Tobacco Documents Online

it Iftrh. Envlron. Contam. Toxicol, 24, 498-503 (1993) A R C H I V E S O F Environmental Contamination a n d Toxicology O 1993 str~Vahr rie,r Yak tnc. Determination of Polycyclic Aromatic Hydrocarbons in the Lung Hiroshi Seto*, Tomoko Ohkubo*, Takako Kanoh*, Morio Koike**, Kyoji Nakamura***, and Yutaka Kawahara' *Deputment of Environtnental Health, The Tokyo Metropolitan Research Laboratory of Public Health, 3-24-1 Hyakunincho, Shinjuku-ku, 169 Tokyo, Japan; $*Deputment of Pathology, The Tokyo Meuopolitan Komagome Hospital, 113 Tokyo, Japan; ***Department of Pathology. The Tokyo MctropliWa Fuchu Hospitd, 183 Tokyo, Japan, and 'Depanment of Pathology, The Tokyo Metropolitan Hno-o Hospital. 150 Tokyo, Japan Abstract. Polycyclic aromatic hydrocarbons (PAHs) accumu- lated in human lung samples from men (n = 236) and women (n - 128) were determined by high-performance liquid chro- matography (HPLC) to examine their association with lung cancer. The mean values for benzo[a]pyrene (BaP), henzo- [k]Buoranthene (BkF), and benzo[g,h,i]perylene (BghiP) in lungs (ng/g dry lung) of Japanese autopsied patients were 0.54, 0.44, and 0.87, respectively. Tlu modal values were 0.3, 0.3 and 0.5, respectively. Each of the PAH concentrations was highly correlated with the others (r > 0.83). PAH concentra- tions in the lungs showed age-related increases with low corre- lation-coeffrcient values. BaP, BkF and BghiP concentrations in lungs of various subgroups were in the following order. male > female; and lung cancer > all cancers > non-cancer among male not female group. Only BghiP concentration in the lungs of the male smoker group is significantly higher (P t 0.10) than that of the male non-smoker group. Even among non-smoker groups, PAH concentrations in the lungs of male group were significantly higher than those of female group. In the male population, excess exposure to PAHs to- gether with fine carbon particles, such as tobacco smoke or diesel exhaust, correlated with increased prevalence of lung cancer. Tomingas et at. (1976) detected ng/g levels of BaP in human bronchial carcinoma tissues. However, their analytical method using thin-layer chromatography-Fluorophotometry is inade- quate in terms of selectivity and sensitivity for subjects exposed to low levels of PAHs. Furthermore, circumstances ofexposure to PAHs and cancer incidence are different from country to country. Thus, we attempted to examine the relationship be- tween lung cancer and residual PAHs in lung tissue of au- topsied patients in Tokyo, Japan, using high performance liquid chromatography (HPLC). Materials and Methods Subjects We utilized 364 lungs from autopsies of Japanese conducted at three hospitals of the Tokyo Metropolitan Government during 1986 to 1989. The sampling was done at random. For the determination of PAH content, a 50 g sample of fonnalin-in0ated and fixed lung was obtained from the lower lobe that was free of tumor. The diagnosis was con- firmed histologically in every case. Smoking histories were obtained by interview with the patient or relatives; personal histories ofoccupa- tion and residence were not obtained. Polycyclic aromatic hydrocarbons (PAHs) are considered to be a potential human health hazard. Benzo[a]pyrene (BaP), one of the most extensively studied carcinogenic PAHs (Phillips 1983), is found in diesel engine exhaust (Hayano et at. 1985; Tong et al. 1984) and in tobacco smoke (Appel et al. 1990). Human exposure to BaP (average daily intake) can be estimated at 17 ng from ambient air (Matsumoto and Kashimoto 1985) and 50 ng from food (Obana et a!. 1984), in Japan. It is important to know how much of this carcinogenic compound accumulates in human tissues, and whether it represents a prac- tical risk factor for cancer. It is noteworthy that inhaling PAHs with fine carbon particles is persistent, and is accumulated in the lung (Ohkubo et al. 1988; Sun et al. 1984). However, BaP administered orally is hardly accumulated in the body (Forth et al. 1988). Thus, the amount of PAHs in lung represents the amount of alveolar exposttre to PAHs, for example, through diesel soot and tobacco smoke. Chemicals Polycyclic aromatic hydrocarbon standards were obtained from com- mercial sources; BkF from RK Chemical Co. (Hanville, OH), BaP from Wako Pure Chemicals Ind. (Osaka, Japan) and BghiP from Ald- rich (Milwaukee, WI). Alumina and silica-gel were obtained from ~ Merck (Darmstadt, Germany); alumina, Merck art 1097, non acti- p vated, silica-gel, Merck art 7734, 130°C 4 h activation. The solvents ~ used were of either HPLC grade or "pesticide grade" from Wake Pure ~ Chemicals Ind. 4 ~ ~ Apparatus Analytical HPLC for quantitation was carried out with a Jasco (Tokyo, Japan) 880.PU pump, a Rheodyne (Berkeley, CA) 7125 injector, a GL W

Jdic Aromatic Hydmcarbons in the Lung Science (Tokyo, Japan) Inertsll ODS-2 eolumn (250 mrn x 4.6 mm I.D.), a Jasco 880-CO column oven (40°C) and a Hitachi (Tokyo, Japan) F-1000 fluorescence detoctor (excitation at 365 nm, emission at 430 nm). The flow rate of the mobile phase (75% (v/v) acetonitrile- water) was L3 mUmin, . Analytical Method We previously reported that fonnalin-fixed lung samples are available for detcnnination of residual PAHs (Seto et al. 1987). The analytical procedure was based on the reported HPLC method (Obana et al. 1981), with modifications as follows. A lung sample (50 g) was ho- mogenized in dhanol (200 ml) using an Ultra-Turrax TPI8/30S2 (Bre- isgau, Gcrmany) followed by the addition of 16.8 g of potassium hydroxide and refluxed for 1.5 h. Water (200 ml) was added, and the mixture was transferred to a funnel. Polycyclic aromatic hydrocarbons were extracted from the mixture with hexane (150 ml, 100 ml x 2). The hexane fractions wem combined and washed with water (200 ml X 3), then dried over anhydrous sodium sulfate and concentrated to 5 ml in a rotary evaporator. The solution was applied to a double- layered open glass column (tower, alumina 5 g, upper, silica-ge) 5 g, 300 mm X 15 mn Lb,), The column was eluted with SO nd of hexane then 100 mi of 15%(v/v) diethyl ethenhexanemixture. The latereluate was reduced in volume to I ml under a gentle atream of highly purified nitrogen and an aliquot (50 µl) of this solution was injected into the HPf.C column. The PAH content of the sample was calculated from a standard eurve preparcd with authentic PAH. The glassware was non- transparent (bown color). The recoveries of BaP, BkF and BghiP from lung samples averaged 87.8 t 5,4%, 8Z6 ~ 3. 1% and 93.7 m 4.3 (n = 4), respectivety, when 50 ng of BaP or BkF, or 250 ng of BghiP was added. The data presentcd in this study were uncorrected by the percent recovery. Analyses of aeven separate determinatiuns af BaP in the human lung on different days gave values of 0.19 *_ 0,03 ny/g wet lung, The detection limit of each PAH with this HPLC system was as follows: BaP t.0, BkF 2,0 and BghiP 5,0 pg at a signal-to-noise ratio of 3. An additional piece (I g) of lung was dried (105°C) to constant weight to obtain the wct:dry ratio and to allow expression of the