1
176
Synlett
M. Moreno et al.
Letter
Most of these strains are made responsible for nosoco-
Br
O
Br
a) BBr3, CH2Cl2, –30 °C to r.t., 17 h (95%)
b) NaBH4, THF, –40 °C to r.t., 30 min (91%)
c) PhCH(OMe)2, pTsOH⋅H2O, THF, r.t., 5 d (56%)
2
1
mial infections. Preliminary studies showed that the cys-
tobactamids target bacterial type IIa topoisomerases which
are validated antibacterial targets. However, as quinolones
are not suited anymore to serve as template for new inhibi-
tors the cystobactamids offer new opportunities in search
H
O
OH
O
Ph
OMe
OH
4
5
3
e) i-PrOH, DEAD, Ph P, THF,
for new anti-infectives, especially as this novel structural
3
r.t., 17 h (85%)
Br
scaffold and the limited cross-resistance found make the
cystobactamids promising lead structures. Structurally,
cystobactamid 507 (3) is the simplest member. It was re-
ported to exert similar but lower inhibitory activity than
cystobactamids 1 and 2.
En route to 1 and 2 we synthesized cystobactamid 507
3) which additionally would allow us to further test its bio-
logical properties. These tests would show whether the tris-
aryl unit is an essential element for all cystobactamids and
furthermore would clarify if minor impurities present in
the natural sample did alter the assay read-outs.
d) Ni(NO3)2⋅5H2O, pTsOH⋅H2O,
acetone, r.t., 2.5 h (74%)
f) Pd2(dba)3, (PhO)3P, i-PrOH,
dioxane, 80 °C, 1.5 h (70%)
O
O2N
O
Ph
OH
6
g) camphor-10-sulfonic acid,
O
CH2Cl2–MeOH (1:2), 0 °C to r.t., 17 h (90%)
h) MnO2, CH2Cl2, r.t., 17 h (81%)
(
O
OH
OH
i) 2-methyl-2-butene, NaClO2, NaH2PO4,
t-BuOH, r.t., 17 h (75%)
O2N
O
Ph
O2N
O
O
7
8
Although cystobactamids only contain p-aminobenzoic
acids we experienced two major synthetic challenges: a)
accessing the tetrasubstituted arene unit and b) the lack of
reactivity of anilines and the lack for reactivity of ortho-
substituted phenolic and isopropoxy benzoic acids in am-
ide formations. This amide formation can only proceed un-
der conditions that are different from those established in
peptide synthesis.
O
j) TMSCHN2, MeOH–PhMe, 0 °C to r.t., 1 h (80%)
k) BnOH, DEAD, Ph3P, THF, r.t., 17 h (70%)
l) LiOH, THF–H2O (1:1), r.t., 17 h (95%)
OH
OBn
O2N
O
9
Scheme 1 Synthesis of tetrasubstituted benzoic acid 9
The synthesis of the tetrasubstituted arene 9 com-
menced with o-bromobenzaldehyde 4. The bromo substitu-
ent served as ‘dummy’ group which can be removed after
the selective introduction of the nitrogen functionality at
C4. In addition, this starting material allows for differentiat-
ing between the two phenolic groups and thus enables se-
lective introduction of the isopropyl group at C3 (Scheme
tablished in peptide chemistry gave poor coupling yields.
We made the bulky isopropoxy substituents with ortho ori-
entation to the amino groups as well as the reduced reactiv-
ity of the aromatic amino groups responsible for the diffi-
culties to achieve amide formation.
We found that Ghosez’s reagent 136 is best suited to
couple benzoates with anilines (Scheme 2). First, aniline 11,
which straightforwardly is accessible from benzoic acid 10,
1).
7
O-Demethylation of 4, aldehyde reduction and protec-
was coupled with tetrasubstituted benzoic acid 9 to yield
tion of the benzylic alcohol and one phenolic group as ben-
zylidene acetal yielded phenol 5. This protection paved the
way for introducing a) the nitro group to yield nitroarene 64
and b) the isopropyl group. Palladium-catalyzed debromi-
amide 12 after hydrogenation of the nitro group and with
concomitant removal of the benzyl protecting group. Next,
8
the second amide coupling between 12 and p-nitrobenzoic
acid 14 yielded cystobactamid 507 (3) after simultaneous
5
9
nation yielded nitroarene 7. After removal of the ben-
reduction of the nitro group and removal of the tert-butyl
zylidene protection, the benzylic alcohol was transformed
into the carboxylate under standard conditions. The result-
ing benzoic acid 8 was temporarily methylated in order to
protect the o-phenolic group, which finally furnished the
desired p-nitrobenzoic acid 9. These final steps were crucial
as successful amide formation could only be achieved if the
o-phenolic group is protected.
With this key building block in hand, we could finalize
the synthesis of cystobactamid 507 (3) by coupling three p-
aminobenzoic acid units. However, we had to search for
conditions that allow for creating an amide bond between
two arene moieties. Common reagent systems such as
HOAt, EDC, or a mixture of HOBt and EDC that are well es-
ester. It has to be noted that it became necessary to switch
from a methyl to a tert-butyl ester (10 → 11) because the
final ester hydrolysis with the corresponding methyl ester
(step under basic conditions) led to simultaneous amide
hydrolysis of the p-aminobenzoic acid moiety.
The NMR spectra and chromatographic parameter
(HPLC) for the synthetic material were identical to those
collected for natural cystobactamid 507 (3).
The synthetic hurdles that we encountered in this syn-
thesis stemmed us to prepare a structurally simplified deri-
vate 18 in which both isopropoxy groups of cystobactamid
507 are replaced by the smaller methoxy groups. Starting
from o-vanillin the tetrasubstituted central arene unit 15
©
Georg Thieme Verlag Stuttgart · New York — Synlett 2015, 26, 1175–1178