Aryltricyclospirodienones; novel steroid mimics as inhibitors of aromatase

Article abstract of DOI:10.1039/a608338b

The aryltricyclospirodienones 16a, 16b, 16c and 18 have been designed as potential inhibitors of aromatase and are synthesised from 6-methoxytetralone; ? compounds 16a and 18 have proved effective inhibitors with activities of the same order as aminoglutethimide.

Full text of DOI:10.1039/a608338b

View Article Online / Journal Homepage / Table of Contents for this issue
Aryltricyclospirodienones; novel steroid mimics as inhibitors of
aromatase
David Hobbs-Mallyon, Warren Li and Donald A. Whiting*
Chemistry Department, University of Nottingham, Nottingham, UK NG7 2RD
The aryltricyclospirodienones 16a, 16b, 16c and 18 have been designed as potential inhibitors of
aromatase and are synthesised from 6-methoxytetralone;† compounds 16a and 18 have proved effective
inhibitors with activities of the same order as aminoglutethimide.
Aromatase is a key enzyme in the biogenesis of mammalian sex
hormones, conducting the conversion of androgens 1a–c into
estrogens 2a–c. The mode of action of this enzyme, which
belongs to the cytochrome P-450 group, has been the subject of
intense study1 and the mechanism has been demonstrated to
involve three sequential oxidative stages. The first two of these
are C19 hydroxylations, with excision of C19 as formate and
aromatisation as the final phase. Interest in this enzyme focuses
both on sexual development in various organisms and on
human diseases in which estrogens play an important role, not-
ably certain types of carcinoma—a significant proportion of
breast cancers are hormone related. In suitable circumstances
reduction of estrogen levels can control such cancers and inhib-
ition of the activity of aromatase is an attractive target enzyme
for this purpose in view of its unique role. A considerable num-
ber of synthetic inhibitors have been investigated,2a,2b both ster-
oidal and non-steroidal, some of which have clinical value and
there is much current activity in this area. Various modes of
action have been noted, including (a) reversible competitive
inhibition with weak binding into the active site, (b) quasi-
reversible inhibition, with strong complexation to the haem
iron, (c) irreversible binding to the protein and (d) mechanism-
based ‘suicide’ inhibition. The different types of binding can be
distinguished by spectroscopic methods.
Fig. 1
pounds,4 with clinical gain. We selected as our synthetic target
the structural type 3, containing a tricyclic spirodienone sub-
unit, for the following reasons. In compounds of type 3 the
steroid angular methyl (C19) is replaced by the ring residue. The
repeated oxidation at C19 in the steroid requires free C–Me
rotation, prevented in 3, with potential blocking of the second
or third oxidation step. Also the dienone moiety echoes the
androgen ring A, and the ∆1 double bond promotes irreversible
inhibition in some steroidal inhibitors.5 Finally, we hoped that
ring B would be a platform to control the spatial disposition of
substituents Y in 3 close to the haem iron, thus offering possible
enhancement of selectivity. Only two relative stereochemistries
are feasible, those with the aryl substituent trans or cis to the
bridging ring C residue. As shown in Fig. 1, the trans geometry
4a 6 mimics well the natural androgen substrate geometry 4b,
allowing for the low energy barrier to rotation of the aryl sus-
tituent; interestingly the cis form 4c, with a quasi-boat ring B, is
also a fair match.
In our contribution to this field we chose to concentrate on
relatives of 2-aryltetralins† as potential inhibitors. Such com-
pounds are known to have estrogen antagonist properties3 and
frequently metabolise at a slower rate than steroidal com-
† IUPAC names for tetralin, tetralol and tetralone are 1,2,3,4-
tetrahydronaphthalene, 1,2,3,4-tetrahydronaphthalen-1-ol and 3,4-
dihydronaphthalen-1(2H)-one, respectively. IUPAC names are given in
the Experimental section for the compounds prepared in this paper.
J. Chem. Soc., Perkin Trans. 1, 1997
1511

View Article Online
In some preliminary studies we explored various routes to
tricyclospirodienones derived from 1-substituted tetralones.7 In
this paper we show that the desired aryl substituted analogues
can be synthesised in racemic form, and that two examples
display inhibitory activity comparable with that of amino-
glutethimide.
The methodology for introducing functionalised alkyl chains at
C-1 in tetralins is limited, but we have found that an allyl group
can be efficiently inserted by reaction of a 1-tetralol with allyl
trimethylsilane–zinc iodide. Using these reagents, tetralols 12a–
c were converted into the 1-allyl derivatives 13a–c in good yield.
However the cis tetralol 12c gave a mixture of two geometric
isomers (ca. 2:1) and we were unable to find reaction con-
ditions under which the reaction proceeded stereoselectively.
Disappointingly we did not discover a way to separate the
diastereoisomers of the allyl compounds 13a–c using HPLC.
