Designed Bifunctional Phosphine Ligand-Enabled Gold-Catalyzed Isomerizations of Ynamides and Allenamides: Stereoselective and Regioselective Formation of 1-Amido-1,3-dienes

Article abstract of DOI:10.1021/acs.orglett.7b02624

By using designed biphenyl-2-ylphosphines functionalized with a remote basic groups as ligands, N-alkynyl-o-nosylamides are directly converted to (1E,3E)-1-amido-1,3-dienes with excellent diastereoselectivities under gold catalysis. With allenamides as su

Full text of DOI:10.1021/acs.orglett.7b02624

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX-XXX
Designed Bifunctional Phosphine Ligand-Enabled Gold-Catalyzed
Isomerizations of Ynamides and Allenamides: Stereoselective and
Regioselective Formation of 1‑Amido-1,3-dienes
Xingguang Li,†,‡ Zhixun Wang,† Xu Ma,† Pei-nian Liu,*,‡ and Liming Zhang*,†
†Department of Chemistry and Biochemistry University of California, Santa Barbara, California 93106, United States
‡Shanghai Key Laboratory of Functional Materials Chemistry Key Lab for Advanced Materials and Institute of Fine Chemicals, East
China University of Science and Technology, 130 Meilong Road, Shanghai 200237, P. R. China
S
* Supporting Information
ABSTRACT: By using designed biphenyl-2-ylphosphines functionalized with a
remote basic groups as ligands, N-alkynyl-o-nosylamides are directly converted to
(1E,3E)-1-amido-1,3-dienes with excellent diastereoselectivities under gold catalysis.
With allenamides as substrates, the gold-catalyzed isomerizations are high yielding
and applicable to a broad substrate scope including various nitrogen protecting
groups and exhibit unprecedented (3E)-selectivities for the distal C−C double bond
and good regioselectivities. Combining this gold catalysis with one-pot Diels−Alder
reactions leads to rapid assembly of valuable bicyclic compounds.
Scheme 1. Formations of 1-Amido-1,3-dienes and Our
Design on Gold-Catalyzed Isomerization
1-Amido-1,3-diene is an attractive N-delivering structural motif
owing to its general stability over its amino counterparts, and
synthetic intermediates featuring it could participate in inter- or
intramolecular Diels−Alder reactions en route to valuable N-
containing cyclic structures. This synthetic potential, however,
has been seldom reported, likely due to their limited access in
synthetically desirable stereoselective and regioselective man-
ners. In contrast, the related 1-amidofurans are frequently
employed in total synthesis.1 The reported synthetic methods
include the following: (a) the thermal rearrangements of
propargylic trichloroacetimidates to form (1Z)-1-amido-1,3-
dienes,2 which is less desirable for the D−A reaction, and upon
Et3N treatment affords a mixture of (1E) and (1Z) isomers
(e.g., 85/15); (b) metal-mediated/-catalyzed amidative3 or
C(sp2)−C(sp2)4 cross-couplings, which, however, suffers from
the need for halodiene substrates of preinstalled regio- and
stereochemistry in the former and the need for controlling
stereochemistry of both reacting partners in the latter (Scheme
1A); (c) allenamide isomerization under excessive heating (e.g.,
135 °C5) or in the presence of an acid catalyst (Scheme 1B),
which is limited in scope to products that avoid regio- or
stereochemical issues at the distal C−C double bond.5,6
Herein, we disclose the results including the combination of
gold catalysis with a one-pot Diels−Alder reaction to construct
valuable N-containing bicyclic structures.
