Plasma antioxidant vitamins, chronic hepatitis B virus infection and urinary aflatoxin B1-DNA adducts in healthy males

Plasma antioxidant vitamins, chronic hepatitis B virus infection

and urinary aflatoxin B

1

–DNA adducts in healthy males

Ming-Whei Yu

1

_{, Yi-Ching Chiang}

1

_{, Ju-Ping Lien}

1

_and

_{factors for the development of hepatocellular carcinoma (HCC)}

Chien-Jen Chen

1,2

_{in Taiwan and other parts of southeastern Asia and Africa}

(1–5). AFB

1

is carcinogenic in many animal species and its

School of Public Health, College of Public Health, National Taiwan

covalent binding to hepatic DNA has been shown to be a

University, Taipei 100, Taiwan and1Graduate Institute of Epidemiology,

critical step in hepatocarcinogenesis (6).

College of Public Health, National Taiwan University, No. 1 Jen-Ai Rd.

Sec. 1, Taipei 100, Taiwan

The major AFB

1

–DNA adduct formed in vivo is AFB

1

-N

7

-guanine (7). Treatment of AFB

1

to rats resulted in the urinary

2_{To whom correspondence should be addressed}

excretion of AFB

1

-N

7

-guanine adducts. The adduct excretion

Epidemiological evidence indicates that aflatoxin B

1

(AFB

1

)

following AFB

1

administration was shown to be proportional

intake is associated with an increased risk of hepatocellular

to the total amount of AFB

1

-N

7

-guanine initially formed in

carcinoma (HCC). The hepatocarcinogenesis is initiated by

the hepatic DNA (8). Although the presence of AFB

1

–DNA

covalent binding of AFB

1

to cellular DNA. To determine

adducts in liver specimens is readily detected by

immunohisto-whether nutritional factors and hormonal status may

influ-chemical methods (3), liver tissues are usually difficult to

ence the binding of AFB

1

to hepatic DNA, a cross-sectional

obtain. Measurement of the adduct levels in urine provides a

study was performed on a total of 42 male asymptomatic

non-invasive means of estimating the degree of AFB

1

binding

hepatitis B surface antigen (HBsAg) carriers and 43 male

to hepatic DNA. It has been used as an intermediate endpoint

non-carriers in a cohort study on the multistage

develop-in cancer prevention trials for assessdevelop-ing the efficacy of

chemo-ment of HCC in Taiwan. The major AFB

1

–DNA adduct

preventive agents that may have an inhibitory effect at early

in vivo, AFB

1

-N

7

-guanine, was measured by

high-perform-stages of AFB

1

-induced hepatocarcinogenesis (6,9). The

rela-ance liquid chromatography in urine. Urinary AFB

1

-N

7

-tionship between urinary AFB

1

-N

7

-guanine adducts and HCC

guanine was detectable in 40% of the subjects. HBsAg

risk was recently investigated in a prospective study and the

carriers had a higher detection rate of urinary AFB

1

–

presence of the DNA adduct in urine has been shown to be a

DNA adducts than non-carriers and the difference was

significant predictor for the cancer risk (4).

statistically significant after multivariate adjustment. After

In addition to chronic HBV infection and aflatoxin exposure,

taking into account the total AFB

1

urinary metabolite

our previous epidemiological studies have linked many risk

level, chronic HBsAg carrier status, and other potential

factors, including hepatitis C virus (HCV), alcohol drinking,

confounders, plasma levels of cholesterol,

α-tocopherol,

cigarette smoking, elevated serum level of endogenous

and

α- and β-carotene were positively associated with

testosterone, low serum retinol level and/or vegetable

consump-the detection rate of consump-the AFB

1

–DNA adducts in a

dose-tion, as well as genetic susceptibility, to the development of

dependent manner, whereas plasma lycopene level was

HCC (10–15). There may be a complex interaction among

inversely related to the presence of the adducts in urine.

multiple HCC risk factors at various stages of

hepatocarcino-The association of urinary AFB1–DNA adducts with the

genesis. We previously examined the effect of chronic HBV

plasma levels of cholesterol,

α-tocopherol, lycopene, and

infection, sociodemographic characteristics, and habits of

cigar-α- and β-carotene was observed at both low and high

ette smoking and alcohol drinking on AFB

1

–DNA adduct

exposure levels of AFB

1

. There was a synergistic interaction

of plasma

α-tocopherol with α- and β-carotene on the

formation (16). Age and habits of cigarette smoking and

adduct levels. No association with the adducts was found

alcohol drinking were found to be positively associated with

for plasma levels of retinol and testosterone. This study

the adduct levels. Whether other factors may influence the

demonstrated different associations of antioxidant vitamins

binding of AFB

1

to hepatic DNA have yet to be identified.

