Synthesis of trans-3,4-dihydroxy-3,4-dihydrophenanthro[3,2-b]-[1]benzothiophene, A potentially carcinogenic metabolite of sulfur heterocycle phenanthro[3,2-b][1]benzothiophene

Article abstract of DOI:10.1039/b100345n

The present study describes the synthesis of trans-3,4-dihydroxy-3,4-dihydrophenanthro[3,2-b][1]benzothiophene (3), which is a potential proximate carcinogen of the environmentally occurring potent carcinogen phenanthro-[3,2-b][1]benzothiophene (1). Three approaches were investigated for the synthesis of 3. In the first approach, the diarylalkene 9, which was prepared by the Wittig reaction of ylide 7 and aldehyde 8, was subjected to an oxidative photocyclization reaction to produce a mixture of compounds from which 5, a synthetic precursor to 3, was obtained only in trace amounts. The major cyclized product was 10. The second approach entailed the Suzuki cross-coupling reaction of 2-(dihydroxyboryl)-5-methoxybenzaldehyde (12) with 3-bromodibenzothiophene (11) to produce 13 in 92% yield. The aldehyde function of 13 was elongated with trimethylsulfonium iodide in the presence of KOH to generate the ethylene oxide 14, which, after methanesulfonic acid treatment, produced a 1 : 1 mixture of 3-methoxyphenanthro[3,2-b][1]benzothiophene (6) and 3-methoxyphenanthro[1,2-b][1]benzothiophene (15). Only the undesired compound 15 could be isolated in pure form by extracting the mixture with hot ethanol, leaving behind a 7 : 3 mixture of 6 and 15. In the third approach, the Wittig reaction of 7 with 2-bromo-5-methoxybenzaldehyde (16) produced 17 predominantly as the Z-isomer in quantitative yield. Cyclodehydrobromination of 17 with KOH-quinoline at elevated temperature produced a mixture from which 6 and 3-methoxyphenanthro[3,4-b][1]benzothiophene (18) were easily separated by column chromatography in 23 and 51% yields, respectively. The intermediate 6 was conveniently processed to 3 following the reaction sequence: methoxy phenol→o-quinone→dihydrodiol. A relatively polar dihydrodiol obtained by similar processing of the mixture of 6 and 15 was identified as 3.

Full text of DOI:10.1039/b100345n

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis of trans-3,4-dihydroxy-3,4-dihydrophenanthro[3,2-b]-
[1]benzothiophene, a potentially carcinogenic metabolite of
sulfur heterocycle phenanthro[3,2-b][1]benzothiophene
1
Subodh Kumar*
Environmental Toxicology and Chemistry Laboratory, Great Lakes Center,
State University of New York College at Buffalo, 1300 Elmwood Avenue, Buffalo,
NY 14222, USA
Received (in Cambridge, UK) 8th January 2001, Accepted 28th February 2001
First published as an Advance Article on the web 30th March 2001
The present study describes the synthesis of trans-3,4-dihydroxy-3,4-dihydrophenanthro[3,2-b][1]benzothiophene
(3), which is a potential proximate carcinogen of the environmentally occurring potent carcinogen phenanthro-
[3,2-b][1]benzothiophene (1). Three approaches were investigated for the synthesis of 3. In the first approach, the
diarylalkene 9, which was prepared by the Wittig reaction of ylide 7 and aldehyde 8, was subjected to an oxidative
photocyclization reaction to produce a mixture of compounds from which 5, a synthetic precursor to 3, was obtained
only in trace amounts. The major cyclized product was 10. The second approach entailed the Suzuki cross-coupling
reaction of 2-(dihydroxyboryl)-5-methoxybenzaldehyde (12) with 3-bromodibenzothiophene (11) to produce 13
in 92% yield. The aldehyde function of 13 was elongated with trimethylsulfonium iodide in the presence of KOH
to generate the ethylene oxide 14, which, after methanesulfonic acid treatment, produced a 1 : 1 mixture of
3-methoxyphenanthro[3,2-b][1]benzothiophene (6) and 3-methoxyphenanthro[1,2-b][1]benzothiophene (15). Only
the undesired compound 15 could be isolated in pure form by extracting the mixture with hot ethanol, leaving behind
a 7 : 3 mixture of 6 and 15. In the third approach, the Wittig reaction of 7 with 2-bromo-5-methoxybenzaldehyde (16)
produced 17 predominantly as the Z-isomer in quantitative yield. Cyclodehydrobromination of 17 with KOH–quino-
line at elevated temperature produced a mixture from which 6 and 3-methoxyphenanthro[3,4-b][1]benzothiophene
(18) were easily separated by column chromatography in 23 and 51% yields, respectively. The intermediate 6 was
conveniently processed to 3 following the reaction sequence: methoxy phenol→o-quinone→dihydrodiol. A relatively
polar dihydrodiol obtained by similar processing of the mixture of 6 and 15 was identified as 3.
