A stereoselective synthesis of BMS-262084, an azetidinone-based tryptase inhibitor

Article abstract of DOI:10.1021/jo010757o

A highly stereoselective synthesis of the novel tryptase inhibitor BMS-262084 was developed. Key to this synthesis was the discovery and development of a highly diastereoselective demethoxycarbonylation of diester 12 to form the trans-azetidinone 13. BMS-262084 was prepared in 10 steps from D-ornithine in 30% overall yield.

Full text of DOI:10.1021/jo010757o

J . Org. Chem. 2002, 67, 3595-3600
3595
A Ster eoselective Syn th esis of BMS-262084, a n Azetid in on e-Ba sed
Tr yp ta se In h ibitor
Xinhua Qian,* Bin Zheng, Brian Burke, Manohar T. Saindane, and David R. Kronenthal
Process Research and Development, Bristol-Myers Squibb Pharmaceutical Research Institute,
1 Squibb Drive, P.O. Box 191, New Brunswick, New J ersey 08903-0191
xinhua.qian@bms.com
Received J uly 26, 2001
A highly stereoselective synthesis of the novel tryptase inhibitor BMS-262084 was developed. Key
to this synthesis was the discovery and development of a highly diastereoselective demethoxycar-
bonylation of diester 12 to form the trans-azetidinone 13. BMS-262084 was prepared in 10 steps
from D-ornithine in 30% overall yield.
Sch em e 1
In tr od u ction
Utilization of tryptase inhibitors for the treatment of
asthma has been the focus of recent attention.1 BMS-
262084 (1) is a sub-nanomolar inhibitor of tryptase and
Sch em e 2
suppresses induced inflammation in animal lungs. Early
routes (Scheme 1) to this development candidate utilized
the alkylation of the dianion derived from the known,
homochiral, azetidinone acid 2 to establish the requisite
trans-stereochemistry.2,3 As scale-up of this step proved
difficult to optimize, and faced with the need for large
quantities of material for development purposes, we
sought to identify and develop a practical synthesis which
would not proceed through dianion intermediates. In this
paper, we describe a new approach to 1 starting from
D-ornithine, the most salient feature of which is the
highly stereoselective demethoxycarbonylation of diester
12 to afford the trans-azetidinone 13. This new synthesis
proceeds in 30% overall yield.
of diastereoselectivity would have to be achieved in the
dealkoxycarbonylation step leading to trans-azetidinone
13 while simultaneously preserving the integrity of the
existing stereogenic center. While dealkoxycarbonylation
of azetidinone C4-geminal esters is well documented,4c-e,5
a highly diastereoselective process providing >95% trans-
selectivity was unknown. Moreover, little useful informa-
tion has been reported regarding the propensity for
racemization during the formation of azetidinones from
R-bromoacids and aminomalonates. By overcoming these
Resu lts a n d Discu ssion
As outlined in Scheme 2, we envisioned a synthesis of
BMS-262084 arising from D-ornithine and an appropri-
ately protected aminomalonate, and recognized that, for
this approach to be successful, an unprecedented degree
(4) (a) For a review see: Georg, G. I. The Organic Chemistry of
â-Lactams; VCH: New York, 1993. (b) Sheehan, J . C.; Bose, A. K. J .
Am. Chem. Soc. 1950, 72, 5158. Sheehan, J . C.; Bose, A. K. J . Am.
Chem. Soc. 1951, 73, 1761. (c) Shiozaki, M.; Ishida, N.; Hiraoka, T.;
Yanagisawa, H. Tetrahedron Lett. 1981, 22, 5205. Shiozaki, M.; Ishida,
N.; Maruyama, H.; Hiraoka, T. Tetrahedron 1983, 39, 2399. Shiozaki,
M.; Masuko, H. Heterocycles 1984, 22, 1725. (d) Simig, G.; Fetter, J .;
Hornya´k, G.; Zauer, K.; Doleschall, G.; Lempert, K.; Nyitrai, J .;
Gombos, Z.; Gizur, T.; Barta-Szalai, G.; Kajta´r-Peredy, M. Acta Chim.
Hung. 1985, 119, 17. Bertha, F.; Lempert, K.; Kajta´r-Peredy, M. Acta
Chim. Hung. 1985, 120, 111. (e) Madara´sz, Z.; Ne´meth, I.; Toscano,
P.; Welch, J .; Nyitrai, J . Tetrahedron Lett. 1995, 36, 45.
* To whom correspondence should be addressed.
(1) (a) Ono, S.; Kuwahara, S.; Takeuchi, M.; Sakashita, H.; Naito,
Y.; Knodo, T. Bioorg. Med. Chem. Lett. 1999, 9, 3285. (b) Wright, C.
D.; Havill, A. M.; Middleton, S. C.; Kashem, M. A.; Dripps, D. J .;
Abraham, W. M.; Thomson, D. S.; Burgess, L. E. Biochem. Pharmacol.
1999, 58, 1989. (c) Rice, K. D.; Spencer, J . Expert Opin. Ther. Pat. 1999,
9, 1537. Rice, K. D.; Sprengeler, P. A. Curr. Opin. Drug Discovery Dev.
1999, 2, 463.
(2) Baldwin, J . E.; Adlington, R. M.; Gollins, D. W.; Schofield, C. J .
