Flavonoid dimers as novel, potent antileishmanial agents

Article abstract of DOI:10.1021/jm301172v

The present study found that synthetic flavonoid dimers with either polyethylene glycol linker or amino ethyleneglycol linker have marked leishmanicidal activity. Compound 39 showed very consistent and promising leishmanicidal activity for both extracellu

Full text of DOI:10.1021/jm301172v

Article
pubs.acs.org/jmc
Flavonoid Dimers as Novel, Potent Antileishmanial Agents
Iris L. K. Wong,§,† Kin-Fai Chan,§,† Tak Hang Chan,*,†,‡ and Larry M. C. Chow*,†
†Department of Applied Biology and Chemical Technology and the State Key Laboratory for Chirosciences, The Hong Kong
Polytechnic University, Hung Hom, Hong Kong SAR, China, and State Key Laboratory in Chinese Medicine and Molecular
Pharmacology, Shenzhen, China
‡Department of Chemistry, McGill University, Montreal, Quebec, H3A 2K6 Canada
S
* Supporting Information
ABSTRACT: The present study found that synthetic flavonoid
dimers with either polyethylene glycol linker or amino ethyleneglycol
linker have marked leishmanicidal activity. Compound 39 showed
very consistent and promising leishmanicidal activity for both
extracellular promastigotes (IC50 ranging from 0.13 to 0.21 μM) and
intracellular amastigotes (IC50 = 0.63 μM) irrespective of the drug-
sensitivity of parasites. Moreover, compound 39 displayed no toxicity
toward macrophage RAW 264.7 cells (IC50 > 100 μM) and primary
mouse peritoneal elicited macrophages (IC50 > 88 μM). Its high
value of therapeutic index (>140) was better than other highly
potent antileishmanials such as amphotericin B (therapeutic index =
119). Compound 39 is therefore a new, safe, and effective
antileishmanial candidate compound which is even effective against
drug-refractory parasites.
characterized a series of flavonoid dimers, as illustrated in Chart
1, to be used in reversing drug resistance in cancer and
Leishmania by inhibiting membrane transporters.5−10 In the
course of the investigation, it became necessary to assess the in
vitro leishmanicidal activity of these compounds and draw
structure−activity relationships, if possible.
In the present study, we have selected structurally related
flavonoid dimers to assess their antileishmanial activity by
screening against insect stage parasites (promastigotes) as well as
measuring their toxicity toward mouse macrophage cell line.
Compounds with promising antipromastigotes activity and low
toxicity were then screened against human host stage
(amastigotes). At the same time, medicinal chemistry was carried
out to further modify structure of promising lead compounds to
examine structure−activity relationships. Through these pro-
cesses, we have discovered novel compounds with potent
antileishmanial activity and high therapeutic index which was
comparable or superior to other current antileishmanials.
INTRODUCTION
■
Leishmaniasis is one of the six major parasitic diseases targeted by
the World Health Organization. It is endemic in 88 countries.
More than 350 million people worldwide are at risk of infection.
Out of the annual 2 million cases, 500 000 of them belong to
visceral leishmaniasis (VL). VL is caused by Leishmania donovani
amastigotes infecting macrophages in a patient’s reticuloendo-
thelial organs such as liver and spleen. VL could be fatal if left
untreated.1 Currently, there are no effective vaccines for
leishmaniasis. Primary means of chemotherapy include pentava-
lent antimonials (SbV) which is a decade-old drug. Antimonial
treatment, however, is less than ideal due to the difficulty in
administration (slow intravenous injection), severe side effects,
poor patient compliance, and emergence of antimonials-resistant
cases. Newer antileishmanials include amphotericin B (with or
without liposomal formulations), miltefosine, and paromomycin.
All of them, however, have different types of limitations including
nephrotoxicity, incomplete efficacy, and expensive or incon-
venient treatment protocol. There is an urgent need to speed up
the development of a new generation of more effective and safe
antileishmanials.
Recent findings suggested that plant-derived products have
potent and selective antiparasitic properties.2 Among these plant-
derived products, flavonoids represent a large family of
polyphenolic compounds found in vegetables and fruits. As
humans consume large amounts of flavonoids every day, it is
generally accepted that flavonoids are safe and not toxic. The
most common flavonoids are flavone and isoflavone.3 Naturally
occurring flavones have been reported to have potent
leishmanicidal activity.4 We have previously synthesized and
RESULTS
■
In Vitro Antipromastigote Activity of Flavonoid Dimers
Reported Previously. A. Polyethylene Glycol (PEG) Linked
Flavonoid Dimers 1−30. We investigated if synthetic flavonoid
dimers 1−30 have any antipromastigotes activity toward L.
donovani (wild-type LdAG83, sodium stibogluconate (SSG)-
resistant Ld39 and pentamidine-resistant LdAG83PentR50)
Received: August 9, 2012
Published: September 18, 2012
© 2012 American Chemical Society
8891
dx.doi.org/10.1021/jm301172v | J. Med. Chem. 2012, 55, 8891−8902

Journal of Medicinal Chemistry
Article
a
Chart 1. Chemical Structures of Previously Reported Flavonoid Dimers 1−50
a
PEG 600 with average number of EG unit of 13 was used to synthesize compound 15.
promastigotes. Cytotoxicity toward mouse macrophage cell line
RAW 264.7 was also determined as an indicator for the toxicity of
flavonoid dimers. Flavonoid dimers 1−30 contained two
flavonoid moieties linked together by a PEG linker. Structural
differences of these flavonoid dimers included different
substituents in the flavone ring A and C as well as different
number of ethylene glycol (EG) units.
As shown in Table 1, all flavonoid dimers in series A contain
OH groups in the ring A or ring C with linker lengths varied from
3 to 6 EG units. Compounds 1−3 with 3−5 EG units displayed
antipromastigotes activity toward both drug sensitive and
resistant L. donovani at micromolar level. Their IC50 ranges
from 1.8 to 6.7 μM. No cytotoxicity was observed toward the
macrophage RAW 264.7 cells (IC50 > 200 μM). Compound 4
with 6 EG units was less effective in killing wild-type and
pentamidine-resistant LdAG83 promastigotes. Compound 5
with hydroxyl group at C-3 position of C ring or compound 6
with hydroxyl group at C-5 position of A ring did not have
significant antipromastigotes activity.
Flavonoid dimers in series B have all hydroxyl groups removed
from A ring. As shown in Table 1, flavonoid dimers with shorter
linker length (compounds 7−9 with 2−4 EG units) displayed no
significant antipromastigotes activity and cytotoxicity toward
RAW 264.7 macrophages. On the other hand, dimers with longer
linker length (compounds 10−15 with 5−13av EG units)
showed antipromastigotes activity toward Leishmania promasti-
gotes as well as cytotoxicity toward RAW 264.7 macrophages
with IC50 smaller than 11 μM. Thus, removal of hydroxyl groups
from A ring and/or having linker length longer than 5 EG units
seems not favorable.
In flavonoid dimers series C of Table 1, hydrophobic
substitutions at 3-, 6-, or 7-position, such as methyl group
(compounds 16−18, 22−27), ethyl group (compounds 19−
21), or fluoro group (compound 28−30) were introduced into
ring A or C. Compounds 20, 26, 28, and 29 showed potent
antipromastigotes activity (IC50 from 2.3 to 7.9 μM) without
cytotoxicity toward macrophage RAW 264.7 (IC50 higher than
80 μM). Some structural modifications, particularly with methyl
groups at the 3-position of ring C (compounds 22−24), resulted
in cytotoxic effect on both Leishmania promastigotes and RAW
264.7 macrophages.
Although some flavonoid dimers listed in Table 1 have
significant antipromastigotes activity without cytotoxicity toward
macrophage RAW 264.7 cells, these compounds are too
hydrophobic and are only sparingly soluble in aqueous medium.
Attempts to use some of these compounds in in vivo animal
experiments proved to be impractical.5 In order to improve their
physicochemical properties as potential drug candidates, we have
to introduce some hydrophilic groups into the structure.
B. Amine-Linked Flavonoid Dimers. We have previously
reported the synthesis of a new class of flavonoid dimers
containing 4 EG units with an amine group in the middle of the
linker.10 The amino group should confer better aqueous
solubility and thus better physicochemical properties. Com-
pounds 31−50, with different substituents on the amine
nitrogen, were tested for their antipromastigotes activity
(Table 2). Compound 31, with R = H, was cytotoxic to both
promastigotes and macrophage RAW 264.7 cells with IC50 value
1.1−3.3 μM, respectively. Comparing compound 31 with 9, it is
clear that the replacement of central oxygen by an amine in the
linker changed the activity dramatically. Replacing the hydrogen
with an ethyl group (compound 32), hydroxyethyl group
(compound 33), or ethyl propanoate group (compound 34)
on the amine group did not relieve toxicity toward macrophage
RAW 264.7 cells (IC50 for RAW 264.7 = 6.4−16.0 μM).
Interestingly, when bulkier R groups were introduced to the
amine nitrogen, the flavonoid dimers thus generated (compound
35−44) were generally nontoxic to RAW 264.7 cells with IC50
values ranging from 45.7 μM to greater than 100 μM (Table 2).
For example, the tert-butyloxycarbonyl (Boc) group (compound
8892
dx.doi.org/10.1021/jm301172v | J. Med. Chem. 2012, 55, 8891−8902

