Catalytic, enantioselective N-acylation of lactams and thiolactams using amidine-based catalysts

Article abstract of DOI:10.1021/ja306766n

In contrast to alcohols and amines, racemic lactams and thiolactams cannot be resolved directly via enzymatic acylation or classical resolution. Asymmetric N-acylation promoted by amidine-based catalysts, particularly Cl-PIQ 2 and BTM 3, provides a convenient method for the kinetic resolution of these valuable compounds and often achieves excellent levels of enantioselectivity in this process. Density functional theory calculations indicate that the reaction occurs via N-acylation of the lactim tautomer and that cation-π interactions play a key role in the chiral recognition of lactam substrates.

Full text of DOI:10.1021/ja306766n

Article
pubs.acs.org/JACS
Catalytic, Enantioselective N‑Acylation of Lactams and Thiolactams
Using Amidine-Based Catalysts
Xing Yang,† Valentina D. Bumbu,† Peng Liu,‡ Ximin Li,† Hui Jiang,† Eric W. Uffman,† Lei Guo,†
Wei Zhang,† Xuntian Jiang,† K. N. Houk,*,‡ and Vladimir B. Birman*,†
†Department of Chemistry, Washington University, One Brookings Drive, St. Louis, Missouri 63130, United States
‡Department of Chemistry and Biochemistry, University of California, Los Angeles, California 90095-1569, United States
S
* Supporting Information
ABSTRACT: In contrast to alcohols and amines, racemic
lactams and thiolactams cannot be resolved directly via
enzymatic acylation or classical resolution. Asymmetric N-
acylation promoted by amidine-based catalysts, particularly Cl-
PIQ 2 and BTM 3, provides a convenient method for the
kinetic resolution of these valuable compounds and often
achieves excellent levels of enantioselectivity in this process. Density functional theory calculations indicate that the reaction
occurs via N-acylation of the lactim tautomer and that cation−π interactions play a key role in the chiral recognition of lactam
substrates.
INTRODUCTION
■
Chiral lactams (and thiolactams)1 find numerous applications
as pharmaceuticals,2 chiral auxiliaries,3 ligands,4 and synthetic
intermediates.5 Many lactams can be found among bioactive
natural products.6 As a consequence, methods for their
preparation in enantiopure form continue to be in high
demand. Most available asymmetric catalytic methods that form
Figure 2. ABC-promoted KR of benzylic alcohols.
the lactam ring directly from achiral precursors7,8 deliver N-
substituted lactams, which usually need to be deprotected prior
particularly interested in exploring the potential of ABCs in the
catalytic, enantioselective acylation of nitrogen nucleophiles
(Figure 3).15 Most examples of this transformation come from
the formidable body of work on enzymatic resolution of
amines.16 Enantioselective acylation of amines using non-
enzymatic catalysts,17 however, is made particularly challenging
by the high nucleophilicity of these substrates, which results in
their rapid background reaction with conventional acyl donors.
to further synthetic manipulations. Most often, chiral lactams
are synthesized via cyclization of enantioenriched acyclic
precursors3,6e which simply relegates the problem of controlling
the absolute configuration to an earlier step in the synthesis.
However, in those cases when N-unsubstituted lactams are
easily available in racemic form,9 their resolution10 into
individual enantiomers becomes a viable alternative to
asymmetric synthesis.
Since 2003, our group’s efforts have been focused on the
development of a new class of enantioselective amidine-based
catalysts (ABCs) and their synthetic applications.11−13 The
structures of the most successful ABCs are shown in Figure 1.
In 2004, having demonstrated their efficacy in the kinetic
resolution (KR)14 of benzylic secondary alcohols (Figure 2), we
turned our attention to other classes of substrates. We became
Figure 3. Scope of catalytic asymmetric acylation.
Received: July 13, 2012
Published: October 2, 2012
Figure 1. Amidine-based catalysts.
© 2012 American Chemical Society
17605
dx.doi.org/10.1021/ja306766n | J. Am. Chem. Soc. 2012, 134, 17605−17612

Journal of the American Chemical Society
Article
Fortunately, several creative solutions to this problem have
been found over the past decade, most notably by the groups of
Fu,18 Seidel,19 and Bode.20 In addition, noncatalytic methods
are available for the resolution of racemic amines, such as
classical resolution and related methods, i.e., crystallization of
their diastereomeric salts with chiral acids,10,21 and acylation
with stoichiometric chiral acylating agents.22
Table 1. KR of 4-Substituted Oxazolidinones
We realized that the enantioselective N-acylation of
secondary amides with simple anhydrides should be inherently
less problematic than that of amines, due to the lower
nucleophilicity of the former. Surprisingly, prior to our first
publication in this area in 2006,12a there had been no reports of
this transformation achieved using either enzymes or non-
enzymatic catalysts. At the same time, we reasoned, it would be
even more practically valuable than the enantioselective N-
acylation of amines, because, in contrast to the latter, amides
cannot be resolved via classical resolution.
Generally speaking, resolution of amidesboth acyclic and
cyclic ones (lactams)is underdeveloped and relies mostly on
the use of stoichiometric resolving agents or indirect methods.