Our initial reaction discovery and conditions optimization
were conducted with the ynamide 1a-Ns(o) (eq 1; for details,
please see Supporting Information (SI)). Under the optimal
conditions, i.e., L1AuCl (5 mol %), NaBARF (20 mol %),
DCE, 3 Å MS, 60 °C, and 36 h, the desired 1-amido-1,3-diene
Recently we7 reported gold-catalyzed isomerizations of
arylalkynes and certain types of aliphatic alkynes into dienes8
via the intermediacy of allenes. The chemistry is enabled by
designed biphenyl-2-ylphosphine ligands7,9 featuring a basic 3′-
amino group. We envision that 1-amido-1,3-dienes could be
prepared via our alkyne isomerization chemistry directly from
ynamides or from the intermediary allenamides (Scheme 1C).
Moreover, our gold-specific ligands might enable unprece-
dented control of the stereo- and regiochemical outcomes.
Received: August 23, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b02624
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
and a 5-OTIPS group were less reactive, and the catalyses
required longer reaction times and higher catalyst loadings to
afford the corresponding dienes in 78% and 71% yield,
respectively (entries 2 and 3). Intriguingly, the reaction of 1e
possessing a 4-OTIPS exhibited good reactivity, and the
amidodiene 2e was formed with mostly (Z)-configuration on
the distal C−C double bond (entry 4). However, ynamide 1f
derived from the related secondary alcohol afforded the
amidodiene 2f with mostly (3E)-configuration in 83% yield
(entry 5). The ynamide 1g with a tethered electron-deficient
olefin was also successfully isomerized into the corresponding
ω-amidotrienoate product 2g, albeit with a 10 mol % catalyst
loading and in a moderate yield (entry 6). Finally, the N-alkyl
substrate 1h participated well in the reaction, affording 2h in
81% yield (entry 7). It remains to be pointed out that these
reactions display excellent (1E)-selectivity and the (1Z)-
isomers were not observed.
2a-Ns(o) was formed in 89% isolated yield and exhibits
exclusive (1E) geometry and a synthetically valuable 8.8:1 (3E)-
selectivity at the distal C−C double bond. Notably, ynamides
have been extensively studied in gold catalysis10 and invariably
exhibit pronounced polarization upon interaction with electro-
philic species (e.g., LAu+) in the form of ketenimium-type
intermediates, which is opposite to what occurs in our
chemistry. Remarkably the conventional gold ligands, e.g.,
Ph3P, IPr, JohnPhos, are completely incapable of promoting
this amidodiene formation.
The ynamide isomerization, however, does not work with
substrates possessing additional substitution at the propargylic
position [e.g., N-benzyl-N-(cyclohexylethynyl)-2-nitrobenzene-
sulfonamide] and is limited to o-nosylamides. We surmised that
the culprit must be the first proton shuttling in the gold
catalysis, i.e., the isomerization of the ynamide into an
allenamide intermediate.7 As such, allenamides, also available
via Cu catalysis,11 would permit a broadened reaction scope if
employed as substrates. To this end, a range of allenamides
were subjected to the gold catalysis (Table 2). To our delight,
N-tosylallenamide 3a reacted efficiently in the presence of only
0.5 mol % L1AuCl and 2.0 mol % NaBARF at ambient
temperature and in 2 h, affording the terminal 1-amido-1,3-
diene 4a with exclusive (1E)-geometry and in 96% yield.
Similarly, 3b with the distal allene end fully substituted
underwent highly efficient transformation under mild con-
ditions (entry 2). The reaction of 3c featuring an n-propyl at
the allene distal end was again high yielding, and most notably
the 1-amido-1,3-diene product 4c exhibits excellent (3E)-
selectivity (>50:1) in addition to the exclusive (1E)-geometry
(entry 3). In contrast, the treatment of 3c with CSA5 led to a
nearly equal mixture of (3E)- and (3Z)-products and a lower
yield, highlighting the extraordinary utility of the designed
ligand. In addition, with JohnPhos as the gold ligand, only trace
4c (<5%) was formed under the same conditions (i.e., rt, 2 h)
and an 11% yield was produced at 40 °C for 12 h (68%
conversion), again highlighting the enabling nature of our
designed bifunctional ligand. Variation of the amide moiety to
include benzamide (entry 4), oxazolidinone (entry 5), and
lactam (entry 6) were inconsequential, and the gold catalyses
uniformly exhibit >90% yields and exceptional (3E)-selectiv-
ities, in contrast to the poor or no geometric selectivities with
the distal alkene in CSA-catalyzed counterparts.
