Carcinogenic Activity of Cigarette Smoke Gas Phase and Its Modulation by Beta-Carotene and N-Acetylcysteine
http://www.100md.com
《毒物学科学杂志》
Center for Health and the Environment, University of California, Davis, One Shields Avenue, Davis, California 95616
ABSTRACT
Male strain A/J mice were exposed for six hours a day, five days a week for six months to either full tobacco smoke or to tobacco smoke drawn through a HEPA filter that removed more than 99% of particulate matter. After another four months in air, the animals were sacrificed and lung tumors were counted for calculation of multiplicities and incidences. Analysis of the chamber atmospheres showed that in the filtered smoke the concentrations of polycyclic aromatic hydrocarbons and tobacco smoke specific nitrosamines were reduced to from below 18% to even nondetectable levels of the original values measured in the unfiltered smoke. Aldehydes and other volatile organic compounds such as 1,3-butadiene, benzene, or acrolein were reduced to about 50 to 90% of the concentrations found in unfiltered smoke. Some potentially carcinogenic metals reached levels in filtered smoke ranging from 77% to less than 1% found in full smoke. The mice exposed to the filtered smoke atmosphere had practically identical lung tumor multiplicities and incidence as had the animals exposed to full smoke, significantly higher than in air exposed controls. Diets containing 0.5% -carotene or 0.4% N-acetylcysteine afforded some chemoprevention. It was tentatively concluded that 1,3-butadiene might be an important contributor to lung tumorigenesis in this mouse model of tobacco smoke carcinogenesis.
Key Words: tobacco smoke; mice; strain A/J mice; lung tumors; 1,3-butadiene.
INTRODUCTION
The first evidence that inhalation of tobacco smoke caused lung cancer was obtained in human studies, during the 1930's and 1940's in Germany and in the 1950's in Great Britain and the U.S. (Doll, 1998). Evidence for the carcinogenicity of tobacco smoke in experimental animals was more difficult to obtain. Only a few inhalation studies were conclusive, the most convincing ones being large experiments with hamsters where tobacco smoke produced laryngeal tumors, but no tumors in the lower respiratory tract (IARC, 2004). On the other hand, skin painting studies in mice with tobacco smoke condensate produced early on evidence that the "tar" or particulate phase of tobacco smoke was carcinogenic. From then on, it was generally assumed that the carcinogenic potential of tobacco smoke was mostly, if not exclusively, associated with the particulate phase and its polycyclic aromatic hydrocarbons and nitrosamines.
Interestingly, some dissenting opinions were aired already in 1961: ‘Nearly all investigations on the tobacco problem have been started with the assumption that 1) the carcinogenic agents are produced only in the process of smoking and 2) that higher aromatic hydrocarbons, especially benzpyrene compounds, are the main causative substances. This however, seems to be a prejudice’ (Druckrey, 1961). Nevertheless, during the ensuing years considerable efforts were made to mitigate adverse health effects caused by smoking. It was anticipated that the development of low tar–low nicotine filter cigarettes would have a beneficial effect. Unfortunately, this does not seem to have been the case. Changes in cigarette design and manufacturing failed to benefit public health and no evidence of reduced risk could be found (NCI, 2001). On the other hand, Lee and Sanders (2004) concluded, after an extensive analysis of the available epidemiological literature, that the smoking of filter and low tar cigarettes carried 23 to 36% less risk than the smoking of unfiltered ones. They still considered the risk to develop lung cancer as being high.
While thus the putative role in carcinogenesis of the tobacco smoke tar phase, particularly of its polycyclic aromatic hydrocarbons and tobacco specific nitrosamines, was widely studied, much less attention was paid to the gas phase. The gas phase of cigarette smoke contains some 400 to 500 compounds (Hoffmann and Wynder, 1999). The bulk of gases are oxygen, nitrogen, carbon dioxide, and carbon monoxide. Some well-known carcinogens such as benzene, formaldehyde, acetaldehyde, and 1,3-butadiene are also present. Initial attempts to establish the carcinogenic potential of the gas phase were negative in mice; in hamsters, where full smoke produces laryngeal tumors, no such lesions were observed after gas phase inhalation only (IARC, 2004). However, in 1974 studies in Snell's mice showed unequivocally that tobacco smoke gas phase was as active in producing benign and malignant lung tumors as was full tobacco smoke (Leuchtenberger and Leuchtenberger, 1971, 1974). This was later confirmed in strain A mice where tumor development was correlated with a partial analysis of the chamber atmospheres for polycyclic aromatic hydrocarbons and nitrosamines (Witschi et al., 1997b). In the present experiments, we reproduced the previous findings and expanded analysis of chamber atmospheres. Analysis covered 100% of the gas phase carcinogens listed by Hoffmann and Wynder (1999), 70% of the carcinogenic inorganic compounds, 80% of carcinogenic polycyclic aromatic hydrocarbons present in more than 5 ng quantity per cigarette and four of the most important tobacco specific nitrosamines. In addition, we reexamined the effects of two putative chemopreventive agents on the carcinogenicity of cigarette smoke gas phase.
MATERIALS AND METHODS
Materials. Kentucky 2R4F reference cigarettes were obtained from the Tobacco Research Institute, University of Kentucky (Lexington, KY). They deliver on average 0.81 mg of nicotine and 11.6 mg of particulate material per cigarette. Diets AIN-93G and AIN-93M were purchased from DYETS (Bethlehem, PA). The -carotene diet with gelatine beadlets containing 10% -carotene was premixed by the supplier. Diets containing N-acetylcysteine (Sigma, St. Louis, MO) were prepared weekly by mixing in a Hobart blender. Test diets were stored at 4°C until used.
Animals. Male strain A/J mice, 5 to 6 weeks old were purchased from Jackson Laboratories, Bar Harbor, ME. After arrival, a few animals were selected at random for a standard rodent health surveillance screen by the Comparative Pathology Laboratory, UC Davis. No evidence for infectious disease (pathogenic agents) or presence of parasites or ova in pelage and cecum were reported. Due to the absence of significant macroscopic lesions in all animal examined, no histopathology was processed. Serology was negative for mouse hepatitis virus, Sendai virus, Reovirus type 3, pneumonia virus, parvo, ectromelia, and Mycoplasma pulmonis. After a two-week acclimatization period, the mice were assigned at random to the different treatment groups. They were housed in accordance with institutional guidelines four to a cage, in polypropylene cages with tightly fitting wire screen lids on a paper-based bedding material (Tek-fresh, Harland). They were fed diet AIN-93G until five months old and from then on diet AIN-93 M. At all times during the experiment, including during smoke exposure, water and food were freely accessible. The animals were observed daily and weights were recorded weekly.
Exposure system. The tobacco smoke exposure system was the one used in our previous studies on the effects of full smoke and filtered smoke (Witschi et al., 1997a,b). Briefly, two Hinners-type stainless steel inhalation chambers (volume 0.44 m3) were used. In chamber 1, mice were exposed to a mixture of tobacco smoke (TS) generated from 89% cigarette sidestream smoke and 11% mainstream smoke. In chamber 2, the TS was drawn before entering the chamber through a HEPA filter that removed >99.8% of particulate material.
Experimental design. All protocols had been approved by the UC Davis Animal Use and Care Administrative Advisory Committee. Animals were exposed to full or filtered tobacco smoke 6 h a day, five days a week, for a total of 25 weeks. Animals in the full smoke chamber were kept on AIN-93G/M diet throughout. In the filtered smoke chamber, animals were divided into three groups: Ain-93G/M diet alone, AIN93-G/M diet containing 0.5% -carotene and AIN-93G/M diet containing 0.4% N-acetylcysteine (NAC). Control animals, fed the same diets, were housed in inhalation chambers ventilated with filtered air. After 25 weeks, all animals were removed into a conventional animal facility and killed 18 weeks later.
Analysis of chamber atmospheres. Total suspended particulates, CO, and nicotine were measured in both chambers during the entire TS exposure period. CO concentrations were monitored every 30 min with a model 880 non-dispersive-infrared (NDIR) analyzer and total suspended particulate mater (TSP) was measured three times a day in the full smoke chamber and once a day in the filtered smoke chamber by weighing material collected on Pallflex filters. Nicotine was measured once a week by drawing 10 l of air through XAD4 cartridges and subsequent analysis of the extraction liquid with gas chromatography/thermionic selective detector.
