Molecular mechanics calculations as predictors of enantioselectivity for chiral nucleophile catalyzed reactions

Article abstract of DOI:10.1016/S0040-4020(02)00987-0

We present molecular mechanics (MM) calculations as models of activated complexes for the β-lactam forming [2+2] cycloaddition between imino ester 4 and the zwitterionic intermediates derived from ketenes and various chiral nucleophilic catalysts. Our met

Full text of DOI:10.1016/S0040-4020(02)00987-0

Tetrahedron 58 (2002) 8351–8356
Molecular mechanics calculations as predictors of
enantioselectivity for chiral nucleophile catalyzed reactions
*
Andrew E. Taggi, Ahmed M. Hafez, Travis Dudding and Thomas Lectka
Department of Chemistry, Johns Hopkins University, 3400 North Charles Street, Baltimore, MD 21218, USA
Received 18 June 2002; accepted 8 July 2002
Abstract—We present molecular mechanics (MM) calculations as models of activated complexes for the b-lactam forming [2þ2]
cycloaddition between imino ester 4 and the zwitterionic intermediates derived from ketenes and various chiral nucleophilic catalysts. Our
method employs the use of Monte Carlo conformational searches utilizing the MMFF force field contained within the Macromodel program.
These models accurately predict the sense of stereochemical induction observed experimentally. Also, the predicted energetic differences for
minima leading to (R) or (S)-derived ketene facial selectivity correlate in a general sense with the magnitude of the enantioselectivity. This
work establishes that our approach represents a viable method for the design of new nucleophilic catalysts a priori using MM
calculations. q 2002 Elsevier Science Ltd. All rights reserved.
1. Introduction
to model activated complexes for reactions catalyzed by
chiral nucleophiles and to account for both the sense and
relative degree of optical induction.
The design and screening of new catalysts for asymmetric
synthesis is a very important, yet extremely time consuming
undertaking. Often the search for a new catalyst involves
running an extensive number of reactions followed by chiral
HPLC or GC analysis to determine its efficacy, which
amounts to a very tedious undertaking. Recently, there have
been many developments intended to increase the speed of
this process by employing high throughput screening
techniques during the search for new catalysts.1 This
process could also be accelerated through the development
of a straightforward computational model for the catalytic
system being examined. Potential catalysts could then be
screened computationally to remove the least likely
candidates before the catalyst is screened experimentally.
In some cases, especially where researchers often do not
have access to the expensive equipment required to perform
some of the latest screening techniques, this pre-screen
could save valuable time and resources. Not only would this
expedite the screening process, it would also provide
information as to what controls the sense of induction in
the catalyst system. This additional information could be
utilized to determine which catalyst motifs can be best
exploited. A simple but effective computational technique is
expected to work best on metal-free catalyst systems that
lead to all-organic activated complexes. In this paper, we
propose the use of molecular mechanics (MM) calculations
The sheer importance of b-lactams has made this structural
motif a worthwhile goal for the synthetic organic chemist.2
Recently b-lactams (especially non-natural ones) have
achieved many important non-antibiotic uses such as their
development as mechanism-based serine protease inhibi-
tors,3 as well as being useful precursors to b-amino acids.
As a testament to the continuing importance of b-lactams to
pharmaceutical science, a recent issue of Tetrahedron was
devoted exclusively to b-lactam chemistry and synthesis.4
We recently reported a mechanistically distinct, chiral
nucleophile catalyzed b-lactam forming reaction5 by
making the ketene nucleophilic (through generation of a
zwitterionic intermediate),6 and the imine component
electrophilic (Eq. (1)).7–9 Herein we expand upon our
initial discussion by modeling ketenes with different
substituents through a series of MM calculations using the
Macromodel program,10 and by screening other chiral
nucleophiles with the intent to lead to the de novo design of
a novel nucleophilic catalyst for asymmetric synthesis of
b-lactams. By comparing the results of our ‘theoretical
screening’ to our experimental results, we establish the
overall validity of the process.