results in terms of PAH (ng) pergram dry lung. Quality Control The quality control (QC) sample was specially prepared. The QC sample consisted of [reeze-dned fish meat and BaP-augmcnted diesel sont. Polycyclic uomntic hydrocarbons in the QC sample (I g) was measured according to the previously outlined procedure once every five lung samples. Avenged values (ng/g) of BaP, BkP and BghiP in the QC sample were 109 m 8.9, 31.f t 2.4, and 23.7 't 2,7, respec- tively.'Phe QC data were ch<eked by using an X- o x control chart. Periodical changes of the values were not observed during the inves- tigation, Statistical Analysis Values In the text and tables represent the mean ± standard deviation. Concentration data showed logarithmic distribution curves; the data were normalized by log transformation, Statistical significance was calculated by routincs available in the MULTI-TOKBI program pack- age (SRI lnc, Tokyo, Japan). 499 €' x i 1 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 AGE Fig. l. Age distribution of subjects. Dotted bar for males, open bar for females. Each bar covers the range of values from the number at the left foo[ to less than that at the right faot Results Age Distribution of Samples Figure l shows a histogram of the age distribution of samples- The average age t standard deviation of the mate group was 65.9 :L 12.8 (n = 236), and that of the female group was 67.3= 13.1(n=128), Distribution of PAH Content of Samples A typical chromatogmm of PAHs in human lung is shown in Figure 2. Benzo[a]pyrene and BghiP were completely sepa- rated from other peaks. Fluorescence spectrum of the peak obtained was similar to that of authentic compound (BkF, BaP and BghiP, data not shown). A histogram of BaP content in lung is shown in Figure 3. The histogram of BaP as well as BkF and BghiP is asymmetric. However, the data log-transformed gave normal distribution curves. The mean values for BaP, BkF and BghiP in lung were 0.54, 0.44 and 0.87 (ng/g dry lung), respectively. The modal values were 0.3, 0.3 and 0.5 (ng/g dry lung), respectively. Correlation between PAH Concentrations of Lung Samples Correlation coefficients among PAHs concentrations were cal- culated and the results are shown in Table 1: high coefficient N values were obtained (r> 0.83, P < 0.01). 0 ~ V Comparison of PAH Concentration between Male and y Fe,mafe Groups ~ ~ Each of BaP, BkF and BghiP concentrations in the lungs of the male group was significandy higher than that of the female group (P < 0.001), as shown in Figure 4.

500 H. Sete - Table 1. Correlation between polycyclic aromatic hydrocarbon ceu- centradons of lung samples r . E E n=3tKt r BaP' BkF° BghiP` r tn .[ Bemo[a]pyrene - 0.888* 0.898" Benzo[k]fluoranthene 0.888* - 0.833* Z W Benzo[g,h,ijpcrylene 0.898* 0.833* - r 2 'Benzo[a]pyrene W U Z W U tn W tz bBenzo[k]Ouoranthene `Benzo[g,h,i]perylene Values are comelatian coefficients Level of significance by Ztesq *P < 0.01 ~,,~ Y l„JJ~__ 0 5 10 15 20 TIME (mjn) Flg. 2. Typical chromatogram of PAHs in human lung. Column, Inensii ODS.2 (250 mm x 4.6 mm LD.); mobile phase, 75% (vlv) accloniirile-water, flow-rate, 1.3 ml/min; column oven temp., 40°C; delection, fluorescence emission at 430 nm when excited at 365 nm. Pcak 1, bcnzojkjfluoranlhcnr 2, bcnzo(ajpyrene; 3, benzo[g,h,ij perytene 2 0 - 0 ' '~ ' r 0 0.5 1.0 1.5 2.0 CONCENTRATION (ng/g dry lung) Flg. 3. Histogratn of BaP cantent in lung Effect of Smoking on PAH Concentration in Lung We obtained 246 personal data about the smoking habit of persons who died from 1986 to 