Inhibition of FLT3 and PDGFR tyrosine kinase activity by bis(benzo[b]furan-2-yl)methanones

Article abstract of DOI:10.1016/j.bmc.2006.12.011

A series of bis(benzo[b]furan-2-yl)methanones was synthesized and tested for inhibition of FLT3 and PDGFR autophosphorylation. Mostly, C-5 substitution leads to PDGFR selectivity, which was strongest in the case of the 5,5′-dimethoxy derivative. The 5,5′-

Full text of DOI:10.1016/j.bmc.2006.12.011

Bioorganic & Medicinal Chemistry 15 (2007) 2187–2197
Inhibition of FLT3 and PDGFR tyrosine kinase activity
by bis(benzo[b]furan-2-yl)methanonesq
a
a
Siavosh Mahboobi,a, , Andrea Uecker,b, Christophe Cenac, Andreas Sellmer,
*
´
Emerich Eichhorn,a Sigurd Elz,a Frank-D. Bo¨hmerb, and Stefan Dovea,*,
aInstitute of Pharmacy, Faculty of Chemistry and Pharmacy, University of Regensburg, D-93040 Regensburg, Germany
bInstitute of Molecular Cell Biology, Jena University Hospital, D-07747 Jena, Germany
Received 1 June 2006; revised 29 November 2006; accepted 8 December 2006
Available online 12 December 2006
Abstract—A series of bis(benzo[b]furan-2-yl)methanones was synthesized and tested for inhibition of FLT3 and PDGFR autophos-
phorylation. Mostly, C-5 substitution leads to PDGFR selectivity, which was strongest in the case of the 5,50-dimethoxy derivative.
The 5,50-diamino and the 6,60-dihydroxy compounds are more active at FLT3. At both kinases, the potency of the best inhibitors
approaches IC50 values of ca. 0.5 lM. Molecular modeling studies suggest that the bisbenzofuranylmethanones are able to fit into
the same binding site as their indolyl analogues which have been suggested to form a bidentate hydrogen bridge with the backbone
in the hinge regions. The loss of one H bond by the NH–O exchange might be partially compensated by, for example, the weak
interaction of one furanyl oxygen with FLT3 Cys-828.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
review see lit.4). The FLT3 receptor is crucial for the
maintenance, proliferation, and differentiation processes
in haematopoiesis. FLT3 is expressed by normal mye-
loid and lymphoid early progenitors as well as by leuke-
mic cells.5 Aberrant expression and/or mutation of
FLT3 in acute myeloid leukemia (AML) and acute lym-
phoblastic leukemia (ALL) are the reasons for the great
interest currently shown in the development of FLT3
inhibitors (for review, see lit.6,7).
The platelet-derived growth factor receptor (PDGFR)
and the Fms-like tyrosine kinase 3 (FLT3) have been
implicated in a number of pathologies, especially in var-
ious cancers, and are therefore considered as drug tar-
gets (for review, see lit.1,2). Both are members of the
class III receptor tyrosine kinase (RTK) family, which
also includes the receptors for the stem cell factor
(c-KIT), and for the colony stimulating factor 1
(CSF1R). The PDGF receptor is involved in wound
healing and regulation of homeostasis of the connective
tissue compartment. It is expressed on early stem cells,
mast cells, myeloid cells, mesenchymal cells, and smooth
muscle cells.3 Overactivity/-expression of PDGFR has
been implicated in malignancies as well as in different
diseases involving excessive cell growth such as athero-
sclerosis and fibrosis. Therefore, a number of inhibitors
targeting PDGFR are presently under development (for
Considering the structural similarities between FLT3
and PDGFR it is not too surprising that most of the
small molecule inhibitors targeting one kinase are also
more or less active against the other. Among these com-
pounds are the quinoxalines AG1295,8 AG1296,8 and
AGL2043,8 the piperazinylquinazoline MLN518 (Tan-
dutinib),9 the indolocarbazole PKC412 (Midostaurin),10
the indolinones SU5416 (Semaxanib)11 and SU11248
(Sunitinib)12 as well as the diarylurea BAY43-9006
(Sorafenib)13 (Fig. 1). However, the indolocarbazole
CEP701 (Lestaurtinib)14 and the benzimidazolylquinoli-
none CHIR25815 show highly significant selectivity for
FLT3 versus PDGFR. On the other hand, the Abl/
Bcr-Abl, c-KIT, and PDGFR inhibitor STI571 (Imati-
nib) is inactive against FLT3.16,17
Keywords: Bisbenzofuranylmethanone; Receptor tyrosine kinase; FL T3;
PDGFR; Bisindolylmethanone; Benzofuranylindolylmethanone.
q
This paper is dedicated to Professor Wolfgang Wiegrebe on the
occasion of his 75th birthday.
*
Corresponding authors. Tel.: +49 0941 943 4824; fax: +49 0941 943
1737; e-mail addresses: siavosh.mahboobi@chemie.uni-regensburg.
de; stefan.dove@chemie.uni-regensburg.de
We have previously identified bis(1H-indol-2-yl)metha-
none (1a) as a lead for the development of novel class
These authors contributed equally to this work.
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2006.12.011

2188
S. Mahboobi et al. / Bioorg. Med. Chem. 15 (2007) 2187–2197
O
N
R
R
N
N
N
N
S
N
N
H
N
N
N
N
AG1295 (R = Me)
AG1296 (R = OMe)
AGL2043
MLN518
O
R2
O
N
H
H
N
N
R1
N
O
O
H
O
N
H
N
N
N
N
SU5416 (R1 = H, R2 = H)
SU11248 (R1 = F, R2 = CONHC2H4NEt2)
O
N
O
CEP701
HO
O
O
OH
FF
H
N
H
N
PKC412
F
N
H
N
O
Cl
O
N
O
N
BAY43-9006
F
N
H2N
O
N
STI571
N
N
HN
N
N
O
N
H
NH
HN
CHIR258
Figure 1. Small molecule inhibitors of FLT3 and/or PDGFR.
III RTK inhibitors18 (cf. Fig. 2 and Table 1). PDGFR
inhibition could be improved by hydroxy, methoxy,
and carboxy substituents in 5-position of one indole
antiviral agents,21–24 as possible aldose reductase inhib-
itors, and platelet aggregation inhibitors,23 or as charge-
generating compounds in the photosensitive layer of
photoreceptors.25 As described here, PDGFR and
FLT3 can also be inhibited by bisbenzofuranylmetha-
nones. Most of these compounds are less potent than
their corresponding bisindolylmethanone homologues,
however, the unsubstituted leads are nearly equiactive.
Therefore, one of the suggested hydrogen bonds which
is disabled by the NH–O exchange should be replaced
by another interaction. This may be an H bond of one
of the oxygens, since the corresponding bisbenzothi-
ophenyl analogue is inactive.18 The resulting specific
binding mode of the bis(benzo[b]furan-2-yl)methanones
alters the structure–activity and structure–selectivity
relationships observed in the bis(1H-indol-2-yl)metha-
none series.
moiety18 (e.g., 1b: IC50 = 0.3 lM and 1c: IC50
=
0.2 lM). Further studies on disubstituted derivatives
resulted in potent FLT3 inhibitors bearing at least one
polar group at C-5, as for example, compound 1e. The
enhanced FLT3 inhibition is associated with a reason-
able selectivity versus PDGFR19 (cf. Table 1).
