| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Advances in Brief |
Georgetown University Medical Center, Lombardi Cancer Center, Washington, DC 20007 [R. G., L. E., S. S. K., P. G. S.]; Research Triangle Institute, Research Triangle Park, North Carolina 27709 [E. P., J. B. B.]; and Laboratory of Human Carcinogenesis, National Cancer Institute, NIH, Bethesda, Maryland 20892 [E. D. B.]
| ABSTRACT |
|---|
|
|
|---|
2-fold. The risk for
increasing carcinogenic PAHs (odds ratio, 8.20; 95% confidence
interval, 2.3928.09) was 3-fold compared with noncarcinogenic PAHs
(odds ratio, 2.61; 95% confidence interval, 0.759.12). A higher
concentration of PAHs was detected in the lung tissue of males,
although the estimated smoking was similar in males and females. Race
was not associated with PAH concentrations overall, but PAH
concentrations appeared to be higher in African-American males than in
any other group. Age was weakly correlated with an increase in
fluoranthene and pyrene. The measurement of PAHs in human lung tissue
can be used to estimate the actual dose to the target organ. | Introduction |
|---|
|
|
|---|
The carcinogenicity of PAH compounds is mediated by DNA damage (3) . The best-characterized pathway for genotoxicity of PAHs consists of covalent binding of the metabolically activated carcinogens to DNA bases to form DNA adducts (6) . The DNA adducts are promutagenic lesions that can lead to mutations in oncogenes or tumor suppressor genes, if not repaired or otherwise eliminated (7) . Concentration of DNA adducts is considered the biologically effective dose of the parent PAH (8) . Because accumulation of parent PAH compounds in the lung tissue precedes formation of DNA adducts, it is intuitive that the lipophilic parent PAHs would be a good estimate of dose in the target organ.
Smoking yields of PAH compounds vary considerably based on the type of
cigarette and human differences in smoking behavior. It was estimated
by the FTC machine method that a smoker is exposed to about 130
µg/day/pack of cigarettes (9)
. PAH exposure also occurs
from non-tobacco sources and can be assessed by environmental sampling
or personal monitoring (10)
. Certain occupational
exposures, e.g., coke-oven work, present a burden comparable
with smoking. The air in urban areas can contain higher percentage of
PAH compounds than the air in the countryside, especially in the
winter. Our diet also contributes to exposure, especially from
overcooked foods; it has been estimated that diet contributes 3
µg/day of total PAH intake,
70% for a nonsmoker
(11)
. The PAH exposure pattern is therefore complex, and
analysis of the lung tissue can provide direct information difficult to
estimate by other methods.
Our study determined the concentration of 11 PAHs in 70 lung samples of cancer-free subjects using a sensitive and robust GC-MS method (12) . Six of the analyzed PAHs were classified previously as laboratory animal carcinogens (1) . The study reports the analysis of individual PAHs as well as the sum of carcinogenic, noncarcinogenic, and total PAHs. The size (70 individuals) and scope (11 PAHs) of the study provide an evaluation of the effect of smoking, race, gender, age, and fat content on the accumulation of PAHs in the lung tissue.
| Materials and Methods |
|---|
|
|
|---|
Cotinine.
Smoking data were not consistently available at the time of death.
Thus, we classified smokers on the basis of serum cotinine. This might
lead to the misclassification of some smokers as nonsmokers because of
the 48-h half-life of cotinine. But because these individuals died
unexpectedly (e.g., trauma), it is unlikely that they
changed their smoking patterns before death. Serum cotinine was
determined by a RIA kit (STC Technologies, Bethlehem, PA) with
modifications as reported (13)
. Subjects with a serum
cotinine concentration >14 ng/ml were classified as smokers
(14)
.
PAH Analysis.
The GC-MS method was described elsewhere (12)
. Briefly,
PAHs were extracted into hexane from 4 g of tissue after
saponification. The extract was separated using a HP 6890 gas
chromatograph (Agilent Technologies, Palo Alto, CA) on a Restek Rtx-5
MS 5% diphenyldimethylpolysiloxane column and analyzed by GC-MS HP
5973 mass spectrometer (Agilent Technologies) using selected ion
monitoring mode. The GC-MS was calibrated by standard solutions
containing all 11 target compounds at concentrations ranging from 0.5
to 50 pg/µl. The standard solutions also contained internal standards
at fixed concentrations. Second-order calibration curves were fitted to
this data using the least-squares method, weighting the data by the
inverse of the concentration level to obtain an optimal fit at the
lower levels.
