α-Halo Amides as Competent Latent Enolates: Direct Catalytic Asymmetric Mannich-Type Reaction

Article abstract of DOI:10.1021/jacs.7b03291

α-Halogenated carbonyl compounds are susceptible to dehalogenation and thus largely neglected as enolate precursors in catalytic enantioselective C-C bond-forming reactions. By merging the increased stability of the α-C-halogen bond of amides and the direct enolization methodology of the designed amide, we explored a direct catalytic asymmetric Mannich-type reaction of α-halo 7-azaindoline amides with N-carbamoyl imines. All α-halo substituents, α-F, -Cl, -Br, -I amides, were tolerated to provide the Mannich-adducts in a highly stereoselective manner without undesirable dehalogenation. The diastereoselectivity switched intriguingly depending on the substitution pattern of the aromatic imines, which is ascribed to stereochemical differentiation based on the open transition-state model. Functional group interconversion of the 7-azaindoline amide moiety of the Mannich-adducts and further elaboration into a diamide without dehalogenation highlight the synthetic utility of the present protocol for accessing enantioenriched halogenated chemical entities.

Full text of DOI:10.1021/jacs.7b03291

Subscriber access provided by CORNELL UNIVERSITY LIBRARY
Article
#-Halo Amides as Competent Latent Enolates:
Direct Cata-lytic Asymmetric Mannich-type Reaction
Bo Sun, Pandur Venkatesan Balaji, Naoya Kumagai, and Masakatsu Shibasaki
J. Am. Chem. Soc., Just Accepted Manuscript • Publication Date (Web): 22 May 2017
Downloaded from http://pubs.acs.org on May 22, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 9
Journal of the American Chemical Society
1
2
3
4
5
6
α-Halo Amides as Competent Latent Enolates: Direct Catalytic
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Asymmetric Mannich-type Reaction
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Bo Sun, Pandur Venkatesan Balaji, Naoya Kumagai,* and Masakatsu Shibasaki*
Institute of Microbial Chemistry (BIKAKEN), Tokyo, 3ꢀ14ꢀ23 Kamiosaki, Shinagawaꢀku, Tokyo 141ꢀ0021, Japan.
ABSTRACT: αꢀHalogenated carbonyl compounds are susceptible to dehalogenation and thus largely neglected as enolate precurꢀ
sors in catalytic enantioselective C–C bond forming reactions. By merging the increased stability of the αꢀCꢀhalogen bond of amꢀ
ides and the direct enolization methodology of the designed amide, we explored a direct catalytic asymmetric Mannichꢀtype reacꢀ
tion of αꢀhalo 7ꢀazaindoline amides with Nꢀcarbamoyl imines. All αꢀhalo substituents, αꢀF, ꢀCl, ꢀBr, ꢀI amides, were tolerated to
provide the Mannichꢀadducts in a highly stereoselective manner without undesirable dehalogenation. The diastereoselectivity
switched intriguingly depending on the substitution pattern of the aromatic imines, which is ascribed to stereochemical differentiaꢀ
tion based on the open transition state model. Functional group interconversion of the 7ꢀazaindoline amide moiety of the Mannichꢀ
adducts and further elaboration into a diamide without dehalogenation highlight the synthetic utility of the present protocol for acꢀ
cessing
enantioenriched
halogenated
chemical
entities.
Introduction
Chlorinated and brominated compounds comprise a large famꢀ
1
ily of biologically active natural products, leading to the purꢀ
suit of halogenꢀcontaining drug candidates as well as their
fluorinated analogs. The growing interest and demand for halꢀ
ogenated chemical entities motivated the development of new
protocols to incorporate halogen atoms into the carbon frameꢀ
works of interest in a highly stereoselective manner. A comꢀ
monly adopted method of accessing this class of compounds,
particularly those bearing halogen atoms at the stereogenic
carbon, is functionalization of double bonds triggered by haleꢀ
1
g
nium cations and their equivalents. For targetꢀoriented synꢀ
thesis of these halogenated compounds, merging the C–C
bond construction stage and the introduction of halogen atoms
offers a more direct option with better step economy. In light
of the wide applicability of enolate chemistry, enantioselective
