[CANCER RESEARCH 35, 305-309, February 1975]

The Relationship between Metabolism, DNA Binding, and

Carcinogenicity of 15, 16-Dihydro- 11-methylcyclopenta
[a]phenanthren-17-one in the Presence of a Microsomal
Enzyme Inhibitor
Maurice
Chemistry

M. Coombs,
Department,

Tarlochan

Imperial

S. Bhatt,

Cancer Research

and Cohn W. Vose'

Fund Laboratories,

Lincoln's

SUMMARY
The mean latent period for skin tumor production by the
carcinogen
I 5, 16-dihydro- 11-methylcyclopenta
[ajphe
nanthren-l7-one
(Compound
IVb) in the mouse was 30
weeks for a dose of 60 zg/week and about 45 weeks for 60
jzg/week, while at 0.6 zg/week, no tumors were observed
during 100 weeks. Simultaneous
administration
of the
closely related noncarcinogen (IVa) (54 @zg/week)together
with the carcinogen at 60 zg/week had no effect on the
mean latent period. Simultaneous
administration
of a
threefold quantity of the microsomal
enzyme inhibitor
7,8-benzoflavone (I) with the carcinogen at the highest dose
increased the mean latent period to 38 weeks, while at the
intermediate dose it completely suppressed tumor forma
tion.
Neither ketone IVa nor IVb bound covalently to calf
thymus DNA in vitro without prior metabolic activation.
After incubation with rat liver microsomes and NADPH in
the presence of air, both ketones bound covalently to added
DNA in vitro, the noncarcinogen (IVa) about four times
more extensively than the carcinogen (IVb), roughly in
proportion to the overall extents to which these ketones
were metabolized. In contrast, overall metabolism of the
carcinogen (IVb) was somewhat increased by the addition of
a threefold quantity of the inhibitor (I) to the incubation
mixture, but binding to added DNA was almost completely
prevented.
These results are discussed in connection with the hypoth

Inn Fields, London

WC2A 3PX, England

strated by Kinoshita and Gelboin (8, 9). These authors

also discovered that tumor production
by the related
hydrocarbon
benzo(a)pyrene
(III) was not inhibited, but
rather was somewhat enhanced by this flavone. They
showed that metabolism of DMBA was required before
covalent binding of the latter to DNA could be observed in
vitro and that the flavone (I) prevented this binding when
added to the incubation mixture. In addition, binding of
DMBA to the DNA ofmouse skin in vivo was inhibited 60%
by this flavone.
The carcinogen
15,l6-dihydro-ll-methylcyclopenta
[a]phenanthren-l7-one
(IVb) (2, 4) has, in common with
DMBA, a phenanthrene ring system bearing a methyl group
in the same relative ring position. Although this compound
is not

purely

a hydrocarbon,

it was

to investigate

o4

@A@o

w
I

w
II

0
R..
a , R:H

esis that cellular DNA is the target of the carcinogen (IVb)

for tumor initiation.

of interest

whether its capacity to induce skin tumors in mice would be

inhibited by 7,8-benzoflavone. This paper reports the out
come ofthis experiment and ofexperiments on the influence
of the flavone on the in vitro metabolism and binding of this
caltinogen to DNA in the presence of microsomes prepared
from the livers of untreated rats (Chart 1).

or3H

b. RaCH3 orC3H3

Chart 1. Chemicalcompoundsmentionedin the text.

INTRODUCTION

MATERIALS

The ability of 7,8-benzoflavone

(I) to inhibit the produc

tion of skin tumors by simultaneous application of the

potent carcinogen DMBA2 (II) has recently been demon

AND METHODS

Animal Experiments

The compoundsand mixtures weretestedby twice-weekly

dorsal painting of toluene solutions (6 z1) for 50 weeks,

followed by observation of the animals for a 2nd year.
Fund. Present address: MetabolicStudies Laboratory, G. D. Searle & Co.
Groups
of formally randomized TO albino mice, 10 male
Ltd., Lane End Road, High Wycombe,Bucks HPI2 4HL, England.
and 10 female, were used for each solution. The conditions
Received June 17, 1974; accepted October 14, 1974.
were generally as previously described (2, 4).
1 Recipient

2 The

of

research

abbreviation

FEBRUARY

used

1975

bursary

is:

DMBA,

from

the

Imperial

Cancer

Research

7,l2-dimethylbenz(a)anti@racene.

305

M.

M.

Coombs

et

a!.