Fat Content Analysis.
Fat content was determined gravimetrically after extraction into hexane
as described previously (15)
. The lung tissue (2 g) was
homogenized with PBS (pH 7.2; 2 ml; Life Technologies, Inc., Rockville,
MD) and methanol (2 ml; Sigma Chemical Co., St. Louis, MO).
Hexane/ethyl ether (Sigma Chemical Co.), 10 ml of a 1:1 solution, was
used to extract the homogenate overnight on a rotary mixer (Fisher,
Inc., Fair Lawn, NJ). The organic layer was transferred to a new vial,
the samples were extracted twice more for 2 h each as described
above, and organic layers were pooled and dried under a constant stream
of nitrogen.
Statistical Analysis.
Statistical analyses were performed using STATISTICA 99 edition
(StatSoft, Inc., Tulsa, OK) and SAS System for Windows version 8.01
(SAS Institute, Inc., Cary, NC). All of the PAH data entries are
reported as ng/gram of wet tissue (ppb). The entries analyzed by GC-MS
in duplicate were averaged. The background, average of 34 blank
injections, was subtracted from each entry (12)
. PAH
levels were summed for each individual, on a ng/gm wet tissue basis,
where measurements below the detection limit were assigned zero values.
PAHs known to be animal carcinogens (BA, BBF, BKF, BAP, IP, and DA;
Ref. 1
), and those that were not (FL, PY, CH, BEP, and BP)
were summed separately. The distribution of the PAHs in the tissue was
not normal but could be normalized with log transformation. To assess
the effects of smoking on PAH levels, nonparametric methods
(Mann-Whitney U test and Spearman R correlation coefficient)
and logistic regression analysis were used (SAS System for Windows
8.01; SAS Institute Inc.). Dependent (PAH) variables were dichotomized
by the median. Controlling variables for the logistic regression were
gender, race, age, and fat content. Interactions between smoking and
the other four independent variables (gender, race, age, and fat
content) were evaluated. ANOVA with log-transformed PAH concentrations
was used to assess the difference for PAH levels by smoking status,
where nonsmokers were characterized by cotinine <14 ng/ml, and light
smokers (cotinine 14339 ng/ml) and heavy smokers (cotinine >339
ng/ml) were split by the median cotinine concentration among smokers
(median, 339 ng/ml).
| Results |
|---|
|
|
|---|
10-fold between the most (FL) and least (BEP) abundant
PAH. The data show that none of the PAHs was detected in all of the
tissues examined, but all of the PAHs were detected in at least some
samples. Our analyses concentrated on the effect of smoking on PAH
concentration in the lung tissue.
|
|
|
|
The concentration of PAHs in the lung tissue of males compared
with females was uniformly higher for both the carcinogenic and
noncarcinogenic PAHs (Table 4)
, although there were similar percentages of male and female smokers,
and the cotinine levels tended to be higher for the females compared
with the males (borderline statistical significance). Almost every
individual PAH was higher in males compared with females.
Interestingly, the increase in males compared with females was mostly
attributable to the subcategory of African-Americans (Table 5)
.
|
|
The average fat content of the lungs was 1.1%, with a range of 0.1 to 5.7% (data not shown). We did not find a significant correlation of any PAH, or sum of PAHs, with the fat content of the lung tissue.
| Discussion |
|---|
|
|
|---|
2-fold, and those PAHs that are carcinogenic are present at higher
concentrations. The risk for having higher carcinogenic PAHs compared
with noncarcinogenic PAHs was >3-fold. Moreover, there was a
dose-response relationship for smoking characterized by increasing
serum cotinine. Men had a higher concentration of PAHs than women,
although the smoking levels were equivalent. This was mostly explained
by differences in African-Americans compared with Caucasians, although
overall there was no difference between the races. There have been several previous reports of PAH concentrations in tissues, but this study uses a more sensitive detection method, is more comprehensive, and is the first study to detect significant increases of BAP and other PAHs with smoking, including a dose-response relationship (16, 17, 18, 19) . These previous studies of PAHs in human tissues report values that vary by two orders of magnitude, possibly because of different population characteristics (race and smoking status), extraction/quantitation methods (UV-Vis spectroscopy, fluorescence, and mass spectrometry), and other methodological differences (wet or dry tissue weight). Importantly, the majority of previous studies used cancer patients, and cancer status was shown to affect PAH concentration in tissues (17) .
