Temporal patterns of DNA adduct formation and glutathione S-transferase activity in the testes of rats fed aflatoxin B1: A comparison with patterns in the liver

Article abstract of DOI:10.1002/(SICI)1098-2280(1999)33:4<293::AID-EM6>3.0.CO;2-E

Fisher-344 male rats were fed 1.6 ppm of aflatoxin B₁ (AFB₁) continuously and intermittently for several weeks. At various time periods, DNA was isolated from the testes and livers and analyzed for AFB₁-DNA adducts. The ability of the testis to detoxify AFB₁ was also investigated by the glutathione S-transferase (GST) activity assay and compared with that of the liver. The levels of testicular AFB₁-DNA adducts were 2.4 to 8.1 times lower than those of the liver after 4 to 16 weeks of continuous treatment and 2.2 to 46.2 times lower after 8 to 20 weeks of intermittent treatment. The testicular DNA adducts markedly decreased over time. By 16 weeks of continuous and 20 weeks of intermittent exposure, they had decreased 37 and 91%, respectively. In contrast, hepatic AFB₁-DNA adducts increased four- fold from 4 to 16 weeks of continuous treatment but increased at a much slower rate after intermittent exposure. In both the liver and testis, significant levels of AFB₁-DNA adducts persisted for at least 1 month after ending the treatment, suggesting that this type of lesion was poorly repaired. In control rats, the testis showed significantly higher GST activity than the liver. In treated rats, these differences were significant during the first 12 weeks of continuous treatment but not at later times. Tissue-specific differences such as germ-cell depletion and increased testicular detoxification may play an important role in the observed differential pattern of DNA adduct formation between the testis and liver.

Full text of DOI:10.1002/(SICI)1098-2280(1999)33:4<293::AID-EM6>3.0.CO;2-E

Environmental and Molecular Mutagenesis 33:293–302 (1999)
Temporal Patterns of DNA Adduct Formation and
Glutathione S-Transferase Activity in the Testes
of Rats Fed Aflatoxin B1: A Comparison
with Patterns in the Liver
Rene E. Sotomayor,1* Saura Sahu,1 Melissa Washington,1
Dennis M. Hinton,1 and Ming Chou2
1Center for Food Safety and Applied Nutrition, Food and Drug Administration,
Laurel, Maryland
2National Center for Toxicological Research, Food and Drug Administration,
Jefferson, Arkansas
Fisher-344 male rats were fed 1.6 ppm of aflatoxin adducts increased four-fold from 4 to 16 weeks of
B1 (AFB1) continuously and intermittently for sev- continuous treatment but increased at a much
eral weeks. At various time periods, DNA was slower rate after intermittent exposure. In both the
isolated from the testes and livers and analyzed for liver and testis, significant levels of AFB1-DNA ad-
AFB1-DNA adducts. The ability of the testis to de- ducts persisted for at least 1 month after ending the
toxify AFB1 was also investigated by the glutathi- treatment, suggesting that this type of lesion was
one S-transferase (GST) activity assay and com- poorly repaired. In control rats, the testis showed
pared with that of the liver. The levels of testicular significantly higher GST activity than the liver. In
AFB1-DNA adducts were 2.4 to 8.1 times lower treated rats, these differences were significant dur-
than those of the liver after 4 to 16 weeks of ing the first 12 weeks of continuous treatment but
continuous treatment and 2.2 to 46.2 times lower not at later times. Tissue-specific differences such as
after 8 to 20 weeks of intermittent treatment. The germ-cell depletion and increased testicular detoxifi-
testicular DNA adducts markedly decreased over cation may play an important role in the observed
time. By 16 weeks of continuous and 20 weeks of differential pattern of DNA adduct formation between
intermittent exposure, they had decreased 37 and the testis and liver. Environ. Mol. Mutagen. 33:
91%, respectively. In contrast, hepatic AFB1-DNA 293–302, 1999
© 1999 Wiley-Liss, Inc.
Key words: AFB1; testis; liver; DNA adducts; GST; rats
INTRODUCTION
[Gorelick, 1990]. The AFB1-glutathione is then eliminated
through the bile. The conjugation of AFB1-epoxide with
glutathione appears to be an important mechanism of pro-
tection against induction of DNA damage [Gorelick, 1990;
Lotlikar et al., 1984].
Despite the wealth of information on AFB1, little is
known about its damaging effects on DNA in the mamma-
lian testis and its consequences for germ-cell mutagenesis.
AFB1 can cause spermatogenic-cell killing and subsequent
germ-cell depletion in rats [Egbunike et al., 1980; Gopal et
al., 1980; Sahay, 1993]. However, sperm maturation, sperm
production, and capacity for testosterone secretion were
apparently not affected [Egbunike, 1982, 1985]. From these
studies, it can be inferred that AFB1 is likely to cause DNA
damage in spermatogenic cells.
Aflatoxin B1 (AFB1), a secondary metabolite of the fun-
gus Aspergillus flavus, is one of the most powerful food-
borne carcinogens known to man [Wogan, 1973, 1992].
Epidemiological studies in Africa and Asia have shown
strong correlations between hepatocarcinoma and exposure
to AFB1 [Wogan, 1992; Groopman et al., 1991].