coupling of αꢀhalogenated enolate and electrophiles is a viable
option toward achieving this aim. Although catalytic enolizaꢀ
2
tion of highly reactive 2ꢀhaloꢀ1,3ꢀdicarbonyl compounds and
3
3
ꢀhalooxindoles is wellꢀexplored for producing halogenꢀ
bearing enolate precursors, αꢀhalogenated monocarbonyl comꢀ
pounds are rarely exploited in reactions triggered by catalytic
enolization (Scheme 1a). The Cꢀhalogen bond at the αꢀposition
of carbonyl is labile and this has been exploited for catalytic
dehalogenative enolization. Catalytic deprotonative enolizaꢀ
tion of αꢀhalo thioesters with retention of the Cꢀhalogen bond
6
7
bissulfonylethene or fluorination with NFSI. Recently, Trost
et al. documented catalytic enolization of cyclic ketones bearꢀ
ing αꢀchloro and bromo substituents in this Mannichꢀtype reꢀ
8
4
action manifold with high stereoselectivity. The scarcity of
catalytic enantioselective C–C bond forming reactions involvꢀ
ing in situ catalytic enolization of αꢀhalo monocarbonyl comꢀ
pounds, however, prompted us to apply our amide enolization
protocol to produce enantioenriched products with halogens
on a stereogenic carbon. 7ꢀAzaindoline amides, which we
developed as latent enolates operative in a cooperative catalytꢀ
was demonstrated by Coltart et al. for diastereoselective aldol
5
reactions using a MgBr /amine binary catalytic system. Enanꢀ
2
tioselective entry was reported using αꢀchloroaldehydes via
enamine catalysis, which allowed for direct Michael reactions
with reactive 1,1ꢀ
9,10
ic system,
can be readily prepared with an αꢀchloro or αꢀ
Scheme 1. The Art of Enolization of αꢀHalocarbonyls
bromo substituent from inexpensive haloacetyl halides
(Scheme 1b). Amide carbonyl is the least reactive of the seꢀ
ries of carbonyl functionalities, and likely suppresses undeꢀ
sired dehalogenation. Herein we report the proficiency of αꢀ
halogenated 7ꢀazaindoline acetamides 1 for catalytic enolizaꢀ
ACS Paragon Plus Environment

Journal of the American Chemical Society
tion in enantioselective Mannichꢀtype reactions with Nꢀ
Page 2 of 9
1
1
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
carbamoyl imines; all αꢀhalo substituents, F, Cl, Br, and I
were tolerated. In particular, αꢀCl and αꢀBr amide 1ꢀCl and 1ꢀ
Br, which conform to naturally abundant chlorinated and broꢀ
minated compounds, were readily prepared from inexpensive
1
2
commercial sources 3 and 4 with 7ꢀazaindoline 2, which can
be recovered after synthetic elaboration of the Mannich prodꢀ
ucts.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Results and Discussion
Given the abundance of chlorinated natural products, we beꢀ
gan our study with a Mannichꢀtype reaction of αꢀ
a
Table 1. Evaluation of Chiral Ligands
a
5
: 0.2 mmol, 1ꢀCl: 0.1 mmol, 0.4 M on 1ꢀCl. Isolated yields
are reported. Enantioselectivity of the major diastereomers is preꢀ
b
sented. Run with 10 mol% of catalyst.
chloroꢀ7ꢀazaindoline amide 1ꢀCl with NꢀBocꢀimine 5a. Based
on our previous findings that Cu(I)/Barton’s base catalytic
system facilitated the enolization of 7ꢀazaindoline amides,
9
suitable reaction conditions quickly emerged following a brief
survey of chiral ligands (Table 1). While biarylꢀtype bisphosꢀ
phine ligands L1–L4 afforded encouraging enantioselectivity,
the diastereoselectivity was generally poor and variable conꢀ
versions were observed (entries 1–4). Chiral alkylphosphine
ligand (S,S)ꢀPhꢀBPE L5 afforded synꢀ6ꢀCl with excellent enꢀ
antioselectivity, albeit with a small preference for the syn isoꢀ
mer (entry 5). A possible primary reason for the low diastereꢀ
oselectivity is the less tractable prochiral face selection of
imine 5a; the absolute configuration at the αꢀposition was
mostly identical for both syn/anti diastereomers and the
prochiral faceꢀselection of in situꢀgenerated Cu(I)ꢀenolate of
a
b
5
: 0.2 mmol, 1ꢀCl: 0.1 mmol, 0.4 M on 1ꢀCl. Determined by
1
H
NMR analysis of the crude mixture with 3,4,5ꢀ
c
trichloropyridine as an internal standard. Determined by HPLC
analysis. Determined by HPLC analysis. Negative sign indicates
opposite enantiomer was obtained as a major product. Isolated
yield.