with AFB

1

–DNA adduct formation. The data consistent

Synthetic antioxidants have been reported to inhibit AFB

1

–

with our previous finding in cultured woodchuck hepato-

DNA adduct formation and thus the tumorigenicity of AFB

1

cytes that

α-tocopherol and β-carotene enhanced AFB

1

–

in rats (17,18). Experimental studies have also suggested that

DNA adduct formation suggest that prospective investi-

dietary antioxidants such as vitamin C, vitamin E and several

gation of the relationship between plasma micronutrients

carotenoids may suppress chemically induced carcinogenesis

and risk of AFB

1

-related HCC is warranted.

through a variety of mechanisms, including reduction of

oxidative DNA damage, modulation of

carcinogen-metaboliz-ing enzymes, and/or trappcarcinogen-metaboliz-ing of electrophilic forms of

carcino-Introduction

gens (19–21). However, vitamin E and

β

-carotene were

observed to enhance AFB

1

–DNA adduct formation in our

Ingestion of aflatoxin B

1

(AFB

1

*) and chronic infection with

previous in vitro study on cultured woodchuck hepatocytes

hepatitis B virus (HBV) have been identified as major risk

(22). In addition to the environmental agents, hormonal status

may also influence the AFB

1

binding to DNA. The pathway

*Abbreviations: HCC, hepatocellular carcinoma; HBV, hepatitis B virus;

_{leading to metabolic activation of AFB}

1

was reported to be

AFB1, aflatoxin B1; AFB1-N7-guanine, 8,9-dihydro-8-(N7

-guanyl)-9-hydroxy-more active in male than in female rats (23). The role of

aflatoxin B1; HBsAg, hepatitis B surface antigen; HPLC, high-performance

remains to be explored. This study was carried out to elucidate

two HBsAg non-carriers were positive for anti-HCV. AFM

1

the potential roles of endogenous testosterone and nutritional

was observed as the most abundant AFB

1

metabolite excreted.

factors including common antioxidants at the early stage

The mean level of total AFB

1

urinary metabolites (including

of AFB

1

-induced hepatocarcinogenesis by investigating the

AFB

1

, AFM

1

, AFP

1

, and AFB

1

-N

7

-guanine) was 9.47 ng/ml

covalent binding of AFB

1

to hepatic DNA. Urinary AFB

1

-N

7

-

(

616.89 SD). Forty percent (34/85) of the urine samples

guanine was used as a biomarker for assessing AFB

1

–DNA

contained a detectable level of AFB

₁

-N

7

-guanine adducts. The

binding in the liver.

_{mean level of AFB}

1

-N

7

-guanine adduct in the positive urine

samples was 0.52 ng/ml (

61.02 SD). More than 90% of the

Materials and methods

study subjects had a urinary AFB

1

-N

7

-guanine adduct level

Study subjects

_{,10% of the total AFB1}

_{metabolites excreted in the urine.}

The subjects of the present study have been described in a previous publication

In univariate analysis, the odds ratios associated with the

that examined associations of multiple factors with urinary levels of AFB1–

positivity of urinary AFB

1

-N

7

-guanine were elevated with

DNA adducts (16). Initially, a total of 43 male asymptomatic HBsAg carriers

and 43 male HBsAg non-carriers were included in the study. HBsAg non-

_{increasing plasma levels of}

α

_{-tocopherol (P for trend}

5 0.005)

carriers were matched with the carriers on age (65 years) and time of

_and

_β

_{-carotene (P for trend}

_{5 0.018). No significant association}

biospecimen collection (63 months). They were randomly selected from the

with the AFB

1

–DNA adducts in urine was observed for chronic

cohort members without HCC in a prospective study on the multifactorial

etiology of multistage hepatocarcinogenesis in Taiwan. The cohort character-

HBsAg carrier status and plasma levels of testosterone and

istics and method of follow-up have been described in a previous study (15).

_{other nutrients. Since there were correlations among various}

The urine and blood samples from study subjects were consecutively collected

plasma nutrients, logistic regression analysis was adopted

between August 1988 and June 1992. Urine samples were kept frozen at

to estimate multivariate-adjusted ORs associated with the

–30°C and blood samples were kept at –70°C until they were used in the

present analyses. At the time of urine and blood collection, each study subject

_{positivity of urinary AFB}

₁

_{–DNA adducts. After taking into}

was also personally interviewed to obtain information related to habits of

_{account other potential confounders including the total AFB}

1 cigarette smoking and alcohol drinking, dietary pattern and health history. In

urinary metabolites, plasma

α

-tocopherol level retained its

the present analysis, study subjects were 42 HBsAg carriers and 43 HBsAg

non-carriers whose plasma samples were sufficient for analysis of micronutri-

significance. Chronic HBsAg carrier status and plasma

choles-ents and testosterone. All the chronic HBsAg carriers included in this study

_{terol level were also significant factors associated with the}

remained asymptomatic throughout a 5-year follow-up period, except two

_{positivity of urinary AFB}

1

–DNA adducts. Increase in the

who were subsequently affected with liver cirrhosis. The mean duration of

detection rate of the adducts was related to the plasma

α

-biospecimen storage prior to testing is 4.4 years (range: 2.4–6.3 years). There

was no significant difference in seasons of biospecimen collection between

carotene at a borderline significant level (high versus low

HBsAg carriers and non-carriers.