Keywords: molecular mechanics; Monte Carlo conformational searches;
b-lactams; enantioselective; ketenes.
ð1Þ
*
Corresponding author. Tel.: þ1-410-516-6448; fax: þ1-410-516-8420;
e-mail: lectka@jhu.edu
0040–4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(02)00987-0

8352
A. E. Taggi et al. / Tetrahedron 58 (2002) 8351–8356
N
N
H
re fac e
approach
H
PhCO2
O
0.0 kcal
Ph
OMe
proposed
si face
approach
ketene-5a
adduct
2.64 kcal
ketene-5a adduct (macromodel)
Scheme 1. Proposed mechanism for the BQ catalyzed b-lactam synthesis.
Figure 1. Stereochemical model of the putative zwitterionic intermediates
of BQ and phenylketene.
ketene. To understand which specific structural aspects of
the catalyst engendered asymmetry to this process, we
sought a simple computational procedure that would
accurately predict product selectivities. The use of Monte
Carlo MM calculations was judged to be a superior method
of analysis based on the conformational flexibility and
all-organic nature of this system.12
Initially, we investigated the energetic importance of the
ketene geometry in the putative zwitterionic intermediates
formed between BQ and ketenes using the MMFF force
field of Macromodel.13 In all the cases that we studied, the
bridgehead (quinuclidine) nitrogen is preferably trans to the
ketene substituent across the CvC bond (Table 1).
Inspection of the lowest conformational structures of each
entry in Table 1 reveals that the E-ketene geometry is
energetically less favored due to repulsive van der Waals
interactions between the ketene and the adjacent quinoline
ring of the catalyst. Additionally, the presence of noticeable
A1,3 interactions between the terminal substituent of the
ketene and the bridging carbon framework of the quinucli-
dine nucleus are expected to play a role. The energetic
preference for the Z-ketene geometry arises from a
favorable placement of the terminal ketene substituent
away from the quinoline and benzoyl groups of catalyst 5a,
effectively creating a pocket that shields one face of the
ketene. The predominant A1,3 interactions that were present
in the E geometry have also been alleviated in the Z
conformer.
2. Results and discussion
Based on experimental evidence and the known reactivity of
ketenes, we have postulated a reaction mechanism that is
consistent with the pathway depicted in Scheme 1.6 An acid
chloride (1) reacts with benzoylquinine (5a, BQ), or another
amine-based nucleophilic catalyst (5b–e, 8–11), to form an
acylammonium salt that is subsequently deprotonated to
form a zwitterionic enolate 2, generating 1 equiv. of the
hydrochloride of proton sponge (6, PS) which precipitates
from the reaction solution (free ketenes may not be involved
in the reaction).11 The putative enolate then reacts with
imine 4 to form the b-lactam (7). Within this mechanism,
the nucleophilic cinchona alkaloid-derived catalyst serves a
dual role, acting as both a proton shuttle for the in situ
generation of ketenes from acid halides as well as an
asymmetric catalyst for b-lactam formation. Central to this
reaction scenario is the generation of a reactive zwitterionic
intermediate 2, formed by the addition of the nucleophilic
cinchona alkaloid nitrogen to the electrophilic carbon of a
Inspection of the Z isomers of these same models thus
reveals that the re-face of the ketene CvC bond is open to
the approach of an electrophile. A specific example of this
trend is shown in Fig. 1 where the model derived from
re-face (top face) exposure of the Z-BQ–phenylketene
adduct derived from phenylacetyl chloride and BQ is
calculated to be 2.64 kcal/mol lower in energy than the
corresponding model containing the exposed si-face.
Importantly, the observed sense of induction of the b-lactam
products isolated experimentally is consistent with these
models and in accord with other models proposed for
cinchona alkaloid catalyzed processes.