Tetrahedron 1990, 46, 4733.
(3) Bisacchi, G.; Slusarchyk, W. A.; Treuner, U.; Sutton, J . C.;
Zahler, R.; Seiler, S.; Kronenthal, D. R.; Randazzo, M. E.; Xu, Z.; Shi,
Z.; Schwinden, M. D. Pat. Appl. WO 99US13811 99/67215.
(5) (a) For reviews see: Krapcho, A. P. Synthesis 1982, 821. Krapcho,
A. P. Synthesis 1982, 893. (b) Simig, G.; Doleschall, G.; Hornya´k, G.
Tetrahedron 1985, 41, 479. Fetter, J .; Lempert, K.; Kajta´r-Peredy, M.;
Simig, G.; Hornya´k, G.; Horva´th, Z. J . Chem. Res., Synop. 1985, 368;
J . Chem. Res., Miniprint 1985, 3901.
10.1021/jo010757o CCC: $22.00 © 2002 American Chemical Society
Published on Web 04/27/2002

3596 J . Org. Chem., Vol. 67, No. 11, 2002
Qian et al.
Sch em e 3
F igu r e 1. Effect of concentration on the demethoxycarbony-
lation of the trifluoroacetamide 12.
Sch em e 4
challenges, a highly stereoselective and practical method
to prepare the necessary 3,4-trans-azetidinone system
emerged.
Selective protection of the δN atom in D-ornithine
hydrochloride (5) was achieved in 76% yield using ethyl
trifluoroacetate under basic, aqueous conditions, and
producing trifluoroacetamide 6 along with trace amounts
of the bisprotected product6 (Scheme 3). Diazotization-
bromination7 of 6 gave the R-bromoacid 7 in 74% yield
with >99% retention of configuration by chiral HPLC
analysis. Reductive alkylation of dimethyl aminoma-
lonate (9) with 2,4-dimethoxybenzaldehyde (8) under a
hydrogen atmosphere gave the protected malonate 10 in
89% yield after crystallization. This one-pot process is
an improvement over the procedure involving imine
formation followed by borohydride reduction.5b,8
In an initial attempt to carry out a one-pot preparation
of the azetidinone 12, the R-bromoacid 7 was converted
to the acid chloride with oxalyl chloride followed by
coupling with the protected malonate 109 in the presence
of excess triethylamine (Scheme 4). The excess base
promoted an intramolecular cyclization of the resulting
amide 11 at ambient temperature with inversion of
configuration to afford azetidinone 12 in 95% yield.
However, under these conditions, about 13% racemiza-
tion occurred, presumably via the corresponding ketene
derived from the acid chloride. While employment of the
more hindered base diisopropylethylamine, at 0-5 °C,
resulted in only 5% racemization, a significant amount
of elimination was observed during the cyclization.
Employment of less basic amines, such as N-methylpi-
peridine, provided 12 with ee > 96%, but the reaction
was extremely slow and often incomplete. Consequently,
a two-base, one-pot protocol was developed. Coupling of
the acid chloride of 7 and malonate 10 was conducted in
the presence of K2HPO4 in dichloromethane-water to
generate amide 11. Addition of triethylamine (3 molar
equiv) promoted cyclization and gave 12 in 95% yield with
no detectable racemization by chiral HPLC analysis.
The dealkoxycarbonylation of 12 was next studied in
some detail. Dealkoxycarbonylation of similar azetidino-
nes using Krapcho’s conditions (NaCl, DMSO, H2O) is
generally a nonselective process.4d,5a To our pleasant
surprise, when the initial demethoxycarbonylation of 12
was conducted in DMSO-H2O-LiCl at 135 °C (Scheme
4), the trans/cis product ratio (∼20:1) was significantly
higher than expected on the basis of literature prece-
dent.4,5 We speculated that the relatively acidic trifluo-
roacetamide proton in compound 12 might be serving as
an intramolecular proton source, and effectively compet-
ing with H2O to selectively protonate the intermediate
C4-enolate from the R-face. To probe the source of the
high selectivity, the trans/cis product ratio was examined
as a function of the starting reactant concentration. As
indicated by the data in the Figure 1, dilution of the
reaction medium (H2O held constant) led to progressively
higher trans/cis product ratios, supporting an intramo-
lecular proton transfer. At an initial reactant concentra-
tion of 0.08 M, the trans/cis product ratio reached 50:1.
A study of the trans/cis product ratio as a function of
δ
(6) (a) Schallenberg, E. E.; Calvin, M. J . Am. Chem. Soc. 1955, 77,
2779. (b) Curphey, T. J . J . Org. Chem. 1979, 44, 2805. (c) Blacklock,
T. J .; Shuman, R. F.; Butcher, J . W.; Shearin, W. E., J r.; Budavari, J .;
Grenda, V. J . J . Org. Chem. 1988, 53, 836.
(7) Goto, K.; Waki, M.; Mitsuyasu, N.; Kitajima, Y.; Izumiya, N. Bull.
