Asymmetric alkylation of diphenylmethane derivatives using (-)-sparteine

Article abstract of DOI:10.1016/j.tetlet.2003.11.138

Alkylation of 2-oxygenated diphenylmethane derivatives using sec-butyllithium and (-)-sparteine gave enantiomeric excesses of up to 60% with allyl bromide but alkylations with methyl electrophiles were poorly selective. When compounds with a free hydroxy in the 2-position were alkylated the selectivity was reversed.

Full text of DOI:10.1016/j.tetlet.2003.11.138

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 1191–1193
Asymmetric alkylation of diphenylmethane derivatives using
())-sparteine
James A. Wilkinson,a,* Stephen B. Rossington,a John Leonardb and Nigel Husseinc
aCentre for Drug Design, Biosciences Research Institute, Cockcroft Building, University of Salford, Salford M5 4WT, UK
bAstraZeneca, Process R+D, Silk Road, Macclesfield, Cheshire SK10 2NG, UK
cGlaxoSmithKline Chemical Development, Old Powder Mills, Tonbridge, Kent TN11 9AN, UK
Received 15 October 2003; revised 12 November 2003; accepted 28 November 2003
Abstract—Alkylation of 2-oxygenated diphenylmethane derivatives using sec-butyllithium and ())-sparteine gave enantiomeric
excesses of up to 60% with allyl bromide but alkylations with methyl electrophiles were poorly selective. When compounds with a
free hydroxy in the 2-position were alkylated the selectivity was reversed.
Ó 2003 Elsevier Ltd. All rights reserved.
Our group has an interest in agents, which bind to
tubulin at the colchicine binding site.1 Generally these
have as common structural features two aromatic rings,
often oxygenated, with some linking group. The exam-
ples shown are all powerful anticancer agents, which
exert their activity as a result of binding to the colchicine
site on tubulin (Fig. 1).
OR
CH3
O
O
1 R=Me, Et
OMe
gained if enantioenriched compounds could be syn-
thesised and tested and this has led us to investigate new
methodologies for the preparation of enantioenriched
compounds of this form.
Substituted diphenylmethanes structurally related to
podophyllotoxin are of considerable interest as tubulin
binding ligands. Compounds 1 have been synthesised as
racemates and been found to have excellent tubulin
binding properties.2 Single enantiomers have not been
made and groups other than methyl at the benzylic
position were not investigated. Valuable information
about the nature of the colchicine binding site could be
Existing syntheses of simple substituted diphenyl-
methanes are generally based on very classical chemis-
try,3 but examples exist where lateral lithiation with an
alkyllithium has been followed by alkylation.4 Our
O
O
MeO
MeO
MeO
O
OMe
MeO
OH
OMe
NH
OH
MeO
MeO
MeO
MeO
OPO(ONa)2
O
O
combretastatin A-4
Figure 1. Tubulin binding ligands.
ZD6126
podophyllotoxin
Keywords: Asymmetric; Alkylation; Deprotonation; Diphenylmethane; Sparteine.
* Corresponding author. Tel.: +44-161-295-4046; fax: +44-161-295-5111; e-mail: j.a.wilkinson@salford.ac.uk
0040-4039/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2003.11.138

1192
J. A. Wilkinson et al. / Tetrahedron Letters 45 (2004) 1191–1193
intention was to effect lateral lithiation/alkylation to give
an enantioenriched product from a prochiral diphenyl-
methane derivative (Scheme 1).
testing ground. Attempts to deprotonate 2 with lithium
diisopropylamide were unsuccessful (4 equiv LDA in
THF at room temperature for 48 h gave ca. 2% incor-
poration of deuterium). This effectively ruled out any
work with chiral lithium amide bases.8 However,
deprotonation with sec-butyllithium followed by
quenching with excess methyl iodide gave a near quan-
titative yield of racemic 3.9
OR
OR
R'
Ph
i) alkyl Li, ligand*
ii) R'-X
enantioenriched
Ph
Scheme 1.
The results of methylations of 2 in the presence of ())-
sparteine are summarised in Table 1 (entries 1–5).
Methyl iodide in ether was the only system, which gave
an appreciable ee and this was poor. The major enan-
tiomer was tentatively assigned as shown by assuming
that the sense of methylation would be identical to that
of allylation (vide infra). Other methylating reagents
failed to react at all. Attempts to vary the reaction
conditions with regard to temperature or the addition
sequence failed to give improved enantioselectivity as
did the use of a Ôwarm–cool cycleÕ according to the
method of Beak.10
Excellent results have been obtained in asymmetric
alkylation of benzylic CH2 groups by several researchers
using alkyllithiums in conjunction with ())-sparteine.5
We report here the initial results of an investigation into
lateral lithiation/alkylation of oxygenated diphenyl-
methanes using sparteine as a chiral control element.
OMe
Ph
OMe
CH3
i) sec-BuLi, Et2O, -78 °C
98% yield
ii) MeI
Allylations of 2 were also carried out (Table 1, entries
6–9). The reaction was most successful when the mix-
ture of sparteine/s-BuLi and 2 was stirred at )78 °C for
6 h before addition of the electrophile. A warm–cool
cycle followed by addition of allyl bromide gave a
lower ee (49%) in favour of the R-enantiomer. In this
case the products were converted to a compound of
known optical rotation and absolute stereochemistry
5,11 by ozonolysis and oxidation to the carboxylic acid
(Scheme 3). As can be seen the results when using allyl
bromide in ether were much more encouraging with
high yield and an ee of 60% in favour of the R-enan-
tiomer.
Ph
(+/—)-3
2
Scheme 2.
The initial starting material investigated was compound
2 (Scheme 2), which is readily available from 2-benzyl-
phenol.6 The methylated derivative 3 has a known spe-
cific rotation, thought to be for enantiomerically pure
material, ()49.2) although the absolute stereochemistry
of the two enantiomers has not been determined.7 For
this reason, methylation of 2 was chosen as the first
Table 1.
OMe
OMe
i) sec-BuLi, (-)-sparteine, -78 ºC
R
ii) R-X,-78 ºC to rt
Ph
Ph
3 R = Me
2
4 R = CH2CHCH2
Entry
Product
R–X
Solvent
Yield
Ee
Config.
(+)
1
2
3
4
5
6
7
8
9
3
3
3
MeI
MeI
Diethyl ether
THF
Toluene
83
95
60
0
7
0
0
MeI
MeOTs
MeOTf
AllylBr
AllylBr
AllylOTs
AllylOTf
Diethyl ether
Diethyl ether
Diethyl ether
THF
0
4
4
84
62
0
60
31
(R)
(R)
Diethyl ether
Diethyl ether
0
OMe
OMe
i) O3, CH2Cl2, -78°C
ii) Ag2O, H2O
O
46% yield
Ph
5
OH
Ph
4
Scheme 3.

J. A. Wilkinson et al. / Tetrahedron Letters 45 (2004) 1191–1193
1193
OH
OH
°
i) 2.2 eq. sec-BuLi, (-)-sparteine, -78 C
74% yield
46% ee
°
ii) warm to 0 C for 5 h
Ph
Ph
°
iii) cool to -78 C, add allyl bromide
7
6
Scheme 4.
The hydroxy compound, 2-benzylphenol 6 was also
investigated as a substrate for alkylation using sec-butyl-
lithium/sparteine (Scheme 4). Two equivalents of base
were utilised in order to produce a dianionic species but
only 1.1 equiv of sparteine. The reaction with allyl bro-
mide was attempted and proved fairly successful giving
the alkylated compound 7 in 74% isolated yield and an
ee of 46%. The warm–cool procedure (sec-butyllithium
was added at )78 °C and the solution warmed to 0 °C
and stirred at this temperature for 5 h before cooling to
)78 °C and adding the electrophile) was essential to the
success of the reaction. The ee was obtained by
O-methylation of the product (NaH, THF, MeI)
and submitting it to the same procedure as outlined
above to give the known 5. In this case however the
opposite S-enantiomer had been obtained. Reaction of
the dianion derived from 6 with iodomethane gave a
75% yield of racemic product.12
Work
is
currently
being
carried
out
to
improve these results and probe the origin of the stere-
oselectivity.
Acknowledgements
We thank the EPSRC and GlaxoSmithKline for a
CASE award (S.B.R.).
References and notes
1. Jordan, A.; Hadfield, J. A.; Lawrence, N. J.; McGown, A.
T. Med. Res. Rev. 1998, 18, 259.
2. Getahun, Z.; Jurd, L.; Chu, P. S.; Lin, C. M.; Hamel, E.
J. Med. Chem. 1992, 35, 1058.
This reversal of selectivity opens up the possibility of
obtaining either enantiomer of a given derivative, as
desired, by using a protected or unprotected starting
material. This greatly increases the possibilities for this
chemistry since only one enantiomer of sparteine is
readily available.13
3. Olah, G. A.; Krishnamurti, R.; Suryaprakash, G. K. In
Comprehensive Organic Synthesis; Trost, B. M., Fleming,
I., Eds.; Pergamon: Oxford, 1991; Vol. 3, pp 298–327.
4. Paquette, L. A.; Henzel, R. P.; Wilson, S. E. J. Am. Chem.
Soc. 1972, 94, 7780; Itter, F.-A.; Voegtle, F. Chem. Ber.
1985, 118, 2300.
5. (a) Beak, P.; Basu, A.; Gallagher, D. J.; Park, Y. S.;
Thayumanavan, S. Acc. Chem. Res. 1996, 29, 552; (b)
Thayumanavan, S.; Basu, A.; Beak, P. J. Am. Chem. Soc.
1997, 119, 8209; (c) Nakamura, S.; Nakawaga, R.;
Watanabe, Y.; Toru, T. J. Am. Chem. Soc. 2000, 122,
11340.
6. O-Methylations were achieved using sodium hydride and
iodomethane (3 equiv) in THF at room temperature.
7. Hesse, G.; Hagel, R. Liebigs Ann. 1976, 996.
8. Although they may have potential for selective deproto-
nation of chromium carbonyl complexes derived from 2.
See: Gibson, S. E.; OÕBrien, P.; Rahimian, E.; Smith,
M. H. J. Chem. Soc., Perkin Trans. 1 1999, 909.
9. All new compounds gave satisfactory spectral data
including HRMS. Enantiomeric excesses were determined
by comparison of optical rotations with known com-
pounds.
The reasons for our results are not clear. We assume
these reactions involve a thermodynamic/kinetic reso-
lution of an equilibrating pair of complexes. The closest
literature precedent available is BeakÕs work on ethyl-
aniline derivatives in which the sparteine/lithiated sub-
strate complexes were configurationally stable at )78 °C
but formed in approximately equal amounts (complexes
derived from ethyl benzamide were labile at )78 °C,
however). The best results were obtained when the
mixture was warmed to )25 °C to equilibrate the two
complexes.14 This is not consistent with our results for
allylations, which require either a fairly high de in the
deprotonation step or a dynamic equilibrium between
the two diastereomeric complexes at )78 °C. We suspect
that, as in BeakÕs work, the deprotonation is not espe-
cially selective but that, unlike the ethylaniline work, the
two complexes equilibrate at )78 °C. The large variation
of ee with the electrophile suggests that the two equili-
brating complexes react at similar rates with iodome-
thane but at significantly different rates with allyl
bromide. The change in the sense of stereoselection from
reactions of 2 to those of 6 is difficult to explain. It could
be that a different pathway is followed in reactions of 2
to that followed in 6. Alternatively, a different level of
aggregation or dimerisation may be found in the dilithio
species derived from 6.
10. Basu, A.; Beak, P. J. Am. Chem. Soc. 1996, 118, 1575.
11. ½aꢀ of optically pure S-isomer is )24.8 (c 10.5, CHCl3)
D
Meyers, A. I.; Whitten, C. E. Tetrahedron Lett. 1976,
1947.
12. ½aꢀ of optically pure R-isomer is +27.8 (c 5.77, benzene)
D
Kinoshita, T.; Ueno, T.; Ikai, K.; Fujiwara, M.; Okamoto,
K. Bull. Chem. Soc. Jpn. 1988, 61, 3273.
13. An excellent surrogate for (+)-sparteine is available:
Dearden, M. J.; Firkin, C. R.; Hermet, J.-P. R.; OÕBrien,
P. J. J. Am. Chem. Soc. 2002, 124, 11870.
14. Basu, A.; Gallagher, D. J.; Beak, P. J. Org. Chem. 1996,
61, 5718. See also Refs. 5a,b.