Structure-activity relationships in glycosylated 2-phenyl-indoles, 2-phenyl-benzo[b]thiophenes and 2-phenyl-benzo[b]furans as DNA binding and potential antitumor agents

Article abstract of DOI:10.1016/j.bmc.2011.01.014

In earlier investigations we have described the synthesis and biological evaluation of a panel of novel glycosylated heteroaromatics (1-12). It was found that these compounds can bind to DNA in vitro and are cytotoxic against several cancer cell lines at

Full text of DOI:10.1016/j.bmc.2011.01.014

Bioorganic & Medicinal Chemistry 19 (2011) 1779–1789
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Structure–activity relationships in glycosylated 2-phenyl-indoles,
2-phenyl-benzo[b]thiophenes and 2-phenyl-benzo[b]furans as DNA
binding and potential antitumor agents
⇑
Wei Shi, Todd L. Lowary
Alberta Ingenuity Centre for Carbohydrate Science and Department of Chemistry, The University of Alberta, Gunning-Lemieux Chemistry Centre, Edmonton, Canada AB T6G 2G2
a r t i c l e i n f o
a b s t r a c t
Article history:
In earlier investigations we have described the synthesis and biological evaluation of a panel of novel
glycosylated heteroaromatics (1–12). It was found that these compounds can bind to DNA in vitro and
are cytotoxic against several cancer cell lines at low micromolar concentration. We report here struc-
ture–activity studies of these molecules with respect to DNA binding and cytotoxicity. In particular the
structure of the linker moiety between the carbohydrate and the intercalator, the stereochemistry at
the anomeric position, and the substituents and stereochemistry at C-4 in one of the carbohydrate resi-
Received 28 November 2010
Revised 5 January 2011
Accepted 8 January 2011
Available online 14 January 2011
Keywords:
Anticancer
dues (4-amino-2,3,4,6-tetradeoxy-a-L-threo-hexopyranose) are investigated. All these structural features
were identified to have a clear influence on DNA binding; however, only the substituents at C-4 in the
carbohydrate residue exhibited an obvious impact on cytotoxicity. It was found that the amino group
at C-4 was favored over all other substituents with regard to both DNA binding and cytotoxicity. The
information gathered from these structure–activity investigations suggested that future work on the
preparation of additional analogues should focus on molecules containing an amino sugar moiety.
Ó 2011 Elsevier Ltd. All rights reserved.
Structure–activity relationship
DNA binding
Cytotoxicity
1. Introduction
binding, antibacterial activity, and cytotoxicity. Second, compared
to daunosamine (1–4) and acosamine systems (5–8), the 4-N-TDTH
DNA binding is one of the common modes of action for many
clinically-used anticancer agents, such as dactinomycin, mitomy-
cin, daunorubicin, and trabectedin.1 Using a molecular hybridiza-
tion approach,2 we recently designed and synthesized a small
panel of novel glycoconjugates (1–12, Chart 1).3,4 These molecules
comprise three functional modules: a flat aromatic moiety that can
potentially intercalate between DNA base pairs, a minor-groove
binding carbohydrate moiety,5 and a linker, which we selected to
be a propargylic group. By using direct and indirect fluorescence
measurements, it was confirmed that 1–12 bind to DNA in vitro
with moderate affinity.3 It was also demonstrated that some of
these analogues showed low micromolar antibacterial4 and anti-
cancer6 activity in vitro against one strain of gram-positive bacteria
and three solid-tumor cell lines.
In this paper, we investigated the effect of structural features in
the carbohydrate and the propargyl linker in these compounds on
cytotoxicity and DNA binding. Among the three configurations of
the carbohydrate domain, we decided to concentrate our struc-
ture–activity relationship (SAR) analysis on the 4-amino-2,3,4,6-
tetradeoxy-a-L-threo-hexopyranoside (4-N-TDTH) system (9–12)
for two reasons. First, and most importantly, this series of
analogues exhibited representative properties in terms of DNA
system has one stereogenic center fewer, which simplifies SAR
analysis. It should be noted that because of the dramatically lower
cytotoxicity of free indole analogues (4, 8, and 12) compared to
molecules with the other three aromatic domains,6 the SAR analy-
sis did not include analogues possessing a free indole core.
2. Results and discussion
2.1. Effect of stereochemistry at the anomeric position in the
carbohydrate residue
2.1.1. DNA binding
It has been found that the configuration at the anomeric carbon
can have a significant influence on the biological activity of carbo-
hydrate-containing compounds, such as the anthracycline antibi-
otics.7,8 Whereas the
a-configured anthracyclines usually have
potent anticancer activity and strong DNA binding affinity, their
b-configured counterparts exhibit both lower cytotoxicity and
much weaker DNA binding.7,8 With enough quantities of the
b-configured analogues 18–25 in hand (Chart 2),4 we first investi-
gated the effect of the anomeric configuration on biological
activity.
Using a previously described ethidium bromide (EtBr) fluores-
cent intercalator displacement (FID) assay,3 the DNA binding abil-
ity of analogues 13–25 were determined at 50 lM. As presented in
⇑
Corresponding author.
E-mail address: tlowary@ualberta.ca (T.L. Lowary).
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.01.014

1780
W. Shi, T. L. Lowary / Bioorg. Med. Chem. 19 (2011) 1779–1789
O
O
O
O
O
O
X
X
X
HO
NH2
HO
NH2
NH2
1 X = O
2 X = S
5 X = O
6 X = S
9
X = O
10 X = S
3 X = NTs
4 X = NH
7 X = NTs
8 X = NH
11 X = NTs
12 X = NH
Chart 1. Structures of anthracycline analogues 1–12.
O
O
O
X
X
O
AcO
HO
15 X = O
16 X = S
17 X = NTs
13 X = O
14 X = S
O
O
O
O
O
HO
AcO
O
X
X
X
NH2
20 X = O
21 X = S
22 X = NTs
23 X = O
24 X = S
25 X = NTs
18 X = O
19 X = S
Chart 2. Structures of 13–25.
Table 1, those compounds possessing an amino group (23–25),
were shown to displace EtBr from DNA, thus indicating their ability
to bind to DNA. However, no DNA binding could be detected by this
assay for those compounds in which the amino group is replaced
with a hydroxyl functionality (15–17 and 20–22). The O-acetylated
analogues, 13, 14, 18, and 19, were not evaluated due to their poor
solubility in the aqueous buffer required for the assay.
amount of EtBr was displaced) than their
a
-configured equivalents
9–11. Compared to the N-tosyl protected indole core (11 vs. 25),
the enhancement is more pronounced for the benzo[b]furan (9
vs. 23) and benzo[b]thiophene systems (10 vs. 24). Subsequently,
an FID titration experiment was performed to obtain the relative
binding constant for 23–25 using the binding constant of EtBr as
a reference (Fig. 1). Due to relatively weak DNA binding affinity,
From these data, it is clear that the b-configured amino ana-
logues 23–25 exhibited stronger DNA binding affinity (a larger
a
linear relationship between fluorescence intensity and
Table 1
The effect of
a
and b configuration on DNA binding
affinitya
Compd
% Remaining EtBr
9 (
a
)
68.5 1.9
79.9 1.0
38.6 1.9
96.7 3.2
102.6 4.7b
111.2 3.6b
103.2 3.4b
99.9 1.8
109.2 5.1b
26.9 1.1
26.6 1.4
19.8 1.0
10 (
11 (
15 (
16 (
17 (
20 (b)
21 (b)
22 (b)
23 (b)
24 (b)
25 (b)
a
a
a
a
a
)
)
)
)
)
a
All the compounds were tested at 50
The percentage is slightly over 100% presumably
l
M.
b
due to the weak fluorescence enhancement effect
between the assayed compounds and free EtBr or the
EtBr–DNA complex.
Figure 1. FID titration curves of b-configured analogues 23–25.

