Efficient synthesis of (Z)- and (E)-methyl 2-(methoxyimino)-2-phenylacetate

Article abstract of DOI:10.1016/j.tetlet.2010.02.073

Direct oximation of 2-oxo-2-phenylacetate (3) gave the (Z)-methyl 2-(methoxyimino)-2-phenylacetate (1) in 71% yield, while the E oxime 2 was prepared from 3 in 65% yield via oxime isomerization of 2-(methoxyimino)-2-phenylacetic acid (5). Computational st

Full text of DOI:10.1016/j.tetlet.2010.02.073

Tetrahedron Letters 51 (2010) 2144–2147
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Efficient synthesis of (Z)- and (E)-methyl 2-(methoxyimino)-2-phenylacetate
b
c
d
e
e
Yong-Jin Wu a, , Stella Huang , Alicia Ng , Qi Gao , S. Roy Kimura , David R. Langley
*
a Department of Neuroscience Chemistry, Research and Development, Bristol-Myers Squibb Company, 5 Research Parkway, Wallingford, CT 06492, USA
b Analytical Research and Development, Research and Development, Bristol-Myers Squibb Company, 5 Research Parkway, Wallingford, CT 06492, USA
c Bioanalytical and Discovery Analytical Sciences, Research and Development, Bristol-Myers Squibb Company, 5 Research Parkway, Wallingford, CT 06492, USA
d Analytical Research and Development, Bristol-Myers Squibb Company, 1 Squibb Drive, New Brunswick, NJ 08901, USA
e Department of Computer-Aided Drug Design, Research and Development, Bristol-Myers Squibb Company, 5 Research Parkway, Wallingford, CT 06492, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Direct oximation of 2-oxo-2-phenylacetate (3) gave the (Z)-methyl 2-(methoxyimino)-2-phenylacetate
(1) in 71% yield, while the E oxime 2 was prepared from 3 in 65% yield via oxime isomerization of 2-
(methoxyimino)-2-phenylacetic acid (5). Computational studies suggest that the isomerization of 5 is
thermodynamically driven, while the direct oximation of ketoester 3 is kinetically controlled.
Ó 2010 Elsevier Ltd. All rights reserved.
Received 15 January 2010
Revised 4 February 2010
Accepted 12 February 2010
Available online 19 February 2010
Alkyloximes are valuable intermediates in organic synthesis
and are found in a variety of drugs on the market, including fluvox-
amine (antidepressant), oxiconazole (antifungal), gemifloxacin
(antibacterial), and cefetamet (antibacterial) (Fig. 1).
amine hydrochloride as demonstrated by Uda et al. (Scheme 1).1
Thus, the treatment of 3 with methoxyamine hydrochloride affor-
ded the two oxime isomers 1 and 2 in 70% and 20% yields, respec-
tively. We carried out this reaction on a multigram scale and
obtained similar yields of both isomers. The method did allow
us to deliver a sizable amount of the major Z isomer 1 for our
medicinal chemistry program, but we were unable to scale up a
sufficient quantity of the minor E isomer 2. Oxime 2 can also be
prepared in moderate yield via a palladium-catalyzed cross-cou-
pling reaction of phenylboronic acid with (Z)-methyl 2-chloro-2-
(methoxyimino)acetate (7) as described by Ziegler et al. (Scheme
1).2 As chloride 7 is not readily available, this method has limited
application for our purpose. We also looked into the oximation of
2-oxo-2-phenylacetic acid (4) with methoxyamine hydrochloride
as the oxime acid 6 can be converted to its methyl ester 2, but
O
N
O
N
NH2
OMe
N
Cl
Cl
Cl
N
F3C
Oxiconazole (OxistatTM
)
Fluvoxamine (LuvoxTM
)
Cl
OMe
H
O
N
CO2H
N
S
again, the
E isomer was formed as a very minor product
S
MeO
N
(5:6 = 20:1) when oximation of 4 was carried out in methanol at
pH 7 and À20 °C as reported recently by Tsukiyama and Sato
(Scheme 1).3 Therefore, there is still a need to develop efficient
methods for the preparation of the E oximes of 2-arylacetates.
