Rhodium-Catalyzed Domino Hydroformylation/Double-Cyclization Reaction of Arylacetylenecarboxamides: Diastereoselectivity Studies and Application in the Synthesis of 1-Azabicyclo[x.y.0]alkanes

Article abstract of DOI:10.1002/asia.201801193

A domino method for the rapid syntheses of 1-azabicyclo[x.y.0]alkane scaffolds, such as indolizidines, quinolizidines, decahydropyridoazepines, and their derivatives, has been developed. This strategy involved a rhodium-catalyzed hydroformylation of allyl

Full text of DOI:10.1002/asia.201801193

CHEMISTRY
AN ASIAN JOURNAL
www.chemasianj.org
Accepted Article
Title: Rh-Catalyzed Domino Hydroformylation Double Cyclizations
of Arylacetylenecarboxamides: Diastereoselectivity Study and
Application for the synthesis of 1-Azabicyclo[x.y.0]alkanes
Authors: Wen-Hua Chiou, Jui-Chi Tsai, Yi-Huei Lin, Guei-Tang Chen,
Yu-Kai Gao, Yu-Che Tseng, and Chien-Lun Kao
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Asian J. 10.1002/asia.201801193
Link to VoR: http://dx.doi.org/10.1002/asia.201801193
A Journal of
A sister journal of Angewandte Chemie
and Chemistry – A European Journal

10.1002/asia.201801193
Chemistry - An Asian Journal
FULL PAPER
Rh-Catalyzed Domino Hydroformylation Double Cyclizations of
Arylacetylenecarboxamides: Diastereoselectivity Study and
Application for the synthesis of 1-Azabicyclo[x.y.0]alkanes
Jui-Chi Tsai, Yi-Huei Lin, Guei-Tang Chen, Yu-Kai Gao, Yu-Che Tseng, Chien-Lun Kao and Wen-Hua
Chiou*
Abstract:
A
domino method for the rapid syntheses of 1-
reaction. Hopefully, we are able to obtain a clear understanding
of the methodology and apply it to syntheses of natural alkaloid
products.
azabicyclo[x.y.0]alkane scaffolds such as indolizidines, quinolizidines,
decahydropyridoazepines, and their derivatives, has been developed.
The strategy used for the construction of the alkaloid scaffolds
involves Rh-catalyzed hydroformylation of allyl-, 3-butenyl-, or
homoallyl amides, followed by two spontaneous ring closures under
mild conditions. By changing the substituent position at amide
substrates and altering the carbon chain length of the substrates, the
reaction scope and the diastereoselectivity have been fully explored.
This method can be applied to the syntheses of natural alkaloid
tashiromine and epilupinine. The obvious difference in reactivity of two
isomeric amide substrates, 3-butenamide 1j and homoallylamide 1i,
could be attributed to a better HOMO-LUMO overlapping in the TS
derived from butenamides during cyclization. The explanations have
been supported by the corresponding DFT calculations.
Indolizidine and quinolizidine alkaloid families have been widely
[5]
distributed among a large number of natural products.
Many
compounds in the family possess a wide range of important
pharmacological activities. For example, epilupinine,
a
quinolizidine alkaloid isolated from Lupinus leteus, exhibits
significant activity against leukaemia P-388 and L1210 leukaemia.
[6]
Additional pharmaceutical activities such as antiviral,
antiarrhythmic and platelet anti-aggregating properties have also
[7]
been reported.
The ring conformations of azabicyclic
compounds with different ring sizes also dictate the orientation of
the nitrogen lone pair in space, resulting important consequences
[8]
on the activities.
Therefore, physiological activities of
azabicyclic compounds have prompted many synthetic chemists
to develop various elegant approaches. [9]
In the preliminary communication, we demonstrate the domino
reaction strategy by using allylamides for readily preparation of
indolizidines. In this work, we describe our whole findings as a full
account, which includes the reaction scope, the limitation, the
application for syntheses of natural alkaloids, and the rationale for
the reactivity difference in isomeric homoallylamide and 3-
butenamide in the process.
Introduction
Recently, employment of a domino reaction design has proven to
be an efficient and popular strategy to synthesize complicate
[1]
structures.
Such a process does not only allow rapid
transformations from simple starting materials to a complex
molecule, but also reduces the production costs such as
consumption of solvents, reagents, and energy, which become
[2]
increasingly crucial issues in years.
The design of a domino
Results and Discussion
reaction to synthesize specific sophisticated targets with
considerable structural and stereochemical complexity appeals to
Our synthetic strategy is to take advantage of the domino reaction
sequence: hydroformylation of olefin 1 to the corresponding
aldehyde 2, intramolecular condensation with the amide group to
hemiaminal 3, dehydration to transient N-acyliminium 4,
cyclization by the adjacent acetylene group as a π carbon
nucleophile to the aromatic-stablized vinyl cation 5. The cation 5
is best described as an ethenylidenylaryl cation species.
Subsequent addition with an acetic acid to the final enolacetate 6.
Two classes of amides have been designed for investigation on
the domino reaction. One is the allyl/homoallyl amides, in which
the alkene part is located on the amine side and the nucleophile
moiety is on the carbonyl side. Only difference between
allylamides and homoallylamides is the length of the carbon chain.
The substrates are readily available from couplings of the
corresponding acids with either allylamines or homoallylamines.
The other is the 3-butenamides, in which the nucleophile part is
on the amine side and the alkene moiety is on the carboxyl side.
be
a significant intellectual challenge. Therefore, domino
reactions have received considerable attention from the synthetic
organic community. We have developed a strategy of aromatic-
mediated domino hydroformylation double cyclization, and
applied to synthesis of tricyclic alkaloid crispine A over three steps.
