Single-Step Synthesis of 5,6,7,8-Tetrahydroindolizines via Annulation of 2-Formylpiperidine and 1,3-Dicarbonyl Compounds

Article abstract of DOI:10.1021/acs.orglett.5b01671

An expedient single-step synthesis of 5,6,7,8-tetrahydroindolizines has been achieved via the annulation of commercially available 2-formylpiperidine hydrochloride and 1,3-dicarbonyl compounds in THF in the presence of pyrrolidine and 4 ? molecular sieves

Full text of DOI:10.1021/acs.orglett.5b01671

Letter
pubs.acs.org/OrgLett
Single-Step Synthesis of 5,6,7,8-Tetrahydroindolizines via Annulation
of 2‑Formylpiperidine and 1,3-Dicarbonyl Compounds
Simona S. Capomolla, Ngiap-Kie Lim, and Haiming Zhang*
Small Molecule Process Chemistry, Genentech Inc., 1 DNA Way, South San Francisco, California 94080, United States
S
* Supporting Information
ABSTRACT: An expedient single-step synthesis of 5,6,7,8-tetrahy-
droindolizines has been achieved via the annulation of commercially
available 2-formylpiperidine hydrochloride and 1,3-dicarbonyl
compounds in THF in the presence of pyrrolidine and 4 Å
molecular sieves. A variety of β-ketoesters, ketones, and amides participated in this annulation chemistry, affording the desired
5,6,7,8-tetrahydroindolizines in moderate to good yields.
Scheme 1. Synthetic Strategies to Tetrahydroindolizines
5,6,7,8-Tetrahydroindolizine is a structural motif found in many
biologically active compounds, and as such, is of general interest
to the synthetic and medicinal chemistry community.1−4 For
example, 5,6,7,8-tetrahydroindolizine cores are found in the
anticancer natural alkaloids rhazinilam and rhazinicine,1 and the
antimicrobial agents polygonatine A, polygonatine B, and
kinganone.3 CMV423, a 5,6,7,8-tetrahydroindolizine derivative,
shows promise for the treatment of human cytomegalovirus
(HCMV) infections (Figure 1).4
Figure 1. Biologically active 5,6,7,8-tetrahydroindolizines.
There are several reported approaches to the synthesis of
5,6,7,8-tetrahydroindolizines in the literature, each of which
centered on certain substitution pattern of the tetrahydroindo-
lizine core.5−8 For example, 5,6,7,8-tetrahydroindolizines are
strategy employs high loading of an expensive ruthenium catalyst
(3 mol % Ru3(CO)12), high CO pressure (20 bar), and high
temperature (140 °C). Therefore, the synthetic utility of these
aforementioned methods is limited, and we believe that a more
efficient and economic synthesis of 5,6,7,8-tetrahydroindolizines
is highly desirable.
In order to support an internal drug research and development
program, we were required to develop an efficient synthesis of a
wide variety of 5,6,7,8-tetrahydroindolizines for structure−
activity relationship (SAR) studies and further process develop-
ment. We decided to focus on a synthetic strategy involving an
annulation reaction of commercially available 2-formylpiperidine
hydrochloride (2a) and 1,3-dicarbonyl compounds. Herein, we
wish to report a facile one-step synthesis of 5,6,7,8-
synthesized via a 1,3-dipolar cycloaddition of munchnones9 with
̈
acetylenic dipolarophiles followed by elimination of CO2
(Scheme 1A).5 However, this process not only requires multistep
synthesis and high reaction temperature (125−140 °C) but also
affords low overall yields and poor regioselectivity when
unsymmetrical acetylenes are employed. Recently, Coelho
reported that 5,6,7,8-tetrahydroindolizines were prepared from
selective hydrogenation of indolizines that were derived from the
intramolecular cyclization of Morita−Baylis−Hillman (MBH)
adducts prepared from acrylates or α,β-unsaturated ketones and
substituted 2-pyridinecarboxaldehydes (Scheme 1B).6 Similarly,
this chemistry also suffers from multistep operations and low
overall yields. A single step synthesis of 5,6,7,8-tetrahydroindo-
lizines has been previously achieved by a ruthenium-catalyzed
multicomponent reaction (Scheme 1C).7 Unfortunately, this
Received: June 8, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.5b01671
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
tetrahydroindolizines by annulating 2a with β-ketoesters,
ketones, or amides (Scheme 1D).
