Novel access to 2-substituted quinolin-4-ones by nickel boride-mediated reductive ring transformation of 5-(2-nitrophenyl)isoxazoles

Article abstract of DOI:10.1016/j.tetlet.2019.151327

Reductive ring transformation of 3-substituted 5-(2-nitrophenyl)isoxazoles, readily accessible via 1,3-dipolar cycloaddition of 2-ethinylnitrobenzene with nitrile oxides, opens a novel access to 2-substituted quinolin-4-ones. Nickel boride, generated in situ from nickel chloride and sodium borohydride, allows, via simultaneous reduction of the nitro group and reductive cleavage of the isoxazole ring, the one-step conversion into the target quinolin-4-ones. This protocol tolerates various functional groups, except olefins, and thus is complementary to the reductive ring transformation with iron/acetic acid, which predominantly tolerates olefins.

Full text of DOI:10.1016/j.tetlet.2019.151327

Journal Pre-proofs
Novel access to 2-substituted quinolin-4-ones by nickel boride-mediated reduc-
tive ring transformation of 5-(2-nitrophenyl)isoxazoles
Bernhard Lohrer, Franz Bracher
PII:
S0040-4039(19)31114-1
DOI:
https://doi.org/10.1016/j.tetlet.2019.151327
TETL 151327
Reference:
Tetrahedron Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
30 August 2019
22 October 2019
24 October 2019
Please cite this article as: Lohrer, B., Bracher, F., Novel access to 2-substituted quinolin-4-ones by nickel boride-
mediated reductive ring transformation of 5-(2-nitrophenyl)isoxazoles, Tetrahedron Letters (2019), doi: https://
doi.org/10.1016/j.tetlet.2019.151327
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will
undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing
this version to give early visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2019 Published by Elsevier Ltd.

Novel access to 2-substituted quinolin-4-ones by nickel boride-mediated reductive ring
transformation of 5-(2-nitrophenyl)isoxazoles
Bernhard Lohrer and Franz Bracher*
Ludwig-Maximilians University, Department of Pharmacy – Center for Drug Research, Butenandtstr.
5-13, 81377 Munich, Germany
*Corresponding author.
e-mail address: franz.bracher@cup.uni-muenchen.de (F. Bracher)
Keywords:
Quinolin-4-ones
Alkaloids
Isoxazoles
Ring transformation
GRAPHICAL ABSTRACT
N
[3+2] cycloaddition
with in situ-prepared
nitrile oxide R-CNO
O
O
NaBH4 (7 equ.)
NiCl2 (1 equ.)
R
NO2
NO2
N
H
R
Abstract
Reductive ring transformation of 3-substituted 5-(2-nitrophenyl)isoxazoles, readily accessible via 1,3-
dipolar cycloaddition of 2-ethinylnitrobenzene with nitriles oxides, opens a novel access to 2-
substituted quinolin-4-ones. Nickel boride, generated in situ from nickel chloride and sodium
borohydride, allows, via simultaneous reduction of the nitro group and reductive cleavage of the
isoxazole ring, the one-step conversion into the target quinolin-4-ones. This protocol tolerates various
functional groups, except olefins, and thus is complementary to the reductive ring transformation with
iron/acetic acid, which predominantly tolerates olefins.
Introduction
Quinolin-4-ones are a class of heterocyclic compounds with significant relevance in medicinal
chemistry. Substituted quinolin-4-one-3-carboxylic acids (topoisomerase inhibitors) like ciprofloxacin
are among the most important antibiotics [1], and the closely related elvitegravir is an important HIV
integrase inhibitor [2]. Further, a large number of bioactive quinolin-4-ones have been isolated from
plants and microorganisms, and antibacterial, antiplasmodial, and cytotoxic activities have been
reported for many of them [3] (Figure 1). Typically, these quinolin-4-one alkaloids bear an alkyl,
alkenyl or aryl residue at C-2, and some of them are further substituted at C-3. A number of 2-
substituted quinolin-4-ones are important quorum sensing signal molecules in Pseudomonas species,
which open a completely new option for the treatment of Pseudomonas infections [4]. Further,
naturally occuring quinolones served as lead structures for the development of antimycobacterial [5],
antimalarial [6] and antimitotic compounds for tumor therapy [7].