For example, Kuneida et al. reported in 1992 that racemic
oxazolidinones could be resolved by chromatographic separa-
tion of their diastereomeric N-camphanoyl derivatives.23a Later,
the same group achieved catalytic KR of N-acyl-oxazolidinones
and N-acyl-imidazolidinones via enantioselective reductive
deacylation mediated by CBS.23b Transformation of secondary
amides into their N-hydroxymethyl derivatives followed by
enzymatic O-acylation and deprotection has been reported by
several groups.24 Acyclic amides and β-lactams have been
resolved via enzymatic hydrolysis of the amide bond.25 An
analogous nonenzymatic transformation, enantioselective alco-
holysis of N-acyl-β-lactams, was recently developed in our
group using catalyst 3.13d Apart from our studies in this area
described below, Miller et al. reported KR of formamides and
thioformamides in 2010.26 Their study has provided the first
successful examples of catalytic, enantioselective N-acylation of
acyclic amides.
In the present Article, we provide a detailed account of our
studies on the KR of lactams and thiolactams promoted by
ABCs. Our focus shall be on the correlation between the
structures of our substrates and catalysts and their reactivity and
enantioselectivity in KR. Therefore, the discussion of varying
reaction parameters (solvent, concentration, temperature, time,
acylating agent, base, etc.) shall be kept to a minimum. The
results shown in Tables 1−5 are the best obtained for each
given substrate−catalyst combination. We also provide the first
density functional theory (DFT) study on the mechanism of
ABC-catalyzed N-acylation of lactams and the origins of
enantioselectivity in this process.
General conditions: 0.2 M ( )-substrate concentration, 0.75 equiv (i-
PrCO)2O, 0.75 equiv i-Pr2NEt, (R)-2 was used at in tert-amyl alcohol
at 0° C; (R)-3 was used in chloroform at rt (23° C), unless specified
otherwise. Absolute configuration of the fast-reacting enantiomer is
a
b
c
shown. 0.05 M substrate. 1.5 equiv (i-PrCO)2O. 0.1 M substrate,
d
e
1
tert-amyl alcohol At rt. Conversion estimated by H NMR.
phenyl-imidazo[1,2-a]quinoline (Cl-PIQ) 2,11b on the other
hand, typically produces lower selectivity factors29 and besides,
reaches its peak performance in tert-amyl alcohol, which is less
than ideal from the practical standpoint. The efficacy of KR
depends on the presence of all of the fast-reacting enantiomer
of the substrate in solution. Therefore, it is usually critical to
ensure complete dissolution of the racemic substrate prior to
the start of the reaction. For example, oxazolidinone 21, which
is singularly insoluble in tert-amyl alcohol, could not be resolved
at all using Cl-PIQ in this solvent. Fortunately, it dissolved
relatively easily in chloroform at room temperature, which
permitted its successful KR using BTM (Table 2, entry 8). This
is not to suggest, however, that Cl-PIQ can be dismissed in
favor of BTM. In fact, we have found a number of
oxazolidinone substrates that fail under BTM catalysis, but
produce respectable results using Cl-PIQ (Table 1, entries 6a
and 6b, Table 2, entries 9a and 9b). Some enhancement of
enantioselectivity in Cl-PIQ-catalyzed reactions is observed at
RESULTS AND DISCUSSION
■
Oxazolidinones. In 2006, we disclosed the first examples of
KR of oxazolidinones via enantioselective N-acylation.12a Our
inspiration came from prior reports27 that N-acylation of
oxazolidinones with anhydrides is effectively promoted by
DMAP, the classical achiral acyl transfer catalyst.28 The results
of our initial investigation of this class of substrates, as well as
subsequent, previously unpublished findings are summarized in
Tables 1 and 2. Generally speaking, benzotetramisole (BTM)
311c has emerged as the catalyst of choice for most of the
oxazolidinones studied, due to its outstanding enantioselectivity
and the convenience of the optimal reaction conditions
(chloroform at room temperature). 7-Chloro-1,2-dihydro-2-
17606
dx.doi.org/10.1021/ja306766n | J. Am. Chem. Soc. 2012, 134, 17605−17612

Journal of the American Chemical Society
Article
There are, however, several notable differences between
oxazolidinones and alcohols.11f For example, increasing the size
of the a-alkyl group on benzylic alcohol substrates (Me→ Et→i-
Pr→t-Bu) leads to progressively higher selectivity factors. In the
case of oxazolidinones, the trend is not so clear. Although gem-
dimethyl substitution of C5 usually increases enantioselectiv-
ities when C4 substituent is an aryl (phenyl, 1-naphthyl or 2-
naphthyl), it makes little difference when R is heteroaryl (2-
furyl or 2-thienyl). Furthermore, KR of 1-thienylethanol is
much less enantioselective than that of 1-phenylethanol (s = 19
vs 80 under similar conditions). In contrast, BTM-catalyzed KR
of 2-thienyl-oxazolidinone yielded one of the highest selectivity
factors obtained in this series, considerably outperforming that
of its phenyl counterpart (Table 1, entries 5b vs 1c, s = 430 vs
170). An even more intriguing difference is found in the
following example. The ring constrained benzylic alcohol 1-
indanol has proved to be an unsuitable substrate for our
catalysts. Its lack of reactivity has been explained by its inability
to adopt the reactive conformation, which would allow the
benzene ring to stack on top of the N-acylated catalyst.
