Boronic Acid-Catalyzed, Highly Enantioselective Aza-Michael Additions of Hydroxamic Acid to Quinone Imine Ketals

Article abstract of DOI:10.1021/jacs.5b11518

Boronic acid is one of the most versatile organic molecules in chemistry. Its uses include organic reactions, molecular recognition, assembly, and even medicine. While boronic acid catalysis, which utilizes an inherent catalytic property, has become an important research objective, it still lags far behind other boronic acid chemistries. Here, we report our discovery of a new boronic acid catalysis that enables the aza-Michael addition of hydroxamic acid to quinone imine ketals. By using 3-borono-BINOL as a chiral boronic acid catalyst, this reaction could be implemented in a highly enantioselective manner, paving the way to densely functionalized cyclohexanes.

Full text of DOI:10.1021/jacs.5b11518

Communication
pubs.acs.org/JACS
Boronic Acid-Catalyzed, Highly Enantioselective Aza-Michael
Additions of Hydroxamic Acid to Quinone Imine Ketals
Takuya Hashimoto, Alberto Osuna Galvez, and Keiji Maruoka*
́
Department of Chemistry, Graduate School of Science, Kyoto University, Sakyo, Kyoto 606-8502, Japan
S
* Supporting Information
Scheme 1. Use of Different Chiral Catalysts
ABSTRACT: Boronic acid is one of the most versatile
organic molecules in chemistry. Its uses include organic
reactions, molecular recognition, assembly, and even
medicine. While boronic acid catalysis, which utilizes an
inherent catalytic property, has become an important
research objective, it still lags far behind other boronic acid
chemistries. Here, we report our discovery of a new
boronic acid catalysis that enables the aza-Michael addition
of hydroxamic acid to quinone imine ketals. By using 3-
borono-BINOL as a chiral boronic acid catalyst, this
reaction could be implemented in a highly enantioselective
manner, paving the way to densely functionalized
cyclohexanes.
s represented by Suzuki−Miyaura coupling,1 there are a
Amyriad of organic transformations that rely on boronic
acid as a key functionality.2 The unique property of boronic
acidsforming covalent bonds reversibly with alcoholshas
also been extensively applied in molecular recognition, sensors,
and self-assembly.3,4 More recently, this property of boronic
acid catalysis by disclosing the operation of a catalyst dimer in
acids has even found use in medicine, such as in the
commercialized drugs bortezomib and tavaborole.5,6 Far less
developed, compared with these applications, is the use of
boronic acid itself as a catalyst in organic synthesis, although it
has received increasing attention recently due to its distinct
mode of action, facilitating transformations unachievable by
other catalyses.7−9 From the practical point of view, the ready
availability, stability, and low toxicity of boronic acid also justify
its use as a catalyst. One obstacle to expanding boronic acid
catalysis is the limited knowledge about its mode of action, as
the electrophilic activation of carboxylic acids, reported as early
as 1996, still stands as the only reliable protocol.10 Another
challenge is the lack of a boronic acid catalyst that controls the
stereoselectivity rigorously. In the field of asymmetric catalysis,
boronic acid catalysis has not been recognized as a viable
option, and only a few examples exist in the literature, with
limited success reported.11,12
Here, we report our discovery of a new boronic acid-
catalyzed reaction which turned into a highly enantioselective
catalysis. It was revealed that the aza-Michael addition of
hydroxamic acid to quinone imine ketals is uniquely promoted
by a catalytic amount of boronic acid, and the reaction could be
implemented smoothly with enantioselectivities >90% by use of
3-borono-BINOL. The asymmetric reaction proceeded even at
2 mol% catalyst loading, which is unusually low in boronic acid
catalysis. We also shed new light on the mechanism of boronic
the reaction.
