Au(I)-Catalyzed Synthesis of Trisubstituted Indolizines from 2-Propargyloxypyridines and Methyl Ketones

Article abstract of DOI:10.1021/acs.orglett.9b01929

A new Au(I)-catalyzed method for the preparation of trisubstituted indolizines from easily accessible 2-propargyloxy-pyridines is reported. The reaction tolerates a wide range of functionality, allowing for diversity to be introduced in four distinct regions of the product (R, R¹, R², and Ar). The proposed mechanism proceeds via enol addition to an allenamide intermediate and explains the observed increase in yields when electron poor methyl ketones are utilized.

Full text of DOI:10.1021/acs.orglett.9b01929

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Au(I)-Catalyzed Synthesis of Trisubstituted Indolizines from
2‑Propargyloxypyridines and Methyl Ketones
Matthew D. Rossler,† Colin T. Hartgerink,† Emily E. Zerull,† Benjamin L. Boss, Abigail K. Frndak,
Miles M. Mason, Leslie A. Nickerson,‡ Evan O. Romero, Jaimie E. Van de Burg, Richard J. Staples,§
and Carolyn E. Anderson*
Department of Chemistry and Biochemistry, Calvin College, Grand Rapids, Michigan 49546, United States
S
* Supporting Information
ABSTRACT: A new Au(I)-catalyzed method for the preparation of trisubstituted indolizines from easily accessible
2‑propargyloxy-pyridines is reported. The reaction tolerates a wide range of functionality, allowing for diversity to be introduced
in four distinct regions of the product (R, R1, R2, and Ar). The proposed mechanism proceeds via enol addition to an
allenamide intermediate and explains the observed increase in yields when electron poor methyl ketones are utilized.
core largely fall into one of three categories: 1,3-dipolar
cycloaddition to pyridinium ions,5 cyclocondensations,6 or
transition metal catalyzed cycloisomerization reactions.7
As these are the common approaches to indolizine
formation, we were surprised to observe trisubstituted
indolizine 4 as a significant byproduct during the synthesis
of pyridonyl enol ether 2 from 2-propargyloxypyridine 1
(Scheme 1).8 To the best of our knowledge, this is the first
synthesis of trisubstituted indolizines from a 2-alkoxypyridine,
and it represents a unique approach to the indolizine core.
Formation of compound 4 was confirmed by NMR and X-ray
analysis. Given the medicinal importance of indolizines and the
simplicity of the starting materials, efforts were undertaken to
explore this reaction further.
ndolizines are versatile heterocyclic cores that are
Icommonly found within pharmaceutical targets for a
broad array of therapeutic indications (Figure 1).1 The
medicinal value of indolizines has resulted in the development
of a number of methods for the synthesis of this privileged
motif.2 Beyond the classical strategies of Scholtz3 and
Tschichibabin,4 methods for the synthesis of the indolizine
Optimization
Initial investigations focused on the 1-phenylethanol solvent.
While indolizine 4 was originally observed under conditions
where only 1-phenylethanol, MgSO4, and catalyst 5a were
thought to be present, closer inspection of product 4 suggested
that an oxidized form of the alcohol, specifically acetophenone,
1
was being incorporated. This was confirmed by H NMR
analysis of the original lot of 1-phenylethanol, which was found
to contain ∼10% acetophenone. When the reaction was
performed with 1-phenylethanol that lacked this contamin-
ation, no indolizine was observed (Table 1, entry 1). By
intentionally adding acetophenone to the reaction, it was
possible to recreate the original conditions (entries 2 and 3).
Received: June 4, 2019
Figure 1. Pharmaceutical targets containing indolizine cores.
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.9b01929
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 1. First Observation of Trisubstituted Indolizines
Scheme 2. Catalyst Screen
pyridine 1 toward nucleophilic addition;8,9 however, in this
case, it was found to be ineffective. For simplicity, catalyst 5c
was utilized for all further studies.
a
The 1-phenylethanol solvent was later found to contain ∼10%
1
acetophenone, as determined by H NMR (vide infra).
