EFFECT OF METHYLATION ON THE CONFORMER STABILITY AND REACTIVlTY OF THE BAY-REGION DIOL-EPOXIDES OF CHRYSENE

B.D. SILVERMAN

IBM Thomas J. Watson Research Center Yorktown Heights, NY 10598 (U.S.A.)

(Received April 13th, 1984)

(Revision received December 14th, 1984)

(Accepted December 28th, 1984)

SUMMARY

The relative stabilities of conformers of the bay-region tetrahydroepoxide

of methylated chrysene have been calculated. From these calculations on
tetrahydroepoxides, one infers that substitution of a methyl group in the

same bay-region as the epoxide should destabilize both syndiaxial and anti-

diequatorial bay-region diol-epoxide diastereomers with respect to the syn-

diequatorial and antidiaxial diastereomers. The results of these calculations,
together with recent experimental observations, suggest that the enhanced

in vivo binding to DNA of the isomer having the methyl group and the
epoxide in the same bay-region (1,2diol-3,4-epoxide of 5-MeC) might be

partially due to this destabilization of the syn-diaxial diastereomer. The

carbocation delocalization energies associated with epoxide ring opening

of the methylated bay-region tetrahydroepoxide isomers of chrysene are also given.

Key words: Methylation – Chrysene - Reactivity - Conformer - Stability -.

Carcinogenesis

INTRODUCTION
Methylation is known to significantly influence the tumorigenic/carcino-
genie potency of polycyclic aromatic hydrocarbons (PAH) [ 1,2] . In
particular, substitution of a methyl group in the bay-region, at a molecular
position that is not on the terminal six-membered ring, generally yields
either the most, or one of the most tumorigenically active monomethyl
isomers [ 3-71. For example, 5-methyl chrysene (5-MeC, Fig. 1) the bay-
region methylated isomer of chrysene has been found to be the most active
of the monomethyl chrysene isomers [8-lo]. Investigations [4,11-181,
similar to those implicating the bay-region diol-epoxide of benzo [a] pyrene
[19]and of other PAH [20] as ultimate carcinogens, have implicated the
0009-2797/85/$03.30
@ 1985 Elsevier Scientific Publishers Ireland Ltd.
Printed and Published in Ireland

314

12 I

IO”0 0 :

,900 4

@
7 6 5

Fig. 1. 5-Methyl chrysene.

1,2diol-3,4-epoxide of 5-MeC(bay-region diol-epoxide of 5-MeC) as an
ultimate tumorigen on mouse skin. 5-MeC is of particular interest since the
molecule has two different bay-regions and consequently can be metabolized
to two different bay-region diol-epoxides. Recent evidence [18] suggests
that the major ultimate carcinogen of 5-MeC has the methyl group and the
epoxide in the same bay-region, Fig. 2a. The other diol-epoxide (Fig. 2b)
apparently plays a less active role in 5-MeC induced tumorigenesis.
The presence of a methyl group can affect the tumorigenic activity of
PAH in a number of different ways. It can alter the distribution of meta-
bolites [21-231 and in particular, the amounts of detoxification products
[24,25]. It can induce a diaxial conformation of an adjacent diol [21,26]
and thereby significantly inhibit metabolism to tumorigenic electrophilic
intermediates [21,27]. The methyl group might also affect the rate of DNA
repair after covalent interaction between the PAH and DNA, as well as
affecting the rate of such interaction. It is of interest that examination of
the levels of 5-MeC metabolites in mouse skin, of the DNA adducts formed,
and the persistence of these adducts, indicates that the difference in the
rate of adducts formed with the two different bay-region diol-epoxides of
5-MeC is apparently due to differences in the rates of reaction of the 5-MeC

diol-epoxides with DNA [ 181.

Different reaction rates of PAH diol-epoxide isomers can depend upon a

number of factors. An examination of two of these factors in connection

with the methylated chrysenes forms the subject matter of the present paper.

Since calculated ease of epoxide ring opening has apparently yielded a

useful index of carcinogenic potency [28], particularly among isomeric

metabolites of the same parent PAH [29], previous investigations have
attempted to correlate the carcinogenic potency of a set of methylated

(a) \ Cb)

Fig. 2.(a) 1,2-Diol-3,4-epoxide of 5-MeC. (b) 1,2-Dial-3,4-epoxide of ll-MeC (7,8-
dial-9,10-epoxide of 5-MeC).