d
e
1
ꢀCl was generally wellꢀcontrolled. This tendency led us to
use structurally distinct chiral phosphine ligands L6–L8 with a
ferrocene framework. Although a poor outcome was realized
with (R,S )ꢀJosiphos L6 (entry 6), Walphosꢀtype ligands L7
p
Table 2. Substrate Scope of Monosubstituted NꢀBoc Imines
and L8 outperformed other chiral ligands, and in particular,
L8, having a dicyclohexylphosphine unit, delivered synꢀ6aꢀCl
in 87% isolated yield with high diastereoꢀ and enantioselecꢀ
tivity (entries 7,8). Unexpectedly, aromatic imine 5b with a
Clꢀsubstituent at the oꢀposition switched the diastereoselectiviꢀ
ty to predominantly afford antiꢀ6bꢀCl with high enantioselecꢀ
5
in Direct Catalytic Asymmetric Mannichꢀtype Reaction
a
of 1ꢀCl
13
tivity. The identical absolute configuration at the αꢀposition
ACS Paragon Plus Environment

Page 3 of 9
Journal of the American Chemical Society
of the major enantiomers implied that the prochiral face selecꢀ
tion was altered by the presence of an oꢀsubstituent (vide-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
infra). The catalyst comprising Cu(I)/L8/Barton’s base was
widely effective for the direct Mannichꢀtype reactions of 1ꢀCl
with both aromatic and aliphatic imines (Table 2–4). In conꢀ
trast to the highly labile nature of αꢀCl carbonyl compounds,
the thusꢀobtained αꢀClꢀβꢀNHBoc amides exhibited sufficient
stability, and no substitution/βꢀelimination/aziridination was
detected during the workꢀup and chromatographic separation.
Table 2 summarizes the scope of aromatic imines preferentialꢀ
ly affording the synꢀadduct when imines 5 without oꢀ
substituents were used. Generally high enantioselectivity was
observed with imines bearing halogen (5c,d,f–h), nitro (5i),
Me (5k), and vinyl (5l) substituents at the m or pꢀpositions,
while MeO substituted imines (5e,j) afforded somewhat lower
enantioselectivity. The present Mannich protocol was scalable
to at least a gram scale without any detrimental effects on
conversion or stereoselectivity (5c). As noted above, the presꢀ
ence of oꢀsubstituents altered stereoselection to give the antiꢀ
adduct as the major diastereomer, not only for oꢀCl (5b) but
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
also for oꢀF (5m), oꢀBr (5n), oꢀNO (5o), and 5p, having an
2
electronꢀdonating oꢀMeO substituent.
To gain further insight into the switched diastereoselectiviꢀ
ty, an inꢀdepth study of the substituent effect was performed as
summarized in Table 3. Using oꢀClꢀsubstituted imine 5b as a
reference, a series of dichloroꢀsubstituted imines 5q–s was
subjected to the optimized conditions of the Mannichꢀtype
reaction. This comparative study revealed that the position of
another Clꢀsubstituent with respect to the anchoring oꢀCl subꢀ
stituent affected the diastereoselectivity; p’ꢀCl substitution (5s)
1
4
led to a significant loss in antiꢀselectivity, whereas o’ꢀ or m’ꢀ
Cl substitution (5q,r) gave the products with a comparable
level of diastereoꢀ and enantioselectivity (o’, m’, and p’ denote
the relative position of the additional substituent with respect
to the anchoring oꢀsubstituent). This tendency was also obꢀ
served in the series of disubstituted imines 5tꢀv having oꢀClꢀ
o’ꢀOMe, oꢀClꢀm’ꢀOMe, and oꢀClꢀp’ꢀOMe substitution patterns.
The magnitude of the decrease in antiꢀselectivity was largely
proportional to the sterics of the p’ꢀsubstituents, from syn/anti
= 15/85 for p’ꢀF (5w) to syn/anti = 33/67 for p’ꢀBr (5y). The
oꢀF substituent (5m) exerted a smaller anchoring effect to
cause antiꢀdiastereoselectivity (syn/anti = 19/81), which was
weakened by p’ꢀF substituents (5z) in a similar manner as
observed for disubstituted imines bearing the oꢀCl substituent.
Intriguingly, larger p’ꢀCl (5aa) and p’ꢀBr (5ab) substituents
overrode the anchoring effect of the oꢀF substituent, preferenꢀ
tially affording the synꢀproduct.
a
5
: 0.2 mmol, 1ꢀCl: 0.1 mmol. 0.4 M on 1ꢀCl. Isolated yields
are reported. o’, m’, and p’ denote the relative position of the adꢀ
ditional substituent with respect to the anchoring orthoꢀ
substituents. Enantioselectivity of the major diastereomers is preꢀ
sented.
The deterioration of antiꢀselectivity by the presence of the p’ꢀ
substituent was similarly valid for imines 5ac and 5ad sharing
identical substitution patterns with oꢀBrꢀp’ꢀCl and oꢀNO ꢀp’ꢀ
2
Cl, respectively.