The carcinogen (IVb) was applied at 3 dose levels: 0.5%

w/v, the dose arbitrarily
chosen for all previous work
(Group 1); 0.05% w/v in 2 separate experiments (Groups 2
and 3); and 0.005% w/v (Group 4). This corresponds to 60,
6, and 0.6 zg of IVb weekly; the highest dose level is known
to produce a high tumor incidence with a mean latent period
of about 30 weeks (2, 4). Four other solutions, namely, 0.5%
IVb + 1.5% w/v of 7,8-benzoflavone (I) (Group 6), 0.05%
IVb + 0.15% I (Group 7), 0.5% I alone (Group 8), and
0.05% IVI, + 0.45% w/v ofthe noncarcinogen (IVa) (Group

[1 l-3H]ketone

(IVa)

was synthesized

as follows:

17,17-

ethylenedioxy - 11,12,13,14,15,16 - hexahydrocyclopenta[a]

phenanthren-ll-one
(100 mg, 0.34 mmole) (I) was reduced
with sodium borotritide (54.6 mg, 1.44 mmoles, 100 mCi!
mmole) in dry tetrahydrofuran,
under reflux. Exchange
able tritium was removed from the product dissolved in
diethyl ether, by washing with N NaOH, and the crude
11-hydroxyketal
(1 11.4 mg, 33.8 mCi/mmole)
was de
hydrated and aromatized
by heating with glacial acetic
acid (4 ml), nitrobenzene (1 ml), and concentrated HC1 (1
ml), as previously described (I). The product was purified
by chromatography
on Welm Grade II alumina (10 g) to
yield pale yellow needles of the [1 l-3H]ketone (IVa) (88
mg; specific activity, 20.4 mCi/mmole),
m.p. l99_20l0
[literature, 202203
(1)], UV and infrared spectra identical
with those of an authentic, unlabeled specimen.

5), were also used. Groups painted with toluene (Group 9),
or clipped but otherwise untreated (Group 10), served as
negative controls. The dates of first appearance of skin
tumors were recorded, and the tumors were generally
allowed to grow to 8 to 10 mm in diameter before the
animals were killed. A few sick animals were also killed to
avoid loss of skin tumors. Autopsies were performed
whenever possible, but histological examination was con
fined to the skin tumors; no other tumors were observed by
visual inspection. In previous work with this carcinogen (2,
4), skin tumors of this size usually involved the panniculus
carnosus muscle and were, therefore, classified as carcino
mas rather than papillomas. With this criterion, the pathol
ogy ofthe tumors induced with IVb in Groups 1, 2, 3, 5, and
6 is presented (Table I).

Metabolism and DNA Binding

Calf thymus DNA and NADPH were purchased from
Sigma Chemical Co., London, England. A Spinco prepara
tive ultracentrifuge
was used for high-speed sedimenta
tions. Radioactivity
was counted in a Nuclear-Chicago
Model 2 liquid scintillation counter using Triton phosphor

(Triton X-lOO, 500 ml; PPO, 4 g; toluene, 1 liter); count

ing times were adjusted to give l0@ counts and were cor
rected for quenching. Thin-layer
chromatography
was
carried out on glass plates (20 x 20 cm) coated to a thick
ness of 0.25 mm with Silica Gel G (E. Merck AG, Darm
stadt, Germany) and dried for at least 20 hr at ambient

Preparation of Tritiated Compounds

The [I l-3H3]methyl ketone (IVb) (specific activity, 24.6

mCi/mmole)

was prepared

as recently described

(3). The

Table I
Number

of mice surviving

without tumors and histology

of induced tumors at site of application

Each group initially consisted of 10 male and 10 female animals.

withSqua
No. ofmice

Squa
oftumorless
mo.1lVb,6O@ig/wk14
GroupTreatmentNo.

cinoma6
mo.

12 mo.

survivors atmous
18 mo.

0410@b2IVb,6gig/wk(lst20
27Iaexperiment)3IVb,6pg/wk(2nd19

4
13

1
7

262experiment)4IVb,0.6pg/wk20

11

65IVb,6@ig/wk
073c54@ig/wk6IVb,60@g/wk+I,19
+ IVa,20

19
10

17
2

039@l80@ig/wk-7IVb,6@g/wk+I,20

18

13

18
19

12
17

18

18

518
@g/wk8I,
59Toluenecontrol20
l80pg/wk19
410Clipping

control20
a The number
Group 6,2.

of tumors

b One

additional

animal

C One

additional

skin

unavailable
had

tumor

a tumor
of

latent

for histology
(spindle
period

cell
67

were in Group
sarcoma)

weeks

regressed

car
pap
illomamous

24

at

the

1, 2; in Group 2, 1; and in
site

when

of
it

application.
had

reached

mm

in

diameter.

