FUNDAMENTAL AND APPLIED TOXICOLOGY 32, 2 5 0 - 2 5 9 (1996)
ARTICLE NO. 0128
Protection against Aflatoxin B-,-Induced Hepatic Toxicity as a ShortTerm Screen of Cancer Chemopreventive Dithiolethiones1
YULIA Y. MAXUITENKO,* THOMAS J. CURPHEY,| THOMAS W. KENSLER,$ AND B. D. ROEBUCK* 2
* Department of Pharmacology and Toxicology and ^Department of Pathology, Dartmouth Medical School. Hanover. New Hampshire 03755; and
^Department of Environmental Health Sciences, Johns Hopkins School of Hygiene and Public Health, Baltimore, Maryland 21205
Received February 20, 1996; accepted April 6, 1996
fective chemopreventive agent in many different rodent
models of experimental carcinogenesis in target organs
such as the small intestine, pancreas, lung, forestomach,
MAXUITENKO, Y. Y., CURPHEY, T. J., KENSLER, T. W., AND ROEcolon, urinary bladder, trachea, liver, mammary gland,
BUCK, B. D. (1996). Fundam. Appl. Toxicol. 32, 250-259.
and skin (see Kensler and Helzlsouer, 1995, for referDithiolethiones are an important class of cancer chemopreven- ences). OLT has undergone phase I clinical trials to detertive agents. More than 50 new dithiolethione analogs were synthemine its pharmacokinetics and dose-limiting side effects
sized for structure-activity studies. Using selected dithiolethiones,
during
chronic administration to humans (Benson, 1993).
studies were designed to measure protection against the hepatotoxPhase
II
clinical trials are underway in Qidong, Jiangsu
icity of aflatoxin B, (AFB,) and relate it to the protection against
Province,
People's Republic of China. This population is
carcinogenicity. Young male F344 rats were pretreated with 0.1
or 0.3 mmol dithiolethiones/kg body wt and challenged with toxic exposed to aflatoxin B, (AFB,) in their food supply and
doses of AFBi (50 ^/g/100 g rat/day) on 2 successive days. One day the trial is evaluating the effectiveness of OLT in modulatlater, the protection from hepatotoxicity was assessed by measur- ing AFB i metabolism and disposition in vivo.
ing serum hepatic enzymes, hepatic necrosis, and degree of bile
In humans several dithiolethiones are known to possess
duct cell proliferation. The ability of these dithiolethiones to pre- pharmaceutical properties other than cancer chemoprevenvent AFB! -induced tumorigenicity was assessed by quantifying the
tion. For example, OLT has been used as an orally effective
hepatic burden of putative preneoplastic lesions [placental glutathione S-transferase (GST-P)-positive foci]. Significant correla- drug for the treatment of schistosomiasis. Single doses of
tions (p < 0.01) were observed between these toxicological indices OLT have achieved cure rates greater than 90% (Archer,
and GST-P focal burden (alanine aminotransferase, r = 0.943; 1985). OLT is also an irreversible inhibitor of HIV reverse
sorbitol dehydrogenase, r = 0.897; histological index, r = 0.893; transcriptase and is currently in a clinical trial in AIDS pabile duct cell proliferation, r = 0.933). These results imply that tients (Prochaska et ai, 1995). 5-(4-MethoxyphenyI)-3tfinhibition of hepatotoxicity affords protection against hepatocar- 1,2-dithiole-3-thione (ADT) is used currently as a choleretic
cinogenicity. The extent of protection from acute hepatotoxicity drug and to stimulate salivary secretion (Hausler and Ritschoffers a simple, short-term biological endpoint to screen dithioleth- ard, 1979). Thus, general tolerance of dithiolethiones and
iones and related compounds for their chemopreventive properties. their acceptance in humans have been broadly examined.
O 1996 Society of Toxicology
While studying the antischistosomal properties of OLT,
Bueding and associates (1982) noted that exposure of mice
Dithiolethiones hold great promise as cancer chemo- infected with Schistosoma mansoni to OLT increased levels
prevention drugs. For example, 4-methyl-5-pyrazinyl-3//- of glutathione in host tissues while reducing glutathione
l,2-dithiole-3-thione (OLT)3 has been shown to be an ef- stores of the parasite. These observations were expanded to
show that OLT and other dithiolethiones were potent in1
ducers
of enzymes involved in the maintenance of the rePreliminary accounts of this work were presented at the 1996 Annual
duced glutathione pools as well as enzymes involved in elecMeeting of the American Association for Cancer Research, Washington,
DC.
trophile detoxication, such as NAD(P)H:quinone reductase.
1
To whom requests for reprints should be addressed at Department of
epoxide hydrolase, UDP-glucuronosyl transferase, and glutaProtection against Aflatoxin B,-Induced Hepatic Toxicity as a
Short-Term Screen of Cancer Chemopreventive Dithiolethiones.
Pharmacology and Toxicology, Dartmouth Medical School, 7650 Remsen,
Hanover, NH 03755.
3
Abbreviations used: OLT, 4-methyl-5-pvrazinyl-3//-l,2-dithiole-3-thione (oltipraz); AFB,, aflatoxin B,; ADT, 5-<4-methoxyphenyl)-3//-I,2dithiole-3-thione; D3T, 3/Y-l,2-dithiole-3-thione; 5mD3T, 5-methyl-3//1,2-dithiole-3-thione; D2T, 1,3-dithiole-2-thione; SALT, serum alanine ami-
0272-0590/96 $18 00
Copyright C 1996 by the Society of Toxicology
All rights of reproduction in any form reserved
notransferase, sSDH. serum sorbitol dehydrogenase; BrdU. bromodeoxyundine; GST-P, placental form of glutathione 5-transferase; H&E. hematoxylin
and eosin; BDC, bile duct cell.
250
CORRELATION OF AFLATOXIN B, TOX1C1TY WITH ITS TUMORIGENICITY
thione S-transferase (Ansher et al, 1986). The elevation of
electrophile detoxication (phase 2) enzymes has been recognized as a characteristic action of many chemopreventive
agents (Prochaska et al, 1985). In contrast to marked induction of phase 2 enzymes, cytochrome P450 levels and other
phase 1 enzyme activities were only slightly elevated by
OLT (Kensler et al, 1985). The initial confirmation of the
predicted cancer chemopreventive activity of OLT came
from demonstration that OLT afforded protection against
the development of forestomach and/or pulmonary cancers
induced by benzo[a]pyrene, diethylnitrosamine, or uracil
mustard in female ICR/Ha mice (Wattenberg and Bueding,
1986). Later, chemopreventive properties of OLT against
AFB r induced hepatocarcinogenesis in male F344 rats were
demonstrated (Roebuck et al., 1991). OLT also affords protection against acute and chronic toxic effects. For example,
pretreatment of rats or mice with OLT resulted in the inhibition of acute hepatotoxic effects of carbon tetrachloride (Ansher et al., 1983), acetaminophen (Ansher et al., 1983), ally]
alcohol (Davies et al., 1987), and AFB, (Liu et al., 1988).