Journal of Medicinal Chemistry
Article
antipromastigotes activity slightly when compared to com-
pounds 45−48. Placing the phenyl group further away from the
nitrogen (compound 44) resulted in low antipromastigotes
activity similar to that of compound 36.
Table 1. Antipromastigotes Activity of Synthetic Flavonoid
Dimers with PEG Linkers
a
IC50 (μM)
promastigotes
Ld39
macrophages
RAW 264.7
Flavonoid dimers 39 and 40 are of particular interest.
Introduction of a pyridine ring as part of R resulted in a very
strong selective antipromastigotes activity. Compound 39 with
nitrogen on position 4 of the pyridine ring has the highest
antipromastigotes activity among all flavonoid dimers. No
toxicity to the RAW 264.7 cells was observed. Compound 40,
containing nitrogen at position 2 of the pyridine ring, displayed
about 20-fold lower antipromastigotes activity than compound
39, indicating that the position of nitrogen atom on the pyridine
ring is critical in antipromastigotes activity. Flavonoid dimer with
N-mesylate group (compound 42) also displayed a significant
antipromastigotes activity. Interestingly, replacing the N-
mesylate with N-tosylate (compound 43) resulted in almost
complete loss of antipromastigote activity with IC50 values
greater than 100 μM. Similarly, a succinamide group (compound
41) gave poor activities.
Design, Synthesis, and Biological Activity of New
Flavonoid Dimers Based on Compound 39. A. Chemistry.
Compound 39 was found to display promising antipromastigotes
activity. Using compound 39 as a template, we synthesized a
series of flavonoid dimer to study structure−activity relation-
ships. Structural modifications were made by (1) shortening the
linker length, (2) changing the attachment position of the two
flavones to the amino PEG linker, (3) modifying the two flavones
with different substitutions, and (4) further modifying pyridine
ring at the amino PEG linker. As shown in Scheme 1, selective
alkylation of amino group of compound 51 furnished compound
52, which was then subjected to Mitsunobu reaction by using 2
equiv of 4′-hydroxyflavone, triphenyl phosphine, and diisopropyl
azodicarboxylate (DIAD) in tetrahydrofuran (THF). Com-
pound 53 with shorter linker length was obtained in reasonable
yield. Similarly, by selective alkylation of amino group of
compound 54 followed by Mitsunobu reaction with different
hydroxyflavones, compounds 56−61 were furnished in good
yield (Scheme 2). With the same chemistry but varying the
pyridyl methyl halides used in the selective alkylation of amino
group of compound 54, followed by Mitsunobu reaction with 4′-
hydroxyflavones, compounds 68−73 were furnished in good
yield (Scheme 3).
B. In Vitro Antipromastigote Activity. As illustrated in Table
2, compound 53 (with shorter amino PEG linker than
compound 39) has lost the antipromastigotes activity against
SSG-resistant Ld39. This result reiterated the importance of PEG
linker length in antipromastigotes activity of flavonoid dimers.
Attachment position of the two flavones to the amino PEG linker
is also important. Attachment at C-3′ position of B-ring
(compound 56) did not change the antipromastigotes activity
significantly compared to compound 39 which has linker
attached at C-4′ position. However, attachment at C-2′ position
of B-ring (compound 57) or C-3 position of C-ring (compound
58) resulted in at least 10-fold reduction in antipromastigotes
activity as compared to parent compound 39.
C3′-methoxy substitution in the B-ring (compound 59)
caused a remarkable cytotoxicity toward both promastigotes and
host peritoneal elicited macrophage (PEM) cells. On the
contrary, C3-methoxy substitution in the C-ring (compound
60) or addition of a fluorine atom at C-6 position of A-ring
(compound 61) did not cause any toxic effect toward the PEM
cells and retained the promising antipromastigotes activity.
b
compd
LdAG83
LdAG83 PentR50
Series A
2.7 0.3
1.9 0.7
1.8 0.5
1.4
1
2
3
4
5
6
5.6 3.8
2.2 0.7
2.3 0.7
>24.0
6.7 2.2
3.1 1.4
2.6 1.7
11.8
>200.0
>200.0
>200.0
ND
>24.0
>24.0
ND
ND
>24.0
>24.0
ND
ND
Series B
>24.0
7
>24.0
>24.0
>100.0
8
>24.0
>24.0
>24.0
>95.0
9
>24.0
>24.0
>24.0
>100.0
10
11
12
13
14
15
5.3 0.3
4.8 0.7
6.2 1.3
5.0 0.2
9.4 0.7
9.2 0.8
7.8 1.5
4.3 0.5
10.3 0.4
4.6 0.4
8.6 0.4
8.2 1.0
Series C
>24.0
10.4 1.4
4.9 0.7
7.3 2.0
6.0 2.0
10.6
8.9 2.3
7.1 1.2
4.8 0.3
4.4 0.4
8.1 1.2
8.4 2.4
7.9 2.8
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
>24.0
>24.0
>100.0
>24.0
4.3 1.3
3.6 1.3
>24.0
>24.0
>100.0
2.2 0.3
>24.0
>24.0
>100.0
>24.0
>100.0
4.7 0.3
3.1 0.4
6.9 1.4
4.8 1.7
1.9 0.3
>24.0
4.6 0.4
2.5 0.2
>19.0
6.3 2.4
3.3 0.8
>20.0
>200.0
5.5 0.7
8.1 0.6
3.1 1.3
2.3 0.4
88.6 4.5
>80.0
4.2 0.9
2.8 0.4
>24.0
4.6 2.1
2.5 0.3
>24.0
4.4 0.6
6.7 1.1
2.6 0.5
4.1 0.6
4.2 0.7
7.9 3.3
5.3 1.1
2.4 0.6
4.8 0.6
4.2 0.3
3.8 0.3
4.1 0.7
2.3 0.5
3.7 0.9
4.6 1.4
5.3 2.1
>200.0
>100.0
7.8 1.1
a
Synthetic flavonoid dimers were divided into three series (A−C) with
various linker lengths containing different EG units (n). Substitutions
were made at A-ring of flavone. IC50 values for antipromastigotes
activity towards LdAG83, Ld39, and LdAG83PentR50 were shown.
Toxicity towards macrophage RAW 264.7 cell line was also included.
The values were presented as mean standard error of mean. N = 1−
b
4 independent experiments. ND = not determined. All compounds
were dissolved in DMSO, and the highest % of DMSO used was 1% at
which no toxicity to promastigotes and RAW 264.7 was observed.
35) displayed a marked antipromastigotes activity without
toxicity toward RAW 264.7 cells. Introduction of a benzyl
group (compound 36) into the amino linker generated a
compound with low antipromastigotes activity without any toxic
effect to RAW 264.7 cells. However, when the benzyl group
contained polar nitro group (compound 37) or carboxylic ester
group (compound 38) at C-4 position, the antipromastigotes
activity was completely lost. In addition, the benzyl group
containing a fluorine atom at C-2 (compound 45), C-3
(compound 46), C-4 (compound 47) positions or two fluorine
atoms at C-3 and C-4 (compound 48) positions also resulted in
no improved antipromastigotes activity. Nevertheless, a benzyl
group with three fluorine atoms (compound 49) or a
trifluoromethyl group at C-4 position (compound 50) improved
8893
dx.doi.org/10.1021/jm301172v | J. Med. Chem. 2012, 55, 8891−8902

Journal of Medicinal Chemistry
Article
a
Table 2. Antipromastigotes Activities of Synthetic Flavonoid Dimers with Amino PEG Linkers
IC50 (μM)
promastigotes
macrophages
b
compd
LdAG83
3.3
Ld39
LdAG83PentR50
RAW 264.7
PEM
ND
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
53
56
57
58
59
60
61
68
69
70
71
72
73
2.3
1.1
2.1 0.9
8.3 1.0
6.4 1.5
16.0 6.2
>100.0
>79.0
70.3 9.7
45.7 19.1
>100.0
53.0 12.7
65.0
3.2 0.8
3.2 1.1
3.8 0.8
3.2
5.5 1.1
7.2 2.2
6.6 0.2
1.6
ND
ND
ND
1.1
ND
ND
ND
ND
8.4 6.1
>100.0
12.3 6.0
>100.0
ND
ND
ND
0.13 0.04
ND
27.0
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
>100.0
>100.0
ND
0.20 0.03
5.0 1.5
>100.0
0.21 0.06
4.2 1.5
>100.0
>88.0
>33.0
ND
0.75 0.25
100.0
0.90 0.60
>100.0
96.0 4.0
>100.0
85.0
ND
ND
8.9
18.4
ND
>100.0
ND
ND
ND
>100.0
ND
ND
ND
>100.0
ND
ND
ND
>100.0
ND
ND
ND
33.3
ND
ND
ND
14.1
ND
ND
ND
1.4 0.3
0.19 0.02
2.3 0.2
2.5 0.3
0.63 0.14
0.98 0.10
0.31 0.03
0.56 0.07
1.8 0.5
1.1 0.1
0.65 0.07
0.44 0.05
20.4 8.8
>50.0
ND
>58.0
>50.0
>11.0
>33.0
2.7 0.3
>50.0
>100.0
0.44 1.04
4.0 1.1
3.1 0.7
0.69 0.25
0.98 0.17
0.36 0.07
0.98 0.19
>50.0
ND
ND
ND
ND
ND
ND
ND
33.1 18.9
>100.0
>92.0
ND
>50.0
ND
>50.0
ND
>100.0
>33.0
0.58 0.11
ND
ND
ND
ND
a
Flavonoid dimers containing amino EG linker (compounds 31−73) with different substitutions on the linker were synthesized and tested for their
cytotoxicity towards promastigotes (LdAG83, Ld39, and LdAG83PentR50) and RAW 264.7 cells. IC50 values were presented as mean standard
error of mean. N = 1−4 independent experiments. ND = not determined. All compounds were dissolved in DMSO, and the highest percentage of
DMSO used was 1% at which no toxicity to promastigotes and RAW 264.7 macrophages was observed.
b
a
We tried to modify the pyridine ring on the amino linker in
order to improve the potency of compound 39 (Table 2). First,
we changed the position of the nitrogen atom in the pyridine ring
and found that the rank order of antipromastigotes activities was
as follows: para-position (compound 39; IC50 = 0.20 μM against
LdAG83) > meta-position (compound 68; IC50 = 0.56 μM) ≫
ortho-position (compound 40; IC50 = 5.0 μM). Second, a bromo
substitution at meta-position (compound 69) or ortho-position
(compound 71) and a cyano group at meta-position (compound
70) of the pyridine ring reduced the antipromastigotes activity,
especially in the Ld39 promastigotes. Third, replacement of
pyridine ring by pyrimidine ring (compound 72) exhibited at
least 2-fold lower antipromastigotes activity as compared to the
parent compound 39. Finally, compound 73 completely lost the
antipromastigotes activity probably due to bulkiness of quinidine
ring. Among all the flavonoid dimers, 39 has the strongest
antipromastigotes activity (IC50 ranging from 0.13 to 0.21 μM
against LdAG83, Ld39, and LdAG83PentR50) and very low
toxicity toward RAW 264.7 cells (IC50 > 100 μM). Importantly,
Scheme 1. Synthesis of Compound 53
a
Reagents and conditions: (a) 4-(chloromethyl)pyridine hydro-
chloride, K2CO3, acetonitrile, reflux, 3 h; (b) 4′-hydroxyflavone,
DIAD, PPh3, THF, reflux, 12 h.
8894
dx.doi.org/10.1021/jm301172v | J. Med. Chem. 2012, 55, 8891−8902

Journal of Medicinal Chemistry
Article
a
Scheme 2. Synthesis of Compounds 56−61
a
Reagents and conditions: (a) 4-(chloromethyl)pyridine hydrochloride, K2CO3, acetonitrile, reflux, 3 h; (b) corresponding hydroxyflavone, DIAD,
PPh3, THF, reflux, 12 h.
a
Scheme 3. Synthesis of Compounds 68−73
a
Reagents and conditions: (a) for compounds 62−67, RX, K2CO3, acetone, reflux, 3 h; (b) for compounds 68−73, 4′-hydroxyflavone, DIAD, PPh3,
THF, reflux, 12 h.
a
Table 3. Antipromastigotes Activities and Therapeutic Index of Standard Antileishmanials Compared to Flavonoid Dimer 39
IC50 (μM)
therapeutic index
promastigotes
promastigotes
Ld39
macrophages
compd
LdAG83
LdAG83PentR50
RAW 264.7
PEM
LdAG83
Ld39
LdAG83PentR50
b
SSG
pentamidine
2215.5 166.2
13.9 2.1
7089.6 747
5.3 0.8
ND
>11 000
30.0 5.0
12.0 2.5
>11 000
30.4 10.5
7.40 0.40
75.3 9.4
>100
>5.0
2.2
>1.6
5.7
ND
0.6
b
54.8 9.5
0.051 0.00
9.3 2.4
c
amphotericin B
0.09 0.02
4.3 1.1
0.095 0.02
17.3 2.1
24.5 2.4
89.0 7.8
>100.0
82.2
17.5
>1.6
0.7
77.9
4.4
145.1
8.1
b
miltefosine
20.0
>100
4
b
paromomycin
62.6 9.5
85.5 14
90.9 17
69.6 10.8
>4.1
0.9
>1.2
0.9
c
luteolin
114.3 20.6
102.9 9.0
0.2 0.03
43.2 8.0
95.6 2.8
>100
77.7 7.7
>100
c
quercetin
>1.0
>440
>1.0
>419
>1.4
>677
c
39
0.21 0.06
0.13
0
>88.0
a
IC50 values of current antileishmanials and flavonoid dimer 39 towards promastigotes (LdAG83, Ld39 and LdAG83PentR50) and macrophages
(RAW 264.7 and PEM) were determined. IC50 values were presented as mean standard error of mean. N = 1-4 independent experiments.
Therapeutic index was defined as the ratio of IC50 of antileishmanials towards PEM over promastigotes (LdAG83, Ld39 and LdAG83PentR50). ND
b
c
= not determined. Compounds were dissolved in sterile H2O. Compounds were dissolved in DMSO. No toxicity to the PEM cells was observed at
1% DMSO.
compound 39 was effective against SSG-resistant Ld39 and
pentamidine-resistant LdAG83PentR50.
Comparison of Antipromastigotes Activity of 39 with
Other Antileishmanial Agents. We have compared 39 with
other current antileishmanials, namely, SSG, pentamidine,
miltefosine, paromomycin, and amphotericin B (Table 3). SSG
has the lowest antipromastigotes activity against LdAG83 (IC50 =
2216 μM) and was essentially ineffective against SSG-resistant
8895
dx.doi.org/10.1021/jm301172v | J. Med. Chem. 2012, 55, 8891−8902

