Asymmetric synthesis of CDP840 by Jacobsen epoxidation. An unusual syn selective reduction of an epoxide

Article abstract of DOI:10.1021/jo971476c

An asymmetric synthesis of the PDE IV inhibitor, CDP840 (3) is reported. The absolute stereochemistry was controlled by a Jacobsen epoxidation of the Z triaryl olefin 8 (89% ee) or the E triaryl olefin 9 (48% ee). The disparate results in stereocontrol were interpreted in terms of the 'skewed side-on approach model' proposed by Jacobsen. LiBH₄ · BH₃ reduction of the epoxides was found to proceed with retention of configuration.

Full text of DOI:10.1021/jo971476c

J . Org. Chem. 1997, 62, 9223-9228
9223
Asym m etr ic Syn th esis of CDP 840 by J a cobsen Ep oxid a tion . An
Un u su a l Syn Selective Red u ction of a n Ep oxid e
J . E. Lynch,* W.-B. Choi, H. R. O. Churchill, R. P. Volante, R. A. Reamer, and R. G. Ball†
Department of Process Research and Department of Molecular Design and Diversity,
Merck Research Laboratories, PO Box 2000, Rahway, New J ersey 07065
Received August 8, 1997X
An asymmetric synthesis of the PDE IV inhibitor, CDP840 (3) is reported. The absolute
stereochemistry was controlled by a J acobsen epoxidation of the Z triaryl olefin 8 (89% ee) or the
E triaryl olefin 9 (48% ee). The disparate results in stereocontrol were interpreted in terms of the
“skewed side-on approach model” proposed by J acobsen. LiBH4‚BH3 reduction of the epoxides was
found to proceed with retention of configuration.
Sch em e 1
The phosphodiesterases (PDE) are a family of enzymes
in which at least five different isozymes have been
characterized. These isozymes are distinct in their
cellular distribution and also in their affinity for cyclic
AMP and cyclic GMP. The discovery of isozyme-selective
inhibitors has facilitated the elucidation of the separate
roles each isotype plays in cellular function. PDE IV in
particular is believed to be the dominant isozyme present
in inflammatory cells and in airway smooth muscle.
Inhibition of PDE IV leads to an increase in the concen-
tration of cAMP resulting in suppression of a broad range
of cellular function in inflammatory cells.1 This has
generated a great deal of interest in selective PDE IV
inhibitors as potential antiasthmatic agents.
We wish to report an asymmetric synthesis of CDP840
(3), a selective inhibitor of PDE IV.2 3 has previously
been obtained in enantiomerically pure form by the 1,4
addition of phenylmagnesium bromide to the chiral
diarylacrylate 1 (Scheme 1) or by additions to other chiral
Michael acceptors.3
We were interested in probing other synthetic routes
that did not require the use of a stoichiometric, covalently
bound chiral auxiliary. The J acobsen epoxidation of
olefin 8 or 9 seemed an attractive possibility.4 Although
this introduces two carbon-oxygen bonds where we
require two carbon-hydrogen bonds, the well prece-
dented stereocontrol of the oxidation and the general ease
of benzylic C-O bond hydrogenolysis attenuated our
concern about the circuitousness of this strategy. Mix-
tures of E and Z olefins were prepared in a straightfor-
ward manner as shown in Scheme 2.
Grignard addition to aldehyde 4 followed by oxidation5
and picoline addition gave carbinol 7. Dehydration of 7
gave variable ratios of Z and E olefin isomers, depending
on solvent. When the elimination was performed in
toluene a 4:1 ratio (9:8 ) was obtained; when the elimina-
tion was performed in THF a 2.5:1 ratio was obtained.
The olefins, 8 and 9, were efficiently separated by
fractional crystallization. Crystallization of the mixture
of methanesulfonic acid salts gave pure E olefin 9 (MSA
salt), 70% from the toluene elimination; treatment of the
mother liquors with aqueous NaOH followed by crystal-
lization of the resulting free base from ethyl acetate gave
the pure Z olefin (26% from the THF elimination). The
† Department of Molecular Design and Diversity.
X Abstract published in Advance ACS Abstracts, December 1, 1997.
(1) Banner, K. H.; Page, C. P. Eur. Respir. J . 1995, 8, 996-1000.
Raeburn, D. Methylxanthines Phosphodiesterase Inhib. Treat. Airways
Dis., Proc. Meet.; Costello, J . F.; Piper, P. J ., Eds.; Parthenon Publ.
Group: Carnforth, UK. 1994; pp 101-17. Christensen, Siegfried B.;
Torphy, Theodore J . Annu. Rep. Med. Chem. 1994, 29, 185-94. Silver,
P. J .; Gordon, R. J .; Buchholz, R. A.; Dundore, R. L.; Ferguson, E. W.;
Harris, A. L.; Pagani, E. D. Adv. Second Messenger Phosphoprotein
Res. (Biol. Cyclic Nucleotide Phosphodiesterases) 1992, 25, 341-51.
Torphy, T. J .; Livi, G. P.; Balcarek, J . M.; White, J . R.; Chilton, F. H.;
Undem, B. J . Adv. Second Messenger Phosphoprotein Res. (Biol. Cyclic
Nucleotide Phosphodiesterases) 1992, 25, 289-305. Giembycz, Mark
A. Biochem. Pharmacol. 1992, 43, 2041-51.
1
configuration of each olefin was determined by H NOE
difference experiments. The olefin isomers were readily
interconvertible under strongly acidic conditions (12 N
HCl) and in principle a process to convert the mixture to
(2) Celltech Therapeutics Ltd, US Patent No. 5,608,070.