_{tertile, P}

_{5 0.0548). Although a positive association was also}

Laboratory analyses

_{observed between plasma}

_β

_{-carotene levels and urinary AFB}

1

–

Serum HBsAg was assayed using a radioimmunoassay (Abbott Laboratories,

_{DNA adducts, this association was not statistically significant.}

North Chicago, IL). All but six study subjects were also tested for antibodies

Plasma lycopene level was inversely associated with the

against HCV (anti-HCV). Anti-HCV was examined by a second-generation

enzyme immunoassay (Abbott Laboratories, North Chicago, IL). The urine

positivity of the adducts. No significant association with

samples were assayed for levels of AFB1, two of its main metabolites (AFM1

_{urinary AFB}

1

–DNA adducts was found for plasma retinol and

and AFP1) and the AFB1-N7-guanine adducts. Urinary AFB1metabolites were

testosterone. Because plasma levels of

α

-carotene and

β

-quantified by high-performance liquid chromatography (HPLC) as described

previously (16). In short, aflatoxin metabolites were extracted from a 5-ml

carotene were highly correlated (r

5 0.64) and the two

aliquot of urine. Analysis of various aflatoxin metabolites was done on an

_{carotenoids associated with urinary AFB}

1

–DNA adducts in a

HPLC gradient liquid chromatograph with a Waters model 470 fluorescence

_{similar manner, the levels of the two carotenoids were therefore}

detector (Waters Associates, Milford, MA). The HPLC column was a C18

combined in the analysis. This made almost no change in the

10-µm (3.9 mm inside diameter3 30 cm length)µBondpak column (Waters).

For analysis of AFM1, AFB1-N7-guanine, and AFB1, chromatographic separa-

ORs of the positivity of urinary AFB

1

–DNA adducts with

tion was achieved by elution for 12 min with 15% acetonitrile followed by

_{other variables. There was a significant positive association}

22% acetonitrile in water. The flow rates were 1.5, 0.8, 0.3 and 1.0 min/ml

between the adducts and plasma levels of

α

- and

β

-carotene

at 0–12, 13–21, 22–35 and 36–41 min, respectively. Elutes were measured by

(Table I).

fluorescence detection with 365 nm excitation and 430 nm emission wavelength

for AFM1, and 500 nm emission for AFB1and AFB1-N7-guanine. The HPLC

Table II shows the detection rate of AFB

₁

-N

7

-guanine in

analysis for AFP1 was done with a 30-min elution with 12% acetonitrile

_{urine by levels of total AFB}

1

urinary metabolites and plasma

followed by a 12–22% acetonitrile linear gradient generated over 12 min,

nutrients. Due to the sample size not being large, study subjects

then elution at 22% acetonitrile. The flow rates were 1.5, 0.3, 1.5, 0.3, 1.0

and 0.3 ml/min at 0–17, 18–19, 20–29, 30–41, 42–44 and 45–52 min,

were categorized into only two groups according to the median

respectively. AFP1was measured by monitoring the fluorescence emission at

_{of their distribution. The association of urinary AFB}

1

–DNA

500 nm with the excitation wavelength at 365 nm. Authentic aflatoxin

adducts with the plasma levels of cholesterol,

α

-tocopherol,

standards were used to determine chromatographic retention times. All

aqueous mobile phases before use were adjusted by orthophosphoric acid and

and

α

- and

β

-carotene was observed at both low and high

triethylammonium formate buffer to pH 3.0. Plasma micronutrients were also

_{exposure levels of AFB}

1

. The multivariate-adjusted ORs

associ-measured by HPLC using a modification of the method by Miller and Yang

_{ated with the positivity of the AFB}

1

–DNA adducts in urine

(24). Plasma testosterone levels were determined by a radioimmunoassay kit

suggested a strong synergistic interaction of AFB

1

exposure

(Biomerieux, Marcy l’Etoile, France).

with plasma levels of these nutrients. In contrast, lycopene

Statistical methods

Odds ratios (ORs) and their 95% confidence intervals (CIs) were computed

appeared to have an inhibitory effect on AFB

1

–DNA adduct

to examine the associations of the positivity of urinary AFB1-N7-guanine with

formation at low and high AFB

₁

exposure. Interactive effect

various variables. Mantel’s chi-square test for a trend was performed to

_{between plasma}

_α

_{-tocopherol and}

_α

_{- and}

_β

_{-carotene on the}

examine the dose–response relationship for the odds ratios. Logistic regression

adduct levels is depicted in Table III. All the ORs associated

was used to estimate ORs when adjusting covariates.

with the positivity of urinary AFB

1

–DNA adducts in relation

Results

_{to various variables in this study were not materially changed}

when adjustment was also made for storage time of

biospeci-The age range of the subjects was 33 to 66 years, with a mean

Table I. Associations of the positivity of urinary AFB1-N7-guanine adducts with HBsAg carrier status and plasma levels of testosterone and various nutrients