Journal of Medicinal Chemistry
Article
Ld39 (IC50 = 7090 μM). Second-line antileishmanials like
pentamidine and miltefosine displayed moderate antipromasti-
gote activity (IC50 values ranging from 5.3 to 54.8 μM and 4.3 to
17.3 μM, respectively) and moderate cytotoxicity toward
macrophage RAW 264.7 cell line (IC50 values of 30.0 μM and
20.0 μM, respectively). As expected, pentamidine was less
effective toward pentamidine-selected strain, LdAG83PentR50,
with IC50 3.9-fold higher than that of wild type LdAG83.
Paromomycin also displayed a poor antipromastigotes activity
(IC50 ranging from 24.5 to 85.5 μM) although it is nontoxic to
RAW 264.7 (IC50 > 100 μM). Amphotericin B has the highest
antipromastigotes activity (IC50 values ranging from 0.051 to
0.095 μM). However, it was also the most toxic compound for
RAW 264.7 cells (IC50 of 12.0 μM) (Table 3). The flavonoids
luteolin and quercetin have a low antipromastigotes activity with
IC50 ranging from 70 to 114 μM (Table 3). In comparison,
compound 39 has very potent antipromastigotes activity with
IC50 ranging from 0.13 to 0.21 μM, which is comparable to that of
the most potent antileishmanial, amphotericin B (Table 3).
Importantly, compound 39 was not toxic to RAW 264.7 and
PEM cells (IC50 > 100 μM and >88 μM, respectively) (Table 3).
We have also determined the therapeutic index (ratio of IC50
against PEM cells to IC50 against promastigotes). A therapeutic
index less than 10 would indicate probable nonselective
cytotoxicity for the tested compounds.11 As shown in Table 3,
compound 39 has the highest therapeutic index (from greater
than 419 to greater than 677), followed by amphotericin B
(77.9−145.1). Other compounds displayed a significantly lower
therapeutic index ranging from 0.6 to 17.5.
In Vitro Antiamastigotes Activity of Flavonoid Dimer
39 and Its Derivatives. Here, we tested if compound 39 and its
derivatives have antiamastigotes activity. Intracellular amasti-
gotes of Ld39 were obtained by infecting PEM cells with late-log
phase Ld39 promastigotes. A very pronounced antiamastigote
activity was noted for compound 39 compared to solvent
control. Numerous amastigotes were observed in host macro-
phages in control (Figure 1A) whereas only a few amastigotes
were noted in the compound 39-treated case (5 μM) (Figure
1B). Percentage of macrophage infected was decreased from
100% in solvent control to 20% in compound 39-treated case
(Figure 1C). Number of amastigotes per 100 macrophages was
also reduced from 887 to 48 after adding 5 μM of compound 39
(Figure 1C). Similar antiamastigote activities were determined
for a number of structural derivatives of compound 39 (Table 4).
We have compared compound 39 and its derivatives with
other antileishmanial compounds in terms of their antiamasti-
gotes activity and therapeutic index. As shown in Table 4, we
found that antiamastigotes activity of compound 39 and its
derivatives followed a similar trend as antipromastigotes, with
compound 39 being the most active (IC50 = 0.63 0.12 μM).
Compound 53 (with shorter amino PEG linker than compound
39) showed significantly lower antiamastigotes activity. Attach-
ment position of the two flavones to the amino PEG linker is also
important. Compounds 56 (linker connected to flavones at C-3′
position of B-ring), compound 57 (linker connected to flavones
at C-2′ position of B-ring), and compound 58 (linker connected
to flavones at C-3 position of C-ring) have an IC50 3.7-fold, 11.1-
fold, and 4.9-fold lower than compound 39 (linker connected at
C-4′ position of B-ring). Antiamastigotes activity of compound
59 (C3′-methoxy substitution in the B-ring) cannot be measured
because it was toxic to host PEM cells (IC50 = 2.7 0.3 μM). In
contrast, C3-methoxy substitution in the C-ring (compound 60)
or addition of a fluorine atom at C-6 position of A-ring
Figure 1. Antiamastigotes activity of compound 39 against SSG-
resistant L. donovani Ld39 amastigotes grown in PEM cells. PEM cells
were infected with late-log stage promastigotes of Ld39 for 24 h at 37 °C.
Infected macrophages were then treated with either DMSO (A) or 5 μM
compound 39 (B) for 3 days at 37 °C. Two representative microscopy
pictures are shown in each treatment group. After 3 days, the coverslips
were stained with Giemsa. Percentage of macrophage infected and the
number of amastigotes per 100 macrophages were determined (C). The
black column represents the percentage of macrophage infected, and the
striped column represents the number of amastigotes per 100
macrophages.
(compound 61) did not cause toxic effect toward the PEM cells.
However, the antiamastigotes activity of compounds 60 and 61
were still 8.3-fold and 4.9-fold lower than the unsubstituted
compound 39. Therefore, substitution at two flavones did not
improve antiamastigotes activity of compound 39.
When we changed the position of the nitrogen atom in the
pyridine ring, we found that the rank order of antiamastigotes
activities was as follows: para-position (compound 39; IC50
=
0.63 μM against LdAG83) ≫ meta-position (compound 68; IC50
= 2.4 μM) > ortho-position (compound 40; IC50 = 4.1 μM). A
bromo substitution at meta-position (compound 69) and ortho-
positions (compound 71) or a cyano group at meta-position
(compound 70) of the pyridine ring also dramatically reduced
antiamastigotes activity. Replacement of pyridine ring by
pyrimidine ring (compound 72) exhibited about 2-fold lower
antiamastigotes activity as compared to the parent compound 39.
In terms of therapeutic index, compound 39 is still the safest
compound among the synthetic derivatives because of its high
antiamastigotes activities and the lower toxicity toward the PEM
cells (therapeutic index >139.7). We have also determined
antiamastigotes activity and therapeutic index of various
antileishmanials. Although SSG, luteolin, and quercetin dis-
played a very poor antipromastigote activity (Table 2), they were
found to possess an improved antiamastigotes activity with IC50
values of 675, 7.7, and 6.0 μM, respectively (Table 4). This
represents 11-, 12-, and at least 17-fold increase in
8896
dx.doi.org/10.1021/jm301172v | J. Med. Chem. 2012, 55, 8891−8902

Journal of Medicinal Chemistry
Article
Table 4. Antiamastigotes Activities and Therapeutic Index of
Flavonoid Dimer 39 and Its Derivatives Compared to
a
Standard Antileishmanials
IC50 (μM)
therapeutic index
compd
amastigotes Ld39 macrophages PEM amastigotes Ld39
39
40
53
56
57
58
59
60
61
68
69
70
71
72
SSG
0.63 0.12
4.1 2.1
5.8 3.3
2.3 1.0
7.0
>88.0
>33.0
>139.7
>8.0
>58.0
>10.0
>21.7
>1.6
>50.0
>11.0
3.1 0.1
toxic
>33.0
>10.6
ND
2.7 0.3
>50.0
5.2 1.7
3.1 2.7
2.4 0.1
>10.0
>9.6
>100.0
33.1 18.9
>100.0
>92.0
>32.3
13.8
>10.0
>12.1
>13.1
>23.6
>16.3
<1.0
7.6 1.0
>7.6
>100.0
>33.0
1.4 0.8
675 75
>30.0
>11 000
30.4 10.5
75.3 9.4
>100.0
7.4 0.4
77.7 7.7
>100.0
Figure 2. Dose response curve for antileishmanial activities of 39,
amphotericin B, and SSG against Ld39 parasites. Ld39 promastigotes
(A) or amastigotes grown in PEM (B) were treated with different
concentrations of 39, amphotericin B, or SSG, and their percentages of
survival are plotted against log concentration of compounds used (in
molar). Data is presented as mean standard error of mean.
pentamidine
miltefosine
paromomycin
amphotericin B
luteolin
16.0 6.4
41.0
4.7
>2.4
0.062 0.00
7.7 3.6
6.0 1.1
119.4
10.1
quercetin
>16.7
a
PEM cells were infected with late-log Ld39 promastigotes of SSG-
resistant Ld39 for 24 h at 37 °C. Infected macrophages were then
treated with various antileishmanials and incubated for 3 days at 37 °C.
After 3 days, the coverslips were stained with Giemsa. The number of
amastigotes per 100 macrophages was determined and used to
calculate IC50 values. The values were presented as mean standard
error of mean. N = 1−3 independent experiments. Therapeutic index
was determined by dividing IC50 towards PEM cells over IC50 towards
Ld39 amastigotes. Therapeutic index smaller than 10 suggests
probable non-selective cytotoxicity for the tested compounds.11
DISCUSSION
■
In the present study, we demonstrate that a group of 63 flavonoid
dimers displayed novel and promising antipromastigotes and
antiamastigotes activities. We are able to draw some structure−
activity relationships. (1) Comparison of series A (with OH-
group on the two flavone rings) and series B (removal of OH-
group from the two flavone rings) reveals that the linker chain
length is crucial for selective antileishmanial activity and
noncytotoxicity of flavonoid dimers (Table 1). In series A,
good antipromastigotes activity and nontoxicity toward RAW
264.7 cells were observed for PEG linker length with 3−5 EG
units. On the other hand, series B showed that, for linker with 5−
13 EG units, high toxicity toward RAW 264.7 cells were observed
and that rendered them unsuitable as selective antileishmanials
(Table 1). (2) Series C clearly demonstrates that some
hydrophobic substitutions including ethyl (compound 20) at
C-6 position, methyl (compound 26), or fluorine (compound 28
and 29) at C-7 position of ring A of flavone would help in
regaining the selective leishmanicidal activity as compared to the
parent compounds 9 and 10 with 4 and 5 EG units (Table 1). (3)
The amine linked flavonoid dimers reveal that an amine group in
the linker has a strong influence on both leishmanicidal activity
and cytotoxicity. The selectivity depends critically on the nature
of the R substituent on the amine group. Substituents including
Boc- (compound 35), pyridine ring- (compound 39 and 40), and
mesylate- (compound 42) on the amino linker of dimers
markedly enhance the leishmanicidal activity (Table 2).
antiamastigotes activity compared to that of antipromastigotes,
respectively. This stage-dependent leishmanicidal activity
suggests that their targets are only present in intracellular
amastigotes but not in extracellular promastigotes. Pentamidine,
miltefosine, and paromomycin displayed a moderate antiamas-
tigotes activity with IC50 values of >30.0, 16.0, and 41.0 μM,
respectively (Table 4). Therapeutic indexes of these six
compounds were smaller than 17. Amphotericin B has the
highest antiamastigotes activity (0.062 μM), followed by
compound 39 (0.63 0.12 μM) (Table 4). Amphotericin B,
however, was also the most toxic to PEM cells (IC50 = 7.4 μM).
In contrast, compound 39 was almost nontoxic to PEM cells
(IC50 > 88 μM). Put together, the therapeutic index of
amphotericin B (therapeutic index =119.4) was slightly lower
than that of compound 39 (therapeutic index >139.7),
suggesting that both of them possess very high selective
cytotoxicity for Leishmania amastigotes.
In summary, we found that flavonoid dimer 39 was highly
potent against both promastigotes (IC50 = 0.21 μM) and
amastigotes (IC50 = 0.63 μM) (Figure 2). Its activity was
comparable to or slightly lower than that of amphotericin B (IC50
= 0.09 μM for promastigotes and 0.062 μM for amastigotes) and
much better than SSG (Figure 2). Together with its low toxicity
against macrophages, 39 is a potential candidate compound for
further development into antileishmanial drug.
A good antileishmanial compound should be both potent and
safe to use. Among all flavonoid dimers tested, compound 39
exhibited the strongest antipromastigotes activity with IC50
ranging from 0.13 to 0.21 μM (Table 2), and it is also very
active against Ld39 amastigotes with IC50 of 0.63 μM (Table 4).
It displayed no toxic effect to macrophage cell line RAW 264.7
(IC50 > 100 μM) (Table 2) or primary mouse PEM cells (IC50 >
88 μM) (Tables 4 and 5). Importantly, compound 39 displayed
8897
dx.doi.org/10.1021/jm301172v | J. Med. Chem. 2012, 55, 8891−8902

Journal of Medicinal Chemistry
Article
consistent efficacy against both drug-sensitive Leishmania as well
as SSG-resistant Ld39 and pentamidine-resistant LdAG83-
PentR50 (Tables 2 and 3). In comparison with other clinically
used antileishmanials, compound 39 showed stronger anti-
promastigotes (Table 3) and antiamastigotes (Table 4) activity
than most antileishmanials including SSG, pentamidine,
miltefosine, and paromomycin (Table 3). Compound 39 was
about 2-fold (Table 3) and 10-fold (Table 4) less potent than the
second-line antileishmanial amphotericin B (Tables 3 and 4)
against promastigotes and amastigotes, respectively. However,
amphotericin B is also the most toxic compound toward
macrophages RAW 264.7 (IC50 = 12.0 μM) and primary
mouse PEM cells (IC50 = 7.4 μM) (Tables 3 and 4). Overall,
therapeutic index of compound 39 (>139.7) was higher than that
of amphotericin B (119.4) (Table 4), thus rendering compound
39 suitable as a potential leishmanicidal agent.
Flavonoids are widely distributed in the plant kingdom and
have long been studied for their antiparasitic activities.3,12−15
Similar to SSG, monomeric flavonoids such as luteolin and
quercetin were demonstrated to solely exhibit promising
leishmanicidal activity for intracellular amastigotes (IC50 = 7.7
and 6.0 μM), but not for extracellular promastigotes (IC50 = 89−
114 μM and 70−103 μM) (Tables 3 and 4), suggesting that their
specific targets are present only in the amastigote form.
Quercitrin (quercetin 3-O-α-L-rhamnopyranoside), one of the
constituents of the biologically active aqueous extract obtained
from Kalanchoe pinnata, has been shown to have potent
antileishmanial activity toward amastigote form of L. ama-
zonensis, but not to promastigote form.16 It has been previously
reported that some polyphenols did not show selective toxicity
for the Leishmania extracellular form, but displayed pronounced
effect against Leishmania amastigotes.17,18 In the present study,
synthetic flavonoid dimer, 39, was demonstrated to possess very
potent antipromastigotes and antiamastigotes activity, suggesting
that its target is likely different from the monomeric flavonoids.
The level of that putative target is therefore predicted to be
constitutively expressed in both Leishmania stages.
Biflavonoids are rich in many species of plant and reported to
have significant antiviral and antiprotozoal activity.11,19−22
Biflavonoids are characterized by two flavonoid monomeric
units (flavone or flavanone) covalently linked either with C−C or
C−O−C bonds.11 In vitro studies by Weniger et al. revealed that
six out of eight biflavonoids tested showed potent antiaxenic
amastigote activity including lanaroflavone (IC50 = 7.2 μM),
bilobetin (IC50 = 2.7 μM), ginkgetin (IC50 = 2.8 μM),
isoginkgetin (IC50 = 1.9 μM), and sciadopitysin (IC50 = 8.2
μM).11 In comparison, synthetic flavonoid dimer 39 (IC50 = 0.63
μM) is about 3-fold more active in killing amastigotes than the
most active biflavonoid, isoginkgetin. However, isoginkgetin was
moderately toxic to normal L6 cells (IC50 = 11 μM), therefore
giving a relatively low therapeutic index of about 5.8.
Comparatively, compound 39 showed very selective and
promising leishmanicidal activity as indicated by the high value
of therapeutic index (>139.7, Table 4).
The possible mechanism of action of flavonoids in killing
Leishmania has been discussed recently.17,23−25 Instead of direct
antiparasitic activity, polyphenol and proanthcyanidins could
induce nitric oxide and tumor necrosis factor-α (TNF) in
macrophages.17 Quercetin and luteolin are found to be inhibitor
of topoisomerase and induce cell cycle arrest leading to apoptosis
in Leishmania.23−25 A biflavonoid, 2″,3″-diidroochnaflavone,
isolated from the leaves of Luxemburgia nobilis, was cytotoxic to
murine Ehrlich carcinoma and human leukemia K562 cells and
found to inhibit topoisomerase I and II-α.26 Morelloflavone, a
biflavonoid, was found to inhibit tumor angiogenesis by targeting
Rho GTPase and extracellular signal-regulated kinase signaling
pathway.27 However, the mode of action of antileishmanial
biflavonoid remains unclear. We are currently trying to identify
the potential target(s) of compound 39.
In summary, the present study suggested that flavonoid
dimers, especially those with amino linker, have marked
leishmanicidal potency. Compound 39 showed very consistent
and promising leishmanicidal activity irrespective of the drug-
sensitivity of parasite for both extracellular promastigotes (IC50
ranging from 0.13 to 0.21 μM) and intracellular amastigotes
(IC50 = 0.63 μM). Moreover, it displayed no toxicity for
macrophages, and its high value of therapeutic index was
comparable to other current antileishmanials like amphotericin
B. With its amine and pyridine functions, compound 39 or its
salts have reasonable solubility in aqueous media. Therefore,
compound 39 can be potentially developed into a new, safe, and
effective drug for the treatment of leishmaniasis, especially the
drug refractory parasites.
EXPERIMENTAL SECTION
■
Chemicals. The flavonoid dimers 1−50 and amino diol 54 were
prepared according to the reported procedures.5,6,8,10 Amino diol 51 and
3-hydroxyflavone were commercially available. 2′-Hydroxyflavone, 3′-
hydroxyflavone, 4′-hydroxyflavone, 3′-methoxy-4′-hydroxyflavone, 3-
methoxy-4′-hydroxyflavone, and 6-fluoro-4′-hydroxyflavone were pre-
pared according to the reported procedures. All NMR spectra were
recorded on a Bruker MHz DPX400 spectrometer at 400.13 MHz for 1H
and 100.62 MHz for 13C. All NMR measurements were carried out at
room temperature, and the chemical shifts are reported as parts per
million (ppm) in unit relative to the resonance of CDCl3 (7.26 ppm in
1
the H, 77.0 ppm for the central line of the triplet in the 13C modes,
respectively). Low-resolution and high-resolution mass spectra were
obtained on a Micromass (U.K.) Q-TOF-2 by electron spray ionization
(ESI) mode. Melting points were measured using Electrothermal
IA9100 digital melting point apparatus and were uncorrected. All
reagents and solvents were reagent grade and were used without further
purification unless otherwise stated. The plates used for thin-layer
chromatography (TLC) were E. Merck silica gel 60F254 (0.25-mm
thickness), and they were visualized under short (254-nm) and long
(365-nm) UV light. Chromatographic purifications were carried out
using MN silica gel 60 (230−400 mesh). The purity of tested
compounds was determined by HPLC, which was performed by using
an Agilent 1100 series installed with an analytic column of Agilent Prep-
Sil Scalar column (4.6 mm ×250 mm, 5 μm) at UV detection of 320 nm
(reference at 450 nm) with isocratic elution of hexane (50%)/ethyl
acetate (25%)/methanol (25%) at a flow rate of 1 mL/min. All tested
compounds were shown to have >95% purity according to HPLC.
Pentamidine, amphotericin B, paromomycin, luteolin, and quercetin
were purchased from Sigma. Miltefosine was from Cayman. SSG was a
generous gift from Glaxo Smith Kline.
General Procedure I for Synthesis of Amino Diols 52, 55, and 62−
67. To a round-bottom flask was added amino diol 51 or 54,
corresponding arylmethyl halides, excess K2CO3, and acetone or
acetonitrile. The reaction mixture was stirred at refluxing temperature
for 3 h. When TLC indicated complete consumption of starting
material, the reaction mixture was cooled and filtered to remove excess
K2CO3.The obtained filtrate was evaporated to give a crude reaction
mixture, which was subjected to purification by flash column
chromatography on silica gel with 3−20% MeOH in CH2Cl2 as eluent
to furnish the desired compound.
General Procedure II for the Synthesis of Flavonoid Dimers 53,
56−61, and 68−73. To a round-bottom flask, containing the
corresponding amino diol, monohydroxyflavone, triphenylphosphine,
and tetrahydrofuran (THF), was added diisopropyl azodicarboxylate
(DIAD) dropwise. The reaction mixture was stirred at refluxing
8898
dx.doi.org/10.1021/jm301172v | J. Med. Chem. 2012, 55, 8891−8902

