Near room temperature cross-coupling reactions of arene boronic acids with a quinoxaline 1,4-dioxide benzylsulfanyl derivative

Article abstract of DOI:10.1002/ejoc.201200433

A new quinoxaline 1,4-dioxide benzylsulfanyl derivative was tested under mild conditions in copper-mediated Liebeskind-Srogl (LS) cross-coupling reactions with arene boronic acids. These first organometallic cross-coupling reactions performed on a quinoxaline 1,4-dioxide derivative open new perspectives in medicinal chemistry. They represent also the first LS cross-couplings run with an electrophile equipped with an aryl N-oxide moiety. A mild an efficient protocol was developed to prepare quinoxaline 1,4-dioxide derivatives. A benzylsulfanyl quinoxaline 1,4-dioxide derivative was engaged as an electrophilic partner in copper-mediated Liebeskind-Srogl cross-coupling reactions with a wide range of arene boronic acids. These experiments represent the first organometallic coupling reactions on quinoxaline 1,4-dioxide derivatives. Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.

Full text of DOI:10.1002/ejoc.201200433

Job/Unit: O20433
/KAP1
Date: 31-05-12 17:33:53
Pages: 6
SHORT COMMUNICATION
DOI: 10.1002/ejoc.201200433
Near Room Temperature Cross-Coupling Reactions of Arene Boronic Acids
with a Quinoxaline 1,4-Dioxide Benzylsulfanyl Derivative
Samir Dahbi[a] and Philippe Bisseret*[a,b]
Keywords: Heterocycles / Nitrogen oxides / Cross-coupling / Boron / Copper
A new quinoxaline 1,4-dioxide benzylsulfanyl derivative was
tested under mild conditions in copper-mediated Liebes-
kind–Srogl (LS) cross-coupling reactions with arene boronic
acids. These first organometallic cross-coupling reactions
performed on a quinoxaline 1,4-dioxide derivative open new
perspectives in medicinal chemistry. They represent also the
first LS cross-couplings run with an electrophile equipped
with an aryl N-oxide moiety.
reaction can rapidly become tedious, as it implies the repeti-
tion of condensation reactions from the starting material.
A more elegant approach, which has not been explored yet,
would rely on the transition-metal-catalyzed derivatization
of an electrophilic quinoxaline 1,4-dioxide precursor. To the
best of our knowledge, only either non-oxygenated 2-halo-
quinoxaline derivatives like 2 or quinoxaline monoxide 3
have been involved so far in transition-metal-catalyzed
cross-coupling reactions: compound 2 has been engaged in
Introduction
Heteroaryls derived from the quinoxaline 1,4-di-N-oxide
scaffold such as compounds 1 are well represented in medic-
inal chemistry. These compounds are endowed with many
biological properties that mainly encompass the properties
of their mono- or non-oxygenated homologs.[1,2] Apart
from their antibacterial and antiprotozoal properties, mem-
bers of this group have recently been in the limelight be-
cause of their potential as prodrugs against hypoxic cancer-
ous cells associated with increased resistance to radiation
and chemotherapy.[3] For many years, the synthesis of quin-
oxaline dioxides has mainly relied on the condensation of
benzofuroxan derivatives with 1,3-diketones or β-keto-es-
ters under basic conditions (Scheme 1), as originally de-
scribed by Issidorides and Haddadin from Beirut Univer-
sity.[4] The so-called Beirut reaction was extended further
to the synthesis of 2-sulfanyl-[5] and 2-dialkylphosphonyl-[6]
quinoxaline 1,4-dioxide derivatives.
a
few Suzuki, Stille, and Sonogashira reactions
(Scheme 2a),[7] whereas 3 has recently been subjected to pal-
ladium-catalyzed cross-coupling reactions involving direct
C–H activation of its most acidic C–H bond (Scheme 2b).[8]
Although quite elegant, this latter method initiated by Fag-
nou et al. is not recommended for the preparation of ther-
mally sensitive molecules, as it requires temperatures over
100 °C.
Scheme 1. Formation of quinoxaline 1,4-dioxides by the Beirut re-
action.
