| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Carcinogenesis |
Division of Environmental and Occupational Health and Cancer Center, University of Minnesota, Minneapolis, Minnesota 55455
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We have been investigating the interactions between the two activation pathways of the tobacco-specific nitrosamine, NNK.3
This lung-specific carcinogen is activated by 
-hydroxylation to either a methylating agent or a pyridyloxobutylating agent (Fig. 1)
. The methylation pathway generates methyl DNA adducts such as 7-mG and O6-mG (1, 2, 3)
. The pyridyloxobutylating agent reacts with DNA to form adducts; the majority of these adducts are unstable to DNA hydrolysis conditions and decompose to release HPB (2
, 3)
.
|
Our studies in A/J mice indicate that there are interactions between the two activation pathways of NNK (3) . The model methylating agent, AMMN, was more tumorigenic when given in combination with NNKOAc. The ability of NNKOAc to increase the tumorigenic activity of AMMN was attributed to its ability to enhance the persistence of O6-mG in lung DNA.
We have been exploring the possibility that NNKOAc increases the levels of O6-mG in lung DNA as a result of inhibition of O6-mG repair. The primary repair pathway for O6-mG is mediated by AGT. AGT facilitates the transfer of the methyl group from the O6-position of guanine to a thiol group at the protein’s active site (8) . This reaction inactivates AGT, resulting in depletion of cellular AGT levels. When AGT is depleted, O6-mG accumulates, and the potential for tumor initiation by this promutagenic adduct increases (9) . Therefore, if the pyridyloxobutylation pathway generates adducts that can compete with O6-mG for repair by AGT, less O6-mG repair should occur when both pathways are present.
Our laboratory has demonstrated that NNKOAc impedes O6-mG repair by AGT in vitro (10 , 11) . This interference results from either direct alkylation of the protein or as a result of DNA alkylation to generate AGT substrate adducts; these adducts then compete with O6-mG for repair by AGT. One AGT substrate adduct has been identified as O6-[4-oxo-4-(3-pyridyl)-butyl]guanine (12) .
The experiments described in this study investigate the ability of NNKOAc to deplete AGT in A/J mouse lung. In addition, we compared the consequences of AGT depletion by NNKOAc to those observed with a known AGT inhibitor, O6-bG (13) . To determine the effect that AGT depletion has on O6-mG persistence, we measured the ability of these two compounds to alter the levels of O6-mG derived from two model methylating agents, MNU or AMMN. Because an increased persistence of O6-mG in lung DNA should result in an increase in tumor yield, we conducted an A/J mouse bioassay to determine the outcome of coadministration of NNKOAc or O6-bG on the tumorigenic activity of MNU or AMMN.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Animals.
Female A/J mice, age 6–7 weeks, were purchased from Jackson Labs (Bar Harbor, ME). They were housed in groups of 5 under standard conditions (15)
and maintained on American Institute of Nutrition-76A-modified diet (Dyets, Inc., Bethlehem, PA) throughout each experiment. The animals were given i.p. injections of NNKOAc in saline (0.2 ml) or O6-bG in 40% PEG-400 in PBS (0.2 ml) immediately followed by i.p. injections of NNK, AMMN, or MNU in saline (0.2 ml). The first injection was on the mouse’s right side, and the second injection was on the mouse’s left side. Saline or 40% PEG-400 in PBS was substituted for the compounds in control groups. The treatment groups are displayed in Table 1
.
|
. Five animals in each group were sacrificed at 4 or 96 h postinjection. Lung and liver tissues were removed, freeze clamped at liquid nitrogen temperatures, and stored at -80°C for tissue extract preparation.
Tissue Extract Preparation.
