Regioselective bromination of fused heterocyclic N-oxides

Article abstract of DOI:10.1021/ol3034675

A mild method for the regioselective C2-bromination of fused azine N-oxides is presented, employing tosic anhydride as the activator and tetra-n-butylammonium bromide as the nucleophilic bromide source. The C2-brominated compounds are produced in moderate to excellent yields and with excellent regioselectivity in most cases. The potential extension of this method to other halogens, effecting C2-chlorination with Ts₂O/TBACl is also presented. Finally, this method could be incorporated into a viable one-pot oxidation/bromination process, using methyltrioxorhenium/urea hydropgen peroxide as the oxidant.

Full text of DOI:10.1021/ol3034675

ORGANIC
LETTERS
2013
Vol. 15, No. 4
792–795
Regioselective Bromination of Fused
Heterocyclic N‑Oxides
Sarah E. Wengryniuk,† Andreas Weickgenannt,† Christopher Reiher,‡
Neil A. Strotman,‡ Ke Chen,*,‡ Martin D. Eastgate,*,‡ and Phil S. Baran*,†
Department of Chemistry, The Scripps Research Institute, 10550 North Torrey Pines
Road, La Jolla, California 92037, United States, and Chemical Development,
Bristol-Myers Squibb, One Squibb Drive, New Brunswick, New Jersey, 08903, United States
pbaran@scripps.edu; ke.chen@bms.com; martin.eastgate@bms.com
Received December 18, 2012
ABSTRACT
A mild method for the regioselective C2-bromination of fused azine N-oxides is presented, employing tosic anhydride as the activator and tetra-n-
butylammonium bromide as the nucleophilic bromide source. The C2-brominated compounds are produced in moderate to excellent yields and
with excellent regioselectivity in most cases. The potential extension of this method to other halogens, effecting C2-chlorination with Ts2O/TBACl
is also presented. Finally, this method could be incorporated into a viable one-pot oxidation/bromination process, using methyltrioxorhenium/
urea hydropgen peroxide as the oxidant.
C2-substituted heteroarenes are common motifs in
bioactive molecules. The synthesis of these compounds
commonly relies on CÀC bond formation through cross-
coupling reactions or the introduction of heteroatoms
through SNAr techniques, both requiring prior generation
of the C2-halogenated derivative.1 Unfortunately, direct
incorporation of halogens onto heterocycles (Figure 1A,
1 f3) proves challenging, as issues of poor regioselectivity
and over halogenation often arise. C2-halogenation via the
corresponding N-oxides (readily available via oxidation of
the parent arenes) provides a popular alternative strategy
(Figure 1A). Thus, chlorination is commonly achieved
using POCl3 or PCl5 whereas iodination often requires a
metalation based strategy.2,3 Comparatively, there is a
dearth of mild, reliable methods for N-oxide bromination.
There have been a few reports on the bromination of
7-azaindoles, at both the 2- and 4-position,4 but these
reported methods have not found wide application, most
likely due to lack of generality. By far the most common
methods use POBr3 or Br2 at elevated temperatures.
(3) (a) Clark, R. B.; He, M. S.; Fyfe, C.; Lofland, D.; O’Brien, W. J.;
Plamondon, L.; Sutcliffe, J. A.; Xiao, X. Y. J. Med. Chem. 2011, 54 (5),
1511–1528. (b) Heitman, L. H.; Goblyos, A.; Zweemer, A. M.; Bakker,
R.; Mulder-Krieger, T.; van Veldhover, J. P. D.; de Vries, H.; Brussee, J.;
Ijzerman, A. P. J. Med. Chem. 2009, 52 (4), 926–931. (c) Lumeras, W.;
Caturla, F.; Vidal, L.; Esteve, C.; Balague, C.; Orellana, A.; Dominguez,
M.; Roca, R.; Huerta, J. M.; Godessart, N.; Vidal, B. J. Med. Chem.
2009, 52 (17), 5531–5545. (d) Paudler, W. W.; Jovanovic, M. V. J. Org.
Chem. 1983, 48 (7), 1064–1069. (e) Petitjean, A.; Kyritsakas, N.; Lehn,
J. M. Chem.;Eur. J. 2005, 11 (23), 6818–6828. (f) Richter, H. G. F.;
Adams, D. R.; Benardeau, A.; Bickerdike, M. J.; Bentley, J. M.; Blench,
T. J.; Cliffe, I. A.; Dourish, C.; Hebeisen, P.; Kennett, G. A.; Knight,
A. R.; Malcolm, C. S.; Mattei, P.; Misra, A.; Mizrahi, J.; Monck,
N. J. T.; Plancher, J. M.; Roever, S.; Roffey, J. R. A.; Taylor, S.;
Vickers, S. P. Bioorg. Med. Chem. Lett. 2006, 16 (5), 1207–1211. (g)
Yang, Y. H. Synth. Commun. 1989, 19 (5À6), 1001–1008. (h) Zhang,
Y. Q.; Guiguemde, W. A.; Sigal, M.; Zhu, F. Y.; Connelly, M. C.;
Nwaka, S.; Guy, R. K. Bioorgan. Med. Chem. 2010, 18 (7), 2756–2766.
(4) (a) Minakata, S.; Komatsu, M.; Ohshiro, Y. Synthesis-Stuttgart
1992, 7, 661–663. (b) Thibault, C.; L’Heureux, A.; Bhide, R. S.; Ruel, R.
Org. Lett. 2003, 5 (26), 5023–5025.
† The Scripps Research Institute.
‡ Bristol-Myers Squibb.
(1) (a) Taylor, E. C.; Crovetti, A. J. J. Org. Chem. 1954, 19 (10), 1633–
1640. (b) Inglis, S. R.; Stojkoski, C.; Branson, M. M.; Cawthray, J. F.;
Fritz, D.; Wiadrowski, E.; Pyke, S. M.; Booker, G. W. J. Med. Chem.
2004, 47 (22), 5405–5417. (c) Chambers, R. J.; Marfat, A. Synth.
Commun. 1997, 27 (3), 515–520. (d) Ji, J. G.; Li, T.; Bunnelle, W. H.
Org. Lett. 2003, 5 (24), 4611–4614. (e) Nicolaou, K. C.; Namoto, K.;
Ritzen, A.; Ulven, T.; Shoji, M.; Li, J.; D’Amico, G.; Liotta, D.; French,
C. T.; Wartmann, M.; Altmann, K. H.; Giannakakou, P. J. Am. Chem.
Soc. 2001, 123 (38), 9313–9323. (f) You, Y. M.; Park, S. Y. J. Am. Chem.
Soc. 2005, 127 (36), 12438–12439.
(2) (a) Mongin, F.; Trecourt, F.; Gervais, B.; Mongin, O.; Queguiner,
G. J. Org. Chem. 2002, 67 (10), 3272–3276. (b) Mongin, O.; Rocca, P.;
Thomasditdumont, L.; Trecourt, F.; Marsais, F.; Godard, A.; Queguiner,
G. J. Chem. Soc., Perkin Trans. 1 1995, No. 19, 2503–2508. (c) Trecourt,
F.; Gervais, B.; Mongin, O.; Le Gal, C.; Mongin, F.; Queguiner, G.
J. Org. Chem. 1998, 63 (9), 2892–2897.
r
10.1021/ol3034675
Published on Web 01/25/2013
2013 American Chemical Society

However, regiocontrol is largely substrate dependent and
des-bromo product, as well as over bromination, are
normally observed as major side reations; in addition,
exposureofN-oxidestoprolongedheating can causesafety
concerns regarding thermal stability and the liability of
exothermic degradation.
The lack of mild, reliable, and scalable methods for
N-oxide bromination came to our attention during recent
efforts to prepare kilogram quantities of a drug candidate.
In this letter, the invention of a simple and scalable method
for the synthesis of C2-bromo fused azines, from the
corresponding N-oxide precursors, is described. This
new method proves applicable to both electron-deficient
and electron-rich heterocycles, including quinolines, iso-
quinolines, and azaindoles, possessing a variety of sub-
stitution patterns. The potential extension of this method
to chlorinations is also demonstrated as well as a one-pot
oxidation/bromination sequence that obviates the need for
isolation of the N-oxide intermediates.
followed by attack of a bromide ion and subsequent
deprotonation/aromitization to give the C2-substituted pro-
ducts, similar to a ReissertÀHenze reaction (Figure 1C).5
This approach is widely used in the context of N-oxides6 and
has been used for chlorination;7 however issues of regios-
electivity, observation of des-oxygenation, and, in many
cases, competing addition of the activator counterion are
problematic. Two recent encouraging reports in the litera-
ture, by Wei6a and Yin,6b showed that a range of heteroatom
nucleophiles were added selectively at the C2-position
when PyBroP (PyBroP = Bromotripyrrolidinophosphonium
hexafluorophosphate) or Ts2O was used as the activating
agent. By adopting a similar approach but substituting the
nucleophile with an appropriate bromide source, we hoped
to obtain the C2-brominated compounds while avoiding the
aforementioned competing side reactions (Figure 1C).
Using commercially available 6-methoxyquinoline
N-oxide (8), a variety of activating agent/bromide combina-
tions were screened for selective C2-bromination (Table 1).
As expected, when standard bromination conditions, such
as POBr3, were utilized (Table 1, entries 1À6) multiple
competing processes occurred. However, two new alter-
native conditions were identified, enabling the desired C2-
bromination with muchimproved efficiencyand selectivity
(Table 1, entries 7 and 10). We were pleased to discover
that PyBroP can function as both the activating agent and
the external bromide source, delivering the C2-bromide in
69% isolated yield, with 5:1 regioselectivity. It was hy-
pothesized that the regioselectivity arose from prior co-
ordination to the N-oxide and delivery of the bromide
intramolecularly through a pentavalent phosphorus inter-
mediate. Alternatively, the use of Ms2O/TBABr gave both
a significantly higher yield and regioselectivity (Table 1,
entry 8) while maintaining the mild reaction conditions.
We then investigated the use of other sulfonyl anhydrides,
and while Tf2O gave no desired product, the use of Ts2O
Table 1. Screening of Activator/Bromine Source Combinations
Figure 1. (A) Common methods for heteroarene and N-oxide
bromination. (B) Functionalization of N-oxides via activating
agent/nucleophile. (C) Novel method for mild, regioselective
bromination.
entry
conditions
PPh3, Br2
C2:C8
conversion
1
2
3
4
5
6
7
1:1.03
2.2:1
n.d.
n.d.
n.d.
n.d.
5:1
22
PPh3, NBS
17
POBr3
<1
SOBr2
<1
The design of a new method built upon promising
precedents is summarized in Figure 1B. In such a method,
coordination of an activating agent to the N-oxide would
enhance the electrophilic character of the C2-position,
oxalic bromide
<1
acetyl bromide
<1
84 (69)a
PyBroP, 4 A MS, CF3Ph,
K3PO4
8
Ms2O, TBABr, THF, 4 A MSb
Tf2O, TBABr, THF, 4 A MSb
Ts2O, TBABr, THF, 4 A MSb
15:1
n.d.
15:1
90
(5) Henze, M. Ber. Dtsch. Chem. Ges. 1936, 69, 1566–1568.
(6) (a) Londregan, A. T.; Jennings, S.; Wei, L. Q. Org. Lett. 2011,
13 (7), 1840–1843. (b) Yin, J. J.; Xiang, B. P.; Huffman, M. A.; Raab,
C. E.; Davies, I. W. J. Org. Chem. 2007, 72 (12), 4554–4557.
(7) (a) L’heureux, A.; Thibault, C.; Ruel, R. Tetrahedron Lett. 2004,
45 (11), 2317–2319. (b) Schneider, C.; Goyard, D.; Gueyrard, D.;
Joseph, B.; Goekjian, P. G. Eur. J. Org. Chem. 2010, 34, 6665–6677.
9
<1
100 (97)a
10
a Isolated yield. b TBABr and substrate stirred with molecular sieves
10 min prior to addition of activator.
Org. Lett., Vol. 15, No. 4, 2013
793

Unfortunately, when this same reagent system was
applied in the context of 6-fluoroquinoline N-oxide
(10) and 3-methoxyquinoline N-oxide (11), significant
formation of byproducts was observed, including regioi-
somers (3), des-bromo product (12), and a C2-O-tosyl
derivative (13) (Table 2, entries 1 and 7).
Table 2. Optimization of Reaction Conditions with 10 and 11
In order to address these issues, a range of solvents and
reaction concentrations were examined (Table 2). In the
case of 11, we hypothesized reduction product 12 to be a
result of hydride donation from THF, and indeed 12 was
not observed in any other solvent in our screen (Table 2,
entries 2À6). Interestingly, the use of DCM not only
eliminated reduction but also gave exclusive formation of
the C2 isomer (as determined by 1H NMR). Turning our
attention to 6-fluoroquinoline (10), it was hypothesized
that 13 was forming as a result of a dimerization reaction,
since molecular sieves were present to eliminate water.
Alternatively, attack of a second N-oxide to the 2-position
of an activated species would lead to generation of the
amide after deoxygenation. Tosylation of the resulting
amide under the reaction conditions would then give 13.
To circumvent this issue, we first attempted slow inverse
addition of the N-oxide; however this was unsuccessful.
Gratifyingly, running the reaction at lower concentration
(0.01 M in DCM) led to 71% isolated yield of the desired
compound with none of the undesired byproduct ob-
served. When these optimized conditions were applied in
the context of the first model substrate, 6-methoxyquinoline
N-oxide (8), we saw no loss in conversion and now obtained
asingle regioisomer (Table 2, entry 9). Using these optimized
entry
1
compd
solvent
THF
[M]
0.1
conv
100
3
12
13
11
82
18
0
[1:1]a
100
100
100
100
100
0
2
3
4
5
6
7
8
11
11
11
11
11
10
10
CPME
Et2O
0.1
0.1
0.1
0.1
0.1
0.1
0.01
95
0
0
0
0
0
0
0
0
80
0
CF3Ph
PhMe
DCM
DCM
DCM
100
100
100
100
100
(71)b
100
(97)b
0
0
0
100
0
100
9
8
DCM
0.01
100c
0
0
[>30:1]
a [1:1] C2:C4 regioisomers. b Isolated yield. c Only single regioisomer
detected by 1H NMR.
gave further improvement (Table 1, entry 10), affording
the product with both the highest yield (97%) and regio-
selectivity (15:1); as such these conditions were chosen for
further exploration.
Figure 2. Scope of N-oxide bromination. All compounds were obtained as a single regioisomer. aReaction run with 2.5 equiv of Ts2O
and 4 equiv of TBABr. bFor a comprehensive list of simple pyridine derivatives examined, see Supporting Information.
794
Org. Lett., Vol. 15, No. 4, 2013

conditions, the rate of the reaction was examined with both
8 and 10, and it was found that the reaction was complete
within 30 min in both cases.
substrate, we found TBACl to be very effective, giving the
2-chloro product (14) in 96% yield, with 10:1 regioselec-
tivity (Figure 3A).
Finally, performing both the oxidation and bromination
in a one-pot process was explored, thus eliminating the
need for prior oxidation and isolation of the N-oxide, thus
affecting the equivalent of a direct CÀH bromination
(Figure 3B). A range of oxidation conditions were screened
using 6-methoxyquinoline (8), and it was found that the
use of MTO/UHP,8 followed by the addition of molecular
sieves, TBABr, and Ts2O gave the desired brominated
compound in 70% yield in a one-pot fashion.9
In conclusion, a simple method for the regioselective
bromination of a range of fused heterocyclic N-oxides has
been developed. The method employs Ts2O as the activiat-
ing agent and TBABr as the nucleophilic bromide source.
The procedure has been conducted on a gram scale, is
applicable to complex molecules (as demonstrated by the
successful bromination of quinine), and has been extended
to the incorporation of chlorine using TBACl. Finally, a
viable one-pot oxidation/bromination procedure was de-
veloped employing MTO/UHP as the oxidant. The use of
thisactivation methodin other contexts suchasiodination,
fluorination, and other functionalizations are clear areas
for additional study.
With improved conditions in hand the scope of the
bromination was examined (Figure 2). A range of quino-
line and isoquinoline derivatives were brominated in
modest to high yields and with excellent regioselectivity for
the isomers shown. Particularly compelling is the bromina-
tion of the antimalarial agent quinine (31), which proceeded
in high yield despite the presence of an olefin, tertiary alcohol,
and tertiary amine. Both electron-rich and -poor substrates
reacted well; furthermore, in the case of halo-substituted
derivatives (20, 21, 26, and 27) no interconversion with the
existing halogen was observed. This methodology also
proved applicable to azaindole 28 as well as fused pyridine
derivative 29, which possesses a sensitive acyl moiety.
Another attractive feature of this method is scalability:
the bromination of 8 was performed on a gram scale with
no loss of regioselectivity and in 75% isolated yield (see
Supporting Information for details). However, to date and
despite our best efforts, the extension of Ts2O/TBABr to
the bromination of simple substituted pyridines has not
been successful and is the subject of ongoing investigation
(Figure 2, 32).
We recognized the potential extension of this method
to the introduction of other halogens simply by switching
the tetrabutylammonium salt. Again using 8 as our model
Acknowledgment. We would like to thank the Deutsche
Forschungsgemeinschaft for a postdoctoral fellowship
(WE 5022/1-1) for A.W. This work was supported by a
collaboration between Bristol-Myers Squibb Company
and The Scripps Research Institute.
Supporting Information Available. Experimental pro-
cedures, copies of all spectral data, and full characteriza-
tion. This material is available free of charge via the
Internet at http://pubs.acs.org.
(8) (a) Coperet, C.; Adolfsson, H.; Khuong, T. A. V.; Yudin, A. K.;
Sharpless, K. B. J. Org. Chem. 1998, 63 (5), 1740–1741. (b) Zhang, Y. Q.;
Guiguemde, W. A.; Sigal, M.; Zhu, F. Y.; Connelly, M. C.; Nwaka, S.;
Guy, R. K. Bioorg. Med. Chem. 2010, 18 (7), 2756–2766.
(9) The bromination conditions are tolerant of up to 5 wt % of the
rhenium catalyst.
Figure 3. (A) Analogous chlorination of 8. (B) One-pot oxidation/
bromination sequence.
The authors declare no competing financial interest.
Org. Lett., Vol. 15, No. 4, 2013
795