Together, these data led us to propose a plausible rationale
for the observed divergent diastereoselectivity (Figure 1). Simꢀ
9
ilar to other previously studied 7ꢀazaindoline amides, abnorꢀ
Table 3. Substrate Scope of Disubstituted NꢀBoc Imines 5
mally downfielded αꢀprotons (4.92 ppm) were observed for 1ꢀ
in Direct Catalytic Asymmetric Mannichꢀtype Reaction of
1
a
Cl in H NMR (THFꢀd
8
) due to hydrogen bonding with the
1
ꢀCl
inherent pyridyl nitrogen, which is indicative of the preferred
Eꢀconformation of 1ꢀCl in solution. Activation of 1ꢀCl through
bidentate coordination with the Cu(I) complex was evidenced
by the upfield shift of the αꢀprotons (4.71 ppm) as well as
NOE correlation with the protons at the 2ꢀposition of the indoꢀ
15
line. Subsequent deprotonation by Barton’s base generated
the corresponding enolate, which engaged in a Mannichꢀtype
addition to NꢀBoc imines 5. In the simplest case (i), given the
coordinatively saturated Cu(I)ꢀenolate and steric bulk of
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 9
neighboring ligand L8, mꢀ or pꢀsubstituted imines 5ꢀ(i) preꢀ
sumably reacted via open transition state model I, where the
smallest steric barrier and dipole interactions were anticipated.
In model I, the reaction proceeded through the Reꢀface of
imine 5ꢀ(i), affording the product 6ꢀCl in a synꢀselective fashꢀ
ion. On the other hand, imines 5ꢀ(ii) bearing an oꢀsubstituent
followed a different scenario (case (ii)). Due to electrostatic
repulsion with the neighboring imine nitrogen, 5ꢀ(ii) prefers
the sꢀtrans conformation over the sꢀcis counterpart, which
eventually rendered model III for Reꢀface attack considerably
less attainable because of the peripheral steric repulsion. Inꢀ
stead, model IV for the C–C bond formation via the Siꢀface of
the imine 5ꢀ(ii) was more likely, predominantly affording antiꢀ
oꢀCl imines 5s,v,w–y, oꢀBrꢀp’ꢀCl imine 5ac, and oꢀNO ꢀp’ꢀCl
imine 5ad, the Siꢀface model VI was generally favored to give
2
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
antiꢀ6ꢀCl with gradual erosion of the antiꢀselectivity proporꢀ
tional to the steric bias of the p’ꢀsubstituents. In contrast, for
imines 5m,z,aa,ab with the smaller oꢀF substituent, antiꢀ
selectivity was generally lower and the reaction through model
V for the Reꢀface attack was a major pathway to give a higher
fraction of synꢀ6ꢀCl when the p’ꢀsubstituent was p’ꢀCl or p’ꢀBr,
larger than a fluoroꢀsubstituent (imines 5aa,ab).
The present Cu(I)/Barton’s base catalytic system accomꢀ
modated the use of aliphatic imines 5ae–aj as applicable subꢀ
strates by changing the chiral bisphosphine ligand from L8 to
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(S,S)ꢀPhꢀBPE L5 (Table 4). All imines tested, including nonꢀ
6
ꢀCl. The unexpectedly lower antiꢀselectivity of imine 5p with
branched (5ae–ag), αꢀbranched (5ah,ai), and βꢀbranched (5aj)
imines, were converted to the corresponding Mannich prodꢀ
an oꢀOMe group is ascribed to the fact that the hydrogen
Figure 1. Schematic representation of the origin of diastereoselectivity based on the substitution pattern of aromatic imines 5.
i) synꢀSelective reaction of mꢀ or pꢀmonosubstituted imines 5ꢀ(i). (ii) antiꢀSelective reaction of o,o’ꢀ or o,m’ꢀdisubstituted
(
imines 5ꢀ(ii). (iii) o,p’ꢀdisubstituted imines 5ꢀ(iii) showed divergent diastereoselectivity depending on the steric factor of subꢀ
stituents. o’, m’, and p’ denote the relative position with respect to the oꢀsubstituent.
bonding interaction between the imine nitrogen and the oꢀ
OMe group mediated by a proton rendered the sꢀcis conforꢀ
mation more feasible, which prefers a Reꢀface attack, as
shown in model III’, to deliver a considerable amount of the
synꢀconfigured product. The Cu(I)ꢀenolate ligated with L8
effectively exposed one face accessible for the reaction with
imines, and the identical 2R configuration was favored for
both synꢀ and antiꢀ6ꢀCl. The reaction via the Siꢀface for antiꢀ
selectivity was compromised by the presence of the p’ꢀ
substituent (case (iii)), as delineated in model VI for the Siꢀ
face attack in which steric repulsion was predicted based on
the presence of the p’ꢀsubstituent. In the case of p’ꢀsubstituted
ucts in a synꢀselective fashion, likely through model I depicted
in Figure 1. To overcome the inherent lower reactivity, both
higher catalyst loading (10 mol%) and higher temperature
(0 °C)
Table 4. Substrate Scope of Aliphatic NꢀBoc Imines 5 in
Direct Catalytic Asymmetric Mannichꢀtype Reaction of 1ꢀ
a
Cl
ACS Paragon Plus Environment

Page 5 of 9
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
5
: 0.3 mmol, 1ꢀCl: 0.1 mmol. 0.33 M on 1ꢀCl. Isolated yields
are reported. Enantioselectivity of the major diastereomers is preꢀ
sented.