The reaction scope was then studied under the optimal
conditions. As shown in Table 1, 1b with a cyclohexylmethyl
group at the alkyne end reacted smoothly to afford 1-amido-
1,3-diene 2b in 93% yield. The effect of having a functional
group close to the triple C−C bond was then probed.
Compared to 1a-Ns(o), substrates possessing a 6-OAc group
a
Table 1. Scope of Ynamide Isomerization
Allenamide 3g with two different substituents at its distal
allene terminus poses an additional and significant challenge to
the intended isomerization, i.e. regioselectivity. As shown in
entry 7, with L1 as the ligand, the reaction remained highly
efficient and (3E)-selective in the case of 4g, but the ratio of the
regioisomers 4g and 4g′ was only 1.5/1, favoring the former.
With CSA as the catalyst, both the regio- and diastereose-
lectivity were poor, despite the methylene product 4g′ is
slightly favored. Recently we reported that by positioning the
remote aniline nitrogen in L1 further away from the pendent
phenyl ring in the form of more basic tertiary amines, the
isomerization of propargylic esters can be realized with
controlled regioselectivity.9b When one of the ligands, i.e., L2,
a
Reactions were carried out with L1AuCl (5 mol %), NaBARF (20
b
c
mol %), 3 Å MS in DCE (0.05 M) for 36 h. Isolated yield. L1AuCl
(7.5 mol %), NaBARF (30 mol %) were used. L1AuCl (10 mol %),
NaBARF (30 mol %) were used.
d
B
DOI: 10.1021/acs.orglett.7b02624
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
excellent geometric selectivity (entry 9); in contrast, both the
regioselectivity and geometric selectivity were again poor in the
Brønsted acid catalysis. Finally, when the allenamide 3i was
subjected to our gold catalysis, the L1Au+-catalyzed reaction led
to regioselective deprotonation at the smaller methyl group and
hence afforded the kinetic product 4i′ in 92% yield (entry 10).
In contrast, the acid-catalyzed transformation yielded mostly
the thermodynamic product 4i, in which the deprotonation
occurs at the much bigger isopropyl group. This result
demonstrates the sensitivity of our design ligand to steric
hindrance en route to hard-to-achieve regioselectivity.
Table 2. Scope of Allenylamide Isomerization
The observed regio- and stereoselectivities in these gold
catalyses can be readily rationalized by invoking the essential
role of the ligand remote basic group. For details, see the SI.
We anticipated that the synthesis of 1-amido-1,3-dienes from
the gold catalyzed ynamide isomerizations could be rendered in
tandem with Diels−Alder reactions in one pot, therefore
enabling rapid assembly of versatile cyclic structures. We first
examined intramolecular D−A reactions. As shown in Table 3,
Table 3. Scope of Ynamide Isomerization−Cycloaddition
a
Domino Reaction
a
Reactions were carried out with L1AuCl (10 mol %), NaBARF (30
mol %), 3 Å molecular sieves in DCE (0.05 M) at 80 °C for 48 h, and
b
then treated with BHT in mesitylene at 180 °C for 12 h. L1AuCl (5
mol %), NaBARF (20 mol %), 3 Å molecular sieves in DCE (0.05 M)
c
at 60 °C for 36 h, and then 100 °C for 12 h. Isolated yield.
a
Reactions were carried out with LAuCl (0.5 mol %), NaBARF (2.0
b
mol %) in DCE (0.10 M). Reactions were carried out with 10% CSA,
3 Å MS in DCM (0.10 M) at rt for 20 min. NMR yield was reported
using triisopropylbenzene as internal reference. Isolated yield. LAuCl
(2.5 mol %), NaBARF (10 mol %) were used. LAuCl (7.5 mol %),
ynamides 1i and 1j featuring α,β-unsaturated sulfones tethered
at the nitrogen atom indeed underwent this tandem process to
afford the hexahydroindole 5a and the octahydroquinoline 5b
with exquisite relative stereochemistry in synthetically useful
yields (entries 1 and 2). The Diels−Alder reactions were
performed at 180 °C and displays excellent endo selectivity.
