Synthesis and cytotoxic properties of a series of bicyclo[3.2.1]octane α-methylene ketones

Article abstract of DOI:10.1016/S0040-4039(01)00235-0

The 8-chlorobicyclo[3.2.1]oct-6-ene 3 has been prepared by a [3+2] cycloaddition route, and converted to the bicyclic α-methylene ketones 6 and 8-12, some of which showed cytotoxic properties.

Full text of DOI:10.1016/S0040-4039(01)00235-0

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 2729–2731
Synthesis and cytotoxic properties of a series of
bicyclo[3.2.1]octane a-methylene ketones
J. A. Miller,a,* G. M. Ullah,a G. M. Welsha and P. Mallonb
aDepartment of Medicinal Chemistry, Wellcome Research Laboratories, Langley Court, Beckenham, Kent BR3 3BS, UK
bChemistry Department, University of St. Andrews, St. Andrews, Fife, Scotland KY16 9AL, UK
Received 3 November 2000; revised 26 January 2001; accepted 7 February 2001
Abstract—The 8-chlorobicyclo[3.2.1]oct-6-ene 3 has been prepared by a [3+2] cycloaddition route, and converted to the bicyclic
a-methylene ketones 6 and 8–12, some of which showed cytotoxic properties. © 2001 Elsevier Science Ltd. All rights reserved.
Over the years, the literature has provided many exam-
ples of a,b-unsaturated ketones which possess interest-
ing biological properties. These range in structure from
simple monocyclics, like the antibiotic methylenomycin
B1 to more complex and usually more lipophilic sub-
stances, like the punaglandins2 and the clavulones,3
which have led to new approaches to anti-viral and
anti-inflammatory drugs.4 However, there is also a huge
number of polycyclic terpenoid structures, e.g. the
diterpenes of the Rabdosia family, such as enmein 1 and
the ent-kaurane umbrosin A 2, in which an a-methylene
cyclopentanone is embedded in a bicyclo[3.2.1]octane
framework.5 Compounds such as 1 and 2 exhibit a
range of anti-tumour and anti-microbial properties, and
there is good evidence that the enone group is essential
for their biological activity.5
an a-methylene cyclopentanone sub-unit.6 Arising from
observations made in our [3+2] cycloaddition approach
to the helminthosporanes,7 it seemed reasonable to
speculate that a bicyclo[3.2.1]oct-6-ene, such as 3,
would act as a one-pot precursor for an a-methylene
ketone functionality, since the double bond in 3 is part
of BOTH an allyl ether and a vinyl sulphide.8 More-
over, 3 was likely to be accessible via a [3+2] cycloaddi-
tion of the allylic chloride 4 to the alkyne 5—a
particular example of a general reaction described some
years ago.9 It was also envisaged that this de novo
synthetic approach would allow facile incorporation of
different groups, including other rings, around the
cyclohexane ring of the bicyclics, e.g. the 8-chloro
group in 3 would, in principle, allow another function-
ality to be introduced at the site equivalent to where 2
has a hydroxyl group. In principle, this direct and
flexible route10 to a-methylene cyclopentanones could
offer significant advantages over existing methodology,
which is largely based on interconversions of natural
products,5 or on 1,2-rearrangements of synthetic bicy-
clo[4.2.0] systems.6b This report describes the synthesis
of several of our targets via [3+2] cycloaddition chem-
istry, and briefly describes their anti-tumour
properties.11
This background suggested that there would be poten-
tial interest in simpler lipophilic structures containing
The synthesis of 5 and its [3+2] cycloaddition to the
allylic chloride 4 (which contains about 15% of its
allylic isomer 3-chloro-3-methylcyclohexene—both give
the same allyl cation) are outlined in Scheme 1. The
cycloaddition goes in reasonable yield, and is amenable
to scale-up to the 50 g level. Thereafter, we established
conditions for the selective hydrolyses of each of the
functionalities of 3—the ether/sulphide and the chlo-
rine. The first of these, hydrolysis to the enone 6, was
Keywords: [3+2] cycloaddition; Hg(II) mediated hydrolysis; alkynyl
and vinyl sulphide; bicyclo[3.2.1]octane a-methylene ketones; cytotox-
icity.