Tissue extracts were prepared according to the method of Gerson et al. (16)
. Briefly, a pair of A/J mouse lungs or one liver was homogenized in ice cold cell extract buffer [70 mM HEPES, 1.6 mM EDTA, 1 mM DTT, and 5% glycerol (pH 7.8); 1 or 4 ml, respectively] at 4°C in a Tenbroeck tissue grinder (Pyrex, Corning, NY). The homogenate was sonicated 3 x 10 s (setting 4, Sonic Dismembrator 60; Fisher Scientific, Chicago, IL) at 4°C. The homogenates were centrifuged 2 min (12000 x g) at 4°C. The resulting supernatants were assayed immediately for AGT activity and total protein.
Protein Assay.
The protein content of each extract was determined by the Bradford method with Bio-Rad Protein Assay Kit II (Bio-Rad, Hercules, CA).
AGT Assay.
AGT activity of the tissue extracts was determined by measuring the transfer of [3H]methyl groups from the O6 position of guanine in [3H]MNU-alkylated DNA to the protein (10
, 17)
. Tissue extracts (1–2 mg of protein) were incubated with [3H]MeDNA containing 55 pmol of O6-mG in 60 mM HEPES, 1.3 mM EDTA, 1.6 mM DTT, 40 µM spermidine, and 4% glycerol for 60 min at 37°C (total volume: 1 ml). BSA (1 mg) was substituted for tissue extract in the controls. Each incubation was performed in triplicate. The reactions were stopped by the addition of 50% trichloroacetic acid (200 µl). The samples were heated at 80°C for 60 min to release purines and any unrepaired O6-mG from the DNA. The precipitate, which contains methylated AGT and depurinated DNA, was collected by centrifugation (40 min, 16,000 x g). Pellets were washed with 1-ml cold 5% TCA and collected by centrifugation (30 min, 16,000 x g). They were dissolved in 0.1 N NaOH (0.3 ml) and transferred to a scintillation vial. The microfuge tubes were washed with H2O (0.3 ml) and 0.2 mM Tris-HCl (0.6 ml). These washes were combined with the pellet and scintillation fluid (Ecoscint A; National Diagnostics, Atlanta, GA). The extent of [3H]methyl groups transferred to protein was determined by scintillation counting (Beckman LS 6500 scintillation counter; Beckman, Schaumburg, IL). The fmol AGT/ml was related to the total protein content of the tissue extract, yielding a final AGT activity in fmol/mg protein.
The Effect of AGT Inhibitors on O6-mG Levels in Mouse Lung DNA.
Groups of 30 mice were treated as described in Table 1
. Each group was divided into two subgroups of 15 mice for sacrifice at 4 and 96 h after treatment. Lungs of 5 mice were pooled for each time point, freeze clamped at liquid nitrogen temperatures, and stored at -80°C for DNA isolation. DNA was isolated according to methods published previously (2)
.
Measurement of O6-mG Levels in Lung DNA.
Neutral thermal and mild acid hydrolysates of lung DNA were prepared as described previously (1)
. Levels of 7-mG, O6-mG, and guanine were determined by high-pressure liquid chromatography analysis with fluorescence detection (Waters 470 scanning fluorescence detector with SAT/IN enhancement; Waters Corp., Milford, MA) and separated on a single Partisil 10 SCX column (Phenomenex). Neutral thermal hydrolysates were eluted with 100 mM ammonium phosphate buffer (pH 2.0; 1 ml/min). Mild acid hydrolysates were eluted with the same buffer plus 10% methanol (1 ml/min). Standard curves were constructed for each analysis to determine adduct concentration. The amount of 7-mG and O6-mG were related to the guanine concentration of each sample.
Effects of AGT Inhibitors on Tumorigenic Activity of Methylating Agents.
