A regioselective PCl5 mediated dehydration for preparing Δ9,11 corticosteroids

Article abstract of DOI:10.1016/S0040-4039(01)00274-X

A regioselective PCl₅-induced dehydration of 11α-hydroxy corticosteroids was invented to provide the corresponding Δ^9,11 double bond trienes in excellent yield (>90%) with 99:1 regioisomeric ratio to the Δ^11,12 isomers. A syn-elimination mechanism was proposed for this reaction.

Full text of DOI:10.1016/S0040-4039(01)00274-X

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 2639–2642
A regioselective PCl5 mediated dehydration for preparing D9,11
corticosteroids
Xiaoyong Fu,* Chou-Hong Tann, T. K. Thiruvengadam, Junning Lee and Cesar Colon
Synthetic Chemistry, Schering-Plough Research Institute, 1011 Morris Ave., Union, NJ 07083, USA
Received 21 December 2000; revised 1 February 2001; accepted 5 February 2001
Abstract—A regioselective PCl5-induced dehydration of 11a-hydroxy corticosteroids was invented to provide the corresponding
D9,11 double bond trienes in excellent yield (>90%) with 99:1 regioisomeric ratio to the D11,12 isomers. A syn-elimination
mechanism was proposed for this reaction. © 2001 Elsevier Science Ltd. All rights reserved.
The pharmacologically important corticosteroid drugs
such as betamethasone (I, X=F), beclomethasone (I,
X=Cl), dexamethasone (II, X=F, R1=OH, R2=H),
and mometasone (II, X=Cl, R1=Cl, R2=furoyl) have
been well received since their introduction into the
world-wide market. Corticosteroids continue to play
major roles in dermatological and allergy practices.1
The 9,11-halohydrin is crucial for biological activity.2
Introduction of the D9,11 double bond is the key to
functionalizing both the C-9 and C-11 positions. The
existing manufacturing process involves conversion of
the 11a-hydroxyl group into its corresponding mesylate
and subsequent elimination to the desired 9,11 double
bond. The elimination step, however, produces a con-
siderable amount of a side product, the D11,12 isomer
(15%). Both the yield and the purity suffer at this step.
Minimizing or eliminating the D11,12 isomer formation
has been a long standing challenge in order to boost the
overall chemical yield and product quality. We wish to
report a PCl5 mediated elimination approach to selec-
tively dehydrate the 11a-hydroxyl group to form the
delta 9,11 double bond.
It is well understood that the problem in the elimina-
tion step is associated with the configuration of carbon
C-11. With an 11b-OH,3 the base catalyzed anti E2
elimination works well to provide the D9,11 double bond
in greater than 90% yield. Obviously, with the 11a-OH,
the leaving group cannot be anti-coplanar with the H-9
to exclusively form the D9,11 double bond. Although
11b-hydroxylation by fermentation is documented in
the literature,4 attempts to introduce the 11b-OH on the
corresponding C-11 deoxy substrate through biotrans-
formation have proved fruitless. Based on the above
analyses, this situation could be resolved by a syn-elim-
ination approach, or by converting the 11a-OH to an
11b-leaving group to conduct the anti-elimination.
Attempts to convert the 11a-mesylate into the 11b-Br
analogue by refluxing the mesylate with NaBr in ace-
tone were unsuccessful. Treatment of 11a-hydroxy (III)
in THF with NaH followed by addition of CS2 and
CH3I did not produce the desired 11-xanthate5 for the
proposed syn-elimination in a reasonable yield. Bern-
stein et al.6 reported that reaction of a 3,17-bisketal-
11a-OH-steroid with phosphorous oxychloride in
pyridine gave the D9,11 double bond in 90% ‘crude’
yield. Indeed the desired D9,11 (VI) was formed with
excellent ratio to D11,12 isomer (98:2 by HPLC area) by
treatment of III under the Bernstein conditions.
Although the solution yield of the desired product VI
was less than 15% with remaining unidentified impuri-
ties, the D9,11 to D11,12 ratio was encouraging. Careful
analysis of this reaction indicated that under the
employed conditions, D9,11 (VI) was formed very slowly,
and heating was required to push the disappearance of
the starting material. Alternatives to POCl3, such as
PCl5 were used in order to optimize the reaction yield.
Shoppee et al.7 reported that reaction of 5a-androstan-
CH2OH
CH2R1
O
O
HO
HO
OH
OR2
X
X
O
O
II
I
Keywords: dehydration; corticosteroids; betamethasone; mometasone;
syn-elimination; phosphorous pentachloride.
* Corresponding author. Fax: (908)820-6620; e-mail: xiaoyong.fu@
spcorp.com
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)00274-X

2640
X. Fu et al. / Tetrahedron Letters 42 (2001) 2639–2642
Table 1. Solvent and temperature effects on the PCl5
observed from this reaction than that detected from the
reaction run in CH2Cl2/py. This result clearly suggested
that the elimination did not go through the 11b-Cl
intermediate.
dehydration
Solvent
Temp. (°C)
HPLC area (%)
D9,11
D11,12
11b-Cl
The better ratio (Table 1) of D9,11 to D11,12 isomers
might have resulted from a solvent effect. With this
idea in mind, the PCl5 mediated dehydration was car-
ried out in different solvents and at different tempera-
tures. After various experiments, THF was found to be
the solvent choice and −78 to −90°C was observed to be
the optimum reaction temperature (see Table 1 for
results of solvent and temperature effects). With these
conditions, D9,11 (VI) was isolated in greater than 90%
yield with a ratio of D9,11 (VI) to D11,12 (VII) of about
99:1 (Scheme 1).
CH2Cl2/py
py
Rt
Rt
70.4
52.3
74.4
81.0
92.3
95.5
97.5
98.4
11.2
3.8
14.5
7.4
4.9
3.3
1.8
1.3
18.4
44.0
11.0
11.3
2.8
1.2
0.7
0.3
CH2Cl2
THF
−20
Rt
THF
THF
THF
THF
−20
−45
−78
−85
11a-ol with PCl5 in CHCl3 at rt gave 5a-androst-9(11)-
ene in 65% isolated yield with a major by-product
identified as 9a,11b-dichloride. Although the isomeric
ratio of D9,11 to D11,12 was not analyzed by HPLC, the
reported facile elimination was attractive. When 11a-
hydroxy (III) was reacted with PCl5 in CH2Cl2/py at rt,
as expected, the starting material was consumed in less
than five minutes and HPLC analysis of the reaction
mixture indicated three main peaks with an 86:14 ratio
of D9,11 to D11,12. The third peak was isolated and
identified as 11b-Cl compound (V) by mass spectrum
and NMR experiments. The C-11 b configuration was
determined by observation of only small coupling con-
stants (<3 Hz) of 11a-H with H-9 and H-12. The
corresponding 9a,11b-dichloride7 was not detected
under the conditions employed.
The mechanism of this reaction is not clear to us at this
moment, however, it can be speculated (Scheme 1) from
experimental observations. Reaction of 11a-hydroxyl
group with PCl5 produces an 11a-chlorophosphate
intermediate. The phosphorous function should be a
facile leaving group to provide a partial positive charge
at position C-11. At the same time, the chloride would
abstract the proton at C-9 since the two groups can be
arranged in the proximity of the same face. This would,
in return, increase the elimination rate to provide a
D9,11 double bond. It is known that the D9,11 (VI) is
more stable than the D11,12 (VII) isomer because the
former is a tri-substituted double bond. Furthermore,
hydrogens attached to tertiary carbons have weaker
CꢀH bonds, which preferentially undergo intramolecu-
lar elimination. However, the activation energy differ-
ence may not be large enough for the D9,11 to be a
dominating product at high temperature. But at lower
temperature, the ratio of the two olefins can be dramat-
ically increased to favor the desired product. This is
clear from the results of the reactions from room
temperature to −85°C in THF (Table 1). On the other
hand, identification of the 11b-Cl impurity (V) provides
One can speculate that the D9,11 double bond is formed
by the base catalyzed E2 elimination via the 11b chloro
intermediate since the latter compound was indeed
detected in the reaction. In order to minimize the level
of the 11b-Cl, the reaction was carried out in pyridine
so that all of the 11b-Cl formed would be converted to
the D9,11 (VI). However, a higher level of 11b-Cl was
H
CH3
CH2OCO2Et
O
CH2OCO2Et
O
OH
HO
PCl4O
OH
11
O
C
PCl5/THF
12
9
D
-85°C
(95%)
PCl3
H
H
H
O
O
Cl
III
IV
CH2OCO2Et
CH2OCO2Et
CH2OCO2Et
O
OH
O
OH
O
OH
Cl
Cl-
-POCl3
+
O
O
O
VI (∆9,11
(98.4%)
)
VII (∆11,12
)
(1.3%)
V (11β-Cl)
(0.3%)
Scheme 1.

X. Fu et al. / Tetrahedron Letters 42 (2001) 2639–2642
2641
CH2OCO2Et
CH2OCO2Et
CH2OCO2Et
O
OH
O
OH
O
HO
OH
PCl5/THF
OPCl4
CH3
H
+
H
11
C
12
D
O
O
9
H
O
H
Cl-
VI (∆9,11
)
VII (∆11,12
)
9,11
∆
r.t.
91.6
96.7
97.5
8.6
3.3
2.5
11α-OH
-45ºC
-60ºC
(III)
Cl3P
Cl
O
H
CH3
r.t.
98.2
93.7
92.0
1.8
6.7
8.0
11
H
H
12
C
D
11β-OH
(VIII)
-45ºC
-60ºC
H
11,12
∆
*The above data are HPLC area% ratios for VI and VII only. The 11-β-Cl from III is left out for comparison.
Scheme 2.
good evidence to exclude the possibility of the forma-
tion of a free carbonium ion at C-11. The attack of a
chloride ion on the carbonium C-11 would be mainly
from the sterically less hindered a-face to furnish the
corresponding 11a-Cl. However, there is no evidence
for the formation of 11a-Cl from the PCl5 dehydration
reaction.
with PCl5 in THF failed to generate an olefinic signal
by NMR study.
In summary, a regioselective PCl5-induced dehydration
reaction has been discovered and successfully imple-
mented in the manufacturing processes of beta-
methasone and mometasone. The solvent and tempera-
ture effects on the selectivity have been studied. The
new processes produced the corresponding products in
higher yields and superior quality, resulting in increased
manufacturing efficiency.
Furthermore, if the carbonium ion were indeed the
reactive intermediate, the product distribution from the
11b hydroxy analogue would be similar to that from
the 11a hydroxy under the same reaction conditions.
However, treatment of the 11b-OH compound with
PCl5 in THF at −60°C furnished D9,11 (VI) with a ratio
of 92.0:8.0 to its D11,12 isomer. While reaction at room
temperature, as expected from its anti-coplanar orienta-
tion of H-9 with the leaving group, generated D9,11 as
the dominant product. These results clearly suggest that
in the case of the 11b-OH there are two competing
elimination pathways. At higher temperature, the E2
elimination mechanism dominates the reaction. The
main pathway rate is slower at lower temperature, and
the competing syn-elimination on the other hand gener-
ates more D11,12 isomer, as depicted in Scheme 2. These
experimental results are also contrary to the hypothesis
of the formation of a free carbonium ion at C-11
(Scheme 2).
Acknowledgements
The authors wish to thank Dr. Derek Walker and Mr.
Luis Gil for helpful discussions. The authors are also
indebted to Dr. Birendra Pramanik for providing mass
spectra and Dr. Tze-Ming Chan for NMR analysis.
References
1. (a) Herzog, H.; Oliveto, E. P. Steroids 1992, 57, 617; (b)
Hogg, J. A. Steroids 1992, 57, 593; (c) Hirschmann, R.
Steroids 1992, 57, 579.
2. (a) Rausser, R.; Lyncheski, A. M.; Harris, H.; Grocela, R.;
Murrill, N.; Bellamy, E.; Ferchinger, D.; Gebert, W.;
Herzog, H. L.; Hershberg, E. B.; Oliveto, E. P. J. Org.
Chem. 1966, 31, 26; (b) Oliveto, E. P.; Rausser, R.; Her-
zog, H. L. J. Am. Chem. Soc. 1958, 80, 6687.
Extension of the new chemistry to the 16a methyl series
would provide an efficient synthesis for the mometa-
sone triene intermediate. The reaction of the 16a ana-
logue of 11a-hydroxy (III) with PCl5 was carried out in
THF at −85°C. The corresponding dehydrated product
D9,11 was isolated in greater than 90% yield with a 99:1
ratio to the D11,12 isomer. The scope of this reaction
seems to be limited to the C-11-hydroxy, the most
sterically hindered site in the steroid nucleus. Treatment
of the corresponding 12b-hydroxy with PCl5 did not
produce any elimination product. Reaction of menthol
3. (a) Sulfur dioxide/acid chlorides/py to convert 11b-OH to
D
9,11, see: Hazen, G. G.; Rosenburg, D. W. J. Org. Chem.
1964, 29, 1930; (b) Methanesulfonyl chloride/base to trans-
form 11b-OH to D9,11, see: Fried, J.; Florey, K.; Sabo, E.
F.; Herz, J. E.; Restivo, A. R.; Borman, A.; Singer, F. M.
J. Am. Chem. Soc. 1955, 77, 4181. Arth, G. E.; Fried, J.;
Johnston, D. B. R.; Hoff, D. R.; Sarett, L. H.; Silber, R.

2642
X. Fu et al. / Tetrahedron Letters 42 (2001) 2639–2642
H.; Stoerk, H. C.; Winter, C. A. J. Am. Chem. Soc. 1958,
Marshall, C. W.; Pincus, G.; Schenker, V. J. Am.
Chem. Soc. 1949, 71, 3261; (b) Collingsworth, D. R.;
Brunner, M. P.; Haines, W. J. J. Am. Chem. Soc. 1952, 74,
2381; (c) Hanson, F. R.; Mann, K. M.; Nielson, E. D.;
Anderson, H. V.; Brunner, M. P.; Karnematt, J. N.;
Collingsworth, D. R.; Haines, W. J. J. Am. Chem. Soc.
1953, 75, 5369.
80, 3161. Chamberlin, E. M.; Tristram, E. W.; Utne, J.;
Chemerda, J. M. J. Org. Chem. 1960, 25, 295; (c) Burgess
reaction to eliminate 11b-OH to D9,11, see: Crabbe, P.;
Leon, C. J. Org. Chem. 1970, 35, 2594; (d) From 9a-OH to
D
9,11, see: Beaton, J. M.; Huber, J. E.; Padilla, A. G.;
Breuer, M. E. US Pat. 1978, No. 4,127,596. Bergstrom, C.
G.; Dodson, R. B. Chem. Ind. 1961, 1530; (e) From 3a or
17a-OH to D9,11, see: Breslow, R.; Corcoran, R. J.; Snider,
B. B.; Doll, R. J.; Khanna, P. L.; Kaleya, R. J. Am. Chem.
Soc. 1977, 99, 905. Hyatt, J. A.; McCombs, C. A. US Pat.
1982, No. 4,328,162.
5. O’Connor, G. L.; Nace, H. R. J. Am. Chem. Soc. 1953, 75,
2118.
6. Bernstein, S.; Lenhard, R. H.; Williams, J. H. J. Org.
Chem. 1954, 19, 41.
7. Shoppee, C. W.; Nemorin, J. J. Chem. Soc., Perkin Trans.
4. (a) Hechter, O.; Jacobsen, R. P.; Jeanloz, R.; Levy, H.;
1973, 542.
.
.