Download Differential Expression of a-Fetoprotein and y

[CANCERRESEARCH39, 3495-3501, September19791
0008-5472/79/0039-0000$02.00
Differential Expression of a-Fetoprotein and y-Glutamyltranspeptidase
in Chemical and Spontaneous Hepatocarcinogenesis
Hannu Jalanko1 and Erkki Ruoslahti2
Division of Immunology, City of Hope National Medical Center, Duarte, California 91010
ABSTRACT
The expression of two markers of fetal liver, a-fetoprotein
(AFP) and -y-glutamyltranspeptidase (GGT), was studied in
chemical and spontaneous hepatocarcinogenesis in mice. Se
rum AFP concentration increased within 3 weeks from the
commencement of feeding of o-aminoazotoluene. This early
elevation subsided about 3 months after the beginning of the
administration of the carcinogen. A new, sustained elevation of
the serum AFP level followed at 5 to 6 months accompanied by
the appearance of liver tumors. In immunofluorescence, some
small oval cells and scattered adult-type hepatocytes contained
AFP during the early stage of chemical carcinogenesis. During
the later phase, AFP was detected in a few of the nodular
areas, in solitary hepatocytes, and in groups of carcinoma
cells.
GGT activity in the liver increased within 1 week after the
carcinogen regimen was started, preceding the early increase
of AFP production. At the final stage, the chemically induced
hepatomas contained about 80 times more GGT than did
normal liver. In histochemical staining, proliferating oval cells
and small areas of hepatocytes stained for GGT during the
early weeks, and later most nodules, small areas of nonnodular
parenchyma, and carcinomas contained GOT.
During spontaneous carcinogenesis in male C3HeB/FeJ
mice, premalignant lesions, accompanied by a slight increase
of serum AFP, precede the appearance of liver tumors. No
cells staining for AFP were detected during this early stage.
Once overt liver cancers had developed, AFP was readily
detectable in the tumors and was localized to some but not all
carcinoma
cells. The corresponding
serum AFP levels were
highly elevated. In contrast to the high levels of GOT found
during chemical carcinogenesis, no elevation of GOT was
found in livers at various stages of spontaneous carcinogene
sis, including cancers in eight individual mice.
These results indicate that the production of AFP and GOT
is not turned on as a single ‘
‘genetic
package,' ‘
and that these
two markers differ in their behavior in liver carcinogenesis.
in the circulation and of GOT in the lesions of the liver serve as
useful markers for the progress of the carcinogenic process
(1, 4, 8, 22).
In rats and mice, liver cell cancer (hepatocellular carcinoma,
hepatoma) can be induced by various chemicals. Several stud
ies have previously shown that in rats serum AFP concentration
increases early during the chemical hepatocarcinogenesis, and
the elevation shows a biphasic pattern (14, 15, 26, 31 ). The
first, transient increase coincides with the appearance of small
oval cells; some of these cells contain AFP as shown by
immunofluorescence and immunoperoxidase stainings (16, 18,
25, 30). The second increase of serum AFP concentration is
believed to indicate appearance of neoplastic cells (14, 16).
GOT is a membrane-bound enzyme which catalyzes the
transfer of the -y-glutamyl moiety of glutathione to an amino
acid acceptor (10). Normal adult hepatocytes contain only
small amounts of GGT activity, but in the fetal and neonatal rat
liver GOT activity is substantially higher (4). In contrast to
normal hepatocytes, neoplastic and preneoplastic liver cells
seem to contain increased amounts of GOT as studied in rats
during chemical hepatocarcinogenesis (4, 7, 8, 13, 28). While
the expressions of both AFP and GOT have been studied
separately, no studies are available on simultaneous measure
ment of both these markers.
Two models for hepatocarcinogenesis, chemically induced
tumors and spontaneous hepatomas, are available in mice.
Some strains of mice spontaneously develop hepatocellular
carcinomas with a high frequency, and this process is associ
ated with an increase of the concentration of serum AFP (3,
12). In this report, we show that the expression of GOT and
AFP, while similar in chemical carcinogenesis, differs in spon
taneous hepatocarcinogenesis.
MATERIALS AND METHODS
Mice. Female C3H/A/BOM mice from Cl. Bomholtgard, Ltd.
(Ry, Denmark) were used for the study of chemical hepatocar
cinogenesis. In our hands, females of the C3H/A/BOM strain
have had a relatively low frequency of spontaneous hepatomas.