315

PAH’s with the effect of a methyl group on ease of epoxide ring opening

[30-331. The present paper will present results of such calculations per-

formed for methylated chrysenes.
The other factor affecting diol-epoxide reactivity that will be addressed is

the effect of a bay-region methyl group on the relative stability of different
diol-epoxide conformers. It has been shown [34] that steric crowding due

to the presence of a bay-region methyl group in benz[a]anthracene and
cyclopenta[a]phenanthren-17-one is expected to destabilize the commonly
observed diol-epoxide conformers in much the same manner that is expected

for benzo [c] phenanthrene [ 35,361. Calculations discussed in the present
paper demonstrate a similar effect for 5MeC. The enhanced tumorigenic

activity of the syn-diastereomer of the ‘fjord’ or bay-region diol epoxide
of benzo[c]phenanthrene is apparently due to such steric crowding and

consequent destabilization of the syn-diaxial conformer [ 371.
The previous [34] as well as the present computational results can be

summarized as follows: if calculated ease of epoxide ring opening is taken

as the measure of molecular reactivity, the bay-region methylated isomer

is not found to be the most reactive isomer for the three PAH studied. It
is, however, found to be among the isomers that are calculated to be the

most reactive. Furthermore, conformers of the PAH diol-epoxide diastere-
omers commonly encountered in the absence of significant molecular

steric crowding, e.g., the syndiaxial (Fig. 3c) and antidiequatorial (Fig. 3d)

conformers, are found to be destabilized with respect to the syn-diequatorial

(Fig. 3e) and anti-diaxial (Fig. 3f) conformers when the methyl group and the epoxide are in the same bay-region.

For 5-MeC, these results suggest that the observed enhanced in vivo
binding to DNA of the 1,2-diol-3,4-epoxide, compared with the binding

of the 7,8diol-9,10-epoxide [18], is not solely a consequence of relative

ease of carbocation formation but could be partially due to conformer destabilization.

METHODS

All molecular energies have been determined by a two-step procedure

that has been previously described [34]. Molecular geometries are deter-

mined by the Allinger MMPI force-field program [ 381. This program accepts
as input, an idealized initial or starting geometry. The molecular structure

is then varied until a minimum or stationary point in the force-field energy

is found. Relevant PAH geometries determined in this manner, parallel

X-ray determined structures closely [36,39]. Calculations are performed for

tetrahydroepoxides and not for diol-epoxides, e.g., diol-epoxides are simu-

lated by replacing the hydroxyl groups with hydrogen atoms. It is known

that the presence of the hydroxyl groups plays an important role in the

interaction between the antidiequatorial diastereomer of the bay-region
diol-epoxide of benzo[a] pyrene and DNA [40,41]. It is not known whether

the hydroxyl groups are important in determining the relative stability of

316

Fig. 3. (a) Perp Conformer; 3,4-epoxide of chrysene. (b) Nonperp conformer; 3,4-epoxide

of chrysene. (c) Syn-diaxial diastereomer of the 1,2-diol-3,4-epoxide of chrysene (CHDE).
(d) Anti-diequatorial diastereomer of CHDE. (e) Syn-diequatorial diastereomer of CHDE.

(f) Anti-diaxial diastereomer of CHDE.

317

conformers of the diol-epoxide diastereomers. Furthermore, one might

expect hydroxyl group orientation to depend upon the molecular environ-

ment. In any event, calculations that treat the effect of the hydroxyl

groups on the relative molecular energies to the required accuracy, are

beyond our present computational capabilities. The hydroxyl groups are
therefore neglected in the present calculations and we focus primarily on

energetic differences arising from interactions between the epoxide and the

adjacent bay-region methyl group. The terminology ‘diol-epoxide’ is used, however, in the title and in the major body of the text. This is done since

the calculations on the tetrahydroepoxides have been primarily motivated by

what one can infer from them concerning relative diol-epoxide stability.

The terminology ‘tetrahydroepoxide’ and ‘epoxide’ are used interchangeably in the text.

Force-field molecular geometries have also been determined for the
carbocations that arise from epoxide ring opening. The force-field program

has not been explicitly parametrized to yield carbocation geometries, how-

ever, the results that have been obtained with use of this program are found

to parallel results obtained from the Gaussian-70 molecular orbital program
(STO-3G basis) [42] .

