Novel stereoselective synthesis of 4-acetoxyazetidinone from methyl 6,6-dibromopenicillanate: Key intermediate for the preparation of carbapenem antibiotics

Article abstract of DOI:10.1080/00397910902883637

A novel and practical process for the completely stereoselective synthesis of carbapenem key intermediate (3R,4R)-4-acetoxy-3-[(R)-1-((t- butyldimethylsilyl)oxy)ethyl]-2-azetidinone has been developed by starting from methyl 6,6-dibromopenicillanate. Aldol condensation of the magnesium enolate derived from the-sulfoxide with acetaldehyde allows for the stereospecific introduction of a 1-R-hydroxyethyl substituent as C-6. The hydroxyethylated product was reduced with Zn-NH4OAc in tetrahydrofuran (THF) efficiently to give methyl 6-(1-hydroxyethyl)-penicillanate-1-oxide. Protection of the hydroxy group followed by treatment with 2-mercaptobenzothiazole afforded the dithioazetidinone, which was easily reduced and methylated to give 4-methylthioazetidinone. Then this compound was oxidized with permanganate to selectively remove the side chain at the N-position to afford the 4-methylthioazetidinone derivative. A facile conversion of the methylthio group to acetoxy for a practical synthesis of 4-acetoxyazetidinone was also reported. This method provided an efficient and cost-effective process with good overall yield and excellent stereoselectivity.

Full text of DOI:10.1080/00397910902883637

This article was downloaded by: [University of Minnesota Libraries, Twin
Cities]
On: 17 April 2013, At: 22:09
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954
Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH,
UK
Synthetic Communications: An
International Journal for Rapid
Communication of Synthetic
Organic Chemistry
Publication details, including instructions for
authors and subscription information:
http://www.tandfonline.com/loi/lsyc20
Novel Stereoselective Synthesis
of 4-Acetoxyazetidinone
from Methyl 6,6-
Dibromopenicillanate:
Key Intermediate for the
Preparation of Carbapenem
Antibiotics
Bohua Long a & Jiannan Xiang a
a College of Chemistry and Chemical Engineering,
Biomedical Engineering Center, State Key Laboratory
of Chemo/biosensing and Chemometrics, Hunan
University, Changsha, China
Version of record first published: 16 Oct 2009.
To cite this article: Bohua Long & Jiannan Xiang (2009): Novel Stereoselective
Synthesis of 4-Acetoxyazetidinone from Methyl 6,6-Dibromopenicillanate:
Key Intermediate for the Preparation of Carbapenem Antibiotics, Synthetic
Communications: An International Journal for Rapid Communication of Synthetic
Organic Chemistry, 39:22, 4019-4029
To link to this article: http://dx.doi.org/10.1080/00397910902883637
PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-
and-conditions
This article may be used for research, teaching, and private study purposes.
Any substantial or systematic reproduction, redistribution, reselling, loan,
sub-licensing, systematic supply, or distribution in any form to anyone is
expressly forbidden.
The publisher does not give any warranty express or implied or make any
representation that the contents will be complete or accurate or up to
date. The accuracy of any instructions, formulae, and drug doses should be
independently verified with primary sources. The publisher shall not be liable
for any loss, actions, claims, proceedings, demand, or costs or damages
whatsoever or howsoever caused arising directly or indirectly in connection
with or arising out of the use of this material.