Journal of Medicinal Chemistry
Article
temperature under nitrogen atmosphere for 12 h. When TLC indicated
complete consumption of starting material, the reaction mixture was
cooled and filtered through a short pad of silica gel. The obtained filtrate
was evaporated to give a crude reaction mixture, which was subjected to
purification by flash column chromatography on silica gel with 10−30%
acetone in CH2Cl2 as eluent to furnish the desired compound.
2,2′-((Pyridin-4-ylmethyl)azanediyl)diethanol (52). This com-
pound was obtained as a pale brown oil (1.6 g, 18%) from 51 (4.8 g,
45.7 mmol), 4-(chloromethyl)pyridine hydrochloride (7.5 g, 45.7
mmol), K2CO3 (13.0 g, 94.2 mmol), and acetonitrile (140 mL)
according to the general procedure I described above. 1H NMR
(CDCl3) δ 2.60 (t, J = 5.20 Hz, 4 H), 3.55 (t, J = 5.20 Hz, 4H), 3.64 (s,
2H), 4.25 (br, 2H), 7.25 (d, J = 6.0 Hz, 2H), 8.33 (d, J = 6.0 Hz, 2H). 13C
NMR (CDCl3) δ 56.2, 58.4, 59.5, 123.9, 149.1, 149.4. LRMS (ESI) m/z
197 (M+ + H, 100), 219 (M+ + Na, 18). HRMS (ESI) Calcd for
C10H17N2O2 (M+ + H) 197.1290. Found 197.1281.
1,7-Bis[4′-(4H-chromen-4-on-2-yl)phenyl]-4-(4-pyridylmethyl)-
1,7-dioxa-4-azaheptane (53). This compound was obtained as a white
foam (0.81 g, 23%) from 52 (1.08 g, 5.51 mmol), 4′-hydroxyflavone
(2.65 g, 11.1 mmol), triphenylphosphine (3.10 g, 11.8 mmol), THF (30
mL), and DIAD (2.40 g, 11.8 mmol) according to the general procedure
II described above. 1H NMR (CDCl3) δ 3.13 (t, J = 4.8 Hz, 4H), 3.93 (s,
2H), 4.16 (t, J = 4.8 Hz, 4H), 6.73 (s, 2H), 6.97 (d, J = 8.6 Hz, 4H), 7.34
(dd, J = 7.2, 8.0 Hz, 2H), 7.38 (d, J = 6.0 Hz, 2H), 7.52 (d, J = 8.4 Hz,
2H), 7.67 (dd, J = 7.2, 8.0 Hz, 2H), 7.86 (d, J = 8.6 Hz, 4H), 8.19 (dd, J =
1.2, 8.0 Hz, 2H), 8.57 (d, J = 6.0 Hz, 2H). 13C NMR (CDCl3) δ 53.6,
59.0, 67.0, 106.2, 114.9, 117.9, 123.5, 123.9, 124.2, 125.1, 125.6, 128.0,
133.6, 149.0, 149.7, 156.1, 161.4, 163.2, 178.3. LRMS (ESI) m/z 637
(M+ + H, 100), 659 (M+ + Na, 12). HRMS (ESI) Calcd for C40H33N2O6
(M+ + H) 637.2339. Found 637.2330.
N-(Pyridin-4′-ylmethyl)-3,9-dioxa-6-azaundecane-1,11-diol (55).
This compound was obtained as a pale brown oil (3.6 g, 31%) from
54 (6.0 g, 31.1 mmol), 4-(chloromethyl)pyridine hydrochloride (6.6 g,
40.2 mmol), K2CO3 (11.2 g, 81.2 mmol), and acetonitrile (70 mL)
according to the general procedure I described above. 1H NMR
(CDCl3) δ 2.75 (t, J = 4.8 Hz, 4H), 3.55 (t, J = 4.8 Hz, 4H), 3.61 (t, J =
4.8 Hz, 4H), 3.71 (t, J = 4.8 Hz, 4H), 3.72 (s, 2H), 4.20 (br, 2H), 7.34 (d,
J = 6.0 Hz, 2H), 8.56 (d, J = 6.0 Hz, 2H). 13C NMR (CDCl3) δ 54.7, 58.7,
61.6, 68.7, 72.5, 124.0, 147.8, 149.7. LRMS (ESI) m/z 285 (M+ + H,
100), 307 (M+ + Na, 60). HRMS (ESI) Calcd for C14H24N2O4Na (M+ +
Na) 307.1634. Found 307.1635.
1,13-Bis[3′-(4H-chromen-4-on-2-yl)phenyl]-7-(4-pyridylmethyl)-
1,4,10,13-tetraoxa-7-azatridecane (56). This compound was obtained
as a white foam (0.25 g, 35%) from 55 (0.28 g, 0.99 mmol), 3′-
hydroxyflavone (0.48 g, 2.02 mmol), triphenylphosphine (0.54 g, 2.06
mmol), THF (20 mL), and DIAD (0.42 g, 2.08 mmol) according to the
general procedure II described above. 1H NMR (CDCl3) δ 2.87 (t, J =
5.6 Hz, 4H), 3.70 (t, J = 5.6 Hz, 4H), 3.82 (t, J = 4.4 Hz, 4H), 3.83 (s,
2H), 4.17 (t, J = 4.4 Hz, 4H), 6.78 (s, 2H), 7.05 (dd, J = 2.4, 8.0 Hz, 2H),
7.33 (d, J = 5.2 Hz, 2H), 7.37 − 7.44 (m, 6H), 7.49 (d, J = 8.0 Hz, 2H),
7.55 (d, J = 8.0 Hz, 2H), 7.69 (dd, J = 8.4, 8.4 Hz, 2H), 8.22 (dd, J = 1.2,
7.6 Hz, 2H), 8.50 (d, J = 5.6 Hz, 2H). 13C NMR (CDCl3) δ 54.2, 58.8,
67.7, 69.5, 70.2, 107.8, 112.5, 117.7, 118.1, 118.9, 123.6, 124.0, 125.7,
130.1, 133.1, 133.8, 149.6, 156.2, 159.2, 163.1, 178.3. LRMS (ESI) m/z
725 (M+ + H, 100), 747 (M+ + Na, 35). HRMS (ESI) Calcd for
C44H41N2O8 (M+ + H) 725.2863. Found 725.2842.
(M+ + H, 100), 747 (M+ + Na, 18). HRMS (ESI) Calcd for C44H41N2O8
(M+ + H) 725.2863. Found 725.2849.
1,13-Bis[2-phenyl-4H-chromen-4-on-3-yl]-7-(4-pyridylmethyl)-
1,4,10,13-tetraoxa-7-azatridecane (58). This compound was obtained
as a white foam (0.23 g, 32%) from 55 (0.28 g, 0.99 mmol), 3-
hydroxyflavone (0.48 g, 2.02 mmol), triphenylphosphine (0.54 g, 2.06
mmol), THF (20 mL), and DIAD (0.42 g, 2.08 mmol) according to the
general procedure II described above. 1H NMR (CDCl3) δ 2.60 (t, J =
5.6 Hz, 4H), 3.40 (t, J = 5.6 Hz, 4H), 3.61 (s, 2H), 3.61 (t, J = 4.8 Hz,
4H), 4.24 (t, J = 4.8 Hz, 4H), 7.20 (d, J = 6.0 Hz, 2H), 7.33 (dd, J = 5.6,
8.0 Hz, 2H), 7.39−7.42 (m, 6H), 7.48 (d, J = 8.0 Hz, 2H), 7.62 (dd, J =
5.6, 8.0 Hz, 2H), 8.09−8.12 (m, 4H), 8.20 (dd, J = 1.2, 8.0 Hz, 2H), 8.43
(d, J = 8.0 Hz, 2H). 13C NMR (CDCl3) δ 54.0, 58.6, 69.6, 70.2, 71.4,
118.0, 123.5, 124.1, 124.6, 125.7, 128.3, 128.7, 130.6, 130.9, 133.4, 140.4,
149.5, 149.5, 155.2, 155.4, 175.0. LRMS (ESI) m/z 725 (M+ + H, 100),
747 (M+ + Na, 45). HRMS (ESI) Calcd for C44H41N2O8 (M+ + H)
725.2863. Found 725.2890.
1,13-Bis[4′-(4H-chromen-4-on-2-yl)-2-methoxyphenyl]-7-(4-pyri-
dylmethyl)-1,4,10,13-tetraoxa-7-azatridecane (59). This compound
was obtained as a white foam (0.26 g, 23%) from 55 (0.42 g, 1.48 mmol),
3′-methoxy-4′-hydroxyflavone (0.80 g, 2.99 mmol), triphenylphosphine
(0.82 g, 3.13 mmol), THF (20 mL), and DIAD (0.62 g, 3.07 mmol)
according to the general procedure II described above. 1H NMR
(CDCl3) δ 2.78 (t, J = 5.6 Hz, 4H), 3.64 (t, J = 5.6 Hz, 4H), 3.74 (s, 2H),
3.80 (t, J = 4.8 Hz, 4H), 3.88 (s, 6H), 4.17 (t, J = 4.8 Hz, 4H), 6.68 (s,
2H), 6.95 (d, J = 8.4 Hz, 2H), 7.28 (d, J = 6.0 Hz, 2H), 7.31−7.37 (m,
4H), 7.44−7.51 (m, 4H), 7.64 (dd, J = 3.2, 8.4 Hz, 2H), 8.15 (d, J = 8.0
Hz, 2H), 8.46 (d, J = 4.4 Hz, 2H). 13C NMR (CDCl3) δ 54.1, 56.0, 58.7,
68.5, 69.3, 70.2, 106.4, 109.3, 112.9, 117.9, 119.8, 123.5, 123.8, 124.5,
125.1, 125.5, 133.6, 149.5, 149.6, 151.4, 156.1, 163.2, 178.2. LRMS
(ESI) m/z 785 (M+ + H, 65), 807 (M+ + Na, 100). HRMS (ESI) Calcd
for C46H44N2O10Na (M+ + Na) 807.2894. Found 807.2899.
1,13-Bis[4′-(3-methoxy-4H-chromen-4-on-2-yl)phenyl]-7-(4-pyri-
dylmethyl)-1,4,10,13-tetraoxa-7-azatridecane (60). This compound
was obtained as a white foam (0.18 g, 26%) from 55 (0.25 g, 0.88 mmol),
3-methoxy-4′-hydroxyflavone (0.48 g, 1.79 mmol), triphenylphosphine
(0.50 g, 1.91 mmol), THF (20 mL), and DIAD (0.38 g, 1.88 mmol)
according to the general procedure II described above. 1H NMR
(CDCl3) δ 2.84 (t, J = 5.6 Hz, 4H), 3.67 (t, J = 5.6 Hz, 4H), 3.79 (s, 2H),
3.82 (t, J = 4.8 Hz, 4H), 3.88 (s, 6H), 4.16 (t, J = 4.8 Hz, 4H), 7.02 (d, J =
7.2 Hz, 4H), 7.33 (d, J = 4.8 Hz, 2H), 7.36 (dd, J = 3.2, 7.6 Hz, 2H), 7.51
(d, J = 7.6 Hz, 2H), 7.65(dd, J = 3.2, 7.6 Hz, 2H), 8.10 (d, J = 7.2 Hz,
4H), 8.24 (dd, J = 1.2, 7.2 Hz, 2H), 8.51(br, 2H). 13C NMR (CDCl3) δ
54.2, 58.8, 59.9, 67.5, 69.4, 70.2, 114.5, 117.9, 123.4, 123.6, 124.2, 124.6,
125.7, 130.2, 133.3, 140.8, 149.5, 149.6, 155.1, 155.4, 160.7, 174.9.
LRMS (ESI) m/z 785 (M+ + H, 100), 807 (M+ + Na, 40). HRMS (ESI)
Calcd for C46H45N2O10 (M+ + H) 785.3074. Found 785.3063.
1,13-Bis[4′-(6-fluoro-4H-chromen-4-on-2-yl)phenyl]-7-(4-pyridyl-
methyl)-1,4,10,13-tetraoxa-7-azatridecane (61). This compound was
obtained as a white foam (0.13 g, 12%) from 55 (0.42 g, 1.48 mmol), 6-
fluoro-4′-hydroxyflavone (0.77 g, 3.01 mmol), triphenylphosphine
(0.82 g, 3.13 mmol), THF (20 mL), and DIAD (0.62 g, 3.07 mmol)
according to the general procedure II described above. 1H NMR
(CDCl3) δ 2.83 (t, J = 6.0 Hz, 4H), 3.66 (t, J = 6.0 Hz, 4H), 3.78 (s, 2H),
3.80 (t, J = 4.8 Hz, 4H), 4.14 (t, J = 4.8 Hz, 4H), 6.67 (s, 2H), 6.98 (d, J =
8.8 Hz, 4H), 7.32 (d, J = 5.6 Hz, 2H), 7.38 (dd, J = 3.2, 7.6 Hz, 2H), 7.38
(dd, J = 4.4, 9.2 Hz, 2H), 7.81(m, 6H), 8.49 (d, J = 5.6 Hz, 2H). 13C
NMR (CDCl3) δ 54.3, 58.8, 67.7, 69.3, 70.2, 105.4, 110.4, 110.6, 115.0,
120.0, 120.0, 121.5, 121.8, 123.5, 123.8, 125.0, 125.1, 128.0, 149.5, 149.6,
152.3, 158.3, 160.7, 161.7, 163.5, 177.4. LRMS (ESI) m/z 761 (M+ + H,
100), 783 (M+ + Na, 30). HRMS (ESI) Calcd for C44H39N2O8F (M+ +
H) 761.2674. Found 761.2659.
1,13-Bis[2′-(4H-chromen-4-on-2-yl)phenyl]-7-(4-pyridylmethyl)-
1,4,10,13-tetraoxa-7-azatridecane (57). This compound was obtained
as a white foam (0.21 g, 29%) from 55 (0.28 g, 0.99 mmol), 2′-
hydroxyflavone (0.48 g, 2.02 mmol), triphenylphosphine (0.54 g, 2.06
mmol), THF (20 mL), and DIAD (0.42 g, 2.08 mmol) according to the
general procedure II described above. 1H NMR (CDCl3) δ 2.77 (t, J =
5.6 Hz, 4H), 3.60 (t, J = 5.6 Hz, 4H), 3.69 (s, 2H), 3.77 (t, J = 4.8 Hz,
4H), 4.17 (t, J = 4.8 Hz, 4H), 7.00 (d, J = 8.4 Hz, 2H), 7.08 (dd, J = 7.2,
8.0 Hz, 2H), 7.13 (s, 2H), 7.20 (d, J = 5.6 Hz, 2H), 7.36 (dd, J = 7.2, 8.0
Hz, 2H), 7.41 (dd, J = 7.2, 8.0 Hz, 2H), 7.49 (d, J = 8.0 Hz, 2H), 7.64
(dd, J = 7.2, 8.0 Hz, 2H), 7.87 (d, J = 8.0 Hz, 2H), 8.19 (d, J = 7.2 Hz,
2H), 8.35 (d, J = 8.0 Hz, 2H). 13C NMR (CDCl3) δ 54.1, 58.5, 68.4, 69.2,
70.2, 112.7, 113.0, 118.0, 121.0, 121.2, 123.5, 123.8, 124.8, 125.5, 129.3,
132.3, 133.5, 149.4, 156.4, 157.2, 160.8, 178.6. LRMS (ESI) m/z 725
N-(Pyridin-3′-ylmethyl)-3,9-dioxa-6-azaundecane-1,11-diol (62).
This compound was obtained as a pale brown oil (1.1 g, 36%) from
54 (2.0 g, 10.4 mmol), 3-(bromomethyl)pyridine hydrobromide (2.7 g,
10.7 mmol), K2CO3 (3.1 g, 22.5 mmol), and acetone (40 mL) according
to the general procedure I described above. 1H NMR (CDCl3) δ 2.74 (t,
J = 5.38 Hz, 4 H), 3.52−3.61 (m, 8 H), 3.67−3.74 (m, 6 H), 4.26 (br, 2
H), 7.26−7.30 (m, 1 H), 7.75−7.79 (m, 1 H), 8.49−8.54 (m, 2 H). 13C
NMR (CDCl3) δ 54.4, 56.8, 61.6, 68.7, 72.5, 123.4, 133.7, 137.0, 148.7,
8899
dx.doi.org/10.1021/jm301172v | J. Med. Chem. 2012, 55, 8891−8902