O
N
O
F
COOH
N
N
H
HN
Burkholone
Ciprofloxacin
O
O
N
H
R
N
OH
R = n-C3H7 (1)
R = n-C5H11 (2)
YM-30059
Figure 1. Selected quinolin-4-ones of pharmaceutical interest: The synthetic antibiotic ciprofloxacin
[1] and representative quinolin-4-one alkaloids, burkholone [8], YM-30059 [9], 2-propylquinolin-4-
one (1) [10] and 2-pentylquinolin-4-one (2) [10-11].
Various effective methods have been developed for the large scale production of quinolone antibiotics
targeting topoisomerases [12]. The synthesis of quinolin-4-ones bearing alkyl or aryl residues at C-2
(and/or C-3) following established strategies like Conrad-Limpach synthesis (from anilines and β–
ketoesters), Camps reaction (cyclization of ortho-acylaminoacetophenones), Niementowski and
related reactions (from anthranilic acid-derived imines) has been thoroughly reviewed [12-13] (Figure
2).
Conrad-Limpach
Niementowski
Camps
O
O
OR''/NR''2
COOR'
O
N
N
N
R
base
R
R
H/R'
H/R'
heat
or acid
base
O
O
O
metal
cat.
reduction
NR2
R
N
R
X
NO2
X = NH2/NHR'
X = Br/I (+ amine)
H/R'
Leimgruber-Batcho
Figure 2. Established methods for the construction of the quinolin-4-one ring system.
Among the newer methods, those involving enaminoketone intermediates have attracted considerable
interest [14]. Whereas the modified Leimgruber-Batcho synthesis [15] and newer protocols involving
chelation-controlled C-H amidation reactions prior to cyclization with the enaminoketone [16] are
rather suited for the preparation of 2-unsubstituted quinolin-4-ones, several approaches starting from
acetylenic ketones, and partly going through enaminoketones (arising from addition of amines to the
alkyne) have been worked out in the recent years [17].
In continuation of our research on the construction of heterocyclic ring systems via enaminoketone
intermediates [18] we developed a convenient approach to canthin-4-ones, which can be regarded as