W. Shi, T. L. Lowary / Bioorg. Med. Chem. 19 (2011) 1779–1789
1781
concentration was not seen when the concentration of titrants was
larger than 60 M. Cognizant of the fluorescence interference for
compounds 23 and 25 at 20 M, which is indicated by a fluores-
cence intensity of over 100%, only the percentages of remaining
EtBr between 40 and 60 M were used to estimate the concentra-
tion at which there is a 50% fluorescence loss of EtBr. This value is
approximately 55 M for 23, 45 M for 24, and 55 M for 25. Using
the equation KLigand = KEtBr[EtBr]/[Ligand]50% the DNA binding
affinity of the b-configured analogues is 8–10-fold less than EtBr,
2.2. Effect of substituents and stereochemistry at C-4 in the
carbohydrate domain
l
l
In previous studies with 1–12 it was found that different sub-
stituents at C-4 in the carbohydrate domain, such as O-acetyl, hy-
droxyl, azido, and amino groups, had some effect on both DNA
binding and cytotoxicity.6 These trends are also observed from
the data presented above. For example, in Table 1, the replacement
of an axial amino group (11 and 23–25) with an equatorial
hydroxyl group (17 and 20–22) led to the total loss of DNA binding
affinity. Furthermore, in Table 2, the compounds with an equatorial
O-acetyl (14 and 19) or an equatorial hydroxyl group (16 and 21) at
C-4 exhibited a pronounced decrease in cytotoxicity in comparison
with the amino analogues (10 and 24). However, to make SAR
interpretation more conclusive, it was necessary to synthesize
additional analogues with these groups installed in the opposite
orientation at C-4 (27 and 29, Scheme 1). No significant differences
in biological activity were observed for the different aromatic cores
when the attached sugar was 4-N-TDTH.6 Therefore, we chose only
the benzo[b]thiophene system for these investigations and, given
l
l
l
l
9,10
,
but is at least 2–4-fold stronger than the corresponding
ured isomers.
a-config-
2.1.2. Cytotoxicity
Next, compounds 13–25 were screened against three cancer cell
lines. Because the observed trends are the same for all the three
aromatic systems, only the data for the benzo[b]thiophene ana-
logues are listed in Table 2 for the sake of brevity; data for the
other ring systems can be found in Table S2. The effect of the ano-
meric configuration on the cytotoxicity is subtle. Compared to the
a-configured 4-amino analogues, the strong enhancement of DNA
binding affinity but rather constant cytotoxicity of the b-glycoside
analogues further confirmed our previous conclusion using 1–12
that DNA binding alone is not the major mode of action of this class
the results discussed above comparing the a- and b-stereochemis-
try, we propose that similar trends will be observed for the other
aromatic rings.
of molecules.6 In addition, compared to the compounds with an
a
configuration, the cancer cell selectivity of b-configured analogues
did not show any significant improvement (Table S1 in
Supplementary data).
2.2.1. Chemistry
The preparation of 27 and 29 is shown in Scheme 1. Under Mits-
unobu conditions with benzoic acid, alcohol 16 was converted in
Table 2
Cytotoxicity of
a
- and b-configured analogues of the benzo[b]thiophene system against three solid-tumor cell linesa
Compd
Cell viability (%) at 25
HT29
lM
IC50
(lM)
MCF7
HepG2b
MCF7
HT29
HepG2
10 (
14 (
16 (
19 (b, 4-OAc)
21 (b, 4-OH)
24 (b, 4-NH2)
a
a
a
, 4-NH2)
, 4-OAc)
, 4-OH)
<1
<1
<1
6 (0.7)
<1
2 (0.7)
<1
<1
<1
25 (0.3)
8 (2.1)
7 (0.9)
<1
7.1 1.2
5.9 0.5
7.2 1.0
11.1 1.1
10.0 0.7
12.8 0.8
4.6 0.5
8.8 0.9
12.3 1.4
16.4 0.4
15.7 1.6
13.6 1.0
4.9 0.5
27 (0.9)
51 (3.2)
40 (3.5)
23 (1.2)
<1
—
—
—
c
—
7.7 0.4
a
b
c
The numbers in parentheses represent the standard deviation.
HepG2/C3A is simplified as HepG2.
‘—’ = not determined.
O
O
b
a
O
S
S
S
O
O
O
HO
OBz
OH
26
27
16
c
d
O
O
S
S
O
O
N3
H2N
28
29
Scheme 1. Synthesis of 4-N-TDTH compounds 29 with an equatorial amino group. Reagents and conditions: (a) PPh3, benzoic acid, DIAD, ꢀ20 °C?rt, 79%; (b) NaOCH3
methanol, 99%; (c) PPh3, diphenylphosphoryl azide, DIAD, ꢀ20 °C?rt, 56%; (d) PPh3, H2O, THF, 70 °C, 75%.

1782
W. Shi, T. L. Lowary / Bioorg. Med. Chem. 19 (2011) 1779–1789
Table 3
benzoate ester 26 in 79% yield. The removal of the benzoyl group in
26 was achieved in nearly quantitative yield and a second Mitsun-
obu reaction with diphenylphoshoryl azide was used to install the
azido group in 28. The yield for this conversion (56%) was rather
low due to the generation of a significant amount of the two pos-
sible elimination products as was observed earlier for a similar
reaction.4 Finally, Staudinger reduction led to the production of
29 with the desired equatorial amino group in 75% yield.
The effects of the stereochemistry at C-4 on cytotoxicity against three solid-tumor cell
linesa
O
O
S
O
S
O
R
R
2.2.2. DNA binding
With enough quantities of 29 in hand, we first assayed its DNA
binding affinity using the FID assay. Originally, we speculated that
compared to the anomeric configuration (10 vs. 24, Table 1), the
stereochemistry at C-4 would have a less pronounced effect on
DNA binding because the relative spatial relationship between
the carbohydrate and aromatic domains would be less influenced
by such an apparently small substituent change. However, to our
surprise, at a concentration of 100 lM, the percentage of remaining
EtBr for 29 is comparable to that for b-configured analogue 24
(Fig. 2). This indicates much stronger binding of 29 to DNA than
14 R = OAc
16 R = OH
26 R = OBz
27 R = OH
29
30
10
31
R = NH2
R = NH2
R = HCOO
R = HCOO
Compd
Cell viability (%) at 25
l
M
IC50 (lM)
MCF7
HT29
HepG2b MCF7
HT29
HepG2
10
14
16
26
27
29
30
31
<1
<1
<1
6 (1.1)
25 (0.3)
7.1 1.2
5.9 0.5
7.2 0.9
11.1 1.1
—
8.8 0.9
12.3 1.4
16.4 0.4
—
15.6 1.7
4.7 0.8
c
27 (3.5)
51 (3.2)
<1
6 (0.7)
—
—
—
—
100 (7.3) 81 (2.1) 99 (1.2)
56 (1.9)
<1
10 (1.1) 10 (1.2)
<1
13.4 1.1
5.3 0.7
the
the titration curves, the concentration at which there is a 50%
fluorescence loss of EtBr is approximately 65 M for 29 and
45 M for 24, which indicates that compared to EtBr, the binding
constant of 24 is around 1.5-fold larger than 29. When 10 was
tested in the FID assay, minimal EtBr displacement at 100
a-configured analogue 10 (ꢁ80% remaining EtBr). Based on
<1
<1
17 (1.1)
6.6 1.0
37 (0.7)
59 (2.1)
<
—
—
7.7 0.6 17.4d
12.6 1.3 19.8d
2 (0.3)
l
l
a
b
c
The numbers in parentheses represent the standard deviation.
HepG2/C3A is simplified as HepG2.
‘—’ = not determined.
lM
d
Only assayed once.
was observed, and consequently no titration was carried out.
Therefore, the order of DNA binding affinity is 10 ꢂ 29 < 24 and
based on these results we propose that the relative position of
the C-1 and C-4 substituents on the carbohydrate ring appears to
be more important than their individual orientations. In particular,
stronger DNA binding affinity is achieved when these substituents
are cis on the ring.
which this group was axial. For the same substituent, the equato-
rial orientation seems slightly favored over the axial for most of
the cell lines. The significant loss of the cytotoxicity for 26 could
be due to the bulkiness of the benzoyl group, but this remains to
be investigated. Regardless of the orientation at C-4, the introduc-
tion of an amino group dramatically improved the in vitro antican-
cer activity, especially for MCF-7 cells (compounds 29 and 10).
2.2.3. Cytotoxicity
The stereochemistry and identity of different substituents at
C-4 also showed some effect on cytotoxicity (Table 3). When the
substituent at C-4 was oriented equatorially, esterification of the
hydroxyl group improved the cytotoxicity (14 and 30 vs. 16).6 On
the other hand, esterification of the free hydroxyl group appeared
to have little effect on the cytotoxicity of 31 compared to 27,6 in
2.3. Effect of the intercalator–carbohydrate linker
2.3.1. Analogues without carbohydrate moieties
After the investigation of the 4-N-TDTH carbohydrate moiety,
we turned our attention to another structural component, the
Figure 2. FID titration results of 10, 24, and 29.

W. Shi, T. L. Lowary / Bioorg. Med. Chem. 19 (2011) 1779–1789
1783
linker moiety. The easiest modification of the propargyl linker is to
reduce the triple bond to a single bond by hydrogenation. Before
applying this modification to the target compounds, a simpler sys-
tem was used to explore both the chemistry required and the effect
of alkyne reduction on cytotoxicity. In our previous studies, com-
pounds 32 and 33 (Scheme 2), which contain only the aromatic do-
main and the propargyl linker, showed potent cytotoxicity (cell
afforded 40 and 41 in 93% and 96% yield, respectively. Mitsunobu
reaction with diphenyl phosphoryl azide was then used to convert
the hydroxyl group in 40 and 41 to an azido group in 42 and 43.
The yields of this transformation (96% for 42 and 95% for 43) was
significantly (ꢁ40%) higher than the previous corresponding reac-
tion (56% for 28 in Scheme 1) presumably because the leaving
group is equatorial and therefore less prone to elimination. Indeed,
smaller amounts of elimination products were detected by TLC.
