An anionic nucleophilic catalyst system for the diastereoselective synthesis of trans-β-lactams

Article abstract of DOI:10.1021/ol0511070

(Chemical Equation Presented) Trans-disubstituted β-lactams show increasing utility and prominence in numerous pharmaceutical applications, making their asymmetric synthesis an attractive goal for chemists. We introduce an anionic, nucleophilic catalyst s

Full text of DOI:10.1021/ol0511070

ORGANIC
LETTERS
2005
Vol. 7, No. 16
3461-3463
An Anionic Nucleophilic Catalyst
System for the Diastereoselective
Synthesis of trans-â-Lactams
Anthony Weatherwax, Ciby J. Abraham, and Thomas Lectka*
Department of Chemistry, New Chemistry Building, Johns Hopkins UniVersity,
3400 North Charles Street, Baltimore, Maryland 21218
lectka@jhu.edu
Received May 12, 2005
ABSTRACT
Trans-disubstituted
synthesis an attractive goal for chemists. We introduce an anionic, nucleophilic catalyst system that provides an efficient, diastereoselective
route to trans-disubstituted -lactams, a complement to our previously described catalytic methodology for generating the corresponding cis
â-lactams show increasing utility and prominence in numerous pharmaceutical applications, making their asymmetric
â
diastereomers. This catalytic, “switch mechanism” process allows for flexibility in the stereoselective synthesis of
cis or trans products as desired from the same substrates.
â-lactams, producing either
Long at the forefront of antibiotic development,1 â-lactam
chemistry has begun to branch out forcefully into new areas
of research. Many of the new uses of â-lactams, most notably
as serine protease inhibitors,2 particularly of thrombin,3
chymase,4 and tryptase,5 as well as â-lactamase inhibitors,6
have one common requirement: they all necessitate a trans
relationship between the ring substituents of an R,â-disub-
stituted â-lactam core. While we have previously developed
catalytic methodology to produce disubstituted cis-â-lactams
in excellent yields and high enantio- and diastereoselectivity,7
the disubstituted trans-â-lactams have remained a more
elusive target, with only a few catalyst-mediated routes
known.8 A hallmark of these published works is the careful
choice of substrates to obtain the desired selectivities.
(1) For a review on â-lactam antibiotics, see: (a) Rolinson, G. N. J.
Antimicrob. Chemother. 1998, 41, 589-603. (b) The Organic Chemistry
of â-Lactams; Georg, G. I., Ed.; VCH Publications: New York, 1993 and
references therein.
(2) Wilmouth, R. C.; Kassamally, S.; Westwood, N. J.; Sheppard, R. J.;
Claridge, T. D.; Alpin, R. T.; Wright, P. A.; Pritchard, G. J.; Schofield, C.
J. Biochemistry 1999, 38, 7989-7998.
(3) Han, W. T.; Trehan, A. K.; Wright, J. J. K.; Federici, M. E.; Seiler,
S. M.; Meanwell, N. A. Bioorg. Med. Chem. 1995, 3, 1123-1143.
(4) Aoyama, Y.; Uenaka, M.; Kii, M.; Tanaka, M.; Konoike, T.;
Hayasaki-Kajiwara, Y.; Naya, N.; Nakajima, M. Bioorg. Med. Chem. 2001,
9, 3065-3075.
(5) Ono, S.; Kuwahara, S.; Takeuchi, M.; Sakashita, H.; Naito, Y.; Kondo,
T. Bioorg. Med. Chem. Lett. 1999, 9, 3285-3290.
In this letter, we report the development of an anionic,
nucleophilic catalyst system based on a 2-aryl-2-imidazoline
scaffold that has allowed us to obtain trans-â-lactams in good
to excellent diastereoselectivity. We employ acid chlorides
(as ketene precursors),9 imino esters, and proton sponge (as
(7) (a) Taggi, A. E.; Hafez, A. M.; Wack, H.; Young, B.; Drury, W. J.,
III; Lectka, T. J. Am. Chem. Soc. 2000, 122, 7831-7832. (b) Taggi, A. E.;
Hafez, A. M.; Wack, H.; Young, B.; Ferraris, D.; Lectka, T. J. Am. Chem.
Soc. 2002, 124, 6626-6635.
(6) (a) Deziel, R.; Malenfant, E. Bioorg. Med. Chem. Lett. 1998, 8, 1437-
1442. (b) Yoakim, C.; Ogilvie, W.; Cameron, D.; Chabot, C.; Grande-Matre,
C.; Guse, I.; Hache, B.; Naud, J.; Kawai, S.; O’Meara, J.; Plante, R.; Deziel,
R. AntiViral Chem. Chemother. 1998, 9, 379-387.
(8) (a) Tanaka, H.; Hai, A. K. M. A.; Sadakane, M.; Okumoto, H.; Torii,
S. J. Org. Chem. 1994, 59, 3040-3046. (b) Townes, J. A.; Evans, M. A.;
Queffelec, J.; Taylor, S. J.; Morken, J. P. Org. Lett. 2002, 4, 2537-2540.
(9) Tidwell, T. T. Ketenes; John Wiley & Sons: New York, 1995.
10.1021/ol0511070 CCC: $30.25
© 2005 American Chemical Society
Published on Web 07/06/2005