fused analogues of 5-aza-quinolin-4-ones, via cyclization of enaminoketones obtained through
reductive cleavage of isoxazoles [19]. Isoxazoles are convenient precursors for enaminoketones, as on
the one side numerous methods have been published for the construction of the isoxazole ring (most
prominent are [3+2] cycloadditions of alkynes with nitrile oxides and on the other side reductive
cleavage of isoxazoles to give the enaminoketones as versatile intermediates can be accomplished with
a broad spectrum of reducing agents [20]).
This prompted us to investigate a route through generation of enaminoketones from isoxazoles for the
synthesis of quinolin-4-ones. A literature search revealed that very few reactions of this type had been
published before: Yamanaka [21] reported on the synthesis of 2-methyl-quinolin-4-one via Raney
nickel-catalyzed reduction of 3-methyl-5-(2-nitrophenyl)isoxazole, and Kurth [22] prepared four aryl-
quinolin-4-ones from nitrophenyl isoxazoles through reduction with iron powder in acidic media, but
observed unequivocal outcome when using different acidic reagents. In both of these conversions,
reductive cleavage of the isoxazole is accompanied by reduction of the nitro group to give a primary
aromatic amine. So it was challenging to search for a reliable novel method for the synthesis of 2-
substituted quinolin-4-ones via reductive ring transformation of 5-(2-nitrophenyl)isoxazoles. Having
the most important chemotypes of natural 4-quonolones in mind, it appeared worthwhile to investigate
approaches to 2-aryl- and 2-alkyl-quinolin-4-ones, as well as analogues bearing unsaturated side
chains. Only a limited number of methods is available for the synthesis of quinolin-4-ones bearing
olefinic residues at C-2, and very few publications describe the synthesis of 2-(1-alkenyl)-quinolin-4-
ones other than styryl derivatives [5, 23].
Results and discusssion
For orienting evaluations we selected compound 6, bearing an unconjugated unsaturated side chain at
the isoxazole ring, as substrate for reductions. Isoxazole 6 was conveniently obtained from trans-hex-
3-ene aldoxime (3) via NCS-mediated conversion into the corresponding nitrile oxide 4 and [3+2]
cycloaddition with 2-ethinylnitrobenzene (5). Following Yamanaka’s protocol [21] we subjected
isoxazole 6 to Raney nickel-catalyzed hydrogenation at 5 bar, and obtained an inseparable about 2.6:1
mixture (determined by NMR) of the 2-substituted quinolin-4-one 7 and its saturated analogue,
alkaloid 2 (see Figure 1), in 80% yield. This rendered Raney nickel-catalyzed reductive ring
transformations poorly attractive. As an alternative reducing agent we investigated iron powder in
glacial acetic acid, and fortunately, the olefinic moiety was conserved under these conditions, and
alkenyl-quinolin-4-one 7 was obtained in 89% yield, with no indication for over-reduction (Scheme
1). This outcome is of special interest, since related allyl-type side-chains are found in the biologically
active quinolin-4-one natural products burkholone and YM-30059 (Figure 1), as well as some novel
antibacterial quinolin-4-ones from Tenacibaculum maritimum [24].
We further investigated the compatibility of iron-mediated reductive ring transformations with two
other olefinic substrates, the styryl isoxazole 8 and the 1-pentenyl analogue 9, both readily available
via cycloaddition of 2-ethinylnitrobenzene (5) with nitrile oxides derived from trans-cinnamic
aldehyde and trans-2-hexenal oximes, respectively. Whereas the styryl isoxazole 8 was converted into
quinolin-4-one 10 under conservation of the olefinic bond (50% yield), the attempted analogous
preparation of 2-pentenyl quinolin-4-one 11 gave an inseparable 1:1 mixture (determined by NMR
analysis) of the target compound and its saturated analogue, alkaloid 2 (about 80% yield) (Scheme 1).
So, for the synthesis of 2-(1-alkenyl)-4-quinolones our recently developed method using Horner-
Wadsworth-Emmons olefination of in situ-prepared phosphonates [23] appears more reliable.

NCS, Et3N
pyridine,
CHCl3
5
N
NO2
OH
CNO
3
4
N
O
O
Fe, AcOH
89%
N
H
NO2
6
7
N
O
O
R
Fe, AcOH
N
H
R
NO2
8
9
R = phenyl
R = n-C3H7
10
11
R = phenyl (50%)
2
R = n-C3H7 (1:1 mixture with
)
Scheme 1. Fe/AcOH-mediated reductive conversion of (2-nitrophenyl)isoxazoles into quinolin-4-
ones bearing unsaturated side chains.
These results indicate that iron-mediated reductive ring transformations of 5-(2-nitrophenyl)isoxazoles
has some potential for the synthesis of quinolin-4-ones bearing unsaturated side chains, but depending
on the nature of the olefinic side chain, over-reduction to give the (poorly separable) saturated
analogues is an obstacle of this protocol.
A literature search for alternative protocols for reductive ring cleavage of isoxazoles brought us to the
nickel(II) salt/sodium borohydride couple. It is well known that this couple generates nickel boride
(Ni2B) and hydrogen in situ, and diverse functional groups, among them nitroarenes, can be reduced
with this reagent system [25]. Only two reports in literature deal with reductive ring cleavage of
isoxazoles with nickel(II) salts/sodium borohydride: Koroleva et al. [26] claimed chemoselective
reductive cleavage using nickel(II) sulfate in a letter, and Oliver and Lusby described the ring cleavage
of a bicyclic isoxazole using nickel(II) chloride/sodium borohydride [27]. This prompted us to
investigate the underexplored nickel boride-mediated reduction of isoxazoles (combined with
reduction of the nitro group) for application in the synthesis of 2-substituted quinolin-4-ones. For this
purpose we selected the 3-phenylisoxazole 12 (obtained in the usual manner from benzaldoxime and
2-ethinylnitrobenzene 5) for optimization of the reaction conditions. This model compound was
regarded as representative, since undesired side-reactions like over-reduction were not expected here.
Surprisingly, first experiments using nickel(II) sulfate heptahydrate gave confusing results, so we
moved to nickel(II) chloride hexahydrate, and even first experiments using methanol-THF mixtures as
solvent were promising. We modified a couple of parameters (amounts of nickel salt and sodium
borohydride, reaction temperature and time; see Supporting Information, Table S1) and ended up with
a protocol giving the desired 2-phenyl-quinolin-4-one (23) in up to 82% yield. A considerable excess
of sodium borohydride was required, since with as few as 1.5 equivalents of reducing agent starting
material was completely consumed, but almost no conversion to the quinolin-4-one took place (in one
of these experiments, a hydroxylamine resulting from partial reduction of the nitro group was obtained
as the major product; see Supporting Information). With 3.5 equivalents of sodium borohydride yields
above 60% were obtained, and the optimized amount was 7 equivalents. Catalytic amounts (0.2
equivalents) of nickel salt, in combination with the mentioned excess of sodium borohydride) already
gave noteworthy conversion, but equimolar amounts of nickel were more effective. Slightly enhanced
yields were obtained upon performing the reaction in a closed vessel (82% compared to 79% in an

open vessel), but most likely loss of gaseous hydrogen (formed in the reaction of nickel(II) salt and
borohydride) is not a limiting factor, and mechanistically this reductive cleavage is not likely to be a
classical catalytic hydrogenation [28]. Further, only slight improved yields were obtained when
starting the reaction at -30 °C compared to ice-bath cooling.
In order to investigate the scope and limitations of this novel nickel boride-mediated approach to 2-
substituted quinolin-4-ones, we prepared (by standard alkyne-nitrile oxide cycloadditions) a collection
of additional 5-(2-nitrophenyl)isoxazoles 13-20 bearing alkyl, alkenyl, substituted aryl and heteroaryl
residues at C-3 of the isoxazole ring (yields 48-80%), and subjected them to the optimized conditions
for nickel boride-mediated reductive ring transformation. The results are presented in Table 1.
Table 1. 2-Substituted quinolin-4-ones, obtained by reductive ring transformation of 3-substituted 5-
(2-nitrophenyl)isoxazoles (isolated yields given in parentheses).
O
NH2
R
O
NaBH4 (7 equ.)
NiCl2·6 H2O (1 equ.)
N
O
R
MeOH, THF,
0 °C  RT
NH2
N
H
R
NO2
Entry
Starting
isoxazole
Entry
Product
Starting isoxazole
Product
O
1
7
N
O
O
O
N
O
OMe
N
H
N
H
NO2
OMe
NO2
O
12
21 (79%)
18
26 (64%)
O
2
3
8
9
O
N
O
N
O
S
S
N
N
H
NO2
NO2
H
13
19
1 (76%)
O
27 (82%)
O
N
O
N
O
N
F
N
H
N
H
N
NO2
NO2
F
14
20
28*
22 (74%)
O
O
4
5
10
11
N
O
N
O
Cl
N
H
N
H
NO2
NO2
Cl
15
8
6
29 (62%)
23 (74%)
O
N
O
O
N
O
OMe
N
H
NO2
N
H
OMe
NO2
2 (75 %)
16
24 (72 %)
O
6
N
O
OEt
N
H
NO2
OEt
17
25 (70%)
* detected by NMR analysis, but not isolated in pure form due to solubility issues