The bisindolylmethanones presumably dock into the
ATP-binding pocket of PDGFR and FLT3 by forming
a bidentate hydrogen bond with the backbone of
PDGFR Cys-684 and FLT3 Cys-694, respectively,
involving one indole NH and the CO of the methanone
bridge.18,19 Symmetrical bis(benzo[b]furan-2-yl)metha-
nones are already known as Tilorone analogues20 and
R1
R 2
H
2. Chemistry
R1
R2
H
1a
1b
1c
1d
1e
The first stage of our study consisted in the synthesis of
the oxygen isostere of bisindolylmethanone 1a, and a
series of methoxy-disubstituted bisbenzofuranylmetha-
nones. Chemical modification was then performed in
positions 5, 6, and 7. Salicylaldehyde as well as its 6-,
5-, 4-, 3-methoxy and 5-nitro derivatives were condensed
with 1,3-dihaloacetone in order to obtain the corre-
sponding symmetric bisbenzofuranylmethanones 2–6
OMe
OH
H
N
H
N
H
O
OMe
OH
OMe
OH
Figure 2. Bis(1H-indol-2-yl)methanone (1a) and selected derivatives
with tyrosine kinase inhibition properties.

S. Mahboobi et al. / Bioorg. Med. Chem. 15 (2007) 2187–2197
2189
Table 1. Inhibition of PDGFR and FLT3 tyrosine kinases by bis(benzo[b]furan-2-yl)methanones and related compounds
Compound
X
Position
R
PDGFRa IC50 (lM)
FLT3b IC50 (lM)
R
R
X
X
O
1ac,d
1dc
1ed
2
NH
NH
NH
O
—
5
H
1
4.6
OCH3
OH
10–30
0.3
1.2
>10
0.04
3.5
5
—
4
H
3
O
OCH3
OCH3
OCH3
OCH3
OH
>30
1.0
>30
>30
>30
>30
3.6
4
O
5
5
O
6
>30
>30
0.6
6
O
7
7
O
5
8
O
6
OH
OH
9.7
1.8
9
O
7
20–30
0.7
1.2
>30
1.7
10
12
O
5
OCOCH3
NH2
O
5
0.6
R
R
N
H
O
O
23
25
H
OH
1.07
0.20
0.40
0.27
N
N
AG1295
5.4
1.1
a Assayed with endogenous PDGFR in Swiss 3T3 fibroblasts.
b Assayed with endogenous FLT3 in EOL-1 cells.
c Previously published results on PDGFR inhibition.18
d Previously published results on PDGFR and FLT 3 inhibition.19
and 11 in one step according to Royer et al.20 (see
Scheme 1). The 5-, 6-, and 7-OMe-disubstituted
bis(benzo[b]furan-2-yl)methanones were then demethy-
lated with pyridine hydrochloride under microwave irra-
diation in order to obtain their respective
dihydroxylated derivatives 7–9. Subsequent acetylation
of the 5,50-dihydroxy derivative led to the diester 10.
Hydrogenolytic palladium-catalyzed reduction of the di-
nitro-functionalized compound 11 led to the diamine 12.
21 and 22, which were deprotected by NaOH to yield the
desired hybrid derivatives 23 and 24. In case of 24 hydrog-
enolytic cleavage of both benzyloxy groups18,19 was
subsequently performed to give the 5,50-diOH compound
25 in excellent yield.
The respective benzofurancarbonyl chlorides used for the
synthesis (1727 and 18) were prepared by chlorination of
their corresponding carboxylic acids (Scheme 2). The
5-benzyloxybenzo[b]furan-2-carboxylic acid 1628 hereby
was accessible in four steps (see Scheme 3) from 5-hy-
droxy-2-methoxybenzaldehyde as follows: ring closure
with bromoacetic acid ethyl ester led to the 5-meth-
oxybenzo[b]furan-2-carboxylic acid ethyl ester 13,29 sub-
sequent demethylation with boron tribromide, followed
by benzylation of the phenol intermediate 14,29 led to
15.30 Alkaline cleavage of the ethyl ester group afforded
the desired compound 16 in the following.
In addition, we prepared hybrid benzofuranylindolyl-
methanones to validate the proposed SARs and binding
mode (derivatives 23 and 25). The synthetic route26 is
shown in Scheme 2: metalation of the suitable N-protect-
ed indoles (19 and 20) by use of n-butyllithium in THF
solution, followed by coupling with the respective benzo
[b]furan-2-carbonyl chlorides, yielded the (benzo[b]-
furan-2-yl)-(1-phenylsulfonyl-1H-indol-2-yl)methanones

2190
S. Mahboobi et al. / Bioorg. Med. Chem. 15 (2007) 2187–2197
R
6
3
4
R
R
R
H
2
CHO
OH
5
4
5
6
a
OMe
OH
3-6
7-9
10
11
12
b
c
O
O
7
O
5-OAc
5-NO2
5-NH2
d
Scheme 1. Synthesis of bisbenzofuranylmethanones. Reagents and conditions: (a) (ClCH2)2CO or (BrCH2)2CO, K2CO3, EtCOMe, D; (b) pyridinium
chloride, MW; (c) AcCl, DMAP, AcOMe, pyridine; (d) HCO2NH4, Pd/C, iPrOH, D.
3. Results and discussion
R
R
The biological activities of the bis(benzo[b]furan-2-
yl)methanones are summarized in Table 1, with excep-
tion of the 5,50-dinitro derivative 11 whose solubility
was too low to proceed with the usual testing proce-
dures. The (benzo[b]furan-2-yl)-(1H-indol-2-yl)metha-
none (23) and its 5,50-dihydroxy derivative 25 were
also tested for RTK inhibition in order to probe the
binding mode hypothesis. The quinoxaline AG12958
was used as a standard.
O
CO2H
N
H
i
j
R
R
O
COCl
N
PhO2S
R = H
17
R = OBn 18
R = H 19 ; R = OBn 20
The unsubstituted bisbenzofuranylmethanone 2 inhibit-
ed both kinases with similar potency and with the same
selectivity for PDGFR as the bisindolylmethanone 1a.
The PDGFR inhibition of the 5,50-diOMe-substituted
bisbenzofuranylmethanone derivative 4 is in the same
range (IC50 = 1.0 lM) and associated with significant
selectivity versus FLT3 (IC50 > 30 lM). In contrast, its
bisindolyl analogue 1d is inactive at both kinases. The
4,40-, 6,60-, or 7,70-diOMe-substituted bisbenzofuranyl-
methanones 3, 5 and 6 did not inhibit PDGFR and
FLT3 in concentrations up to 30 lM. As in the case of
the bisindolylmethanones, 5- or 6-dihydroxy is more
favorable than 5- or 6-dimethoxy substitution. The
6,60-diOH derivative 8 is slightly FLT3 selective, attrib-
uted to reduced PDGFR inhibition (IC50 = 9.7 lM),
and a twofold higher potency against FLT3 compared
to the unsubstituted compound. In contrast the 5,50-
diOH positional isomer 7 is sixfold more potent at
PDGFR than at FLT3. These results are opposed to
the data of the corresponding bisindolylmethanones
where 5,50-diOH substitution leads to significant FLT3
selectivity (cf. 1e), whereas polar 6-substituents mostly
reduce FLT3 inhibition.19 The 7,70-diOH derivative 9
was inactive at FLT3 and PDGFR. Surprisingly acetyla-
tion of the 5,50-diOH derivative did not significantly
k
R
R
R = H
R = OBn
21
22
O
N
O
O
O
SO2Ph
l
R
R
R = H
23
R = OBn 24
O
N
H
m
HO
OH
25
(from 24)
O
N
H
Scheme 2. Synthesis of benzofuranylindolylmethanones. Reagents and
conditions: (i) SOCl2 (1.04 equiv), Pyridine (1.1 equiv), DCM; (j)
1—NaH, THF; 2—PhSO2Cl, 0 ꢁC; (k) nBuLi, THF, ꢀ78 ꢁC; (l) 1—
NaOH, MeOH, THF, D; 2—HCl; (m) HCO2NH4, Pd/C, MeOH, D.