Variable AFB1-N7-guanine Crude OR Multivariate-adjusted OR P

(95% CI)a Negative Positive

HBsAg carrier status

Negative 28 15 1.0 1.0 Positive 23 19 1.54 7.52 (1.51–37.53) 0.0138 Testosterone (ng/ml) ø4.7b _{17 11 1.0 1.0} 4.8–5.9 16 12 1.16 1.89 (0.33–10.77) 0.4739 .5.9 18 11 0.94 0.93 (0.16–5.50) 0.9342 Cholesterol (mg/dl) ø178.7b _{19 10 1.0 1.0} 178.8–216.7 20 9 0.86 0.75 (0.15–3.71) 0.7287 .216.7 12 15 2.38 6.99 (1.24–39.44) 0.0276 Retinol (µg/dl) ø43.0b _{18 10 1.0 1.0} 43.1–58.0 19 9 0.85 0.57 (0.10–3.13) 0.5169 .58.0 14 15 1.93 1.86 (0.32–10.91) 0.4911 α-Tocopherol (µg/dl) ø621.0b _{22 7 1.0}c _1.0 621.1–829.0 17 9 1.66 1.88 (0.28–12.71) 0.5162 .829.0 12 18 4.71 11.20 (1.47–85.59) 0.0199 Lycopene (µg/dl) ø14.3b _{20 8 1.0 1.0} 14.4–21.7 15 13 2.17 0.22 (0.02–2.68) 0.2343 .21.7 16 13 2.03 0.04 (0.002–0.79) 0.0342 α-Carotene1β-carotene (µg/dl) ø13.4b _{21 7 1.0}c _1.0 13.5–22.7 17 10 1.76 7.77 (0.65–93.56) 0.1064 .22.7 13 17 3.92 33.12 (1.58–695.66) 0.0243

a_{All the variables shown in the table, total AFB}

1urinary metabolites, age and habits of cigarette smoking and alcohol drinking were also included in the

multiple logistic regression model.

b_{Categorized according to the tertile distribution of subjects.} c_{Test for trend was statistically significant.}

Table II. Detection rate of urinary AFB1-N7-guanine adducts by total AFB1metabolites in urine and plasma levels of selected nutrients

Variable Total no. Detection rate of Crude OR Multivariate-adjusted OR P

AFB1-N7-guanine (%) (95% CI)a

Total urinary AFB1/α-tocopherol

Low/lowb _{22 4.5 1.0}c _1.0

Low/high 20 45.0 17.18 22.56 (1.64–309.74) 0.0197

High/low 21 47.6 19.09 41.83 (2.99–584.57) 0.0055

High/high 22 63.6 36.75 73.19 (4.56–1173.76) 0.0024

Total urinary AFB1/α-carotene1β-carotene

Low/low 24 12.5 1.0c 1.0

Low/high 18 38.9 4.45 8.63 (1.03–72.34) 0.0470

High/low 19 47.4 6.30 8.10 (1.25–52.29) 0.0280

High/high 24 62.5 11.67 27.68 (3.33–230.41) 0.0021

Total urinary AFB1/lycopene

Low/high 21 28.6 1.0c _1.0

Low/low 21 19.0 0.59 3.02 (0.27–34.35) 0.3725

High/high 21 57.1 3.33 3.45 (0.62–19.02) 0.1556

High/low 22 54.5 3.00 32.55 (3.22–329.40) 0.0032

Total urinary AFB1/cholesterol

Low/low 23 21.7 1.0c _1.0

Low/high 19 26.3 1.29 1.54 (0.21–11.31) 0.6701

High/low 21 47.6 3.27 4.04 (0.69–23.52) 0.1204

High/high 22 63.6 6.30 11.03 (1.78–68.16) 0.0098

a_{Age, HBsAg carrier status, habits of cigarette smoking and alcohol drinking and plasma levels of testosterone and other nutrients were also included in the}

multiple logistic regression model.

b_{High and low level was categorized according to the median of study subjects.} c_{Test for trend was statistically significant.}

Table III. Detection rate of urinary AFB1-N7-guanine adducts by plasma levels ofα-tocopherol andα-carotene plusβ-carotene

α-Tocopherol/α-carotene Total no. Detection rate of Crude OR Multivariate-adjusted OR P

1β-carotene AFB1-N7-guanine (%) (95% CI)a

Low/lowb _{25 20.0 1.0}c _1.0

Low/high 18 33.3 2.00 8.19 (0.74–90.77) 0.0867

High/low 18 38.9 2.55 7.30 (0.96–55.30) 0.0543

High/high 24 66.7 8.00 29.43 (2.94–294.27) 0.0040

a_{Age, total AFB}

1urinary metabolites, habits of cigarette smoking and alcohol drinking, HBsAg carrier status and plasma levels of testosterone and other

nutrients were also included in the multiple logistic regression model.