Synthetic Communications1, 39: 4019–4029, 2009
Copyright # Taylor & Francis Group, LLC
ISSN: 0039-7911 print=1532-2432 online
DOI: 10.1080/00397910902883637
Novel Stereoselective Synthesis
of 4-Acetoxyazetidinone from Methyl
6,6-Dibromopenicillanate: Key Intermediate
for the Preparation of Carbapenem Antibiotics
Bohua Long and Jiannan Xiang
College of Chemistry and Chemical Engineering, Biomedical Engineering
Center, State Key Laboratory of Chemo=biosensing and Chemometrics,
Hunan University, Changsha, China
Abstract: A novel and practical process for the completely stereoselective
synthesis of carbapenem key intermediate (3R,4R)-4-acetoxy-3-[(R)-1-((t-
butyldimethylsilyl)oxy)ethyl]-2-azetidinone has been developed by starting from
methyl 6,6-dibromopenicillanate. Aldol condensation of the magnesium enolate
derived from the b-sulfoxide with acetaldehyde allows for the stereospecific
introduction of a 1-R-hydroxyethyl substituent as C-6. The hydroxyethylated
product was reduced with Zn-NH4OAc in tetrahydrofuran (THF) efficiently to
give methyl 6-(1-hydroxyethyl)-penicillanate-1b-oxide. Protection of the hydroxy
group followed by treatment with 2-mercaptobenzothiazole afforded the dithioa-
zetidinone, which was easily reduced and methylated to give 4-methylthioazetidi-
none. Then this compound was oxidized with permanganate to selectively remove
the side chain at the N-position to afford the 4-methylthioazetidinone derivative.
A facile conversion of the methylthio group to acetoxy for a practical synthesis of
4-acetoxyazetidinone was also reported. This method provided an efficient and
cost-effective process with good overall yield and excellent stereoselectivity.
Keywords: Aldol condensation, carbapenem antibiotics, debromination,
dithioazetidinone, stereoselective synthesis
Received November 12, 2008.
Address correspondence to Jiannan Xiang, College of Chemistry and
Chemical Engineering, Biomedical Engineering Center, State Key Laboratory
of Chemo=biosensing and Chemometrics, Hunan University, Changsha 410082,
China. E-mail: jnxiang@hnu.cn
4019

4020
B. Long and J. Xiang
INTRODUCTION
The carbapenem b-lactam antibiotics represented by imipenem and
panipenem have been the target of recent synthetic endeavors because
of their prominent antibacterial activities and broad spectra.[1] In several
syntheses of these antibiotics reported thus far,[2,3] the title compound,
(3R,4R)-4-acetoxy-3-[(R)-1-((t-butyldimethyllsilyl)oxy)ethyl]-2-azetidinone
(1), holds a pivotal position as one of the most versatile key synthetic
intermediates. Thus, various types of carbon chains required to construct
the five-membered ring fused with b-lactam can be readily introduced
into 1 by substituting its acetoxy group with nucleophiles.[4] The same
synthetic strategy has been also applied for producing the key inter-
mediates of 1b-methylcarbapenems such as ertapenem, chemically and
metabolically stable carbapenem antibiotics that show excellent anti-
bacterial activities.[5,6] Accordingly, a number of synthetic methods for
1 have hitherto been explored by employing various chiral compounds
as starting materials.[7]
To provide carpenems that have the correct absolute configuration
necessary for full antibacterial activity, an enantiospecific synthesis of
the azetidinone intermediate from the easily available and inexpensive
6-aminopenicillanic acid (6-APA) had been developed.[8–10] However,
several fundamental problems were persistent in the penicillin-based
routes. Aldol reaction with the sulfide oxidation level of the penicillin
gave a mixture of isomers that required recrystallization and=or chroma-
tography to isolate the desired 8R isomer. Subsequent isomerization of
the cis C5–C6 relationship, in the dominate 8R aldol isomer, to trans
was not complete (91:9) and required chromatography to isolate the pure
trans compound. Moreover, the ring cleavage of the thiazolidine ring
required the use of poisonous mercury(II) acetate, which is unsuitable
for large-scale pharmaceutical synthesis.
Therefore, it was necessary to reinvestigate the latter process to
develop practical methods for preparation of the stereochemically pure
4-acetoxyazetidinone (1) with appreciable yield and excellent stereoselec-
tivity. In this report, a reliable method for stereospecific introduction of
the trans-R-hydroxyethyl side chain onto 6,6-dibromopenicillin was
developed. Acetaldehyde condensation involving the b-sulfoxide contain-
ing Grignards reagent resulted in a significant improvement in stereocon-
trol because increased addition occured from the more congested b-face. A
greatly improved method for debromination of the intermediate, methyl
(3S,5R,6S)-6-bromo-6-[(R)-1-hydroxyethyl]penicillanate-1b-oxide (5), was
also developed after a systematic study of various reduction methods. We
also employed a recently reported method for conversion of the azetidinone
derivative (10) with a 4-methylthio group into 4-acetoxyazetidinone 1 using

Synthesis of 4-Acetoxyazetidinone
4021
Scheme 1. Synthesis of 4-acetoxyazetidinone 1. Reagents and conditions: (a)
CH3MgCl, THF, ꢀ78ꢁC; then CH3CHO (90% yield); (b) Zn, THF, aq. NH4OAc,
rt (80% yield); (c) TBDMSCl, imidazole, DMF, rt (88% yield); (d) 2-mercapto-
benzothiazole, toluene, reflux, 2 h; then Et3N (0.1 mol equiv.), toluene, 5ꢁC, 2 h
(100% yield); (e) NaBH4, CH3I, DMF, 0ꢁC (80% yield); (f) KMnO4, buffer
Me2CO, 0ꢁC (85% yield); and (g) NBS, CH3CO2H, CH2Cl2, 0ꢁC (82% yield).
N-bromosuccinimide as the reactant rather than the heavy-metal reactant.
Except for the final recrystallization, no chromatographic purification was
required for the process. This article described the improvements that were
necessary to achieve a practical synthesis of 1 in good overall yield with
excellent stereoselectivity (Scheme 1).
RESULTS AND DISCUSSION
6-Aminopenicillanic acid (6-APA) was diazotized and brominated by a
reported procedure[11] to give 6,6-dibromopenicillanic acid in 85% yield.
Conversion into its methyl ester (2) was readily achieved on a large scale
with sodium hydrogen carbonate and methyl iodide in an aprotic
solvent.[11] Methyl 6,6-dibromopenicillanate (2) was of interest as a start-
ing point in the preparation of other potentially interesting penicillanate
derivatives. Oxidation of the compound 2 with 40% peracetic acid in

4022
B. Long and J. Xiang
methylene chloride gave only one isomer, whose configuration was
determined by 1H NMR and 13C NMR spectra. The physical and
spectral characteristics showed this compound from the peroxy-acid
oxidation was in fact the 1b-sulfoxide (3), which had been reported in
prior literature[11]: mp 132–134ꢁC (lit. mp 133–135ꢁC).
Since the initial report of DiNinno et al.[12] on the regiospecific
hydroxyethylation of the C-6 position of penicillinates, many research
groups had used this approach to prepare penem and carbapenem
precursors. This method proved to be useful even though the aldol con-
densation suffered from a lack of stereospecificity. Subsequently, Brown
and Volkmann gave a thorough discussion of aldol selectivity with
various halogens at C-6 and sulfur oxidation state.[13] This work sug-
gested that complete stereoselectivity in the aldol reaction could be
achieved if the 1b-sulfoxide was involved in acetaldehyde condensations.
Indeed, transmetalation of the 1b-sulfoxide (3) with methyl magne-
sium chloride in tetrahydrofuran (THF) at ꢀ78ꢁC and subsequent addi-
tion of excess acetaldehyde after workup produced only a single aldol
adduct 5 in good yield, which had the desired R configuration at the
OH bearing carbon. This reaction was also markedly influenced by the
nature of the solvent and the type of the metal cation present in the solu-
tion as well as the halide atom of the molecule involved.[13] For compar-
ison, in the sulfide series, the aldol reaction of 2 produced a mixture of all
four diastereomers in the ratio of 72:19:5:3.[12] The dominant sulfide aldol
isomer 4 had been converted into acetoxyazetidinone 1.[10] Sulfide oxida-
tion of this dominant aldol isomer 4, methyl (3S,5R,6S)-6-bromo-6-
[(R)-1-hydroxyethyl]penicillanate, with 40% peracetic acid in methylene
chloride produced a compound identical to 1b-sulfoxide aldol adduct 5
by high-field proton and carbon NMR.
We believed the THF Grignard results were consistent with the
intermediacy of the THF-stabilized tetrahedral conbanions. Carbanionic
localization in the more hindered b-face of the 1b-sulfoxide would be sta-
bilized by sulfoxide chelation and furthermore would eliminate strain
caused by interaction of the remaining bromine atom, the thiazolidine
b-methyl group, and the proximal sulfoxide moieties.
Reduction of the bromine atom in methyl (3S,5R,6S)-6-bromo-6-
[(R)-1-hydroxyethyl]penicillanate-1b-oxide (5) was attempted by catalytic
hydrogenation employing palladium on carbon, but a mixture of cis and
trans isomers was generated with varying cis=trans ratios at C5=C6 from
3=1 to 1=1. Other reduction methods by Zn-AcOH, Zn-AcOH-MeOH, or
tri-n-butytin hydride gave poor yields (less than 60%). The zinc–silver
coupling reduction of 5 in methanol gave the desired trans isomer 6 in
70% yield with excellent stereoselectivity. However, this method was less
desirable on a large scale. Consequently we studied the debromination in

Synthesis of 4-Acetoxyazetidinone
4023
detail and found that simply stirring with 4 molar equivalents of Zn in a
mixture of THF and aqueous ammonium acetate at room temperature
afforded 6 almost exclusively in 80% yield.
The protection of the hydroxyl group in 6 was achieved by treatment
with t-butyldimethylsilyl chloride and imidazole in dimethylformamide
(DMF) at room temperature to give 7 in 90% yield.
Dithioazetidinones, easily obtained from reactions of penicillin
sulfoxides with heteroaromatic thiols, had been proved to be key inter-
mediates, which could be easily reduced and further alkylated=acylated
to give various sulfides at a later stage in the synthetic sequence.[14]
Treatment of the penicillin sulfoxide (7) with 2-mercaptobenzothiazole
in refluxing toluene for 1 h afforded a crystalline disulfide, which was
transformed into the conjugated ester disulfide 8 by base-catalyzed
double-bond isomerization (quantitative yield from 7).
Sodium borohydride (NaBH4) was a readily available, mild reducing
agent and was easy to manipulate. On the other hand, reduction of dis-
ulfide to the corresponding thiol was an important reaction both in
bio- and organic chemistry.[15] Reductive alkylation of the disulfide (8)
by sodium borohydride in dimethylformamide with subsequent addition
of CH3I afforded the 4-methyl thioazetidinone (9) in one step in 85%
yield. Similarly, 4-acetylthioazetidinone could be prepared if an appropri-
ate acyl halide or anhydride was added. For the present synthesis, we
sought to apply this method to this more complex side chain. The details
of this study will be disclosed later.
As shown in Scheme 1, a convenient procedure for the removal of
ab-unsaturated N-substituent from the azetidinone 9 was treatment with
potassium permanganate at ice-bath temperature. When the compound 9
was oxidized with aqueous potassium permanganate in acetone contain-
ing a little acetic acid, removal of the N-substituent was accompanied by
oxidation at sulfur, the product being the sulfone. However, oxidation at
sulfur did not occur if acidic reaction conditions were avoided. Finally, it
was convenient to carry out the oxidation in phosphate buffer to give the
desired 4-methylthioazetidinone 10 in 85% yield. Permanganate oxida-
tion complements these procedures and had proved to be a key step in
selective removal of N-isopropylidineacatate group to give the desired
simple azetidinone in a single operation.
Traditionally, the solvolytic conversion of the methylthio group of 10
to acetoxy group was facilitated by poisonous mercury salt as a con-
sequence of its high affinity toward sulfur. The method using copper salt
had been proposed as an alternative method.[16] Although the copper salt
was less poisonous than mercury salt, the copper itself was a heavy metal,
and thus the problem related to the disposal of reaction waste could
not be completely solved. Moreover, such heavy-metal reactants are

4024
B. Long and J. Xiang
expensive and thus are not preferable. To overcome these disadvantages,
a novel and efficient process had been recently reported for preparing
4-acetoxyazetidinone 1 by reacting the azetidinone derivative 10 with
N-bromosuccinimide in the presence of an organic carboxylic acid, which
was economical and could be safely handled.[17]
CONCLUSIONS
In conclusion, a highly efficient and convenient synthesis of the very
useful carbapenem precursor 4-acetoxyazetidinone 1 was described in
about 32% overall yield. 6,6-Dibromopenicillanate-1b-sulfoxide was
employed as a key intermediate and involved in the crucial acetaldehyde
condensation with complete stereoselectivity. In addition, a greatly
improved Zn-NH4OAc-THF reduction method was developed in this
process. N-Bromosuccinimide was used to overcome the drawbacks of
the heavy-metal reactant to convert the methylthio group to acetoxy.
Each step did not use chromatographic purification. Its ease of workup,
good yields, excellent stereoselectivity, and mild reaction conditions made
this process a very practical alternative to the existing methods.
EXPERIMENTAL
General
Melting points were measured with a XRC-1 melting-point apparatus
and are uncorrected. Fourier transform (FT)–infrared (IR) spectra were
recorded in KBr disks using an FD-5DX spectrometer. Mass spectra
1
were carried out on a ZAB-HS mass spectrometer. H NMR and 13C
NMR spectra were recorded in CDCl3 at room temperature on a Varian
Inova-400 spectrometer (400 MHz). The chemical-shift scale is based on
internal tetramethylsilane (TMS).
Methyl 6,6-Dibromopenicillanate-1b-oxide (3)
Peracetic acid (40%, 26.6 g, 0.14 mol) was added dropwise over 30 min
to a stirred and ice–cold mixture consisting of methyl 6,6-dibromopenicil-
lanate 2 (37.4 g, 0.10 mol), methylene chloride (200 mL), and water
(100 mL). The mixture was stirred at 0ꢁC for 2 h. The organic layer
was separated, and the aqueous layer was back-extracted with methylene
chloride (3 ꢂ 200 mL). The combined organic layers were washed with

Synthesis of 4-Acetoxyazetidinone
4025
cold water (200 mL) and stirred for 30 min with saturated sodium
hydrogen carbonate (200 mL) at 10ꢁC. The organic layer was washed
with brine, dried with MgSO4, and concentrated to give the product as
a light yellow solid (35.0 g, 90%). Mp 120–122ꢁC. IR (KBr): 1794,
1
1754 cmꢀ1; H NMR (CDCl3): d 1.43 (s, 3H, CH3), 1.58 (s, 3H, CH3),
3.84 (3H, s, OCH3), 4.60 (s, 1H, 3-H), 5.32 (s, 1H, 5a-H).
Methyl (3S,5R,6S)-6-Bromo-6-[(R)-1-hydroxyethyl]penicillanate-
1b-oxide (5)
Methyl 6,6-dibromopenicillanate-1b-oxide 3 (23.4 g, 0.06 mol) and THF
(200 mL) were combined at 25ꢁC in a three-neck, round-bottom flask
equipped with a low-temperature thermometer. Upon cooling to ꢀ78ꢁC,
methylmagnesium chloride (28 ml, 2.5 M in THF, 0.07 mol) was added
dropwise, maintaining the temperature at less than ꢀ75ꢁC. Thirty min
after Grignard addition, distilled acetaldehyde (10 mL) was added over
10 min and stirred for 30 min. The reaction was quenched at ꢀ78ꢁC with
200 mL of saturated aqueous NH4Cl. After the solution was warmed to
20ꢁC, ethyl acetate (200 mL) and water (200 mL) were added. The layers
were separated, and the aqueous layer was extracted with ethyl acetate
(2 ꢂ 200 mL). The organic layers were combined and washed with satu-
rated NaHCO3 (200 mL) and brine (200 mL). Treatment of the organic
layer with MgSO4 and removal of solvent under vacuum produced an
off-white solid (19.2 g, 90%). Mp 135–137ꢁC. IR (KBr): 3570, 1807,
1
1120 cmꢀ1. H NMR (400 MHz, CDCl3): d 1.37 (d, J ¼ 6 Hz, 3H), 1.52
(s, 3H), 1.61 (s, 3H), 2.05 (s, 3H), 3.83 (s, 3H), 4.26 (q, J ¼ 6 Hz, 1H),
4.48 (s, 1H), 4.89 (s, 1H). Anal. calcd. for C11H16BrNO5S (354.22): C,
37.30; H, 4.55; N, 3.95; S, 9.05. Found: C, 37.18; H, 4.40; Br, N, 3.95;
S, 9.37.
Methyl (3S,5R,6S)-6-[(R)-1-Hydroxyethyl]penicillanate-1b-oxide (6)
A solution of 14.2 g (0.04 mol) of the aldol 5 in 150 mL of THF was
treated with 200 mL of 1 M aqueous NH4OAc and 10.4 g (0.16 mol) of
powdered zinc. The resulting mixture was stirred at room temperature
for 1 h and then filtered through a bed of celite. The organic layer was
separated from the aqueous layer, which was extracted twice with
methylene chloride (3 ꢂ 200 mL). The combined organic extracts were
combined, washed with brine (200 mL), dried with MgSO4, and then
concentrated under reduced pressure to give 6 in about 80% yield as an
oil (8.82 g). This alcohol was used directly in the next step. It could be

4026
B. Long and J. Xiang
separated by column chromatography and its individual characteristics
are listed here. IR (KBr): 3585, 1765, 1175, 1120 cmꢀ1 1H NMR
.
(400 MHz, CDCl3): d 1.39 (d, J ¼ 6 Hz, 3H), 1.54 (s, 3H), 1.62 (s, 3H),
3.21 (bs, 1H), 3.80 (dd, J ¼ 2, J ¼ 6 Hz, 1H), 3.94 (s, 3H), 4.48 (m, 1H),
4.52 (s, 1H), 4.86 (d, J ¼ 2 Hz, 1H). Anal. calcd. for C11H17NO5S
(275.32): C, 47.99; H, 6.22; N, 5.09; S, 11.65. Found: C, 48.09; H, 6.18;
N, 5.05; S, 11.36.
Methyl (3S,5R,6S)-6-[(R)-l-((t-Butyldimethylsi1yl)oxy)ethyl]
penicillanate-1b-oxide (7)
A solution of 8.26 g (0.03 mol) of the alcohol 6 in 100 mL dry DMF was
treated with imidazole (4.10 g, 0.06 mol) and t-butyldimethylsilyl chloride
(5.4 g, 0.036 mol). The resulting mixture was stirred at room temperature
overnight, diluted with 400 ml of methylene chloride, and washed with
water (3 ꢂ 100 mL) and 100 mL brine. The methylene chloride solution
was dried over MgSO4, filtered, and evaporated under vacuum to provide
7 as an oil (10.28 g, 88%). This silyl ether was employed directly in the
next step. It could be separated by column chromatography, and its
individual characteristics are listed here.
1
IR (KBr): 2950, 2920, 1800, 1460,1118 cmꢀ1. H NMR (400 MHz,
CDCl3): d 0.07 (s, 6H), 0.89 (s, 9H), 1.30 (d, 6 Hz, 3H), 1.38 (s, 3H),
1.54 (s, 3H), 3.55 (dd, J ¼ 2, J ¼ 2.4, 1H), 3.78 (s, 3H), 4.38 (m, 1H),
4.41 (s, 1H), 4.74 (d, J ¼ 1.6 Hz, 1H). Anal. calcd. for C17H31NO5SSi
(389.58): C, 52.41; H, 8.02; N, 3.60; S, 8.23. Found: C, 52.49; H, 8.00;
N, 3.60; S, 8.29.
(3R,4R)-4-(Benzothiazol-2-yldithio)-3-[(R)-1-((t-butyldimethylsilyl)oxy)
ethyl]-1-(l-methoxycarbonyl-2-methyl-1-propenyl)-2-azetidinone (8)
A mixture of the sulfoxide 7 (7.8 g, 0.02 mol) and 2-mercaptoben-
zothiazole (3.34 g, 0.02 mol) was heated in refluxing toluene (200 mL)
with provision for the azeotropic removal of water (Dean–Stark appa-
ratus) for 1 h. The mixture was cooled to 5ꢁC and treated with triethy-
lamine (0.2 g, 2 mmol). After being stirred at 5ꢁC for 2 h, the mixture
was washed successively with 5% aq. citric acid (2 ꢂ 200 mL), water
(200 mL), and brine (200 mL). The dried (MgSO4) organic layer was
evaporated to give the disulfide 8 (10.78 g, 100%) as an oil. This
disfide was employed directly in the next step. It could be separated
by column chromatography, and its individual characteristics are listed
here.

Synthesis of 4-Acetoxyazetidinone
4027
1
IR (neat): 1770, 1726 cmꢀ1. H NMR (400 MHz, CDCl3): d 0.07
(6H, s), 0.87 (9H, s), 1.27 (3H, d, J ¼ 6 Hz), 1.89 (3H, s), 2.07 (3H, s),
3.34 (1H, dd, J ¼ 2.8, 2.4 Hz), 3.73 (3H, s), 4.28 (1H, m), 5.54 (1H, d,
J ¼ 2.4 Hz).
(3R,4R)-4-Methylthio-3-[(R)-1-((t-butyldimethylsilyl)oxy)ethyl]-
1-(1-methoxycarbonyl-2-methyl-1-propenyl)-2-azetidinone (9)
A solution of 1.0 g of NaBH4 in 20 mL of DMF was added to a
stirred solution of the disulfide 8 (10.78 g, 0.02 mol) in 100 mL of
DMF at 0ꢁC. After 30 min at 0ꢁC, 3.8 mL (0.06 mol) of CH3I was
introduced, and the resulting reaction mixture was stirred at room
temperature for 40 min. Ethyl acetate was then added, and some inso-
luble material was removed by filtration. After washing the filtrate
several times with ethyl acetate, evaporation of the solvent afforded
3.1 g (80%) of the compound 9 as a colorless oil. This azetidinone
derivative was employed directly in the next step. It could be sepa-
rated by column chromatography, and its individual characteristics
are listed here.
IR (neat): 1780, 1730, 1640 cmꢀ1. 1H NMR (400 MHz, CDCl3) d: 0.08
(6H, s), 0.84 (9H, s), 1.20 (3H, d, J ¼ 6 Hz), 1.98 (3H, s), 2.12 (3H, s), 2.23
(3H, s), 3.11(1H, dd, J ¼ 5, 2.5 Hz), 3.76 (3H, s), 4.27 (1H, m), 5.13 (1H,
d, J ¼ 2.5 Hz). Anal. calcd. for C18H33NO4SSi (387.61): C, 55.81%; H,
8.53%; N, 3.62%; S, 8.29%. Found: C, 55.44%; H, 8.70%; N, 3.42%;
S, 8.45%.
(3R,4R)-4-Methylthio-3-[(R)-1-((t-butyldimethylsilyl)oxy)ethyl]-
2-azetidinone (10)
A solution of 6.20 g (0.016 mol) of the azetidinone derivative 9 in acetone
(100 mL), cooled in ice, was treated with potassium permanganate (4.0 g)
in water (40 mL) and pH 7 phosphate buffer (40 mL) in portions over 1 h.
The product was extracted with ethyl acetate (3 ꢂ 200 mL), then washed
with brine (200 mL), dried (MgSO4), and evaporated to give a colorless
crystalline substance 10 (3.74 g, 85%). Mp 75–82ꢁC. IR (KBr): 3060,
1
1755, 1710, 1250 cmꢀ1. H NMR (400 MHz, CDCl3): d 0.10 (s, 6 H),
0.90 (s, 9 H), 1.26 (d, J ¼ 6 Hz, 3H), 2.12 (s, 3H), 3.35 (dd, J ¼ 1.5,
J ¼ 4 Hz, 1H), 4.40 (m, 1H),5.83 (d, J ¼ 4.4 Hz, 1H), 6.80 (brs, 1H).
[a]D: þ26.2 (c ¼ 0.73, CHCl3). Anal. calcd. for C14H25NO2SSi (299.5):
C, 52.32; H, 9.15; N, 5.08; S, 11.64. Found: C, 52.24; H, 9.08; N, 4.91;
S, 10.95.

4028
B. Long and J. Xiang
(3R,4R)-4-acetoxy-3-[(R)-1-((t-butyldimethylsilyl)oxy)ethyl]-
2-azetidinone (1)
3.74 g (0.014 mol) of 4-methylthioazetidinone 10 and 3.30 ml (0.028 mol)
of acetic acid were dissolved in 100 mL of dichloromethane and 2.66 g
(0.015 mmol) of N-bromosuccinimide was added thereto at 0ꢁC. The
reaction solution was then stirred for 30 min at room temperature and
then aqueous solution of sodium thiosulfate, washed successively with
water, aqueous sodium bicarbonate solution and brine, dried over
MgSO4 and then filtered. The filtrate was concentrated under reduced
pressure and recrystallized from hexane to give 3.20 g (82%) of the
4-acetoxyazetidinone 1 as fluffy white needles. mp 105–106ꢁC. [a]D:
þ48.0 (c ¼ 0.56, CHCl3). IR (KBr): 3600–3100, 1790, 1750 cmꢀ1
.
1H
NMR (400 MHz, CDCl3): d 0.06 (s, SiCH3), 0.08 (s, SiCH3), 0.88
(s, Si-t-C4H9), 1.26 (d, J ¼ 6.4 Hz, CH3CH), 2.11(s, CH3CO2), 3.19 (dd,
J ¼ 1.3 Hz and 3.4 Hz, H3), 4.22 (dq, J ¼ 3.4 Hz and 6.3 Hz, CH3CH),
6.48 (br, s, NH); 5.84 (d, 1H, J ¼ 1.3 Hz, H4). 13C NMR (100 MHz,
CDCl3): d 17.88, 20.84, 22.28, 25.64, 63.83, 65.02, 75.10, 77.00,
166.48, 171.22.
ACKNOWLEDGMENTS
The authors are grateful for the financial support of the 985 Foundation
of the Ministry of Education. This work was supported by the National
Natural Science Foundation of China (Grant No. 20472018), the Natural
Science Foundation of Hunan (Key Project No. 07JJ3019), and the
Doctoral Fund of the Ministry of Education of China (Grant No.
20060532022).
REFERENCES
1. Southgate, R.; Branch, C.; Coulton, S.; Hunt, E. In Recent Progress in the
Chemical Synthesis of Antibiotics and Related Microbial Products; G. Lukas
(Ed.); Springer-Verlag: Berlin, 1993; vol. 2, p. 621.
2. Humphrey, G. R.; Miller, R. A.; Pye, P. J.; Rossen, K.; Reamer, R. A.;
Maliakal, A.; Ceglia, S. S.; Grabowski, E. J. J.; Volante, R. P.; Reider, P. J.
Efficient and practical synthesis of a potent anti-MRSA b-methylcarbapenem
containing a releasable side chain. J. Am. Chem. Soc. 1999, 121, 11261–11266.
3. Miyazaki, S.; Hosoyama, T.; Furuya, N. In vitro and in vivo antibacterial
activities of L-084, a novel oral carbapenem, against causative organisms
of respiratory tract infections. Antimicrob. Agents Chemother. 2001, 45,
203–207.

