A catalytic asymmetric route to carbapenems

Article abstract of DOI:10.1021/ol901269d

Image Presented Efficient syntheses of N-acetyl thienamycin and epithienamycin A in their readily deprotected form are reported where three contiguous stereocenters are established in a single catalytic asymmetric azetidinone-forming reaction. These examples are a template for synthesizing C-5/C-6 cis or trans carbapenems with independent control of the C-8 stereocenter. A library of oxidatively and sterochemically defined azetidinone precursors to a variety of naturally occurring carbapenems and potential biosynthetic intermediates has been prepared to facilitate studies of carbapenem antibiotic biosynthesis.

Full text of DOI:10.1021/ol901269d

ORGANIC
LETTERS
2009
Vol. 11, No. 16
3606-3609
A Catalytic Asymmetric Route to
Carbapenems
Micah J. Bodner,† Ryan M. Phelan,† and Craig A. Townsend*
Department of Chemistry, Johns Hopkins UniVersity, 3400 North Charles Street,
Baltimore, Maryland 21218
ctownsend@jhu.edu
Received June 6, 2009
ABSTRACT
Efficient syntheses of N-acetyl thienamycin and epithienamycin A in their readily deprotected form are reported where three contiguous
stereocenters are established in a single catalytic asymmetric azetidinone-forming reaction. These examples are a template for synthesizing
C-5/C-6 cis or trans carbapenems with independent control of the C-8 stereocenter. A library of oxidatively and sterochemically defined
azetidinone precursors to a variety of naturally occurring carbapenems and potential biosynthetic intermediates has been prepared to facilitate
studies of carbapenem antibiotic biosynthesis.
Thienamycin and related carbapenem antibiotics show high
potency, a broad spectrum of activity, and comparative
stability to many clinically encountered ꢀ-lactamases that
confer resistance to penicillins and cephalosporins. Efficient,
scalable routes for the preparation of carbapenems are a
valuable, practical goal because they are not available by
large scale fermentation and/or semisynthesis. Most carbap-
enem syntheses proceed through azetidinone intermediates,
and the majority of the synthetic effort is expended in
establishing properly functionalized stereocenters at C-5, C-6,
and C-8 (Figure 1). Previous reports have shown that desired
C-5 and C-6 configurations can be derived from chiral
precursors.1a-d Some common methods of azetidinone
formation are ester enolate-imine condensations,2a-d
[2 + 2] cycloaddition reactions of olefins with isocyanates3
or ketenes with imines.4a-c The stereochemical outcome of
the latter is highly dependent upon the substrates and reaction
conditions giving C-3 and C-4 cis or trans diastereoselec-
tively, but not enantioselectively. Optically active material
has been obtained in specific cases by using ketenes or imines
bearing chiral auxiliaries, but these reactions are not ap-
plicable to general carbapenem synthesis.5a-c Azetidinones
with a carboxylic acid 8, carboxymethylene 9, or an
equivalent group at C-4 can be converted by several methods
to carbapenems.6a-c
† These authors contributed equally to this work.
(1) (a) Iimori, T.; Takahashi, Y.; Izawa, T.; Kobayashi, S.; Ohno, M.
J. Am. Chem. Soc. 1983, 105, 1659–1660. (b) Bateson, J. H.; Robins, A. M.;
Southgate, R. J. Chem. Soc., Perkin Trans. 1 1991, 29–35. (c) Melillo,
D. G.; Shinkai, I.; Liu, T.; Ryan, K.; Sletzinger, M. Tetrahedron Lett. 1980,
21, 2783–2786. (d) Karady, S.; Amato, J. S.; Reamer, R. A.; Weinstock,
L. M. J. Am. Chem. Soc. 1981, 103, 6765–6767.
(2) (a) Hart, D. J.; Ha, D. C. Chem. ReV. 1989, 89, 1447–1465. (b) Hart,
D. J.; Lee, C. S.; Pirkle, W. H.; Hyon, M. H.; Tsipouras, A. J. Am. Chem.
Soc. 1986, 108, 6054–6056. (c) Cainelli, G.; Giacomini, D.; Panunzio, M.
Tetrahedron Lett. 1985, 26, 937–940. (d) Georg, G. I.; Kant, J.; Gill, H. S.
J. Am. Chem. Soc. 1987, 109, 1129–1135.
To carry out biosynthetic studies of the carbapenems, we
required syntheses of naturally occurring carbapenams/ems
(3) Ohashi, T.; Kan, K.; Ueyama, N.; Sada, I.; Myama, A.; Watanabe,
K. U.S. patent 4791198, 1988.
10.1021/ol901269d CCC: $40.75
Published on Web 07/17/2009
 2009 American Chemical Society