(3) (a) Celltech Therapeutics Ltd, US Patent No. 5,622,977. (b) Use
of the acrylamide with cis-2-aminoindanolacetonide as the Michael
acceptor, Choi, W.-B; Churchill, H.; Lynch, J . E. Unpublished results.
Use of a vinyl phenyl sulfoxide as the chiral Michael acceptor, Houpis,
I.; Molina, A.; Dorsiotis, I.; Reamer, R. A.; Volante, R. P.; Reider, P. J .
Tetrahedron Lett. 1997, 38, 7131.
(5) Murahashi, S.-I.; Naota, T. Adv. in Metal-Org. Chem. 1994, 3,
225 and refs therein. Tanaka, M.; Kobayashi, T.; Sakakura, T. Angew.
Chem., Int. Ed. Engl. 1984, 23, 518. Agarwal, D. D.; J ain, R.; Sangha,
P.; Rastogi, R. Indian J . Chem. 1993, 32B, 381. Miller, R. A.; Li, W.;
Humphrey, G. R. Tetrahedron Lett. 1996, 37, 3429.
(4) Brandes, B. D.; J acobsen, E. N. J . Org. Chem. 1994, 59, 4378.
S0022-3263(97)01476-X CCC: $14.00 © 1997 American Chemical Society

9224 J . Org. Chem., Vol. 62, No. 26, 1997
Lynch et al.
Sch em e 2
Sch em e 3
Sch em e 4
a single isomer could be developed; however, such a
process was not reduced to practice.6
catalyst in a skewed manner such that interactions with
the disubstituted end of the double bond are minimized.
In the case of the E olefin the additional steric hindrance
of the large cyclopentyloxy substituent in the “favored”
approach raises the activation energy of this transition
state and results in both lower enantioselectivity and also
the observed slower rate of reaction.
Attempted hydrogenolyses of the epoxides with Ni(Ra)8
were nonselective giving complex mixtures of products.
However, a 1:1 mixture of LiBH4 and BH3-THF cleanly
reduced epoxide 11 at the more substituted carbon
providing the alcohol 13 as the pyridine borane complex.9
It was found that crystallization of 13 from hexane: ethyl
acetate (8:3) improved the enantiomeric purity to >99%
ee (46% yield from 11). Treatment of 12 under identical
conditions produced the corresponding pyridine complex
of alcohol 14 as the major product, however, in much
lower yield (Scheme 5). HPLC analysis of the crude
reduction mixtures showed that 11 was reduced with 66:1
selectivity (13:ent-14) and 12 was reduced with 8:1
selectivity (14:ent-13). Heating 13 or 14 in 4:1 toluene:
J acobsen epoxidation reactions of both 8 and 9 (free
bases) were stereospecific i.e., the Z olefin gave only the
syn epoxide and the E olefin gave only the anti epoxide
(the relative configurations of each epoxide were likewise
determined by 1H NOE difference measurements). Treat-
ment of 8 and 9 with NaOCl solution in the presence of
S,S salen Mn(III) complex 10 (1 mol %) and 4-(3-
phenylpropyl)pyridine N-oxide7 at -10 °C to 0 °C pro-
vided the epoxides 11 and 12, respectively. Although
somewhat sensitive to silica gel, the purified epoxides
were isolated by flash chromatography. The enantiomers
of 11 and 12 were each prepared by the same method
using the R,R salen complex. Chiral HPLC showed a
94.5:5.5 ratio of enantiomers in the case of 11 (89% ee)
and 48% ee in the case of 12. In addition to showing a
lesser degree of stereoselectivity, epoxidation of 9 also
proceeded more slowly compared to epoxidation of 8; after
3 h reaction time HPLC showed 80% conversion of 8
versus only 10% conversion of 9 under identical condi-
tions.
The disparate results in asymmetric induction can be
understood in terms of the “skewed side-on approach”
model proposed by J acobsen (Scheme 4).4 In this model
that face of the olefin which can avoid severe steric
interactions with the catalyst ligand approaches the
(8) King, S. B.; Sharpless, K. B. Tetrahedron Lett. 1994, 35, 5611.
Cope, A. C.; McKervey, M. A.; Weinshenker, N. M.; Kinnel, R. B. J .
Org. Chem. 1970, 35, 2918. Roberts, R. M.; Bantel, K. H.; Low, C. E.
J . Org. Chem. 1973, 38, 1903.
(9) Brown, H. C.; Yoon, N. M. J . Am. Chem. Soc. 1968, 90, 2686.
Other methods for reduction of epoxides at the more substituted carbon
have also been reported. Hutchins, R. O.; Taffer, I. M.; Burgoyne, W.,
J . Org. Chem. 1981, 46, 5214. Ranu, B. C.; Das, A. R. J . Chem. Soc.,
Chem Commun. 1990, 1334. Guanti, G., Banfi, L.; Merlo, V.; Narisano,
E. Tetrahedron 1994, 50, 2219. Taber, D. F.; Houze, J . B. J . Org. Chem.
1994, 59, 4004. Taber, D. F.; Louey, J . P. Tetrahedron 1995, 51, 4495
and refs therein.
(6) Dissolution of either pure olefin in 12 N HCl leads to a rapid
equilibration to ∼60:40 mixture of the two olefins (∼30 min, 22 °C).
(7) Senanayake, C.; Smith, G. B.; Ryan, K. M.; Fredenburg, L.; Liu,
J .; Roberts, F. E.; Hughes, D. L., Larsen, R. D.; Verhoeven, T. R.;
Reider, P. J . Tetrahedron Lett. 1996, 37, 3271.

Asymmetric Synthesis of CDP840
J . Org. Chem., Vol. 62, No. 26, 1997 9225
Sch em e 5
Sch em e 6
reduction would necessarily give the same diastereomeric
ratio of products.
Some solvolysis reactions of phenyl-substituted ep-
oxides, and particularly of 4-methoxyphenyl-substituted
epoxides are known to occur predominantly with reten-
tion of configuration (syn substitution).11,12 To our knowl-
edge this is the first documented case of a hydride
reduction of an epoxide with retention of configuration.13
We thus chose to investigate the reduction of other
phenyl substituted epoxides under identical conditions
as used for 11 and 12 (Scheme 5). In contrast to 11 and
12, reduction of phenylcyclohexene oxide (15a ) by this
method did not occur at -20 °C, even after 5 h. After an
overnight age at ambient temperature HPLC analysis
showed 74% of cis-2-phenylcyclohexanol (the result of
inversion). Inspection of the crude 1H NMR spectrum
revealed about a 10:1 ratio of cis:trans isomers (Scheme
6). Reduction of 4-methoxyphenylcyclohexene oxide
(15b),12 however, was rapid at -20 °C. HPLC showed
two major products in an 8:1 ratio. After the usual
MeOH or in ethanol at reflux for 2 h followed by vacuum
concentration provided the free hydroxypyridines. Chiral
HPLC showed no loss in enantiomeric purity when
compared to the starting epoxides.
1
aqueous workup, H NMR confirmed the major isomer
was in fact the trans-cyclohexanol,14 the product of
retention of configuration.
The final deoxygenation was accomplished by one-pot
mesylation (Et3N, MsCl) followed by Zn/acetic acid reduc-
tion, providing CDP840 in 85% yield. We were surprised
to find that the enantiomer of 13 which was prepared
using the R,R salen complex ultimately gave the undes-
ired enantiomer of CDP840, and that the isomer gener-
ated with the S,S catalyst had the necessary configura-
tion to give CDP840. We had incorrectly predicted the
opposite result based on the J acobsen model for the
epoxidation and that the borane/borohydride reduction
would proceed with inversion of stereochemistry at the
reacting center. At the time it was not clear which of
these assumptions was incorrect. However, recrystalli-
zation of 13 from ethanol provided a crystal suitable for
X-ray analysis17 which allowed assignment of the relative
stereochemistry of the two chiral centers. This result
coupled with NOE determination of the epoxide structure
proved that the reduction had proceeded with retention
of configuration! Brown has reported that reduction of
1-phenylcyclopentene oxide with BH3-THF and BF3 gave
predominantly trans-2-phenylcyclopentanol (an apparent
syn reduction).10 However, the use of the strong acid,
BF3, is likely to lead to rearrangement of the epoxide to
2-phenylcyclopentanone and Brown reports that, indeed,
reduction of this ketone gave the same ratio of cis- and
trans-2-phenylcyclopentanols (1:4.6) as obtained by re-
duction of the epoxide.10 The reductions of 11 and 12
reported here cannot proceed via a ketone intermediate
since both epoxides would give the same ketone and on
Based on previous reports on the “syn” substitution of
epoxides and on the dramatic difference in the rate of
reduction of 15a vs 11, 12, and 15b, it is likely that a
carbonium ion intermediate is involved in the reductions
with retention of configuration. Apparently, the ad-
ditional stability provided by the p-methoxy group is
required to make the carbonium ion accessible under
these conditions. One can invoke alkoxide delivery of the
reducing agent as an explanation for the observed syn
selectivity. However, it is not clear what species delivers
the hydride; borohydride and borane are known to react
reversibly to form a diborohydride (B2H7-) in diglyme.15
It is possible this species may be involved in the reaction,
although free borane is more Lewis acidic and would
more easily lead to carbonium ion formation. Control
experiments showed the reduction requires both BH3 and
LiBH4, no reaction was observed with the latter alone
and although a slow reaction with BH3 alone occurred,
it did not produce the same products.
In summary, a catalytic asymmetric synthesis of
CDP840 has been demonstrated. The results of the
(11) Battistini, C.; Balsamo, A.; Berti, G.; Crotti, P.; Macchia, B.;
Macchia, F. J . Chem. Soc., Chem. Commun. 1974, 712. Imuta, M.;
Ziffer, H. J . Am. Chem. Soc. 1979, 101, 3990.
(12) Balsamo, A.; Crotti, P.; Macchia, B.; Macchia, F. Tetrahedron
1973, 29, 2183.
(13) Reductions of chiral benzylic alcohols with Raney Ni are known
to proceed with retention of configuration in some cases. See ref 8.
(14) Alexakis, A.; J achiet, D.; Normant, J . F. Tetrahedron 1986, 42,
5607.
(15) Brown, H. C.; Tierney, P. A. J . Am. Chem. Soc. 1958, 80, 1552.
Brown, H. C.; Stehle, P. F.; Tierney, P. A. J . Am. Chem. Soc. 1957, 79,
2020. Duke, B. J .; Howarth, O. W.; Kenworth, J . G. Nature 1964, 202,
81.
(10) Brown, H. C.; Yoon, N. M. Chem. Commun. 1968, 1549.

9226 J . Org. Chem., Vol. 62, No. 26, 1997
Lynch et al.
22 °C, and ketone 6 (5.05 g, 17.0 mmol) in THF (50 mL) was
added. The solution was aged 2 h; LC analysis showed 0.5%
ketone remaining. Hexanes (100 mL) and water (50 mL) were
added, the mixture was stirred vigorously for 10 min, and the
layers were separated. The organic layer was concentrated
to a white solid that was then triturated in MTBE (40 mL).
The resulting slurry was stirred overnight, filtered, washed
with MTBE (15 mL), and dried, giving a white solid (6.10 g,
92%). 1H NMR δ 8.13 (m, 2H); 7.38 (m, 2H); 7.33-7.99 (m,
3H); 6.88 (m, 2H); 6.32-6.22 (m, 3H); 4.61 (m, 1H); 3.79 (s,
4H, includes hydroxyl H); 3.52 (s, 2H); 1.87-1.65 (m, 6H);
1.62-1.45 (m, 2H). 13C NMR δ 149.0, 148.6, 147.1, 146.6,
146.4, 138.9, 128.1, 127.1, 126.3, 118.5, 114.1, 111.2, 80.4, 77.7,
56.0, 47.6, 32.74, 32.66, 24.0. IR: 3600. HRMS, m/z Calcd for
C25H27NO3 (M+) 389.1991, found 389.1978.
(E-1-[3-(Cyclop en tyloxy)-4-m eth oxyp h en yl]-1-p h en yl-
2-(4-p yr id yl)eth en e, 9. Carbinol 7 (38.9 g, 100 mmol) was
suspended in toluene (200 mL) at 18 °C. Methanesulfonic acid
(19.6 mL, 300 mmol) was added in one portion, T ) 30 °C after
addition. The mixture was stirred 2 h, and LC showed
complete conversion. THF (200 mL) was added, and after
stirring 1 h the mixture was filtered. The solid was washed
with 1:1 THF:toluene (70 mL) and dried to afford 29.5 g of
pure E olefin MSA salt (63%). The salt (11.9 g, 25.4 mmol)
J acobsen epoxidation of olefins 8 and 9 are in accord with
the “skewed side-on approach” model both in terms of
the direction and degree of asymmetric induction and the
relative rate of reaction. A highly selective hydride
reduction of an epoxide with retention of configuration
has been documented.
Exp er im en ta l Section
Gen er a l. Solvents and reagents were used as received from
the vendor. “Dry” refers to solvents stored over 3A molecular
sieves for at least 16 h prior to use. HPLC analysis were
performed on a Zorbax C8-RX column (250 × 4.6 mm) eluting
with CH3CN and 0.1% aqueous H3PO4 mixtures in proportions
such that the desired analyte was eluted in 3-15 min with
UV detection at 210 nm. NMR and IR spectra were recorded
in CDCl3 solution unless otherwise specified.
r-[3-(Cyclopen tyloxy)-4-m eth oxyph en yl]ben zen em eth -
a n ol, 5. Isovanillin (100 g, 0.65 mol) and potassium carbonate
(138.2 g, 1.12 mol) were added to DMF (600 mL) portionwise
at rT. The mixture was heated to 60 °C, and cyclopentyl
bromide (112 mL, 1.12 mol) was added over 30 min at 60 °C.
The mixture was aged at 60 °C for 14 h. The mixture was
cooled to rT, and water (600 mL) was added in one portion.
The solution was stirred for 30 min. The mixture was
extracted twice with toluene (800, 400 mL). The combined
organic layers were washed with a dilute HCl aqueous solution
(0.2 N, 800 mL) and with water (2 × 600 mL). The organic
layer was concentrated to 1 L in volume. To the above toluene
solution was added phenylmagnesium bromide solution in
ether (230 mL, 3 M) at -10 °C, and the mixture was aged for
30 min below 0 °C. Aqueous HCl (1 N) solution (700 mL) and
ethyl acetate (700 mL) were added successively, and the layers
were separated. The organic layer was washed with water (2
× 500 mL) and concentrated to dryness. Crystallization of the
crude alcohol in 1:6 ethyl acetate:hexanes mixture (1.4 L) gave
167.8 g of a white solid (87% yield). 1H NMR δ 7.4-7.2 (m,
5H); 6.92 (d, J ) 2 Hz, 1H); 6.88 (dd, J ) 7, 2 Hz, 1H); 6.81 (d,
J ) 7 Hz, 1H); 5.26 (d, J ) 4 Hz, 1H); 4.75 (m, 1H); 3.82 (s,
3H); 2.49 (d, J ) 4 H, 1H (O-H)); 2.0-1.75 (m, 6H); 1.7-1.5
(m, 2H). 13C NMR δ 149.5, 147.7, 144.1, 136.6, 128.4, 127.4,
126.5, 119.0, 113.6, 111.9, 80.4, 76.0, 56.1, 32.83, 32.79, 24.10.
IR 3600(s). Anal. Calcd for C19H22O3: C, 76.48; H, 7.43.
Found: C, 76.41; H, 7.45.
3-(Cyclop en t yloxy)-4-m et h oxyb en zop h en on e, 6. Ru-
Cl3‚H2O (0.63 g) was added to a solution of alcohol 5 (50.0 g,
0.168 mol) in 10:1 THF:H2O (198 mL). The solution was
heated to 40 °C in an oil bath, and 70% tert-butyl hydroper-
oxide (10 mL, 73 mmol) was added. The ensuing exothermic
reaction caused the solution to warm slowly to 60 °C and tert-
butyl hydroperoxide (80 mL, 0.584 mol) addition continued at
a rate (45 min) sufficient to maintain a gentle reflux. LC
analysis showed 74% conversion to ketone 6. Additional tert-
butyl hydroperoxide (52 mL, 0.380 mol) was added at 45 °C
(without any observed temperature increase), and the solution
was allowed to stir at ambient temperature for 24 h. LC
analysis indicated 97% conversion. The solution was diluted
with MTBE (250 mL) and quenched with 10% NaHSO3
solution (250 mL), and the organic layer was then washed with
5% NaHCO3 (100 mL) and water (100 mL), dried (MgSO4),
filtered, and concentrated to a thick slurry (160 mL). Hexanes
(300 mL) was added slowly, and the slurry was stirred at
ambient temperature for 30 min, cooled to 0 °C, filtered,
washed with hexane (100 mL), and dried in vacuo overnight,
giving 44.2 g of white solid (89% yield). 1H NMR δ 7.78 (m,
2H); 7.57 (m, 1H); 7.48 (m, 3H); 7.38 (m, 1H); 6.89 (d, J ) 8
Hz, 1H); 4.85 (m, 1H); 3.92 (s, 3H), 2.05-1.75 (m, 6H); 1.70-
1.55 (m, 2H). 13C NMR δ 195.8, 154.0, 147.5, 138.4, 131.8,
130.1, 129.7, 128.1, 125.1, 115.6, 110.4, 80.5, 56.1, 32.8, 24.1.
IR 1650. Anal. Calcd for C19H20O3: C, 77.00; H, 6.80.
Found: C, 76.82; H, 6.80.
was partitioned between EtOAc (100 mL), saturated Na2
CO3
solution (50 mL), and water (50 mL). The organic layer was
washed with water (50 mL) and concentrated to an oil (10.0
g, 105%). The oil was crystallized from EtOAc (50 mL) by
addition of hexanes (200 mL), giving an off-white solid (5.9 g,
62%); a second crop gave 2.8 g (30%) of comparable purity. 1H
NMR δ 8.33 (m, 2H); 7.36 (m, 3H); 7.18 (m, 2H); 6.90-6.78
(m, 6H); 4.69 (m, 1H); 3.86 (s, 3H); 1.90-1.74 (m, 6H); 1.65-
1.51 (m, 2H). 13C NMR δ 150.6, 149.5, 147.4, 147.0, 145.0,
139.5, 135.1, 130.1, 128.8, 128.1, 123.7, 123.5, 120.8, 114.8,
111.5, 80.6, 56.1, 32.8, 24.1. IR: 1600, 1515. Anal. Calcd for
C25H25NO2: C, 80.83; H, 6.78; N, 3.77. Found: C, 80.58; H,
6.76; N, 3.67.
(Z)-1-[3-(Cyclop en tyloxy)-4-m eth oxyp h en yl]-1-p h en yl-
2-(4-p yr id yl)eth en e, 8. MSA (7.0 mL, 108 mmol) was added
to carbinol 7 (20.9 g, 53.6 mmol) in THF (150 mL). The
solution was refluxed 4 h and was then concentrated to
dryness. The residue was triturated in 10:1 methyl isobutyl
ketone (MIBK):MeOH (220 mL), and the resulting solid was
filtered (12.3 g of E isomer, 49%). The mother liquors were
concentrated to dryness, and the residue was dissolved in CH2-
Cl2 (150 mL). NaOH (2 N) (50 mL) was added, and the
mixture was stirred for 10 min. The organic layer was
concentrated to a thick oil; EtOAc (45 mL) was added, and
the mixture was stirred for 2 h at 22 °C. The resulting slurry
was cooled to 0 °C, stirred an additional 30 min, filtered,
washed with cold EtOAc, and dried giving an off-white solid,
5.18 g (26%). 1H NMR δ 8.38 (d, J ) 7 Hz, 2H); 7.36 (m, 5H);
6.92 (d, J ) 7 Hz, 2H); 6.83 (d, J ) 8 Hz, 1H); 6.81 (s, 1H);
6.68 (m, 2H); 4.57 (m, 1H); 3.88 (s, 3H); 1.8-1.65 (m, 6H);
1.57-1.48 (m, 2H). 13C NMR δ 150.1, 149.6, 147.7, 147.1,
145.2, 142.6, 131.6, 128.4, 128.3, 127.9, 124.7, 123.8, 122.6,
117.1, 112.0, 80.3, 56.0, 32.6, 24.1. IR 1600, 1510. Anal.
Calcd for C25H25NO2: C, 80.83; H, 6.78; N, 3.77. Found: C,
80.50; H, 6.77; N, 3.60.
syn -(1S,2S)-1-[3-(Cyclop en t yloxy)-4-m et h oxyp h en yl]-
1-p h en yl-2-(4-p yr id yl)-eth ylen e Oxid e, 11. CH2Cl2 (8 mL)
was added to a mixture of olefin 8 (2.40 g, 6.46 mmol), S,S
salen Mn complex 10 (48 mg, 0.075 mmol), and 4-(3-phenyl-
propyl)pyridine N-oxide (64 mg, 0.03 mmol), and the dark
mixture was cooled to 0 °C. A solution of NaOCl (1.72 M, 13
mL, 22.4 mmol) and 5 N NaOH (0.4 mL, 2 mmol) was added
over 1.5 h (via syringe pump), and the mixture was stirred at
0 °C for 22 h. The mixture was diluted with MTBE (10 mL),
the resulting emulsion was filtered, and the layers were
separated. The organic layer was washed with water, dried
(Na2SO4), and concentrated to a dark oil (2.17 g). Flash
chromatography (1:1 EtOAc:hexanes) gave the epoxide (1.58
g) in 91% purity (3.72 mmol, 58%); the epoxide coelutes with
unreacted 8 and an overoxidized byproduct. LC on Chiracel
OD (75:25 hexane:iPA 1 mL/min showed a major enantiomer
at 5.5 min (80 A%) and a minor one at 11.1 min (4.4 A%) 89.6%
1-[3-(Cyclop en t yloxy)-4-m et h oxyp h en yl]-1-p h en yl-2-
(4-p yr id yl)eth a n ol, 7. NaHMDS (2.21 M in THF, 11.5 mL,
25.4 mmol) was added to 4-picoline (2.46 mL, 25.3 mmol) in
dry THF (50 mL) at 22 °C. The solution was aged for 1 h at

Asymmetric Synthesis of CDP840
J . Org. Chem., Vol. 62, No. 26, 1997 9227
ee. 1H NMR δ 8.41 (d, J ) 6 Hz, 2H); 7.33 (bs, 5H); 6.99 (d, J
) 6 Hz, 2H); 6.72 (s, 2H); 6.65 (s, 1H); 4.56 (m, 1H); 4.28 (s,
1H); 3.79 (s, 3H); 1.9-1.48 (m, 8H). 13C NMR δ 149.7, 149.3,
147.0, 144.7, 140.4, 128.5, 128.2, 127.0, 126.3, 121.73, 121.66,
116.0, 111.4, 80.4, 68.8, 66.7, 55.9, 32.7, 32.5, 24.1, 24.0. IR
1260. HRMS, m/z Calcd for C25H25NO3: 387.1834. Found:
387.1817. A nearly enantiomerically pure sample was ob-
tained from an oxidation which proceeded to only 85% conver-
sion. Following chromatography, a crystalline solid separated
from the concentrated fractions; this mixture was slurried in
4:1 hexane:EtOAc, and the solid was filtered. LC analysis
showed this solid was a 2:1 mixture of 11 (racemic):8. Careful
preparative TLC of the mother liquors provided a sample of
13-F r ee Ba se. The borane complex (338 mg, 80% ee, oil
from above) was refluxed in 4:1 toluene:MeOH (10 mL) for 8
h, and LC analysis showed complete conversion. The solution
was then concentrated redissolved in fresh 4:1 toluene:metha-
nol and concentrated again giving an oil (311 mg, 95%). Chiral
LC (Chiracel OJ , 60:40 hexane: EtOH 1 mL/min, 35 °C, 280
nm) showed a major isomer at 5.4 min (87.7%) and a minor
isomer at 8.4 min (9.3%), 80% ee. Treatment of the crystalline
material from above in the same manner provided an oil in
100% yield, and chiral LC showed a major isomer at 5.4 min
1
(94.8%) and the minor isomer at 8.4 min (0.4%), 99.2% ee. H
NMR δ 8.45 (m, 2H), 7.39 (m, 4H), 7.35-7.28 (m, 1H), 7.10
(m, 2H), 6.69 (d, J ) 8 Hz, 1H), 6.65 (dd, J ) 8, 2 Hz, 1H),
6.61 (d J ) 2 Hz, 1H), 5.33 (d, J ) 9 Hz, 1H), 4.58 (m, 1H),
4.10 (d, J ) 9 Hz, 1H), 3.77 (s, 3H), 2.25 (bs, 1H), 2.0-1.5 (m,
8H). 13C NMR δ 152.0, 149.2, 148.9, 147.3, 140.4, 133.3, 128.8,
128.5, 127.1, 122.0, 120.8, 116.0, 111.9, 80.5, 75.7, 59.4, 56.0,
99.7% ee. [R]36 -16.2 (c 1.34, EtOH).
D
a n ti-(1R,2S-1-[3-(Cclop en tyloxy)-4-m eth oxyp h en yl]-1-
p h en yl-2-(4-p yr id yl)-eth ylen e Oxid e, 12. NaOCl (1.72 M)
(26 mL, 44.7 mmol) containing 5 N NaOH (0.8 mL, 4 mmol)
was added to a solution of olefin 9 (4.80 g, 12.9 mmol), S,S
Mn salen complex 10 (96 mg, 0.15 mmol), and 4-(3-phenyl-
propyl)pyridine N-oxide (128 mg, 0.6 mmol) in CH2CL2 (16 mL)
over 40 min at 0 °C. After 18 h, additional 10 (48 mg) and
4-(3-phenylpropyl)pyridine N-oxide (128 mg) was added; 4 h
later a third portion of catalyst (48 mg) was added and the
mixture was stirred an additional 3 h. LC analysis showed
87:5 ratio of 12:9. Hexane (16 mL) was added, and the layers
were separated. The organic layer was washed with water (2
× 10 mL), dried over MgSO4, and concentrated to an oil. Flash
chromatography gave the epoxide 12 (3.60 g, 69% yield). LC
on Chiracel OJ , 80:20 hexane:EtOH showed a major isomer
at 5.50 min, 73.5%, and a minor isomer at 7.22 min, 25.5%
(48% ee). The sample crystallized on standing; the solid was
slurred in 4:1 hexane:EtOAc (25 mL) and filtered, giving 1.90
g of crystalline epoxide. Chiral LC analysis showed 10% ee.
The mother liquors were evaporated to an oil (1.70 g) shown
to be 95% ee. 1H NMR δ 8.48 (d, J ) 6 Hz, 2H); 7.22 (m, 5H);
6.98 (d, J ) 6 Hz, 2H); 6.89 (s, 1H); 6.80 (s, 2H); 4.72 (m, 1H);
4.39 (s, 1H); 3.83 (s, 3H); 2.0-1.7 (m, 6H); 1.7-1.5 (m, 2H).
13C NMR δ 150.0, 149.2, 147.8, 144.6, 135.0, 132.4, 128.9,
128.1, 128.0, 121.6, 119.3, 112.7, 111.5, 80.5, 69.1, 66.5, 56.1,
32.84, 32.77, 24.1. IR 1265. Anal. Calcd for C25H25NO3: C,
77.49; H, 6.50; N, 3.61. Found: C, 77.26; H, 6.47; N, 3.54.
(1R,2R-2-[3-(cyclopen tyloxy)-4-m eth oxyph en yl]-2-ph en -
yl-1-(4-p yr id yl)eth a n ol BH3 Com p lex, 13. LiBH4 (66 mg,
3 mmol) was dissolved in 1 M BH3‚THF (10 mL, 10 mmol). A
solution of the 1S,2S epoxide 11 (1.23 g, 3.17 mmol) in THF
(5 mL) was cooled to -20 °C, and 6.5 mL of the LiBH4/BH3
solution was added dropwise such that the temperature did
not rise above -10 °C. The solution was aged at -20 °C for 3
h, and then MeOH (2.5 mL) was added dropwise. Hexanes
(10 mL), EtOAc (10 mL), and water (20 mL) were added with
vigorous stirring. The organic layer was separated and
concentrated to a yellow solid. The solid was broken up and
stirred with 8:3 hexanes: EtOAc (11 mL) at 22 °C for 1 h,
cooled to 0 °C, filtered, washed with hexanes, and dried, giving
a white solid (590 mg) (mp 83-93 °C dec, 1H NMR analysis
showed 18 mol % ) 5 wt % EtOAc which could not be removed
by drying in vacuum at 25 °C, 44% yield). The mother liquors
and wash were concentrated and flash chromatographed (2:1
32.9, 32.6, 24.1. [a]33 +119 (c 1.65, EtOH) for the 99.2% ee
D
sample. HRMS, m/z Calcd for C25H27NO3: 389.1991. Found:
389.1993.
(1S,2R-2-[3-(Cyclopen tyloxy)-4-m eth oxyph en yl]-2-ph en -
yl-1-(4-p yr id yl)eth a n ol, 14. The LiBH4-BH3 solution (pre-
pared as for 13, 6 mL) was added to epoxide 12 (1.03 g, 2.66
mmol) in dry THF (3 mL) at -30 °C at such a rate that the
temperature did not exceed -15 °C. The solution was aged 1
h at -30 °C, and then 2 h at -20 °C, and then MeOH (2.5
mL) was added dropwise. Hexanes (10 mL), EtOAc (10 mL),
and water (20 mL) were added with vigorous stirring. The
organic layer was dried (Na2SO4) and concentrated to an oil;
LC analysis showed 55 area % 14-BH3 and 6.7% 13. The
crude oil was chromatographed on silica gel (2:1 hexane:
EtOAc) and the purified oil, 89:11 14-BH3:13 (331 mg, 31%)
was dissolved in EtOH (10 mL) and heated at reflux 10 h,
concentrated, redissolved in EtOH (10 mL), and concentrated
again. Chiral LC showed 93% ee. 1H NMR δ 8.25 (m, 2H),
7.23-7.10 (m, 5H), 7.08 (m, 2H), 6.39-6.25 (m, 3H), 5.30 (d,
J ) 8 Hz, 1H), 4.69 (m, 1H), 4.10 (d, J ) 8 HZ, 1H), 3.8 (bs,
1H), 3.78 (s, 3H), 1.9-1.7 (m, 6H), 1.7-1.5 (m, 2H). 13C NMR
δ 152.3, 149.1, 149.0, 147.6, 141.4, 132.2, 128.51, 128.46, 126.7,
121.9, 121.1, 116.3, 112.1, 80.5, 75.3, 59.1, 56.0, 32.9, 32.8, 24,1.
IR:3585, 1602, 1510, 1266. Combustion analysis was obtained
for the benzenesulfonic acid salt. Benzenesulfonic acid (105
mg, 0.66 mmol) was added to the alcohol 14 (235 mg, 0.60
mmol) in EtOAc (1.5 mL), giving a clear solution. The solution
was allowed to stand at ambient temperature for 1 h. The
resulting crystalline salt was collected by filtration, washed
with EtOAc (5 mL), and dried overnight in vacuo (305.4 mg,
0.55 mmol, 92%). Anal. Calcd for C31H33N06S: C, 67.99; H,
6.07; N, 2.56. Found: C, 67.55; H, 6.04; N, 2.40.
(2R)-1-[3-(Cyclop en tyloxy)-4-m eth oxyp h en yl]-1-p h en -
yl-2-(4-p yr id yl)eth a n e, 3, CDP 840. Triethylamine (15.6 mg,
0.15 mmol) was added to the free base 13 (89.6% ee, 30 mg,
0.077 mmol) in CH2Cl2 (0.25 mL) at 0 °C. Methanesulfonyl
chloride (13.2 mg, 0.116 mmol) was added, and the solution
was aged 30 min. LC showed 95.4% mesylate and 4% 13. The
CH2Cl2 was evaporated, the residue was dissolved in acetic
acid (0.5 mL), and zinc dust (15 mg, 0.23 mmol) was added.
LC analysis showed 72% conversion to CDP840 after 10 min.
The mixture was allowed to stir 10 h, giving 95% CDP840:5%
13. The mixture was diluted with methyl tert-butyl ether (5
mL) and water (2 mL), and 5 N NaOH was added dropwise to
pH 11. The layers were separated, the organic layer was
concentrated to an oil (40 mg) that was purified by prep TLC
(2:1 hexane:EtOAc), giving 24.4 mg of oil (85%). Chiral LC
showed two peaks 6.36 min, 95% and 9.48 min, 5% (90% ee).
An authentic sample of CDP840 showed a single peak on the
same LC system at 6.35 min. A sample of 13 of 99.0% ee
similarly treated gave CDP840 also in 99.0% ee.
1
hexanes:EtOAc), giving an additional 338 mg of oil (27%). H
NMR δ 8.31 (d, J ) 7 Hz, 2H); 7.35 (m, 4H); 7.29 (m, 1H);
7.23 (d, J ) 7 Hz, 2H); 6.70 (d, J ) 8 Hz, 1H); 6.63 (dd, J ) 8,
2 Hz, 1H); 6.57 (d, J ) 2 Hz, 1H); 5.41 (dd, J ) 9, 3 Hz, 1H);
4.59 (m, 1H); 4.02 (d, J ) 9 Hz, 1H); 3.76 (s, 3H); 2.68 (d, J )
3 Hz, 1 H); 2.9-2.1 (bs, 3H); 1.9-1.68 (m, 6H); 1.68-1.50 (m,
2H). 13C NMR δ 155.7, 149.2, 147.5, 146.7, 139.4, 132.2, 129.1,
128.7, 127.5, 123.3, 120.7, 115.8, 112.1, 80.6, 75.1, 59.4, 56.0,
32.8, 32.6, 24.1. IR 3700-3100, 2370. Anal. Calcd for
C25H30BNO3: C, 74.45; H, 7.50; N, 3.47. Found: C, 73.53; H,
7.56; N, 3.34 (consistent with 5 wt % EtOAc contamination).
HRMS, m/z Calcd for C25H30BNO3: 403.2319. Found: 403.2369.
A crystal suitable for X-ray analysis was obtained from the
above crystalline salt. A slurry of 13 (10 mg) in EtOH
(absolute, 0.3 mL) was warmed briefly in a 45 °C oil bath (10
s); a small amount remained undissolved. The mixture was
allowed to stand overnight at ambient temperature, giving
clear, prism-shaped crystals, mp 123 °C dec.
1-(4-Meth oxyp h en yl)cycloh exen e Oxid e (15b).12 K2CO3
(1 g) and then 1-(4-methoxyphenyl)cyclohexene (448 mg, 2.38
mmol) were added to a 0.14 M solution of dimethyldioxirane
(34 mL, 4.76 mmol) in acetone (prepared as described by Adam
et al.16) at 0 °C. After 10 min, HPLC showed complete reaction.
(16) Adam, W.; Chan, Y.-Y.; Cremer, D.; Gauss, J .; Scheutzow, D.;
Schindler, M. J . Org. Chem. 1987, 52, 2800.

9228 J . Org. Chem., Vol. 62, No. 26, 1997
Lynch et al.
The mixture was filtered and concentrated. The residue was
dissolved in CDCl3 (3 mL) and dried over K2CO3; concentration
gave a clear oil that crystallized on standing, mp 45-46 °C.
1H NMR 7.30 (d, J ) 9.7 Hz, 2H); 6.88 (d, J ) 9.7 Hz, 2H);
3.79 (s, 3H); 3.07 (m, 1H); 2.25 (ddd (ABXY), J ) 14, 9, 6 Hz,
1H); 2.11 (m (ABXY), 1H); 1.98 (m, 2H); 1.7-1.4 (m, 3H); 1.4-
1.2 (m, 2H). 13C NMR δ 158.8, 134.7, 126.6, 113.7, 62.0, 59.9,
55.3, 29.0, 24.8, 20.2, 19.9.
Ack n ow led gm en t. We thank Ms. Amy Bernick for
mass spectroscopic measurements, Professor Eric J a-
cobsen for valuable conversations, and reviewer “C” for
bringing reference 10 to our attention.
Su p p or tin g In for m a tion Ava ila ble: X-ray data for com-
pound 13 (6 pages). This material is contained in libraries on
microfiche, immediately follows this article in the microfilm
version of the journal, and can be ordered from the ACS; see
any current masthead page for ordering information.
(17) The author has deposited atomic coordinates for this structure
with the Cambridge Crystallographic Data Centre. The coordinates
can be obtained, on request, from the Director, Cambridge Crystal-
lographic Data Centre, 12 Union Road, Cambridge, CB2 1EZ, UK.
J O971476C