1989. The rates of smoking (including cx-smoking) were: male 74% and female 24%. However, exposure to passive smoking could not be estimated. Only BghiP concentration in the lungs of the male smoker group is significantly higher (P C 0.10) than that of the male 2.5 rr ~ ~ 2.0 V -Y A I _V M F M F M F BaP BkF BghiP Fig. 4. Comparison of PAH concentrations in lungs of the male group (M) with those of the female group (F). Numerals beside the bars are percen4le values . , non-smoker group (Table 2). But no significant effect of smok- ing on PAH concentration in lung was observed among female group. Even among non-smoker groups, PAH concentrations in lung of male group were higher than those of female group (P < 0.01 for BaP and BkF, and P< 0.05 for BghiP). Correlation between PAH Concentration and Age N tl1 Correlation coefficients between PAH concentration and age of 40 persons from whom samples were obtained are shown in Table y 3, classified by sex and smoking history. Values of the correla- y fion coefficient were relatively low. Higher correlation coeffL 0* cienf values of the female group than the male group, and of the -~ non-smoker group than the smoker group were obtained. Rela- ~ tionship between age and BaP level of lung among nonsmokers classifed by sex is shown in Figure 5. No significance on the

4 I x Aromatic Hydrocartans in the Lung Tabb 2. Comparison of polycyclic aromatic hydrocarbon levels in human lung, classified according to sex and smoking habit 501 Item n age BaP' BkF° BghiP` Whole 364 663 ~ 12.8 0.54 ± 0.35 0.44µ0.26 0.87~0.56 Male 236 65.9 t 12,8 0.62 ± 0.38 0.49 ± 0.28 0,99 *- 0.62 Petnile 128 67.3 t 13.1 0.40 ± 0.22 0.36!0.21 0.66~0.35 Significmtce* - 0.001 0.001 0,001 Smoking n age BaP BkF BghiP Malc Smoker Non-smoker 122 42 66.2 ± 12.5 66.9 ~ 13.4 0.66 0.57 ± 0.41 ~ 0.32 0.51 ± 0.29 0.48 m 0.21 1.05 ± 0.64 0.87 :t 0.54 Significance - - - 0.t0 Female Smoker Nan•smoker 20 62 63.7 t 14,6 68.3 ~ 12.1 0.39 0.41 ± 0.23 ~ 0.24 0.35 ! 0.I8 0.39 z 0.27 0.67 x 0.38 0.67 '_ 0.36 Significance aBcnzolalpyrcnc °Benzo[k]Buoranthene Benzo{g,h,i]perylcne The data ere expressed as mean ± standard deviation. Unit of concentration is nglg dry lung. Significance, significance level of the difference between the means by Mast. The data of PAIls are analyzed a0er log transformation, -, Not significant. •Male to female group Table 3. Correlation between age and polycyclic aromatic hydrocar- bon conccnttation of lung n BaP' BkF° BghiP° - 1.D c ~ Whole 364 0.151 0.197 0.152 >` 0.8 Male 236 0.145 0.184 0.132 ~ F-emale 128 0,223 0.271 0.248 Smoker 142 0.187 0.161 0.150 Non-smokcr 104 0.211 0.241 0.247 •Benzola]pyrcne "Benzo[k]Buoranthcne ' Benzolg,h,ijpcrylene Values ue correlatton coefficients effect of age to lung BoP level is obtained by analysis of variance. In the case of other PAIls such as BkF or BghiP, similar results were obtained. Comparisons of PAFI Concentrations in Lungs among Cancer, Lung Cancer and Non-Cancer Groups The samples were divided into all cancers, lung cancer and non-eancer groups after pathological determination and classi- tied by sex. There was no significant difference in age among these groups, Table 4 shows that the levels of residual PAHs in the lung cancer group and all cancer group are significantly higher than those in the non-cancer group among mate. Discuss3on The histograms of residual PAHs were asymmetric (non-nor- mal). Concentrations of BaP in human lungs ranged from 0.10 to 2.00 ng/g dry tung, The result implies that they were given extensive exposure and/or