choice would determine the C-5 and C-6 stereocenters and
enantiopure ketenes could be employed to establish that at
C-8.11,12 Various ketenes could be used to make cis azeti-
dinones, which could be epimerized to trans, to afford
precursors of cis or trans carbapenems.
We investigated the azetidinone-forming reaction of simple
alkyl ketenes with imine 7 catalyzed by BQ or BQd (Table
1). Ketenes were generated from the corresponding acid
Table 1. Azetidinone Precursors of Carbapenemsa
Figure 1. (a) Naturally occurring carbapenems and potential
biosynthetic intermediates. (b) Azetidinone forming reactions of
ketenes and imines.
and potential pathway intermediates to serve as enzyme
substrates and reference standards. The naturally occurring
carbapenems number about 50 and constitute a native
combinatorial library exhibiting modulated biological activi-
ties expressed in substituent and oxidation state variation at
C-2 and C-6.7 Moreover, advanced investigations of the
simplest carbapenem, carbapen-2-em-3-carboxylic acid,8 and
thienamycin9 have established that an epimerization event
occurs at C-5 (from S to R) during the course of biosynthesis.
Thus, an approach was needed to establish either the 5S- or
5R-configuration, establish the relative cis or trans stereo-
chemistry of the C-6 substituent, and set the historically
troublesome C-8 hydroxyl stereocenter.
a Reaction conditions: 10 mol % catalyst, 10 mol % In(OTf)3, 1.2 equiv
Given the variety of products desired, we were limited by
the scope of existing methods. We chose to build upon a
recently developed catalytic, asymmetric azetidinone-forming
reaction. This robust, scalable method uses the cinchona
alkaloid derivatives o-benzoylquinine (BQ) or its pseudoe-
nantiomer o-benzoylquinidine (BQd) as catalysts to give cis-
substituted azetidinones with excellent enantioselectivity (ee)
and diastereoselectivity (dr).10a-e We reasoned that if this
system could be applied to carbapenem synthesis, catalyst
acid chloride, 1.2 equiv Et3N, 1.0 equiv imine 7, toluene, -78 °C.
chlorides by treatment with triethylamine in situ. The reaction
produced cis-azetidinones with excellent enantioselectivity,
and only a trace of the trans diastereomer. Propionyl chloride,
butyryl chloride and isovaleryl chloride were used to
synthesize azetidinone precursors of deshydroxy northiena-
mycin 5, PS-5 (3) and PS-6 (4). To synthesize azetidinone
precursors of thienamycin (1) and epithienamycin A (2), ethyl
(3S)-hydroxybutanoate and ethyl (3R)-hydroxybutanoate