[3]
Therefore, based on the useful concept, we have reported an
alkyne-mediated domino hydroformylation/double cyclization for
preparation of indolizidine alkaloids.[4] The success encourages
us to study thoroughly the scope and the limitation of the domino
[a]
J.-C. Tsai, Y.-H. Lin, G.-T. Chen, Y.-K. Gao, Y.-C. Tseng, C.-L. Kao
Prof. W.-H. Chiou*
Department of Chemistry
Institution National Chung Hsing University
Kuokuang road, Taichung 402, Taiwan, R. O.C.
E-mail:wchiou@dragon.nchu.edu.tw
Supporting information for this article is given via a link at the end of
the document.
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

10.1002/asia.201801193
Chemistry - An Asian Journal
FULL PAPER
It is obvious that the ability of the nucleophile to trap the resulting
N-acyliminium is very critical in the double cyclization (Scheme 1).
Exposure of E-enolacetate 6b in a basic methanol solution
afforded only single trans-ketone 7b in 81% yield. It is noteworthy
that treatment of enolacetates in acidic conditions led to formation
of messy residues. A large coupling constant of 10.8 Hz was
observed between the bridgehead methine and the methine next
to the ketone group, i.e., H-8a and H-8, indicating a trans
configuration between two adjacent methines. The preliminary
DFT calculations on stability indicate trans-ketone 7b is more
stable than cis-ketone 7b by 4.1 kcal/mol in vacuo, which
suggests that the trans isomer is preferentially favored under
thermodynamic control in the basic methanolysis of indolizidine
enolacetate 6b.
AcO
m
R1
R1
[Rh]/CO/H2
H+/ AcOH
H
N
1
N
m
n
n
n = 0 or 1
m = 1 or 2
R1 = aryl
6
O
O
+Sol, -H+
[Rh]/CO/H2
δ+
R1
R1
R1
R1
+
Cδ
CHO
OH
Study on the Diastereoselectivity of Substituted Allylamides:
With substrates bearing the optimized carbon nucleophile, i.e.,
para-methoxyphenylacetylene, we turn our attention on how an
additional substituent effects the relative stereochemistry in the
product. Therefore, we designed a series of amides substrates
bearing with an alkyl group at available positions for the double
cyclization reactions. First of all, amide 1d, a gem-disubstituted
H
N
N
N
N
n
m
m
m
m
n
n
n
O
5
3
O
2
O
O
4
Scheme 1. The Domino hydroformylation Double Cyclization Reactions for the
allyl-/homoallylamides.
alkene bearing a methyl group at the 2 position (using the
[12]
indolizidine numbering system),
was subjected within the
Substrate Optimization: Our investigation commenced with
pressurized double cyclized conditions, in which a higher
pressure is required for hydroformylation of gem-disubstituted
olefins. Moreover, due to a significant environmental difference of
two terminus of the gem-alkene group, the use of
triphenylphosphite is able to produce an excellent selectivity for
hydroformylation on N-Allylamide 1a bearing a phenylacetylene
[10]
group.
Treatment of amide 1a under hydroformylation
conditions, [3] i.e., Rh-BIPHEPHOS [11] (0.5 mol%) and pTSA (10
mol%) at 60 ˚C under 2 atm of CO and 2 atm of H2 in acetic acid
afforded a mixture of cyclized product 6a and unidentified side
products. After purification, enolacetate 6a was obtained in 32%
isolated yield (Scheme 2). Although the yield is not satisfactory,
the preliminary results demonstrate the reaction is feasible for the
construction of the indolizidine skeleton. The poor yield may
attribute to that the second cyclization proceeds incompletely.
Hence, we anticipate that introduction of an electron donating
group on the phenyl moiety should enhance the nucleophilicity of
the acetylene moiety to afford the better yield. Based on the
suggestion, amide 1b bearing a para-methoxyphenyl group was
subjected to the conditions, which give only single enolacetate 6b
in 83% isolated yield with an E configuration. The configuration is
unambiguously determined by the ROESY spectra, which give a
strong correlation between the phenyl protons (H-2') and the
bridgehead proton (H-8a).
the linear aldehyde.
A slightly more Rh catalyst loading
guarantees the complete reaction within 24 h. Gratifyingly,
treatment of amide 1d in the conditions afforded only a single E-
anti product 6d in 73% yield. Here the term of "anti" is used to
describe the relative configuration between the C-2 position with
the bridgehead methine. Subsequent basic methanolysis of E-anti
enolacetate 6d produced a single anti-trans ketone 7d in 69%
yield, showing an excellent diastereoselectivity (Scheme 3).
PMP
AcO
PMP
O
H
PMP
H
H
N
(b)
(a)
Me
H
Me
N
N
69%
73%
O
1d
anti-trans-7d
O
E-anti-6d
O
AcO
PMP
H
AcO
PMP
PMP
(a')
OAc
PMP
H
H
R
R
O
H
AcO
R
H
N
+
+
N
H
N
N
(a)
(b)
H
N
n-Pr
n-Pr
n-Pr
Z,anti-6e, 22%
H
O
O
n-Pr
O
O
O
N
N
E,syn-6e, 14%
1e
E,anti-6e, 41%
O
O
(b)
O
PMP
O
H
PMP
(b)
(b)
(a'), (b)
7a, 79%
6a, 32%
6b, 83%
1a, R = Ph
H
7b, 81%
1b, R = 4-MeOC6H4
1c, R = BnOCH2
+
H
O
H
N
N
n-Pr
(a) Rh(acac)(CO)2 (0.5 mol%); BIPHEPHOS (1.0 mol%) CO (2 atm) H2
(2 atm) pTSA (10 mol%), AcOH, 60 OC. (b) K2CO3 (25 mol%) MeOH, rt.
n-Pr
O
anti-trans-7e, 50%
syn-trans-7e, 15%
(a) Rh(acac)(CO)2 (2 mol%); P(OPh)3 (4 mol%) CO (20 atm) H2 (20
atm) pTSA (10 mol%), AcOH, 60 OC. (a') Rh(acac)(CO)2 (1 mol%);
BIPHEPHOS (2 mol%) CO (2 atm) H2 (2 atm) pTSA (10 mol%), AcOH,
60 OC.(b) K2CO3 (25 mol%) MeOH, rt.
Scheme 2. Results of three nucleophiles in the double cyclization reaction and
the subsequent hydrolysis.
