Carcinogenesis vol.19 no.7 pp.1231–1237, 1998
Wild-type p53 expression in liver tissue and in enzyme-altered
foci: an in vivo investigation on diethylnitrosamine-treated rats
Pia Lennartsson1, Johan Högberg1,2 and Ulla Stenius1,3
1Institute
of Environmental Medicine, Karolinska Institutet, Box 210,
S-17177 Stockholm, and 2National Institute for Working Life, S-171
84 Solna, Sweden
3To
whom correspondence should be addressed
Email: [email protected]
Previous reports have documented an attenuated p53
response to DNA damage in hepatocytes isolated from
enzyme-altered foci (EAF). Here, we have studied this p53
response in vivo in rats with EAF. These animals received
repeated doses of diethylnitrosamine (DEN) for 6 weeks
and a challenging dose 24 h before death. Liver sections
were then analysed using an immunohistological procedure
for p53, or a double-staining procedure for p53 and
glutathione-S-transferase pi (GST-P). In control rats or
rats with EAF not given the challenging dose of DEN,
there was no p53 staining. In control rats, only given the
challenging dose of DEN, there was a centrilobular p53
nuclear staining that co-localized with TUNEL staining. In
an experiment involving four rats with EAF 389 K 39
hepatocytes/mm2 of non-EAF tissue stained positively for
p53, while the corresponding value for EAF tissue was
27.6 K 7.5. Thus, p53-positive cells were 14.6-fold more
frequent in non-EAF than in EAF tissue. In many EAF no
p53-positive cells were seen at all and 83% of the EAF
demonstrated <20% of the number of p53-positive cells
seen in non-EAF tissue. Very few EAF had as high a
proportion of p53-positive cells as did the average nonEAF tissue. EAF >0.06 mm2 had significantly fewer p53positive cells than smaller EAF. The ratio of p53 expression
in non-EAF tissue and large EAF was 32.6. In a control
experiment, four EAF-bearing rats were used as donors to
prepare primary cultures of hepatocytes. After 24 h of
exposure to DEN, many of the cultured cells became p53positive. Among GST-P-negative hepatocytes, 12.8% were
p53-positive, whereas only 0.25% of the GST-P-positive
hepatocytes were p53-positive. Literature data suggest that
the altered xenobiotic metabolism in EAF may give rise to
a 3–4-fold difference in DNA damage between non-EAF
and EAF tissues. It is concluded that GST-P-positive EAF
hepatocytes have an attenuated p53 response to DNA
damage. This attenuated response may facilitate clonal
expansion of EAF under stress induced by DNA-damaging
chemicals.
Introduction
p53 is a transcription factor that accumulates in the nucleus
in response to DNA damage induced by, for example, chemical
carcinogens. This protein activates the transcription of many
genes, including p21WAF1 which encodes a cyclin-dependent
*Abbreviations: DEN, diethylnitrosamine; EAF, enzyme-altered foci; GSTP, glutathione-S-transferase pi.
© Oxford University Press
kinase inhibitor. p53-induced p21WAF1 may prevent cells with
damaged DNA from replicating (1,2). p53 may also trigger
apoptosis, presumably by activating genes related to ROS
production and subsequent GSH depletion (3). The apparent
importance of p53 as a tumour suppressor is indicated by the
findings that ~50% of all human tumours carry a mutated p53
gene and that certain cancer-prone families demonstrate a
mutated p53 allele (4).
Treatment of rats with a carcinogen results in the development of numerous liver lesions called enzyme-altered foci
(EAF*) (5). There are many indications that these early lesions
are adaptive in nature (6). One of the most conspicuous of
these indications are alterations in xenobiotic metabolism,
which render EAF hepatocytes resistant to toxicants. One
xenobiotic metabolizing enzyme, glutathione-S-transferase pi
(GST-P), which can inactivate many reactive xenobiotic metabolites, is not detectable in normal rat liver tissue, but is
expressed at a high level in most EAF. Thus, this enzyme is
commonly used as a marker for EAF (5). A few EAF
may eventually develop into hepatocellular carcinomas and a
surprising finding is that p53 mutations are rarely found in
such tumours in the rat (7,8).
Previously, we have studied p53 expression in hepatocytes
isolated from EAF-bearing rats. We found that EAF hepatocytes
in primary culture did not respond to DNA-damaging agents,
such as diethylnitrosamine (DEN), with an increased expression
of p53 (9–11). Non-EAF hepatocytes in the same culture
responded readily with induction of p53 within 24 h and none
of these cells replicated its DNA. In agreement with the
hypothesis that an altered xenobiotic metabolism confers
resistance to xenobiotics on EAF hepatocytes, we found
indications of DNA damage four times more often in non-EAF
hepatocytes than in EAF hepatocytes. However, a significant
number of EAF hepatocytes appeared to have damaged DNA
and some of these cells replicated their DNA. These data were
in agreement with in vivo data on ‘replicative foci’ induced
by 2-AAF in rat liver, and with the finding that cells in these
foci containing DNA adducts could replicate (12).
