v>EPA
United States
Environmental Protection
Agency
Health Effects Research
Laboratory
Research Triangle Park IMC 27711
Research and Development
EPA-600/S1-81-046 July 1981
Project Summary
Improved Scoring of Chemical
Transformation of
C3H/10T1/2 Cells
Charles Heidelberger
This research program was under-
taken to improve the scoring of the
transformation by chemical carcino-
gens of C3H/10T1/2 mouse embryo
fibroblasts.
1) A probabilistic view of the trans-
formed focus formation in these
cells induced by methylcholan-
threne (MCA) treatment has been
formulated and validated. The
authors define PI as the proba-
bility that a cell will be activated
by carcinogen treatment, P2 as
the probability per cell generation
that an activated cell will be
transformed, and P3 as the prob-
ability per cell generation that an
activated cell will be deactivated.
The equation derived is
log(F/N) = log [2p,p2 (1 -p3)/2(1 -p3)-1 ] +
nlog (1-p3),
where F = mean no. of foci per dish; N =
no. of cells at confluence; and n = no.
of cell generations to confluence.
2) 5-Azacytidine induces differen-
tiation of C3H/10T1/2 cells
into both muscle cells and adi-
pocytes. Although phorbol-ester
related tumor promoters inhibit
muscle cell formation, this is not
affected by inhibitors of tumor
promotion; moreover, other
classes of tumor promoters do
not inhibit this differentiation.
Effects of promoters on adipo-
cyte formation were inconsistent.
Hence, this assay cannot be used
to screen for tumor promoters.
3) The powerful tumor promoter,
12-0-tetradecanoyl phorbol-13-
acetate {TPA), produces tempo-
rary and reversible rounding up
of the cells and loosened adhe-
sion to the substratum. Although
TPA does not affect the growth
rate of these cells in 10% fetal
calf serum, it does stimulate their
growth rate in 1 % fetal calf serum.
4) A quantitative study of the "nat-
ural history" of clones derived
from various morphological types
of transformed foci was carried
out. All tumorigenic Type III
clones formed colonies in soft
agarose, this growth efficiency
increased with passage number,
but only one Type II and no Type I
clones grew in soft agarose. Type
I clones often progressed to Type
II, and many Type II clones
progressed to Type III upon fur-
ther passaging.
5) The cell-surface morphology was
studied in the scanning electron
microscope (SEM). Nontrans-
formed cells had flat, smooth
surfaces. Transformed cells in
logarithmic growth exhibited
rough surfaces with ruffles, blebs,
and microvilli; however, the sur-
faces of transformed cells post
confluence became smooth.
"Mini-foci" of rough-surfaced
-------
cells were visible within five days
after MCA treatment.
6) Preliminary experiments revealed
that monoclonal antibodies com-
mon to transformed clones, which
are probably against oncofetal
antigens, can be prepared, and
should be useful for scoring trans-
formation.
7) Mouse peritoneal macrophages
activated by BCG treatment se-
lectively kill chemically trans-
formed, but not nontransformed
C3H/10T1/2 cells.
8) By mutagenesis of atumorigenic
transformed clone of these cells,
followed by selection post-
confluence at 39.5° with FUdR
and extensive cloning, six mu-
tants temperature-sensitive for
transformed phenotypes were
isolated and characterized.
This Project Summary was devel-
oped by EPA's Health Effects Research
Laboratory, Research Triangle Park.
NC, to announce key findings of the
research project that is fully docu-
mented in a separate report of the
same title (see Project Report ordering
information at back).
Introduction
It is now generally accepted that 70-
90% of human cancers are environ-
mentally caused. Since our environment
contains over 500,000 naturally oc-
curring and manmade chemicals that
have not been tested for carcinogenicity,
and since it costs about $500,000 per
compound for a conventional large
scale carcinogenesis test in hundreds of
rats and mice kept throughout their
lifespan, it would cost $250,000,000,000
to test them all conventionally. This is
clearly prohibitive. Thus, the inescapable
conclusion is that much quicker and
less expensive methods of prescreening
must be developed and validated.
The widely heralded Ames test ap-
pears to be highly promising as such a
prescreen. There is an impressive cor-
relation between the carcinogenic ac-
tivies of 300 chemicals and their
microsome-activated mutagenesis to
tester strains of Salmonella typhimurium.