306

CANCER RESEARCH

VOL. 35

Effect of 7,8-Benzoflavone

temperature.
Before use, the plates were washed by as
cending elution with chloroform:methanol
(3:1, w/v), and
material washed to the solvent front was isolated by draw
ing a line across the silica gel layer 1 cm from the top of the
plate. After the plates were redried in air, the metabolite
mixture, dissolved in ethanol, was applied as a thin line I
cm from the base of the plate, which was then developed
by ascending elution with a mixture of toluene:ethyl
acetate:methanol
( 15:5: 1, by volume) to the line drawn 1
cm from the top. The plate was divided into bands cor
responding to fluorescent zones seen in UV at 254 nm, and
each band was carefully scraped off. The metabolites were
eluted from the recovered silica gel with ethanol and their
radioactivity
was determined
by scintillation
counting.
When possible, the UV absorption of the ethanolic solution
was determined with a Perkin-Elmer Model 402 UV spec
trophotometer.
In each case, the unmetabolized ketone ran
as the least polar band, and was identified on the basis of
its RF value, and its fluorescent and UV spectra.
Preparation of the Microsomal Fraction of Rat Liver.
All work was carried out at 4,following the method of
Gelboin and Sokoboff (6). Liver from male 75-g Sprague
Dawley rats, freshly killed by stunning, followed by cervical
dislocation, was homogenized in ice-cold 0.25 M sucrose
(10 ml/2 g liver). After removal of cell debris at 9000 x g
for 10 mm, the supernatant was centrifuged at 16,000 x g
for 10 mm to remove mitochondria,
and the microsomes
were aggregated at 90,000 x g for 40 mm. The microsomes
were washed by resuspension in 0. 1 M sodium phosphate
buffer (pH 7.4) (5 ml/2 g liver), centrifuged at 90,000 x g
for 40 mm, and finally resuspended in fresh buffer (1.4 ml!
1 g of liver) for the binding experiments.
DNA Binding in Vitro. Incubations were carried out in
0. 1 M phosphate buffer at 37for 30 mm with gentle shaking
in air. Each tube contained DNA (2 mg), NADPH (2 mg),
microsome suspension (0.2 ml), and 3H-labeled ketone (IVa
or b) (30 ;zg) in pure dimethyl sulfoxide (20 jul) in a total
volume of 3 ml. Metabolism was quenched by cooling in ice,
the microsomes were removed by centrifugation at 80,000 x
g for 1 hr at 4, and the supernatant was extracted with
water-saturated
phenol (3 ml). The upper, aqueous layer
was separated, treated with ethoxy-ethanol (3 ml), and the
precipitated DNA was sedimented at 3,000 x g. The DNA
was washed with 1:1 v/v ethoxyethanol:water
(3 ml), with
absolute ethanol (3 x 3 ml), and with hot absolute ethanol
(3 x 3 ml). The residue was finally dissolved in l0@ M
sodium chloride solution (1 ml), and DNA phosphorus was
determined colorimetrically
(7). Aliquots were each made
up to 1 ml with water, and radioactivity was counted in 10
ml of Triton scintillant. Each value quoted in Table 2 is a
mean of 5 replicate determinations.
In Vitro Metabolism. Incubations were carried out with
gentle shaking at 37in 250-mi conical flasks open to air.
Flasks contained microsome suspension (5.6 ml), 0. 1 M
phosphate buffer, pH 7.4 (56 ml), NADPH (56 mg, 60.7
zmoles), and either labeled ketone (IVa) (1.0 mg, 4.3
jzmoles), or the latter plus 7,8-benzoflavone (3.3 mg, 12.1
jsmoles) in dimethyl sulfoxide (0.3 ml). Incubations were

FEBRUARY

1975

on Carcinogenicity

Table 2
Extent ofcovalent binding of the ketones, IVa and IVb, to calf thymus
DNA in vitro
Each value rcpresents the mean of 5 replicate determinations, and each
section contains results from 1 batch of microsomes. For further details see
the text.
bindingindex
DNA
(jzmolcsketone/moleMicrosomalof
DNASubstratepreparationphosphorus)IVbPreparation
I
3.OaIVbPreparation