* Corresponding author. Present address: Protherics PLC, Beechfield
House, Lyme Green Business Park, Macclesfield, Cheshire SK11
0JL, UK. Tel.: +44 (0)1625 500555; fax: +44 (0)1625 500666; e-mail:
allen.miller@protherics.com
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)00235-0

2730
J. A. Miller et al. / Tetrahedron Letters 42 (2001) 2729–2731
achieved with mercury(II) chloride in refluxing aqueous
acetone, whereas the latter was accomplished with mer-
cury(II) perchlorate at room temperature, and yielded
the alcohol 7—see Scheme 1. Availability of 7 then
allowed entry to the analogues 8–10. The similarity in
these hydrolysis conditions12 suggested that all of the
functionalities in 3 might be altered in the same opera-
tion, and this was indeed the case, e.g. the preparation
of 11 and 12 from 3, as shown in Scheme 1.
was found to be active against two tumour cell lines-
P388 (IC50 0.53 mM) and DLD-1 (IC50 9.7 mM), both
run as proliferative assays. The analogues 8 and 9 were
considerably weaker (about 10% growth inhibition at
10 mM in each case). The chloride 6 was then found to
kill mice after dosing at levels ranging from 65–225
mg/kg. Furthermore, when the 8-fluoro derivative 10
was subject to the in vitro assays, it was observed that
it was so volatile that it killed cells in neighbouring
wells. At this point work on this series of compounds
ceased, and we would caution other investigators to
bear in mind the potential for toxicity and volatility in
these compounds and their analogues.
Structural assignment of the enones in Scheme 1 is
based on appropriate analytical data, but the enone
functionality was easily recognised13 from the consistent
broadened singlets (J<0.5 Hz) that appeared at l 5.2
0.08 and 5.99 0.06 ppm in the proton NMR spectra,
and from the strong infrared absorptions at 1720 5 and
1635 5 cm−1. Similarly, the proton NMR spectra are
diagnostic for the relative configuration at C-8 and all
the compounds reported here show a doublet (J=5 Hz)
for the C-8 proton. This is well established in the
literature for the endo series (i.e. in which the functional
group at C-8 is anti to the enone group), and is readily
distinguished from the exo series, for which the vicinal
coupling constant (<1 Hz) is considerably smaller.14
The endo configuration of the starting chloride 3, and
the observed retention of configuration in its analogues,
are expected15 in bicyclo[3.2.1]oct-6-enes, on account of
participation by the 6-ene in stabilisation of positive
charge development at the 8-position, e.g. in solvolyses,
or Lewis acid-mediated substitutions at C-8.
Details of key experimental procedures are given below.
Preparation of 1-methyl-6-methoxymethyl-7-phenyl-
sulphenyl-8-chlorobicyclo[3.2.1]oct-6-ene 3. The cyclo-
adduct was prepared by addition at −30°C of 1-methyl-
3-chlorocyclohexene 4 (containing about 15% of its
allylic isomer, 3-chloro-3-methylcyclohexene) (1.40 g,
10.7 mmol) to a stirred mixture of 1-phenylsulphenyl-2-
methoxymethylethyne 5 (1.78 g, 10.0 mmol) and
reagent grade zinc chloride (3.0 g, 22.0 mmol) sus-
pended in dichloromethane (20 ml). Thereafter the
sequence followed the published procedure,9 and, after
chromatographic purification, 3 (52–60%) was isolated
1
as a very pale yellowish oil. Key H NMR data include
l 0.85 (s, 3H, 1-CH3), 3.32 (s, 3H, CH2OCH3), 4.12 (d,
2
J=5 Hz, 1H, 8-CHCl) and 4.35 (d, J=12 Hz, 1H, one
of CH2OCH3) and 4.15 (d, 2J=12 Hz, 1H, one of
CH2OCH3) ppm.
Having established a general route to 8-substituted
bicyclo[3.2.1]octan-7-ones, the ‘parent’ compound 6
Scheme 1. Reagents: (i) nBuLi, PhSCl, THF, −78°C; (ii) ZnCl2, CH2Cl2; (iii) HgCl2, aq. acetone, D; (iv) Hg(ClO4)2, aq. DME,
20°C; (v) Hg(ClO4)2, ROH, 20°C; (vi) DAST, CH2Cl2, −50 to 20°C; (vii) Ph3P, CBr4, MeCN, reflux.

J. A. Miller et al. / Tetrahedron Letters 42 (2001) 2729–2731
2731
General procedure for the preparation of 6-methylene
bicyclo[3.2.1]oct-6-enes. Compound 6: The 8-chlorobi-
cyclo[3.2.1]oct-6-ene 3 (3.08 g, 10 mmol) and mer-
cury(II) chloride (4.70 g, 17.3 mmol) were dissolved in
20% aqueous acetone (50 ml) and the mixture refluxed
for 5 days. Solvents were then evaporated and the
residue shaken with water (25 ml) and chloroform (25
ml). The aqueous layer was rewashed with chloroform
(2×20 ml) and the combined chloroform extracts were
washed with water (20 ml). The chloroform solution
was dried (MgSO4) and the solvent evaporated. The
residue was purified by flash chromatography and
enone 6 (1.78 g, 58%) was isolated as a very pale yellow
oil following elution with ether: 40–60°C petrol (1:10).
1999, 12, 395–400; (c) Rossi, A.; Kapahi, P.; Natoli, G.;
Takahashi, T.; Chen, Y.; Karin, M.; Santoro, M. G.
Nature 2000, 403, 103–108.
5. Fujita, E.; Node, M. In Progress in the Chemistry of
Organic Natural Products; Herz, W.; Grisebach, H.;
Kirby, G. W.; Tamm, C., Eds.; 1984; Vol. 46, pp. 77–157.
6. Others have reported similar mimicry, e.g.: (a) Kakiuchi,
K.; Ue, M.; Takeda, M.; Tadaki, T.; Kato, Y.;
Nagashima, T.; Tobe, Y.; Koike, H.; Ida, N.; Odaira, Y.
Chem. Pharm. Bull. 1987, 35, 617–631; (b) Ue, M.;
Kobiro, K.; Kakiuchi, K.; Tobe, Y.; Odaira, Y. Chem.
Lett. 1987, 1903–1904; and (c), in the quadrone series,
Hirota, A.; Nakagawa, M.; Sakai, H.; Isogai, A. J.
Antibiot. 1982, 35, 783–787; Hirota, A.; Nakagawa, M.;
Hirota, H.; Takahashi, T.; Isogai, A. J. Antibiot. 1986,
39, 149–152.
1
Key H NMR data include l 1.04 (s, 3H, 1-CH3), 3.97
(d, J=5 Hz, 1H, 8-CHCl), 5.20 (bs, 1H, one of ꢀCH2)
and 5.93 (bs, 1H, one of ꢀCH2) ppm. Strong infrared
absorptions for the a-methylene cyclopentanone
appeared at 1725 and 1640 cm−1.
7. Gray, B. D.; Miller, J. A. J. Chem. Soc., Chem. Commun.
1987, 1136–1137.
8. (a) Bibang Bi Ekogha, C.; Ruel, O.; Julia, S. A. Tetra-
hedron Lett. 1983, 24, 4825–4828; (b) Ruel, O.; Bibang Bi
Ekogha, C.; Julia, S. A. Tetrahedron Lett. 1983, 24,
4829–4832; (c) Wada, M.; Nakamura, H.; Taguchi, T.;
Takei, H. Chem. Lett. 1977, 345–348.
9. (a) Gray, B. D.; McMillan, C. M.; Miller, J. A.; Moore,
M. Tetrahedron Lett. 1987, 28, 235–238; (b) Gray, B. D.;
McMillan, C. M.; Miller, J. A.; Ullah, G. M. Tetrahedron
Lett. 1987, 28, 689–692. The term [3+2] cycloaddition is
used in a structural sense only and has no implications
about mechanism.
10. For another direct route, see: Tius, M. A.; Patterson, G.
M. L.; Astrab, D. P. J. Antibiot. 1985, 38, 1061–1067.
This [3+2] cycloaddition of an allenyl ether yields the
a-methylene cyclopentanone directly using quite different
chemistry from that described here. For more recent
[3+2] cycloadditions leading to enones, see: Hartley, R.
C.; Caldwell, S. T. J. Chem. Soc., Perkin Trans. 1 2000,
477–501.
11. Anti-tumour and other medicinal properties have also
been well documented in a-methylene g-lactones—see:
Sharma, J. C.; Sharma, R. P. Heterocycles 1986, 24,
441–457 for references. For a review of their synthesis,
see: Grieco, P. A. Synthesis 1975, 67–82.
Mercury(II) perchlorate mediated hydrolysis-preparation
of 8-hydroxybicyclo[3.2.1]oct-6-ene 7. Mercury(II) per-
chlorate was freshly prepared by addition of mercuric
oxide (7.60 g) to 70% perchloric acid (8 ml) in
dimethoxyethane (DME) (90 ml). Gentle heating was
required to dissolve all the solid. After allowing to cool,
water (6.8 ml) was added, followed by a solution of
cycloadduct 3 (10 g, 32.4 mmol) in DME (20 ml). The
reaction mixture was stirred at room temperature for
3.5 hours, then water (200 ml) was added and the
reaction mixture extracted with ether (3×50 ml). The
combined organic extracts were washed with dilute
hydrochloric acid (100 ml) and then with a saturated
sodium chloride solution and dried (MgSO4). After
removing the solvent, the crude product was purified by
flash column chromatography, eluting with ether: 40–
60°C petrol (1:2) and the pure bicyclic alcohol 7 (6.02 g,
64%) was isolated as a clear, thick syrup. Key 1H NMR
data include l 0.82 (s, 3H, 1-CH3), 2.47 (bs, 1H, OH),
3.27 (s, 3H, CH2OCH3), 3.69 (d, J=5 Hz, 1H, 8-CH-
2
OH), 4.17 (d, J=12 Hz, 1H, one of CH2OCH3) and
2
4.37 (d, J=12 Hz, 1H, one of CH2OCH3) ppm. In the
infrared spectrum a strong broad absorption appeared
for the hydroxy group at 3430 cm−1.
12. To appreciate the importance of the mercury counterion,
see: McKillop, A.; Ford, M. E. Tetrahedron 1974, 30,
2467–2475.
13. See references 5, 6a and 6b for many examples within the
same ranges.
References
14. (a) Piers, E.; Britton, E. W.; Geraghty, M. B.; Keziere, R.
J.; Kido, F. Can. J. Chem. 1975, 53, 2838–2848; (b) Piers,
E.; Geraghty, M. B.; Smillie, R. D.; Soucy, M. Can. J.
Chem. 1975, 53, 2849–2864.
1. (a) Haneishi, T.; Kitahara, N.; Takiguchi, Y.; Arai, M.;
Sugawara, S. J. Antibiot. 1974, 27, 386–392; (b) Haneishi,
T.; Terahara, A.; Arai, M.; Hata, T.; Tamura, C. J.
Antibiot. 1974, 27, 393–399.
15. Hess, B. A. J. Am. Chem. Soc. 1971, 93, 1000–1003. The
same phenomenon is studied much more in the 7-substi-
tuted norbornene system, e.g. Winstein, S.; Shatavsky,
M.; Norton C.; Woodward, R. B. J. Am. Chem. Soc.
1955, 77, 4183–4184; Brookhart, M.; Diaz, A.; Winstein,
S. J. Am. Chem. Soc. 1966, 88, 3135–3136.
2. Baker, B. J.; Scheuer, P. J. J. Nat. Prod. 1994, 57,
1346–1353.
3. Grechkin, A. N. J. Lipid Mediators Cell Signalling 1995,
11, 205–218.
4. (a) Santoro, M. G. Trends Microbiol. 1997, 5, 276–281;
(b) Santoro, M. G.; Roberts, S. M. Drug News Perspect.
.
.