A convenient allenoate-based synthesis of 2-quinolin-2-yl malonates and β-ketoesters

Article abstract of DOI:10.1021/ol502554e

N-Protected o-aminobenzaldehydes smoothly react with α,γ-dialkylallenoates under Bronsted basic conditions to yield 2,3-disubstituted quinolines. This three-step reaction cascade of Michael addition, aldol condensation, and 1,3-N → C rearrangement uses the complete protecting group as a building block in a highly efficient C,C-bond formation of a new all-carbon quaternary center. Carbamate protected substrates (N-Boc, N-Cbz, N-Alloc) thus give 2-quinolin-2-yl-malonates, while amide protected substrates (N-Ac, N-Bz) afford 2-quinolin-2-yl-β-ketoesters in high yields.

Full text of DOI:10.1021/ol502554e

Letter
pubs.acs.org/OrgLett
A Convenient Allenoate-Based Synthesis of 2‑Quinolin-2-yl
Malonates and β‑Ketoesters
Philipp Selig*,† and William Raven‡
†Institute of Organic Chemistry, RWTH Aachen University, Landoltweg 1, 52074 Aachen, Germany
‡Institute of Inorganic Chemistry, RWTH Aachen University, Landoltweg 1, 52074 Aachen, Germany
S
* Supporting Information
ABSTRACT: N-Protected o-aminobenzaldehydes smoothly
react with α,γ-dialkylallenoates under Brønsted basic con-
ditions to yield 2,3-disubstituted quinolines. This three-step
reaction cascade of Michael addition, aldol condensation, and
1,3-N → C rearrangement uses the complete protecting group
as a building block in a highly efficient C,C-bond formation of
a new all-carbon quaternary center. Carbamate protected
substrates (N-Boc, N-Cbz, N-Alloc) thus give 2-quinolin-2-yl-malonates, while amide protected substrates (N-Ac, N-Bz) afford 2-
quinolin-2-yl-β-ketoesters in high yields.
he quinoline ring system is an important heteroaromatic
formation of these two products can be rationalized as follows:
Tmoiety which is present in a wide variety of bioactive and N-Boc protection increases the acidity of the NH-proton
pharmaceutically useful substances,1 such as the Cinchona
sufficiently to allow for deprotonation by K2CO3 and formation
of a highly nucleophilic amide anion. Aza-Michael addition of
this anion to allenoate 2 results in intermediate A, which
selectively cyclizes at the γ-position to give the bicyclic
intermediate B. Intermediate B itself was too unstable to be
isolated after aqueous workup and column chromatography. N-
Boc deprotection and aromatization/tautomerization quickly
result in the formation of 2-quinolin-2-yl-propanoate 3. While
the formation of 3 can thus be explained quite intuitively, the
formation of malonic ester 4a was much more surprising at
first. Here, the former N-Boc protecting group has obviously
been transferred to the adjacent nonaromatic position with the
concomitant formation of a new all-carbon quaternary center.
The N-Boc group thus not only activates the aminoaldehyde
substrate for nucleophilic attack but also plays a rather unusual
role as a building block in C,C-bond formation. This protecting
group transfer can be explained easily; however, if intermediate
B is considered in its mesomeric form B′. Aromatization
converts the quinolinyl ring into a cationic leaving group, thus
activating the Boc-group for nucleophilic attack, and simulta-
neously provides a stabilized anionic carbon center in direct
proximity. Despite the usually unreactive nature of the N-Boc
group, this special situation favors an efficient 1,3-N → C acyl
transfer to product 4a.13
alkaloids, the cytotoxic alkaloid camptothecin, quinoline
antibiotics,2 or the new antituberculosis drug Bedaquiline.3 Of
the many classical syntheses of quinolines, the Friedlander
̈
condensation between an aromatic 2-amino carbonyl and an
active methylene compound is still the method of choice for the
synthesis of quinolines substituted on the pyridine ring,4 and
improvements of this methodology continue to be developed
up to this day.5 While a number of related quinoline syntheses
have been developed by variation of the amino component
relatively early on (e.g., the Niementowski and Pfitzinger
syntheses),6 little work has been devoted to variation of the
active methylene compound. Today, alkynes represent the only
major alternative to enolizable carbonyls.7
Following up on our work on α,γ-dialkylallenoate esters as
