Carcinogenesis vol.20 no.8 pp.1641–1644, 1999
SHORT COMMUNICATION
Development of resistance during the early stages of experimental
liver carcinogenesis
Aroon Yusuf, Prema M.Rao, Srinivasan Rajalakshmi and
Dittakavi S.R.Sarma1
Department of Laboratory Medicine and Pathobiology, Faculty of Medicine,
University of Toronto, Medical Sciences Building, Toronto, Ontario, Canada
M5S 1A8
1To whom correspondence should be addressed
Email: [email protected]
The present study was designed to determine whether the
resistant phenotype is acquired at the initiated cell stage
itself or requires further exposure to a promoting regimen
to express resistance. Male Fischer 344 rats were initiated
with diethylnitrosamine (DENA) (200 mg/kg i.p.) and were
subjected to either no further treatment or to the resistant
hepatocyte (RH) model of liver tumor promotion. Six weeks
later, the resistance of the focal lesions generated in these
two groups to the mitoinhibitory effects of 2-acetylaminofluorene (2-AAF) was determined by subjecting the rats to
two-thirds partial hepatectomy (PH) in the presence of a
mitoinhibitory dose of 2-AAF (5 mg/kg i.p.) given at the time
of PH. Labeling index was determined by administering
multiple injections of [3H]thymidine. All rats were killed
48 h post-PH. While only a small percentage (23%) of the
glutathione S-transferase-positive foci generated by DENA
in the absence of an exogenous liver tumor promoting
regimen were resistant to the mitoinhibitory effects of
2-AAF, a majority (85%) of the foci became resistant to
2-AAF following exposure to the RH model of liver tumor
promotion. Further, initiated rats exposed to either 2-AAF
or to CCl4 alone, the two components of the RH model,
resulted in 71% of the foci being resistant to the mitoinhibitory effects of 2-AAF. Similar patterns of results were
obtained when the resistance of the foci to the mitoinhibitory effects of orotic acid, a liver tumor promoter and an
inhibitor of DNA synthesis in normal hepatocytes, was
monitored. These results suggest that the majority of
initiated hepatocytes are not of resistant phenotype, however, they have acquired a unique ability to express resistance upon exposure to certain agents such as 2-AAF and
CCl4 or to a promoting regimen such as the RH model of
liver tumor promotion.
One of the characteristic features of the neoplastic cell is its
resistant phenotype. In addition, in certain systems it has also
been demonstrated that neoplastic cells acquire a unique ability
to express a multi-resistant phenotype upon exposure to certain
chemicals, including cancer chemotherapeutic agents (1–3).
Experimental models of carcinogenesis, especially rat liver
models, not only substantiated the concept that the neoplastic
Abbreviations: 2-AAF, 2-acetylaminofluorene; DENA, diethylnitrosamine;
GST-7,7, glutathione S-transferase; PH, two-thirds partial hepatectomy; RH,
resistant hepatocyte.
© Oxford University Press
cell is of resistant phenotype, but also provided evidence that
‘resistance’ is expressed early in the carcinogenic process (4).
Further, hepatic nodules, a precursor population for hepatocellular carcinoma, exhibited alterations in the biochemical
machinery that are compatible with the concept that the
neoplastic cell is of resistant phenotype, expressing resistance
to cytotoxicity and mitoinhibition induced by a wide variety
of structurally diverse chemicals and also to negative growth
regulation (4–14). However, it is not clearly established whether
different components of resistance are acquired at the initiated
cell stage itself or acquired at different phases of the carcinogenic process. The present approach was developed to address
these questions.
In this study, expression of the resistant phenotype was
monitored by determining the resistance to the mitoinhibitory
effects of 2-acetylaminofluorene (2-AAF) or orotic acid, a
liver tumor promoter. Male Fischer 344 rats were initiated
with a single administration of diethylnitrosamine (DENA)
and later exposed to the resistant hepatocyte (RH) model of
liver tumor promotion (2-AAF coupled with CCl4) or to no
further treatment. Several weeks later the rats were subjected
to two-thirds partial hepatectomy (PH) in the presence of
mitoinhibitory levels of 2-AAF or orotic acid. Proliferating
hepatocytes were labeled with [3H]thymidine. We arbitrarily
defined the resistant focus as the focus that had a labeling
index higher than the highest labeling index in the surrounding
liver. If the foci generated by DENA alone (in the absence of
exposure to the RH protocol) are resistant to the mitoinhibitory
effects of 2-AAF or to orotic acid then it will be reasoned that
this resistance was acquired at the initiated hepatocyte stage
itself. On the other hand, if the foci becomes resistant upon
exposure to the RH protocol then it will be considered that
initiated hepatocytes may not be of resistant phenotype but
have acquired an ability to express resistance upon exposure
to the promoting regimen. In the present discussion, unless
otherwise stated, resistance is defined as resistance to the
mitoinhibitory effects of 2-AAF or orotic acid.