their alveolar clearance systems in- rn C 0.6 3 6 10 20 19 17 13 <50 50-59 60-69 70-79 80+ AGE-GROUP Pig. 5. Comparison of BaP concentration in lung by age. The data represent mean (circle) m standard deviation (bar). (0) Male non- smoker; (0) female non-smoker. Numerals above (below) bars are number of samples of male (female) eluding metabolic systems for lung BaP were differing ability. The data log-transformed could be assumed a lognormal distri- bution of concentrations of PAHs in lung of whole samples as well as male or female samples (data not shown). The correlation between PAH concentrations of lung samples was significant (Table I). This means that these PAHS are probably derived from the same sources and are similar to each other in chemical properties and fate in the environment. BaP determined in this study is a representative carcinogenic com- pound associated with carbon particles insufflated into the

$02 Table 4. Comparison of polycyclic aromatic hydrocarbon levels in human lung, classified according to cancer or non-cancer H. Canecr n age BaP° BkF^ BghiP° Male 1.ung cancer 30 67.3 ± 12.3 0.74 ~ 0.44 0.55m0.25 1.19m0.75 All cancers 149 65.3 t 12.4 0.65 m 0.39 0.50±0.26 L04w0.64 Non-cancer 87 66.9 ~ 13.4 0.55 ~ 0.35 0.46 ± 0.27 0.89 ~ 0.58 Significancc't - 0.05 0.05 0.05 SigniBcance"2 - 0.05 0.10 0.05 Female Lung cancer 13 68.7 ± 9.0 0.47 t 0.31 0,39 :t 0.23 0.68 m 0.43 A1lcancen 74 66.1 ~ 13.1 0.41 t 0.24 0.361-0.27 0.64~0.34 Non-csttcer 54 69.1 = 12.8 0.40 ± 0.21 0.36±0.20 0.68t0.36 SignificanccO Slgnificance'2 'Benzofalpytene °Benzo/kjfluorsnthetx ° Benzo[g, h, i]perylone '1 Lung canccr to non-cancer group. ''Ail eancera to non-canecr group. Other conditions are same as Table 2 lungs. Consequently, residual BaP in lung is related to the amount of insufflated carbonaceous particulate matter. In the clearance study on rats exposed to diesel engine exhaust, the decreases of these PAHs in the lung were correlated with each other and with that of carbon in the lung (Seto et a1. 1990). When BaP was administered orally (gastrointestinal tract), it was rapidly metabolized and hardly accumulated in the body (Forth et al. 1984). However, insufflated PAHs with fine car- bon particles is persistent, and accumulates in the lung (Ohkubo er a!. 1988; Sun el al. 1984). Since only a trace amount of residual PAHs (0.01 ng/g wet lung) was detected in the lung of rats (Fischer-344, male, 64 w old) exposed to clean air, it is suggested that the residual PAHs in lung may reflect continuous exposure to polluted air that is inhaled during life. Although lung PAHs shown in Table 3 and Figure 5 seems to be increas- ing with aging, no significance was observed on the effect of age to lung BaP level by analysis of variance. Actual exposures to these PAHs will vary according to life style factors, such as smoking habit, occupation, residential location, etc. Exposures to PAHs are particularly high for coke oven workers (Bjprseth et al. 1978). Engine exhaust of vehicles, containing a number of PAHs, is one of the main sources of PAHs in the atmosphere in Tokyo, Japan. Tobacco smoking is another source of PAH exposure (Appel et a/. 