In the rat, OLT appeared to protect against both genotoxic
and cytotoxic actions of AFB, when fed before or during
AFB, exposure (Kensler et al., 1987; Roebuck et al., 1991;
Kensler et al., 1992b) or even as a result of a transient
intervention relative to a prolonged AFB, treatment period
(Bolton et al., 1993). Administration of OLT after AFB,
exposure occurred, however, yielded no protective effect
(Maxuitenko et al., 1993). To the contrary, in a model of
azoxymethane-induced colon carcinogenesis, OLT afforded
protection irrespective of time of its administration (Rao et
al, 1993).
Mechanisms of chemoprevention by dithiolethiones are
not fully understood. Mechanistic studies indicate that
chemoprevention afforded by OLT is due primarily to the
enhancement of electrophile detoxication pathways (Kensler
et al, 1987; Roebuck et al, 1991). We have synthesized
more than 50 dithiolethione analogs for structure-activity
studies aimed at identifying the most effective compounds
and molecular structural features among all dithiolethione
analogs. These structure-activity studies might help to understand the mechanism by which dithiolethiones inhibit carcinogenesis. For example, De Long et al (1986) reported
that 3W-l,2-dithiole-3-thione (D3T) was more effective in
the induction of NAD(P)H:quinone reductase in murine
Hepa Iclc7 cells in culture than OLT. The examination of
the effects of dietary administration of several dithiolethione
analogs on inhibition of hepatic AFB,-DNA adduct formation in rats revealed that D3T was the most effective compound in the inhibition of hepatic AFB,-DNA binding and
in the induction of electrophile detoxication enzymes, such
as glutathione S-transferases (Kensler et al, 1987).
The present studies were undertaken to characterize fur-
251
ther the dithiolethione analogs. Because of the large number
of dithiolethiones under investigation and the lack of complete understanding of mechanisms of their biological activity, these studies were undertaken with three specific objectives: (I) to develop an efficient biological screen for more
effective dithiolethiones; (2) to define the significance of
toxicity in AFB,-induced carcinogenesis; and (3) to examine
structure-activity features of dithiolethiones that produce
greater biological activity.
MATERIALS AND METHODS
Chemicals. AFB, was obtained from Aldnch Chemical Company (Milwaukee, WI). Pelleted punfied diet was of the AIN-76A formulation (Bieri,
1980), but without ethoxyquin, and was purchased from Harlan (Madison,
WI). OLT was prepared by a modification of the method described by
Barreau et al. (1977). D3T was prepared by the procedure of Meinetsberger
et al. (1977) 5-Methyl-3//-l,2-dithiole-3-thione (5mD3T) was prepared
from acetone and carbon disulfide by a method to be described elsewhere
(Curphey and Libby, unpublished) ADT was the gift of Solevay Pharma
(Suresnes, France). 1,3-Dithiole-2-thione (D2T) was obtained from Aldnch
Chemical Company Serum alanine aminotransferase (sALT) and serum
sorbitol dehydrogenase (sSDH) were measured spectrophotometrically using Sigma diagnostic kits (Sigma Chemical Co., St. Louis, MO). Bromodeoxyuridine (BrdU) was purchased from Sigma. Anti-BrdU monoclonal
antibody was supplied by Becton Dickinson (San Jose, CA). lmmunostaining of BrdU-labeled DNA was with the peroxidase-antiperoxidase detection system (Signet Laboratories, Inc., Dedham, MA). The primary glutathione S-transferase Yp antibody for immunohistochemical analysis was purchased from Biotrin International (Dublin, Ireland). The locaJization of
the placental form of glutathione S-transferase (GST-P) was done with a
peroxidase—antiperoxidase immunoenzymatic staining kit obtained from
Dako Corporation (Carpinteria, CA).
Animals.
Male F344 rats were purchased from Charles River Breeding Laboratory (Kingston, NY). Rats were housed in individual, suspended, wire-bottomed cages with free access to deionized water and
diet. Fresh diet was fed every other day. Unused diet was stored at 4°C.
The animal room was maintained at 22 ± 2°C, 5 5 - 6 5 % relative humidity, and had a 12-hr light-dark cycle. Rats were allowed to acclimate
to the facilities for 1 week prior to the experiment. They were weighed
once a week. Protection against either the acute toxic effects or the
carcinogenic effects of AFB, was assessed beginning when the rats were
7 weeks of age. In both experiments, a low or high dose of dithiolethiones
(0.1 or 0.3 mmol/kg body wt, respectively) was gavaged three times per
week beginning I week prior to AFB, treatment. The dithiolethiones are
highly insoluble in aqueous solutions and difficult to keep in suspension.
They were gavaged as a powder dispersed in a saturated solution of
sucrose. The density and viscosity of the saturated sucrose maintained
the dithiolethione particles in suspension. AFB, was dissolved in tricaprylin and was administered by gastric incubation.
Acute toxicity experiment.
A schematic of the experimental protocol
is shown in Fig. I. For each of the five dithiolethiones, groups of four rats
received either a low or a high dose of a dithiolethione. The five dithiolethiones were D2T, ADT, OLT, 5mD3T, and D3T (see Fig. 2 for chemical
structures). Two days following the last treatment with dithiolethiones, rats
were challenged with an obviously acutely toxic dose of AFB,. AFB, (50
/ig/100 g body wt) was gavaged for 2 successive days and on the third day
the rats were autopsied. The AFB, control group received the AFB, challenge and the no-AFB, control group received the AFB, vehicle, tricaprylin
only. Because of the number of assays performed, the groups treated with
252
MAXUITENKO ET AL.