For additional measurements of selected tobacco smoke constituents, chamber air samples were collected on five different time points, spaced approximately one month apart from each other, and sent for analysis to commercial laboratories. For determination of metals, 8 l of air were collected on a 25 mm (3 μm) TEFLO filter (Pall Corp) and analyzed by XRF (X-ray Fluorescence) at Chester Labnet (Tigard, OR). Polycyclic aromatic hydrocarbons (PAH) were measured by collecting a total of 60 l of air on 2 XAD-40 cartridges and subsequent analysis by gas chromatography/mass spectrometry (GC/MS) at Alta Analytical Laboratory Inc. (El Dorado Hills, CA). For the determination of acrolein, benzene, and 1,3-butadiene, air samples were collected into a 6 l SUMMA canister and analyzed by GC/MS; for acetaldehyde and formaldehyde determination 10 l of air were collected on SEPPAK LPDNPH cartridges and analyzed by HPL at Air Toxics (Folsom, CA). For the determination of tobacco specific nitrosamines, 120 l of air were collected on Cambridge filter pads, extracted with 10 ml of 100 mM ammonium acetate and then analyzed for selected tobacco smoke-specific nitrosamines by LC/MS at Arista Laboratories (Richmond, VA).
Tissue preparation. After killing the animals by pentobarbital overdose, the lungs were manually expanded to total lung capacity through instillation of Tellyesniczky's fluid. After at least 24 h of fixation, the number of tumor nodules visible on the lung surface was counted under a magnifying glass. The results were expressed as tumor incidence, i.e., percentage of animals with one or several lung tumors and as tumor multiplicity. Tumor multiplicity was calculated as average number of tumors per lung of all animals, including nontumor bearing ones (Stoner and Shimkin, 1985). For histopathological evaluations, lung lobes containing tumors were dehydrated, embedded in paraffin, sectioned at 5 μm and stained with hematoxylin and eosin (H&E).
Statistical analysis. All numerical data were calculated as mean and SD or SEM. Comparisons of body weights and of tumor multiplicity between tobacco smoke-exposed and air-exposed controls were made by ANOVA followed by Tukey-Kramer's or Dunnetts post test. Tumor incidences were compared using Fisher's exact test. A p value of 0.05 or less was considered to be significant.
RESULTS
Exposure conditions for the two chambers are given in Table 1. The insertion of a HEPA filter between chamber 1 and chamber 2 effectively lowered the concentrations of total suspended particulate matter to 0.12% of chamber 1 and of nicotine to 16%, whereas CO was only reduced to 86%. In both chambers, the animals tolerated the tobacco smoke exposure well and only one death was observed during the exposure period. During the exposure to filtered and unfiltered smoke, the animals failed to gain weight as rapidly as did control animals kept in air (Fig. 1). After the six months exposure, the body weights of the animals kept in full smoke was significantly lower than those of all other groups, indicating that inhalation of the gas phase appeared to be less stressful than exposure to full smoke, although it still reduced body weight gain to a significant extent in comparison with controls kept in air. At the end of the experiment, the weights of the animals exposed to filtered or unfiltered tobacco smoke were not significantly different from the weight of control animals kept in air throughout.
Tables 2 and 3 show the results of analysis of the chamber atmospheres for several of the known carcinogens in tobacco smoke. Filtering the smoke resulted in a reduction of several polycyclic aromatic hydrocarbons, among them benzo(a)pyrene, to below detection levels. The concentrations of others were reduced to less than 10%, with the exception of naphthalene and methylnaphthalene which were reduced to only 69% resp. 51% of the full smoke atmosphere. Tobacco specific nitrosamines declined to 18% for 4-(methylnitrosamino)-1-(pyridyl)-1-butanone (NNK) and to 5% for N-nitrosonornicotine (NNN), with N-nitrosoanabasine (NAB) and N-nitrosoanatabine (NAT) becoming undetectable. Concentrations of the volatile organic compounds decreased much less and fell only to between 50 and 90% of the values measured in unfiltered smoke. The exception was formaldehyde where a bigger reduction was seen (to 17%). Metal concentrations were in general lower in filtered smoke compared to full smoke, but also were quite variable in the unfiltered smoke chamber where, with the exception of Cd and Ni, they were not detected in all samples.
Lung tumor data are shown in Table 4. In animals exposed to air and fed control diet or diets with -carotene or NAC, lung tumor multiplicities ranged from 0.9 to 1.1. Full tobacco smoke produced a significant increase in both lung tumor multiplicities and lung tumor incidence. Filtered smoke had exactly the same effect: lung tumor multiplicities and incidences were practically identical to the ones found in animals exposed to the full tobacco smoke. Ingestion of the -carotene and NAC diets, during and after smoke exposure, significantly reduced lung tumor multiplicities and lung tumor incidences, although not quite to the levels seen in animals kept in air.
From each group, 50 to 70% of the grossly visible tumors were sectioned and examined under the microscope. The tumors were classified into beingn tumors (adenomas) and malignant tumors (adenomas with carcinomatous foci or frank adenocarcinomas). Adenomas had a solid growth pattern with alveolar spaces being filled by contiguous neoplastic cells. Carcinomatous foci arising within adenomas often showed a papillary growth pattern and displayed a substantially higher percentage of abnormal mitotic figures. On average, 20 to 30% of all tumors showed features of malignancy (Table 5). The difference between the lowest rate found in the animals exposed to full smoke (9%) and the highest rate seen in the animals exposed to filtered smoke and fed -carotene (33%) was statistically not significant.
DISCUSSION
The present results confirm earlier findings (Leuchtenberger and Leuchtenberger, 1971, 1974; Witschi et al., 1997b). Although there were substantial differences in the concentrations of several carcinogens in the chamber atmospheres before and after HEPA filtration of cigarette smoke (a mixture of 11% cigarette mainstream and 89% sidestream smoke), tumor response was not affected. This might have some implications for experimental design. Most studies demonstrating enhanced lung tumor development by tobacco smoke and underlying mechanisms have been conducted by whole-body exposure of mice (De Flora et al., 2003). It has on occasion been suggested that under such conditions smoke particles deposit on the fur of the animals and carcinogens are absorbed orally through grooming rather than by inhalation. Since the gas phase has as much carcinogenic activity as has whole smoke, oral exposure to carcinogens seems unlikely.
Histopathological evaluation of more than 50% of all the grossly visible tumors revealed in about one-fourth signs of malignancy: carcinomatous foci arising within an adenoma or malignant features through the entire tumor, as described in detail before (Witschi et al., 1997a,b). The tumor type distribution was also similar to previous findings where control animals had 20 to 23% and tobacco smoke exposed animals 13 to 17% malignant tumors There was no indication that the chemoprevention regimens would result in a shift in tumor type. Such shifts are usually seen in studies that examine progression of lesions in the presence of chemopreventive agents (Estensen et al., 2004), an experimental design not adopted in the current study.
Our observations raise the question what agents are responsible for tumor development in the lungs of A/J mice. The general consensus appears to be that, at least in man, polycyclic aromatic hydrocarbons, particularly benzo(a)pyrene, and tobacco smoke specific nitrosamines are the main causative agents of lung cancer (Hecht, 1999). By inference, it is thought that this applies to animals, too. Our data show that, with the exception of naphthalene and methylnaphthalene, many polycyclic aromatic hydrocarbons were no longer detectable in the filtered smoke. Others were reduced to concentrations of less than 10% of the full smoke atmosphere. This makes it unlikely that PAHs were major contributors. Concentrations of NNK and NNN also fell below 20%. We have previously discussed how dosimetric calculations do not allow us to assume that NNK is a major carcinogen in strain A/J mice exposed to tobacco smoke (Witschi et al., 1997b).