Table 2. Scope of 1,3-Dicarbonyl Compounds 3
We initiated our studies by examining the annulation reaction
of 2-formylpiperidine hydrochloride (2a) and ethyl acetoacetate
(3a) to form 2,3-disubstituted 5,6,7,8-tetrahydroindolizine 1a
(Table 1).10 Under the best conditions described in Table 1, the
Table 1. Effect of Reaction Parameters on Annulation of 2a
a
and 3a to Generate Tetrahydroindolizine 1a
b
c
entry variation from the “standard” conditions conv (%) yield (%)
d
1
none
94
90
92
0
80(73)
61
2
23 °C instead of 0−5 °C
no 4 Å molecular sieves
K2CO3 as base
3
74
4
0
5
Et3N as base
85
100
100
100
93
75
90
85
87
<5
55
6
piperidine as base
azepane as base
CH2Cl2 as solvent
2-MeTHF as solvent
EtOH as solvent
PhMe as solvent
2 equiv of pyrrolidine
1 equiv of 2a
7
69
8
59
9
79
10
11
12
13
29
28
46
50
a
Reactions were performed using 3a (2.0 mmol, 260 mg) in solvent
b
(3.9 mL, 15 mL/g) in 15 mL vials under N2 for 2 h. Determined by
HPLC analysis. Assay yields were obtained by quantitative HPLC
analysis. Isolated yield.
c
d
reaction afforded 94% conversion in 80% HPLC assay yield and
73% isolated yield by employing 1.4 equiv of 2a, 2 mmol of 3a, 4
equiv of pyrrolidine as the base, and 150 wt % of 4 Å molecular
sieves as the dehydrating agent in THF at 0−5 °C for 2 h (Table
1, entry 1). Performing the reaction at 23 °C or in the absence of
molecular sieves generated slightly lower assay yields (Table 1,
entries 2−3). The reactions carried out using inorganic base
K2CO3 or organic base Et3N resulted in much inferior conversion
or assay yield (Table 1, entries 4−5). The use of cyclic secondary
amine bases such as piperidine and azepane gave quantitative
conversion, albeit in lower assay yields (Table 1, entries 6−7). A
screening of solvents such as dichloromethane, 2-methyltetrahy-
drofuran, ethanol, and toluene did not afford any advantage over
the solvent of choice THF (Table 1, entries 8−11). Finally,
reduction of the stoichiometry of pyrrolidine base from 4 equiv
to 2 equiv or that of compound 2a from 1.4 equiv to 1 equiv both
resulted in decreased conversion and assay yields (Table 1,
entries 12−13).
With a set of optimized conditions in hand, we next examined
the scope and limitations of this annulation reaction by reacting
2-formylpiperidine hydrochloride (2a) with various 1,3-
dicarbonyl compounds (Table 2). A cyclopropyl group can be
readily incorporated in the 3-position of tetrahydroindolizine 1b
in 50% yield when ethyl 3-cyclopropyl-3-oxopropanoate (3b)
was employed (Table 2, entry 2). Disappointingly, a sterically
bulky tert-butyl substituted β-ketoester 3c only generated 12%
yield of product 1c, indicating that this annulation reaction is
sensitive to the steric property of the 1,3-dicarbonyl compounds
(Table 2, entry 3). Interestingly, β-ketoesters substituted with an
a
Reactions were performed using 2.0 mmol of 3, 1.4 equiv of 2a, 4
equiv of pyrrolidine, 150 wt % of 4 Å MS in THF (15 mL/g) in 15 mL
b
c
vials under N2 at 0−5 °C for 2 h. Isolated yield. Reaction was
performed using 4.0 mmol of 3m. The annulation product was
contaminated with an unknown impurity after silica gel column
chromatography. Therefore, the reaction mixture was directly carried
to N-Boc deprotection and 1m was isolated.
aryl group, whether electron-neutral (3d), rich (3e), or deficient
(3f), all produced the desired tetrahydroindolizines 1d−f in
about 40% yield (Table 2, entries 4−6). Pyridine-substituted β-
ketoester 3g also gave the intended product 1g in a moderate
41% yield (Table 2, entry 7). The relatively low yields of these
B
DOI: 10.1021/acs.orglett.5b01671
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
reactions could be attributed to possible competitive decom-
position of the free 2-formylpiperidine under the reaction
conditions as the electrophilicity and corresponding reactivity of
the carbonyl group decreased. In fact, the reactions did not
proceed further after 2 h with starting materials 3d−g remaining.