AFB1 is activated by the cytochrome P450 mono-oxyge-
nase system to AFB1-8,9-epoxide, which binds covalently
to DNA, RNA, and proteins [Groopman et al., 1991;
Gorelick, 1990]. The primary adduct in DNA and RNA was
identified as 8,9-dihydro-8-(N7-guanyl)9-hydroxy-AFB1
(AFB1-N7-Gua) [Essigman et al., 1977; Lin et al., 1977;
Croy et al., 1978]. Under alkaline conditions, the opened
imidazole ring derivatives of AFB1-N7-Gua are also formed
[Lin et al., 1977; Hertzog et al., 1982]. These adducts appear
to account for 95–98% of the AFB1 residues bound to DNA
[Groopman et al., 1991].
*Correspondence to: Rene E. Sotomayor, Ph.D., Food and Drug Admin-
istration (HFS-509), MOD-1, 8301 Muirkirk Rd., Laurel, MD 20708.
E-mail: res@cfsan.fda.gov
AFB1-8,9-epoxide can also react with glutathione, a
Received 25 February 1999; provisionally accepted 19 March 1999; and in
reaction catalyzed by glutathione S-transferases (GST) final form 4 May 1999
© 1999 Wiley-Liss, Inc.

294
Sotomayor et al.
Fig. 1. Chemical structures of AFB1-N7-Gua, AFB1-FAPyr1 and AFB1-FAPyr2 according to Hertzog et al.
[1982].
Tris-saturated phenol/chloroform/isoamyl alcohol, and anhydrous ethanol
were purchased from Amresco Inc. (Solon, OH).
On the other hand, mammalian testes are rich in gluta-
thione and GST activity [Kraus and Kloft, 1980; Guthen-
berg et al., 1983]. Several authors have shown that gluta-
thione and GST play an important role in protecting the
testis against DNA damage by various genotoxic agents
[Teaf et al., 1987; Mori et al., 1989; Gandy et al., 1990].
This mechanism may also play an important role in protect-
ing the testis against DNA damage by AFB1.
We investigated whether AFB1 causes the formation of
DNA adducts in the rat testis, the temporal pattern of DNA
adduct formation, and the ability of the testis to detoxify
AFB1. For a comparison, these endpoints were also assessed
in the liver, the primary target organ for AFB1 carcinogen-
esis.
Synthesis of AFB1-DNA Standards
The AFB1-8,9-epoxide was prepared from the peroxidation of AFB1 by
dimethyldioxirane as described by Baertschi et al. [1988] and modified by
Chen et al. [1995]. Briefly, approximately 1.5 equivalent weight of freshly
prepared dimethyldioxirane (concentration range 0.05–0.12 mM) was
added to AFB1 dissolved in anhydrous acetone. The reaction mixture was
stirred at room temperature for 20 min and the resulting AFB1-8,9-epoxide
was formed. Solvent and excess dimethyldioxirane were removed by
vacuum evaporation. The residue was redissolved in anhydrous acetone.
The adduct, 8,9-dihydro-8-(7-guanyl)9-hydroxy aflatoxin B1 (AFB1-N7-
Gua), was prepared by reacting AFB1-8,9-epoxide with calf-thymus DNA
(1 mg/ml) in 50 mM phosphate buffer, pH 7.4. After 30 min incubation
with vigorous stirring, the DNA was precipitated with ethanol. The purified
AFB1-DNA was hydrolyzed with 0.15 N HCl at 90°C for 30 min. The ring
opened imidazole derivatives 8,9-dihydro-8-(2,6-diamino-4-oxo-3,4-dihy-
dropyrimid-5-yl-formamido) 9-hydroxy AFB1 (AFB1-FAPyr1) and 8,9-
dihydro-8-(2-amino-6-formamido-4-oxo-3,4-dihydropyrimid-5-yl)9-
MATERIALS AND METHODS
Chemicals
AFB1 (CAS No 1162-65-8), RNAses A and T1 were purchased from hydroxy-AFB1 (AFB1-FAPyr2) were prepared by incubating AFB1-N7-
Sigma Chemical Co. (St. Louis, MO). Proteinase K, Tris-saturated phenol, Gua with 15 mM sodium carbonate and 30 mM/L sodium bicarbonate, pH

AFB1 Test Comparison of Testes and Liver
295
9.5, for 30 min (see Fig. 1 for chemical structures). The AFB1-N7-Gua and precipitated with chilled absolute ethanol and the DNA was spooled out
the AFB1-FAPyr adducts were separated by reverse-phase chromatography with a sealed glass Pasteur pipette. The DNA was rinsed three times in 70%
and analyzed by mass spectrometry. The mass spectrum of AFB1-N7-Gua ethanol, dried, and redissolved in distilled water.
showed cationized Na at m/e 502 (M ϩ Na)ϩ and the mass spectra of the
Approximately 3 g of liver per rat were homogenized and extracted as
AFB1-FAPyr adducts showed either protonated parent ions at m/e 498 described above for the testis. The DNA concentration was measured by
(M ϩ H)ϩ or cationized Na at m/e 520 (M ϩ Na ϩ H)ϩ.
UV in a Beckman DU-650 spectrophotometer (Beckman Instruments,
Fullerton, CA) using a standard curve prepared from calf thymus DNA.
The purity of DNA was determined by UV scanning between 200 and 300
nm and by the 260/280 nm ratio. The latter was Ն1.85.
Animals, Diets, and Treatment
Fisher-344/N male rats from the breeding colony of the National Center
for Toxicological Research (NCTR, Jefferson, AR) were used in this study.