However, the correlation is not perfect,
there is yet no conclusive proof that
carcinogenesis results from a somatic
mutation, and (trite as is sounds) bacte-
ria do not get cancer. Hence, it would be
folly to rely solely on one prescreening
system.
At the present time, several systems
have been developed for obtaining
quantitative data on the oncogenic
transformation of cultured cells by
chemical carcinogens. It seems clear
that such systems, although more com-
plicated and tedious than a bacterial
mutagenesis assay, may be more rele-
vant as prescreens for carcinogenic
activities. They also provide valuable
test materials for studying the cellular
and molecular mechanisms of chemical
carcinogenesis; such information will
be of practical value in providing means
to.prevent human cancer.
In this study, a system was developed
for studying chemical oncogenesis in
vitro with the C3H/10T1/2 mouse
embryo fibroblasts. However, before
this system can be validated as a poten-
tial prescreen by testing in it a larger
number of carcinogenic and noncarcin-
ogenic chemicals, additional research
must be done to perfect its quantitative
application.
The perfection and validation of this
system that determines and quantitates
oncogenic transformation will produce
the following benefits:
1) The availability of a reliable pre-
screen for environmental carcino-
gens (or mixtures thereof), which
is one of the highest priorities in
environmental surveillance.
2) Such a pre-screen can be widely
applied to test large numbers of
samples (pure compounds and
mixtures) of environmental pol-
lutants for their carcinogenic ac-
tivity.
3) The determination of the oncogenic
transformation of cultured animal
cells is much more relevant to
human carcinogenesis than mea-
suring mutagenesis in bacteria or
other biological or biochemical
parameters.
4) The cost, space, and manpower
requirements to test individual
samples would be expected to be
less than 1 % of those necessary to
carry out adequate tests for car-
cinogenesis in living rodents.
5) In addition to surveillance of gen-
eral environmental samples for
carcinogenic activity, this system
could be used to monitor the
changes in carcinogenic activities
of specifically collected specimens.
6) The perfection of this pre-screen
will also provide important funda-
mental information on the mecha-
nisms of chemical carcinogenesis
that, if intelligently applied, might
lead to prevention of many types of
human cancer.
7) Such a surveillance could lead to
major steps in the abatement of
carcinogenic pollutants.
Results and Discussion
Probabilistic Theory of
Transformed Focus Formation
One of the original aims of the project
was to apply to C3H/10T1/2 cells the
transformation of single individual cells
in dishes or microtiter plates, such as
had been previously done with mouse
prostate fibroblasts. It soon became
evident, however, that the effects of cell
density on transformation were so
major as to obscure the very meaning of
the concept of transformation frequency.
This was reinforced by early reconstruc-
tion experiments in which focus forma-
tion was measured after mixing of
known numbers of nontransformed and
transformed cells. Here it became evi-
dent that the numbers of foci in the
dishes did not follow Poisson distribu-
tions, and that there was considerable
migration of transformed cells to form
satellite foci. Moreover, we found that
the use of microtiter plates for trans-
formation experiments, even with single
cells, was not suitable because of
variations in the background and indis-
tinctness of the transformed foci that
precluded their classification as Type I,
II, or III, as originally defined and studied.
It soon became clear that the enormous
amount of time-consuming labor that
went into the preparation and inspection
of the dishes containing single cells, as
well as the enormous number of dishes
that would be required to obtain statisti-
cally significant results, precluded this
approach to improved quantitation and
scoring of this bioassay system. These
complexities led to a theoretical and
experimental examination of a proba-
bilistic theory of transformed focus
formation, which should have major
impact on the field.
Quantitative Study of the
Properties of Cells Derived
from Various Types of Foci
In this study, a rigorous investigation
was made of various of the transformed
phenotypes in a series of cell lines
derived as defined below, in order to
relate the "natural history" and evolu-
tion of these cultures to the phenomenon
of tumor progression. In a variety of cell
-------
culture systems an extremely tight
correlation has been observed between
the ability of a transformed cell to grow
when cultured in semi-solid media
(anchorage independence for growth)
and its ability to grow as a tumor in vivo.
At the inception of this study, very little
was known about the relationship be-
tween C3H/10T1/2 transformation
and anchorage independent- growth.