In medicinal chemistry, the preparation of a wide range
of quinoxaline 1,4-dioxide analogs by making use of this
[a] Université de Haute-Alsace
École Nationale Supérieure de Chimie de Mulhouse
Laboratoire de Chimie Organique et Bioorganique
3 rue A. Werner, 68093 Mulhouse Cedex, France
Fax: +33-3-89336860
E-mail: philippe.bisseret@uha.fr
[b] Institut de Science des Matériaux de Mulhouse LRC CNRS
7228,
Mulhouse, France
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201200433.
Scheme 2. Organometallic cross-coupling reactions involving quin-
oxaline derivatives: (A) see ref.[7]; (b) see ref.[8]; (c) our work.
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1

Job/Unit: O20433
/KAP1
Date: 31-05-12 17:33:53
Pages: 6
S. Dahbi, P. Bisseret
SHORT COMMUNICATION
We present here our organometallic cross-coupling reac-
We began our study by investigating Suzuki–Miyaura
tions aimed at preparing 2-arylated quinoxaline 1,4-dioxide cross-coupling reactions between tolylboronic acid and 2-
derivatives of type 6 mainly from novel benzylsulfanyl de- chloro-3-methylquinoxaline 1,4-dioxide (5) by using sodium
rivative 4 (Scheme 2c). This sulfurylated derivative was en- carbonate as a base and Pd(PPh3)4 as a catalyst, as advo-
gaged in palladium-catalyzed cross-coupling reactions with cated for the cross-couplings of 2,3-dichloroquinoxaline[7f]
arene boronic acids in the presence of an excess amount of or 2-chloropyridine N-oxide derivatives.[13] Despite our ef-
copper(I) thiophene-2-carboxylate (CuTC), that is, by using forts, by varying the solvent (THF or DMF), the tempera-
conditions initially developed by Liebeskind and Srogl for ture, and the reaction time we could not detect any trace
the cross-couplings of (hetero)aromatic thioethers.[9] Al- amounts of the desired 2-tolyl-quinoxaline 1,4-dioxide. As
though no sulfurylated heteroaryl N-oxides have been inves- similar negative results were obtained when we used
tigated so far in Liebeskind–Srogl (LS) reactions, nonaro- Pd(OAc)2 in the presence of PPh3 or dppf in place of
matic thioimidate N-oxides have been shown to react with Pd(PPh3)4, we did not explore this kind of cross-coupling
phenylboronic acid in the presence of 2 equiv. of copper(I) any further.
3-methylsalicylate (CuMeSal).[10] For comparison, we also
We rather turned our attention to the iron-catalyzed
describe here the organometallic cross-couplings of already cross-coupling reactions of 5 with phenylmagnesium brom-
known 2-chloride 5.[5a,5b]
ide by using either THF alone or a mixture of THF and N-
methylpyrrolidone (NMP) as the solvent and [Fe(acac)3] as
the catalyst, as recommended for the arylation of heteroaryl
chlorides (including 2-chloroquinoxaline) by Grignard rea-
gents.[14]
Results and Discussion
Although we ran many experiments, varying the tem-
perature (from –10 to 60 °C), the number of equivalents of
phenylmagnesium bromide (2–4 equiv.), as well as the time
of the reaction (1 h–18 h), the results were also quite disap-
pointing, as in most cases only complex mixtures were ob-
tained. In the most favorable experiment run at 60 °C over
1 h, we could isolate expected dioxide 11a in low yield
(15%) next to 2-methyl-3-phenylquinoxaline (8) and mon-
oxides 9 and 10 (Scheme 4).[15] This tendency to yield
deoxygenated derivatives was also observed in trials run at
lower temperatures.