Based on our findings, we proposed that p53 plays an
important role in adaptive responses induced by genotoxic
carcinogens in rat liver. We suggest that a lack of induction
of p53 in EAF hepatocytes in response to DNA damage may
facilitate the clonal expansion of these cells. This suggestion
is based on the assumption that p53 can induce a permanent
stop in the cell cycle in many non-EAF hepatocytes (cf. ref.
13) and/or trigger apoptosis, whereas EAF hepatocytes may
replicate even when exposed to high levels of genotoxic
compounds. If such EAF cells can be eliminated by apoptotic
mechanisms after the exposure and if the exposure to the
toxicant is relatively short, the risk for malignancy should
remain low.
In this investigation we have challenged EAF-bearing rats
with a single dose of DEN 24 h before death, and examined
the p53 response in non-EAF and EAF tissue using
1231
P.Lennartsson, J.Högberg and U.Stenius
immunohistological procedures. We have also examined the
responses of non-EAF and EAF hepatocytes to DEN treatment
obtained in vitro.
Materials and methods
Treatment of donor animals
Female Sprague–Dawley rats were injected i.p. with DEN (0.30 mmol/kg
body wt) 24 h after birth. At 3 weeks of age, the animals were weaned and
thereafter injected i.p. with the same dose of DEN once each week. After 6–
10 doses of DEN, the rats were used as donor animals or for immunohistological
studies. Most rats used for immunohistology also received an injection of DEN
(0.60 mmol/kg body wt) 24 h prior to death. In control immunohistological
experiments untreated rats were challenged by an injection of DEN
(0.60 mmol/kg body wt) 24 h prior to death.
Immunohistological staining
Rats were subjected to surgery under light ether anaesthesia. Their livers
were perfused with 3.7% buffered formaldehyde at 37°C for 1.5 h and
subsequently placed in formalin for 24 h. After fixation the sections were
processed for paraffin sectioning. The slides (in citrate buffer) were treated
with microwaves five times for 5 min each time prior to staining. Sections
were single- or double-stained with polyclonal antibodies directed towards
GST-P and monoclonal antibodies (Pab 240) against p53 (Oncogene Science)
as follows. After incubation with anti-p53 (with a dilution of 1/100 under
standard conditions) and anti-GST-P overnight, the slides were exposed to
anti-rabbit secondary antibodies (dilution 1/100) for 1.5 h. GST-P staining
was visualized using an alkaline phosphatase-conjugated secondary antibody
and New Fuchsin. p53 was visualized using the EnVision1TM peroxidase kit
(Dako, Denmark) with 3-diaminobenzidine tetrahydrochloride as substrate.
Some sections were stained with terminal deoxynucleotidyl transferase using
TdT FragELTM apoptosis kit (Calbiochem).
Primary cultures and immunocytology
Hepatocytes were isolated using the collagenase perfusion technique and
cultured as described previously (9,14). In brief, the cells were cultured for
1.5 h in complete medium containing serum. This medium was then replaced
by serum free medium containing DEN (0.3 mM) for 24 h. Fixed cells were
double-stained with polyclonal rabbit anti-GST-P antibodies and monoclonal
antibodies (Pab 240) towards p53 (Oncogene Science; at a dilution of 1/100).
Secondary antibodies were then used, and finally, peroxidase-rabbit antiperoxidase complexes and alkaline phosphatase-mouse anti-alkaline phosphatase complexes. Peroxidase activity was visualized with 3-diaminobenzidine tetrahydrochloride and alkaline phosphatase with New Fuchsin (Dako,
Sweden) as substrates. GST-P was used as a marker for distinguishing between
EAF (positive) and non-EAF (negative) hepatocytes. The fraction of cells that
expressed the GST-P marker varied from 5 to 17% in the different animals.
The percentage marker-positive cells was determined by counting at least 500
cells with typical hepatocyte morphology from randomly-selected fields on
each plate. Immunoprecipitation and Western blot analysis was performed
essentially as described in (15). p53 was detected with monoclonal antibody
Pab 240 (Oncogene Science) and rabbit p53 antibody CM-1 (Novocastra
Laboratories, Newcastle).
Evaluation of the in situ data presented in Tables I–IV
Stained liver sections (a total of at least 46 mm2 from each rat) were analysed
microscopically. The area, volume and volume fraction of EAF were calculated
using software based on the method of Pugh et al. (16). All GST-P-positive
EAF .0.003 mm2 were analysed and the total EAF area analysed per rat
varied between 1.96 and 3.03 mm2. Twenty randomly selected GST-P negative
fields (non-EAF tissue), each spanning at least one periportal and one
perivenous zone, were also analysed in each section. In each rat at least
3.43 mm2 non-EAF tissue was analysed. p53 expression was assessed by
counting the number of p53-positive cells in each EAF and non-EAF fields.