Even though the force-field program determines accurate molecular

geometries, the difference in force-field energies of closely related mole-

cular structures is not always a reliable measure of relative molecular

stability [43]. To determine a more reliable measure, the force-field opti-

mized molecular geometries have been used as input to a molecular orbital
program to determine total energies. In the present paper the INDO [44]

semi-empirical and Gaussian-70 (STO-3G basis) [ 451 ab-initio molecular
orbital programs have been used. The difference in energy between the

epoxide and resultant carbocation, namely, the delocalization energy, should provide a relative measure of ease of epoxide ring-opening.

RESULTS
Figures 3a and 3b show the two different bay-region epoxideconformer
geometries considered. Figure 3a, which we will call the perp epoxide
conformer, has the plane of the epoxide ring approximately perpendicular
to a plane through the aromatic portion of the molecule. This is the epoxide
ring geometry adopted by the two commonly encountered diol-epoxide
diastereomers, namely the syn-diaxial (Fig. 3c) andanti-diequatorial
(Fig. 3d), bay-region diol-epoxides. Figure 3b, the conformer which we will
call the nonperp epoxide conformer, has the plane of the epoxide ring tilted
away from the perpendicular to the aromatic plane. This is the epoxide
ring geometry adopted by the two uncommonly encountered diol-epoxide
diastereomers, namelythe syn-diequatorial (Fig. 3e) and anti-diaxial (Fig. 3f)
diol-epoxides.
Table I lists the total molecular energies of the two epoxide conformers
for each of their monomethyl isomers. Isomers for which the particular

318

TABLE I
TOTAL MOLECULAR ENERGY OF CONFORMERS OF THE BAY-REGION
EPOXIDE(3,4-EPOXIDE) OF CHRYSENE AND METHYLATED CHRYSENE
P refers to the perp conformer of thebay-region epoxide (Fig. 3a);NP refers to the non-
perp conformer (Fig. 3b); 5-P, for example, refers to the perp 3,4-epoxide (bay-region
epoxide) conformer of 5-MeC (Fig. 4a).Total molecular energies in thetable are given
in Atomic Units (Au.). Energy differences between the two conformers are given in
electron volts (eV). This energy difference is placed next to the more stable conformer
entry.
INDO Gaussian-70 (STO-3G)
P -151.87925 (Au.) (0.065eV) -755.28913 (Au.)
NP -151.87685 -755.28948 (0.0095eV.)
6-P -160.32106 (0.079) -793.86879
6-NP -160.31816 -793.86940 (0.017)
7-P -160.32273 (0.12) -793.86994 (0.022)
7 -NP -160.31818 -793.86913
8-P -160.32353 (0.10) -793.87293
8-NP -160.31977 -793.87306 (0.0035)
9-P -160.32246 (0.090) -793.87203
9-NP -160.31915 -793.87281 (0.021)
5-P -160.30854 -793.85405
5-NP -160.31009 (0.042) -793.85554 (0.041)
10-P -160.30919 (0.10) -793.85423 (0.018)
lo-NP -160.30546 -793.85357
11-P -160.30848 (0.11) -793.85440 (0.049)
ll-NP -160.30437 -793.85260

methyl group substitution induces steric crowding have been placed at the

end of the list. Calculations have not been performed for the isomer with
the methyl group at the peri or 12 position. For this isomer, realistic simula-

tion of the interaction between the methyl and hydroxyl groups is not

achieved by replacing the hydroxyl groups with hydrogen atoms. One

generally expects peri methyl substitution to inhibit carcinogenesis induced

by PAH [4,21]. The energy difference between the two epoxide conformers

is also listed in Table I for each of the monomethyl isomers as well as for

the unsubstituted epoxide.

Even though these energy differences are small it should be noted that
they exhibit systematic behavior. All of the monomethyl INDO conformer

energies, except the ones associated with 5-MeC, show the epoxide with
its plane approximately perpendicular to the aromatic plane, namely, the

per-p conformer, to be lower in energy, by about 2 kcal/mol, than the non-

319

perp conformer (see Table I). The only exception to this ordering of energies

occurs for the conformers of 5MeC that have both the methyl groups and

epoxide in the same bay-region. For these conformers the energies are
reversed, namely the nonperp conformer, is calculated to be the stable

conformer. This is the conformer with ring geometry similar to the ring

geometries of the syn-diequatorial and anti-diaxial diastereomers of the

1,2-diol-3,4 epoxide of 5-MeC (see Figs. 3b, 3e and 3f). It is of interest that

such energy reversal of epoxide conformer stability is not observed for the

other isomers that have a methyl group in a bay-region, i.e., at positions 10 and 11.