Accordingly, to determine whether the initiated hepatocyte
is a resistant phenotype, male Fischer 344 rats (Charles Rivers
Breeding Laboratories, St Constant, Quebec, Canada) weighing
120–130 g were initiated with a single necrogenic dose of
DENA (200 mg/kg i.p.; Sigma Chemical Co., St Louis, MO).
Two weeks later the rats were divided into two groups. The
rats in the first group (group 1) were not subjected to any
further treatment, whereas those in group 2 were subjected to
the RH model of liver tumor promotion (15): 20 mg/kg
intragastral 2-AAF daily for 3 days and on day 4, 2 ml/kg
intragastral CCl4, 1:1 with corn oil. Six weeks later all animals
were subjected to PH in the presence of 2-AAF (5 mg/kg i.p.)
given at the time of PH. This dose of 2-AAF inhibits DNA
synthesis by nearly 100% in normal hepatocytes. [3H]Thymidine (50 µCi/injection, sp. act. 84.6 Ci/mmol) was administered
every 4 h beginning 16 h after PH to determine the labeling
index in the surrounding liver and the foci. At 48 h post-PH,
1641
A.Yusuf et al.
Fig. 1. Percent labeling index of hepatocytes exposed to a mitoinhibitory
dose of 2-AAF (5 mg/kg) in foci/nodules and surrounding non-nodular liver
in animals initiated with a single necrogenic dose of DENA (200 mg/kg)
and exposed to no further treatment (A) or to a complete RH protocol of
liver tumor promotion (B). Every focus/nodule in the section and 30
random fields in the surrounding non-nodular liver in group A and 60
random fields in the surrounding liver in group B were counted. Every
hepatocyte, both labeled and unlabeled, in the focus/nodule and on average
100 hepatocytes in each field in the surrounding non-nodular liver were
counted. Percent labeling represents the percent of labeled hepatocytes per
total hepatocytes within the focus or within the field. Each triangle
represents one focus or one field from the surrounding liver. Data shown are
for three animals in group A and six animals in group B. The dashed line
represents the highest labeling index in the surrounding liver. Further details
are given in the text.
all animals were killed and livers were sectioned and fixed in
cold acetone for immunohistochemical staining for glutathione
S-transferase (GST-7,7). The stained sections were then processed for autoradiography (16).
The results presented in Figure 1 indicate that 23% of the
foci generated by DENA alone (group 1) in the absence of an
exogenous liver tumor promoting regimen were resistant to
the mitoinhibitory effects of 2-AAF. In contrast, 85% of the
foci initiated by DENA and promoted by the RH model were
resistant to the mitoinhibitory effects of 2-AAF (Figure 1).
The latter observation is in agreement with the conclusions
drawn from the RH model of liver tumor promotion (4). Since
the RH model has two components, i.e. 2-AAF and CCl4, it
1642
Fig. 2. Percent labeling index of hepatocytes exposed to a mitoinhibitory
dose of 2-AAF (5 mg/kg) in foci/nodules and surrounding non-nodular liver
in animals initiated with a single necrogenic dose of DENA (200 mg/kg)
and exposed to either only 2-AAF (A) or only CCl4 (B). Every focus in the
section and 50 random fields in the surrounding liver were counted. Each
triangle represents one focus or one field from the surrounding liver. Percent
labeling index was calculated as described in the legend to Figure 1. Data
shown are for five animals in each group. The dashed line represents the
highest labeling index in the surrounding liver. Further details are given in
the legend to Figure 1 and in the text.