Dithiolethiones
0.1 mmol/kg bw
Low
|
I
I
I
I
I
AFBj
i BrdU
I
I
I
Number of r^/group
Dithiolethiones
4
4
AFB i control group
4
4
No-AFB i control group
4
AFBj
I
I
I
I
I
I
I
I
I
I
I
I
I
I
j
I
i
I
BidU
I
0.3 mmol/kg bw
High
BrdU
I
I
I
1
2
3
I
I
4 5 6 7 8
Days of Experiment
I
9
FIG. 1. Experimental protocol for evaluation of the inhibition of AFBrinduced hepatotoxicity by selected dithiolethiones. Rats were gavaged with
0.1 or 0.3 mmol/kg body wt of D2T, ADT, OLT, 5mD3T, or D3T at 6, 4, and 2 days prior to the treatment with two doses of 50 jig AFB, /100 g rat/
day given by gavage 24 hr apart. One day after the AFB, treatment animals were killed, serum enzyme activities measured, and BDC labeling index
and extent of hepatic damage determined. Arrows indicate single treatments of 50 jig AFB ,/l00 g body wt; asterisks indicate single doses of dithiolethione.
high and low doses of dithiolethiones were evaluated separately, each with
their own AFB, and no-AFB, control groups.
All rats were given two doses of 10 mg BrdU/100 g body wt 5 and 2 hr
prior to autopsy and were killed 28 hr after the last dose of AFB,. Blood
was collected at autopsy and serum separated, and sALT and sSDH activities were determined within 48 hr. Livers were removed and weighed.
Multiple 3-mm sections of hepatic tissue were cut from the right lateral
lobe of the liver. These sections were fixed in acetone and embedded in
paraffin; one section was stained with hematoxylin and eosin (H&E) and
a serial section was stained immunohistochemically for the expression of
3tf-1.2-Dilbiole-3-llilo*
(D3T)
h
l,3-Dithlol«-2-lblo»«
(D2T)
(SmDJT)
Ckifi
4-Mcliyl-5-p7Tailiiyl]tf-1,2-41Uilolc-)-llUone
(.OLT)
FIG. 2.
3H-1.2-dllhlole3-lhiono
(ADT)
Chemical structures of selected dithiolethiones.
BrdU. The extent of hepatic damage was evaluated on H&E slides. The
following histological grading system was used to characterize hepatic parenchymal toxicity: 0 = normal appearance; 1 = scattered, individual cells
showing pyknosis, karyolysis, and/or karyorrhexis, between 1 and 10 cells/
hepatic lobule; 2 = scattered, individual cells showing pyknosis, karyolysis,
and/or karyorrhexis, more than 10 cells/lobule; 3 = confluent areas of
necrosis, less than 10% of analyzed area; 4 = confluent areas of necrosis,
between 10 and 50% of analyzed area; 5 = confluent areas of necrosis,
more than 50% of analyzed area. All slides were scored twice by the same
investigator blinded to the identity of the slides. For each rat an average
of the two scores was reported.
The percentage of bile duct cells (BDCs) labeled with BrdU was determined by microscopic examination of the immunohistochemically stained
sections. A cell was defined as a BDC if its basal surface rested on a
basement membrane and had a closed circular or longitudinal profile (Kossor et aL, 1995). Slides were scored blindly as to the identity of the treatment On average, 700 to 900 BDCs were counted in a total of 20 to 30
randomly selected microscopic fields. Based on preliminary observations
and to decrease the total number of BDCs that must be counted, two doses
of BrdU were used. Labeling index was calculated by dividing the number
of labeled nuclei by the total number of nuclei counted.
Tumorigtnesis experiment
A schematic of the experimental protocol
is given in Fig. 3. The same five dithiolethiones were evaluated at the same
doses as in the acute toxicity experiment. The schedule of treatment was
similar to that described above except that it extended throughout the 2
weeks of AFB, treatment (25 /ig/rat given five times a week for 2 successive
weeks) to conform to the previous tumorigenesis experiments (Kensler et
al., 1987, 1992a; Liu et al., 1988; Maxuitenko et al., 1993). Both the lowdose and the high-dose (0.1 and 0.3 mmol/kg body wt, respectively) groups
were evaluated simultaneously. The AFB, control group received the AFB,
challenge and the no-AFB, control group received the AFB, vehicle, tricaprylin only. All groups were composed of eight rats.
All rats were killed 5 weeks after the dithiolethione and AFB, treatments.
At autopsy, livers were removed and weighed. Multiple 3-mm sections of
hepatic tissue were cut from the right lateral lobe of the liver. These sections
were fixed in acetone, embedded in paraffin, stained immunohistochemically for the expression of GST-P by the peroxidase-antiperoxidase com-
CORRELATION OF AFLATOXIN B, TOXICITY WITH ITS TUMORIGENICITY
253
Dithiolethiones
0.1 mmol/kg bw
Low
AFBj
i.t.i.i.i.
0J mmol/kg bw
High
1.1.1.1.1.
vfm TTm
*»• ** * * * »
I
I
I
I
TTtTr t m f
I
I
I
Number of ran/group
Dithiolethiones
8
8
I ,, I
AFBj
1.1.1.1.1.
.M.I.I.I.
I
J
I
1
2
I
L
AFB i control group
L
No-AFB i control group
I
3
4
8
Weeks of Experiment
9
FIG. 3. Expenmental protocol for evaluation of the chemopreventive effects of pretreatment with selected dithiolethiones against AFB,-induced
carcinogenesis. Rats were gavaged with 0.1 or 0.3 mmol/kg body wt of D2T, ADT, OLT, 5mD3T, or D3T 3 times a week for 3 weeks starting 1 week
prior to the AFB, treatment. Rats received 25 /jg AFB,/rat/day by gavage, five times a week for two successive weeks. Livers were analyzed for GSTP-positive foci 5 weeks after the last dose of AFB,. Arrows indicate single treatments of 25 /xg AFB, /rat; asterisks indicate single doses of dithiolethione
plex method (Kaku et at., 1983), and examined by light microscopy. Focal
transectional areas of the GST-P-positive foci and areas of the tissue sections
were determined with an image analysis system (Kensler et al., 1987). The
minimal size of a GST-P-positive focus that was counted contained at least
five nucleated hepatocytes. Reproducibility of the counts was checked by
evaluation of the same sections by two different investigators.
Statistical analysis. Data are expressed as means ± SE. Body weights,
liver weights, toxicological indices, and focal data were analyzed by oneway ANOVA followed by Bonferroni multiple comparison tests. The degree
of closeness of the linear relationship between two parameters was measured
by the correlation coefficient r.
RESULTS
Acute Toxicity Experiment
Using the same tumorigenic protocol as in this report,
Liu et al. (1988) determined that maximum AFB r induced
hepatic toxicity was sustained within the first two daily doses
of AFB,. Therefore, we chose to evaluate toxicity following
two doses of AFB,. Proliferation of the hepatobiliary system
is a classic observation following exposure to AFB! (Wogan,
1976; Roebuck and Maxuitenko, 1994). In a preliminary
experiment, 25 fig AFB!/100 g body wt for 2 successive
days yielded a labeling index on the third day of 8% for the
BDCs. Increasing the dose to 50 fig AFB,/100 g body wt
for 2 days resulted in a labeling index of 42%. This latter
labeling index provided a large dynamic range for an assay
dependent on the measurement of the diminution of proliferation. It also should be noted that the change in the BDC
labeling index as a result of the AFB, treatment was more
pronounced than the change in the activity of serum enzymes. While both sALT and sSDH activities rose only 1.7-
fold 24 hr after two doses of 25 fig of AFB,, the change in
BDC labeling index was 6.6-fold.
Although low doses of dithiolethiones (0.1 mmol/kg body
wt) did not reduce the body weight gain, at the high dose
(0.3 mmol/kg body wt) there was some inhibition in normal
weight gain, especially in the D2T group (see Table 1). Rats
fed the low doses of ADT and OLT gained more weight
than rats in both control groups. AFB, treatment resulted in
body weight losses in all groups receiving low doses of
dithiolethiones when compared with no-AFB, control rats.
Pretreatment with D3T and 5mD3T partially protected
against this effect. With the high doses of dithiolethiones,
protection against AFB i-induced body weight loss was increased. At both doses of dithiolethiones, 5mD3T and D3T
were more effective than D2T, OLT, and ADT.
AFB, treatment resulted in a statistically significant increase in sSDH and sALT activities when compared with
no-AFB, controls (see Fig. 4). When low doses of dithiolethiones were tested, only D3T produced a significant reduction
in sALT activity. sSDH and sALT activities were significantly reduced by high doses of OLT, 5mD3T, and D3T.
The high doses of 5mD3T and D3T offered almost complete
protection against AFB,-induced elevation of hepatic enzyme activities in serum.
At autopsy no gross abnormalities of the livers were seen.
Under a light microscope, liver architecture and hepatocyte
cytology of the no-AFB, control group appeared normal
(see Table 1, histological index). The AFB, treatment led to
degeneration and necrosis of the liver in the periportal and
into the midzonal region of the lobule. Cellular damage,
distortion of hepatic architecture, and hemorrhage, as well
254
MAXUITENKO ET AL.
TABLE 1
Effect of Pretreatment with 0.1 or 0.3 mmol/kg body wt of Selected Dithiolethiones on AFBi-Induced Hepatotoxiciry
Treatment
Dithiolethiones
Body weight gain" (g/day)
AFB,
A. 0.1 mmol dithiolethiones/kg
—
D2T
ADT
OLT
5mD3T
D3T
—
B. 0.3 mmol dithiolethiones/kg
—
D2T
ADT
OLT
5mD3T
D3T
body wt
+
+
+
+
+
+
body wt
+
+
+
+
+
+
—
Dithiolethiones*
+4.0
+4.1
+4.7
+4.4
+3.8
+4.2
+4.0
±
±
±
±
±
±
±
0.2
0.2
0.2
0.1
0.4
0.2
0.6
+4.8 ± 0.2
+2.9 ± 0.2^
+4.1 ± 0.1
+4.4 ± 0.1
+3 3 ± 0.2*'
+4 1 ± 0.2
+4.5 ± 0.1
AFB,'
Histological
index"
BDC labeling
index" (%)
42.6 ± 2.6'
39 1 ± 2.1'
47.7 ± 1.1'
36.2 ± 3.7'
26.1 ± 2.0*'
19.4 ± 1.7*'
0.3 ± 0.1'
-4.5
-4.0
-5.3
-5.1
-1.1
-0.6
+2.4
± 0.7'
± 0.2'
± 0.6'
±0.3'
± 0.4*'
± 0.5*'
± 0.7'
4.4
4.0
4.8
4.6
3.9
3.6
0.1
± 0.2*
± 0*
± 0.1'
± 0.2*
± 0 1'
± 0 3'
± 0.1'
-4.4
-3.0
-4.4
-2.5
+1.5
+ 1.8
+ 2.5
± 0.7'
± 0.4'
± 0.4'
± 0.7'
± 0 5'
± 0.4'
± 0.4'
49 ± 0.1'
4 3 ± 0.3'
4.8 ± 0.1'
4.1 ± 0.2*
3.0 ± 0*'
1.3 ± 0.3'
0.8 ± 0.1'
44.2
44.8
44.6
34.0
14.6
3.2
0.5
± 1.7*
± 1.1'
± 1.1*
± 3.5*
± 4.1*'
± 1.2'
± 0 1'
° Mean ± SE, n = 4.
* Mean value for 6 days of pretreatment with dithiolethiones.
c
Mean value for 2 days of treatment with AFB,.
'Statistically different from no-AFB, control group (p < 0.05) for each dose of dithiolethiones separately.
' Statistically different from AFB, control group (p < 0.05) for each dose of dithiolethiones separately.
as proliferation of portal bile ducts, were observed. Pretreatment of rats with dithiolethiones afforded differing degrees
of protection from AFB r induced hepatotoxicity. At the low
doses of dithiolethiones, only D3T afforded protection,
though, it was not statistically significant. At the high dose
of dithiolethiones, however, protection afforded by 5mD3T
and D3T was significant (see Table 1).
BrdU labeling of the livers of the AFB,-treated rats revealed prominent labeling of the cells in the periductal space
and hepatobiliary system, whereas the hepatocytes largely
remained unlabeled. The BDC labeling index was increased
from less than 1% in the no-AFB i control groups to greater
than 40% in the AFB, control groups. This similar response
was expected and observed in the two experiments since the
same dose of AFB| was used. At both the low and high
doses of dithiolethiones, 5mD3T and D3T reduced the labeling index, with the high dose offering a greater reduction
(Table 1).
Tumorigenesis Experiment
As expected, all rats survived the treatment and grew at
comparable rates. Body weights at the termination of the
experiment did not differ significantly between the groups,
nor did the liver weights (data not shown).
The influence of pretreatment with dithiolethiones on the
observed number of GST-P-positive foci is shown in Table
2. GST-P-positive foci are rarely encountered in no-AFB,
control rats, whereas the treatment with AFB, resulted in
observation of approximately 30 foci when 2 to 3 cm2 of
tissue was examined. Pretreatment with dithiolethiones decreased the observed number of GST-P-positive foci. The
high dose was more protective, and ADT, OLT, 5mD3T, and
D3T afforded considerable protection. Statistical analysis of
the "observed" focal data is subject to sampling biases
(Pugh et ai, 1983) since large foci are more likely to be
observed than small foci. For this reason statistical approaches cannot be applied directly to the "observed" data.
To avoid this bias, morphometric transformation of the data
is necessary prior to the analysis (Pugh et ai, 1983).
The influence of pretreatment with the dithiolethiones on
the number of GST-P-positive foci per unit volume of liver
and focal size is summarized in Table 2. The low doses of
D3T or 5mD3T reduced the number of GST-P-positive foci
per cubic centimeter by more than fivefold and threefold,
respectively, while OLT, ADT or D2T had no significant
effect. Pretreatment with the high doses of dithiolethiones
significantly decreased the number of AFB,-induced GSTP-positive foci per cubic centimeter in all groups. While
it seemed that some treatment groups had hepatic GST-Ppositive foci that were distinctly smaller than others, the
difference in focal size was not statistically significant.
The influence of pretreatment with the dithiolethiones on the
CORRELATION OF AFLATOXIN B, TOXICITY WITH ITS TUMORIGENICITY
A.
0.1 mmol/kg body weight
255
0.3 mmol/kg body weight
BO
tn
60
X
o
o
o
40
o
o
20
a
C/J
a
-
D2T AOT OLT 5 m - D 3 T
D3T
AFB. +
B.
+
+
+
-
-
02T ADT OLT 5m-D3T
D3T
AFB, +
+ -
0.1 mmol/kg body weight
+
+
+
-
+ -
0.3 mmol/kg body weight
14
to
X
o
o
o
>>
">
o
o
a
1FD
D2T ADT OLT 5 m - D 3 T
D3T
4.
4-
4-
4-
-
4-
AFB
D2T ADT OLT 5 m - D 3 T
D3T
4.
+
+
+
4-
-
FIG. 4. Effect of pretreatment with 0.1 or 0.3 mmol/kg body wt of selected dithiolethiones on serum sorbitol dehydrogenase, sSDH (A) and serum
alanine aminotransferase, sALT (B), activities after two doses of 50 fj.% AFB,/100 g body wl. Rats were treated as described in Fig. I. Measurements
were made 28 hr after the last dose of AFB,. D, No-AFB, control group; • , AFB, control group; • , dithiolethione-pretreated groups Results are means
± SE, n = 4 "Statistically different from no-AFB, control group (p < 0.05); 'statistically different from AFB, control group (p < 0 05).
volume percentage of GST-P-positive foci and the extent of
inhibition of the focal burden is summarized in Table 3. It
should be noted that the mean focal diameter and the number
of foci per cubic centimeter of the liver must be calculated
using statistical probabilities (Pugh et ai, 1983) and are less
reliable parameters than the volume percentage of the liver
occupied by foci. Volume percentage can be directly calculated
from the number and transectional areas of foci as seen by
light microscopy. The low doses of D3T or 5mD3T resulted
in inhibition of GST-P-positive foci, while ADT or OLT did
not have a statistically significant inhibitory effect on focal
burden. D2T pretreatment had an enhancing effect, yielding a
50% increase in the focal burden. The high doses of D3T,
5mD3T, OLT, and ADT produced even larger reductions in
the focal burden than the low doses. The high dose of D2T
did not have any effect on the volume percentage of GST-Ppositive foci. A comparison of the volume percentage of GSTP-positive foci in the dithiolethione groups to the volume percentage in the AFB, control group revealed that inhibition of
focal burden by greater than 50% was afforded by low doses of
ADT, 5mD3T, and D3T and by high doses of all dithiolethiones
except D2T.
DISCUSSION
Because of concerns regarding the stability of some dithiolethiones in the diet and limited quantities of many dithio-
256
MAXUITENKO ET AL.
TABLE 2
Effect of Pretreatment with 0.1 or 0.3 mmol/kg body wt of Selected Dithiolethiones on Hepatic GST-P-Positive Foci
Number of foci/cm3 of liver"
Number of foci observed4*
Mean focal diameter", (jtm)
Treatment
Dithiolethiones
_
0.1 mmol/
kg body wt
AFB,
+
+
+
+
+
D2T
ADT
OLT
5mD3T
D3T
—
27
24
21
22
7
2
+
-
±
±
±
±
±
0.3 mmol/
kg body wt
0.1 mmol/
kg body wt
16
9
4
1
0.4
289
402
325
112
69
± 4
2
2
4
2
1
±
±
±
±
±
3
1
1
0.4
0.3
±
±
±
±
±
0.1 :± 0.1
0.3 mmol/
kg body wt
0.1 mmol/
kg body wt
386 ± 45'
31C
244 ± 4 7 ^
4T
170 ± 25~*
6T
82 ± 18*
3tf
20 ± 6'
20*
12 ± 9*
2 ±
0.3 mmol/
kg body wt
306 ± 38
429
225
315
308
117
364
268
222
218
159
± 48
± 25
±41
± 53
± 18
±
±
±
±
±
77
53
37
25
97
163
" Mean ± SE, n = 8.
6
On average 2 - 3 cm2 of hepatic tissue per animal was analyzed. Statistical analysis is inappropriate before morphometric transformation.
c
Statistically different from no-AFB, control group (p < 0.05) for each dose of dithiolethiones separately.
' Statistically different from AFB, control group (p < 0.05) for each dose of dithiolethiones separately.
lethiones synthesized, we chose to gavage the dithiolethiones
instead of mixing them in the diet and feeding as we traditionally have done (Kensler et al., 1987, 1992a; Liu et al,
1988; Bolton et al, 1993; Maxuitenko et al., 1993). The
additional advantage of gavage is the assurance that all the
rats receive equal amounts of the compound. The rats were
gavaged with the dithiolethiones three times a week. The
doses were designed to approximate the weekly consumption
of dithiolethiones from the diet (Kensler et al., 1987, 1992a).
That is, gavage with 0.1 mmol OLT or D3T/kg body wt in
this experiment is approximately equivalent to 0.01% (100
ppm) or 0.006% (60 ppm) in the diet, respectively. Gavage
with 0.3 mmol of OLT or D3T is approximately equivalent
to 0.03% (300 ppm) or 0.02% (200 ppm) in the diet, respec-
tively. It should be noted that gavage with dithiolethiones,
despite all the advantages mentioned earlier, has one disadvantage: it seems that delivery of dithiolethiones by gavage
as a cumulative dose of what the animal would have eaten
in the diet over approximately 48 hr results in some toxicity.
In the preliminary experiment, we found that a dose of 0.5
mmol D3T/kg body wt (0.03% when given in the diet) was
toxic to the rats when given by gavage, but was tolerated
when given in the diet (Kensler et al., 1992a). At 0.5 mmol/
kg body wt, OLT and ADT showed no adverse effects when
given by gavage, but 5mD3T and D2T were toxic. Like
D3T, no toxicity was seen with repeated doses of these
selected dithiolethiones at the 0.3 mmol/kg level. Despite
these adverse effects which partly appear to be related to
TABLE 3
Effect of Pretreatment with 0.1 or 0.3 mmol/kg body wt of Selected EHthiolethiones on the Volume Percentage
of the Liver Occupied with GST-P-Positive Foci and the Extent of Inhibition of the Focal Burden
Volume percentage of GST-P-positive foci"
Percentage inhibition
Treatment
Dithiolethiones
AFB,
D2T
ADT
OLT
5mD3T
D3T
—
+
+
+
+
+
+
-
0.1 mmol/Kg
body wt
0.3 mmol/kg
body wt
1.80
2.69
0.70
1.12
0.27
0.012
±
±
±
±
±
± 0.6*
±0.3
o.4i ± o.r
0.10 ± 0.04r
0.01 ± 0.005C
o.r
0.004r
0.001
0.3 mmol/kg
body wt
0
1.03
0.5"
0.2
0.2
0.1 mmol/kg
body wt
o.oo3 ± o.oor
± o.oor
" Mean ± SE, n = 8.
* Statistically different from no-AFB, control group (p < 0.05) for each dose of dithiolethiones separately.
' Statistically different from AFB, control group (p < 0.05) for each dose of dithiolethiones separately.
(-50)
61
38
85
99.4
43
77
94
99.4
99.8
257
CORRELATION OF AFLATOXIN B, TOXICITY WITH ITS TUMORIGENICITY
B.
A.
120 _
E
N^
A•
r-0.943. p < 0 01
''A
100
" /
in
H
o
o
ao
-
#
• ' / • '
60
i
'
/.A
40
u
a
0
20
<
n
0
/ •
y
// y
- oS*-'
10 •*
10" 1
10" 1
10" 1
10*
101
10°
Volume X of GST-P+ foci
C.
10 1
Volume X of GST-P+ foci
D.
X
x
•a
•o
3
_o
o
a
.a
o
a
m
0.0
10
4
10" 1
10" 1
10" 1
10§
101
Volume X of GST-P+ foci
101
Volume X of GST-P+ foci
FIG. 5. Correlation between the volume percentage of liver occupied by GST-P-positive foci and markers of hepatotoxicity: (A) sALT activity; (B)
sSDH activity; (C) histological index; (D) BDC labeling index. Measurements were made 28 hr after the last dose of AFB,. Each symbol represents
one treatment group. O, No-AFB, control group; • , AFB, control group; A, groups that received 0.1 mmol/kg body wt of selected dithiolethiones; • ,
groups that received 0.3 mmol/kg body wt of selected dithiolethiones. Dashed line (- -) represents 95% confidence interval of the regression line (—).
the bolus delivery of the drugs, gavage is advantageous for
limited quantities of drugs, uniform dosage, and stability.
Classically, bile duct hyperplasia has been associated with
AFBi exposure. BDC proliferation is seen in chickens,
ducks, turkeys (Newberne, 1973), swine (Miller et al., 1981),
dogs (Newberne et al., 1966), horses (Angsubhakorn et al,
1981), and rats (Wogan, 1976; Roebuck and Maxuitenko,
1994) exposed to AFB,. We have quantified BDC proliferation resulting from two doses of 50 /j,g of AFB,. As was
mentioned earlier, the acute treatment protocol of two doses
of 50 /ig of AFB, was chosen to achieve a wider range of
toxic responses against which the efficacy of dithiolethiones
might be judged. The total dose of AFB, approximates the
quantity of AFB, a rat would receive during the first week
of a tumorigenic treatment protocol, when most of hepatic
toxicity occurs (Liu et al., 1988). The extent of BDC proliferation significantly (p < 0.01, all 56 rats) correlated with
traditional markers of hepatic toxicity, such as sALT (r =
0.835), sSDH (r = 0.829), and extent of hepatic damage (r
= 0.902). These findings clearly allow BDC proliferation to
be used as an index of AFB,-induced hepatotoxicity. Under
conditions of low toxicity, it appears that the BDC labeling
index is a more sensitive marker of hepatotoxicity and will
be superior to serum markers of toxicity in discriminating
among effective compounds. For example, treatment with
the high dose of D3T resulted in a 6.4-fold increase in the
BDC labeling index; whereas, sALT and sSDH activities
were increased only 3- and 1.4-fold, respectively when compared with the no-AFB, control group.
Previous studies demonstrated that ingestion of OLT or
258
MAXUITENKO ET AL.
D3T during AFB, treatment diminished hepatotoxicity (Liu
et al., 1988), retarded the development of putative preneoplastic lesions (Kensler et al, 1987, 1992a; Bolton et al,
1993; Maxuitenko et al, 1993), and ultimately prevented
cancer (Roebuck et al., 1991). However, neither the relationship between hepatotoxicity and hepatocarcinogenicity associated with AFBj exposure nor the mechanisms responsible
for BDC hyperplasia by AFB, are understood. It was proposed (Liu et al., 1988; Kensler et al., 1992c) that OLT may
act indirectly through the activation of electrophile detoxication enzymes to decrease the hepatotoxicity of AFB, and
thereby reduce parenchyma! cell proliferation, thus conferring cancer chemoprevention. To explore any relationship
between hepatotoxicity and tumorigenicity of AFB,, all toxicological indices were correlated to the volume percentage
of the liver occupied by GST-P-positive foci. These correlations were statistically significant (p < 0.01) for all toxicological indices used (Fig. 5). While correlation is certainly
not causation, the degree of association is impressive and
consistent. These results imply that inhibition of AFB r induced hepatotoxicity affords protection against putative preneoplastic focal development. The explanation could be that
cell death resulting from AFBj exposure causes loss of liver
cell mass which in turn results in compensatory cell proliferation. Hepatotumorigenicity studies have demonstrated that
proliferating cells are more susceptible to carcinogens (Solt
and Farber, 1976), and cell proliferation has been recognized
as an important factor in cancer development (Cohen and
Ellwein, 1990). It is likely to play an important role in fixation of carcinogen-induced DNA damage and in the clonal
expansion of initiated cells. The important role of cell regeneration induced by hepatotoxins requires further investigation to understand the mechanisms that control cell division
in postnecrotic liver regeneration.
Of the more than 50 dithiolethione analogs we have synthesized and the many others that are already available, 5
were chosen to span a wide range of chemopreventive responses for the development of this approach. Based on
the activities of these 5 compounds, it seems clear that the
presence of the 3//-l,2-dithiole-3-thione structure is necessary for the compound to exhibit protective properties. Thus,
the analog lacking this moiety, D2T, was devoid of chemopreventive activity. The requirement of the 3W-l,2-dithiole3-thione nucleus for activity, as well as the relative ranking
among the compounds in our study, agrees reasonably well
with previous observations of chemoprevention by these
agents in vivo. Thus, Bueding and co-workers found that
OLT and ADT afforded similar protection against hepatotoxicity induced by carbon tetrachloride or acetaminophen in
female CD-I mice, as measured by survival and by effect
on serum enzyme, glutathione, and glutathione S-transferase
levels (Ansher et al, 1983). In a more extensive study which
examined phase 2 enzyme induction in various tissues of
mice and rats, these workers again found OLT and ADT to
have similar activities, although OLT appeared somewhat
more active than ADT (Ansher et al., 1986). In our own
study of the effects of 8 dithiolethione analogs on AFB,
metabolism and DNA adduct formation in male F344 rats,
we found D3T to be considerably more effective at induction
of phase 2 enzymes and suppression of AFB,-DNA adduct
formation than either OLT or ADT, with OLT being somewhat more effective than ADT (Kensler et al., 1987). Structure-activity information is also available from two in vitro
studies employing the Hepa Iclc7 murine hepatoma cell
line. The Bueding group found the parent D3T to be more
effective than either OLT or ADT at induction of
NAD(P)H:quinone reductase in this cell line, with OLT and
ADT displaying similar, although somewhat dose-dependent, activity (De Long et al., 1986). Under some conditions,
however, OLT was found to be more effective at elevating
cellular glutathione levels than was D3T. In a study that
measured the ability of 25 dithiolethiones and analogs to
induce NAD(P)H:quinone reductase in Hepa Iclc7 cells
transfected with the p41-285GH plasmid, D3T was found
to be a more potent inducer than OLT, ADT, or 5mD3T,
whose inducing potencies were very similar (Egner et al.,
1994). In agreement with the current study, D2T was found
to be devoid of inducing activity. Given the many different
measures of chemoprevention that have been employed, the
correspondence between previous structure-activity studies
and the present study is quite good, suggesting that the methods presented here constitute a valid approach to screening
the many available dithiolethiones for cancer chemopreventive activity.
To date, we have identified 10 to 20 dithiolethiones that
have greater enzyme induction properties in vitro and in vivo
than does OLT (Kensler and Curphey, unpublished data). It
is anticipated that many of them will be found to be more
effective than OLT, which is an extremely effective cancer
chemopreventive agent (Roebuck et al., 1991; Bolton et al.,
1993).
ACKNOWLEDGMENTS
The authors express their appreciation to the following individuals who
contributed to the completion of the study: Karen Baumgartner, Laune
Bergeron, Adam Libby. and Denise MacMillan. This work was supported
by NIH Grant CA-39416
REFERENCES
Angsubhakorn. S.. Poomvises. P.. Romruen, K . and Newberne. P. M.
(1981). Aflatoxicosis in horses J. Am. Vet. Med Assoc. 178, 274-278
Ansher, S. S., Dolan, P,, and Bueding. E. (1983). Chemoprotective effects
of two dithiolethiones and of butylhydroxyanisole against carbon tetrachloride and acetaminophen toxicity. Hepatology 3, 932-935.
CORRELATION OF AFLATOXIN B, TOXICITY WITH ITS TUMORIGENICITY
Ansher, S. S., Dolan, P., and Bueding, E. (1986) Biochemical effects of
dithiolthiones. Food Chem. Toxicol. 25, 405-415.
Archer, S. (1985). The chemotherapy of schistosomiasis. Annu. Rev. Pharmacol. Toxicol. 25, 485-508.
Barreau, M., Cotrel, C , and Jeanmart, C. (1977). German Patent 2705641.
Benson, A. B., III. (1993). Oltipraz: A laboratory and clinical review. J.
Cell. Biochem. Suppl. 17F, 278-291.
Bieri, J. G. (1980). Second report of the ad hoc committee on standards
for nutritional studies. J. Nutr. 110, 1726.
Bolton, M. G., Mufioz, A., Jacobson, L. P., Groopman, J. D., Maxuitenko,
Y. Y., Roebuck, B. D., and Kensler, T. W. (1993). Transient intervention
with oltipraz protects against afiatoxin-induced hepatic tumorigenesis.
Cancer Res. 53, 3499-3504.
Bueding, E., Dolan, P., and Leroy, J.-P. (1982) The antischistosomal activity of oltipraz. Res. Commun. Chem. Pathol Pharmacol. 37, 293-303.
Cohen, S. M., and Ellwein, L. B (1990). Cell proliferation in carcinogenesis. Science 249, 1007-1011.
Davies, M. H., Schamber, G. J., and Schnell, R. C. (1987). Role of glutathione in oltipraz-induced protection in acetaminophen and allyl alcohol
hepatotoxicity in the male hamster. Toxicohgist 7, 219.
De Long, M J., Dolan, P., Santamana, A. B , and Bueding, E. (1986). 1,2Dithiol-3-thione analogs. Effects on NAD(P)H:quinone reductase and
glutathione levels in murine hepatoma cells. Carcmogenesis 7, 977-980.
Egner, P. A., Kensler, T. W., Prestera, T., Talalay, P., Libby, A. H., Joyner,
H. H , and Curphey, T J. (1994). Regulation of phase 2 enzyme induction
by oltipraz and other dithiolethiones. Carcinogenesis 15, 177-181.