We therefore continued to tolerate diastereoisomeric mixtures in
the sequence, hoping to effect separation or selective reaction at
a later stage. Hydroboration–oxidation afforded the corres-
ponding primary alcohols 14a–c. Treatment with trimethylsilyl
chloride–sodium iodide in acetonitrile then provided the phen-
olic iodides 15a–c, via concurrent demethylation and hydroxy
substitution. Finally the phenolic iodides, as (2:1) diastereo-
isomeric mixtures, were refluxed with potassium tert-butoxide
in tert-butyl alcohol to afford the desired aryl tricyclospiro-
dienones 16a–c. These were obtained in low yield (10–15%),
possibly since the spirocyclisations were sluggish (24 h reflux
required) and other reactions competed, e.g. elimination. How-
ever in compensation the final products 16a–c were obtained as
single diastereoisomers, as shown unequivocally by 13C NMR
spectroscopy. It seems likely that the compounds have the aryl
substituent trans to bridging ring C. In support, MM2 calcula-
tions6 show that for 16a the trans geometry is substantially
more stable (ca. 15 kJ molϪ1) than the cis isomer, while the trans
iodide 15a is only ca. 6 kJ molϪ1 more stable than its cis coun-
terpart. Since a product-like transition state is to be expected,
the trans iodide 15a would have a lower activation energy for
spirocyclisation than the cis form and the trans system should
cyclise significantly faster. Given the slow cyclisation rate and
loss of material in side reactions, it is not surprising that only
one diastereoisomer could be detected, almost certainly then
the trans compounds 16 as shown. It should be said that at
The initial requirement was a satisfactory route to 2-
aryltetralones of type 5 (Scheme 1). For the fluorophenyltetra-
Scheme 1 Reagents and conditions: i, Na, benzene, 48 h reflux; ii,
MeOH, H2SO4, 24 h reflux; MeOH, KOH, 2 h reflux; HF, 24 h, room
temp.; iii, MgII, reflux
lone 5a, we adapted the route devised for 2-phenyltetralone by
Campbell and Kidd. 8 Thus, 3-methoxyphenethyl bromide 6 was
condensed with the 4-fluorophenylacetonitrile 7a, using sodium
sand in benzene (Scheme 1); a parallel reaction was carried out
with phenylacetonitrile 7b, to provide the nitriles 8a (84%)
and 8c (61%) respectively. Nitrile 8a was hydrolysed to the corre-
sponding acid 8b (66%) an intramolecular Friedel–Crafts
cyclisation then afforded the desired 2-aryltetralone 5a (100%);
the most convenient method for this step proved to be reaction
in anhydrous hydrofluoric acid at ambient temperature. Direct
cyclisation of the nitrile 8a to tetralone 5a using Houben–
Hoesch conditions proved unsatisfactory.9
A more direct and convenient route to 5a was also investi-
gated, based on the work of Hussey and Herr,10 and Stille and
Newsom.11 This chemistry involves the addition of an aryl
Grignard reagent to 2-bromotetralone; refluxing the initial
product induces magnesium salt catalysed rearrangement to
yield a mixture from which 2-aryltetralones can be isolated. We
were able to effect selective 2-bromination (74%) of 6-
methoxytetralone, using cupric bromide, and the bromide 9 was
treated at 0 ЊC with 4-fluorophenylmagnesium bromide 10 in
diethyl ether. The reaction mixture was then refluxed for 2 h and
column chromatography of the product led to the isolation of
2-(4-fluorophenyl)-6-methoxytetralone 5a, albeit in only 15%
yield. The corresponding 2-phenyl compound 5b was similarly
prepared in 13% yield. The reaction proceeds through the cis
and trans halohydrins 11.11 Although these are poor yields, they
compare favourably with the overall return from the stepwise
procedure and are much more direct.
The route adopted to the target spirodienones is outlined in
Scheme 2. Reduction of the tetralones 5a–c gave the corres-
ponding tetralols 12a–c; using sodium borohydride, mixtures
(ca. 2:1) of the cis and trans aryl alcohols were obtained. Stereo-
selectivity could be achieved with other reagents; thus, reduc-
tion of e.g. 2-(4-methoxyphenyl)-6-methoxytetralone 5c,12 with
L-Selectride at 0 ЊC afforded exclusively the cis tetralol 12c.†
Scheme 2 Reagents and conditions: i, Me3SiCH2CH᎐CH2, ZnI2; ii,
᎐
BH3Me2S; NaOH, H2O2; iii, Me3SiCl, NaI, MeCN; iv, ButOK,
ButOH, 24 h, reflux
1512
J. Chem. Soc., Perkin Trans. 1, 1997

View Article Online
Table 1 Inhibition of aromatase by various compounds
present no spectroscopic evidence can be advanced to substan-
tiate the stereochemical conclusions, the H NMR spectra at
400 MHz being too complex to elucidate.
1
Compound
% Inhibition
The iodide 15c has two possible pathways for spirocyclisa-
tion, leading either to 16c or to the isomer 17. The 13C NMR
data for the product show clear parallels with those for 16a and
16b and support the assignment as 16c.
Aminoglutethimide 23
Tricyclospirodienone 16b
Tricyclospirodienone 16a
Tricyclospirodienone 18
85
20
86
77
We also set out to synthesise a further member of this group of
tricyclospirodienone, 18, with a 6,6,6-ABC ring system. Our ini-
tial plan involved addition of a C4 side chain to a tetralol piva-
loate, as in development work.7 Such an approach was frus-
trated by the facile accompanying elimination reaction to the
tions of the androgens, and is likely to be a requirement for
activity. These results are extremely encouraging and suggest
that further work in this area would be rewarding, especially
since all the new compounds were tested as racemates. The syn-
thesis of homochiral samples, and compounds functionalised in
ring C are the clear future targets.
Experimental
General details
Unless otherwise stated the following apply. Melting points
were recorded using a Kofler hot-stage apparatus and are uncor-
1
rected. H NMR and 13C NMR spectra were measured in deu-
teriochloroform using internal TMS standard; multiplicities in
13C NMR were obtained using a DEPT sequence. Mass spectra
were obtained using electron impact or chemical ionisation.
Infrared spectra were collected from thin films (oils) or potas-
sium bromide discs (solids). All solvents were dried by standard
methods before use; ‘light petroleum’ was the fraction bp 40–
50 ЊC. ‘Evaporation’ refers to evaporation under reduced pres-
sure; ‘washed’ indicates the use of water and aqueous sodium
hydrogen carbonate as appropriate; ‘dried’ implies the use of
magnesium sulfate. ‘Chromatography’ means column chroma-
tography using silica gel 60; eluting solvents are listed.
2-Bromo-6-methoxy-3,4-dihydronaphthalen-1(2H)-one 9
6-Methoxy-1-tetralone (10 g, 0.057 mol) in chloroform (70 ml)
was added slowly to a refluxing stirred suspension of cupric
bromide (22.5 g, 0.1 mol) in ethyl acetate (75 ml), and the mix-
ture was refluxed for 2 h. The mixture was filtered, washed with
water and aq. sodium hydrogen carbonate, dried and evapor-
ated. The residual solid was crystallised from light petroleum
(bp 60–80 ЊC)–dichloromethane to yield the title bromo ketone
(9.45 g, 74%), mp 80–81 ЊC (lit.,15 mp 83 ЊC) (Found: C, 51.91;
H, 4.26; Br, 31.23%; m/z 253.992. C11H11BrO2 requires C, 51.79;
H, 4.35; Br, 31.32%; Mϩ, 253.994).