Active34.4

1.2IVbPreparation

Active18.1

2.5IVbHeat

Active46.5

inactivated (2
100)0IVb

mm at

equivalentsActive07,8-benzoflavone
+ 3 mole
(I)IVbPreparation
4
2.0IVbAbsent0IVb

Active18.1

IActive0.87IVaActive77.3
+ 3 mole equivalents
5.0IVaAbsent0a

Mean

SE.

terminated by cooling in ice, followed by extraction with

ethyl acetate (5 x 25 ml). Emulsions were separated by
low-speed centrifugation,
the organic solutions were corn
bined, dried over anhydrous Na2SO4, and evaporated to
dryness under reduced pressure below 40. The residues
were dissolved in ethanol (10 ml) for chromatographic
and
spectral examination. The percentages of the total radioac
tivity recovered in the ethyl acetate extracts were as follows:
IVa, 73%; IVb, 67%; and IVb + 7,8-benzoflavone, 69%.
RESULTS
Table 1 and Chart 2 summarize the skin-painting experi
ments. At the highest dose, the carcinogen (IVb) induced
tumors at the site of application in 17 of 20 mice alive at the
appearance ofthe 1st tumor, with a mean latent period of 29
weeks in agreement with previous results (2, 4). Reduction
of the dose to one-tenth prolonged the mean latent period by
about 50%, to 43 weeks in the 1st experiment, and to 46
weeks in the 2nd, and reduced the tumor incidence to 9 of 20
and 8 of 19, respectively. At this dose, simultaneous
painting ofa 9-fold amount ofthe inactive ketone (IVa) had
little effect on the latent period, although the tumor
incidence was slightly higher ( 11 of 20). No tumors were
observed in the animals painted with one-hundredth of the
highest dose, and their survival was similar to the toluene
treated and untreated controls.

The ratio of carcinomas to papillomas amongst the tu

307

A@@

M.

M.

Coombs

et

a!.

E
0

z
6pg +l8pg BF@0 tumors
orO6pg alone

Chart 2. Date of first appearance of skin tumors and number of mice

with tumors produced by the carcinogen (IVb) at 60, 6, and 0.6 @tg/week,

and also on simultaneous administration ofthe noncarcinogen (IVa) and of

the inhibitor 7,8-benzoflavone (BF, I) at the doses indicated. Experimental
details are given in Materials and Methods. Neither

the differences

in

tumor incidence (a) with IVb at 60 ag/week in the presence and absence of
inhibitor, or (b) in the 3 experiments with IVb at 6 ag/week are statistically
significant (p > 0.05). For the difference in tumor incidence between the

groups treated with IVb at 60 and 6@zg/week,p < 0.01. expt., experiment.

mors produced appeared to be dose dependent. At 60 zg/

week (Group 1), carcinomas predominated,
whereas at 6
ag/week (Groups 2, 3, and 5), the reverse was true. A
marked tendency was also noticed for the rate of growth
of the skin tumors, as determined by the time taken for
them to grow to 8 to 10 mm from their 1st appearance, to
increase with increasing latent period. Thus for Group I
this time was 10 to 30 weeks for tumors appearing before
the 25th week, but was more than 40 weeks for those ap
pearing after the 40th week. Also, histology demonstrated
that tumors of this size were usually carcinomas, whereas
smaller tumors were often papillomas. Most of the tumors
induced with the carcinogen at the lower dose (Groups 2,
3, and 5) grew slowly, and more than one-half of them did
not reach 10 mm. This may at least partly account for the
observed differences in the carcinoma:papilloma
ratio.
7,8-Benzoflavone was not found to be carcinogenic in this
test system, as was previously established also for mice of a
different strain (9). When 7,8-benzoflavone
was painted,
together with the carcinogen, at 60 ag/week (Group 6) at 3
times the dose of the carcinogen, it prolonged the mean
latent period by almost 30% (8 weeks) and reduced the
tumor incidence to 14 to 20. The carcinoma:papilloma
ratio and rate of growth of the tumors were similar to those
observed with the carcinogen above (Group I). Tumor
production was totally inhibited by this flavone in the same
relative excess with the carcinogen at 6 @tg/week (Group 7).
The general survival of these animals was similar to that of
the controls.
In a number of experiments, it has been observed con
sistently that the noncarcinogen (IVa) is metabolized about
3 times faster than the related carcinogen (IVb) under
standard conditions. In the present experiment (Table 2,
Preparation 4), summarized diagrammatically
in Chart 3,
IVa and IVb were metabolized to the extent of64 and 24%,

308

respectively, and each gave rise to a similar band of me

tabolites at RF 0.7 to 0.75. In addition, both ketones gave
smaller quantities of a number of more polar metabolites,
several of which, by comparison of patterns of R@values
and fluorescence colors in UV, appeared to possess homol
ogous structures. Incubation of IVb in the presence of a 3fold amount of the inhibitor 7,S-benzoflavone (I) did not
appreciably alter the pattern of these metabolites, but ra
ther more of IVb was metabolized (35%).
No covalent binding ofeither IVa or IVb was observed in
vitro (Table 2) without metabolism, for the radioactivity
initially associated with added DNA was readily extracted
with organic solvents. After incubation of both ketones with
rat liver microsomes and NADPH, nonextractabbe radioac
tivity remained associated with the DNA. This is inter
preted as covalent binding of the ketones, since there is no
evidence to suggest loss of tritium from the molecules under
the mild conditions of these experiments.
Although the
binding index of IVb varied appreciably
with different
batches of microsomes, as did the extent of its metabolism,
there was fairly close agreement among replicate assays
using the same microsome preparation. Under these condi
tions (Table 2, Preparation
4), the noncarcinogen
(IVa)
bound to DNA about 4 times more than the carcinogen
(IVb), while binding of the latter was almost completely
prevented in the presence of a 3-fold quantity of 7,8-benzo
flavone in the incubation mixture.

DISCUSSION
A 10-fold reduction in the highest dose (60 pg/week) of
the carcinogen increased the latent period by 50%, corn
pared with an increase of 30% when 7,8-benzoflavone was
applied together with the carcinogen at this high dose. The
inhibitor therefore substantially reduces the effectivedose
of the carcinogen in vivo. The complete suppression of
tumor formation by the inhibitor with the carcinogen at 6

40
0

IVa
@20

S
,;,o

Jbflb

y py

?71

IVb

pyb

bbb

0.6 0.5

04

py

@-@-@----

RF

1P 0.9

0.8

0.7

03

0.2

Chart 3. Profiles of metabolites produced by in vitro incubation with

rat liver microsomes of the ketones, IVa and IVb, as disclosed by thin-layer
chromatography. For details, see Materials
and Methods.The following
letters refer to fluorescence colors seen in UV of wavelength 254 mm: b,
blue; g, greenish yellow; o, orange; p, purple; y, yellow. In each case, the
unmetabolized

substrate

formed the least polar band (RF 0.9 to 0.95).

CANCER RESEARCH

VOL. 35

Effect of 7,8-BenzoJlavone
jzg/week, and the failure of the latter to produce tumors
when administered alone at 0.6 zg/week seem to be in line
with this interpretation.
The manner in which 7,8-benzoflavone reduces the effec
tive dose can be rationalized if we accept that chemical
attack leading to covalent binding with a crucial, cellular,
informational
macromolecule
(DNA, RNA, or protein) is
required for initiation of the carcinogenic process (10). It is
then apparent from Table 2 that the reactive form of the
carcinogen is generated by metabolism, for no binding to
added DNA occurs in vitro in the absence of active mi
crosomes. If the inhibitor functions by diminishing the
amount of proximate carcinogen formed in this way and
available for covalent binding, it does not do so by generally
inhibiting metabolism,
for enhancement
of the overall
metabolism of (IVb)@is observed.

In the absenceofthe inhibitor, the in vitro binding of both

on Carcinogenicity

understanding of the nature of proximate

carcinogens of related chemical structure.

forms of these 2

ACKNOWLEDGMENTS
We are indebted to Dr. L. Pang for confirming the pathology of the
induced

skin tumors.

We also thank

M. Hall

for valuable

technical

assistance.

REFERENCES
I. Coombs, M. M. Potentially Carcinogenic Cyclopenta[a]phenan
threnes. I. A New Synthesis of l5,16-Dihydro-I7-oxocyclopenta[a]phenanthrene and the Phenanthrene Analogue of 18-Noroestrone
Methyl Ether. J. Chem. Soc. C, 955962,1966.
2. Coombs,

M. M., Bhatt, T. S., and Croft, C. J. Correlation

between

Carcinogenicity and Chemical Structure in Cyclopenta[a]phenan

threnes. Cancer Res., 33: 832837,1973.

the carcinogen (IVb) and the noncarcinogen (IVa) to DNA

is roughly proportional to the extent to which these ketones

3. Coombs, M. M., and Crawley, F. E. H. Characterisation

of an
Epoxide-dibenzoxpin
as a Major Urinary Metabolite of the Carcino

are metabolized. Thus in both casesmetabolism appearsto

gen 15,16-Dihydro- I l-methyl-cyclopenta[a]phenanthren- 17-one, J.