Groups of 20 mice were treated as indicated in Table 1
. After 16 weeks, mice were sacrificed, pulmonary tissue was removed and fixed in 10% phosphate-buffered fixative, and lung adenomas were counted. For statistical analysis, the tumor per mouse was converted to a log10 (tumor count + 1) value. The Student t test was performed on these log values to determine statistical differences between groups.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We chose experimental conditions similar to our earlier study (3) . In those studies, we determined that a 4.2-µmol dose of NNKOAc (i.p. in saline) yielded levels of lung DNA pyridyloxobutylation comparable with those produced by a tumorigenic dose of NNK (10 µmol). AMMN was used as the model methylating agent in the previous study. There was a concern that NNKOAc and AMMN competed with one another for hydrolysis by esterase and that this competition could account for some of the interactions observed between these two compounds. Therefore, we also included MNU in these studies. MNU spontaneously decomposes to the same methylating agent as AMMN and eliminates any complications arising from competition for activation enzymes. Preliminary studies established that 0.75 µmol of MNU methylates lung DNA to a similar extent as 0.5 µmol of AMMN (data not shown). The 4- and 96-h time points were selected because our previous study indicated that the levels of DNA methylation peaked at 4 h and that there was a strong correlation between lung tumor formation and pulmonary levels of O6-mG at 96 h (3) . Lung adenomas were counted 16 weeks after exposure to the chemicals as described previously (3 , 15) .
AGT Depletion by NNKOAc.
The ability of NNKOAc to deplete lung and liver AGT levels was determined by treating mice with 4.2 µmol of NNKOAc (i.p., in saline) and measuring AGT levels in lung and liver tissue extracts. We compared these data to those obtained with 10 µmol of NNK, 2.5 µmol of O6-bG, or the methylating agents AMMN and MNU in the presence or absence of NNKOAc or O6-bG. The 2.5-µmol dose of O6-bG (
30 mg/kg; i.p., 40% PEG-400 in PBS) was established in preliminary studies (data not shown).
The results from this study are summarized in Table 2
. NNKOAc reduced lung AGT levels at both 4 and 96 h postinjection. Liver AGT levels were not significantly affected by NNKOAc treatment. O6-bG depleted both liver and lung AGT levels at 4 h, but the levels at 96 h were not different from the vehicle controls. Similar effects have been reported for O6-bG in other mouse strains (13
, 18
, 19)
. NNK lowered both lung and liver AGT levels at 4 h. Whereas the lung AGT levels were still depressed at 96-h posttreatment, liver AGT levels returned to normal by 96 h after NNK treatment. The ability of the liver AGT levels to recover to normal levels by 96 h is consistent with the rapid repair of O6-mG in the liver (Table 3)
and the decreased sensitivity of this tissue to the carcinogenic effects of NNK (20)
.
|
|
Effect of O6-bG and NNKOAc on DNA Methylation by AMMN or MNU.
To compare the abilities of NNKOAc and O6-bG to increase the levels of O6-mG derived from AMMN or MNU, levels of O6-mG were measured in lung DNA isolated from mice treated with AMMN or MNU in the presence or absence of O6-bG or NNKOAc. The levels at 4 and 96 h in each treatment group are displayed in Table 3
. The levels of O6-mG detected in lung DNA from NNK-treated mice were similar to our earlier study (3)
. Coadministration of NNKOAc and AMMN increased the levels of O6-mG at 4 and 96 h over those observed with AMMN alone. A similar effect was observed in our previous study (3)
. Simultaneous administration of O6-bG and AMMN lead to a similar increase in the levels of O6-mG in lung DNA from AMMN-treated animals. There were no significant effects of either NNKOAc or O6-bG on liver O6-mG levels.
In contrast, neither compound significantly altered the levels of O6-mG in lung DNA from MNU-treated mice. There was an increase in the levels of O6-mG in liver DNA when MNU was given with O6-bG, but this increase was not statistically significant.
Effect of O6-bG and NNKOAc on Tumor Induction by AMMN or MNU.
The ability of NNKOAc and O6-bG to influence lung tumor multiplicity in AMMN- or MNU-treated mice was compared to see if the increases in lung O6-mG levels correlated with increased tumor yield. The bioassay results are summarized in Table 4
. NNK (10 µmol) was included as a positive control, and the tumor multiplicity obtained in this study is similar to previous reports (3
, 4
, 15) . As observed in the original study, NNKOAc increased the tumor multiplicity in animals receiving both AMMN and NNKOAc (3)
. However, this increase was not significant at the highest dose of AMMN in this study. NNKOAc also significantly augmented the tumor multiplicity of mice receiving 1 µmol of MNU. These data confirm our previous finding that NNKOAc increases the tumorigenicity of methylating agents when the compounds are coadministered (3)
.
|
).