However, its oxazolidinone counterpart, the indane-fused
substrate 22, produced fairly good enantioselectivity when
acylated in the presence of 10 mol % of Cl-PIQ (BTM was
completely ineffective in this case). The mechanistic origin of
the differences noted above is not clear. The only definite
conclusion they point to is that the structure-selectivity trends
observed for one class of substrates cannot be extrapolated
reliably to another one.
Table 2. KR of 4,5-Substituted Oxazolidinones
Survey of Structurally Related Lactams and Acyclic
Amides. Upon completion of our initial study of oxazolidi-
nones, we decided to map out the substrate scope of our new
enantioselective N-acylation methodology. We were disap-
pointed to find that pyrrolidinone 23 and imidazolidinone 24,
despite being close structural analogues of 4-phenyl-oxazolidin-
2-one 5, are completely unreactive in the presence of 10 mol %
catalyst loadings of either BTM or Cl-PIQ (Table 3, entries 1
and 2). On the other hand, benzyl ( )-pyroglutamate 25 did
react, albeit very slowly, in the presence of 10 mol % of Cl-PIQ
and afforded excellent enantioselectivity (entry 3).
Comparison of the results obtained in the oxazolidinone and
pyrrolidinone series (Table 1, entries 1 and 8; Table 3, entries 2
and 3, respectively) suggested that the reactivity of lactams
toward N-acylation might be dependent on the pKa of the N−
H bond.31 Indeed, while substrate 23 was totally unreactive,
replacing the amide moiety in its structure with a carbamate (cf.
5), or even the α-phenyl with the more electron-withdrawing
ester group (cf. 25), was sufficient to permit acylation.
Combination of the two, as expected, resulted in the higher
reactivity of substrate 12 (Table 1, entry 8). With this in mind,
we turned our attention to another structural type of 5-
membered lactams bearing an additional heteroatom in the ring
and thus predicted to have increased N−H acidity: oxazolidin-
4-one 26, thiazolidin-4-one 27, and imidazolidin-4-one 28.
Gratifyingly, all three of them underwent acylation in the
presence of BTM with excellent selectivity factors and,
pleasingly, much higher reaction rates than their oxazolidin-2-
one prototype 5. It is noteworthy that these three compounds
are closely related to Seebach’s chiral enolate precursors.5
Although they are easily obtained in racemic form via acid-
catalyzed condensation of benzaldehyde, their asymmetric
synthesis would not be trivial. Therefore, the KR procedure
described herein should be of practical value for their
preparation in enantiopure form. We also examined 6-phenyl-
a
General conditions: same as Table 1, except as noted. 1.5 equiv (i-
b
c
PrCO)2O. Substrate incompletely dissolved at 0.1 M. 0.05 M
substrate.
low substrate concentrations (Table 1, entries 1a vs 1b, 0.2 and
0.05 M, respectively). Homobenzotetramisole (HBTM) 4a,
which was developed a few years later,11d also produced a
respectable selectivity factor when used in tert-amyl alcohol
(Table 1, entry 1d). However, it proved to be less active than
either 2 or 3 and therefore was not pursued further.
The absolute sense of enantioselection and the general
structure-selectivity trends noted in this study were consistent
with the cation−π interaction model proposed earlier in the
context of enantioselective O-acylation.11a,e As we had
expected, an aryl or heteroaryl group next to the nitrogen of
the substrates was important not only for the enantioselectivity,
but even reactivity. Thus, isopropyl-substituted oxazolidinone
13 simply failed to react under our standard reaction conditions
(entry 9). Later, we demonstrated that nonaromatic sub-
stituents, e.g., an alkene, alkyne, or ester group can also be quite
effective as chiral recognition elements (Table 1, entries 6−8),
which, again, is in accord with the successful KR of the
analogous alcohols by the same catalysts.11f,30
17607
dx.doi.org/10.1021/ja306766n | J. Am. Chem. Soc. 2012, 134, 17605−17612

Journal of the American Chemical Society
Article
Table 3. KR of Related Lactams and Amides
Table 4. KR of Thiolactams
a
General conditions: same as Table 1, except as noted. 0.1 M
b
c
substrate. Substrate incompletely dissolved. Conversion estimated by
d
1H NMR. Product decomposed during chromatography.
reached only low conversions during acylation. Furthermore,
the acylated product proved to decompose easily on silica gel.
Similarly slow reaction was observed in the case of N-(α-
phenethyl)thioformamide 36. Ee’s of the products in both of
these cases were low, which suggested that further efforts in this
direction were unlikely to produce a practically useful result.
β-Lactams. In a recent parallel development in our group,
we demonstrated that N-aroyl-β-lactams could be efficiently
resolved via BTM-catalyzed enantioselective alcoholysis.13d
While this method produced good to excellent enantioselectiv-
ities and displayed a broad substrate scope, we were aware that
the N-benzoyl group could not be removed without opening
the four-membered ring. Therefore, it was not suitable for
producing N-unsubstituted β-lactams themselves in enantio-
pure form. This consideration prompted us to revisit the N-
acylation chemistry. We were worried, of course, that, by
analogy with 5-phenylpyrrolidinone mentioned above, the pKa
of the amide moiety of β-lactams might be too high to permit
acylation with ABCs. Indeed, both BTM and HBTM, proved to
be ineffective. Fortunately, Cl-PIQ lived up to our expectations
(Table 5). After some optimization of the reaction conditions,
we were able to obtain modest to good selectivity factors with a
range of substrates bearing an aryl or heteroaryl group at C4.