Attempts to improve the reaction yield by adding water
scavengers (e.g., Na2SO4, CaSO4, or molecular sieves) resulted
in no enhancement in yield over reactions run in the absence
of any such additives (see Supporting Information for full
details). Further, augmenting the reaction with either acidic
(e.g., TsOH) or basic additives (e.g., pyridine, morpholine,
K2CO3) gave either similar or decreased yields of indolizine 4.
Table 1. Solvent Studies
a
entry
1-phenylethanol
acetophenone
yield 4 (%)
1
2
3
4
5
6
7
0.5 M
0.5 M
0.5 M
0.5 M
5 equiv
−
0
23
34
32
42
39
20
1 equiv
3 equiv
3 equiv
0.5 M
0.5 M
0.5 M
b
Reaction Scope
The reaction scope was then evaluated, focusing initially on
different methyl ketones (Table 2). In these cases, yields were
found to be higher when the reactions were performed at a
lower concentration (0.25 M versus 0.5 M). Utilizing a series
of substituted acetophenone derivatives, it was observed that
electron-poor methyl ketones provided yields of indolizine 7
that were higher than those of either acetophenone or
electron-rich variants (entries 9−13 versus entries 1−5).
This suggests that a more electrophilic carbonyl correlates to
higher reaction yields. Interestingly, fluorinated acetophenone
derivatives do not seem to follow this electronic trend, as only
3-fluoroacetophenone gives yields in the expected range for an
electron deficient methyl ketone (entries 6−8). This may be
due to the ability of ortho- and para-fluorine substituents to
serve as π-electron donors rather than as strictly σ-withdrawing
groups.10 Similarly, pyridinyl ketones, which would also be
expected to be electron-poor, give only slightly improved yields
of the corresponding indolizines 7m and 7n (entries 14 and
15). Attempts to utilize alkyl methyl ketones failed to yield an
isolable amount of the corresponding indolizine 7o due to its
rapid decomposition (entry 16).
2.5 equiv
a
b
Conditions: catalyst 5a (5 mol %), 100 °C, 18 h. MgSO4 (1 equiv)
added.
Further improvement could be realized if acetophenone was
used as the solvent; however, the reaction still benefitted from
a small amount of alcohol (entries 5−7). These studies also
revealed that MgSO4 was unnecessary (entry 4 versus entry 3)
and that the product could be obtained reproducibly when
heated conventionally at 100 °C for 18 h. Increasing the
duration of the reaction further had no impact on the yield of
compound 4, as the remainder of the mass was converted into
ketone 3 under the reaction conditions.
Using this combination of solvents, additional Au(I) and
Pt(II) catalysts were screened in the reaction (Scheme 2).
Electron-rich biaryl triflimide Au(I) catalysts 5c and 5e and
PtCl2 were found to work best, providing compound 4 in 54,
52, and 50%, respectively. Biaryl Au(I) catalysts in which gold
was ligated to chlorine, as well as NHC catalysts, were found to
be inefficient in the reaction. Improved yields could be
achieved with these catalysts upon addition of Ag salts (e.g., 5b
plus AgSbF6, 6a or 6d plus AgOTf, Scheme 2); however,
catalysts that were precomplexed with weakly coordinating
groups (e.g., CH3CN, NTf2) still provided the highest yields.
NaAuCl4·2H2O was also screened in the indolizine formation,
as it has been shown previously to activate 2-propargyloxy-
Extension of the indolizine formation to differently
substituted 2-propargyloxypyridines 8 was also accomplished
(Table 3). In general, substitution at the terminal position was
well tolerated with even large cyclohexyl substituents
proceeding in moderate to good yields, depending on the
coupling partner (entries 5−7). As was observed previously in
other gold-catalyzed reactions of 2-propargyloxypyridines,8
B
DOI: 10.1021/acs.orglett.9b01929
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Table 2. Viability of Diverse Methyl Ketones
Table 4. Substitution on the Pyridine Ring
a
entry
Ar
yield (%)
1
2
3
4
5
6
7
8
Ph
4
54
40
40
38
38
41
66
51
73
76
80
62
71
51
59
−
yield
a
entry SM
R1
X
Ar
product (%)
4-Me-C6H4
4-OMe-C6H4
3-OMe-C6H4
2,3-OCH2O-C6H3
2-F-C6H4
3-F-C6H4
4-F-C6H4
4-NO2-C6H4
3-NO2-C6H4
3,5-(CF3)2C6H3
2-naphthyl
7a
7b
7c
7d
7e
7f
7g
7h
7i
7j
7k
7l
7m
7n
7o
1
2
3
4
5
6
7
8
9
10a
10b
10c
10c
10d
10d
10d
10e
10e
COOH
CO2Me
NHPiv
NHPiv
NHPiv
CH2OTIPS
CH2OTIPS
H
CH
CH
CH
CH
CH
CH
CH
N
Ph
Ph
Ph
11a
11b
11c
11cc
11ccc
11d
11dd
11e
−
−
65
56
65
62
86
8
3-F-C6H4
3-NO2-C6H4
Ph
3,5-(CF3)2C6H3
4-F-C6H4
3,5-(CF3)2C6H3
b
9
10
11
12
13
14
15
16
H
N
11ee
27
a
Isolated yields. Mean values from multiple experiments ( 2%).