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
7% of the total HPB-releasing adducts in NNKOAc-treated DNA. The percentage of total AGT substrate adducts ranges from 3 to 30% of the total HPB-releasing adducts, depending on the DNA preparation (10
, 11)
. A study investigating the levels of AGT substrate adducts formed in lungs and livers of NNK- or NNKOAc-treated mice is currently underway in our laboratory. The ability of NNKOAc to reduce AGT activity supports our hypothesis that this compound increases the levels and persistence of O6-mG in lung DNA from AMMN-treated mice as a result of interference of O6-mG repair. Consistently, we observed a trend toward enhanced AGT depletion when the methylating agents were administered with NNKOAc; however, this difference was not statistically significant.
O6-bG has been a useful tool in assessing the role of O6-alkylguanine adducts in the carcinogenic or toxic properties of a variety of compounds (13 , 21, 22, 23, 24) . In our studies, O6-bG was an effective inactivator of AGT in both lungs and liver of A/J mice. The effects of this compound were transient, with recovery back to control levels by 96 h after administration. Similar effects have been observed previously in livers of O6-bG-treated mice or rats (13 , 18 , 19 , 25) . As with NNKOAc, there was a trend toward increased AGT depletion when O6-bG was coadministered with the methylating agents, but the differences were not statistically significant.
The depletion of lung AGT levels was accompanied by increased O6-mG levels in lung DNA from mice receiving both AMMN and NNKOAc or O6-bG at both 4 and 96 h (Table 3)
. The results with NNKOAc were similar to our previous study (3)
. Other groups have reported the ability of O6-bG to increase the levels of O6-mG from a variety of methylating agents (21
, 26)
. O6-bG and NNKOAc were comparable in their ability to increase O6-mG levels in lung DNA from 0.5- or 1-µmol AMMN-treated mice (Table 3)
. Simultaneous treatment of NNKOAc or O6-bG with MNU did not significantly affect lung O6-mG levels. However, there was a trend toward higher amounts of O6-mG at the 96-h time point. The lower levels of DNA methylation observed in MNU-treated mice probably account for the diminished effects of O6-bG or NNKOAc on O6-mG levels in the MNU-treated mice. As a result of the lower DNA alkylation, AGT levels recovered more quickly in the MNU-treated mice (Table 2)
. Consistently, O6-bG or NNKOAc is more effective at increasing O6-mG levels in the 1-µmol AMMN-treated mice than in those receiving 0.5 µmol of AMMN.
The results from the bioassay indicate that the pyridyloxobutylating agent, NNKOAc, can influence the tumorigenic activity of methylating agents in two ways. One mechanism of interaction is that both compounds are tumorigenic, and the increase in tumors per mouse results from an addition of these tumorigenic activities. This mechanism seems to dominate at low doses of AMMN or MNU because NNKOAc, but not O6-bG, significantly increased adenomas per mouse under these treatment conditions (Table 4)
. NNKOAc is weakly tumorigenic in this model (this study and Ref. 3
), whereas O6-bG is not. The NNKOAc-induced increase in tumor yields at these doses of MNU and AMMN appears to be additive.
A second way NNKOAc can enhance the tumorigenic activity of methylating agents is through enhanced depletion of the repair protein, AGT, leading to reduced O6-mG repair. This mechanism seems to dominate at the higher doses of AMMN, because NNKOAc and O6-bG increased the tumorigenic activity of 0.75 µmol of AMMN to a similar extent. A similar trend was observed with 1 µmol of AMMN; however, the increase in tumors per mouse in the mice receiving both 1 µmol of AMMN and NNKOAc was not statistically significant. The increases were more than additive and correlate well with the increased levels of O6-mG detected at 96 h in the lung DNA of these animals (r = 0.92). This observation is similar to that reported in our initial study (3) .
The studies with the model pyridyloxobutylating agent, NNKOAc, and the model methylating agents, MNU and AMMN, suggest that similar interactions are probably important for the organospecific effects of NNK. Compounds that only methylate or pyridyloxobutylate DNA are not potent lung carcinogens in laboratory animals (1 , 3 , 27) . Whereas there are multiple mechanisms that could be proposed for these differences such as cell-selective metabolic activation (28) , our studies with model compounds indicate that the simultaneous presence of both pathways leads to a more active tumorigen.
The DNA repair capacity of the tissue likely contributes to the organoselectivity of NNK because the formation and persistence of O6-mG appear to be important for NNK’s lung tumorigenic activity in experimental animals (3 , 4 , 29) . In A/J mice, studies with deuterated NNK analogues demonstrated that DNA methylation is necessary for lung tumor formation (4) . A strong correlation exists between O6-mG levels at 96 h postinjection and lung tumors after various doses of NNK (3) , indicating that O6-mG persistence is important for tumor formation. Such a relationship was not observed at earlier time points.
The reduced O6-mG repair capacity of the lung likely contributes to the marked persistence of O6-mG in this tissue. The levels of lung AGT are roughly 20% of the levels of liver AGT (Ref. 16
and this study). Consistently, lung AGT levels were suppressed longer than liver AGT levels after treatment with the various alkylating agents (Table 2)
. This depletion results in increased persistence of O6-mG in lung versus liver DNA (Table 3)
.
It is our belief that the added presence of the pyridyloxobutylation pathway extends the depletion of AGT, allowing the mutagenic adduct O6-mG to persist longer than it would in the absence of this pathway. Consistently, O6-mG persistence is greater in lung DNA from NNK-treated mice than in animals receiving simple methylating agents (3) . Thus, the increased persistence of O6-mG in lung DNA results in increased susceptibility to tumorigenesis.
Our studies with methylating agents in A/J mice support this concept (Ref. 3 and this study). In addition, high levels of AGT protect against NNK-induced lung tumorigenesis because NNK is a less potent lung carcinogen in transgenic mice containing the human AGT transgene (30) . This decrease in tumorigenesis was accompanied by a significant reduction in frequency of activated K-ras mutations (30) , indicating that AGT protects against the activation of this oncogene by NNK. It would be interesting to explore the carcinogenic activity of NNK in an AGT-deficient mouse. Our data suggest that there should be only modest effects on lung carcinogenicity because AGT is essentially depleted. However, there might be an increase in tumorigenesis in tissues that have high metabolic activity, such as liver.
All of the data support the hypothesis that the pyridyloxobutylation pathway of NNK contributes to the tumorigenic activity of NNK in A/J mice in at least two ways: (a) the pyridyloxobutylation pathway generates adducts that contribute to the initiating activity of NNK; and (b) the pyridyloxobutylating metabolite reduces the repair of O6-mG by increasing the depletion of lung AGT. This decreased repair increases the probability of K-ras activation by the DNA methylation pathway, resulting in lung adenoma initiation.
The relevance of these findings to other species susceptible to NNK-induced pulmonary carcinogenesis will require additional study. It is clear that the mechanism of NNK-induced pulmonary tumorigenesis is complicated. For example, although K-ras activation is an important mechanism of tumor induction by NNK in A/J mice (5
, 6)
, NNK-induced K-ras oncogene activation does not occur in rats (31)
. In addition, there was residual pulmonary tumorigenic activity of NNK in AGT transgenic mice, implicating other forms of damage by NNK (30)
. Studies in rats suggest that the relative contribution of these two pathways in the rat lung will be difficult to elucidate. There was no effect of -carbon-deuterium substitution on lung tumor formation in NNK-treated rats (32)
. While a linear relationship was observed between O6-mG levels in Clara cells and tumor incidence in NNK-treated rats (29)
, a similar relationship was observed for HPB-releasing adducts and lung tumor formation in type II lung cells in NNK-treated rats (33)
. Morphological studies suggest that NNK-induced rat lung tumors arise primarily from type II lung cells (29)
. All these reports indicated that both activation pathways of NNK contribute significantly to the lung tumorigenic activity of NNK.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported by Grant CA-59887 from the National Cancer Institute. The core facilities at the University of Minnesota Cancer Center were funded by the National Cancer Institute Grant CA-77598. 