The acute need for cell culture cor-
relates with in vivo tumorigenicity
prompted the initiation of this work. The
ability of cells to grow in semi-solid
medium was assayed following a modi-
fication of the basic procedures of
Macpherson for soft agar culture.
Agarose was found to be superior for
the culturing of 58MCA Cl 16 cells,
permitting the formation of large colo-
nies with a high plating efficiency. An
agarose overlay concentration of 0.25%
was found to be optimal. Higher overlay
concentrations reduced colony sizes.
Lower agarose concentrations were too
dilute and colonies were observed to
grow, migrate, fragment and fuse.
Once the optimal conditions for soft
agarose culture had been determined,
C3H/10T1/2 Cl 8 and a number of
transformed variants were assayed for
anchorage independence of growth. All
eight tumorigenic cell lines tested grew
in soft agarose culture. In general, all
transformed cell lines fell into one of
two classes with respect to the size of
the colonies formed in soft agarose
culture. Most of the cell lines grew to
form only small colonies (25-200 cells)
but some cell lines grew to form very
large colonies containing 1000 or more
cells.
The ability of cells to grow as tumors
was assayed in syngeneic C3H mice.
Mice were X-irradiated (450r) 24 hours
prior to the injection of cells. Cells
(2x106) were then inoculated subcuta-
neously and the animals examined
weekly for six months for the presence
of tumors.
Of the five nontumorigenic cell lines
tested here, three (including C3H/1OT1 /
2) showed no, or very poor, capacity for
growth in soft agarose.
The ability of a cell line to grow in soft
agarose increased as a function of the
number of passages it is maintained in
culture, since the plating efficiency and
colony size of cells grown in agarose
increased with passage number. The
results with OMBA Cl III are particularly
noteworthy. These cells were nontu-
morigenic when tested at passage 7, but
were tumorigenic when tested at pas-
sage 15. The acquisition of tu morigenicity
in these cells appears to be accompanied
by an increase in the ability to grow in
soft agarose.
The fact that none of these early
transformed cell lines were tumorigenic
at early passages, whereas 58 MCA Cl
16 was highly tumorigenic, suggests
that oncogenic transformation occurs in
several steps or that the transformed
cell lines are too strongly antigenic to
induce tumors- in C3H mice that were
immunosuppressed by 450 radX-irradi-
ation. The feasibility of using intracranial
injections or nude mice for tumorgenicity
studies is now under study.
Morphology of Nontransformed
and Transformed Cells in the
Scanning Electron Microscope
(SEM)
Subconfluent cultures of C3H/10T1 /2
cells taken at various times following
treatment for 24 hr with 1.0 Aig/ml of
MCA were examined in the SEM. The
surface morphology of untreated cells is
flat and featureless.
This study recently discovered that
these carcinogen-treated cultures form
a new transformation parameter, "mini-
foci," which are seen also in cultures
derived from Type II and Type III foci.
Mini-foci are groups of cells growing on
top of each other in a criss-cross fashion
and are detected under subconfluent
culture conditions. They have been
detected in MCA exposed cultures as
early as four and seven days following
treatment. Similarly grouped cells have
also been found in 7-day cultures treated
with MCA (0.1 fjg/m\, 24 hr) alone or
MCA (0.1 Aig/ml, 24 hr) plus TPA (0.1
//g/ml, four days after MCA treatment)
but apparently in much lower frequency.
Mini-foci have not been detected in
control-cultures.
SEM studies of carcinogen treated
confluent cultures proved uninteresting
and did not reveal differences in cell
surface morphology that could be defi-
nitely identified as similar to or charac-
teristic of transformed (Cl 16) cells. This
was true even with cells in foci-like
areas identifiable by light microscopy.
Because of these rather disappointing
results, the author proposed to reexam-
ine certain assumptions and start with
cells from foci of transformed cells
detectable by light microscopy. Sublines
of cells isolated from the three different
types of foci - Type I, Type II and Type III
were studied. A "Type 0" subline,
isolated from the "flat" areas between
foci of cells in carcinogen exposed
cultures, was also available. These, as
well as controls (subline from acetone
treated cells) were coded and processed
for SEM examination. The cultures
were seeded as usual (5000 cells per
dish) and fixed for SEM at 4, 5 (subcon-
fluent), 8 (confluent), and 14 days. The
identity of each sample was checked
only after completion of SEM examina-
tion.