Mean values from multiple experiments ( 3%).
b
furyl
3-pyridonyl
2-pyridonyl
CH2CH2CH3
a
the electronics of the substituent at C-3 of the pyridine directly
impacted the observed yield. When electron withdrawing
groups were present, formation of the indolizine was not
observed (entries 1 and 2). Conversely, when electron
donating substituents were added, good to excellent yields of
substituted indolizines 11c, 11cc, 11ccc, 11d, and 11dd were
observed (entries 3−7). Pyrimidine 10e was also susceptible to
the reaction; however, only low yields of dinitrogen analogues
11e and 11ee were observed due to their instability.
Substitution could also be tolerated at the C-6 position,
although in this case, the sterics of the reaction seem to
dominate, as no change in yield was observed with a more
electron deficient methyl ketone (Scheme 3).
Isolated yields. Mean values from multiple experiments ( 2%).
silylated substrate 8d, bearing a two carbon linker, is more
reactive than the analogous substrate 8e with only a methylene
linker (entries 8−10). The methyl ketone coupling partner
continued to be the largest predictor of reaction yield with
more electron-poor methyl ketones providing higher yields of
indolizine 9, regardless of the substitution pattern (entries 2, 4,
6, 7, 9, and 12 versus entries 1, 3, 5, 8, 10, 11, and 13).
Substituents at the propargylic position were also tolerated,
providing indolizines 9f, 9ff, and 9g, bearing extended chains at
the C-2 position, in reasonable yields (entries 11−13).
Substitution on the pyridine ring was also evaluated (Table
4). Unlike substitution on the propargyl chain of the substrate,
Table 3. Evaluation of Side Chain Substitution
a
entry
SM
R1
R2
Ar
product
yield (%)
1
2
3
4
5
6
7
8
8a
8a
8b
8b
8c
8c
8c
8d
8d
8e
8f
(CH2)3Ph
(CH2)3Ph
(CH2)2Ph
(CH2)2Ph
Cy
H
H
H
H
H
H
H
H
H
H
Me
Me
Et
Ph
9a
9aa
9b
9bb
9c
9cc
9ccc
9d
9dd
9e
9f
9ff
9g
49
72
49
64
40
57
71
45
59
5
4-NO2-C6H4
Ph
3,5-(CF3)2C6H3
Ph
3-F-C6H4
3,5-(CF3)2C6H3
Ph
3-F-C6H4
Cy
Cy
(CH2)2OTIPS
(CH2)2OTIPS
CH2OTIPS
Et
Et
C4H9
9
10
11
12
13
Ph
Ph
37
69
41
8f
8g
3,5-(CF3)2C6H3
Ph
a
Isolated yields. Mean values from multiple experiments ( 2%).
C
DOI: 10.1021/acs.orglett.9b01929
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
been previously implicated in the synthesis of pyridonyl ketone
3.9 Attempts to synthesize intermediate 14 independently were
unsuccessful due, we believe, to its high reactivity. It was,
however, possible to prepare saturated allenamide analogue 15.
Upon subjecting compound 15 to the reaction conditions, a
small but isolable amount of analogue 16 was obtained
(Scheme 6). The low yields of compound 16 can be attributed
Scheme 3. Substitution at the C-6 Position
Scheme 6. Evaluation of Allenamides as Intermediates
Solvent Recycling
While utilizing substituted acetophenones as the reaction
solvent is unusual, it has been shown that the solvent mixture
can be recovered after Kugelrohr distillation of the crude
reaction mixture (containing both methyl ketone and
1‑phenylethanol). This mixture can then be reused in
subsequent reactions without any loss in efficiency (Scheme
4).
to its instability, as it was observed to decompose both during
purification and on the benchtop. Despite this, the formation
of heterocycle 16 suggests that allenamide 14 and its analogues
are indeed likely intermediates in the formation of indolizine 4.
Conversion of allenamide 14 into indolizine 4 is proposed to
occur as shown in Scheme 7. Beginning with the addition of
Scheme 4. Solvent Recycling
Scheme 7. Proposed Mechanism for Indolizine Formation
Mechanism
Formation of indolizine 4 and its analogues was originally
thought to proceed via ketone 3. However, subjecting ketone 3
to the reaction conditions provided no evidence of indolizine
formation (Scheme 5). Adding a small amount of 2‑prop-
argyloxypyridine 1 provided indolizine 4 but only in
proportion to the amount of starting material 1 that was
added. Given these results, ketone 3 was eliminated as a
possible intermediate, and allenamide 14 was pursued, as it had
Scheme 5. Formation of Pyridonyl Ketones 39 and
Indolizine 4
enol 17 to the central carbon of allenamide 14, subsequent
protodeauration, either from the alcohol or methyl ketone,
would then lead to the formation of β,γ-unsaturated ketone 19.
The proposed enol 17 is expected to form more rapidly when
the aryl group is electron-poor, leading to higher yields with
electron-poor methyl ketones as reported (see Table 2).
Enolization of compound 19 followed by aldol reaction and
loss of water would then provide indolizine 4.
The protodeauration of compound 18 is expected to occur
through a combination of pathways involving either deproton-
ation of the alcohol or the methyl ketone. In order to
determine the relative rates of these processes, the reaction was
performed in d3-acetophenone in the presence of proteo
alcohol (Scheme 8). In this case, 23% deuterium incorporation
was observed at the exocyclic methyl group, suggesting that
protodeauration is ∼3× faster from the alcohol than the
methyl ketone (see Supporting Information).
D
DOI: 10.1021/acs.orglett.9b01929
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
spectrometers were provided by Major Research Instrumenta-
tion grants of the National Science Foundation under Grants
CHE-0922973 and CHE-0741793, respectively. The CCD
based X-ray diffractometer at Michigan State University was
upgraded and/or replaced by departmental funds. We thank
Dr. C. D. Anderson (tesa tape), Dr. J. B. Johnson (Hope
College), Dr. D. E. Benson (Calvin College), and Mr. J. Lear
(Ohio State University) for helpful insights and support of this
work.
Scheme 8. Isotope Study
REFERENCES
Summary
■
(1) (a) Gundersen, L.-L.; Charnock, C.; Negussie, A. H.; Rise, F.;
Teklu, S. Eur. J. Pharm. Sci. 2007, 30, 26−35. (b) Hsieh, H.-P.; Chao,
Y.-S.; Liou, J.-P.; Chang, J.-Y.; Tung, Y.-S. Preparation of bicyclic
heterocyclo phenyl ketones as antitumor agents. U.S. Pat. Appl. US
20060148801 A1, 2006. (c) Nekkar, P.; Rao, P.; Kabir, S. N.
Preparation of indolizine derivatives as inhibitors of lipoprotein
associated phospholipase A2 and 15-lipoxygenase. PCT Int. Appl.
WO 2013116948 A1, 2013. (d) Hynd, G.; Ray, N. C.; Finch, H.;
Montana, H. G.; Cramp, M. C.; Harrison, T. K.; Arienzo, R.; Blaney,
P.; Griffon, Y.; Middlemiss, D. Indolizine derivatives as CRTH2
receptor modulators, their preparation, pharmaceutical compositions,
and use in therapy. PCT Int. Appl. WO 2007031747 A1, 2007.
(e) Oku, T.; Sawada, K.; Kuroda, A.; One, K. Preparation of
indolizines as cyclic guanosine 3′,5′-monophosphate phosphodiester-
ase inhibitors. Jpn. Kokai Tokkyo Koho JP 10120680 A, 1998.