were silated and saponified to give the corresponding acids,
which were converted to the acid chlorides for ketene
generation.13 The expected azetidinones were produced on
a multigram scale.
It was postulated that the chiral transition state of the
azetidinone-forming reaction might be used to amplify
diastereoselection at C-8 from a pool of racemic starting
material. This possibility was investigated using silyl-
protected 3-hydroxy butyryl chloride as the ketene precursor.
The racemic 3-hydroxybutyrate was protected as the t-
butyldimethyl silyl, t-butyldiphenyl silyl, or triisopropyl silyl
(4) (a) Brandi, A.; Cicchi, S.; Cordero, F. M. Chem. ReV. 2008, 108,
3988–4035. (b) Fu, N.; Tidwell, T. T. Tetrahedron 2008, 64, 10465–10496.
(c) Claudio Palomo, J. M. A.; Ganboa, I.; Oiarbide, M. Eur. J. Org. Chem.
1999, 322, 3–3235.
(5) (a) Sunagawa, M.; Matsumura, H.; Enomoto, M.; Inoue, T.; Sasaki,
A. Chem. Pharm. Bull. 1991, 39, 1931–1938. (b) Sasaki, A.; Goda, K.;
Enomoto, M.; Sunagawa, M. Chem. Pharm. Bull. 1992, 40, 1094–1097.
(c) Colombo, M.; Crugnola, A.; Franceshi, G.; Lombardi, P. U.K. Patent
Appl. GB 2144419, 1985.
(6) (a) Shih, D. H.; Baker, F.; Cama, L.; Christensen, B. G. Heterocycles
1984, 21, 29–40. (b) Reider, P. J.; Grabowski, E. J. J. Tetrahedron Lett.
1982, 23, 2293–2296. (c) Devries, J. G.; Hauser, G.; Sigmund, G.
Tetrahedron Lett. 1984, 25, 5989–5992.
(7) Fischbach, M. A.; Clardy, J. Nat. Chem. Biol. 2007, 3, 353–355.
(8) Li, R. F.; Stapon, A.; Blanchfield, J. T.; Townsend, C. A. J. Am.
Chem. Soc. 2000, 122, 9296–9297.
(9) Hamed, R. B.; Batchelar, E. T.; Mecinovic, J.; Claridge, T. D. W.;
Schofield, C. J. ChemBioChem 2009, 10, 246–250.
(10) (a) Taggi, A. E.; Hafez, A. M.; Wack, H.; Young, B.; Ferraris, D.;
Lectka, T. J. Am. Chem. Soc. 2002, 124, 6626–6635. (b) France, S.; Wack,
H.; Hafez, A. M.; Taggi, A. E.; Witsil, D. R.; Lectka, T. Org. Lett. 2002,
4, 1603–1605. (c) France, S.; Shah, M. H.; Weatherwax, A.; Wack, H.;
Roth, J. P.; Lectka, T. J. Am. Chem. Soc. 2005, 127, 1206–1215. (d) France,
S.; Weatherwax, A.; Taggi, A. E.; Lectka, T. Acc. Chem. Res. 2004, 37,
592–600. (e) Shah, M. H.; France, S.; Lectka, T. Synlett 2003, 1937–1939.
(11) Georg, G. I.; Kant, J.; Gill, H. S. J. Am. Chem. Soc. 1987, 109,
1129–1135
.
(12) Cainelli, G.; Contento, M.; Giacomini, D.; Panunzio, M. Tetrahe-
dron Lett. 1985, 26, 937–940
.
(13) Lengweiler, U. D.; Fritz, M. G.; Seebach, D. HelV. Chim. Acta
1996, 79, 670–701.
Org. Lett., Vol. 11, No. 16, 2009
3607