The initial clue for the development of new chiral boronic
acid catalysis emerged serendipitously in the course of our
study using quinone imine ketals as prochiral electrophiles in
chiral Brønsted acid catalysis.13,14 Stimulated by a recent report
on the base-mediated addition of hydroxamic acid to quinone
ketals,15 we became interested in the development of oxa-
Michael addition of hydroxamic acid 2 to quinone imine ketals
1 catalyzed by chiral Brønsted acid (Scheme 1). As a
preliminary experiment, we set up two reactions using different
chiral Brønsted acid catalysts, 3 and 4, both developed in this
group.16,17 The use of axially chiral dicarboxylic acid 3 gave the
expected bicyclic compound 7, derived from the oxa-Michael
addition (1a to 5) and sequential intramolecular cyclization. On
the other hand, the reaction catalyzed by the boronate ester-
assisted chiral carboxylic acid 4 gave rise to the product 8a,
derived from the initial aza-Michael addition (1a to 6), in a
racemic form.18 The fact that the product was obtained in a
racemic form indicated that the chiral diol moiety of catalyst 4
is dissociated and does not interfere in this catalysis, and that 2-
boronobenzoic acid is acting as a unique boronic acid catalyst.
To confirm this assumption, we screened some achiral
boronic acids, as shown in Table 1. As anticipated, 2-
Received: November 4, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.5b11518
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
a
Table 1. Optimization of the Reaction Conditions
entry
RB(OH)2
RB(OH)2, amount
o-NBA amount
yield of 6 (%)
ee of 6 (%)
yield of 8a (%)
ee of 8a (%)
1
2
3
4
5
6
7
8
9a
9b
10 mol%
10 mol%
10 mol%
10 mol%
10 mol%
10 mol%
10 mol%
10 mol%
10 mol%
10 mol%
10 mol%
2 mol%
trace
10
67
37
9
9c
trace
9
9d
38
55
78
69
76
84
42
19
91
10a
10a
10a
10a
10a
10b
10c
10a
10a
17
77
77
88
91
95
97
45
36
86
94
10 mol%
25 mol%
50 mol%
50 mol%
50 mol%
50 mol%
10 mol%
30 mol%
trace
trace
trace
trace
trace
trace
trace
trace
b
9
b
10
b
11
bc
,
12
13
bd
,
2 mol%
92
a
b
Performed with 1a (0.1 mmol), 2 (0.13 mmol), and a catalytic amount of RB(OH)2 in CH2Cl2 (1.0 mL) at 0 °C for 24 h. Performed at −10 °C.
c
d
Performed for 120 h on 1.0 mmol scale. Performed for 72 h on 1.0 mmol scale.
Scheme 2. Plausible Reaction Pathway
boronobenzoic acid 9a promoted the reaction efficiently, giving
8a in good yield (entry 1). In addition, 2-boronophenol 9b,
phenylboronic acid 9c, and even cyclohexylboronic acid 9d
accelerated the reaction to give 8a, albeit in poorer yields
(entries 2−4). In these latter cases, the uncyclized aza-Michael
adduct 6 was also isolated in a substantial amount.
With these preliminary results in hand, we commenced the
development of a chiral boronic acid catalyst. By screening a
variety of chiral boronic acids, we found out that easily
accessible 3-borono-BINOL 10a is a promising chiral catalyst,
with which the product was obtained with modest enantio-
selectivity (entry 5). The aza-Michael adduct 6 was also isolated
with the same level of enantioselectivity. In consideration of the
fact that 2-boronobenzoic acid 9a, which has a boronic acid and
a carboxylic acid in the molecule, promoted the reaction
smoothly (entry 1), we decided to add an achiral Brønsted acid
as a co-catalyst to improve the reactivity. For this purpose, o-
Figure 1. Investigation of the catalytic species.
nitrobenzoic acid (o-NBA) was identified as a proper co-
catalyst with which exclusive formation of 8a was achieved
(entry 6). Moreover, an increase in the enantioselectivity was
observed when using more of the co-catalyst (entries 7 and 8).
In the end, optimized conditions for this enantioselective
reaction were set to the use of 10 mol% catalyst 10a and 50 mol
% co-catalyst at −10 °C in CH2Cl2, by which the reaction
afforded the product 8a in 84% yield with 97% ee (entry 9).
