Synthesis of 11β-perfluorohexylestradiol

Article abstract of DOI:10.1021/jo051424k

We report the synthesis of 11β-perfluorohexylestradiol 1e using a perfluoroorganometallic reagent for the introduction of the fluorous part. This compound is useful for biological studies and for imaging the ERa estradiol receptor distribution in the whol

Full text of DOI:10.1021/jo051424k

Synthesis of 11â-Perfluorohexylestradiol
Vangelis Agouridas, Jean-Claude Blazejewski,* Emmanuel Magnier, and Matthew E. Popkin†
SIRCOB, UMR CNRS 8086, Universite´ de Versailles, 45 Avenue des Etats-Unis, 78035 Versailles, France
jcblaz@chimie.uvsq.fr
Received July 11, 2005
We report the synthesis of 11â-perfluorohexylestradiol 1e using a perfluoroorganometallic reagent
for the introduction of the fluorous part. This compound is useful for biological studies and for
imaging the ERR estradiol receptor distribution in the whole cell by secondary ion mass spectrometry
(SIMS). The key step of this synthesis involves the radical reduction of an 11â-oxalate derivative.
The stereochemical outcome of this reaction was studied for a range of C11 substituents, and we
attempted to rationalize the apparent abnormal behavior of the phenyl group.
CHART 1. Estrogens and Antiestrogens
Estrogens play a critical role in the growth, develop-
ment, and maintenance of a diverse range of tissues.
They exert their physiological effects via activation of the
estrogen receptor (ER), which functions as a ligand-
activated transcriptional regulator.1 The estrogen recep-
tor, which is known to exist as two distinct subtypes, ERR
and ERâ,2 has become an enormously important target
for chemotherapeutic drugs against certain reproductive
cancers.
A wide repertoire of structurally distinct compounds
bind to the ER with differing degrees of affinity and
potency. Estrogens, such as estradiol 1a (Chart 1), act
solely as receptor agonists, whereas antiestrogens, such
as fulvestrant 1b, function as pure antagonists. A third
category of compounds, termed selective ER modulators
(SERMs) or partial antiestrogens, have the ability to act
as agonists or antagonists, depending upon the exact
cellular context and the ER isoform targeted.3,4 The latter
category includes some compounds that are in clinical
† Present address : Glaxo Smith Kline plc, Old Powder Mills,
Tonbridge TN11 9AN, United Kingdom.
(1) (a) Evans, R. M. Science 1988, 240, 889-895. (b) Green, S.;
Chambon, P. Nuclear Hormone Receptors; Parker, M. G., Ed.; Academic
Press: London, 1991; Chapter 2, pp 15-38.
(2) (a) Kuiper, G. G. J. M.; Enmark, E.; Pelto-Huikko, M.; Nilsson,
S.; Gustafsson, J.-A° . Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 5925-
5930. (b) Ogawa, S.; Inoue, S.; Watanabe, T.; Hiroi, H.; Orimo, A.;
Hosoi, T.; Ouchi, Y.; Muramatsu, M. Biochem. Biophys. Res. Commun.
1998, 243, 122-126. (c) Kuiper, G. G. J. M.; Carlsson, B.; Grandien,
K.; Enmark, E.; Ha¨ggblad, J.; Nilsson, S.; Gustafsson, J.-A° . Endocri-
nology 1997, 138, 863-870.
use, such as tamoxifen 2 and raloxifene 3, which are
administered for the treatment of osteoporosis and
hormone-dependent breast cancer. The widespread use
(3) Wijayratne, A. L.; Nagel, S. C.; Paige, L. A.; Christensen, D. J.;
Norris, J. D.; Fowlkes, D. M.; McDonnell, D. P. Endocrinology 1999,
140, 5828-5840.
(4) (a) Jordan, V. C. J. Med. Chem. 2003, 46, 883-908. (b) Jordan,
V. C. J. Med. Chem. 2003, 46, 1081-1111.
10.1021/jo051424k CCC: $30.25 © 2005 American Chemical Society
Published on Web 09/30/2005
J. Org. Chem. 2005, 70, 8907-8912
8907

Agouridas et al.
SCHEME 1a
of these drugs for the treatment of ER-positive breast
cancer, both in the advanced disease and as an adjuvant
therapy, has demonstrated the utility of partial anties-
trogens. Unfortunately, following an initial period of
effective clinical application, most breast cancers will
begin to develop resistance to this therapy after a few
months.3 Consequently, the development of more effective
pharmaceutical agents for the treatment of estrogen-
receptor-positive breast cancer, and the elucidation of this
inactivation process, remains essential.
a (a) C6F13I, CuCl, ethanolamine, tert-butyl alcohol, 80 °C, 8 h.
receptor ERR-substrate complex in the cell by ionic
microscopy (SIMS).11 We have recently published our
work in this area concerning the preparation of 7R-
perfluorohexylestradiol 1d.12 On the basis of the new
results described above, we report here the synthesis of
11â-perfluorohexylestradiol 1e.
We initially chose to prepare the requisite compound
1e by perfluoroalkylation of a readily available 9(11)-
dehydroestradiol derivative 4.13 Not too unexpectedly for
a sterically crowded double bond, attempts to introduce
the perfluorohexyl moiety by a radical chain reaction with
perfluorohexyliodide failed under a variety of conditions
and initiators.14
The solution of the crystal structure of the ligand
binding domain of ER complexes with various estrogens
has helped illuminate the molecular basis for estrogenic
or antiestrogenic activity.5
At the same time, ligand binding domain complexes
with partial antiestrogens such as raloxifene 3 have also
been crystallized.5,6 The picture of antiestrogenic activity
that has emerged suggests the key is the displacement
of the 12 helix of the ligand binding domain by the “side
chain” of the antiestrogens, which occupy what would be
the 11â position in estradiol 1a.6
Recently, the emergence of new pure antiestrogens,
antagonists devoid of any estrogenic activity, has raised
the hope for pharmaceuticals of increased effectiveness.
Two such antiestrogens, RU 58668 1c (11â-substituted)7
and ICI 182,780 1b (a 7R-substituted steroid, fulvestrant
as generic name),8 neatly encapsulate the findings of ER
structure-activity relationships.9
Prior to the establishment of the exact mode of estra-
diol binding to the ER, a detailed pharmacophore had
already emerged from the wealth of structure-binding
affinity relationships that had been acumulated.9 It has
been repeatedly shown that the 11â and 7R positions of
the steroid may be substituted by large, nonpolar groups
without a loss of affinity.9 In fact, certain compounds,
such as RU 58668 1c and fulvestrant 1b cited above, can
show superior binding affinity. It is assumed, though
unproven, that these pure antiestrogens demonstrate the
ability to displace the same H12 helix as the partial
antiestrogens tamoxifen 2 and raloxifene 3.6 The equiva-
lency of positions 7R and 11â was further demonstrated
in the case of fulvestrant 1b; this derivative undergoes
a complete rotation of its steroid core so that its 7R long
chain can be accommodated inside the ligand binding
cavity of ERâ.10
Use of the conditions described by Burton15 (CuCl,
ethanolamine, tert-butyl alcohol) gave incomplete success.
Although a fluorinated product was isolated (albeit in a
modest 18% yield), its structure was determined to be
that of an 11R-perfluorohexyl-8(9)-dehydroestradiol de-
rivative 5 (Scheme 1), useless for our purpose.16
More gratifying results were obtained with FITS-6,
tridecafluorohexylphenyliodonium trifluoromethanesul-
fonate (PhI(C6F13)(CF3SO3)), as an electrophilic perfluo-
ralkylating agent.17 Using pyridine as a base,18 we
observed the formation of a mixture of the dibenzylated
11-perfluorohexyl-9(11)-dehydroestradiol 6 (46%)19 and
the 8(9)-dehydroestradiol derivative 5 (3.5%) already
described in the previous experiment (Scheme 2).
(11) (a) Chandra, S. Appl. Surf. Sci. 2003, 203-204, 679-683. (b)
Guerquin-Kern, J.-L.; Coppey, M.; Carrez, D.; Brunet, A. C.; Nguyen,
C. H.; Rivalle, C.; Slodzian, G.; Croisy, A. Microsc. Res. Tech. 1997,
36, 287-295.
(12) Blazejewski, J.-C.; Wilmshurst, M. P.; Popkin, M. E.; Waksel-
man, C.; Laurent, G.; Nonclercq, D.; Cleeren, A.; Ma, Y.; Seo, H.-S.;
Leclercq, G. Bioorg. Med. Chem. 2003, 11, 335-345.
(13) Labaree, D. C.; Zhang, J.-x.; Harris, H. A.; O’Connor, C.;
Reynolds, T. Y.; Hochberg, R. B. J. Med. Chem. 2003, 46, 1886-1904.
(14) (a) Dolbier, W. R., Jr. Chem. Rev. 1996, 96, 1557-1584. (b)
Dolbier, W. R., Jr. In Organofluorine Chemistry: Fluorinated Alkenes
and Reactive Intermediates; Chambers, R. D., Ed.; Topics in Current
Chemistry Vol. 192; Springer-Verlag: Berlin, 1997; pp 97-163. (c)
Brace, N. O. J. Fluorine Chem. 1999, 93, 1-25.
(15) (a) Burton, D. J.; Kehoe, L. J. J. Org. Chem. 1970, 35, 1339-
1342. (b) Burton, D. J.; Kehoe, L. J. J. Org. Chem. 1971, 36, 2596-
2599.
(16) The 11R structure of 5 was confirmed a posteriori when the
stereochemistries of compounds 1e and 12 were unambiguously
determined. In this work, 11R-substituted estradiol derivatives showed
consistently negative optical rotations, whereas those of 11â compounds
are positive. Moreover, the angular methyl group appears as a sharp
singlet in 1H and 13C NMR spectra of 11R derivatives, and as a broad
line or a triplet in the 11â-perfluoroalkylated spectra.
(17) Umemoto, T.; Kuriu, Y.; Shuyama, H.; Miyano, O.; Nakayama,
S.-i. J. Fluorine Chem. 1986, 31, 37-56.
(18) Umemoto, T.; Kuriu, Y.; Nakayama S.-i. Tetrahedron Lett. 1982,
23, 1169-1172.
The introduction of highly fluorinated chains into the
estradiol nucleus in the key 7R or 11â positions could give
compounds useful for mapping the distribution of the
(5) Brzozowski, A. M.; Pike, A. C. W.; Dauter, Z.; Hubbard, R. E.;
Bonn, T.; Engstro¨m, O.; O¨ hman, L.; Greene, G. L.; Gustafsson, J.-A° .;
Calquist, M. Nature 1997, 398, 753-758.
(6) Pike, A. C. W.; Brzozowski, A. M.; Hubbard, R. E.; Bonn, T.;
Thorsell, A.-G.; Engstro¨m, O.; Ljunggren, J.; Gustafsson, J.-A° .; Calquist,
M. EMBO J. 1999, 18, 4608-4618.
(7) Van de Velde, P.; Nique, F.; Bouchoux, F.; Bre´maud, J.; Hameau,
M.-C.; Lucas, D.; Moratille, C.; Viet, S.; Philibert, D.; Teutsch, G. J.
Steroid Biochem. Mol. Biol. 1994, 48, 187-196.
(8) (a) Wakeling, A. E.; Bowler, J. J. Steroid Biochem. Mol. Biol.
1992, 43, 173-177. (b) Howell, A.; Osborne, C. K.; Morris, C.; Wakeling,
A. E. Cancer 2000, 89, 817-825. (c) Robertson, J. F. R.; Come, S. E.;
Jones, S. E.; Beex, L.; Kaufmann, M.; Makris, A.; Nortier, J. W. R.;
Possinger, K.; Rutqvist, L.-E. Eur. J. Cancer 2005, 41, 346-356.
(9) Anstead, G. M.; Carlson, K. E.; Katzenellenbogen, J. A. Steroids
1997, 62, 268-303.
(19) The 9(11)-dehydro structure of 6 mainly relies on the presence
of an isolated AB system for protons H12R and H12â (δ 2.31 and 2.35,
respectively) identified by HCOR and DEPT (δ C12 ) 40.8 ppm), and
on the absence of coupling with other protons in a COSY experiment.
The signals corresponding to these protons, although partially ob-
scured, were more apparent in the debenzylated derivative 7 of 6 (δ
2.02 and 2.35 ppm, J ) 14.6 Hz) and showed no further coupling.
(10) (a) Pike, A. C. W.; Brzozowski, A. M.; Walton, J.; Hubbard, R.
E.; Thorsell, A.-G.; Li, Y.-L.; Gustafsson, J.-A° .; Carlquist, M. Structure
2001, 9, 145-153. (b) Tedesco, R.; Katzenellenbogen, J. A.; Napolitano,
E. Bioorg. Med. Chem. Lett. 1997, 7, 2919-2924.
8908 J. Org. Chem., Vol. 70, No. 22, 2005

Synthesis of 11â-Perfluorohexylestradiol
SCHEME 2a
SCHEME 3a
a (a) FITS-6, CH2Cl2, additive (see Table 1), rt, 2 h; (b) H2, 10%
Pd/C, MeOH, rt, 14 h.
TABLE 1. Perfluoroalkylation of
9(11)-Dehydroestradiol Dibenzyl Ether 4 with FITS-6
recovered
entry
additives (equiv) a
6 (%)b 5 (%)b
4 (%) b
1
2
3
4
5
pyridine (1)
46
3.5
4
4
11
3.5
3.5
18
80
nBu4NI (1.1)
H2O (1.1), NaHCO3 (1.1)
Et3SiH (5)
34
24
a (a) C6F13MgBr, Et2O, -60 °C, 2 h; (b) (i) nBuLi, THF/hexane,
-60 °C, 10 min, (ii) methyl chlorooxoacetate, -60 °C to room
temperature, 1 h; (c) Bu3SnH, AIBN, toluene, reflux, 14 h; (d) H2,
10% Pd/C, MeOH, rt, 14 h.
4
34
75c
27
Et3SiH (2.5), dBMPd (2.5)
a One equivalent of 4 and 1.1 equiv of FITS-6 were used in all
experiments, rt, 2 h. b Isolated yield. c 4 was reduced to estradiol
dibenzyl ether. d 2,6-Di-tert-butyl-4-methylpyridine.
to a mixture of the 9R-ketone 8 and its 9â epimer,25 by
elimination of the perfluoroalkyl group, at temperatures
above -20 °C.
Attempts to capture the putative carbocationic inter-
mediate with various nucleophiles were unsuccessful
(Table 1) but showed interesting variations in the dis-
tribution of the elimination products. In particular,
changing the base from pyridine (entry 1) to sodium
bicarbonate (entry 3) or a non-nucleophilic amine (entry
5) inverted the relative proportions of 6 and 5.
Unfortunately, the double bond in 6 proved to be
unreactive under a variety of catalytic as well as ionic
reduction conditions.
We next turned to a route devised by Napolitano et
al.20 involving the condensation of an organometallic
derivative with a 11-ketoestradiol followed by deoxygen-
ation of the resulting alcohol. In this way, reaction of
perfluorohexylmagnesium bromide21 with the 9R-ketone
813,20 readily afforded the perfluoroalkylated tertiary
alcohol 9 in 51% isolated yield (Scheme 3).
The next step, deoxygenation of alcohol 9, proved to
be troublesome. Not unexpectedly, in view of the desta-
bilizing influence of a perfluoroalkylated chain on an
adjacent incipient carbocation,22 this reaction failed under
the usual ionic protocol using boron trifluoride-diethyl
etherate and triethylsilane even under forcing condi-
tions.20 The formation of a conventional alcohol derivative
suitable for further reduction (xanthate23 or pentafluo-
rophenylthionocarbonate)24 was impeded by the thermal
instability of the intermediate alcoholate, and also by the
sluggish reactivity of this encumbered and quite acidic
alcohol at low temperature. We thus observed that the
lithium or potassium alcoholate derived from 9 reverts
Although we were able to prepare the corresponding
trifluoroacetate (BuLi, (CF3CO)2O, -78 °C, 30% yield),
no fluorinated product could be isolated after workup
(aqueous HF) of the reduction mixture (Ph2SiH2, (t-BuO)2,
120 °C, 20 h).26 Recently, Ste´phan et al. were faced with
a cognate problem in the androstene series, and elegantly
solved it by reduction of an oxalate derivative.27 However,
they observed that the reduction was accompanied by
respectable amounts of elimination and/or rearrangement
products.
In our case, the intermediate oxalate 10 was obtained
with a modest (41%) isolated yield, but we were pleased
to find that the ensuing reduction proceeded quite
cleanly, giving a 50:50 mixture of the epimeric 11-
perfluorohexylestradiols 11a and 11b in 75% isolated
yield, the recovery of the starting alcohol 9 accounting
for the remaining mass balance. No elimination product
5 or 6 could be detected in this reaction. The two isomers
11a and 11b could be easily separated by chromatogra-
phy. Further deprotection of the benzyl ether groups by
catalytic hydrogenation (10% Pd/C, H2) proceeded un-
eventfully to give 11â-perfluorohexylestradiol 1e (94%
yield) as well as its 11R isomer 12, thus achieving the
projected synthesis. Assignment of the stereochemistry
at center 11 by NMR with both isomers at hand was
relatively straightforward. In compound 12, the H9
proton could be located at δ 2.31, and appears as a triplet
with a coupling constant of 10.2 Hz, pointing to a trans
relationship with both H8 and H11. In compound 1e, the
H9 proton appears at δ 2.81 as a doublet of doublets (J
) 10.5 and 4.2 Hz), as expected for a trans relationship
with H8 and a cis relationship with H11. Moreover, in
this later compound, the angular methyl group (δ 0.78)
(20) Tedesco, R.; Fiaschi, R.; Napolitano, E. J. Org. Chem. 1995,
60, 5316-5318.
(21) Burton, D. J.; Yang, Z.-Y. Tetrahedron 1992, 48, 189-275.
(22) van Alem, K.; Belder, G.; Lodder, G.; Zuilhof, H. J. Org. Chem.
2005, 70, 179-190 and references therein.
(23) Barton, D. H. R.; Parekh, S. I.; Tse, C.-L. Tetrahedron Lett.
1993, 34, 2733-2736.
(25) Liang, C. D.; Baran, J. S.; Allinger, N. L.; Yu, J. H. Tetrahedron
1976, 32, 2067-2069.
(24) (a) Barton, D. H. R.; Blundell, P.; Dorchak, J.; Jang, D. O.;
Jaszberenyi, J. Cs. Tetrahedron 1991, 47, 8969-8984. (b) Barton, D.
H. R.; Dorchak, J.; Jaszberenyi, J. Cs. Tetrahedron 1992, 48, 7435-
7446.
(26) Kim, J.-G.; Cho, D. H.; Jang, D. O. Tetrahedron Lett. 2004, 45,
3031-3033.
(27) Lecomte, V.; Ste´phan, E.; Rager, M.-N.; Jaouen, G. J. Org.
Chem. 2004, 69, 3216-3219.
J. Org. Chem, Vol. 70, No. 22, 2005 8909

Agouridas et al.
CHART 2a
FIGURE 1. Partially fluorinated radicals.
claimed to be as large as an isopropyl group in 1,3-diaxial
interactions in cyclohexane as well as in other systems.32
The bulk of a longer fluorinated chain may be thought
to be slightly larger and thus closer to that of a phenyl
group in this context.33
As expected for a nonsterically demanding group, the
reduction of the allyl compound 13a was fully stereose-
lective. But at this stage, the only apparent factor that
could explain the diverging behavior between the per-
fluorohexyl group and the phenyl group appears to be
the difference in the rate of hydrogen abstraction from
the hydrogen donor Bu3SnH.
Thanks to their high electrophilic character, fully
fluorinated radicals are widely used for the easy intro-
duction of a perfluoralkyl chain onto aromatic or olefinic
substrates.14 For the same reason, these radicals are also
strong hydrogen atom abstractors.34 The behavior of
partially fluorinated radicals 16a-d (Figure 1) is less
well understood.
Thus it was shown that the radical 16a formed from
2,2,2-trifluoroethyliodide adds with difficulty to electron-
rich olefins.35 We have observed that tertiary trifluoro-
methylated radical 16b, derived from xanthates, reacted
sluggishly with allyltributyltin under atypical radical
reaction conditions.36 In a related example, it has been
demonstrated that the radical 16c is only 4.3 times more
reactive for addition to olefins and 5.8 times more
reactive for hydrogen abstraction than a conventional
primary alkyl radical.37 These observations point to a
reduced electrophilicity of the radical center in these
species compared to that of the perfluorinated series, and
consequently a lower hydrogen abstraction rate.38 How-
ever, these figures remain still higher than in the
nonfluorinated case. In particular, the absolute hydrogen
abstraction rate from tributyltin hydride by the benzyl
radical was measured to be 3.6 × 104 M-1 s-1 at 298 K.39
The value reported for 16c is 1.4 × 107 M-1 s-1 at the
same temperature,37 pointing to a ca. 400-fold increased
rate in favor of the partially fluorinated radical. For a
more fluorinated (and thus more polar) radical like 16d,
this effect may be expected to be greater.37
a Relative percentage of 11R and 11â isomers formed.
is broadened by space coupling with the fluorine atoms
of the 11â-perfluorohexyl side chain.
We were intrigued by the lack of stereoselectivity
observed during the reduction step of the fluorinated
alcohol 9. On the basis of ample evidence in the estradiol
series,20 attack of bulky reagents at the 11 position occurs
mainly by the less hindered R face of the steroid nucleus,
as exemplified here by the condensation of perfluoro-
hexylmagnesium bromide with ketone 8 leading exclu-
sively to the 11â-alcohol 9. Moreover, it has been shown
that, in the sugar series, the reduction of a related oxalate
occurs with full inversion of the fluorinated center;28 we
thus expected the same behavior in our particular case.
To gain some insight into the stereochemical outcome of
this reaction in our system, we also investigated the
reduction of oxalate derivatives of the 11R-allyl alcohol
13a,13 and of the 11R-phenyl alcohol 13b (Chart 2).20
The reduction of oxalate 14a was fully selective toward
the 11â-allylestradiol derivative 15a;13 no 11R epimer
could be detected within experimental error.29 A con-
trasting result was obtained with the phenyl oxalate
derivative 14b. In this case, we observed a 92:8 ratio in
favor of the R stereoisomer 15b (Chart 2). It appears that,
under the conditions used here, the behavior of a per-
fluorohexyl group is intermediate between that of an allyl
and a phenyl group.
Two factors must be accounted for in order to explain
the stereoselectivity observed during the reduction of
oxalates 10, 14a, and 14b: the steric bulk of the
substituent on one hand and the rate of hydrogen
abstraction by the intermediate radical on the other
hand.
With regard to the first point, perfluoroalkyl chains
may be considered bulky, rigid rods.31 Insofar as it may
be envisaged that the steric interactions involved here
are confined to the beginning of the chain, it is notewor-
thy that a trifluoromethyl group has repeatedly been
One may thus be tempted to attribute the observed
stereoselectivities to the reactivity-selectivity prin-
(32) Schlosser, M.; Michel, D. Tetrahedron 1996, 52, 99-108.
(33) Eliel, E. L.; Wilen, S. H.; Mander, L. N. Stereochemistry of
Organic Compounds; John Wiley & Sons: NewYork, 1994.
(34) Dolbier, W. R., Jr.; Rong, X. X. J. Fluorine Chem. 1995, 72,
235-240.
(35) (a) Cloux, R.; Kova´ts, E. sz. Synthesis 1992, 409-412. (b) Long,
Z.-Y.; Chen, Q.-Y. Tetrahedron Lett. 1998, 39, 8487-8490.
(36) Blazejewski, J.-C.; Diter, P.; Warchol, T.; Wakselman, C.
Tetrahedron Lett. 2001, 42, 859-861.
(37) Bartberger, M. D.; Dolbier, W. R., Jr.; Lusztyk, J.; Ingold, K.
U. Tetrahedron 1997, 53, 9857-9880.
(38) The reactivity of such a radical could, however, be enhanced
by the use of an R heteroatom substituent. See: (a) Gagosz, F.; Zard,
S. Z. Org. Lett. 2003, 5, 2655-2657. (b) Tournier, L.; Zard, S. Z.
Tetrahedron Lett. 2005, 46, 455-459.
(39) Franz, J. A.; Suleman, N. K.; Alnajjar, M. S. J. Org. Chem. 1986,
51, 19-25.
(28) Lavaire, S.; Plantier-Royon, R.; Portella, C. J. Carbohydr. Chem.
1996, 15, 361-370.
(29) The major products formed, reminiscent of those described in
the androstene case, resulted from an intramolecular cyclization of the
allyl moiety with the oxalyl residue.27,30
(30) Lecomte, V.; Ste´phan, E.; Jaouen, G. Tetrahedron Lett. 2005,
46, 1123-1126.
(31) (a) Kirsch, P. Modern Fluoroorganic Chemistry: Synthesis,
Reactivity, Applications; Wiley-VCH: Weinheim, Germany, 2004. (b)
Ellis, D. A.; Denkenberger, K. A.; Burrow, T. E.; Mabury, S. A. J. Phys.
Chem. A 2004, 108, 10099-10106. (c) Liantonio, R.; Metrangolo, P.;
Pilati, T.; Resnati, G. Cryst. Growth Des. 2003, 3, 355-361.
8910 J. Org. Chem., Vol. 70, No. 22, 2005

Synthesis of 11â-Perfluorohexylestradiol
NaHCO3 (5 mL), extraction with dichloromethane, and drying
(MgSO4), PTLC on silica gel (CH2Cl2:pentane 2:3) afforded pure
5 (62 mg, 18%).
SCHEME 4
1H NMR (300 MHz, CDCl3): δ 0.62 (s, 3H), 1.43-1.77 (m,
4H), 2.04-2.23 (m, 4H), 2.42-2.72 (m, 3H), 3.58 (t, J ) 7.8
Hz, 1H), 3.96 (dt, J ) 27.0, 7.2 Hz, 1H), 4.42-4.51 (AB system,
JAB ) 12.1 Hz, 2H), 4.95 (s, 2H), 6.69 (m, 2H), 6.99 (m, 1H),
7.17-7.36 (m, 10H). 13C NMR (75 MHz, CDCl3): δ 12.4, 21.4,
26.7, 28.4, 28.6, 35.3, 36.3 (dd, JCF ) 21, 17 Hz), 44.5, 46.3,
70.0, 71.8, 87.7, 111.5, 114.4, 121.7, 122.7, 127.4, 127.5, 127.6,
128.0, 128.4, 128.6, 129.2, 137.0, 137.2, 138.9, 141.8, 157.0.
19F NMR (188 MHz, CDCl3): δ -81.3 (t, J ) 9 Hz, 3F), -107.0
and -116.7 (AB system, J ) 276 Hz, 2F), -120.2 (m, 2F),
-122.2 to -123.4 (m, 4F), -126.7 (m, 2F). MS pos. ESI (m/z):
ciple: the most reactive fluorinated radical is also the
least selective.
791 [M + Na+], 722 [M + Na+ - CF3]. [R]25 -86 (c 0.5,
To check this hypothesis, we also investigated the
reduction of both compounds with a large excess of
tributyltin hydride (80 equiv).40 Under these conditions,
we observed an increase of the selectivity in the reduction
of the fluorinated oxalate 10 (11a/11b ) 20/80 vs 50/50)
and practically no change with the phenyl derivative 14b
(15c/15b ) 10/90 vs 8/92). Thus, in the case of the
fluorinated radical, hydrogen abstraction competes ef-
fectively with radical inversion.
However, this hardly explains the very strong prefer-
ence of the phenyl group for the R position upon radical
reduction. It should be recalled that, by contrast, upon
ionic reduction (BF3, Et2O/Et3SiH) of the benzyl alcohol
13b, the 11R-phenyl group (like other common substit-
uents) goes exclusively to the â position to give 15c (Chart
2).20 Although ionic reduction may not involve a true free
carbocationic species, this shows that the phenyl group
could be well-accommodated in the 11â position. On the
other hand, Bu3SnH may be considered bulkier than Et3-
SiH, and thus should show a stronger tendency to deliver
its hydrogen atom from the less hindered R face of the
steroid nucleus. One possible clue to these observations
could be that the delivery of the hydrogen atom from Bu3-
SnH occurs directly to the nucleus of the phenyl group
rather than to the crowded 11-benzylic center.39 This
hypothesis is consistent with the insensitivity of the
selectivity to hydride concentration. The resulting semi-
benzene derivative thus obtained is then able to evolve
to the more stable 11R-phenylestradiol by a hydrogen
shift (Scheme 4).41
D
dichloromethane). HRMS: obsd 768.2280, calcd 768.2273
(C38H33O2F13).
11-Tridecafluorohexyl-3,17â-dibenzyloxyestra-
1,3,5(10),9(11)-tetraene (6). FITS-6 (327 mg, 0.49 mmol, 1.1
equiv) was added in one portion to a solution of 9(11)-
dehydroestradiol 4 (200 mg, 0.44 mmol) and pyridine (25 µL,
0.44 mmol, 1 equiv) in dichloromethane (12 mL). After the
mixture had been stirred for 2 h at room temperature, the
reaction was quenched by the addition of water (10 mL). The
organic layer was washed with water (2 × 5 mL), and dried
over MgSO4. After removal of the solvent, the residue was
purified by PTLC on silica gel (CH2Cl2:pentane 2:3) to afford
5 (12 mg, 3.5%) (vide supra) and 6 (155 mg, 46%) as glasses.
1H NMR (300 MHz, CDCl3): δ 0.94 (s, 3H), 1.30-1.89 (m,
6H), 2.02-2.19 (m, 3H), 2.55-2.70 (m, 3H), 3.64 (t, J ) 8.3
Hz, 1H), 4.61 (s, 2H), 5.08 (s, 2H), 6.80 (m, 2H), 7.20-7.45
(m, 11H). 13C NMR (75 MHz, CDCl3): δ 10.8, 24.3, 26.8, 28.1,
28.2, 40.8, 41.6, 42.2, 47.7, 69.9, 71.7, 87.2, 111.2, 112.9, 118.8,
127.4, 127.47, 127.52, 127.9, 128.3, 128.5, 129.2, 129.7, 136.5,
138.5, 140.0, 146.2, 159.0. 19F NMR (188 MHz, CDCl3): δ -81.3
(t, J ) 10 Hz, 3F), -100.5 and -103.2 (AB system, J ) 269
Hz, 2F), -118.1 (m, 2F), -123.5 (m, 4F), -126.6 (m, 2F). MS
pos. ESI (m/z): 791 [M + Na+], 722 [M + Na+ - CF3]. HRMS:
obsd 768.2264, calcd 768.2273 (C38H33O2F13).
11r-Tridecafluorohexyl-3,17â-dibenzyloxyestra-
1,3,5(10)-trien-11â-ol (9). Freshly distilled perfluorohexylio-
dide (3.24 mL, 15 mmol) was slowly added, at -60 °C under
an argon atmosphere, to a 3 M solution of methylmagnesium
bromide in diethyl ether (5 mL, 15 mmol). After being stirred
for 30 min at -60 °C, followed by careful addition of a solution
of ketone 8 (1.12 g, 2.40 mmol) in THF (5 mL), the mixture
was stirred again for 2 h at -60 °C and allowed to reach room
temperature. After addition of a saturated solution of am-
monium chloride (10 mL), we filtered the mixture on a pad of
Celite, and washed it with CH2Cl2 (20 mL). The organic phase
was separated, and the aqueous layer was extracted twice with
CH2Cl2 (2 × 20 mL). The combined organic extracts were
washed with saturated NaHCO3 solution (2 × 10 mL) and
water (2 × 10 mL), dried over MgSO4, and concentrated under
reduced pressure. The residue was purified by chromatography
on a silica gel column (eluant: CH2Cl2) to afford 962 mg of 9
as a white solid (51% yield). Mp: 126.2-126.4 °C. 1H NMR
(300 MHz, CDCl3): δ 0.99 (s, 3H), 1.13-1.37 (m, 3H), 1.43-
1.78 (m, 7H), 1.94-2.06 (m, 1H), 2.13-2.25 (m, 2H), 2.55-
2.58 (m, 3H), 3.40 (t, J ) 7.8 Hz, 1H), 4.41-4.51 (AB system,
J ) 12 Hz, 2H), 4.96 (s, 2H), 6.69 - 6.73 (m, 2H), 7.17-7.37
(m, 10H), 7.83 (t, J ) 7.5 Hz, 1H). 13C NMR (75 MHz, CDCl3):
δ 13.5, 23.9, 26.0, 27.4, 28.2, 35.7, 42.0, 44.9, 46.1, 50.3, 69.9,
71.7, 81.2 (t, JCF ) 21 Hz), 88.5, 110.9, 113.8, 127.4, 127.5,
127.6, 127.9, 128.4, 128.6, 129.1, 129.2, 131.3, 137.4, 138.9,
141.5, 156.8. 19F NMR (188 MHz, CDCl3): δ -81.4 (t, J ) 10
Hz, 3F), -111.3 and -115.7 (AB system, J ) 286 Hz, 2F),
-118.2 (m, 2F), -122.1 to -123.3 (m, 4F), -126.8 (m, 2F). IR
(cm-1, KBr): 3615, 3026, 2939, 2862, 1613, 1577, 1450, 1234.
MS pos. ESI (m/z): 825 [M + K+], 809 [M + Na+], 787 [M +
Contrary to the case of the phenyl oxalate, the stereo-
chemical outcome of the radical reduction of the perfluo-
roalkylated oxalate 10 appears to be governed by a
balance between kinetic and steric factors. With the
proper choice of reaction conditions, we were able to form
substantial amounts of the â isomer during the reduction,
necessary for achieving the synthesis of the requisite 11â-
perfluorohexylestradiol 1e. Biological testing of this
substance is currently underway.
Experimental Section
11r-Tridecafluorohexyl-3,17â-dibenzyloxyestra-
1,3,5(10),8(9)-tetraene (5). A mixture of ethanolamine (0.11
mL, 1.82 mmol, 4.1 equiv), 9(11)-dehydroestradiol 4 (200 mg,
0.44 mmol), tBuOH (0.54 mL), CuCl (15 mg, 0.15 mmol, 0.35
equiv), and perfluorohexyliodide (490 µL, 2.20 mmol, 5 equiv)
was stirred at 80 °C for 8 h. After dilution with saturated
(40) We greatly acknowledge one of the reviewers for this very
interesting suggestion.
(41) Maeda, H.; Huang, Y.; Hino, N.; Yamauchi, Y.; Ohmori, H.
Chem. Commun. 2000, 2307-2308.
H+]. [R]25 -15 (c 0.44, dichloromethane). Anal. Calcd for
D
C38H37F13O3: C, 58.02; H, 4.48. Found: C, 57.92; H, 4.19.
J. Org. Chem, Vol. 70, No. 22, 2005 8911

Agouridas et al.
General Procedure for the Preparation of Oxalates.
A solution of n-butyllithium (0.6 mL, 1.2 M in hexane, 1.2
equiv) was added to a solution of the steroidal alcohol (0.59
mmol) in THF (2 mL), and cooled to -60 °C under an argon
atmosphere. After stirring the solution for 10 min at -60 °C,
we added methyl chlorooxoacetate (144 mg, 2 equiv). The
mixture was allowed to reach room temperature, and was then
diluted with CH2Cl2 (5 mL) and water (2 mL). The organic
phase was separated, washed with water and brine, dried over
MgSO4, and concentrated under reduced pressure. Flash
chromatography of the residue (eluant: CH2Cl2) afforded the
oxalate derivative.
Hz, 1H), 2.57 (m, 1H), 2.71-2.83 (m, 2H), 3.30 (t, JHF ) 19.8
Hz, 1H), 3.40 (t, J ) 6.9 Hz, 1H), 4.42-4.53 (AB system, J )
12.3 Hz, 2H), 4.94 (s, 2H), 6.59-6.70 (m, 2H), 6.99 (d, J ) 8.4
Hz, 1H), 7.18-7.37 (m, 10H). 13C NMR (75 MHz, CDCl3): δ
13.5, 23.0, 27.3, 28.2, 30.6, 35.7, 38.1 (t, J ) 19 Hz), 38.4, 42.6,
48.5, 53.0, 69.8, 71.6, 89.5, 112.3, 114.4, 127.4, 127.5, 127.6,
127.9, 128.3, 128.5, 137.2, 138.6, 139.0, 156.4. 19F NMR (188
MHz, CDCl3): δ -81.7 (t, J ) 9 Hz, 3F), -102.1 (m, 2F), -119.1
(m, 2F), -120.6 to -122.7 (m, 4F), -125.9 (m, 2F). MS pos.
ESI (m/z): 819 [M + K]+, 793 [M + Na]+. [R]25 82.6 (c 0.38,
D
dichloromethane). Anal. Calcd for C38H37F13O2: C, 59.22; H,
4.58. Found: C, 59.33; H, 4.53.
Oxalic Acid 11r-Tridecafluorohexyl-3,17â-dibenzyloxy-
estra-1,3,5(10)-triene-11â-yl Ester Methyl Ester (10). Yield
of 41%. 1H NMR (300 MHz, CDCl3): δ 0.97 (s, 3H), 1.29-1.79
(m, 10H), 1.93-2.15 (m, 2H), 2.57-2.73 (m, 3H), 3.54 (t, J )
7.2 Hz, 1H), 3.74 (dd, J ) 15.7, 2.2 Hz, 1H), 3.93 (t, 3H), 4.58
(s, 2H), 5.07 (s, 2H), 6.81-6.83 (m, 2H), 7.30-7.48 (m, 10H),
7.68 (dd, J ) 9.6, 3.4 Hz, 1H). 13C NMR (75 MHz, CDCl3): δ
12.3, 23.9, 25.5, 27.5, 28.2, 34.7, 40.0, 42.4, 46.6, 50.5, 53.6,
69.8, 71.8, 88.5, 90.6, 111.2, 113.4, 127.4, 127.5, 127.6, 127.9,
128.3, 128.5, 129.5, 137.2, 138.7, 141.5, 155.4, 156.9, 157.8.
19F NMR (188 MHz, CDCl3): δ -81.3 (t, J ) 10 Hz, 3F), -104.2
and -106.2 (AB system, J ) 284 Hz, 2F), -117.2 to -128.7
(m, 8F).
General Procedure for the Radical Reduction of
Oxalates. To a solution of the oxalate (0.15 mmol) in dry,
degassed toluene (2 mL) were added Bu3SnH (120 µL, 0.45
mmol, 3 equiv) and AIBN (5 mg, 0.03 mmol, 0.2 equiv). The
mixture was refluxed overnight. The solution was cooled and
partitioned between CH2Cl2 (5 mL) and water (5 mL). The
aqueous layer was extracted twice with CH2Cl2. The combined
organic extracts were washed with water (2 × 10 mL), dried
over MgSO4, and concentrated under reduced pressure. The
residue was purified by PTLC on silica gel (eluant: CH2Cl2/
pentane 3/2).
General Procedure for Catalytic Hydrogenations. To
a solution of the steroid (0.045 mmol) in MeOH (2 mL) was
added 10% Pd/C (6 mg). The substrate was hydrogenated
overnight under 1 atm. The mixture was then filtered on
Celite, and the filtrate was evaporated to afford the product.
11â-Tridecafluorohexyl-3,17â-dihydroxyestra-1,3,5(10)-
triene (1e). Yield of 94%. Mp: 196.3-196.5 °C. 1H NMR (300
MHz, methanol-d4): δ 0.78 (broad s, 3H), 1.00-2.05 (m, 10H),
2.30 (d, J ) 14.6 Hz, 1H), 2.45-2.69 (m, 2H), 2.81 (dd, J )
10.5, 4.5 Hz, 1H), 3.24 (s, 1H), 3.42 (broad t, JHF ) 21 Hz,
1H), 3.58 (t, J ) 7.3 Hz, 1H), 6.36 (s, 1H), 6.43 (dd, J ) 8.7,
2.7 Hz, 1H), 6.95 (d, J ) 8.7 Hz, 1H). 13C NMR (75 MHz,
methanol-d4): δ 13.5, 24.0, 28.5, 31.0, 31.7, 37.6, 38.4, 39.6 (t,
JCF ) 19 Hz), 43.7, 49.7, 53.8, 83.8, 113.7, 115.9, 128.3, 130.1,
139.7, 155.7. 19F NMR (188 MHz, methanol-d4): δ -80.0 (t, J
) 9.1 Hz, 3F), -100.1 (m, 2F), -117.1 (m, 2F), -119.2 (m, 2F),
-120.4 (m, 2F), -123.9 (m, 2F). IR (cm-1, KBr): 3328, 2954,
2910, 1234. MS neg. ESI (m/z): 635 [M + formate], 625.2 [M
+ Cl-], 589 [M - 1]. [R]25D 92.8 (c 0.54, methanol-d4). HRMS:
obsd 590.1483, calcd 590.1490 (C24H23O2F13).
Supporting Information Available: Experimental de-
tails and characterization data of compounds 7, 11b-15b.
Copies of 1H, 19F (if relevant), and 13C NMR spectra of
compounds 1e, 5-7, 9-12, 14, 15b. This material is available
free of charge via the Internet at http://pubs.acs.org.
11â-Tridecafluorohexyl-3,17â-dibenzyloxyestra-
1,3,5(10)-triene (11a). 1H NMR (300 MHz, CDCl3): δ 0.92
(broad s, 3H), 1.05-1.26 (m, 3H), 1.37-1.64 (m, 5H), 1.79 (dd,
J ) 12.6, 2.1 Hz, 1H), 1.94-2.02 (m, 2H), 2.43 (d, J ) 14.7
JO051424K
8912 J. Org. Chem., Vol. 70, No. 22, 2005