b_{High and low level was categorized according to the median of study subjects.} c_{Test for trend was statistically significant.}

Discussion

adducts with DNA, than female rats (23). However, variation

in plasma testosterone levels was not correlated with AFB

1

–

The data on urinary aflatoxins were based on spot urine

DNA adducts in this study.

samples. Although no adjustment was made for urine

concen-Low serum retinol level has been associated with the

tration in the calculation of AFB

1

-N

7

-guanine excretion in this

_{development of various human cancers, including HCC (14).}

study, this may not pose significant problems for investigation

Liver fractions from rats with vitamin A-deficiency formed a

of the associations between various variables and the positivity

higher level of DNA adducts by AFB1

(32). Our previous

of urinary AFB

1

–DNA adducts, because total quantity of AFB

1

in vitro study with cultured woodchuck hepatocytes

demon-metabolites excreted in urine was included in the multivariate

strated a potent inhibitory effect of retinol on AFB

1

–DNA

analyses as a covariate for adjusting its effect.

adduct formation (22). However, we failed to find a significant

Epidemiological studies suggest that HBV and AFB

1

may

association of AFB

1

–DNA adducts with plasma retinol levels

exert a synergistic effect on the development of HCC (4). The

in this cross-sectional study. The reason for the discrepancy

mechanisms responsible for this interaction remain to be

between our in vitro and human study is unclear but may be

elucidated. In woodchucks, chronic infection with woodchuck

explained by the difference in the effect of retinol between

hepatitis virus, a virus similar in characteristics to HBV,

species. On the other hand, since this study was conducted in

produced an enhanced metabolic activation of chemical

carci-a well-nourished populcarci-ation with low prevcarci-alence of vitcarci-amin

nogen, including AFB

1

(25). However, epidemiological studies

A-deficiency, whether a more striking association between

of the relationship between HBsAg carrier status and the

urinary AFB

1

–DNA adducts and plasma retinol level may

formation of AFB

1

–DNA adducts, using AFB

1

-albumin adduct

be shown up in malnourished populations requires further

or urinary AFB

1

-N

7

-guanine as the surrogate dosimeter for

investigations.

estimating hepatic DNA binding by AFB

1

, have been

inconsist-Lycopene,

α

-carotene and

β

-carotene are carotenoids with

ent (16,26–28). We have previously demonstrated that the

a similar chemical structure. They are antioxidant contents of

detection rate of urinary AFB

1

-N

7

-guanine adducts was higher

vegetables and fruits (21). Experimental and epidemiological

in HBsAg carriers than in non-carriers, but this association

studies on the potential role of

α

-carotene and lycopene

was not statistically significant in univariate analysis or in a

in carcinogenesis are limited (33–37). The influence of

β

-multivariate analysis including the total AFB

1

urinary

metabol-carotenoids on susceptibility to various forms of cancers have

ites, age, and habits of cigarette smoking and alcohol drinking,

been evaluated in a number of epidemiological studies (38–

as covariates (16). In the present study, HBsAg carrier status

40). Although the majority of results have shown that high

was strongly associated with urinary AFB

1

–DNA adduct levels

intake and/or high serum level of

β

-carotene were associated

after adjustment was made for plasma nutrients and other

with a reduced risk of cancer at several sites (38,39), some

potential confounders. The discrepancy between our two

stud-data are conflicting and the significance of

β

-carotene in

ies may be due to the difference in the control of plasma

carcinogenesis remains unclear (40). The most marked finding

nutrients that were significantly associated with the AFB

1

–

of this study was the strong and extremely diverse associations

DNA adduct levels. The possible biological mechanisms for

of various carotenoids with AFB

1

-N

7

-guanine adducts in urine.

HBV involvement in the genesis of HCC have been extensively

α

-Carotene and

β

-carotene were positively associated with the

reviewed (1). This study provides evidence that chronic HBV

DNA adducts, while the adducts appeared to reduce with

infection may interact with AFB

1

intake in the initiation

increasing plasma levels of lycopene. Difference in the ability

process of hepatocarcinogenesis and thus result in an increased

of diverse carotenoids to suppress the development of

spontan-risk of AFB

1

-related HCC.

eous HCC and chemically-induced neoplastic transformation,

HCC is two to three times more frequent in men than in

enhance gap junctional communication and regulate gene

women (1). The marked sex difference in susceptibility to

expression have been shown in experimental studies (34–

hepatocarcinogenesis was also observed in various animal

37). Although the underlying mechanism for the molecular

models (29–31). A relationship between elevated serum

testo-specificity of each of the carotenoids to influence AFB

1

–DNA

sterone level and HCC risk in humans has been documented

adduct formation is unclear, this study provides additional

(12). In experimental studies, the mechanism for the action of

insight into the complexity of the biological function of

testosterone in hepatocarcinogenesis may be through its effects

carotenoids.

on promotion of the growth of tumor and/or modulation of

α

-Tocopherol is among the most potent antioxidants from

the activity of enzymes involved in metabolism of

hepatocarci-natural source (21,36). Several experimental and

epidemiolog-nogens (23,30). For a given level of AFB

1

exposure, male rats

Groopman,J.D. (1986) Modulation of aflatoxin metabolism, aflatoxin-N7

-(41). The potential use of

α

-tocopherol in cancer

chemopreven-guanine formation and hepatic tumorigenesis in rats fed ethoxyquin: role

tion has been evaluated in lung cancer, oral leukoplakia and

of induction of glutathione S-transferases. Cancer Res., 46, 3924–3931.

colorectal polyps. However, epidemiologic evidence for its

_{7. Croy,R.G., Essigmann,J.M., Reinhold,V.N. and Wogan,G.N. (1978)}

chemopreventive action is inconsistent (40,42,43). In this

Identification of the principal aflatoxin B1–DNA adduct formed in vivo in

rat liver. Proc. Natl Acad. Sci. USA, 75, 1745–1749.

study, the detection rate of the AFB

1

–DNA adducts in urine

8. Bennett,R.A., Essigmann,J.M. and Wogan,G.N. (1981) Excretion of an

was significantly elevated with increasing plasma levels of

α

-aflatoxin-guanine adduct in the urine of aflatoxin B1-treated rats. Cancer

tocopherol. There was a synergistic interaction between

α

-Res., 41, 650–654.

tocopherol and

α

- and

β

-carotene on the DNA adduct levels.

_{9. Groopman,J.D., Wogan,G.N., Roebuck,B.D. and Kensler,T.W. (1994)}

The data demonstrating a positive association between

_{Molecular biomarkers for aflatoxins and their application to human cancer}

urinary AFB

1

–DNA adducts and the plasma levels of

α

-

prevention. Cancer Res., 54, 1907s–1911s.

10. Yu,M.W., You,S.L., Chang,A.S., Lu,S.N., Liaw,Y.F. and Chen,C.J. (1991)

tocopherol and

β

-carotene are in accordance with our previous

Association between hepatitis C virus antibodies and hepatocellular

in vitro study in which

α

-tocopherol and

β

-carotene enhanced

carcinoma in Taiwan. Cancer Res., 51, 5621–5625.

the adduct formation in cultured woodchuck hepatocytes (22).

_{11. Chen,C.J., Yu, M.W., Wang,C.J., Huang,H.Y. and Lin,W.C. (1993) Multiple}

Urinary AFB

1

–DNA adduct levels can be influenced by carci-

risk factors of hepatocellular carcinoma: a cohort study of 13737 male

nogen metabolism and DNA repair. However, it was shown

adults in Taiwan. J. Gastroenterol. Hepatol., 8, 83s–87s.

12. Yu,M.W. and Chen,C.J. (1993) Elevated serum testosterone levels and risk

that a supplement with antioxidant vitamins had no significant

of hepatocellular carcinoma. Cancer Res., 53, 790–794.

effect on DNA repair activity (44,45). Thus, our findings cast

13. Yu,M.W., Chen,C.J., Luo,J.C., Brandt-Rauf,P.W., Carney,W.P. and

doubts on the potential role of certain dietary antioxidants,

Santella,R.M. (1994) Correlations of chronic hepatitis B virus infection

such as

α

-tocopherol,

α

-carotene and

β

-carotene, in prevention

_{and cigarette smoking with elevated expression of neu oncoprotein in the}

of AFB

1

-induced DNA damage. However, it is premature to

development of hepatocellular carcinoma. Cancer Res., 54, 5106–5110.

14. Yu,M.W., Hsieh,H.H., Pan,W.H., Yang,C.S. and Chen,C.J. (1995) Vegetable

conclude that these antioxidant vitamins have an adverse effect

consumption, serum retinol level and risk of hepatocellular carcinoma.

on the development of HCC. The pathogenesis of cancer is a

Cancer Res., 55, 1301–1305.

multistage process including DNA alteration and cell

prolifera-15. Yu,M.W., Gladek-Yarborough,A., Chiamprasert,S., Santella,R.M.,

tion. Vitamins may have multiple biological mechanisms to

_{Liaw,Y.F. and Chen,C.J. (1995) Cytochrome P-450 2E1 and glutathione}

inhibit or retard one or more stages of carcinogenesis. The

_{S-transferase M1 polymorphisms and susceptibility to hepatocellular}

carcinoma. Gastroenterology, 109, 1266–1273.

association of HCC risk with plasma levels of various

antioxid-16. Yu,M.W., Lien,J.P., Liaw,Y.F. and Chen,C.J. (1996) Effects of multiple

ant vitamins remains to be elucidated. On the other hand, our

risk factors for hepatocellular carcinoma on formation of aflatoxin B1–

prospective study has demonstrated that an increased risk of

DNA adducts. Cancer Epidemiol. Biomark. Prev., 5, 613–619.

