Crystallization-induced diastereoselection: Asymmetric synthesis of substance P inhibitors

Article abstract of DOI:10.1002/1521-3765(20020315)8:6<1372::AID-CHEM1372>3.0.CO;2-6

A novel three-component condensation followed by a crystallization-induced asymmetric transformation is used to build this key substance P inhibitor intermediate in a short synthetic sequence.

Full text of DOI:10.1002/1521-3765(20020315)8:6<1372::AID-CHEM1372>3.0.CO;2-6

FULL PAPER
Crystallization-Induced Diastereoselection: Asymmetric Synthesis
of Substance P Inhibitors
Philip J. Pye,*[a] Kai Rossen,*[a] Steven A. Weissman,[a] Ashok Maliakal,[a]
Robert A. Reamer,[a] Richard Ball,[b] Nancy N. Tsou,[b] R. P. Volante,[a] and Paul J. Reider[a]
Keywords: asymmetric synthesis ¥
Abstract: A novel three-component condensation followed by a crystallization-
induced asymmetric transformation is used to build this key substance P inhibitor
intermediate in a short synthetic sequence.
crystallization-driven transforma-
tion ¥ dihydroxylation ¥ Mitsunobu
reaction ¥ substance P inhibitors
Introduction
CF3
CF3
HO
LG
O
N
O
Although the neuropeptide substance P was isolated from
brain cells as early as 1931,[1] its functional role was not well
understood until the advent of orally-available substance P
inhibitors.[2] These compounds show great promise for the
treatment of chemotherapy-induced emesis, but more excit-
ingly, a recent clinical trial has shown the efficacy of a Merck
substance P antagonist as an antidepressant in patients with
major depressive disorder. This landmark discovery is even
more significant as it demonstrates a novel mode of action for
antidepressive therapy.[3]
O
3
CF3
CF3
F
H
N
N
O
1
Bn
N
N
H
F
2
CF3
CF3
Hofmann
elimination
O
O
O
N
O
CF3
N
CF3
F
X
Bn
Bn
F
4
5
Results and Discussion
OH
NH
O
N
OH
Structurally, the Merck substance P antagonists consist of a
C2/C3 cis-substituted morpholine acetal core linked to a
heterocycle, such as the 3-oxo-1,2,4-triazole moiety in 1.[4] An
initial retrosynthetic analysis of 1 suggested an apparently
trivial disconnection at the acetal center (Scheme 1) leading
to 3,5-bis(trifluoromethyl)-sec-phenethyl alcohol 3 and the
activated morpholine 2. Unfortunately, this glycosidation-
OH
HO
F
F3C
CF3
F3C
CF3
7
6
Scheme 1. Retrosynthetic analysis of 1.
[a] Dr. P. J. Pye, Dr. K. Rossen, Dr. S. A. Weissman, A. Maliakal,
R. A. Reamer, Dr. R. P. Volante, Dr. P. J. Reider
Department of Process Research
type approach failed, as either facile elimination of the leaving
group (LG) or substitution from the b-face resulted in the
undesired trans stereochemistry, presumably due to steric
blocking by the adjacent 4-fluorophenyl group. Alternatively,
vinyl ether 4 was a desirable intermediate, since diastereose-
lective hydrogenation of the vinyl ether is known to afford the
requisite stereochemistry at the newly formed chiral center.[4]
Importantly, 4 is retrosynthetically transformed to the bicyclic
quaternary ammonium salt 5 by a regioselective Hofmann
elimination. The critical cis-stereochemistry in 5 is set by an
Merck Research Laboratories, P.O. Box 2000, Rahway
New Jersey 07065 (USA)
Fax : (‡1)908-594-5190
E-mail: philip_pye@merck.com, kai.rossen@degussa-huels.de
[b] Dr. R. Ball, N. N. Tsou
Department of Molecular Design and Diversity
Merck Research Laboratories, P.O. Box 2000, Rahway
New Jersey (USA)
Supporting information for this article is available on the WWW under
http://www.wiley-vch.de/home/chemistry/ or from the author.