2 To whom requests for reprints should be addressed, at Division of Environmental and Occupational Health and Cancer Center, University of Minnesota, Minneapolis, MN 55455. Phone: (612) 626-0164; Fax: (612) 626-5135; E-mail: peter431{at}umn.edu
3 The abbreviations used are: NNK, 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone; AGT, O6-alkylguanine-DNA alkyltransferase; AMMN, acetoxymethylmethylnitrosamine; 7-mG, 7-methylguanine; HPB, 4-hydroxy-1-(3-pyridyl)-1-butanone; MNU, N-methyl-N-nitrosourea; NNKOAc, 4-(acetoxymethylnitrosamino)-1-(3-pyridyl)-1-butanone; O6-bG, O6-benzylguanine; O6-mG, O6-methylguanine; PEG, polyethylene glycol.
Received 2/15/01. Accepted 5/25/01.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-hydroxylation of N'-nitrosonornicotine and 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone: evidence for a cyclic oxonium ion intermediate in the alkylation of nucleophiles. Chem. Res. Toxicol., 3: 350-356, 1990.[Medline]
-deuterium substitution on the tumorigenicity of 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone in F344 rats. Carcinogenesis (Lond.), 8: 291-294, 1987.This article has been cited by other articles:
|
|
 |
|
  P. Upadhyaya, B. R. Lindgren, and S. S. Hecht Comparative Levels of O6-Methylguanine, Pyridyloxobutyl-, and Pyridylhydroxybutyl-DNA Adducts in Lung and Liver of Rats Treated Chronically with the Tobacco-Specific Carcinogen 4-(Methylnitrosamino)-1-(3-pyridyl)-1-butanone Drug Metab. Dispos., June 1, 2009; 37(6): 1147 - 1151. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. E. Sandercock, J. N. Hahn, L. Li, H.A. Luchman, J. L. Giesbrecht, L. A. Peterson, and F. R. Jirik Mgmt deficiency alters the in vivo mutational spectrum of tissues exposed to the tobacco carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) Carcinogenesis, April 1, 2008; 29(4): 866 - 874. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Wang, H. Liu, Z. Zhang, M. R. Spitz, and Q. Wei Association of Genetic Variants of O6-Methylguanine-DNA Methyltransferase with Risk of Lung Cancer in Non-Hispanic Whites Cancer Epidemiol. Biomarkers Prev., December 1, 2006; 15(12): 2364 - 2369. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. S. Mijal, S. Kanugula, C. C. Vu, Q. Fang, A. E. Pegg, and L. A. Peterson DNA Sequence Context Affects Repair of the Tobacco-Specific Adduct O6-[4-Oxo-4-(3-pyridyl)butyl]guanine by Human O6-Alkylguanine-DNA Alkyltransferases. Cancer Res., May 1, 2006; 66(9): 4968 - 4974. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Zwickenpflug, S. Tyroller, and E. Richter METABOLISM OF MYOSMINE IN WISTAR RATS Drug Metab. Dispos., November 1, 2005; 33(11): 1648 - 1656. [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 |
, NNK; 
, AMMN alone; •, AMMN + 4.2 µmol of NNKOAc; 
, AMMN + 2.5 µmol of O6-bG; 
, MNU alone; 
, MNU + 4.2 µmol of NNKOAc; 
, MNU + 2.5 µmol of O6-bG.