This Letter describes a practical synthesis of the E oxime 2 via
oxime isomerization.
N
N
O
N
O
Me
O
H2N
H2N
O
O
O
Gemifloxacin (FactiveTM
)
Cefetamet
t-Bu
At the outset of this program, we wondered if the E oxime ester
2 can be prepared from the Z isomer 1 by means of isomerization,
and not surprisingly, we observed no appreciable formation of 2
when 1 was subjected to a variety of thermal and acidic conditions.
The lack of isomerization is consistent with the literature finding
that O-alkyloximes rarely isomerize under standard conditions,
while hydroxyloximes can undergo facile isomerization. For exam-
ple, Narasaka reported that oxime 8a isomerizes quickly upon
exposure to trifluoromethanesulfonic acid, but no isomerization
is observed with the corresponding O-methyloxime 8b under the
same conditions (Scheme 2).4
Figure 1.
In a recent medicinal chemistry program, we desired to have
easy access to both (Z)- and (E)-methyl 2-(methoxyimino)-2-
phenylacetate 1 and 2. A simple approach to these compounds
involves oximation of 2-oxo-2-phenylacetate (3) with methoxy-
* Corresponding author. Tel.: +1 203 677 7485; fax: +1 203 677 7702.
E-mail address: yong-jin.wu@bms.com (Y.-J. Wu).
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.02.073

Y.-J. Wu et al. / Tetrahedron Letters 51 (2010) 2144–2147
2145
OMe
MeO
O
O
N
N
MeONH2•HCl
MeOH, 80 °C, 12 h
OMe
OMe
OMe
OMe
MeONH2•HCl
O
+
O
O
O
O
O
this study
Ref. 1
3
3
2 (20%)
1 (70%)
OMe
MeO
N
N
OMe
N
MeO
N
OMe
OMe
O
MeONH2•HCl,
+
OH
OH
OH
O
O
pH 7, -20 °C
+
1 (71%)
2 (25%)
O
3 : 1
Ref. 3
4
5
6
(20 : 1))
LiOH;
HCl
LiOH;
HCl
100%
100%
K2CO3,
MeI, DMF,
rt, 12 h
PhB(OH)2,
MeO
MeO
Cl
N
PdCl2(PPh3)2,
MeO
N
OMe
N
N
OMe
OMe
2: 65% yield
from 3
EtOH, aq. Na2CO3,
45-50 °C
45%
OH
OH
O
O
O
O
1: 24% yield
from 3
Ref. 2
2
7
5
6
Scheme 1.
CHCl3
MeO
OMe
N
N
RO
OR
OH
OH
N
N
CF3SO3H
(2 equiv)
O
O
Me
Me
5
Ref. 4
6
1 : 3.5
anti
9a/b
8a/b
syn
Scheme 3.
a: R = H; b: R = Me
R
Time
anti/syn
OMe
O
MeO
N
N
H
Me
<20 min
12 h
2 : 1
100 : 0
OH
OH
+
O
5
6
Scheme 2.
We next turned our attention to the isomerization of oxime acids
5 and 6 (Scheme 3). These acids were prepared in quantitative yields
from their corresponding esters upon treatment with aqueous lith-
ium hydroxide in THF followed by neutralization with 1 N aqueous
hydrochloric acid. Under these conditions, no isomerization
occurred. Both oxime acids were very stable at room temperature
as white solids presumably due to intermolecular hydrogen bonding
interactions as observed in the X-ray crystal structures (data not
shown). However, whendissolved inanorganicsolventsuch aschlo-
roform, dichloromethane, and methanol at room temperature for 6–
12 h, both 5 and 6 isomerized to the same mixture of 5 and 6 in a
1:3.5 ratio as shown in the 1H NMR spectrum of the equilibrium mix-
ture (Fig. 2). This ratio remained unchanged with an addition of acid
(e.g., hydrochloric acid) or use of polar solvents (e.g., methanol) or at
higher temperature (e.g., 80 °C). Under basic conditions such as
aqueous 1 N lithium hydroxide, both 5 and 6 showed no detectable
isomerization. Thus, the isomerization observed with both 5 and 6
is most likely self-acid-catalyzed.
OMe
N
OH
O
5
MeO
N
OH
O
6
7.6 7.4 7.2 7.0 6.8 6.6 6.4 6.2 6.0 5.8 5.6 5.4 5.2 5.0 4.8 4.6 4.4 4.2 4.0
f1 (ppm)
Figure 2. 1H NMR spectra of 6 (bottom), 5 (middle), and the equilibrium mixture of
5 and 6 (top).
The observation that oxime acids 5 and 6 readily isomerize led
to an efficient preparation of the E oxime 2. Thus, ketoester 3 was
treated with methoxyamine hydrochloride in methanol at 80 °C for
12 h to give a 3:1 mixture of 1 and 2. This crude mixture was
hydrolyzed with lithium hydroxide and then neutralized with
1 N hydrochloric acid. The resulting 3:1 mixture of 5 and 6 was dis-
solved in chloroform and the ratio changed to 1:3.5 after overnight
at room temperature. The solvent was removed, and the residue
was treated with potassium carbonate and methyl iodide in DMF
to afford 2 and 1 in 65% and 24% overall yields, respectively, from
3. We have utilized this equilibration method to scale up multi-
gram quantities of oxime 2.
In our current studies, ketoester 3 was used for the sizable prep-
aration of both oxime isomers 1 and 2. Certainly, the E isomer 2 can
be made directly from ketoacid 4 in three steps: oximation, isom-
erization, and methylation.
The oxime geometry of compounds 1, 2, and 5 was assigned by
comparing 1H NMR data with those reported in the literature.1,3
With compound 5 assigned as Z configuration, compound 6 must
have E configuration. In addition, we obtained 15N chemical shifts
of the four oximes from 1H–15N correlation spectra as shown in
Table 1. These stereochemical assignments are consistent with
the general observation that the Z oximes exhibit more downfield

2146
Y.-J. Wu et al. / Tetrahedron Letters 51 (2010) 2144–2147
Table 1
interest was defined by the phenyl ring, the tetrahedral carbon, the
15N chemical shifts of oximesa
oxime nitrogen, and the methoxy group. Calculations of the force-
field (OPLS20056) dihedral potential profile with several simplified
implicit solvent models (Generalized Born11 and simple dielectrics)
all consistently yielded a global torsional minimum near 300° (data
not shown). Since the E and Z oxime isomers have torsion angles 0°
and 180°, respectively, these simplified models predicted that the
system would favor the E isomer 2, which is closer to the torsional
minimum. However, this does not agree with experimental observa-
tions, suggesting the need for a more accurate solvent description
and proper ensemble averaging of conformational populations. We
therefore performed molecular dynamics simulations of intermedi-
ate 10 in a periodic 38 Â 38 Â 38 Å box of explicit methanol solvent.
Simulations were carried out using the AMBER GAFF force-field12 at
300 K in the NPT ensemble. Simulation temperatures were periodi-
cally ramped up to 600 K and annealed back down to 300 K in the
NVT ensemble to allow for escape from potentially stable intra-
molecular hydrogen bonds and other conformational wells. Tor-
sional data were collected only during the 300 K NPT intervals, and
ten separate 200 ps data collection intervals were used (total of
2 ns of production simulation time). The resulting torsion angle dis-
tribution shown in Table 3 predicted that the Z oxime 1 is preferred
over the E oxime 2 by an approximately 3.5 to 1 ratio which is in
excellent agreement with the observed ratio (3:1).
Compd Observed dN
Configuration
assignment
X-ray
confirmation
(ppm)
1
2
5
6
401.2
371.0
402.8
376.4
Z
E
Z
E
Yes
No
Yes
Yes
The 15N chemical shift was referenced to NH3 (liquid) at À380.2 ppm.
a
nitrogen chemical shifts than their corresponding Z oximes.5 Final-
ly, we obtained X-ray crystal structures of 1, 5, and 6, thus confirm-
ing our stereochemical assignments.
The dominance of 6 over 5 in the equilibrium is not well under-
stood. The hydrogen bonding hypothesis would favor the Z isomer
5, while the argument based on the steric repulsion between meth-
oxy and carboxylic acid or phenyl does not seem to be obvious
(Fig. 3). Other factors such as dipole–dipole interaction may play
important roles.
Me
MeO
O
MeO
OMe
H
O
N
N
H
O
N
N
O
OH
OH
O
H
O
O
N
OMe
HO
6
5
5
6
disfavored
θ
O
favored
Figure 3.
O
10
To better understand the oxime isomerization of 5 and 6, we
carried out preliminary computational studies. Since thermody-
namic equilibration was observed for the oxime acids 5 and 6,
we sought to determine whether their quantum mechanical
(QM) ground state energies agreed with the observed ratios. Calcu-
lations were performed by first taking the OPLS20056 molecular
mechanics force-field minima in vacuum for each,7 then subjecting
them to density functional theory geometry optimizations using
the 6-31G* basis set.8 The conformational minima identified
through this procedure agreed with the expected internally hydro-
gen bonded conformations. Single-point ground state energies
were then calculated at the RIMP2 level of theory with Dunning’s
aug-cc-pvtz9 basis set in both vacuum and water dielectric
Figure 4.
Table 3
Torsion angles
Proportion
90 < h < 270
h < 90 or h > 270
0.778
0.222
In summary, we have developed an efficient synthesis of (E)-
methyl 2-(methoxyimino)-2-phenylacetate (2) using a facile isom-
erization of the Z oxime acid 5.13 This methodology should be
applicable to other E oximes of 2-arylacetates. Our preliminary com-
putational studies suggest that the isomerization of 5 is thermody-
(e = 78) environments. All QM calculations were carried out using
the QCHEM package.10 The resulting calculated energies (Table 2)
correctly identified oxime 6 as the preferred isomer over 5 but
over-estimated the energy difference, predicting a 1:91 ratio of 5
to 6 (cf. observed ratio: 1:3.5). The discrepancy is likely due to ne-
glect of other minor conformational populations as well as micro-
scopic solvent effects.
namically driven, while direct oximation of ketoester
3 is
kinetically controlled. Future efforts will be directed toward delin-
eating the full scope of this isomerization, understanding the isom-
erization mechanism, and applying this methodology to the
synthesis of biologically important compounds.
Since both oximes 1 and 2 do not isomerize under standard con-
ditions, we hypothesized that the intermediate 10 (Fig. 4) (which
would lead to 1 and 2 in the oximation of keto ester 3) exhibits con-
formational preferences that kinetically dictate the final oxime iso-
mer ratio. In our computational studies, the dihedral angle of
Acknowledgments
We thank Drs. Lorin Thompson and John Macor for their
encouragement and support.
Table 2
Supplementary data
D
E vacuum
D
E dielectric
e
= 78
Calculated ratio of 5:6
25°C, vacuum
(kcal/mol)
(kcal/mol)
Supplementary data (X-ray crystal structures of compounds 1, 5
and 6) associated with this article can be found, in the online ver-
sion, at doi:10.1016/j.tetlet.2010.02.073.
E5–
E6
1:90
2.67
2.61

Y.-J. Wu et al. / Tetrahedron Letters 51 (2010) 2144–2147
2147
12. Cornell, W. D.; Cieplak, P.; Bayly, C. I.; Gould, I. R.; Merz Jr, K. M.; Ferguson, D.
M.; Spellmeyer, D. C.; Fox, T.; Caldwell, J. W.; Kollman, P. A. J. Am. Chem. Soc.
1995, 117, 5179.
References and noes
1. Uda, J.; Okamoto, Y.; Iwabuchi, T.; Inoue, T.; Nakano, H.; Ishikawa, T.; Kawasaki,
N.; Katoh, M.. PCT Int. Appl., WO 2005068415 A1, 2005.
13. All new compounds have spectral and analytical data in agreement with the
indicated structures. Compound 1: 1H NMR (400 MHz, CDCl3) d 3.93 (3H, s),
4.01 (3H, s), and 7.40 (3H, m), and 7.55 (2H, m). 13C NMR (125 MHz, CDCl3)
d (attached H’s) 164.2 (0), 150.6 (0), 130.5 (1), 130.2 (0), 128.9 (2C, 0), 126.3
(2C, 0), 63.1 (3), 52.5 (3). HRMS calcd m/z for C10H12NO3 194.0817 (M+H)+,
found 194.0811. Compound 2: 1H NMR (400 MHz, CDCl3) d 3.88 (3H, s), 4.05
(3H, s), and 7.40 (5H, s). 13C NMR (125 MHz, CDCl3) d (attached H’s) 164.0
(0), 149.2 (0), 129.7 (1), 129.2 (2C, 0), 129.5 (0), 128.1 (2C, 0), 63.8 (3), and
53.1 (3). HRMS calcd m/z for C10H12NO3 194.0817 (M+H)+, found 194.0810.
Compound 5: 1H NMR (400 MHz, CDCl3) d 4.08 (3H, s), 7.43 (3H, m), and
7.65 (2H, m). 13C NMR (125 MHz, CDCl3) d (attached H’s) 63.3 (3), 126.5 (2C,
1), 128.8 (2C, 1), 129.8 (0), 130.6 (0), 149.6 (0), and 167.4 (0). HRMS calcd
m/z for C9H10NO3 180.0660 (M+H)+, found 180.0654. Compound 6: 1H NMR
(400 MHz, CDCl3) d 4.08 (3H, s), 7.43 (3H, m), and 7.49 (2H, m). 13C NMR
(125 MHz, CDCl3) (attached H’s) d 64.3 (3), 127.4 (0), 128.0 (2C, 1), 129.6
(2C, 1), 130.3 (1), 147.4 (0), and 162.7 (0). HRMS calcd m/z for C9H10NO3
180.0661 (M+H)+, found 180.0654.
2. Ziegler, H.; Neff, D.; Stutz, W. PCT Int. Appl., WO 9520569 A1, 1995.
3. Tsukiyama, T.; Sato, K. PCT Int. Appl., WO 2002059078 A1, 2002.
4. Kitamura, M.; Yoshida, M.; Kikuchi, T.; Narasaka, K. Synthesis 2003, 15, 2415.
5. Berger, S.; Braum, S.; Kalinowski, H. NMR Spectroscopy of the None-Metallic
Elements; John Wiley & Sons: New York, 1996. P206.
6. Kaminski, G. A.; Friesner, R. A.; Tirado-Rives, J.; Jorgensen, W. J. J. Phys. Chem. B
2001, 105, 6474.
7. Conformational searches conducted using Monte Carlo Multiple Minima
Method with Default Settings in Maestro, Version 8.5, Schrodinger, LLC, New
York, NY, 2008.
8. Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta. 1973, 28, 213.
9. Dunning, T. H., Jr.; Hay, P. J. In Modern Theoretical Chemistry; Schaefer, E. D., III,
Ed.; Plenum: New York, 1976; pp 1–28.
10. Shao, Y.; and 65 co-workers. Phys. Chem. Chem. Phys. 2006, 8, 3172.
11. Still, W. A.; Tempczyk, A.; Hawley, R. C.; Hendrickson, T. J. Am. Chem. Soc. 1990, 112,
6127–6129; Onufriev, A.; Bashford, D.; Case, D. A. J. Comput. Chem. 2002, 23, 1297.