became of interest to determine whether an exposure to either
2-AAF or CCl4 itself is sufficient for the initiated hepatocyte
to express the resistant phenotype. To answer this question, in
the next experiment male Fischer 344 rats (120–130 g) were
initiated with DENA as described in the previous experiment
and were then divided into two groups. Two weeks after the
administration of DENA, rats in group 1 received only 2-AAF
(20 mg/kg intragastrally, daily for 3 days), whereas the rats in
group 2 received only CCl4 (1:1 with corn oil, 2 ml/kg
intragastrally). Six weeks later, all the rats were subjected to
PH in the presence of 5 mg/kg 2-AAF. Rats initiated with
DENA alone without any further treatment (group 1 in the
previous experiment) were used as the control for this experiment. Resistance of the foci was determined as described
above. The results presented in Figure 2 indicate that a large
number of foci (71%) exhibited resistance to the mitoinhibitory
effects of 2-AAF. Resistance or lack of it did not appear to
depend on the size and zonal distribution of foci.
Resistance during early liver carcinogenesis
Table I. Percent of GST-7,7-positive foci/nodules resistant to the
mitoinhibitory effects of orotic acid in the livers of rats exposed to different
treatment protocols
Treatment
Resistant foci/nodule (%)
DENA
DENA
DENA
DENA
5
80
60
48
alone
1 RH protocol (2-AAF 1 CCl4)
1 2-AAF
1 CCl4
Experimental details are identical to those given in the legends to Figures 1
and 2, except that instead of 2-AAF (5 mg/kg), a tablet of 300 mg of orotic
acid methyl ester was implanted i.p. at the time of PH. This dose of orotic
acid inhibits DNA synthesis by nearly 100% in normal hepatocytes.
The values are generated from six rats in each group.
In the second series of experiments, the above-described
studies were repeated by determining the resistance of foci to
the mitoinhibitory effects of orotic acid, a liver tumor promoter
and a mitoinhibitor for normal hepatocytes. Similar to the
results obtained above, only a small percentage (5%) of foci
generated by DENA alone in the absence of any further
treatment were resistant to the mitoinhibitory effects of orotic
acid, whereas a majority (80%) of the foci initiated by DENA
and promoted by the RH model of liver tumor promotion were
resistant to the mitoinhibitory effects of orotic acid (Table I).
Further, exposure of initiated hepatocytes to 2-AAF alone or
to CCl4 alone resulted in a greater percentage (60 and 40%,
respectively) expressing resistance to the mitoinhibitory effects
of orotic acid (Table I).
The results presented above are interpreted to indicate that
a majority of initiated hepatocytes are not a resistant phenotype,
however, they have acquired a unique ability to express
resistance upon exposure to either the promoting regimen or
to agents such as 2-AAF and CCl4. Since 2-AAF and CCl4
are genotoxic, it would be interesting to determine whether
the second exposure has to be a genotoxic insult for the
initiated hepatocytes to express the resistant phenotype. The
fact that the surrounding non-initiated hepatocytes under identical conditions did not express resistance to either orotic acid
or 2-AAF suggests that the initiated cells are altered in such
a fashion that they have acquired a unique potential to express
resistance upon exposure to certain tumor promoters and/or
genotoxic agents. This unique property, i.e. the ability to
express resistance upon exposure to certain chemicals, seen as
early as in initiated cells, can be observed throughout the
neoplastic process as well. Development of multi-drug resistance during cancer chemotherapy is a classical example of
this phenotype. The expression of resistance does not seem to
be an adaptive response targeted specifically to the chemical
or the drug to which the preneoplastic or the neoplastic cell
is exposed. For example, in these studies resistance was
monitored to two different mitoinhibitory agents, orotic acid
and 2-AAF. These two agents require entirely different metabolic pathways for them to exert mitoinhibitory effects. Orotic
acid needs to be metabolically converted to uridine nucleotides
and the accumulation of uridine nucleotides is essential for it
to exert its mitoinhibitory effects (17,18). Conversely, 2-AAF
needs to be hydroxylated and esterified for it to exert its
mitoinhibitory effects (19). The fact that a similar pattern of
results was obtained irrespective of whether we used orotic
acid or 2-AAF as an index to monitor the resistant phenotype
suggests that the acquisition of resistance may reflect a
Fig. 3. Schematic representation implying a possible role of induction of
resistance in the carcinogenic process.
significant event in the pathogenesis of the carcinogenic
process. It will be of interest to identify the alteration(s)
relatable to resistance at the genetic level and to determine by
molecular approaches whether a resistant phenotype with a
potential to progress to cancer can be induced.