SEM studies of these samples yielded
encouraging results with subconfluent
(4, 5 days) cultures. Cells with altered
morphology and distinguishable from
controls were easily seen in Type III and
Type II cultures. These cultures had cells
that often occurred in groups and tended
to overlap each other in a criss-cross
fashion. These "mini-foci" were taken
to be significant in view of their similarity
to the foci seen in the light microscope,
and are considered to be diagnostic of
transformed cells. Additionally, many
cells in these early cultures showed a
profusion of microvilli and blebs on their
surfaces, and presented a morphology
similar to Cl 16 or other oncogenic
transformed cell populations. Type I
cells were mostly non-overlapping and
had fewer altered cells with microvilli
and blebs, but their sparseness made
scoring of these cultures as transformed
difficult and equivocal. Type 0 cultures
were similarly equivocal. Control, ace-
tone treated cultures at similar subcon-
fluency showed little or no detectable
criss-cross cells, but contained cells
with microvilli and ruffles. These cells at
the relatively low (less than 400x)
magnifications commonly used could
not be distinguished as being different
from those seen in the Type 0 and Type I
cultures. Thus, distinctions between
Type I, Type 0, and A (acetone) controls
are equivocal and subjective. Experience
suggests that quantitative analyses
should be attempted in the future to
make clear the distinctions, if any,
among these cultures.
In other experiments, cells from Cl 16
(MCA-transfprmed, tumorigenic) were
examined after growth on glass or
plastic surfaces under confluent and
subconfluent conditions. The results of
this study showed easily detectable
cells with transformed morphologies in
Cl 16 cultures. Subconfluent cultures
were best for detection of the trans-
formed cells. Cl 16 cells in comparison
to control cells appeared less tightly
adherent to the growing surface (more
-------
rounded). Cell surfaces often contained
blebs and microvilli. Microvillar distri-
bution and size often appear to be
heterogeneous. Cell criss-crossing was
also noted. This overall morphology was
maintained even in five-day-old cultures,
a time at which control cultures are
well-spread, actively growing and often
nearly confluent. Confluent Cl 16 cul-
tures were much less remarkable in
appearance, had many fewer cells with
large blebs or microvilli, and generally
were difficult to distinguish from control
10T1/2 cells. Nonetheless, closer ex-
amination showed subtle changes in
cell-cell associations and cell shape.
In summary, the experiments done so
far permit the conclusion that trans-
formed cells are recognizable by scan-
ning electron microscopy at early times
(within one week of carcinogen expo-
sure) in terms of changes in cell surface
morphology, and/or cell shape and/or
cell to cell association (close, tight
adhesion vs. loose contact; criss-cross,
piled vs. parallel oriented, contact-
inhibited). Secondly, the expression of
these changes is dependent upon cul-
ture conditions such as cell density.
Thirdly, it was noted that treatment of
C3H/10T1/2 cells with two different
chemical carcinogens can result in dis-
similar but transformed phenotypes.
Development of Monoclonal
Antibodies to Oncofetal
Antigens on Transformed Cells
For the prototype experiments, two
oncogenically transformed clones were
chosen that had previously been shown
to exhibit the oncofetal antigen: MCA Cl
15 and DMBA Cl 2. C3H mice were
immunized three times at weekly inter-
vals with 5x105 DMBA Cl 2 cells by
intraperitoneal injections and were bled
three days after the last immunization.
The binding capacity of the antibodies to
the transformed cells was determined
by the binding of protein A. This protein,
which is a single polypeptide isolated
from the cell wall of Staphylococcus
aureus, has an extraordinary affinity of
binding to the Fc region of the immuno-
globulin IgG. Thus, the assay measures
the amount of protein A (labeled with
125I) that binds to the Fc portion of the
antibody, which in turn binds to the
surface antigen of the intact cell through
the Fab portion of the antibody. The
transformed cells (1 xl O5) were plated in
Falcon 96 well microtiter plates and
allowed to attach for 12 hours. They
were then washed three times with
Basal Medium Eagle's (BME) with 1%
bovine serum albumin. The cells were
then treated for 30 minutes at4°C with
the mouse antiserum or various dilutions
thereof in BME. The cells were again
washed 3 times with BME and incubated
for 1 hr at room temperature with 105
cpm of 1Z5l-protein A (Pharmacia) radio-
iodinated with chloramine T (86) in
phosphate buffered saline (PBS), and
then washed three times with PBS. The
wells of the microtiter plate were sawed
out, placed in liquid scintillation vials,
and the bound 125I was counted. The
antiserum obtained from OMBA Cl 2
immunized mice bound to those cells
and to MCA Cl 15 cells to approximately
the same extent and to a significantly
higher extent than to the non-immune
serum control at dilutions of 1:5. This
similarity in extents of binding suggests
that the antibody bound to a common
cell-surface antigen on both transformed
cells.
There was no significant binding of
nontransformed C3H/10T1 /2 Cl 8 cells
to antisera prepared against those cells
or to the antiserum prepared against the
transformed DMBA Cl 2 cells. Moreover,
the transformed DMBA Cl 2 cells reacted
with the antiserum directed against the
same cells, but not with the antiserum
to the nontransformed cells. These two
experiments demonstrated that the
cells and assays used were sufficiently
selective and sensitive to allow us to
proceed to prepare monoclonal anti-
bodies from hybridomas.
C3H mice were immunized three
times once weekly with 5x106 DMBA Cl
2 transformed cells. Three days after the
last immunization, spleen cells were
collected and fused with a mouse
myeloma cell line, P3 x 63/Ag8, that
lacks thymidine kinase using polyethyl-
ene glycol. After completion of the cell
fusion and selection in HAT medium for
two weeks, the cells were then cultured
in regular medium. After the fourth
week in culture, the supernatants of the
hybridomas were added to target DMBA
Cl 2 cells plated in microtiter wells as
described previously and assayed by 125I
protein A binding. The wells were either
counted for 125I binding, or the pattern of
bound 125I was determined by autoradi-
ography at -70°C. In a typical experi-
ment, 4 out of 120 hybridoma wells
contained antibodies against the trans-
formed cells. Experiments are now in
progress to prepare pure clones of the
hybridomas that secrete antibodies of
the desired properties.
Specific Killing of Transformed
Cells by Activated Peritoneal
Macrophages
The first approach to this problem
involved setting up an assay for the
killing effect of activated peritoneal
macrophages. Since much of our work
involves determination of the reproduc-
tive capacity of cells by measuring their
plating efficiency, and since this assay
has not been applied to macrophage
killing, this method was used and
compared with the 125IUdR release
assay.
The peritoneal macrophages of C3H
mice were activated by injecting IP 1 ml
of BCG containing 2 x 107 live organisms.
Nine days later, the mice were injected
IP with 2 ml of thioglycollate 5 days
before harvesting the macrophages.
After harvesting, the cells were counted
and plated in 100 mm dishes (2x106/
dish) with 10 ml of Basal Medium Eagle
(BME) + 10% fetal calf serum (PCS).
After one hour's incubation, the medium
was aspirated out; the adherent cells
are practically all (90-95%) macrophages.
We then plated 5000 of our test cells
with 10 ml of BME + 10% FCS and incu-
bated for 48 hr. At the end of this time all
the cells were scraped from individual
dishes and replated, without counting,
into 10 100 mm dishes with 10 ml of
medium in each. The dishes were incu-
bated for 10 days and fixed with metha-
nol, stained with Giemsa and scored for
the development of colonies. The results
demonstrate that the activated macro-
phages recognize the malignant DMBA
Cl 2 and MCA Cl 16 cells and kill them,
as shown by the decrease in number of
colonies. In contrast, the nontransformed
C3H/10T1/2 cells were not killed by
the macrophages and the number of
colonies was actually increased as
compared to the dishes without macro-
phages. This is due to a feeder effect.
An exactly comparable experiment
was then set up with the nontrans-
formed C3H/10T1/2 Cl 8 cells, and
with acetone-treated and Types 0, I, II,
and Ill-derived cell lines at passage 9.
None of these cell lines at this passage
number grew significantly in soft agarose.
The activated macrophages again en-
hanced the plating efficiency of the
C3H/1OT1 /2 Cl 8 cells, but produced no
significant effect on the reproductive
capacity of the others except for a
marginal killing effect on the Type III cell
line.