Chem. Soc. J pn. 1982, 55, 261.
the N-protecting group also supported the intramolecu-
lar proton-transfer mechanism. Protection as a phthal-
imido group should effectively shut down the possibility
of intramolecular proton transfer and, in fact, resulted
in a 1.25:1 trans/cis product mixture (Table 1). Employ-
ment of a δN-Cbz protecting group10 resulted in a 25:1
(8) Husinec, S.; J uranic, I.; Llobera, A.; Porter, A. E. A. Synthesis
1988, 721.

A Stereoselective Synthesis of BMS-262084
J . Org. Chem., Vol. 67, No. 11, 2002 3597
Ta ble 1. Effects of P r otectin g Gr ou p s on th e
Dem eth oxyca r bon yla tion Rea ction a
Sch em e 5
protecting groups
substrate
concn (M)
trans/cis
product ratio
entry
R1
R2
a
b
c
d
e
DMB
H
DMB
DMB
DMB
CF3CO
CF3CO
phthalyl
Cbz11
0.15
0.15
0.09
0.09
0.09
28:1
9:1
1.25:1
25:1
CF3CO
46:1
a
Sch em e 6
All studies were carried out with 2 equiv of H2O.
trans/cis product ratio, whereas the more electron-
withdrawing trifluoroacetamide protecting group pro-
vided a 46:1 trans/cis mixture under identical conditions.
On the other hand, without protection N1 provided a
proton11 that would carry out a nonstereoselective in-
tramolecular proton transfer which competed with the
desired intramolecular δN-proton transfer and resulted
in a 9:1 trans/cis product ratio. These observations clearly
indicate that the δN-proton is crucial to the diastereo-
selectivity of the demethoxycarbonylation reaction de-
scribed here.
We next studied the effect of added water on the
reaction diastereoselectivity. When the reaction was
carried out in DMF12 in the presence of 1 equiv of water
relative to the substrate 12 (Scheme 4), the trans/cis
product ratio reached 36:1 at a 0.2 M reaction concentra-
tion of 12, providing azetidinone 13 in 93% yield.
However, the selectivity dropped to 17:1 when the
amount of water was increased to 4 equiv. The erosion
of the trans/cis selectivity observed with increasing levels
of water again supports the intramolecular proton-
transfer hypothesis.
With compound 13 in hand, we next addressed removal
of the dimethoxybenzyl and trifluoroacetamide protecting
groups and guanylation of the side chain nitrogen. The
order of the deprotection operations proved to be critical.
Initially, the N1-DMB group in 13 was removed (Scheme
5) under oxidative conditions (K2S2O8/K2HSO4) to give
azetidinone 14 in 83% isolated yield. However, difficulties
δ
were encountered in the subsequent removal of the N-
trifluoroacetyl group in 13. Using aqueous alkaline
conditions, cleavage occurred along with concomitant
hydrolysis of the methyl ester.13 However, under these
conditions, a substantial portion of the product under-
went cyclization to the δ-lactam 16. This intramolecular
cyclization could not be suppressed, even using mildly
basic (aqueous carbonate) conditions, and ∼1:1 mixtures
of 15 and 16 were generally observed.
Interestingly, when 13 was first treated under aqueous
alkaline conditions to effect ester hydrolysis and depro-
tection of the δN-trifluoroacetyl group, lactamization was
not observed as a competitive side reaction (Scheme 6).
The crude amino acid was isolated as the acetic acid salt
17.14 Subsequent reaction with N,N′-bis-Cbz-1-guanylpyra-
zole (18)15 in the presence of TEA gave guanidine 19.
Reesterification16 of 19 provided benzyl ester 20 in 83%
(9) It is worth mentioning that without protection of the nitrogen
in the malonate 10, the subsequent intramolecular cyclization step did
not proceed.
(10) In our initial studies, the δN atom in D-ornithine hydrochloride
(5) was protected with a Cbz group. However, during the diazotization-
bromination of the RN atom, nitrosation at the δN atom occurred. This
observation offered an additional reason for using a more electron-
withdrawing protecting group such as trifluoroacetamide at the δN
atom.
(14) Guanidine displacement with 18 is significantly faster when
17 is isolated and the reaction is carried out in the absence of water.
(15) Bernatowicz, M. S.; Wu, Y.; Matsueda, G. R. Tetrahedron Lett.
1993, 34, 3389. Wu, Y.; Matsueda, G. R.; Bernatowicz, M. S. Synth.
Commun. 1993, 23, 3055.
(16) Earlier studies had indicated that the subsequent acylation of
the azetidinone nitrogen was not successful in the presence of a C4-
carboxylic acid.
(11) For the method of preparing a -N1-nonprotected counterpart
of compound 9 see ref 4d, Simig et al.
(12) Screening experiments had indicated DMF to be a more suitable
solvent than DMSO.
(13) Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic
Synthesis, 2nd ed.; J ohn Wiley & Sons: New York, 1991.

3598 J . Org. Chem., Vol. 67, No. 11, 2002
Qian et al.
overall yield from 13. Oxidative cleavage4c,5b,17 of the N1-
DMB protecting group followed by chromatographic
purification gave the deprotected azetidinone 21 in 76%
isolated yield. Acylation of the N1-position in 21 with
carbamoyl chloride 223,18 in the presence of triethylamine
and catalytic amounts of DMAP gave the advanced
intermediate 23 in almost quantitative yield. Finally,
hydrogenolysis under neutral conditions removed the
benzyl and Cbz protecting groups and gave 1 (BMS-
262084) in quantitative yield and with ee > 99%.
In conclusion, we have developed a highly stereoselec-
tive synthesis of the tryptase inhibitor 1. Key to this was
the development of a highly trans-selective dealkoxycar-
bonylation reaction.
mL, 10 µL injection, 210 nm; RT ) 26.7 min (enantiomer, RT
) 31.1 min).
N-DMB Am in om a lon a te 10. Deoxygenated methanol (1.0
L) was added to a mixture of Pt/C (10%, 5.0 g), dimethyl
aminomalonate hydrochloride (75.0 g, 410 mmol), and 2,4-
dimethoxybenzaldehyde (68.0 g, 410 mmol) under an atmo-
sphere of N2. Triethylamine (40 mL, 287 mmol) was then
added to the resulting suspension. The reaction mixture was
sparged with H2 for 4.5 h followed by N2 for 15 min and was
filtered through a Celite pad. The filtrate was concentrated
to ca. 250 mL. Dichloromethane (500 mL) was then added to
the concentrated residue, and the resulting solution was
washed with satd NaHCO3 (250 mL) and water (250 mL). After
phase splitting, the combined aqueous layers were extracted
with dichloromethane (100 mL). The dichloromethane layers
were combined, dried over anhyd Na2SO4, filtered, and con-
centrated to give a yellow solid which was subsequently
crystallized from 300 mL of 2-propanol to afford 10 as a yellow
solid (108.5 g, 89% yield). Mp: 62-64 °C. 1H NMR: δ 7.11 (d,
J ) 8.9, 1H), 6.44-6.39 (m, 2H), 4.08 (d, J ) 1.7 Hz, 1H), 3.81
(s, 3H), 3.79 (s, 3H), 3.75 (d, J ) 1.5 Hz). 3.71 (s, 6H), 3.02 (s
br, 1H). 13C NMR: δ 169.0, 160.3, 158.6, 130.6, 119.2, 103.7,
98.3, 64.0, 55.2, 55.1, 52.6, 46.8. IR (KBr): 3350, 3012, 2950,
2838, 2361, 2054, 1757, 1726, 1608, 1506, 1440, 1296, 1209,
Exp er im en ta l Section
1H and 13C NMR spectra (300 MHz) were recorded in CDCl3
and are referenced to TMS unless otherwise noted. Melting
points were taken on a Thomas-Hoover capillary melting point
apparatus and are not corrected. Chiral HPLC analyses were
carried out with a Shimadzu LC-10AD HPLC system or a
Waters Alliance 2690 module with a 996 PDA detector and
Millennium 2010 data system.
1035, 984, 938, 846, 964, 723 cm-1. HRMS: calcd for C14H20
-
NO6 (M+ + H) 298.1291, found 298.1294.
Am id e 11 a n d Cycliza tion to Azetid in on e 12. To a
solution of bromoacid 7 (53.4 g, 183 mmol) in dichloromethane
(214 mL) were added oxalyl chloride (18.3 mL, 210 mmol) and
DMF (5.67 mL, 73.2 mmol) at 0 °C. The mixture was warmed
to 23 °C, stirred for 1 h, and then recooled to 0 °C. The
resulting cold solution was slowly added to a heterogeneous
mixture of 33% K2HPO4 (295 mL) [aq, 20 wt %, pH 9.17 (0 °C)
or 9.26 (at ambient temperature)] and a solution of 10 (51.6
g, 174 mmol) in dichloromethane (214 mL) at 0-5 °C. Upon
completion of the addition, the reaction mixture was stirred
for 30 min, followed by addition of triethylamine (76.4 mL,
549 mmol) at 0-10 °C. The reaction mixture was then warmed
to 23 °C and stirred for 5 h. The organic layer was separated,
washed with 3.0 N HCl (200 mL) and satd NaHCO3 (200 mL),
dried over anhyd Na2SO4, filtered, and concentrated to provide
Am in o Acid 6. To a solution of D-ornithine hydrochloride
(50.6 g, 300 mmol) in 3.0 N NaOH (aq, 100 mL, 300 mmol)
was added ethyl trifluoroacetate (35.8 mL, 300 mmol) at 23
°C. A white precipitate was formed within 20 min. The
suspension was stirred overnight and filtered to provide 6 (52.0
g, 76%) as a white solid after drying in vacuo. Mp: 212.0 °C
1
dec, lit.6a (DL) 228-232 °C dec. H NMR (D2O): δ 3.56 (dd, J
) 11.8, 5.7 Hz, 1H), 3.17 (d, J ) 6.84 Hz, 1H), 2.85 (d, J )
7.53 Hz, 1H), 1.78-1.42 (m, 4H). 13C NMR (D2O, internal
standard CH3CN, δ 1.30, CH3-): δ 174.7, 159.5 (q, J ) 37.1
Hz), 116.3 (q, J ) 283 Hz). 54.7, 39.5, 28.0, 23.9. IR (KBr):
3298, 2940, 2084, 1711, 1583, 1486, 1445, 1414, 1219, 1055,
886, 763, 727 cm-1. HRMS: calcd for C7H12F3N2O3 (M+ + H)
229.0800, found 229.0820. Chiral HPLC: ee > 99%; Chiralpak
WH column, 250 × 4.6 mm, 10 µm; mobile phase 0.5 mM
CuSO4 (aq); isocratic at 40 °C, 1.0 mL/min, 210 nm; concentra-
tion 0.5 mg/mL, 10 µL injection; RT ) 20.7 min (enantiomer,
RT ) 24.2 min).
1
azetidinone 12 as a yellow oil (81.1 g, 95%). H NMR: δ 7.41
(s br, 1H), 7.06 (d, J ) 8.8 Hz, 1H), 6.43-6.34 (m, 2H), 4.68
(d, J ) 15.0 Hz, 1H), 4.39 (d, J ) 15.0 Hz, 1H), 3.89-3.62 (m,
10 H), 3.46-3.32 (m, 5H), 1.80-1.54 (m, 4H). 13C NMR: δ
168.6, 167.4, 167.2, 160.8, 158.6, 157.4 (q, J ) 36.4 Hz), 131.0,
115.9 (q, J ) 286 Hz), 115.0, 103.8, 98.1, 67.4, 57.5, 55.4, 55.3,
53.0, 52.8, 40.4, 39.1, 26.4, 22.8. IR (film): 3309, 3099, 2953,
Br om oa cid 7. Potassium bromide (219.9 g, 1.85 mol) was
added to a solution of 6 (52.6 g, 231 mmol) in 462 mL of 4.0 N
H2SO4 at 0 °C, followed by the slow addition of a solution of
sodium nitrite (19.9 g, 289 mmol) in water (40 mL) at 0-8 °C.
The mixture was stirred at 0-5 °C for 4 h, and N2 was bubbled
into the brown reaction mixture to discharge the color (about
15 min). The reaction was extracted with MTBE (600 mL),
washed with 15% brine (500 mL), 10% brine (500 mL)
containing 2.5 g of Na2S2O3, and 10% brine (300 mL), dried
over anhyd Na2SO4, filtered, and concentrated to give bro-
2838, 1757, 1608, 1501, 1453, 1301, 1194, 1041, 835, 718 cm-1
.
HRMS: calcd for C21H25F3N2O8 (M+) 490.1563, found 490.1556.
Chiral HPLC: ee 98%; Chiralpak AD column, 250 × 4.6 mm,
10 µm; mobile phase hexane/2-propanol (95:5, v/v); isocratic
at ambient temperature, 1.0 mL/min, 210 nm; concentration
0.5 mg/mL, 10 µL injection; RT ) 34.2 min (enantiomer, RT
26.5 min).
)
1
moacid 7 as a colorless oil (49.9 g, 74% yield). H NMR (CD3-
Azetid in on e 13. A mixture of LiCl (8.25 g, 194 mmol) and
DMF (310 mL, water content 0.012%) was heated to 130 °C,
followed by the addition of a solution of azetidinone 12 (38.1
g, 77.8 mmol) in DMF (35 mL, water content 0.012%) and
water (1.40 mL, 77.8 mmol) over a period of 2 h at 128-132
°C. Upon completion of the addition, the mixture was stirred
at 130 °C for 1.5 h and then evaporated under reduced
pressure at 60 °C to about 200 mL. The residue was diluted
with MTBE (200 mL) and washed with 3.0 N HCl (200 mL)
and satd NaHCO3 (200 mL). Activated carbon (1.0 g) was
added to the ether layer, and the resulting suspension was
filtered through a Celite pad. The filtrate was dried over anhyd
Na2SO4, filtered, and concentrated to give the azetidinone 13
as a yellow oil (31.3 g, 93%; trans:cis ) 36.3:1). 1H NMR: δ
7.63 (s br, 1H), 7.05 (d, J ) 8.9 Hz, 1H), 6.49-6.40 (m, 2H),
4.57 (d, J ) 14.3 Hz, 1H), 4.11 (d, J ) 14.3 Hz, 1H), 3.77 (s,
3H), 3.73 (s, 3H), 3.71 (s, 3H), 3.54 (d, J ) 2.3 Hz, 1H), 3.40-
3.29 (m, 2H), 3.12-3.04 (m, 1H), 1.81-1.60 (m, 4H). 13C
NMR: δ 170.8, 168.5, 161.0, 158.5, 157.3 (q, J ) 36.7 Hz),
131.3, 115.8 (q, J ) 286 Hz), 115.0, 104.2, 98.3, 56.6, 55.3,
CN): δ 7.61 (s, br, 1H), 4.33 (dd, J ) 8.8, 6.5 Hz, 1H), 3.29
(dd, J ) 13.1, 6.6 Hz, 2H), 2.14-1.84 (m, 2H), 1.79-1.60 (m,
2H). 13C NMR (CD3CN): δ 171.1, 157.9 (q, J ) 36.2 Hz), 117.1
(q, J ) 285 Hz), 46.5, 39.5, 32.6, 27.1. IR (film): 3309, 2961,
2551, 1931, 1742, 1445, 1178, 1066, 1020, 840, 712, 651 cm-1
.
HRMS: calcd for C7H9BrF3NO3 (M+) 290.9718, found 290.9732.
Chiral HPLC: ee 98%; Chiralpak AS column, 250 × 4.6 mm,
10 µm; mobile phase hexane/2-propanol (96:4, v/v); isocratic
at ambient temperature, 1.0 mL/min; concentration 0.5 mg/
(17) (a) For a possible mechanism see: Needles, H. L.; Whitefield,
R. E. J . Org. Chem. 1964, 29, 3632. (b) Finkelstein, J .; Holden, K. G.;
Perchonock, C. D. Tetrahedron Lett. 1978, 19, 1629. (c) Heck, J . V.;
Szymonifka, M. J .; Christensen, B. G. Tetrahedron Lett. 1982, 23, 1519.
(d) Hakimelahi, G. H.; Khalafi-Nezhad, A. Helv. Chim. Acta 1984, 67,
18. (e) Kayakiri, H.; Oku, T.; Hashimoto, M. Chem. Pharm. Bull. 1991,
39, 1397. (f) Fetter, J .; Lempert, K.; Kajta´r-Peredy, M.; Simig, G. J .
Chem. Soc., Perkin Trans. 1 1988, 1135. Fetter, J .; Bertha, F.; Czuppon,
T.; Kajta´r-Peredy, M.; Lempert, K. J . Chem. Res., Miniprint 1995, 2738.
(18) McGhee, W. D.; Pan, Y.; Talley, J . J . Tetrahedron Lett. 1994,
35, 839.

A Stereoselective Synthesis of BMS-262084
J . Org. Chem., Vol. 67, No. 11, 2002 3599
55.1, 54.7, 52.2, 40.3, 39.1, 26.2, 25.5. IR (film): 3324, 3083,
2950, 1761, 1619, 1516, 1296, 1209, 1148, 1040, 927, 840, 715
cm-1. HRMS: calcd for C19H23F3N2O6 (M+) 432.1508, found
432.1495. Anal. Calcd for C19H23F3N2O6: C, 52.78; H, 5.36; N,
6.48. Found: C, 52.71; H, 5.44; N, 6.14. Chiral HPLC: ee 98%;
Chiralpak AD column, 250 × 4.6 mm, 10 µm; mobile phase
hexane/2-propanol (95:5, v/v); isocratic at ambient tempera-
ture, 1.0 mL/min, 210 nm; concentration 0.5 mg/mL, 10 µL
injection; RT ) 24.8 min (enantiomer, RT ) 20.9 min).
in three steps) of 20 as a white foam. [R]25 ) +25.9 (c 0.97,
D
CHCl3). 1H NMR: δ 11.72 (s, 1H), 8.29 (t, J ) 5.3 Hz, 1H),
7.39-7.25 (m, 15H), 7.03 (m, 1H), 6.39-6.36 (m, 2H), 5.16 (s,
2H), 5.14 (s, 2 H), 5.11 (s, 2H), 4.61 (d, J ) 14.2 Hz, 1H), 4.10
(d, J ) 14.1 Hz, 1H), 3.77 (s, 3H), 3.63 (s, 3H), 3.57 (d, J ) 2.3
Hz, 1H), 3.40 (dd, J ) 12.1, 6.1 Hz, 2H), 3.11 (m, 1H), 1.82-
1.60 (m, 4H). 13C NMR: δ 170.3, 167.8, 163.6, 160.9, 158.6,
155.9, 153.7, 136.7, 135.1, 134.5, 131.4, 128.7, 128.61, 128.58,
128.48, 128.4, 128.3, 128.0, 127.8, 115.3, 104.1, 98.3, 104.1,
98.3, 68.1, 67.1, 66.9, 56.6, 55.3, 55.2, 55.0, 40.4, 40.2, 26.3,
25.5. HRMS: calcd for C40H43N4O9 (M+ + H) 723.3030, found
723.3068. IR (KBr): 3337, 2941, 1756, 1639, 1508, 1383, 1264,
1209, 1045, 744, 698 cm-1. Anal. Calcd for C40H42N4O9: C,
66.47, H, 5.86, N, 7.75. Found C, 66.21, H, 5.72, N, 7.93.
Gu a n id in e Azetid in on e 21. To a 1-L, round-bottom flask
equipped with a condenser, a thermocouple, and a mechanical
stirrer were added 162 mL of water and 223 mL of CH3CN.
The solution was heated to reflux followed by the addition of
a solution of 20 (9.31 g, 12.88 mmol, in 20 mL of CH3CN). The
mixture was heated to reflux again followed by adding K2S2O8-
(s) (30.28 g, 111 mmol) and K2HPO4(s) (9.02 g, 51.2 mmol).
After 15 min, the heat was removed, and the reaction mixture
was cooled to 0 °C with vigorous stirring for 30 min and
filtered. The filtrate was neutralized with NaHCO3 to pH 7.2,
and the organic layer was separated and evaporated. The
residue was redissolved in EtOAc (350 mL) followed by
washing with satd NaHCO3 (100 mL), water (100 mL), and
brine (100 mL), separated, dried over anhyd Na2SO4, filtered,
and evaporated to give an orange-colored oil. Chromatographic
purification of the crude product with a gradient of 40-60%
EtOAc in hexanes (v/v, 5%/500 mL intervals) gave the product
N1-Dep r otected Azetid in on e 14. To a 2-L, three-neck,
round-bottom flask equipped with a mechanical stirrer, a
thermocouple, and an addition funnel were added 500 mL of
CH3CN and 1244 mL of water. The mixture was heated to
reflux followed by the addition of a solution of 13 (51.5 g, 0.119
mol) in 330 mL of CH3CN. The reaction mixture was heated
to reflux again followed by the addition of K2S2O8 (257.6 g,
0.95 mol) and K2HPO4 (83.0 g, 0.48 mol). The resulting reaction
mixture rapidly turned an orange color and was heated for 38
min with the internal temperature maintained at 91-92 °C
under vigorous stirring. The heat was removed, and the
reaction mixture was allowed to cool at ambient temperature
for 7 min and then cooled in an ice-water bath to <3 °C. All
solids were removed by filtration, and the filtrate was evapo-
rated to remove CH3CN. The aqueous residue was cooled to 0
°C, and the pH was adjusted to 8-9 with NaHCO3. The
resulting mixture was then saturated with solid NaCl and
extracted with ethyl acetate (3 × 400 mL). The combined
organic layers were washed with satd NaHCO3 (2 × 200 mL)
and brine (2 × 200 mL), separated, and dried over anhyd Na2-
SO4 overnight. Filtration and evaporation of the solution
followed by crystallization of the residue from ethyl acetate
and dichloromethane gave 14 as plate-shaped white crystalline
solids (22.8 g, 73% yield). An additional 3.2 g of 14 was
obtained from the mother liquor by chromatography, giving a
21 as a white foam (5.65 g, 76% yield). [R]25 ) +7.94 (c 1.02,
D
CHCl3). 1H NMR: δ 11.73 (s, 1H), 8.34 (t, J ) 5.2 Hz, 1H),
7.39-7.26 (m, 15H), 6.25 (s, 1H), 5.18 (s, 2H), 5.17 (s, 2 H),
5.11 (s, 2H), 4.11 (d, J ) 2.5 Hz, 1H), 3.45 (dd, J ) 12.3, 6.4
Hz, 2H), 3.23 (m, 1H), 1.90-1.70 (m, 4H). 13C NMR: δ 170.6,
168.8, 163.6, 156.0, 153.8, 136.7, 134.9, 134.5, 128.8, 128.6,
128.4, 128.36, 128.1, 127.9, 68.1, 67.3, 67.1, 57.1, 53.6, 40.3,
26.4, 25.6. HRMS: calcd for C31H33N4O7 (M+ + H) 573.2349,
found 573.2384. IR (KBr): 3334, 1767, 1735, 1639, 1426, 1325,
1207, 1047, 743, 697 cm-1. Anal. Calcd for C31H32N4O7: C,
65.02, H, 5.63, N, 9.78. Found C, 64.76, H, 5.41, N, 9.73.
N1-Acyla ted Azetid in on e 23. A solution of 21 (4.88 g, 8.52
mmol) in 80.0 mL of dichloromethane was cooled to 0 °C
followed by addition of the carbamoyl chloride 22 (2.53 g, 10.2
mmol),3,17 triethylamine (1.68 mL, 11.9 mmol), and DMAP (200
mg, 1.6 mmol). The reaction mixture was stirred at 0 °C for
30 min, then warmed to ambient temperature, and stirred for
another 4.65 h. The reaction mixture was diluted with 400
mL of EtOAc and washed with 100 mL of 0.5 N HCl, water
(100 mL), satd NaHCO3 (100 mL), water (100 mL), and brine
(160 mL). The organic layer was separated, dried over anhyd
Na2SO4, filtered, and evaporated to give a white foam.
Chromatographic purification of the crude product with a
gradient of 40-65% EtOAc in hexanes (v/v, 5%/250 mL
total yield of 82%. Mp: 116-117 °C. [R]25 ) +39.5 (c 0.99,
D
1
CHCl3). H NMR: δ 7.10 (s, br, 1H), 6.23 (s, 1H), 3.91 (d, J )
2.34 Hz, 1H), 3.80 (s, 3H), 3.44 (dd, J ) 12.7, 6.3 Hz, 2H), 3.3
(m, 1H), 1.94-1.75 (m, 4 H). 13C NMR (CD3OD): δ 173.6,
172.6, 159.4 (q, J ) 36.7 Hz), 118.0 (q, J ) 286.6 Hz), 58.0,
54.9, 53.3, 40.6, 27.7, 27.2. HRMS: calcd for C10H14F3N2O4 (M+
+ H) 283.0905, found 283.0905. IR (KBr): 3350, 3171, 1787,
1740, 1703, 1560, 1242, 1181, 1155, 662 cm-1. Anal. Calcd for
C
10H13F3N2O4: C, 42.56, H, 4.64, N, 9.93. Found C, 42.64, H,
4.57, N, 9.98.
Ben zyl Ester 20. A solution of 13 (9.55 g, 22.1 mmol) in
90 mL of methanol was cooled to 0 °C followed by addition of
2.0 N NaOH(aq) (34.5 mL, 69.0 mmol). The reaction mixture
was stirred at 0 °C for 20 min, then warmed to ambient
temperature, and stirred for an additional 2 h. Acetic acid (2.63
mL, 2 equiv) was then added to the reaction mixture followed
by evaporation to dryness. The resulting solids 17 were stored
in vacuo overnight and then redissolved in methanol.14 The
clear solution was cooled to 0 °C followed by the addition of
N,N′-bis-Cbz-1-guanylpyrazole15 (18; 8.27 g, 21.9 mmol) and
triethylamine (9.72 mL, 69.0 mmol). The reaction mixture was
stirred at 0 °C for 20 min, warmed to ambient temperature,
and stirred for an additional 6 h. The volatiles were removed
under reduced pressure, and the oily residue was diluted with
400 mL of EtOAc and washed with 0.5 N HCl (2 × 50 mL),
water (80 mL), and brine (100 mL). The organic layer was
separated, dried over anhyd Na2SO4, filtered, and evaporated
to give an off-white foam (19; 13.75 g). After being dried in
vacuo overnight, the crude product was dissolved in dichlo-
romethane followed by the addition of benzyl alcohol (2.41 mL,
23.06 mmol), 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide
hydrochloride (4.83 g, 25.2 mmol), and DMAP (500 mg, 4
mmol). The resulting solution was stirred at ambient temper-
ature for 1.9 h followed by the removal of all volatiles under
reduced pressure. The oily residue was diluted with 400 mL
of MTBE followed by washing with 100 mL of 0.5 N HCl (2 ×
90 mL), water (100 mL), satd NaHCO3 (100 mL), water (100
mL), and brine (120 mL). The organic layer was separated and
evaporated to give a white foam (20; 15.6 g). Chromatographic
purification (40-60% gradient of EtOAc in hexanes, v/v, 5%/
500 mL intervals) of the crude product gave 13.3 g (83% yield
intervals) gave 23 as a white foam (6.42 g, 96% isolated yield).
1
[R]25 ) -32.2 (c 0.99, CHCl3). H NMR: δ 11.73 (s, 1H), 8.32
D
(t, J ) 5.4 Hz, 1H), 7.40-7.26 (m, 15H), 5.239 (d, J ) 12.2 Hz,
1H), 5.17 (s, 2H), 5.14 (d, J ) 12.2 Hz, 1H), 5.12 (s, 2H), 4.38
(d, J ) 3.6 Hz, 1H), 4.29 (s, 1H), 3.59-3.18 (m, 11H), 1.93-
1.72 (m, 4H), 1.35 (s, 9H). 13C NMR: δ 169.1, 165.0, 163.6,
156.6, 156.0, 153.8, 149.7, 136.7, 134.8, 134.5, 128.8, 128.65,
128.61, 128.4, 128.3, 128.2, 128.0, 127.9, 68.2, 67.5, 67.1, 54.4,
52.0, 50.9, 43.5, 40.1, 29.3, 26.3, 25.3. HRMS: calcd for
C
41H50N7O9 (M+ + H) 784.3670, found 784.3667. IR (KBr):
3338, 2962, 1788, 1736, 1640, 1427, 1261, 1205, 1138, 1047,
994, 748, 697 cm-1. Anal. Calcd for C41H49N7O9: C, 62.82, H,
6.30, N, 12.51. Found: C, 62.62, H, 6.25, N, 12.33.
1 (BMS-262084). A solution of 23 (5.93 g, 7.6 mmol) in 50
mL of ethanol was degassed with a vacuum/argon, followed
by adding 10% Pd/C (50% water, 2.4 g). The resulting black
suspension was degassed again with a vacuum. The reaction
mixture was then exposed to hydrogen (via a balloon) at 1 atm
of pressure. After 1.1 h, the catalyst was removed by filtration,
and the filtrate was evaporated. The residue was kept in vacuo

3600 J . Org. Chem., Vol. 67, No. 11, 2002
Qian et al.
overnight to give 1 as a white powder (3.18 g, 99% yield). Mp:
213-215 °C dec. [R]25D ) -65.9 (c 0.99, MeOH). 1H NMR (CD3-
OD): δ 4.17 (d, J ) 3.29 Hz, 1H), 3.61-3.11 (m, 11H), 1.94-
1.75 (m, 4H), 1.32 (s, 9H). 13C NMR (CD3OD): δ 176.6, 168.7,
159.4, 158.7, 152.3, 58.7, 53.2, 51.8, 46.5, 45.0, 41.8, 29.6, 27.4,
26.3. HRMS: calcd for C18H32N7O5 (M+ + H) 426.2465, found
426.2470. IR (KBr): 3385, 3184, 1775, 1657, 1535, 1395, 1259,
1207, 996, 763 cm-1. Anal. Calcd for C18H31N7O5: C, 50.81, H,
7.34, N, 23.04. Found: C, 50.65, H, 7.42, N, 22.72. Chiral
HPLC: ee 99.6%; Chiralpak OD column, 250 × 4.6 mm, 10
µm; mobile phase hexane/EtOH (85:15, v/v); isocratic at
ambient temperature, 1.0 mL/min, 220 nm; concentration 0.25
mg/mL, 10 µL injection; RT ) 18.6 min (enantiomer, RT ) 15.7
min).
Ack n ow led gm en t. We thank Ms. J acqueline Shag-
am for her help with the chiral HPLC analyses. We
thank Dr. Percy Manchand for his suggestions and
proof-reading of this manuscript.
Su p p or tin g In for m a tion Ava ila ble: NMR spectra of the
isolated compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
J O010757O