The results of the present study raise several questions. For
example, what is the significance of the two populations of
lesions, i.e. the foci that express resistance and those that
express resistance upon exposure to the tumor promoting
regimen, in terms of carcinogenesis? Is it that the lesions that
have not yet acquired resistance are promoter dependent for
their growth and progression and, by contrast, those that
express resistance have perhaps become promoter independent?
Another question pertains to the heterogeneity within the
population of the resistant lesions. Could it be that the degree
of resistance reflects the ability or the inability to remodel, the
least resistant ones being those with greater potential to
remodel? In other words, this heterogeneity may play an
important role in the different fates of these preneoplastic
lesions in the pathogenesis of the carcinogenic process. These
considerations will also be of significance in developing
chemopreventive and therapeutic strategies for liver cancer.
The eventual question, however, is whether acquisition of
resistance is a prerequisite for the initiated cell to progress
through the carcinogenic process. The studies carried out by
Farber and co-workers clearly point out that acquisition of
resistance is one mechanism for the initiated cell to progress
through the carcinogenic process (4,20). Indeed the RH model
of liver tumor promotion was developed on the premise
that the early putative preneoplastic hepatocyte is a resistant
phenotype (4). Furthermore, rats initiated with DENA alone
developed no hepatocellular carcinomas, while those similarly
initiated with DENA and subsequently exposed to the complete
RH protocol or to dietary 2-AAF alone had incidences of
70 and 45% hepatocellular carcinoma, respectively, within
8 months (21). Whether acquisition of resistance is the only
mechanism for cancer development is still an open question.
Nevertheless, it may be rationalized that acquisition of resistance in its broadest sense provides the preneoplastic and the
neoplastic hepatocytes with a growth and survival advantage
in an otherwise mitoinhibitory and cytotoxic environment
created by tumor promoters and carcinogens.
Based on the results of the present study, it may also be
speculated that certain tumor promoters, especially those tumor
promoters that are mitoinhibitors, may induce resistance in
initiated cells and selectively amplify those that acquired
resistance (Figure 3). More importantly, it is likely that there
may be agents which are not necessarily initiators or promoters
1643
A.Yusuf et al.
but still participate in the carcinogenic process by virtue of
their ability to induce resistance in initiated hepatocytes. It
would be interesting to identify this group of compounds and
determine their role in the carcinogenic process.
Acknowledgements
We wish to thank Dr E.Laconi for helpful discussions. This work was
supported in part by USPHS grant CA 37077 and funds from National Cancer
Institute, Canada. A.Y. was supported by the Canadian Liver Foundation.
References
1. Haddow,A. (1938) Cellular inhibition and origin of cancer. Acta Unio. Int.
Contra Cancrum, 3, 342–352.
2. Baldini,N. (1997) Multidrug resistance—a multiplex phenomenon [news;
comment]. Nature Med., 3, 378–380.
3. Endicott,J.A. and Ling,V. (1989) The biochemistry of P-glycoproteinmediated multidrug resistance. Annu. Rev. Biochem., 58, 137–171.
4. Solt,D.B. and Farber,E. (1976) New principle for the analysis of chemical
carcinogenesis. Nature, 263, 701–703.
5. Sarma,D.S.R., Rajalakshmi,S. and Rao,P.M. (1997) Neoplasia—a
pathological aberration or physiological adaptation. In Kisilevsky,R.,
Lebovitz,R.M., Lee,R.E. and Lieberman,M.W. (eds) The Molecular
Pathology of Disease. http://www.bcm.tmc.edu/pathology/dept/researchs/
lieberman/festschrift/fest.htm.
6. Farber,E., Parker,S. and Gruenstein,M. (1976) The resistance of putative
premalignant liver cell populations, hyperplastic nodules, to the acute
cytotoxic effects of some hepatocarcinogens. Cancer Res., 36, 3879–3887.