versatile building blocks in the synthesis of heterocycles,8 we
became interested in the possible application of allenoates for
quinoline synthesis. Allenes, just like isomeric alkynes, can
formally be considered as dehydrated carbonyl compounds.9 α-
Allenic esters exhibit strong electrophilicity on the central β-
carbon and are latent nucleophiles at the α- and γ-position. A
Friedlander-type quinoline synthesis with o-aminobenzaldehyde
̈
and allenoates would therefore proceed as an aza-Michael
addition−aldol condensation sequence.10 A related allenoate-
based approach to the chromane skeleton with o-hydrox-
ybenzaldehyde has in fact already been investigated in detail.11
We began our investigations with the reaction of free o-
aminobenzaldehyde (1) and α,γ-dimethylallenoate 2, but to our
disappointment, no reaction could be observed under a variety
of conditions.12 N-Boc protected o-aminobenzaldehyde (1a),
on the other hand, smoothly reacted with allenoate 2 under
Brønsted basic conditions at rt to give two different quinoline
products 3 and 4a in various proportions (see Scheme 1). The
The relative amounts of the two isolable products 3 and 4a
were strongly dependent on the reaction conditions. Heating of
the reaction mixture both drastically reduced reaction times and
increased the selectivity for the formation of rearranged product
4a. After a short optimization (see Supporting Information), we
identified DMSO as the optimal solvent, freshly ground K2CO3
Received: August 29, 2014
Published: September 12, 2014
© 2014 American Chemical Society
5192
dx.doi.org/10.1021/ol502554e | Org. Lett. 2014, 16, 5192−5195

Organic Letters
Scheme 1. Overview of the Allenoate-Based Friedlander Quinoline Synthesis
Letter
̈
i
t
as a cheap and efficient Brønsted base, and 100 °C as the
optimum temperature, which allowed for the isolation of 4a in
72% yield (Table 1, entry 1).
were all well tolerated. Branched substituents Pr and Bu
increased the required reaction times from a few minutes to
t
several hours, but even with the extremely bulky Bu-allenoate
10, product 16a bearing two directly adjacent quaternary
centers on the quinoline ring, was still available in 69% yield.
N-Cbz and N-Alloc protected o-aminobenzaldehyde (1b, 1c)
reacted in the very same fashion and gave benzyl ethyl
malonates 4b, 11b−16b (Table 1, entries 8−14) as well as allyl
ethyl malonates 4c, 11c−16c (Table 1, entries 15−21) in good
to excellent yields. The reactions were very clean, and only
occasionally could small amounts of protecting group free
quinolines (cf. 3) be observed as side products. All three widely
used carbamate protecting groups N-Boc, N-Cbz, and N-Alloc
were thus readily employed as O-protected malonate building
blocks.
We next investigated a number of readily available α,γ-
dialkylallenoates 5−10 in the reaction with 1a and were pleased
to find that all expected tert-butyl ethyl malonates 11a−16a
were formed in good yields without any further optimization
(Table 1, entries 2−7). α-Me, α-nPr, α-Bn, γ-Me, γ-nBu, γ-Bn,
branched γ-iPr, and even γ-tBu substituents on the allenoate
Table 1. Synthesis of 2-Quinolin-2-yl-malonates from
a
Carbamate Protected o-Aminobenzaldehydes
In light of these results, we became interested if other
noncarbamateN-protecting groups were suitable for the 1,3-
N → C rearrangement as well, and N-acetyl and N-benzoyl
protected o-aminobenzaldehyde (1d, 1e) were chosen as
additional substrates (see Table 2). Again, the reactions
proceeded very smoothly. Transfer of the N-Ac protecting
group from 1d gave rise to aliphatic β-ketoesters 4d, 11d−16d
(Table 2, entries 1−7), while transfer of the N-Bz protecting
group from 1e afforded aromatic β-ketoesters 4e, 11e−16e
(Table 2, entries 8−14) in high yields. Only the most
b
entry substrates
PG
Rα
Rγ
R
product yield [%]
1
2
3
4
5
6
7
1a
1a
1a
1a
1a
1a
1a
2
Boc
Boc
Boc
Boc
Boc
Boc
Boc
Me
nPr
Bn
Me
Me
Me
nBu
Bn
iPr
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
4a
72
71
68
83
79
78
69
5
11a
12a
13a
14a
15a
16a
6
7
Me
Me
Me
Me
8
t
9
unreactive Bu allenoate 10 gave comparably low yields of β-
10
ketoesters 16d/e. As amides 1d/e are more labile to basic
hydrolysis than carbamates 1a−c, prolonged reaction times >1
h lead to N-deprotection and deactivation of the substrates.