ether in anticipation that steric bulk would accentuate any
stereoselection observed resulting from bias in the transition
state. One equivalent of each ketene (16-18) was used under
the standard reaction conditions showing only modest
selectivity (ca. 2:1) when the bulky triisopropyl silyl group
was used (Scheme 1). When the reaction was repeated with
Scheme 2
.
Syntheses of Protected N-Acetyl Thienamycin and
Epithienamycin A
Scheme 1.
Effect of C-8 Stereocenter on Catalysisa
a Reaction conditions: 10 mol % BQd, 10 mol % In(OTf)3, 1.0 equiv
acid chloride, 1.0 equiv Et3N, 1.0 equiv imine 7, toluene, -78 °C.
two and three equivalents of ketene, selectivity did not
improve and yields decreased. This observation further
implied that the catalyst has minimal bias for (S) or (R)
3-hydroxy butyrates and efficiently catalyzed cis-azetidinone
formation with either enantiomer.
Having made azetidinones with absolute configurational
control of three contiguous stereocenters, we synthesized
p-nitrobenzyl (PNB) N-acetyl thienamycin (24) and PNB
epithienamycin A (32). Initially the C-4 acids of azetidinones
13 and 14 were accessed by hydrogenolysis of the benzyl
esters, but the relatively reactive N-tosyl azetidnones were
not compatible with downstream reactions. We resolved to
remove the tosyl group early in the synthesis. After smooth
detosylation with samarium(II) diiodide14a,b the resulting
ꢀ-lactams were amenable to efficient completion of carbap-
enem synthesis (Scheme 2).
For the preparation of 24 and C-5/C-6 trans carbapenems,
the benzyl ester of 19 was removed by hydrogenation. The
resulting acid 20 was oxidatively decarboxylated with lead
tetraacetate to epimerize the C-5 stereocenter and produce a
known thienamycin precursor, acetoxy azetidinone 21.15 This
can be readily coupled with TBS enol 33 and desilylated to
give diazo azetidinone 23. The final stages of carbapenem
synthesis were achieved by cyclization of 23 to the 2-oxo-
carbapenam with rhodium acetate. After filtration of the
rhodium catalyst, the 2-oxocarbapenam was activated as its
enolphosphate, which underwent heteroconjugate addition
with N-acetylcysteamine (HSNAc) to give PNB N-acetyl
thienamycin (24). The PNB ester, conventionally used in
carbapenem syntheses can be readily removed by hydro-
genolysis to afford the natural product.16
hydrogenolysis of the C-5 benzyl ester to the acid 27. In
this case the Arndt-Eistert homologation was employed to
introduce a methylene and preserve the cis stereochemistry.
Azetidinone carboxylic acid 27 was activated as the acid
chloride, and this was substituted with diazomethane to give
diazoketone 28. The diazoketone was unstable to silica gel
chromatography and was subjected directly to light-promoted
Wolff rearrangement to give homologated acid 29.17 This
intermediate was activated with carbonyldiimidazole and the
final acetate unit was added by the method of Masamune.18
The silyl protecting groups were removed followed by the
introduction of the diazo moiety to give 31, directly
analogous to 23. Compound 31 was converted to PNB
epithienamycin A (32) in the same manner as 23 was
cyclized to PNB N-acetyl thienamycin (24) and similarly can
be deprotected by hydrogenation.19a,b
Here we demonstrate an efficient approach to carbapenem
synthesis with control of the three contiguous stereocenters
C-5, C-6, and C-8. Azetidinone precursors of deshydroxy
northienamycin (5), PS-5 (3), and PS-6 (4) have been
produced on a multigram scale with excellent dr and ee. The
Syntheses of 32 and C-5/C-6 cis carbapenems start with
TBS protection of the azetidinone nitrogen followed by
(17) Fetter, J.; Lempert, K.; Gizur, T.; Nyitrai, J.; Kajtanperedy, M.;
Simig, G.; Hornyak, G.; Doleschall, G. J. Chem. Soc., Perkin Trans. 1 1986,
221–227.
(14) (a) Vedejs, E.; Lin, S. Z. J. Org. Chem. 1994, 59, 1602–1603. (b)
Hasegawa, E.; Curran, D. P. J. Org. Chem. 1993, 58, 5008–5010.
(15) Berks, H. A. Tetrahedron 1996, 53, 2, 331–375.
(16) Corbett, D. F.; Coulton, S.; Southgate, R. J. Chem. Soc., Perkin
Trans. 1 1982, 3011–3016.
(18) Brooks, D. W.; Lu, L. D. L.; Masamune, S. Angew. Chem., Int.
Ed. 1979, 18, 72–74.
(19) (a) Kametani, T.; Huang, S. P.; Nagahara, T.; Ihara, M. J. Chem.
Soc., Perkin Trans. 1 1981, 2282–2286. (b) Kametani, T.; Nagahara, T.;
Ihara, M. J. Chem. Soc., Perkin Trans 1 1981, 3048–3052.
3608
Org. Lett., Vol. 11, No. 16, 2009

ease of stereocontrolled azetidinone formation simplified
syntheses of protected N-acetyl thienamycin (24) and epi-
thienamycin A (32) as well as provided a general route to
carbapenems possessing C-5/C-6 cis or trans configurations
with independent control of C-8 hydroxyl stereochemistry.
These efficient and flexible approaches allow access to the
many naturally occurring carbapenems and will enable
investigation of their biosynthetic relationships.
encouragement. We also acknowledge Dr. I. P. Mortimer
(JHU) for performing mass spectrometric analysis and the
NIH for financial support (AI014937).
Supporting Information Available: Experimental pro-
cedures and characterization for all compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
Acknowledgment. We thank Professor T. Lectka and
members of his laboratory for insightful discussions and
OL901269D
Org. Lett., Vol. 11, No. 16, 2009
3609