1990). The levels of residual PAHs in lung of the male group were obviously higher than those of the female group (Figure 4). Residual PAIi (BghiP) in lung is increased by smoking among male. However, no significant difference was observed in the case of female. This means that amounts of exposure to PAHs by smoking among female smokers were less than those of male smoken. And the levels of PAHs in lung of male group were significantly higher than those of female group among non-srnokers. The data suggest that exposutn to PAHs through alveolar other than by smoking is greater among male than female. However, the difference between male and female is not completely elucidated by "exposure-level" alone. There are certainly other mcchanisms which would have to be considered to explain the differences between male and female, such as different levels of induction of metabolizing enzymes because of hormonal differences. The relative contents of BaP in the lungs were as follows: lung cancer > all cancer > non-cancer among male. Our re- sults showed that prevalence of cancer of male, especially lung cancer, increases with increasing BaP content in lung. As men- tioned before, in this population the contaminant levels in lung tissue are also slightly associated with age and smoking. How- ever, these confounders are negligible, because of their very little effects on the analysis. Relationship of PAH levels in lung tissue with cancers was clearly demonstrated among male but not among female. Thus, another approach to elucidate preva- lence of female lung cancer must be used. Inhaled BaP from heavily polluted air is activated and binds to DNA (Haugen et al. 1986). The covalent binding of reactive metabolites to DNA is considered to be an important step in tumor initiation by these carcinogens (Miller 1978). Epidemio- logical studies indicate an exess relative risk for lung cancer for smokers (Doll and Peto 1981; Redmond 1983) and coke oven workers (Redmond, 1983) with prolonged exposure. Experi- mental animal studies such as studies of carcinogenicity of tobacco smoke (Auerbach et al. 1970; Hoffman and Wynder 1971) or diesel exhausts (Ishinishi et al. 1986; Mauderly et al. 1986; Stdber 1986) have also supported the validity of the epidemiological conclusions. The interpretation that lung cancer is associated with the PAH levels found in the lung tissue must be limited, because the data were from autopsied patients. The autopsy population is a unique group and is often not representative of all deaths at a given hospital, or in a given community. Acknawledgments. We are grateful to all doctors and patients who provided lung specimens. [teferences Appel BR, Guirguis G, Kim [n-Suk, Garbin 0, Fraccia M, Flessel CP, Kizer KW, Book SA, WatnrinerTE (1990) Benzene, henao(a)py-

. Aromatic Hydrocarbons in the Lung rene, and lead in smoke from tobacco products other than ciga- rcttea. Am I Public Health 80:560-564 Auerbach 0, Hammond EC, Kirman D, Garfinkel L (1970) Effects of cigarette smoking on dogs. 11. Pulmonary neoplasms, Arch Envi- ron Heahh 21:754-768 Auerbach 0, Hammond EC, Garfinkel L(1979) Changes in bronchial epithelium in relation to cigarette smoking. 1955d960 vs. 1970- 1977. N Engl 3 Med 300:381-386 Bjlrseth A, Dj4rseth 0, Fjcldstad PE (1978) Polycyclic aromatic hy- tlrocarbons in work atmosphcres. lI. Determination in a coke plant. Scand J Work Environ Health 4:224-236 Doll R, and Pelo R(1981) The causes of cancer, quantitative estimates of avoidable risks of eancer in the United States today. 