The rats were weaned at 21 days of age, housed singly, acclimated to
NIH-31 meal diet for 1 week, and randomly allocated to control and
experimental groups at 4 weeks of age. Rats in the control group received
AFB1-free NIH-31 meal diet and water ad libitum. Rats in the treatment
group were fed with NIH-31 meal diet premixed with 1.6 ppm of AFB1.
The AFB1-NIH-31 meal diets were prepared by mixing 40 kg of the
NIH-31 meal diet with 64 mg of AFB1 dissolved in 500 ml of ethanol. The
latter was evaporated from the diet by reducing the pressure during mixing.
The concentration of AFB1 in each diet preparation was measured accord-
ing to the method of Park et al. [1990]. Recovery was Ͼ90%.
GST Activity Assay
The GST activity of cytoplasmic supernates from testicular and liver
homogenates was measured by the procedure of Habig et al. [1974] using
glutathione and 1-chloro-2,4-dinitrobenzene as substrates.
Hydrolysis and Solid-Phase Extraction (SPE) of DNA
Groups of rats were fed continuously for 4, 8, 12, or 16 weeks. Other
groups were fed intermittently for 8, 12, 16, or 20 weeks with alternate
4-week dosing periods and 4-week no-dosing periods. Rats were sacrificed
at the end of the treatment periods by CO2 asphyxiation. Various tissues,
including testes and livers, were removed for different studies and kept
frozen at Ϫ80°C until analysis.
Separate groups of rats were used to estimate food consumption. Four
rats per group were housed singly in hanging cages. Food spilled out by the
rats was collected underneath the cage and weighed. The food consumption
was corrected by adding the weight of wasted food to the weight difference
before and after feeding. There were no significant differences in food
consumption between control and treated rats.
Two to three mg of testicular or liver DNA per rat were used for
mild-acid hydrolysis. Acid hydrolysis was carried out according to Hertzog
et al. [1980]. Briefly, the DNA was dissolved in 0.1 N HCl and hydrolyzed
at 70–75°C for 30 min. The DNA hydrolysate was placed on ice and the
pH was brought to approximately 4.5 with 5N NaOH.
The acid-hydrolyzed DNA was loaded onto 3-ml Oasis C18 extraction
cartridges (Waters Corp., Milford, MA) previously conditioned with 3 ml
of HPLC-grade methanol followed by 3 ml of glass-distilled water. The
cartridges were eluted with 15 ml of 10% methanol in 50 mM K-acetate,
pH 4.5, followed by 12 ml of 85% methanol in 50 mM potassium acetate,
same pH. Recovery of the DNA adducts was Ն95%, as determined by
elution of known amounts of [3H]-AFB1-N7-Gua adducts. The 85% meth-
anol eluate containing the DNA adducts was evaporated using a centrifugal
vacuum evaporator (Jouan, Winchester, VA). The dry eluate was either
kept frozen at Ϫ20°C or was dissolved in 300 ␮l of 35% methanol for
immediate HPLC analysis.
DNA Extraction and Purification
The testes were placed on a petri dish containing phosphate buffered
saline (PBS) and 20 mM of Tris-EDTA buffer, pH 7.8, on ice. The tunica
albuginea was removed and the mass of seminiferous tubules transferred to
a 50 ml centrifuge tube on ice. Ten ml of cold Tris-EDTA buffer, pH 7.8,
were added. The mass of seminiferous tubules was homogenized with a
Polytron PT3000 homogenizer (Brinkman Instruments, Westbury, NY) at
5,000 to 6,000 rpm for approximately 2 min. The homogenate was then
centrifuged at 1,500g for 15 min. The supernate was kept on ice and rapidly
frozen at Ϫ20°C for glutathione S-transferase (GST) analysis. The nuclear
pellet was resuspended in 10 ml of Tris-EDTA buffer and the centrifuga-
tion was repeated. The final nuclear pellet was resuspended in approxi-
mately 7 ml of Tris-EDTA buffer. RNAses A and T1 were added to a final
concentration of 150 ␮g/ml and 1,000 U/ml, respectively. The nuclear
suspension was then vortexed and incubated for 1 hr at 40°C in a water
bath. Fresh proteinase K (150 ␮g/ml final concentration) was added and the
incubation continued for another hour. SDS was added to a final concen-
tration of 0.5% and the suspension was incubated for an additional 0.5 hr.
At the end of the incubation period, an equal volume of Tris-saturated
phenol, pH 7.8, was added. The centrifuge tube was rocked for 5 min, the
suspension was transferred to a Phase-Lock Gel III centrifuge tube (5
Prime 6 3 Prime, Boulder, CO) and centrifuged at 1,500g for 12 min. The
top aqueous layer containing the DNA was transferred to a fresh centrifuge
tube and an equal volume of Tris-saturated phenol/chloroform/isoamyl
alcohol (25:24:1) added. The suspension was rocked for 5 min and cen-
trifuged as described above. The top aqueous phase was transferred to a
fresh centrifuge tube and an equal volume of chloroform/isoamyl alcohol
HPLC Analysis of AFB1-DNA Adducts
The DNA adduct standards, testicular and liver DNA hydrolysates were
analyzed using a Waters HPLC system. This consisted of a 991 model
photodiode array detector, a 470 model spectrofluorometer, a 600-E system
controller and a U6K model injector. A Supelcosil LC-18S column (250 ϫ
4.6 mm, 5-␮m particle size) coupled with a Supelguard LC-18S guard
column (20 ϫ 4.6 mm, 5-␮m particle size, Supelco, Bellefonte, PA), were
used. The column was eluted isocratically with 35% methanol in water at
a flow of 1 ml/min for 20 min at room temperature. The effluent was
monitored for fluorescence at 365 nm excitation and 428 nm emission
[Autrop et al., 1983; Weaver et al., 1994]. The fluorometer was set at a gain
of 1,000x and an attenuation of 2 for maximum fluorescence detection.