The final conversion of the azido group to the amino group was
achieved by the Staudinger reaction to give 36 and 37 in 88% and
85% yield, respectively.
viability <10%) at 100
HepG2/C3A liver cancer cells.6 However, potent cytotoxicity was
lost when the concentration was decreased to 25 M (Table 4).
lM, especially against HT29 colon and
l
Therefore, we first chose to determine if a change in cytotoxicity
would be observed upon reduction of the alkyne in 32 and 33 to
an alkane (34 and 35).
Using standard palladium-catalyzed hydrogenation conditions,
the conversion of 32–34, and 33–35 proceeded smoothly in yields
of 82% and 89%, respectively. When 34 was screened against three
cancer cell lines, no significant variation in cytotoxicity was ob-
served compared to the parent alkyne (Table 4). However, a very
pronounced improvement was achieved with 35, specifically
2.3.2.2. DNA binding. The relative DNA binding affinity of 36 and
37 was assayed by the EtBr FID assay (Table 5). From the assay re-
sults, there was a moderate increase of the DNA binding affinity for
both 36 and 37 measured as a decrease (ꢁ25%) in the percentage of
remaining EtBr. Although the effect is small, we postulate that the
greater flexibility of the alkane chain helps the propyl analogues
adopt conformations that allow stronger interaction with DNA.
against HT29 colon cancer cells (IC50 ꢁ16
lM). We next therefore
investigated the effect of replacing the proparyl linker with a pro-
2.3.2.3. Cytotoxicity. Next, all intermediates and final products
with the propargyl or propyl linker were assayed for their
in vitro anticancer activity. Because similar trends were obtained
for the both benzo[b]furan and benzo[b]thiophene systems, only
the cytotoxicity for the benzo[b]thiophene system is summarized
in Table 6; data for the benzo[b]furan system can be found in
Tables S3. In contrast to the previous observation for 33 versus
35 (Table 4) that the propyl linker benefited cytotoxicity, it was
found that, to some extent, compounds containing a propargyl
linker exhibited better cytotoxicity than their equivalents with
the propyl linker. This trend is most pronounced for analogues
lacking an amino group (14 vs. 39, 16 vs. 41) against HepG2/C3A
liver cancer cells, but becomes ambiguous for the amino-contain-
ing analogues. The overall conclusion is that modification of the
propargyl group to the propyl group does not change the cytotox-
icity in a predictable manner. Moreover, the contrast between the
increase of DNA binding affinity and the fluctuation of the cytotox-
icity change from alkyne 10 to alkane 37 further suggests that DNA
binding is not the major biological target of this class of
compounds.
pyl group on compounds containing a carbohydrate residue.
2.3.2. Analogues with carbohydrate moieties
2.3.2.1. Chemistry. Unfortunately, under the hydrogenation con-
ditions used for the synthesis of 34 and 35, the direct conversion
of 4-N-TDTH analogues 9 and 10 to their propyl-linked counter-
parts 36 and 37 (Scheme 3) was not very efficient regardless of
the reaction time and amount of catalyst used. In particular, it
was difficult to obtain the desired products pure, and they were al-
ways contaminated by minor impurities of unknown structure.
This is presumably due to the poisoning effect of the amino group
through the coordination to the palladium catalyst. We therefore
synthesized 38 and 39 from 13 and 14, respectively (Scheme 3).
Under standard hydrogenation conditions, the alkyne was re-
duced to the alkane in 97% yield for the synthesis of 38 from 13,
and in 99% yield for the synthesis of 39 from 14. Using potassium
carbonate in methanol, removal of the acetyl group in 38 and 40
OH
OH
3. Conclusions
a
In summary (Fig. 3), the linker moiety, the stereochemistry at
the anomeric position, and the substituents and stereochemistry
at C-4 in the 4-N-TDTH system all influence DNA binding affinity
of this family of compounds. On the other hand, except for the sub-
stituent at C-4, the other structural features examined did not have
a significant impact on cytotoxicity. The information gathered from
these SAR studies further confirmed our previous conclusions6 that
the cytotoxicity of this class of compounds does not arise as a re-
sult of DNA binding. In view of the importance of the amino group
at C-4 for both DNA binding and cytotoxicity, further SAR studies
on the modification of the amino group is warranted.
X
X
32 X = O
33 X = S
34 X = O
35 X = S
Scheme 2. Synthesis of compounds 34 and 35 via reduction of the alkyne to an
alkane. Reagents and conditions: (a) H2, 10% Pd/C, ethyl acetate, 82% for 34 and 89%
for 35.
Table 4
Cytotoxicity of 32–35 against three solid-tumor cell linesa
Compd
Cell viability (%)
4. Experimental section
4.1. Chemistry
100
l
M
25 lM
MCF7
HT29
HepG2b
MCF7
HT29
HepG2
32
33
34
35
2 (0.3)
37 (3.2)
16 (1.1)
<1
<1
8 (1.9)
<1
<1
17 (2.0)
<1
86 (3.7)
90 (5.8)
94 (2.1)
79 (2.9)
86 (5.7)
68 (3.4)
76 (2.1)
<1c
88 (6.4)
97 (3.8)
91 (6.2)
87 (7.1)
All reagents were purchased from commercial sources and were
used without further purification unless noted. Before use, reaction
solvents were purified by successive passage through columns of
alumina and copper in a PURESOLV-400 System from Innovative
Technology Inc. under argon atmosphere. Unless stated otherwise,
all reactions were carried out under a positive pressure of argon
<1
<1
a
b
c
The numbers in parentheses represent the standard deviation.
HepG2/C3A is simplified as HepG2.
IC50 = 16.3 1.5 lM.

1784
W. Shi, T. L. Lowary / Bioorg. Med. Chem. 19 (2011) 1779–1789
O
b
a
O
O
O
X
X
X
AcO
O
O
AcO
HO
38 X = O
39 X = S
40 X = O
41 X = S
13 X = O
14 X = S
d
c
O
O
X
X
O
O
N3
NH2
42 X = O
43 X = S
36 X = O
37 X = S
Scheme 3. Synthesis of 4-N-TDTH compounds 36 and 37 with the propyl linker. Reagents and conditions: (a) H2, 10% Pd/C, ethyl acetate, 97% for 38 and 99% for 39; (b) K2CO3,
methanol, 93% for 40, 96% for 41; (c) PPh3, DPPA, DIAD, ꢀ20 °C?rt, 96% for 42, 95% for 43; (d) PPh3, H2O, THF, 70 °C, 88% for 36, 85% for 37.
acetonitrile and water, the ratio of which depends on the
compound. For all basic amino compounds, 0.1% by volume of tri-
fluoroacetic acid was added to facilitate elution and avoid aggrega-
tion. When the purity derived from HPLC analysis is greater than
99.5%, it is reported as >99%.
Table 5
The percentage of remaining EtBr of 9 versus 36 and 10 versus 37 at 100lM
Compd
Linker
% Remaining EtBr
9
10
Propargyl
68.5 1.9
79.9 1.0
36
37
Propyl
43.5 1.0
51.9 1.0
4.1.1. General procedures
4.1.1.1. Deacetylation. The acetyl protected compound (1 equiv)
was dissolved in CH3OH (30–40 mL per mmol). K2CO3 (30 mol %)
was added, and then the reaction mixture was stirred at rt and fol-
lowed by TLC (ꢁ12 h). After the evaporation of the solvent, the res-
idue was diluted with water and extracted with CH2Cl2. The
combined organic layer was dried (Na2SO4), filtered, and concen-
trated to yield the crude product, which was purified by column
chromatography.
and were monitored by TLC on silica gel G-25 UV254 (0.25 mm,
Macherey–Nagel). Spots were detected under UV light and/or by
charring with 10% H2SO4 in ethanol, or in acidified ethanolic anis-
aldehyde or vanillin. Optical rotations were measured at 22 2 °C.
Melting points are uncorrected. 1H NMR spectra were recorded on
VARIAN INOVA-NMR spectrometers at 400, 500 or 600 MHz, and
chemical shifts are referenced to either TMS (0.0, CDCl3), or the sol-
vent residual CHCl3 (7.26, CDCl3), or CD2HOD (4.78, CD3OD). 13C
NMR spectra were recorded at 100 or 125 MHz, and 13C chemical
shifts are referenced to CDCl3 (77.23 ppm, CD3Cl3) or CD3OD
(49.00 ppm, CD3OD). 1H data are reported as though they were first
order. The errors between the coupling constants for two coupled
protons were less than 0.5 Hz, and the average number was re-
ported. Assignment of NMR resonances was done based on
1H–1H COSY, HMQC, and in some cases HMBC experiments. In
the interpretation of the NMR data for methylene protons on the
carbohydrate ring, ‘a’ and ‘e’ refer to axial and equatorial orienta-
tion, respectively. In the cases where no clear assignment of these
hydrogens could be made based on all NMR data, assignments
were made taking into consideration the anisotropy effect of a ring
4.1.1.2. Mistunobu reaction. To a solution of the hydroxyl com-
pound (1 equiv) in THF (15–20 mL per mmol) was added PPh3
(3 equiv) at ꢀ20 °C. To this mixture was added a solution of DIAD
(2.5 equiv) and DPPA (2.5 equiv) in THF (5–7 mL per mmol) at
ꢀ20 °C. The reaction mixture was then allowed to warm to room
temperature and followed by TLC (ꢁ12 h). The resulted solution
was then diluted with Et2O, washed with brine, dried (Na2SO4), fil-
tered, and concentrated to yield the crude product, which was
purified by column chromatography.