HCC is associated with low vegetable intake (14). Vegetables

_{17. Jhee,E.C., Ho,L.L. and Lotlikar,P.D. (1988) Effect of butylated}

contain a wide variety of phytochemicals with the potential

_{hydroxyanisole pretreatment on in vitro hepatic aflatoxin B}

1–DNA binding

to modulate carcinogenesis (21). Whether constituents in

and aflatoxin B1–glutathione conjugation in rats. Cancer Res., 48, 2688–

2692.

vegetables other than

α

-carotene and

β

-carotene may be

18. Mandel,H.G., Manson,M.M., Judah,D.J. and Simpson,J.L. (1987)

important protective factors for AFB

1

-induced HCC required

Metabolic basis for the protective effect of antioxidant ethoxyquin on

further studies.

aflatoxin B1hepatocarcinogenesis in the rat. Cancer Res., 47, 5218–5223.

To date we have not found any reports on the effect of

_{19. Duthie,S.J., Ma,A., Ross,M.A. and Collins,A.R. (1996) Antioxidant}

cholesterol on AFB

1

–DNA adduct formation. The finding of

supplementation decreases oxidative DNA damage in human lymphocytes.

Cancer Res., 56, 1291–1295.

the positive association of plasma cholesterol with urinary

20. Van Lieshout,E.M.M., Peters,W.H.M. and Jansen,J.B.M.J. (1996) Effect

AFB

1

–DNA adducts in this study warrants further

investi-of oltipraz,α-tocopherol,β-carotene and phenethylisothiocyanate on rat

gation.

oesophageal, gastric, colonic and hepatic glutathione, glutathione S-transferase and peroxidase. Carcinogenesis, 17, 1439–1445.

Acknowledgements

21. Frei,B. (1994) Natural Antioxidants in Human Health and Disease.

Academic Press, San Diego, California. This study was supported by grants from the National Science Council (NSC

22. Yu,M.W., Zhang,Y.J., Blaner,W.S. and Santella,R.M. (1994) Influence of 84-2331-B-002-183, NSC 85-2331-B-002-264) and National Institute of

vitamins A, C and E andβ-carotene on aflatoxin B1binding to DNA in

Health, Department of Health, Executive Yuan (DOH86-HR-627) of the

woodchuck hepatocytes. Cancer, 73, 596–604. Republic of China.

23. O’Brien,K., Mose,E., Judah,D. and Neal,G. (1983) Metabolic basis of the species difference to aflatoxin B1-induced hepatotoxicity. Biochem.

References

Biophys. Res. Commun., 114, 813–827.

24. Miller,K.W. and Yang,C.S. (1985) An isocratic high-performance liquid 1. Yu,M.W. and Chen,C.J. (1994) Hepatitis B and C viruses in the development

chromatography method for the simultaneous analysis of plasma retinol, of hepatocellular carcinoma. Crit. Rev. Oncol. Hematol., 17, 71–91.

α-tocopherol and various carotenoids. Anal Biochem., 145, 21–26. 2. Chen,C.J., Yu,M.W., Liaw,Y.F., Wang,L.W., Chiamprasert,S., Matin,F.,

25. De Flora,S., Hietanen,E., Bartsch,H., Camoirano,A., Izzotti,A., Hirvonen,A., Bell,D.A. and Santella,R.M. (1996) Chronic hepatitis B

Bagnasco,M. and Millman,I. (1989) Enhanced metabolic activation of carriers with null genotypes of glutathione S-transferase M1 and T1

chemical hepatocarcinogens in woodchucks infected with hepatitis B virus. polymorphisms who are exposed to aflatoxin are at increased risk of

Carcinogenesis, 10, 1099–1106. hepatocellular carcinoma. Am. J. Hum. Genet., 59, 128–134.

26. Allen,S.J., Wild,C.P., Wheeler,J.G., Riley,E.M., Montesano,R., Bennett,S., 3. Chen,C.J., Zhang,Y.J., Lu,S.N. and Santella,R.M. (1992) Aflatoxin B1

Whittle,H.C., Hall,A.J. and Greenwood,B.M. (1992) Aflatoxin exposure, DNA adducts in smeared tumor tissue from patients with hepatocellular

malaria and hepatitis B infection in rural Gambian children. Trans. R. Soc. carcinoma. Hepatology, 16, 1150–1155.

Trop. Med. Hyg., 86, 426–430. 4. Qian,G.S., Ross,R.K., Yu,M.C., Yuan,J.M., Gao,Y.T., Henderson,B.E.,

27. Wild,C.P., Fortuin,M., Donato,F., Whittle,H.C., Hall,A.J., Wolf,C.R. and Wogan,G.N. and Groopman,J.D. (1994) A follow-up study of urinary

Montesano,R. (1993) Aflatoxin, liver enzymes and hepatitis B virus markers of aflatoxin exposure and liver cancer risk in Shanghai, People’s

infection in Gambian children. Cancer Epidemiol. Biomark. Prev., 2, Republic of China. Cancer Epidemiol. Biomark. Prev., 3, 3–10.