Alkaloid 2-propylquinolin-4-one (1, see Figure 1) was obtained in 76% yield, and 2-aryl analogues 22-
26, irrespective of the electronegativity of the substituents, were obtained in 64-74% yields. The
thienyl derivative 27 was obtained in 82% yield, and based on NMR data of the crude product, 3-
pyridyl analogue 28 was formed as well, but due to its extremely poor solubility this product could not
be isolated in sufficient purity. In both experiments involving alkenyl isoxazole precursors (6, 8),
clean reduction of the olefinic groups was observed, rendering phenylethyl compound 29 and alkaloid
2-pentylquinolin-4-one (2, see Figure 1) in good yields.
So the nickel boride-mediated reductive ring transformation of 5-(2-nitrophenyl)isoxazoles represents
a novel and effective approach to 2-substituted quinolin-4-ones. This reduction protocol is compatible
with a number of substituents located at aromatic rings (halogens, ethers, esters), further it is
compatible with thiophene rings, which are easily desulfurized by Raney nickel [29]. However,
conjugated and isolated olefinic bonds are gently reduced by this reagent. In this respect, the nickel
boride method is fairly complementary to the reduction with iron in acidic medium.
In conclusion we could demonstrate that the hitherto only rudimentarily investigated iron-mediated
reductive ring transformation of 5-(2-nitrophenyl)isoxazoles to 2-substituted quinolin-4-ones has some
potential applications in the synthesis of target compounds bearing olefinic residues, and further we
introduced nickel boride mediated transformations as a novel, cheap and easy-to-perform approach to
quinolin-4-ones.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the online version, at
https://doi.org/10.1016/xxxxx.
References
[1] T.D. Gootz, K.E. Brighty, Med. Res. Rev. 16 (1996) 433-486.
[2] E.D. Deeks, Drugs 74 (2014) 687-697.
[3] (a) S. Heeb, M.P. Fletcher, S.R. Chhabra, S.P. Diggle, P. Williams, M. Cámara, FEMS Microbiol.
Rev. 35 (2011) 247-274; (b) M.F. da Silva, M.S. Soares, J.B. Fernandes, P.C. Vieria, Alkaloids Chem.
Biol. 64 (2007) 139-214.
[4] S. Wagner, R. Sommer, S. Hinsberger, C. Lu, R.W. Hartmann, M. Empting, A. Titz, J. Med.
Chem. 59 (2016) 5929-5969.
[5] A.A. Wube, A. Hufner, C. Thomaschitz, M. Blunder, M. Kollroser, R. Bauer, F. Bucar, Bioorg.
Med. Chem. 19 (2011) 567-579.
[6] A. Monastyrskyi, D.E. Kyle, R. Manetsch, Curr. Top. Med. Chem. 14 (2014) 1693-1705.
[7] Y.H. Chang, J.S. Yang, S.C. Kuo, J.G. Chung, Anticancer Res. 29 (2009) 3139-3148.
[8] T. Mori, T. Yamashita, K. Furihata, K. Nagai, K. Suzuki, Y. Hayakawa, K. Shin-Ya, J. Antibiot.
60 (2007) 713-716.
[9] K. Kamigiri, T. Tokunaga, M. Shibazaki, B. Setiawan, M. Morioka, K.-I. Suzuki, R.M.
Rantiatmodjo, J. Antibiotics 49 (1996) 823-825.
[10] C. Agier, M. Bury, J. Aquette, R. Hocquemiller, P.G. Waterman, Nat. Prod. Res. 21 (2007) 698-
703.
[11] S.J. Wratten, M.S. Wolfe, R.J. Andersen, D.J. Faulkner, Antimicrob. Agents Chemother. 11
(1977) 411-414.
[12] A. Naeem, S.L. Badshah, M. Muska, N. Ahmad, K. Khan, Molecules 21 (2016) 268.
[13] A.A. Boteva, O.P. Krasnykh, Chem. Heterocycl. Comp. 45 (2009) 757.
[14] (a) D.V. Dar'in, P.S. Lobanov, Russ. Chem. Rev. 84 (2015) 601-633; (b) F. Bracher, Mini-Rev.
Org. Chem. (2019) in press; doi: 10.2174/1570193X16666181204122143.

[15] (a) J. Tois, M. Vahermo, A. Koskinen, Tetrahedron Lett. 46 (2005) 735-737; (b) J. Joussot, A.
Schoenfelder, L. Larquetoux, M. Nicolas, J. Suffert, G. Blond, Synthesis 48 (2016) 3364-3372.
[16] (a) F. Wang, L. Jin, L. Kong, X. Li, Org. Lett. 19 (2017) 1812-1815; (b) P. Shi, L. Wang, K.
Chen, J. Wang, J. Zhu, Org. Lett. 19 (2017) 2418-2421.
[17] (a) R. Bernini, S. Cacchi, G. Fabrizi, A. Sferrazza, Synthesis (2009) 1209-1219; (b) T. Zhao, B.
Xu, Org. Lett. 12 (2010) 212-215; (c) J. Shao, X. Huang, X. Hong, B. Liu, B. Xu, Synthesis 44 (2012)
1798-1808; (d) L. Åkerbladh, P. Nordeman, M. Wejdemar, L.R. Odell, M. Larhed, J. Org. Chem. 80
(2015) 1464-1471; (e) O. Seppänen, M. Muuronen, J. Helaja, Eur. J. Org. Chem. (2014) 4044-4052;
(f) Z. Zheng, Q. Tao, Y. Ao, M. Xu, Y. Li, Org. Lett. 20 (2018) 3907-3910; (g) X.-D. Fei, Z. Zhou, W.
Li, Y.-M. Zhu, J.-K. Shen, Eur. J. Org. Chem. (2012) 3001-3008; (h) V. O. Iaroshenko, S. Mkrtchyan,
A. Vill, Synthesis 45 (2013) 205-218.
[18] (a) F. Bracher, D. Hildebrand, Liebigs Annalen der Chemie (1993) 837-839; (b) B. Strödke, A.P.
Gehring, F. Bracher, Arch. Pharm. (Weinheim) 348 (2015) 125-131; (c) A. Puzik, F. Bracher, J.
Heterocycl. Chem. 46 (2009) 770-773.
[19] (a) T. Tremmel, F. Bracher, Tetrahedron 71 (2015) 4640-4646; (b) F. Bracher, T. Tremmel, Mini-
Rev. Org. Chem. 14 (2017) 92-98.
[20] (a) T. Morita, S. Yugandar, S. Fuse, H. Nakamura, Tetrahedron Lett. 59 (2018) 1159-1171; (b) F.
Hu, M. Szostak, Advanced Synthesis & Catalysis 357 (2015) 2583-2614.
[21] T. Sakamoto, Y. Kondo, D. Uchiyama, H. Yamanaka, Tetrahedron 47 (1991) 5111-5118.
[22] K.C. Coffman, T.A. Palazzo, T.P. Hartley, J.C. Fettinger, D.J. Tantillo, M.J. Kurth, Org. Lett. 15
(2013) 2062-2065.
[23] B. Lohrer, F. Bracher, Tetrahedron Lett. 59 (2018) 3632-3635.
[24] D. Li, N. Oku, A. Hasada, M. Shimizu, Y. Igarashi, Beilstein J. Org. Chem. 14 (2018) 1446-1451.
[25] B. Ganem, J.O. Osby, Chem. Rev. 86 (1986) 763-780.
[26] E.V. Koroleva, F.A. Lakhvich, T.V. Yankova, Chem. Heterocycl. Comp. 23 (1987) 1267-1267.
[27] Y. Gaoni, Tetrahedron Lett. 29 (1988) 1591-1594.
[28] J.O. Osby, B. Ganem, Tetrahedron Lett. 26 (1985) 6413-6416.
[29] (a) J. Rentner, M. Kljajic, L. Offner, R. Breinbauer, Tetrahedron 70 (2014) 8983-9027; (b) F.
Bracher, T. Papke, J. Chem. Soc. Perkin Transactions 1 (1995) 2323-2326.

Declaration of interests
☒ The authors declare that they have no known competing financial interests or personal
relationships that could have appeared to influence the work reported in this paper.
☐The authors declare the following financial interests/personal relationships which may be
considered as potential competing interests:

Graphical Abstract
Novel access to 2-substituted quinolin-4-ones by
nickel boride-mediated reductive ring transformation
of 5-(2-nitrophenyl)isoxazoles
Leave this area blank for abstract info.
Bernhard Lohrer and Franz Bracher
[3+2] cycloaddition with
in situ-prepared nitrile oxide
R-CNO
N
O
O
NaBH4 (7 equ.)
R
NiCl2 (1 equ.)

Highlights
• Access to quinolin-4-ones by reductive ring transformation of nitrophenyl isoxazoles
• Nickel boride identified as reducing agent with high chemoselectivity
• High functional group compatibility
• This method opens novel applications for the total synthesis of quinolin-4-one alkaloids