Of 20 mice observed by repeated determinations of AFP level
INTRODUCTION
and autopsy at sacrifice, until the age of 20 months, one
In experimental hepatocarcinogenesis, the liver undergoes developed a hepatoma. Twenty-two of the 54 female C3H/A/
a series of morphological and biochemical alterations leading BOM mice used in the feeding experimentswere control mice
ultimately to overt liver cell cancer (6). The increases of AFP3 from an immunotherapy experiment (1 1). They were immunized
with complete or incomplete Freund's adjuvant without any
antigen added, once a month as detailed in Ref. 11. Such
the Finnish Cancer Society. To whom requests for reprints should be addressed,
immunization has been found to affect neither the serum AFP
at the Department of Bacteriology and Immunology, University of Helsinki, 00290
level
nor the histology of the liver of C3H mice (12). Since no
Helsinki 29, Finland.
2 Supported
by Grants
CA
1 9894
and
CA
1 6434
from
National
Cancer
differences were seen in the kinetics of serum AFP during
Institute, Department of Health, Education, and Welfare.
carcinogenesis or in the appearance of GOT in the liver be
3 The abbreviations
used
are: AFP,
a-fetoprotein;
GGT,
y-glutamyltranspepti
tween the immunization control mice and mice that did not
dase; 0-AT, o-aminoazotoluene.
Received January 17, 1979: accepted June 7. 1979.
receive any additional treatments, the results from the 2 groups
1 Recipient
of
grants
SEPTEMBER1979
from
the
Orion
Foundation,
the
Lääke
Foundation,
and
3495
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1979 American Association for Cancer Research.
H. Ja!anko and E. Ruos!ahti
have been pooled. Male C3HeB/FeJ mice from The Jackson
Laboratory (Bar Harbor, Maine) were used for the study of
spontaneous liver cancer. The mice were toe-marked individ
ually, kept in plastic cages, and fed standard mouse pellets
and water ad !ibitum.
Carcinogen Feeding. o-AT (ICN Pharmaceuticals, Plainview,
N. V.) was dissolved in maize oil and mixed with standard
laboratory meal (Sotka, OTK, Finland) to obtain a diet contain
ing 0.06% (w/w) o-AT. Feeding of the carcinogen was started
when the mice were 4.5 months old. The feeding was continued
for 8 months until the animals were 12.5 months old.
Bleedings. The mice were bled from the upper tail vein at 2to 6-week intervals. Blood (100 @@l)
was collected in calibrated
capillary tubes and diluted with 0.4 ml of phosphate-buffered
saline (0.01 55 M NaH2PO4.H2O,0.01 55 M NaHPO4.7H2O, pH
7.4).
Quantitation of AFP. Measurement of serum AFP was done
by competitive double-antibody radioimmunoassay (20, 21).
The monitoring of AFP to obtain livers with premalignant and
malignant lesions during spontaneous hepatocarcinogenesis in
male C3HeB/FeJ mice has been described previously (12).
Immunofluorescence. The mice were exsanguinated under
pellet was suspended in 0.6 ml of 100 m@iTris-HCI (pH 7.6)
containing 10 mM MgCl2, and used for assay.
GOT was assayed with ‘y-glutamyl-p-nitroanilide
(Sigma) as
substrate and glycylglycine as acceptor using previously pub
lished procedures with slight modifications (19, 27). Tissue
aliquots of 0.25 ml were incubated with 0.75 ml of substrate
mixture at 37°for 15 mm. The final reaction mixture contained
4 mM L-'y-glutamyl-p-nitroanilide, 50 mM glycylglycine, 10 mt.i
MgCI2, and 100 mM Tris-HCI (pH 7.6). The reaction was ter
minated by addition of 3 ml of 1.3 M acetic acid, and the
solution was clarified by centrifugation at 105,000 x g for 15
mm. Enzyme activity (nmol of p-nitroaniline liberated per mg of
protein per 15 mm) was calculated using a molar extinction
coefficient of p-nitroaniline at 405 nm of 9900 M1 cm1.
Sample aliquots to which acetic acid was added before sub
strate solution were used as blanks. Protein was determined
according
to the method of Lowry et a!. (1 7). Normal and
pathological tissue samples were measured simultaneously in
every assay.
RESULTS
ether anesthesia, and a careful autopsy was performed on
each mouse. The liver was sliced, and macroscopic tumors
and suspect areas were separately processed. For immunoflu
orescence staining, each block was fixed in 95% cold ethanol
containing 1% acetic acid and processed according to the
Serum AFP during Chemical Hepatocarcinogenesis. Dur
ing chemical carcinogenesis in female C3H/A/BOM mice,
elevated serum AFP was first observed at 3 weeks after com
mencement of o-AT feeding (Chart 1). This early increase
method of Sainte-Marie (24). The sections were labeled with
months. At the first peak, AFP levels ranged from 295 to 2,600
the use of a double-antibody
ng/mI and, ultimately, from 600 up to 34,000 ng/mI. The AFP
procedure with goat anti-mouse
AFP as the first antibody and IgO of a horse anti-goat IgO
serum labeled with fluorescein (Sigma Chemical Co., St. Louis,
Mo.) as the second antibody. A serial section was labeled with
rabbit anti-mouse lgG as the first antibody and fluorescein
subsided
and was followed
by a sustained
increase
at 5
levels prior to the carcinogen treatment varied between 10 and
280 ng/ml. All but 2 of the 54 mice treated with the carcinogen
10
labeled goat anti-rabbit lgG as the second antibody to control
for nonspecific staining caused by the presence of serum
proteins in the sections. The same or serial sections were
5.0
afterwards stained with hematoxylin and eosin for histological
study.