The concentration of the DNA adducts was calculated from standard
curves prepared using pmol concentrations of the authentic AFB1-N7-Gua
standards.
Statistical Analysis
Statistical differences between control and treated rats were determined
(24:1) was added. The suspension was centrifuged as before and the top by separate two-way ANOVAs, including the one- and two-tailed lsd-
aqueous phase transferred to a centrifuge tube kept on ice. The DNA was t-tests.

296
Sotomayor et al.
TABLE I. Mean Body, Testis and Liver Weights ؎ SEM (g) of Rats Fed Continuously and Intermittently
with 1.6 ppm of AFB1
Treatment
(Weeks)
Continuous
TW
Intermittent
TW
Controls
TW
BW
LW
BW
LW
BW
LW
4
215 Ϯ 5
(5)
2.6 Ϯ 0.03
(5)
9.9 Ϯ 0.1
(5)
199 Ϯ 9
(5)
2.4 Ϯ 0.08
(5)
8.6 Ϯ 0.5
(5)
8
281 Ϯ 8
(5)
2.9 Ϯ 0.01
(5)
11.7 Ϯ 0.4
(5)
277 Ϯ 7
(5)
2.7 Ϯ 0.05
(5)
11.2 Ϯ 0.3
(5)
289 Ϯ 6
(5)
2.6 Ϯ 0.11
(5)
11.8 Ϯ 0.4
(5)
12
16
20
316 Ϯ 12*
(5)
346 Ϯ 7*
(4)
2.7 Ϯ 0.14
(5)
2.9 Ϯ 0.08
(4)
11.8 Ϯ 0.5
(5)
12.2 Ϯ 0.1
(4)
302 Ϯ 13*
(5)
365 Ϯ 11
(5)
367 Ϯ 13*
(5)
2.8 Ϯ 0.09
(5)
3.0 Ϯ 0.04
(5)
2.9 Ϯ 0.11
(5)
10.6 Ϯ 0.4
(5)
13.0 Ϯ 0.5
(5)
13.5 Ϯ 0.6
(5)
355 Ϯ 14
(5)
379 Ϯ 12
(5)
400 Ϯ 9
(5)
2.8 Ϯ 0.05
(5)
2.9 Ϯ 0.16
(5)
3.0 Ϯ 0.07
(5)
12.1 Ϯ 0.6
(5)
13.6 Ϯ 0.6
(5)
13.7 Ϯ 0.4
(5)
Note: Figures marked with asterisks differ significantly from controls: *P Ͻ 0.05. In testing for significance, TW and LW were adjusted for BW
differences. The number of rats per group is shown in parenthesis. BW ϭ body weight; TW ϭ testis weight; LW ϭ liver weight; SEM ϭ standard error
of the mean.
Fig. 2. Linear relationship between fluorescence intensity and adduct concentration for AFB1-FAPyr1 ᮀ,
AFB1-FAPyr2 F and AFB1-N7-Gua E.
RESULTS
(r Ն 0.99) between concentration (pmol) and fluorescence
intensity was obtained for all the standards. The limit of
detection was 1.4 pmol for AFB1-FAPyr1 and 3.2 pmol for
AFB1-FAPyr2 per injection. AFB1-N7-Gua was approxi-
mately an order of magnitude less fluorescent than the
opened imidazole-ring adducts.
The fluorescent chromatographic profiles of the standards
AFB1-N7-Gua, AFB1-FAPyr1, and AFB1-FAPyr2 are
shown in Figure 3. AFB1-N7-Gua and AFB1-FAPyr1 eluted
as single peaks with similar retention times (Fig. 3A,B).
AFB1-FAPyr2 eluted as two fluorescent peaks at 7.5 Ϯ 0.2
min for the major peak and 9.5 Ϯ 0.3 min for the minor
peak (Fig. 3C). The latter did not show a corresponding UV
peak at 260 nm.
Body, Liver, and Testis Weights
Continuous feeding of rats with 1.6 ppm of AFB1 resulted
in a significant decrease of their body weights (P Յ 0.05)
from 12 weeks onward (Table I). Rats exposed intermit-
tently showed significantly lower body weights only when
they were dosed during the last 4 weeks of treatment, i.e., at
12 and 20 weeks (P Յ 0.05). There were no significant
differences in testis or liver weights between treated and
control rats at any time period (Table I). No deaths were
observed during this study.