We first prepared benzylsulfanyl derivative 4 by using the
Beirut reaction, as already advocated for the synthesis of
quinoxaline thioether derivatives.[5] Although ammonia gas
was previously recommended for this transformation,[5] we
obtained a very good yield of 4 by simply making use of
concentrated aqueous ammonia in THF (Scheme 3). With
other bases like cesium carbonate or calcium hydroxide in
THF, only very poor yields of 4 were obtained. At this
stage, direct and high-yielding access to 2-chloride 5 was
easily obtained simply by treating 4 with 2 equiv. of 2,4-
dichloro-5,5-dimethylhydantoin under conditions originally
reported for the mild conversion of benzylic sulfides into
the corresponding chlorosulfonyl derivatives.[11] In our case,
chlorosulfonyl quinoxaline 1,4-dioxide 7 could not be de-
tected by analyzing the reaction crude mixture by NMR
spectroscopy.[12] This new procedure for the preparation of
5 appears much milder than the previously reported two-
step protocol, which relies on the oxidation of the quinox-
aline thioether derivative into the corresponding sulfone
prior to treatment with hot concentrated HCl.[5a,5b]
Scheme 4. Iron-catalyzed cross-coupling reaction.
At this stage, rather than investigating cross-coupling re-
actions starting from other 2-haloquinoxaline 1,4-dioxides
in place of 5,[16] we decided to study the reactivity of benzyl-
sulfanyl derivative 4 in LS cross-coupling reactions with bor-
onic acids, in the hope that the mild and neutral conditions
used would prevent the formation of deoxygenated com-
pounds. Optimization of the reaction conditions was per-
formed with phenylboronic acid (Table 1).
Our preliminary experiment (Table 1, Entry 1) conducted
at 50 °C with about 1 equiv. of both the boronic acid and
CuTC was encouraging, as it gave the expected cross-cou-
pling compound in the same yield as the best experiment
run with chloride 5. We were in particular pleased to find
Scheme 3. Preparation of electrophilic quinoxaline 1,4-dioxides 4
and 5.
2
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O20433
/KAP1
Date: 31-05-12 17:33:53
Pages: 6
Cross-Coupling Reactions of Arene Boronic Acids
Table 1. Optimization of the LS cross-coupling reaction.
whereas starting benzylsulfanyl derivative 4 was entirely re-
covered in the absence of CuTC (Table 1, Entry 6). By em-
ploying 1.1 equiv. of the boronic acid in the presence of
3.5 equiv. of CuTC, a good yield of 11a was also obtained,
though not as high as that obtained with 2.2 equiv. of the
boronic acid (Table 1, Entry 4). Switching from CuTC to
CuMeSal (another copper reagent sometimes used in LS
couplings) was detrimental, as we only could isolate 11a
in low yield (Table 1, Entry 5). At room temperature, with
2.2 equiv. of phenylboronic acid and 3.5 equiv. of CuTC,
the reaction appeared much more sluggish, leading to 11a
in only modest yield after 2 d (Table 1, Entry 7), and by
using potassium phenyltrifluoroborate in place of phen-
ylboronic acid, the reaction practically did not occur
(Table 1, Entry 8). Finally, when the reaction was run with
2.2 equiv. of phenylboronic acid and 3.5 equiv. of CuTC at
30 °C over only 5 h, starting quinoxaline dioxide 4 was al-
most completely consumed and the desired cross-coupling
product was obtained in excellent yield (Table 1, Entry 9).
Applying these optimized conditions, we examined the
scope of the LS cross-coupling reaction between 4 and ar-
ene boronic acids (Scheme 5).
Entry
x/y
t [h]
T [°C]
4/11a[a]
Yield [%][b]
1
2
3
1.1:1.3
2.2:2.2
2.2:3.5
1.1:3.5
2.2:3.5
2.2:0.0
2.2:3.5
2.2:3.5
2.2:3.5
18
18
18
18
18
18
48
48
5
50
30
30
30
30
30
20
20
30
68:32
66:34
0:100
5:95
15
20
89
4
72
5[c]
6
5:95
26[d]
n.d.[e]
25
100:0
60:40
95:5
7
8[f]
9
3
93
5:95
[a] Estimated by NMR spectroscopy. [b] Isolated yield. [c] CuMe-
Sal was used instead of CuTC. [d] Several unidentified quinoxaline
derivatives were present. [e] Starting compound 4 was entirely reco-
vered. [f] PhBF3 K+ was used in place of PhB(OH)2.