1 Natl Cancer Inst 66:1192-1308 Forth H, Kohl R, Kahl GP (1988) Phannacokinetias of low doses of bcnzo(a)pyrene in the rat. Food ChemToxicol 26:45-51 Haugen A, Becher G, Bonestad C, Vahakangas K, Trivers GE, New- man Ml, Hanis CC (1986) Detemtination of polycyclic aromatic hydrocarbons !n the urine, txnzo(a)pyrene diot epoxidcDNA ad- ducts in lymphocyte DNA, and antibodies to the adducts in sera from coke oven workers exposed to tnwuru( amounts of polycy- clic aromatic hydrocarbons in the work atmosphere. Cancer Res 46:4178-4183 Hayano S, I.te JH, Furuy. K, Kikuchi T, Oikawa C, lida Y, Matsush- ita H, Kinouchi T, Manabc Y, Ohnishi Y(1985) Formation of hazardous substances and mutagenicity of PAH produced during the combustion process in  diesel engine. Atmos Environ 19:1009-1015 Hoffman D, Wynder EL (1971) A study of tobacco carcinogenesis. XI. Tumor initiators, tumor accelerators, and tumor promoting activ- ity of condensate fractions. Cancer 27:848-864 Ishinishi N, Kuwabara N, Nagasc S, Suzuki T, Ishiwata S, Khonu T (1986) Long4erm inhalation studies on effects of exhaust from hcavy and light duty diesel engines on P344 rats. In: Ishinishi N, Koizuml A, McClellan RO, Slbber W (eds) Carcinogenic and mutagenic effects of diesel engine exhaust. Elsevier Sci Pub, Amsterdam, pp 329-348 Matsumoto H, Kashfmoto T(1985) Average daily respiratory intake of polycyclic aromatic hydrocarbons in ambient air determined by capillary gas chromatography. Bull Environ Contarn Toxicol 34:17-23 Mauderly )L, Jonu RK, McClellan RO, Henderson RF, Griffith WC (1986) Carcinogenicily of diesel exhaust inhaled chmnically by tats, In: Ishinishi N, Koizumi A, McClellan RO, StBber W (eds) 503 Carcinogenic and mutagenic effects of diesel engine exhaust. Elsevier Sci Pub, Amsterdam, pp 397-409 Miller EC (1978) Some current perspectives on chemical carcinogene sis in humans and experimental animals: presidential address. Cancer Res 38:1479-1496 Obana H, Hori 5, Kashimoto T, Kunita N(1981) Polycyctic aromatic hydrocarbons in human fat and liver. Bull Environ Contam Toxi- co127:23-27 Obana H, Hori 5, Tanaka R, Kashimoto T(1984) Dietary intakes of polycyclic aromatic hydrocarbons- 1 Food Hyg Soc Japan 25:35-40 Ohkubo T, Seto H, Ohsawa M, Suzuki T, Kanoh T (1988) Polycyclic aromatic hydrocavbons vetained in rat lung exposed to diesel en- gine exhaust. Ann Rep Tokyo Metr Res Lab Public Health 39:193-197 Phillips DH (1983) Fifty years of benzo(a)pyrene. Nature 303:468-472 Redmond CK (1983) Cancer monality among coke oven workers. Env'uon Health Perspect 52:67-73 Seto H, Nagashima C, Kanoh T(1987) Availability of formalin-fixed lung as a sample for determination of polycyclic aromatic hydro- carbon residues. Ann Rep Tokyo Metr Res Lab Public Health 38:282-285 Seto H, Ohkubo T, Suzuki T, Ohsawa M, Ikeda S, Watanabe N, Kanoh T (1990) Clearance of polycyclic aromatic hydrocar- bons and elemental carbon from rat lungs after exposure to diesel engine exhaust. Ann Rep Tokyo Metr Res Lab Public Health 41:226-228 Sttiber W (1986) Experimental induction of tumors in hamsters, mice and rats after long-term inhalation of filtered and unfiltered diesel engine exhaust. In: Ishinishi N, Koizumi A, McClellan RO, Std- ber W (eds) Carcinogenic and mutagenic effects of diesel engine exhaust. Elsevier Sci Pub, Amsterdam, pp 421-439 Sun JD, Wolff RK, Kanapi0y GM, McClellan RO (1984) Lung reten- tion and metabolic fate of inhaled benzo(a)pyrene associated with diesel exhaust particles. Toxicol Appl Pharmacol73:4g-59 Tomingas R, Pon F, Dehnen W (1976) Polycyclic aromatic hydrocar- bons in human bronchial carcinoma. Cancer Lett 1:189-196 Tong HY, Sweetman JA, Karasek FW, Jellum E, Thorsrud AK (1984) Quantitative analysis of polycyclic aromatic compounds in diesel exhaust particulate extracts by combined chromatngraphic tech- niques. J Chromatogr 312:183 -102 Manuscript received September 15, 1992 and in revised form Novem- ber 16, 1992.