Potentially carcinogenic volatile organic compounds (IARC classes 1, 2a, and 2b) were reduced much less in the filtered smoke. Data on the pulmonary carcinogenicity in mouse lungs of some of these agents are available and may be compared to our exposure concentrations as measured in filtered smoke. When strain A/J mice inhaled for 20 weeks an atmosphere of 157 mg/m3 of naphthalene (IARC 2B), only 11% of the mice developed lung tumors and multiplicities were 0.4 (Adkins et al., 1986). In a two-year study with B6C3F1 mice, the same concentration of naphthalene produced only a weak tumorigenic response (Abdo et al., 1992). The level of naphthalene measured in our study (0.177 mg/m3) was thus far below the one expected to produce tumors. Similar considerations apply to some of the other potentially carcinogenic volatiles (IARC class 1 and 2) such as acetaldehyde, formaldehyde, and benzene. After a 16-week exposure of CBA mice to an airborne benzene concentration of 957 mg/m3, 36% developed 18 months later pulmonary adenomas (Farris et al., 1993). After long term exposure of mice to 17.5 mg/m3 to formaldehyde, only nasal tumors were found (Kerns et al., 1983). No tumor data are available for mice following inhalation of acetaldehyde. In rats, exposure for 28 months to 1350 mg/m3 caused only nasal tumors (Woutersen et al., 1986). The airborne concentrations of these agents measured in our experiments did not nearly approximate such values. As far as acrolein is concenred (IARC class 3), oral administration to rats or mice failed to produce tumors (IARC, 1995). Considering metals, inhaled Cd is not a pulmonary carcinogen in mice (Oberdorster et al., 1994) and oral or subcutaneous exposure to Cd compounds have given conflicting results (Waalkes and Rehm, 1994). No Cr was detected in filtered smoke and concentrations of Ni were much lower than had been in positive inhalation assays with several nickel compounds in mice (0.3 mg/m3 vs. 2.5 mg/m3; Dunnick et al., 1995). While the concentrations of single agents in the unfiltered and filtered smoke cannot account for the overall tumor response, it is of course entirely possible that additive, synergistic, or tumor-promoting interactions between chemicals could have substantially enhanced tumorigenicity.
There is one agent that by itself might account for the carcinogenicity of full tobacco smoke or filtered tobacco smoke in strain A/J mice: 1,3-butadiene. A two-year inhalation study in C57Bl/6 x C3H F1 mice found increased neoplasms of the lung after exposure to 1,3-butadiene at concentrations as low as 13.8 mg/m3 (6.25 ppm) (Melnick et al., 1990). Strain A/J mice are more prone to develop chemically induced alveolar tumors than other strains and this probably also applies to 1,3-butadiene. The two exposure concentrations that yielded a tumorigenic response in our experiment are reasonably close to each other (3.6 mg/m3 for full smoke or 2.9 mg/m3 for filtered smoke) and not much lower than 13.8 mg/m3.1,3-Butadiene itself has been shown to inhibit its own metabolism to the highly active and carcinogenic diepoxide (Dahl and Henderson, 2000). It is conceivable that other organic volatiles might interfere with 1,3-butadiene metabolism. Partial removal of them by the filter might thus enhance butadiene activation and carcinogenesis. Since mice are much more sensitive to the toxic and carcinogenic action of 1,3-butadiene than are rats and even primates (Henderson, 2001), it is difficult to extrapolate lung tumor data induced by cigarette smoke from mouse to man based on 1,3-butadiene. Furthermore there is no evidence that 1,3-butadiene produces lung cancer in occupationally exposed humans (IARC, 1999). Nevertheless, it is interesting to note that 1,3-butadiene present in tobacco smoke has been shown to enhance the development of arteriosclerotic plaques in animal models of arteriosclerosis, whereas the tar phase does not seem to have such an effect (Penn et al., 1996; Penn and Snyder, 1996).
-Carotene and NAC are two agents who, among other activities, have strong antioxidant properties. In many experimental systems they have chemopreventive action and are capable to significantly reduce tumor development. Multiple mechanistic studies with biomarkers of effect have suggested that both agents should provide chemoprevention against tobacco smoke (De Flora et al., 1999, 2001). Experiments to show this have not been fully successful. In strain A mice, -carotene and NAC had no chemopreventive action against full cigarette smoke (Obermueller-Jevic et al., 2002; Witschi et al., 1998). In SWR mice, a small, but statistically not significant effect of NAC was observed (de Flora et al., 2003). The protective effect of both agents against the gas phase was thus unexpected. It would seem reasonable to assume that the oxidative stress placed onto the lung by filtered tobacco smoke is as intense as it would be with full smoke. If the main carcinogenic stimulus in filtered smoke was oxidative stress, a plausible explanation for the protective effects of the two antioxidants could be postulated. However, this makes it difficult to explain why neither -carotene nor NAC were effective (in strain A mice) against full tobacco smoke. It has been discussed before that two main classes of carcinogens found in full smoke, polycyclic aromatic hydrocarbons and nitrosamines, might not be as important contributions to carcinogenesis as is generally assumed. This allows the speculation that the full smoke contains some other as of yet unidentified carcinogen(s) that could overcome any beneficial effects of -carotene or NAC. It also has been postulated that polycyclic aromatic hydrocarbons found in tobacco tar act mostly as promoting agents (Rubin, 2001). Such processes could counteract or even nullify critical protective mechanisms of -carotene and NAC against full smoke. It is also conceivable that some elements of full tobacco smoke may inactivate or interfere with the bioavailability of the two chemopreventive agents. Initial studies with cell culture systems will allow to test these different hypotheses.
As surprising as they are, the present findings probably will not have much practical impact on chemoprevention of lung cancer in man, given the disappointing results of major clinical trials (Omenn, 1998; van Zandwijk et al., 2000). On the other hand, both -carotene and NAC have been found to mitigate biomarkers of exposure and possibly of effect in phase I and phase III clinical trials (de Flora et al., 2003; IARC, 1998). Given the right circumstances, -carotene and NAC may yet have some beneficial effects. Our observations may lead to future experiments where mechanistic events that relate to chemoprevention can be separated from events that are inconsequential.
In conclusion, experimental evidence shows that the gas phase of tobacco smoke, containing both combustion products and volatile agents extracted from the tobacco, produces lung tumors in strain A mice. In human smokers, the reduction of PAH and TSN in cigarettes were not necessarily as effective in lowering lung cancer rates as was originally hoped for. The evaluation of cytotoxic, mutagenic, and carcinogenic properties of cigarette mainstream and sidestream smoke or of potentially less active cigarettes are often done with cigarette smoke condensate (Eclipse Expert Panel, 2000; IARC 2004; Stratton et al., 2001). The present investigation suggests that additional evaluations should include the gas phase of tobacco smoke.
ACKNOWLEDGMENTS
I would like to thank Imelda Espiritu, Dale Uyeminami, and Man Ly for their most competent assistance in conducting these experiments. Excellent animal care was provided by Shanie McCarty and Loreli Coleman. The advice of Dr. O. G. Raabe and Dr. J. A. Last in dosimetry calculations is appreciated and so is the help provided by Dr. R. D. Cardiff for the classification of lung tumors. This publication was made possible through grants No. ES05707 and CA96217 from the National Institutes of Health. Its contents are solely the responsibility of the author and do not necessarily represent the official views of the NIH.
REFERENCES
Abdo, K. M., Eustis, S. L., McDonald, M., Jokinen, M. P., Adkins, B. J., and Haseman, J. K. (1992). Naphthalene: A resipratory tract toxicant and carcinogen for mice. Inhal. Toxicol. 4, 393–409.
Adkins, B. J., Van Stee, E. W., Simmons, J. E., and Eustis, S. L. (1986). Oncogenic response of strain A/J mice to inhaled chemicals. J. Toxicol. Environ. Health 17, 311–322.
Dahl, A. R., and Henderson, R. F. (2000). Comparative metabolism of low concentrations of butadiene and its monoepoxide in human and monkey hepatic microsomes. Inhal. Toxicol. 12, 439–451.
De Flora, S., Bagnasco, M., and Vainio, H. (1999). Modulation of genotoxic and related effects by carotenoids and vitamin A in experimental models: Mechanistic issues. Mutagenesis 14, 153–172.