Additional alkyl substituents at C3 were tolerated, although the
enantioselectivity and the reaction rates were diminished in
these cases (cf. entries 8 vs 1a, 9 vs 6). Indene-derived substrate
46 (entry 10) produced an encouraging selectivity factor 13,
which is in line with the result obtained in the oxazolidinone
series (cf. 22, Table 2) and in contrast to the corresponding
alcohol, 1-indanol.11f At the same time, substrate 47 failed to
undergo acylation completely, which again suggests the
importance of π-interactions (entry 11). Finally, we explored
the possibility of rendering this class of substrates more reactive
by replacing the lactam oxygen with sulfur (cf. Table 4).
Indeed, thio-β-lactam 48 (Table 5, entry 12) reacted very
a
b
General conditions: same as Table 1, except as noted. At rt. 0.1 M
substrate.
oxazinone 29, the ring-expanded analogue of 5, as well as
acyclic secondary amides, N-(α-phenethyl)-formamide 30 and
-trifluoroacetamide 31. Unfortunately, all of these failed to react
under our standard conditions.
Thiolactams. The apparent correlation between pKa and
reactivity noted above suggested that thiolactams, known to be
more acidic than the analogous lactams, might display
enhanced activity. Indeed, 4-phenyloxazolidine-2-thione 32
(Table 4, entry 1) underwent more rapid acylation than its
oxygen counterpart 5 (Table 1, entry 1c). Although the
enantioselectivity was reduced relative to the latter, it was still
amply sufficient for practical purposes. The corresponding
dithio derivative 33, unfortunately, proved to be virtually
insoluble in chloroform and other suitable reaction solvents and
dissolved only gradually in the course of acylation. Thus, the
calculated values for the selectivity factor and the conversion
may not reflect the true enantioselectivity and rate of the
reaction. We were especially pleased to find that 34, the thio
analogue of the completely unreactive pyrrolidinone 23,
underwent BTM-catalyzed KR smoothly and with an excellent
selectivity factor. Oxazine-2-thione 35, on the other hand,
17608
dx.doi.org/10.1021/ja306766n | J. Am. Chem. Soc. 2012, 134, 17605−17612

Journal of the American Chemical Society
Article
finer details of the reaction mechanism and transition state
geometry in this process still remained unclear.
Table 5. KR of β-Lactams
In the acylation of alcohols, the hydroxyl attacking the acyl
carbonyl is hydrogen-bound to the carboxylate counterion,
which acts as a general base. Thus, the C−O bond formation
and the O−H bond cleavage occur in a concerted fashion. Does
the N-acylation of lactam substrates also happen in the same
way?
We have considered three possible pathways illustrated
schematically in Figure 4 for 4-phenyloxazolidinone 5: (a) N-
Figure 4. Three possible pathways of catalytic N-acylation of 5.
acylation proceeding via concerted deprotonation of the lactam
nitrogen by the carboxylate anion, in direct analogy to the
acylation of alcohols; (b) Initial acylation of the lactam oxygen
with subsequent O-to-N acyl transfer, and (c) N-acylation of
the lactim tautomer 5a hydrogen-bound to the carboxylate
anion. M06-2X calculations32,33 on a simplified model (using
des-phenyl-BTM as catalyst and acetic anhydride) indicate that
the transition states TS-A and TS-B representing the first two
scenarios are 7.9 and 15.4 kcal/mol higher in energy,
respectively, than the isomeric transition state TS-C. The latter
is favored by the higher nucleophilicity of the lactim nitrogen
than of either the nitrogen or the oxygen of the lactam form.34
Using transition state TS-C as the starting point, we analyzed
the origin of enantioselectivity in the acylation of 4-phenyl-
oxazolidinone 5 using M06-2X. The geometries of the
competing diastereomeric transition states with (R)-BTM are
shown in Figure 5. Both transition states involve a concerted
C−N bond formation and deprotonation by the acetate
General conditions: 0.05 M ( )-substrate concentration, 2.0 equiv (i-
PrCO)2O, 2.0 equiv i-Pr2NEt, tert-amyl alcohol at 0 °C, unless
a
b
1
specified otherwise. In CDCl3 at rt. Conversion estimated by H
c
NMR. 4.0 equiv (i-PrCO)2O, 4.0 equiv i-Pr2NEt.
rapidly under BTM catalysis. Alas, virtually no enantioselectiv-
ity was observed in this reaction.
Computational Studies. Taking into account that the N-
acylation project was initially conceived by analogy with KR of
alcohols, it is not surprising that we interpreted all our
experimental data in accord with the previously proposed
transition state model for the latter process11a,e (Figure 2).
Indeed, all successfully resolved lactam substrates exhibited the
same absolute sense of enantioselection as their alcohol
counterparts. This general observation was consistent with π-
interactions playing a major role in the transition states leading
to the N-acylation of 5- and 4-membered lactams. However,
Figure 5. Transition states of BTM-catalyzed acylation of 5 (in
CHCl3).
17609
dx.doi.org/10.1021/ja306766n | J. Am. Chem. Soc. 2012, 134, 17605−17612

Journal of the American Chemical Society
Article
counterion (representing the isobutyrate anion in the actual
reaction). In both structures shown, the oxazolidinone oxygen
forms a weak hydrogen bond to the C2-position of the catalyst
(O−H distances are both ∼2.3 Å). The acetate counterion also
forms additional hydrogen bonds to the C2 and C3 of the
catalyst. In the fast-reacting diastereomer, the phenyl ring of the
substrate is stacked over the thiazolium moiety of the catalyst
due to the cation-π interactions, as expected by analogy with
the acylation of alcohols. The computed free energy difference
was in a good agreement with the experimental result
(selectivity factor s = 170 at room temperature corresponds
to ΔG298K = 3.0 kcal/mol).
With this result in hand, we proceeded to analyze the
transition states for the Cl-PIQ-catalyzed KR of three
representative lactams: 4-phenyloxazolidinone 5, its ring-
constrained analogue, the Indane-fused oxazolidinone 22, and
4-phenylazetidinone 37 (Figures 6−8, respectively). Qualita-
transition states. Strong cation-π interactions clearly contrib-
uted to the stabilization of all the fast-reacting enantiomers,
even 22, despite the rigidly constrained orientation of its
benzene ring (Figure 7). In fact, our calculations overestimated
the free energy differences for all Cl-PIQ-based structures,
especially the one with 4-phenyloxazolidin-2-one 5 (Figure 6),
in contrast to the good agreement obtained for the BTM-based
transition states (Figure 5). In the actual Cl-PIQ-catalyzed KR,
substrate 5 produced s = 17 in CHCl3 and s = 41 in tert-amyl
alcohol at 0 °C, i.e., ΔG273K = 1.5 or 2.0 kcal/mol, respectively.
The reason for this discrepancy is unclear at present.
We speculate that the ease of tautomerization is likely to be
one of the key factors determining the rate of acylation. What
we do know is that the more acidic substrates generally get
acylated faster. Since experimental pKa values for many of these
compounds have not been reported, we have used instead the
M06-2X computed proton affinities of the corresponding
anions relative to the 4-phenyl-oxazolidinone anion (Figure 9).
Figure 6. Transition states of Cl-PIQ-catalyzed acylation of 5 (in
CHCl3).
Figure 9. Computed relative proton affinities of conjugate bases of
lactams and thiolactams. Negative values indicate that the N−H bond
is more acidic than the N−H bond in 5.
Figure 7. Transition states of Cl-PIQ-catalyzed acylation of 22 (in
CHCl3).
A qualitative correlation between the acidity and the reactivity
of the substrates was observed. The only obvious outlier in the
series displayed in Figure 9 is the oxazine-2-thione 35, which
displays poor reactivity despite its high acidity, apparently due
to its different geometry. Even though this approach provides
only a rough estimate of reactivity, it does allow us to predict
with some confidence the likelihood of success in the acylation
of new substrates.
CONCLUSION
■
Figure 8. Transition states of Cl-PIQ-catalyzed acylation of 37 (in
CHCl3).
Our studies described above demonstrate that several classes of
4- and 5-membered lactams and thiolactams are amenable to
catalytic enantioselective N-acylation using amidine-based
catalysts. This methodology provides a means for achieving
their kinetic resolution under simple experimental conditions,
using only inexpensive reagents, and often with excellent
tively similar results were obtained in each of these cases, except
that the acyl carbonyls were twisted markedly out of the plane
of the catalyst and pyramidalized in both diastereomeric
17610
dx.doi.org/10.1021/ja306766n | J. Am. Chem. Soc. 2012, 134, 17605−17612

Journal of the American Chemical Society
Article
selectivity factors.35 The presence of an aryl, alkenyl, alkynyl or
ester moiety adjacent to the NH group is apparently critical for
the enantioselectivity. Computational studies indicate that
cation-π interactions play a key role in the chiral recognition
of lactam substrates. There is no simple answer to the question
“which catalyst is the best?” Based on available data, Cl-PIQ has
broader substrate scope (see results with substrates 10, 22, 25,
and 37), while BTM exhibits higher enantioselectivity in the
KR of those substrates where a direct comparison has been
made (see Tables 1 and 2). As a rule, we have given priority to
BTM when exploring new substrates with proton affinities
equal to or lower than that of the conjugate base of 4-phenyl-
oxazolidin-2-one 5.
(4) E.g., chiral dirhodium carboxamidates: Doyle, M. P.; Duffy, R.;
Ratnikov, M.; Zhou, L. Chem. Rev. 2010, 110, 704.
(5) (a) Imidazolidin-4-ones as chiral enolate equivalents: Seebach,
D.; Sting, A, R.; Hoffmann, M. Angew. Chem., Int. Ed. Engl. 1996, 35,
2708 and references cited therein. (b) Vince lactam (2-
azabicyclo[2.2.1]hept-5-en-3-one) in the synthesis of pharmaceuticals:
Singh, R.; Vince, R. Chem. Rev. 2012, 112, 4642 and references cited
therein.
(6) Pyrrolidinones: (a) Proteasome inhibitor Lactacystin: Omura, S.;
Fujimoto, T.; Otoguro, K.; Matsuzaki, K.; Moriguchi, R.; Tanaka, H.;
Sasaki, Y. J. Antibiot. 1991, 44, 113. (b) Proteasome inhibitor
Salinosporamide A: Feling, R. H.; Buchanan, G. O.; Mincer, T. J.;
Kauffman, C. A.; Jensen, P. R.; Fenical, W. Angew. Chem., Int. Ed. 2003,
42, 355. (c) Angiogenesis inhibitor Streptopyrrolidine: Shin, H. J.;
Kim, T. S.; Lee, H.-S.; Park, J. Y.; Choi, I.-K.; Kwon, H. J.
Phytochemistry 2008, 69, 2363. (d) Cytokine modulator Cytoxazone
(oxazolidin-2-one): Kakeya, H.; Morishita, M.; Koshino, H.; Morita,
T.-i.; Kobayashi, K.; Osada, H. J. Org. Chem. 1999, 64, 1052. (e)
Tetramic acids (pyrrolidine-2,4-diones): Royles, B. J. L. Chem. Rev.
1995, 95, 1981. See also natural β-lactam antibiotics (ref 2d).
(7) Asymmetric synthesis of β-lactams: (a) C−H insertion: Doyle,
M. P.; Kalinin, A. V. Synlett 1995, 1075. (b) Staudinger reaction:
Hodous, B. L.; Fu, G. C. J. Am. Chem. Soc. 2002, 124, 1578. France, S.;
Weatherwax, A.; Taggi, A. E.; Lectka, T. Acc. Chem. Res. 2004, 37, 592.
Kinugasa reaction: Shintani, R.; Fu, G. C. Angew. Chem., Int. Edit.
2003, 42, 4082. (c) NHC-catalyzed annulation: He, M.; Bode, J. W. J.
Am. Chem. Soc. 2008, 130, 418.
(8) Dihydropyridones: (a) He, M.; Struble, J. R.; Bode, J. W. J. Am.
Chem. Soc. 2006, 128 (26), 8418. (b) Wanner, B.; Mahatthananchai, J.;
Bode, J. W. Org. Lett. 2011, 13, 5378. (c) Simal, C.; Lebl, T.; Slawin, A.
M. Z.; Smith, A. D. Angew. Chem., Int. Edit. 2012, 51, 3653.
(9) See, e.g., (a) rac-Imidazolidin-4-ones: Zehavi, U.; Ben-lshai, D. J.
Org. Chem. 1961, 26, 1097. See also ref 5a. rac-β-Lactams:
(b) Rasmussen, J. K.; Hassner, A. Chem. Rev. 1976, 76, 389. rac-
Oxazolidinones: (c) Ishibuchi, S.; Ikematsu, Y.; Ishizuka, T.; Kunieda,
T. Tetrahedron Lett. 1991, 32, 3523. (d) Espino, C. G.; Du Bois, J.
Angew. Chem., Int. Ed. 2001, 40, 598. (e) He, M.; Bode, J. W. Org. Lett.
2005, 7, 3131. See also ref 5b.
ASSOCIATED CONTENT
■
S
* Supporting Information
General experimental details, previously unreported kinetic
resolution data, characterization and preparation data for
previously unreported compounds, details of computational
studies. This material is available free of charge via the Internet
at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
houk@chem.ucla.edu; birman@wustl.edu
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Dr. Yuhua Zhang for providing preliminary data with
catalyst 4b. Financial support for these studies has been
provided by National Institutes of Health (R01 GM072682 to
V.B.B.) and National Science Foundation (CHE-1059084 to
K.N.H.; CHE-1012979 to V.B.B.). Calculations were per-
formed on the Hoffman2 cluster at UCLA and the Extreme
Science and Engineering Discovery Environment (XSEDE),
which is supported by the National Science Foundation (OCI-
1053575).
(10) For a review of resolution methods, see: Faigl, F.; Fogassy, E.;
Nog
519.
́
rad
́
i, M.; Pal
́
ovics, E.; Schindler, J. Tetrahedron Asymm. 2008, 19,
(11) Catalyst development and enantioselective O-acylation: (a) Bir-
man, V. B.; Uffman, E. W.; Jiang, H.; Li, X.; Kilbane, C. J. J. Am. Chem.
Soc. 2004, 126, 12226. (b) Birman, V. B.; Jiang, H. Org. Lett. 2005, 7,
3445. (c) Birman, V. B.; Li, X. Org. Lett. 2006, 8, 1351. (d) Birman, V.
B.; Li, X. Org. Lett. 2008, 10, 1115. (e) Li, X.; Liu, P.; Houk, K. N.;
Birman, V. B. J. Am. Chem. Soc. 2008, 130, 13836. (f) Li, X.; Jiang, H.;
Uffman, E. W.; Guo, L.; Zhang, Y.; Yang, X.; Birman, V. B. J. Org.
Chem. 2012, 77, 1722 and references cited therein.
(12) Enantioselective N-acylation: (a) Birman, V. B.; Jiang, H.; Li, X.;
Guo, L.; Uffman, E. W. J. Am. Chem. Soc. 2006, 128, 6536. (b) Yang,
X.; Bumbu, V. D.; Birman, V. B. Org. Lett. 2011, 13, 4755.
(13) Enantioselective alcoholysis of acyl donors: (a) Yang, X.; Lu, G.;
Birman, V. B. Org. Lett. 2010, 12, 892. (b) Yang, X.; Birman, V. B.
Angew. Chem., Int. Ed. 2011, 50, 553. (c) Yang, X.; Birman, V. B.
Chem.Eur. J. 2011, 17, 11296. (d) Bumbu, V. D.; Birman, V. B. J.
Am. Chem. Soc. 2011, 133, 13902. (e) Liu, P.; Yang, X.; Birman, V. B.;
Houk, K. N. Org. Lett. 2012, 14, 3288. (f) Yang, X.; Liu, P.; Houk, K.
N.; Birman, V. B. Angew. Chem., Int. Ed. 2012, 51, 9776.
REFERENCES
■
(1) We define the terms “lactams” and “thiolactams” broadly, to
describe any structure containing an amide −CO−NR− or thioamide
−CS−NR− moiety within a ring that may or may not contain
additional heteroatoms.
(2) (a) Phosphodiesterase inhibitor Rolipram (pyrrolidinone):
Schwabe, U.; Miyake, M.; Ohga, Y.; Daly, J. W. Mol. Pharmacol. 1976,
12, 900. (b) Antibiotic Linezolid (oxazolidin-2-one): Brickner, S. J.;
Hutchinson, D. K.; Barbachyn, M. R.; Manninen, P. R.; Ulanowicz, D.
A.; Garmon, S. A.; Grega, K. C.; Hendges, S. K.; Toops, D. S.; Ford, C.
W.; Zurenko, G. E. J. Med. Chem. 1996, 39, 673. (c) Cholesterol-
lowering drug Ezetimibe (β-lactam): Rosenblum, S. B.; Huynh, T.;
Afonso, A.; Davis, H. R.; Yumibe, N.; Clader, J. W.; Burnett, D. A. J.
Med. Chem. 1998, 41, 973. (d) Testero, S. A.; Fisher, J. F.; Mobashery,
S. β-Lactam antibiotics. In Burger’s Medicinal Chemistry, Drug Discovery
and Development, 7th ed.; Abraham, D. J., Rotella, D. P., Eds.; Wiley:
Hoboken, NJ, 2010; Vol. 7 (Antiinfectives), pp 259−404.
(14) Kagan, H. B.; Fiaud, J. C. Top. Stereochem. 1988, 18, 249.
(15) For a recent review, see: Krasnov, V. P.; Gruzdev, D. A.; Levit,
G. L. Eur. J. Org. Chem. 2012, 1471.
(16) For a review of enzymatic KR of amines: van Rantwijk, F.;
Sheldon, R. A. Tetrahedron 2004, 60, 501.
(3) Oxazolidin-2-ones: (a) Evans, D. A. Aldrichimica Acta 1982, 15,
23. (b) Ager, D. J.; Prakash, I.; Schaad, D. R. Aldrichimica Acta 1997,
(17) For recent reviews of nonenzymatic kinetic resolution and
́
30, 3. Oxazolidine-2-thiones and thiazolidine-2-thiones: (c) Velazquez,
enantioselective acylation catalysts, see: (a) Muller, C. E.; Schreiner, P.
F.; Olivo, H. F. Curr. Org. Chem. 2002, 6, 1. (d) Delaunay, D.; Toupet,
L.; Corre, M. L. J. Org. Chem. 1995, 60, 6604. Camphor-derived
lactams: (e) Boeckman, R. K.; Boehmler, D. J.; Musselman, R. A. Org.
Lett. 2001, 3, 3777.
̈
R. Angew. Chem., Int. Ed. 2011, 50, 6012. (b) Pellissier, H. Adv. Synth.
Catal. 2011, 353, 1613. (c) Spivey, A. C.; Arseniyadis, S. Top. Curr.
Chem. 2010, 291, 233. (d) Denmark, S. E.; Beutner, G. L. Angew.
17611
dx.doi.org/10.1021/ja306766n | J. Am. Chem. Soc. 2012, 134, 17605−17612

Journal of the American Chemical Society
Article
Chem., Int. Ed. 2008, 47, 1560. (e) Vedejs, E.; Jure, M. Angew. Chem.,
Int. Ed. 2005, 44, 3974.
(18) (a) Arai, S.; Bellemin-Laponnaz, S.; Fu, G. C. Angew. Chem., Int.
Ed. 2001, 40, 234. (b) Arp, F. O.; Fu, G. C. J. Am. Chem. Soc. 2006,
128, 14264.
(19) (a) De, C. K.; Klauber, E. G.; Seidel, D. J. Am. Chem. Soc. 2009,
131, 17060. (b) Klauber, E. G.; De, C. K.; Shah, T. K.; Seidel, D. J. Am.
Chem. Soc. 2010, 132, 13624. (c) Klauber, E. G.; Mittal, N.; Shah, T.
K.; Seidel, D. Org. Lett. 2011, 13, 2464. (d) De, C. K.; Seidel, D. J. Am.
Chem. Soc. 2011, 133, 14538. (e) Mittal, N.; Sun, D. X.; Seidel, D. Org.
Lett. 2012, 14, 3084.
(35) Isobutyric anhydride was demonstrated to provide the highest
enantioselectivity in our previous investigations12 and thus has been
used as the acylating agent throughout this study. Previously described
deacylation of N-acylated oxazolidin-4-ones, oxazolidin-4-thiones, and
thiazolidin-4-thiones (see, e.g., refs 3a−3d) should permit easy
recovery of the parent compounds. Solvolysis of several other KR
products (e.g., N-isobutyryl-β-lactams or N-acyl derivative of
imidazolidin-4-one 28) is expected to result in their ring-opening.
(20) Binanzer, M.; Hsieh, S. Y.; Bode, J. W. J. Am. Chem. Soc. 2011,
133, 19698.
(21) Vries, T.; Wynberg, H.; van Echten, E.; Koek, J.; ten Hoeve, W.;
Kellogg, R. M.; Broxterman, Q. B.; Minnaard, A.; Kaptein, B.; van der
Sluis, S.; Hulshof, L.; Kooistra, J. Angew. Chem., Int. Ed. 1998, 37, 2349.
(22) For examples of noncatalytic methods of enantioselective N-
acylation, see: (a) Al-Sehemi, A. G.; Atkinson, R. S.; Fawcett, J.;
Russell, D. R. Tetrahedron Lett. 2000, 41, 2239. (b) Arseniyadis, S.;
Valleix, A.; Wagner, A.; Mioskowski, C. Angew. Chem., Int. Ed. 2004,
43, 3314. (c) Maezaki, N.; Furusawa, A.; Uchida, S.; Tanaka, T.
Tetrahedron 2001, 57, 9309. (d) Karnik, A. V.; Kamath, S. S.
Tetrahedron Asymm. 2008, 19, 45.
(23) (a) Ishizuka, T.; Kimura, K.; Ishibuchi, S.; Kunieda, T. Chem.
Lett. 1992, 991. (b) Hashimoto, N.; Ishizuka, T.; Kunieda, T.
Tetrahedron Lett. 1998, 39, 6317.
(24) (a) Jouglet, B.; Rousseau, G. Tetrahedron Lett. 1993, 34, 2307.
(b) Csomos
Asymmetry 1996, 7, 1789.
́ ́
, P.; Kanerva, L. T.; Bernath, G.; Fulop, F. Tetrahedron:
̈
̈
(25) (a) Evans, C.; McCague, R.; Roberts, S. M.; Sutherland, A. G.;
Wisdom, R. J. Chem. Soc., Perkin Trans. 1 1991, 2276. (b) Brieva, R.;
Crich, J. Z.; Sih, C. J. J. Org. Chem. 1993, 58, 1068. (c) Forro,
T.; Tasnadi, G.; Fulop, F. Adv. Synth. Catal. 2006, 348, 917. (d) For a
review of available enzymatic methods, see: Forro,
Rev. Org. Chem. 2004, 1, 93.
́ ́
E.; Paal,
́
̈
̈
́
E.; Fulop, F. Mini-
̈
̈
(26) Fowler, B. S.; Mikochik, P. J.; Miller, S. J. J. Am. Chem. Soc.
2010, 132, 2870.
(27) (a) Kano, S.; Yokomatsu, T.; Shibuya, S. Heterocycles 1990, 31,
1711. (b) Ager, D. J.; Allen, D. R.; Schaad, D. R. Synthesis 1996, 1283.
(28) Hofle, G.; Steglich, W.; Vorbruggen, H. Angew. Chem., Int. Edit.
Engl. 1978, 17, 569.
(29) Enantioselectivity in KR of racemic mixtures is expressed in
terms of a selectivity factor (s) defined as the ratio of reaction rates of
the fast- and the slow-reacting enantiomers of the starting material: s =
kfast/kslow. Kagan’s equations (ref 14) are used to calculate it from the
ee values of the product and the unreacted starting material:
conversion C = eeSM/(eeSM + eePR); selectivity factor s = ln[(1 −
C)(1 − eeSM)]/ln[(1 − C)(1 + eeSM)]. All C and s values reported in
Tables 1−5 were obtained by averaging the results of duplicate runs
(see Supporting Information). However, due to the calculation error
associated with high ee’s, large s values should be viewed as
approximate.
(30) Shiina, I.; Nakata, K.; Ono, K.; Sugimoto, M.; Sekiguchi, A.
Chem.Eur. J. 2010, 16, 167.
(31) A similar observation has been made in the context of DMAP-
catalyzed acylation of lactams with Boc2O: Hansen, M. M.; Harkness,
A. R.; Coffey, D. S.; Bordwell, F. G.; Zhao, Y. Tetrahedron Lett. 1995,
36, 8949.
(32) (a) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215.
(b) Zhao, Y.; Truhlar, D. G. Acc. Chem. Res. 2008, 41, 157.
(33) Geometries were optimized with M06-2X/6-31G(d) with the
SMD solvation model in chloroform. Single point energies were
calculated with M06-2X/6-311++G(d,p). See Supporting Information
for details of computational studies.
(34) Under the reaction conditions, the tautomerization between 5
and 5a is expected to occur via a base-assisted mechanism and to be
faster than the subsequent acylation step. Thus, the choice between
the three pathways is determined by the energy differences between
transition states TS-A, TS-B, and TS-C. See Supporting Information.
17612
dx.doi.org/10.1021/ja306766n | J. Am. Chem. Soc. 2012, 134, 17605−17612