GGT Histochemistry. GOT was detected according to the
2.5
method of Autenberg et a!. (23). Small tissue samples were
frozen in liquid nitrogen, cut in a cryostat into 5-@sections, air
dried, fixed in cold acetone for 30 mm, and incubated in a
freshly prepared solution containing L-y-gIutamyl-4-methyl-f@naphthylamide (Vega-Fox Biochemicals, Tucson, Ariz.) as the
substrate, Fast Blue BB salt (Sigma) as the diazonium coupling
reagent, and glycylglycine
(Fluka AG, Buchs SG, Switzerland)
0.50
as the acceptor.
For histological study, the same or serial sections were later
fixed with 10% formalin and stained with hematoxylin and
eosin. In some cases, a Carnoy fixation followed by staining
with toluidine blue (pH 5) was used (5).
Assay for GGT. Tissue samples were frozen in liquid nitrogen
and stored at —20°.
In livers containing large macroscopic
tumors, these tumors were first isolated and assayed sepa
rately. For the assay, samples were homogenized with 12
strokes of a Teflon homogenizer in 5 ml of an ice-cold solution
containing 0.25 M sucrose, 100 mM Tris-HCI (pH 7.6), and 10
mM MgCI2. The homogenate was centrifuged at 9,1 50 x g for
10 mm, and the supernatant was collected. This was further
centrifuged at 105,000 x g for 60 mm (Beckman Model L5-65
centrifuge) to pellet the microsomal membrane fraction. The
3496
I
0.25
0.10
0
1
2
3
4
5
6
7
8
9
10 11
MONTHS
Chart 1. Serum AFP concentration during chemical hepatocarcinogenesis in
female C3H/A/BOM mice. Feeding of 0.06% o-AT was started immediately after
the first bleeding and continued for 8 months. AFP levels were monitored for the
first 5 months in 54 mice and thereafter in 27 mice. The line was drawn using the
mean values: bars, SE. The elevations of the AFP levels at the first peak and
from the fifth month on are statistically highly significant ( p < 0.001).
CANCER RESEARCH VOL. 39
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1979 American Association for Cancer Research.
AFP and GOT in Hepatocarcinogenesis in Mice
showed a value higher than 300 ng/ml at some point between
3 and 10 weeks of the treatment, and 70% of the mice showed
a greater than 5-fold increase of serum AFP level during this
period.
Chemically Induced Tumors. After 8 months with carcino
gen followed by 2.5 months without carcinogen, 25 of the 28
(89%) of the female C3H/A/BOM mice observed for this length
contrast to the findings in chemical carcinogenesis,
the sur
rounding parenchyma was relatively normal, and no AFP-con
taming cells could be observed in these areas. Staining for AFP
in the early phase of spontaneous carcinogenesis did not reveal
positive cells. The livers of 4 old (1 4 to 18 months) male
C3HeB/FeJ mice with slightly elevated serum AFP levels (380
to 760 ng/ml), but no macroscopic tumor, were cut in small
surface of the liver was covered by 10 to 50 small nodules with
pieces, and several sections from different pieces were studied.
Histologically, areas of hyperplasia, groups of atypical hepa
tocytes, and in one case strong inflammation were found, but
diameters ranging from barely visible to 3 mm. In addition to
no cells positive for AFP were present.
of time had grossly visible liver tumors.
In all tumor mice, the liver contained multiple tumors. The
this, 1 1 livers (39%) contained one large hepatoma varying
from 5 to 12 mm in diameter. Two livers (7%) contained 2 large
tumors each. Ten of the 13 large tumors (77%) were invasive,
showed areas of papilliform and adenoid growth, necrotic
areas, abundant mitoses, and various nuclear abnormalities.
The remaining 3 large tumors and about 80% of the smaller
tumors were well differentiated. In these tumors, the normal
architectural structure of the liver was preserved, no invasion
of the surrounding parenchyma was observed, and the cells
looked relatively normal. Many tumors contained alternating
areas with malignant and benign appearance. Microscopic
nodular areas, variation of cell size, nuclear abnormalities,
and
GGT in the Liver. Biochemical determination of GOT from
tissue homogenates was performed at different stages of chem
cal and spontaneous hepatocarcinogenesis. GOT activity was
low in normal mouse liver (Table 1). Fetal liver contained from
100 to 1000 times more GOT than did adult liver. During
carcinogenesis induced by o-AT, a significant increase was
found within 1 week after the carcinogen regimen had been
started. This increase preceded the elevation of the serum AFP
level. In the late phase, GOT activity was on the average of 80
times higher than in normal liver, both in carcinomas and in
livers showing various premalignant lesions.