Scheme 3. Results of amide 1d and 1e in the reactions.
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

10.1002/asia.201801193
Chemistry - An Asian Journal
FULL PAPER
In the same manner, the reaction of 1e, bearing an n-propyl group
at the 3 position, affords Z-anti enolacetate 6e (22%), E-anti
enolacetate 6e (41%), E-syn enolacetate 6e (14%). While the
ROESY between the phenyl protons (H-2') and the bridgehead
proton (H-8a) is observed in the E enol-acetates, the observation
of a strong ROESY correlation between the phenyl protons and
yield and the minor product syn-cis-ketone 7g in 8% yield. A small
amount of syn-cis-ketone 7g has been obtained in spite of
appearance of anti-trans-ketone 7g and syn-trans-ketone 7g as
the major products. Therefore, substitution at the 7 position is
quite detrimental for the diastereoselectivity in the second
cyclization, and adverse to the formation of trans ketone in
the methylene group, i.e., H-7 proton, confirms the Z configuration. methanolysis reaction. Reaction of 1h bearing a hex-5-ynamide
Subsequent individual methanolysis for either Z-anti enolacetate
6e or E-anti enol acetate 6e led to the formation of the same
product, which was identified as anti-trans ketone 7e. In a similar
manner, methanolysis of the other E-syn enolacetate 6e afforded
as syn-trans ketone 7e. Direct treatment of the crude reaction
mixture of 1e in the basic solution has been carried out, yielding
anti-trans ketone 7e (50%) and syn-trans ketone 7e (15%). Thus,
the results are consistent with those in the individual hydrolysis.
Different from the reaction of 1e, reaction of amide 1f which bears
a methyl group at the 6 position, produced only single syn product
6f in 87% yield. The results indicate that the 6 position was an
excellent stereogenic center. Subsequent methanolysis of
enolacetate 6f at room temperature proceeded smoothly to yield
single ketone syn-trans-7f in 72% yield (Scheme 4).
moiety (m = 2) afforded enol tosylate 8h and an inseparable
mixture of E and Z enol acetate 6h (73%, ~ 1: 1, Scheme 4).
Different from the single product formation after methanolysis of
the enol acetate 6b, the methanolysis of the E/Z enol acetate
mixture 6h yielded two ketones, identified as trans ketone 7h and
cis ketone 7h. Direct treatment of the crude product in the basic
solution yielded trans ketone 7h (27%) and cis ketone 7h (30%),
accompanied with enol tosylate 8h (4%). The results also imply
that cis-ketones 7h, is no less stable than trans-ketones 7h in the
1-azabicyclo[5.3.0] system, which is different from that trans-
ketones is more stable in the 1-azabicyclo[4.3.0] system.
The observed selectivity for the substituted allylamides has been
summarized in the figure 2. Both C-2 and C-6 Substituted
allylamides reacts efficiently to provide single isomer as the
product. The reaction of gem-disubstituted substrate 1d required
more pressure to give the 2,8a-anti product 6d, while the reaction
of allylamide 1f proceeds at low pressure to yield the 6,8a-syn
product 6d. The results have shown that the C-3 position is not a
good stereogenic center because of the formation of three
different products in the reaction of the C-3 substituted allylamide
1e. Although the dominated products with E- and Z- 3,8a-anti-
enolacetate products are able to be hydrolysed to the identical
ketone derivative anti-trans ketone 7e, a small amount of 3,8a-
syn product has been observed. As regards the C-7 substituted
substrate 1g, the substitution at the C-7 position has deteriorated
the selectivity in the reaction. Both syn and anti enolacetate
products have been formed in equal amount as an inseparable
mixture. Hence, we conclude that the diastereoselectivity in the
double cyclization of allylamides is excellent in substitution on the
C-2 and C-6 position, mediocre in substitution on the C-3 position
and poor in substitution on the C-6 position.
PMP
AcO
PMP
O
H
PMP
H
(a)
(b)
H
N
H
O
N
N
87%
72%
Me
Me
Me
Me
Me
O
O
E,syn-6f
1f
syn,trans-7f
PMP
AcO
PMP
H
N
(a)
(b)
H
N
73%
6g
O
1g
O
PMP
H
O
H
PMP
O
PMP
O
H
H
H
H
Me
Me
Me
+
+
(a), (b)
N
N
N
62% trans-7g
8% cis-7g
O
O
O
anti-trans-7g syn-trans-7g
syn-cis-7g, 8%
66% inseparable,~ 6:5
PMP
OMe
PMP
O
TsO
PMP
poor
both anti and syn
PMP
O
AcO
H
H
H
N
(a), (b)
H
H
+
+
7
6
H
N
N
N
8a
N
2
anti favored
completely
3
O
O
O
O
syn favored
completely
O
1h
8h, 4% trans-7h, 27% cis-7h, 30%
(a) Rh(acac)(CO)2 (0.5 mol%); BIPHEPHOS (1.0 mol%) CO (2 atm) H2
(2 atm) pTSA (10 mol%), AcOH, 60 OC. (b) K2CO3 (25 mol%) MeOH, rt.
anti favored
Figure 1. Summaries of the observed diastereoselectivity of various substituted
allylamide substrates 1d, 1e, 1f, and 1g.
Scheme 4. Results of amide 1f, 1g and 1h in the reactions.