Scheme 3 Reagents and conditions: i, PCC, CH2Cl2, room temp.; ii,
Ph3P᎐CH2, THF, Ϫ78 ЊC; iii, BH3–Me2S; NaOH, H2O2; Me3SiCl, NaI,
᎐
MeCN; iv, ButOK, ButOH, reflux 24 h
stilbene-like product 19 and we were forced to employ a more
circuitous route via alcohol 14c. Thus standard oxidation to the
aldehyde 20 followed by Wittig reaction to form alkene 21,
enabled hydroboration to the homologated alcohol 22a.
Demethylation–iodination afforded 22b, which cyclised to yield
the desired spirodienone 18; NMR data indicate that the
phenolic ring D is still intact.
The compounds described above were then tested in vivo for
anti-aromatase activity using human term placental micro-
somes.13 Aminoglutethimide 23,14 an inhibitor which has been
2-(4-Fluorophenyl)-4-(3-methoxyphenyl)butanonitrile 8a
4-Fluorophenylacetonitrile (2.51 g, 18.6 mmol) was added to
sodium sand (0.43 g, 18.6 mmol) under dry diethyl ether (30
ml). This mixture was stirred under reflux (nitrogen) for 2 h and
then cooled to 0 ЊC when 3-methoxyphenethyl bromide (2.0 g,
9.3 mmol) was added. The reaction mixture was stirred over-
night when the diethyl ether was evaporated and replaced with
benzene (50 ml). The resulting suspension was refluxed for 48 h,
cooled and quenched with dil. hydrochloric acid. The organic
layer was washed, dried and evaporated and the residue was
chromatographed to afford the title nitrile (2.11 g, 84%)
(Found: m/z 269.124. C17H16NFO requires Mϩ, 269.122); νmax
/
cmϪ1 2241; δH 2.05–2.32 (2 H, m, 3-CH2), 2.77 (2 H, t, J 6.3, 4-
CH2), 3.72 (1 H, t, J 6.3, 2-CH), 3.80 (3 H, s, OCH3), 6.76 (3 H,
m, 3 × CH), 7.06 (2 H, m, 2 × CH), 7.20–7.32 (3 H, m,
3 × CH).
used in the clinic, was tested at the same time. The table shows
the activity of a selected few compounds including aminoglut-
ethimide, under parallel conditions, expressed as percentage
inhibition.
2-(4-Fluorophenyl)-6-methoxy-3,4-dihydronaphthalen-1(2H)-one
5a
It can be seen that three of the tricyclospirodienones are
effective inhibitors, and that two, compounds 16a and 18, dis-
play activities comparable to that of aminoglutethimide. The
dienone with the lowest activity of the three, 16b, lacks a polar
function in ring D; the latter mimicks the ring D oxygen func-
The nitrile 8a was hydrolysed with 10% methanolic sulfuric acid
at reflux for 24 h. Standard isolation procedures afforded 2-(4-
fluorophenyl)-4-(3-methoxyphenyl)butanoic acid (66%), mp
101–103 ЊC (Found: C, 71.00; H, 5.98%; m/z 288.117.
C17H17FO3 requires C, 70.82; H, 5.94%; Mϩ, 288.116);
J. Chem. Soc., Perkin Trans. 1, 1997
1513

View Article Online
νmax/cmϪ1 3700–2400, 1708. This acid (250 mg, 0.87 mmol) was
dissolved in anhydrous hydrofluoric acid (4.5 g) and the solu-
tion was set aside overnight. The hydrofluoric acid was then
evaporated and the residue was dissolved in diethyl ether–
dichloromethane (2:3); this solution was then washed (aq.
sodium hydroxide), dried and evaporated. The residue was
chromatographed to yield the title ketone (200 mg, 80%), mp
133–134 ЊC (Found: C, 75.39; H, 5.71%; m/z 270.170.
(c) 2-(4-Fluorophenyl)-6-methoxy-3,4-dihydronaphthalen-
1(2H)-one 5a (1.65 g, 6.11 mmol) was treated with sodium
borohydride as in the preceding experiment, to afford 2-(4-
fluorophenyl)-6-methoxy-1,2,3,4-tetrahydronaphthalen-1-ol 12a
(1.48 g, 89%) as a 2:1 mixture of cis and trans isomers (Found:
m/z 272.120 C17H17FO2 requires Mϩ, 272.12); νmax(KBr)/cmϪ1
3415 (br); δH 1.85–2.15 (2 H, m, 3-CH2), 2.8–3.1 (3 H, br m, 4-
CH2, 2-CH), 3.81 (3 H, s, OCH3), 4.71 (d, J 2.6, 1-CH, cis
isomer), 4.76 (d, J 9.0, 1-CH, trans isomer), 6.65 (m), 6.8 (m)
and 7.25 (m) (total 7 H, ArH).
C17H15FO2 requires C, 75.54; H, 5.59%; Mϩ, 270.106); νmax
/
cmϪ1 1672; δH 2.3–2.4 (2 H, m, 3-CH2), 2.9–3.2 (2 H, m, 4-CH2),
3.74 (1 H, dd, J 8, 8, 2-CH), 3.87 (3 H, s, OCH3), 6.72 (1 H, d, J
2.6, 5-CH), 6.85 (1 H, dd, J 8.7, 2.6, 7-CH), 7.02 (2 H, m,
2 × CH), 7.14 (2 H, m, 2 × CH), 8.06 (1 H, d, J 8.7, 8-CH); δC
29.15 (3-CH2), 31.39 (4-CH2), 53.37 (2-CH), 55.47 (OCH3),
1-Allyl-2-aryl-6-methoxy-1,2,3,4-tetrahydronaphthalenes 13
(a)
cis-2-(4-Methoxyphenyl)-6-methoxy-1,2,3,4-tetrahydro-
naphthalen-1-ol 12c (1.42 g, 5.0 mmol) in dichloromethane (5
ml) was added dropwise to a mixture of allyltrimethylsilane
(2.85 g, 25 mmol) and boron trifluoride–diethyl etherate (0.75
g, 5.3 mmol) in dichloromethane (10 ml) under nitrogen at
Ϫ78 ЊC. The mixture was stirred for 3 h and then quenched with
aq. sodium hydrogen carbonate. The organic phase was separ-
ated, washed, dried and evaporated, and the residue was chro-
matographed (10% diethyl ether–light petroleum) to yield 1-
allyl-2-(4-methoxyphenyl)-6-methoxy-1,2,3,4-tetrahydronaph-
thalene 13c (1.10 g, 72%) as a mixture of diastereoisomers (ca.