Introduction
For the past two decades, numerous studies have indicated that
“bay-region”1 diol epoxides and their dihydrodiol precursors
are the major carcinogenic metabolites of a number of poly-
cyclic aromatic hydrocarbons (PAHs).2–4 The importance of
these metabolites in cancer induction has stimulated strong
interest in their synthesis and their chemical and biological
properties.4 The analogous polycyclic sulfur heterocycles (thia-
PAHs) are also common environmental contaminants5 and
many have been shown to possess high mutagenic/carcinogenic
activity.5–7 Although a number of thia-PAHs exhibit mutagenic
and/or carcinogenic activity, not much attention has been paid
to determining the nature of the intermediate(s) involved in
the metabolic activation of such thia-PAHs. Thia-PAHs are
metabolized8,9 by cytochrome P-450 catalyzed oxidation
reactions to sulfoxides and sulfones, as well as to dihydrodiols
analogous to those produced by PAHs.3 At the same time, a
number of thia-PAHs that are capable of being metabolized to
bay-region diol epoxides have been found to be considerably
more carcinogenic than their carbocyclic analogues (PAHs).6,7
These observations are consistent with the speculation that the
bay-region concept of metabolic activation of PAHs10,11 may be
extended to thia-PAHs because Hückel MO calculations have
predicted bay-region diol epoxides of thia-PAHs to be more
electrophilic than their carbocyclic analogues.12 Recently,
synthesis and subsequent biological studies of dihydrodiols of
a weak carcinogen, naphtho[1,2-b][1]benzothiophene (NBT, a
thia-analogue of chrysene) have demonstrated that, among two
major dihydrodiol metabolites of NBT,8 trans-3,4-dihydroxy-
3,4-dihydronaphtho[1,2-b][1]benzothiophene (NBT 3,4-diol), a
metabolic precursor to the bay-region diol epoxide of NBT,
is significantly more mutagenic than the isomeric trans-1,2-
dihydroxy-1,2-dihydronaphtho[1,2-b][1]benzothiophene (NBT
1,2-diol).13 However, in contrast to a significant difference
noted between the mutagenicity of chrysene and its 1,2-diol,14
no significant difference was noted between the mutagenicity
of NBT and NBT 3,4-diol. These findings suggest that the
weak carcinogenicity of NBT may be attributed to its metab-
olism either to the weakly carcinogenic NBT 3,4-diol or to
an unknown minor metabolite(s) of high mutagenicity/
carcinogenicity.
1018
J. Chem. Soc., Perkin Trans. 1, 2001, 1018–1023
This journal is © The Royal Society of Chemistry 2001
DOI: 10.1039/b100345n

View Article Online
As a part of our research goal to identify the mutagenic and/
or carcinogenic metabolites of thia-PAHs, and to investigate
the applicability of the bay-region concept to these thia-PAHs
with more certainty, we initiated our studies with an environ-
mentally occurring15 pentacyclic thia-PAH, phenanthro[3,2-
b][1]benzothiophene (1). In contrast to NBT, 1 is a potent
carcinogen,6 and its carcinogenic activity is higher than that of
its carbocyclic isostere, dibenzo[a,h]anthracene (2), which is
considered to be metabolically activated via bay-region diol
epoxide (4).3 The present study was undertaken to synthesize
trans-3,4-dihydroxy-3,4-dihydrophenanthro[3,2-b][1]benzothio-
phene (3), a precursor of the bay-region diol epoxide of 1, in
order to establish whether 3 is a mutagenic/carcinogenic
metabolite of 1. Synthesis of 3 has allowed us to confirm the
structure of one of the major metabolites of 116 and to evaluate
its potential as a proximate mutagen/carcinogen of 1.17,18
cyclization route, a second approach involving a synthetic
application of the Suzuki cross-coupling reaction22 was investi-
gated for the synthesis of 6 (Scheme 2). Synthetic application of
Scheme 2 Reagents: i, Pd(PPh3)4, CsF–DME; ii, Me3SϩIϪ, KOH–
MeCN; iii, MeSO3H–CH2Cl2.
the Suzuki cross-coupling reaction has recently been developed
in our laboratory for a concise, regiospecific synthesis of
methoxy-substituted PAHs,19 which are analogous to 6. One of
the key starting materials, 2-(dihydroxyboryl)-5-methoxybenz-
aldehyde (12), has been synthesized in the literature23 from the
dimethyl acetal of 2-bromo-5-methoxybenzaldehyde (16) in less
than 30% yield. A significant improvement in the yield (>90%)
of 12 was achieved by replacing the dimethyl acetal of 16
with the ethylene acetal of 16. The coupling reaction of
3-bromodibenzothiophene (11)24 with 2-(dihydroxyboryl)-5-
methoxybenzaldehyde (12), in the presence of anhydrous CsF
and a catalytic amount of Pd(PPh3)4, produced 3-(2-formyl-4-
methoxyphenyl)dibenzothiophene (13) in 87% yield. The alde-
hyde functionality of 13 was modified to ethylene epoxide by
reacting it with trimethylsulfonium iodide and powdered KOH
in acetonitrile at 65 ЊC for 12 h to produce the relatively
unstable ethylene epoxide derivative 14 as an oil. This homolo-
gation procedure, which was initially developed by Borredon
et al.,25 was found to be superior to the method that employed
a phase-transfer catalyst.19a The NMR spectrum of 14 which
indicated the lack of an aldehydic proton at 10.01 ppm, and the
appearance of three proton signals between 2.8 and 3.9 ppm as
double doublets confirmed the presence of oxiranyl side chain
in 14. The acid-catalyzed cyclodehydration of 14 with MeSO3H
in CH2Cl2 at room temperature took place smoothly. However,
the reaction was not regiospecific, and produced a nearly 1 : 1
mixture of 6 and 2-methoxyphenanthro[1,2-b][1]benzothio-
phene (15). The presence of 6 in the mixture was indicated
by NMR, which showed two singlets corresponding to two
protons (H-7 and H-13) of meso-acenic type (vide infra). Separ-
ation of 6 from 15 by chromatography was not successful.
