Addition of aryllithiums to an 11-oxo-steroid

Article abstract of DOI:10.1016/S0040-4039(01)01033-4

The addition of aryllithiums to an 11-oxo-steroid is possible in non-polar medium and at room temperature. A series of protected 11α-aryl-11β-hydroxy-androsten-diones have been prepared.

Full text of DOI:10.1016/S0040-4039(01)01033-4

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 5409–5411
Addition of aryllithiums to an 11-oxo-steroid
Vincent Lecomte, Nicolas Foy, Franck Le Bideau, Elie Ste´phan* and Ge´rard Jaouen
Laboratoire de chimie organome´tallique, Ecole Nationale Supe´rieure de Chimie et CNRS, 11 rue Pierre et Marie Curie,
75005 Paris, France
Received 2 May 2001; accepted 12 June 2001
Abstract—The addition of aryllithiums to an 11-oxo-steroid is possible in non-polar medium and at room temperature. A series
of protected 11a-aryl-11b-hydroxy-androsten-diones have been prepared. © 2001 Elsevier Science Ltd. All rights reserved.
Steroids substituted at position 11 by aryl groups are
very much in demand for their biological properties,
and their synthesis continues to attract considerable
interest1–5 particularly in an industrial context.6 Access
to derivatives of this type via organometallic additions
onto 11-oxo-steroids has always remained problematic.
In fact, 11-oxo-steroids possessing two angular methyl
groups (pregnane or pregnene series, androstane or
androstene, cortisone etc.) are fairly unreactive towards
organometallic derivatives. The literature reports little
more than the addition of methyllithium or methylmag-
nesium halides to steroids of this type,7 as well as, more
marginally, ethynylation8 or allylation9 reactions.
Fonken10 on the other hand has noted the lack of
reactivity of phenyllithium and 4-methoxyphenyllithium
towards a protected pregnan-11-one. Arylations in
position 11 have therefore been performed via other
routes. For example, Cleve5 recently described the syn-
thesis of 11-arylated compounds in the androstene
series via a Suzuki coupling.
(2a–d) to an 11-oxo-steroid in the androstene series.
Adducts 3 can be obtained in yields between 30 and
60% in non-polar media and at room temperature. The
aryllithium addition reaction was performed on 1, the
diketal of adrenosterone11 (Scheme 1).
The reaction was initially studied for the addition of
phenyllithium 2a onto the keto-steroid 1 in order to
establish the most favorable parameters for this addi-
tion. The addition reaction may of course occur in
competition with the classical enolization reaction,
which after hydrolysis can lead to an epimerization of
the steroid 1 in position 9 giving the product 4.
Table 1 shows the results obtained by the reaction of
phenyllithium with the steroid 1 under varying condi-
tions of solvent, temperature and reaction time. The
influence of lithium salt, present in commercial phenyl-
lithium at a level of around 3%, was also tested by
starting from a phenyllithium prepared by halogen/
metal exchange,12 and then used as such (entry 7) or
with the addition of LiBr (entry 8).
Here we describe the addition of a series of aryllithiums
Scheme 1.
Keywords: arylation; lithium compounds; 11-oxo-steroids.
* Corresponding author. E-mail: stephan@ext.jussieu.fr
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)01033-4

5410
V. Lecomte et al. / Tetrahedron Letters 42 (2001) 5409–5411
Table 1. Reaction of steroid 1 with phenyllithiuma
Entry
Solventb
Temperature (°C)
Time (h)
% 1
% 3a
% 4
1
2
3
4
THF
Toluene
Toluene
Toluene
20
20
20
60
20
20
20
20
2
2
24
24
2
2
2
2
100
50
46
30
34
100
47
23
0
50
54
30
42
0
0
0
0
40
24
0
10
31
5
Et2O/toluene (60/40)
Et2O/toluene (60/40) + TMEDAc
Et2O/toluene (45/55)
Et2O/toluene (45/55)
6
7d
8d
43
46
a The reaction is conducted with 3 equivalents of phenyllithium/steroid 1. Commercial phenyllithium was used in experiments 1–6. In all cases,
phenyllithium was added to a steroid solution.
b The reaction medium also contained approximately 10% saturated hydrocarbons. These are either cyclohexane when commercial phenyllithium
is used (entries 1–6), or hexanes when phenyllithium is prepared by exchange starting from commercial BuLi (entries 7 and 8).
c Experiment conducted in the presence of TMEDA in 1/1 PhLi/TMEDA.
d Experiments 7 and 8 were carried out using phenyllithium prepared by reaction of bromobenzene with butyllithium at room temperature in
ether. Experiment 7 was done without addition of LiBr, experiment 8 with 3 equivalents of LiBr/steroid 1.
The percentage of the addition product 3a is deter-
exploratory experiment with silylated enol ether, in
agreement with the lack of formation of 4. Prolonged
heating in this solvent (entry 4) leads to a lower rate of
addition and increased enolization. In this case the
presence of lithium salt in the medium affects the
addition rate only slightly, but does increase the
epimerization rate (especially in entry 8).
1
mined from the H NMR spectrum of the crude reac-
tion product recorded in CDCl3, taking into
consideration the integration of the angular methyl
signals. These signals are strongly deshielded for the
adduct (CH3-18: signal at 0.82 ppm for 1, 1.19 ppm for
3a; CH3-19: signal at 1.22 ppm for 1, 1.34 ppm for 3a)
and thus indicate an 11b configuration for the OH
group. Besides this well-known effect of deshielding of
angular methyls by an OH in 11b13 there is also a
contribution from the aryl group in 11a, which may be
estimated at +0.1 ppm for each of the two signals. The
The structure of phenyllithium in solution has been
studied by Jackman14 and Bauer.15 This lithium com-
pound exists in the form of a dimer in equilibrium with
its monomer in THF, and in the form of a tetramer in
toluene or hydrocarbon–ether. The mechanisms of
addition of lithium compounds onto ketones also vary
according to the medium.16 In ether media, the kinetics
causes dissociation of the dimer followed by addition of
the monomer onto the ketone, while in a hydrocarbon
medium a complex between the polymer and the ketone
appears to form first, and the addition product is
formed from this complex.
1
9b epimer, 4, was characterized by H and 13C NMR in
CDCl3. The formation of the epimer is signaled by the
1
appearance in H NMR of two further angular methyls
(0.82 and 1.29 ppm) and a doublet corresponding to
1
protons in C-7 or C-12 according to a H–13C shift-cor-
relation (3.2 ppm).
The addition reaction does not occur in THF (entry 1).
It does however take place in toluene or toluene–ether
medium at room temperature (entries 2, 3, 5, 7 and 8)
with the adduct 3 obtained in 42–54% yield. The pres-
ence of ether in the medium leads to increased competi-
tion from the enolization reaction at room temperature,
as can be seen from the percentages found for the
epimer 4 (entries 5, 7 and 8), without any increase in
the rate of addition. Enolization does not take place
after 2 h of reaction in toluene, according to an
The addition reaction in THF of a phenyllithium
monomer onto the hindered ketosteroid 1 would there-
fore not occur. However, it would seem that the forma-
tion of a tetramer–ketosteroid complex, followed by
conversion to an adduct 3 is possible. When the lithium
tetramer monomerizes on addition of a chelating ligand
(Table 1, entry 6) there is logically no reaction.
Table 2. 11a-Aryl-11b-hydroxy-steroids 3a–d
R
Yieldsa (%)
Mp (°C)
1H NMR: l (ppm)b
CH3-18
CH3-19
H aromatics
H (3a)
Me (3b)
43
40
20
25
182
171
153
–
1.19
1.19
1.18
1.22
1.34
1.34
1.33
1.48
7.18–7.47 (m)
7.08 (d); 7.31c
6.81 (d); 7.33c
7.58; 7.88
MeO (3c)
NMe2 (3d)d
a As an isolated product.
b In CDCl3 for 3a–c and in CD3OD for 3d.
c Broad signal.
d In this case the steroid deprotected at positions 3 and 17 is obtained as the chlorhydrate.

V. Lecomte et al. / Tetrahedron Letters 42 (2001) 5409–5411
5411
mercial adrenosterone to an analog of mifepristone20
(RU 486), the first known progesterone antagonist.
The highest reactivity of tetrameric lithiums had
already been observed in alkyllithium additions to ethy-
lene in pentane.17 The solvent effect observed here can
also be compared to the work of Canonne18 on addi-
tion of organomagnesium reagents to hindered ketones,
in which it was possible to increase the rate of addition
by using hydrocarbon diluents, to the detriment of the
secondary reduction reaction. To explain the observed
results, the author postulates a coordination effect of
the Grignard reagent on the alkoxide thus formed.
Acknowledgements
We wish to thank B. McGlinchey for her assistance in
translating the manuscript.
References
A series of steroids 3 was then prepared by the addition
of 3 equivalents of aryllithium to the steroid 1 under
optimal conditions as previously established. Tolyl-
lithium and 4-(N,N-dimethyl) phenyllithium were pre-
pared by reaction of the brominated derivatives with
lithium in ether.19 Anisyllithium was prepared by an
exchange reaction between 4-iodo-anisole and butyl-
lithium in ether. The extent of addition of the aryllith-
1. Geisler, J.; Cleve, A.; Harre, M. Tetrahedron 2000, 56,
6489–6492.
2. Rao, P. N.; Acosta, C. K.; Bahr, M. L.; Burdett, J. E.;
Cessac, J. W.; Morrison, P. A.; Kim, H. K. Steroids 2000,
65, 395–400.
3. Hazra, B. G.; Basu, S.; Pore, V. S.; Joshi, P. L.; Pal, D.;
Chakrabarti, P. Steroids 2000, 65, 157–162.
4. Rao, P. N.; Wang, Z.; Cessac, J. W.; Rosenberg, R. S.;
Jenkins, D. J. A.; Diamandis, E. P. Steroids 1998, 63,
523–530.
5. Cleve, A.; Fritzemeier, K. H.; Heinrich, N.; Klar, U.;
Mu¨ller-Fahrnow, A.; Neef, G.; Ottow, E.; Schwede, W.
Tetrahedron 1996, 52, 1529–1542.
1
iums (as established by H NMR) varies between 31
and 60%. Table 2 shows the yields obtained after
purification as well as some of the characteristics of the
adducts. Elemental analysis of the latter is in agreement
with the proposed structure.
Commercial phenyllithium and anisyllithium prepared
by exchange were added to a solution of the steroid 1 in
toluene, while the steroid in toluene solution was added
to the lithiums 2b and 2d. The medium was hydrolyzed
after 5–17 h at room temperature. After extraction by
dichloromethane and the usual work-up, the steroids
3a–c were separated from the starting material by chro-
matography on a silica column with a mixture of
AcOEt/petroleum ether (4/6) as eluent. This purifica-
tion method was not possible for the addition product
3d, which shows the same RF as the starting steroid.
The addition product must therefore be obtained in two
steps: crystallization of steroids 1 and 3d in cyclohex-
ane, followed by filtration (to eliminate N,N-dimethyl-
aniline and eventually 4-bromo-N,N-dimethylaniline,
which has not been transmetalated-formed during
hydrolysis) then treatment of a solution of the steroid
mixture in THF by dilute HCl to obtain the chlorhy-
drate of 3d. The two ketone functions are deprotected
by this acid treatment.
6. As shown by the number of patents cited in Chemical
Abstracts.
7. Kirk, D. N.; Petrow, V. J. Chem. Soc. 1961, 2091–2098
and references cited therein.
8. Shoppee, C. W.; Lack, R. E. J. Chem. Soc. 1962, 3624–
3629.
9. Wieland, P.; Kalvoda, J. Tetrahedron Lett. 1983, 24,
5603–5606.
10. Fonken, G. S. J. Org. Chem. 1965, 30, 2095–2096.
11. Bernstein, S.; Littell, R.; Williams, J. H. J. Am. Chem.
Soc. 1953, 75, 1481–1482.
12. Gilman, H.; Langham, W.; Moore, F. W. J. Am. Chem.
Soc. 1940, 62, 2327–2335.
13. Bridgeman, J. E.; Cherry, P. C.; Clegg, A. S.; Evans, J.
M.; Jones, E. R. H.; Kasal, A.; Kumar, V.; Meakins, G.
D.; Morisawa, Y.; Richards, E. E.; Woodgate, P. D. J.
Chem. Soc. (C) 1970, 250–257.
14. Jackman, L. M.; Scarmoutzos, L. M. J. Am. Chem. Soc.
1984, 106, 4627–4629.
15. Bauer, W.; Winchester, W. R.; Schleyer, P. von R.
Organometallics 1987, 6, 2371–2379.
In conclusion, the addition of aryllithiums to a pro-
tected adrenosterone allowed synthesis of a series of
protected 11a-aryl-11b-hydroxy-androsten-diones in
overall acceptable yields, considering the low reactivity
of 11-oxo-steroids. These steroids should allow access
to 11-arylated testosterones in three steps: a deoxygena-
tion step at position 11, currently under study, a depro-
tection and a selective reduction in position 17. We are
of course hoping to find an 11b stereochemistry for aryl
groups, specifically to obtain a short route from com-
16. Al-Aseer, M. A.; Smith, S. G. J. Org. Chem. 1984, 49,
2608–2613.
17. Kammermeier, B. O. T.; Klumpp, G. W.; Kolthof, K.;
Vos, M. Tetrahedron Lett. 1991, 32, 3111–3114.
18. Canonne, P.; Foscolos, G.; Caron, H.; Lemay, G. Tetra-
hedron 1982, 38, 3563–3568.
19. Gilman, H.; Zoellner, E. A.; Selby, W. M. J. Am. Chem.
Soc. 1933, 55, 1252–1257.
20. Teutsch, G.; Philibert, D. Hum. Reprod. 1994, 9 (Suppl.
1), 12–31.
.