK DW, MORTON MJ, et al.: Gap junction intercellular communication and cytotoxicity in normal human cells after exposure to smoke condensates from cigarettes that burn or primarily heat tobacco. Toxicol In Vitro 2000; 14: 41–51. RAHMAN I, LI XY, DONALDSON K, et al.: Glutathione homeostasis in alveolar epithelial cells in vitro and lung in vivo under oxidative stress. Am J Physiol 1995; 269: L285–292. RAHMAN I, SMITH CA, LAWSON MF, et al.: Induction of gamma-glutamylcysteine synthetase by cigarette smoke is associated with AP-1 in human alveolar epithelial cells. FEBS Lett 1996; 396: 21–25. RITTER D, KNEBEL JW, AUFDERHEIDE M, et al.: Development of a cell culture model system for routine testing of substances inducing oxidative stress. In Vitro Toxicology 1999; 13: 745–751. RITTER D, KNEBEL JW, AUFDERHEIDE M: In vitro exposure of isolated cells to native gaseous compounds – development and validation of an optimised system for human lung cells. Exp. Toxic. Pathol. 2001; 53: 373–386. ROEMER E, TEWES FJ, MEISGEN TJ, et al.: Evaluation of the potential effects of ingredients added to cigarettes. Part 3: in vitro genotoxicity and cytotoxicity. Food Chem Toxicol 2002; 40: 105–111. WILLEY JC, GRAFSTROM RC, MOSER JR CE, et al. : Biochemical and morphological effects of cigarette smoke condensate and its fractions on normal human bronchial epithelial cells in vitro. Cancer Res 1987; 47: 2045–2049. WINKELMEIER P, GLAUNER B, LINDL T: Quantification of cytotoxicity by cell volume and cell proliferation. ATLA 1993; 21: 269–280.
Exp Toxic Pathol 55 (2003) 1
57