1372
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0806-1372 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 6

1372 1376
intramolecular displacement of the lactol hydroxyl group of 6,
thus overcoming the bias towards the trans arrangement. We
envisioned deriving lactol 6 from chiral aminodiol 7 using a
new Mannich boronic acid condensation. It was imperative to
the success of this strategy that the single stereocenter of
aminodiol 7 controls both stereocenters in the morpholine
ring of 6.
OH
NH
O
O
H
H
(HO)2B
+
+
HO
F
10
F3C
CF3
tert-amylalcohol
7
Synthesis of aminodiol 7 in racemic or enantiomerically
pure form was readily achieved by using known chemistry. A
Sharpless asymmetric dihydroxylation of 3,5-bis(trifluoro-
methyl)styrene (8) set up the necessary absolute stereochem-
istry, and subsequent selective activation of the primary
alcohol as the mesylate followed by displacement with
ethanolamine lead cleanly to crystalline 7 (Scheme 2; see
also Experimental Section). An ambitious new three-compo-
nent condensation was then utilized to assemble the core
morpholine ring system in one step from 7.[5]
Ar-CF3
O
OH
O
N
OH
N
Ar-F
Ar-F
O
N
OH
HO
HO
Ar-F
Ar-CF3
13
O
HO
11
O
Ar-CF3
6
OH
Ar-CF3
OH
N
Ar-F
N
Ar-F
Interconverting
mixture of
isomers
HO
HO
Ar-CF3
12
14
OH
OH
NH
Crystallization
HO
HO
O
OH
O
N
OH
b
a
N
HO
HO
F3C
CF3
F3C
CF3
F3C
CF3
F
F
8
9
7
Scheme 2. Synthesis of aminodiol 7. a) (DHQ)2PHAL, K2OsO4, NMNO,
tBuOH, water; b) MeSO2Cl, 2,6-lutidine, MeCN.
F3C
CF3
F3C
CF3
rac-6
12
Enantiomerically Pure
Racemic
Heating 7 with 4-fluorophenylboronic acid (10) and aque-
ous glyoxal afforded, in quantitative yield, a mixture of lactol
diastereomers: 6, 11, and 12 (50:10: <2 area%) and
regioisomers 13 and 14 (10 and 20 area%) (Scheme 3). This
poor selectivity would severely limit the utility of this
morpholine synthesis, if not for the fact that these isomers
are interconvertible, presumably through the ring-opened
aldehyde/diol. Thus, a chromatographically isolated mixture
of 11 and 12 returns to the initially observed isomeric mixture
of 6, 11, 12, 13, and 14 by addition of DBU, catalytic H3PO4, or
by simple heating. This facile equilibration of the isomers in
solution, coupled with a crystallization of the desired diaste-
reomer, could lead to a crystallization-induced transformation
that would funnel the complex mixture into a single crystalline
isomer.[6] Indeed, this crystallization-induced asymmetric
transformation was first demonstrated in the racemic series
by seeding the mixture of 6, 11, 12, 13, and 14 with crystalline
rac-6 to afford the desired rac-6 in 65% yield (90 area%).
Crystal properties of enantiomerically pure compounds are
expected to differ from those of the racemate.[7] Thus, the
dependence of our synthesis on a crystallization-induced
asymmetric transformation gives this often inconsequential
issue great significance. Repeating the boronic acid conden-
sation with (S)-7, we obtained the same expected ratio of
isomers in solution. However, attempts to crystallize enantio-
merically-pure trans-lactol 6 were unsuccessful and resulted in
the isolation of a minor component of the reaction mixture,
the cis-lactol 12. Pure 12 was obtained in 86% yield by slow
Ar-CF3 = 3,5-bis(trifluoromethyl)phenyl; Ar-F = 4-fluorophenyl
Scheme 3. Interconversion of diastereomers.
crystallization from the reaction mixture with methylcyclo-
hexane. Evidently, the trans-lactol 6 is still the thermodynami-
cally preferred species in solution and, consequently, brief
exposure of a solution of the cis-lactol 12 to a trace amount of
H3PO4 rapidly establishes a 87:13 equilibrium of trans/cis-
lactols 6 and 12 (Scheme 4). Single-crystal X-ray structures
were obtained for rac-6 and enantiomerically pure 12 and
show extensive hydrogen-bonding networks for both cases.[8]
While the X-ray result confirms the structural assignments, a
simple explanation for the forces that stabilize the minor
species 12 of a mixture in the crystalline state of the
enantiomerically pure product is not readily apparent (see
Supporting Information).
As the subsequent steps required the use of the trans-lactol
6 (vide infra), 12 was first converted to trans-lactol 6. This was
accomplished in 98% yield by equilibration of a solution of 12
to the 87:13 trans/cis mixture and slow precipitation of the
trans-lactol 6 as the HCl salt. A simple conversion to the free
base gives a solution of clean trans-lactol 6, which is stable at
208C for extended periods of time in the absence of strong
bases or acids.
Although a number of the lactol isomers could potentially
cyclize to bicyclic acetal 15, attempts to achieve this trans-
formation under traditional acetal-forming conditions result-
Chem. Eur. J. 2002, 8, No. 6
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0806-1373 $ 17.50+.50/0
1373

P. J. Pye, K. Rossen et al.
FULL PAPER
ide.[10] The choice of iodide counterion for salt 5 allowed the
direct isolation of the key vinyl ether 4 in 90% yield with
rejection of the sodium iodide byproduct in the liquors.
In conclusion, we have developed a short practical asym-
metric synthesis of vinylether intermediate 4, a key inter-
mediate for the synthesis of substance P inhibitor 1, in 58%
overall yield from readily available starting materials. The
morpholine core is assembled rapidly in a novel three-
component cyclization, and a subsequent crystallization-
induced asymmetric transformation is used to funnel the
complex mixture of diastereomers into a single species, thus
establishing the morpholine stereochemistry. A diastereose-
lective formation of the bicyclic acetal 15, quaternization to 5,
and a subsequent regioselective Hofmann elimination lead to
4. Notably, all chiral centers in 4 and subsequently in 1 are
ultimately derived from the single chiral center in diol (S)-7,
prepared by the Sharpless catalytic asymmetric dihydroxyla-
tion.
O
N
OH
O
OH
H3PO4
THF
N
HO
HO
F
F
87 : 13
F3C
CF3
F3C
CF3
12
6
Enantiomerically
Pure
O
N
OH
HCl(g)
HO
F
HCl
F3C
CF3
6. HCl salt
Scheme 4. cis trans Isomerization.
ed in complex mixtures. We then focused on the cyclization of
trans-lactol 6 through selective activation of the lactol
hydroxyl group and SN2 type cyclization with the side chain
alcohol from the a-face. The different steric and electronic
environment of both alcohols made it likely that a regiose-
lective activation of the lactol hydroxyl would be possible, and
the neutral conditions required to prevent the facile elimi-
nation of the activated lactol 2 directed us to the Mitsunobu
reaction.[9] Treatment of a solution of 6 with tributylphosphine
in THF at À308C followed by the addition of diisopropyl
diazodicarboxylate (DIAD) and warming to ambient temper-
ature led to the formation of crystalline bicyclic acetal 15 in
86% isolated yield from 6 ¥ HCl (Scheme 5).
Experimental Section
General: All reagents were purchased from common commercial sources
and used as received. All reactions were performed in a dry nitrogen
atmosphere. Column chromatography was performed with EM Science
silica gel 60 (230 400 mesh).
Unless otherwise stated all reactions were monitored by HPLC analysis by
using a Thermo Separations AS1000 with a Zorbax Rx-C8 reverse phase
column. Optical rotations and microanlyses were determined by QTI
(Whitehouse, NJ). NMR spectra were obtained on Bruker Avance 600, 500,
and 400 MHz spectrometers. Spin spin coupling constants (J) are reported
in hertz.
Preparation of 3,5-bistrifluoromethylstyrene (8): A solution of 3,5-bis(tri-
fluoromethyl)bromobenzene (75 g, 0.256 mol), tetrabutylammonium chlor-
ide (71.4 g, 0.256 mol), triethylamine (71.3 mL, 0.97 mol), and palladium(ii)
acetate (113 mg) in acetonitrile (360 mL) was deoxygenated with
a
vigorous flow of nitrogen. The reaction was pressurized with ethylene
(950 psi) and heated to 808C for 36 hours. Pentane (400 mL) was added,
and, after washing with water (3 Â 200 mL), the organic layer was dried
over Na2SO4 and the solvent was removed in vacuo to afford 8 as a pale
yellow oil (58.2 g, 0.242 mol, 95% yield). 1H NMR/13C NMR consistent
with that purchased from Aldrich.
F
O
N
OH
N
O
b
a
O
HO
F
HCl
CF3
Preparation of (S)-bis-3,5-(trifluoromethyl)phenyl ethyleneglycol (9):
Hydroquinine 1,4-phthalazinediyl diether ((DHQ)2PHAL) (13.3 g,
17.0 mmol) was added to a solution of potassium osmate dihydrate
(5.68 g, 15.4 mmol) in tert-butanol (2.3 L)/water (1.97 L). An aqueous
solution of N-methylmorpholine N-oxide (596 mL, 50 wt.%, 2.88 mol) was
then added. After 15 minutes styrene 8 (468 g, 1.95 mol) was added over
3.5 hours, while a temperature of 158C was maintained. The reaction
mixture was warmed to 238C for 30 minutes and quenched by the addition
of 10% aq. Na2SO3. After aging for 18 hours, the phases were separated
and the aqueous layers extracted with ethyl acetate (1 Â 2 L). The organic
layers were combined, washed with 0.4n H2SO4 saturated with Na2SO4
(2.4 L). After drying over Na2SO4 the solvent was removed in vacuo to
afford crude 9 as an off white solid (87% ee). Recrystallization from ethyl
F3C
CF3
CF3
6. HCl salt
15
F
CF3
Bn
N
O
O
c
O
I
O
1
N
CF3
CF3
F
Bn
acetate/hexanes gave 9 as a white solid (395 g, 74%, >99% ee). M.p. 142
5
4
25
1448C; [a] ˆ 19.858 (c ˆ 2.833 in MeOH); 1H NMR (400.13 MHz,
CF3
589
[D6]DMSO): d ˆ 8.02 (s, 2H), 7.94 (s, 1H), 5.69 (d, J ˆ 4.8 Hz, 1H), 4.84
(t, J ˆ 5.6 Hz, 1H), 4.77 (q, J ˆ 5.3 Hz, 1H), 3.59(m, 1H), 3.50 (m, 1H);
13C NMR (100.61 MHz, [D6]DMSO): d ˆ 147.7, 130.1 (q, J ˆ 32.8 Hz), 127.5
(q, J ˆ 3.2 Hz), 123.9(q, J ˆ 273.1 Hz), 120.8 (septet, J ˆ 4.0 Hz), 72.6, 66.8;
IR(KBr): nÄ ˆ 3312, 2896, 1622, 1473 cmÀ1; elemental analysis calcd (%) for
C10H8F6O2: C 43.81, H 2.94, F 41.58; found: C 43.45, H 2.77, F 41.93.
Scheme 5. a) i) aq. K2CO3, EtOAc; ii) Bu3P, DIAD, THF, À308C to 208C,
86%; b) BnI, acetone, 508C, 89%; c) 1.1 equiv NaOH, EtOH, H2O, 90 %.
The bicyclic acetal was quaternized in acetone with benzyl
iodide at 508C to yield 89% of iodide salt 5. Regioselective
Hofmann elimination proceeded as expected by heating 5 in
ethanol/water (3:1) with 1.1 equivalents of sodium hydrox-
Preparation of (S)-2-imino[1-(bis-3,5-(trifluoromethyl)phenyl)ethanol]-2'-
ethanol (7): Methanesulfonyl chloride (89mL, 1.26 mol) was added over
1 hour to a solution of diol 9 (300 g, 1.09mol) in acetonitrile (900 mL)
1374
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0806-1374 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 6

Substance P Inhibitors
1372 1376
containing 2,6-lutidine (600 mL, 5.2 mol). HPLC analysis indicated com-
plete consumption of diol. After addition of 2-aminoethanol (1 L), the
acetonitrile and 2,6-lutidine were removed by reduced pressure distillation
at 908C over 3 hours. The remaining oil was partitioned between 10%
aqueous sodium carbonate (1 L) and ethyl acetate (1.2 L). The organic
layer was washed with water (3 Â 300 mL) and then with brine (1 Â
300 mL). After drying with MgSO4, the solvent was removed to afford an
143.9, 132.9 (d, J ˆ 3.1 Hz), 131.7 (q, J ˆ 33.2 Hz), 129.4 (d, J ˆ 8.0 Hz),
126.0 (q, J ˆ 2.5 Hz), 123.1 (q, J ˆ 272.6 Hz), 121.6 (m), 115.4 (d, J ˆ
20.9Hz), 89.0, 71.5, 61.2, 56.8, 52.7, 52.6; IR (KBr): nÄ ˆ 3073, 2976, 1016,
935 cmÀ1; elemental analysis calcd (%) for C20H16F7NO2: C 55.18, H 3.70, F
30.55, N 3.22; found: C 55.10, H 3.60, F 30.66, N 3.20.
Preparation of bicyclic acetal quaternary ammonium iodide 5: Benzyl
chloride (0.7 mL, 6.08 mmol) was added to a solution of sodium iodide
(880 mg, 587 mmol) in acetone (10 mL), and the reaction mixture stirred in
the dark for 18 hours. The resulting slurry was filtered, bicyclic acetal (15)
(2.00 g, 4.60 mmol) was added, and the resulting solution was heated at
508C for 8 hours. The solvent was switched into cyclohexane (10 mL), and
oil. Crystallization from MBTE/heptane afforded 59% of (S)-7 as a white
25
solid (205 g, 0.647 mol). M.p. 80 828C; [a] ˆ 34.48 (c ˆ 2.64 in MeOH);
589
1H NMR (400.13 MHz, [D6]DMSO): d ˆ 8.03 (s, 2H), 7.93 (s, 1H), 5.71
(brs, 1H), 4.85 (m, 1H), 4.46 (brs, 1H), 3.43 (t, J ˆ 5.6 Hz, 2H), 2.75 (m,
1H), 2.69(m, 1H), 2.59(m, 2H), 1.95 (brs, 1H); 13C NMR (100.61 MHz,
[D6]DMSO): d ˆ 148.7, 130.1 (q, J ˆ 32.0 Hz), 127.1 (q, J ˆ 3.2 Hz), 123.8 (q,
J ˆ 272.3 Hz), 120.7 (m), 70.8, 60.8, 57.2, 51.8; IR (KBr): nÄ ˆ 2973, 2868,
1427, 1029cm À1; elemental analysis calcd (%) for C12H13F6NO2: C 45.43, H
4.13, F 35.93, N 4.42; found: C 45.15, H 3.77, F 35.89, N 4.12.
the product was removed by filtration to afford quaternary salt 5 as a pale
25
yellow solid (2.76 g, 89%). M.p. 169 171 8C; [a] ˆ À0.868 (c ˆ 2.05 in
589
1
MeOH); H NMR (500.13 MHz, CD3CN): d ˆ 8.25 (m, 2H), 8.15 (s, 2H),
8.09(s, 1H), 7.55 (m, 2H), 7.51 (m, 1H), 7.44 (m 2H), 7.32 (m, 2H), 6.03 (s,
1H), 5.91 (dd, J ˆ 11.9, 3.6 Hz, 1H), 5.77 (s, 1H), 5.62 (d, J ˆ 12.3 Hz,), 4.82
(m, 1H), 4.70 (m, 1H), 4.45 (m, 1H), 3.75 (d, J ˆ 12.3 Hz, 1H), 3.68 (m,
2H), 3.57 (t, J ꢀ 12 Hz, 1H); 13C NMR (125.77 MHz, CD3CN): d ˆ 165.3 (q,
J ˆ 249.8 Hz), 140.1, 136.5 (vbr), 134.6, 132.6 (q, J ˆ 33.8 Hz), 131.7, 130.0,
128.3 (q, J ˆ 3.7 Hz), 126.8, 124.8 (d, J ˆ 3.7 Hz), 124.4 (q, J ˆ 272.0 Hz),
124.0 (m), 117.2 (d, J ˆ 22.1 Hz), 94.7, 69.3, 67.1, 65.0, 60.2, 58.2, 57.8; IR
Preparation of lactol 12: A 40% aqueous glyoxal solution (0.46 mL,
2 equiv) were added to a solution of aminodiol (S)-7 (632 mg, 2.00 mmol) in
tert-amylalcohol (10 mL). After
a 1 hour, 4-fluorophenylboronic acid
(335 mg, 2.54 mmol) was added, and the reaction mixture was heated to
408C for 15 hours. The reaction mixture was diluted with cyclohexane
(80 mL) and washed with water (3 Â 30 mL). Drying (MgSO4) and
evaporation gave a glassy solid, consisting of the diastereomeric mixture
6, 11, 12, 13, and 14 (1.06 g), which partially crystallized on standing. After
addition of methylcyclohexane (4 mL) and EtOAc (0.15 mL), the resulting
(KBr): nÄ ˆ 3278, 3059, 1098, 959 cmÀ1
.
Preparation of cis-vinyl ether 4: A slurry of quaternary salt 5 (2.00 g,
2.98 mmol) in ethanol (3 mL) was heated at 408C. Sodium hydroxide (2n,
1.64 mL, 3.28 mmol) was added and heating was continued until dissolution
occurred. After 20 minutes the reaction was seeded with 4 (20 mg). After
40 minutes at 408C precipitation of 4 occurred. The reaction was heated at
758C for a further 4 hours and then allowed to cool. An ethanol/water
mixture (4:3, 7 mL) was added over 45 minutes. After 1 hour the product 4
was isolated by filtration and washed twice with 1:1 ethanol/water. After
drying 1.46 g 4 was obtained as a white powder (90%). Supercritical-fluid
slurry was aged for 18 hours in a 458C bath to give the crystalline cis-lactol
25
12 (775 mg, 95 A%, 86% yield). M.p. 130 1328C; [a] ˆ 1.148 (c ˆ 1.075
589
in MeOH); 1H NMR (600.13 MHz, CD3CN): d ˆ 7.83 (s, 1H), 7.78 (s, 2H),
7.40 (m, 2H), 7.05 (m, 2H), 4.96 (dd, J ˆ 9.8, 3.4 Hz, 1H), 4.80 (dd, J ˆ 7.9,
1.9Hz, 1H), 4.29(d, J ˆ 7.9Hz, 1H), 4.19(td, J ˆ 11.7, 2.6 Hz, 1H), 3.87
(brs, 1H), 3.63 (m, 1H), 3.59(d, J ˆ 1.9Hz, 1H), 3.17 (dt, J ˆ 11.7, 2.6 Hz,
1H), 2.53 (td, J ˆ 11.3, 3.8 Hz, 1H), 2.38 (dd, J ˆ 12.8, 9.8 Hz, 1H), 2.21 (dd,
J ˆ 12.8, 3.4 Hz, 1H); 13C NMR (150.90 MHz, CD3CN; selected data): d ˆ
chromatographic analysis (OD column, 3% MeOH isocratic) indicated
25
ꢁ99% ee of the desired enantiomer. M.p. 101 1038C; [a] ˆ 114.18
589
94.0, 70.5, 68.8, 62.4, 59.8, 52.2; IR (KBr): nÄ ˆ 3395, 1029, 800 cmÀ1
;
(c ˆ 1.18 in MeOH); IR (KBr): nÄ ˆ 3028, 2882, 1466, 753 cmÀ1; elemental
analysis calcd (%) for C27H22F7NO2: C 61.72, H 4.22, F 25.31, N 2.67; found:
C 61.22, H 4.13, F 25.09, N 2.61.
elemental analysis calcd (%) for C20H18F7NO3: C 52.99, H 4.00, F 29.33, N
3.09; found: C 52.62, H 3.91, F 29.25, N 3.03.
Preparation of 6 ¥ HCl salt: A solution of the cis-lactol 12 (705 mg,
1.57 mmol) in EtOAc (10 mL) was heated to reflux. After cooling to room
temperature, the solution was saturated with HCl gas. To the clear solution
of the salt, methylcyclohexane (10 mL) was added. The slow crystallization
was allowed to take place over 18 hours, hen the slurry was concentrated to
[1] U. S. von Euler, J. H. Gaddum, J. Physiol. (London) 1931, 72, 74 78.
[2] a) R. M. Snider, J. W. Constantine, J. A. Lowe, K. P. Longo, W. S.
Lebel, H. A. Woody, S. E. Droda, M. C. Desai, F. J. Vinick, R. W.
Spencer, H. J. Hess, Science 1991, 251, 435 437; b) Z. Gao, N. P. Peet,
Curr. Med. Chem. 1999, 6, 375 388.
[3] a) M. S. Kramer, N. Cutler, J. Feighner, R. Shrivastava, J. Carman, J. J.
Sraek, S. A. Reines, G. Liu, D. Snavely, E. Wyatt-Knowles, J. J. Hale,
S. G. Mills, M. MacCoss, C. J. Swain, T. Harrison, R. G. Hill, F. Hefti,
E. M. Scolnick, M. A. Cascieri, G. G. Chicchi, S. Sadowski, A. R.
Williams, L. Hewson, D. Smith, E. J. Carlson, R. J. Hargreaves, N. M.
Rupniak, Science 1998, 281, 1640 1645; b) K. A. Maubach, N. M. J.
Rupniak, M. S. Kramer, R. G. Hill, Curr. Opin. Chem. Biol. 1999, 3,
481 488.
[4] a) J. J. Hale, S. G. Mills, M. MacCoss, P. E. Finke, M. A. Cascieri, S.
Sadowski, E. Ber, G. G. Chicchi, M. Kurtz, J. Metzger, G. Eiermann,
N. N. Tsou, F. D. Tattersall, N. M. Rupniak, A. R. Williams, W.
Rycroft, R. Hargreaves, D. E. MacIntyre, J. Med. Chem. 1998, 41,
4607 4614; b) R. J. Alabaster, A. W. Gibson, S. A. Johnson, J. S.
Edwards, I. F. Cottrell, Tetrahedron: Asymmetry 1997, 8, 447 450.
[5] For condensations of organoboronic acids with imines see: a) N. A.
Petasis, I. A. Zavialov, I. A J. Am. Chem. Soc. 1997, 119, 445 446;
b) N. A. Petasis, I. A. Zavialov, J. Am. Chem. Soc. 1998, 120, 1179 8
11799; c) L. M. Harwood, G. S. Currie, M. G. B. Drew, R. W. A. Luke,
Chem. Commun. 1996, 1953 1954; d) G. S. Currie, M. G. B. Drew,
L. M. Harwood, D. J. Hughes, R. W. A. Richard, R. J. Vickers, Perkin I
2000, 2982 2990; for a related strategy, see: e) C. Agami, F. Couty, M.
Poursoulis, J. Vaissermann, Tetrahedron 1992, 48, 431.
[6] a) B. S. Green, R. Arad-Yellin, M. D. Cohen, Top. Stereochem. 1986,
16, 131 218; b) E. Vedejs, R. W. Chapman, S. Lin, M. M¸ller, D. R.
Powell, J. Am. Chem. Soc. 2000, 122, 3047 3052; c) Y. J. Shi, K. M.
Wells, P. J. Pye, W. B. Choi, H. R. O. Churchill, J. E. Lynch, A.
Maliakal, J. W. Sager, K. Rossen, R. P. Volante, P. J. Reider, Tetrahe-
dron, 1999, 55, 909 918; d) R. S. Ward, Tetrahedron Asymmetry 1995,
about half its volume and filtered to give the desired 6 ¥ HCl salt (747 mg,
25
94 A%, 98% yield). M.p. 176 1788C; [a] ˆ 778 (c ˆ 4.7 in MeOH);
589
1H NMR (600.13 MHz, CD3CN): d ˆ 12.01 (brs, 1H), 7.89(s, 1H), 7.86 (s,
2H), 7.80 (brm, 2H), 7.21(m, 2H), 6.26 (br, 1H), 5.39(d, J ˆ 10.2 Hz, 1H),
5.26 (d, J ˆ 7.9Hz, 1H), 5.08 (br, 1H), 4.63 (t, J ˆ 12.5 Hz, 1H), 4.21 (dd,
J ˆ 13.2, 2.6 Hz, 1H), 3.98 (d, J ˆ 12.5 Hz, 1H), 3.89(d, J ˆ 7.9Hz, 1H),
3.32 (m, 1H), 3.12 (d, J ˆ 13.2 Hz, 1H), 2.98 (dd, J ˆ 13.2, 10.2 Hz, 1H);
13C NMR (150.90 MHz, CD3CN): d ˆ 164.4 (d, J ˆ 247.8 Hz), 143.5, 133.2
(br), 132.0 (q, J ˆ 33.6 Hz), 127.8 (q, J ˆ 3.1 Hz), 124.5 (q, J ˆ 271.6 Hz),
122.8 (m), 117.0 (d, J ˆ 22.0 Hz), 95.4, 72.6, 66.1, 64.3, 61.8, 53.4; IR (KBr):
nÄ ˆ 3278, 3059, 1098, 959 cmÀ1
; elemental analysis calcd (%) for
C20H19ClF7NO3: C 49.04, H 3.91, Cl 7.24, F 27.15, N 2.86; found: C 48.65,
H 3.77, Cl 7.36, F 26.9, N 2.81.
Preparation of bicyclic acetal 15: A slurry of lactol HCl salt 6 (4.25 g,
8.68 mmol) in ethyl acetate (50 mL) was washed with dilute aqueous
potassium carbonate (0.7 g in 70 mL water). The organic layer was further
washed with water and dried over MgSO4. The solvent was switched into
THF (40 mL), and the solution cooled to À308C. Tributylphosphine
(2.65 mL, 10.4 mmol) was added followed by the addition of DIAD
(2.0 mL, 9.98 mmol) over 30 minutes, while a temperature below À258C
was maintained. After the addition, the reaction was allowed to warm to
158C over 3 hours, and the solvent was removed in vacuo. The residue was
filtered through a short silica gel column (20% ethylacetate in hexane) and
recrystallized from EtOH/water (1:1) to afford bicyclic acetal 15 as a white
25
solid (3.23 g, 7.43 mmol, 86% yield). M.p. 99 1018C; [a] ˆ 23.58 (c ˆ
589
1
1.36 in MeOH); H NMR (500.13 MHz, CDCl3): d ˆ 7.72 (s, 1H), 7.58 (m,
4H), 7.11 (m, 2H), 5.61 (s, 1H), 5.58 (dd, J ˆ 10.3, 6.4 Hz, 1H), 4.46 (ddd,
J ˆ 11.9, 10.0, 6.0 Hz, 1H), 4.22 (s, 1H), 3.98 (ddd, J ˆ 11.9, 7.0, 3.4 Hz, 1H),
3.91 (ddd, J ˆ 13.9, 10.0, 7.0 Hz, 1H), 3.47 (ddd, J ˆ 13.9, 6.0, 3.4 Hz, 1H),
2.90 (m, 2H); 13C NMR (125.77 MHz, CDCl3): d ˆ 161.1 (d, J ˆ 247.2 Hz),
Chem. Eur. J. 2002, 8, No. 6
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0806-1375 $ 17.50+.50/0
1375

P. J. Pye, K. Rossen et al.
FULL PAPER
6, 1475 1490; e) L. J. Silverberg, S. Kelly, P. Vemishetti, D. H.
Vipond, F. S. Gibson, B. Harrison, R. Spector, J. L. Dillon, Org. Lett.
2000, 2, 3281 3283.
[8] R. J. Davey, N. Blagden, G. D. Potts, R. Docherty, J. Am. Chem. Soc.
1997, 119, 1767 1772.
[9] a) S. S. Bhagwat, P. R. Hamann, W. C. Still, J. Am. Chem. Soc. 1985,
107, 6372 6376; b) Reviews: D. L. Hughes, Org. React. 1992, 42, 335
656; D. L. Hughes, Org. Prep. Proced. Int. 1996, 28, 127 164.
[10] Products from alternative fragmentation pathways were not observed.
[7] a) E. L. Eliel, S. H. Wilen, L. N. Mander, Stereochemistry of Organic
Compounds, Wiley, New York, 1994, p. 364; b) J. Jaques, A. Collet,
S. H. Wilen, Enantiomers, Racemates and Resolutions, Krieger,
Malabar, FL, 1994.
Received: January 4, 2002 [F3779]
1376
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0806-1376 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 6