3:1). HPLC (reversed phase) gave a sample of the major iso-
mer (Found: m/z 208.180. C21H24O2 requires Mϩ, 308.178); δH
2
112.54, 113.35 (5-CH, 7-CH), 115.30 (2 × CH, d, JCF 20.7, 3Ј-
CH, 5Ј-CH), 126.31 (4a-C), 129.94 (2 × CH, d, 3JCF 7.4, 2Ј-CH,
6Ј-CH), 130.29 (8-CH), 135.73 (1Ј-C, d, JCF 3.6), 146.47 (8a-C),
161.76 (4Ј-C, d, 1JCF 244.1), 163.68 (6-C), 196.74 (C᎐O).
᎐
Reaction of 2-bromo-6-methoxy-3,4-dihydronaphthalen-1(2H)-
one 9 with aryl Grignard reagents
(a) 4-Fluorophenylmagnesium bromide (2  in diethyl ether,
15.2 ml) was added dropwise to a solution of 2-bromo-6-
methoxy-3,4-dihydronaphthalen-1(2H)-one
9 (7.58 g, 30
mmol) in toluene (40 ml) at 0 ЊC and the solution was allowed
to warm to ambient temperature before refluxing for 2 h. The
reaction mixture was quenched with aq. ammonium chloride
and the organic layer was washed, dried and evaporated. The
residuewaschromatographed[dichloromethane–light petroleum
(3:1)] to yield 2-(4-fluorophenyl)-6-methoxy-3,4-dihydronaph-
thalen-2(1H)-one 5a (1.2 g, 15%), indistinguishable from the
above sample.
1.95–2.05 (4 H, m, 3-CH , CH ᎐CHCH ), 2.95 (3 H, m, 2-CH,
᎐
2
2
2
4-CH2), 3.20 (1 H, dt, J 12.2, 3.9, 1-CH), 3.77 and 3.78 (each 3
H, s, OCH ), 4.72 (1 H, d, J 17.2, CH ᎐CH, trans), 4.82 (1 H, d,
᎐
2
3
J 10.2, CH ᎐CH, cis), 5.56 (1 H, m, CH ᎐CH), 6.65 (2 H, m, 5-
᎐
᎐
2
2
CH, 7-CH), 6.86 (2 H, d, J 8.6, 2 × 11-CH), 7.01 (1 H, d, J 9.2,
8-CH), 7.12 (2 H, d, J 8.6, 2 × 10-CH); δC 23.18 (3-CH2), 29.47
(4-CH ), 36.12 (CH ᎐CHCH ), 42.77, 43.95 (1-CH, 2-CH),
᎐
2
2
2
(b) In parallel fashion, phenylmagnesium bromide was
employed to provide 2-phenyl-6-methoxy-3,4-dihydronaph-
thalen-2(1H)-one 5b (13%), mp 118–119 ЊC (lit.,16 mp 113–
116 ЊC (Found: C, 81.2; H, 6.44%; m/z 252.112. C17H16O2
requires C, 80.93; H, 6.39%; Mϩ, 252.115).
55.13, 55.20 (OCH3), 111.21, 113.42, 113.53 (5-CH, 7-CH, 11-
CH), 115.51 (CH ᎐CH), 128.82 (10-CH), 130.46 (8-CH),
᎐
2
132.79, 136.46, 137.12 (4a-C, 8a-C, 9-C), 138.08 (CH ᎐CH),
᎐
2
157.75, 157.86 (6-C, 12-C).
(b) 2-(4-Fluorophenyl)-6-methoxy-1,2,3,4-tetrahydronaph-
thalen-1-ol 12a (0.36 g, 1.32 mmol) in dichloromethane (10 ml)
at 0 ЊC was treated with allyltrimethylsilane (0.18 g, 1.6 mmol)
and zinc iodide (0.51 g, 1.6 mmol) for 30 min. Isolation of the
product as in the above experiment provided 1-allyl-2-(4-
2-Aryl-6-methoxy-1,2,3,4-tetrahydronaphthalen-1-ols 12
(a) L-Selectride (10.2 ml, 1.0  in THF) was diluted with THF
(20 ml) and the solution was cooled to 0 ЊC, when (4-
methoxyphenyl)-6-methoxy-1-tetralone12 (2.82 g, 0.1 mol) in
THF was added over 5 min. The reaction mixture was stirred for
4 h when aq. sodium hydroxide (15 ml, 4 ) and aqueous hydro-
gen peroxide (5 ml, 30%) were added. Stirring was continued
overnight when the mixture was diluted with water (150 ml) and
extracted with diethyl ether. The organic extracts were washed,
dried and evaporated, and the residue was chromatographed
(30% diethyl ether–light petroleum) to yield cis-2-(4-
methoxyphenyl)-6-methoxy-1,2,3,4-tetrahydronaphthalen-1-ol
12c (1.78 g, 63%) as a single diastereoisomer, mp 103–104 ЊC
(Found: C, 76.27; H, 7.23%; m/z 284.147. C18H20O3 requires C,
76.03; H, 7.09%; Mϩ, 284.141); νmax(KBr)/cmϪ1 3422 (br); δH
1.63 (1 H, br s, OH), 1.86 (1 H, m, 3-CHax), 2.36 (1 H, m, Jgem
12.8, 3-CHeq), 2.78–3.00 (2 H, m, 4-CH2), 3.00 (1 H, ddd, J
10.0, 2.8, 2.8, 2-CH), 3.77 and 3.79 (each 3 H, s, OCH3), 4.62
(1 H, d, J 2.8, 1-CH), 6.67 (1 H, d, J 2.6, 5-CH), 6.75 (1 H, dd, J
2.6, 8.4, 7-CH), 6.90 (2 H, d, J 8.8, 2 × 11-CH), 7.23 (1 H, d, J
8.4, 8-CH) and 7.23 (2 H, d, J 8.8, 2 × 10-CH). Reduction with
sodium borohydride in methanol gave a 2:1 mixture of the cis
and trans diastereoisomers.