555–561. 5. Van Rensburg,S.J., Cook-Mozaffari,P., Van Schalkwyk,D.J., Van Der

28. Groopman,J.D., Hall,A.J., Whittle,H., Hudson,G.J., Wogan,G.N., Watt,J.J., Vincent,T.J. and Purchase,I.F. (1985) Hepatocellular carcinoma

Montesano,R. and Wild,C.P. (1992) Molecular dosimetry of aflatoxin-N7

-and dietary aflatoxin in Mozambique -and Transkei. Br. J. Cancer, 51,

guanine in human urine obtained in the Gambia, west Africa. Cancer 713–726.

29. Sawaki,M., Enomoto,K., Takahashi,H., Nakajima,Y. and Mori,M. (1990) Phenotype of preneoplastic and neoplastic liver lesions during spontaneous liver carcinogenesis of LEC rats. Carcinogenesis, 11, 1857–1861. 30. Moore,M.R., Drinkwater,N.R., Miller,E.C., Miller,J.A. and Pitot,H.C.

(1981) Quantitative analysis of the time-dependent development of glucose-6-phosphatase-deficient foci in the livers of mice treated neonatally with diethylnitrosamine. Cancer Res., 41, 1585–1593.

31. Kim,C.M., Koike,K., Saito,I., Miyamura,T. and Jay,G. (1991) HBx gene of hepatitis B virus induces liver cancer in transgenic mice. Nature, 351, 317–320.

32. Bhattacharya,R.K., Prabhu,A.L. and Aboobaker,V.S. (1989) In vivo effect of dietary factors on the molecular action of aflatoxin B1: role of vitamin

A on the catalytic activity of liver fractions. Cancer Lett., 44, 83–88. 33. Ziegler,R.G., Colavito,E.A., Hartge,P., McAdams,M.J., Schoenberg,J.B.,

Mason,T.J., Fraumeni,J.F. Jr (1996) Importance ofα-carotene,β-carotene and other phytochemicals in the etiology of lung cancer. J. Natl Cancer Inst., 88, 612–615.

34. Murakoshi,M., Nishino,H., Satomi,Y. et al. (1992) Potent preventive action ofα-carotene against carcinogenesis: spontaneous liver carcinogenesis and promoting stage of lung and skin carcinogenesis in mice are suppressed more effectively by α-carotene than by β-carotene. Cancer Res., 52, 6583–6587.

35. Bertram,J.S., Pung,A., Churley,M., Kappock IV,T.J., Wilkins,L.R. and Cooney,R.V. (1991) Diverse carotenoids protect against chemically induced neoplastic transformation. Carcinogenesis, 12, 671–678.

36. Zhang,L.X., Cooney,R.V. and Bertram,J.S. (1991) Carotenoids enhance gap junctional communication and inhibit lipid peroxidation in C3H/10T1/ 2 cells: relationship to their cancer chemopreventive action. Carcinogenesis, 12, 2109–2114.

37. Zhang,L.X., Cooney,R.V. and Bertram,J.S. (1992) Carotenoids up-regulate Connexin43 gene expression independent of their provitamin A or antioxidant properties. Cancer Res., 52, 5707–5712.

38. Byers,T. and Perry,G. (1992) Dietary carotenes, vitamin C and vitamin E as protective antioxidants in human cancers. Annu. Rev. Nutr., 12, 139–159. 39. Comstock,G.W., Bush,T.L. and Helzlsouer,K. (1992) Serum retinol, beta-carotene, vitamin E and selenium as related to subsequent cancer of specific sites. Am. J. Epidemiol., 135, 115–121.

40. The Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study Group. (1994) The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. N. Engl. J. Med., 330, 1029–1035.

41. Bertram,J.S., Kolonel,L.N. and Meyskens,F.L. Jr (1987) Rationale and strategies for chemoprevention of cancer in humans. Cancer Res., 47, 3012–3031.

42. DeCosse,J.J., Miller,H.H. and Lesser,M.L. (1989) Effects of wheat fiber and vitamins C and E on rectal polyps in patients with familial adenomatous polyposis. J. Natl Cancer Inst., 81, 1290–1297.

43. Benner,S.E., Wargovich,M.J., Lippman,S.M., Fisher,R., Velasco,M.,Winn,R.J. and Hong,W.K. (1994) Reduction in oral mucosa micronuclei frequency following α-tocopherol treatment of oral leukoplakia. Cancer Epidemiol. Biomark. Prev., 3, 73–76.

44. Wei,Q., Matanoski,G.M., Farmer,E.R., Hedayati,M.A. and Grossman,L. (1994) DNA repair related to multiple skin cancers and drug use. Cancer Res., 54, 437–440.

45. Hu,J.J., Roush,G.C., Berwick,M., Dubin,N., Mahabir,S., Chandiramani,M. and Boorstein,R. (1996) Effects of dietary supplementation ofα-tocopherol on plasma glutathione and DNA repair activities. Cancer Epidemiol. Biomark. Prev., 5, 263–270.

Received on November 4, 1996; revised on February 7, 1997; accepted on February 20, 1997