Table 5. Substrate Scope of NꢀBoc Imines 5 in Direct Cataꢀ
lytic Asymmetric Mannichꢀtype Reaction of 1ꢀBr
a
Figure 2. Reaction profile of amides 1ꢀCl, 1ꢀBr, and 1ꢀI in
catalytic asymmetric Mannichꢀtype reaction. 0.2 M on 1ꢀX.
2
.0 equiv of imine 5a was used.
were essential for completion. Enantioselectivity was generalꢀ
ly high, while synꢀselectivity was moderate, presumably due to
the increased sterics of the flanking alkyl chains.
Notably, αꢀBr and αꢀI 7ꢀazaindoline amides (1ꢀBr, 1ꢀI)
served as competent latent enolates in the catalytic asymmetric
Mannichꢀtype reactions. Although a somewhat slower reaction
rate was observed, faceꢀselectivity of the in situꢀgenerated
enolate of 1ꢀBr had the same propensity observed for 1ꢀCl;
products synꢀ6ꢀBr were predominantly obtained with mꢀ or pꢀ
substituted imines 5a,c,d,f,h,ak via a Reꢀface attack (model I,
Figure 1), while antiꢀ6ꢀBr was the major diastereomer when oꢀ
substituted imines 5b,n,r,t were used. Cu(I)/L5/Barton’s base
catalytic system accommodates aliphatic imine 5ag at a higher
reaction temperature (–20 °C). For comparison, the kinetic
profile of the Mannichꢀtype reaction was traced for 1ꢀCl, 1ꢀBr,
and 1ꢀI using imine 5a (Figure 2). Under identical conditions
with 10 mol% of catalyst loading at –60 °C and 0.2 M concenꢀ
tration, 1ꢀCl produced the highest reaction rate and 1ꢀBr folꢀ
lowed to reach completion within 24 h, while the reactivity of
1
ꢀI was insufficient. A higher concentration (0.3 M) and exꢀ
tended reaction time allowed 1ꢀI to provide the desired prodꢀ
uct 6ꢀI in a reasonable yield (Table 6). The empirical tendency
for diastereoselectivity shown in Figure 1 was also valid for 1ꢀ
I; imine 5a with no substituent and 5n with an oꢀBr substituent
exhibited synꢀ and antiꢀselectivity, respectively. αꢀF azaindoꢀ
line amide 1ꢀF has a stable C–F bond and its use in the direct
Mannichꢀtype reaction was previously addressed with NꢀCbzꢀ
a
5: 0.2 mmol, 1ꢀBr: 0.1 mmol, 0.33 M on 1ꢀBr. Isolated yields
are reported. Enantioselectivity of the major diastereomers is preꢀ
sented. Run using L5 instead of L8 at –20 °C for 72 h with 0.3
mmol of imine 5ag.
b
16,17
imines 7 (Scheme 2).
Compared with other αꢀhalo azaindoꢀ
line amides, 1ꢀF exhibited very slow reaction kinetics, and no
product was obtained at –60 °C. A higher reaction temperature
with Taniaphosꢀtype ligand L9 having a similar architecture
afforded the corresponding products 8 in an antiꢀselective
manner. It is noteworthy that the absolute configuration at the
αꢀposition was opposite that observed for 1ꢀCl, 1ꢀBr, and 1ꢀI,
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 9
which could be ascribed to the reaction through Eꢀenolate in
the specific case of 1ꢀF. The involvement of the Eꢀenolate is
Chart 1. Unsuccessful Latent Enolates
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
rationalized by minimal steric factor of the fluorine atom and
1
8
a
smaller dipole moment. Subsequent addition to imine 7 bearꢀ
ing m- or pꢀsubstituents likely proceeded through model VII
Scheme 3. Transformation of Mannichꢀadducts
19
to preferentially afford antiꢀconfigured product antiꢀ8ꢀF.
Table 6. Substrate Scope of NꢀBoc Imines 5 in Direct Cataꢀ
a
lytic Asymmetric Mannichꢀtype Reaction of 1ꢀI
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
5: 0.2 mmol, 1ꢀI: 0.1 mmol. 0.33 M on 1ꢀI. Isolated yields are
a
reported. Enantioselectivity of the major diastereomers is presentꢀ
ed.
Reagents and conditions: (a) 6N HCl aq., 60 °C, 24 h; (b)
CuCl, MeOH, 80 °C (bath temp.), 48 h; (c) BH •NH , LDA, THF,
3
3
0
°C to rt, 1 h. (d) TFA, CH Cl , rt, 20 min; (e) FmocꢀAlaꢀOH,
2 2
PyBOP, 2,6ꢀlutidine, CH Cl , rt, 18 h; (f) CuCl, MeOH, 80 °C
2
2
Chart 1 shows the exclusive nature of 7ꢀazaindoline amide
for the present direct enolization methodology. Derivatives of
the most reactive 1ꢀCl among 1ꢀX were synthesized and evalꢀ
uated in a Mannichꢀtype reaction with imine 5a under optiꢀ
mized conditions. All of the derivatives tested, including 7ꢀ
azaindole amide 9, indoline
(
bath temp.), 42 h; (g) CuCl, CH CN/MeOH, 60 °C, 60 h; (h)
3
Cs CO , CH CN, rt, 12 h.