In histochemical staining for GGT, small proliferating oval
mitoses were present in the parenchyma surrounding the tu
mors. No clear-cut correlation was found between the circulat
ing AFP level and the pathology of the liver lesions.
cells and small areas of normal-looking hepatocytes (Figs. 9
and 10) were positive during the early weeks of chemical
AFP in the Liver. During the early weeks of chemical hepa
tocarcinogenesis,
damage of hepatocytes, cell hyperplasia,
areas stained for GOT, either partially or completely (Figs. 1 1
and 12). Small GOT-positive areas were found also in the
nonnodular parenchyma in this late phase (Fig. 13). Some
carcinomas stained uniformly with GOT (Fig. 14); in others,
variation in the intensity of staining was observed.
In contrast to the marked increase of GGT activity found in
and appearance of small, proliferating oval cells were seen
histologically.
Immunofluorescence
staining revealed AFP in
some of the oval cells. This is similar to previous findings in
rats (16, 18, 25, 30). Some scattered cells, not clearly different
from normal adult hepatocytes,
and 2).
also stained for AFP (Figs. 1
During the second increase of serum AFP, when nodular
areas, bizarre cytological changes, and frank carcinomas ap
peared, various patterns of AFP staining were observed. Stain
ing for AFP was found in the nonnodular parenchyma in small
groups of cells or in solitary hepatocytes (Figs. 3 and 4). A vast
majority of nodular areas were negative for AFP, but in about
10% of the nodules in the late phase, all or some cells con
tamed AFP (Figs. 5 and 6). The cells in the AFP-positive
nodules showed cytological changes, but no clear histological
difference was found between the few nodules containing AFP
positive cells and the well-differentiated nodules in which no
staining for AFP was observed. The large carcinomas resulting
from carcinogen treatment were positive for AFP. There were
groups of cells which stained more brightly than the rest of the
tumor. In one of the mice, an infiltrating
reticuloendothelial
neoplasm was observed. This tumor was negative for AFP.
Hepatomas that developed spontaneously in male C3HeB/
FeJ mice also contained AFP. Eight macroscopic tumors, which
varied from 1 to 12 mm in diameter, were studied. Histologi
cally, 3 of these tumors were clearly invasive, had lost the
normal architectural structure, and showed cellular atypia. In
5 tumors, the cells looked less abnormal and the normal
architectural structure of the tissue was preserved. AFP-con
taming cells were found in both types of tumors. The intensity
of the staining for AFP varied in different parts of the tumors,
and the AFP-positive cells were mainly found in groups (Figs.
7 and 8) as was the case with chemically
SEPTEMBER 1979
induced tumors. In
cardinogenesis. Later on, approximately 80% of the nodular
the liver lesions during carcinogenesis induced by o-AT, no
increase of GOT activity was found in spontaneous liver lesions
whether malignant or benign. All the 8 macroscopic tumors
described above were negative for GOT, both biochemically
and histochemically (Figs. 15 and 16; Table 1). The livers of
the 4 mice with benign lesions and slightly elevated serum AFP
levels also had normal GOT levels.
DISCUSSION
The behavior of 2 carcinofetal markers, AFP and GGT, has
been compared in chemical and spontaneous hepatocarcino
genesis in mice. We found that, while the levels of both of these
markers increased early during the process of chemical carci
nogenesis, only AFP was present in spontaneous liver card
nomas.
Both AFP and GGT were present in some oval cells and
adult-type hepatocytes during the early stage of chemical
carcinogenesis. For technical reasons, it was not possible to
obtain evidence on the question of whether the same cells
contained both AFP and GGT during this or any other situation
where both markers were present. Due to leakage of AFP,
staining for AFP could not be done reliably on frozen sections,
although we attempted
this using several different
fixatives.
GOT, on the other hand, became inactivated under conditions
where AFP could be stained. Differences in the expression of
AFP and GOT became apparent during the later phases of
chemical carcinogenesis.
A great majority of nodules with no
signs of invasiveness contained large amounts of GOT,
3497
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1979 American Association for Cancer Research.