De Flora, S., Izzotti, A., D' Agostini, F., and Balansky, R. (2001). Mechanisms of N-acetylcysteine in the prevention of DNA damage and cancer, with special reference to smoking- and oxidative stress-related end-points. Carcinogenesis 22, 999–1013.
De Flora, S., D'Agostini, F., Balansky, R., Camoirano, A., Bennicelli, C., Bagnasco, M., Cartiglia, C., Tampa, E., Longobardi, M. G., Lubet, R. A., and Izzotti, A. (2003). Modulation of cigarette smoke-related end-points in mutagenesis and carcinogenesis. Mutat. Res. 523–524, 237–252.
Doll, R. (1998). Uncovering the effects of smoking: Historical perspective. Stat. Methods Med. Res. 7, 87–117.
Druckrey, H. (1961). Experimental investigations on the possible carcinogenic effects of tobacco smoking. Acta Med. Scand. (Suppl. 369), 24–41.
Dunnick, J. K., Elwell, M. R., Radovsky, A. E., Benson, J. M., Hahn, F. F., Nikula, K. J., Barr, E. B., and Hobbs, C. H. (1995). Comparative carcinogenic effects of nickel subsulfide, nickel oxide, or nickel sulfate hexahydrate chronic exposures in the lung. Cancer Res. 55, 5251–5256.
Eclipse Expert Panel (2000). A safer cigarette A comparative study. A consensus report. Inhal. Toxicol. 12(Suppl. 5), 1–48.
Estensen, R. D., Jordan, M. M., Wiedmann, T. S., Galbraith, A. R., Steele, V. E., and Wattenberg, L. W. (2004). Effect of chemopreventive agents on separate stages of progression of benzo[alpha]pyrene induced lung tumors in A/J mice. Carcinogenesis 25, 197–201.
Farris, G. M., Everitt, J. I., Irons, R. D., and Popp, J. A. (1993). Carcinogenicity of inhaled benzene in CBA mice. Fundam. Appl. Toxicol. 20, 503–507.
Hecht, S. S. (1999). Tobacco smoke carcinogens and lung cancer. J. Natl. Cancer Inst. 91, 1194–1210.
Henderson, R. (2001). Species differences in the metabolism of olefins: Implications for risk assessment. Chem. Biol. Interact. 135–135, 53–64.
Hoffmann, D., and Wynder, E. L. (1999). Active and passive smoking. In Toxicology (H. Marquardt, S. G. Schaefer, R. McClellan, and F. Welsch, Eds.), pp. 879–898. Academic Press, New York.
IARC (International Agency for Research on Cancer). (1998). IARC Handbook on Cancer Prevention. Carotenoids. IARC, Lyon, France.
IARC (International Agency for Research on Cancer). (1995). IARC Monographs on the Evaluation of Carcinogenic Risk to Humans, Volume 63: Dry Cleaning, Some Chlorinated Solvents and Other Industrial Chemicals. IARC, Lyon, France.
IARC (International Agency for Research on Cancer). (1999). IARC Monographs on the Evaluation of Carcinogenic Risk to Humans, Volume 71: Re-evaluation of Some Organic Chemicals, Hydrazine and Hydrogen Peroxide. IARC, Lyon, France.
IARC (International Agency for Research on Cancer) (2004). IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans, Volume 83: Tobacco Smoke and Involuntary Smoking. IARC, Lyon, France.
Kerns, W. D., Pavkov, K. L., Donofrio, D. J., Gralla, E. J., and Swenberg, J. A. (1983). Carcinogenicity of formaldehyde in rats and mice after long-term inhalation exposure. Cancer Res. 43, 4382–4392.
Lee, P. N., and Sanders E. (2004). Does increased cigarette consumption nullify any reduction in lung cancer risk associated with low-tar filter cigarettes Inhal. Toxicol. 16, 817–833.
Leuchtenberger,C., and Leuchtenberger, R. (1971) Einfluss von frischem Zigarettenrauch auf die Entwicklung von Lungentumoren und auf Lungenkulturen bei der Snell-Maus. Schweiz. Med Wochenschr. 101, 1374–1381.
Leuchtenberger, C., and Leuchtenberger, R. (1974). Differential response of Snell's and C57 black mice to chronic inhalation of cigarette smoke. Pulmonary carcinogenesis and vascular alterations in lung and heart. Oncology 29, 122–138.
Melnick, R. L., Huff, J., Chou, B. J., and Miller, R. A. (1990). Carcinogenicity of 1,3-butadiene in C57BL/6 x C3H F1 mice at low exposure concentrations. Cancer Res. 50, 6592–6599.
NCI (2001). Smoking and Tobacco Control Monographs. No. 13: Risks Associated with Smoking Cigarettes with Low Tar Machine Measured Yield of Tar and Nicotine. US National Institutes of Health, Washington, DC.
Oberdorster, G., Cherian, M. G., and Baggs, R. B. (1994). Importance of species differences in experimental pulmonary carcinogenicity of inhaled cadmium for extrapolation to humans. Toxicol. Lett. 72, 339–343.
Obermueller-Jevic, U. C., Espiritu, I., Corbacho, A. M., Cross, C. E., and Witschi, H. (2002). Lung tumor development in mice exposed to tobacco smoke and fed beta-carotene diets. Toxicol. Sci. 69, 23–29.
Omenn, G. S. (1998). Chemoprevention of lung cancer: The rise and demise of beta-carotene. Annu. Rev. Public Health 19, 73–99.
Penn, A., Keller, K., Snyder, C., Nadas, A., and Chen, L. Ch. (1996). The tar fraction of cigarette smoke does not promote arteriosclerotic plaque development. Environ. Health Persp. 104, 1108–1113.
Penn, A., and Snyder, C. A. (1996). 1,3 Butadiene, a vapor phase component of environmental tobacco smoke, accelerates arteriosclerotic plaque development. Circulation 93, 552–557.
Rubin, H. (2001). Synergistic mechanisms in carcinogenesis by polycyclic aromatic hydrocarbons and by tobacco smoke: A bio-historical perspective with updates. Carcinogenesis 22, 1903–1930.
Stoner, G. D., and Shimkin, M. B. (1985). Lung tumors in strain A mice as a bioassay for carcinogenicity. In Handbook of Carcinogen Testing (H. Milman and E. K. Weisburger, Eds.), pp. 179–214. Noyes Publications, Park Ridge, NJ.
Stratton, K., Shetty, P., Wallace, R., and Bondurant, S. (2001). Clearing the Smoke. Assessing the Science Base for Tobacco Harm Reduction. National Academy Press, Washington, D.C.
van Zandwijk, N., Dalesio, O., Pastorino, U., De Vries, N., and van Tinteren, H. (2000). EUROSCAN, a randomized trial of vitamin A and N-acetylcysteine in patients with head and neck cancer or lung cancer. J. Natl. Cancer Inst. 92, 977–986.
Waalkes, M. P., and Rehm, S. (1994). Chronic toxic and carcinogenic effects of cadmium chloride in male DBA/2NCr and NFS/NCr mice: Strain-dependent association with tumors of the hematopoietic system, injection site, liver, and lung. Fundam. Appl. Toxicol. 23, 21–31.
Witschi, H. P., Espiritu, I., Peake, J. L., Wu, K., Maronpot, R. R., and Pinkerton, K. E. (1997a). The carcinogenicity of environmental tobacco smoke. Carcinogenesis 18, 575–586.
Witschi, H. P., Espiritu, I., Maronpot, R. R., Pinkerton, K. E., and Jones, A. D. (1997b). The carcinogenic potential of the gas phase of environmental tobacco smoke. Carcinogenesis 18, 2035–2042.
Witschi, H., Espiritu, I., Yu, M., and Willits, N. H. (1998). The effects of phenethyl isothiocyanate, N-acetylcysteine and green tea on tobacco smoke-induced lung tumors in strain A/J mice. Carcinogenesis 19, 1789–1794.