Journal of Medicinal Chemistry
Article
150.4. LRMS (ESI) m/z 285 (M+ + H, 97), 307 (M+ + Na, 100). HRMS
(ESI) Calcd for C14H24N2O4Na (M+ + Na) 307.1634. Found 307.1631.
N-(2′-Bromopyridin-4′-ylmethyl)-3,9-dioxa-6-azaundecane-1,11-
diol (63). This compound was obtained as a pale brown oil (0.7 g, 38%)
from 54 (1.0 g, 5.2 mmol), 4-(bromomethyl)-2-bromopyridine (1.3 g,
5.1 mmol), K2CO3 (0.8 g, 5.9 mmol), and acetone (40 mL) according to
the general procedure I described above. 1H NMR (CDCl3) δ 2.77 (t, J =
5.14 Hz, 4 H), 3.55−3.64 (m, 8 H), 3.67−3.80 (m, 6 H), 7.28−7.47 (m,
1 H), 7.59 (s, 1 H), 8.31 (d, J = 4.89 Hz, 1 H). 13C NMR (CDCl3) δ 54.6,
58.1, 61.5, 68.8, 72.4, 122.8, 127.9, 142.4, 149.9, 152.1. LRMS (ESI) m/z
363 (M+ + H, 100), 385 (M+ + Na, 38). HRMS (ESI) Calcd for
C14H24N2O4Br (M+ + H) 363.0919. Found 363.0921.
N-(2′-Cyanopyridin-4′-ylmethyl)-3,9-dioxa-6-azaundecane-1,11-
diol (64). This compound was obtained as a pale brown oil (0.9 g, 36%)
from 54 (1.5 g, 7.8 mmol), 4-bromomethyl-2-pyridinecarbonitrile (1.6
g, 8.1 mmol), K2CO3 (1.2 g, 8.7 mmol), and acetone (50 mL) according
to the general procedure I described above. 1H NMR (CDCl3) δ 2.75 (t,
J = 5.14 Hz, 4 H), 3.50−3.61 (m, 8 H), 3.65−3.74 (m, 4 H), 3.77 (s, 2
H), 7.58−7.60 (m, 1 H), 7.82 (s, 1 H), 8.61 (d, J = 5.87 Hz, 1 H). 13C
NMR (CDCl3) δ 54.7, 58.0, 61.6, 68.8, 72.4, 117.4, 126.8, 128.5, 133.8,
150.9, 150.9. LRMS (ESI) m/z 310 (M+ + H, 100), 332 (M+ + Na, 18).
HRMS (ESI) Calcd for C15H24N3O4 (M+ + H) 310.1767. Found
310.1755.
N-(3′-Bromopyridin-4′-ylmethyl)-3,9-dioxa-6-azaundecane-1,11-
diol (65). This compound was obtained as a pale brown oil (1.0 g, 43%)
from 54 (1.5 g, 7.8 mmol), 4-(bromomethyl)-3-bromopyridine (1.6 g,
6.4 mmol), K2CO3 (1.2 g, 8.7 mmol), and acetone (40 mL) according to
the general procedure I described above. 1H NMR (CDCl3) δ 2.75 (t, J =
5.14 Hz, 4 H), 3.44−3.66 (m, 8 H), 3.72 (s, 2 H), 7.64 (d, J = 4.89 Hz, 1
H), 8.40 (d, J = 5.38 Hz, 1 H), 8.54 (s, 1 H). 13C NMR (CDCl3) δ 54.9,
58.4, 61.5, 69.1, 72.5, 124.9, 126.7, 128.5, 148.2, 151.3. LRMS (ESI) m/z
363 (M+ + H, 78), 385 (M+ + Na, 15). HRMS (ESI) Calcd for
C14H24N2O4Br (M+ + H) 363.0919. Found 363.0930.
1,13-Bis[4′-(4H-chromen-4-on-2-yl)phenyl]-7-(4-(2-
bromopyridyl)methyl)-1,4,10,13-tetraoxa-7-azatridecane (69). This
compound was obtained as a white foam (1.40 g, 37%) from 63 (1.70 g,
4.68 mmol), 4′-hydroxyflavone (2.23 g, 9.37 mmol), triphenylphos-
phine (2.65 g, 10.11 mmol), THF (30 mL), and DIAD (2.02 g, 10.00
mmol) according to the general procedure II described above. 1H NMR
(CDCl3) δ 2.77 (t, J = 5.38 Hz, 4 H), 3.60 (t, J = 5.38 Hz, 4 H), 3.68−
3.77 (m, 6 H), 4.09 (t, J = 4.65 Hz, 4 H), 6.63 (s, 2 H), 6.92 (d, J = 8.80
Hz, 4 H), 7.21 (d, J = 4.89 Hz, 1 H), 7.31 (td, J = 7.46, 1.22 Hz, 2 H), 7.44
(d, J = 8.80 Hz, 2 H), 7.52 (s, 1 H), 7.59 (ddd, J = 8.56, 7.09, 1.47 Hz, 2
H), 7.73−7.78 (m, 4 H), 8.11 (dd, J = 8.07, 1.71 Hz, 2 H), 8.17 (d, J =
4.89 Hz, 1 H). 13C NMR (CDCl3) δ 54.2, 58.1, 67.6, 69.3, 70.1, 106.0,
114.9, 117.9, 122.6, 123.8, 124.0, 125.0, 125.5, 127.5, 127.9, 133.5, 142.4,
149.8, 153.4, 156.0, 161.5, 163.1, 178.1. LRMS (ESI) m/z 803 (M+ + H,
68), 825 (M+ + Na, 91). HRMS (ESI) Calcd for C44H40N2O8Br (M+ +
H) 803.1968. Found 803.1975.
1,13-Bis[4′-(4H-chromen-4-on-2-yl)phenyl]-7-(4-(2-
cyanopyridyl)methyl)-1,4,10,13-tetraoxa-7-azatridecane (70). This
compound was obtained as a white foam (0.64 g, 30%) from 64 (0.89 g,
2.88 mmol), 4′-hydroxyflavone (1.38 g, 5.80 mmol), triphenylphos-
phine (1.59 g, 6.07 mmol), THF (20 mL), and DIAD (1.22 g, 6.04
mmol) according to the general procedure II described above. 1H NMR
(CDCl3) δ 2.82 (t, J = 5.38 Hz, 4 H), 3.65 (t, J = 5.38 Hz, 4 H), 3.79 (t, J
= 5.38 Hz, 4 H), 3.85 (s, 2 H), 4.12−4.16 (m, 4 H), 6.69 (s, 2 H), 6.89−
6.99 (m, 4 H), 7.34−7.39 (m, 2 H), 7.48−7.52 (m, 2 H), 7.62−7.67 (m,
2 H), 7.77−7.84 (m, 4 H), 8.17 (dd, J = 8.07, 1.71 Hz, 2 H), 8.54 (d, J =
4.89 Hz, 1 H). 13C NMR (CDCl3) δ 54.3, 58.1, 67.6, 69.4, 70.2, 106.1,
115.0, 117.6, 118.0, 123.9, 124.2, 125.1, 125.5, 126.5, 128.0, 128.3, 133.6,
133.8, 150.7, 152.1, 156.1, 161.5, 163.2, 178.2. LRMS (ESI) m/z 750
(M+ + H, 100), 772 (M+ + Na, 35). HRMS (ESI) Calcd for C45H40N3O8
(M+ + H) 750.2815. Found 750.2803.
1,13-Bis[4′-(4H-chromen-4-on-2-yl)phenyl]-7-(4-(3-
bromopyridyl)methyl)-1,4,10,13-tetraoxa-7-azatridecane (71). This
compound was obtained as a white foam (0.39 g, 26%) from 65 (0.67 g,
1.85 mmol), 4′-hydroxyflavone (0.88 g, 3.70 mmol), triphenylphos-
phine (1.06 g, 4.05 mmol), THF (20 mL), and DIAD (0.82 g, 4.06
mmol) according to the general procedure II described above. 1H NMR
(CDCl3) δ 2.84 (t, J = 5.38 Hz, 4 H), 3.62 (t, J = 5.62 Hz, 4 H), 3.71−
3.77 (m, 4 H), 3.81 (s, 2 H), 4.04−4.10 (m, 4 H), 6.63 (s, 2 H), 6.91 (d, J
= 9.29 Hz, 4 H), 7.28−7.35 (m, 2 H), 7.44 (d, J = 8.31 Hz, 2 H), 7.56−
7.65 (m, 3 H), 7.73−7.77 (m, 4 H), 8.12 (dd, J = 8.07, 1.71 Hz, 2 H),
8.37 (d, J = 4.89 Hz, 1 H), 8.54 (s, 1 H). 13C NMR (CDCl3) δ 54.7, 58.5,
67.6, 69.3, 70.1, 106.0, 114.9, 117.9, 121.9, 123.8, 124.0, 124.8, 125.0,
125.5, 127.9, 133.5, 148.1, 148.9, 151.3, 156.0, 161.5, 163.1, 178.1.
LRMS (ESI) m/z 803 (M+ + H, 95), 825 (M+ + Na, 34). HRMS (ESI)
Calcd for C44H40N2O8Br (M+ + H) 803.1968. Found 803.1937.
1,13-Bis[4′-(4H-chromen-4-on-2-yl)phenyl]-7-(4-pyrimidinyl-
methyl)-1,4,10,13-tetraoxa-7-azatridecane (72). This compound was
obtained as a white foam (0.29 g, 29%) from 66 (0.39 g, 1.37 mmol), 4′-
hydroxyflavone (0.65 g, 2.73 mmol), triphenylphosphine (0.74 g, 2.82
mmol), THF (20 mL), and DIAD (0.57 g, 2.82 mmol) according to the
general procedure II described above. 1H NMR (CDCl3) δ 2.91 (t, J =
5.38 Hz, 4 H), 3.66−3.87 (m, 8 H), 3.94 (s, 2 H), 4.11−4.25 (m, 4 H),
6.70−6.75 (m, 2 H), 6.97−7.06 (m, 4 H), 7.36−7.43 (m, 2 H), 7.52 (d, J
= 8.31 Hz, 2 H), 7.63−7.71 (m, 3 H), 7.81−7.90 (m, 4 H), 8.20 (dd, J =
7.83, 1.47 Hz, 2 H), 8.59−8.62 (m, 1 H), 9.10 (s, 1 H). 13C NMR
(CDCl3) δ 54.7, 60.7, 67.6, 69.3, 70.1, 106.2, 115.0, 117.9, 120.0, 123.9,
124.2, 125.1, 125.6, 128.0, 133.6, 156.1, 156.9, 158.3, 161.6, 163.2, 169.7,
178.3. LRMS (ESI) m/z 726 (M+ + H, 100), 748 (M+ + Na, 28). HRMS
(ESI) Calcd for C43H40N3O8 (M+ + H) 726.2815. Found 726.2789.
1,13-Bis[4′-(4H-chromen-4-on-2-yl)phenyl]-7-(4-quinolinylmeth-
yl)-1,4,10,13-tetraoxa-7-azatridecane (73). This compound was
obtained as a white foam (0.43 g, 26%) from 67 (0.71 g, 2.13 mmol),
4′-hydroxyflavone (1.03 g, 4.33 mmol), triphenylphosphine (1.20 g,
4.58 mmol), THF (20 mL), and DIAD (0.92 g, 4.55 mmol) according to
the general procedure II described above. 1H NMR (CDCl3) δ 2.88 (t, J
= 5.62 Hz, 4 H), 3.63−3.76 (m, 8 H), 4.03 (t, J = 5.38 Hz, 4 H), 4.20 (s, 2
H), 6.63−6.66 (m, 2 H), 6.84−6.91 (m, 4 H), 7.28−7.34 (m, 2 H),
7.43−7.48 (m, 3 H), 7.54−7.67 (m, 4 H), 7.71−7.76 (m, 4 H), 8.06−
8.09 (m, 1 H), 8.14 (dd, J = 8.07, 1.71 Hz, 2 H), 8.19 (d, J = 7.82 Hz, 1
N-(Pyrimidin-4′-ylmethyl)-3,9-dioxa-6-azaundecane-1,11-diol
(66). This compound was obtained as a pale brown oil (0.44 g, 50%)
from 54 (0.74 g, 3.83 mmol), 4-bromomethylpyrimidine (0.53 g, 3.07
mmol), K2CO3 (0.50 g, 3.62 mmol), and acetone (40 mL) according to
the general procedure I described above. 1H NMR (CDCl3) δ 2.74 (t, J =
5.14 Hz, 4 H) 3.42−3.55 (m, 8 H), 3.56−3.70 (m, 4 H), 3.80 (s, 2 H),
7.56−7.59 (m, 1 H), 8.60 (d, J = 5.38 Hz, 1 H), 9.01 (d, J = 1.47 Hz, 1
H). 13C NMR (CDCl3) δ 54.6, 59.9, 61.4, 68.6, 72.3, 120.4, 157.1, 158.1,
168.5. LRMS (ESI) m/z 286 (M+ + H, 100), 308 (M+ + Na, 81). HRMS
(ESI) Calcd for C13H24N3O4 (M+ + H) 286.1767. Found 286.1764.
N-(Quinolin-4′-ylmethyl)-3,9-dioxa-6-azaundecane-1,11-diol
(67). This compound was obtained as a pale brown oil (1.10 g, 46%)
from 54 (1.50 g, 7.78 mmol), 4-bromomethylquinoline (1.60 g, 7.21
mmol), K2CO3 (1.20 g, 8.70 mmol), and acetone (40 mL) according to
1
the general procedure I described above. H NMR (CDCl3) δ 2.82−
2.87 (m, 4 H), 3.46−3.63 (m, 4 H), 3.63−3.88 (m, 4 H), 4.13 (s, 2 H),
7.55−7.60 (m, 2 H), 7.69−7.74 (m, 1 H), 8.12 (dd, J = 8.31, 1.47 Hz, 1
H), 8.26−8.30 (m, 1 H), 8.85−8.88 (m, 1 H). 13C NMR (CDCl3) δ
54.8, 56.5, 61.4, 69.1, 72.4, 121.2, 123.9, 126.4, 127.4, 129.2, 129.6,
145.3, 147.9, 150.0. LRMS (ESI) m/z 335 (M+ + H, 100), 357 (M+ + Na,
45). HRMS (ESI) Calcd for C18H27N2O4 (M+ + H) 335.1971. Found
335.1965.
1,13-Bis[4′-(4H-chromen-4-on-2-yl)phenyl]-7-(3-pyridylmethyl)-
1,4,10,13-tetraoxa-7-azatridecane (68). This compound was obtained
as a white foam (1.20 g, 36%) from 62 (1.30 g, 4.58 mmol), 4′-
hydroxyflavone (2.20 g, 9.24 mmol), triphenylphosphine (2.50 g, 9.54
mmol), THF (30 mL), and DIAD (2.00 g, 9.90 mmol) according to the
general procedure II described above. 1H NMR (CDCl3) δ 2.78−2.83
(m, 4 H), 3.62−3.66 (m, 4 H), 3.75−3.79 (m, 6 H), 4.11 (t, J = 4.65 Hz,
4 H), 6.65−6.67 (m, 2 H), 6.94−6.98 (m, 4 H), 7.19 (dd, J = 7.83, 4.89
Hz, 1 H), 7.32−7.36 (m, 2 H), 7.45−7.49 (m, 2 H), 7.59−7.64 (m, 2 H),
7.68 (d, J = 7.82 Hz, 1 H), 7.76−7.80 (m, 4 H), 8.13−8.17 (m, 2 H),
8.45−8.47 (m, 1 H), 8.57 (s, 1 H). 13C NMR (CDCl3) δ 53.8, 57.1, 67.6,
69.3, 70.2, 106.1, 115.0, 117.9, 123.2, 123.9, 124.1, 125.0, 125.5, 127.9,
133.5, 135.1, 136.4, 148.4, 150.2, 156.1, 161.6, 163.2, 178.2. LRMS
(ESI) m/z 725 (M+ + H, 100), 747 (M+ + Na, 29). HRMS (ESI) Calcd
for C44H41N2O8 (M+ + H) 725.2863. Found 725.2883.
8900
dx.doi.org/10.1021/jm301172v | J. Med. Chem. 2012, 55, 8891−8902

