Synthesis of N-alkoxybenzimidazoles and N-alkoxypyrimidazoles

Article abstract of DOI:10.1016/S0040-4039(02)01841-5

A variety of novel N-alkoxy aromatic-fused imidazoles have been prepared in a simple two-step process from 2-fluoro nitroaromatics and 2-chloro-3-nitropyridine. The imidazole forming step involves tandem heterocyclisation and O-alkylation with an in situ alkylating agent, and supports prior mechanistic proposals. Additional mechanistic experiments are described. The protocol is versatile with respect to both substrate halo-nitro aromatics and to the nature of the added electrophile (halides) used in the second step, and thereby significantly extends the scope of this reaction and its applicability to diverse synthesis. This methodology can also be used to generate various types of novel N-alkoxypyrimidazoles (4-deazapurine analogues), and an X-ray structure of one such pyrimidazole is presented.

Full text of DOI:10.1016/S0040-4039(02)01841-5

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 7707–7710
Synthesis of N-alkoxybenzimidazoles and N-alkoxypyrimidazoles
John M. Gardiner,a,* Andrew D. Goss,a Tahir Majid,b Andrew D. Morley,c,† Robin G. Pritcharda
and John E. Warrena
aDepartment of Chemistry, UMIST, Manchester M60 1QD, UK
bAventis Pharma US, Route 202-206, Bridgewater, NJ 08807, USA
cAventis Pharma, Rainham Road South, Dagenham, Essex RM10 7XS, UK
Received 5 June 2002; revised 20 August 2002; accepted 30 August 2002
Abstract—A variety of novel N-alkoxy aromatic-fused imidazoles have been prepared in a simple two-step process from 2-fluoro
nitroaromatics and 2-chloro-3-nitropyridine. The imidazole forming step involves tandem heterocyclisation and O-alkylation with
an in situ alkylating agent, and supports prior mechanistic proposals. Additional mechanistic experiments are described. The
protocol is versatile with respect to both substrate halo-nitro aromatics and to the nature of the added electrophile (halides) used
in the second step, and thereby significantly extends the scope of this reaction and its applicability to diverse synthesis. This
methodology can also be used to generate various types of novel N-alkoxypyrimidazoles (4-deazapurine analogues), and an X-ray
structure of one such pyrimidazole is presented. © 2002 Elsevier Science Ltd. All rights reserved.
The 1-hydroxybenzimidazole ring system was first
reported in 1887 (though initially isomerically misas-
signed as an oxaziridine),1 and is in tautomeric equi-
librium with the alternative benzimidazole-N-oxide
structure.2 Syntheses have generally been indirect
because benzimidazoles do not undergo direct N-oxida-
tion to the benzimidazole-N-oxide. Reports of O-alkyl-
ation generating N-alkoxy benzimidazoles have been
rare, with only a handful of prior examples, most giving
N-alkoxy benzimidazoles only as low-yielding side
products.3 We previously described efficient syntheses
of a variety of N-alkoxy benzimidazoles from 2-
nitroanilines in a one-pot tandem process.4 A number
of these compounds showed low or sub-mM anti-HIV
or antitumour activity.5 This process differs substan-
tively from standard condensation and reductive cycli-
sations used almost universally for benzimidazole
construction, and also provides the advantage of con-
comitant introduction of extra functionality on N1.
2-position and the alkoxy group be derived from the
same alkylating agent. We have thus been seeking to
modify this general methodology to provide the 2-sub-
stituent and N-alkoxy group from diverse different
sources. This would establish that a matrix of C2 and
O-alkyl groups could be employed through such a
methodology.
We recently demonstrated that certain N-alkoxybenz-
imidazoles 2 can be O-dealkylated and then re-alkyl-
ated with various types of new electrophiles, providing
a method for diversification to C2, N-alkoxy differenti-
ated systems 3 (Scheme 1). Although this method
involved essentially quantitative O-dealkylation and
realkylation steps without requiring intermediate purifi-
cation, it does require introduction of an N-alkoxy
group and then its removal, and this puts limitations on
the nature of R (i.e. substituents suitable for selective
This protocol has proven versatile for variations in
both nitroaniline substitutions and across a range of
alkylating agents constituting various aryls, allylics and
simple alkyls (Scheme 1), but the sequential/tandem
nature of the process required that substituents at the
* Corresponding author. Tel.: +44 (0)161 200 4530; fax: +001 (510)
217 2371; e-mail: john.gardiner@umist.ac.uk
† Current address: AstraZeneca, Mereside, Alderley Park, Mac-
clesfield, Cheshire, SK10 4TG. Tel: +44 (0)1625 512782; fax: +44
(0)1625 510823.
Scheme 1. O-Dealkylation–realkylation synthesis of N-
alkoxybenzimidazoles.6
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01841-5