Gratifyingly, besides β-ketoesters, 1,3-diketones can also
participate in the annulation chemistry and produce tetrahy-
droindolizine derivatives in good yields. For example, acyclic 1,3-
diketones pentane-2,4-dione (3h) and 1-phenylbutane-1,3-dione
(3i) reacted with 2-formylpiperidine hydrochloride (2a),
forming tetrahydroindolizines 1h and 1i in 60% and 76% yields,
respectively (Table 2, entries 8−9). It is worth noting that in the
latter reaction mixture, only trace amount of the minor isomer
was observed by LCMS analysis. Cyclic 1,3-diketones, such as
cyclohexane-1,3-dione (3j), 5,5-dimethylcyclohexane-1,3-dione
(3k), and cycloheptane-1,3-dione (3l), also produced the desired
products 1j−l in 58%, 71%, and 39% yields (Table 2, entries 10−
12).11 In addition, N-Boc protected piperidine-2,4-dione 3m
underwent the annulation reaction with 2-formylpiperidine
hydrochloride (2a) and afforded the heterotricyclic product
1m in 62% yield after direct N-Boc deprotection (Table 2, entry
13).12 Similarly, p-methoxybenzyl (PMB) protected piperidine-
2,4-dione 3n and N-Boc protected 5-ethylpiperidine-2,4-dione
3o generated products 1n and 1o in 68% and 61% yields,
respectively (Table 2, entries 14−15).
Figure 2. Unproductive 1,3-dicarbonyl compounds.
and 3v), or higher activation energy arising from five−five ring
strain in the transition states of the reaction (for 3s and 3t).
Finally, the scalability of this annulation chemistry was
successfully demonstrated using 2-formylpiperidine hydro-
chloride (2a) and N-Boc piperidine-2,4-dione (3m) at 5.0 g
scale. The reaction went smoothly and provided the desired
tricyclic compound 1m in a slightly lower 53% yield than the
small scale (4.0 mmol) reaction after direct Boc deprotection (eq
1).
It is noteworthy that, besides 2,3-disubstituted tetrahydroin-
dolizines, 1,2,3-trisubstituted tetrahydroindolizine could also be
attained by this annulation chemistry when a 2-piperidinyl
ketone was employed. For example, 2-piperidinylethanone (2b)
reacted with ethyl acetoacetate (3a) under our standard
conditions, affording 58% yield of the desired 1,2,3-trisubstituted
tetrahydroindolizine 1p (Scheme 2). Thus, one can readily
We propose a mechanism for this annulation process
exemplified by the reaction of 2-formylpiperidine hydrochloride
(2a) and ethyl acetylacetate (3a) (Scheme 3). It is believed that
Scheme 3. Proposed Mechanism
Scheme 2. Synthesis of Highly Substituted Pyrroles
envision that a wide spectrum of 1,2,3-trisubstituted tetrahy-
droindolizines can be achieved in a single step from 2-piperidinyl
ketones and 1,3-dicarbonyl compounds. Furthermore, 1,2,3,4-
tetrasubstituted pyrrole 1q could also be prepared using N-
methylaminoacetone (2c) and ethyl acetoacetate (3a), albeit in a
low yield of 21% (Scheme 2).13
A few observed limitations to this annulation reaction included
the use of certain 1,3-dicarbonyl compounds shown in Figure 2.
