4614
S. J. Teague et al. / Tetrahedron Letters 46 (2005) 4613–4616
problematical. For instance, a 2-chloro substituent
failed to be displaced by thiol 7 under a range of condi-
tions. This may be the result of competing deprotona-
tion of the 1H-benzimidazole. However displacement
of a 2-methylsulfone moiety proceeded smoothly to give
the required compound 1a in acceptable yield.
required oxindole 11 even though intramolecular
displacement of the fluoro group by the ylide to give
an alternative oxindole is plausible. The oxindole 11
could be reduced to a separable mixture of hydroxy
derivatives 12 and 13. These reacted in differing ways.
The syn-disposition of the H-3 and the 2-hydroxyl
resulted in imine formation and reduction to give the
3-methanesulfanyl-dihydroindole 14. anti-Disposition
resulted in the elimination of water to give the vinyl
sulfide 15.
The quinolinone 1a had unacceptable solubility in the
aqueous buffers used for biological testing. Compounds
were required incorporating a solubilising group in the
5-position of the quinolinone, necessitating the synthesis
of 1b (Scheme 1). The functionalisation of this position
also enabled an alternative disconnection of the thio-
ether linkage to be employed involving nucleophilic
attack of a 2-thiobenzimidazole at the 8-position of
the quinolinone.
Oxidation of 15 gave the required vinyl sulfone 16
(Scheme 5). Displacement of the fluoro group by the
2-thiobenzimidazole and reduction of the nitro group
completed the required synthesis of 2.
Incorporation of a 4-substituent on the benzimidazole
was anticipated to be acceptable from knowledge of
the binding orientation of these compounds relative to
that of ATP. This was based upon X-ray crystallo-
graphy of the compounds complexed with the target
protein. The group in the 4-position utilises the space
occupied by the ribose unit of ATP. However, the method-
ologies available to elaborate 4-substituted 2-thiobenz-
imidazoles were limited.
The required quinolinone 9 (Scheme 3) was synthesised
by annulation of the aniline 8 with 5. This formally elec-
trophilic cyclisation was surprisingly still effective when
carried out upon the very electron deficient aniline 8.
Displacement of the fluoro group to give 10 using the
benzimidazole thiolate gave the required product even
though deprotonation of the acidic 1H-quinolone might
have been anticipated. Reduction of the nitro group
gave 1b which was reductively aminated with a variety
of polar aldehydes to introduce solubilising substituents.
The required 4-nitrobenzimidazole 18 (Scheme 6) was
prepared using the known regio-selective nitration of
the seleno-protected dianiline 17.7
Replacements for the quinolinone moiety were required
including indole 2 (Scheme 1).
Reduction and diazotisation gave the iodide 19 (Scheme
7). This is the first reported synthesis of a 4-diazo-benz-
imidazole. The iodide underwent Heck type reactions.
The Gassman indole synthesis5 (Scheme 4) when car-
ried out upon the fluoro-nitroaniline 8 gave the
O
O
H
N
Cl
SH
NO2
S
O
O
O
NO2
O
NO2
5
N
MeO
(1b)
N
H
a, b
c
d
H
N
N
H
NH2
Cl
F
9
F
8
10
N
Scheme 3. Reagents, conditions and yields: (a) 5, MeCN, rt, 20 h, 100%; (b) Ph2O, 61%;4 (c) 2-thiobenzimidazole, 10 M NaOH, EtOH, 100 °C, 20 h,
40%; (d) H2, 10%Pd/C, AcOH, 5bar, 8 days, 77%.
O2N
O2N
SMe
NO2
NO2
F
SMe
OH
SMe
H
CO2Et
-
N
S+
O
N
H
d
c
N
H
N
a
b
H
H
F
F
F
14
H
12
H
(8)
11
O2N
SMe
NO2
O2N
SMe
OH
N
H
H
H
N
N
H
S+
F
F
F
-
13
15
CO2Et
Scheme 4. Reagents, conditions and yields: (a) MeSCH2CO2Et, SO2Cl2, Proton-Sponge, DCM;6 (b) Et3N, À70 °C, 1 h then rt, 20 h, 70% over two
steps; (c) 1 M BH3, THF, 0 °C, 15min; (d) rt, 20 h, 77% (41% 14, 37% 15).
Teague, Simon J.