The concentration of BAP in lung tissue has been reported most frequently. Lodovici et al. (16) reported BAP and five other PAHs in 20 lung autopsy tissues, where BAP concentration correlated with BAP DNA-adducts. Seto et al. (17) analyzed BAP, BKF, and BP in 364 lung samples from both cancer-free and cancerous formalin-fixed autopsies. The authors found that cancer tissue contains higher PAH concentration than noncancerous tissue, and that, similar to this study, males have higher PAH concentrations in the lung than females. Tokiwa et al. (18) analyzed BAP in 158 lung tumor resections and proposed a correlation of the PAHs with environmental exposure. Tomingas et al. (19) analyzed 12 PAHs in 33 bronchial carcinoma tissues but detected only four of the PAHs. Only BAP was found in all tissues at high concentrations in that study.
It is interesting that four of five PAHs that were increased in smokers were classified previously as carcinogenic to animals (1) . The steady-state in the lung tissue is defined by complex exposure, distribution, and metabolism processes that cannot be unequivocally defined in this study. There are several possible explanations why smoking increases carcinogenic PAHs more that the noncarcinogenic PAHs. A smoking-related increase in noncarcinogenic PAHs such as FL and PY, which seem to accumulate in the tissue with age, is too small and cannot be detected. Another explanation is that metabolic turnover is an important determinant of the steady-state and that smoking affects clearance of the PAHs indirectly through induction of metabolic enzymes. Smoking could also be a better source of carcinogenic PAHs than diet and the environment. However, a conclusion about the source of PAH cannot be made without detailed knowledge of individual exposures because the concentration of PAHs in the environment and in cigarette smoke vary considerably, depending on several factors (source of environmental pollution, type of cigarette, and smoking topography; ref. 9 ). Because we do not know what the brand types were of the smokers, we could not determine PAH levels in relation to FTC tar yields, to assess whether lower yield cigarettes actually produced lower dose levels. PAH concentrations were consistently higher in males than in females with significant differences for several individual hydrocarbons. This might be attributable to differences in the type of cigarettes smoked by men compared with women. Men tended to smoke heavier cigarettes with higher tar content at the time when exposures occurred.
Lung cancer rates differ by gender and race. The highest rates are observed for African-American males (112/100,000), followed by Caucasian males (73/100,000) and African-American and Caucasian females (46 and 43/100,000 respectively; Ref. 20 ). The PAH concentration in lung tissue from this study follows the same trend, although the difference by race was not statistically significant. African-American males have the highest PAHs (0.784 ng/g), followed by Caucasian males (0.407 ng/g), African-American females (0.308 ng/g), and Caucasian females (0.279 ng/g). In this study, African-American males had the highest concentration of almost every PAH examined, with Caucasian males second in many cases. This ranking suggests that African-American males have higher doses compared with other groups, or that metabolism/distribution of the PAHs in AfricanAmerican males is different. It is known, for example, that African-Americans smoke mentholated cigarettes more frequently than Caucasians (21) , which could affect exposures and inhalation patterns.
Fat content did not correlate with the PAH concentration in the lung tissue. We had expected to find a correlation because PAHs are lipophilic. This indicates that the PAHs do not accumulate in the fat of the lung. This result is consistent with a previous study that reported low PAH content in human fat tissue (22) . The concentrations (<1 ppb) are much lower than, for example, concentrations of some polychlorinated biphenyl compounds (>1 ppm), which were shown to accumulate in the fat. It was shown that only the polychlorinated biphenyls, which are not metabolized by cytochrome P-450, accumulate in the adipose tissue (23) . The available information on the metabolism of PAHs in human lung suggests that the metabolic activity might be an important determinant of the steady-state of the PAHs in the tissue (24) .
There are several limitations to this study. The use of cotinine as a marker of smoking is well accepted, and higher levels are indicative of greater smoking levels (14) . However, cotinine is only a marker of recent smoking and does not reflect lifetime exposure, whereas the PAH levels might reflect long-term exposure. Also, some individuals might be misclassified as nonsmokers if they stopped smoking recently. In this study group, because most died unexpectedly from trauma or cardiac disease, it is unlikely that the subjects actually changed their smoking habits. Another limitation is the small size of the study, making it difficult to assess race and gender issues combined, although it was large enough to reveal overall patterns. Finally, we were unable to actually obtain lifetime smoking histories, brand use, and other factors that would affect exposure. This information would be needed for additional interpretations of the data.