Examples included ethyl trifluoroacetoacetate (3p), β-ketoa-
mides 3q and 3r, 5-membered cyclic 1,3-dicarbonyl compounds
3s and 3t, as well as unprotected piperidine-2,4-dione (3u) and
1-phenylquinoline-2,4-dione (3v), which all gave <10% product
by HPLC and LCMS analysis, even after prolonged reaction time
and/or elevated reaction temperature. We attributed the lack of
success for these reactions to decomposition/polymerization of
the reactant under the standard reaction conditions (for 3p and
3u), decreased electrophilicity of the 1,3-dicarbonyl (for 3q, 3r
compounds 3a and 2a, upon treatment with pyrrolidine and the
dehydrating reagent molecular sieves, gave rise to enamine 414
and iminium species 5, respectively. In fact, enamine 4 was
observed by HPLC and LCMS in the reaction mixture. The
1
formation of enamine 4 was further confirmed by H NMR
spectroscopy when the reaction was performed in THF-d8
solvent under the standard reaction conditions absent of
compound 2a. Similarly, treatment of 2-formylpiperidine
hydrochloride (2a) with pyrrolidine (4 equiv) and molecular
C
DOI: 10.1021/acs.orglett.5b01671
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(4) Snoeck, R.; Andrei, G.; Bodaghi, B.; Lagneaux, L.; Daelemans, D.;
de Clercq, E.; Neyts, J.; Schols, D.; Naesens, L.; Michelson, S.; Bron, D.;
Otto, M. J.; Bousseau, A.; Nemecek, C.; Roy, C. Antiviral Res. 2002, 55,
413−424.
(5) (a) Harju, K.; Manevski, N.; Yli-Kauhaluoma, J. Tetrahedron 2009,
65, 9702−9706. (b) Dalla Croce, P.; La Rosa, C. Heterocycles 2001, 55,
1843−1858. (c) Nayyar, N. K.; Hutchison, D. R.; Martinelli, M. J. J. Org.
Chem. 1997, 62, 982−991.
(6) Teodoro, B. V. M; Correia, J. T. M.; Coelho, F. J. Org. Chem. 2015,
80, 2529−2538.
(7) Biletzki, T.; Imhof, W. Eur. J. Org. Chem. 2012, 6513−6516.
(8) (a) Bowie, A. L., Jr.; Trauner, D. J. Org. Chem. 2009, 74, 1581−
1586. (b) Ghosh, S. K.; Buchanan, G. S.; Long, Q. A.; Wei, Y.; Al-Rashid,
Z. F.; Sklenicka, H. M.; Hsung, R. P. Tetrahedron 2008, 64, 883−893.
(c) Rocchiccioli, S.; Settambolo, R.; Lazzaroni, R. J. Organomet. Chem.
2005, 690, 1866−1870. (d) Marchalin, S.; Cvopova, K.; Pham-Huu, D.-
P.; Chudik, M.; Kozisek, J.; Svoboda, I.; Daich, A. Tetrahedron Lett.
2001, 42, 5663−5667. (e) Lehmann, T.; Gmeiner, P. Heterocycles 2000,
53, 1371−1378. (f) Wang, M. D.; Alper, H. Tetrahedron Lett. 1995, 36,
6855−6858. (g) Gmeiner, P.; Lerche, H. Heterocycles 1990, 31, 9−12.
(h) Tanis, S. P.; Raggon, J. W. J. Org. Chem. 1987, 52, 819−827.
(9) (a) Huisgen, R.; Gotthardt, H.; Bayer, H. O. Angew. Chem., Int. Ed.
Engl. 1964, 3, 135−136. (b) Huisgen, R.; Gotthardt, H.; Bayer, H. O.;
Schaefer, F. C. Chem. Ber. 1970, 103, 2611−2624. (c) Santiago, B.;
Dalton, C. R.; Huber, E. W.; Kane, J. M. J. Org. Chem. 1995, 60, 4947−
4950. (d) Gribble, G. W. In Oxazoles: Synthesis, Reactions, and
Spectroscopy, A; Palmer, D. C., Ed.; Wiley: New York, 2003; Vol. 60,
pp 473−576.
sieves in THF-d8 also showed the disappearance of the
characteristic aldehyde peak. However, the mixture was too
complex on 1H NMR to conclusively identify iminium species 5.
We reasoned that the complex 1H NMR of the mixture could be a
result of equilibration of iminium species 5 and enamine 5′, and/
or potential decomposition of 5 and 5′. Intermediates 4 and 5
then react to generate isomeric iminium species 6a and 6b.
Under the reaction condition, species 6b converts to 6a, which
cyclizes to afford the bicyclic intermediate 7. Intermediate 7 then
extrudes pyrrolidine to produce the desired 5,6,7,8-tetrahy-
droindolizine 1a.