However, the extraction of the mixture with hot ethanol
produced a residue, which, after recrystallization from CH2Cl2,
yielded pure 15. The product that crystallized out from the
ethanol extract contained nearly 70% of 6 and 30% of 15 (1H
NMR). An attempt to isolate pure 6 in a sufficient amount from
the ethanol-extracted mixture by fractional recrystallization
from various solvents was unsuccessful. However, the chemical
transformation of the mixture containing 6 as the major
product (following the procedure shown in Scheme 4) produced
a mixture of two dihydrodiols from which 3 was separated by
analytical HPLC. Although this procedure afforded 3 in pure
form by analytical HPLC, however, due to the poor loadability
of the dihydrodiol mixture, preparative HPLC was not very
Results and discussion
The most recent advances13,19–21 in the synthesis of various
dihydrodiol derivatives of PAHs and thia-PAHs prompted us to
select 3,4-dimethoxyphenanthro[3,2-b][1]benzothiophene (5)
or 3-methoxyphenanthro[3,2-b][1]benzothiophene (6) as the
potential intermediates for the synthesis of 3. Three synthetic
strategies for the preparation of 5 or 6 were explored. The
oxidative photocyclization route,20 which is widely used in the
synthesis of PAH derivatives analogous to 5 and which required
simpler starting materials, was first investigated (Scheme 1).
Scheme 1 Reagents and conditions: i, 50% NaOH–CH2Cl2; ii, hν.
However, this strategy was not productive since the oxidative
photocyclization of 1,2-diarylethylene 9 produced a mixture of
chromatographically similar compounds containing mostly 10
and trace amounts of 5 (1H NMR). Because of the lack of
formation of 5 in a sufficient amount via the oxidative photo-
J. Chem. Soc., Perkin Trans. 1, 2001, 1018–1023
1019

View Article Online
successful in isolating pure 3 in the sufficient amounts for
biological studies.
Finally, an attempt to synthesize 6 via cyclodehydrobromin-
ation of an o-bromostilbene, previously applied in constructing
phenanthrene building blocks,26,27 afforded a 1 : 2 mixture of
6 and 18 (Scheme 3), which were easily separable by column
Scheme 4 Reagents: i, BBr3–CH2Cl2; ii, Fremy’s salt; iii, NaBH4,
O2–EtOH.
1
a moderate yield (40%). The H NMR spectrum of 3 was
consistent with its assigned structure. The large value for the
coupling constant (~11.7 Hz) between the carbinol protons
H-3 and H-4 of 3 indicates that the conformation of the trans-
dihydrodiols with a pseudo-diaxial orientation of the carbinol
protons is strongly populated. An HPLC analysis and a com-
parison of the UV spectra indicated that the major dihydrodiol
obtained from the mixture of 6 and 15 (see above) was 3.
Our preliminary studies16–18 have indicated that 3 is a major
metabolite of 1, and exhibits mutagenic activity in Salmonella
typhimurium strain TA100 that is higher than that of 1. Details
of these and other biological studies with 3 and other deriv-
atives of 1 will be reported elsewhere.
Scheme 3 Reagents: i, 50% NaOH–CH2Cl2, ii, KOH–quinoline.
chromatography. The Wittig reaction of the phosphonium salt
7 (prepared from triphenylphosphine and 2-bromomethyl-
dibenzothiophene28) with 2-bromo-5-methoxybenzaldehyde
(16)29 produced 1-(2-bromo-5-methoxy)-2-(2-dibenzothienyl)-
ethylene (17) in nearly quantitative yield as a mixture of Z- and
E-isomers in the ratio of Z : E ≈ 4. The Z and E-isomers of 17
were separated by preparative TLC (2% EtOAc–hexanes), and
Experimental
1
their structures were tentatively assigned by their H NMR on
3-Bromodibenzothiophene (11),24 2-bromo-5-methoxybenz-
aldehyde (16),29 and 2-bromomethyldibenzothiophene28 were
synthesized according to the published procedures. All reagents
and solvents (anhydrous or otherwise) were used as received
without further purification. Dry column grade silica gel was
purchased from E. Merck. The 1H and 13C NMR spectra were
recorded on 300, 400 or 500 MHz NMR spectrometers in an
appropriate solvent with tetramethylsilane (TMS) as internal
standard. Chemical shifts are in ppm relative to internal TMS
the basis of the relatively larger coupling constant exhibited
by the olefinic protons of the E-isomer (δ 7.19, J = 16 Hz)
versus the relatively less polar Z-isomer (δ 6.68, J = 12 Hz). The
olefinic protons, and one phenyl proton (H-5) of the E-isomer
were significantly deshielded compared to the corresponding
protons of the Z-isomer. The strong preference for the forma-
tion of the Z-isomer in the Wittig reaction with o-substituted
benzaldehyde is consistent with previous observations.30,31 The
treatment of 17 as a mixture with powdered KOH in refluxing
quinoline (redistilled) produced a mixture of two compounds,
which were easily separated by column chromatography in 23
and 51% yields, respectively. Both of these compounds showed
a molecular ion (Mϩ) at 318 in their mass spectrum con-
firming that they were isomers. A comparison of the 1H NMR
spectra allowed us to identify these two compounds as 6 and
3-methoxyphenanthro[3,4-b][1]benzothiophene (18). The 1H
NMR spectrum of the relatively nonpolar product was con-
sistent with structure 6. This spectrum showed the presence
of two downfield sharp singlets at 8.60 and 9.02 ppm of
the meso-acenic protons H-7 and H-13 of 6. One of these
resonances was deshielded by interaction with proton H-7.
1
for H NMR spectra and relative to solvent signals for 13C
NMR spectra. Mass spectral data, HRMS (EI), were obtained
by the mass spectral facility of the Department of Chemistry,
State University of New York at Buffalo. All the melting points
were uncorrected.
CAUTION: substituted phenanthro[b]benzothiophenes (5, 6,
10, 15, 18, and 19), o-quinone 20, and dihydrodiol 3 are potential
carcinogens and should be handled in accordance with NIH
Guidelines for the Laboratory Use of Chemical Carcinogens.