Foci were also divided into three groups according to size and the volume
fraction of each group was determined.
Fig. 1. Immunochemical staining for p53 in liver slices from DEN-treated
rats. (A) Many p53-positive hepatocytes (brown nuclei) are seen in a
perivenous area (left), while no immunoreactivity is detected in a periportal
area (right). (B) The region close to a central vein at higher magnification.
Many p53-positive hepatocytes (brown nuclei) are seen, some with two
nuclei. Several hepatocytes which do not demonstrate staining are also
present.
Fig. 2. TUNEL staining of liver slices from DEN-treated rats. TUNELpositive hepatocytes (brown nuclei or nuclear fragments) are seen in the
perivenous area.
Results
In control experiments involving untreated rats there was no
staining for p53. In rats treated with DEN 24 h before death
several p53-positive hepatocytes were seen. The staining was
localized to nuclei (Figure 1A and B) in perivenous hepatocytes
(Figure 1A). Pab 240 was used in most stainings. CM-1 was
also tested and gave similar results. Serial sections without
1232
p53 antibodies were entirely negative. In sections from livers
fixed with standard procedures (i.e. not perfused with formalin)
the p53 staining was weaker or absent. Figure 2 shows that
there was a co-localization in the lobuli between p53 positive
cells and cells stained by TdT activity. The number of
p53 expression in liver tissue and EAF
Fig. 3. A p53-negative EAF from a DEN-treated rat. The liver section was
double-stained for GST-P and p53. GST-P-positive EAF (red cytoplasm)
with surrounding non-EAF tissue are seen. In the non-EAF tissue, several
p53-positive hepatocytes (brown nuclei) can be seen. In EAF tissue no p53positive nuclei can be seen.
TUNEL-positive cells were few as compared with the number
of p53-positive cells.
In control experiments involving EAF-bearing animals
which did not receive a challenging dose of DEN, there was
no staining for p53. Both non-EAF tissue and EAF were
totally negative for p53. A recent study demonstrated that
single exposure to DEN-induced DNA adducts 6 h later, with
maximal levels of adducts being obtained at 24 h (17). We
administered the challenging dose 24 h prior to sacrifice.
GST-P was used as a marker for EAF and liver sections
were double-stained for GST-P and p53. In the double-stained
sections, GST-P-positive areas were clearly seen, as were p53
positive nuclei. p53-positive nuclei were observed in GST-Pnegative, as well as in GST-P-positive areas. On a cellular
level p53 staining was distributed in the same fashion as that
seen in EAF-free rats, i.e. such staining was only present in
nuclei of hepatocytes. In serial sections subjected to staining
for p53 only or to double-staining, there were no apparent
differences in the number of p53-positive cells. The number
of p53-positive cells varied with the size of the challenging
dose of DEN. Necrogenic doses induced a very high frequency
of p53 staining.
In non-EAF tissue in the livers of EAF-bearing rats the
distribution of p53-positive cells in the lobules was uneven
and similar to that observed in EAF-free rats. p53-positive
hepatocytes were present primarily outside GST-P-positive
areas. Figure 3 shows an EAF in which no p53-positive nuclei
can be seen, as can the relatively frequent occurence of p53positive nuclei in non-EAF tissue. There were also EAF with
p53-positive cells. Figure 4 depicts a GST-P-positive EAF
containing several p53-positive nuclei.
Even doses which induced signs of necrosis in large nonEAF areas and a p53 response in the majority of non-EAF
hepatocytes, induced p53 in only a few EAF hepatocytes.
Figure 5 shows part of an EAF from a rat challenged with a
necrogenic dose (1.8 mmol/kg) of DEN. In this case the section
was stained for p53 only and a more concentrated solution of
the primary antibody against p53 was used, which resulted in
Fig. 4. A p53-positive EAF from a DEN-treated rat. The liver section was
double-stained for GST-P and p53. (A) GST-P-positive EAF (red
cytoplasm), close to a central vein and surrounding non-EAF tissue are
seen. In the non-EAF tissue several p53-positive hepatocytes (brown nuclei)
can be seen, mainly in the vicinity of the vein. In EAF tissue several p53positive nuclei can also be seen, again close to the vein. (B) The same EAF
at higher magnification.
faint staining of non-EAF cytoplasm. The EAF can be seen as
virtually unstained islands in heavily stained non-EAF tissue.
In Figures 3 and 4 and Tables I–IV we present data from
an experiment involving four rats. Each rat received 6 weekly
doses of DEN and was then challenged with a single DEN
dose of 0.60 mmol/kg body weight 24 h prior to death. Only
GST-P-positive EAF .0.003 mm2 were examined. There were
no macroscopic nodules or tumours present in the livers of
these rats. Microscopic examination revealed, on the other
hand, many GST-P-positive EAF. No EAF was .0.3 mm2. In
Table I certain parameters for the GST-P-positive EAF in these
rats are presented. The numbers and the volume fractions were
as expected from previous investigations.
Table II presents quantitative data about p53 expression. It
can be seen that the number of positive cells per mm2 nonEAF tissue varied relatively little from rat to rat. The lowest
1233
P.Lennartsson, J.Högberg and U.Stenius
value was 333 and the highest 422 positive cells/mm2. The
corresponding figures for all EAF varied more widely, i.e.
from 18.5 to 36.3 positive cells/mm2. From rat to rat, there
was a certain co-variation of the frequency of p53-positive
cells in these two types of tissues. Thus, the ratios of the nonEAF to EAF values were slightly more stable, varying between
11.2 and 18.0. The mean of these four ratios was 14.6. Thus,
Fig. 5. A p53-negative EAF from a DEN-treated rat. This section was
immunostained only for p53. Non-EAF tissue is seen to contain many p53positive hepatocytes (dark brown nucleus and light brown cytoplasm).
Necrotic cells surround a central vein (left) and part of an EAF with a few
light brown nuclei is seen (right). Dilution of the primary p53 antibody was
1/50.
Table I. GST-P-positive EAF in DEN-treated rats
Rata
No. of
EAFb
No. of
EAF/cm2
Mean EAF
area (mm2)
Volume
fraction (%)
1
2
3
4
Σ
Mean 6 SD
52
59
73
68
252
–
113
87
137
80
–
104 6 26
0.039
0.033
0.042
0.032
–
0.036 6 0.005
4.41
2.91
5.67
2.60
–
3.9 6 1.4
aLiver
sections from four rats were analysed. Each rat received six doses of
DEN and was then challenged with a dose of DEN 24 h prior to death (see
Materials and methods).
bNumber of GST-P-positive EAF .0.003 mm2.
assuming that EAF tissue contains on average as many
hepatocytes/unit area as non-EAF tissue, p53-positive hepatocytes were 14.6 times more frequent in non-EAF than in
EAF tissue.
Table II also documents the number of EAF which can be
regarded as ‘p53-negative’. All EAF with ,75 p53-positive
cells/mm2 (which is ~20% of the average number in non-EAF
tissue) were assigned to this category (Figure 3). In these four
rats 83.3% of the EAF were p53-negative. Almost all of the
rest also demonstrated a lower proportion of p53-positive cells
than that seen in non-EAF tissue (Figure 4). Very few EAF
per rat liver exhibited a frequency of p53-positive cells as
high as that of the average non-EAF tissue (390 p53-positive
cells/mm2). Among EAF with a higher frequency, all except
one were ,0.06 mm2. This large EAF (found in rat no. 2)
showed signs of regression, with weak GST-P-staining and
many cells demonstrating signs of degenerative alterations.
There was no consistent correlation between the volume
fraction of EAF (Table I) and the ratio of p53 expression
presented in Table II. However, the highest ratio was found
for the rat with the largest volume fraction, while the lowest
ratio was associated with the rat with the smallest volume
fraction. This suggests a relationship between p53 expression
and EAF size, so for further analysis we subdivided the EAF
into categories by size.
In Table III, p53 expression in three groups of EAF, called
here small, medium and large EAF, are given. In three of the
four rats, large EAF (0.06–0.294 mm2) had fewer p53-positive
hepatocytes than did small and medium EAF. The ratios of
p53-positive cells in non-EAF tissue, on the one hand, and in
these three groups of EAF, on the other, are also given. For
small and medium EAF, the ratios were similar and ranged
between 6.9 and 15.9. This variation reflected differences
between rats, rather than differences between the small and
medium EAF within a single rat liver. For large EAF the ratio
varied between 15.3 and 51.1. In three of the rats this ratio
was markedly higher than for the other two groups. If a single
large EAF (in rat no. 2; as mentioned above), containing a
very high number of p53-positive cells, was excluded, the
average number of p53-positive cells in large EAF was altered
from 25.8 to 8.1 cells/mm2 (Table III).
Table IV summarizes the data presented in Table III. The
mean values for the four rats with respect to the number of
p53-positive hepatocytes/mm2 EAF tissue are given. It can be
seen that the number of p53-positive cells was similar in small
and medium EAF, but significantly lower in the large EAF.
The mean values for the four rats with respect to the ratios of
Table II. p53 expression in non-EAF and EAF liver tissues from DEN-treated rats
Rat
p531/mm2
non-EAF tissuea
p531/mm2
EAF tissueb
Ratioc
p53-negative
EAFd (%)
p53-positive
EAFe (%)
1
2
3
4
Mean 6 SD
421.7
394.7
332.5
408.3
389.3 6 39.4
29.9
26.0
18.5
36.3
27.6 6 7.5
14.1
15.2
18.0
11.2
14.6 6 2.8
84.6
81.4
87.9
79.4
83.3 6 3.6
15.4
18.6
12.4
20.6
16.7 6 3.6
aNumber
of p53-positive cells/mm2 non-EAF tissue area (see Materials and methods).
of p53-positive cells/mm2 total EAF tissue area (see Materials and methods).
p53 expression 5 number of p53-positive cells in non-EAF tissue divided by number of p53-positive cells in EAF tissue.
dEAF with ,75 p53-positive cells/mm2.
eEAF with .75 p53-positive cells/mm2.
bNumber
cRatio of
1234
p53 expression in liver tissue and EAF
Table III. p53 expression in EAF of different sizes
Rat
Size (mm2)a
No. of
EAF
p531/mm2
EAF tissueb
p531/mm2
non-EAF tissuec
Ratiod
1
0.003–0.02
.0.02–0.06
.0.06–0.294
0.003–0.02
.0.02–0.06
.0.06–0.294
0.003–0.02
.0.02–0.06
.0.06–0.294
0.003–0.02
.0.02–0.06
.0.06–0.294
27
16
9
28
24
7
33
29
11
28
31
9
45.8
54.4
15.1
29.4
24.8
25.8
29.5
32.0
6.5
59.1
47.0
11.3
421.7
421.7
421.7
394.7
394.7
394.7
332.5
332.5
332.5
408.3
408.3
408.3
9.2
7.7
27.8
13.4
15.9
15.3
11.3
10.4
51.1
6.9
8.7
36.1
2
3
4
aEAF
were divided into three categories with the indicated sizes. In the text these are called small, medium and large EAF.
of p53-positive cells/mm2 total EAF tissue area.
of p53-positive cells/mm2 in non-EAF tissue area (see Table II).
dRatio of p53 expression in non-EAF and EAF tissues.
bNumber
cNumber
Table IV. Summary of p53 expression in EAF of different sizes
Size (mm2)
No. of EAFa
p531/mm2
EAF tissueb
Ratioc
Volume
fraction (%)d
% of total
foci volumee
0.003–0.02
.0.02–0.06
.0.06–0.294
116
100
36
41.0 6 14.4
39.6 6 13.6
14.7 6 8.2*
10.2 6 2.8
10.7 6 3.6
32.6 6 15.0
0.59
1.35
1.72
16
37
47
aData for
bNumber
four rats.
of p53-positive cells/mm2 total EAF tissue area. Mean 6 SD of the data for four rats presented in Table III.
cRatio of p53 expression in non-EAF and EAF tissues. Mean 6 SD of the ratios presented in Table III.
dMean volume fraction of GST-P-positive EAF for four rats.
ePercentage of total EAF volume.
*Significantly different compared with values for small and medium EAF (Students t-test, P , 0.05).
p53 expression in non-EAF and EAF tissue (as shown in
Table II) are also given. Again, there was a noteworthy
similarity between the two smaller groups, whereas the larger
size category demonstrated a higher ratio. The contribution of
the different groups to the total volume fraction is also shown.
The large EAF accounted for about half (47%) of the total
volume fraction.
The material was also divided into other size categories.
For example, the large EAF were subdivided into two groups,
but this resulted in data from fewer areas, with greater
variability and without obvious trends. When the small and
medium EAF were subdivided into other categories, results
similar to those found for the whole group were obtained. The
use of a higher upper size limit for inclusion in the group of
medium EAF tended to increase the ratio for this group.
In order to compare directly these in vivo data with in vitro
findings we isolated hepatocytes from EAF-bearing rats and
studied p53 expression in primary cultures. These experiments
were performed essentially as described previously (9,10). The
challenging dose of DEN was added to the culture medium
and p53 staining assessed 24 h after this addition. At this
time-point the highest number of positive cells was obtained.
In control experiments immunoprecipitated protein from control and DEN treated primary cultures were prepared for
Western blot analysis. A single band was detected by the p53
CM-1 antibody, as shown in Figure 6. Other antibodies,
including Pab 240, were also used and gave similar results
(not shown). Parallel cultures with cells from the same rat
Fig. 6. Western blot analysis of immunoprecipitated p53 from primary
cultures of hepatocytes. Cells were isolated from a DEN-treated rat with
EAF. Primary cultures with or without DEN were incubated for 24 h. CM-1
antibody was used to detect p53.
were immunostained with Pab 240. The fraction of p53 positive
cells were 5.5% in the control culture and 18.2% in the
DEN culture.
Table V documents data obtained from four donor animals.