Synthesis of 4-Acetoxyazetidinone
4029
4. Tewari, N.; Nizar, H.; Rai, B. P.; Mane, A.; Prasad, M. Development of a
scalable process for 1-b-methyl azetidinone: A carbapenem key Intermediate.
Org. Process Res. Dev. 2005, 9, 827–829.
5. Sunagawa, M.; Sasaki, A. The structure aspects of carbapenem antibiotics.
Heterocycles 2001, 54, 497–528.
6. Cunha, B. A. A review of its microbiologic, pharmacokinetic, and clinical
aspects. Drugs Today 2002, 38, 195–213.
7. Berks, A. H. Preparations of two pivotal intermediates for the synthesis of
1-b-methyl carbapenem antibiotics. Tetrahedron 1996, 52, 331–375, and
references cited therein.
8. Leanza, W. J.; DiNinno, F.; Muthard, D. A.; Wilkening, R. R.; Wildonger,
K. J.; Ratcliffe, R. W.; Christensen, B. G. An efficient synthesis of
2-substituted-thio-6-hydroxyethyl-penem-3-carboxylic acids via 2-thioxope-
nams. Tetrahedron 1983, 39, 2505–2513.
9. Fujimoto, K.; Iwano, Y.; Hirai, K.; Sugawara, S. From penicillin to penem
and carbapenem, VII: Synthesis and antibacterial activity of penem deriva-
tives. Chem. Pharm. Bull. 1986, 34, 999–1014.
10. Goo, Y. M.; Lee, Y. B.; Kim, H. H.; Lee, Y. Y.; Lee, W. Y. Synthetic studies
on penems and carbapenems: Practical preparation of (3R,4R)-4-acetoxy-3-
[(1R)-1-hydroxyethyl] azetidin-2-one derivatives from 6-aminopenicillanic
acid. Bull. Korean Chem. Soc. 1987, 8, 15–19.
11. Micetich, R. G.; Maiti, S. N.; Tanaka, M.; Yamazaki, T.; Ogawa, K. Studies
on 6-halo- and 6,6-dihalopenicillins: The rearrangement of methyl 6,6-
dibromopenicillanate to 1,4-thiazepine. J. Org. Chem. 1986, 51, 853–858.
12. DiNinno, F.; Beattie, T. R.; Christensen, B. G. Aldol condensations of
regiospecific penicillanate and cephalosporanate enolates: Hydroxyethylation
at C-6 and C-7. J. Org. Chem. 1977, 42, 2960–2965.
13. Brown, B. B.; Volkmann, R. A. Stereoselectivity in 6-halopenicillanate
Grignard reactions. Tetrahedron Lett. 1986, 27, 1545–1548.
14. Lakouraj, M. M.; Movassagh, B.; Fadae, Z. Synthesis of organic sulfides via
Zn=AlCl3 system in aqueous media. Synth. Commun. 2002, 32, 1237–1242.
15. Ookawa, A.; Yokoyama, S.; Soai, K. Chemoselective reduction of diaryl
bisulfides to thiols with sodium borohydride in mixed solvent containing
methanol. Synth. Commun. 1986, 16, 819–825.
16. Nakatsuka, T.; Iwata, H.; Tanaka, R.; Lmajo, S.; Ishiguro, M. A facile con-
version of the phenylthio group to acetoxy by copper reagents for a practical
synthesis of 4-acetoxyazetidin-2-one derivatives from (R)-butane-1,3-diol.
J. Chem. Soc., Chem. Commun. 1991, 662–664.
17. Moon, C. J.; Baik, K. U.; Oh, S. H.; Kim, J. W.; Lee, J. H. Acyloxylation
process for preparing 4-acyloxy-2-azetidinone derivatives. US Patent
5,639,878, 1997.