Dr. Michael Fisher of the Frederick
Cancer Center has performed on our
-------
ells a newly developed microassay,
tiich is carried out in 96-well microtiter
lates. Log phase target cells are pre-
abeled for 24 hours with 0.2 //Ci/ml of
25IUdR washed three times, and trypsin-
zed and counted. Mouse peritoneal
tacrophages were activated in culture
/ith macrophage activating factor (MAP)
78) for 24 hours. Then 105 viable
trypan blue) macrophages (MP) are
dded to each microtiter well. After 45
Tin the macrophages were washed
hree times to remove nonadherent
ells, and 5 x103 target cells were added
o each well so that the ratio of macro-
ihages to target cells added was 20:1.
tfter one day, the cells were washed to
emove nonadherent cells, and incuba-
ion was carried on for three additional
'ays when the cells were washed three
imes and incubated for one hour with
):1 N NaOH, and the radioactivity of the
JaOH lysate was determined in a gamma
:ounter. In all experiments, controls
vere run in which the target cells were
ncubated with nonactivated peritoneal
nacrophages. The target cells were
;ent to Dr. Fisher at the 9th passage,
ind were tested by him at the 14th
>assage.
In these experiments there was no
:ytolysis produced by the nonactivated
nacrophages of any of the cell lines, or
>y the activated macrophages on the
lontransformed C3H/10T1/2 Cl 8,
icetone-treated, or Type I cells. How-
iver, there was significant lysis of the
Vpes II and III cells, as well as of the
>ncogenically transformed MCA Cl 16
:ells. It is not known whether the
differences in results between these
sxperiments and the ones done by
jlating efficiency are due to the dif-
erence in the assay, the difference in
he method of activation of the macro-
phages, or the difference in the passage
lumber of the Types II and 111 cells. It is
:lear, however, that the micro 125IUdR
assay is quicker, easier, requires less
sells and media, and is more accurate
nan the plating efficiency assay. Hence,
all future experiments will be done by
he micro radioactivity assay.
Conclusions
All work supported by this grant was
aimed at improving the scoring, and
making it more rapid and quantitative, of
the C3H/10T1/2 mouse embryo fibro-
blast system for oncogenic transforma-
tion and its initiation and promotion.
The following conclusions have been
drawn:
1 ) A probabilistic view of transformed
focus formation in these cells in-
duced by methylcholanthrene
(MCA) treatment has been formu-
lated and validated. We define Pi
as the probability that a cell will be
activated by carcinogen treatment,
P2 as the probability per cell gen-
eration that an activated cell will
be transformed, and Pa as the
probability per cell generation that
an activated cell will be deactivated.
The equation derived is
log(F/N) = log [2Plp8
nlog (1
where F = mean no. of foci per dish; N =
no. of cells at confluence; and n = no. of
cell generations to confluence.
2) 5-Azacytidine induces differentia-
tion of C3H/10T1/2 cells into
both muscle cells and adipocytes.
Although phorbol-ester related
tumor promoters inhibit muscle
cell formation, this is not affected
by inhibitors of tumor promotion;
moreover, other classes of tumor
promoters do not inhibit this dif-
ferentiation. Effects of promoters
on adipocyte formation were in-
consistent. Hence, this assay can-
not be used to screen for tumor
promoters.
3) The powerful tumor promoter, 1 2-
0-tetradecanoyl phorbol-13-ace-
tate (TPA), produces temporary
and reversible rounding up of the
cells and loosened adhesion to the
substratum. Although TPA does
not affect the growth rate of these
cells in 10% fetal calf serum, it
does stimulate their growth rate in
1% fetal calf serum.
4) A quantitative study of the "nat-
ural history" of clones derived
from various morphological types
of transformed foci was carried
out. All tumorigenic Type III clones
formed colonies in soft agarose,
this growth efficiency increased
with passage number, but only
one Type II and no Type I clones
grew in soft agarose. Type I clones
often progressed to Type II, and
many Type II clones progressed to
Type III upon further passaging.