H. Ja!anko and E. Ruos!ahti
@
Table 1
Expression of AFP and GGT during chemical and spontaneous hepatocarcinogenesis in mice
Chemical carcinogenesis was induced in female C3H/BOM mice by feeding the mice 0.06% o-AT for 8 months. Early phase means here the early weeks of the
carcinogen feeding, when serum AFP concentration first became increased. Late phase means the time from 5 months on (Chart 1) when a sustained increase of
serum AFP level occurred. Spontaneous carcinogenesis was followed in male C3HeB/FeJ mice by monitoring the serum AFP concentration and sacrificing the mice
either soon after increase of AFP was observed (early phase) or later when carcinomas were found in the liver (late phase). Fetal blood and livers were taken from
fetuses 1 to 3 days before estimated time of delivery. Normal samples were obtained from adult C3HeB/FeJ and C3H/A/BOM mice. Measurements and stainings
Methods.AFPGOTType
were
done as described in Materials and
in liveraCells
in serum (@zg/ml)Cells
of liver
RangeNormal0.18
tissueAmount
containing AFPAmount
GOTMeanRangeMean
containing
0.01—0.28 (25)bNo
staining0.40—2.8
1100—4900 (4)Hepatocytes436250—623
cellsFetal1970
(1 1)bScattered
(5/2)Ductal
ductal
cells
HepatocytesChemicalcarcino
genesisEarly
cellsLate
phase0.77
0.29—2.6
(54)Some
of small oval cells60.3—1
hepatocytes
(27)Scattered
Some late nodular areas
Solitary hepatocytes and
groups of hepatocytes
outside the nodules
Groups of carcinoma cells31
0.24—34
phase3.70
7
—32
2731
5—87
cellsSpontaneouscarcino
genesis
Early phase
Late phase0.42
0.27—i.4
(19)e
staining observedt
0.05—420 (30)@NoGroups of carcinoma cells0.8
1.30
(5)Most
of small oval
areas of hepatocytes
Most nodular areas
Small groups of hepatocytes
outside the nodules
(5/3f
(7/3)C. dSmall
Uniform or uneven staining of
carcinoma
.6
(4/2)C
0.5—0.6 (8/3)CNo
0.50—1
staining of hepatocytes
No staining of carcinoma cells
a Nmol of p-nitroaniline liberated per mg protein per 15 mm.
b Numbers
in parentheses,
C Individual
samples
d Isolated
a Some
I Four
large
of the
livers
number
were
pooled
of mice
to
yield
tested.
enough
tissue
for
the
assay.
The
number
of
mice
used
as
source
of
samples/number
of
pooled
samples
tested
is
shown.
carcinomas.
AFP
from
values
mice
were
with
obtained
elevated
from
serum
Ref.
12.
AFP
concentration
were
stained.
Hyperplasia,
cell
atypia,
and
in one
case
strong
inflammation
were
observed
histologically.
whereas only few nodules in the late phase stained for AFP. In
addition to the lack of AFP staining in most nodules, a major
dissimilarity in the expression of AFP and GOT concerns the
sion of such tumors during their maintenance by transplantation
(9).
One could argue that the expression of GOT in chemically
staining patterns observed.The benign and malignant lesions
tended to stain uniformly with OGT, clearly contrasting the
induced tumors may be due to their more malignant nature
and/or faster proliferation rate as compared to spontaneous
hepatomas, rather than to difference in the etiology of the
tumor. However, many of the spontaneous tumors are histolog
pathological
lesion
against
the
surrounding
parenchyma,
whereas AFP was mostly found in scattered cells or in groups
of cells within the lesions. This difference may at least partly
depend on the fast secretion of AFP and its dependence on
the phase of the cell cycle (26). It is also possible that the
ically poorly differentiated, produce tumors when transplanted
to syngeneic animals (2), and, judged from the changes in
serum AFP level (12), progress with a speed at least equal to
production of the 2 markers is associated with different stages
of cell differentiation.
that of chemically induced tumors.
The main finding of this study concerns the different expres
sion of GOT in chemical and spontaneous carcinogenesis.
in spontaneous carcinogenesis, it is possible that the induction
of GOT is mainly associated with the exposure of these animals
Since the expression of GOT does not seem to be increased
localization of AFP did not reveal any
to chemical carcinogens. An increase of GGT activity can be
differences between the chemically induced and spontaneous
liver tumors. In both cases, patches of AFP-positive cells were
seen in the liver within 24 to 48 hr after a single large dose of
a carcinogen (8), and it is known that neoplastic liver cells are
seen. In contrast to this, we found GOT to increase only during
more resistant to toxic effects of carcinogens than are normal
liver cells (6). This is in agreement with the role GOT is thought
to play in the detoxification of compounds such as carcinogens
(29). The increased resistance to the toxicity of the carcinogen
Immunohistochemical
chemical hepatocarcinogenesis. No increase of GOT activity
was seen during the early stages of spontaneous carcinogen
esis or in the resulting hepatomas. It seems that increased GOT
is rarely, if ever, a characteristic of primary spontaneous hep
atomas in C3H mice. Fiala et a!. (7) reported increased GOT
activity in 1 of 4 transplantable tumors, originally obtained from
spontaneous hepatomas in C3H x V mice. However, primary
hepatomas such as the ones we have studied are likely to be
more representative of spontaneous hepatomas than are trans
planted tumors since changes may occur in the gene expres
3498
could convey a selective advantage to the cells containing GOT
and could explain the presence of GOT in the proliferative
lesions induced by carcinogens. However, it is obvious that the
presence of large quantities of GOT in fetal liver cells cannot
be explained on the basis of the above considerations.
Also,
the fact that we have studied spontaneous hepatomas in only
one highly inbred mouse strain does not warrant general con
CANCERRESEARCHVOL. 39
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1979 American Association for Cancer Research.
AFP
@
clusions on the role of GOT in malignant liver cells.
How the expression of AFP and GOT is regulated is not 14.
known, but the present results indicate that these 2 carcinofetal
markers are not turned on simultaneously as a single “genetic
package. ‘
‘
They possibly reflect different aspects of malig
nancy or the development of liver carcinomas through different
pathogenetic processes (25). The difference in the expression 16.
of AFP and GOT in chemical and spontaneous carcinogenesis,
if true in other species, might allow conclusions to be made on
17.
the etiology (chemical carcinogenesis versus other causes) of
a given liver tumor.
18.
REFERENCES
19.
1. Abelev, 0. I. a-Fetoprotein as a marker of embryo-specific differentiations in
normal and tumor tissues. Transplant. Rev., 20: 3—37,1974.
20.
2. Andervont,H. B., and Dunn,T. B. Transplantationof spontaneousand
21 .
3.
4.
5.
6.
induced hepatomas in inbred mice. J. NatI. Cancer Inst., 13: 455-503,
1952.
Becker, F. F., Stillman, D., and Sell, S. Serum a-fetoprotein in a mouse strain
(C3H-A'@fB)with spontaneous hepatocellular carcinomas. Cancer Aes., 37:
870—872,1977.
Cameron, A., Kellen, J., Kolin, A., Malkin, A., and Farber, E. @Glutamyltrans
ferase in putative premalignant liver cell populations during hepatocarcino
genesis. Cancer Res., 38: 823—829,1978.
Daoust, A., and Calamai, R. Hyperbasophilic foci as sites of neoplastic
transformation in hepatic parenchyma. Cancer Aes., 31: 1290—1
296, 1971.
Farber, E. The pathology of experimental liver cell cancer. In: M. H. Cameron,
D. A. Linsell,and0. P. Warwick(ed). LiverCellCancer,Vol. 1, pp. 243277. Amsterdam: Elsevier Press, Inc., 1976.
7. Fiala, S., Fiala, A. E., and Dixon, B. y-Glutamyl transpeptidase in transplant
able,chemicallyinducedrat hepatomasand spontaneousmousehepato
mas. J. Natl. Cancer Inst., 48: 1393-1401 , 1972.
8. Fiala, S., Mohindru, A., Kettering, W. 0., Fiala, A. E., and Morris, H. P.
Glutathione and gamma-glutamyl transpeptidase in rat liver during chemical
carcinogenesis. J. NatI. Cancer Inst., 5 7: 591 —598,1976.
9. Hancock, R. L., and Grunau, J. Soluble RNA methylase activity of transplant
able mouse hepatomas. J. NatI. Cancer Inst., 45: 971 —978,1970.
10. Hanes, C. S., Hird, F. J. A., and Isherwood, F. A. Synthesis of peptides in
enzymic reactions involving glutathione. Nature (Lond.), 166: 288—292,
1950.
11. Jalanko, H., Engvall, E., and Ruoslahti, E., Immunochemical properties of
alpha-fetoprotein (AFP) and autologous antibodies to AFP. Immunol. Com
mun. 7: 209—222,1978.
12. Jalanko, H., Virtanen, I., Engvall, E., and Ruoslahti, E. Early increase of
serum alpha-fetoprotein in spontaneous hepatocarcmnogenesisin mice. lnt.
J. Cancer,21: 453-459, 1978.
13. Kalengayi, M. M. A., Ronchi, 0., and Desmet, ‘I.
J. Histochemistry of
and
GGT
in Hepatocarcinogenesis
in Mice
gamma-glutamyl transpeptidase in rat liver during aflatoxin B-induced car
cmnogenesis.J. NatI. Cancer Inst., 55: 579—588,1975.
Kitagawa, T., Yokochi, T., and Sugano, H. Alpha-fetoprotein and hepato
carcinogenesis in rats fed 3'-methyl-4-dimethylammnoazobenzeneor N-2fluoroenylacetamide. lnt. J. Cancer, 10: 368—381
, 1972.