The effect of bay-region methylation on molecular planarity is shown

in Fig. 4. This figure includes an edge on view looking into the particular

molecular bay-region which is sterically crowded as a result of methylation.

In the absence of the steric crowding between the bay-region methyl group

and adjacent epoxide, the per-p conformer (Fig. 3a) is apparently stabilized

since certain bonds on the angular benzo-ring of this conformer adopt a
staggered as opposed to the eclipsed geometry of the nonperp conformer

(Fig. 3b) [34,35,46,47]. When the methyl group and the epoxide are in the
same bay region (Fig. 4a) steric crowding between the hydrogen atoms of

the methyl group and the hydrogen atom coupledto position4 destabilizes
the perp (Fig. 4a) with respect to the nonperp conformer (Fig. 4b). Energies
obtained with the Gaussian-70 (STO-3G) ab-initio program exhibit a similar
systematic trend as do the INDO energies, however, the energy differences
are smaller. The Gaussian-70 results do not, however, indicate a conformer
preference for any of the conformers except for those of 5-MeC.
Table II lists the calculated carbocation delocalization energies for the
monomethyl isomers and for the unsubstituted molecule. Entries in the
table have been listed in order of decreasing Gaussian-70 delocalization
energy. The epoxide conformer calculated to be the most stable has been
used in each of the delocalization energy calculations. One sees that methy-
lation generally enhances carbocation stability, as expected. The isomer
with the methyl group and the epoxide in the same bay-region (3,4-tetrahy-
droepoxide of 5-MeC) is calculated to be more reactive than all of the
isomers except one, namely, the 3,4-tetrahydroepoxide of ll-MeC (9,10-
tetrahydroepoxide of 5-MeC) which has the methyl group and epoxide
in separate bay-regions. The present calculations of relative carbocation
stability do not, therefore, rationalize the predominance of DNA adducts
formed with the 1,2diol-3,4-epoxide of 5-MeC compared with those formed
upon interaction with the 7,8diol-9,10-epoxide of 5-MeC [18]. The calcu-
lations are, however, consistent with the observation that both isomers are
reactive with mouse skin and calf thymusDNA.
The significant delocalization energy gained by methylation at position 11
(Table II) is consistent with the value of the lowest unoccupied carbocation
molecular orbital (LUMO) amplitude at this position [32,33]. The value at
this position is greater than that at any other position of potential methylation.
In the simple Huckel approximation the value of this amplitude at position

320

(b)
(d)

Fig. 4. Sterically crowded epoxides. (a) 3,4-Epoxide of 5-MeC; Perp conformer, (b)

3,4-Epoxide of 5-MeC; Nonperp conformer. (c) 3,4-Epoxide of ll-MeC; Perp conformer.

(d) 3,4-Epoxide of lo-MeC; Perp conformer.

5 is exactly zero. Methylation at position 5, e.g., in the same bay-region as

the epoxide stabilizes the resulting carbocation in a way that is different

from methylation at ‘remote’ positions. The carbocation delocalization

energy for this case depends upon the direct ‘through-space’ interaction of

the methyl group with the carbocation as well as with the epoxide. One

expects it to depend sensitively upon the local bay-region geometry of the

carbocation and diol-epoxide. It is of interest that calculations of the effect

of bay-region methylation on the carbocation stabilization of benz[a]-

anthracene and cyclopenta[a] phenanthren-17-one have yielded similar

321

TABLE II
REACTIVITY OF 3,4-EPOXIDE OF CHRYSENE AND METHYLATED CHRYSENE
Carbocation delocalization energy (Au.)
INDO Gaussian-70 (STO-3G)
Chrysene 0.3841 0.4107
11 -MeC 0.3970 0.4182
5-MeC 0.3915 0.4156
9-MeC 0.3851 0.4135
lo-MeC 0.3877 0.4129
8-MeC 0.3837 0.4123
7 -MeC 0.3835 0.4118
6-MeC 0.3847 0.4117

results [34]. These calculations show that methyl substitution at the posi-

tion of greatest LUMO amplitude yields the greatest carbocation stabilization

of all the monomethyl substitutions. In each of these two other calculations,

bay-region methylation provided the second largest carbocation stabilization energy.

