Chemical synthesis of (S)-spiro(estradiol-17,2′-[1,4]oxazinan)- 6′-one derivatives bearing two levels of molecular diversity

Article abstract of DOI:10.1002/ejoc.200701077

The present study shows the development of a strategy that can afford different estradiol derivatives bearing a 17-spiro-δ-lactone moiety with a nitrogen atom inserted in the lactone ring. Such [1,4]oxazinan-6′-one derivatives contain two levels of molecular diversity introduced to modulate biological activity. The strategy employed to prepare these compounds includes the stereoselective formation of a 17β-oxirane from the carbonyl group of estrone, the regioselective opening of the resulting oxirane with various hydrophobic amino acids (giving the first level of diversity), the spirolactonization, the alkylation of nitrogen by different functional groups (producing the second level of diversity), and final deprotection, when necessary. A series of (S)-spiro-(estradiol-17,2′-[1,4]oxazinan)-6′- one derivatives was generated to illustrate the usefulness of this diversification strategy, which could easily be extended to other steroidal and nonsteroidal ketone scaffolds for interacting with additional biological targets. Wiley-VCH Verlag GmbH & Co. KGaA, 2008.

Full text of DOI:10.1002/ejoc.200701077

FULL PAPER
DOI: 10.1002/ejoc.200701077
Chemical Synthesis of (S)-Spiro(estradiol-17,2Ј-[1,4]oxazinan)-6Ј-one
Derivatives Bearing Two Levels of Molecular Diversity
François Rouillard,[a] Jenny Roy,[a] and Donald Poirier*[a]
Keywords: Steroids / Lactones / Amino acids / Estradiol derivatives / Enzyme inhibitors
The present study shows the development of a strategy that
can afford different estradiol derivatives bearing a 17-spiro-
δ-lactone moiety with a nitrogen atom inserted in the lactone
ring. Such [1,4]oxazinan-6Ј-one derivatives contain two
levels of molecular diversity introduced to modulate bio-
logical activity. The strategy employed to prepare these com-
pounds includes the stereoselective formation of a 17β-
oxirane from the carbonyl group of estrone, the regioselec-
spirolactonization, the alkylation of nitrogen by different
functional groups (producing the second level of diversity),
and final deprotection, when necessary. A series of (S)-spiro-
(estradiol-17,2Ј-[1,4]oxazinan)-6Ј-one derivatives was gener-
ated to illustrate the usefulness of this diversification strat-
egy, which could easily be extended to other steroidal and
nonsteroidal ketone scaffolds for interacting with additional
biological targets.
tive opening of the resulting oxirane with various hydro- (© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
phobic amino acids (giving the first level of diversity), the
Germany, 2008)
HSD (Figure 1A). It was later demonstrated that this lac-
tone was the most potent inhibitor among a series of spiro-
γ-lactones having a C19-steroid scaffold.[6] From these re-
sults, it was established that a spiro-γ-lactone is an impor-
tant pharmacophore for the inhibition of type 2 17β-HSD.
The same group later reported that the spiro-δ-lactone of
E2 (Figure 1B) demonstrated the best inhibition over a
series of differently substituted lactones.[7] It has also been
shown that spiro-δ-lactones inhibited type 5 17β-HSD.[8]
More interestingly, introduction of a chemical group in the
4Ј- and 5Ј-positions of the lactone ring can modulate bio-
logical activity.[9] In light of these results, our group has
developed a strategy that can afford different compounds
bearing a δ-lactone moiety with a nitrogen atom inserted in
the spirolactone ring. Such [1,4]oxazinan-6Ј-one derivatives
(Figure 2) contain two levels of chemical diversity repre-
sented by the R1 and R2 groups. The strategy employed
Introduction
17β-hydroxysteroid dehydrogenases (17β-HSDs) play an
important role in the regulation of steroid hormones, such
as estrogens and androgens, by catalyzing the reduction of
17-ketosteroids or the oxidation of 17β-hydroxysteroids by
using NAD(P)H or NAD(P)+ as cofactor, respectively.[1]
Because it is involved in the last step of sex steroid hormone
biosynthesis from cholesterol, the 17β-HSD enzyme family
is an interesting target for controlling the concentration of
estrogens and androgens. The 17β-HSDs activities are wide-
spread in human tissues, not only in classic steroidogenic
tissues, such as testis, ovary, and placenta, but also in a
large series of peripheral intracrine tissues.[2] Since the
1990s, new types of 17β-HSDs were reported,[3] indicating
that a fine regulation is carried out. In a therapeutic per-
spective, 17β-HSD inhibitors are useful tools in the treat-
ment of estrogen-sensitive pathologies (breast, ovarian, and
endometrium cancers) and androgen-sensitive pathologies
(prostate cancer, benign prostatic hyperplasia, hirsutism,
etc.).[4] Synthetic inhibitors can also help to clarify their
role when not fully understood.
In a previous study,[5] a spiro-γ-lactone derivative of C18-
steroid estradiol (E2) was reported to inhibit type 2 17β-
[a] Medicinal Chemistry Division, Oncology and Molecular Endo-
crinology Laboratory, CHUL Research Center and University
Laval,
Figure 1. Chemical structures of two type-2 17β-HSD inhibitors
with a C18-steroid (estradiol) nucleus; one contains a spiro-γ-
lactone (A) and the other one contains a spiro-δ-lactone (B) as
pharmacophore. The stereogenic centers are illustrated only for ste-
roid A, but they are the same for all the other steroid derivatives
reported in this paper. Partial numbering of carbon atoms is repre-
sented on steroid B.
CHUQ – Pavillon CHUL, Québec, Québec G1V 4G2, Canada
Fax: +1-418-654-2761
E-mail: donald.poirier@crchul.ulaval.ca
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
2446
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 2446–2453

Chemical Synthesis of (S)-Spiro(estradiol-17,2Ј-[1,4]oxazinan)-6Ј-one Derivatives
to prepare these compounds included the formation of a methodology.[10] A regioselective aminolysis of this oxirane
steroidal C17-oxirane, the opening of the resulting oxirane was next performed to introduce the first level of molecular
with various hydrophobic amino acid derivatives to give the diversity by using an amino acid methyl ester as a versatile
first level (R1) of diversity, the spirolactonization, the alky- building block.
lation of nitrogen by different groups to produce the second
level (R2) of diversity, and the final deprotection. We thus
report the development of a versatile route that allows the
chemical synthesis of different (S)-spiro(estradiol-17,2Ј-
[1,4]oxazinan)-6Ј-one derivatives with two sites of diversifi-
cation suitable for modulating biological activities.