4.1.1.3. Staudinger reaction. To a solution of azide compound
(1 equiv) in THF (30–40 mL per mmol) and H2O (>50 equiv) was
added PPh3 (2.5 equiv), and the reaction was stirred under reflux
and followed by TLC (ꢁ10 h). After evaporation, the residue was
purified by column chromatography to yield the pure amino
compound.
r
bond, which results in equatorial hydrogens resonating more
downfield than axial hydrogens.11 Electrospray mass spectra were
recorded on samples suspended in mixtures of THF with CH3OH
and added NaCl. Optical rotations were measured on a Perkin–
Elmer 241 Polarimeter at the sodium D line (589 nm). Optical rota-
tions are in units of deg mL(dm g)ꢀ1. IR spectra were recorded on
the Nicolet Magna 750 FTIR spectrometer. The reported purity
values were obtained with a Varian HPLC system, using an evapo-
rative light scattering detector (ELSD) 2000ES from Alltech, and a
Varian Microsorb-MV 100-5 C18 column. The eluant consisted of
4.1.1.4. Palladium-catalyzed hydrogenation. A solution of the
alkyne in EtOAc (10 mL) was hydrogenated in the presence of
10% Pd/C at room temperature and normal pressure. After stirring
for 3–8 h, the reaction mixture was filtered through a Celite pad
and concentrated. The crude product was purified by column
chromatography.

W. Shi, T. L. Lowary / Bioorg. Med. Chem. 19 (2011) 1779–1789
1785
Table 6
Cytotoxicity of benzo[b]thiophene analogues with the propyl linker and their counterparts with the propargyl linker against three solid-tumor cell linesa,b
O
O
O
O
S
S
S
O
O
AcO
HO
NH2
10
14
16
O
O
S
S
O
S
O
O
O
AcO
HO
NH2
37
39
41
Compd
Cell viability (%) at 25
HT29
lM
IC50
(
lM)
MCF7
HepG2c
MCF7
HT29
HepG2
10
14
16
37e
<1
<1
<1
6 (0.7)
<1
<1
7.1 1.2
—
—
5.9 0.5
7.2 0.9
11.1 1.1
8.2 1.1
8.8 0.9
12.3 1.4
16.4 0.4
7.5 1.1
d
27 (3.5)
51 (3.2)
<1
6 (1.1)
25 (0.3)
<1
11.8 1.7
47 (2.4)
68 (4.0)
9 (1.7)
72 (1.2)
84 (6.1)
—
—
10.2 1.5
15.3f
—
—
39
19 (2.2)
41
a
b
c
d
e
f
The numbers in parentheses represent the standard deviation.
The compounds are grouped based on the heteroatom and the linker.
HepG2/C3A is simplified as HepG2.
‘—’ = not determined.
The underlined number indicates an analogue with a propyl linker.
Only assayed once.
SAR on DNA binding
SAR on cytotoxicity
minor difference between
alkyne and alkane
alkyne over alkane
minor difference between
axial and equatorial
cis- to C-1 is favored
O
O
O
X
O
X
R
R
minor difference between
β
is favored
amino is required
amino is favored
α and β
X = O, S, NTs
Figure 3. Schematic representation of structure–activity relationships on DNA binding and cytotoxicity.
4.1.2. 3-(2-Phenyl-benzo[b]thiophen-3-yl)-prop-2-ynyl 4-O-
benzyl-2,3,6-trideoxy-a-L-threo-hexopyranoside (26)
4.64 (s, 2H, OCH2C„C), 4.20 (br q, 1H, J5,6 = 6.6 Hz, H-5),
2.20–2.28 (m, 1H, H-3e), 2.06–2.14 (m, 1H, H-2e), 1.95–2.01 (m,
1H, H-3a), 1.68–1.74 (m, 1H, H-2a), 1.19 (d, 3H, J5,6 = 6.6 Hz, H-
6); 13C NMR (125 MHz, CDCl3, dC) 166.1 (C@O), 146.9 (Ar), 141.1
(Ar), 137.5 (Ar), 133.7 (Ar), 133.0 (Ar), 130.4 (Ar), 129.7 (2, Ar),
128.8 (Ar), 128.7 (2, Ar), 128.4 (4, Ar), 125.2 (Ar), 125.0 (Ar),
123.3 (Ar), 122.1 (Ar), 113.0 (Ar), 95.9 (C-1), 90.4 („C), 80.4
(„C), 70.0 (C-4), 65.8 (C-5), 55.1 (OCH2), 24.1 (C-2), 23.0 (C-3),
17.2 (C-6). HRMS (ESI) Calcd for (M+Na) C30H26O4SNa: 505.1444.
Found: 505.1445. Purity: >99%.
To a solution of compound 16 (87 mg, 0.23 mmol) in THF (5 mL)
was added PPh3 (181 mg, 0.692 mmol) at ꢀ20 °C. To this mixture
was added a solution of DIAD (117 mg, 0.582 mmol) and benzoic
acid (70 mg, 0.58 mmol) in THF (3 mL) at ꢀ20 °C. The reaction
was stirred for 12 h while warming to room temperature, and then
diluted with Et2O. The organic layer was washed with brine, dried
(Na2SO4), filtered, and concentrated to yield the crude product,
which was purified by column chromatography (12:1, hexanes–
EtOAc) to yield pure 26 (88 mg, 79%) as a colorless foamy solid:
Rf 0.27 (12:1 hexanes–EtOAc); IR:
m
2216 (C„C), 1716 (C@O)
4.1.3. 3-(2-Phenyl-benzo[b]thiophen-3-yl)-prop-2-ynyl 2,3,6-
trideoxy-a-L-threo-hexopyranoside (27)
Compound 26 (67 mg, 0.14 mmol) was dissolved in CH3OH
(10 mL). A catalytic amount of 1 M sodium methoxide in CH3OH
was added, and then the solution was stirred for 12 h. The solvent
cmꢀ1; [
a
]
D
ꢀ79.9 (c 1.0, CH2Cl2); 1H NMR (500 MHz, CDCl3, dH)
8.13–8.16 (m, 2H, Ar), 7.99–8.03 (m, 2H, Ar), 7.96–7.99 (m, 1H,
Ar), 7.80–7.83 (m, 1H, Ar), 7.57–7.61 (m, 1H, Ar), 7.37–7.51 (m,
7H, Ar), 5.26 (br d, 1H, J1,2a = 4.0 Hz, H-1), 5.09 (br s, 1H, H-4),

1786
W. Shi, T. L. Lowary / Bioorg. Med. Chem. 19 (2011) 1779–1789
was evaporated and the residue was purified by column chroma-
for
C17H16O2: 252.1150. Found: 252.1152. Anal. Calcd for
tography (2:1, hexanes–EtOAc) to acquire pure 27 (53 mg, 99%)
C17H16O2: C, 80.93; H, 6.39. Found: C, 81.01; H, 6.36.
as a yellowish syrup: Rf 0.21 (2:1 hexanes–EtOAc); IR:
m 3445
(O–H), 2216 (C„C) cmꢀ1; [
a]
D
ꢀ97.9 (c 1.1, CH2Cl2); 1H NMR
4.1.7. 3-(2-Phenylbenzothiophen-3-yl)-propanol (35)
(500 MHz, CDCl3, dH) 7.96–8.00 (m, 2H, Ar), 7.92–7.96 (m, 1H,
Ar), 7.79–7.82 (m, 1H, Ar), 7.36–7.49 (m, 5H, Ar), 5.13 (br s, 1H,
H-1), 4.58 (s, 2H, OCH2C„C), 4.04 (br q, 1H, J5,6 = 6.6 Hz, H-5),
3.60 (br s, 1H, H-4), 1.99–2.10 (m, 2H, H-2e, H-3e), 1.74–1.84 (m,
2H, OH, H-3a), 1.59–1.66 (m, 1H, H-2a), 1.19 (d, 3H, J5,6 = 6.6 Hz,
H-6); 13C NMR (125 MHz, CDCl3, dC) 146.9 (Ar), 141.1 (Ar), 137.5
(Ar), 133.7 (Ar), 128.8 (Ar), 128.7 (2, Ar), 128.4 (2, Ar), 125.2 (Ar),
124.9 (Ar), 123.3 (Ar), 122.0 (Ar), 113.0 (Ar), 95.9 (C-1), 90.4
(„C), 80.3 („C), 67.3 (C-4), 66.6 (C-5), 54.9 (OCH2), 25.7 (C-2),
23.2 (C-3), 17.1 (C-6). HRMS (ESI) Calcd for (M+Na) C23H22O3SNa:
401.1182. Found: 401.1183. Purity: >99%.
This compound was synthesized as a white foamy solid from 33
(26 mg, 0.10 mmol) and 10% Pd/C (21 mg, 20 mol % in palladium) in
89% yield by following procedure D: Rf 0.47 (2:1 hexanes–EtOAc); 1H
NMR (500 MHz, CDCl3, dH) 7.84–7.87 (m, 1H, Ar), 7.78–7.81 (m, 1H,
Ar), 7.53–7.56 (m, 2H, Ar), 7.45–7.49 (m, 2H, Ar), 7.38–7.43 (m, 2H,
Ar), 7.35–7.38 (m, 1H, Ar), 4.70 (t, 2H, J = 5.7 Hz, CH2OH), 4.70 (dd,
2H, J = 7.7 Hz, J = 6.1 Hz, CH2CH2CH2OH), 1.90–1.96 (m, 2H,
CH2CH2OH), 1.39 (br s, 1H, OH); 13C NMR (100 MHz, CDCl3, dC)
140.2 (Ar), 139.3 (Ar), 138.7 (Ar), 134.6 (Ar), 131.6 (Ar), 129.7 (2,
Ar), 128.7 (2, Ar), 128.1 (Ar), 124.3 (Ar), 124.2 (Ar), 122.3 (Ar),
122.2 (Ar), 62.3 (CH2OH), 32.9 (CH2CH2OH), 20.4 (CH2CH2CH2OH).
HRMS (EI) Calcd for C17H16OS: 268.0922. Found: 268.0920. Purity:
99.3%.
4.1.4. 3-(2-Phenyl-benzo[b]thiophen-3-yl)-prop-2-ynyl 4-azido-
2,3,4,6-tetradeoxy-a-L-erythro-hexopyranoside (28)
This compound was synthesized as a colorless oil from 27
(60 mg, 0.16 mmol), PPh3 (126 mg, 0.481 mmol), DIAD (81 mg,
0.40 mmol), and DPPA (110 mg, 0.402 mmol) in 56% yield by fol-
4.1.8. 3-(2-Phenyl-benzo[b]furan-3-yl)-propyl 4-amino-2,3,4,6-
tetradeoxy-a-L-threo-hexopyranoside (36)
This compound was synthesized as a colorless syrup from 42
(82 mg, 0.21 mmol), H2O (0.08 mL), and PPh3 (94 mg, 0.36 mmol)
in 88% yield by following procedure C: Rf 0.32 (10:1 CH2Cl2–
lowing procedure B: Rf 0.24 (20:1 hexanes–EtOAc); IR:
m
2217
cmꢀ1 (C„C), 2099 cmꢀ1 (N@N@N); [
a
]
D
ꢀ100.1 (c 0.7, CH2Cl2);
1H NMR (500 MHz, CDCl3, dH) 7.96–8.00 (m, 2H, Ar), 7.92–7.96
(m, 1H, Ar), 7.79–7.82 (m, 1H, Ar), 7.36–7.49 (m, 5H, Ar), 5.14 (br
d, 1H, J1,2a = 3.3 Hz, H-1), 4.57 (s, 2H, OCH2C„C), 3.72 (dq, 1H,
J4,5 = 9.8 Hz, J5,6 = 6.2 Hz, H-5), 3.00–3.08 (m, 1H, H-4), 1.92–2.00
(m, 3H, H-3a, H-3e, H-2e), 1.79–1.87 (m, 1H, H-2a), 1.27 (d, 3H,
J5,6 = 6.2 Hz, H-6); 13C NMR (125 MHz, CDCl3, dC) 147.0 (Ar),
141.1 (Ar), 137.5 (Ar), 133.7 (Ar), 128.8 (Ar), 128.7 (2, Ar), 128.4
(2, Ar), 125.2 (Ar), 125.0 (Ar), 123.2 (Ar), 122.0 (Ar), 112.9 (Ar),
94.6 (C-1), 90.1 („C), 80.5 („C), 67.9 (C-5), 62.5 (C-4), 54.8
(OCH2), 29.1 (C-2), 23.7 (C-3), 18.6 (C-6). HRMS (EI) Calcd for
CH3OH); IR:
m
3368, 3298 (N–H) cmꢀ1; [
a
]
D
ꢀ45.1 (c 0.8, CH3OH);
1H NMR (500 MHz, CD3OD, dH) 7.77–7.80 (m, 2H, Ar), 7.53–7.58 (m,
1H, Ar), 7.41–7.46 (m, 3H, Ar), 7.32–7.36 (m, 1H, Ar), 7.24–7.28 (m,
1H, Ar), 7.18–7.22 (m, 1H, Ar), 4.61 (br d, 1H, J1,2a = 3.4 Hz, H-1),
3.84 (br q, 1H, J5,6 = 6.5 Hz, H-5), 3.64 (dt, 1H, J = 9.8 Hz,
J = 5.9 Hz, OCH2), 3.35 (dt, 1H, J = 9.8 Hz, J = 5.8 Hz, OCH2),
2.97–3.08 (m, 2H, CH2CH2CH2O, 2.58 (br s, 1H, H-4), 1.94–2.04
(m, 3H, CH2CH2O, H-3e), 1.76–1.85 (m, 1H, H-2e), 1.52–1.58 (m,
1H, H-3a), 1.39–1.45 (m, 1H, H-2a), 0.98 (d, 3H, J5,6 = 6.5 Hz,
H-6); 13C NMR (125 MHz, CD3OD, dC) 155.4 (Ar), 152.0 (Ar), 132.6
(Ar), 131.6 (Ar), 129.8 (2, Ar), 129.3 (Ar), 127.9 (2, Ar), 125.6 (Ar),
123.6 (Ar), 120.7 (Ar), 116.9 (Ar), 111.9 (Ar), 98.5 (C-1), 67.1(9)
(C-5), 67.1(7) (OCH2), 49.4 (C-4), 30.5 (CH2CH2O), 26.9 (C-3), 24.5
(C-2), 21.7 (CH2CH2CH2O), 17.8 (C-6). HRMS (ESI) Calcd for
(M+H) C23H28NO3: 366.2064. Found: 366.2066. Purity: >99%.
C23H21N3O2S: 403.1354. Found: 403.1351.
4.1.5. 3-(2-Phenyl-benzo[b]thiophen-3-yl)-prop-2-ynyl 4-
amino-2,3,4,6-tetradeoxy- -threo-hexopyranoside (29)
a-L
This compound was synthesized as a colorless oil from 28
(20 mg, 0.051 mmol), H2O (0.02 mL), and PPh3 (26 mg,
0.099 mmol) in 75% yield by following procedure C: Rf 0.19 (15:1
4.1.9. 3-(2-Phenyl-benzo[b]thiophen-3-yl)-propyl 4-amino-
2,3,4,6-tetradeoxy-a-L-threo-hexopyranoside (37)
CH2Cl2–CH3OH); IR:
m
3378 (N–H), 2215 (C„C) cmꢀ1
;
[
a]
D
ꢀ116.0 (c 2.4, CH2Cl2); 1H NMR (500 MHz, CD3OD, dH) 7.94–7.98
(m, 2H, Ar), 7.86–7.89 (m, 1H, Ar), 7.82–7.84 (m, 1H, Ar),
7.36–7.48 (m, 5H, Ar), 5.03 (br s, 1H, H-1), 4.56 (s, 2H, OCH2C„C),
4.06 (dq, 1H, J4,5 = 9.4 Hz, J5,6 = 6.2 Hz, H-5), 2.64 (ddd, 1H,
J3a,4 = 11.4 Hz, J4,5 = 9.4 Hz, J3e,4 = 4.3 Hz, H-4), 1.70–1.82 (m, 3H,
H-2a, H-2e, H-3a), 1.58–1.67 (m, 1H, H-3e), 1.18 (d, 3H,
J5,6 = 6.2 Hz, H-6); 13C NMR (125 MHz, CD3OD, dC) 147.8 (Ar),
142.3 (Ar), 138.8 (Ar), 134.9 (Ar), 130.1 (Ar), 129.8 (2, Ar), 129.4
(2, Ar), 126.6 (Ar), 126.2 (Ar), 124.1 (Ar), 123.2 (Ar), 114.1 (Ar),
96.5 (C-1), 92.1 („C), 80.9 („C), 72.2 (C-5), 55.4 (OCH2), 54.3
(C-4), 30.7 (C-2), 27.9 (C-3), 18.5 (C-6). HRMS (ESI) Calcd for
(M+H) C23H24NO2S: 378.1522. Found: 378.1524. Purity: 99.3%.
This compound was synthesized as a yellowish syrup from 43
(76 mg, 0.17 mmol), H2O (0.06 mL), and PPh3 (94 mg, 0.29 mmol)
in 85% yield by following procedure C: Rf 0.31 (10:1 CH2Cl2–
CH3OH); IR:
m
3364 (N–H) cmꢀ1; [
a]
ꢀ26.7 (c 2.5, CH3OH); 1H
D
NMR (500 MHz, CD3OD, dH) 7.76–7.82 (m, 2H, Ar), 7.49–7.52 (m,
2H, Ar), 7.41–7.46 (m, 2H, Ar), 7.35–7.40 (m, 2H, Ar), 7.29–7.34
(m, 1H, Ar), 4.62 (br s, 1H, H-1), 3.88 (br q, 1H, J5,6 = 6.6 Hz, H-5),
3.57 (dt, 1H, J = 9.8 Hz, J = 5.8 Hz, OCH2), 3.27–3.33 (m, 1H,
OCH2), 2.95–3.06 (m, 3H, H-4, CH2CH2CH2O), 2.00–2.09 (m, 1H,
H-3e), 1.86–1.92 (m, 2H, CH2CH2O), 1.69–1.78 (m, 2H, H-2e, H-
3a), 1.40–1.47 (m, 1H, H-2a), 1.04 (d, 3H, J5,6 = 6.6 Hz, H-6); 13C
NMR (125 MHz, CD3OD, dC) 141.7 (Ar), 140.6 (Ar), 139.7 (Ar),
136.1 (Ar), 133.0 (Ar), 130.7 (2, Ar), 129.8 (2, Ar), 129.2 (Ar),
125.5 (Ar), 125.3 (Ar), 123.4 (Ar), 123.2 (Ar), 98.1 (C-1), 67.6
(OCH2), 64.9 (C-5), 50.1 (C-4), 31.0 (CH2CH2O), 24.3 (C-3), 24.1
(C-2), 24.0 (CH2CH2CH2O), 17.6 (C-6). HRMS (ESI) Calcd for
(M+H) C23H28NO2S: 382.1835. Found: 378.1834. Purity: 98.3%.
4.1.6. 3-(2-Phenylbenzofuran-3-yl)-propanol (34)
This compound was synthesized as a colorless oil from 32 (25 mg,
0.11 mmol) and 10% Pd/C (21 mg, 20 mol % in palladium) in 82%
yield by following procedure D: Rf 0.27 (3:1 hexanes–EtOAc); 1H
NMR (600 MHz, CDCl3, dH) 7.82–7.85 (m, 2H, Ar), 7.59–7.62 (m,
1H, Ar), 7.47–7.52 (m, 3H, Ar), 7.36–7.43 (m, 1H, Ar), 7.29–7.33 (m,
1H, Ar), 7.24–7.28 (m, 1H, Ar), 4.70 (t, 2H, J = 6.3 Hz, CH2OH), 4.70
(t, 2H, J = 7.7 Hz, CH2CH2CH2OH), 2.01–2.07 (m, 2H, CH2CH2OH),
1.46 (br s, 1H, OH); 13C NMR (125 MHz, CDCl3, dC) 153.9 (Ar), 150.9
(Ar), 131.3 (Ar), 130.4 (Ar), 128.7 (2, Ar), 128.2 (Ar), 126.9 (2, Ar),
124.4 (Ar), 122.4 (Ar), 119.5 (Ar), 115.5 (Ar), 111.1 (Ar), 62.3
(CH2OH), 32.4 (CH2CH2OH), 20.4 (CH2CH2CH2OH). HRMS (EI) Calcd
4.1.10. 3-(2-Phenyl-benzo[b]furan-3-yl)-propyl 4-O-acetyl-
2,3,6-trideoxy-a-L-erythro-hexopyranoside (38)
This compound was synthesized as a colorless oil from 13
(126 mg, 0.311 mmol) and 10% Pd/C (33 mg, 10 mol % in palla-
dium) in 97% yield by following procedure D: Rf 0.50 (10:1 hex-
anes–EtOAc); IR: 1735 (C@O) cmꢀ1; [
a]
ꢀ51.3 (c 0.8, CH2Cl2);
D
1H NMR (500 MHz, CDCl3, dH) 7.85–7.88 (m, 2H, Ar), 7.60–7.63

W. Shi, T. L. Lowary / Bioorg. Med. Chem. 19 (2011) 1779–1789
1787
(m, 1H, Ar), 7.47–7.54 (m, 3H, Ar), 7.37–7.41 (m, 1H, Ar), 7.30–7.34
(m, 1H, Ar), 7.25–7.29 (m, 1H, Ar), 4.72 (br s, 1H, H-1), 4.54
(ddd, 1H, J4,5 = 9.7 Hz, J3a,4 = 9.7 Hz, J3e,4 = 4.4 Hz, H-4),
3.76–3.86 (m, 2H, H-5, OCH2), 3.48 (dt, 1H, J = 9.8 Hz, J = 6.0 Hz,
OCH2), 3.02–3.16 (m, 2H, CH2CH2CH2O), 2.05–2.12 (m, 5H,
CH2CH2O, O@CCH3), 1.94–2.01 (m, 1H, H-3e), 1.76–1.93 (m, 3H,
H-2a, H-2e, H-3a), 1.16 (d, 3H, J5,6 = 6.3 Hz, H-6); 13C NMR
(125 MHz, CDCl3, dC) 170.3 (C@O), 154.0 (Ar), 150.8 (Ar), 131.3
(Ar), 130.5 (Ar), 128.7 (2, Ar), 128.1 (Ar), 126.9 (2, Ar), 124.4 (Ar),
122.4 (Ar), 119.5 (Ar), 115.7 (Ar), 111.1 (Ar), 96.3 (C-1), 73.6
(C-4), 66.6 (C-5), 66.3 (OCH2), 29.6 (CH2CH2O), 29.3 (C-2), 24.2
(C-3), 21.2 (O@CCH3), 21.0 (CH2CH2CH2O), 17.9 (C-6). HRMS (ESI)
Calcd for (M+Na) C25H28O5Na: 431.1834. Found: 431.1836. Anal.
Calcd for C25H28O5: C, 73.51; H, 6.91. Found: C, 73.29; H, 7.07.
J = 9.8 Hz, J = 6.0 Hz, OCH2), 3.51 (dq, 1H, J4,5 = 9.0 Hz, J5,6 = 6.2 Hz,
H-5), 3.38 (dt, 1H, J = 9.8 Hz, J = 6.2 Hz, OCH2), 3.20–3.27 (m, 1H,
H-4), 2.96–3.10 (m, 2H, CH2CH2CH2O), 1.94–2.01 (m, 2H,
CH2CH2O), 1.68–1.88 (m, 4H, H-2a, H-2e, H-3a, H-3e), 1.51 (br s,
1H, OH), 1.22 (d, 3H, J5,6 = 6.2 Hz, H-6); 13C NMR (125 MHz, CDCl3,
dC) 140.4 (Ar), 139.3 (Ar), 138.6 (Ar), 134.8 (Ar), 131.8 (Ar), 129.7
(2, Ar), 128.6 (2, Ar), 127.9 (Ar), 124.2 (Ar), 124.1 (Ar), 122.2(7)
(Ar), 122.2(6) (Ar), 96.1 (C-1), 72.1 (C-4), 69.4 (C-5), 66.3 (OCH2),
30.1 (CH2CH2O), 29.7 (C-2), 27.7 (C-3), 23.6 (CH2CH2CH2O), 17.9
(C-6). HRMS (ESI) Calcd for (M+Na)
Found: 405.1501. Purity: >99%.
C23H26O3SNa: 405.1500.
4.1.14. 3-(2-Phenyl-benzo[b]furan-3-yl)-propyl 4-azido-2,3,4,6-
tetradeoxy- -threo-hexopyranoside (42)
a-L
This compound was synthesized as a colorless oil from 40
(99 mg, 0.27 mmol), PPh3 (212 mg, 0.808 mmol), DIAD (137 mg,
0.678 mmol), and DPPA (187 mg, 0.678 mmol) in 96% yield by fol-
4.1.11. 3-(2-Phenyl-benzo[b]thiophene-3-yl)-propyl 4-O-acetyl-
2,3,6-trideoxy-a-L-erythro-hexopyranoside (39)
This compound was synthesized as a colorless oil from 14
(133 mg, 0.321 mmol) and 10% Pd/C (34 mg, 10 mol % in palla-
dium) in 99% yield by following procedure D: Rf 0.29 (10:1
lowing procedure C: Rf 0.26 (20:1 hexanes–EtOAc); IR: m 2100
cmꢀ1 (N@N@N); [
a]
ꢀ17.9 (c 0.5, CH2Cl2); 1H NMR (500 MHz,
D
CDCl3, dH) 7.82–7.85 (m, 2H, Ar), 7.58–7.61 (m, 1H, Ar), 7.45–7.53
(m, 3H, Ar), 7.35–7.40 (m, 1H, Ar), 7.29–7.34 (m, 1H, Ar), 7.23–
7.28 (m, 1H, Ar), 4.76 (br s, 1H, H-1), 3.91 (dq, 1H, J5,6 = 6.5 Hz,
J4,5 = 1.7 Hz, H-5), 3.72 (dt, 1H, J = 9.8 Hz, J = 6.0 Hz, OCH2), 3.45
(dt, 1H, J = 9.8 Hz, J = 6.1 Hz, OCH2), 3.39 (br s, 1H, H-4), 3.01–
3.13 (m, 2H, CH2CH2CH2O), 2.02–2.18 (m, 3H, CH2CH2O, H-3e),
1.88–1.97 (m, 2H, H-2e, H-3a), 1.53–1.60 (m, 1H, H-2a), 1.16 (d,
3H, J5,6 = 6.5 Hz, H-6); 13C NMR (125 MHz, CDCl3, dC) 154.0 (Ar),
150.9 (Ar), 131.3 (Ar), 130.4 (Ar), 128.7 (2, Ar), 128.1 (Ar), 126.9
(2, Ar), 124.4 (Ar), 122.4 (Ar), 119.5 (Ar), 115.5 (Ar), 111.1 (Ar),
96.9 (C-1), 66.3 (OCH2), 65.2 (C-5), 60.0 (C-4), 29.4 (CH2CH2O),
24.2 (C-2), 23.0 (C-3), 20.9 (CH2CH2CH2O), 17.9 (C-6). HRMS (ESI)
Calcd for (M+Na) C23H25N3O3Na: 414.1794. Found: 414.1792.
hexanes–EtOAc); IR: 1737 (C@O) cmꢀ1; [
a]
D ꢀ61.8 (c 0.8, CH2Cl2);
1H NMR (400 MHz, CDCl3, dH) 7.80–7.87 (m, 2H, Ar), 7.56–7.59 (m,
2H, Ar), 7.34–7.49 (m, 5H, Ar), 4.65 (br s, 1H, H-1), 4.47–4.53 (m,
1H, H-4), 3.78 (dq, 1H, J4,5 = 9.5 Hz, J5,6 = 6.2 Hz, H-5), 3.70 (dt,
1H, J = 9.8 Hz, J = 6.0 Hz, OCH2), 3.40 (dt, 1H, J = 9.8 Hz, J = 6.2 Hz,
OCH2), 2.97–3.12 (m, 2H, CH2CH2CH2O), 2.07 (s, 3H, O@CCH3),
1.91–2.06 (m, 3H, CH2CH2O, H-3e), 1.74–1.86 (m, 3H, H-2a, H-2e,
H-3a), 1.14 (d, 3H, J5,6 = 6.2 Hz, H-6); 13C NMR (100 MHz, CDCl3,
dC) 170.3 (C@O), 140.4 (Ar), 139.3 (Ar), 138.7 (Ar), 134.8 (Ar),
131.8 (Ar), 129.7 (2, Ar), 128.6 (2, Ar), 128.0 (Ar), 124.2 (Ar),
124.1 (Ar), 122.3 (Ar), 122.2 (Ar), 96.2 (C-1), 73.6 (C-4), 66.5
(C-5), 66.4 (OCH2), 30.2 (CH2CH2O), 29.2 (C-2), 24.2 (C-3), 23.6
(CH2CH2CH2O), 21.2 (O@CCH3), 17.9 (C-6). HRMS (ESI) Calcd for
(M+Na) C25H28O4SNa: 447.1606. Found: 447.1609. Anal. Calcd for
4.1.15. 3-(2-Phenyl-benzo[b]thiophen-3-yl)-propyl 4-azido-
C25H28O4S: C, 70.73; H, 6.65; S, 7.55. Found: C, 70.34; H, 6.89; S,
2,3,4,6-tetradeoxy-a-L-threo-hexopyranoside (43)
7.67.
This compound was synthesized as a colorless oil from 42
(96 mg, 0.25 mmol), PPh3 (197 mg, 0.752 mmol), DIAD (127 mg,
0.629 mmol), and DPPA (173 mg, 0.629 mmol) in 95% yield by
4.1.12. 3-(2-Phenyl-benzo[b]furan-3-yl)-propyl 2,3,6-trideoxy-
a
-
L
-erythro-hexopyranoside (40)
following procedure C: Rf 0.29 (20:1 hexanes–EtOAc); IR: m
This compound was synthesized as a colorless oil from 38
2097 cmꢀ1 (N@N@N);
[
a]
ꢀ29.2 (c 0.6, CH2Cl2); 1H NMR
D
(123 mg, 0.301 mmol) and K2CO3 (12 mg, 0.088 mmol) in 93% yield
(400 MHz, CDCl3, dH) 7.84–7.87 (m, 1H, Ar), 7.78–7.81 (m, 1H,
Ar), 7.54–7.58 (m, 2H, Ar), 7.33–7.48 (m, 5H, Ar), 4.70 (br s, 1H,
H-1), 3.88 (dq, 1H, J5,6 = 6.5 Hz, J4,5 = 1.7 Hz, H-5), 3.65 (dt, 1H,
J = 9.9 Hz, J = 6.0 Hz, OCH2), 3.37–3.43 (m, 2H, H-4, OCH2), 2.97–
3.10 (m, 2H, CH2CH2CH2O), 2.04–2.14 (m, 1H, H-3e), 1.85–2.01
(m, 4H, H-2e, H-3a, CH2CH2O,), 1.47–1.52 (m, 1H, H-2a), 1.16 (d,
3H, J5,6 = 6.5 Hz, H-6); 13C NMR (125 MHz, CDCl3, dC) 140.4 (Ar),
139.3 (Ar), 138.7 (Ar), 134.8 (Ar), 131.7 (Ar), 129.7 (2, Ar), 128.6
(2, Ar), 128.0 (Ar), 124.2 (Ar), 124.1 (Ar), 122.3 (Ar), 122.2 (Ar),
96.8 (C-1), 66.5 (OCH2), 65.1 (C-5), 60.0 (C-4), 30.1 (CH2CH2O),
24.1 (C-2), 23.5 (C-3), 23.0 (CH2CH2CH2O), 18.0 (C-6). HRMS (ESI)
Calcd for (M+Na) C23H25N3O2SNa: 430.1565. Found: 430.1568.
by following procedure A: Rf 0.26 (3:1 hexanes–EtOAc); IR:
m 3422
(O–H) cmꢀ1; [
a
]
D
ꢀ39.4 (c 1.5, CH2Cl2); 1H NMR (500 MHz, CDCl3,
dH) 7.83–7.86 (m, 2H, Ar), 7.58–7.61 (m, 1H, Ar), 7.46–7.53 (m, 3H,
Ar), 7.35–7.40 (m, 1H, Ar), 7.26–7.33 (m, 2H, Ar), 4.68 (br s, 1H,
H-1), 3.76 (dt, 1H, J = 9.8 Hz, J = 6.0 Hz, OCH2), 3.55 (dq, 1H,
J4,5 = 9.1 Hz, J5,6 = 6.3 Hz, H-5), 3.45 (dt, 1H, J = 9.8 Hz, J = 6.1 Hz,
OCH2), 3.22–3.28 (m, 1H, H-4), 3.01–3.14 (m, 2H, CH2CH2CH2O),
2.03–2.10 (m, 2H, CH2CH2O), 1.70–1.90 (m, 4H, H-2a, H-2e, H-3a,
H-3e), 1.47 (br s, 1H, OH), 1.23 (d, 3H, J5,6 = 6.3 Hz, H-6); 13C
NMR (125 MHz, CDCl3, dC) 154.0 (Ar), 150.8 (Ar), 131.3 (Ar),
130.5 (Ar), 128.7 (2, Ar), 128.1 (Ar), 126.9 (2, Ar), 124.3 (Ar),
122.4 (Ar), 119.6 (Ar), 115.6 (Ar), 111.1 (Ar), 96.2 (C-1), 72.1
(C-4), 69.5 (C-5), 66.1 (OCH2), 29.7 (CH2CH2O), 29.5 (C-2), 27.7
(C-3), 20.9 (CH2CH2CH2O), 17.9 (C-6). HRMS (ESI) Calcd for
(M+Na) C23H26O4Na: 389.1723. Found: 389.1721. Anal. Calcd for
4.2. DNA binding fluorescence assay
The fluorescence was measured according to the literature re-
ported procedure.3,4 Sheared herring sperm DNA (10 mg/mL in
10 mM Tris–HCl, 10 mM NaCl, 1 mM EDTA, pH 7.5) was purchased
from Promega and diluted with BPE (bis-phosphate EDTA) buffer to
2 mg/mL as the stock solution. BPE buffer consists of 6 mM Na2H-
PO4, 2 mM NaH2PO4, 1 mM Na2EDTA, pH 7.0 in Milli-Q water.
Molecular-biological grade DMSO and ethidium bromide (EtBr)
were purchased from Sigma–Aldrich. A stock solution of EtBr
(1 mM) was prepared in BPE buffer. The test analogues were
dissolved in DMSO to make 10 mM stock solutions and further
diluted with DMSO, as needed, to the desired concentration prior
C23H26O4: C, 75.38; H, 7.15. Found: C, 75.73; H, 7.05.
4.1.13. 3-(2-Phenyl-benzo[b]thiophene-3-yl)-propyl 2,3,6-
trideoxy- -erythro-hexopyranoside (41)
a-L
This compound was synthesized as a colorless oil from 39
(127 mg, 0.303 mmol) and K2CO3 (12 mg, 0.088 mmol) in 96% yield
by following procedure A: Rf 0.44 (2:1 hexanes–EtOAc); IR: 3403
(O–H) cmꢀ1; [
a]
ꢀ45.0 (c 1.6, CH2Cl2); 1H NMR (400 MHz, CDCl3,
D
dH) 7.83–7.86 (m, 1H, Ar), 7.78–7.82 (m, 1H, Ar), 7.54–7.58 (m, 2H,
Ar), 7.33–7.48 (m, 5H, Ar), 4.62 (br s, 1H, H-1), 3.68 (dt, 1H,

1788
W. Shi, T. L. Lowary / Bioorg. Med. Chem. 19 (2011) 1779–1789
to the assay. Fluorescence was measured on a PTI (Photon Technol-
ogy International) MP1 fluorescence system.
4.3.1. Cell culture
Three cancer cell lines (MCF-7, HT-29, and HepG2/C3A) and
three non-tumor cell lines (CRL-7761, Vero, and NIH/3T3) were
purchased from the American Type Culture Collection (ATCC).
NIH/3T3 mouse lung cells were maintained in DMEM/HIGH culture
4.2.1. Single-point fluorescent intercalator displacement (FID)
assay
media supplemented with 10% calf serum, 2 mM
unit penicillin, and 100 g streptomycin. Vero monkey kidney cells
were maintained in DMEM/HIGH culture media supplemented
with 5% FBS, 2 mM -glutamine, 10 mM HEPES (N-2-hydroxyethyl-
piperazine-N0-2-ethane sulfonic acid), 100 unit penicillin, and
100 g streptomycin. All the rest four cell lines were maintained
in DMEM/HIGH culture media supplemented with 10% FBS and
2 mM -glutamine. Cell cultures were grown in monolayers in a
L-glutamine, 100
Three individual fluorescence measurements were required for
l
this assay. First, to a fluorescence microcell was added 98.5
lL of
an ethidium bromide (EtBr) solution in PBE buffer (5 M), and its
l
L
emission curve was measured from 540 nm to 700 nm at an exci-
tation wavelength of 520 nm. Second, to the above EtBr solution
was added 1 lL of a 2 mg/mL stock solution of hsDNA in PBE buf-
fer. After mixing and a short-time equilibration (ꢁ10 s), the emis-
sion scan was performed under the same set of parameters until
the error between the maxima of two continuous emission curves
was <10%. The intensity difference at the emission maximum from
the two previous measurements was set as 100% for normalization.
l
L
humidified atmosphere of 5% CO2 and 95% air at 37 °C. The culture
media were changed every 2–3 days. Cell cultures were passaged
once a week using trypsin-EDTA (0.25%) to detach the cells from
their culture flasks.
Finally, 0.5
DMSO was added to the mixture in the cell, which gave a final con-
centration of 50
M. After mixing and equilibration (ꢁ3 min), the
lL of a 10 mM stock solution of a test compound in
4.3.2. MTS non-radioactive cell proliferation assay
l
Rapidly growing cells were counted and seeded at a concentra-
emission scan was performed until steady measurements were ob-
tained. The normalized intensity difference at the emission maxi-
mum between the first measurement and this measurement was
used to represent the percentage of remaining EtBr. Alternatively,
the normalized intensity difference between the second and third
measurements was used to represent the percentage of displaced
EtBr. Under the above assay conditions, no fluorescence was ob-
served for all the reagents and their combinations except for EtBr.
Each compound was assayed in triplicate, and the values shown in
the table reflect an average of this data. In cases where the percent-
age is slightly over 100%, this is presumably due to the weak fluo-
rescence enhancement between the assayed compounds and free
EtBr or the EtBr–DNA complex.
tion of 1 ꢃ 104 cells/well in 100
lL total volume per well into a
96-well microtiter plate. After incubation at 37 °C for 24 h, the cul-
ture medium was removed and the cell assay media (DMEM-F12
media with 1.0 mg/mL human albumin, 5.0 mg/L human transfer-
in, and 5.0 mg/L bovine insulin) with or without tested compounds
in 100 lL total volume were added to each well containing the
exponentially growing cancer cells. Culture wells (triplicate to
quintuplicate per sample) were incubated for 3 days at 37 °C (5%
CO2, 95% air). Then 20 lL of MTS dye solution was added to each
sample well. After 2–2.5 h of incubation at 37 °C, the absorbance
of formazan was recorded at 490 nm with a SPECTRAmax absor-
bance microtitre plate reader. Cell viability was calculated as a per-
centage of control wells, which contained the cell assay media
only. Statistical and graphical analysis information was deter-
mined using Origin Pro 7.5 software (OriginLab Corp.) and Micro-
soft Excel (Microsoft Corp.). The determination of IC50 values was
performed using nonlinear regression analysis based on the Boltz-
mann function as follows:
4.2.2. Multiple-point fluorescent intercalator displacement
(FID) assay
The assay is nearly identical to the above single-point assay. The
major difference is that in step 3, 0.2 lL aliquots of a 10 mM stock
solution of a test compound in DMSO was added to the fluorescing
solution in the cell, and the fluorescence was measured after each
addition until a 50% reduction of fluorescence occured, correspond-
ing to a 50% displacement of EtBr by the test compound. The appar-
ent binding constant of the test ligand was then calculated from
the equation:
A1 ꢀ A2
y ¼ A2 ꢃ
0Þ=dx
1 þ eðxꢀx
where y is the dependent variable (percentage of growth inhibi-
tion), x is the independent variable (concentration), and A1, A2, x0,
and dx are the parameter values (A1 is initial value, A2 is final value,
x0 is center, and dx is time constant).
½EtBrꢄ
KLigand ¼ KEtBr
ꢃ
½Ligandꢄ50%
Acknowledgments
where [Ligand]50% is the concentration of ligand that gives a 50%
reduction of fluorescence, [EtBr] is the concentration of EtBr, and
KEtBr is the binding constant for ethidium bromide.
This work was supported by the University of Alberta, the Nat-
ural Sciences and Engineering Research Council of Canada (NSERC)
and the Alberta Ingenuity Centre for Carbohydrate Science. W.S. is
the recipient of a Studentship from the Alberta Ingenuity Fund.
4.3. Cytotoxicity assay
CellTiter 96Ò aqueous one solution MTS dye, a mixture of
[3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-
sulfophenyl)-2H-tetrazolium (MTS), inner salt and an electron
coupling reagent (phenazine ethosulfate, PES), was obtained from
Promega. DMEM high glucose culture media with sodium
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2011.01.014.
pyruvate, DMEM-F12 culture media, trypsin-EDTA, L-glutamine,
References and notes
fetal bovine serum (FBS), human transferin, bovine insulin, and
phosphate-buffered saline (PBS, calcium and magnesium free), a
solution of penicillin and streptomycin were purchased from Invit-
rogen Corp. Human albumin was purchased from Sigma. Dauno-
mycin and Doxorubicin were obtained from IFFECT ChemPhar
(Hong Kong) Company Limited.
1. Wishart, D. S.; Knox, C.; Guo, A. C.; Cheng, D.; Shrivastava, S.; Tzur, D.; Gautam,
B.; Hassanali, M. Nucleic Acids Res. 2008, 36, D901.
2. Viegas-Junior, C.; Danuello, A.; da Silva Bolzani, V.; Barreiro, E. J.; Fraga, C. A. M.
Curr. Med. Chem. 2007, 14, 1829.
3. Shi, W.; Coleman, R. S.; Lowary, T. L. Org. Biomol. Chem. 2009, 7, 3709.
4. Shi, W.; Marcus, S. L.; Lowary, T. L. Carbohydr. Res. 2010, 345, 10.

W. Shi, T. L. Lowary / Bioorg. Med. Chem. 19 (2011) 1779–1789
1789
5. Wang, A. H.; Ughetto, G.; Quigley, G. J.; Rich, A. Biochemistry 1987, 26,
1152.
9. Boger, D. L.; Fink, B. E.; Brunette, S. R.; Tse, W. C.; Hedrick, M. P. J. Am. Chem. Soc.
2001, 123, 5878.
6. Shi, W.; Marcus, S. L.; Lowary, T. L. Bioorg. Med. Chem. 2011, 19, 603.
7. Di Marco, A.; Casazza, A. M.; Gambetta, R.; Supino, R.; Zunino, F. Cancer. Res.
1976, 36, 1962.
10. Bartulewicz, D.; Bielawski, K.; Bielawska, A. Arch. Pharm. (Weinheim) 2002, 335,
422.
11. Silverstein, R. M.; Webster, F. X. Spectrometric Identification of Organic
Compounds; John Wiley & Sons, 1998.
8. Britt, M.; Zunino, F.; Chaires, J. B. Mol. Pharmacol. 1986, 29, 74.