(b)2-Phenyl-6-methoxy-3,4-dihydronaphthalen-2(1H)-one5b
(0.3 g, 1.2 mmol) was dissolved in methanol (20 ml) with sodium
borohydride (0.1 g, 2.63 mmol). After 2 h at ambient temperature
the reaction mixture was poured into water and the product was
isolated via diethyl ether extraction and purified by column chro-
matography (1% methanol–dichloromethane) to yield 2-phenyl-
6-methoxy-1,2,3,4-tetrahydronaphthalen-1-ol12b(0.23g, 76%)as
a 2:1 mixture of cis and trans isomers (Found: m/z 254.132.
C17H18O2 requires Mϩ, 254.130); νmax(film)/cmϪ1 3414 (br).
fluorophenyl)-6-methoxy-1,2,3,4-tetrahydronaphthalene
13a
(0.332 g, 85%) as a mixture of diastereoisomers (ca. 2:1)
(Found: m/z 296.160. C20H21FO requires Mϩ, 296.158); δH 1.8–
2.3 and 2.5–3.3 (total 8 H), 3.794 and 3.796 (total 3 H, OCH3),
4.7 (1 H, br d, J 18.4, CH ᎐CH, trans), 4.8 (1 H, d, J 10,
᎐
2
CH ᎐CH, cis), 5.55 (1 H, m, CH ᎐CH), 6.6–6.8 and 6.9–7.2
᎐
᎐
2
2
(total 7 H, ArH).
(c) In an experiment parallel to (b) above, 2-phenyl-6-
methoxy-1,2,3,4-tetrahydronaphthalen-1-ol 12b was reacted
with allyltrimethylsilane to yield 1-allyl-2-phenyl-6-methoxy-
1,2,3,4-tetrahydronaphthalene 13b (78%) as a mixture of dia-
stereoisomers (ca. 2:1) (Found: m/z 278.164. C20H22O requires
Mϩ, 278.167).
2-Aryl-1-(3-hydroxypropyl)-6-methoxy-1,2,3,4-tetrahydronaph-
thalenes 14
(a) 1-Allyl-2-(4-methoxyphenyl)-6-methoxy-1,2,3,4-tetrahydro-
naphthalene 13c (0.311 g, 1.01 mmol) in THF at 0 ЊC was treat-
ed with borane–dimethyl sulfide complex (1 , 0.108 ml). The
solution was then stirred at ambient temperature for 3 h before
refluxing for 1 h. The cooled mixture was diluted with ethanol
and aq. sodium hydroxide (2 , 0.7 ml) and hydrogen peroxide
(30%, 0.12 ml) were added. After refluxing the mixture for 1 h, it
was poured into water and extracted with diethyl ether. The
organic extracts were washed, dried and evaporated. The resi-
due was chromatographed (70% diethyl ether–light petroleum)
to afford 1-(3-hydroxypropyl)-2-(4-methoxyphenyl)-6-methoxy-
1,2,3,4-tetrahydronaphthalene 14c (1.151 g, 46%) (Found: m/z
326.188. C21H26O3 requires Mϩ, 326.188); δC(major diastereo-
1514
J. Chem. Soc., Perkin Trans. 1, 1997

View Article Online
isomer) 22.75, 27.03, 29.09, 31.29 (4 × CH2), 42.63, 43.59 (1-
CH, 2-CH), 55.06 (2 × OCH3), 63.00 (CH2OH), 111.14, 113.44,
113.55 (5-CH, 7-CH, 11-CH), 128.64 (10-CH), 130.03 (8-CH),
133.60, 136.41, 136.93 (4a-C, 8a-C, 9-C), 157.11 and 157.68
(6-C, 12-C).
(b) In parallel experiments, 1-allyl-2-(4-fluorophenyl)-6-
methoxy-1,2,3,4-tetrahydronaphthalene 13a and 1-allyl-2-
phenyl-6-methoxy-1,2,3,4-tetrahydronaphthalene 13b yielded,
respectively, 2-(4-fluorophenyl)-1-(3-hydroxypropyl)-6-methoxy-
1,2,3,4-tetrahydronaphthalene 14a (59%) (Found: m/z 314.71.
C20H23FO2 requires Mϩ, 314.168) and 1-(3-hydroxypropyl)-6-
methoxy-2-phenyl-1,2,3,4-tetrahydronaphthalene 14b (34%)
(Found: m/z 296.180. C20H40O2 requires Mϩ, 296.178), both as
(2:1) mixtures of diastereoisomers.