Woutersen, R. A., Appelman, L. M., Garderen-Hoetmer, A., and Feron, V. J. (1986). Inhalation toxicity of acetaldehyde in rats. III. Carcinogenicity study. Toxicology 41, 213–231.(Hanspeter Witschi)
ABSTRACT
Male strain A/J mice were exposed for six hours a day, five days a week for six months to either full tobacco smoke or to tobacco smoke drawn through a HEPA filter that removed more than 99% of particulate matter. After another four months in air, the animals were sacrificed and lung tumors were counted for calculation of multiplicities and incidences. Analysis of the chamber atmospheres showed that in the filtered smoke the concentrations of polycyclic aromatic hydrocarbons and tobacco smoke specific nitrosamines were reduced to from below 18% to even nondetectable levels of the original values measured in the unfiltered smoke. Aldehydes and other volatile organic compounds such as 1,3-butadiene, benzene, or acrolein were reduced to about 50 to 90% of the concentrations found in unfiltered smoke. Some potentially carcinogenic metals reached levels in filtered smoke ranging from 77% to less than 1% found in full smoke. The mice exposed to the filtered smoke atmosphere had practically identical lung tumor multiplicities and incidence as had the animals exposed to full smoke, significantly higher than in air exposed controls. Diets containing 0.5% -carotene or 0.4% N-acetylcysteine afforded some chemoprevention. It was tentatively concluded that 1,3-butadiene might be an important contributor to lung tumorigenesis in this mouse model of tobacco smoke carcinogenesis.
Key Words: tobacco smoke; mice; strain A/J mice; lung tumors; 1,3-butadiene.
INTRODUCTION
The first evidence that inhalation of tobacco smoke caused lung cancer was obtained in human studies, during the 1930's and 1940's in Germany and in the 1950's in Great Britain and the U.S. (Doll, 1998). Evidence for the carcinogenicity of tobacco smoke in experimental animals was more difficult to obtain. Only a few inhalation studies were conclusive, the most convincing ones being large experiments with hamsters where tobacco smoke produced laryngeal tumors, but no tumors in the lower respiratory tract (IARC, 2004). On the other hand, skin painting studies in mice with tobacco smoke condensate produced early on evidence that the "tar" or particulate phase of tobacco smoke was carcinogenic. From then on, it was generally assumed that the carcinogenic potential of tobacco smoke was mostly, if not exclusively, associated with the particulate phase and its polycyclic aromatic hydrocarbons and nitrosamines.
Interestingly, some dissenting opinions were aired already in 1961: ‘Nearly all investigations on the tobacco problem have been started with the assumption that 1) the carcinogenic agents are produced only in the process of smoking and 2) that higher aromatic hydrocarbons, especially benzpyrene compounds, are the main causative substances. This however, seems to be a prejudice’ (Druckrey, 1961). Nevertheless, during the ensuing years considerable efforts were made to mitigate adverse health effects caused by smoking. It was anticipated that the development of low tar–low nicotine filter cigarettes would have a beneficial effect. Unfortunately, this does not seem to have been the case. Changes in cigarette design and manufacturing failed to benefit public health and no evidence of reduced risk could be found (NCI, 2001). On the other hand, Lee and Sanders (2004) concluded, after an extensive analysis of the available epidemiological literature, that the smoking of filter and low tar cigarettes carried 23 to 36% less risk than the smoking of unfiltered ones. They still considered the risk to develop lung cancer as being high.
While thus the putative role in carcinogenesis of the tobacco smoke tar phase, particularly of its polycyclic aromatic hydrocarbons and tobacco specific nitrosamines, was widely studied, much less attention was paid to the gas phase. The gas phase of cigarette smoke contains some 400 to 500 compounds (Hoffmann and Wynder, 1999). The bulk of gases are oxygen, nitrogen, carbon dioxide, and carbon monoxide. Some well-known carcinogens such as benzene, formaldehyde, acetaldehyde, and 1,3-butadiene are also present. Initial attempts to establish the carcinogenic potential of the gas phase were negative in mice; in hamsters, where full smoke produces laryngeal tumors, no such lesions were observed after gas phase inhalation only (IARC, 2004). However, in 1974 studies in Snell's mice showed unequivocally that tobacco smoke gas phase was as active in producing benign and malignant lung tumors as was full tobacco smoke (Leuchtenberger and Leuchtenberger, 1971, 1974). This was later confirmed in strain A mice where tumor development was correlated with a partial analysis of the chamber atmospheres for polycyclic aromatic hydrocarbons and nitrosamines (Witschi et al., 1997b). In the present experiments, we reproduced the previous findings and expanded analysis of chamber atmospheres. Analysis covered 100% of the gas phase carcinogens listed by Hoffmann and Wynder (1999), 70% of the carcinogenic inorganic compounds, 80% of carcinogenic polycyclic aromatic hydrocarbons present in more than 5 ng quantity per cigarette and four of the most important tobacco specific nitrosamines. In addition, we reexamined the effects of two putative chemopreventive agents on the carcinogenicity of cigarette smoke gas phase.
MATERIALS AND METHODS
Materials. Kentucky 2R4F reference cigarettes were obtained from the Tobacco Research Institute, University of Kentucky (Lexington, KY). They deliver on average 0.81 mg of nicotine and 11.6 mg of particulate material per cigarette. Diets AIN-93G and AIN-93M were purchased from DYETS (Bethlehem, PA). The -carotene diet with gelatine beadlets containing 10% -carotene was premixed by the supplier. Diets containing N-acetylcysteine (Sigma, St. Louis, MO) were prepared weekly by mixing in a Hobart blender. Test diets were stored at 4°C until used.
Animals. Male strain A/J mice, 5 to 6 weeks old were purchased from Jackson Laboratories, Bar Harbor, ME. After arrival, a few animals were selected at random for a standard rodent health surveillance screen by the Comparative Pathology Laboratory, UC Davis. No evidence for infectious disease (pathogenic agents) or presence of parasites or ova in pelage and cecum were reported. Due to the absence of significant macroscopic lesions in all animal examined, no histopathology was processed. Serology was negative for mouse hepatitis virus, Sendai virus, Reovirus type 3, pneumonia virus, parvo, ectromelia, and Mycoplasma pulmonis. After a two-week acclimatization period, the mice were assigned at random to the different treatment groups. They were housed in accordance with institutional guidelines four to a cage, in polypropylene cages with tightly fitting wire screen lids on a paper-based bedding material (Tek-fresh, Harland). They were fed diet AIN-93G until five months old and from then on diet AIN-93 M. At all times during the experiment, including during smoke exposure, water and food were freely accessible. The animals were observed daily and weights were recorded weekly.
Exposure system. The tobacco smoke exposure system was the one used in our previous studies on the effects of full smoke and filtered smoke (Witschi et al., 1997a,b). Briefly, two Hinners-type stainless steel inhalation chambers (volume 0.44 m3) were used. In chamber 1, mice were exposed to a mixture of tobacco smoke (TS) generated from 89% cigarette sidestream smoke and 11% mainstream smoke. In chamber 2, the TS was drawn before entering the chamber through a HEPA filter that removed >99.8% of particulate material.
Experimental design. All protocols had been approved by the UC Davis Animal Use and Care Administrative Advisory Committee. Animals were exposed to full or filtered tobacco smoke 6 h a day, five days a week, for a total of 25 weeks. Animals in the full smoke chamber were kept on AIN-93G/M diet throughout. In the filtered smoke chamber, animals were divided into three groups: Ain-93G/M diet alone, AIN93-G/M diet containing 0.5% -carotene and AIN-93G/M diet containing 0.4% N-acetylcysteine (NAC). Control animals, fed the same diets, were housed in inhalation chambers ventilated with filtered air. After 25 weeks, all animals were removed into a conventional animal facility and killed 18 weeks later.
Analysis of chamber atmospheres. Total suspended particulates, CO, and nicotine were measured in both chambers during the entire TS exposure period. CO concentrations were monitored every 30 min with a model 880 non-dispersive-infrared (NDIR) analyzer and total suspended particulate mater (TSP) was measured three times a day in the full smoke chamber and once a day in the filtered smoke chamber by weighing material collected on Pallflex filters. Nicotine was measured once a week by drawing 10 l of air through XAD4 cartridges and subsequent analysis of the extraction liquid with gas chromatography/thermionic selective detector.