1-(2,3-Dimethoxyphenyl)-2-(dibenzo[b,d]thiophen-2-yl)ethylene
(9)
1
The H NMR spectrum in which the most downfield protons
H-1 and H-13 were centered at 8.68 and 8.94 ppm, respectively,
as doublets was consistent with the structure assigned to 18.
The presence of the two most upfield aromatic proton signals,
one as a doublet with a small coupling (~2 Hz), and the other as
a double doublet with a small (~2.5 Hz) and a large coupling
(~9 Hz) confirmed the presence of the methoxy group at
position 3 of 6 and 18.
The chemical transformation of 6 to its trans-dihydrodiols 3
is illustrated in Scheme 4. Thus, demethylation of 6 to the
corresponding phenol 19 was achieved in a nearly quantitative
yield (96%) with boron tribromide in CH2Cl2. The oxidation
of the phenol 19 with Fremy’s salt produced highly colored
o-quinone 20 in 95% yields. The reduction of the o-quinone 20
with sodium borohydride in 95% ethanol while bubbling oxygen
took place stereospecifically to provide 3,4-dihydrodiol 3 in
2,3-Dimethoxybenzaldehyde (8) (1.8 g, 11 mmol) and the
phosphonium salt 7 (5.4 g, 10 mmol, prepared from 2-bromo-
methyldibenzothiophene and triphenylphosphine) were dis-
solved in 125 mL of CH2Cl2, and the resulting solution was
treated with 40 mL of 50% NaOH. The mixture was stirred
under Ar at room temperature for 3 h by which time TLC
indicated the absence of starting aldehyde 8. The reaction
mixture was diluted with water (100 mL), and the organic layer
was separated. The aqueous layer was extracted twice with
CH2Cl2, and the combined organic layer was washed with
water, dried over Na2SO4, and concentrated to dryness to
produce a solid. Recrystallization of the solid from EtOH
afforded 2.91 g (85%) of 9 as colorless needles, mp 134–135 ЊC.
1H NMR (300 MHz, CDCl3) δ 3.89 (s, 3 H), 3.90 (s, 3 H),
6.80–6.85 (m, 5 H), 7.34 (dd, 1 H, J = 1.7 and 8.3 Hz), 7.40–7.45
1020
J. Chem. Soc., Perkin Trans. 1, 2001, 1018–1023

View Article Online
(m, 2 H), 7.65 (d, 1 H, J = 8.3 Hz), 7.80–7.85 (m, 1 H),
7.93–7.98 (m, 1 H), 8.03 (d, 1 H, J = 1.4 Hz). Anal. Calcd for
C22H18O2S: C, 76.3; H, 5.2. Found: C, 76.1; H, 4.9%.
134.98, 134.64, 132.30, 127.08, 126.87, 124.64, 124.21, 122.94,
121.78, 121.46, 121.32, 110.01, 55.68; MS (m/z) 318 (Mϩ, 100).
Anal. Calcd for C20H14O2S: C, 75.5; H, 4.4. Found: C, 75.3; H,
4.7%.
Oxidative photocyclization of 9
3-Methoxyphenanthro[3,2-b][1]benzothiophene (6) and 3-meth-
oxyphenanthro[1,2-b][1][benzothiophene (15)
A solution of 9 (0.58 g, 1.67 mmol), iodine (0.42 g), and 1,2-
epoxybutane (5 mL) in anhydrous benzene was irradiated by
UV (Hanovia lamp) for 12 h. The solution was concentrated to
250 mL and washed with 10% NaHSO3 followed by water.
After drying over anhydrous Na2SO4, the benzene solution was
evaporated under reduced pressure. The crude product was
dissolved in EtOAc–hexanes, and kept at Ϫ25 ЊC for a week to
To a suspension of 13 (0.64 g, 2 mmol) and trimethylsulfonium
iodide (0.44 g, 2.16 mmol) in acetonitrile (30 mL) containing a
trace of water (0.1 mL) was added powdered KOH (0.2 g, 5.1
mmol), and the mixture was stirred at 60–65 ЊC for 20 h under
argon. The mixture was poured on to ice-cold water, and
extracted with EtOAc. The organic phase was washed with
water, dried over anhydrous Na2SO4 and the solvent was
removed by rotary evaporation to afford 0.65 g (100%) of
sufficiently pure ethylene epoxide 14 as a light yellow thick oil.
1H NMR (400 MHz, CDCl3) δ 2.82 (dd, 1 H, J = 2.6 and 5.9
Hz), 3.10 (dd, 1 H, J = 4.0 and 5.5 Hz), 3.85 (s, 3 H), 3.87 (dd,
1 H, J = 2.6 and 3.7 Hz), 6.89 (d, 1 H, J = 2.6 Hz), 6.92 (dd, 1 H,
J = 2.6 and 8.4 Hz), 7.31 (d, 1 H, J = 8.4 Hz), 7.45–7.52 (m,
3 H), 7.84 (d, 1 H, J = 1.1 Hz), 7.83–7.88 (m, 1 H), 8.15–8.22
(m, 2 H).