HSusler, R., and Ritschard, J. (1979). Clinical study with a sialagogue
drug (Sulfarlem S 25 = TTPT) in the treatment of xerostomia. Schweiz,
Rundsch. Med. (Prax.) 68, 1063-1068
Kaku, T., Ekem, J. K., Lindayen, C , Bailey, D. J., van Nostrand,
A. W. P , and Farber, E (1983). Comparison of formalin- and acetonefixation for immunohistochemical detection of carcinoembryonic antigen
(CEA) and keratin. Am. J. Clin. Pathol. 80, 806-815.
Kensler, T. W., Egner, P. A., Dolan, P. M., Groopman, J. D., and Roebuck,
B. D. (1987). Mechanism of protection against aflatoxin tumorigenicity
in rats fed 5-(2-pyrazinyl)-4-methyl-l,2-<iithiol-3-thione (oltipraz) and
related 1,2-dithiol-3-thiones and 1,2-dithiol-3-ones. Cancer Res. 47,
4271-4277.
Kensler, T. W., Egner, P. A., Trush, M. A., Bueding, E., and Groopman,
J. D (1985). Modification of aflatoxin B, binding to DNA in vivo in rats
fed phenolic antioxidants, ethoxyquin and a dithiothione. Carcinogenesis
6, 759-763.
Kensler, T. W., and Helzlsouer, K. J. (1995). Oltipraz: Clinical opportunities
for cancer chemoprevention. J. Cell. Biochem. Suppl. 22, 101-107.
Kensler, T W., Groopman, J. D., Eaton, D. L., Curphey, T. J., and Roebuck,
B. D. (1992a). Potent inhibition of afiatoxin-induced hepatic tumorigenesis by the monofunctional enzyme inducer 1,2-dithiole-3-thione. Carcinogenesis 13, 95-100.
Kensler, T. W., Groopman, J. D., and Roebuck, B. D. (1992b). Chemoprotection by oltipraz and other dithiolethiones. In Cancer Chemoprevention
(L Wattenberg, M. Lipkin, C. W Boone, and G. J. Kelloff, Eds.), pp.
205-226. CRC Press, Boca Raton, FL.
Kensler, T., Styczynski, P., Groopman, J., Helzlsouer, K., Curphey, T.,
259
Maxuitenko, Y., and Roebuck, B. D. (1992c). Mechanisms of chemoprotection by oltipraz. J. Cell. Biochem. Suppl. 161, 167-172.
Kossor, D C , Goldstein, R. S., Ngo, W., DeNicola, D. B., Leonard, T. B.,
Dulik, D. M , and Meunier, P. C. (1995). Biliary epithelial cell proliferation following a-naphthylisothiocyanate (ANIT) treatment: Relationship
to bile duct obstruction. Fundam. Appl. Toxicol. 26, 51-62.
Liu, Y -L.. Roebuck, B. D., Yager, J. D., Groopman, J. D., and Kensler,
T. W. (1988). Protection by 5-<2-pyrazinyl)-4-rnethyl-l,2-dithiol-3-thione
(oltipraz) against the hepatotoxicity of aflatoxin B, in the rat. Toxicol.
Appl. Pharmacol. 93, 442-451.
Maxuitenko, Y. Y., MacMillan, D. L., Kensler, T. W., and Roebuck, B. D.
(1993). Evaluation of the post-initiation effects of oltipraz on aflatoxin
B i-induced preneoplastic foci in a rat model of hepatic tumorigenesis.
Carcinogenesis 14, 2423-2425.
Meinetsberger, E., Schoffer, A., and Behringer, H. (1977). Eine einfache
und ergiebige Herstellung von 3-Thioxo-l,2-dithiol (1,2-Trithion). Synthesis, 802-803.
Miller, D. M., Stuart, B P., and Crowell, W. A. (1981). Experimental
aflatoxicosis in swine: Morphological and clinical pathological results.
Can. J. Comp. Med. 45, 343-351.
Newbeme, P M (1973). Chronic aflatoxicosis. J Am. Vet. Med. Assoc.
163, 1262-1269.
Newbeme, P. M., Russo, R., and Wogan, G N. (1966). Acute toxicity of
aflatoxin B, in the dog. Path. Vet. 3, 331-340.
Prochaska, H. J., Chavan, S. J., Baron, P., and Polsky, B. (1995). Oltipraz,
a novel inhibitor of human immunodeficiency virus type I (HIV-1) replication. J. Cell. Biochem. Suppl. 22, 117-125.
Prochaska, H. J., De Long, M. J., and Talalay, P. (1985). On the mechanism
of induction of cancer protective enzymes. A unifying proposal. Proc.
Nail. Acad Sci. USA 82, 8232-8236.
Pugh, T. D., King, J. H., Koen, H., Nychka, D., Chover, J., Wahba, G.,
He, Y., and Goldfarb, S (1983). Reliable stereological method for estimation the number of microscopic hepatocellular foci from their transections. Cancer Res. 43, 1261-1268.
Rao, C. V., Rivenson, A., Katiwalla, M., Kelloff, G. J., and Reddy, B. S.
(1993). Chemopreventive effect of oltipraz during different stages of
experimental colon carcinogenesis induced by azoxymethane in male
F344 rats. Cancer Res. 53, 2502-2506.
Roebuck, B. D., Liu, Y.-L., Rogers, A. E., Groopman, J. D., and Kensler,
T. W. (1991). Protection against aflatoxin B,-induced hepatocarcinogenesis in F344 rats by 5-<2-pyrazinyl)-4-methyl-1,2-dithiole-3-thione (oltipraz): Predictive role for short-term molecular dosimetry. Cancer Res.
51, 5501-5506.
Roebuck, B. D , and Maxuitenko, Y. Y. (1994). Biochemical mechanisms
and biological implications of the toxicity of aflatoxins as related to
aflatoxin carcinogenesis In The Toxicology of Aflatoxins: Human Health,
Veterinary, and Agricultural Significance (D. L. Eaton and J. D. Groopman, Eds.), pp. 27-44. Academic Press, San Diego.
Solt, D., and Farber, E. (1976). New principle for the analysis of chemical
carcinogenesis. Nature 263, 701-703.
Wattenberg, L. W., and Bueding, E. (1986). Inhibitory effects of 5-(2pyrazinyI)-4-methyl-l,2-dithiole-3-thione (oltipraz) on carcinogenesis induced by benzolajpyrene, diethylnitrosamine and uracil mustard Carcinogenesis 7, 1379-1381.
Wogan, G. N. (1976). The induction of liver cell cancer by chemicals. In
Liver Cell Cancer (H. M. Cameron, D. A. Linsell, and G. P. Warwick,
Eds.), pp. 121-152. Elsevier Scientific, Amsterdam/New York.