R
Y
R
O
CHO
OH
OEt
OEt
OMe
OH
13
14
15
16
e
f
Y
O
g
h
OBn
OBn
OEt
OH
O
Scheme 3. Synthesis of the 5-benzyloxybenzo[b]furan-2-carboxylic acid 16. Reagents and conditions: (e) BrCH2CO2Et, K2CO3, DMF, D; (f) BBr3
(2.5 equiv), DCM, 0 ꢁC; (g) BnBr (1.2 equiv), K2CO3 (4 equiv), Me2CO, D; (h) KOH [2 N], H2O, MeOH, THF.

S. Mahboobi et al. / Bioorg. Med. Chem. 15 (2007) 2187–2197
2191
ylmethanone derivatives may bind to the ATP site in a
mode resembling the interaction of PD-173074 in a crys-
tallized complex with the FGFR-132 (see Fig. 3). The
key interactions are two hydrogen bonds of an indole
NH and the methanone oxygen with the CO and the
NH moiety, respectively, of the backbone of FLT3
Cys-694 (PDGFR Cys-684). Bisindolylmethane deriva-
tives are inactive.19 The bidentate hydrogen bond do-
nor–acceptor system results in one indole ring pointing
inwards and the other pointing outwards, and thereby
enables the definition of an inner and an outer binding
pocket. The outer indole is open for substitution in
positions 5 and 6 even with larger polar, for example,
piperidinoethoxy or quinolylcarboxy groups.18,19 The
substituents are surrounded by short chains of the b1
strand and of the hinge region, and protrude into the
solvent.
The question arises whether the bis(benzo[b]furan-2-
yl)methanones may bind to the ATP site of FLT3 and
PDGFR similarly as the bisindolylmethanones even
though one of the H bonds, namely the one disabled by
the NH–O exchange, cannot be formed. Since the unsub-
stituted leads 1a and 2 are nearly equiactive, this hydro-
gen bond should be replaced by another interaction,
possibly an H bond of one of the oxygens, since the cor-
responding bisbenzothiophenyl analogue is inactive.18
To check this hypothesis, the 5,50-diOH derivative 7
was docked into the putative bisindolylmethanone bind-
ing site of the FLT3 crystal structure 1rjb (see Fig. 4).
Figure 3. Comparison of (A) the binding mode of PD-173074 at the
FGFR-1 with (B) the suggested binding mode of the bisindolylmeth-
anone 1e at FLT3. The crystal structure of the FGFR-1 PD-173074
complex (PDB 2fgi) was aligned with the minimized FLT3 structure
(from PDB 1rjb) containing 1e. For the superposition, the backbone
˚
atoms of the FLT3 residues lying within 4 A around 1e were used as
template for the corresponding FGFR-1 amino acids (rms without the
Generally each benzofuranyl moiety may be synperipla-
nar (sp, cis) or antiperiplanar (ap, trans) with the car-
bonyl function if the OC–CO torsion angles are taken
as reference. AMPAC 8.15 calculations of compound
7 with the COSMO water solvation model result in the
lowest energy for a ap/sp planar conformation. Howev-
er, the sp/sp and ap/ap conformers are only 0.12 and
0.72 kcal/mol, respectively, higher in energy. The rota-
tional barrier around one of the OC–CO bonds amounts
to 2.7 kcal/mol. Also in the unsolvated state, the ap/sp
conformer represents the global minimum. In conclu-
sion, each binding conformation with nearly coplanarity
of both benzofuranyl moieties is energetically possible,
but an ap/sp geometry is preferred.
˚
strongly deviating Phe-830/Phe-642 in the DFG motif: 0.96 A). FLT3
Phe-691 (FGFR-1 Val-561) is the only residue which differs from
PDGFR (Thr-681). Arrows, salt bridge between Glu and Lys; dotted
lines, bidentate hydrogen bonds.
change activity against both kinases (compd 10). 5,50-
DiNH2 substitution (compd 12) improved FLT3 inhibi-
tion, and maintained PDGFR inhibition compared to
the unsubstituted and the 5,50-diOH analogues.
The docking mode in Figure 4 shows the outer benzofu-
ranyl moiety in ap orientation since the furanyl oxygen
should not point to the backbone O atom of Cys-694.
Then, however, the inner benzofurane ring may be sp
with the carbonyl function. This conformation may en-
able the inner furanyl oxygen to form a weak H bond
with the SH group of Cys-828 preceding the DFG motif
The hybrid benzofuranylindolylmethanone 23 was
significantly more potent as FLT3 and equiactive as
PDGFR inhibitor compared to its ‘pure’ indolyl and
benzofuranyl analogues (1a and 2). 5,50-Dihydroxy
substitution (compd 25) slightly improves activity at
both kinases, but does neither lead to FLT3 nor to
PDGFR selectivity like in the case of the corresponding
bisindolylmethanone 1e and bisbenzofuranylmethanone
7, respectively.
˚
(Oꢁ ꢁ ꢁHS distance 2.25 A) and the lipophilic edge of the
benzofurane moiety to fit to a hydrophobic pocket
(Val-624, Ala-642).
More detailed discussions of the SAR must consider the
probable binding mode and the symmetry of the
bis(benzo[b]furan-2-yl)methanone scaffold. As suggested
from a homology model of the PDGFR,18 from the re-
cent crystal structure (PDB 1rjb) of FLT3,31 and from
SAR of bis(1H-indol-2-yl)methanones,18,19 the bisindol-
Compound 7 fits to a bank of four residues, Phe-691 to
Cys-694, belonging to b5 and the hinge region. Within
˚
3 A around the ligand, Phe-691 is the only mutated ami-
no acid compared to the PDGFR (Thr-681) and, there-
fore, the primary candidate for explaining selectivity.
Lys-644 (b3) and Glu-661 (aC), forming a salt bridge

2192
S. Mahboobi et al. / Bioorg. Med. Chem. 15 (2007) 2187–2197
(b3) from the top and of Val-624 (b2), Leu-818 (b7),
and Phe-830 (DFG-motif) from the bottom. The outer
benzofuranyl moiety interacts with the N-terminal
domain residues Leu-616 (b1) and Tyr-693 (hinge)
as well as with Phe-830.
The suggested ap/sp binding mode of the bisbenzofura-
nylmethanones is directly opposed to the proposed sp/
ap mode of the bisindolyl compounds (compare
Fig. 3B and Fig. 4B) which best explains the struc-
ture–selectivity relationships in this series.19 The differ-
ent conformations formally result in an alignment of
the indolyl-5 with the benzofuranyl-6 position and vice
versa. In some respects this corresponds to the opposite
impact of 5- and 6-hydroxy groups on PDGFR and
FLT3 inhibition: 5-OH indolyl and 6-OH benzofuranyl
structures tend to FLT3, 5-OH benzofuranyl and many
6-substituted indolyl moieties19 to PDGFR selectivity.
The preference of the 5-hydroxy (7), 5-methoxy (4),
and 5-acetoxy (10) benzofuranyl moieties for PDGFR
may be explained by electrostatic interaction of the O
atoms with the OH group of Thr-681. On the other
hand, the inactivity of compounds 4 and 5 at FLT3
should be based on repulsion between the methoxy
substituents and Phe-691.
Figure 4 shows that the outer benzofuranyl moiety pro-
trudes into the solvent. Obviously, the 5- and 6-positions
are free and remain solvated in the bound state. Substit-
uents in 4- and 7-positions are sterically hindered by b1
and the hinge region like in the case of the bisindolyl-
methanones,19 leading to inactivity of compounds 3, 6
and 9 at both kinases.
The higher FLT3 inhibition of the hybrids (compare 23
to 1a and 2, 25 to 7) indicates an sp/sp binding mode
with formation of the bidentate H bond like in the case
of the bisindolylmethanones. The inner 50-OH-benzofu-
ranyl moiety of compound 25 does probably not similar-
ly fit to the binding site like the 50-OH-indolyl moiety of
1e suggested to bind in sp/ap mode, since 1e is by a fac-
tor of seven more active than the hybrid 25. As men-
tioned above, this ap conformation is unfavorable for
an inner benzofuranyl moiety (no H bond with the SH
of Cys-828). In the case of PDGFR inhibition, the hy-
brids 23 and 25 do not improve the activity of the
‘‘pure’’ analogues. This is not surprising since the poten-
cy of the latter is very similar, too (cf. 1a with 2, 1e with
7). Thus, no additional information about the PDGFR
binding mode can be obtained from the present data
of the hybrids.
Figure 4. Suggested binding mode of the bisbenzofuranylmethanone
derivative 7 at FLT3 demonstrated by a minimized complex with the
FLT3 crystal structure 1rjb.31 (A) Overview of the model with compd 7
and with representation of some important amino acids. Specific
regions are highlighted on the ribbon/tube backbone: light cyan—
inhibitor binding site consisting of parts of b5 and the hinge region,
yellow—nucleotide binding loop, red orange—catalytic loop, green—
activation loop. Other important secondary structure elements are
indicated at the backbone. The view is from the C-terminus and
corresponds approximately to the views in Fig. 3. The C-terminal part
after the activation loop is not displayed. (B) Detailed view of the
binding site of compd 7 with all amino acids lying within a sphere of
The model depicted in Figure 4 does not explain all of
the structure–activity and structure–selectivity relation-
ships. Detailed topological differences of the binding
sites remain unsolved due to the lack of a PDGFR crys-
tal structure. The relatively high activity of the 5,50-
diacetoxy derivative 10 at FLT3 is somewhat surprising
compared to its inactive 5,50-dimethoxy analogue 4.
However, the binding site offers some degrees of free-
dom in the direction of the 5- and 6-positions of the
inner benzofuranyl moiety, as shown by the high inhib-
itory potency of STI-571 at the FLT3 Phe-691-Thr
˚
3.5 A around the ligand. Backbone atoms are dark, side chains light
cyan. C atoms orange—Phe-691. Yellow lines—hydrogen bonds of
Cys-694 and Cys-828 with the ligand. Purple line—salt bridge.
as in other tyrosine kinases, are close to the 5- and
6- positions of the inner benzofuranyl moiety whose
binding site is completed by the side chains of Ala-642

S. Mahboobi et al. / Bioorg. Med. Chem. 15 (2007) 2187–2197
2193
mutant.17 Also the lower FLT3 activity of the 5,50-dihy-
droxy derivative 7 versus the 5,50-diamino analogue 12
cannot be explained by the model since both compounds
should be able to interact with Glu-661 (distance ca.
was filtered, the resulting solution was washed with
water and evaporated. The obtained residue was sub-
jected to column chromatography (SiO2, ethyl acetate)
and then recrystallized.
˚
2.8 A). The SAR as well as the modeling of the new ser-
ies do not provide further reasons for the extraordinary
high and selective inhibition of FLT3 by the 5,50-diOH
bisindolylmethanone 1e. Generally the possibility can-
not be ruled out that in some cases the measured IC50
values are based on a mixture of different bound confor-
mations and/or even different binding modes at both
kinases.
5.1.2.2. Bis(benzo[b]furan-2-yl)methanone (2). Prepa-
ration from salicylaldehyde (Lancaster) as described
above, using 1,3-dichloropropan-2-one. Yield: 9.62 g
(60%) of colorless crystals. Mp 156 ꢁC (from dichloro-
methane), lit.33 155–157 ꢁC. 1H NMR (DMSO-d6): d
(ppm) = 7.40 (dd, 2H, J = 7.7 Hz, 7.1 Hz), 7.59 (dd,
2H, J = 8.2 Hz, 7.1 Hz), 7.79 (d, 2H, J = 8.2 Hz), 7.90
(d, 2H, J = 7.7 Hz), 8.22 (s, 2H).
4. Conclusions
5.1.2.3. Bis(4-methoxybenzo[b]furan-2-yl)methanone (3).
Preparation from 2-hydroxy-6-methoxybenzaldehyde
(Lancaster) as described above, using 1,3-dibromopro-
pan-2-one. Yield: 1.59 g (15%) of light yellow crystals.
Mp 195–196 ꢁC (from acetone), lit.22 181–183 ꢁC. 1H
NMR (DMSO-d6): d (ppm) = 3.97 (s, 6H), 6.92 (d, 2H,
J = 8.0 Hz), 7.40 (d, 2H, J = 8.5 Hz), 7.54 (dd, 2H,
J = 8.5 Hz, 8.0 Hz), 8.10 (s, 2H).
Bis(benzo[b]furan-2-yl)methanones are FLT3 and
PDGFR tyrosine kinase inhibitors exhibiting IC50 values
in the low micromolar range. The unsubstituted com-
pound 2 is as active as the corresponding bis(1H-indol-
2-yl)methanone 1a. In contrast to the former bisindolyl-
methanone series, the potency could not be strongly im-
proved by introduction of substituents. Most effective
was a 5,50-diamino substitution (compd 12 with sixfold
lower IC50 value than 2 against FLT3). A high degree of
selectivity for PDGFR versus FLT3 was observed for
the 5,50-dimethoxy-substituted derivative 4. At both ki-
nases, the bisbenzofuranyl- and their bisindolylmetha-
none analogues are suggested to fit to the same binding
site, but with different conformations resulting from the
formation of individual hydrogen bonds and accounting
for the opposite impact of 5- and 6-hydroxy groups on
PDGFR and FLT3 inhibition: 5,50-diOH bisindolyl-
and 6,60-diOH bisbenzofuranylmethanones tend to
FLT3, 5,50-diOH bisbenzofuranyl derivatives to PDGFR
selectivity. Further modifications of the structures, in par-
ticular the investigation of unsymmetric substitution pat-
terns, will possibly lead to more potent inhibitors with
higher selectivity for one of both kinases.
5.1.2.4. Bis(5-methoxybenzo[b]furan-2-yl)methanone (4).
Preparation from 2-hydroxy-5-methoxybenzaldehyde
(Acros and Lancaster) as described above, using 1,3-
dichloropropan-2-one. Yield: 8.70 g (50%) of light yellow
crystals. Mp 175 ꢁC (from dichloromethane), lit.22 173–
174 ꢁC, lit.20 176 ꢁC. 1H NMR (DMSO-d6):
d
(ppm) = 3.82 (s, 6H), 7.20 (dd, 2H, J = 9.0 Hz, 2.5 Hz),
7.35 (d, 2H, J = 2.5 Hz), 7.71 (d, 2H, J = 9.0 Hz), 8.12
(s, 2H).
5.1.2.5. Bis(6-methoxybenzo[b]furan-2-yl)methanone (5).
Preparation from 2-hydroxy-4-methoxybenzaldehyde
(Acros) as described above, using 1,3-dichloropropan-2-
one. Yield: 6.07 g (29%) of yellow crystals. Mp 203–
204 ꢁC (from toluene), lit.22 191–193 ꢁC, lit.20 200 ꢁC.
1H NMR (DMSO-d6): d (ppm) = 3.87 (s, 6H), 7.04 (d,
2H, J = 8.8 Hz), 7.40 (s, 2H), 7.76 (d, 2H, J = 8.8 Hz),
8.13 (s, 2H).
5. Experimental protocols
5.1. Chemical procedures
5.1.2.6. Bis(7-methoxybenzo[b]furan-2-yl)methanone (6).
Preparation from 2-hydroxy-3-methoxybenzaldehyde
(Aldrich) as described above, using 1,3-dichloropropan-
2-one. Yield: 10.31 g (64%) of colorless crystals. Mp
146–147 ꢁC (from methanol), lit.22 143–144 ꢁC. 1H
NMR (DMSO-d6): d (ppm) = 4.00 (s, 6H), 7.18 (d, 2H,
J = 8.0 Hz), 7.32 (dd, 2H, J = 8.0 Hz, 8.0 Hz), 7.45 (d,
2H, J = 8.0 Hz), 8.13 (s, 2H).