In conclusion, this study shows that smoking increases the concentration of BAP and other carcinogenic PAHs in the human lung tissue with a dose-response relationship. These methods could provide for direct assessment of actual exposure to PAHs in target organs, rather than estimating exposure through FTC tar yields or self-reported histories. Also, additional data would be useful to study the impact of metabolic phenotypes and/or cigarettes with different FTC tar yields and the exposures from cigarette-like devices on the steady-state of PAH in the lung tissue.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
1 Supported by NIH, National Cancer Institute
intramural funds, and by Department of Defense Grant
DAMD17-99-1-9309. ![]()
2 To whom requests for reprints should be
addressed, at Cancer Genetics and Epidemiology Program, Lombardi Cancer
Center, Georgetown University Medical Center, Research Building W315,
3970 Reservoir Road, NW, Washington, DC 20007. Phone: (202) 687-0003;
Fax: (202) 687-0004; E-mail: pgs2{at}georgetown.edu ![]()
3 The abbreviations used are: OR, odds
ratio; CI, confidence interval; BAP, benzo(a)pyrene;
PAH, polycyclic aromatic hydrocarbon; FTC, Federal Trade
Commission; GC-MS, gas chromatography-mass spectrometry; BA,
benzo(a)pyrene; BBF, benzo(b)anthracene;
BKF, benzo(k)fluoranthene; DA,
dibenzo(a,h)anthracene; IP, indeno(1,2,3)pyrene; FL,
fluoranthene; PY, pyrene; CH, chrysene; BEP,
benzo(e)pyrene; BP,
benzo(g,h,i)perylene. ![]()
Received 4/27/01. Accepted 7/16/01.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
P. Suwan-ampai, A. Navas-Acien, P. T. Strickland, and J. Agnew Involuntary Tobacco Smoke Exposure and Urinary Levels of Polycyclic Aromatic Hydrocarbons in the United States, 1999 to 2002 Cancer Epidemiol. Biomarkers Prev., March 1, 2009; 18(3): 884 - 893. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. A. Fowler, S. Cassie, S. M. Rhind, M. J. Brewer, J. M. Collinson, R. G. Lea, P. J. Baker, S. Bhattacharya, and P. J. O'Shaughnessy Maternal Smoking during Pregnancy Specifically Reduces Human Fetal Desert Hedgehog Gene Expression during Testis Development J. Clin. Endocrinol. Metab., February 1, 2008; 93(2): 619 - 626. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. S. Neal, J. Zhu, A. C. Holloway, and W. G. Foster Follicle growth is inhibited by benzo-[a]-pyrene, at concentrations representative of human exposure, in an isolated rat follicle culture assay Hum. Reprod., April 1, 2007; 22(4): 961 - 967. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Sasaki, T. Kondo, F. Sata, Y. Saijo, S. Katoh, S. Nakajima, M. Ishizuka, S. Fujita, and R. Kishi Maternal smoking during pregnancy and genetic polymorphisms in the Ah receptor, CYP1A1 and GSTM1 affect infant birth size in Japanese subjects Mol. Hum. Reprod., February 1, 2006; 12(2): 77 - 83. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Culea, O. Cozar, and E. Culea PAHs in Cigarette Smoke by Gas Chromatography-Mass Spectrometry Indoor and Built Environment, June 1, 2005; 14(3-4): 283 - 292. [Abstract] [PDF] |
||||
![]() |
A. J. Marrogi, L. E. Mechanic, J. A. Welsh, E. D. Bowman, M. A. Khan, L. Enewold, P. G. Shields, and C. C. Harris TP53 Mutation Spectrum in Lung Cancer Is Not Different in Women and Men Cancer Epidemiol. Biomarkers Prev., April 1, 2005; 14(4): 1031 - 1033. [Abstract] [Full Text] [PDF] |
||||
![]() |
O. I. Kalantzi, R. Hewitt, K. J. Ford, L. Cooper, R. E. Alcock, G. O. Thomas, J. A. Morris, T. J. McMillan, K. C. Jones, and F. L. Martin Low dose induction of micronuclei by lindane Carcinogenesis, April 1, 2004; 25(4): 613 - 622. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Ito, M. Son, S. Sato, T. Ohashi, M. Kondo, K. Shimokata, and H. Kume Effects of Fluoranthene, a Polycyclic Aromatic Hydrocarbon, on cAMP-Dependent Anion Secretion in Human Airway Epithelia J. Pharmacol. Exp. Ther., February 1, 2004; 308(2): 651 - 657. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |