Synthesis of novel optically pure β-lactams

Article abstract of DOI:10.1139/v04-159

Several new β-lactams were synthesized as racemates via a Staudinger reaction. The corresponding optically pure compounds were obtained in subsequent biotransformation steps either through baker's yeast reduction or lipase resolution. Their absolute confi

Full text of DOI:10.1139/v04-159

28
Synthesis of novel optically pure ␤-lactams
Yan Yang, Fengyun Wang, Fernande D. Rochon, and Margaret M. Kayser
Abstract: Several new β-lactams were synthesized as racemates via a Staudinger reaction. The corresponding optically
pure compounds were obtained in subsequent biotransformation steps either through baker’s yeast reduction or lipase
resolution. Their absolute configurations were established. The X-ray crystal structures of three new substituted β-lac-
tams are reported here. These compounds represent key building blocks for a variety of medicinally important mole-
cules, including inhibitors of aspartyl proteases and Taxol® analogues.
Key words: optically pure β-lactams, lipase resolutions, baker’s yeast reductions, Staudinger reaction.
Résumé : Faisant appel à la réaction de Staudinger, on a réalisé la synthèse de plusieurs nouvelles β-lactames racémi-
ques. On a obtenu les composés correspondants optiquement purs par le biais de biotransformations subséquentes im-
pliquant une réduction par la levure de boulanger ou par résolution avec une lipase. On en a déterminé les
configurations absolues. On a déterminé les structures cristallines de trois β-lactames substituées par diffraction des
rayons X. Ces composés sont des unités pouvant servir à la synthèse de plusieurs molécules médicinalement importan-
tes, y compris des inhibiteurs de protéases de l’aspartyle et des analogues du Taxol®.
Mots clés : β-lactames optiquement pures, résolutions à la lipase, réductions avec de la levure de boulanger, réaction
de Staudinger.
[Traduit par la Rédaction] Yang et al. 36
Introduction
easier to administer, fewer side effects, and active on multi-
drug-resistant (MDR) cells is an on-going process (6). We
have been interested in developing the synthesis of optically
pure C-13 side chains in which 3′-phenyl is replaced with
bulky, polar groups. The preparation of side chains in the
form of 3-hydroxy-β-lactams is particularly appealing since
it is in this form that the prospective side chains are gener-
ally coupled with baccatin III. Here, we report the syntheses
and characterizations of several isomers of 4-(1-
ethoxycarbonyl-1-methylethyl)-3-hydroxy-β-lactam (1), the
4-(1-carboxy-1-methylethyl)-3-hydroxy-β-lactam (2), and
the corresponding lactone (3).
The β-lactam skeleton represents a core structure of
numerous natural and synthetic antibiotics (1). The 3-
hydroxy-substituted β-lactams are convenient precursors of
β-amino-α-hydroxy esters that constitute key fragments of
many biologically important natural products such as
aminosugars, α-hydroxy-β-amino acids, glycosphingolipids,
peptidic enzyme inhibitors, and alkaloids (2). Furthermore,
3-hydroxy-β-lactams are used for the construction of ana-
logues of the anticancer drug Taxol® (3). It is not surprising,
therefore, that the development of enantioselective syntheses
of such β-lactams has been pursued by numerous laborato-
ries around the world.
The extensive structure–activity relationship (SAR) stud-
ies have shown that the phenylisoserine C-13 side chain of
paclitaxel (Taxol®, Fig. 1), essential for anticancer activity,
tolerates a variety of modifications (4), and even small
changes significantly influence the bioactive properties of
the drug (3, 5). The therapeutic potential of paclitaxel is lim-
ited by its high hydrophobicity and development of resis-
tance. Search for analogues with better water solubility,
A number of methods for the preparation of β-lactams,
such as ketene–imine cycloaddition (1a, 7), metalloester
enolate–imine
condensation
(8),
isocyanate–alkene
Received 22 March 2004. Published on the NRC Research
Press Web site at http://canjchem.nrc.ca on 5 February 2005.
cycloaddition (9), and alkyne–nitrone cycloaddition (10), have
been developed over the years. Among these, the most
widely used is the Staudinger reaction, in which ketene–
imine addition leads to a β-lactam ring system. Optically
pure β-lactams prepared via asymmetric synthesis employed
either a combination of chiral imine and achiral ketenes,
achiral imines and chiral ketenes (1a, 11), or a direct
cyclization of ketene–imine in the presence of a chiral cata-
lyst (12). Only a few studies resorted to biocatalytic ap-
proaches such as lipase-catalyzed resolution (13) or baker’s
yeast reduction (14). Since our preliminary goal was to syn-
Dedicated to Professor Peter Stanetty on the occasion of his
60th birthday.
Y. Yang, F. Wang, and M.M. Kayser.1 Department of
Physical Sciences, University of New Brunswick, Saint John,
NB E2L 4L5, Canada.
F.D. Rochon. Département de chimie, Université du Québec
à Montréal, C. P. 8888, Succ. Centre-Ville Montréal, QC
H3C 3P8, Canada.
1Corresponding author (e-mail: kayser@unbsj.ca).
Can. J. Chem. 83: 28–36 (2005)
doi: 10.1139/V04-159
© 2005 NRC Canada

Yang et al.
29
Scheme 1.
Fig. 1. Structure of paclitaxel (Taxol®).
dered ester function undergoes hydrolysis aided by the
strategically located 3-OH group. The two compounds were
easily separated during the extraction process (see Experi-
mental). This turned out to be very useful in the character-
ization of 7-trans and 7-cis, since the original mixture from
the Staudinger reaction was inseparable by chromatography.
The pure sample of 7-trans was prepared by acylation of 1-
trans (Scheme 1), while 7-cis was obtained from baker’s
yeast reduction of the α-keto-β-lactam (8, Scheme 2). Their
stereochemistry was assigned on the basis of coupling con-
stants for the cis- and trans-β-lactams (16). Oxidation of
compound 1 with DMSO/phosphorus pentoxide provided the
α-keto-β-lactam (8). Under the same conditions, 2-cis was
dehydrated to a racemic mixture of fused lactone–lactams 3
as shown in Scheme 2.
thesize and characterize all the diastereomers and enantio-
mers, we decided to prepare the racemic β-lactams through
Staudinger cyclization (Scheme 1) and to obtain enantiopure
β-lactams in the subsequent biotransformation steps.
In an initial attempt to obtain optically pure 3-hydroxy-β-
lactams (1), we performed baker’s yeast reduction of the
racemic 3-oxo-β-lactam (8). This strategy was successful for
the tert-butyl analogue 9, where it yielded easily separable
3-hydroxy diastereomers 3R,4S-cis and 3R,4R-trans (14).
Unfortunately, in the case of ketone 8, the reduction product
consisted of an inseparable mixture of essentially racemic 1-
cis and an optically pure 1-trans isomer, tentatively identi-
fied as 3R,4R (Scheme 2). The ratio of cis to trans (5:1) was
Results and discussion
Condensation of 3,3-diethoxy-2,2-dimethylpropionic acid
ethyl ester (4) (15) with p-anisidine gave 3-(4-methoxy-
phenylimino)-2,2-dimethylpropionic acid ethyl ester (5),
which was used in the following step without purification.
Cycloaddition of 5 with the in-situ-generated acetoxyacetyl
chloride in the presence of anhydrous triethylamine gave
acetyl β-lactam 7 (PMP = p-methoxyphenyl) in 53% yield
after two steps. The β-lactam product 7 consisted of trans
1
measured from the H NMR spectrum of the product mix-
ture.
1
and cis diastereomers in the ratio of 2:1 based on the H
NMR spectrum of the product mixture. Hydrolysis of 7 (2 N
KOH) gave two compounds as shown in Scheme 1. It was
interesting to note that while the trans-acetyl β-lactam was
selectively hydrolyzed to the corresponding trans-3-hydroxy
β-lactam (1) as expected, its cis diastereomer was hydro-
lyzed at both acetyl and acetate groups, yielding cis-3-
hydroxy β-lactam (2). Apparently in the cis isomer, the hin-
Since a racemic mixture of cis-3-acetoxy-4-phenyl-β-lac-
tams was reported to be enantioselectively hydrolyzed by
Amano lipase (13a), we decided to test the ability of this en-
zyme to resolve the mixture of lactams 7. It turned out that
the mixture of the isomers of 7 could be resolved efficiently,
particularly since the 7-trans is not a suitable substrate for
the lipase PS “Amano” (Scheme 3).
© 2005 NRC Canada

30
Can. J. Chem. Vol. 83, 2005
Scheme 2.
Scheme 3.
Of the four products formed in the course of the lipase
resolution, compounds 1 and 3 were easily separated by
chromatography and the absolute configuration of 1 (3S,4R)
was established with certainty by X-ray diffraction methods.
The whole diffraction sphere was measured with Cu-Kα ra-
diation. The absolute structure parameter (Flack) determined
using the option TWIN/BASF 0.5 indicated the right config-
uration (Table 1).
85.4(1)° and 95.1(2)°. The N—C and O—C bond distances
close to the carbonyl groups are shorter than the others. For
example, the distances N—C1 = 1.360(3) Å and O4—C5 =
1.335(3) Å, while N—C3 = 1.488(3) Å and O4—C6 =
1.461(3) Å.
The lactone–lactam 3 (1R,5S), apparently resulting from
hydrolysis of the ethoxy carbonyl group followed by
cyclization, was isolated and its absolute configuration was
initially established by chiral-phase HPLC analysis. This
was possible because the four isomers of 7 are clearly re-
The molecular structure of 1 (3S,4R) is shown in Fig. 2.
The internal angles in the four-membered ring vary between
© 2005 NRC Canada

Yang et al.
31
Table 1. Crystal data for compounds 1, 3, and 11.
Compound
1
3
11
Empirical formula
C16H21NO5
C14H15NO4
C23H18BrNO4
Formula weight
Temperature (K)
Wavelength (Å)
Crystal system
Space group
307.34
223(2)
1.541 78
Monoclinic
P21
261.27
223(2)
1.541 78
Monoclinic
P21
452.29
293
0.710 73
Orthorhombic
P212121
a (Å)
b (Å)
c (Å)
11.383 2(2)
5.895 9(1)
12.270 9(2)
110.431(1)
771.74(2)
2
5.896 2(1)
18.571 0(2)
11.529 2(1)
90.762(1)
1 262.32(3)
4
8.901(3)
10.392(4)
21.842(8)
90.
2020.4(13)
4
β (°)
Volume (Å3)
Z
Dcalcd (Mg m–3)
1.323
1.375
1.487
Absorption coefficient (mm–1)
F(000)
0.815
328
0.842
552
2.064
920
θ range for data collection (°)
Reflections collected
3.84–72.9
9298
3.83–72.77
15 210
1.86–25.0
2052
Independent reflections (Rint)
2711 (0.051 1)
4 802 (0.029 8)
2052
Observed reflections (I > 2σ(I))
Goodness-of-fit on F2
Final R1 [I > 2σ(I)]
2656
1.041
0.051 5
0.139 6
4 570
1.045
0.035 0
0.073 0
899
1.048
0.063 6
0.108 2
wR2 (all data)
Absolute structure parameter
0.1(2)
0.00(17)
0.00(3)
Fig. 2. Crystal structure of cis-3-hydroxy-β-lactam (1, 3S,4R).
solved on the chiral-phase HPLC. Furthermore, in the course
of the HPLC-monitored lipase resolution we were able to
observe that only one of the cis enantiomers was hydro-
lyzed; its antipode and the two trans diastereomers were un-
affected. Additional evidence for the configuration of 3
(1R,5S) was obtained when a mixture of 1 (3S,4R) and 3
(1R,5S) under extended treatment with an excess of the
lipase was transformed entirely into the lactone 3 (1R,5S).
Finally, we were able to obtain a suitable crystal, and the
proposed absolute configuration of 3 (1R,5S) was confirmed
by X-ray diffraction. Again, the absolute structure parameter
(Flack) confirmed the correct enantiomer (Table 1). The
compound crystallized with two independent molecules in
the unit cell. The two molecules are very similar as shown
by identical equivalent bond distances and angles. The mo-
lecular structure of one of the two molecules is shown in
Fig. 3.
The conformations of the two molecules are slightly dif-
ferent, especially in the region of the five-membered ring.
For example, the torsion angles in the first molecule N1-C3-
C4-C5 = –91.9(2)°, N1-C3-C4-C7 = 27.5(3)°, and N1-C3-
C4-C6 = 153.8(2)°, while the corresponding angles in the
second molecule are N2-C17-C19-C18 = –82.4(2)°, N2-
C17-C19-C20 = 38.7(3)°, and N2-C17-C19-C27 = 165.3(2)°.
Again, the N—C and O—C bond distances close to the car-
bonyl bonds are shorter than the others.
The remaining products from the lipase resolution, 7
(3R,4S) and 7-trans (racemic), were hydrolyzed with 2 N
© 2005 NRC Canada

32
Can. J. Chem. Vol. 83, 2005
Fig. 3. Crystal structure of the fused lactone–lactam 3 (1R,5S).
Fig. 4. Crystal structure of trans-β-lactam (11, 3R,4R).
KOH to give 2 (3R,4S) and 1-trans (racemic), which were
separated during the work-up. The subsequent treatment of 2
(3R,4S) with DMSO/phosphorous pentoxide gave lactone 3
(1S,5R), which is clearly distinguishable from its antipode
on the chiral-phase HPLC and has opposite specific rotation
(Scheme 3).
The attempts to resolve racemic 1-trans using lipases PS
“Amano”, PS-C “Amano” II, AY “Amano”, and AK
“Amano” were not successful. To characterize fully and to
establish unambiguously the absolute configurations of the
trans products, we decided to separate and identify the 1-
trans enantiomer from the baker’s yeast reduction
(Scheme 2). We were unable to separate the overlapping cis
and trans isomers, however, on treatment with 2 N KOH,
only the racemic 1-cis was hydrolysed to the corresponding
acid and the enantiopure 1-trans was separated during the
work-up. Unfortunately, neither 1-trans nor its p-
bromobenzoyl derivative 10 could produce crystals suitable
for X-ray crystallographic analysis. Thus, the (3R,4R) abso-
lute configuration was tentatively assigned by analogy to
other related trans-β-lactams obtained from yeast-catalyzed
reductions: the phenyl-substituted 11 (3R,4R) (17) whose
crystal structure was determined and is shown in Fig. 4 (Ta-
ble 1), and tert-butyl-substituted lactam 12 (3R,4R) reported
earlier (14).
In summary, we have synthesized and fully characterized
several new β-lactams. The absolute configurations were
deduced from chiral-phase HPLC and confirmed through
X-ray crystal structure determination. We are currently de-
veloping and optimizing more efficient and better-yielding
© 2005 NRC Canada

Yang et al.
33
lene chloride was freshly prepared by distillation from cal-
cium hydride. Triethylamine was freshly distilled from so-
dium hydride. Commercially available DMSO was treated
with molecular sieves (4 Å). Thin layer chromatography was
performed on Sigma-Aldrich 0.2 mm aluminum-backed sil-
ica gel plates with UV indicator and was visualized under a
UV lamp. Flash chromatography was performed on 230–400
mesh silica gel (Silicycle). Chiral-phase HPLC analyses
were performed on a Chiracel OD-H column (4.6 mm ×
150 mm) using hexane:iso-propanol (90:10) as the mobile
phase. A UV detector set at 254 nm was used to monitor the
reactions. Capillary gas chromatography was performed on a
DB-1301 (15 m × 0.53 mm × 1.0 µm) column from J & W
Scientific. Lipases were generous gifts from Amano Enzyme
USA Co., Ltd. Commercial baker’s yeast was obtained from
a local grocery chain.
Melting points were determined on a Fisher–Johns melt-
ing point apparatus and are uncorrected. Specific rotation
([α]D) was measured on a PerkinElmer 241 polarimeter at
room temperature with energy source Na 589 and is given in
10–1° cm2 g–1. IR spectra were recorded as thin films on a
Mattson Satellite FT IR spectrometer. 1H NMR and 13C
NMR spectra were recorded in CDCl3 solution, except
where otherwise specified, at room temperature either on a
Bruker AC-250 FT-NMR spectrometer or a Varian Unity
400 FT-NMR spectrometer; chemical shifts (δ) are reported
in ppm using Me4Si as the internal standard. J values are ex-
pressed in Hz. The high-resolution mass spectra were ob-
tained on a Kratos MS50TC mass spectrometer.
protocols for the preparations of these and several other re-
lated compounds.
Experimental
The X-ray crystallographic data for compounds 1 and 3
were collected using a Bruker Platform diffractometer,
equipped with a Bruker SMART 2K charged-coupled device
(CCD) area detector using the program SMART and normal
focus sealed tube source graphite-monochromated Cu-Kα ra-
diation. The crystal-to-detector distance was 4.908 cm, and
the data collection was carried out in 512 pixel × 512 pixel
mode, utilizing 4 pixel × 4 pixel binning. The initial unit cell
parameters were determined by a least-squares fit of the an-
gular setting of strong reflections, collected by a 9.0° scan in
30 frames over four different parts of the reciprocal space
(120 frames total). One complete sphere of data was col-
lected, to better than 0.8\Å resolution. Upon completion of
the data collection, the first 101 frames were recollected to
improve the decay correction analysis. The crystal structure
of the bromo compound (11) was determined on a Siemens
P4 instrument using graphite-monochromated Mo-Kα radia-
tion and the XSCANS program (18). The intensity data were
corrected for the effects of Lorentz and polarization. An ab-
sorption correction based on the equations of the crystal
faces was applied for crystal 11. The coordinates of the first
set of atoms (Br for 11) were determined by direct methods
and the positions of all the other atoms were found by the
standard Fourier methods. The structures were refined on
( )-3-Acetoxy-4-(1-ethoxycarbonyl-1-methylethyl)-1-(4-
methoxyphenyl)-2-azetidinone (7)
A mixture containing p-anisidine (15.4 g, 0.125 mol),
ethyl 3,3-diethoxy-2,2-dimethylpropanoate (4) (30.0 g, 0.137
mol), and p-toluenesulfonic acid monohydrate (0.100 g,
0.526 mmol) was heated at reflux for 5 h. The reaction mix-
ture was cooled and evaporated under vacuum to remove the
resulting ethanol. The imine 5 thus obtained was used in the
next step without purification. Phosphorus oxychloride
(28.0 g, 0.182 mol) in dry methylene chloride (100 mL) was
added dropwise to a solution of imine 5, acetoxyacetic acid
6 (23.4 g, 0.198 mol), and triethylamine (69.9 mL, 0.502
mol) in methylene chloride (500 mL) maintained at 5 °C.
The stirred reaction was warmed to room temperature and
monitored by TLC until complete disappearance of the
imine (48 h). The reaction was quenched with 2 N HCl
(400 mL) solution, and extracted with methylene chloride
(3 × 400 mL). The combined organic layers were washed se-
quentially with a saturated solution of NaHCO3 and brine,
dried over magnesium sulfate, and concentrated under re-
duced pressure. The residue was purified by flash chroma-
tography on silica gel to give compound 7 as a mixture of
diastereomers (trans:cis = 2:1) with a yield of 23.0 g (53%)
over two steps from p-anisidine. 7-trans: IR (CHCl3, cm–1):
2
F0 using all reflections. The H atoms were fixed in ideal-
ized positions using a riding model with displacement
parameters = 1.2Ueq (1.5 for methyl groups) of the parent
atom. The SHELXTL system (19) was used for all calcula-
tions and drawings. The experimental details and crystal
data are listed in Table 1. The absolute structure of the three
compounds was established with certainty. For crystals 1
and 3, the data were measured with Cu-Kα radiation on the
entire sphere. Crystal 11 was measured with Mo-Kα, but it
contains a heavy atom, which makes it very easy to deter-
mine the absolute configuration. The absolute Flack parame-
ter was obtained using the option TWIN/BASF 0.5 in the
refinement program. The crystal of compound 3 contains
two independent molecules in the unit cell.
1
2981, 2940, 1765, 1731, 1513, 1372, 1221, 1149. H NMR
(400 MHz, CDCl3) δ: 1.17 (3H, t, J = 7.2, CH2CH3), 1.21
(3H, s, CCH3), 1.29 (3H, s, CCH3), 2.17 (3H, s, COCH3),
3.78 (3H, s, OCH3), 4.03 (2H, q, J = 7.0, CH2CH3), 4.49
(1H, d, J = 1.8, NCH), 5.74 (1H, d, J = 1.8, OCH), 6.86
(2H, d, J = 9.1, ArH), 7.21 (2H, d, J = 8.9, ArH). 13C NMR
(100 MHz, CDCl3) δ: 13.9 (CH2CH3), 20.0 (CCH3), 20.7
All chemicals were purchased from commercial suppliers
and were used as received except where noted. Dry methy-
© 2005 NRC Canada

34
Can. J. Chem. Vol. 83, 2005
(CCH3), 22.9 (COCH3), 43.5 (CCH3), 55.5 (OCH3), 61.3
(OCH2), 66.7 (NCH), 75.4 (OCH), 114.4, 123.1, 128.7,
157.8, 162.5 (CO), 169.2 (CO), 175.3 (CO). 7-cis: H NMR
ganic phases were washed with water (5 × 150 mL), fol-
lowed by brine and dried over sodium sulfate. The solvent
was removed under vacuum. The residue was purified by
flash chromatography on silica gel and crystallized from
ethyl acetate – hexane to give 5.3 g (71%) of 8 as yellow
crystals; mp 64 to 65 °C. IR (CHCl3, cm–1): 2981, 1815,
1
(400 MHz; CDCl3) δ: 1.26 (3H, t, J = 7.3, CH2CH3), 1.29
(3H, s, CCH3), 1.32 (3H, s, CCH3), 2.10 (3H, s, COCH3),
3.80 (3H, s, OCH3), 4.11 (2H, q, J = 7.0, CH2CH3), 4.76
(1H, d, J = 5.3, NCH), 5.74 (1H, d, J = 5.3, OCH), 6.89
(2H, d, J = 9.0, ArH), 7.33 (2H, d, J = 9.2, ArH). 13C NMR
(100 MHz, CDCl3) δ: 13.9 (CH2CH3), 20.0 (CCH3), 20.7
(CCH3), 22.9 (COCH3), 44.5 (CCH3), 54.2 (OCH3), 61.0
(OCH2), 63.2 (NCH), 73.3 (OCH), 114.4, 121.5, 129.7,
157.3, 163.2 (CO), 168.8 (CO), 175.0 (CO). HR-MS calcd.
for C18H23NO6 (M+): 349.1525; found: 349.1511.
1
1764, 1513, 1253, 1149. H NMR (400 MHz, CDCl3) δ:
1.18 (3H, t, J = 7.0, CH2CH3), 1.27 (3H, s, CCH3), 1.37
(3H, s, CCH3), 3.82 (3H, s, OCH3), 4.07 (2H, q, J = 7.2,
CH2), 4.95 (1H, s, NCH), 6.92 (2H, d, J = 9.0), 7.38 (2H, d,
J = 9.2). 13C NMR (100 MHz, CDCl3) δ: 13.8 (CH2CH3),
21.0 (CCH3), 23.6 (CCH3), 44.9 (CCH3), 55.5 (OCH3), 61.7
(OCH2), 76.7 (NCH), 114.6, 121.5, 128.9, 158.4, 161.0
(CO), 174.4 (CO), 193.3 (CO). HR-MS calcd. for
C16H19NO5 (M+): 305.1263; found: 305.1274.
3-Hydroxy-1-(4-methoxyphenyl)-2-azetidinone (1)
A solution of 7 (mixture) (5.0 g, 0.014 mol) in THF
(100 mL) was slowly treated with 2 N KOH (50 mL) at
0 °C. The reaction was stirred at 0 °C until TLC indicated
complete conversion (2 h). The solution was extracted with
ethyl acetate (3 × 80 mL). The combined organic layers
were washed with brine, dried over magnesium sulfate, fil-
tered, and evaporated to dryness to afford racemic 1-trans
(2.8 g) as a colorless solid. The water layer was acidified
with 2 N HCl to pH 2 ~ 3, and then extracted with ethyl ace-
tate. The organic layers were combined and washed with
brine, dried over magnesium sulfate, filtered, and evaporated
to dryness to afford racemic 2-cis (1.2 g) as a colorless solid.
( )-trans-4-(1-Ethoxycarbonyl-1-methylethyl)-3-hydroxy-1-
(4-methoxyphenyl)-2-azetidinone (1): mp 78 to 79 °C. IR
(CHCl3, cm–1): 3387, 2939, 1730, 1513, 1249, 1150. 1H
NMR (400 MHz, CDCl3) δ: 1.16 (3H, s, CCH3), 1.18 (3H, t,
J = 7.2, CH2CH3), 1.27 (3H, s, CCH3), 3.78 (3H, s, OCH3),
4.05 (2H, q, J = 7.2, CH2), 4.38 (1H, d, J = 1.7, NCH), 4.73
(1H, d, J = 1.7, HOCH), 6.84 (2H, d, J = 9.0, ArH), 7.16
(2H, d, J = 8.9, ArH). 13C NMR (100 MHz, CDCl3) δ: 14.0
(CH2CH3), 20.4 (CCH3), 22.8 (CCH3), 43.6 (CCH3), 55.5
(OCH3), 61.2 (OCH2), 68.6 (NCH), 76.7 (OHCH), 114.4,
123.0, 129.0, 157.7, 167.2 (CO), 175.6 (CO). HR-MS calcd.
for C16H21NO5 (M+): 307.1420; found: 307.1428. ( )-cis-4-
(1-Carboxy-1-methylethyl)-3-hydroxy-1-(4-methoxyphenyl)-
2-azetidinone (2): mp 168 to 169 °C. IR (CHCl3, cm–1):
3373, 2979, 2839, 1724, 1513, 1248, 1170. 1H NMR
(400 MHz, d6-acetone) δ: 1.32 (3H, s, CCH3), 1.35 (3H, s,
CCH3), 3.79 (3H, s, OCH3), 4.74 (1H, d, J = 5.3, NCH),
5.16 (1H, d, J = 5.3, OHCH), 6.92 (2H, d, J = 9.0), 7.40
(2H, d, J = 9.0). 13C NMR (100 MHz, d6-acetone) δ: 22.1
(CCH3), 26.8 (CCH3), 45.6 (CCH3), 56.5 (OCH3), 65.7
(NCH), 77.7 (OCH), 115.7, 122.7, 132.8, 158.4, 168.8 (CO),
178.6 (CO). HR-MS calcd. for C14H17NO5 (M+): 279.1107;
found: 279.1113.
( )-cis-7-(4-Methoxyphenyl)-2,2-dimethyl-4-oxa-7-aza-
bicyclo[3.2.0]heptane-3,6-dione (3)
Phosphorus pentoxide 1.8 g (0.013 mmol) was added to
dry DMSO (35 mL) and stirred at room temperature for
30 min. Racemic 2-cis (2.4 g, 0.0090 mol) in DMSO
(15 mL) was added over 30 min. The resulting mixture was
stirred at room temperature until TLC indicated complete
conversion (48 h). The reaction was quenched with a cold
saturated aqueous solution of sodium bicarbonate (40 mL)
and extracted with ethyl acetate (3 × 40 mL). The combined
organic phases were washed with water (5 × 80 mL) and
brine, dried over sodium sulfate, and concentrated under re-
duced pressure. Crystallization of the crude residue from
ethyl acetate – hexane afforded 1.9 g (85%) of 3 as a color-
less crystalline solid; mp 169–169.5 °C. IR (CHCl3, cm–1):
1
2976, 2936, 2838, 1781, 1513, 1385, 1247, 1149. H NMR
(250 MHz, CDCl3) δ: 1.28 (3H, s, CCH3), 1.39 (3H, s,
CCH3), 3.80 (3H, s, OCH3), 4.49 (1H, d, J = 4.6, NCH),
5.44 (1H, d, J = 4.6, OCH), 6.90 (2H, d, J = 8.9), 7.31 (2H,
d, J = 8.9). 13C NMR (62.9 MHz, CDCl3) δ: 20.4 (CCH3),
25.1 (CCH3), 42.8 (CCH3), 55.5 (OCH3), 61.2 (NCH), 79.8
(OCH), 114.6, 118.7, 130.3, 157.1, 160.7 (CO), 180.1 (CO).
HR-MS calcd. for C14H15NO4 (M+): 261.1001; found:
261.1006.
Lipase-catalyzed hydrolysis of 7
The reaction was carried out in a 0.2 mol L–1 potassium
phosphate buffer (pH 7.5, 50 mL) containing a mixture of
cis and trans diastereomers of 7 (2.1 g, 0.006 mol) and
lipase PS “Amano” (2.1 g). The reaction mixture was stirred
at room temperature. The kinetic resolution of 7-cis was
monitored by chiral HPLC (the 7-trans diastereomers were
not hydrolysed by the lipase). After 26 h, 50% conversion of
7-cis was observed. The mixture was extracted with ethyl
acetate, and the combined ethyl acetate layers were washed
with brine and dried over magnesium sulfate. Removal of
the solvent afforded a mixture of unreacted 7 (3R,4S) and
racemic 7-trans as well as the compounds 1 (3S,4R) and 3
(1R,5S). Separation on a silica gel column gave unreacted
material 7 (1.2 g), optically pure 3-hydroxy-β-lactam (1)
(3S,4R) as a colorless crystal (0.080 g), and lactone 3
(1R,5S) as a colorless crystal (0.18 g). The recovered mix-
ture 7 (7.5 g, 0.022 mol) was hydrolyzed with 2 N KOH, as
described above, to yield optically pure 2 (3R,4S) as a white
solid (1.1 g) and racemic 7-trans as a colorless solid (5.0 g).
( )-4-(1-Ethoxycarbonyl-1-methylethyl)-1-(4-
methoxyphenyl)-3-oxo-2-azetidinone (8)
Phosphorus pentoxide (6.0 g, 0.042 mol) was added to dry
DMSO (90 mL) and stirred at room temperature for 30 min.
Racemic 1-trans (7.5 g, 0.025 mol) in DMSO (45 mL) was
added over an 1 h period. The resulting mixture was stirred
at room temperature until TLC showed complete conversion
(48 h). The reaction was quenched with a cold saturated
aqueous solution of sodium bicarbonate (100 mL) and ex-
tracted with ethyl acetate (3 × 100 mL). The combined or-
© 2005 NRC Canada

Yang et al.
35
In Enantioselective synthesis of β-amino acids. Edited by E.
Juaristi. Wiley-VCH, New York. 1997; C. Palomo, J.M.
Aizpurua, I. Ganboa, and M. Oiarbide. Pure Appl. Chem. 72,
1763 (2000); R. Herranz, J. Castro-Pichel, S. Vinuesa, and
M.T. Garcia-Lopez. J. Org. Chem. 55, 2232 (1990), and refs. cited
therein; Y. Kobayashi, Y. Takemoto, Y. Ito, and S. Terashima.
Tetrahedron Lett. 31, 3031(1990); K. Iizuka, T. Kamijo, T.
Kubota, K. Akahane, H. Umeyama, and Y. Kiso. J. Med. Chem.
31, 701 (1988); T. Aoyagi, H. Tobe, F. Kojima, M. Hamada,
T. Takeuchi, and H. Umezawa. J. Antibiot. 31, 636 (1978).
1 (3S,4R): mp 139–139.8 °C. [α]D = –170 (c 1.00 in
CH2Cl2). 3 (1R, 5S): mp 140 to 141 °C. [α]D = –110 (c 1.00
in CH2Cl2). 3 (1S, 5R) prepared from 2 (3R,4S) in 91%
yield: mp 139 to 140 °C. [α]D = +108 (c 1.00 in CH2Cl2). 2
(3R, 4S): mp 158 to 159 °C. [α]D = +13.4 (c 1.00 in
CH3COCH3).
Procedure for baker’s yeast reduction of 8
A suspension of baker’s yeast (10 g) and dextrose (40 g)
in sterilized water (500 mL) was stirred at 30 °C in a
fermentor with aeration for 15 min. Compound 8 (1.00 g)
was added and stirring continued for 26 h. The reaction was
monitored by chiral HPLC. After complete conversion, the
reaction mixture was saturated with salt and centrifuged at
3000× g for 10 min to remove yeast cells. The supernatant
was extracted with ethyl acetate. The cell pellet was ex-
tracted three times with ethyl acetate. The combined organic
layers were washed with brine, dried over magnesium sul-
fate, and filtered. Vacuum evaporation afforded the crude
mixture (0.82 g) of optically pure 1-trans, and mixtures of 1-
cis and 3. The crude reaction product (0.29 g) was hydro-
lyzed with 2 N KOH as described before. Optically pure 1-
trans (3R,4R) was isolated as a colorless oil (0.029 g); 1-cis
and 3 were hydrolyzed to the racemic acid 2 and were iso-
lated from the aqueous layer as a colorless solid (0.17 g). 1
(3R,4R): [α]D = +84.9 (c 1.65 in CH2Cl2). 1H NMR
(250 MHz, CDCl3) δ: 1.17 (3H, s, CCH3), 1.18 (3H, t, J =
7.10, CH2CH3), 1.28 (3H, s, CCH3), 3.78 (3H, s, OCH3),
4.05 (2H, q, J = 7.05, CH2CH3), 4.35 (H, d, J = 1.70, NCH),
4.75 (1H, d, J = 1.6, OHCH), 6.82 (2H, d, J = 9.2, ArH),
7.14 (2H, d, J = 9.2).2
3. R.A. Holton, R.J. Biedinger, and P.D. Boatman. Semisynthesis
of Taxol® and Taxotere. In Taxol® science and applications.
Edited by M. Suffness. Boca Raton, CRC Press, Florida. 1995.
4. S. Lin, X. Geng, C. Qu, R. Tynebor, D.J. Gallagher, E. Pollina,
J. Rutter, and I. Ojima. Chirality, 12, 431 (2000); I. Ojima, S.
Lin, J.C. Slater, T. Wang, P. Pera, R.J. Bernacki, C. Ferlini,
and G. Scambia. Bioorg. Med. Chem. 8, 1619 (2000); T.
Yamaguchi, N. Harada, K. Ozaki, H. Arakawa, K. Oda, N.
Nakanishi, K. Tsujihara, and T. Hashiyama. Bioorg. Med.
Chem. Lett. 9, 1639 (1999); S.M. Ali, M.Z. Hoemann, J.
Aube, G.I. Georg, and L.A. Mitscher. J. Med. Chem. 40, 236
(1997); I. Ojima, J.C. Slater, E. Michaud, S.D. Kuduk, P.-Y.
Bounaud, P. Vrignaud, M.-C. Bissery, J.M. Veith, P. Pera, and
R.J. Bernacki. J. Med. Chem. 39, 3889 (1996); G.I. Georg,
T.C. Boge, Z.S. Cheruvallath, J.S. Clowers, G.C.B. Harriman,
M. Hepperle, and H. Park. The medicinal chemistry of Taxol®.
In Taxol® science and applications. Edited by M. Suffness.
Boca Raton, CRC Press, Florida. 1995; S.M. Ali, M.Z.
Hoemann, J. Aubé, L.A. Mitscher, and G.I. Georg. J. Med.
Chem. 38, 3821 (1995); G.I. Georg, G.C.B. Harriman, M.
Hepperle, and R.H. Himes. Bioorg. Med. Chem. Lett. 4, 1381
(1994).
5. For recent reviews see: X. Wang, H. Itokawa, and K.-H. Lee.
In Taxus: The genus taxus. Medical and aromatic plants — In-
dustrial profiles. Vol. 32. Edited by H. Itokawa and K.H. Lee.
Taylor and Francis, London. 2003; D.G.I. Kingston. Chem.
Commun. 867 (2001); D.G.I. Kingston. J. Nat. Prod. 63, 726
(2000).
6. T. Konno, J. Watanabe, and K. Ishihara. J. Biomed. Mater.
Res. 65A, 210 (2003); A.L. Seligson, R.C. Terry, J.C. Bressi,
J.G. Douglass III, and M. Sovak. Anti-Cancer Drugs, 12, 305
(2001), and refs. cited therein; M.I. Nicoletti, T. Colombo, C.
Rossi, C. Monardo, S. Stura, M. Zucchetti, A. Riva, P.
Morazzoni, M.B. Donati, E. Bombardelli, M. D’Incalci, and R.
Giavazzi. Cancer Res. 60, 842 (2000); D. Sampath, C.M.
Discafani, F. Loganzo, C. Beyer, H. Liu, X. Tan, S. Musto, T.
Annable, P. Gallagher, C. Rios, and L.M. Greenberger. Mol.
Cancer Ther. 2, 873 (2003). For a review see: I. Ojima, S. Lin,
and W. Tao. Curr. Med. Chem. 6, 927 (1999).
Acknowledgements
Financial support by the Natural Sciences and Engi-
neering Research Council of Canada (NSERC) (FDR and
MMK) and Medical Research Fund of New Brunswick
(MMK) are gratefully acknowledged. The NMR spectra
were recorded at the Atlantic Regional Magnetic Resonance
Centre at Dalhousie University, Halifax, Nova Scotia, Can-
ada. The authors thank Dr. D.L. Hooper for the spectra with
chiral solvating agents. The help of Stéphane M. Foulem, an
undergraduate participant, is gratefully acknowledged.
References
1. For reviews on β-lactam antibiotics see: (a) G.I. Georg and
V.T. Ravikumar. In The organic chemistry of β-lactams. Edited
by. G.I. Georg. VCH, New York. 1993; (b) R. Southgate, C.
Branch, S. Coulton, and E. Hunt. In Recent progress in the
chemical synthesis of antibiotics and related microbial prod-
ucts. Vol. 2. Edited by G. Lukacs. Springer-Verlag, Berlin.
1993. p. 621; (c) F.H. Van der Steen and G. van Koten. Tetra-
hedron, 47, 7503 (1991).
7. For reviews see: C. Palomo, J.M. Aizpurua, I. Ganboa, and M.
Oiarbide. Eur. J. Org. Chem. 3223 (1999). Some recent exam-
ples: M.D. Mihovilovic, A. Feicht, and M.M. Kayser. J. Prakt.
Chem. (Weinheim, Ger.), 342, 585 (2000); B.K. Banik and F.F.
Becker. Tetrahedron Lett. 41, 6551 (2000).
8. For reviews see: M. Benaglia, M. Cinquini, and F. Cozzi. Eur.
J. Org. Chem. 563 (2000); D.J. Hart and D.-C. Ha. Chem. Rev.
89, 1447 (1989). A recent example: S. Schunk and D. Enders.
Org. Lett. 2, 907 (2000).
2. C. Palomo, J.M. Aizpurua, I. Ganboa, and M. Oiarbide.
Synlett, 1813 (2001); C. Palomo, J.M. Aizpurua, and I. Ganboa.
2 Supplementary data may be purchased from the Directory of Unpublished Data, Document Delivery, CISTI, National Research Council
Canada, Ottawa, ON K1A 0S2, Canada (http://www.nrc.ca/cisti/irm/unpub_e.shtml for information on ordering electronically). CCDC
220029, 220030, and 225739 contain the crystallographic data for this manuscript. These data can be obtained, free of charge, via
www.ccdc.cam.ac.uk/conts/retrieving.html (or from the Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ,
UK; fax +44 1223 336033; or deposit@ccdc.cam.ac.uk).
© 2005 NRC Canada

36
9. J.R. Malpass. In Comprehensive organic chemistry: The syn-
Can. J. Chem. Vol. 83, 2005
W.J. Drury III, A.E. Taggi, D. Ferraris, and T. Lectka. Org.
Lett. 1, 1985 (1999).
thesis and reactions of organic compounds. Vol. 2. Edited by
D. Barton and W.D. Ollis. Pergamon Press, Oxford. 1979. p.
22; M. Chmielewski, Z. Kaluza, and B. Furman. Chem.
Commun. 2689 (1996).
13. (a) R. Brieva, J.Z. Crich, and C.J. Sih. J. Org. Chem. 58, 1068
(1993); (b) E. Forró and F. Fülöp. Tetrahedron: Asymmetry,
12, 2351 (2001); (c) W. Adam, P. Groer, H.-U. Humpf, and
C.R. Saha-Möller. J. Org. Chem. 65, 4919 (2000); (d) A.
Basak, T. Mahato, G. Bhattacharya, and B. Mukherjee. Tetra-
hedron Lett. 38, 643 (1997); (e) J.A. Carr, T.F. Al-Azemi, T.E.
Long, J.-Y. Shim, C.M. Coates, E. Turos, and K.S. Bisht. Tet-
rahedron, 59, 9147 (2003).
10. A. Basak, S.C. Ghosh, T. Bhowmick, A.K. Das, and V.
Bertolasi. Tetrahedron Lett. 43, 5499 (2002); M.M.-C. Lo and
G.C. Fu. J. Am. Chem. Soc. 124, 4572 (2002); M. Miura, M.
Enna, K. Okuro, and M. Nomura. J. Org. Chem. 60, 4999
(1995); M. Kinugasa and S. Hashimoto. J. Chem. Soc. Chem.
Commun. 466 (1972).
14. M.M. Kayser, Y. Yang, M.D. Mihovilovic, A. Feicht, and F.D.
11. M. Arun, S.N. Joshi, V.G. Puranik, B.M. Bhawal, and
A.R.A.S. Deshmukh. Tetrahedron, 59, 2309 (2003); R.
Fernández, A. Ferrete, J.M. Lassaletta, J.M. Llera, and E. Mar-
tin-Zamora. Angew. Chem. Int. Ed. 41, 831(2002); S.N. Joshi,
A.R.A.S. Deshmukh, and B.M. Bhawal. Tetrahedron: Asym-
metry, 11, 1477 (2000); S. Brown, A.M. Jordan, N.J. Law-
rence, R.G. Pritchard, and A.T. McGown. Tetrahedron Lett.
39, 3559 (1998); I. Ojima and F. Delaloge. Chem. Soc. Rev.
26, 377 (1997); S. Matsui, Y. Hashimoto, and K. Saigo. Syn-
thesis, 1161 (1998); Y. Hashimoto, A. Kai, and K. Saigo. Tet-
rahedron Lett. 36, 8821 (1995); T. Fujisawa, A. Shibuya, D.
Sato, and M. Shimizu. Synlett, 1067 (1995).
Rochon. Can. J. Chem. 80, 796 (2002).
15. N.C. Deno. J. Am. Chem. Soc. 69, 2233 (1947).
16. It is generally recognized that cis- β-lactams have larger cou-
pling constants than the corresponding trans diastereomers. C.
Palomo, J.M. Aizpurua, F. Cabré, J.M. Garcia, and J.M.
Odriozola. Tetrahedron Lett. 35, 2721 (1994). Also see B.K.
Banik and F.F. Becker, in ref. 7. In the case of compound 7,
the J coupling constant for cis isomers is 5.3 Hz, whereas that
for the trans is 1.8 Hz.
17. M.M. Kayser, M.D. Mihovilovic, J. Kearns, A. Feicht, and
J.D. Stewart. J. Org. Chem. 64, 6603 (1999).
18. XSCANS. PC Version 5 [computer program]. Bruker AXS
Inc., Madison, Wisconsin. 1995.
19. SHELXTL. Release 5.10 [computer program]. Bruker AXS
Inc., Madison, Wisconsin. 1997.
12. P.A. Magriotis. Angew. Chem. Int. Ed. 40, 4377 (2001); B.L.
Hodous and G.C. Fu. J. Am. Chem. Soc. 124, 1578 (2002);
A.E. Taggi, A.M. Hafez, H. Wack, B. Young, W.J. Drury III,
and T. Lectka. J. Am. Chem. Soc. 122, 7831 (2000); H. Wack,
© 2005 NRC Canada