7. Feo,F., Canuto,R.A., Garcea,R., Brossa,O. and Caselli,G.C. (1978)
Phenobarbital stimulation of cytochrome P-450 and aminopyrine
N-demethylase in hyperplastic liver nodules during DL-ethionine
carcinogenesis. Cancer Lett., 5, 25–30.
8. Hanigan,M.H. and Pitot,H.C. (1985) Activities of benzphetamine
N-demethylase and aryl hydrocarbon hydroxylase in cells isolated from
gamma-glutamylpeptidase-positive foci and surrounding tissue. J. Natl
Cancer Inst., 75, 1107–1112.
9. Roomi,M.W., Ho,R.K., Sarma,D.S.R. and Farber,E. (1985) A common
biochemical pattern in preneoplastic hepatocyte nodules generated in four
different models in the rat. Cancer Res., 45, 564–571.
1644
10. Farber,E. and Rubin,H. (1991) Cellular adaptation in the origin and
development of cancer. Cancer Res., 51, 2751–2761.
11. Mansbach,J.M., Mills,J.J., Boyer,I.J., De Souza,A.T., Hankins,G.R. and
Jirtle,R.L. (1996) Phenobarbital selectively promotes initiated cells with
reduced TGF beta receptor levels. Carcinogenesis, 17, 171–174.
12. Zhang,X., Wang,T., Batist,G. and Tsao,M.S. (1994) Transforming growth
factor beta 1 promotes spontaneous transformation of cultured rat liver
epithelial cells. Cancer Res., 54, 6122–6128.
13. Factor,V.M., Kao,C.-Y., Santoni-Rugiu,E., Woitach,J.T., Jensen,M.R. and
Thorgeirsson,S.S. (1997) Constitutive expression of mature transforming
growth factor β1 in the liver accelerates hepatocarcinogenesis in transgenic
mice. Cancer Res., 57, 2089–2095.
14. Schulte-Hermann,R., Bursch,W., Grasl-Kraupp,B., Torok,L., Ellinger,A.
and Mullauer,L. (1995) Role of active cell death (apoptosis) in multi-stage
carcinogenesis. Toxicol. Lett., 82–83, 143–148.
15. Semple-Roberts,E., Hayes,M.A., Armstrong,D., Becker,R.A., Racz,W.J.
and Farber,E. (1987) Alternative methods of selecting rat hepatocellular
nodules resistant to 2-acetylaminofluorene. Int. J. Cancer, 40, 643–645.
16. Sheikh,A., Yusuf,A., Laconi,E., Rao,P.M., Rajalakshmi,S. and
Sarma,D.S.R. (1993) Effect of orotic acid on in vivo DNA synthesis in
hepatocytes of normal rat liver and in hepatic foci/nodules. Carcinogenesis,
14, 907–912.
17. Rao,P.M., Vasudevan,S., Laconi,E., Rajalakshmi,S. and Sarma,D.S.R.
(1986) Imbalances of pyrimidine and purine nucleotides and their
implications in liver carcinogenesis. In Scarpelli,D. and Migaki,G. (eds)
Nutritional Diseases: Research Directions. Alan R. Liss, New York, NY,
pp. 411–420.
18. Pichiri-Coni,G., Coni,P., Laconi,E., Schwarze,P.E., Seglen,P.O., Rao,P.M.,
Rajalakshmi,S. and Sarma,D.S.R. (1990) Studies on the mitoinhibitory
effect of orotic acid on hepatocytes in primary culture. Carcinogenesis,
11, 981–984.
19. Weisburger,J.H. and Williams,G.M. (1982) Metabolism of chemical
carcinogens. In Becker,F.E. (ed.) Cancer: A Comprehensive Treatise, 2nd
Edn. Plenum, New York, NY, Vol. 1, pp. 241–333.
20. Solt,D.B., Medline,A. and Farber,E. (1977) Rapid emergence of carcinogeninduced hyperplastic lesions in a new model for the sequential analysis of
liver carcinogenesis. Am. J. Pathol., 88, 595–618.
21. Solt,D.B., Cayama,E., Tsuda,H., Enomoto,K., Lee,G. and Farber,E. (1983)
Promotion of liver cancer development by brief exposure to dietary
2-acetylaminofluorene plus partial hepatectomy or carbon tetrachloride.
Cancer Res., 43, 188–191.
Received January 12, 1999; revised April 14, 1999; accepted April 27, 1999