2
3
3
amide 10, 7ꢀazaindoline sulfonamide 11, NꢀBoc 2ꢀ
pyridylamide 12, and Weinreb’s amide 13, as well as simple
amide and ester 14,15 failed to give the desired Mannich
products.
Scheme 2. Direct Catalytic Asymmetric Mannichꢀtype Reꢀ
a
action of 1ꢀF with NꢀCbz Imines 7
Transformation of the enantioenriched Mannich products
with retention of the labile C–halogen bond at the stereogenic
carbon highlights the synthetic utility of the present catalytic
protocol (Scheme 3). The 7ꢀazaindoline amide moiety of the
diastereomerically pure synꢀ6cꢀCl was hydrolyzed by 6N HCl
aq. at 60 °C to give βꢀamino acid 16 in 79% yield with 92%
recovery of 7ꢀazaindoline 2, albeit with marginal loss of the
stereochemical integrity (syn/anti = 94/6). Transesterification
of synꢀ6cꢀCl mediated by CuCl in refluxing MeOH delivered
20
the corresponding methyl ester 17. Myers’ protocol using
lithium amidotrihydroborate enabled direct reduction of the
21
amide to primary alcohol 18. Of particular note is that a free
amino group could be exposed without aziridination by treatꢀ
ment with TFA to remove the Boc group at room temperature,
and subsequent addition of FmocꢀAlaꢀOH using PyBOP furꢀ
nished amide 19. CuClꢀmediated transesterification of 19 gave
methyl ester 20, providing a general procedure to access pepꢀ
22
tides with an αꢀchloroꢀβꢀamino acid unit. As expected, Manꢀ
nich product synꢀ6cꢀBr bearing αꢀBr substituent was more
susceptible to undesired reactions and a tiny amount of epiꢀ
merized product was observed in the conversion to ester 21
2
0
under CuClꢀmediated transesterification conditions. The
labile C–Br bond was exploited to furnish cisꢀsubstituted
8
aziridine 22.
ACS Paragon Plus Environment

Page 7 of 9
Journal of the American Chemical Society
Int. Ed. 2009, 48, 7604; (e) Xuan, Y.; Nie, S.; Dong, L.; Zhang,
1
J.; Yan, M. Org. Lett. 2009, 11, 1583; (f) Hatano, M.; Horibe, T.;
Ishihara, K. Org. Lett. 2010, 12, 3502; (g) Hatano, M.; Moriyama,
K.; Maki, T.; Ishihara, K. Angew. Chem. Int. Ed. 2010, 49, 3823;
(h) Chauhan, P.; Chimni, S. S. Adv. Synth. Catal. 2011, 353,
Conclusion
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
In conclusion, we determined that αꢀhalogenated 7ꢀazaindoline
amides were amenable to catalytic enolization and subsequent
enantioselective addition to Nꢀcarbamoylimines. All αꢀhalo
substituents, F, Cl, Br, and I, were successfully accommodatꢀ
ed; considering that αꢀhalogenated monocarbonyl compounds
have been underexplored as latent enolates, the observed
broad applicability of αꢀhalogenated 7ꢀazaindoline amides is
noteworthy. The divergent diastereoselectivity depending on
the substitution pattern of the aromatic imines was rationalized
by the plausible open transition state models. Divergent funcꢀ
tional group interconversion of the 7ꢀazaindoline moiety of the
Mannich product as well as Nꢀacylation without dehalogenaꢀ
tion highlight the synthetic utility of the reaction to access
enantioenriched halogenated building blocks. Further develꢀ
opments toward an aldol reaction manifold and the application
to natural product synthesis are underway.
3
203; (i) Liu, W. B.; Reeves, C. M.; Stoltz, B. M. J. Am. Chem.
Soc. 2013, 135, 17298; (j) Hong, S.; Kim, M.; Jung, M.; Ha, M.
W.; Lee, M.; Park, Y.; Kim, M. H.; Kim, T. S.; Lee, J.; Park, H.
G. Org Biomol Chem 2014, 12, 1510; (k) Bae, H. Y.; Song, C. E.
ACS Catalysis 2015, 5, 3613.
(a) Noole, A.; Järving, I.; Werner, F.; Lopp, M.; Malkov, A.;
Kanger, T. Org. Lett. 2012, 14, 4922; (b) Dou, X.; Yao, W.;
Zhou, B.; Lu, Y. Chem. Commun. 2013, 49, 9224.