The reactions were readily scalable under the same
conditions to give excellent yields and gram-amounts of
products 13b/e (Tables 1 and 2, entry 11). Our newly found
allenoate based quinoline synthesis thus offers a convenient
c
8
9
1b
1b
1b
1b
1b
1b
1b
2
Cbz
Cbz
Cbz
Cbz
Cbz
Cbz
Cbz
Me
nPr
Bn
Me
Me
Me
nBu
Bn
iPr
tBu
Bn
Bn
Bn
Bn
Bn
Bu
Bn
4b
85
97
82
88
92
60
59
5
11b
12b
13b
14b
15b
16b
10
11
12
13
14
6
d
7
Me
Me
Me
Me
8
alternative to the classical Friedlander methodology, especially
̈
9
for the formation of sterically congested products which are
difficult to access with conventional synthetic methods, and
often require special, prefunctionalized substrates or multistep
syntheses.14
10
15
16
17
18
19
20
21
1c
1c
1c
1c
1c
1c
1c
2
Alloc
Alloc
Alloc
Alloc
Alloc
Alloc
Alloc
Me
nPr
Bn
Me
Me
Me
nBu
Bn
iPr
tBu
allyl
allyl
allyl
allyl
allyl
allyl
allyl
4c
88
92
92
87
94
61
68
5
11c
12c
13c
14c
15c
16c
The structure of the rearranged products was unambiguously
confirmed by X-ray crystal diffraction of crystalline β-ketoester
15e (Figure 1).
The now easy availability of quinolinyl malonates and β-
ketoesters prompted us to explore the potential for further
functional group manipulations in these products (Figure 2).
Benzyl malonates 13b and 16b were readily converted to their
corresponding quinolinyl-2-propanoates 17 and 18 in excellent
yields by hydrogenolysis and spontaneous decarboxylation.
Despite this known disposition for decarboxylation,14c manip-
6
7
Me
Me
Me
Me
8
9
10
a
Reactions were carried out on a 1.0 mmol (1a) or 0.5 mmol (1b/c)
scale, using 1.25 equiv of allenoate and 1.0 equiv of K2CO3. Yields of
isolated products. Reaction with 0.36 mmol of 1b. 94% (1.97 g) 13b
b
c
d
on a 5.0 mmol scale.
5193
dx.doi.org/10.1021/ol502554e | Org. Lett. 2014, 16, 5192−5195

Organic Letters
Letter
ulations of the malonate or β-ketoester moiety which keep the
quaternary center intact were also possible. Reduction of β-
ketoester 4d with NaBH4 afforded 51% of diastereomerically
pure alcohol 19, and LiAlH4 reduction of malonate 13b gave
the 1,3-diol 20 in 40% unoptimized yield. Finally, we could
show that 3-Me substituted products such as 4a were easily
elaborated for further reactions on the benzylic position by
straightforward radical bromination with NBS/AIBN.
In conclusion, we developed a quick and convenient method
for the synthesis of densely substituted 2-quinolin-2-yl
malonates and β-ketoesters from N-protected o-aminobenzal-
dehydes and α,γ-dialkylallenoates. The reaction features a
highly effective, aromatization-initiated 1,3-N → C rearrange-
ment of carbamate or amide protecting groups with the
formation of a new quaternary carbon center. The reaction
succeeds with high yields and under very simple conditions
with Boc, Cbz, Alloc, Ac, and Bz protecting groups.
Investigations regarding the functional group transfer of other
N-protecting groups are currently ongoing in our laboratory.
