The effect of water and phenol on the chiral oxazaborolidine-catalyzed reduction of a prostaglandin enone derivative

Article abstract of DOI:10.1021/op900118n

The alcohol 2, a key intermediate in the synthesis of the highly selective EP4-receptor agonist ONO-4819, was synthesized by (R)- methyloxazaborolidine ((R)-Me-CBS)-catalyzed asymmetric reduction of enone 1 with borane dimethylsulfide complex. Addition of

Full text of DOI:10.1021/op900118n

Organic Process Research & Development 2009, 13, 933–935
The Effect of Water and Phenol on the Chiral Oxazaborolidine-Catalyzed Reduction
of a Prostaglandin Enone Derivative
Chisa Ohta,* Shin-itsu Kuwabe, Tai Shiraishi, Shuichi Ohuchida,† and Takuya Seko†
Fukui Research Institute, Ono Pharmaceutical Co., Ltd., 1-5-2 Technoport, Yamagishi, Mikuni, Sakai, Fukui 913-8538, Japan
Scheme 1
Abstract:
The alcohol 2, a key intermediate in the synthesis of the highly
selective EP4-receptor agonist ONO-4819, was synthesized by (R)-
methyloxazaborolidine ((R)-Me-CBS)-catalyzed asymmetric reduc-
tion of enone 1 with borane dimethylsulfide complex. Addition of
water and phenol to the reduction of enone 1 using (R)-Me-CBS
as catalyst changed the chemoselectivity of the reduction.
Introduction
Prostaglandins have been shown to have a broad range of
biological activities in diverse tissues through binding to specific
receptors.1 In particular, those of the E series are multifunctional
regulators of bone metabolism.2 Recently, it was reported that
the highly selective agonist for prostaglandin E receptor subtype
EP4 ONO-4819 in combination with risedronate could be an
effective treatment for osteoporosis.3
Table 1. Screening of the stereoselective reduction of 1a
One of the key steps in the synthesis of ONO-4819 is
stereoselective reduction of enone 1 to chiral alcohol 2 (Scheme
1). We report here that addition of water and phenol changed
chemoselectivity in the synthesis of key (S)-hydroxy intermedi-
ate 2 catalyzed by (R)-methyloxazaborolidine ((R)-Me-CBS).
Results and Discussion
The asymmetric reduction of carbonyl compounds with
binaphthol-modified lithium aluminum hydride (BINOL-H) and
chiral methyloxazaborolidine catalysts (CBS) has been dem-
onstrated as a powerful tool for the synthesis of enantiomerically
pure alcohols from prochiral ketones.4,5 Initially, we demon-
strated the reduction of 1 with stoichiometric amounts of
BINOL-H reagent and obtained 2 with good diastereoselectivity
(80% de), but these reduction conditions required improvement
for scale-up (stoichiomeric amount of chiral reagent, tedious
removal procedure for Al, and low concentration conditions
were needed). Two unique structural features of enone 1 are
entry
PhOH equiv
solvent
2 yield %b
2 de %
1
none
1.0
1.0
none
1.0
toluene
toluene
toluene
THF
65
95
85
80
85
75
80
62
75
45
2
3c
4
5
THF
a All reactions were conducted using the general procedure unless otherwise
noted. To the mixture of BH3-Me2S (1 equiv), phenol and (R)-Me-CBS catalyst
(0.1 equiv) at 0 °C was added a solution of 1 over 10 min. b The combined yield
of (S)- and (R)-alcohols was determined by HPLC using an absolute calibration
method. c BH3-Me2S was added to the mixture of 1, catalyst, and phenol in
toluene.
* Corresponding author. Telephone: +81-776-82-6161. Fax: +81-776-82-
8420. E-mail: ch.ohta@ono.co.jp.
† Current address: Minase Research Institute, Ono Pharmaceutical Co., Ltd.,
3-1-1 Shimamoto, Minase, Osaka 618-8585, Japan.
worth noting: (a) the lactone moiety in the system of two
5-membered rings; (b) enone 1 has four chiral centers, one of
which may affect the stereoselectivity for the reduction of the
carbonyl moiety. We next screened the conditions of stereo-
selective reduction using Me-CBS catalyst with borane dim-
ethysulfide complex (BMS), and the results are summarized in
Table 1. Reduction of 1 with BMS in the presence of (R)-Me-
CBS as a catalyst provided (S)-chiral alcohol 2 with moderate
enantioselectivity (75% de, entry 1). As an earlier report had
already mentioned that the enantioselectivity was enhanced by
(1) Coleman, R. A.; Smith, W. L.; Narumiya, S. Pharmacol. ReV. 1994,
46, 205.
(2) Raisz, L. G. Osteoarthritis Cartilage 1999, 7, 419.
(3) (a) Maruyama, T.; Kuwabe, S.; Kawanaka, Y.; Shiraishi, T.; Shinagawa,
Y.; Sakata, K.; Seki, A.; Kishida, Y.; Yoshida, H.; Maruyama, T.;
Ohuchida, S.; Nakai, H.; Hashimoto, S. Bioorg. Med. Chem. 2002, 10,
2103. (b) Ito, M.; Nakayama, K.; Konaka, A.; Sakata, K.; Ikeda, K.;
Maruyama, T. Bone 2006, 39, 453.
(4) (a) Corey, E. J.; Helal, C. J. Angew. Chem., Int. Ed. 1998, 37, 1986.
(b) Corey, E. J.; Bakshi, R. K.; Shibata, S. J. Am. Chem. Soc. 1987,
109, 5551.
(5) Noyori, R.; Tomino, I.; Ymamada, M.; Nishizawa, M. J. Am. Chem.
Soc. 1984, 106, 6717.
10.1021/op900118n CCC: $40.75  2009 American Chemical Society
Published on Web 07/13/2009
Vol. 13, No. 5, 2009 / Organic Process Research & Development
•
933

Table 3. Effect of water and phenol in the reduction of
enone 1a
HPLC area ratio
PhOH H2O time
entry equiv equiv
h
1
2
3
4
5
6
2 de %
1
2
3
4
1.0
1.0
1.1
1.0
none
0.1
none
0.05
3
5
4
3
3.2 87.7 2.1 1.3 4.4 1.3
3.0 82.0 1.0 0.3 9.4 4.3
NDb 71.2 0.2 0.5 20.9 7.2
ND 98.3 0.7 0.1 0.7 0.2
78
77
78
80
Figure 1
Table 2. Effect of water for chemoselectivity on reduction of
a To the mixture of BH3-Me2S (1.0 equiv), phenol, catalyst (0.1 equiv) and H2O
enone 1a
in toluene was added 1 solution at 0 °C over 10 min. b ND: not detected.
HPLC area ratio
BMS
H2O
timeb
h
entry equiv equiv
1
2
3
4
executed in the presence of 0.2 equiv of water in a solution
containing a large excess of BMS, catalyst and PhOH improved
the chemoselectivity to be controlled to 5.1% of triol 3 and
lactol 4 (entry 2). Since we obtained satisfied results in the
condition that large amount of reductant was present in the
mixture, we applied this promising condition to the normal one.
When addition period of 1 was extended in the presence of
water, chemoselectivity was found to be improved (entry 3).
Addition of 0.1 equiv of water led to formation of almost no
overreductive byproducts and afforded (S)-chiral alcohol 2 with
good stereoselectivity (77% de) even when 1 was added over
2.5 h.
1c
2c
5.0
5.0
1.0
none
0.2
0
3
0
3
5
2.8
0
95.9 0.07
2.6
0.6
71.1 26.2
74.9 23.2
1.1
3.7
1.2
0.8
1.4
0.3
2.2
3.5
92.7
95.1
,
3d e
0.1
a All reactions were conducted using the general procedure unless otherwise
noted. To the mixture of BH3-Me2S, phenol, catalyst, and H2O in toluene was
added enone 1 at 0 °C over 10 min. b Reaction time after completion of adding
enone 1. c 5 equiv of phenol and 0.5 equiv of catalyst were used. d 1 equiv of
phenol and 0.1 equiv of catalyst were used. e 1 was added over 2.5 h.
using alcohol as additives for this reduction,6 addition of phenol
slightly increased the stereoselectivity in toluene (entry 2). These
two results described above were obtained by slow addition of
enone 1 into a premixed solution of BMS, phenol and catalyst.
It is good to mention that the stereoselectivity was decreased
to 62% de when BMS was added to the mixture of enone 1
and catalyst (entry 3). We also need to note that addition of
phenol sometimes caused negative impact on the stereoselec-
tivity, for example, the reduction in THF media which you can
find the results in entries 4 and 5. From these initial experiments,
we decided to select the entry 2 condition in Table 1 for further
optimization work for a scale-up in the reduction of 1.
In a small-scale experiment, an extended addition time of 1
caused no problems for the reaction based on the selected
conditions. However, a kilogram lab-scale synthesis (1, 769 g)
led to overreduction of the lactone moiety on 1 to give a
substantial amount (∼30%) of triol 3 and lactol 4 (Figure 1)
without completion of the desired ketone reduction. To over-
come this chemoselectivity issue, we investigated the effect of
the addition period of 1 to the reaction mixture, temperature,
and purity of catalyst, but no difference was observed in the
reduction of lactone moiety of enone on a smalll scale (∼3%
of triol 3 and lactol 4).
We also found interesting chemoselectivity between 1,2- and
1,4-reduction of the enone moiety of 1 (Table 3). When 1.0
equiv of phenol was used in the absence of water in toluene,
5.7% of 5 and 6 was obtained, which was formed Via 1,4- and
1,4- plus 1,2-reduction of the enone moiety of 1 (entry 1).
Additional water promoted the 1,4-reduction of the enone
moiety of 1 (entry 2). Using 1.1 equiv of phenol in the absence
of water also increased the byproducts 5 and 6 (entry 3).
However, addition of 0.05 equiv of water to the reaction mixture
in the presence of 1.0 equiv of phenol decreased the formation
of these byproducts 5 and 6, and desired product 2 was obtained
with good stereoselectivity. Also, the amount of triol 3 and lactol
4 was well controlled to less than 1% (entry 4). It is thought
that adding alcohol to an oxazaborolidine-catalyzed reduction
accelerates the recycling of the active catalyst species which
leads to the improvement in stereoselectivity.6a In the reduction
of enone 1, addition of water and phenol led not only to the
prevention of the reduction of the lactone moiety but also to
the promotion of 1,4-reduction of the enone moiety. From these
observations, although the reaction mechanism is unclear, the
active reductant species is likely to be a soft hydride source
compared to conditions involving no additives. Because slight
difference of water equivalent affected the regioselectivity in
the reduction of enone 2, we kept the attention to the water
content of the starting material 1 which was controlled no less
than 0.05 wt % and the solvent was used as dehydrated grade
in small-scale experiments. From the point of the purity of (R)-
Me-CBS catalyst, the catalyst purchased from BASF showed
good reproducibility in stereoselectivity and regioselectivity. For
further scale-up manufacturing, we believe that water content
of the reaction mixture should be controlled to the appropriate
amount for a desired reaction since it is a very important factor
for chemoselectivity of the reduction, and the evaluation in the
case that addition period of reagents is extended should be
conducted for the stereoselective reduction of high functional-
ized compound using Me-CBS catalyst with BMS.
It is thought that the reductant was in excess to 1 at an early
stage of the addition period of enone 1, so we decided to
examine the effect of addition of water in the condition that a
larger excess of BMS and catalyst were used (Table 2). Almost
all the lactone moiety was reduced to triol 3 and lactol 4 in the
presence of large amount of reducing reagent (entry 1) in
anhydrous condition. Ordinarily, it is important to keep the
reaction mixture under anhydrous conditions due to the sensitiv-
ity of the (R)-MeCBS catalyst, however, the reaction which was
(6) (a) Tschaen, D. M.; Abramson, L.; Cai, D.; Desmond, R.; Dolling, U.-
H.; Frey, L.; Karady, S.; Shi, Y.; Verhoeven, T. R. J. Org. Chem. 1995,
60, 4324. (b) Shi, Y.; Cai, D.; Dolling, U.-H.; Douglas, A. W.; Tschaen,
D. M.; Verhoeven, T. R. Tetrahedron Lett. 1994, 35, 6409. (c) Kagawa,
T.; Kawanami, Y. Asymmetric reduction catalyst, its solution/prepara-
tion procedure and manufacturing method of optically active alcohol
using this catalyst. JP2008073591 (A), 2008.
934
•
Vol. 13, No. 5, 2009 / Organic Process Research & Development

toluene (30 mL) was added BH3-Me2S (3.28 mL, 34.5 mmol)
at room temperature. The resulted mixture was stirred for 30
min. Then (R)-Me-CBS catalyst (1.09 mol/L in toluene, 3.45
mL, 3.45 mmol) was added at 0 °C, and the mixture was stirred
for another 30 min. To the mixture was added enone 1 (15 g,
34.5 mmol) in toluene (75 mL), and stirring was continued at
0 °C. After stirring for 1.5 h, to the solution was added MeOH
(7.0 mL, 172.5 mmol) and aq 15% NH4Cl solution (150 mL),
and the resulted mixture was stirred for 30 min at room
temperature. The separated aqueous layer was extracted with
toluene (30 mL), and the combined organic layers were washed
with aq 20% NaCl solution (30 mL) and dried over anhydrous
MgSO4. The solvent was removed by evaporation, and the
residue was purified by column chromatography on silica gel
(Fuji silysia BW-235S, 557 g, toluene/EtOAc ) 3:2) to give 2
Figure 2
Conclusion
In conclusion, we have performed the stereoselective reduc-
tion of prostaglandin enone derivative using BMS in the
presence of (R)-Me-CBS as a catalyst. In scale-up manufactur-
ing of this reaction, a small amount of water was crucial to
prevent the over-reduction of the lactone moiety. We also found
interesting changes in chemoselectivity to be brought about by
addition of water and phenol. The mechanism of changing
chemoselectivity for the reduction in the presence of water
remains unclear, and further studies are needed to understand
the reaction pathway in detail.
as a colorless oil (11.8 g, 79%, 90.2% de, 99.3 area %). Rf
1
0.34 (hexane/EtOAc, 1:2); [R]20 -35.2 (c 1.04, EtOH); H
D
NMR (200 MHz, CDCl3) δ 8.00 (dd, 2H, J ) 8.42, 1.28 Hz),
7.59 (m, 1H), 7.57 (m, 1H), 7.45 (m, 2H), 5.55 (ddd, 1H, J )
15.47, 7.23, 1.10 Hz), 5.20 (m, 1H), 5.03 (td, 1H, J ) 6.50,
1.83 Hz), 4.42 (s, 2H), 4.35 (m, 1H), 3.40 (s, 3H), 2.78 (m,
5H), 2.51 (m, 2H), 2.21 (ddd, 1H, J ) 15.52, 4.90, 1.92 Hz);
13C NMR (50 MHz, CDCl3) δ 176.5, 166.1, 138.6, 137.7, 135.0,
133.4, 129.7, 129.6, 129.0, 128.92 (2C), 128.88 (2C), 128.6
(2C), 83.4, 79.3, 74.7, 72.9, 58.4, 54.0, 44.0, 42.7, 37.7, 35.0;
IR (liquid film) 3449, 2927, 1771, 1715, 1275, 1110, 714 cm-1;
Mass (ESI, Pos., 20 V) m/z 459 (M + Na); Anal. Calcd for
C26H28O6: C, 71.54; H, 6.34; Found: C, 71.71; H, 6.34.
HPLC conditions for the determination of diastereoselec-
tivity: CHIRALCEL OD-RH; MeCN/H2O ) 30:70 (0-90
min); detection, 210 nm; flow rate, 1.0 mL/min; retention
time of 2 was 63.4 min, and that of ꢀ-OH was 53.9 min.
HPLC conditions for analysis of reaction: YMC-Pack-
ODS-A-302; MeCN/H2O ) 45:55; detection, 210 nm;
column temperature, 40 °C; flow rate, 1.0 mL/min;
retention time of 2 was 9.2 min, and those of 1, 3, 4, 5,
6 were 13.6, 4.2, 6.4, 14.0, 10.2 min, respectively.
Experimental Section
General. Infrared (IR) spectra were recorded on a Jasco FT-
IR-430 spectrometer. 1H NMR spectra were measured with a
Varian Gemini-200 (200 MHz) spectrometer. The chemical
shifts (δ) on 1H NMR spectra were reported in parts per million
relative to tetramethylsilane. Splitting patterns are designated
as s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet; br,
broad. The optical rotations were measured with a JASCO DIP-
1000 polarimeter. High-performance liquid chromatography
(HPLC) was carried out using a Hitachi LC-Organizer,
L-4000UV Detector, L-6200 Intelligent Pump, and D-2500
Chromato-Integrator or Shimazu LC-2010. Analytical TLC was
performed on Merck preparative TLC plates (silica gel 60 F254,
0.25 mm). Column chromatography was carried out with silica
gel [Fuji Silysia BW-235S]. All reactions were carried out under
argon atmosphere unless otherwise noted.
Dehydrated solvents (THF and toluene) were purchased from
Kanto Kagaku and were used as received. (R)-Methyl-CBS-
oxazaborolidine toluene solution (1.0 mol/L) was purchased
from BASF.
Acknowledgment
We thank Ms. M. Sugioka for performing the microanalyses.
Preparation of (3aR,4R,5R,6aS)-4-{(1E,3S)-3-hydroxy-4-
[3-(methoxymethyl)phenyl]-1-buten-1-yl}-2-oxohexahydro-
2H-cyclopenta[b]furan-5-yl Benzoate 2. To a mixture of
phenol (3.25 g, 34.5 mmol) and water (31 mg, 1.73 mmol) in
Received for review May 7, 2009.
OP900118N
Vol. 13, No. 5, 2009 / Organic Process Research & Development
•
935