Kroes, A., Sontag, J. M., Sell, S., Williams, 0. M., and Wiesburger, J. H.
Elevated concentrations of serum alpha-fetoprotein in rats with chemically
induced liver tumors. Cancer Res., 35: 1214—1
217, 1975.
Kuhlmann, W. 0., Localization of alpha,-fetoprotein and DNA-synthesis in
liver cell populations during experimental hepatocarcmnogenesisin rats. Int.
J. Cancer, 21: 368—380,1978.
Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, A. J. Protein
measurement with the Folin phenol reagent. J. Biol. Chem., 193: 265—275,
1951.
Onoe, T., Kaneko, A., Dempo, K., Ogawa, K., and Minase, T. Alpha-fetopro
tom and early histological changes of hepatic tissue in DAB-hepatocarcino
genesis. Ann. N. V. Aced. Sci., 259: 168—180,1975.
Orlowski, M., and Meister, A. Isolation of y-glutamyl transpeptidase from
hog kidney. J. Biol. Chem., 240: 338-347, 1965.
Pihko, H., and Ruoslahti, E. High level of alpha-fetoprotein in the sera of
adult mice. Int. J. Cancer, 12: 354-360, 1973.
Ruoslahti, E. Antigen-antibody precipitates and immunoadsorbents: appli
cation to purification of alpha-fetoprotein. Scand. J. lmmunol., Suppl. 3, 3649, 1976.
22. Ruoslahti,E., andSeppálä,
M. Alpha-fetoprotein
in cancerandfetaldevel
opment. Adv. Cancer Res., in press, 1979.
23. Rutenberg, A. M., Kim, H., Fischbein, J. W., Hank, J. S.. Wasserkrug, H. L.,
and Seligman, A. M. Histochemical and ultrastructural demonstration of ‘y
glutamyl transpeptidase activity. J. Histochem. Cytochem., 17: 51 7—526,
1969.
24. Sainte-Marie, 0. A paraffin embedding technique for studies employing
immunofluorescence. J. Histochem. Cytochem., 10: 250—256,1962.
25. Sell, S. Distribution of a-fetoprotein- and albumin-containing cells in the
livers of Fischer rats fed four cycles of N-2-fluorenylacetamide. Cancer Res.,
38: 3107-31
13, 1978.
26. Sell, S., Becker, F. F., Leffert, H., and Watabe, H. Expression of an
oncodevelopmental gene product (a-fetoprotein) during fetal development
and adult onco9enesis. Cancer Res., 36: 4239—4249,1976.
27. Szasz,0. A kineticphotometricmethodfor serumgamma-glutamyl
trans
peptidase. dIm. Chem., 15: 124-136, 1969.
28. Taniguchi, N., Tsukada, Y., Mukuo, K., and Hirai, H. Effect of hepatocarcin
ogenic azo dyes on glutathione and related enzymes in rat liver. Gann, 65:
381-387, 1974.
29. Tate, S. S., Thompson, G. A., Miester, A. Recent studies on gamma-9lutamyl
transpeptidase. In: I. M. Arias and N. B. Jakoby (eds.), Glutathione: Metab
olism and Function, pp. 45—55.New York: Raven Press, 1976.
30. Tchipysheva, T. A., Guelstein, V. I., and BAnnikov, G. A., Alpha-fetoprotein
containing cells in the early stages of liver carcmnogenesisinduced by 3'methyl-4-dimethylaminoazobenzene and 2-acetylaminofluorene. lnt. J. Can
cer (Phila.), 20: 388-393, 1977.
31 . Watabe, H. Early appearance of embryonic a-globulin in rat serum during
carcinogenesis with 4-dimethylaminoazobenzene. Cancer Aes., 3 1: 1192—
1194, 1971.
Fig. 1. AFP-containing hepatocytes in the fifth week of chemical hepatocarcinogenesis in female C3H/A/BOM mice. All the AFP-positive cells in this figure and
in Figs. 3, 5, and 7 were negative when stained with anti-mouse IgG. lmmunofluorescence, x 400.
Fig. 2. The same section as in Fig. 1. The AFP-positive cells are histologically indistinguishable from normal adult type hepatocytes. H & E, x 400.
Fig. 3. A solitary AFP-positive hepatocyte in the 8th month of carcino9enesis induced by o-AT. Immunofluorescence, x 400.
Fig. 4. Same area as in Fig. 3 showing the atypical AFP-positive hepatocyte (arrow) in nonnodular parenchyma. H & E, x 250.
Fig. 5. Nodular area of AFP-positive cells during ninth month of chemical carcinogenesis. Immunofluorescence, x 100.
Fig. 6. Same area as in Fig. 5 with higher magnification showing the small nodular area (arrows). Cellular hypertrophy, mitoses, and bizarre cytological changes
are seen. H & E, x 250.