–
in the NMR spectrum of the crude mixture that essentially
As the results shown in Scheme 5 attest, this method is
only starting material 4 was present and no trace amounts compatible with a variety of arene boronic acids possessing
of deoxygenated derivatives 8–10 were observed. About the either electron-donating or electron-withdrawing groups in
same result was obtained at 30 °C with nearly 2 equiv. of the meta or para position. Notably, the tolerance for chlor-
both CuTC and phenylboronic acid (Table 1, Entry 2). ide or bromide on the arene moiety in this LS reaction of-
Raising the number of equivalents of CuTC to 3.5 was key fers an opportunity for envisioning subsequent cross-cou-
to obtaining a very good yield of 11a (Table 1, Entry 3), pling reactions.
Scheme 5. Synthesis of 2-methyl-3-aryl-quinoxaline 1,4-dioxides. Reactions were monitored by NMR spectroscopy. Yield of the isolated
product is reported.
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3

Job/Unit: O20433
/KAP1
Date: 31-05-12 17:33:53
Pages: 6
S. Dahbi, P. Bisseret
SHORT COMMUNICATION
d) D. Benitez, M. Cabrera, P. Hernández, L. Boiani, M. L. Lav-
aggi, R. Di Maio, G. Yaluff, E. Serna, S. Torres, M. E. Ferreira,
N. Vera de Bibao, E. Torres, S. Pérez-Silanes, B. Solano, E.
Moreno, I. Aldana, A. López de Ceráin, H. Cerecetto, M.
González, A. Monge, J. Med. Chem. 2011, 54, 3624–3636.
For a review, see: a) A. Carta, P. Corona, M. Loriga, Curr.
Med. Chem. 2005, 12, 2259–2272; b) M. González, H. Cere-
cetto, A. Monge, Top. Heterocycl. Chem. 2007, 11, 179–211.
a) K. M. Amin, M. F. Isamil, E. Noaman, D. H. Soliman, Y.
Ammar, Bioorg. Med. Chem. 2006, 14, 6917–6923; b) U. Das,
H. N. Pati, A. K. Panda, E. De Clercq, J. Balzarini, J. Molnár,
Z. Baráth, I. Ocsovski, M. Kawase, L. Zhou, H. Sakagami,
J. R. Dimmock, Bioorg. Med. Chem. 2009, 17, 3909–3915; c)
M. L. Lavaggi, M. Cabrera, M. De Los Ángeles Aravena, C.
Olea-Azar, A. López de Ceráin, A. Monge, G. Pachón, M. Ca-
scante, A. M. Bruno, L. I. Pietrasanta, M. González, H. Cere-
cetto, Bioorg. Med. Chem. 2010, 18, 4433–4440.
a) C. H. Issidores, M. J. Haddadin, J. Org. Chem. 1966, 31,
4067–4068; b) M. L. Ewards, R. E. Bambury, H. W. Ritter, J.
Med. Chem. 1976, 19, 330–333; c) M. J. Haddadin, C. H. Issid-
ores, Heterocycles 1993, 35, 1503–1525.
a) E. Abushanab, J. Org. Chem. 1973, 38, 3105–3107; b) A.
Carta, G. Paglietti, M. E. R. Nikookar, P. Sanna, L. Sechi, S.
Zanetti, Eur. J. Med. Chem. 2002, 37, 355–366; c) R. Sheng, Y.
Xu, Q. Weng, Q. Xia, Q. He, B. Yang, Y. Hu, Drug Discov.
Ther. 2007, 1, 119–123.
Arene boronic acids bearing a substituent in the ortho
position did not effectively undergo the cross-coupling reac-
tion with 4. We were unable to isolate 11d, 11i, or 11s after
18 h of reaction at 30 °C and a good yield of 11e was real-
ized only when forcing conditions were used. This sensitiv-
ity of LS couplings to steric hindrance has already been
reported.[8c]
[2]
[3]
Conclusions
In summary, we have for the first time developed an or-
ganometallic cross-coupling reaction involving a quinoxal-
ine 1,4-dioxide partner. For that purpose, new sulfurylated
quinoxaline 1,4-dioxide electrophilic derivative 4 was read-
ily prepared from commercially available compounds.
Whereas chloride 5 did not give satisfactory results, either
in Suzuki–Miyaura or in iron-catalyzed cross-coupling in-
volving tolylboronic acid or phenylmagnesium bromide,
respectively, much more convincing results were obtained
with benzylsulfanyl derivative 4. Indeed, this sulfurylated
compound was found to react quite efficiently under neu-
tral conditions and at near room temperature in LS-type
cross-couplings with a variety of arene boronic acids, in the
presence of a catalytic amount of Pd(PPh3)4 and at least
3 equiv. of CuTC. The necessity, in our case, of a large ex-
cess amount of a copper(I) salt contrasts with typical LS
cross-couplings, which are usually operational with only
1 equiv. of CuTC and can probably be explained by the
ability of the N-oxide groups to form a complex with a cop-
per(I) salt.[17] Apart from offering an extension to the chem-
istry of quinoxaline 1,4-dioxides, our work widens the scope
of LS cross-coupling reactions which, to the best of our
knowledge, have not been performed yet with an electro-
philic partner possessing an aryl N-oxide moiety. We will
continue to explore the scope of our method by testing het-
eroarylboronic acids in place of arene boronic acids.
[4]
[5]
[6]
[7]
S. Dahbi, E. Methnani, P. Bisseret, Tetrahedron Lett. 2010, 51,
5516–5520.
For Sonogashira couplings, see: a) A. Dinsmore, J. H. Birks,
C. D. Garner, J. A. Joule, J. Chem. Soc. Perkin Trans. 1 1997,
801–808; b) C.-S. Kim, K. C. Russel, J. Org. Chem. 1998, 63,
8229–8234; c) M. Armengol, J. A. Joule, J. Chem. Soc. Perkin
Trans. 1 2001, 154–158; d) B. Bradshaw, A. Dinsmore, D. Coli-
son, D. Garner, J. A. Joule, J. Chem. Soc. Perkin Trans. 1 2001,
3232–3238; for Suzuki couplings, see: e) J. J. Li, W. S. Yue, Tet-
rahedron Lett. 1999, 40, 4507–4510; f) H. Sakurai, T. Hirao,
Synthesis 2004, 2535–2539; for Stille couplings, see: g) A. Din-
smore, C. D. Garner, J. A. Joule, Tetrahedron 1998, 54, 3291–
3302.
a) L.-C. Campeau, S. Rousseaux, K. Fagnou, J. Am. Chem.
Soc. 2005, 127, 18020–18021; b) J.-P. Leclerc, K. Fagnou, An-
gew. Chem. 2006, 118, 7945; Angew. Chem. Int. Ed. 2006, 45,
7781–7786; c) Z. Wang, K. Li, D. Zhao, J. Lan, J. You, Angew.
Chem. Int. Ed. 2011, 50, 5365–5369.
[8]
[9]
a) L. S. Liebeskind, J. Srogl, Org. Lett. 2002, 4, 979–981; b) M.
Koley, L. Wimmer, M. Schnürch, M. D. Milhovilovic, Eur. J.
Org. Chem. 2011, 1972–1979.
J. Schleiss, P. Rollin, A. Tatibouët, Angew. Chem. 2010, 122,
587; Angew. Chem. Int. Ed. 2010, 49, 577–580.
Supporting Information (see footnote on the first page of this arti-
1
cle): Experimental procedures and copies of the H NMR and 13C
[10]
[11]
[12]
NMR spectra for all new compounds.
Y.-M. Pu, A. Christesen, Y.-Y. Ku, Tetrahedron Lett. 2010, 51,
418–421.
Heterocyclic sulfonyl chlorides are often quite unstable and
rapidly decompose with the loss of sulfur dioxide into the cor-
responding chloride derivatives: R. O. Roblin, J. W. Clapp, J.
Am. Chem. Soc. 1950, 72, 4890–4892.
O. Lohse, P. Thevenin, E. Waldvogel, Synlett 1999, 45–48.
a) A. Fürstner, A. Leitner, M. Méndez, H. Krause, J. Am.
Chem. Soc. 2002, 124, 13856–13863; b) J. Quintin, X. Franck,
R. Hocquemiller, B. Figadère, Tetrahedron Lett. 2002, 43,
3547–3549; c) L. Boully, M. Darabantu, A. Turck, N. Plé, J.
Heterocycl. Chem. 2005, 42, 1423–1428.
Identification of 8 was readily supported by comparison with a
commercially available reference. The structures of quinoxaline
derivatives 10 and 11a were confirmed by oxidation of 8 with
a mixture of H2O2 (30%)/AcOH/Ac2O as already reported: a)
S. Landquist, G. J. Stacey, J. Chem. Soc. 1953, 2822–2830; b)
E. C. Taylor, G. W. H. Cheeseman, J. Am. Chem. Soc. 1964, 86,
1830–1835. Quinoxaline monoxide 9 could be obtained as a
minor product next to 10 when 8 was oxidized as before with
the omission of acetic anhydride.
Acknowledgments
The authors thank the Centre National de la Recherche Sci-
entifique (CNRS) for financial support. S.D. is grateful to the
“Ministère de la Recherche et de lЈÉducation Nationale” for a fel-
lowship. We thank Dr. Didier Le Nouen for his help with NMR
recording and spectra interpretation and Dr. Cécile Joyeux for run-
ning the MS experiments. Karim Dahbi is thanked for his partici-
pation to the optimization study of the LS cross-coupling reaction.
[13]
[14]
[15]
[1] a) A. Jaso, B. Zarranz, I. Aldana, A. Monge, Eur. J. Med.
Chem. 2003, 38, 791–800; b) A. Carta, M. Loriga, G. Paglietti,
A. Mattana, P. L. Fiori, P. Mollicotti, L. Sechi, S. Zanetti, Eur.
J. Med. Chem. 2004, 39, 195–203; c) S. Ancizu, E. Moreno, B.
Solano, R. Villar, A. Burguete, E. Torres, S. Pérez-Silanes, I.
Aldana, A. Monge, Bioorg. Med. Chem. 2010, 18, 2713–2719;
4
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O20433
/KAP1
Date: 31-05-12 17:33:53
Pages: 6
Cross-Coupling Reactions of Arene Boronic Acids
[16] Contrary to the general trend in organometallic reactions, aryl
chlorides are preferred substrates compared to iodides or
bromides in iron-catalyzed cross-couplings: B. D. Sherry, A.
Fürstner, Acc. Chem. Res. 2008, 41, 1500–1511.
C.-Y. Shen, Z.-Q. Pan, Y.-N. Wang, Q.-H. Luo, Synth. React.
Inorg. Met. Org. Chem. 2003, 33, 1263–1274; b) S. H. Hong,
A. K. Gupta, W. B. Tolman, Inorg. Chem. 2009, 48, 6323–6325.
Received: April 4, 2012
[17] Although most of the reported copper complexes with N-oxide
Published Online: ᭿
derivatives are CuII species, a few CuI variants are known: a)
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5

Job/Unit: O20433
/KAP1
Date: 31-05-12 17:33:53
Pages: 6
S. Dahbi, P. Bisseret
SHORT COMMUNICATION
Cross-Coupling
A mild an efficient protocol was developed
to prepare quinoxaline 1,4-dioxide deriva-
tives. A benzylsulfanyl quinoxaline 1,4-di-
oxide derivative was engaged as an electro-
philic partner in copper-mediated Liebes-
kind–Srogl cross-coupling reactions with a
wide range of arene boronic acids. These
experiments represent the first organomet-
allic coupling reactions on quinoxaline 1,4-
dioxide derivatives
S. Dahbi, P. Bisseret* ........................ 1–6
Near Room Temperature Cross-Coupling
Reactions of Arene Boronic Acids with a
Quinoxaline 1,4-Dioxide Benzylsulfanyl
Derivative
Keywords: Heterocycles / Nitrogen oxides /
Cross-coupling / Boron / Copper
6
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0