In conclusion, we have developed an expedient single-step
synthesis of 5,6,7,8-tetrahydroindolizines via the annulation of
commercially available 2-formylpiperidine hydrochloride and
1,3-dicarbonyl compounds under very mild conditions. A variety
of β-ketoesters, ketones, and amides participated in this
annulation chemistry, affording the desired 5,6,7,8-tetrahydroin-
dolizines in moderate to good yields. In addition, the formation
of 1,2,3-trisubstituted 5,6,7,8-tetrahydroindolizine and 1,2,3,4-
tetrasubstituted pyrrole was also demonstrated using 2-
piperidinylethanone and N-methylaminoacetone. We anticipate
that this practical method will provide rapid access to useful
quantities of versatile 5,6,7,8-tetrahydroindolizines.
ASSOCIATED CONTENT
■
S
* Supporting Information
(10) Tetrahydroindolizine 1a was previously synthesized in 12% yield
by the condensation of pyridine-2-carboxaldehyde and ethyl acetoace-
tate (3a), followed by hydrogenation; see: Sprake, J. M.; Watson, K. D. J.
Chem. Soc., Perkin Trans. 1 1976, 5−8.
General experimental procedures, characterization of new
1
compounds, and copies of H NMR and 13C NMR spectra.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.5b01671.
(11) Compound 1j was previously prepared by three different methods
with each involving a multiple-step synthesis; see: refs 5c and 6, and
Zheng, Z.; Tu, H.; Zhang, L. Chem. - Eur. J. 2014, 20, 2445−2448.
(12) For a previous multistep synthesis, see: Crawford, J. J.; Ortwine,
D. F.; Wei, B.; Young, W. B. PCT Int. Appl. 2013067274, 10 May 2013.
(13) For a previous synthesis, see: Grob, C. A.; Camenisch, K. Helv.
Chim. Acta 1953, 36, 49−58.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: zhang.haiming@gene.com.
Notes
(14) (a) McMurry, J. E. Org. Synth. 1973, 53, 59−62. (b) Gravel, D.;
Labelle, M. Can. J. Chem. 1985, 63, 1874−1883.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors would like to thank Prof. Scott Denmark (University
of Illinois at Urbana−Champaign), Drs. Chong Han and Alex
Andrus (Genentech, Inc.) for helpful discussion, Dr. Christine
Gu (Genentech, Inc.) for collecting HRMS data, and Drs. Francis
Gosselin, Diane Carrera, and Lauren Sirois (Genentech, Inc.) for
proofreading the manuscript.
REFERENCES
(1) (a) Baudoin, O.; Guen
■
́
ard, D.; Guer
́
itte, F. Mini-Rev. Org. Chem.
2004, 1, 333−341. (b) Linde, H. H. A. Helv. Chim. Acta 1965, 48, 1822−
1842. (c) Abraham, D. J.; Rosenstein, R. D.; Lyon, R. L.; Fong, H. H. S.
Tetrahedron Lett. 1972, 13, 909−912. (d) Gerasimenko, I.; Sheludko, Y.;
Stockigt, J. J. Nat. Prod. 2001, 64, 114−116. (e) Kam, T.-S.;
̈
Subramaniam, G.; Chen, W. Phytochemistry 1999, 51, 159−169.
́ ́
(f) David, B.; Sevenet, T.; Morgat, M.; Guenard, D.; Moisand, A.;
Tollon, Y.; Thoison, O.; Wright, M. Cell Motil. Cytoskeleton 1994, 28,
317−326. (g) Sugimoto, K.; Toyoshima, K.; Nonaka, S.; Kotaki, K.;
Ueda, H.; Tokuyama, H. Angew. Chem., Int. Ed. 2013, 52, 7168−7171.
(2) (a) Schroder, F.; Franke, S.; Francke, W.; Baumann, H.; Kaib, M.;
̈
Pasteels, J. M.; Daloze, D. Tetrahedron 1996, 52, 13539−13546.
(b) Movassaghi, M.; Ondrus, A. E. Org. Lett. 2005, 7, 4423−4426.
(3) (a) Sun, L.-R.; Li, X.; Wang, S.-X. J. Asian Nat. Prod. Res. 2005, 7,
127−130. (b) Dinsmore, A.; Mandy, K.; Michael, J. P. Org. Biomol.
Chem. 2006, 4, 1032−1037.
D
DOI: 10.1021/acs.orglett.5b01671
Org. Lett. XXXX, XXX, XXX−XXX