DISCUSSION

The present calculations indicate that, due to steric crowding in the bay-

region of the molecule, both syndiaxial and antidiequatorial conformers

of the 1,2-diol-3,4-epoxide of 5-MeC found a quasi-diequatorial conformation

the syndiequatorial and antidiaxial conformers. Such destabilization had

been previously found for bay-region methylation of benz[a] anthracene and

cyclopenta[a] phenanthrene-17-one [34]. Calculations have shown a similar

effect, albeit more pronounced, for benzo [c] phenanthrene [ 351. For benzo-
[c] phenanthrene,in vitro NMR studies have, however, found the syn as well
as the anti-diastereomerto have hydroxylgroups with a quasi-diequatorial
orientation[48] . A recent NMR study of the synthesizedanti-diastereomer
of the 1,2diol-3,4-epoxideof 5-MeC found a quasidiequatorialconformation
of the diol [49].One expects the relative stabilityof diol-epoxideas well as
diol conformers[ 501 to depend sensitivelyupon the molecularenvironment.

A recent X-ray crystallographic study of the syndiastereomer of the bay-

region diol-epoxide of benzo [ alpyrene has revealed the syn-diequatorial

conformer [ 511. This molecule should be subject to no significant intra-

molecular steric crowding.
These experimental observations suggest that the relative stability of syn
conformers depends more sensitively than anti conformers upon the intra-
and/or intermolecular environment. It would be of interest to determine
hydroxyl group orientation, under physiological conditions, of the syn-
diastereomer of the 1,2diol-3,4-epoxide of 5-MeC. If indeed, the diol
adopts a diequatorial conformation, this might partially account for the

322

enhanced in vivo DNA binding of this metabolite, compared with the meta-

bolite having the methyl group and epoxide in two separate bay-regions

[181.
Comparable numbers of adducts involving both syn and anti-dihydro-diol-

epoxides in the binding of 7,12dimethylbenz[a]anthracene to DNA in

mouse skin embryo cell cultures have recently been observed [52 J . The

present calculations also suggest that destabilization of the syndiaxial
conformer by a bay-region methyl group might be partially responsible for this observation.

REFERENCES

1 J.C. Arcos and M.F. Argus, in: Chemical Induction of Cancer, Academic Press, New

York and London, 1974, Sect. 5.1.1.2.
2 D.M. Jerina and J.W. Daly, Oxidation at carbon, in: D.V. Parke and R.L. Smith
(Eds.), Drug Metabolism, Taylor and Francis Ltd., London, 1977, pp. 13-32.
3 M.M. Coombs, T.S. Bhatt and C.J. Croft, Correlation between carcinogenicity and
chemical structure in cyclopenta(a)phenanthrenes, Cancer Res., 33 (1973) 832.
4 S.S. Hecht, S. Amin, A. Rivenson and D. Hoffman, Tumor initiating activity of 5,11-
dimethylchrysene and the structural requirements favoring carcinogenicity of methy-
lated polynuclear aromatic hydrocarbons, Cancer Lett., 8 (1979) 65.
5 R.P. Iyer, J.W. Lyga, J.A. Secrest III, G.H. Daub and T.J. Slaga, Comparative tumor-
initiating activity of methylated benzo(a)pyrene derivatives on mouse skin, Cancer
Res., 40 (1980) 1073.
6 P.G. Wislocki, K.M. Fiorentini, P.P. Fu, S.K. Yang and A.Y.H. Lu, Tumor-initiating
ability of the twelve monomethylbenz(a)anthracenes, Carcinogenesis, 3 (1982) 2 15.
7 J. DiGiovanni, L. Diamond, R.G. Harvey and T.J. Slaga, Enhancement of the skin
tumor-initiating activity of polycyclic aromatic hydrocarbons by methyl-substitution
at non-benzo ‘bay-region’ positions, Carcinogenesis, 4 (1983) 403.
8 D. Hoffman, W.E. Bondinell and E.L. Wynder, Carcinogenicity of methylchrysenes,
Science, 183 (197-l) 215.

9 S.S. Hecht, W.E. Bondinell and D. Hoffmann, Chrysene and methylchrysenes: pres-
ence in tobacco smoke and carcinogenicity, J. Natl. Cancer Inst., 53 (1974) 1121.

10 S.S. Hecht, M. Loy and D. Hoffmann, On the structure and carcinogenicity of the

methylchrysenes, in: R.I. Freudenthal and P.W. Jones (Eds.), Carcinogenesis, Vol. 1,

Polynuclear Aromatic Hydrocarbons, Metabolism and Carcinogenesis, Raven Press, New York, 1976, pp. 325-340.

11 S.S. Hecht, N. Hirota, M. Loy and D. Hoffmann, Tumor-initiating activity of fluori-

nated 5-methylchrysenes, Cancer Res., 38 (1978) 1694.
12 S.S. Hecht, M. Loy, R. Mazzarese and D. Hoffman, Synthesis and mutagenicity of

modified chrysenes related to the carcinogen, 5-methylchrysene, J. Med. Chem., 21
(1978) 38.

13 S.S. Hecht, E. LaVoie, R. Mazzarese, S. Amin, V. Bedenko and D. Hoffmann, 1,2-
Dihydro-1,2-dihydroxy-5-methylchrysene, a major activated metabolite of the environ-
mental carcinogen, 5-methylchrysene, Cancer Res., 38 (1978) 2191.
14 S.S. Hecht, E. LaVoie, R. Mazzarese, N. Hirota, T. Ohmori and D. Hoffmann,
Comparative mutagenicity, tumor initiating activity, carcinogenicity and in vitro
metabolism of fluorinated 5-methylchrysenes, J. Natl. Cancer Inst., 63 (1979) 855.
15 S.S. Hecht, R. Mazzarese, S. Amin, E. LaVoie and D. Hoffmann, On the metabolic
activation of 5-methylchrysene, in: P.W. Jones and P. Leber (Eds.), Polynuclear
Aromatic Hydrocarbons, Ann Arbor Science Publishers, Inc., Ann Arbor, 1979,
pp. 733-751.

323
16 S.S. Hecht, A. Rivenson and D. Hoffmann, Tumor initiating activity of dihydrodiols
formed metabolically from 5-methylchrysene, Cancer Res., 10 (1980)1396.
17 A.A. Melikian, E.J. LaVoie, S.S. Hecht and D. Hoffmann, Influenceof a bay-region
methyl group on formation of 5-methylchrysene dihydrodiol epoxide: DNA adducts
18 in mouse skin, Cancer Res., 42 (1982)1239.
A.A. Melikian, E.J. LaVoie, S.S. Hecht and D. Hoffmann, 5-Methylchrysene meta-
bolism in mouse epidermis in vivo, dial-epoxide-DNA adduct persistence, and diol
epoxide reactivity with DNA as potential factors influencing the predominance of
5-methylchrysene-1,2-diol-3,4-epoxide-DNA adducts in mouse epidermis, Carcino-
genesis, 4 (1983) 843.
19 D.H. Phillips, Fifty years of benzo(a)pyrene, Science, 303 (1983) 468.
20 A.H. Conney, Induction of microsomal enzymes by foreign chemicals and carcino-
genesis by polycyc!ic aromatic hydrocarbons: G.H.A. Clowes Memorial Lecture,
Cancer Res., 42 (1982) 4875.
21 S.K. Yang, M.W. Chou and P.P. Fu, Metabolic and structural requirements for the
carcinogenic potencies of unsubstituted and methyl-substituted polycyclic aromatic
hydrocarbons, in: B. Pullman, P.O.P. Ts’o and H. Gelboin (Eds.), Carcinogenesis:
Fundamental Mechanisms and Environmental Effects, D. Reidel Publishing Co.,
22 Dordrecht, Holland, 1980, pp. 113-156.
M.W. Chou, S.K. Yang, W. Sydor and C.S. Yang, Metabolism of 7,12-dimethylbenz-
(a)anthracene and 7-hydroxymethyl-12-methylbenz(a)anthracene by rat liver nuclei
and microsomes, Cancer Res., 41 (1981) 559.
23 P.P. Fu, M.W. Chou and S.K. Yang, In vitro metabolism of 12-methylbenz(a)anthra-
cene: effect of the methyl group on the stereochemistry of the 5,6-dihydrodiol
24 metabolite, Biochem. Biophys. Res. Commun., 106 (1982) 940.
M.S. Newman and R.F. Cunico, The synthesis of 5-hydroxymethyl-, 5-acetoxymethyl-
and 5-methylmercapto-7,12-dimethylbenz(a)anthracenes, and of 5,7,12-trimethyl-
benz(a)anthracene, J. Med. Chem., 15 (1972) 323.
25 E.J. LaVoie, V. Bedenko, L. Tulley-Freiler and D. Hoffmann, Tumor initiating activ-
ity and metabolism of polymethylated phenanthrenes, Cancer Res., 42 (1982) 4045.
26 B. Tierney, P. Burden, A. Hewer, 0. Ribeiro and C. Walsh, High-performance liquid
chromatography of isomeric dihydrodiols of polycyclic hydrocarbons, J. Chromatogr.,

I76 (1979) 329.
27 K.L. Hamernik, P. Chiu, M.W. Chou, P.P. Fu and S.K. Yang, Metabolic activation of

6-methylbenzo(a)pyrene, in: M. Cooke and A.J. Dennis (Eds.), Polynuclear Aromatic

Hydrocarbons, Formation, Metabolism and Measurement, Battelle Press, Columbus, Ohio, 1983, pp. 583-597.

28 D.M. Jerina, R.E. Lehr, H. Yagi, 0. Hernandez, P.M. Dansette, P.G. Wislocki, A.W.

Wood, R.L. Chang, W. Levin and A.H. Conney, Mutagenicity of benzo(a)pyrene

derivatives and the description of a quantum mechanical model which predicts the

ease of carbonium formation from diol-epoxides, in: F. deSerres et al. (Eds.), In

Vitro Metabolic Activation and Mutagenesis Testing, Elsevier Press, Amsterdam, 1976, pp. 159-177.

29 D.M. Jerina, J.M. Sayer, D.R. Thakker, H. Yagi, W. Levin, A.W. Wood and A.H.

Conney, Carcinogenicity of polycyclic aromatic hydrocarbons: the bay-region theory,

in: B. Pullman, P.O.P. Ts’o and H. Gelboin (Eds.), Carcinogenesis: Fundamental

Mechanisms and Environmental Effects, D. Reidel Publishing Co., Dordrecht,

30

Holland, 1980, pp. l-12.

G.D. Berger and P.G. Seybold, Substituent effects in chemical carcinogenesis:

chrysene and its methyl derivatives, Int. J. Quant. Chem. Quant. Biol. Symp., 6

(1979) 305.

31 M.T. Poulsen and G.H. Loew, Quantum chemical studies of methyl and fluoro analogs

of chrysene: metabolic activation and correlation with carcinogenic activity, Cancer
Biochem. Biophys., 55 (1981) 81.

324

32 B.D. Silverman and J.P. Lowe, Carcinogenicity of methylated hydrocarbons: effect

of methylation on the calculated dial-epoxide reactivity, Cancer Biochem. Biophys., 6 (1982) 89.

33 B.D. Silverman and J.P. Lowe, Diol-epoxide reactivity of methylated polycyclic
aromatic hydrocarbons (PAH): ranking the reactivity of the positional monomethyl
isomers, in: M. Cooke, A.J. Dennis and G.L. Fisher (Eds.), Polynuclear Aromatic
Hydrocarbons. Physical and Biological Chemistry, Battelle Press, Columbus, Ohio,
1982, pp. 713-753.
31 B.D. Silverman, Dial-epoxide conformer stability of polycyclic aromatic hydro-
carbons (PAH): the effect of a hay-region methyl group, in: M. Cooke and A.J. Dennis
(Eds.), Polynuclear Aromatic Hydrocarbons. Formation, Metabolism and Measure-
ment, Battelle Press, Columbus, Ohio, 1983, pp. 1113-1122.
35 B.D. Silverman, Steric crowding and conformer stability of the dial-epoxides of
henzo(c)phenanthrene, Cancer Biochem. Biophys., 6 (1982) 23.
36 B.D. Silverman and S.J. LaPlaca, Bay region epoxides of benzo[c]phenanthrene:
force-field molecular structures, J. Chem. Sot., Perkin II (1982) 415.
37 W. Levin, A.W. Wood, R.L. Chang, Y. Ittah, M. Croisy-Delcey, H. Yagi, D.M. Jerina
and A.H. Conney, Exceptionally high tumor-initiating activity of benzo(c)phenan-
threne bay-region dial-epoxides on mouse skin, Cancer Res.. 40 (1980) 3910.
38 N.L. Allinger and J.T. Sprague, Calculation of the structure of hydrocarbons contain-
ing delocalized electronic systems by the molecular mechanics method, Quantum
Chemistry Program Exchange, Indiana University, Program No. 318, J. Am. Chem.
Sot., 95 (1973) 3893.
39 B.D. Silverman, Determination of the molecular structure of the bay-region epoxides
of benzo[a]pyrene by a force-field calculation, Cancer Biochem. Biophys., 6 (1983)
131.
40 N.E. Geacintov, A.G. Gagliano, V. Ibanez and R.G. Harvey, Spectroscopic characteri-
zations and comparisons of the structures of the covalent adducts derived from the
reactions of 7,8-dihydroxy-7,8,9,1O-tetrahydrobenzo(a)pyrene-9.1O-oxide, and the
9,10-epoxides of 7,8,9,10tetrahydrobenzo(a)pyrene and 9,10,11,12tetrahydro-
benzo(e)pyrene with DNA, Carcinogenesis, 3 (1982) 247.
41 T. Kinoshita, H.M. Lee, R.G. Harvey and A.M. Jeffrey, Structures of covalent adducts
derived from the reactions of the 9,10-epoxides of 7,8,9,10tetrahydrobenzo(a)-
pyrene and 9,10,11,12tetrahydrobenzo(e)pyrene with DNA, Carcinogenesis, 3
(1982) 255.
42 W.J. Hehr, Carbonium ions: structural and energetic investigations, in: H.F. Schaefer
III (Ed.), Applications of Electronic Structure Theory, Plenum Press, New York and
London, 1977, pp. 217-331.
D.A. Dougherty and K. Mislow, A combination of empirical force-field and extended
Huckel molecular orbital calculations as a computational approach to conformational
analysis, J. Am. Chem. Sot., 104 (1979)1401.
44 J.A. Pople and D.L. Beveridge, Approximate Molecular Orbital Theory, Quantum
Chemistry Program Exchange, Indiana University, Program No. 141, McGraw Hill
Book Co., New York, 1979.
45 W.J. Hehre, R.F. Stewart and J.A. Pople, Self-consistent molecular-orbital methods. I.
Use of Gaussian expansions of Slater type atomic orbitals, Quantum Chemistry
Program Exchange, Indiana University, ProgramNo. 236, J. Chem. Phys., 51 (1969)
2657.
46 C.Y. Yeh, P.P. Fu, F.A. Beland and R.G. Harvey, Application of CNDO/2 theoretical
calculations to interpretation of the chemical reactivity and biological activity of the
syn and anti diolepoxides of benzo(a)pyrene, Bioorg. Chem., 7 (1978) 497.
47 0. Kikuchi, A.J. Hopfinger and G. Klopman, Electronic structure and reactivity of
four stereoisomers of benzo(a)pyrene, 7,8-diol-9,10-epoxide, Cancer Biochem. Bio-
phys., 4 (1979) 1.

325

48 J.M. Sayer, H. Yagi, M. Croisy-Delcey and D.M. Jerina, Novel bay-region diol
epoxides from benzo(c)phenanthrene, J. Am. Chem. Sot., 103 (1981) 4970.

49 J. Pataki, H. Lee and R.G. Harvey, Carcinogenic metabolites of 5-methylchrysene,
50 Carcinogenesis, 4 (1983) 399. of dihydrodiol and other derivativesof benz(c)-
R.E. Lehr and S. Kumar, Synthesis
acridine,J. Org. Chem., 46 (1981)3675.
51 S. Neidle and S.D. Cutbush, X-ray crystallographic analysis of the (+)-7p,8ol-dihy-
droxy-9p,lOp-epoxy-7,8,10-tetrahydrobenzo(a)pyrene: molecular structure ofa
‘syn’diol epoxide, Carcinogenesis, 4 (1983) 415.
52 J.T. Sawicki, R.C. Moschel and A. Dipple, Involvement of both syn- and anti-dihydro-
diol-epoxides in the binding of 7,12-dimethylbenz(a)anthracene to DNA in mouse
embryo cell cultures., Cancer Res., 43 (1983)3212. BAY-3827