Table 2. Synthesis of 2-Quinolin-2-yl-β-ketoesters from
Acetyl and Benzoyl Protected o-Aminobenzaldehydes
a
b
entry
substrates
PG
Rα
Rγ
R
product yield [%]
1
2
3
4
5
6
7
1d
1d
1d
1d
1d
1d
1d
2
Ac
Ac
Ac
Ac
Ac
Ac
Ac
Me
nPr
Bn
Me
Me
Me
nBu
Bn
Me
Me
Me
Me
Me
Me
Me
4d
88
91
79
84
86
69
21
5
11d
12d
13d
14d
15d
16d
6
7
Me
Me
Me
Me
8
9
iPr
10
tBu
8
1e
1e
1e
1e
1e
1e
1e
2
Bz
Bz
Bz
Bz
Bz
Bz
Bz
Me
nPr
Bn
Me
Me
Me
nBu
Bn
Ph
Ph
Ph
Ph
Ph
Ph
Ph
4e
96
97
83
90
95
90
21
9
5
11e
12e
13e
14e
15e
16e
10
11
12
13
14
6
ASSOCIATED CONTENT
* Supporting Information
c
7
Me
Me
Me
Me
■
S
8
9
iPr
Syntheses of starting materials, full synthetic procedures,
optimization data, details on X-ray crystal analysis, full analytical
10
tBu
1
data and copies of H and 13C NMR spectra of quinoline
a
Reactions were carried out on a 0.5 mmol scale, using 1.25 equiv of
b
c
products and derivatives. This material is available free of
charge via the Internet at http://pubs.acs.org.
allenoate and 1.0 equiv of K2CO3. Yields of isolated products. 97%
(1.90 g) 13e on a 5.0 mmol scale.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: Philipp.Selig@rwth-aachen.de.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the Fonds der Chemischen Industrie (FCI) for
financial support (Liebig-Scholarship to P.S.). Special thanks go
to Prof. Dr. D. Enders, Institute of Organic Chemistry, RWTH
Aachen University, for the generous provision of laboratory
space and chemicals.
Figure 1. X-ray crystal structure of β-ketoester 15e. Ellipsoids at 50%
probability.15
REFERENCES
■
(1) (a) Musiol, R.; Magdziarz, T.; Kurczyk, A. In Science against
microbial pathogens: Communicating current research and technological
́
advances; Mendez-Villas, A., Ed.; Formatex: Badajoz, 2011; Vol. 1, pp
72−83. (b) Afzal, O.; Kumar, S.; Haider, M. R.; Ali, M. R.; Kumar, R.;
Jaggi, M.; Bawa, S. Eur. J. Med. Chem. 2014, DOI: 10.1016/
j.ejmech.2014.07.044.
(2) Tatsuta, K.; Tamura, T. J. Antobiotics 2000, 53, 418−421.
(3) (a) Matteelli, A.; Carvalho, A. C. C.; Dooley, K. E.; Kritski, A.
Future Microbiol. 2010, 5, 849−858. (b) Diacon, A. H.; Pym, A.;
Grobusch, M.; Patientia, R.; Rustomjee, R.; Page-Shipp, L.; Pistorius,
C.; Krause, R.; Bogoshi, M.; Churchyard, G.; Venter, A.; Allen, J.;
Palomino, J. C.; De Marez, T.; van Heeswijk, R. P. G.; Lounis, N.;
Meyvisch, P.; Verbeeck, J.; Parys, W.; de Beule, K.; Andries, K.;
Neeley, D. F. M. New Engl. J. Med. 2009, 360, 2397−2405.
(4) (a) Friedlander, P. Chem. Ber. 1882, 15, 2572−2575.
̈
a
Figure 2. Derivatization of quinoline products: From 13b/16b with
(b) Muchowski, J. M.; Maddox, M. L. Can. J. Chem. 2004, 82, 461−
478.
b
c
d
H2, Pd/C; From 4d with NaBH4; From 13b with LiAlH4; From
4a with NBS, AIBN. All yields are of isolated products and are
unoptimized.
́
(5) Review: (a) Marco-Contelles, J.; Perez-Mayoral, E.; Samadi, A.;
Carreiras, M. d. C.; Soriano, E. Chem. Rev. 2009, 109, 2652−2671.
Recent examples: (b) Bandyopadhyay, P.; Prasad, G. K.; Sathe, M.;
Sharma, P.; Kumar, A.; Kaushik, M. P. RSC Adv. 2014, 4, 6638−6645.
5194
dx.doi.org/10.1021/ol502554e | Org. Lett. 2014, 16, 5192−5195

Organic Letters
Letter
́ ́
(c) Banon-Caballero, A.; Guillena, G.; Najera, C. J. Org. Chem. 2013,
̃
78, 5349−5356. (d) Shen, Q.; Wang, L.; Yu, J.; Liu, M.; Qiu, J.; Fang,
L.; Guo, F.; Tang, J. Synthesis 2012, 44, 389−392. (e) Wu, J.; Xia, H.-
G.; Gao, K. Org. Biomol. Chem. 2006, 4, 126−129.