Analysis of AFB1-DNA Adducts
Typical chromatograms of testicular DNA hydrolysates
The fluorescence intensity of each of the AFB1-DNA
adduct standards is shown in Figure 2. A linear relationship from treated and control rats are shown in Figure 4. Most

AFB1 Test Comparison of Testes and Liver
297
Fig. 3. HPLC profiles of the standards AFB1-N7-Gua (A), AFB1-FAPyr1 (B), AFB1-FAPyr2 (C), and a
mixture of AFB1-N7-Gua/AFB1-FAPyr2. Fluorescence measured at 365 nm excitation and 428 nm emission.
hydrolysis products, including adenine, unmodified gua- did (Fig. 3D). Therefore, the single peak observed in all the
nine, and partially degraded DNA, eluted between 2 and 6 testicular chromatograms most likely represents AFB1-FA-
min. Only one fluorescent peak containing the lipophilic Pyr1. No other significant peaks were observed. No peaks
AFB1-DNA adducts was observed in all the chromatograms corresponding to AFB1-DNA adducts were observed in
(Fig. 4B,C,D). This peak eluted at 11.8 Ϯ 0.2 min. Its hydrolyzed DNA from control testes (Fig. 4A).
identity was confirmed by co-elution with the AFB1-N7-
The temporal changes in AFB1-DNA adduct formation in
Gua standard. Under the HPLC conditions used, AFB1-N7- the testis and liver are shown in Figure 5. Large differences
Gua did not separate from AFB1-FAPyr1 but AFB1-FAPyr2 were observed between these tissues. In the testis, DNA

298
Sotomayor et al.
Fig. 4. HPLC profiles of testicular DNA hydrolysates from a control rat (A), rat treated continuously for 4
weeks (B), rat treated continuously for 8 weeks (C), and a rat treated intermittently for 20 weeks (D).
Fluorescence measured at 365 nm excitation and 428 nm emission. Arrows show the peaks containing the
AFB1-DNA adducts.
adduct formation markedly decreased over time, regardless after ending the treatment (at 8 weeks of intermittent expo-
of the type of exposure. In continuously treated rats, the sure). Overall, the extent of DNA adduct formation in the
levels of adducts decreased 42 and 37% at 12 and 16 weeks, testis was significantly lower than that in the liver (P Յ
respectively, compared to the previous 8 weeks. In inter- 0.05) at similar time periods.
mittently treated rats (alternating 4 weeks of dosing and 4
In the liver, the levels of DNA adducts increased approx-
weeks of no-dosing), the levels of DNA adducts fell 82 and imately 4-fold from 4 to 16 weeks of continuous treatment.
91% at 16 and 20 weeks compared to the previous 12 Intermittently treated rats showed significantly lower (P Յ
weeks, respectively. However, significant levels of AFB1- 0.05) but fluctuating levels of DNA adducts. The levels
DNA adducts (P Յ 0.05) persisted in the testis for 4 weeks increased only in those rats fed AFB1 for the last 4 weeks

AFB1 Test Comparison of Testes and Liver
299
Fig. 5. A comparison of the patterns of DNA adduct formation in testis
and liver after continuous (f) and intermittent exposures (o) to 1.6 ppm of
AFB1.
before sacrifice, i.e., at 12 and 20 weeks. Like in the testis,
significant levels of DNA adducts persisted 4 weeks after
removal of rats from treatment, i.e., at 8 and 16 weeks (P Յ
0.05).
GST Activity
The effects of AFB1 on testicular and liver GST activities
are shown in Figure 6. In control rats, the testicular GST
activity increased steadily during the first 12 weeks of
normal feeding but remained relatively constant thereafter
(Fig. 6A). In comparison, the control liver GST activity did
not increase over time. The liver GST activity was signifi-
Fig. 6. A comparison of the GST activity (␮mol/min/mg protein) in testis
(f) and liver (o) in controls (A) and after intermittent (B) and continuous
exposures (C) to 1.6 ppm of AFB1.
cantly lower than that of the testis at 8, 12, 16, and 20 weeks rats, significant differences (P Ͻ 0.05) were observed at 4,
(P Յ 0.0005) but not at 4 weeks (P ϭ 0.0548). 8, and 12 weeks for the testis and at 4, 8, 12, and 16 weeks
In treated rats, both testicular and liver GST activities for the liver. On average, GST activity in the testis of
were generally higher than those of controls. In intermit- continuously and intermittently treated rats was higher than
tently treated rats, significant differences were observed that in the liver at equivalent time periods (Fig. 6B,C).
only at 8 weeks (P ϭ 0.0037) for the testis and at 12, 16, and However, significant differences (P Ͻ 0.05) were observed
20 weeks for the liver (P Ͻ 0.05). In continuously treated only at 8 and 12 weeks of continuous exposure and at 12

300
Sotomayor et al.
weeks of intermittent exposure. Like the long persistence of 82 and 91% at 16 and 20 weeks, respectively, compared
DNA adducts described previously, the GST activity also with the previous weeks. Unfortunately, there are no reports
persisted for 4 weeks after ending the treatment in both the on the testis, to our knowledge, that would allow us to
testis and liver at 8 and 16 weeks of intermittent exposure, compare them with our data. Our observations in the liver
respectively.
are somewhat different from those of Buss et al. [1990].
These investigators found no significant increase in rat liver
AFB1-DNA binding after 8 weeks of exposure in the drink-
ing water.
DISCUSSION
The main focus of this study was to evaluate DNA
The liver has been well studied with regard to its capa-
damage and detoxification in the testes of rats fed AFB1 bility to metabolize AFB1 to its epoxide [see Gorelick,
continuously and intermittently. The liver was also studied 1990, for review]. This process is poorly understood in the
to make comparisons with the testis. To the best of our testis. AFB1-8,9-epoxide is highly reactive and it is unlikely
knowledge, no reports have been published on the damag- that it can reach the testis in significant concentrations.
ing effects of AFB1 on testis DNA despite its importance for Thus, our data showing that AFB1 can cause significant
germ-cell mutagenesis. Continuous and intermittent feeding levels of DNA adducts in the testis suggest that the testis is
with AFB1 were chosen to generally imitate human expo- quite capable of metabolizing AFB1 to its epoxide.
sure to this carcinogen.
The pattern of DNA adduct formation in the testis was
HPLC with fluorescence detection was used to measure clearly different from that in the liver (Fig. 5) and suggests
DNA adducts in the testis and liver of rats. This method has tissue-specific differences. Several mechanisms unique to
been used previously to measure DNA adducts in nonger- the testis may contribute to this difference. Spermatogenesis
minal tissues [Autrop et al., 1983; Weaver et al., 1994; is a continuous process involving multiple cell divisions and
Harris et al., 1986]. In our case, fluorescence intensity was differentiation that result in the release of mature sperm.
linear over a large range of the AFB1-DNA adduct concen- After an acute or brief exposure to AFB1, any adducts
trations (Fig. 2). The limits of detection were higher for the formed are likely to decrease over time due to the contin-
AFB1-N7-Gua adduct and at least an order of magnitude uous production of new germ cells. This was not the case in
lower for the AFB1-FAPyr adducts than those reported our study since rats were exposed continuously or intermit-
previously [Harris et al., 1986]. AFB1-N7-Gua did not sep- tently for several weeks. Germ-cell depletion is one possible
arate from AFB1-FAPyr1 under our HPLC conditions but mechanism. Germ-cell depletion occurs as a consequence of
did from AFB1-FAPyr2 (Fig. 3). Since AFB1-N7-Gua has a germ-cell killing by cytotoxic agents and can result in
half-life of 7.5 hr in rat liver [Croy and Wogan, 1981], it is partial or complete depletion of the germ cells from the
highly unlikely that this adduct persisted for weeks after testis [Russell et al., 1981; Oakberg, 1984]. Extensive sper-
cessation of the treatment. Thus, the single fluorescent peak matogenic-cell depletion has been reported in rats treated
shown in Figure 4 most likely represents AFB1-FAPyr1. with AFB1 by intraperitoneal injection [Egbunike et al.,
The possibility that AFB1-diol could have formed in our 1980], by intratesticular injection [Gopal et al., 1980], and
samples and co-chromatographed with the fluorescent peak by gavage [Sahay, 1993]. As much as 50% reduction of the
shown in Figure 4 is remote for several reasons. First, testis weight has been reported [Gopal et al., 1980]. We
Weaver and Groopman [1994] showed that very harsh hy- cannot rule out the possibility that germ-cell depletion oc-
drolysis conditions are required to release the diol. The curred in our study since no histological analysis was per-
optimal conditions were 12.4 M HCl at 95°C for 60 min. formed in our case. However, no significant differences in
Our acid hydrolysis conditions were much milder. Second, the testis weights between treated and control rats were
if AFB1-diol had been released, the HPLC analysis (Fig. 4) observed at any time period (Table I). If, for instance,
would have shown at least two fluorescent peaks. AFB1- extensive germ-cell depletion would have happened after
FAPyr and AFB1-N7-Gua adducts have different retention continuous treatment, this would have been reflected in the
times from that of the diol [Hertzog et al., 1980; Harris et weight of the testes. Extensive germ-cell depletion appears
al., 1986].
not to have occurred in our case. Therefore, the large
Significant levels of AFB1-DNA adducts were measured decrease in DNA adducts observed particularly after inter-
in the testis, specially during the first 12 weeks of continu- mittent exposure cannot be explained by germ-cell deple-
ous and intermittent exposures. Large differences were ob- tion alone.
served in the pattern and extent of DNA adduct formation
Another possibility is the removal of DNA adducts by
between the testis and liver (Fig. 5). Whereas in the liver the repair mechanisms. Unscheduled DNA synthesis (UDS), a
levels of DNA adducts steadily increased throughout con- form of DNA repair, has been well documented in rodent
tinuous treatment or fluctuated during intermittent exposure, spermatogenic cells [Sega and Sotomayor, 1982; Bentley
in the testis they decreased regardless of the type of treat- and Working, 1988; Sotomayor et al., 1996]. To our knowl-
ment (Fig. 5). This decrease in testicular adducts was more edge, no information exists on repair of AFB1-induced
pronounced after intermittent exposure, with levels falling DNA damage in spermatogenic cells. The present study

AFB1 Test Comparison of Testes and Liver
301
nonmutagens in the rat spermatocyte UDS assay. Mutat Res 203:
135–142.
Buss P, Caviezel M, Lutz WK. 1990. Linear dose-response relationship for
DNA adducts in rat liver from chronic exposure to aflatoxin B1.
Carcinogenesis 11:2133–2135.
Chen W, Nichols J, Zhou Y, Chung KT, Hart RW, Chou MW. 1995. Effect
of dietary restriction on glutathione S-transferase activity specific
toward aflatoxin B1-8,9-epoxide. Toxicol Lett 78:235–243.
Croy RG, Wogan GN. 1981. Temporal patterns of covalent DNA adducts
in rat liver after single and multiple doses of aflatoxin B1. Cancer
Res 41:197–203.
Croy RG, Essigmann JM, Reinhold VN, Wogan GN. 1978. Identification
of the principal aflatoxin B1-DNA adduct formed in vivo in rat liver.
Proc Natl Acad Sci USA 75:1745–1749.
Di Biasio KW, Silva MH, Shull LR, Overstreet JW, Hammock BD, Miller
MG. 1991. Xenobiotic metabolizing enzyme activities in rat,
mouse, monkey, and human testes. Drug Metab Dispos 19:227–
232.
Egbunike GN. 1982. Steroidogenic and spermatogenic potentials of the
male rat after acute treatment with aflatoxin B1. Andrologia 14:
440–446.
Egbunike GN. 1985. Sperm maturation and storage in the male rat after
acute treatment with aflatoxin B1. Andrologia 17:379–382.
Egbunike GN, Emerole GO, Aire TA, Ikegwuonu FI. 1980. Sperm pro-
duction rates, sperm physiology and fertility in rats chronically
treated with sublethal doses of aflatoxin B1. Andrologia 12:467–
475.
Essigman JM, Croy RG, Nadzan AM, Busby WF, Reinhold VN, Bu¨chi G,
Wogan GN. 1977. Structural identification of the major DNA
adduct formed by aflatoxin B1 in vitro. Proc Natl Acad Sci USA
74:1870–1874.
Gandy J, Millner GC, Bates HK, Casciano DA, Harbison RD. 1990.
Effects of selected chemicals on the glutathione status in the male
reproductive system of rats. J Toxicol Environ Health 29:45–57.
Gopal T, Oehme FW, Liao TF, Chen CL. 1980. Effects of intratesticular
aflatoxin B1 on rat testes and blood estrogens. Toxicol Lett 5:263–
267.
Gorelick NJ. 1990. Risk assessment for aflatoxin. I. Metabolism of afla-
toxin B1 by different species. Risk Anal 10:539–559.
Groopman JD, Sabbioni G, Wild CP. 1991. Molecular dosimetry of human
aflatoxin exposures. In: Groopman JD, Skipper PL, editors. Molec-
ular dosimetry and human cancer: analytical, epidemiological, and
social considerations. Boca Raton, FL: CRC Press. p 303–324.
Guthenberg C, Astrand IM, Alin P, Mannervik B. 1983. Glutathione
transferases in rat testis. Acta Chem Biol 37:261–262.
Habig, WH, Pabst, MJ, Jakoby WB. 1974. Glutathione S-transferases.
J Biol Chem 249:7130–7139.
shows that in intermittently treated rats significant levels of
AFB1-DNA adducts persisted in both the testis and the liver
(Fig. 5). The long persistence of these adducts, demon-
strated previously in rat liver [Croy and Wogan, 1981;
Kensler et al., 1986, 1987], suggests that repair of this type
of DNA lesion occurs very slowly, if at all. Croy and
Wogan [1981] concluded that AFB1-FAPyr is the persistent
DNA adduct in rat liver. Thus, it is likely that AFB1-
FAPyr1 is also the persistent DNA lesion in the testis.
In the liver, GST plays a major role in the detoxification
of AFB1 by catalyzing the conjugation of AFB1-epoxide
with glutathione [Lotlikar et al., 1984; Neal and Working,
1983; Di Biasio et al., 1991] and protects against AFB1-
DNA binding [Raj et al., 1984; Kensler et al., 1986, 1987;
Gorelick, 1990]. Our results show that rat testes are also rich
in GST activity (Fig. 6), confirming studies by others [Kraus
and Kloft, 1980; Guthenberg et al., 1983; Di Biasio et al.,
1991; Papp et al., 1994; Hermo et al., 1994]. Furthermore,
in control rats the testis showed significantly higher GST
activity than the liver during the entire length of the study,
i.e., 20 weeks (Fig. 6A). In both intermittently and contin-
uously treated rats, the testis showed significantly higher
GST activity than the liver during the first 12 weeks of
treatment, but not thereafter. These results appear to be
consistent with the hypothesis that GST also plays a major
role in detoxifying AFB1 in the testis. Other researchers
have shown that conjugation of AFB1-epoxide with gluta-
thione by GST protects against genetic damage. In rats,
chemically induced depression of testicular glutathione
caused a significant increase in dominant lethal mutations
by ethyl methanesulfonate [Teaf et al., 1987] and enhanced
binding of this mutagen to sperm heads [Gandy et al., 1990].
Also, GST is thought to play a major role against testicular
toxicity by ethylene oxide in the rat [Mori et al., 1989].
Thus, it can be expected that a similar mechanism may
protect spermatogenic cells against genetic damage by
AFB1.
ACKNOWLEDGMENTS
Harris CC, La Veck G, Groopman J, Wilson VL, Mann D. 1986. Mea-
surement of aflatoxin B1, its metabolites, and DNA adducts by
synchronous fluorescence spectrophotometry. Cancer Res 46:
3249–3253.
The authors thank Dr. Phyllis N. Windle for editorial
review and valuable comments, Dr. Gary A. Sega for sci-
entific review, and Mr. Stuart J. Chirtel, FDA, for statistical
analysis.
Hermo L, Papp S, Robaire B. 1994. Developmental expression of the Yf
subunit of glutathione S-transferase P in epithelial cells of the testis,
efferent ducts, and epididymis of the rat. Anat Rec 239:421–440.
Hertzog PJ, Lindsay Smith JR, Colin Garner R. 1980. A high pressure
liquid chromatography study on the removal of DNA-bound afla-
toxin B1 in rat liver and in vitro. Carcinogenesis 1:787–793.
Hertzog PJ, Lindsay Smith JR, Colin Garner R. 1982. Characterization of
the imidazole ring-opened forms of trans-8,9-dihydro-8-(7-gua-
nyl)9-hydroxy aflatoxin B1. Carcinogenesis 3:723–725.
Kensler TW, Egner PA, Davidson NE, Roebuck BD, Pikul A, Groopman
JD. 1986. Modulation of aflatoxin metabolism, aflatoxin-N7-gua-
nine formation, and hepatic tumorigenesis in rats fed ethoxyquin:
role of induction of glutathione S-transferases. Cancer Res 46:
3924–3931.
REFERENCES
Autrop H, Bradley KA, Shamsuddln AKM, Wakhisi J, Wasunna A. 1983.
Detection of putative adduct with fluorescence characteristics iden-
tical to 2,3-dihydro-2-(7Ј-guanyl)-3-hydroxyaflatoxin B1 in human
urine collected in Murang’a district, Kenya. Carcinogenesis
4:1193–1195.
Baertschi SW, Raney KD, Stone MP, Harris TM. 1988. Preparation of the
8,9-epoxide of the mycotoxin aflatoxin B1: The ultimate carcino-
genic species. J Am Chem Soc 110:7929–7931.
Bentley KS, Working PK. 1988. Activity of germ-cell mutagens and

302
Sotomayor et al.
Kensler TW, Egner PA, Dolan PM, Groopman JD, Roebuck BD. 1987.
Mechanism of protection against aflatoxin tumorigenicity in rats
and G2 in corn and peanut products: a collaborative study. J AOAC
Int 73:260–266.
fed 5-(2-Pyrazinyl)-4-methyl-1,2-dithiol-3-thione (Oltipraz) and re- Raj HG, Clearfield MS, Lotlikar PD. 1984. Comparative kinetic studies on
lated 1,2-dithiol-3-thiones and 1,2-dithiol-3-ones. Cancer Res 47:
4271–4277.
aflatoxin B1-DNA binding and aflatoxin B1-glutathione conjugation
with rat and hamster livers in vitro. Carcinogenesis 5:879–884.
Kraus P, Kloft HD. 1980. The activity of glutathione S-transferase in Russell LD, Malone JP, Karpas SL. 1981. Morphological pattern elicited
various organs of the rat. Enzyme 25:158–160.
Lin JK, Miller JA, Miller EC. 1977. 2,3-Dihydro-2-(guan-7-yl)-3-hydroxy-
by agents affecting spermatogenesis by disruption of its hormonal
stimulation. Tissue Cell 13:369–380.
aflatoxin B1, a major acid hydrolysis product of aflatoxin B1-DNA Sahay M. 1993. Aflatoxin induced testicular degeneration and decreased
or -ribosomal RNA adducts formed in hepatic microsome-mediated
reactions and in rat liver in vivo. Cancer Res 37:4430–4438.
Lotlikar PD, Jhee EC, Insetta SM, Clearfield MS. 1984. Modulation of
microsome-mediated aflatoxin B1 binding to exogenous and endog-
enous DNA by cytosolic glutathione S-transferases in rat and
hamster livers. Carcinogenesis 5:269–276.
Mori K, Kasido M, Fujishiro K, Inoue N. 1989. Testicular toxicity and
alterations of glutathione metabolism resulting from chronic inha-
lation of ethylene oxide in rats. Toxicol Appl Pharmacol 101:299–
309.
meiotic index in Rattus norvegicus. Cytobios 75:191–195.
Sega GA, Sotomayor RE. 1982. Unscheduled DNA synthesis in mamma-
lian germ cells: its potential use in mutagenicity testing. In: de
Serres FJ, Hollander A, editors. Chemical mutagens, vol. 7. New
York: Plenum Press. p 421–445.
Sotomayor RE, Jackson R, Nourbakhsh A. 1996. A procedure to study
unscheduled DNA synthesis in rat spermatogenic cells in vivo:
strain differences between Fisher-344 and Sprague-Dawley rats.
Toxicol Lett 89:77–81.
Teaf CM, Bishop JB, Harbison RD. 1987. Depression of glutathione in
male reproductive tissues and potentiation of EMS-induced germ
cell mutagenesis by L-buthionine sulfoximine. Teratog Carcinog
Mutagen 7:497–513.
Neal GE, Working JA. 1983. The requirements for glutathione S-trans-
ferase in the conjugation of activated aflatoxin B1 during aflatoxin
hepatocarcinogenesis in the rat. Chem Biol Interact 45:259–275.
Oakberg EF. 1984. Germ cell toxicity: significance in genetic and fertility Weaver VM, Groopman JD. 1994. Fluorescence quantification of aflatoxin
effects of radiation and chemicals. In: Chu HY, Generoso WM,
editors. Mutation, cancer and malformation. New York: Plenum
Press. p 549–590.
N7-guanine adducts. Cancer Epidemiol Biomarkers Prev 3:669–
674.
Wogan GN. 1973. Aflatoxin carcinogenesis. Methods Cancer Res 8:309–
344.
Papp S, Robaire B, Hermo L. 1994. Developmental expression of the
glutathione S-transferase Yo subunit in the rat testis and epididymis Wogan GN. 1992. Aflatoxins as risk factors for hepatocellular carcinoma
using light microscope immunocytochemistry. Anat Rec 240:345–
357.
Park DL, Nesheim S, Trucksess NW, Stack ME, Newell RF. 1990. Liquid
chromatographic method for determination of aflatoxins B1, B2, G1
in humans. Cancer Res (Suppl) 52:2114–2118.
Accepted by—
P. Stambrook