Journal of Medicinal Chemistry
Article
H), 8.80 (d, J = 4.40 Hz, 1 H). 13C NMR (CDCl3) δ 54.6, 56.8, 67.6,
69.2, 70.2, 106.0, 114.9, 117.9, 121.0, 123.8, 124.0, 125.0, 125.5, 126.1,
127.4, 127.8, 129.0, 129.9, 133.5, 145.8, 148.2, 150.2, 156.0. LRMS
(ESI) m/z 775 (M+ + H, 100), 797 (M+ + Na, 30). HRMS (ESI) Calcd
for C48H43N2O8 (M+ + H) 775.3019. Found 775.3005.
Cell Lines and Cell Culture. Promastigotes of Leishmania donovani
(LdAG83, Ld39, and LdAG83PentR50) were employed in this study.
The promastigotes were cultured in Schneider’s Drosophila Medium
(Invitrogen), pH 6.9 supplemented with 10% (v/v) heat inactivated
fetal bovine serum (Hyclone) with 4 mM glutamine (Sigma) and 25 μg/
mL gentamicin solution (Invitrogen) at 27 °C for 4 days.28
Promastigotes of LdAG83PentR50 (pentamidine-resistant, IC50 of
pentamidine = 74.7 μM) were selected by gradually increasing the
pentamidine (Sigma) pressure to the wild-type promastigotes of L.
donovani LdAG83.9 Ld39 was SSG-resistant strain and its IC50 of SSG
was 7090 μM.7
In Vitro Antipromastigote Activity. Antipromastigotes activity
was determined by Cell Titer 96 Aqueous Assay (Promega) that
employed a tetrazolium compound (MTS, 2-(4,5-dimethylthiazol-2-yl-
)-5-[3-(carboxymethoxy)phenyl]-2-(4-sulfophenyl)-2H-tetrazolium)
and coupling reagent, phenazine methosulfate (PMS) (Sigma).7
Promastigotes were seeded into 96-well flat bottom microtiter plate at
1 × 105 cells per well in a final volume of 100 μL medium and incubated
with a series concentration of synthetic flavonoid dimers or known
antileishmanials. Parasites were incubated at 27 °C for 72 h. Fresh
solution of 2 mg/mL MTS and 0.92 mg/mL PMS were prepared in a
ratio of 20:1 (MTS:PMS). After 72 h of incubation, 10 μL of MTS:PMS
mixture was added into each well of microtiter plate. The plate was then
incubated at 27 °C for 4 h for color development. After 4 h of incubation,
the OD values were determined at 490 nm using automatic microtiter
plate reader (Bio-Rad). The results were presented as percentage of
survivors (OD value with test compound divided by that of untreated
control). IC50 was defined as the concentration of antileishmanial
compounds needed to reduce OD value by half.
In Vitro Antiamastigotes Activity. Mouse peritoneal elicited
macrophages (PEM) were obtained as previously described.9 A round
coverslip (12 mm in diameter) was placed into each well of 24-well
culture plate. Mouse PEM were resuspended in supplemented DMEM
media containing 10% heat inactivated fetal bovine serum (v/v), 100 U/
mL penicillin and 100 μg/mL streptomycin and seeded into each well at
a cell density of 1 × 105 cells per 500 μL. Macrophages were allowed to
attach overnight. Nonadherent cells were removed by gentle washing
with unsupplemented DMEM media twice. Adherent macrophages
were infected with late-log promastigotes at a parasite-to-macrophage
ratio of 20:1 overnight at 37 °C with 5% CO2. Noninternalized
promastigotes were removed by washing twice with unsupplemented
DMEM media. Infected macrophages were further incubated in 500 μL
of supplemented DMEM media in the presence or absence of flavonoid
dimers or known antileishmanials for 72 h at 37 °C. After incubation,
coverslips were stained with Giemsa, and the percentage of macrophages
infected and number of amastigotes per 100 macrophages were
enumerated.
ASSOCIATED CONTENT
* Supporting Information
Proton and carbon NMR spectra of compound 39, 52, 53, 56−
61, 62−73 and HPLC chromatogram of compound 39. This
material is available free of charge via the Internet at http://pubs.
acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
*E-mail: bcchanth@polyu.edu.hk (T.H.C.), bclchow@polyu.
edu.hk (L.M.C.C.). Phone: (852)-34008670 (T.H.C.), (852)-
34008662 (L.M.C.C.). Fax: (852)-23649932 (T.H.C.), (852)-
23649932 (L.M.C.C.).
■
Author Contributions
§These authors contributed equally. The manuscript was written
through contributions of all authors. All authors have given
approval to the final version of the manuscript.
Notes
The authors declare the following competing financial
interest(s): While there is no patent-licensing arrangement in
place, we are in active discussion with interested parties on
potential option/licensing arrangement of the intellectual
property covering the results of this manuscript.
ACKNOWLEDGMENTS
■
The work described in this paper was supported by the Hong
Kong Research Grant Council General Research Fund (B-
Q21B) and Hong Kong Polytechnic University internal grant
(GU383). I.L.K.W. was supported by a Postdoctoral Fellowship
from the Hong Kong Polytechnic University. The authors
declare that while there is no patent-licensing arrangement in
place, we are in active discussion with interested parties on
potential option/licensing arrangement of the intellectual
property covering the results of this manuscript.
ABBREVIATIONS USED
■
VL, visceral leishmaniasis; PEG, polyethylene glycol; EG,
ethylene glycol; PEM, peritoneal elicited macrophage; SSG,
sodium stibogluconate
REFERENCES
■
(1) Ashford, R. W.; Desjeux, P.; Deraadt, P. Estimation of population at
risk of infection and number of cases of Leishmaniasis. Parasitol. Today
1992, 8, 104−105.
(2) Kayser, O.; Kiderlen, A. F.; Croft, S. L. Natural products as
antiparasitic drugs. Parasitol. Res 2003, 90 (Suppl 2), S55−62.
(3) Tasdemir, D.; Kaiser, M.; Brun, R.; Yardley, V.; Schmidt, T. J.;
Tosun, F.; Ruedi, P. Antitrypanosomal and antileishmanial activities of
flavonoids and their analogues: in vitro, in vivo, structure-activity
relationship, and quantitative structure-activity relationship studies.
Antimicrob. Agents Chemother. 2006, 50, 1352−1364.
In Vitro Cytotoxicity Assay of Macrophages. Mouse macro-
phage cell line, RAW 264.7, or primary mouse PEM cells were grown in
supplemented DMEM media in an atmosphere of 95% air with 5% CO2
at 37 °C. RAW 264.7 or PEM cells were seeded into 96-well flat bottom
microtiter plate at 1 × 104 cells and 2 × 104 cells per well in a final volume
of 100 μL media, respectively. A graded dose of flavonoid dimers or
antileishmanials was added into each well. The plate was incubated in an
atmosphere of 95% air with 5% CO2 at 37 °C for 72 h. Percentage of
survivors was determined using the MTS assay as described previously
and used to calculate IC50 values. Therapeutic index was calculated by
dividing (IC50 value for cytotoxicity of the compounds toward
macrophages) over (IC50 value for antileishmanial activity of the
compounds).
(4) Harborne, J. B.; Williams, C. A. Advances in flavonoid research
since 1992. Phytochemistry 2000, 55, 481−504.
(5) Chan, K. F.; Zhao, Y.; Burkett, B. A.; Wong, I. L.; Chow, L. M.;
Chan, T. H. Flavonoid dimers as bivalent modulators for P-
glycoprotein-based multidrug resistance: synthetic apigenin homo-
dimers linked with defined-length poly(ethylene glycol) spacers
increase drug retention and enhance chemosensitivity in resistant
cancer cells. J. Med. Chem. 2006, 49, 6742−6759.
(6) Chan, K. F.; Zhao, Y.; Chow, T. W.; Yan, C. S.; Ma, D. L.; Burkett,
B. A.; Wong, I. L.; Chow, L. M.; Chan, T. H. Flavonoid dimers as
bivalent modulators for p-glycoprotein-based multidrug resistance:
structure-activity relationships. ChemMedChem 2009, 4, 594−614.
8901
dx.doi.org/10.1021/jm301172v | J. Med. Chem. 2012, 55, 8891−8902