3
.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4. Sauer, S. J.; Garnsey, M. R.; Coltart, D. M. J. Am. Chem. Soc.
2010, 132, 13997.
5
.
(a) Yost, J. M.; Alfie, R. J.; Tarsis, E. M.; Chong, I.; Coltart, D.
M. Chem. Commun. 2011, 47, 571; (b) Alfie, R. J.; Truong, N.;
Yost, J. M.; Coltart, D. M. Tetrahedron Lett. 2017, 58, 185.
Quintard, A.; Alexakis, A. Chem. Commun. 2010, 46, 4085
6
.
7. Shibatomi, K.; Yamamoto, H. Angew. Chem. Int. Ed. 2008, 47,
796.
Trost, B. M.; Saget, T.; Hung, C.ꢀI. J. Angew. Chem. Int. Ed.
017, 56, 2440.
5
8
9
.
.
2
(a) Weidner, K.; Kumagai, N.; Shibasaki, M. Angew. Chem. Int.
Ed. 2014, 53, 6150; (b) Yin, L.; Brewitz, L.; Kumagai, N.;
Shibasaki, M. J. Am. Chem. Soc. 2014, 136, 17958; (c) Weidner,
K.; Sun, Z.; Kumagai, N.; Shibasaki, M. Angew. Chem. Int. Ed.
ASSOCIATED CONTENT
Supporting Information
Experimental details and spectroscopic data of new comꢀ
pounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
2
015, 54, 6236; (d) Arteaga, F. A.; Liu, Z.; Brewitz, L.; Chen, J.;
Sun, B.; Kumagai, N.; Shibasaki, M. Org. Lett. 2016, 18, 2391.
10. (a) Kanai, M.; Kato, N.; Ichikawa, E.; Shibasaki, M. Synlett 2005,
1491; (b) Yamamoto, H.; Futatsugi, K. Angew. Chem. Int. Ed.
2005, 44, 1924; (c) Paull, D. H.; Abraham, C. J.; Scerba, M. T.;
AldenꢀDanforth, E.; Lectka, T. Acc. Chem. Res. 2008, 41, 655;
Crystallographic data for antiꢀ6bꢀCl (CIF)
Crystallographic data for synꢀ6dꢀCl (CIF)
Crystallographic data for antiꢀ6mꢀCl (CIF)
Crystallographic data for antiꢀ6rꢀCl (CIF)
Crystallographic data for synꢀ6ahꢀCl (CIF)
Crystallographic data for synꢀ6dꢀBr (CIF)
Experimental procedures and spectral data (PDF)
(
4
d) Kumagai, N.; Shibasaki, M. Angew. Chem. Int. Ed. 2011, 50,
760; (e) Peters, R., Cooperative Catalysis. WileyꢀVCH:
Weinheim, 2015.
11. For reviews on Mannich reactions, see: (a) Marques, M. M.
Angew. Chem. Int. Ed. 2006, 45, 348; (b) Ting, A.; Schaus, S. E.
Eur. J. Org. Chem. 2007, 5797; (c) Verkade, J. M.; van Hemert,
L. J.; Quaedflieg, P. J.; Rutjes, F. P. Chem. Soc. Rev. 2008, 37,
AUTHOR INFORMATION
Corresponding Author
2
9; (d) Arrayas, R. G.; Carretero, J. C. Chem. Soc. Rev. 2009, 38,
1940; (e) Kobayashi, S.; Mori, Y.; Fossey, J. S.; Salter, M. M.
Chem. Rev. 2011, 111, 2626.
nkumagai@bikaken.or.jp
mshibasa@bikaken.or.jp
12. Bulk purchase of 3 or 4 even lowers the cost shown in Scheme 1.
13. Absolute configuration of antiꢀ6bꢀCl was determined by Xꢀray
crystallographic analysis. That of synꢀ6aꢀCl was deduced from
synꢀ6dꢀCl (determined by Xꢀray crystallographic analysis) by
analogy.
4. The use of the corresponding NꢀCbz imine under otherwise
identical conditions led to lower stereoselectivity (83% isolated
yield, dr 55/45, 64% ee/77% ee).
5. Cu(I)/L2 complex was used in NMR analysis to avoid overlapped
peaks.
6. Brewitz, L.; Arteaga, F. A.; Yin, L.; Alagiri, K.; Kumagai, N.;
Shibasaki, M. J. Am. Chem. Soc. 2015, 137, 15929.
7. For catalytic asymmetric Mannichꢀtype reactions, only highly
ACKNOWLEDGMENT
This work was financially supported by ACTꢀC (JPMJCR12YO)
from JST, and KAKENHI (25713002, JP16H01043 in Precisely
Designed Catalysts with Customized Scaffolding) from JSPS. NK
thanks The Naito Foundation for financial support. Dr. Tomoyuki
Kimura is gratefully acknowledged for the Xꢀray crystallographic
analysis. We thank Dr. Ryuichi Sawa, Ms. Yumiko Kubota, and
Dr. Kiyoko Iijima for the NOE analysis.