1
produce ~5–6 mg of 5 as a granular solid, mp 231–233 ЊC. H
NMR (400 MHz, CDCl3) δ 4.03 (s, 3 H), 4.05 (s, 3 H), 7.38 (d,
1 H, J = 9.1 Hz), 7.47–7.56 (m, 2 H), 7.88 (d, 2 H, J = 8.4 Hz),
8.08 (d, 1 H, J = 9.1 Hz), 8.26–8.32 (m, 1 H), 8.47 (d, 1 H, J = 9.1
Hz), 8.59 (s, 1 H), 9.01 (s, 1 H). HRMS obsd mass 344.087112,
calcd for Mϩ, 344.087105. The mother liquor consisted of a
mixture of chromatographically similar compounds including
the starting alkene 9, and a major product 10, which could not
be isolated in pure form from the mixture. 1H NMR of 10 (400
MHz, CDCl3) δ 4.06 (s, 6 H), 7.22 (d, 1 H, J = 9.2 Hz), 7.33 (t,
1 H, J = 8.1 Hz), 7.43 (t, 1 H, J = 9.1 Hz), 7.77–7.85 (m, 2 H),
7.91 (d, 1 H, J = 8.4 Hz), 7.95 (d, 1 H, J = 8.4 Hz), 8.16 (d, 1 H,
J = 9.1 Hz), 8.69 (d, 1 H, J = 8.1 Hz), 8.78 (d, 1 H, J = 9.1 Hz).
To a stirred solution of the epoxide 14 (0.60 g, 1.8 mmol) in
anhydrous CH2Cl2 (50 mL) at 0 ЊC was added dropwise 50%
MeSO3H in CH2Cl2 (10 mL) during 10 min. The mixture was
stirred at rt for 20 h. The CH2Cl2 solution was washed succes-
sively with water, 10% NaOH, and water, and dried over
anhydrous Na2SO4. Evaporation of the solvent gave 0.37 g
(65%) of a nearly 1 : 1 mixture of 6 and 15. The mixture of 6
and 15 (190 mg) was extracted with hot ethanol, and the residue
was recrystallized from CH2Cl2–hexanes to produce 70 mg of
15, mp 282–284 ЊC. 1H NMR (400 MHz, CDCl3) δ 3.98 (s, 3 H),
7.31 (d, 1 H, J = 2.2 Hz), 7.33 (dd, 1 H, J = 11.7 and 2.2 Hz),
7.45–7.58 (m, 2 H), 7.83 (d, 1 H, J = 8.8 Hz), 7.96 (d, 1 H,
J = 7.3 Hz), 8.03 (d, 1 H, J = 8.8 Hz), 8.23 (d, 1 H, J = 7.3 Hz),
8.31 (d, 1 H, J = 8.8 Hz), 8.60–8.70 (m, 2 H). MS m/z: 314 (Mϩ).
2-(Dihydroxyboryl)-5-methoxybenzaldehyde (12)
A mixture of 2-bromo-5-methoxybenzaldehyde 16 (11.2 g,
0.052 mol), redistilled ethylene glycol (70 mL), anhydrous
benzene (350 mL), and toluene-p-sulfonic acid (0.8 g) was
heated under reflux for 20 h. After cooling to rt, the mixture
was poured on to 10% aqueous K2CO3 (500 mL) contained
in a separating funnel. The benzene layer was separated, and
washed successively with 10% aqueous K2CO3, and brine con-
taining K2CO3, and dried over anhydrous Na2SO4. Removal of
the solvent on a rotary evaporator provided 13.5 g (100%) of
the ethylene acetal of 16 as colorless oil.
To a solution of this acetal (13.5 g, 0.052 mol) in dry ether
(60 mL) n-BuLi (1.6 M, 40 mL, 0.064 mol) in hexane was added
in 5 min under nitrogen at Ϫ75 ЊC. After 1 h of stirring of the
mixture containing the solid at Ϫ75 ЊC, triisopropyl borate
(18.5 mL, 0.08 mol) was added in one portion, and the resulting
mixture was allowed to stir at ambient temperature for 2.5 h,
and then decomposed by adding 60 mL of 2 M HCl. The result-
ing mixture was refluxed for 1.5 h with stirring. Most of the
products precipitated directly, the rest was recovered by extrac-
tion of the ethereal phase with 2 M NaOH and acidification of
the extract to produce a combined yield of 8.6 g (92.7%) of 12,
mp 157–161 ЊC (lit.23 mp 158–160 ЊC).
1
Anal. Calcd for C21H14OSؒ–20CH2Cl2: C, 79.4; H, 4.4. Found: C,
79.5; H, 4.6%. The product (~110 mg) isolated from the mother
liquor contained 70% 6 and 30% 15. Further recrystallization
of this mixture did not improve the purity of 6.
1-(2-Bromo-5-methoxyphenyl)-2-(dibenzo[b,d]thiophen-2-yl)-
ethylene (17)
2-Bromo-5-methoxybenzaldehyde (16) (1.35 g, 6.3 mmol) and
the phosphonium salt 7 (3.24 g, 6.0 mmol, prepared from
2-bromomethyldibenzothiophene and triphenylphosphine)
were dissolved in 100 mL of CH2Cl2, and the resulting solution
was treated with 15 mL of 50% NaOH. The mixture was stirred
under Ar at room temperature for 15 h by which time TLC
indicated the absence of starting aldehyde 7. The reaction
mixture was diluted with water (50 mL), and the organic layer
was separated. The aqueous layer was extracted twice with
CH2Cl2, and the combined organic layer was washed with
water, dried over Na2SO4, and concentrated to dryness. The
residue (4.5 g) obtained as a semisolid was stirred with 100 mL
of hexane, and filtered. The filtrate was passed through a small
column of dry column grade silica gel. Elution of the column
with hexane afforded a mixture of the Z- and E-isomers of 17
(2.4 g, 100%) as a colorless oil, Z : E ≈ 4. The mixture of
alkenes was used in the next step. A small sample of the mixture
was separated by preparative TLC using 2% EtOAc–hexane as
3-(2-Formyl-4-methoxyphenyl)dibenzo[b,d]thiophene (13)
A mixture of 11 (1.5 g, 5.7 mmol), 12 (1.14 g, 6.3 mmol),
anhydrous CsF (2.1 g, 13.8 mmol), and Pd(PPh3)4 (0.22 g, 0.18
mmol) in anhydrous DME (70 mL) was heated under reflux for
12–18 h under argon. The reaction was monitored by TLC (5%
EtOAc–hexanes) until only a trace of or no more bromide was
detected. The reaction mixture was cooled and then extracted
with a 1 : 1 mixture of EtOAc and water. The ethyl acetate layer
was washed with 5% NaOH followed by water to remove any
unreacted boronic acid. After drying over anhydrous Na2SO4,
the EtOAc solution was concentrated in vacuo to yield an oil.
Column chromatography of the oily product over dry column
grade silica gel using hexane as an eluant gave initially a trace
amount of 5, followed by 1.67 g (92%) of 13 as a light yellow
1
the developing solvent. Relatively nonpolar Z-isomer: oil. H
NMR (300 MHz, CDCl3) δ 3.46 (s, 3 H), 6.68 (d, 1 H, J = 12.0
Hz), 6.70 (dd, 1 H, J = 3 and 8.6 Hz), 6.78 (d, 1 H, J = 3 Hz),
6.86 (d, 1 H, J = 12.1 Hz), 7.24–7.29 (m, 1 H), 7.39–7.56 (m, 3
H), 7.65 (d, 1 H, J = 8.3 Hz), 7.80–7.85 (m, 1 H), 7.93–8.00 (m,
2 H). HRMS obsd mass 393.999233, calc. for Mϩ, 394.002698.
E-isomer: oil; 1H NMR (300 MHz, CDCl3) δ 3.86 (s, 3 H), 6.73
(dd, 1 H, J = 8.8 and 3.0 Hz), 7.19 (d, 1 H, J = 16.1 Hz), 7.24 (d,
1
solid, mp 148–149 ЊC (CH2Cl2–hexanes). H NMR (300 MHz,
CDCl3) δ 3.93 (s, 3 H), 7.25 (dd, 1 H, J = 2.8 and 8.5 Hz),
7.42–7.57 (m, 5 H), 7.84 (d, 1 H, J = 1.1 Hz), 7.87–7.97 (m,
1 H), 8.19–8.24 (m, 2 H), 10.01 (s, 1 H). 13C NMR (300 MHz,
CDCl3) δ 192.26, 159.31, 139.77, 139.72, 138.63, 136.08, 135.04,
J. Chem. Soc., Perkin Trans. 1, 2001, 1018–1023
1021

View Article Online
1 H, J = 3 Hz), 7.46–7.57 (m, 4 H), 7.69 (dd, 1 H, J = 8.3 and 1.7
Hz), 7.82–7.88 (m, 2 H), 8.19–8.27 (m, 2 H). HRMS obsd mass
393.999233, calcd for Mϩ, 394.002698.
( )-trans-3,4-Dihydroxy-3,4-dihydrophenanthro[3,2-b][1]benzo-
thiophene (3)
To a stirred suspension of 20 (0.35 g, 1.14 mmol) in 150 mL of
ethanol was added NaBH4 (1.2 g, 33.33 mmol) in portions. The
mixture was stirred for 4 days while bubbling oxygen through
the solution. A nearly colorless solution of the reaction mixture
was poured on to ice, and extracted with ethyl acetate three
times. The combined organic phase was washed with water,
dried (Na2SO4), and concentrated. The resulting residue was
triturated with cold ether, and filtered to afford 0.14 g (40%) of
3 as a light grey crystalline solid, mp (sealed tube) 215–218 ЊC
(decomp.). 1H NMR (300 MHz, acetone-d6 ϩ MeOH-d4)
δ 4.48–4.51 (m, 1 H), 4.90 (d, 1 H, J = 11.7 Hz), 6.24 (dd, 1 H,
J = 2.2 and 10.1 Hz), 7.39 (dd, 1 H, J = 2.4 and 10.1 Hz), 7.50–
7.62 (m, 2 H), 7.92 (d, 1 H, J = 8.6 Hz), 7.93–8.01 (m, 1 H), 8.07
(d, 1 H, J = 8.6 Hz), 8.40–8.48 (m, 1 H), 8.81 (s, 1 H), 8.86 (s, 1
H). UV λmax (EtOH) 384 (ε/molϪ1 dm3 cmϪ1 3830), 367 (4734),
348 (5447), 332 (7058), 302 (17175), 286 (38984). HRMS obsd
mass 300.06249, calcd for Mϩ Ϫ H2O, 300.06089. Anal. Calcd
for C20H14O2Sؒ¹Et2O: C, 74.4; H, 5.4. Found: C, 74.2; H, 5.7%.
3-Methoxyphenanthro[3,2-b][1]benzothiophene (6) and 3-meth-
oxyphenanthro[3,4-b][1]benzothiophene (18)
A solution of 17 (4.7 g, 11.88 mmol) and KOH (4.7 g, 83.8
mmol) in 40 mL of redistilled quinoline was heated at reflux for
3 h. The mixture was cooled, poured into ice, acidified carefully
with concentrated sulfuric acid, and extracted with CH2Cl2 four
times. The combined organic layers were washed with water,
dried (Na2SO4), and concentrated, affording the products as a
dark yellow–brown solid. Trituration of the solid with EtOAc
1
produced 0.70 g (19%) of TLC pure 6, mp 222–224 ЊC. H
NMR (500 MHz, CDCl3) δ 3.99 (s, 3 H), 7.30 (d, 1 H, J = 2.4
Hz), 7.32 (dd, 1 H, J = 2.4 and 8.8 Hz), 7.49–7.55 (m, 2 H), 7.68
(d, 1 H, J = 8.8 Hz), 7.87–7.92 (m, 2 H), 8.27–8.31 (m, 1 H),
8.60 (s, 1 H), 8.65 (d, 1 H, J = 9.1 Hz), 9.02 (s, 1 H). MS (m/z,
relative intensity) 314 (Mϩ, 100%), 271 (70). Anal. Calcd for
C21H14OS: C, 80.2; H, 4.5. Found: C, 80.1; H, 4.7%.
¯
²
Analytical HPLC of 3 was achieved on a Zorbax C-18
column (4.6 × 250 mm) using a linear gradient (1% minϪ1) from
70% MeOH–water to 100% MeOH for 30 min with a flow rate
of 1 mL minϪ1, and the eluants were monitored at 254 nm.
These HPLC conditions showed the elution of 3 as a single
peak at 18.3 min. Analytical HPLC of a mixture of the dihy-
drodiols obtained from a 7 : 3 mixture of 6 and 15 (vide supra)
consisted of two peaks at 18.3 and 19.7 min, respectively. The
relatively polar dihydrodiol (18.3 min) that coeluted with 3 had
a UV spectrum identical to that of 3.
Chromatography of the product obtained from the mother
liquor over dry column grade silica gel using hexane as eluant
1
gave first 1.9 g (51%) of 18 [mp 129–130 ЊC. H NMR (500
MHz, CDCl3) δ 4.02 (s, 3 H), 7.15 (dd, 1 H, J = 2.4 and 9.1
Hz), 7.33 (d, 1 H, J = 2.4 Hz), 7.35 (dd, 1 H, J = 7.9 and 7.3 Hz),
7.45 (dd, 1 H, J = 7.0 and 7.9 Hz), 7.76 (d, 1 H, J = 8.5 Hz),
7.80–7.85 (m, 2 H), 7.91 (d, 1 H, J = 8.2 Hz), 7.96 (d, 1 H,
J = 8.0 Hz), 8.68 (d, 1 H, J = 8.2 Hz), 8.94 (d, 1 H, J = 9.1 Hz).
13C NMR (300 MHz, CDCl3) δ 158.87, 140.58, 140.01, 137.30,
135.18, 130.26, 130.15, 130.00, 128.78, 127.92, 127.66, 126.22,
126.20, 125.64, 123.75, 123.38, 121.02, 115.12, 107.92, 55.87.
MS (m/z) 314 (Mϩ, 100%), 298 (12), 282 (28), 269 (49). Anal.
Acknowledgements
1
–
Calcd for C21H14OSؒ5H2O: C, 79.3; H, 4.5. Found: C, 79.0; H,
4.5%] followed by 0.15 g (4.0%) of an additional amount of 6.
This investigation was supported by Grant No. R826192-01-0
awarded to S. K. by the United States Environmental Protec-
tion Agency, Washington, DC.
3-Hydroxyphenanthro[3,2-b][1]benzothiophene (19)
A solution of 1 M BBr3 (4.0 mL, 4.0 mmol) in CH2Cl2 was
added by syringe to a stirred solution of 6 (0.6 g, 1.91 mmol)
in 125 mL of anhydrous CH2Cl2 at 0 ЊC. After stirring for 12 h
at rt, the mixture was hydrolyzed with ice-cold water. The
organic layer was separated, washed with water, dried (Na2SO4),
and evaporated under reduced pressure to afford a solid. The
solid was triturated with hexane and filtered to produce 0.55 g
(96%) of 19 as a light brown crystalline solid, mp 294–296 ЊC.
1H NMR (500 MHz, CDCl3) δ 7.34 (dd, 1 H, J = 2.4 and 8.8
Hz), 7.38 (d, 1 H, J = 2.4 Hz), 7.57–7.62 (m, 2 H), 7.72 (d, 1 H,
J = 8.8 Hz), 7.96 (d, 1 H, J = 9.1 Hz), 8.01–8.05 (m, 1 H),
8.45–8.50 (m, 1 H), 8.84 (d, 1 H, J = 8.8 Hz), 8.86 (s, 1 H), 9.29
(s, 1 H), 9.97 (s, 1 H, exchangeable with D2O). HRMS obsd
mass 300.059422, calcd for Mϩ, 300.060887. Anal. Calcd for
C20H12OSؒ¾H2O: C, 76.5; H, 4.3. Found: C, 76.3; H, 4.0%.
References
1 A bay-region in a polycyclic aromatic hydrocarbons exists when
bonds in two nonfused benzene rings are fixed in an s-cis butadiene
conformation. The prototype of a bay-region is the sterically
hindered area between the 4 and 5 positions in phenanthrene.
Accordingly the regions between the 6 and 7 positions or between
the 1 and 11 positions in NBT and the 7 and 8 positions or 1 and 13
positions in 1 are the bay-regions.
2 R. E. Lehr, S. Kumar, W. Levin, A. W. Wood, R. L. Chang, A. H.
Conney, H. Yagi, J. M. Sayer and D. M. Jerina, in Polycyclic
Aromatic Hydrocarbons and Carcinogenesis, ACS Symposium Series
283, ed. R. G. Harvey, ACS, Washington, DC, 1985, p. 63.
3 P. G. Wislocki and A. Y. H. Lu, in Polycyclic Aromatic Hydrocarbon
Carcinogenesis: Structure–Activity Relationships, Vol. 1, eds. S. K.
Yang and B. D. Silverman, CRC Press, Inc., Boca Raton, FL, 1988,
p. 1.
4 R. G. Harvey, Polycyclic Aromatic Hydrocarbons: Chemistry and
Carcinogenicity, Cambridge University Press, Cambridge, UK, 1991.
5 J. Jacob, Sulfur Analogues of PAHs (Thiaarenes), Cambridge
University Press, Cambridge, UK, 1990 and references therein.
6 A. Croisy, J. Mispelter, J. M. Lhoste, F. Zajdela and P. Jacqignon,
J. Heterocycl. Chem., 1984, 21, 353.
Phenanthro[3,2-b][1]benzothiophene-3,4-dione (20)
1
6
–
To a stirred mixture of 19 (0.55 g, 1.83 mmol), M KH2PO4
(200 mL), and Fremy’s salt (1.5 g, 5.6 mmol) in 100 mL
of CH2Cl2 and 300 mL of benzene was added 5–6 drops of
Adogen 464, and stirring was continued for 12 h. The mixture
was filtered to recover 0.2 g of phenol 19. The filtrate was added
to 0.25 g (0.83 mmol) of Fremy’s salt, and the stirring was
continued for an additional 5 h. The organic layer was separ-
ated, washed with water, dried (Na2SO4), and concentrated
under reduced pressure. The resulting residue was triturated
with CH2Cl2–hexane to yield 0.35 g (95% based on recovered
19) of 20 as a bright purple crystalline solid, mp 267–269 ЊC. 1H
NMR (500 MHz, CDCl3) δ 6.64 (d, 1 H, J = 10.4 Hz), 7.50–7.66
(m, 2 H), 7.90 (d, 1 H, J = 7.6 Hz), 8.16–8.24 (m, 2 H), 8.34 (d, 1
H, J = 7.6 Hz), 8.42 (d, 1 H, J = 10.7 Hz), 8.66 (s, 1 H), 8.75 (s, 1
H). HRMS obsd mass 314.039714, calcd for Mϩ, 314.040151.
7 B. D. Tilak, Tetrahedron, 1960, 9, 76.
8 S. E. Murphy, S. Amin, K. Coletta and D. Hoffmann, Chem. Res.
Toxicol., 1992, 5, 491.
9 J. Jacob, A. Schmoldt, C. Augustin, G. Raab and G. Grimmer,
Toxicology, 1991, 67, 181.
10 D. M. Jerina and J. W. Daly, in Drug Metabolism, eds. D. V. Parke
and R. L. Smith, Taylor and Frances Ltd., London, 1976, p. 13.
11 D. M. Jerina, R. E. Lehr, H. Yagi, O. Hernandez, P. M. Dansette,
P. G. Wislocki, A. W. Wood, R. L. Chang, W. Levin and A. H.
Conney, in In Vitro Metabolic Activation and Mutagenesis Testing,
eds. F. J. deSerres, J. R. Fouts, J. R. Bend and R. M. Philpot,
Elsevier/North-Holland Biomedical, Amsterdam, 1976, 159.
12 J. E. Sinsheimer, B. H. Hooberman, S. K. Das, P. M. Savla and
A. J. Ashe III, Environ. Mol. Mutagen., 1992, 19, 259.
1022
J. Chem. Soc., Perkin Trans. 1, 2001, 1018–1023

View Article Online
13 B. Misra and S. Amin, Chem. Res. Toxicol., 1990, 3, 93.
14 A. W. Wood, W. Levin, D. Ryan, P. E. Thomas, H. Yagi, H. D. Mah,
D. R. Thakker, D. M. Jerina and A. H. Conney, Biochem. Biophys.
Res. Commun., 1977, 78, 847.
15 A. Nakamura, Osaka-furitsu Koshu Eisei Kenkyusho Kenkyu
Hokoku, Shokuhin Eisei Hen, 1983, 14, 69; Chem. Abstr., 1984, 101,
34134r.
16 Z.-X. Yuan, H. C. Sikka, S. Munir and S. Kumar, Proc. Am. Assoc.
Cancer Res., 2000, 41, 570.
17 S. Kumar, Z.-X. Yuan, S. Munir and H. C. Sikka, Proc. Am. Assoc.
Cancer Res., 2001, 42, 471.
18 S. Kumar, J. M. Lin, S. Amin and H. C. Sikka, Proc. Am. Assoc.
Cancer Res., 2001, 42, 165.
19 (a) S. Kumar, J. Org. Chem., 1997, 62, 8535; (b) S. Kumar,
Tetrahedron Lett., 1996, 37, 6271; (c) S. Kumar, J. Chem. Soc.,
Perkin Trans. 1, 1998, 3157.
21 S. Kumar and T.-Y. Kim, J. Org. Chem., 2000, 65, 3883.
22 N. Miyaura and A. Suzuki, Chem. Rev., 1995, 95, 2457.
23 G. M. Keseru, G. Mezey-Vandor, M. Nogradi, B. Vermas and
M. Kajtar, Tetrahedron, 1992, 48, 913.
24 G. Illuminati, J. F. Nobis and H. Gilman, J. Am. Chem. Soc., 1951,
73, 5887.
25 E. Borredon, M. Delmas and A. Gaset, Tetrahedron Lett., 1982, 23,
5283.
26 S. Amin, K. Huie, N. Hussain, G. Balanikas and S. S. Hecht, J. Org.
Chem., 1985, 50, 1948.
27 S. Amin, K. Huie, N. Hussain, G. Balanikas, S. G. Carmella and S.
S. Hecht, J. Org. Chem., 1986, 51, 1206.
28 J. Gourier and P. Canonne, Bull. Soc. Chim. Fr., 1973, 3110.
29 I. Fleming and M. J. Woolias, J. Chem. Soc., Perkin Trans. 1, 1979,
829.
30 J.-T. Zhang, W. Dai and R. G. Harvey, J. Org. Chem., 1998, 63, 8125.
31 R. G. Harvey, J.-T. Zhang, E. Luna and J. Pataki, J. Org. Chem.,
1998, 63, 6405.
20 A. S. Kiselyov, H. Lee and R. G. Harvey, J. Org. Chem., 1995, 60,
6123 and references therein.
J. Chem. Soc., Perkin Trans. 1, 2001, 1018–1023
1023