5) The cell-surface morphology was
studied in the scanning electron
microscope (SEM). Nontra'ns-
formed cells had flat, smooth sur-
faces. Transformed cells in loga-
rithmic growth exhibited rough
surfaces with ruffles, blebs, and
microvilli; however, the surfaces
of transformed cells post con-
fluence became smooth. "Mini-
foci" of rough-surfaced cells were
visible within five days after MCA
treatment.
6) Preliminary experiments revealed
that monoclonal antibodies com-
mon to transformed clones, which
are probably against oncofetal
antigens, can be prepared, and
should be useful for scoring trans-
formation.
7) Mouse peritoneal macrophages
activated by BCG treatment se-
lectively kill chemically trans-
formed, but not nontransformed
C3H/10T1/2 cells.
8) By mutagenesis of a tumorigenic
transformed clone of these cells,
followed by selection post-con-
fluence at 39.5° with FUdR and
extensive cloning, six mutants
temperature-sensitive for trans-
formed phenotypes were isolated
and characterized.
Several of the initial objectives have
been realized, and sufficiently promising
preliminary data have been collected to
make it highly likely that continuation of
this research will lead to the accom-
plishment of all the objectives.
Recommendations
The numbers of these recommenda-
tions pertain to the numbers of the
conclusions previously stated.
1) The probabilistic view of trans-
formed focus formation of C3H/
10T1/2 cells following chemical
carcinogen treatment has been
derived and validated. This is
expected to have a major impact
on the quantisation and interpre-
tation of transformation in this
transformation system that is now
widely used throughout the world.
This research should be continued.
2) The effects of tumor promoters on
the 5-azacytidine-induceddiffer-
entiation of C3H/10T1/2 cells
into muscle cells and adipocytes
may be useful in establishing the
mechanism of action of phorbol
ester-related tumor promoters.
However, inconsistent results pre-
clude this from being used as a
short-term test for tumor promot-
ers. Hence, this approach has
been discontinued.
3) Since the transient rounding-up of
cells following TPA treatment does
not occur on repeated treatments,
which are required for promotion
-------
of transformation, this effect is
unlikely to be involved in tumor
promotion. The growth-stimulating
effect of TPA in 1 % fetal calf serum,
seems unlikely to be involved in
the mechanism of promotion be-
cause promotion experiments are
carried out in 10% fetal calf serum
in which no growth stimulation
was observed. Hence, these ap-
proaches are being discontinued.
4) The study of the "natural history"
of transformed foci has been com-
plicated, and further studies are
required to fully assess the signifi-
cance of the various transformed
phenotypes.
5) The appearance of "mini-foci,"
detectable in the scanning electron
microscope (SEM) only a few days
after chemical carcinogen treat-
ment, is a highly promising lead to
improve the rapidity of scoring.
This led to an application to the
EPA for continuation of this re-
search, which is currently funded
as EPA Grant No. R-808309, and
the research continues.
6) The preliminary finding that
monoclonal antibodies can be pre-
pared that react with various
transformed clones, is extremely
promising. The antibodies can be
radioiodinated and rapid, objective,
and highly sensitive radioimmuno-
assay can be developed for trans-
formation. Alternatively, the anti-
body can be labeled with fluores-
cence and transformed cells can
be visualized in a fluorescence
microscope and can be separated
in a fluorescence activated cell
sorter. This is also currently funded
on EPA Grant No. R-808309, and
the research continues.
7) Although activated mouse perito-
neal macrophages selectively kill
chemically transformed cells, it is
highly unlikely that this test can be
made sufficiently sensitive to be
used as a primary means of scor-
ing for transformation.
The successes of the approaches
described in 1, 5, and 6, together with
interesting data collected from the
research described in 2, 3, 4, and 7
indicate that the results obtained on this
grant, R-805-208, were very successful.
Charles Heidelberger is with the University of Southern California Cancer
Center, LosAngeles, CA 90031.
Michael D. Waters is the EPA Project Officer (see below).
The complete report, entitled "Irriproved Scoring of Chemical Transformation of
C3H/WT1/2 Cells," (Order No. PB 81-209 686; Cost: $6.50, subject to
change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield. VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Health Effects Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
tUS GOVERNMENT PRINTING OFFICE 1981 -757-012/7238
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
i'o 0000 32 J
u s t>viH tMurt.cn
- ^ * i . » j J1
230
ii.
------- |