(6) Reviews: (a) Manske, R. H. Chem. Rev. 1942, 30, 113−144.
(b) Bergstrom, F. W. Chem. Rev. 1944, 35, 77−277.
(7) (a) Khong, S.; Kwon, O. J. Org. Chem. 2012, 77, 8257−8267.
(b) Patil, N. T.; Raut, V. S. J. Org. Chem. 2010, 75, 6961−6964.
(c) Mohammadpoor-Baltork, I.; Tangestaninejad, S.; Moghadam, M.;
Mirkhani, V.; Anvar, S.; Mirjafari, A. Synlett 2010, 3104−3112.
(8) (a) Selig, P.; Turock
2930−2932. (b) Selig, P.; Turock
2535−2539. (c) Selig, P.; Turock
2013, 355, 297−302.
̌
in, A.; Raven, W. Chem. Commun. 2013, 49,
̌
in, A.; Raven, W. Synlett 2013, 24,
in, A.; Raven, W. Adv. Synth. Catal.
̌
(9) Zhu, X.-F.; Henry, C. E.; Wang, J.; Dudding, T.; Kwon, O. Org.
Lett. 2005, 7, 1387−1390.
(10) (a) Wagner, A. M.; Knezevic, C. E.; Wall, J. L.; Sun, V. L.; Buss,
J. A.; Allen, L. T.; Wenzel, A. G. Tetrahedron Lett. 2012, 53, 833−836.
(b) Yamazaki, S.; Takebayashi, M.; Miyazaki, K. J. Org. Chem. 2010,
75, 1188−1196. (c) Makino, K.; Hara, O.; Takiguchi, Y.; Katano, T.;
Asakawa, Y.; Hatano, K.; Hamada, Y. Tetrahedron Lett. 2003, 44,
8925−8929.
(11) (a) Sun, Y.-W.; Guan, X.-Y.; Shi, M. Org. Lett. 2010, 12, 5664−
5667. (b) Dai, L.-Z.; Shi, Y.-L.; Zhao, G.-L.; Shi, M. Chem.−Eur. J.
2007, 13, 3701−3706. (c) Shi, M.; Dai, L.-Z.; Shi, Y.-L.; Zhao, G.-L.
Adv. Synth. Catal. 2006, 348, 967−972. (d) Zhao, G.-L.; Shi, Y.-L.; Shi,
M. Org. Lett. 2005, 7, 4527−4530.
(12) Free o-aminobenzaldehyde has been reported to react with
highly activated allene-1,3-dicarboxylic esters only: (a) Tamara, Y.;
Tsugoshi, T.; Mohri, S.-i.; Kita, Y. J. Org. Chem. 1985, 50, 1542−1544.
(b) Nixon, N. S.; Scheinmann, F.; Suschitzky, J. L. Tetrahedron Lett.
1983, 24, 597−600.
(13) 1,3-N → C shifts have been reported previously for acyl groups,
but not for carbamate groups: (a) Hara, O.; Ito, M.; Hamada, Y.
Tetrahedron Lett. 1998, 39, 5537−5540. (b) Trapani, G.; Reho, A.;
Latrofa, A.; Marlacchi, F.; Liso, G. J. Chem. Res. (S) 1986, 96−97.
(c) Liso, G.; Trapani, G.; Reho, A.; Latrofa, A. Tetrahedron Lett. 1981,
22, 1641−1644. (d) Akasaki, Y.; Ohno, A. J. Am. Chem. Soc. 1974, 96,
1957−1959.
(14) (a) Karatas,̧ B.; Aumann, R. Organometallics 2010, 29, 801−805.
(b) Suginome, M.; Fukuda, T.; Ito, Y. Org. Lett. 1999, 1, 1977−1979.
(c) Baty, J. D.; Jones, G.; Moore, C. J. Org. Chem. 1969, 34, 3295−
3302.
(15) (a) CCDC 1021178 contain the supplementary crystallographic
data for this paper. These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.
uk/data_request/cif. (b) Spek, A. L. Acta Crystallogr. 2009, D65, 148.
5195
dx.doi.org/10.1021/ol502554e | Org. Lett. 2014, 16, 5192−5195