(f) Stearns, B. A.; Roppe, J. R.; Parr, T. A.; Stock, N. S.; Volkots, D.;
Hutchinson, J. H. Preparation of indolizine inhibitors of 5-
lipoxygenase for treating respiratory, cardiovascular, and other
leukotriene-mediated conditions. PCT Int. Appl. WO 2011038086
A2, 2011.
A new method for the synthesis of trisubstituted indolizines
from 2-propargyloxypyridines and aryl methyl ketones has
been discovered. The reaction proceeds in the highest yields in
the presence of electron-poor ketones, presumably due to the
ease of forming enol 17 under the reaction conditions. The
wide scope of this reaction and the simplicity of the starting
materials make this transformation significant within the array
of methods available for the synthesis of indolizine-containing
medicinal targets.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.9b01929.
Representative experimental procedures, further opti-
mization studies, solid-state packing diagrams for 4,
spectroscopic support for the formation of partially
deuterated indolizine 4/4′, and 1H and 13C NMR
spectra for all new compounds (PDF)
(2) Welsch, M. E.; Snyder, S. A.; Stockwell, B. R. Curr. Opin. Chem.
Biol. 2010, 14, 347−361.
(3) Scholtz, M. Ber. Dtsch. Chem. Ges. 1912, 45, 734−746.
(4) Tschitschibabin, A. E. Ber. Dtsch. Chem. Ges. B 1927, 60, 1607−
1617.
Accession Codes
CCDC 1920795 contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge
via www.ccdc.cam.ac.uk/data_request/cif, by emailing data_
request@ccdc.cam.ac.uk, or by contacting The Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge
CB2 1EZ, UK; fax: +44 1223 336033.
(5) (a) Brioche, J.; Meyer, C.; Cossy, J. Org. Lett. 2015, 17, 2800−
2803. (b) Li, F.; Chen, J.; Hou, Y.; Li, Y.; Wu, X.-Y.; Tong, X. Org.
Lett. 2015, 17, 5376−5379. (c) Xu, J.; Hu, H.; Liu, Y.; Wang, X.; Kan,
Y.; Wang, C. Eur. J. Org. Chem. 2017, 2017, 257−261. (d) Wang, W.;
Han, J.; Sun, J.; Liu, Y. J. Org. Chem. 2017, 82, 2835−2842.
(6) (a) Kucukdisli, M.; Opatz, T. Eur. J. Org. Chem. 2014, 2014,
5836−5844. (b) Tverdokhleb, N. M.; Khoroshilov, G. E.; Dotsenko,
V. V. Tetrahedron Lett. 2014, 55, 6593−6595. (c) Mohan, D. C.; Ravi,
C.; Pappula, V.; Adimurthy, S. J. Org. Chem. 2015, 80, 6846−6855.
(7) Review: Wang, L.-X.; Tang, Y.-L. Eur. J. Org. Chem. 2017, 2017,
2207−2213.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: Carolyn.Anderson@calvin.edu.
ORCID
(8) Romero, E. O.; Reidy, C. P.; Bootsma, A. B.; PreFontaine, N.
M.; Vryhof, N. W.; Wierenga, D. C.; Anderson, C. E. J. Org. Chem.
2016, 81, 9895−9902.
Richard J. Staples: 0000-0003-2760-769X
Carolyn E. Anderson: 0000-0001-8797-4120
Present Addresses
(9) Romero, N. A.; Klepser, B. M.; Anderson, C. E. Org. Lett. 2012,
14, 874−877.
(10) Carroll, T. X.; Thomas, T. D.; Bergersen, H.; Borve, K. J.;
Saethre, L. J. Org. Chem. 2006, 71, 1961−1968.
‡L.A.N.: Department of Chemistry, University of California-
Davis, Davis, CA 95616
§R.J.S.: Department of Chemistry, Michigan State University,
East Lansing, MI 48824
Author Contributions
†M.D.R., C.T.H., and E.E.Z. contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the National Science Foundation
under Grants CHE-1266314 and CHE-1665139 and awards
from Calvin College. NMR spectrometers and mass
E
DOI: 10.1021/acs.orglett.9b01929
Org. Lett. XXXX, XXX, XXX−XXX