Scheme 1. Catalytic Formation of trans-â-Lactams
the stoichiometric “thermodynamic” base) in a Staudinger-
type reaction whose selectivity is catalyst mediated but
substrate independent (Scheme 1). This methodology allows
us to obtain either a cis or a trans product, as desired, from
the same starting materials solely on the basis of catalyst
choice.10
As is the case with our complementary, cis-directing
cinchona alkaloid-based systems, the catalyst in this reaction
presumably acts as a “shuttle” base for ketene (enolate)
formation as well. Surprisingly, the effect of a remotely
positioned negative charge on the reactivity of a nucleophile
has not been well investigated in organic chemistry.11 We
hypothesized that in some cases, its reactivity should be
enhanced, and in the presence of a bulky, weakly coordinat-
ing counterion, its selectivity may be altered as well. We
began by testing a number of anionic nucleophiles, ultimately
settling on catalysts based on readily available 2-aryl-2-
imidazolines (Scheme 2). Such compounds have the advan-
tage of being easily synthesized from commercially available
starting materials. In the case of 4a, the desired product was
obtained in one step through methylation of 2-phenyl-2-
imidazoline (obtained from Aldrich) with CH3I. Compound
4b was synthesized in two steps through condensation of
sulfobenzoic anhydride with N-benzylethylenediamine fol-
lowed by ion exchange with tetraheptylammonium chloride
(all reagents are commercially available).
Figure 1. trans-â-Lactams synthesized. All compounds shown
were synthesized using catalyst 4b. All dr’s were determined by
1H NMR data obtained from crude reaction mixtures and are
displayed in a trans:cis format. Yields given are for products purified
by column chromatography.
the same conditions, 4b produced the â-lactam product in
approximately 50% yield and with a 37:1 dr (trans:cis). The
significant difference between the charged catalyst and the
neutral control is the presence of the anionic sulfonate group.
These promising initial tests prompted us to screen a series
of acid chlorides to investigate the scope of this new reaction
(Figure 1).
We discovered that, for a variety of acid halide precursors,
the trans diastereomers predominate (dr’s ranging from 5:1
to 50:1). For example, phenylacetyl chloride (1a) affords
largely the trans product (50% yield, 37:1 trans:cis). Ad-
ditionally, (p-methoxyphenyl)acetyl chloride (1b) leads to
product in 13:1 dr and 70% yield, whereas (p-chlorophenyl)-
acetyl chloride (1d) leads to product in 46% yield and 50:1
dr. Other substituted phenylacetyl halides work similarly to
produce the desired products in good dr and moderate to
good yield. To date, aliphatic acid chlorides work poorly in
the reaction, affording little, if any, â-lactam products at all.
Efforts to optimize reaction conditions for these aliphatic
substrates are ongoing.
Scheme 2. Synthesis and Performance of
2-Aryl-2-imidazoline Catalysts
An important question concerns how the catalyst produces
the trans diastereomers preferentially. In our earlier work
with benzoylquinine (BQ) and other cinchona alkaloid-
derived catalysts, we proposed that the â-lactam-forming
reaction proceeded through the zwitterionic enolate inter-
(10) Change in diastereoselectivity can be termed a “switch mechanism”
(wherein a change in the nature of the catalyst affords a dramatically
different reaction outcome for the same reacting partners), a long-standing
interest of ours. For example, see: Ferraris, D.; Drury, W. J., III; Cox, C.;
Lectka, T. J. Org. Chem. 1998, 63, 4568-4569.
(11) Such a system appears to be unique, the closest analogy evident in
the literature, although not strictly applicable, can be seen in: Ritchie, C.
D.; Hofelich, T. C. J. Am. Chem. Soc. 1980, 102, 7039-7044.
The prototype neutral catalyst 4a, while active in the
reaction of phenylacetyl chloride (1a, R ) Ph) with N-tosyl
imino ester (2) in the presence of proton sponge (PS, 5),
gave no selectivity, producing the â-lactam product with a
1:1 diastereomeric ratio (dr). However, when tested under
3462
Org. Lett., Vol. 7, No. 16, 2005

Scheme 3. Enantioselective cis-â-Lactam Formation
Scheme 4. Flipping Enolate Geometry with a Bulky Cation
A third scenario would involve product epimerization by
the basic catalyst, converting the original product from cis
to trans during the course of the reaction. Although we have
not observed substantial epimerization employing BQ as a
catalyst and proton sponge as a stoichiometric base, the new
anionic catalysts may be better thermodynamic bases than
BQ and/or better kinetic bases than proton sponge. To
examine this possibility, we performed control experiments
that showed that catalyst 4b could indeed epimerize the cis
diastereomer (cis-3a, made via the BQ-catalyzed reaction)
to the trans diastereomer over the course of the reaction time;
however, this resulted in only an approximate 1:1 mixture
of diastereomers.13 This experiment, when considered with
other controls, indicates that epimerization may play only a
minor role in determining the dr of the reaction and that
equilibrium mixtures of cis and trans products brought about
by the action of an appropriate base are significantly higher
in cis content than is observed in our trans-selective reaction.
Further work on this new anionic catalyst system, includ-
ing reaction optimization, mechanistic studies, extension of
scope to include aliphatic and other classes of acid chlorides,
and development of asymmetric variants, will be reported
in due course.
mediate shown in Scheme 3.7 We believe that it is through
attack on the imine by this enolate species that the stereo-
chemistry of the product is determined. Thus, if the enolate
geometry is altered or the reactive imine diastereoface is
flipped, the selectivity of the reaction can change from cis
to trans.
This reasoning led to our working hypothesis that the (Z)-
enolate (which is generally believed to be thermodynamically
more stable than the (E)-enolate) could lead preferentially
to the cis-â-lactam diastereomer, whereas the reaction leading
to the trans isomer may prefer an (E)-enolate geometry in
some cases. In one possible scenario, an anionic nucleophile
may overcome the putative thermodynamic preference for
the (Z)-enolate, especially when countered by a bulky
tetraalkylammonium cation. We theorize that reaction of such
a nucleophile would produce an intermediate in which much
of the negative charge resides on the enolate oxygen; we
expect that this would then bring the oxygen into close
proximity with the bulky counterion, which may result in
an (E)-enolate conformation (Scheme 4).12 In another pos-
sibility, the (Z)-enolate geometry would be retained; however,
the negative charge on the intermediate enolate could serve
to alter the approach of the imino ester (such that the other
diastereoface presents itself), again leading to the trans
product.
Acknowledgment. T.L. thanks the NIH (GM 064559),
the Sloan and Dreyfus Foundations, DuPont, and Merck &
Co. for support.
Supporting Information Available: General experimen-
tal procedures and compound characterization. This material
is available free of charge via the Internet at http://pubs.acs.org.
OL0511070
(13) Epimerization can conceivably occur through proton abstraction on
the formed â-lactam product, either at the R- or â-positions. However,
another plausible mechanism involves the epimerization of the acidified
R-proton on a ring-opened acylammonium intermediate. Further study
should differentiate between these possibilities.
(12) This mechanistic hypothesis can be thought of as a specific example
of the more general phenomenon of ion aggregation and the role that it
may be playing in the diastereoselectivity of the reaction.
Org. Lett., Vol. 7, No. 16, 2005
3463