1
1
1
1
REFERENCES
acidic
αꢀfluoroꢀβꢀketoesters,
αꢀfluoroꢀβꢀketoimides,
αꢀ
fluoromalonates were used; see (a) Han, X.; Kwiatkowski, J.;
Xue, F.; Huang, K.ꢀW.; Lu, Y. Angew. Chem. Int. Ed. 2009, 48,
1. (a) Gribble, G. W. Acc. Chem. Res. 1998, 31, 141; (b) Gribble, G.
W. Chem. Soc. Rev. 1999, 28, 335; (c) Gribble, G. W.
Chemosphere 2003, 52, 289; (d) Gribble, G. W. Heterocycles
7
604. (b) Pan, Y.; Zhao, Y.; Ma, T.; Yang, Y.; Liu, H.; Jiang, Z.;
Tan, C.ꢀH. Chem. Eur. J. 2010, 16, 779. (c) Yoon, S. J.; Kang, Y.
K.; Kim, D. Y. Synlett 2011, 420. (d) Kang, Y. K.; Yoon, S. J.;
Kim, D. Y. Bull. Korean Chem. Soc. 2011, 32, 1195. (e) Kang, Y.
K.; Kim, D. Y. Tetrahedron Lett. 2011, 52, 2356. (f) Zhao, Y.;
Pan, Y.; Liu, H.; Yang, Y.; Jiang, Z.; Tan, C.ꢀH. Chem. Eur. J.
2
012, 84, 157; (e) Wang, L.; Zhou, X.; Fredimoses, M.; Liao, S.;
Liu, Y. RSC Adv. 2014, 4, 57350; (f) Gribble, G. W.
Environmental Chemistry 2015, 12, 396; (g) Chung, W. J.;
Vanderwal, C. D. Angew. Chem. Int. Ed. 2016, 55, 4396.
2. (a) Okino, T.; Hoashi, Y.; Furukawa, T.; Xu, X.; Takemoto, Y. J.
Am. Chem. Soc. 2005, 127, 119; (b) Nichols, P. J.; DeMattei, J.
A.; Barnett, B. R.; LeFur, N. A.; Chuang, T.ꢀH.; Piscopio, A. D.;
Koch, K. Org. Lett. 2006, 8, 1495; (c) Andres, J. M.; Manzano,
R.; Pedrosa, R. Chem. Eur. J. 2008, 14, 5116; (d) Han, X.;
Kwiatkowski, J.; Xue, F.; Huang, K. W.; Lu, Y. Angew. Chem.
2
011, 17, 3571. (g) Wang, H.ꢀY.; Zhang, K.; Zheng, C.ꢀW.; Chai,
Z.; Cao, D.ꢀD.; Zhang, J.ꢀX.; Zhao, G. Angew. Chem. Int. Ed.
015, 54, 1775. (h) Cosimi, E.; Engl, O. D.; Saadi, J.; Ebert, M.ꢀ
O.; Wennemers, H. Angew. Chem. Int. Ed. 2016, 55, 13127.
18. In the case of αꢀCF ꢀαꢀF 7ꢀazaindoline amide, the αꢀF substituent
located close to azaindoline upon complexation with a Cu(I)
2
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 9
complex, which was evidenced by HOESY analysis (see ref 16).
21. Myers, A. G.; Yang, B. H.; Chen, H.; McKinstry, L.; Kopecky, D.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
This observation suggested that αꢀF substituent was
accommodated in the Eꢀenolate geometry without severe steric
repulsion.
9. No epimerization of the Mannich product was observed during
the course of the reaction and the anti configuration was
kinetically determined.
20. (a) Brohmer, M. C.; Mundinger, S.; Brase, S.; Bannwarth, W.
Angew. Chem. Int. Ed. 2011, 50, 6175; (b) Jakob, U.; Mundinger,
S.; Bannwarth, W. Eur. J. Org. Chem. 2014, 6963.
J.; Gleason, J. L. J. Am. Chem. Soc. 1997, 119, 6496.
22. (a) Hanessian, S.; Cantin, L.ꢀD. Tetrahedron Lett. 2000, 41, 787;
(b) Crousse, B.; Narizuka, S.; BonnetꢀDelpon, D. l.; Bégué, J.ꢀP.
Synlett 2001, 679; (c) Colpaert, F.; Mangelinckx, S.; De
Brabandere, S.; De Kimpe, N. J. Org. Chem. 2011, 76, 2204; (d)
Shibatomi, K.; Narayama, A.; Soga, Y.; Muto, T.; Iwasa, S. Org.
Lett. 2011, 13, 2944. See also ref 2d,f, and 17.
1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Page 9 of 9
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
9
ACS Paragon Plus Environment