7708
J. M. Gardiner et al. / Tetrahedron Letters 43 (2002) 7707–7710
removal are required, e.g. O-aryl, allylic), and thus the
final C2 substituent. Alternatively, we reasoned that
N-alkyl nitroanilines are likely intermediates in the
single-pot tandem reactions. However, these putative
intermediates were only rarely observed, and then only
isolable as trace co-products and only in certain cases.4
We could find no evidence of any build-up of this
intermediate in most cases. However, we found that
4-carboxamide substitution of the 2-nitroaniline system
does allow interception of the intermediate N-alkyl
system in good isolable yields, which could then be
converted onwards to N-alkoxybenzimidazoles by in situ
addition of further base and (different) alkylating agents,
in what we described as a staged tandem sequence.6 This
was, however, specific to this substrate type.
Scheme 2. Reagents and conditions: (i) RCH2NH2, K2CO3; (ii)
NaH, R1CH2Br.
Table 1. Diversity of N-alkoxybenzimidazolesa
Alternatively, we considered that if N-alkyl nitroaniline
intermediates were prepared independently by another
route, they may then undergo tandem heterocyclisation-
O-alkylation in a separate second step with potentially
wide general applicability.
Such a 2-stage route would greatly extend the scope of
the heterocyclisation-O-alkylation methodology and
make accessible a substantively wider diversity of novel
structures with different combinations of the ultimate
2-substituent and the alkoxy group than our previous
methodology.
Herein, we report such a protocol, and also on its
extension to heterocyclic systems, namely the first exam-
ples of N-alkoxypyrimidazole systems (N-alkoxy-4-
deazapurines).
The intermediate alkylated nitroanilines were readily
prepared in near-quantitative yields by standard meth-
ods (reacting 2-halonitroaromatic substrates with vari-
ous primary amines). Treatment of the intermediate
N-alkyl nitroanilines with sodium hydride (2–3 equiv.) in
the presence of various alkylating agents (2 equiv.) did
lead to formation of the desired N-alkoxybenzimida-
zoles, obtained pure by column chromatography
(Scheme 2 and Table 1). This thus supports the proposed
intermediacy of 5 in the previously described one-pot
process, and achieves the objective of providing a new
means to generate diverse novel structures 3 where
R"R1.
was obtained as a competing product. Thus, N-benzyl-
ated substrate 8 was reacted with allyl bromide to afford
the corresponding benzimidazole 12 along with N,N-
dialkylated product 10 (2:1 ratio, ꢀ40% purified yield).
The analogous N-butylated substrate 9 gave dialkylated
material 11 in 60% purified yield as the only significant
isolated product (Scheme 3). This is of mechanistic
relevance as it suggests the electron rich ring methoxy
redirects the reaction pathway away from cyclisation.
Alternative products competed in two cases. Firstly,
when the reaction of 5 (R=PMP; R2=H) was carried
out using propargyl bromide as the electrophile, the
anticipated propargyloxy product 6 was formed. How-
ever, similar reaction of 5 (R=Ph; R2=H) gave the
allenyloxy system 7 as the major product, presumably
arising through base catalysed isomerisation.
The mechanism of the sequential heterocyclisation-O-
alkylation may involve base-induced cyclisation to the
N-oxy intermediate which is then alkylated in situ.
However, alternatively, there could be alkylation of a
nitro-derived oxygen during the process, which then
Secondly, in the case of the 4-MeO N-alkylated nitroani-
line substrates (8 and 9), the N,N-dialkylated material

J. M. Gardiner et al. / Tetrahedron Letters 43 (2002) 7707–7710
7709
Scheme 3. Reagents and conditions: (i) NaH, allyl bromide.
allows for the loss of an alkoxy from nitrogen to
facilitate the final functionalisation (via a second oxy-
gen alkylation).
We carried out additional experiments to illuminate the
mechanism using N-benzylnitroaniline 5 [R=Ph, R2=
H] with allyl bromide as the electrophile. Thus, reac-
tions were carried out using 0, 1 and 2 equiv. of
alkyating agent. In the absence of alkylating agent, the
heterocyclisation proceeded yielding the N-hydroxybenz-
imidazole 13 (Scheme 4). Reaction using allyl bromide
as alkylating agent returned similarly good yields (76–
82%) of N-alkoxybenzimidazole 3 [R=Ph, R1=CH2=
CH2, R2=H], irrespective of whether 1 or 2 equiv. of
alkylating agent were added. Together these results
provide compelling evidence that the mechanism does
not require alkylation of either oxygen prior to, or
during, heterocyclisation, and that the O-alkylation is
likely to be the final step.
Scheme 5. Reagents and conditions: (i) NaH, PrI; (ii) TFA/
CH2Cl2, 2–3 h, rt; (iii) NaH, allylbromide or benzyl bromide.
We have also extended the protocol to 2-chloro-3-
nitropyridine (15) as starting material to provide a
number of new N-alkoxypyrimidazoles 16 (Scheme 6).
The structure of 16 [R=Ph; R1=PMP] was confirmed
by X-ray analysis. The versatility of this chemistry with
this example set of reagents indicates that this should
now be applicable to generate a wide diversity of new
In some cases using carboxamide-bearing precursors,
the heterocyclisation-O-alkylation was accompanied by
concomitant amide N-alkylation (see Table 1; entries 10
and 13), providing more elaborate benzimidazoles.
N-alkoxypyridimidazoles
(N-alkoxy-4-deazapurines)
for biological evaluation.
For example, reaction of
5
[R=Ph, R2=
CONHCH2(2,4-(OMe)2C6H3] with propyl iodide gave
14. Removal of the dimethoxybenzyl group then gave 3
[R=Ph, R1=Et, R2=CONHPr]. Similar chemistry
could be applied in other cases with both the 2-sub-
stituent on the imidazole and the alkylating agent being
varied, for example, to targets such as 3 [R=2,4-DMP,
R1=CH2ꢀCH2 or Ph, R2=CONHCH2R1] (Scheme 5).
In the X-ray structure of 16 [R=Ph, R1=PMP] (Fig. 1)
the 2-phenyl is twisted out of planarity with the bicyclic
heteroaromatic system, tipping C14 and C15 away from
the large N-(p-methoxybenzyloxy) group.
The development of this efficient 2-step protocol signifi-
cantly expands the range of compounds of this type
now accessible by this tandem heterocyclisation-O-
alkylation protocol. It appears to be applicable to a
range of aromatic ring systems, and a wide range of
substituent types can be included at the ultimate 2-posi-
Scheme 4. Reagents and conditions: (i) NaH; (ii) NaH, allyl
bromide (1 or 2 equiv.).
Scheme 6. Reagents and conditions: (i) RCH2NH2, K2CO3;
(ii) NaH, R1CH2Br.

7710
J. M. Gardiner et al. / Tetrahedron Letters 43 (2002) 7707–7710
Figure 1. X-Ray structure of pyrdimidazole 16 [R=Ph, R1=PMP].
tion and in the alkoxy group. This versatility lends itself
to the development of parallel/combinatorial syntheses
of libraries of novel N-alkoxy benzimidazoles. We have
also adapted this approach to facilitate solid-supported
synthesis of libraries of benzimidazole N-alkoxides.7
3. (a) Machin, J.; Mackie, R. K.; McNab, H.; Reed, G. A.;
Sagar, A. J. G.; Smith, D. M. J. Chem. Soc., Perkin Trans.
1 1976, 394; (b) Machin, J.; Smith, D. M. J. Chem. Soc.,
Perkin Trans. 1 1979, 1371; (c) McNab, H.; Smith, D. M.
J. Chem. Soc., Perkin Trans. 1 1973, 1310; (d) Stacy, G.
W.; Wollner, T. E.; Oakes, T. R. J. Heterocycl. Chem.
1966, 51; (e) Cafiero, P. A. C.; French, C. S.; McFarlane,
M. D.; Mackie, R. K.; Smith, D. M. J. Chem. Soc., Perkin
Trans. 1 1997, 1375 and references cited therein; (f)
Machin, J.; Smith, D. M. J. Chem. Soc., Perkin Trans. 1
1979, 1371 and references cited therein.
4. (a) Gardiner, J. M.; Loyns, C. R.; Schwalbe, C. H.;
Barrett, G. C.; Lowe, P. R. Tetrahedron 1995, 51, 4101; (b)
Gardiner, J. M.; Loyns, C. R. Synth. Commun. 1995, 25,
819.
5. (a) Gardiner, J. M.; Loyns, C. R.; Burke, A.; Khan, A.;
Mahmood, N. Bioorg. Med. Chem. Lett. 1995, 5, 1251; (b)
Evans, T. M.; Gardiner, J. M.; Mahmood, N.; Smis, M.
Bioorg. Med. Chem. Lett. 1997, 7, 409; (c) Gardiner, J. M.;
Evans, T. M.; Procter, J.; Smis, M.; McGown, A. T. 2000,
unpublished results.
Acknowledgements
We thank Aventis Pharma for supporting this work
through a CASE award (to A.D.G.). NMR, HPLC and
IR characterisation was carried out on instrumentation
funded with support from EPSRC grants GR/L52246
(NMR), GR/M30135 (IR) and GR/L84391 (HPLC).
The EPSRC Mass Spectrometry Service at Swansea is
thanked for carrying out low and high resolution analy-
ses on a number of the new compounds described.
References
1. (a) von Niementowski, St. Berichte 1887, 20, 1874; (b) von
Niementowski, St. Berichte 1910, 43, 3012.
2. Takahashi, S.; Kano, H. Chem. Pharm. Bull. 1963, 11,
1375.
6. Gardiner, J. M.; Procter, J. Tetrahedron Lett. 2001, 42,
5109.
7. Gardiner, J. M.; Goss, A. D.; Majid, T.; Morley, A. D.
2001, unpublished results.