However, treatment of amide 1g, bearing a methyl group at the 7
position, under the standard conditions gave an inseparable
mixture of both anti/syn E-enolacetate 6g (~ 1: 1) in 73% yield.
The following methanolysis of the mixture resulted in the
formation of three different ketones: an inseparable mixture of
anti-trans-ketone 7g and syn-trans-ketone 7g in 66% isolated
The following hydrolysis of enolacetates is also highly
stereoselective in the indolizidine scaffold. In the hydrolysis of the
C-2, C-6 and C-3 substituted enolacetates, all the resulting
ketones possess the 8-8a-trans configuration, even in the cases
of two diastereomeric enolacetate 3,8a-anti and 3,8a-syn
enolacetate 6e. Only small amount of cis ketone has been found
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

10.1002/asia.201801193
Chemistry - An Asian Journal
FULL PAPER
in the hydrolysis of enolacetate 6f. The results suggest that
formation of the trans-ketone is thermodynamically driven.
However, in the 1-azabicyclo[5.3.0] system, the stability
difference between the trans and cis isomer is not so significant
that one of the product dominates in the reaction.
[a] (a) Rh(acac)(CO)2 (0.5 mol%), BIPHEPHOS (1.0 mol%), pTSA (10 mol%),
AcOH, 60 oC. [b] Isolated yield. PMP : 4-paramethoxyphenyl. [c] E/Z ratio was
determined by 1H-NMR.
Encouraged by the successful application to indolizidines using
the allylamide substrates, we prepared homoallylamide 1i, i.e., n
= 1, and wished to achieve the construction of a quinolizidine
using the one-pot domino reaction strategy. However, attempts to
synthesize quinolizidines directly from the reaction of the
homoallylamide 1i were unsuccessful despite intense efforts to
optimize the conditions, and only resulted in a complex residue or
nonspecific decomposition rather than the desired product
(Scheme 5).
Treatment of 3-butenamide 1j in the cyclized conditions, i.e.,
Rh(acac)(CO)2 (0.5 mol%), BIPHEPHOS (1 mol%) catalyst in
acetic acid at 60 ˚C, successfully afforded the desired single
cyclized E-enolacetate 6j in 61% yield (entry 1). Changing the gas
ratio of CO and H2 to 3:1 (4 atm in total) showed slightly increased
the yield to 73% (entry 2), accompanied by the 6j’ in 7% yield,
which resulted from the formation of a branched aldehyde in
hydroformylation. Reaction of 3-butenamide 1k, bearing a 6-
arylhex-5-ynyl group (m = 2), yielded a 2:3 mixture of E/Z
enolacetate 6k in 56% (entry 3), [12] while partial pressure slightly
decreased to 50% yield (entry 4). Reaction of 3-butenamide 1l,
bearing a 4-arylbut-3-ynyl group (m = 0), afforded only E form of
indolizidine-typed enolacetate 6l in 42% yield (entry 5). Changing
the partial pressure did not result in a significant yield change
(entry 6). The results indicated that the length of the carbon chain
in the nucleophile end plays a critical role in the second cyclization.
As a result of our investigation, it is concluded that 3-butenamide
1j is the most suitable for the preparation of the quinolizidine
structures.
Basic methanolysis of enolacetate 6j in the presence of K2CO3
afforded trans- ketone 7j in 83% yield, while that of the E/Z
enolacetate mixture 6k yielded only single product in 97%,
identified as trans-ketone 7k. [14] However, methanolysis of enol-
acetate 6l proceeded very slowly so that an additional amount of
base was required (to 50 mol%), and it took 5 days to give trans-
ketone 7l in 23% (Scheme 6). The structure of trans-ketone 7l
was confirmed by a single crystal X-ray analysis (CCDC no.
987311). Although kinetic controls should not be underestimated
because the axial attack of the proton in methanolysis to lead the
trans ketone, the formation of only trans ketone in the basic
methanolysis of the three enolacetates 6j~l suggest
thermodynamic control favoring the formation of more stable
products.
AcO
PMP
PMP
[Rh]/CO/H
+
2
H
N
1
H / AcOH
N
n
n = 0 ; 6b, 83%
n = 1 ; messy
n
6
O
O
Scheme 5. Reaction of allylamide 1b and homoallylamide 1i.
Study on 3-butenamides: We therefore considered the reactivity
should be related in the geometry of the corresponding TS (vide
infra), in which poor arrangement may cause the instable TS and
prevent the cyclization. Mann et al. have reported the practical
demonstration of the use 3-butenamide in the hydroformylative
[13]
domino reaction.
Therefore we turned our attention to the
corresponding 3-butenamide 1j, which was just simple swapping
the carbonyl group for the methylene group next to the amide
nitrogen in the homoallylamide. 3-butenamides were readily
prepared from the corresponding amines with 3-butenoic acid or
2-methyl-3-butenoic acid. With 3-butenamides in hand, the
substrate scope of the reaction was then explored as listed in
Table 1.
Table 1. The resutls of double cyclization of 3-butenamides 1i~l..
AcO
PMP
N
AcO
PMP
PMP
O
Me
AcO
PMP
PMP
O
H
H
(a)
H
N
K2CO3 (10 mol%)
MeOH, rt.
N
H
N
O
N
m
m
O
6j'
m
1
6
O
m
O
7j, 83%
6j, m = 1.
7k, 97%
6k, m = 2; E/Z~ 2:3
6l, m = 0.
Entry[a]
Substrate
1j, m = 1
1j, m = 1
1k, m = 2
1k, m = 2
1l, m = 0
1l, m = 0
PCO (atm)
PH2 (atm)
6 (%)[b]
61
7l, 23%, (CCDC987311)
1
2
3
4
5
6
2
3
2
3
2
3
2
1
2
1
2
1
Scheme 6. Results of basic hydrolysis of enolacetates 6j~l.