Chem, Soc. Perkin Trans. 1: 2330, 1974.
4. Coombs, M. M., and Croft, J. C. Carcinogenic Cyclopenta[aJ

produce chemically reactive derivatives capable of covalent

macromolecular binding. Consequently, the amount of
binding to DNA in vitro is not positively correlated with
carcinogenicity,
for the noncarcinogen
(IVa) binds more
extensively. This result is similar to that of Rayman and
Dipple (12) who were unable to find a positive correlation
between carcinogenicity and the extent of in vitro binding to
salmon sperm DNA of the strong carcinogen 7-bromo
methyl- l2-methyl-benz[a]anthracene
in comparison with its
weaker homolog, 7-bromomethylbenz[a]anthracene.
These
authors likewise failed to establish a correlation
with
binding to mouse skin DNA in vivo, after treatment of the
animals with these compounds under conditions used for
skin tumor induction (13). They enumerate reasons why
lack of correlation between extent of DNA binding and
carcinogenecity does not necessarily exclude DNA from
consideration as the initial target for cancer induction by
chemicals, and this topic has recently been discussed in a
more general way be Miller and Miller (I 1).
Nevertheless, the parallelism between, on the one hand,
inhibition by 7,8-benzoflavone of skin tumor production
with either the carcinogen (IVb) or DMBA (8), and on the
other hand, inhibition of in vitro macromolecular binding of
these compounds is noteworthy. The inhibitor also protects
ce!ls in culture from the cytotoxic effects of polycyclic
hydrocarbons
(5) and suppresses the microsomal enzyme
system (aryl hydrocarbon hydroxylase) responsible for their
metabolic activation. It seems reasonable to expect that a
closer study of the way in which this inhibitor diminishesthe
carcinogenecity
of IVb and DM BA will lead to a better

FEBRUARY

1975

phenanthrenes.
Progr. Exptl. Tumor Res. II: 69-85, 1969.
5. Diamond, L., and Gelboin, H. V. Alpha-Naphthoflavone:
an Inhibitor

of Hydrocarbon Cytotoxicity and Microsomal Hydroxylase. Science,

166:10231025,
1969.
6. Gelboin, H. V., and Sokoloff, L. Effects of Methylcholanthrene and
Phenobarbital

on Amino Acid Incorporation

into Proteins.

Science,

134:611612,
1961.
7. King, E. J., and Wootton,
I. D. P. Microanalysis
in Medical
Biochemistry, Ed. 3, p. 77. London: J. & A. Churchill Ltd., 1956.
8. Kinoshita, N., and Gelboin, H. V. Aryl Hydrocarbon Hydroxylase and
Polycyclic Hydrocarbon Tumorigenesis:
Effect of the Enzyme Inhibi

tor 7,8-Benzoflavone on Tumorigenesis and Macromolecule Binding.

Proc. NatI. Acad. Sci. U. S., 69: 824-829, 1972.
9. Kinoshita, N., and Gelboin, H. V. The Role of Aryl Hydrocarbon
Hydroxylase in 7, l2-Dimethylbenz[ajanthracene Skin Tumorigenesis:
On the Mechanisms of 7,8-Benzoflavone Inhibition of Tumorigenesis.
Cancer Res., 32: 13291339,1972.
10. Miller, J. A. Carcinogenesis by Chemicals: An OverviewG.H. A.
Clowes Memorial

Lecture. Cancer Res., 30: 559576, 1970.

11. Miller, C. E., and Miller, J. A. Biochemical Mechanisms of Chemical

Carcinogenesis. In: H. Busch (ed), The Molecular Biology of Cancer,
pp. 377402. London: Academic Press, Inc., 1974.
12. Rayman, M. P., and Dipple, A. Structure and Activity in Chemical
Carcinogenesis. Comparison of the Reactions of 7-Bromomethylbenz
[a]anthracene and 7-Bromomethyl-12-methylbenz[a]anthracene
with
Deoxyribonucleic
Acid in vitro. Biochemistry, 12: 12021207, 1973.
13. Rayman, M. P., and Dipple, A. Structure and Activity in Chemical
Carcinogenesis. Comparison of the Reactions of 7-Bromomethylbenz
[a]anthracene
and 7-Bromomethyl-l2-methylbenz[a]anthracene
with

Mouse Skin Deoxyribonucleic Acid in vivo. Biochemistry, 12:

15381543,
1973.

309