Table 1. Calculated energies for several different BQ–ketene adducts
Entry
R
E1 (kcal)
E2 (kcal)
lDEl (kcal)
Previously we had found that the use of benzoylquinidine
(5b, BQd) as the catalyst inverts the stereochemistry of both
chiral centers in the resultant b-lactam.6 In BQ, the
bridgehead chiral center alpha to the quinuclidine amine
and the center attached to the ester are of opposite
configuration to those in BQd. To determine whether one
or both chiral centers are controlling the sense of induction,
1
2
3
4
5
6
Ph (Z)
Ph (E)
Et (Z)
Et (E)
Br (Z)
Br (E)
73.03 re
78.69 si
36.38 re
38.36 si
70.33 re
74.16 si
75.66 re
78.75 si
38.36 re
38.59 si
73.78 re
74.56 si
2.64
0.06
1.98
0.23
3.46
0.39

A. E. Taggi et al. / Tetrahedron 58 (2002) 8351–8356
8353
OMe
re face
approach
N
N
re face
approach
0.0 kcal
0.0 kcal
N
N
H
H
H
H
si face
approach
PhCON
PhCO2
si face
approach
O
H
O
2.51 kcal
Ph
Ph
OMe
2.74kcal
proposed
proposed
ketene-5e
adduct
ketene-5c
adduct
ketene-5e adduct (macromodel)
ketene-5c adduct (macromodel)
Figure 4. Stereochemical model of the zwitterionic intermediate when
using benzamidoquinine 5e.
Figure 2. Stereochemical model of the putative zwitterionic intermediate of
BEQ and phenylketene.
we investigated the consequence of epimerizing the ‘oxy’
stereogenic center alpha to the ester oxygen (affording
benzoylepiquinine, or BEQ, 5c). Macromodel calculations
in this case predict the same sense of induction as that seen
in BQ catalyzed reactions. As seen in model ketene–5c, the
re-face of the ketene zwitterion is still exposed (Fig. 2). The
lowest energy corresponding si-face exposed conformer is
some 2.74 kcal/mol higher in energy. When we performed
the reaction using BEQ (phenylacetyl chloride 1a, imino
ester 4) under standard conditions with proton sponge as the
stoichiometric base, we obtained the cis-diastereomer 7a in
97% ee, confirming our prediction.
is 2.51 kcal/mol higher in energy. The energy difference
between the two faces is smaller than that of BQ, explaining
the decreased selectivity when using 5e (89% ee). This
calculation suggests, however, that hydrogen bonding can
be used as an organizing principle in the catalyst design of
ketene reactions.
New catalysts. To test our model of the catalyst–ketene
complex, we chose several nucleophilic catalysts and
calculated the relative energy differences between approach
from the re and si faces. The catalysts were then screened
under standard reaction conditions to determine the
predictive power of our model. We discuss the results
from natural, semi-synthetic and entirely synthetic nucleo-
philes. As our first example we examined (S)-(2)-nicotine
(8), an inexpensive and readily available alkaloid. Nicotine
shares some similarities with BQ in that it has a tertiary
amine as well as a less nucleophilic pyridine nitrogen.
Unlike BQ, the tertiary amine is not part of a bicyclic
system, and additionally, there is considerably less steric
bulk associated with the molecule. When we ran the MMFF
calculation, we found there to be a small energy difference
of 0.51 kcal/mol between re and si approach (where the
energy difference for BQ was 2.64 kcal/mol) thus predicting
minimal enantioselectivity (Fig. 5). Visual inspection of the
model also suggests that there should be minimal enantio-
selectivity. The catalyst is contained in the vertical plane
while the ketene is projected outward from the assembly,
allowing reaction to occur from either face. Experimentally,
this model held true, yielding b-lactam 7a in 6% ee and 2:1
diastereomeric ratio (dr, cis/trans).
Carrying our investigation one step further, we removed the
benzoyl group altogether, resulting in deoxyquinine (5d,
DOQ) which now has an achiral center where the ester was
attached. Calculations on the adduct of phenylketene and
DOQ predict the same sense of induction (as observed,
Fig. 1) except for the fact that the lowest si-face
conformation is now about 1.31 kcal/mol higher in energy,
a smaller gap reflected in the diminished ee (72%) that we
observed (Fig. 3). This calculation suggests that the
presence of the oxy stereogenic center in BQ is not critical
to the sense of induction, and it is the steric bulk of the
benzoyl group that is most critical in enhancing selectivity.