For additional measurements of selected tobacco smoke constituents, chamber air samples were collected on five different time points, spaced approximately one month apart from each other, and sent for analysis to commercial laboratories. For determination of metals, 8 l of air were collected on a 25 mm (3 μm) TEFLO filter (Pall Corp) and analyzed by XRF (X-ray Fluorescence) at Chester Labnet (Tigard, OR). Polycyclic aromatic hydrocarbons (PAH) were measured by collecting a total of 60 l of air on 2 XAD-40 cartridges and subsequent analysis by gas chromatography/mass spectrometry (GC/MS) at Alta Analytical Laboratory Inc. (El Dorado Hills, CA). For the determination of acrolein, benzene, and 1,3-butadiene, air samples were collected into a 6 l SUMMA canister and analyzed by GC/MS; for acetaldehyde and formaldehyde determination 10 l of air were collected on SEPPAK LPDNPH cartridges and analyzed by HPL at Air Toxics (Folsom, CA). For the determination of tobacco specific nitrosamines, 120 l of air were collected on Cambridge filter pads, extracted with 10 ml of 100 mM ammonium acetate and then analyzed for selected tobacco smoke-specific nitrosamines by LC/MS at Arista Laboratories (Richmond, VA).
Tissue preparation. After killing the animals by pentobarbital overdose, the lungs were manually expanded to total lung capacity through instillation of Tellyesniczky's fluid. After at least 24 h of fixation, the number of tumor nodules visible on the lung surface was counted under a magnifying glass. The results were expressed as tumor incidence, i.e., percentage of animals with one or several lung tumors and as tumor multiplicity. Tumor multiplicity was calculated as average number of tumors per lung of all animals, including nontumor bearing ones (Stoner and Shimkin, 1985). For histopathological evaluations, lung lobes containing tumors were dehydrated, embedded in paraffin, sectioned at 5 μm and stained with hematoxylin and eosin (H&E).
Statistical analysis. All numerical data were calculated as mean and SD or SEM. Comparisons of body weights and of tumor multiplicity between tobacco smoke-exposed and air-exposed controls were made by ANOVA followed by Tukey-Kramer's or Dunnetts post test. Tumor incidences were compared using Fisher's exact test. A p value of 0.05 or less was considered to be significant.
RESULTS
Exposure conditions for the two chambers are given in Table 1. The insertion of a HEPA filter between chamber 1 and chamber 2 effectively lowered the concentrations of total suspended particulate matter to 0.12% of chamber 1 and of nicotine to 16%, whereas CO was only reduced to 86%. In both chambers, the animals tolerated the tobacco smoke exposure well and only one death was observed during the exposure period. During the exposure to filtered and unfiltered smoke, the animals failed to gain weight as rapidly as did control animals kept in air (Fig. 1). After the six months exposure, the body weights of the animals kept in full smoke was significantly lower than those of all other groups, indicating that inhalation of the gas phase appeared to be less stressful than exposure to full smoke, although it still reduced body weight gain to a significant extent in comparison with controls kept in air. At the end of the experiment, the weights of the animals exposed to filtered or unfiltered tobacco smoke were not significantly different from the weight of control animals kept in air throughout.
Tables 2 and 3 show the results of analysis of the chamber atmospheres for several of the known carcinogens in tobacco smoke. Filtering the smoke resulted in a reduction of several polycyclic aromatic hydrocarbons, among them benzo(a)pyrene, to below detection levels. The concentrations of others were reduced to less than 10%, with the exception of naphthalene and methylnaphthalene which were reduced to only 69% resp. 51% of the full smoke atmosphere. Tobacco specific nitrosamines declined to 18% for 4-(methylnitrosamino)-1-(pyridyl)-1-butanone (NNK) and to 5% for N-nitrosonornicotine (NNN), with N-nitrosoanabasine (NAB) and N-nitrosoanatabine (NAT) becoming undetectable. Concentrations of the volatile organic compounds decreased much less and fell only to between 50 and 90% of the values measured in unfiltered smoke. The exception was formaldehyde where a bigger reduction was seen (to 17%). Metal concentrations were in general lower in filtered smoke compared to full smoke, but also were quite variable in the unfiltered smoke chamber where, with the exception of Cd and Ni, they were not detected in all samples.
Lung tumor data are shown in Table 4. In animals exposed to air and fed control diet or diets with -carotene or NAC, lung tumor multiplicities ranged from 0.9 to 1.1. Full tobacco smoke produced a significant increase in both lung tumor multiplicities and lung tumor incidence. Filtered smoke had exactly the same effect: lung tumor multiplicities and incidences were practically identical to the ones found in animals exposed to the full tobacco smoke. Ingestion of the -carotene and NAC diets, during and after smoke exposure, significantly reduced lung tumor multiplicities and lung tumor incidences, although not quite to the levels seen in animals kept in air.
From each group, 50 to 70% of the grossly visible tumors were sectioned and examined under the microscope. The tumors were classified into beingn tumors (adenomas) and malignant tumors (adenomas with carcinomatous foci or frank adenocarcinomas). Adenomas had a solid growth pattern with alveolar spaces being filled by contiguous neoplastic cells. Carcinomatous foci arising within adenomas often showed a papillary growth pattern and displayed a substantially higher percentage of abnormal mitotic figures. On average, 20 to 30% of all tumors showed features of malignancy (Table 5). The difference between the lowest rate found in the animals exposed to full smoke (9%) and the highest rate seen in the animals exposed to filtered smoke and fed -carotene (33%) was statistically not significant.
DISCUSSION
The present results confirm earlier findings (Leuchtenberger and Leuchtenberger, 1971, 1974; Witschi et al., 1997b). Although there were substantial differences in the concentrations of several carcinogens in the chamber atmospheres before and after HEPA filtration of cigarette smoke (a mixture of 11% cigarette mainstream and 89% sidestream smoke), tumor response was not affected. This might have some implications for experimental design. Most studies demonstrating enhanced lung tumor development by tobacco smoke and underlying mechanisms have been conducted by whole-body exposure of mice (De Flora et al., 2003). It has on occasion been suggested that under such conditions smoke particles deposit on the fur of the animals and carcinogens are absorbed orally through grooming rather than by inhalation. Since the gas phase has as much carcinogenic activity as has whole smoke, oral exposure to carcinogens seems unlikely.
Histopathological evaluation of more than 50% of all the grossly visible tumors revealed in about one-fourth signs of malignancy: carcinomatous foci arising within an adenoma or malignant features through the entire tumor, as described in detail before (Witschi et al., 1997a,b). The tumor type distribution was also similar to previous findings where control animals had 20 to 23% and tobacco smoke exposed animals 13 to 17% malignant tumors There was no indication that the chemoprevention regimens would result in a shift in tumor type. Such shifts are usually seen in studies that examine progression of lesions in the presence of chemopreventive agents (Estensen et al., 2004), an experimental design not adopted in the current study.
Our observations raise the question what agents are responsible for tumor development in the lungs of A/J mice. The general consensus appears to be that, at least in man, polycyclic aromatic hydrocarbons, particularly benzo(a)pyrene, and tobacco smoke specific nitrosamines are the main causative agents of lung cancer (Hecht, 1999). By inference, it is thought that this applies to animals, too. Our data show that, with the exception of naphthalene and methylnaphthalene, many polycyclic aromatic hydrocarbons were no longer detectable in the filtered smoke. Others were reduced to concentrations of less than 10% of the full smoke atmosphere. This makes it unlikely that PAHs were major contributors. Concentrations of NNK and NNN also fell below 20%. We have previously discussed how dosimetric calculations do not allow us to assume that NNK is a major carcinogen in strain A/J mice exposed to tobacco smoke (Witschi et al., 1997b).
Potentially carcinogenic volatile organic compounds (IARC classes 1, 2a, and 2b) were reduced much less in the filtered smoke. Data on the pulmonary carcinogenicity in mouse lungs of some of these agents are available and may be compared to our exposure concentrations as measured in filtered smoke. When strain A/J mice inhaled for 20 weeks an atmosphere of 157 mg/m3 of naphthalene (IARC 2B), only 11% of the mice developed lung tumors and multiplicities were 0.4 (Adkins et al., 1986). In a two-year study with B6C3F1 mice, the same concentration of naphthalene produced only a weak tumorigenic response (Abdo et al., 1992). The level of naphthalene measured in our study (0.177 mg/m3) was thus far below the one expected to produce tumors. Similar considerations apply to some of the other potentially carcinogenic volatiles (IARC class 1 and 2) such as acetaldehyde, formaldehyde, and benzene. After a 16-week exposure of CBA mice to an airborne benzene concentration of 957 mg/m3, 36% developed 18 months later pulmonary adenomas (Farris et al., 1993). After long term exposure of mice to 17.5 mg/m3 to formaldehyde, only nasal tumors were found (Kerns et al., 1983). No tumor data are available for mice following inhalation of acetaldehyde. In rats, exposure for 28 months to 1350 mg/m3 caused only nasal tumors (Woutersen et al., 1986). The airborne concentrations of these agents measured in our experiments did not nearly approximate such values. As far as acrolein is concenred (IARC class 3), oral administration to rats or mice failed to produce tumors (IARC, 1995). Considering metals, inhaled Cd is not a pulmonary carcinogen in mice (Oberdorster et al., 1994) and oral or subcutaneous exposure to Cd compounds have given conflicting results (Waalkes and Rehm, 1994). No Cr was detected in filtered smoke and concentrations of Ni were much lower than had been in positive inhalation assays with several nickel compounds in mice (0.3 mg/m3 vs. 2.5 mg/m3; Dunnick et al., 1995). While the concentrations of single agents in the unfiltered and filtered smoke cannot account for the overall tumor response, it is of course entirely possible that additive, synergistic, or tumor-promoting interactions between chemicals could have substantially enhanced tumorigenicity.
There is one agent that by itself might account for the carcinogenicity of full tobacco smoke or filtered tobacco smoke in strain A/J mice: 1,3-butadiene. A two-year inhalation study in C57Bl/6 x C3H F1 mice found increased neoplasms of the lung after exposure to 1,3-butadiene at concentrations as low as 13.8 mg/m3 (6.25 ppm) (Melnick et al., 1990). Strain A/J mice are more prone to develop chemically induced alveolar tumors than other strains and this probably also applies to 1,3-butadiene. The two exposure concentrations that yielded a tumorigenic response in our experiment are reasonably close to each other (3.6 mg/m3 for full smoke or 2.9 mg/m3 for filtered smoke) and not much lower than 13.8 mg/m3.1,3-Butadiene itself has been shown to inhibit its own metabolism to the highly active and carcinogenic diepoxide (Dahl and Henderson, 2000). It is conceivable that other organic volatiles might interfere with 1,3-butadiene metabolism. Partial removal of them by the filter might thus enhance butadiene activation and carcinogenesis. Since mice are much more sensitive to the toxic and carcinogenic action of 1,3-butadiene than are rats and even primates (Henderson, 2001), it is difficult to extrapolate lung tumor data induced by cigarette smoke from mouse to man based on 1,3-butadiene. Furthermore there is no evidence that 1,3-butadiene produces lung cancer in occupationally exposed humans (IARC, 1999). Nevertheless, it is interesting to note that 1,3-butadiene present in tobacco smoke has been shown to enhance the development of arteriosclerotic plaques in animal models of arteriosclerosis, whereas the tar phase does not seem to have such an effect (Penn et al., 1996; Penn and Snyder, 1996).
-Carotene and NAC are two agents who, among other activities, have strong antioxidant properties. In many experimental systems they have chemopreventive action and are capable to significantly reduce tumor development. Multiple mechanistic studies with biomarkers of effect have suggested that both agents should provide chemoprevention against tobacco smoke (De Flora et al., 1999, 2001). Experiments to show this have not been fully successful. In strain A mice, -carotene and NAC had no chemopreventive action against full cigarette smoke (Obermueller-Jevic et al., 2002; Witschi et al., 1998). In SWR mice, a small, but statistically not significant effect of NAC was observed (de Flora et al., 2003). The protective effect of both agents against the gas phase was thus unexpected. It would seem reasonable to assume that the oxidative stress placed onto the lung by filtered tobacco smoke is as intense as it would be with full smoke. If the main carcinogenic stimulus in filtered smoke was oxidative stress, a plausible explanation for the protective effects of the two antioxidants could be postulated. However, this makes it difficult to explain why neither -carotene nor NAC were effective (in strain A mice) against full tobacco smoke. It has been discussed before that two main classes of carcinogens found in full smoke, polycyclic aromatic hydrocarbons and nitrosamines, might not be as important contributions to carcinogenesis as is generally assumed. This allows the speculation that the full smoke contains some other as of yet unidentified carcinogen(s) that could overcome any beneficial effects of -carotene or NAC. It also has been postulated that polycyclic aromatic hydrocarbons found in tobacco tar act mostly as promoting agents (Rubin, 2001). Such processes could counteract or even nullify critical protective mechanisms of -carotene and NAC against full smoke. It is also conceivable that some elements of full tobacco smoke may inactivate or interfere with the bioavailability of the two chemopreventive agents. Initial studies with cell culture systems will allow to test these different hypotheses.
As surprising as they are, the present findings probably will not have much practical impact on chemoprevention of lung cancer in man, given the disappointing results of major clinical trials (Omenn, 1998; van Zandwijk et al., 2000). On the other hand, both -carotene and NAC have been found to mitigate biomarkers of exposure and possibly of effect in phase I and phase III clinical trials (de Flora et al., 2003; IARC, 1998). Given the right circumstances, -carotene and NAC may yet have some beneficial effects. Our observations may lead to future experiments where mechanistic events that relate to chemoprevention can be separated from events that are inconsequential.
In conclusion, experimental evidence shows that the gas phase of tobacco smoke, containing both combustion products and volatile agents extracted from the tobacco, produces lung tumors in strain A mice. In human smokers, the reduction of PAH and TSN in cigarettes were not necessarily as effective in lowering lung cancer rates as was originally hoped for. The evaluation of cytotoxic, mutagenic, and carcinogenic properties of cigarette mainstream and sidestream smoke or of potentially less active cigarettes are often done with cigarette smoke condensate (Eclipse Expert Panel, 2000; IARC 2004; Stratton et al., 2001). The present investigation suggests that additional evaluations should include the gas phase of tobacco smoke.
ACKNOWLEDGMENTS
I would like to thank Imelda Espiritu, Dale Uyeminami, and Man Ly for their most competent assistance in conducting these experiments. Excellent animal care was provided by Shanie McCarty and Loreli Coleman. The advice of Dr. O. G. Raabe and Dr. J. A. Last in dosimetry calculations is appreciated and so is the help provided by Dr. R. D. Cardiff for the classification of lung tumors. This publication was made possible through grants No. ES05707 and CA96217 from the National Institutes of Health. Its contents are solely the responsibility of the author and do not necessarily represent the official views of the NIH.
REFERENCES
Abdo, K. M., Eustis, S. L., McDonald, M., Jokinen, M. P., Adkins, B. J., and Haseman, J. K. (1992). Naphthalene: A resipratory tract toxicant and carcinogen for mice. Inhal. Toxicol. 4, 393–409.
Adkins, B. J., Van Stee, E. W., Simmons, J. E., and Eustis, S. L. (1986). Oncogenic response of strain A/J mice to inhaled chemicals. J. Toxicol. Environ. Health 17, 311–322.
Dahl, A. R., and Henderson, R. F. (2000). Comparative metabolism of low concentrations of butadiene and its monoepoxide in human and monkey hepatic microsomes. Inhal. Toxicol. 12, 439–451.
De Flora, S., Bagnasco, M., and Vainio, H. (1999). Modulation of genotoxic and related effects by carotenoids and vitamin A in experimental models: Mechanistic issues. Mutagenesis 14, 153–172.
De Flora, S., Izzotti, A., D' Agostini, F., and Balansky, R. (2001). Mechanisms of N-acetylcysteine in the prevention of DNA damage and cancer, with special reference to smoking- and oxidative stress-related end-points. Carcinogenesis 22, 999–1013.
De Flora, S., D'Agostini, F., Balansky, R., Camoirano, A., Bennicelli, C., Bagnasco, M., Cartiglia, C., Tampa, E., Longobardi, M. G., Lubet, R. A., and Izzotti, A. (2003). Modulation of cigarette smoke-related end-points in mutagenesis and carcinogenesis. Mutat. Res. 523–524, 237–252.
Doll, R. (1998). Uncovering the effects of smoking: Historical perspective. Stat. Methods Med. Res. 7, 87–117.
Druckrey, H. (1961). Experimental investigations on the possible carcinogenic effects of tobacco smoking. Acta Med. Scand. (Suppl. 369), 24–41.
Dunnick, J. K., Elwell, M. R., Radovsky, A. E., Benson, J. M., Hahn, F. F., Nikula, K. J., Barr, E. B., and Hobbs, C. H. (1995). Comparative carcinogenic effects of nickel subsulfide, nickel oxide, or nickel sulfate hexahydrate chronic exposures in the lung. Cancer Res. 55, 5251–5256.
Eclipse Expert Panel (2000). A safer cigarette A comparative study. A consensus report. Inhal. Toxicol. 12(Suppl. 5), 1–48.
Estensen, R. D., Jordan, M. M., Wiedmann, T. S., Galbraith, A. R., Steele, V. E., and Wattenberg, L. W. (2004). Effect of chemopreventive agents on separate stages of progression of benzo[alpha]pyrene induced lung tumors in A/J mice. Carcinogenesis 25, 197–201.
Farris, G. M., Everitt, J. I., Irons, R. D., and Popp, J. A. (1993). Carcinogenicity of inhaled benzene in CBA mice. Fundam. Appl. Toxicol. 20, 503–507.
Hecht, S. S. (1999). Tobacco smoke carcinogens and lung cancer. J. Natl. Cancer Inst. 91, 1194–1210.
Henderson, R. (2001). Species differences in the metabolism of olefins: Implications for risk assessment. Chem. Biol. Interact. 135–135, 53–64.
Hoffmann, D., and Wynder, E. L. (1999). Active and passive smoking. In Toxicology (H. Marquardt, S. G. Schaefer, R. McClellan, and F. Welsch, Eds.), pp. 879–898. Academic Press, New York.
IARC (International Agency for Research on Cancer). (1998). IARC Handbook on Cancer Prevention. Carotenoids. IARC, Lyon, France.
IARC (International Agency for Research on Cancer). (1995). IARC Monographs on the Evaluation of Carcinogenic Risk to Humans, Volume 63: Dry Cleaning, Some Chlorinated Solvents and Other Industrial Chemicals. IARC, Lyon, France.
IARC (International Agency for Research on Cancer). (1999). IARC Monographs on the Evaluation of Carcinogenic Risk to Humans, Volume 71: Re-evaluation of Some Organic Chemicals, Hydrazine and Hydrogen Peroxide. IARC, Lyon, France.
IARC (International Agency for Research on Cancer) (2004). IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans, Volume 83: Tobacco Smoke and Involuntary Smoking. IARC, Lyon, France.
Kerns, W. D., Pavkov, K. L., Donofrio, D. J., Gralla, E. J., and Swenberg, J. A. (1983). Carcinogenicity of formaldehyde in rats and mice after long-term inhalation exposure. Cancer Res. 43, 4382–4392.
Lee, P. N., and Sanders E. (2004). Does increased cigarette consumption nullify any reduction in lung cancer risk associated with low-tar filter cigarettes Inhal. Toxicol. 16, 817–833.
Leuchtenberger,C., and Leuchtenberger, R. (1971) Einfluss von frischem Zigarettenrauch auf die Entwicklung von Lungentumoren und auf Lungenkulturen bei der Snell-Maus. Schweiz. Med Wochenschr. 101, 1374–1381.
Leuchtenberger, C., and Leuchtenberger, R. (1974). Differential response of Snell's and C57 black mice to chronic inhalation of cigarette smoke. Pulmonary carcinogenesis and vascular alterations in lung and heart. Oncology 29, 122–138.
Melnick, R. L., Huff, J., Chou, B. J., and Miller, R. A. (1990). Carcinogenicity of 1,3-butadiene in C57BL/6 x C3H F1 mice at low exposure concentrations. Cancer Res. 50, 6592–6599.
NCI (2001). Smoking and Tobacco Control Monographs. No. 13: Risks Associated with Smoking Cigarettes with Low Tar Machine Measured Yield of Tar and Nicotine. US National Institutes of Health, Washington, DC.
Oberdorster, G., Cherian, M. G., and Baggs, R. B. (1994). Importance of species differences in experimental pulmonary carcinogenicity of inhaled cadmium for extrapolation to humans. Toxicol. Lett. 72, 339–343.
Obermueller-Jevic, U. C., Espiritu, I., Corbacho, A. M., Cross, C. E., and Witschi, H. (2002). Lung tumor development in mice exposed to tobacco smoke and fed beta-carotene diets. Toxicol. Sci. 69, 23–29.
Omenn, G. S. (1998). Chemoprevention of lung cancer: The rise and demise of beta-carotene. Annu. Rev. Public Health 19, 73–99.
Penn, A., Keller, K., Snyder, C., Nadas, A., and Chen, L. Ch. (1996). The tar fraction of cigarette smoke does not promote arteriosclerotic plaque development. Environ. Health Persp. 104, 1108–1113.
Penn, A., and Snyder, C. A. (1996). 1,3 Butadiene, a vapor phase component of environmental tobacco smoke, accelerates arteriosclerotic plaque development. Circulation 93, 552–557.
Rubin, H. (2001). Synergistic mechanisms in carcinogenesis by polycyclic aromatic hydrocarbons and by tobacco smoke: A bio-historical perspective with updates. Carcinogenesis 22, 1903–1930.
Stoner, G. D., and Shimkin, M. B. (1985). Lung tumors in strain A mice as a bioassay for carcinogenicity. In Handbook of Carcinogen Testing (H. Milman and E. K. Weisburger, Eds.), pp. 179–214. Noyes Publications, Park Ridge, NJ.
Stratton, K., Shetty, P., Wallace, R., and Bondurant, S. (2001). Clearing the Smoke. Assessing the Science Base for Tobacco Harm Reduction. National Academy Press, Washington, D.C.
van Zandwijk, N., Dalesio, O., Pastorino, U., De Vries, N., and van Tinteren, H. (2000). EUROSCAN, a randomized trial of vitamin A and N-acetylcysteine in patients with head and neck cancer or lung cancer. J. Natl. Cancer Inst. 92, 977–986.
Waalkes, M. P., and Rehm, S. (1994). Chronic toxic and carcinogenic effects of cadmium chloride in male DBA/2NCr and NFS/NCr mice: Strain-dependent association with tumors of the hematopoietic system, injection site, liver, and lung. Fundam. Appl. Toxicol. 23, 21–31.
Witschi, H. P., Espiritu, I., Peake, J. L., Wu, K., Maronpot, R. R., and Pinkerton, K. E. (1997a). The carcinogenicity of environmental tobacco smoke. Carcinogenesis 18, 575–586.
Witschi, H. P., Espiritu, I., Maronpot, R. R., Pinkerton, K. E., and Jones, A. D. (1997b). The carcinogenic potential of the gas phase of environmental tobacco smoke. Carcinogenesis 18, 2035–2042.
Witschi, H., Espiritu, I., Yu, M., and Willits, N. H. (1998). The effects of phenethyl isothiocyanate, N-acetylcysteine and green tea on tobacco smoke-induced lung tumors in strain A/J mice. Carcinogenesis 19, 1789–1794.
Woutersen, R. A., Appelman, L. M., Garderen-Hoetmer, A., and Feron, V. J. (1986). Inhalation toxicity of acetaldehyde in rats. III. Carcinogenicity study. Toxicology 41, 213–231.(Hanspeter Witschi)