Rats used for these in vitro experiments received a larger
number of weekly injections of DEN and, thus, had higher
volume fraction of EAF in their livers than did the rats used
in the in vivo experiment. The fact that the mean percentage
1235
P.Lennartsson, J.Högberg and U.Stenius
Table V. p53 expression by EAF and non-EAF hepatocytes upon exposure
to DEN in vitro
Donor animal
treatmenta
GST-P (%)b
p53/GST-P–
cells (%)c
p53/GST-P1
cells (%)c
7
9
10
10
Mean 6 SD
5.7
5.8
6.0
17.0
8.6 6 5.6
15.0
12.0
15.0
9.3
12.8 6 2.7
0.5
0.1
0.2
0.2
0.25 6 0.17
Ratiod
51.2
aFour
animals were treated with DEN. The total number of DEN injections,
administered once weekly, is shown (see also Materials and methods).
bPercentage of all cells in the culture which were GST-P-positive.
cPercentage of p53-positive cells among GST-P-negative and GST-P-positive
cells, respectively.
dThe ratio of the means of p53 expression in non-EAF and EAF cells.
of GST-P-positive cells was higher in the in vitro cultures than
the mean volume fraction of EAF in the livers of the four rats
used in the in vivo experiment (3.9 6 1.4, total volume
fraction, see Table IV) supports this assumption. The number
of GST-P-negative hepatocytes that became positive for p53
staining in vitro is also given. The mean for the cultures from
the four donor animals was 12.8%. On average, 0.25% of the
GST-P-positive cells were positive for p53. The ratio between
these two means is 51.2.
Discussion
These data demonstrate that expression of wild-type p53 in
liver can be studied using immunohistological techniques.
Although previous data from non-transgenic mice indicate the
reverse (18; cf. also ref. 19), gamma radiation was used in
that study, and more recent studies suggest that gamma
radiation is not an efficient inducer of p53 expression detectable
by immunohistology in rodent hepatocytes (20). We have also
employed non-standard procedures for fixation and amplifications, not previously used in the present context. The present
study also demonstrate that double-staining for p53 and GSTP can be used efficiently, which markedly facilitates a detailed
analysis of the distribution of p53-positive hepatocytes. Factors
which indicate that the p53 staining is specific include the
requirement for a challenging dose of DEN to obtain p53positive cells, the time-point of most intense staining, the
necessity for amplification and rapid fixation, the localization
of the staining in the lobule and its co-localization to TUNELpositive cells, selective localization in the hepatocytes (which
metabolize DEN to a reactive intermediate with DNA-binding
capacity), the subcellular localization to the nucleus and the
results of the Western blot analysis.
In tumours, positive staining for p53 is usually taken as an
indication of the presence of mutations, since these often result
in a more stable protein and since mutated cells can be selected
for during tumour growth (4). Nevertheless, we regard the
positive staining discussed here as indicative of an accumulation of wild-type protein (a possible exception being the large
p53-positive EAF found in rat no. 2). The major arguments
for this conclusion are that we observed the positive cells
primarily in non-EAF tissue and only after a challenge with
DEN, and that the response in non-EAF tissue was similar to
that seen in rats free of EAF. Additional considerations include
the fact that mutations are rare in rat liver tumours (7).
We have previously estimated that DNA damage caused by
1236
DEN in vitro is four times more common in GST-P-negative
rat hepatocytes than in GST-P-positive hepatocytes (9). This
value is supported by in vivo data on DNA adducts in
2-acetylaminofluorene-treated rats. Using 32P-post-labelling, it
was found that such adducts were 3.1 times more common in
‘surrounding’ liver tissue than in altered liver tissue (21).
Together, these data suggest that factors inside the cell affecting
kinetics [i.e. altered metabolism of xenobiotics and the possible
expression of the MDR gene (22)], are a significant protective
factor conferring resistance to the formation of DNA adducts
in EAF cells; whereas such factors as possible alterations in
blood flow may not be involved in resistance.
Is the increase in p53 levels in response to a given amount
of DNA damage different in EAF and non-EAF hepatocytes?
The data presented here strongly suggest that this is, indeed,
the case. On an average, non-EAF tissue exhibited an ~15fold higher frequency of p53-positive hepatocytes than did
EAF (Table II). If alterations in drug metabolism can only
account for a 3–4-fold difference, the remainder of the change
must involve other mechanisms. It thus seems reasonable to
propose that there are alterations in the signal transduction
pathway connecting DNA damage to subsequent p53 accumulation in most EAF cells. The finding that the frequency of
p53 expression in large EAF was .30-fold less than that in
non-EAF tissue supports this idea, as does the finding that
even necrogenic doses of DEN do not induce p53 in many
EAF (Figure 5).
In the in vitro cultures the ratio between the level of p53
response in GST-P-negative and GST-P-positive hepatocytes
upon exposure to DEN was ~50. Even though this value is
higher than the ratio calculated for all EAF in situ, we consider
these two ratios to be in relatively good agreement. The in vitro
and in vivo protocols employed here differed in many ways.
The apparent discrepancies observed may easily be explained
by indications that the rats used in the in vitro experiments
had larger EAF or by technical differences with respect to the
challenging dose of DEN employed, and effects arising from
the isolation and culturing procedures themselves.