Fig. 7. Group of AFP-positive cells in spontaneously developed hepatoma in a male d3HeB/FeJ mouse. lmmunofluorescence, x 400.
Fig. 8. The same area as in Fig. 7, showing the well-differentiated carcinoma. H & E, x 400.
Fig. 9. Small GOT-positive area in the fourth week of chemical hepatocarcinogenesis. GGT staining, x 250.
Fig. 10. Same area as in Fig. 9 of a serial frozen section. GOT-positive area corresponds to the small group of hepatocytes (arrow). H & E, x 250.
Fig. 11. GOT-positive area in eighth month of chemical hepatocarcinogenesis. GOT staining, x 200.
Fig. 12. Same area as in Fig. 11 of a serial frozen section. The nodulus (arrows) is composed of 2 types of cells. The left part is made of small basophilic
hepatocytes expressing GGT activity (Fig. 11), and the right half is formed by vacuolated clear hepatocytes not staining with GGT. Toluidine blue, x 100.
Fig. 13. Three GGT-positive areas of different sizes (arrows) in the late phase of chemical hepatocarcinogenesis. GOT staining, x 250.
Fig. 14. Edge of a poorly differentiated hepatoma induced by o-AT showing uniform high GOT activity. GOT staining, x 100.
Fig. 15. GOT staining of spontaneously developed hepatoma in a male C3HeB/FeJ mouse. No activity observed. x 100.
Fig. 16. The same moderately differentiated carcinoma as in Fig. 15. H & E, x 200.
SEPTEMBER 1979
3499
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1979 American Association for Cancer Research.
@
@
@
•:
.@ @-.@@‘@f!
@
@
@
@
@@/
@‘
@:
@‘
@e:.i
:@
I
C
H. Jalanko and E. Ruos!ahti
.@
@
.:
Os
I
1@
@.
@
.
.
.
@,
C
(@)
...
C
4@
tl.
.jQ
@:
1@
;@
%,
-*
0
@
,r
@
.@
@r!‘ ;
.
@
.@
..
@
‘
@.
.
-
@*i―@@
. i. ‘
@
-
.
a.
I
•.i@.'
@,: .
.@
‘@-@‘
‘
,@)
j@,
_
ó:@
.
.
..
-
•
w
.‘
_•@@
@
@
.,
I'
.
:a@r@
-‘S
@‘
;,
. ..
-@
I
,:@c
;
•
.@
c@
‘r
‘@T
@
@!
@f4@
•, .
,
@f
,
,@1
@
@
;@ I've':@).@‘2,4@1'
‘
“I
.@.
..
.
,,
@..
I
:; ::
.
;r.J
t4'
“ •
:‘
@.
C
4
@
@
,
:@ - ,@
9
•:-
p
!@
a4)
-@
@
..@
@
e_) ,
@
:@
.
@
.
C,.
.
•I@'
@
.
@
0
0.
. 0
.@
0
(•
ff@
a'
@‘è' 0
•
.
0
.
@, 0
0
•
C
0
@
•
—I.
3500
•@
.
C
•
•4II@'
CANCERRESEARCHVOL. 39
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1979 American Association for Cancer Research.
@
@
@h
‘@
@. *@
:@p@'
@
.
I
AFP
and
GGT
@I@$
in Hepatocarcinogenesis
@
in Mice
:‘@
@
C
@
@r@%.1r@
I
1@
Ti@;@1.
@
@
‘
@
p
@
.9
?‘p ,
@I
“
•
•‘@@r(.
. ‘0'
*@ ‘P
e@
#
,
@:
,‘:_@.
:@
c
@
.:@
@
•-@-@
11
@‘
.-
I
4;
•@.i@
I
@:12
C
V
‘@
@
,,
@
@
-..
.,,@,
%
@
‘&,
.@
.
@
@S4@
,
‘t':@@ ‘@.$@‘ I@
4.
@
I..
15
I@
,‘•“_
‘@
‘:
‘@?
.
@
!
.:
@
I
,
.,.@
L@v@;..@I.%A@#;
@,.
.
.1
.,‘
‘
..@:“
‘ @%,
4.
:
.
I,
@
‘.@
,.. .
@‘‘@‘
@i*@,,;;
@
@
@I@'4'•
.
.
@1-
@.
>- ...
,
‘. .
•/•@...
.
..
@*..
.
-‘
SEPTEMBER1979
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1979 American Association for Cancer Research.
3501
Differential Expression of α-Fetoprotein and γ
-Glutamyltranspeptidase in Chemical and Spontaneous
Hepatocarcinogenesis
Hannu Jalanko and Erkki Ruoslahti
Cancer Res 1979;39:3495-3501.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/39/9/3495
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1979 American Association for Cancer Research.