With a maleate properly installed at the alkyne end, 1k
smoothly underwent the tandem process to deliver the bicyclic
lactone 5c as a mixture of diastereomeric isomers in a moderate
64% combined yield (entry 3). In the case of an intermolecular
Diels−Alder reaction, we heated 1a in the presence of N-
phenylmaleimide in DCE at 60 °C for 36 h and then raised the
c
d
e
f
NaBARF (20 mol %) were used.
was employed, much to our delight, the isomerization is highly
regioselective, affording 4g with a trisubstituted distal C−C
double bond with exclusive (3E)-geometry in an excellent yield
(entry 8). With the n-octyl in 3g replaced by a TBSOCH2
group in 3h, L2 is again a more effective ligand, resulting in the
formation of (1E,3E)-4h with good regioselectivity and again
C
DOI: 10.1021/acs.orglett.7b02624
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Am. Chem. Soc. 2008, 130, 3633−3644. (e) Maity, P.; Kundu, D.;
Ranu, B. C. Adv. Synth. Catal. 2015, 357, 3617−3626.
(4) (a) Pawluc, P.; Franczyk, A.; Walkowiak, J.; Hreczycho, G.;
́
temperature to 100 °C for 12 h. The endo-cycloadduct 5d was
isolated in 78% yield (entry 4).
In summary, by using a remotely base-functionalized
biphenyl-2-ylphosphine as a ligand, we demonstrated for the
first time that N-alkynyl-o-nosylamides can be converted
directly to synthetically versatile (1E,3E)-1-amido-1,3-dienes
with good geometric selectivities under gold catalysis. With
allenamides as substrates, the gold-catalyzed isomerizations
have excellent yields and exhibit a broad substrate scope
including various nitrogen protecting groups; moreover, in
contrast to the generally poor regio- and stereoselectivities
under Brønsted acid catalysis, the reaction affords excellent
(3E)-selectivities for the distal C−C double bond and good
regioselectivities. For both approaches, typical gold ligands are
either incompetent or substantially less effective, confirming the
unique capacity of the designed ligands and the essential role of
the remoted basic group. Considering the ease of substrate
synthesis, these ligand-enabled isomerizations provide facile
and stereoselective approaches to versatile (1E,3E)-1-amido-
1,3-dienes, the synthetic potential of which is demonstrated in
one-pot Diels−Alder reactions, which offer rapid access to
valuable bicyclic compounds.
Kubicki, M.; Marciniec, B. Tetrahedron 2012, 68, 3545−3551.
(b) Couty, S.; Barbazanges, M.; Meyer, C.; Cossy, J. Synlett 2005,
2005, 0905−0910. (c) Armstrong, I. S.; Bergmeier, S. C. Synthesis
2017, 49, 2733−2742.
(5) Hayashi, R.; Hsung, R. P.; Feltenberger, J. B.; Lohse, A. G. Org.
Lett. 2009, 11, 2125−2128.
(6) Feltenberger, J. B.; Hsung, R. P. Org. Lett. 2011, 13, 3114−3117.
(7) Wang, Z.; Wang, Y.; Zhang, L. J. Am. Chem. Soc. 2014, 136,
8887−8890.
(8) (a) Taylor, R. E.; Diene, C. R.; Daly, E. M. Sci. Synth. 2009, 46,
571−582. (b) Guo, C.; Lu, X. Tetrahedron Lett. 1991, 32, 7549−7552.
(c) Trost, B. M.; Schmidt, T. J. Am. Chem. Soc. 1988, 110, 2301−2303.
(d) Kwong, C. K.-W.; Fu, M. Y.; Lam, C. S.-L.; Toy, P. H. Synthesis
2008, 2008, 2307−2317. (e) Shintani, R.; Duan, W.-L.; Park, S.;
Hayashi, T. Chem. Commun. 2006, 3646−3647. (f) Yasui, H.;
Yorimitsu, H.; Oshima, K. Synlett 2006, 2006, 1783−1785.
(9) (a) Wang, Y.; Wang, Z.; Li, Y.; Wu, G.; Cao, Z.; Zhang, L. Nat.
Commun. 2014, DOI: 10.1038/ncomms4470. (b) Wang, Z.; Ying, A.;
Fan, Z.; Hervieu, C.; Zhang, L. ACS Catal. 2017, 7, 3676−3680. (c) Li,
X.; Liao, S.; Wang, Z.; Zhang, L. Org. Lett. 2017, 19, 3687−3690.
(10) (a) Pan, F.; Shu, C.; Ye, L.-W. Org. Biomol. Chem. 2016, 14,
9456−9465. (b) Evano, G.; Theunissen, C.; Lecomte, M. Aldrichimica
Acta 2015, 48, 59−70. (c) Wang, X.-N.; Yeom, H.-S.; Fang, L.-C.; He,
S.; Ma, Z.-X.; Kedrowski, B. L.; Hsung, R. P. Acc. Chem. Res. 2014, 47,
560−578.
ASSOCIATED CONTENT
■
S
* Supporting Information
(11) (a) Zhang, Y.; Hsung, R. P.; Tracey, M. R.; Kurtz, K. C.; Vera, E.
L. Org. Lett. 2004, 6, 1151−1154. (b) Persson, A. K.; Johnston, E. V.;
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.7b02624.
Backvall, J.-E. Org. Lett. 2009, 11, 3814−3817.
̈
Experimental procedures and compound characterization
and spectra (PDF)
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail: zhang@chem.ucsb.edu (L.Z.).
*E-mail: E-mail:liupn@ecust.edu.cn (P.L.).
ORCID
Liming Zhang: 0000-0002-5306-1515
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank NSF CHE-1301343 for financial support and NIH
shared instrument grant S10OD012077 for the purchase of a
400 MHz NMR spectrometer. Z.W. thanks the Mellichamp
Academic Initiative in Sustainability for a fellowship in support
of his research. X.L. thanks the China Scholarship Council
fellowship in support of his research.
REFERENCES
■
(1) (a) Takao, K.-i.; Munakata, R.; Tadano, K.-i. Chem. Rev. 2005,
105, 4779−4807. (b) Juhl, M.; Tanner, D. Chem. Soc. Rev. 2009, 38,
2983−2992.
(2) (a) Overman, L. E.; Clizbe, L. A. J. Am. Chem. Soc. 1976, 98,
2352−2354. (b) Overman, L. E.; Clizbe, L. A.; Freerks, R. L.;
Marlowe, C. K. J. Am. Chem. Soc. 1981, 103, 2807−15.
(3) (a) Hartwig, J. F. Angew. Chem., Int. Ed. 1998, 37, 2046−2067.
(b) Wolfe, J. P.; Wagaw, S.; Marcoux, J.-F.; Buchwald, S. L. Acc. Chem.
Res. 1998, 31, 805−818. (c) Shen, R.; Lin, C. T.; Bowman, E. J.;
Bowman, B. J.; Porco, J. A. J. Am. Chem. Soc. 2003, 125, 7889−7901.
(d) Nicolaou, K.; Sun, Y.-P.; Guduru, R.; Banerji, B.; Chen, D. Y.-K. J.
D
DOI: 10.1021/acs.orglett.7b02624
Org. Lett. XXXX, XXX, XXX−XXX