diethyl ether) to provide the phenolic iodide 15c (0.1 g, 65%)
(Found: m/z 408.060. C19H21IO2 requires Mϩ, 408.059). This
iodide (68.6 mg, 0.168 mmol) was dissolved in tert-butyl alco-
hol (17 ml) with potassium tert-butoxide (22 mg, 0.196 mmol)
and the solution was refluxed (nitrogen) overnight. The reaction
mixture was then diluted with water (50 ml) and extracted with
ethyl acetate. The organic extracts were dried and evaporated
and the residue (39 mg) was extracted with N,N-dimethyl-
formamide to yield (3aR*,4R*,10aS*)-4-phenyl-1,2,3,3a,4,5,-
6,8-octahydrocyclopenta[d]naphthalen-8-one 16c as a single dia-
stereoisomer with low solubility in common organic solvents
(Found: m/z 280.148. C19H20O2 requires Mϩ, 280.146);
νmax(KBr)/cmϪ1 1652, 1609, 1516; δH ([2H7]DMF) 1.5–3.0 (12 H,
m, 2 × CH, 5 × CH2), 6.04 (1 H, dd, J 1.9, 9.8, 3-CH), 6.13 (1
H, d, J 1.9, 1-CH), 6.78 (2 H, d, J 8.3, 3Ј-CH, 5Ј-CH), 7.05 (2 H,
d, J 8.3, 2Ј-CH, 6Ј-CH), 7.11 (1 H, d, J 9.8, 4-CH), 9.38 (1 H, s,
OH); δC([2H7]DMF) 22.82, 26.43, 28.57, 29.08 (7-CH2, 9-CH2,
10-CH2, 11-CH2), 39.92 (8-CH2), 40.92 (6-CH), 51.75 (4a-C),
53.36 (5-CH), 115.67 (3Ј-CH, 5Ј-CH), 125.01, 127.47 (1-CH, 3-
CH), 129.01 (2Ј-CH, 6Ј-CH), 134.47 (1Ј-C), 156.87 (4Ј-C),
1-(But-3-enyl)-2-(4-methoxyphenyl)-6-methoxy-1,2,3,4-tetra-
hydronaphthalene 21
1-(3-Hydroxypropyl)-2-(4-methoxyphenyl)-6-methoxy-1,2,3,4-
tetrahydronaphthalene 14c (0.3 g, 0.9 mmol) in dichlorometh-
ane (2 ml) was added to a stirred suspension of pyridinium
chlorochromate (0.3 g, 1.35 mmol) in dichloromethane (2 ml).
After 2 h diethyl ether (10 ml) was added. The supernatant,
with diethyl ether washings from the residue, was passed
through a kieselguhr pad and evaporated; column chromato-
graphy (2:1 light petroleum–diethyl ether) gave the aldehyde 20
(0.22 g, 70%) as a (2:1) mixture of diastereoisomers (Found: m/z
324.173. C21H24O3 requires Mϩ, 324.173); νmax/cmϪ1 1723; δH
1.5–2.4 and 2.7–3.3 (total 10 H), 3.786, 3.791, 3.793 and 3.810
(total 6 H, all s, OCH3), 6.6–7.2 (7 H, ArH), 9.50 and 9.65 (total
1 H, both t, J 2, CHO). The aldehyde (0.25 g) was added to
methylene(triphenyl)phosphorane [from methyltriphenylphos-
phonium iodide (0.47 g, 1.16 mmol) and butyllithium (1.28
mmol)] in THF (10 ml) at Ϫ78 ЊC. The reaction mixture was
allowed to warm to ambient temperature over 60 min, when
water (10 ml) was added and the mixture was extracted with
diethyl ether. The extracts were washed, dried and evaporated.
The residue was chromatographed (9:1 light petroleum–diethyl
ether) to afford the title but-3-enyltetralin 21 (0.144 g, 58%) also
as a (2:1) mixture of diastereoisomers (Found: m/z 322.191.
C22H26O2 requires Mϩ, 322.193); δH 1.5–2.2 and 2.5–3.1 (total
10 H), 3.781, 3.789, 3.795 and 3.803 (total 6 H, all s, OCH3),
157.27 (4-CH), 167.15 (8a-C), 186.13 (C᎐O).
᎐
(b)
2-(4-Fluorophenyl)-1-(3-hydroxypropyl)-6-methoxy-
1,2,3,4-tetrahydronaphthalene 14a (0.69 g, 2.2 mmol) was
refluxed in acetonitrile (40 ml) with sodium iodide (1.8 g, 12.1
mmol) and chlorotrimethylsilane (1.21 g, 11.0 mmol) for 24 h.
The mixture was diluted with water (40 ml) and extracted with
diethyl ether. The extracts were washed, dried, evaporated and
the residue was purified by chromatography (2:1 light
petroleum–diethyl ether) to yield the iodide 15a (0.624 g, 67%)
(Found: m/z 410.052. C19H20FIO requires Mϩ, 410.054). This
iodide (0.1 g, 0.246 mmol) was dissolved in tert-butyl alcohol
(10 ml) with potassium tert-butoxide (34 mg, 0.3 mmol) and the
solution was heated at reflux for 48 h, after which it was poured
into water (25 ml). The product mixture was extracted with
dichloromethane and the extracts were washed, dried and
evaporated; the residue was purified by chromatography (1:1 to
1:2 light petroleum–diethyl ether) to afford (3aR*,4R*,10aS*)-
4-(4-fluorophenyl)-1,2,3,3a,4,5,6,8-octahydronaphthalen-8-one
16a as a single diastereoisomer (10.6 mg, 15%) (Found: m/z
282.142. C19H18FO requires Mϩ, 282.142); νmax(KBr)/cmϪ1
1659, 1621, 1601; δC 21.4, 28.5, 33.1, 35.6, 36.7 (5 × CH2), 45.5
2
4.8–5.0 (2 H, m, CH᎐CH ), 5.5–5.8 (1 H, m, CH᎐CH ), 6.6–6.9
(CH), 52.0 (C), 54.0 (CH), 115.4 (3Ј-CH, 5Ј-CH, d, JCF 20.8),
᎐
᎐
2
2
and 7.0–7.2 (total 7 H, ArH).
126.3, 126.6 (2 × CH), 129.2 (2Ј-CH, 6Ј-CH, d, 3JCF 7.4), 139.5
(C), 153.4 (CH), 161.5 (C, d, 1JCF 250), 165 (C), 187 (C᎐O).
᎐
1-(4-Hydroxybutyl)-2-(4-methoxyphenyl)-6-methoxy-1,2,3,4-
tetrahydronaphthalene 22a
(c) 1-(3-Hydroxypropyl)-2-phenyl-6-methoxy-1,2,3,4-tetra-
hydronaphthalene 14b (80 mg, 0.27 mmol) was treated with
chlorotrimethylsilane (200 mg, 1.8 mmol) and sodium iodide
(310 mg, 2.1 mmol) in acetonitrile (5 ml) as in the preceding
experiment: isolation and purification of the product in similar
fashion gave the iodide 15b (40 mg, 38%) [Found: m/z 264.152.
C19H21IO requires (Mϩ Ϫ HI), 264.151]. Cyclisation of this iod-
ide (28 mg, 0.07 mmol) with potassium tert-butoxide (10 mg,
0.09 mmol) in tert-butanol (10 ml) as described above afforded
(3aR*,4R*,10aS*)-4-phenyl-1,2,3,3a,4,5,6,8-octahydronaphthal-
en-8-one 16b as a single diastereoisomer (2 mg, 10%) (Found:
m/z 264.149. C19H20O requires Mϩ, 264.151); νmax(KBr)/cmϪ1
1657; δH 1.5–2.7 (12 H), 6.15–6.3 (2 H, br m, 1-CH, 3-CH), 6.9–
7.35 (6 H, 4-CH, PhH).
(d) Using the above protocol for iodination–demethylation,
1-(4-hydroxybutyl)-2-(4-methoxyphenyl)-6-methoxy-1,2,3,4-
tetrahydronaphthalene 22a (0.25 g, 0.74 mmol) was treated with
chlorotrimethylsilane (0.44 g, 4.05 mmol) to yield after chroma-
tography (4:1 to 1:1 light petroleum–diethyl ether) the phenolic
iodide 22b (0.105 g, 34%) (Found: m/z 422.074. C20H23IO2
requires Mϩ, 422.074). Monodemethylated material was also
isolated. The phenolic iodide 22b (75 mg) was cyclised with
potassium tert-butoxide as in the preceding experiments. Prod-
uct isolation as before and chromatography (3:1 to 1:1 light
petroleum–ethyl acetate) afforded (4aR*,5R*,11aS*)-5-(4-hydr-
oxyphenyl)-2,3,4,4a,5,6,7,9-octahydro-1H-benzo[d]naphthalen-
Borane–dimethyl sulfide (2 , 0.35 ml) was added dropwise to
1-(but-3-enyl)-2-(4-methoxyphenyl)-6-methoxy-1,2,3,4-tetra-
hydronaphthalene 21 (0.4 g, 1.24 mmol) in hexane (20 ml) at
0 ЊC. After 3.5 h at ambient temperature the solution was
refluxed for 1.5 h. Ethanol (10 ml), aq. sodium hydroxide (2 ,
5 ml) and hydrogen peroxide (30%, 0.25 ml) were added sequen-
tially to the cooled (0 ЊC) mixture, which was then refluxed for 1
h when it was cooled, diluted with water and extracted with
dichloromethane. The extracts were washed, dried and evapor-
ated to leave a residue which was chromatographed (1:2 light
petroleum–diethyl ether) to afford the title alcohol 22a (0.319 g,
76%) as a (2:1) mixture of diastereoisomers (Found: m/z
340.025. C22H28O3 requires Mϩ, 340.204); νmax(film)/cmϪ1 3377
(br).
Aryltricyclospirodienones 16 and 18
(a) A mixture of 1-(3-hydroxypropyl)-2-(4-methoxyphenyl)-6-
methoxy-1,2,3,4-tetrahydronaphthalene 14c (0.122 g, 0.375
mmol), sodium iodide (0.31 g, 2.1 mmol), chlorotrimethylsilane
(0.2 g, 1.8 mmol) and acetonitrile (10 ml) were heated together
under reflux overnight (nitrogen) and then poured into ice–
water (50 ml). The product mixture was extracted with diethyl
ether and the extracts were washed, dried and evaporated. The
residue was purified by chromatography (1:1 light petroleum–
J. Chem. Soc., Perkin Trans. 1, 1997
1515

View Article Online
E. Snoeck and R. Decoster, Breast Cancer Res. Treat., 1994, 30, 89;
9-one 18 as a single diastereoisomer (3.4 mg, 6.5%) (Found: m/z
294.161. C20H22O2 requires Mϩ, 294.162; νmax(KBr)/cmϪ1 3435
(br), 1655, 1613, 1515; δH 0.7–3.2 (14 H), 6.17 (1 H, br d, J 2.0,
1-CH), 6.33 (1 H, dd, J 10.4, 2.0, 3-CH), 6.77 (2 H, d, J 8.6, 2Ј-
CH, 6Ј-CH), 7.01 (2 H, d, J 8.6, 3Ј-CH, 5Ј-CH), 7.64 (1 H, d, J
10.4, 4-CH).
P. V. Plourde, M. Dyroff and M. Dukes, Breast Cancer Res. Treat.,
1994, 30, 103; A. Lipton, L. M. Demers, H. A. Harvey, K. B.
Kambic, H. Grossberg, C. Brady, H. Adlercruetz, P. F. Trunet,
R. J. Santen, P. V. Plourde, M. Dyroff and M. Dukes, Cancer, 1995,
75, 2132; N. Oohata, Y. Hori, Y. Yamagishi, T. Fujita, S. Takase,
M. Yamashita, H. Terano and M. Okuhara, J. Antibiot., 1995, 48,
757; M. Lourdusamy, F. Labrie and S. M. Singh, Synth. Commun.,
1995, 25, 3655; H. I. Holland, S. Kumaresan and G. Lakshmaiah,
Can. J. Chem., 1995, 73, 2185; E. J. T. Dasilva, M. L. S. E. Melo and
A. S. C. Neves, J. Chem. Soc., Perkin Trans. 1, 1996, 1649; V. C. O.
Njar, J. Duerkop and R. W. Hartmann, Steroids, 1996, 61, 138.
3 C. Geynet, C. Millet, H. Truong and E. E. Baulieu, Gynaecol.
Invest., 1972, 3, 2; V. C. Jordan and B. Gosden, Mol. Cell.
Endrocrinol., 1982, 27, 291.
4 We acknowledge valuable discussions with, and personal
communications from, Dr F. T. Boyle, Zeneca Pharmaceuticals,
Alderly Park, Macclesfield, UK.
5 D. F. Covey and W. F. Hood, Cancer Res., Suppl., 1982, 42, 3327;
G. A. Flynn, J. O. Johnston, C. L. Wright and B. W. Metcalf,
Biochem. Biophys. Res. Commun., 1981, 103, 913.
References
1 Cytochrome P-450; Structure, Mechanism and Biochemistry, ed.
P. R. Ortiz de Montellano, Plenum Press, 1986, pp. 217–271;
M. Akhtar, D. E. Stevenson and J. N. Wright, J. Chem. Soc., Perkin
Trans. 1, 1988, 2043; J. Y. Kellis and L. E. Vickery, J. Biol. Chem.,
1987, 262, 4413, 8840; M. Ahkhtar and J. N. Wright, Nat. Prod.
Rep., 1991, 527; E. R. Simpson, M. S. Mahendroo, G. D. Means,
M. W. Kilgore, M. M. Hinshelwood, S. Graham-Lorence,
B. Amarneh, Y. Ito, C. R. Fisher, M. D. Michael, C. R. Mendelson
and S. E. Bulun, Endocr. Rev., 1994, 15, 342.
2 (a) J. O. Johnston and B. W. Metcalf, Novel Approaches to
Cancer Chemotherapy, ed. P. S. Sunkara, Academic Press, 1984,
p. 307; A. Manni and R. J. Santen, Pharmacology and Clinical Uses
of Inhibitors of Hormone Secretion, Bailliere-Tindall, 1987, p. 255;
D. F. Covey, Sterol Biosynthesis Inhibitors: Pharmaceutical and
Agrochemical Aspects, ed. D. Berg and M. Plempel, Ellis Horwood,
1988, p. 534; A. M. H. Brodie, Design of Enzyme Inhibitors as Drugs,
ed. M. Sander and H. J. Smith, Oxford University Press, 1989, p. 503;
(b) Inter alia P. K. Siiteri and E. A. Thompson, J. Steroid. Biochem.,
1975, 6, 317; R. W. Bruggemeier, E. E. Floyd and R. E. Councell,
J. Med. Chem., 1978, 21, 1007; D. F. Covey, W. F. Hood and V. D.
Parikh, J. Biol. Chem., 1981, 256, 1076; B. W. Metcalf, C. L. Wright,
J. P. Burkhart and J. O. Johnston, J. Am. Chem. Soc., 1981, 103,
3221; P. A. Marcotte and C. H. Robinson, Biochemistry, 1982, 21,
2773; P. R. Ortiz de Montellano, K. L. Kunze, H. S. Beilan and
C. Wheeler, Biochemistry, 1982, 21, 1331; M. G. B. Drew, J. Mann
and B. Pietrzak, J. Chem. Soc., Chem. Commun., 1985, 1191; J. N.
Wright, M. R. Calder and M. Akhtar, J. Chem. Soc., Chem.
Commun., 1985, 1733; L. Tan and A. Petit, Biochem. Biophys. Res.
Commun., 1985, 128, 613; M. J. Shih, M. H. Carrell, H. L. Carrell,
C. L. Wright, J. O. Johnston and C. H. Robinson, J. Chem.
Soc., Chem. Commun., 1987, 213; W. E. Childers and C. H.
Robinson, J. Chem. Soc., Chem. Commun., 1987, 320; J. T. Kellis, Jr,
W. E. Childers, C. H. Robinson and L. E. Vickery, J. Biol. Chem.,
1987, 262, 4421; M. Akhtar, J. N. Wright, P. T. van Leersum and
S. G. Chamberlin, J. Chem. Soc., Perkin Trans. 1, 1989, 1647; P. A.
Cole and C. H. Robinson, J. Med. Chem., 1990, 33, 2933; J. Am.
Chem. Soc., 1991, 113, 8130; R. W. Bruggemeier, Breast Cancer Res.
Treat., 1994, 30, 31; H. V. Bossche, H. Moereels and L. M. H.
Koymans, Breast Cancer Res. Treat., 1994, 30, 43; J. Geelen, H. J. J.
Loozen, G. H. Deckers, R. Deleeuw, H. J. Kloosterboer and S. W. J.
Lamberts, J. Steroid Biochem. Mol. Biol., 1993, 44, 681; W. Wouters,
6 Energyminimisationswereestimated usingMacroModel, version 4.0,
with MM2 force field parameters, and employing ‘Monte Carlo’
global searches; we thank Dr M. Luszniak for these calculations.
7 O. Hares, D. Hobbs-Mallyon and D. A. Whiting, J. Chem. Soc.,
Perkin Trans. 1, 1993, 1481.
8 N. Campbell and D. Kidd, J. Chem. Soc., 1954, 2154.
9 F. M. Dean and K. B. Hindley, Tetrahedron Lett., 1972, 15, 1445.
10 A. S. Hussey and R. R. Herr, J. Org. Chem., 1959, 24, 843.
11 J. K. Stille and R. A. Newsom, J. Org. Chem., 1961, 26, 1375.
12 A generous gift of this material from Zeneca Pharmaceuticals is
acknowledged.
13 These assays were conducted by courtesy of Professor P. J. Nicholls,
Welsh School of Pharmacy, University of Wales, Cardiff, to whom
we express our thanks. Aromatase activity was measured by the
3
release of H2O from (1β-3H)androstendione (0.4 µ) on incubation
with placental microsomes (0.15 mg protein mlϪ1) at pH 7.4 and
35 ЊC for
5 min (km for aromatase 25 n). Final inhibitor
concentrations 5 × 10Ϫ6  were employed.
14 E. A. Thompson, Jr and P. K. Siiteri, J. Biol. Chem., 1974, 249, 5373;
P. E. Graves and H. A. Salhanick, Endocrinology, 1979, 105, 52;
C. A. Laughton, R. McKenna, S. Neidle, M. Jarman, R. McCague
and M. G. Rowlands, J. Med. Chem., 1990, 33, 2673.
15 T. R. Kasturi and T. Arunachalam, Can. J. Chem., 1968, 46, 3625.
16 D. Lednicer, J. C. Babcock, S. C. Lyster and G. W. Duncan, Chem.
Ind., 1963, 408.
Paper 6/08338B
Received 11th December 1996
Accepted 27th January 1997
1516
J. Chem. Soc., Perkin Trans. 1, 1997