The finding that most larger EAF had a lower level of p53positive cells, than did smaller EAF cannot be explained by,
for example, circulatory differences (cf. ref. 23). This is
indicated by the data on DNA damage discussed above, as
well as by the data suggesting that there were fewer p53positive EAF cells in vitro than in vivo. Instead, the low level
of p53 expression in large EAF supports the hypothesis
that there are alterations in p53 function that confer growth
advantages. The data suggest that hepatocytes in EAF with an
attenuated p53 response to DEN, in particular hepatocytes
located in EAF .0.06 mm2, may replicate and expand their
clones, while other hepatocytes are prevented from replicating
by DEN treatment. This reasoning is also discussed by Gijssel
et al. (15).
It may be speculated that when DEN treatment is terminated,
many of these cells are effectively eliminated by apoptosis. If
the DEN treatment is continued, tumours may develop within
a few months. In light of the present data, it seems quite
possible that these tumours derive from expanding clones of
EAF cells with an attenuated p53 response and inadequate
protection of their genome.
This scenario is supported by studies on the regulation of
hepatocyte growth in the absence of p53. In p53 null mice,
hepatocytes are more prone to replicate in response to stress,
e.g. liver necrosis. At the same time, hepatocytes isolated from
p53 expression in liver tissue and EAF
these mice demonstrate an increased ‘basal’ rate of cell
proliferation in primary culture (24). Also of relevance is the
fact that mutated p53 alleles are rarely found in hepatocarcinomas in rodents (7).
It is possible that the findings presented here are not typical
for all types of EAF. DEN may alkylate DNA, as well as
inducing 8-hydroxyguanine formation, and DEN adducts
remain in the liver for a long period of time (17); such
characteristics might profoundly affect the results obtained.
Preliminary results in our laboratory suggest that EAF in
phenobarbital-treated rats can respond to a challenging dose
of DEN by inducing p53. More data of this kind may provide
additional information about how p53 expression is altered in
DEN-induced EAF. Studies on possible changes in patterns of
DNA methylation might also provide useful information. It
has been shown that DNA methylation can be influenced by
8-hydroxyguanine adducts (25). In addition, there are stagerelated fluctuations in the methylation pattern in the p53 gene
during tumour development in rat liver (26). Furthermore,
WAF expression in DEN-induced EAF seems to be low (27)
and it has recently been proposed that p21WAF might act as a
negative regulator of DNA-methyltransferase activity in early
stages of the cell cycle (28).
Acknowledgement
This work was financially supported by the National Institute
for Working Life.
References
1. Kastan,M.B., Onyekwere,O., Sidransky,D., Vogelstein,B. and Craig., R.W.
(1991) Participation of p53 protein in the cellular response to DNA
damage. Cancer Res., 51, 6304–6311.
2. El-Deiry,W.S., Harper,J.W., O’Connor,P.M., et al. (1994) WAF1/CIP1 is
induced in p53-mediated G1 arrest and apoptosis. Cancer Res., 54,
1169–1174.
3. Polyak,K., Xia,Y., Zweier,J.L., Kinzler,K.W. and Vogelstein,B. (1997) A
model for p53-induced apoptosis. Nature, 389, 300–305.
4. Harris,C.C. (1996) p53 tumour suppressor gene: at the crossroads of
molecular carcinogenesis, molecular epidemiology, and cancer risk
assessment. Env. Health Perspect., 104, 435–439.
5. Pitot,H.C. (1990) Altered hepatic foci: their role in murine
hepatocarcinogenesis. Ann. Rev. Pharmacol. Toxicol., 30, 465–500.
6. Farber,E. (1992) Clonal adaptation during carcinogenesis. Biochem.
Pharmacol., 39, 1837–1846.
7. Barbin,A., Froment,O., Boivin,S., Marion,M.-J., Belpoggi,F., Maltoni,C.
and Montesano,R. (1997) p53 gene mutation pattern in rat liver tumours
induced by vinyl chloride. Cancer Res., 57, 1695–1698.
8. Masui,T., Nakanishi,H., Inada,K., Imai,T., Mizoguchi,Y., Yada,H.,
Futakuchi,M., Shirai,T. and Tatematsu,M. (1997) Highly metastatic
hepatocellular carcinomas induced in male F344 rats treated with
N-nitrosomorpholine in combination with other hepatocarcinogens show
a high incidence of p53 gene mutations along with altered m-RNA
expression of tumour-related genes. Cancer Lett., 112, 33–45.
9. Stenius,U. and Högberg,J. (1995) GST-P-positive hepatocytes isolated
from rats bearing enzyme-altered foci show no signs of p53 protein
induction and replicate even when their DNA contains strand breaks.
Carcinogenesis, 16, 1683–1686.
10. Stenius,U., Martens,U. and Högberg,J. (1996) The use of isolated
hepatocytes from rats with enzyme-altered foci: studies on p53 expression.
ATLA, 24, 393–397.
11. Stenius,U., Ståhl,A. and Högberg,J. (1997) In vitro studies on nongenotoxic carcinogens: resistance to DNA synthesis inhibition in GST-Ppositive hepatocytes isolated from enzyme altered foci bearing rats. Toxicol.
in Vitro (in press).
12. Huitfeldt,H.S., Brandtzaeg,P. and Poirier,M.C. (1990) Reduced DNA adduct
formation in replicating liver cells during continuous feeding of a chemical
carcinogen. Proc. Natl Acad. Sci., 87, 5955–5958.
13. Linke,S.P., Clarkin,K.C. and Wahl,G.M. (1997) p53 mediates permanent
arrest over multiple cell cycles in response to γ-irradiation. Cancer Res.,
57, 1171–1179.
14. Stenius,U., Warholm,M., Martens,U. and Högberg,J. (1994) Isolation of
glutathione S-transferase P-positive hepatocytes from carcinogen treated
rats by use of ethacrynic acid as selecting agent. Carcinogenesis, 15,
1561–1566.
15. Gijssel,H.E., Maasen,C.B.M., Mulder,G.J. and Meerman,J.H.N. (1997) p53
protein expression by hepatocarcinogens in rat liver and its role in
mitoinhibition of normal hepatocytes as a mechanism of hepatic tumour
promotion. Carcinogenesis, 18, 1027–1033.
16. Pugh,T.D., Kim,J.H., Koen,H., Nychka,D., Chover,J., Wahba,G., He,Y. and
Goldfarb,S. (1983) Reliable stereological method for estimating the number
of microscopic hepatocellular foci from their transections. Cancer Res.
43, 1261–1268.
17. Nakae,D., Kobayashi,Y., Akai,H., Andoh,N., Satoh,H., Ohashi,K.,
Tsutsumi,M. and Konishi,Y. (1997) Involvement of 8-hydroxyguanine
formation in the initiation of rat liver carcinogenesis by low dose levels
of N-nitrosodiethylamine. Cancer Res., 57, 1281–1287.
18. MacCallum,D.E., Hupp,T.R., Midgley,C.A., et al. (1996) The p53 response
to ionising radiation in adult and developing murine tissues. Oncogene,
13, 2575–2587.
19. McWhir,J., Selfridge,J., Harrison,D.J., Squires,S. and Melton,D.W. (1993)
Mice with DNA repair gene (ERCC-1) deficiency have elevated levels of
p53, liver nuclear abnormalities and die before weaning. Nature Genet.,
5, 217–224.
20. Bellamy,C.O.C., Prost,S., Wyllie,A.H. and Harrison,D.J. (1997) UV but
not gamma-irradiation induces specific transcriptional activity of p53 in
primary hepatocytes. J. Pathol., 183, 177–181.
21. Gupta,R.C., Earley,K. and Becker,F.F. (1988) Analysis of DNA adducts in
putative premalignant hepatic nodules and non target tissues of rats during
2-acetylaminofluorene carcinogenesis. Cancer Res., 48, 5270–5274.
22. Thorgeirsson,S.S., Huber,B.E., Sorrell,S., Fojo,A., Pastan,I. and
Gottesman,M.M. (1987) Expression of the multidrug-resistant gene in
hepatocarcinogenesis and regenerating rat liver. Science, 236, 1120–1122.
23. Conway,J.G., Popp,J.A. and Thurman,R.G. (1985) Microcirculation of
hepatic nodules from diethylnitrosamine-treated rats. Cancer Res., 45,
3620–3625.
24. Bellamy,C.O.C., Clarke,A.R., Wyllie,A.H. and Harrison,D.J. (1997) p53
deficiency in liver reduces local control of survival and proliferation, but
does not affect apoptosis after DNA damage. FASEB J., 11, 591–599.
25. Weitzman,S.A., Turk,P.W., Milkowski,D.H. and Kozlowski,K. (1994) Free
radical adducts induce alterations in DNA cytosine methylation. Proc.
Natl Acad. Sci., 91, 1261–1264.
26. Pogribny,I.P., Miller,B.J. and James,S.J. (1997) Alterations in hepatic p53
gene methylation patterns during tumour progression with folate/methyl
deficiency in the rat. Cancer Lett., 115, 31–38.
27. Martens,U., Lennartsson,P., Högberg,J. and Stenius,U. (1996) Low
expression of the WAF1/CIP1 gene product, p21, in enzyme altered foci
induced in rat liver by diethylnitrosamine or phenobarbital. Cancer Lett.,
104, 21–26.
28. Chuang,L.S.-H., Ian,H.-I., Koh,T.-W., Ng,H.-H., Xu,G. and Li,B.F.L.
(1997) Human DNA-(cytosine-5) methyltransferase-PCNA complex as a
target for p21WAF1. Science, 277, 1996–2000.
Received on November 24, 1997; revised on March 2, 1998; accepted on
March 25, 1998
1237