73
56
As described above, 3-butenamide, bearing a 5-arylpent-4-ynyl
group (m = 1), is the most suitable among the three various
substrates in the domino reaction to form quinolizidines with E-
enolacetate moiety. The method can be also applied to the
formation of decahydropyridoazepines (m = 2), but within a pair
of E/Z diastereomers. It may attribute to that the difference in the
50C
42
42
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

10.1002/asia.201801193
Chemistry - An Asian Journal
FULL PAPER
two facial environments in the large ring size is not so obvious that
two faces are available for the solvent molecule to produce both
possible enolacetates. However, it will not be a critical issue
because methanolysis of both E- and Z-enolacetates will bring the
formation of the identical trans-ketone products. The results
provide a useful guideline that allylamides are better substrates to
construct indolizidines while 3-butenamides are suitable to
synthesize quinolizidines.
(E/Z, anti)-enol-acetate mixture 6n yielded two ketones, identified
as the major product trans-anti-ketone 7n in 21% and the minor
product cis-anti-ketone 7n in 14%. Treatment of (E, syn)-
enolacetate 6n with the basic methanol gave a single ketone
product in 65%, identified as trans-syn-ketone 7n by single crystal
X-ray analysis (CCDC no. 990279). The same ketone was also
obtained in 60% yield by treatment of (Z, syn)-enolacetate 6n in
the basic conditions.
Study on the Diastereoselectivity of 2-Substituted 3-
Butenamides:Three
different
2-methyl
substituted
3-
butenamides 1m~o have been synthesized as the substrates to
investigate the diastereoselectivity in the reaction. Reaction of 1m
in the cyclization conditions afforded (E, anti)-enolacetate 6m
(12%), (E, syn)-enolacetate 6m (64%), and (Z, syn)-enolacetate
6m (8%) (Scheme 7). [15]
AcO
PMP
H
(b)
N
65%
Me
O
PMP
O
E,syn-6n,10%
+
H
PMP
PMP
OAc
H
AcO
PMP
H
AcO
PMP
H
PMP
OAc
H
H
(b)
(a)
N
H
N
Me
trans,syn-7n (CCDC990279)
PMP PMP
60%
(a)
+
+
Me
N
O
Me
N
O
N
O
O
1n
N
O
Me
Me
Me
O
Z,syn-6n, 22%
PMP
+
PMP
E,anti-6m, 12%
E,syn-6m, 64%
Z,syn-6m, 8%
(b) 78%
~ 5 :1
AcO
O
H
O
H
H
H
N
(b) 60%
(b) 91%
(b)
+
H
H
Me
N
N
N
O
1m
Me
Me
PMP
(a),(b)
O
H
PMP
O
H
PMP
O
H
Me
O
O
O
trans,anti-7n,21%
cis,anti-7n, 14%
E/Z, anti-6n, 12%, ~2:3
+
+
H
H
H
N
N
N
(a) Rh(acac)(CO)2 (0.5 mol%), BIPHEPHOS (1.0 mol%), CO (2 atm), H2
(2 atm), pTSA (10 mol%), AcOH, 60 oC (b) K2CO3 (25 mol%),MeOH, rt.
Me
Me
Me
O
O
O
cis,syn-7m, 6%
trans,syn-7m, 59%
trans,anti-7m, 10%
Scheme 8. Results of 2-methyl-3-butenamide 1n in the reactions.
(a) Rh(acac)(CO)2 (0.5 mol%), BIPHEPHOS (1.0 mol%), CO (2 atm), H2 (2 atm), pTSA (10
mol%), AcOH, 60 oC (b) K2CO3 (10 mol%),MeOH, rt.
Scheme 7. Results of 2-methyl-3-butenamide 1m in the reactions.
In the same manner, the reaction of 1o yielded (E, syn)-enol-
acetate 6o as a single product in 35% isolated yield, accompanied
by a complex mixture of unidentified side products (Scheme 9).
Direct treatment of crude product 6m with K2CO3 in methanol
yielded trans-anti-ketone 7m (10%), trans-syn-ketone 7m (59%)
and cis-syn-ketone 7m (6%). Individual methanolysis of (E, anti)-
enolacetate 6m led to the formation of trans-anti-ketone 7m as a
single product in 60% yield, and reaction of (E, syn)-enol-acetate
6m gave only trans-syn-ketone 7m in 91% yield. However,
reactions of (Z, syn)-enolacetate 6m gave a 5:1 mixture of trans-
syn-ketone 7m and cis-syn-ketone 7m. It should be noted that the
product distribution in the methanolysis of crude enolacetates is
consistent with those of individual enolacetates, which suggest
the product stability may dominate the distribution in the
methanolysis.
Reaction of 3-butenamide, 1n, the homolouge of 1m (m = 2), with
one carbon more in the nucleophile side chain afforded all four
possible diastereomers, (E, syn)-enolacetate 6n (10%), (Z, syn)-
enolacetate 6n (22%) and an inseparable 2:3 mixture of (E/Z,
anti)-enolacetate 6n (12%) (Scheme 8). Treatment of (E, syn)-
enolacetate 6n with the basic methanol gave a single ketone
product in 65%, identified as trans-syn-ketone 7n by single crystal
X-ray analysis (CCDC no. 990279). The same ketone was also
obtained in 60% yield by treatment of (Z, syn)-enolacetate 6n in
the basic conditions. Different from single product formation
derived from (E)- and (Z)- syn-enolacetates, the methanolysis of
PMP
H
O
PMP
PMP
AcO
(a)
H
N
H
(b)
H
N
N
Me
Me
Me
O
O
1o
O
trans, syn-7o, 29%
(CCDC987313)
O
E, syn-6o, 35%
+
PMP
H
H
N
Me
O
cis, syn-7o, 16%
trans,syn-7o, (CCDC987313)
(a) Rh(acac)(CO)2 (0.5 mol%), BIPHEPHOS (1.0 mol%), CO (3 atm), H2 (1
atm), pTSA (10 mol%), AcOH, 60 oC (b) K2CO3 (25 mol%),MeOH, rt.
Scheme 9. Results of 2-methyl-3-butenamide 1o in the reactions.
Treatment of (E, syn)-enolacetate 6o within the basic conditions
proceeded very slowly as the reaction of 1l did. It took one week
to yielded cis-syn-ketone 7o in 16% yield and trans-syn-ketone
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

10.1002/asia.201801193
Chemistry - An Asian Journal
FULL PAPER
7o in 29% yield, which structure has been confirmed by X-ray
crystallography (CCDC no. 987313).
react to yield vinyl cation 5. The results has been modified in the
acetic acid conditions. The results provided the corresponding
pathway profiles, and located all optimized geometries in the two
reactions (Figure 2). The cyclization of 4j is favored with a
relatively lower activation energy barrier of 5.6 Kcal/mol; while the
cyclization of 4i requires a higher activation energy barrier of 7.1
Kcal/mol. An energy difference of 1.8 Kcal/mol between two TSs
implied the transition state stability is concerned with the TS
geometry. The calculated geometries clearly show both TS
geometries possess a boat-like conformation, which results in a
remarkable flagpole interaction between H-7axial and H-9a. The
Based on these results, we have reached conclusions that the C-
3 position is not a good stereocontrol center in 3-butenamide
substrates, the substitution at the C-3 position will result in the
formation of several diastereomers. Compared to the reaction
results of 1j, amide 1m bearing the C-3 methyl group, affords syn-
enolacetates (i.e., H-3 and the bridged head H-9a) as the major
product, and anti-enolacetates as the minor product. Individual
methanolysis results of quinolizidine enolacetates, (E, anti)-6m,
(E, syn)-6m, and (Z, syn)-6m, shows the relative stereochemistry
between H-3 and H-9a has been intact during the basic hydrolysis, distance between H-7axial and H-9a is 2.065 Å in TS4i and 2.333
and the reactions favor the formation of trans-ketones (i.e., H-9
and H-9a) as the reaction of 6j does. The appearance of cis-syn-
ketone 7m as the minor product indicates the syn configuration
between H-3 and H-9a hampers the selectivity for the formation
of trans-ketone. In the independent methanolysis experiments of
enolacetates 6n, the relative stereochemistry between H-3 and H-
10a has been retained in basic hydrolysis as the reaction of
enolacetates 6m did. The presented configuration exerted a
significant effect on the subsequent configuration. Here it is
worthy to point out that syn-enolacetates 6n is “matched” because
of the single trans-ketone formation; while anti- enolacetates 6n
is “mismatched” because of the formation of both trans- and cis-
ketones.
Å in TS4j, respectively. A closer distance suggests a stronger
flagpole conflict, which distorts the HOMO-LUMO intersection
angle, and results in the instability of TS4i.
Applications to tashiroamine and epilupinine: To demonstrate
the ability of the methodology, we devised concise syntheses of
tashiromine and epilupinine. Tashiromine (10b) is a naturally
occurring indolizidine which isolated from an Asian deciduous
[16]
shrub Maackia tashiroi (Leguminosae).
Thus, ketone 7b and
7j underwent the Baeyer-Villiger oxidation using the Uneyamma’s
[17]
protocol,
which carried out the reaction in a mixed solvent of
1,1,1,3,3,3-hexafluoro-2-propanol (HFIP), CH2Cl2 and aqueous
phosphate buffer solution with mCPBA, affording ester 9b and 9j
in 66% and 45% yield, respectively. The X-ray crystal structure
(CCDC no. 992205) of ester 9j unequivocally established the
trans relationship (See supporting information). Simple global
reduction of both the ester and the lactam groups with LiAlH4
afforded tashiroamine (10b) in 73% yield and epilupinine (10j) in
54% yield, whose NMR data were in accordance with literature
value[18] (Scheme 10).
Figure 2. Potential energy diagram of both cyclizations of 4i and 4j.
In addition, the intersection angle between the HOMO plane and
the LUMO plane is 31.5o for TS4i and 26.7 o for TS4j. [19] A smaller
intersection angle guaranteed a better HOMO-LUMO overlapping
in TS4j derived from 3-butenamide, supported by relative stability
of TS4j. Thus, TS4j derived from 3-butenamides is able to undergo
O
O
H
PMP
O
H
HO
PMP
LiAlH4
mCPBA,
H
N
THF, reflux,
HFIP/CH2Cl2
phosphate buffer
H
X
H
X
H
N
N
n
Y
n
n
Y
cyclization to yield
a quinolizidine but TS4i derived from
9b X = CO, Y = CH2, n = 0, 66%
9j X = CH2, Y = CO, n = 1, 45%
10b, n = 0, 73%
10j , n = 1, 54%
7b X = CO, Y = CH2, n = 0
7j X = CH2, Y = CO, n = 1
homoallylamides can not. Thus, acyliminium 4j which derived
from 3-butenamide, is able to undergo the second cyclization to
yield the quinolizidine, but 4i from homoallylamide can not.
Scheme 10. Synthesis of tashiroamine and epilupinine.
Conclusions
Theoretically calculation for the rationale of reactivity
difference: Intrigued by the fact that 3-butenamide 1j could
undergo the domino double cyclization but homoallylamide 1i
could not, we carried out DFT calculations on the critical second
cyclization of N-acyliminium 4, in which the aryl alkyne moiety
We have investigated the alkyne mediated Rh-catalyzed
hydroformylation double cyclization reaction of 3-butenamides,
which provides cascade syntheses of indolizidines, quinolizidines
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

10.1002/asia.201801193
Chemistry - An Asian Journal
FULL PAPER
and
1-azabicyclo[5.3.0]decanes
and
1-azabicyclo[5.4.0]
References:
undecanes. The full investigations on the substituted amide
substrates revealed the corresponding diastereoselecitivity and
the limitation of the double cyclization. It provided useful
information for subsequent syntheses. Indolizidine skeleton can
be easily achieved by using the reactions of the corresponding
allylamides, in which the substitution at both C2 position and C-6
position both provides single product in good yield. Quinolizidine
skeleton can be synthesized in good yields from the
corresponding 3-butenamide derivatives. The methodology has
been successfully applied to 4-step syntheses of tashiromine and
epilupinine without protecting groups. In addition, based on the
DFT calculation results, the appearance of trans-ketone in the
methanolysis of enol acetates should attribute to its
thermodynamic stability. In addition, the theoretical calculations
were also helpful in providing a mechanistic rationale for the
successful formation of quinolizidine scaffold from 3-butenamide
1j due to better HOMO-LUMO overlapping, which changes for the
worse in homoallylamide 1i. Applications of this methodology
towards other complex natural products of interest are currently
underway.
[1]
For selected excellent reviews on domino reactions, see: a) K. C.
Nicolaou, D. J. Edmonds, P. G. Bulger, Angew. Chem. Int. Ed. 2006, 45,
7134-7186; b) L. F. Tietze, G. Brasche, K. M. Gericke, Domino Reactions
in Organic Synthesis, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim,
2006; c) A. Padwa, S. K. Bur, Tetrahedron 2007, 63, 5341-5378; d) A. K.
Lawrence, K. Gademann, Synthesis 2008, 331-351; e) B. B. Toure, D. G.
Hall, Chem. Rev. 2009, 109, 4439-4486.
[2]
[3]
[4]
[5]
a) L. F. Tietze, Chem. Rev. 1996, 96, 115-136; b) L. F. Tietze, N.
Rackelmann, Pure Appl. Chem. 2004, 76, 1967-1983.
W.-H. Chiou, G.-H. Lin, C.-C. Hsu, S. J. Chaterpaul, I. Ojima, Org. Lett.
2009, 11, 2659-2662.
W.-H. Chiou, Y.-H. Lin, G.-T. Chen, Y.-K. Gao, Y.-C. Tseng, C.-L. Kao,
J.-C. Tsai, Chem. Commun. 2011, 47, 3562-3564.
a) M. Hesse, Alkaloids: nature's curse or blessing?, Wiley-VCH, New
York, 2002; b) Modern Alkaloids: Structure, Isolation, Synthesis and
Biology, Wiley-VCH, Weinheim, 2008; c) J.P. Michael, Nat. Prod. Rep.
2008, 25, 139-165.
[6]
[7]
D. H. Hua, S. W. Miao, A. A. Bravo, D. J. Takemoto, Synthesis 1991,
970-974.
For the reports of bioactivities of inlizidines and quinolizidines, see a) A.
d. Meijere, Chem. Rev. 2003, 103, 931-932; b) M. Ercoli, L. Mina, C. C.
Boido, V. Boido, F. Sparatore, U. Armani, A. Piana, II Farmaco 2004, 59,
101-109; c) A. Sparatore, N. Basilico, S. Parapini, S. Romeo, F. Novelli,
F. Sparatorec, D. Taramelli, Bioorg. & Med. Chem. 2005, 13, 5338-5345;
d) I. Vazzana, R. Budriesi, E. Terranova, Pierfranco Ioan, M. P. Ugenti,
B. Tasso, A. Chiarini, F. Sparatore, J. Med. Chem. 2007, 50, 334-343; e)
M. Casagrande, N. Basilico, S. Parapini, S. Romeo, D. Taramelli, A.
Sparatore, Bioorg. & Med. Chem. 2008, 16, 6813-6823; f) M. Tonelli, G.
Paglietti, V. Boido, F. Sparatore, F. Marongiu, E. Marongiu, P. L. Colla,
R. Loddo, Chem. Biodiv. 2008, 5, 2386-2401; g) A. Sparatore, F. Novelli,
F. Sparatore, Helv. Chim. Acta 2004, 87, 580-591; h) M. Tonelli, I.
Vazzana, B. Tasso, V. Boido, F. Sparatore, M. Fermeglia, M. S. Paneni,
P. Posocco, S. Pricl, P. L. Colla, C. Ibba, B. Secci, G. Collu, R. Loddo,
Bioorg. & Med. Chem. 2009, 17, 4425-4440.
Experimental Section
General procedure for Domino Hydroformylation/Double Cyclization:
Rh(acac)(CO)2 (1.3 mg, 5.0 μmol, 0.5 mol%) and BIPHEPHOS (7.9 mg,
10 μmol, 1.0 mol%) were dissolved in acetic acid (1 mL) under argon. The
resulting catalyst solution was degassed by a frozen-thawed procedure at
least three times. Amide 1 (257 mg, 1.0 mmol, 1.0 equiv.) and pTSA (17
mg, 0.1 mmol, 10 mol%) were placed in a 50 mL flask. The catalyst solution
was transferred to the reaction flask containing the substrate by a pipette,
and the total volume was adjusted to 20 mL with acetic acid. The reaction
flask was placed in a 300 mL stainless steel autoclave and then was
pressurized with CO (2 atm) followed by H2 (2 atm). The reaction mixture
was stirred at 60 oC for 16−20 h. Upon completion of the reaction, the gas
was carefully released in a good ventilated hood and the reaction mixture
was concentrated under reduced pressure to give a crude residue. The
residue was partitioned with CH2Cl2 (20 mL) and NaHCO3(aq) (saturated,
10 mL). After separation of the organic layer, the aqueous layer was
extracted with CH2Cl2 (15 mL X 5). The combined organic layers were
dried over anhydrous Na2SO4, filtered, and then concentrated under
reduced pressure to give the crude product. The crude product was
purified by flash chromatography on silica gel using MeOH/CH2Cl2 or
EtOAc/n-Hex as the eluant to give the product.
[8]
[9]
S. Hanessian, A. K. Chattopadhyay, Org. Lett. 2014, 16, 232-235.
For selected syntheses of indolizidines, see: a) N. A. Khatri, H. F.
Schmitthenner, J. Shringarpure, S. M. Weinreb, J. Am. Chem. Soc. 1981,
103, 6387-6389; b) Y.-S. Cheng, J. Andrew T. Lupo, F. W. Fowler, J. Am.
Chem. Soc. 1983, 105, 7696-7703; c) M. Ihara, T. Kirihara, A. Kawaguch,
M. Tsuruta, K. Fukumoto, T. Kametani, J. Chem. Soc. Perkin Trans. 1
1987, 1719-1726; d) L. Song, K. Liu, C. Li, Org. Lett. 2011, 13, 3434-
3437; e) F. K. I. Chio, S. J. J. Guesné, L. Hassall, T. McGuire, A. P.
Dobbs, J. Org. Chem. 2015, 80, 9868-9880.
[10] For selected execellent books on hydroformylation, see:a) P. W. N. M. v.
Leeuwan, C. Claver, Rhodium Catalyzed Hydroformylation, Kluwer
Academic Publishers, Dordrecht, 2000; b) M. Taddei, A. Mann,
Hydroformylation for Organic Synthesis, Vol. 342, Springer, 2013. c) A.
Börner, R. Franke, Hydroformylation - Fundamentals, Processes, and
Applications in Organic Synthesis, Wiley-VCH, 2016.
Acknowledgements
[11] a) E. Billig, A. G. Abatjoglou, D. Bryant, Union Carbide, U.S., 1988; b) G.
D. Cuny, S. L. Buchwald, J. Am. Chem. Soc. 1993, 115, 2066-2068.
[12] The HSQC experiments show the peaks of δ 4.00 and δ 4.27 are
assigned as methines, i.e., H-10a. The nOe irradiation at the peak of δ
4.00 (H-10a) results in an enhancement δ 7.27, while that at the peak of
δ 4.27 does not cause any enhancement in the aromatic region. Thus,
the peak of δ 4.27 is clearly assigned as Z-enolacetate 7c and that of δ
4.00 as E-enolacetate 7c.
The research was supported by the National Science Council,
(NSC101-2113-M-005-009-MY3 and MOST104-2113-M-005-
004). The authors gratefully thank the National Center for High-
Performance Computing and the University for computation
facilities
[13] E. Airiau, T. Spangenberg, N. Girard, A. Schoenfelder, J. Salvadori, M.
Keywords: Domino reaction • Rh-catalyzed Hydroformylation •
Taddei, A. Mann, Chem. Eur. J. 2008, 14, 10938-10948.
Indolizidine • Quinolizidine • 1-azabicyclic
[14] A large 1H-1H coupling constant of 9.6 Hz was observed in both peaks of
δ 3.66 (H-10) and δ 4.14 (H-10a), indicating a trans arrangement
between H-10 and H-10a in ketone 7c.
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

10.1002/asia.201801193
Chemistry - An Asian Journal
FULL PAPER
[15] To describe the relative stereochemistry simply, the term of “trans/cis”
was used to describe the configuration of C-9 (or 10 or 8) with C-9 (or 10
or 8)a, i.e., the methine next the ketone, as the 1st notation; and the term
of “anti/syn” to describe the configuration of C-3 with C-9a, i.e., the
bridgehead methine, as the 2nd notation.
[19] Although it is not easy to read the degree of the HOMO-LUMO
overlapping, the orientation of the MO could be expressed as a normal
vector of the plane determined by critical nearby atoms. The LUMO
orientation could be defined as the normal vector of the plane spanned
by iminium nitrogen N-5, proton H-9a and carbon C-1, while the HOMO
orientation as the vector of the plane spanned by acetylenic carbon C-9
and the two ortho carbons in the phenyl ring.
[16] S. Ohmiya, H. Kubo, H. Otomasu, K. Saito, I. Murakoshi, Heterocycles
1990, 30, 537-542.
[17] S. Kobayashi, H. Tanaka, H. Amii, K. Uneyama, Tetrahedron 2003, 59,
1547-1552.
[18] a) E. Airiau, C. Chemin, N. Girard, G. Lonzi, A. Mann, E. Petricci, J.
Salvadori, M. Taddei, Synthesis 2010, 2901-2914; b) A. C. Cutter, I. R.
Miller, J. F. Keily, R. K. Bellingham, M. E. Light, R. C. D. Brown, Org. Lett.
2011, 13, 3988-3991.
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

10.1002/asia.201801193
Chemistry - An Asian Journal
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 1:
FULL PAPER
Text for Table of Contents
Author(s), Corresponding Author(s)*
Page No. – Page No.
Title
((Insert TOC Graphic here: max.
width: 5.5 cm; max. height: 5.0 cm))
Layout 2:
FULL PAPER
Jui-Chi Tsai, Yi-Huei Lin, Guei-Tang
Chen, Yu-Kai Gao, Yu-Che Tseng,
Chien-Lun Kao and Wen-Hua Chiou*
AcO
R
PMP
H
PMP
R
[Rh]/CO/H2
H+/ AcOH
Page No. – Page No.
H
N
Rh-Catalyzed Domino Hydro-
formylation Double Cyclizations of
Arylacetylenecarboxamides:
Diastereoselectivity Study and
Application for the synthesis of 1-
Azabicyclo[x.y.0]alkanes
m = 0, 1, 2
n = 1, 2
X, Y = CH2, CO
R = Me
n
m
N
Y
6
Text forXTable of Contents
Y
m
n
X
1
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.