Scheme 1. Synthesis of compounds 4 and 5. Reagents and condi-
tions: (a) (CH3)3SI, DMSO, NaH; (b) -phenylalanine methyl ester,
MeOH, Schlenk tube, 90–100 °C; (c) NaH, THF; (d) (iPr)2EtN,
benzyl bromide or 4-iodobenzylbromide, CH2Cl2, Schlenk tube,
70 °C.
Figure 2. The retrosynthesis of (S)-spiro(estradiol-17,2Ј-[1,4]oxaz-
inan)-6Ј-one derivatives containing two levels of molecular diver-
sity; one (R1) coming from the amino acid used for the oxirane
opening and the second (R2) coming from N-alkylation upon the
secondary amine.
Preparation of Free Amino Acids (Neutralization of the
Corresponding Salt)
The trickiest step to achieve in this methodology devel-
opment was to find a method that would be efficient to
obtain the free amino acid methyl ester from its hydrochlo-
ride salt (Table 1). We first tried organic bases in organic
solvents. We thought that the free amino acid methyl ester
would become soluble in the organic solvent and after a
Results and Discussion
The synthesis of [1,4]oxazinan-6Ј-one derivatives targeted simple washing step of the organic layer with water to re-
in our study started from ketosteroid estrone (E1). For move the new salt formed, we would retrieve the desired
chemical strategy development purposes, the phenol of E1 free amino acid methyl ester by simple evaporation. This
was protected as a methyl ether, which is very stable under strategy proved unsuccessful (Table 1, Entries 1 and 2) re-
various reaction conditions and therefore well indicated for gardless of the kind of amino acid methyl ester hydrochlo-
the first part of our work (Scheme 1). The 17-carbonyl of ride salt, type of solvent, and strength of base used (Et3N
3-methyl-O-E1 was thus treated with dimethylsulfonium or DIPEA in EtOAc, CH2Cl2, or acetone). We achieved
methylide to generate the corresponding oxirane 1 with the better results by dissolving the amino acid methyl ester hy-
appropriate (17S)-stereochemistry according to a known drochloride salt in water with the use of a weak inorganic
Table 1. Preparation of free amino acid methyl ester from corresponding HCl salt.[a]
Entry
Experimental conditions[b]
Gly
Ala
Val
Leu
Met
Phe
1
2
3
4
5
6
7
Et3N/EtOAc or CH2Cl2
DIPEA/EtOAc/CH2Cl2/acetone
Cs2CO3 (1.5 equiv.)/H2O/EtOAc/extraction
PMP resin (4 equiv.)/THF/12 h/strong agitation
PMP resin (4 equiv.)/THF/48 h/strong agitation
EPMP resin (4 equiv.)/THF/48 h/strong agitation
EPMP resin (8 equiv.)/THF/48 h/strong agitation
–
–
0
4
6
–
–
N/C[c]
–
–
34
45
57
61
–
–
–
44
49
63
68
67
–
–
71
–
–
75
–
N/C[c]
N/C[c]
75
–
–
N/C[c]
0
6
37
–
79
76
–
[a] Data expressed as % of free -amino acid methyl ester recovered after several extractions (Entry 3) or filtrations (Entries 4–7). [b]
PMP resin: piperidinomethyl polystyrene resin; EPMP resin: ethylpiperazinomethyl polystyrene resin. [c] N/C: Nonconclusive assay.
Eur. J. Org. Chem. 2008, 2446–2453
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2447

F. Rouillard, J. Roy, D. Poirier
FULL PAPER
base such as Cs2CO3 (Table 1, Entry 3). Thus, the free ether) in an aprotic solvent (toluene)[15] were also tried, but
amino acid methyl ester was extracted by using CH2Cl2 or without success. In fact, the yields were substantially lower
EtOAc as solvent. EtOAc was the solvent that offered the when the Lewis acid was used than the previous procedures
best yields. Unfortunately, the yield of extraction dropped without catalysis. The possibility that the use of a Lewis
in a linearly dependent manner as the hydrophobic charac- acid promotes the polymerization of the free amino acid
ter of the amino acid methyl ester used decreased with no methyl ester is a credible hypothesis to explain the very low
possibility at all of extracting the glycine methyl ester or the yields of oxirane opening. At this point, finding conditions
alanine methyl ester. The third strategy used was the best that can lower the amount of amino acid methyl ester for
of all three and involved the use of scavenger base resins the opening of oxirane becomes an important issue.
and THF as aprotic solvent (Table 1, Entries 4–7). The two
The opening of oxirane 1 (Scheme 1) was next performed
scavengers used were piperidinomethyl polystyrene and eth- in a Schlenk tube so that we could substantially raise the
ylpiperazinomethyl polystyrene resins. Different quantities temperature without losing the solvent by evaporation and
of resin were used with different reaction times. The best increasing the pressure in the flask. With this procedure, we
combination of all the parameters tested required the use were able to lower the amount of amino acid methyl ester
of 4 equiv. of ethylpiperazinomethyl polystyrene resin in needed from 30 to 10 equiv. by heating the Schlenk tube
THF and its reaction with the amino acid methyl ester hy- containing the reactants in EtOH at 90–100 °C for a period
drochloride salt over the course of 48 h with strong agita- of 2–4 d. One problem encountered was that some transes-
tion (Table 1, Entry 6). We also increased the amount of terification of the methyl ester occurred in EtOH giving a
ethylpiperazinomethyl polystyrene up to 8 equiv. but with- mixture of two compounds; one having a methyl ester and
out any improvement in the yields. The workup of the reac- the other one having an ethyl ester. This problem was easily
tion only required filtration of the resin followed by evapo- resolved by performing the reaction in MeOH instead of
ration of THF. However, the concentrated free amino acid EtOH with an acceptable yield of 73% for 2. According to
methyl ester starts to polymerize after 24 h, with complete another study,[16] it was possible to lower the amount of
polymerization after 72 h. Therefore, it is important to use amine needed for the opening of a hindered epoxide at the
the free amino acid methyl ester as soon as obtained to 2,3-position of a steroid skeleton to a quantity as low as
perform the next reaction step (oxirane opening) in order 3 equiv. by using gadolinium triflate [Gd(OTf)3] as a cata-
to avoid this polymerization issue.
lyst. We tried this procedure, but it did not give us the ex-
pected results. Contrary to the amines used in this study,
the amino acid methyl ester contains oxygen atoms that can
chelate the catalyst. Consequently, this procedure probably
increases the polymerization problem. In summary, the best
Oxirane Opening
Two procedures were first attempted to perform the oxirane opening conditions were 10 equiv. of freshly gener-
opening of the (17S)-oxirane moiety of 1 with simple pri- ated amino acid methyl ester dissolved in MeOH and
mary amine to verify how the opening of an oxirane at this heated (90–100 °C) in a Schlenk tube for 2–4 d.
position behaves. Indeed, as reported for the low reactivity
of steroidal 17-ketones,[11] it is known that the hindered ste-
Lactonization
roidal (17S)-oxirane is much-less reactive to aminolysis
than other oxiranes.[12] The first method tried was a regiose-
lective opening of the oxirane with heptylamine and lithium
perchlorate for 4 d heated at reflux in acetonitrile.[13] Con-
trary to the opening of an oxirane at the steroidal 3-posi-
tion,[14] we were not able to achieve the reaction unless we
used more then 20 equiv. of amine. We also tried the oxir-
ane opening using the amine in anhydrous EtOH at reflux
for 4 d but, unfortunately, with the same result. This last
procedure was next tested with an amino acid methyl ester.
Explicitly, the procedure tested was to dissolve the oxirane
starting material in EtOH and use a large excess of the -
phenylalanine methyl ester (up to 30 equiv., in our case, to
have full completion of the reaction) and to heat the reac-
tion mixture at reflux (60 °C) for at least 72 h. We could
retrieve the desired compound from this procedure, but the
excess amount of amino acid methyl ester used was a seri-
ous drawback for further purification of the mixture and
also for the waste of precious amino acid methyl ester start-
ing material, particularly because this starting material can-
not be recovered owing to the polymerization problem.
Different lactonization conditions were tested with hy-
droxy ester 2 as a model compound. Two acidic conditions
using p-toluenesulfonic acid in toluene or 5–10% aqueous
HCl in THF at room temperature to 65 °C were first tried,
but they both gave an unwanted reaction: a suprafacial 1,2-
migration of the 18-CH3 with dehydration of the alcohol at
the 17-position of the steroid skeleton, which is a well-
known side reaction in steroid chemistry.[17] The basic con-
ditions using sodium methylate/MeOH or sodium ethan-
olate/EtOH in benzene at 65 °C gave also inconclusive re-
sults. Finally, the only conditions that gave desired com-
pound 3 in one step was the use of NaH in THF at room
temperature. In this case, the alcoholate generated from ter-
tiary alcohol 2 attacks the methyl ester to generate the
spiro-[1,4]oxazinan-6Ј-one derivative 3 in 55% yield.
N-Diversification of [1,4]Oxazinan-6Ј-one Derivative 3
The presence of a secondary amine inserted into the
Conditions using a Lewis acid (boron trifluoride diethyl spiro-δ-lactone ring is an interesting way to introduce dif-
2448
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 2446–2453

Chemical Synthesis of (S)-Spiro(estradiol-17,2Ј-[1,4]oxazinan)-6Ј-one Derivatives
ferent types of functional groups. As examples, we now re- tected as TBDMS derivative 7. The synthesis of the oxirane
port the alkylation of the NH of 3 with benzyl bromide or group has to be performed before protection of the phenolic
4-iodobenzyl bromide to generate tertiary amines 4 or 5, group of E1 otherwise the TBDMS group is removed while
but diversity can be extended to other halogenoalkyls, acyl doing the oxirane synthesis.[10d] The opening of oxirane 7
chlorides, activated carboxylic acids, sulfonyl chlorides, cya- was easily achieved to give alcohols 8a–e in good yields
nates, and thiocyanates in order to obtain tertiary amines, ranging from 69 to 80%. For the lactonization of 8a–e with
amides, sulfonamides, ureas, and thioureas, respectively. NaH in THF, no more than 1 equiv. of NaH should be used
Only one condition was tried to introduce the benzyl group, because the TBDMS is removed and this protecting group
because the yields were good enough (76–91%) when using is absolutely necessary during the next step, which is the
benzyl bromide and diisopropylethylamine as base. The in- diversification of the secondary amine of 9a–e into 10a–e.
troduction of a second level of diversity upon the secondary As a final step, the removal of the TBDMS protecting
amine was easily performed and it led us to believe that it group of 10a–e was easily achieved by a treatment with 2%
would be feasible to link other kinds of groups upon this HCl in methanol. The overall unoptimized yields of 11a–e
amine to obtain other kinds of functionality, as mentioned for the sequence of reactions (six steps) ranged from 18 to
above.
20%. These yields could, however, be improved by optimiz-
ing the lactonization, which is the limiting step with yields
of 52–55%.
Synthesis of a Series of (S)-Spiro(estradiol-17,2Ј-
[1,4]oxazinan)-6Ј-ones
Phenolic steroids generally have better affinity on ste-
roidogenic receptors and enzymes than their corresponding
methyl ether derivatives. Consequently, the same strategy
reported above for the synthesis of 3-methyl-O-E2 deriva-
tives was applied to synthesize E2 derivatives with a 3-OH
group and an [1,4]oxazinan-6Ј-one moiety at the 17-posi-
tion (Scheme 2). Estrone was first treated with dimethylsul-
fonium methylide to afford oxirane 6, which was next pro-
NMR Spectroscopic Analysis of a Representative Final
Compound
The compound 4 was analyzed by NMR spectroscopy to
confirm its expected structure. The assignment of all carbon
signals and assignment of key proton signals are reported
in Figure 3. In the 13C NMR spectrum, the signals of the
E2 nucleus (C-1 to C-18) were easily identified on the basis
of literature data[18] and confirmed by a series of 2D NMR
spectroscopic experiments (COSY, TOCSY, APT, HSQC,
HMBC, and NOESY).[19] The C-17 (or C-2Ј) signal at δ =
91.9 ppm and C-6Ј signal at δ = 170.7 ppm are very charac-
teristic of the expected spiro[1,4]oxazinan-6Ј-one ring, and
these two values are closely related to reported data of the
spiro-δ-lactone shown in Figure 1B.[7] After we identified
the aromatic signals of the two benzyl groups, the only re-
maining CH signal at δ = 64.7 ppm was necessarily CH-
5Ј. From this signal and by using 2D NMR spectroscopic
experiments, it was then possible to identify the CH2-1ЈЈ (δ
= 35.0 ppm), a value that corresponds very well to a ben-
zylic methylene linked to a carbon atom. The two other
CH2 signals remaining to be assigned were located at δ =
53.2 and 58.5 ppm; such chemical shifts are characteristic
1
to an NCH2 group. In the H NMR spectra, the two pro-
tons of each NCH2 are nonequivalent and their chemical
shifts are very different. To discriminate which signals are
from CH2-3Ј and which signals are from CH2-1ЈЈЈ, the
HMBC experiment was useful. In fact, the two proton sig-
nals associated with the carbon signal at δ = 53.2 ppm
(from HSQC experiment) showed no long-range (2JC,H and
3JC,H) correlations with aromatic carbon signals in HMBC
experiment then identifying the CH2-3Ј. On the other hand,
the two proton signals associated with the carbon signal at δ
= 58.5 ppm (from HSQC experiment) clearly showed long-
Scheme 2. Synthesis of compounds 11a–11e. Reagents and condi-
tions: (a) (CH3)3SI, DMSO, NaH; (b) TBDMSCl, imidazole; (c) -
or -amino acid methyl ester, MeOH, Schlenk tube, 90–100 °C; (d)
NaH, THF; (e) (iPr)2EtN, benzyl bromide, CH2Cl2, Schlenk tube,
70 °C; (f) 2% HCl in MeOH.
3
range (2JC,H and JC,H) correlations with aromatic carbon
signals in the HMBC experiment then confirming the CH2-
1ЈЈЈ.
Eur. J. Org. Chem. 2008, 2446–2453
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2449

F. Rouillard, J. Roy, D. Poirier
FULL PAPER
ports in the literature showing that an interacting group
added in proximity to the D-ring of E2 can modulate the
biological activity of 17β-HSDs inhibitors, steroid sulfatase
inhibitors, estrogen receptor antagonists, and estrogen re-
ceptor antagonists.[4–7,21]
Figure 3. NMR (HSQC) spectra of the representative final com-
1
pound 4 showing JC,H correlations between the protons and car-
bon atoms. The assignment of all carbon atoms and key protons
was indicated on the spectra. The experiment was done with CDCl3
as solvent by using a 400 MHz NMR apparatus.
X-ray Analysis of a Representative Final Compound
Compound 11d was crystallized and its 3D structure de-
termined by X-ray analysis.[20] As expected, X-ray analysis
showed the typical shape of an estra-1,3,5(10)-triene steroid
core (C-1 to C-18) and an [1,4]oxazinan-6Ј-one moiety (O-
1Ј to C-6Ј) linked at spirocarbon 17 (Figure 4a). The (17S)
stereochemistry is in accord with the sequence of reactions
starting from (17S)-oxirane 7 and needed for the synthesis
of 11d. The (5ЈR) stereochemistry observed corresponds
with the -phenylalanine methyl ester used as a building
block. The configuration of the [1,4]oxazinan-6Ј-one moiety
and the orientation of the two benzyl groups at N-4Ј and
C-5Ј are better viewed with the 3D-structure representation
reported in Figure 4b. The spiro[1,4]oxazinan-6Ј-one ring
adopts a pseudochair configuration with the two benzyl
groups in trans-4Ј,5Ј-diequatorial position. Being rigid, this
pseudochair conformer is a good scaffold that can direct
the two elements of molecular diversity (herein represented
by the benzyl groups) in two different positions according
Figure 4. X-ray 3D-structure analysis of the representative final
compound 11d crystallized in EtOH.[20] Three different pictures (a–
c) were represented by using the CS Chem3D Std software (Cam-
bridge Soft Corporation) and the pdb data file obtained from the
X-ray data.
Conclusions
A strategy was reported for the synthesis of (S)-spiro-
to the selected amino acid ( or ). As can be seen in Fig- (estradiol-17,2Ј-[1,4]oxazinan)-6Ј-one derivatives having
ure 4c, the benzyl groups of 11d are both positioned in the two levels of molecular diversity. The first level of diversity
same right side of the molecule but are perpendicular to was introduced by an amino acid for opening an oxirane
each other. However, in the case of 11c, which is obtained easily obtained from a steroidal ketone. The second level of
with -Phe, the two benzyl groups would be in opposite diversity was generated by an alkylation of the NH func-
directions: one on the left side and the other on the right tionality inserted in the spiro-δ-lactone ring. Although the
side. The positioning of the two elements of molecular di- aminolysis step generating the first level of diversity is re-
versity differently and in proximity to the steroid D-ring, is stricted to - and -amino acids that can be extracted in
an interesting characteristic, because there are several re- organic solvent after the neutralization of their HCl salt,
2450
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 2446–2453

Chemical Synthesis of (S)-Spiro(estradiol-17,2Ј-[1,4]oxazinan)-6Ј-one Derivatives
warmed to room temperature and left to react for another 0.5 h
the reactivity of the secondary amine offers great possibility
of diversification (tertiary amines, amides, sulfonamides,
and ureas) for the second level of diversity. The spiro[1,4]-
oxazinan-6Ј-one ring also offers a rigidity that directs the
elements of diversity introduced by the two levels of diver-
sity. The strategy was exemplified with a 17-ketosteroid
(estrone) as starting material, but the strategy could be ad-
vantageously used at other steroidal positions or with other
steroid nuclei. Furthermore, all types of ketones could be
before completion. The reaction was quenched at 0 °C with a satu-
rated NH4Cl aqueous solution and extracted with EtOAc. The or-
ganic layers were combined and washed several times with water
before drying over Na2SO4, filtered and evaporated under reduced
pressure. The crude material was then purified by flash column
chromatography (hexanes/EtOAc, 9:1). White solid (4.85 g, 79%).
1H NMR (400 MHz, CDCl3): δ = 0.92 (s, 3 H, 18-CH3), 0.90–2.35
(residual CH and CH2), 2.65 and 2.97 (2 d, J = 5.0 Hz, 2 H, 17α-
CH2), 2.88 (m, 2 H, 6-CH2), 3.78 (s, 3 H, CH3O), 6.64 (d, J =
used, thus constituting a third level of diversity. This ad- 2.6 Hz, 1 H, 4-CH), 6.72 (dd, J1 = 8.6 Hz, J2 = 2.7 Hz, 1 H, 2-
CH), 7.20 (d, J = 8.6 Hz, 1 H, 1-CH) ppm. 13C NMR (75.5 MHz,
CDCl3): δ = 14.3, 23.3, 26.0, 27.2, 29.1, 29.7, 33.9, 38.9, 40.4, 43.9,
ditional level of diversity greatly extends the usefulness of
our strategy of diversification, not only for developing in-
51.8, 53.7, 55.2, 70.6, 111.5, 113.8, 126.3, 132.4, 137.9, 157.4 ppm.
hibitors of steroidogenic enzymes but also for interacting
LRMS: calcd. for C20H27O2 [M + H]+ 299.2; found 298.7.
with a large series of biological targets.
Opening of Oxirane 1. Synthesis of Methyl N-{[(17β)-17-Hydroxy-
3-methoxyestra-1,3,5(10)trien-17-yl]methyl}-L-phenylalaninate (2)
Step 1, Method A: -Phenylalanine methyl ester·HCl was freed
from the salt by dissolving the salt in H2O. Cs2CO3 (1.5 equiv.) was
then added and the mixtures was stirred for 0.5 h before several
extractions with EtOAc. After evaporation of the solvent, the free
-phenylalanine methyl ester was recovered (see Table 1, Entry 3).
Experimental Section
General Remarks: Estrone and 3-methyl-O-estrone were purchased
from Steraloids Inc. (Newport, RI, USA). Chemical reagents of
highest purity and anhydrous solvents were obtained from Sigma–
Aldrich Canada Ltd. (Oakville, ON, Canada) and Fisher Scientific
(Montréal, QC, Canada). The two resins piperidinomethyl polysty-
rene (3.5 mmolg–1, 100–200 mesh) and ethylpiperazinomethyl poly-
styrene (3.7 mmolg–1, 100–200 mesh) were obtained from Matrix
Innovation (Montréal, QC, Canada). Tetrahydrofuran used in an-
hydrous reactions was distilled from benzophenone ketyl. Reac-
tions were run under an inert (argon) atmosphere in oven-dried
glassware. Analytical thin-layer chromatography (TLC) was per-
formed on 0.20-mm silica gel 60 F254 plates (Fisher Scientific),
and compounds were visualized by using UV light or ammonium
molybdate/sulfuric acid/water (with heating). Flash column
chromatography was performed with Silicycle R10030B 230–
400 mesh silica gel (Québec, QC, Canada). Infrared spectra (IR)
were obtained from a thin film of compound usually solubilized in
CDCl3 and deposited upon a NaCl pellet. They were recorded with
a Perkin–Elmer 1600 FTIR spectrometer (Norwalk, CT, USA) and
only characteristic bands are reported. Nuclear magnetic resonance
(NMR) spectra were recorded at 300 MHz (1H) and 75.5 MHz
(13C) with a Bruker AC/F300 spectrometer (Billerica, MA, USA)
or at 400 MHz (1H) and 100.6 MHz (13C) with a Bruker Avance
400 digital spectrometer. The chemical shifts are expressed in ppm
and are referenced to residual chloroform (δ = 7.26 ppm for 1H
NMR and 77.0 ppm for 13C NMR). Low-resolution mass spectra
(LRMS) were recorded with a PE Sciex API-150 ex apparatus (Fos-
ter City, CA, USA). High-resolution mass spectra (HRMS) were
provided by The Mass Spectrometry Unit (McGill University,
Montréal, QC, Canada). The X-ray analysis was performed at the
Laboratoire de diffraction des rayons-X de l’Université de Mon-
tréal (Montréal, QC, Canada). The names of new compounds were
obtained by using the ACD/I-Lab Web service (ACD/IUPAC
Name 8.05).
Step 1, Method B: -Phenylalanine methyl ester·HCl was freed from
the salt by dissolving the salt in THF. The THF solution was then
transferred to a Schlenk tube containing ethylpiperazinomethyl
polystyrene scavenger resin (4 equiv.). The mixture was strongly ag-
itated for 48 h. Then, the mixture was filtered, and the filtrate was
evaporated under reduced pressure to obtain the free -phenylala-
nine methyl ester (see Table 1, Entry 6).
Step 2: Oxirane 1 (304 mg, 0.76 mmol) was dissolved in dry MeOH
in a Schlenk tube. After bubbling argon to saturation, -phenylala-
nine methyl ester (1.11 g, 7.6 mmol) was added, and the solution
was heated at 90–100 °C for 48 h. After completion of the reaction,
the mixture was left to cool to room temperature before silica gel
was added and the solvent evaporated under reduced pressure. The
crude material was purified by flash column chromatography (hex-
anes/EtOAc, 9:1). Viscous oil (265 mg, 73%). IR (film): ν = 3447
˜
(OH), 3327 (NH, weak), 1737 (C=O, ester) cm–1. 1H NMR
(400 MHz, CDCl3): δ = 0.90 (s, 3 H, 18-CH3), 1.20–2.30 (residual
CH and CH2), 2.53 and 2.75 (2 d, J = 11.4 Hz, 2 H, 17α-CH2),
2.84 (m, 2 H, 6-CH2), 2.95 and 3.05 (2 m, 2 H, CH2Ph), 3.53 (t, J
= 3.8 Hz, NCHCOO), 3.72 (s, 3 H, CH3O from ester), 3.78 (s, 3
H, 3-CH3O), 6.63 (d, J = 2.6 Hz, 1 H, 4-CH), 6.71 (dd, J1 = 8.6 Hz,
J2 = 2.4 Hz, 2-CH), 7.18 (d, J = 8.5 Hz, 1 H, 1-CH), 7.24 (m, 5 H,
CH2Ph) ppm. 13C NMR (75.5 MHz, CDCl3): δ = 14.0, 23.3, 26.0,
27.3, 29.6, 31.9, 34.8, 39.0, 39.6, 43.5, 45.2, 49.6, 51.7, 54.0, 55.0,
63.4, 81.3, 111.2, 113.5, 126.1, 126.7, 128.3 (2ϫ), 129.0 (2ϫ), 132.4,
137.0, 137.7, 157.2, 174.7 ppm. LRMS: calcd. for C30H40NO4 [M
+ H]+ 478.3; found 478.2.
Lactonization of 2. Synthesis of (5ЈS,8R,9S,13S,14S,17S)-5Ј-Benzyl-
3-methoxy-13-methyl-6,7,8,9,11,12,13,14,15,16-decahydro-6ЈH-
spiro(cyclopenta[a]phenanthrene-17,2Ј-[1,4]oxazinan)-6Ј-one (3): To
compound 2 (454 mg, 0.835 mmol) dissolved in dry THF (33 mL)
was added NaH (60% in oil, 33 mg, 0.835 mmol). The reaction was
kept at room temperature and completed after 1 h. The reaction
was then quenched at 0 °C with a saturated NH4Cl aqueous solu-
tion and extracted with EtOAc. The organic layers were combined
and washed with water, dried with Na2SO4, filtered, and evaporated
under reduced pressure. The crude material was purified by flash
column chromatography (hexanes/EtOAc, 8:2). White solid (55%).
(8R,9S,13S,14S,17S)-3-Methoxy-13-methyl-6,7,8,9,11,12,13,14,15,16-
decahydrospiro(cyclopenta[a]phenanthrene-17,2Ј-oxirane) (1):[10a–10e]
In a 2-liter flask under an argon atmosphere was added DMSO
(175 mL) followed by NaH (60% in oil, 4.94 g, 123.60 mmol). The
mixture was heated at 75 °C for 1 h and then cooled to room tem-
perature and anhydrous THF (175 mL) was then added. The mix-
ture was cooled to 0 °C and trimethylsulfonium iodine (25.22 g,
123.60 mmol) dissolved in DMSO (175 mL) was added, followed
by 3-methyl-O-estrone (5.85 g, 20.60 mmol) dissolved in anhydrous
THF (175 mL). The mixture was kept at 0 °C for 0.5 h and then
IR (film): ν = 3350 (NH, weak), 1727 (C=O, lactone) cm–1. 1H
˜
NMR (400 MHz, CDCl3): δ = 0.89 (s, 3 H, 18-CH3), 1.10–2.20
Eur. J. Org. Chem. 2008, 2446–2453
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2451

F. Rouillard, J. Roy, D. Poirier
FULL PAPER
(residual CH and CH2), 2.83 and 3.24 (2 m, 4 H, CH2Ph and 17α-
Supporting Information (see also the footnote on the first page of
CH2), 2.85 (m, 2 H, 6-CH2), 3.78 (s, 3 H, CH3O), 3.80 (m, 1 H,
this article): Experimental procedures and spectroscopic data for
NCHCOO), 6.62 (d, J = 2.4 Hz, 1 H, 4-CH), 6.71 (dd, J1 = 8.4 Hz, compounds 6, 7, 8a–8e, 9a–9e, 10a–10e and 11a–11e, 13C NMR
J2 = 2.4 Hz, 1 H, 2-CH), 7.19 (d, J = 8.4 Hz, 1 H, 1-CH), 7.30 (m, spectra for final compounds 4, 5, 11a–11e.
5 H, CH2Ph) ppm. 13C NMR (75.5 MHz, CDCl3): δ = 14.2, 22.9,
25.9, 27.3, 29.6, 32.4, 35.6, 37.5, 39.0, 43.4, 46.8, 47.9, 49.6, 55.1,
57.7, 93.4, 111.5, 113.7, 126.2, 127.3, 128.8 (2ϫ), 129.6 (2ϫ), 131.9,
Acknowledgments
137.0, 137.7, 157.4, 172.0 ppm. LRMS: calcd. for C29H36NO3 [M
We would like to thank the Canadian Institutes of Health Research
+ H]+ 446.3; found 446.1.
(CIHR) for their financial support and Le Fonds de la Recherche
(5ЈS,8R,9S,13S,14S,17S)-4Ј,5Ј-Dibenzyl-3-methoxy-13-methyl-
6,7,8,9,11,12,13,14,15,16-decahydro-6ЈH-spiro(cyclopenta[a]phen-
anthrene-17,2Ј-[1,4]oxazinan)-6Ј-one (4): To compound 3 (90 mg,
0.201 mmol) dissolved in dry CH2Cl2 (5 mL) was added DIPEA
(60 µL, 0.342 mmol) followed by benzyl bromide (41 µL,
0.342 mmol). The resulting mixture was heated in a Schlenk tube
at 75 °C for 12 h. The reaction quenched with a saturated NH4Cl
aqueous solution, extracted with CH2Cl2, and washed with water.
The organic layers were combined, dried with Na2SO4, filtered, and
evaporated under reduced pressure. The resulting crude material
was purified by flash column chromatography (hexanes/EtOAc,
en Santé du Québec (FRSQ) for a Senior Scholarship (D. P.). We
are grateful to Orval Mamer and Alain Lesimple from The Mass
Spectrometry Unit (McGill University) for HRMS analyses. Care-
ful reading of the manuscript by Micheline Harvey is also greatly
appreciated.
[1] a) V. Luu-The, J. Steroid Biochem. Mol. Biol. 2001, 76, 143–
151; b) F. Labrie, V. Luu-The, S. X. Lin, J. Simard, M. El-Alfy,
G. Pelletier, J. Mol. Endocr. 2000, 25, 1–16; c) F. Labrie, V.
Luu-The, S. X. Lin, J. Simard, C. Labrie, TEM 2000, 11, 421–
427; d) N. Moghrabi, S. Andersson, TEM 1998, 9, 265–270; e)
F. Labrie, V. Luu-The, S. X. Lin, J. Simard, C. Labrie, R. Bre-
ton, Steroids 1997, 62, 148–158.
[2] C. Martel, E. Rheaume, M. Takahashi, C. Trudel, J. Couet, V.
Luu-The, J. Simard, F. Labrie, J. Steroid Biochem. Mol. Biol.
1992, 41, 597–603.
[3] a) H. Peltoketo, V. Isomaa, O. Maentausta, R. Vihko, J. Mol.
Endocr. 1999, 23, 1–11; b) V. Luu-The, Y. Zhang, D. Poirier, F.
Labrie, J. Steroid Biochem. Mol. Biol. 1995, 55, 581–587.
[4] a) D. Poirier, Curr. Med. Chem. 2003, 10, 453–477; b) D. Poir-
ier, R. Maltais, Mini-Rev. Med. Chem. 2006, 6, 37–52; c) D.
Poirier, P. Bydal, M. R. Tremblay, K. M. Sam, V. Luu-The,
Mol. Cell Endocrinol. 2001, 171, 119–128; d) M. R. Tremblay,
D. Poirier, J. Steroid Biochem. Mol. Biol. 1998, 66, 179–191.
[5] a) S. Auger, V. Luu-The, K. M. Sam, D. Poirier, Bioorg. Med.
Chem. Lett. 1994, 4, 2045–2048; b) K. M. Sam, F. Labrie, D.
Poirier, Eur. J. Med. Chem. 2000, 35, 217–225.
[6] a) K. M. Sam, S. Auger, V. Luu-The, D. Poirier, J. Med. Chem.
1995, 38, 4518–4528; b) M. R. Tremblay, V. Luu-The, G. Lebl-
anc, P. Noel, E. Breton, F. Labrie, D. Poirier, Bioorg. Med.
Chem. 1999, 7, 1013–1023.
[7] P. Bydal, S. Auger, D. Poirier, Steroids 2004, 69, 325–342.
[8] W. Qiu, M. Zhou, M. Mazumdar, A. Azzi, D. Ghanmi, V. Luu-
The, F. Labrie, S. X. Lin, J. Biol. Chem. 2007, 282, 8368–8379.
[9] a) F. Labrie, A. Belanger, S. Gauthier, V. Luu-The, Y. Merand,
D. Poirier, L. Provencher, S. M. Singh (Endorecherche Inc.,
Canada), US 6541463, 2003; b) P. Bydal, M.Sc. Thesis, Univer-
sity Laval, Québec (Québec), Canada, 1997, ch. 3 and 4, pp.
98–153.
[10] a) C. E. Cook, R. C. Corley, M. E. Wall, J. Org. Chem. 1968,
33, 2789–2793; b) K. Ponsold, M. Hubner, H. Kasch, I. Noack,
Z. Chem. 1971, 11, 106–107; c) M. Hubner, I. Noack, J. Prakt.
Chem. 1972, 314, 667–668; d) T. Arunachalam, E. Caspi, J.
Org. Chem. 1981, 46, 3415–3420; e) J. L. Zhang, H. B. Zhou,
J. S. Huang, C. M. Che, Chem. Eur. J. 2002, 8, 1554–1562; f)
M. Salman, B. R. Reddy, P. Delgado, P. L. Stotter, L. C.
Fulcher, G. C. Chamness, Steroids 1991, 56, 375–387; g) S. Top,
H. El Hafa, A. Vessieres, J. Quivy, J. Vaissermann, D. W.
Hughes, M. J. McGlinchey, J. P. Mornon, E. Thoreau, G. Ja-
ouen, J. Am. Chem. Soc. 1995, 117, 8372–8380.
[11] a) T. L. Feviq, J. A. Katzenellenbogen, J. Org. Chem. 1987, 52,
247–251; b) M. R. Tremblay, S. Auger, D. Poirier, Synth. Com-
mun. 1995, 25, 2483–2495; c) D. Poirier, R. P. Boivin, M. R.
Tremblay, M. Berube, W. Qiu, S. X. Lin, J. Med. Chem. 2005,
48, 8134–8147.
9:1). White solid (82 mg, 76%). IR (film): ν = 1724 (C=O, lactone)
˜
cm–1. 1H NMR (400 MHz, CDCl3): δ = 0.76 (s, 3 H, 18-CH3),
0.70–2.10 (residual CH and CH2), 2.16 and 2.95 (2 d, J = 13.1 Hz,
2 H, 17α-CH2), 2.77 (m, 2 H, 6-CH2), 3.31 (m, 2 H, 1 H of CH2Ph
and 1 H of NCH2Ph), 3.53 (m, 2 H, NCHCOO and 1 H of
CH2Ph), 3.78 (s, 3 H, CH3O), 4.13 (m, 1 H of NCH2Ph), 6.60 (d,
J = 2.4 Hz, 1 H, 4-CH), 6.72 (dd, J1 = 8.4 Hz, J2 = 2.4 Hz, 1 H,
2-CH), 7.15 (d, J = 8.4 Hz, 1 H, 1-CH), 7.30 (m, 10 H, 2ϫCH2Ph)
ppm. 13C NMR (100.6 MHz, CDCl3): δ = 14.1, 22.6, 25.9, 27.2,
29.6, 32.5, 35.0, 36.5, 38.9, 43.3, 46.5, 49.7, 53.2, 55.1, 58.5, 64.7,
91.9, 111.4, 113.6, 126.2, 126.6, 127.7, 128.2 (2 ϫ), 128.3 (2 ϫ),
129.7 (2 ϫ), 129.9 (2 ϫ), 132.3, 136.1, 137.7, 138.1, 157.4,
170.7 ppm. LRMS: calcd. for C36H42NO3 [M + H]+ 536.3; found
536.3. HRMS: calcd. for C36H42NO3 [M + H]+ 536.31592; found
536.31552.
(5ЈS,8R,9S,13S,14S,17S)-5Ј-Benzyl-3-methoxy-4Ј-(4-iodobenzyl)-
13-methyl-6,7,8,9,11,12,13,14,15,16-decahydro-6ЈH-spiro(cyclo-
penta[a]phenanthrene-17,2Ј-[1,4]oxazinan)-6Ј-one (5): To compound
3 (79 mg, 0.177 mmol) dissolved in dry CH2Cl2 (5 mL) was added
DIPEA (52 µL, 0.300 mmol) followed by 4-iodobenzylbromide
(89 mg, 0.300 mmol). The resulting mixture was heated in a
Schlenk tube at 75 °C for 12 h. The reaction was quenched with
a saturated NH4Cl aqueous solution, extracted with CH2Cl2, and
washed with water. The organic layers were combined, dried with
Na2SO4, filtered, and evaporated under reduced pressure. The re-
sulting crude material was purified by flash column chromatog-
raphy (hexanes/EtOAc, 9:1). White solid (69 mg, 63% yield). IR
(film): ν = 1721 (C=O, lactone) cm–1. 1H NMR (400 MHz, CDCl ):
˜
3
δ = 0.79 (s, 3 H, 18-CH3), 0.85–2.10 (residual CH and CH2), 2.20
and 2.95 (2 d, J = 12.9 Hz, 2 H, 17α-CH2), 2.79 (m, 2 H, 6-CH2),
3.25 and 3.50 (2 dd, J1 = 14.1 Hz, J2 = 5.3 Hz, 2 H, CHCH2Ph),
3.40 and 3.97 (2 d, J = 13.2 Hz, 2 H, NCH2PhI), 3.56 (t, J =
5.1 Hz, 1 H, NCHCH2Ph), 3.77 (s, 3 H, CH3O), 6.60 (d, J = 2.4 Hz,
1 H, 4-CH), 6.72 (dd, J1 = 8.4 Hz, J2 = 2.4 Hz, 1 H, 2-CH), 7.01
(d, J = 8 Hz, 2 H, 2ϫCH from CH2PhI), 7.16 (d, J = 8.4 Hz, 1
H, 1-CH), 7.29 (m, 5 H, CH2Ph), 7.65 (d, J = 8.3 Hz, 2 H, 2ϫCH
from CH2PhI) ppm. 13C NMR (75.5 MHz, CDCl3): δ = 14.2, 22.7,
25.9, 27.2, 29.6, 32.4, 34.8, 36.4, 38.9, 43.3, 46.6, 49.7, 52.7, 55.1,
58.0, 64.2, 92.0, 93.2, 111.4, 113.6, 126.2, 126.7, 128.4 (2ϫ), 129.9
(2ϫ), 131.3 (2ϫ), 132.2, 135.8, 137.4 (2ϫ), 137.7, 138.0, 157.4,
172.6 ppm. LRMS: calcd. for C36H41INO3 [M + H]+ 662.2; found
662.7. HRMS: calcd. for C36H41INO3 [M + H]+ 622.21256; found
622.21238.
[12] R. Maltais, M. R. Tremblay, D. Poirier, J. Comb. Chem. 2000,
2, 604–614.
[13] M. Chini, P. Crotti, F. J. Macchia, J. Org. Chem. 1991, 56,
5939–5942.
2452
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 2446–2453

Chemical Synthesis of (S)-Spiro(estradiol-17,2Ј-[1,4]oxazinan)-6Ј-one Derivatives
[14] a) R. Maltais, V. Luu-The, D. Poirier, Bioorg. Med. Chem.
2001, 9, 3001–3011; b) R. Maltais, V. Luu-The, D. Poirier, J.
Med. Chem. 2002, 45, 640–653.
[15] a) J. Iqbal, A. Pandey, Tetrahedron Lett. 1990, 31, 575–576; b)
G. Sekar, V. K. Singh, J. Org. Chem. 1999, 64, 287–289; c) L. R.
Reddy, M. A. Reddy, N. Bhanumathi, K. R. Rao, Synlett 2000,
3, 339–340.
[16] D. Thibeault, D. Poirier, Synlett 2003, 8, 1192–1194.
[17] P. Bydal, K. M. Sam, D. Poirier, Steroids 1996, 61, 349–353.
[18] a) J. D. Pelletier, F. Labrie, D. Poirier, Steroids 1994, 59, 536–
547; b) P. Dionne, D. Poirier, Steroids 1995, 60, 830–836.
[19] T. D. W. Claridge, Tetrahedron Organic Chemistry Series Vol.
19: High-Resolution NMR Techniques in Organic Chemistry
(Eds.: J. E. Baldwin, R. M. Williams), Pergamon, New York,
1999, pp. 382.
from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
[21]
a) G. M. Allan, H. R. Lawrence, J. Cornet, C. Bubert, D. D.
Fischer, N. Vicker, A. Smith, H. J. Tutill, A. Purohit, J. M. Day,
M. F. Mahon, M. J. Reed, B. V. L. Potter, J. Med. Chem. 2006,
49, 1325–1345; b) N. Vicker, H. R. Lawrence, G. M. Allan, C.
Bubert, A. Smith, H. J. Tutill, A. Purohit, J. M. Day, M. F. Ma-
hon, M. J. Reed, B. V. L. Potter, ChemMedChem 2006, 1, 464–
481; c) J. Messinger, H. H. Thole, B. Husen, B. J. Van Steen, G.
Schneider, J. B. E. Hulshof, P. Koskimies, N. Johansson (Solvay
pharmaceuticals GMBH), WO 2005/047303, 2005; d) D. Poir-
ier, R. Maltais, Mini-Rev. Med. Chem. 2006, 6, 37–52; e) R.
Maltais, M. R. Tremblay, L. C. Ciobanu, D. Poirier, J. Comb.
Chem. 2004, 6, 443–456.
Received: November 13, 2007
Published Online: March 25, 2008
[20] CCDC-676120 contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge
Eur. J. Org. Chem. 2008, 2446–2453
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2453