Journal of Medicinal Chemistry
Article
(7) Wong, I. L.; Chan, K. F.; Burkett, B. A.; Zhao, Y.; Chai, Y.; Sun, H.;
Chan, T. H.; Chow, L. M. Flavonoid dimers as bivalent modulators for
pentamidine and sodium stiboglucanate resistance in leishmania.
Antimicrob. Agents Chemother. 2007, 51, 930−940.
(8) Wong, I. L.; Chan, K. F.; Tsang, K. H.; Lam, C. Y.; Zhao, Y.; Chan,
T. H.; Chow, L. M. Modulation of multidrug resistance protein 1
(MRP1/ABCC1)-mediated multidrug resistance by bivalent apigenin
homodimers and their derivatives. J. Med. Chem. 2009, 52, 5311−5322.
(9) Wong, I. L.; Chan, K. F.; Zhao, Y.; Chan, T. H.; Chow, L. M.
Quinacrine and a novel apigenin dimer can synergistically increase the
pentamidine susceptibility of the protozoan parasite Leishmania. J.
Antimicrob. Chemother. 2009, 63, 1179−1190.
(10) Chan, K. F.; Wong, I. L.; Kan, J. W.; Yan, C. S.; Chow, L. M.;
Chan, T. H. Amine linked flavonoid dimers as modulators for p-
glycoprotein-based multidrug resistance: structure-activity relationship
and mechanism of modulation. J. Med. Chem. 2012, 55, 1999−2014.
(11) Weniger, B.; Vonthron-Senecheau, C.; Kaiser, M.; Brun, R.;
Anton, R. Comparative antiplasmodial, leishmanicidal and antitrypano-
somal activities of several biflavonoids. Phytomedicine 2006, 13, 176−
180.
flavonoids, genistein and orobol. Biochem. Pharmacol. 1990, 39, 737−
744.
(26) Oliveira, M. C.; de Carvalho, M. G.; Grynberg, N. F.; Brioso, P. S.
A biflavonoid from Luxemburgia nobilis as inhibitor of DNA
topoisomerases. Planta Med. 2005, 71, 561−563.
(27) Pang, X.; Yi, T.; Yi, Z.; Cho, S. G.; Qu, W.; Pinkaew, D.; Fujise, K.;
Liu, M. Morelloflavone, a biflavonoid, inhibits tumor angiogenesis by
targeting rho GTPases and extracellular signal-regulated kinase signaling
pathways. Cancer Res. 2009, 69, 518−525.
(28) Chow, L. M.; Wong, A. K.; Ullman, B.; Wirth, D. F. Cloning and
functional analysis of an extrachromosomally amplified multidrug
resistance-like gene in Leishmania enriettii. Mol. Biochem. Parasitol.
1993, 60, 195−208.
(12) del Rayo Camacho, M.; Phillipson, J. D.; Croft, S. L.; Marley, D.;
Kirby, G. C.; Warhurst, D. C. Assessment of the antiprotozoal activity of
Galphimia glauca and the isolation of new nor-secofriedelanes and nor-
friedelanes. J. Nat. Prod. 2002, 65, 1457−1461.
(13) Grael, C. F.; Albuquerque, S.; Lopes, J. L. Chemical constituents
of Lychnophora pohlii and trypanocidal activity of crude plant extracts
and of isolated compounds. Fitoterapia 2005, 76, 73−82.
(14) Jordao, C. O.; Vichnewski, W.; de Souza, G. E.; Albuquerque, S.;
Lopes, J. L. Trypanocidal activity of chemical constituents from
Lychnophora salicifolia Mart. Phytother. Res. 2004, 18, 332−334.
(15) Ribeiro, A.; Pilo-Veloso, D.; Romanha, A. J.; Zani, C. L.
Trypanocidal flavonoids from Trixis vauthieri. J. Nat. Prod. 1997, 60,
836−841.
(16) Muzitano, M. F.; Cruz, E. A.; de Almeida, A. P.; Da Silva, S. A.;
Kaiser, C. R.; Guette, C.; Rossi-Bergmann, B.; Costa, S. S. Quercitrin: an
antileishmanial flavonoid glycoside from Kalanchoe pinnata. Planta
Med. 2006, 72, 81−83.
(17) Kolodziej, H.; Kayser, O.; Kiderlen, A. F.; Ito, H.; Hatano, T.;
Yoshida, T.; Foo, L. Y. Proanthocyanidins and related compounds:
antileishmanial activity and modulatory effects on nitric oxide and tumor
necrosis factor-alpha-release in the murine macrophage-like cell line
RAW 264.7. Biol. Pharm. Bull. 2001, 24, 1016−1021.
(18) Kolodziej, H.; Kiderlen, A. F. Antileishmanial activity and immune
modulatory effects of tannins and related compounds on Leishmania
parasitised RAW 264.7 cells. Phytochemistry 2005, 66, 2056−2071.
(19) Ahmed, M. S.; Galal, A. M.; Ross, S. A.; Ferreira, D.; ElSohly, M.
A.; Ibrahim, A. S.; Mossa, J. S.; El-Feraly, F. S. A weakly antimalarial
biflavanone from Rhus retinorrhoea. Phytochemistry 2001, 58, 599−602.
(20) Lin, Y. M.; Flavin, M. T.; Cassidy, C. S.; Mar, A.; Chen, F. C.
Biflavonoids as novel antituberculosis agents. Bioorg. Med. Chem. Lett.
2001, 11, 2101−2104.
(21) Nunome, S.; Ishiyama, A.; Kobayashi, M.; Otoguro, K.; Kiyohara,
H.; Yamada, H.; Omura, S. In vitro antimalarial activity of biflavonoids
from Wikstroemia indica. Planta Med. 2004, 70, 76−78.
(22) Suarez, A. I.; Diaz, M. B.; Delle Monache, F.; Compagnone, R. S.
Biflavonoids from Podocalyx loranthoides. Fitoterapia 2003, 74, 473−
475.
(23) Boege, F.; Straub, T.; Kehr, A.; Boesenberg, C.; Christiansen, K.;
Andersen, A.; Jakob, F.; Kohrle, J. Selected novel flavones inhibit the
DNA binding or the DNA religation step of eukaryotic topoisomerase I.
J. Biol. Chem. 1996, 271, 2262−2270.
(24) Mittra, B.; Saha, A.; Chowdhury, A. R.; Pal, C.; Mandal, S.;
Mukhopadhyay, S.; Bandyopadhyay, S.; Majumder, H. K. Luteolin, an
abundant dietary component is a potent anti-leishmanial agent that acts
by inducing topoisomerase II-mediated kinetoplast DNA cleavage
leading to apoptosis. Mol. Med. 2000, 6, 527−541.
(25) Yamashita, Y.; Kawada, S.; Nakano, H. Induction of mammalian
topoisomerase II dependent DNA cleavage by nonintercalative
8902
dx.doi.org/10.1021/jm301172v | J. Med. Chem. 2012, 55, 8891−8902