The importance of the OH group of 3-borono-BINOL was
confirmed by carrying out the reaction using methyl-capped
catalysts 10b and 10c, with which lower yields and selectivities
were attained (entries 10 and 11). To demonstrate the
B
DOI: 10.1021/jacs.5b11518
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Scheme 3. Derivatization of the Products
Figure 2. (a) Plausible transition state and (b) observation of the
nonlinear effect.
scalability, the reaction was then conducted on 1 mmol scale
using 2 mol% chiral boronic acid catalyst 10a. It was found out
that increasing the ratio of o-NBA is critical to attain good
results. While the use of 10 mol% o-NBA and 2 mol% 10a took
5 days for completion of the reaction with a substantial
decrease of the ee (entry 12), the combination of 30 mol% o-
NBA and 2 mol% 10a gave rise to the product in 92% yield
with 94% ee within 3 days (entry 13). In the additional
experiments, we also confirmed that the reaction proceeds with
N-Boc-quinone imine ketal, while quinone ketal is not reactive
under these conditions.19
At this stage, we moved our attention to the elucidation of
the reaction pathway and the role of the chiral boronic acid and
o-NBA (Scheme 2). For this purpose, the uncyclized aza-
Michael adduct 6 was isolated and treated with either the
boronic acid or o-NBA. On the one hand, the treatment of 6
with chiral boronic acid 10a gave a mixture of substrates 1a and
2, the remaining 6, and the bicyclic product 8a, indicating that
the boronic acid catalyzes the reversible aza-Michael addition
and also promotes cyclization slowly. On the other hand, the
treatment of 6 with o-NBA resulted in the exclusive formation
of 8a, clarifying that the Brønsted acid accelerates the
intramolecular cyclization. These observations account for the
increase of the enantioselectivity in the reaction employing
more o-NBA (see Table 1, entries 6−8, 12, and 13). Additional
insight was obtained by the 1H NMR monitoring of the o-NBA-
catalyzed cyclization of 6, which showed the accumulation of
N,O-acetal 8a′ before the formation of 8a.
To gain insight into the interaction of boronic acid and
hydroxamic acid, we carried out NMR experiments of the
complex (Figure 1). Addition of deuterated benzohydroxamic
acid 2-d5 to 3-borono-BINOL 10a in CDCl3 led to a drastic
shift of the 1H NMR and 11B NMR signals, as shown in Figure
1 (see Supporting Information (SI) for details). 1H NMR of the
complex showed 22 discernible proton signals, indicating the
involvement of two molecules of 10a in an asymmetric manner.
The 11B NMR of the complex showed two peaks at 27.7 and
8.9 ppm, also indicative of a dimeric species having one boron
and one borate. In addition, ESI-MS measurement provided a
peak at m/z 744.2 (M−H)−, which corresponds to two
molecules of 10a and one hydroxamic acid 2 complexed by the
dehydration of three H2O.
Taking these facts into account, we postulated the catalyst
resting state which consists of a dioxazaborole20,21 and a
boronate half-ester (Figure 2a). From this species, acid−base
catalysis may operate to facilitate the aza-Michael addition,
wherein the quinone imine ketal is activated by hydrogen-
bonding with the free boronic acid and attacked by the
nucleophilically activated hydroxamic acid (indicated in gray).
Chart 1. Substrate Scope
C
DOI: 10.1021/jacs.5b11518
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
The alcohol moieties of the BINOLs might form hydrogen
bonds with each other and rigidify the dimeric catalyst
structure. Further support for the formation of a dimeric
species was provided experimentally by the observation of a
positive nonlinear effect (Figure 2b).22
Japanese Government (MEXT) Scholarship Program for the
fellowship.
REFERENCES
■
(1) Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457−2483.
(2) Boronic Acids: Preparation and Applications in Organic Synthesis,
Medicine and Materials, Vols. 1 and 2; Hall, D. G., Ed.; Wiley-VCH:
Weinheim, 2011.
With the optimized reaction conditions in hand, we
examined the substrate scope of this chiral boronic acid
catalysis (Chart 1). A variety of 3-alkyl- and 3-allyl-substituted
quinone imine ketals were applied without difficulty to give
8b−8e in good yields, with enantioselectivities >94%. The use
of 3-unsubstituted quinone imine ketal gave a mixture of N,O-
acetals 8f′ and 8f, presumably due to the equilibration between
these two structural isomers (see Scheme 2). Aromatic
substituents were tolerated to give a mixture of structural
isomers 8g′−8i′ and 8g−8i, the ratios of the isomers depending
on the aryl group. Both isomers were obtained with high
enantioselectivities. The reaction of 3-TBS-ethynyl quinone
imine ketal gave the corresponding N,O-acetal 8j′ exclusively
with 93% ee.