We have proposed that quinine amide 5e engages in
hydrogen bonding with the ketene enolate in the activated
complex ketene–5e,14 yielding product with the same sense
of induction as BQ. Macromodel calculations were
performed on model complex ketene–5e, and a low energy
complex was found in which an intramolecular hydrogen
bond stabilized the oxygen of the ketene enolate, leaving the
re face exposed to attack of an electrophile (Fig. 4). The
corresponding low energy structure with the si-face exposed
We also investigated brucine (9), another inexpensive
commercially available alkaloid.15 Brucine is a consider-
ably larger molecule than nicotine and has a nucleophilic
tertiary amine in a bridgehead position, like BQ. The
transition state model shows a preference for re face attack
re face
approach
N
H
0.51 kcal
H
N
Me
N
re face
approach
H
H
0.0 kcal
Ph
si face
approach
N
O
O
Ph
OMe
proposed
ketene-5d
adduct
0.0 kcal
si face
proposed
8-ketene
adduct
approach
8-ketene adduct
(macromodel)
ketene-5d adduct (macromodel)
1.31 kcal
Figure 3. Stereochemical model of the putative zwitterionic intermediate of
DOQ.
Figure 5. Proposed stereochemical model of the nicotine–phenylketene
adduct.

8354
A. E. Taggi et al. / Tetrahedron 58 (2002) 8351–8356
Ph
H
O
Ph
Ph
N
re face
N
approach
re face
Ph
Ph
H
approach
0.0 kcal
N
N
MeO
MeO
0.0 kcal
H
O
H
si face
O
approach
H
N
O
si face
approach
2.45 kcal
proposed
9-ketene
adduct
H
proposed
11-ketene
adduct
9-ketene adduct
(macromodel)
11-ketene adduct
(macromodel)
1.40 kcal
O
Figure 6. Proposed stereochemical model of the putative zwitterionic
intermediates of brucine and phenylketene.
Figure 8. Stereochemical model of the synthetic catalyst 11–phenylketene
adduct.
of 1.40 kcal (Fig. 6). When used in our system, 7a was
produced in 77% ee and 10/1 dr. Of all our calculations, this
one is the most surprising, as in no case does shielding of
either the re or the si-face look extensive enough to lead to a
compelling result.
Overall, the molecular models (except for that of brucine)
and the empirical results show good correlation, correctly
predicting the magnitude of the enantioselectivity and the
sense of induction. The model held for both cinchona
derived catalysts as well as other tertiary amine systems
and a de novo designed synthetic amidine catalyst as well
(Table 2).
As our next test, we chose to examine the semi-synthetic
quinidine derivative 10, which is synthesized in one step
from quinidine.16 This catalyst has been used very
effectively for a catalytic asymmetric Baylis–Hillman
reaction, forming a-methylene-b-hydroxy esters in high
enantioselectivity.17 When we modeled 10 we found a
larger energy difference (1.79 kcal) between the two
possible modes of attack by the ketene, suggesting higher
enantioselectivity than with nicotine, but less than that from
BQ (Fig. 7). This was confirmed experimentally with the
formation of 7a in 33% ee and 10/1 dr. An interesting point
about this catalyst system is the observed sense of induction.
The b-lactam product has the opposite chirality (R,R) than
the resultant product from the benzoylquinidine reaction
(S,S), even though the two catalysts have the same
configuration at the aforementioned chiral centers. While
somewhat surprising at first inspection, the sense of
induction was in fact predicted by our model.
Table 2. A summary of the calculated and empirical results using
phenylacetyl chloride with several catalysts
Entrya
Catalyst
E1
(kcal)
E2
(kcal)
lDEl
(kcal)
% eeb
1
2
3
4
5
(S)-Nicotine 8
Brucine 9
Mod-quinidine 10
Amidine 11
238.91 si 238.40 si 0.51
6 (R,R)
15.06 si
67.42 si
53.72 si
16.46 si
69.21 si
56.17 si
75.66 si
1.40
1.79
2.45
2.64
77 (R,R)
33 (R,R)
73 (R,R)
99 (R,R)
Benzoylquinine 5a 73.03 si
a
Reactions run under standard reaction conditions outlined in Section 4.
% ee determined by chiral HPLC.
b
To make models like this useful, one must develop a scale to
give the relative energies meaning. Logically, there must
exist a DE (energy X) at which the catalyst will start to
become selective, and consequently there will be a point
(energy Y) where any further increase in energy will not
result in a measurable increase in selectivity (Fig. 9).
Between these two energies is the scale (depicted as, but not
necessarily linear) that relates the relative energy difference
to the theoretical selectivity of the catalyst. These points or
barriers can be approximated experimentally for each
system, resulting in a scale which calculated energies of
new catalysts can be compared to predict selectivity.
Finally, we chose to study the designed synthetic amidine
catalyst 11, which we predicted upon reaction with
phenylketene would afford a reactive complex in which
the si-face is blocked to approach of an electrophile.
Catalyst 11 is readily made from optically pure phenylene-
diamine. Unlike all of our previous models, this catalyst
uses an imino nitrogen as the nucleophile with a proposed
hydrogen bond donor included in the catalyst to provide
additional rigidity, similar to benzamidoquinine 5e.14 When
calculated, 11 shows a strong preference for re-approach
(2.45 kcal), predicting fairly good enantioselectivity (Fig. 8),
a prediction which is confirmed experimentally (73% ee, 3/1
dr). Modifications are being made to the skeleton of the
amidine catalyst system to optimize the enantioselectivity.
3. Conclusion
We have found that molecular mechanics transition state
OH
Me
re face
approach
N
O
0.0 kcal
N
H
si face
approach
O
Ph
proposed
1.79 kcal
10-ketene adduct
(macromodel)
10-ketene
adduct
Figure 7. Proposed stereochemical model of semi-synthetic 10–phenyl-
ketene adduct.
Figure 9. A diagram of the relationship between the DE of two transition
state models and the resultant selectivity.

A. E. Taggi et al. / Tetrahedron 58 (2002) 8351–8356
8355
(c) Korbel, G. A.; Lalic, G.; Shair, M. D. J. Am. Chem. Soc.
2001, 123, 361–362. (d) Reetz, M. T.; Becker, M. H.; Klein,
H.-W.; Stockigt, D. Angew. Chem., Int. Ed. 1999, 38,
1758–1761. (e) Guo, J.; Wu, J.; Siuzdak, G.; Finn, M. G.
Angew. Chem., Int. Ed. 1999, 38, 1755–1758. For general
reviews see: (f) Reetz, M. T. Angew. Chem., Int. Ed. 2001, 40,
284–310. (g) Jandeleit, B.; Schaefer, D. J.; Powers, T. S.;
Turner, H. W.; Weinberg, W. H. Angew. Chem., Int. Ed. 1999,
38, 2494–2532. (h) Kuntz, K. W.; Snapper, M. L.; Hoveyda,
A. H. Curr. Opin. Chem. Biol. 1999, 3, 313–319. (i) Francis,
M. B.; Jamison, T. F.; Jacobsen, E. N. Curr. Opin. Chem. Biol.
1998, 2, 422–428.
calculations on nucleophilic catalyst–ketene adducts give
insight into the physical reason for the observed sense of
induction observed experimentally with the benzoylquinine
catalyzed b-lactam forming reaction. This model system
was then used to predict the magnitude of stereoselectivity
induced by several chiral nucleophilic catalysts with
different structural motifs. Future studies will expand on
the rational design of novel nucleophilic catalysts.
4. Experimental
2. For a review of recent b-lactam chemistry see: Palomo, C.;
Aizpurua, J. M.; Ganboa, I.; Oiarbide, M. Eur. J. Org. Chem.
1999, 12, 3223–3235.
4.1. General procedure for b-lactam 7a using proton
sponge
3. (a) Wilmouth, R. C.; Kassamally, S.; Westwood, N. J.;
Sheppard, R. J.; Claridge, T. D. W.; Aplin, R. T.; Wright, P. A.;
Pritchard, G. J.; Schofield, C. J. Biochem 1999, 38,
7989–7998. (b) Taylor, P.; Anderson, V.; Dowden, J.; Flitsch,
S. L.; Turner, N. J.; Loughran, K.; Walkinshaw, M. D. J. Biol.
Chem. 1999, 274, 24901–24905.
To a solution of benzoylquinine 5a (5.5 mg, 0.0129 mmol)
and proton sponge 6 (31 mg, 0.142 mmol) in toluene (1 mL)
at 2788C was added phenylacetyl chloride 1a (20 mg,
0.129 mmol) in toluene (0.5 mL) immediately followed by
a-imino ester 4 (33 mg, 0.129 mmol) in toluene (0.5 mL).
The reaction was allowed to stir for 5 h as it slowly warmed
to room temperature. The solvent was removed under
reduced pressure and the crude mixture was subjected to
column chromatography (15% EtOAc/hexanes) on a plug of
silica gel to yield 7a (65% yield, 33 mg).
4. Miller, M. J. Tetrahedron 2000, 56. preface.
5. For an extensive review of the use of cinchona alkaloids as
catalysts see: (a) Kacprzak, K.; Gawronski, J. Synthesis 2001,
961–998. For a review on catalysis with organic molecules
see: (b) Dalko, P. I.; Moisan, L. Angew. Chem., Int. Ed. 2001,
40, 3726–3748.
4.1.1. N-(2R,3R-Diphenyl-2,5,6,7S-tetrahydro-3H-pyr-
rolo[1,2-a]imidazol-7-yl)-benzamide 11. To a pressure
tube was added S-(þ)-2-benzoylamine-4-bromobutanoic
acid (674 mg, 2.36 mmol), (1R,2R)-(2)1,2-dipenylethyl-
enediamine (500 mg, 2.36 mmol) and 0.5 mL of toluene and
the mixture heated at 1408C for 3 h. The reaction was cooled
and the solvent removed under reduced pressure. The
resulting oil was chromatographed on silica gel using (20%
EtOAc/EtOH) and the resulting product recrystallized from
acetone to afforded 206 mg (23%) of the desired product as
a white solid. Mp 142–1458C; 1H NMR ((CD3)2SO) d 8.58
(d, 1H), 7.87 (d, 2H), 7.54–7.40 (m, 4H), 7.32–7.14 (m,
8H), 4.92 (m, 1H), 4.55 (s, 2H), 3.58 (t, 2H), 2.20–2.11 (m,
1H), 2.08–1.97 (m, 1H) ppm; 13C NMR ((CD3)2SO)
168.79, 167.99, 146.02, 135.97, 132.75, 129.98, 129.72,
128.98, 128.76, 127.75, 59.38, 47.64, 37.28; IR (KBr plate)
3360, 2984, 1680, 1657, 1520, 1420. Anal. calcd for
C25H23N3O, C, 78.71; H, 6.08; N, 11.02. Found C, 78.32; H,
6.20; N, 11.35.
6. (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.; Witsel,
D.; Lectka, T. Org. Lett. 2002, 4, 1603–1605. (c) Taggi, A. E.;
Wack, H.; Hafez, A. M.; France, S.; Lectka, T. Org. Lett. 2002,
4, 627–629. (d) Hafez, A. M.; Taggi, A. E.; Dudding, T.;
Lectka, T. J. Am. Chem. Soc. 2001, 123, 10853–10859.
(e) Hafez, A. M.; Taggi, A. E.; Wack, H.; Drury, III., W. J.;
Lectka, T. Org. Lett. 2000, 2, 3963–3965. (f) Taggi, A. E.;
Hafez, A. M.; Wack, H.; Young, B.; Drury, III., W. J.; Lectka,
T. J. Am. Chem. Soc. 2000, 122, 7831–7832.
7. Imine 4 was first popularized in work by: Tschaen, D. H.;
Turos, E.; Weinreb, S. M. J. Org. Chem. 1984, 49,
5058–5064.
8. We have also used N-acyl imino esters to form b-lactams as
intermediates in the synthesis of substituted aspartic acids:
Dudding, T.; Hafez, A. M.; Taggi, A. E.; Wagerle, T. R.;
Lectka, T. Org. Lett. 2002, 4, 387–390.
9. We have previously used imine 4 for the catalytic asymmetric
synthesis of amino acids: Ferraris, D.; Young, B.; Cox, C.;
Dudding, T.; Drury, III., W. J.; Ryzhkov, L.; Taggi, A. E.;
Lectka, T. J. Am. Chem. Soc. 2002, 124, 67–77.
Acknowledgements
T. L. thanks the NIH (GM54348), Merck, Eli Lilly, the NSF
Career Program for support, the Dreyfus Foundation for a
Teacher-Scholar Award, and the Alfred P. Sloan Foundation
for a Fellowship. A. T. thanks the Organic Division of the
American Chemical Society for a Graduate Fellowship
sponsored by Organic Reactions, Inc. (2001–2002).
10. Macromodel V. 7.0 copyright Columbia University 1986–
1998, Schrodinger Inc. 1999. See: Mohamadi, F.; Richards,
N. G. J.; Guida, W. C.; Liskamp, R.; Lipton, M.; Caufield, C.;
Chang, G.; Hendricksen, T.; Still, W. C. J. Comput. Chem.
1990, 11, 440–467.
11. It is possible that acid chloride is deprotonated by BQ or PS
(without the initial formation of an acylammonium salt)
forming free ketene that can then react with the catalyst to
form the zwitterionic intermediate (Scheme 1). IR experiments
did not show the formation of free ketene, making this scenario
seem unlikely, but not impossible. See Ref. 6a as well as:
(a) Brady, W. T.; Scherubel, G. A. J. Org. Chem. 1974, 39,
3790–3791. (b) Brady, W. T.; Scherubel, G. A. J. Am. Chem.
References
1. (a) Jarvo, E. R.; Evans, C. A.; Copeland, G. T.; Miller, S. J.
J. Org. Chem. 2001, 66, 5522–5527. (b) Copeland, G. T.;
Miller, S. J. J. Am. Chem. Soc. 2001, 123, 6496–6502.

8356
A. E. Taggi et al. / Tetrahedron 58 (2002) 8351–8356
Soc. 1973, 95, 7447–7449. (c) Walborsky, H. M. J. Am. Chem.
Soc. 1952, 74, 4962–4963.
15. Brucine has previously been used as an asymmetric catalyst:
(a) Kerr, W. J.; Kirk, G.; Middlemiss, D. Synlett 1995, 10,
1085–1086. (b) Griffiths, S. P.; Johnston, P.; Vermeer, W. A.
H.; Wells, P. B. J. Chem. Soc., Chem. Commun. 1994, 21,
2431–2432.
12. (a) Breit, B.; Zahn, S. K. J. Org. Chem. 2001, 66, 4870–4877.
(b) Breit, B. Eur. J. Org. Chem. 1998, 1123–1134.
13. Halgren, T. A. J. Comput. Chem. 1996, 17, 490–519.
14. Hydrogen bond contacts have been postulated to play similar
roles in other catalytic reactions: (a) Ameer, F.; Drewes, S. E.;
Freese, S.; Kaye, P. T. Synth. Commun. 1988, 18, 495–500.
(b) Vasbinder, M. M.; Jarvo, E. R.; Miller, S. J. Angew. Chem.,
Int. Ed. 2001, 40, 2824–2827.
16. Because this catalyst is a derived from quinidine, the
stereochemistry of the resultant b-lactam should be the
opposite of quinine derivatives.
17. Iwabuchi, Y.; Nakatani, M.; Yokoyama, N.; Hatakeyama, S.
J. Am. Chem. Soc. 1999, 121, 10219–10220.