5.1.1. General aspects. The methyl ethers of compounds
4, 5, and 6 were cleaved under microwave irradiation
using a CEM Discover-focused microwave synthesis sys-
¨
tem. Melting points were determined with a BUCHI
Melting Point B-545 device. IR spectra (pure solid if
no contrary indication is given) were measured with a
BRUKER Tensor 27. H NMR spectra were recorded
1
5.1.2.7. Bis(5-nitrobenzo[b]furan-yl)methanone25 (11).
Preparation from 2-hydroxy-5-nitrobenzaldehyde (Lan-
caster) as described above, using 1,3-dichloropropan-2-
one. Yield: 1.4 g (13%) of colorless crystals. Mp
with a BRUKER Avance 300 (300 MHz) at 300 K,
using solvent peak as internal standard. MS spectra
were measured with a FINNIGAN MAT 95.
1
294 ꢁC (washed with acetone). H NMR (DMSO-d6):
5.1.2. Syntheses
d (ppm) = 8.08 (d, 2H, J = 9.1 Hz), 8.45 (s, 2H), 8.47
(dd, 2H, J = 9.1 Hz, 2.5 Hz), 8.89 (d, 2H, J = 2.5 Hz).
5.1.2.1. Procedure a—Synthesis of bis(benzo[b]furan-
2-yl)methanones from 2-hydroxybenzaldehyde derivatives.
Potassium carbonate (17 g, 1 equiv) was added to a hot
butanone solution (250 mL) of salicylaldehyde (15 g,
122.8 mmol) and 1,3-dihaloacetone (7.8 g, 0.5 equiv),
before refluxing the mixture for 4 h. The suspension
5.1.2.8. Procedure b—Demethylation of compounds 4–
6 leading to the corresponding bisbenzofuranylmethanone
dihydroxylated derivatives. A mixture of a dimethoxylat-
ed bisbenzofuranylmethanone derivative (1.53 g,

2194
S. Mahboobi et al. / Bioorg. Med. Chem. 15 (2007) 2187–2197
4.75 mmol) and pyridinium chloride (8.3 g, 15 equiv)
was heated to 160 ꢁC, while stirring, in a focused micro-
wave apparatus (300 W, 5 min). The hot melt was
hydrolyzed to leave a solid residue which was collected
on filter, washed with water, dried, and subjected to col-
umn chromatography (SiO2, ethyl acetate) before
recrystallization.
the residue with a little diethyl ether results in crystalli-
zation of the compound, which was washed with addi-
tional ether.
5.1.2.15. Bis(5-aminobenzo[b]furan-2-yl)methanone25
(12). Yield: 0.54 g (70%) of red-orange crystals. Mp
1
dec 189 ꢁC (from diethyl ether). H NMR (DMSO-d6):
d (ppm) = 5.28 (s, 4H), 6.88–6.92 (m, 4H), 7.46 (dd,
2H, J = 9.6 Hz, J = 0.8 Hz), 7.93 (d, 2H, J = 0.8 Hz).
5.1.2.9. Bis(5-hydroxybenzo[b]furan-2-yl)methanone (7).
Preparation
from
bis(5-methoxybenzo[b]furan-2-
yl)methanone 4 as described above. Yield: 1.90 g (72%)
of yellow crystals. Mp dec 280 ꢁC (from EtOH/toluene),
lit.22 dec 277–278 ꢁC, lit.20 dec 287 ꢁC. 1H NMR
5.1.2.16. Procedure e—Preparation of 5-meth-
oxybenzo[b]furan-2-carboxylic acid ethyl ester (13). The
synthesis was performed according to lit.29 Yield: 9.2 g
(61%) of a colorless solid. 1H NMR (CDCl3): d
(ppm) = 1.43 (t, 3H, J = 7.1 Hz), 3.85 (s, 3H), 4.44 (qu,
2H, J = 7.1 Hz), 7.05–7.08 (m, 2H), 7.47–7.50 (m, 2H).
(DMSO-d6):
d (ppm) = 7.04 (dd, 2H, J = 8.8 Hz,
2.5 Hz), 7.15 (d, 2H, J = 2.5 Hz), 7.60 (d, 2H,
J = 8.8 Hz), 8.05 (s, 2H), 9.58 (s, 2H).
5.1.2.10. Bis(6-hydroxybenzo[b]furan-2-yl)methanone (8).
Preparation
5.1.2.17. Procedure f—Demethylation of compound
13. The desired compound was prepared by a modifica-
tion of lit.29 as follows: 5-methoxybenzo[b]furan-2-car-
boxylic acid ethyl ester (9.0 g, 40.9 mmol) was
dissolved in dichloromethane (300 mL), boron tribro-
mide (2.5 equiv) was added at 0 ꢁC. After a 1 h stirring
period at 0 ꢁC, the mixture was poured into saturated
brine (200 mL), the organic layer separated, dried
(Na2SO4), the solvent removed under reduced pressure,
and the title compound purified by column chromatog-
raphy (SiO2, CH2Cl2).
from
bis(6-methoxybenzo[b]furan-2-
yl)methanone 5 as described above. Yield: 1.77 g (67%)
of yellow crystals. Mp dec 263–264 ꢁC (from acetone),
lit.22 dec 256–258 ꢁC, lit.20 dec 270 ꢁC. 1H NMR
(DMSO-d6):
d (ppm) = 6.90 (dd, 2H, J = 8.5 Hz,
2.0 Hz), 7.06 (m, 2H), 7.67 (d, 2H, J = 8.5 Hz), 8.06 (m,
2H), 10.24 (s, 2H).
5.1.2.11. Bis(7-hydroxybenzo[b]furan-2-yl)methanone (9).
Preparation
from
bis(7-methoxybenzo[b]furan-2-
yl)methanone 6 as described above. Yield: 1.08 g (77%)
of yellow crystals. Mp dec 268–269 ꢁC (from dichloro-
5.1.2.18. 5-Hydroxybenzo[b]furan-2-carboxylic acid
ethyl ester (14). Yield: 5.95 g (71%) of a colorless solid.
1H NMR (CDCl3): d (ppm) = 1.43 (t, 3H, J = 7.1 Hz),
4.44 (q, 2H, J = 7.1 Hz), 5.57 (s, 1H, exchangeable),
7.01 (dd, 1H, J = 9.1 Hz, 2.5 Hz), 7.08 (d, 1H,
J = 2.5 Hz), 7.42–7.44 (m, 2H).
1
methane), lit.22 dec 268–269 ꢁC. H NMR (DMSO-d6):
d (ppm) = 7.00 (d, 2H, J = 7.2 Hz), 7.20 (dd, 2H,
J = 7.2 Hz, 7.2 Hz), 7.32 (d, 2H, 7.2 Hz), 8.21 (s, 2H),
10,46 (s, 2H).
5.1.2.12. Procedure c—Esterification of bis(5-hydrox-
ybenzo[b]furan-2-yl)methanone (7). Acetyl chloride
5.1.2.19. Procedure g—Benzylation of compound 14.
5-Hydroxybenzo[b]furan-2-carboxylic acid ethyl ester
(6.0 g, 20.09 mmol) was dissolved in acetone
(120 mL), benzyl bromide (1.2 equiv) and potassium
carbonate (4 equiv) were added, and the mixture stir-
red at reflux for 3 h. The solvent was removed under
reduced pressure, the mixture extracted with dichloro-
methane, and the title compound purified by column
chromatography (SiO2, CH2Cl2).
(2.2 mL, 6 equiv) was added to a solution of
bis(5-hydroxybenzo[b]furan-2-yl)methanone (1.54 g,
5.2 mmol) and a catalytic quantity of DMAP in methyl
acetate (20 mL) and pyridine (5 mL). After a 12-h stir-
ring time, the solution was poured into an ice/water mix-
ture (25 mL). Extraction with ethyl acetate, drying of
the combined organic layers over magnesium sulfate,
and distillation of the solvent left the crude product
witch was recrystallized from dichloromethane.
5.1.2.20. 5-Benzyloxybenzo[b]furan-2-carboxylic acid
ethyl ester30 (15). Yield: 7.30 g (85%) of a colorless solid.
5.1.2.13. Bis(5-acetoxybenzo[b]furan-2-yl)methanone
(10). Yield: 1.44 g (73%) of colorless crystals. Mp
209.5–210.5 ꢁC (from dichloromethane), lit.24 203–
1
Mp 83.9–85.3 ꢁC. IR (KBr): m (cmꢀ1) = 1609. H NMR
(DMSO-d6): d (ppm) = 1.32 (t, 3H, J = 7.1 Hz), 4.34 (q,
2H,, J = 7.1 Hz), 5.13 (s, 2H), 7.19 (dd, 1H, J = 7.1 Hz,
2.5 Hz), 7.32–7.65 (m, 8H).
1
204 ꢁC. H NMR (DMSO-d6): d (ppm) = 2.31 (s, 6H),
7.37 (dd, 2H, J = 9.1 Hz, J = 2.5 Hz), 7.67 (d, 2H,
J = 2.5 Hz), 7.85 (d, 2H, J = 9.1 Hz), 8.23–8.24 (m, 2H).
5.1.2.21. Procedure h—Deesterification of compound 15.
A mixture of 5-benzyloxybenzo[b]furan-2-carboxylic acid
ethyl ester (7.00 g, 23.6 mmol), THF, EtOH and 2 N
KOH (150 mL, 1:1:1) was stirred at room temperature
for 4 h. The organic solvents were removed under reduced
pressure, the residue dissolved in water (100 mL), acidified
with concd HCl, extracted with ethyl acetate (3· 100 mL),
and the combined organic layers dried (Na2SO4). Most of
the solvent was removed under reduced pressure and the
title compound crystallized by addition of pentane.
5.1.2.14. Procedure d—Preparation of derivate 12 by
reduction of the 5,50-dinitro compound 11. 1.6 g (9 equiv)
of ammonium formate and 0.1 g of activated Pd/C
(10%) were added to a hot suspension of 11 (900 mg)
in propan-2-ol (250 mL) and the mixture was gently re-
fluxed for 48 h. The solution was cooled down to rt, the
catalyst filtered off, and the solvent evaporated. Ethyl
acetate and water were added, the organic layer was
washed with water and dried (Na2SO4). Treatment of

S. Mahboobi et al. / Bioorg. Med. Chem. 15 (2007) 2187–2197
2195
5.1.2.22. 5-Benzyloxybenzo[b]furan-2-carboxylic acid28
(16). Yield: 7.30 g (71%) of colorless crystals. Mp 182.6–
184.3 ꢁC. IR (KBr): m (cmꢀ1) = 2959, 1726. 1H NMR
(DMSO-d6): d (ppm) = 5.13 (s, 2H), 7.13 (dd, 1H, J =
9.1 Hz, 2.7 Hz), 7.31–7.51 (m, 7H), 7.58 (d, 1H,
J = 9.1 Hz), 13.60 (s, br s 1H, exchangeable).
J = 7.7 Hz), 7.94 (d, 1H, J = 1.1 Hz), 7.99–8.02 (m,
2H), 8.08 (dd, 1H, J = 8.5 Hz, 0.8 Hz). EI-MS (70 eV)
m/z (%): 401 (82) [M+], 337 (61), 260 (66), 232 (86),
204 (53), 143 (100), 89 (47), 77 (81). Anal.
(C23H15NO4S): Calcd C, 68.81; H, 3.77; N 3.49. Found:
C, 68.59; H, 3.60: N, 3.44.
5.1.2.23. Procedure i—Preparation of benzo[b]furan-2-
carbonyl chlorides. To a stirred dichloromethane solution
(20 mL, dry) of the respective benzo[b]furan-2-carboxylic
acid (15.42 mmol), pyridine (1.39 mL, 17 mmol) and then
thionylchloride (1.16 mL, 16 mmol) were added at room
temperature. The mixture was stirred for 30 min and
flashed through a short chromatography column (SiO2,
CH2Cl2). The solvent was removed under reduced pres-
sure and the product dried in vacuo.
5.1.2.31. (5-Benzyloxybenzo[b]furan-2-yl)-(5-benzyl-
oxy-1-phenylsulfonyl-1H-indol-2-yl)methanone
(22).
Yield: 3.80 g (53%) of red crystals. Mp 206.5–206.6 ꢁC
(from dichloromethane by addition of diethyl ether).
IR (KBr): m (cmꢀ1) = 1646. 1H NMR (DMSO-d6): d
(ppm) = 5.14 (s, 2H), 5.17 (s, 2H), 7.23 (dd, 1H,
J = 9.2 Hz, 2.6 Hz), 7.28–7.51 (m, 14H), 7.59–7.64 (m,
2H), 7.70–7.73 (m, 2H), 7.79 (d, 1H, J = 0.8 Hz), 7.92–
7.99 (m, 3H). ES-MS (CH2Cl2/MeOH, NH4Ac) m/z
(%): 631 (9) ½M þ NH4þꢂ, 614 (100) [MH+]. Anal.
(C37H27NO6S): Calcd C, 72.41; H, 4.43; N, 2.28; S,
5.23. Found: C, 71.81; H, 4.43; N, 2.23; S, 5.17.
5.1.2.24. Benzo[b]furan-2-carbonyl chloride27 (17).
Preparation from benzo[b]furan-2-carboxylic acid (Ac-
ros) as described above. Yield: 2.16 g (78%) of colorless
crystals. Mp 139.4–139.7 ꢁC. IR (KBr): m (cmꢀ1) = 3128,
1784, 1723.
5.1.2.32. Procedure l—Removal of the phenylsulfonyl
protection group. The respective (1-phenylsulfonyl-1H-
indol-2-yl)benzo[b]furan-2-yl-methanone 21 or 22
(1.49 mmol) was suspended in a mixture of THF,
MeOH, and an aqueous NaOH (10%) solution
(150 mL, 1:1:1). After a 4-h refluxing period, the organic
solvents were removed under reduced pressure. The
formed precipitate was filtered off, washed with water,
dried, and purified by column chromatography (SiO2,
CH2Cl2) and subsequent crystallization.
5.1.2.25. 5-Benzyloxy-benzo[b]furan-2-carbonyl chloride
(18). Preparation from 16 as described above. Yield:
2.61 g (59%) of colorless crystals. Mp 99.3–102.5 ꢁC. IR
(KBr): m (cmꢀ1) = 2875, 1787, 1730. Anal. (C16H11ClO3):
Calcd C, 67.03; H, 3.87. Found: C, 67.74; H, 4.15.
5.1.2.26. Procedure j—Preparation of the N-protected
indoles. The protection was performed according to the
previously published procedure.19
5.1.2.33. (Benzo[b]furan-2-yl)-(1H-indol-2-yl)metha-
none (23). Yield: 0.26 g (67%) of yellow crystals. Mp
241.9–242.0 ꢁC (from dichloromethane). IR (KBr): m
5.1.2.27. 1-Phenylsulfonyl-1H-indole34 (19). From 1H-
indole (Merck). Yield: 8.20 g (76%).
(cmꢀ1) = 3293, 1609. 1H NMR (DMSO-d6):
d
(ppm) = 7.11–7.17 (m, 1H), 7.32–7.44 (m, 2H), 7.51–
7.62 (m, 2H), 7.79–7.91 (m, 4H), 8.05 (d, 1H,
J = 0.8 Hz), 12.09 (s, 1H, exchangeable). EI-MS
(70 eV) m/z (%): 261 (99) [M+], 143 (100), 115 (26), 89
(33). Anal. (C17H11NO2): Calcd C, 78.15; H ,4.24; N,
5.36. Found: C, 78.06; H, 4.25; N, 5.32.
5.1.2.28. 5-Benzyloxy-1-phenylsulfonyl-1H-indole18 (20).
From 5-benzyloxy-1H-indole (Lancaster). Yield: 19.72 g.
(87%).
5.1.2.29. Procedure k—Preparation of the N-protected
benzofuranylindolylmethanones. To a stirred solution of
the respective 1-phenylsulfonyl-1H-indole 19 or 20
(11.07 mmol) in dry THF (30.0 mL), n-butyllithium
(7.61 mL 1.6 M in hexane, 12.17 mmol) was added at
ꢀ78 ꢁC. After 1 h, a pre-cooled solution of the respec-
tive benzo[b]furan-2-carbonyl chloride 17 or 18
(11.07 mmol) in THF (40.0 mL) was added at once.
The mixture was further stirred for 30 min at ꢀ78 ꢁC,
then allowed to reach room temperature, poured into
water (300 mL), and extracted with ethyl acetate (3·
50 mL). The combined organic layers were dried
(Na2SO4) and the solvent removed under reduced pres-
sure. The crude product was purified by column chro-
matography (SiO2, DCM and light petrol 2:1) and
subsequent crystallization.
5.1.2.34. (5-Benzyloxybenzo[b]furan-2-yl)-(5-benzyloxy-
1H-indol-2-yl)methanone (24). Yield: 0.54 g (76%) of
yellow crystals. Mp 183.8–183.9 ꢁC (from dichlorometh-
ane). IR (KBr): m (cmꢀ1) = 3311, 1605. 1H NMR
(DMSO-d6): d (ppm) = 5.14 (s, 2H), 5.18 (s, 2H), 7.09
(dd, 1H, J = 8.9 Hz, 2.3 Hz), 7.25–7.52 (m, 14H), 7.71
(d, 1H, J = 1.4 Hz), 7.75 (d, 1H, J = 9.3 Hz), 7.91 (d, 1H,
J = 0.8 Hz), 11.96 (d, 1H, J = 1.9 Hz, exchangeable).
EI-MS (70 eV) m/z (%): 473 (20) [M+], 382 (23), 91 (100).
Anal. (C31H23NO4): Calcd C, 78.63; H, 4.90; N, 2.96.
Found: C, 78.66; H, 4.99; N, 2.94.
5.1.2.35. Procedure m—Hydrogenolytic preparation of
compound 25. A suspension of 23 (0.76 g; 1.60 mmol),
ammonium formate (1.01 g; 16.0 mmol), and activated
Pd on charcoal (10%, 0.20 g) in a mixture of THF/
MeOH (1:1; 300 mL) was heated to reflux for 1 h. The
catalyst was filtered over Celite, the solvent removed un-
der reduced pressure, and the product purified by col-
umn chromatography (SiO2, DCM and AcOEt 2:1).
Most of the solvent was removed under reduced
5.1.2.30. (Benzo[b]furan-2-yl)-(1-phenylsulfonyl-1H-
indol-2-yl)-methanone (21). Yield: 1.60 g (33%) of color-
less crystals. Mp 144.8–145.5 ꢁC (from methanol). IR
1
(KBr): m (cmꢀ1) = 3112, 1649. H NMR (DMSO-d6): d
(ppm) = 7.36–7.46 (m, 2H), 7.53-7.58 (m, 1H), 7.60–
7.64 (m, 4H), 7.71–7.82 (m, 3H), 7.89 (d, 1H,

2196
S. Mahboobi et al. / Bioorg. Med. Chem. 15 (2007) 2187–2197
6. Schmidt-Arras, D.; Schwable, J.; Bo¨hmer, F. D.; Serve, H.
Curr. Pharm. Des. 2004, 10, 1867–1883.
7. Tickenbrock, L.; Muller-Tidow, C.; Berdel, W. E.; Serve,
¨
pressure and the product crystallized from the solution
by addition of light petrol.
H. Expert Opin. Emerg. Drugs 2006, 11, 153–165.
8. Gazit, A.; Yee, K.; Uecker, A.; Bo¨hmer, F. D.; Sjoblom,
T.; Ostman, A.; Waltenberger, J.; Golomb, G.; Banai, S.;
Heinrich, M. C.; Levitzki, A. Bioorg. Med. Chem. 2003,
11, 2007–2018.
9. Pandey, A.; Volkots, D. L.; Seroogy, J. M.; Rose, J. W.;
Yu, J. C.; Lambing, J. L.; Hutchaleelaha, A.; Hollenbach,
S. J.; Abe, K.; Giese, N. A.; Scarborough, R. M. J. Med.
Chem. 2002, 45, 3772–3793.
10. Stone, R. M.; DeAngelo, D. J.; Klimek, V.; Galinsky, I.;
Estey, E.; Nimer, S. D.; Grandin, W.; Lebwohl, D.;
Wang, Y.; Cohen, P.; Fox, E. A.; Neuberg, D.; Clark, J.;
Gilliland, D. G.; Griffin, J. D. Blood 2005, 105, 54–60.
11. Mesters, R. M.; Padro, T.; Bieker, R.; Steins, M.; Kreuter,
M.; Goner, M.; Kelsey, S.; Scigalla, P.; Fiedler, W.;
Buchner, T.; Berdel, W. E. Blood 2001, 98, 241–243.
12. Fiedler, W.; Serve, H.; Dohner, H.; Schwittay, M.;
Ottmann, O. G.; O’Farrell, A. M.; Bello, C. L.; Allred,
R.; Manning, W. C.; Cherrington, J. M.; Louie, S. G.;
Hong, W.; Brega, N. M.; Massimini, G.; Scigalla, P.;
Berdel, W. E.; Hossfeld, D. K. Blood 2005, 105, 986–
993.
5.1.2.36. 5-Hydroxybenzo[b]furan-2-yl-(5-hydroxy-1H-
indol-2-yl)methanone (25). Yield: 0.39 g (83%) of light yel-
low crystals. Mp 265.3–265.5 ꢁC. IR (KBr):
m
(cmꢀ1) = 3453, 3314, 1599. 1H NMR (DMSO-d6): d
(ppm) = 6.89 (dd, 1H, J = 8.9 Hz, 2.3 Hz), 7.01–7.04 (m,
2H), 7.12 (d, 1H, J = 2.5 Hz), 7.32 (d, 1H, J = 8.8 Hz),
7.60–7.63 (m, 2H), 7.84 (m, 1H, J = 0.8 Hz), 9.04 (s, 1H,
exchangeable), 9.52 (s, 1H, exchangeable), 11.76 (d, 1H,
J = 1.1 Hz, exchangeable). EI-MS (70 eV) m/z (%): 293
(68) [M+], 159 (100). Anal. (C17H11NO4): Calcd C, 69.62;
H, 3.78; N, 4.78. Found: C, 69.59; H, 3.97; N, 4.74.
5.2. Biochemical and cellular assays
All activity tests were performed as previously
described.19
5.3. Molecular modeling
As previously described,19 an initial computer model of
the FLT3 kinase was generated from the PDB crystal
structure 1rjb31 using the molecular modeling package
SYBYL 7.1 (Tripos Inc., St. Louis, USA) on a Silicon
Graphics Octane workstation. The conformers of compd
7, ap/ap, ap/sp, and sp/sp, their energies and rotational
barriers were calculated by the semi-empirical quantum
chemical program AMPAC 8.15 (Semichem Inc., 7128
Summit, Shawnee, KS66216, USA) with the AM1 ham-
iltonian, the TRUSTE minimization algorithm, and the
COSMO water solvation model35 as implied in SYBYL
7.1. After docking of compd 7, the FLT3-ligand com-
plexes were energy-minimized using the AMBER 99
force field36 with AMBER 99 charges for the protein
(distance-dependent dielectricity constant 4, first 50
cycles with constrained backbone and the steepest des-
cent method). For that purpose, compound 7 was pro-
vided with AMBER 99 atom types by analogy with
corresponding amino acid atoms, as well as with Gastei-
ger–Hueckel charges. New parameters describing the
central OC(CH)COC(CH)O moiety had to be added to
the AMBER 99 force field (derived from the Tripos force
field and from comparison with similar AMBER param-
eters). The final minimization up to a RMS gradient of
13. Wilhelm, S. M.; Carter, C.; Tang, L.; Wilkie, D.;
McNabola, A.; Rong, H.; Chen, C.; Zhang, X.; Vincent,
P.; McHugh, M.; Cao, Y.; Shujath, J.; Gawlak, S.;
Eveleigh, D.; Rowley, B.; Liu, L.; Adnane, L.; Lynch,
M.; Auclair, D.; Taylor, I.; Gedrich, R.; Voznesensky, A.;
Riedl, B.; Post, L. E.; Bollag, G.; Trail, P. A. Cancer Res.
2004, 64, 7099–7109.
14. Smith, B. D.; Levis, M.; Beran, M.; Giles, F.; Kantarjian,
H.; Berg, K.; Murphy, K. M.; Dauses, T.; Allebach, J.;
Small, D. Blood 2004, 103, 3669–3676.
15. Lopes de Menezes, D. E.; Peng, J.; Garrett, E. N.; Louie,
S. G.; Lee, S. H.; Wiesmann, M.; Tang, Y.; Shephard, L.;
Goldbeck, C.; Oei, Y.; Ye, H.; Aukerman, S. L.; Heise, C.
Clin. Cancer Res. 2005, 11, 5281–5291.
16. Sattler, M.; Verma, S.; Byrne, C. H.; Shrikhande, G.;
Winkler, T.; Algate, P. A.; Rohrschneider, L. R.; Griffin,
J. D. Mol. Cell. Biol. 1999, 19, 7473–7480.
17. Bo¨hmer, F. D.; Karagyozov, L.; Uecker, A.; Serve, H.;
Botzki, A.; Mahboobi, S.; Dove, S. J. Biol. Chem. 2003,
278, 5148–5155.
18. Mahboobi, S.; Teller, S.; Pongratz, H.; Hufsky, H.;
Sellmer, A.; Botzki, A.; Uecker, A.; Beckers, T.; Baasner,
¨
S.; Scha¨chtele, C.; Uberall, F.; Kassack, M. U.; Dove, S.;
Bo¨hmer, F. D. J. Med. Chem. 2002, 45, 1002–1018.
´
19. Mahboobi, S.; Uecker, A.; Sellmer, A.; Cenac, C.; Ho¨cher,
H.; Pongratz, H.; Eichhorn, E.; Hufsky, H.; Trumpler, A.;
¨
˚
0.05 kcal/(mol A) was performed with the Powell conju-
Sicker, M.; Heidel, F.; Fischer, T.; Stocking, C.; Elz, S.;
Bo¨hmer, F. D.; Dove, S. J. Med. Chem. 2006, 49, 3101–
3115.
gate gradient method. Considerations on the PDGFR ki-
nase were based on the FLT3 crystal structure with
simple mutation of the only different binding site residue,
FLT3 Phe-691, into PDGFR Thr-681.
20. Royer, R.; Guillaumel, J. J. Heterocycl. Chem. 1986, 23,
1277–1282.
21. Nagahara, M.; Nakanishi, T.; Kuriyama, K.; Ikemoto,
M.; Yamamoto, M. (Kaken Pharma Co. Ltd.) Jpn. Kokai
Tokkyo Koho 1984, JP59231080.
References and notes
22. Nagahara, M.; Kuriyama, K.; Ikemoto, M.; Nakanishi, T.
Yakugaku Zasshi 1985, 105, 840–844.
1. Pietras, K.; Sjoblom, T.; Rubin, K.; Heldin, C. H.;
Ostman, A. Cancer Cell 2003, 3, 439–443.
2. Parcells, B. W.; Ikeda, A. K.; Simms-Waldrip, T.; Moore,
T. B.; Sakamoto, K. M. Stem Cells 2006, 24, 1174–1184.
3. Scheijen, B.; Griffin, J. D. Oncogene 2002, 21, 3314–3333.
4. Board, R.; Jayson, G. C. Drug Resist. Updat. 2005, 8, 75–
83.
23. Nagahara, M.; Oishi, Y.; Takehisa, Y.; Yajima, M.;
Kurokawa, S.; Kajikawa, N. (Kaken Pharma Co. Ltd.)
Jpn. Kokai Tokkyo Koho 1985, JP60181081.
24. Nagahara, M.; Kuriyama, K.; Hiyama, Y.; Ito, K.;
Ikemoto, M.; Nakanishi, T. Yakugaku Zasshi 1986, 106,
27–35.
25. Kondo, A. (Sharp Kk.) Jpn. Kokai Tokkyo Koho 1996,
JP8295688.
5. Morgan, M. A.; Reuter, C. W. Ann. Hematol. 2006, 85,
139–163.

S. Mahboobi et al. / Bioorg. Med. Chem. 15 (2007) 2187–2197
2197
26. Mahboobi, S.; Pongratz, H.; Hufsky, H.; Hockemeyer, J.;
Frieser, M.; Lyssenko, A.; Paper, D. H.; Burgermeister, J.;
¨
Bo¨hmer, F. D.; Fiebig, H. H.; Burger, A. M.; Baasner, S.;
Beckers, T. J. Med. Chem. 2001, 44, 4535–4553.
27. Reichstein, T.; Ignaz, R. Helv. Chim. Acta 1930, 13, 1275–
1281.
28. Reichstein, T.; Oppenauer, R.; Grussner, A.; Hirt, R.;
Rhyner, L.; Glatthaar, C. Helv. Chim. Acta 1935, 18, 816–830.
29. Schmeck, C.; Muller, U.; Schmidt, G.; Pernerstorfer, J.;
¨
31. Griffith, J.; Black, J.; Faerman, C.; Swenson, L.; Wynn,
M.; Lu, F.; Lippke, J.; Saxena, K. Mol. Cell 2004, 13, 169–
178.
32. Mohammadi, M.; Froum, S.; Hamby, J. M.; Schroeder,
M. C.; Panek, R. L.; Lu, G. H.; Eliseenkova, A. V.; Green,
D.; Schlessinger, J.; Hubbard, S. R. EMBO J. 1998, 17,
5896–5904.
33. Zahao, D. J.; Huang, J. K.; Shan, Z. X. Chin. Chem. Lett.
1993, 99, 385–386.
Bischoff, H.; Kretschmer, A.; Vo¨hringer, V.; Faeste, C.;
Haning, H.; Hauswald, M.; Schmidt, D.; Zoche, M.;
Apeler, H.; Jonghaus, W.; Reinemer, P. (Bayer AG) Ger.
Offen. 2002, DE10115408.
34. Sundberg, R. J.; Russel, F. J. Org. Chem. 1973, 38, 3324–
3330.
35. Klamt, A. J. Phys. Chem. 1995, 99, 2224–2235.
36. Cornell, W. D.; Cieplak, P.; Bayly, C. I.; Gould, I. R.;
Merz, K. M. J.; Ferguson, D. M.; Spellmeyer, D. C.; Fox,
T.; Caldwell, J. W.; Kollman, P. A. J. Am. Chem. Soc.
1995, 117, 5179–5197.
30. Lu, J.; Ma, T.; Nagpal, S.; Shen, Q.; Warshawsky, A. M.;
Ochoada, J. M.; Yee, Y. K. (Eli Lilly & Co.) PCT Int.
Appl. 2005, WO2005051938.