Considering the importance of chiral functionalized cyclo-
hexylamines as building blocks for aminoglycoside antibiotics
and antiviral drugs, we implemented the synthetic derivatization
of the products (Scheme 3). As one example, we attempted the
SmI2-mediated N−O bond cleavage of 8a to directly give
aminocyclohexanol 11 (Scheme 3a). By carrying out the
reaction at low temperature in the presence of trifluoroethanol
as a proton source, the desired product was obtained in good
yield. As another example, reductive amination was examined
using a mixture of structural isomers 8g′ and 8g (Scheme 3b).
The reduction was accomplished by use of NaBH3CN under
acidic conditions to converge the mixture into cis-diamine 12 in
good yield with high diastereoselectivity.
(3) Kubo, Y.; Nishiyabu, R.; James, T. D. Chem. Commun. 2015, 51,
2005−2020.
(4) Sun, X.; James, T. D. Chem. Rev. 2015, 115, 8001−8037.
(5) Ban, H. S.; Nakamura. Chem. Rec. 2015, 15, 616−35.
(6) Adamczyk-Wozniak, A.; Borys, K. M.; Sporzyn
ski, A. Chem. Rev.
́
2015, 115, 5224−5247.
(7) Zheng, H.; Hall, D. G. Aldrichimica Acta 2014, 47, 41−51.
́
(8) Dimitrijevic, E.; Taylor, M. S. ACS Catal. 2013, 3, 945−962.
(9) Taylor, M. S. Acc. Chem. Res. 2015, 48, 295−305.
(10) Ishihara, K.; Ohara, S.; Yamamoto, H. J. Org. Chem. 1996, 61,
4196−4197.
(11) Lee, D.; Newman, S. G.; Taylor, M. S. Org. Lett. 2009, 11,
5486−5489.
(12) Arnold, K.; Davies, B.; Her
Int. Ed. 2008, 47, 2673−2676.
́
ault, D.; Whiting, A. Angew. Chem.,
(13) Hashimoto, T.; Nakatsu, H.; Takiguchi, Y.; Maruoka, K. J. Am.
Chem. Soc. 2013, 135, 16010−16013.
(14) Hashimoto, T.; Nakatsu, H.; Maruoka, K. Angew. Chem., Int. Ed.
2015, 54, 4617−4621.
(15) Yin, Z.; Zhang, J.; Wu, J.; Liu, C.; Sioson, K.; Devany, M.; Hu,
C.; Zheng, S. Org. Lett. 2013, 15, 3534−3537.
(16) Hashimoto, T.; Maruoka, K. J. Am. Chem. Soc. 2007, 129,
10054−10055.
(17) Hashimoto, T.; Gal
2013, 135, 17667−17670.
(18) Enders, D.; Wang, C.; Liebich, J. X. Chem. - Eur. J. 2009, 15,
11058−11076.
́
vez, A. O.; Maruoka, K. J. Am. Chem. Soc.
In conclusion, we have disclosed that the aza-Michael
addition of hydroxamic acid to quinone imine ketal is uniquely
facilitated by a boronic acid, and the reaction can be
implemented in a highly enantioselective manner by use of
the co-catalyst system composed of 3-borono-BINOL and o-
nitrobenzoic acid. The mechanistic study revealed the dimeric
nature of the boronic acid catalyst, which activates the
hydroxamic acid as nucleophile by the formation of a
dioxazaborole. This study thereby provided a new mechanistic
foundation of boronic acid catalysis, proved its potential as a
new class of asymmetric catalyst, and will stimulate further
research in the field of chiral boronic acid catalysis.
(19) The reaction of N-Boc-3-methylquinone imine ketal afforded
the product in 48% yield with 64% ee (see SI for details).
(20) Ito, M.; Tokitoh, N.; Okazaki, R. Tetrahedron Lett. 1997, 38,
4451−4454.
(21) Vo, C.-V. T.; Mitchell, T. A.; Bode, J. W. J. Am. Chem. Soc. 2011,
133, 14082−14089.
(22) Kagan, H. B. Synlett 2001, 2001, 888−899.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.5b11518.
■
S
Experimental details and characterization data for new
compounds (PDF)
X-ray crystallographic data for 13 (CIF)
AUTHOR INFORMATION
Corresponding Author
*maruoka@kuchem.kyoto-u.ac.jp
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This work was partially supported by a Grant-in-Aid for
Scientific Research from the MEXT (Japan). A.O.G. thanks the
■
D
DOI: 10.1021/jacs.5b11518
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX