The structure-based design, synthesis, and biological evaluation of DNA-binding amide linked bisintercalating bisanthrapyrazole anticancer compounds

Article abstract of DOI:10.1016/j.bmc.2009.04.072

A series of amide-coupled bisanthrapyrazole derivatives of 7-chloro-2-[2-[(2-hydroxyethyl)methylamino]ethyl]anthra[1,9-cd]pyrazol-6(2H)-one (AP9) were designed using molecular modeling and docking and synthesized in order to develop an anticancer drug tha

Full text of DOI:10.1016/j.bmc.2009.04.072

Bioorganic & Medicinal Chemistry 17 (2009) 4575–4582
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
The structure-based design, synthesis, and biological evaluation
of DNA-binding amide linked bisintercalating bisanthrapyrazole
anticancer compounds
a
a
b
b
Brian B. Hasinoff a, , Rui Zhang , Xing Wu , Lynn J. Guziec , Frank S. Guziec Jr. ,
*
Kyle Marshall b, Jack C. Yalowich c
a Faculty of Pharmacy, University of Manitoba, 750 McDermot Avenue, Winnipeg, Manitoba, Canada R3E 0T5
b Department of Chemistry and Biochemistry, Southwestern University, Georgetown, TX 78628, USA
c Department of Pharmacology and Chemical Biology, University of Pittsburgh School of Medicine, Pittsburgh, PA 15261, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of amide-coupled bisanthrapyrazole derivatives of 7-chloro-2-[2-[(2-hydroxyethyl)methyl-
Received 25 February 2009
Revised 28 April 2009
Accepted 30 April 2009
Available online 6 May 2009
amino]ethyl]anthra[1,9-cd]pyrazol-6(2H)-one (AP9) were designed using molecular modeling and dock-
ing and synthesized in order to develop an anticancer drug that formed
a strongly binding
bisintercalation complex with DNA. Concentration dependency for the increase in the DNA melting tem-
perature was used to determine the DNA binding strength and whether bisintercalation occurred for the
newly synthesized analogs. The ability of the compounds to inhibit the growth of the human erythroleu-
Keywords:
kemic K562 cell line and inhibit the decatenation activity of DNA topoisomerase II
Finally, the compounds were evaluated for their ability to act as topoisomerase II poisons by measuring
the topoisomerase II -mediated double strand cleavage of DNA. All of the bisanthrapyrazoles inhibited
K562 cell growth and topoisomerase II in the low micromolar range. Compounds with either two or
a was also measured.
Bisanthrapyrazole
Bisintercalating
DNA-binding
a
a
Topoisomerase II
three methylene linkers formed bisintercalation complexes with DNA and bound as strongly as, or more
strongly than, doxorubicin. In conclusion, a novel group of amide-coupled bisintercalating anthrapyraz-
ole compounds were designed, synthesized, and evaluated for their physico-chemical and biologic prop-
erties as potential anticancer agents.
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
reductively activated. We previously showed3 in a structure-based
3D-QSAR study that a series of substituted anthrapyrazoles that we
Some of the most efficacious anticancer drugs bind to DNA and
in doing so inhibit DNA topoisomerase II (EC 5.99.1.3) which re-
sults in cell growth inhibition and/or cell death. Included among
these drugs are the anthracyclines such as doxorubicin and dauno-
rubicin, and amsacrine, and mitoxantrone.1 These planar aromatic
molecules intercalate into DNA and promote DNA strand breaks
through their interaction with DNA topoisomerase II.1 Because
the clinical use of doxorubicin and the other anthracyclines is lim-
ited by the development of a cumulative dose-dependent and
potentially fatal cardiotoxicity, which is likely due to oxidative
stress on the relatively unprotected cardiac muscle, non-cardiotox-
ic topoisomerase II-targeted anticancer drugs are urgently needed.
Because the anthracyclines and mitoxantrone are quinones they
are susceptible to reductive activation2 by a variety of reducing en-
zymes to produce damaging reactive oxygen species leading to car-
diotoxicity. The anthrapyrazoles are not quinones and cannot be
synthesized had potent cancer cell growth inhibitory effects. Fol-
lowing this we used molecular modeling to design and synthesize
a series of bisanthrapyrazoles that contained a diester linkage join-
ing two anthrapyrazoles together.4 However, the biological data on
these bisanthrapyrazoles suggested that the ester linkages were
susceptible to hydrolysis, possibly through the action of cellular
esterases.4 We showed that these bisanthrapyrazoles had good cell
growth inhibitory activity and the compounds with two or more
methylene linkers formed bisintercalation complexes with DNA.4
In the present work we attempted to improve the stability of these
compounds by replacing the ester linkage with the much more sta-
ble amide linkage (Fig. 1, compounds 1–5).
We previously showed that the strength of DNA binding for the
anthrapyrazoles was strongly correlated with cell growth inhibi-
tion.3 Bisintercalators can potentially bind much more strongly than
their analogous monointercalators because of the favorable entropic
advantage that ensues after the binding of the first intercalating
group. Because of their strong binding they should have a prolonged
residence time on DNA5 allowing them to interfere with DNA pro-
* Corresponding author. Tel.: +1 204 474 8325; fax: +1 204 474 7617.
E-mail address: B_Hasinoff@UManitoba.ca (B.B. Hasinoff).
0968-0896/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2009.04.072

4576
B. B. Hasinoff et al. / Bioorg. Med. Chem. 17 (2009) 4575–4582
Figure 1. Reaction scheme and structures for the synthesis of the bisanthrapyrazoles 1, 2, 3, 4, and 5 (n = 1–5) starting from 6 (AP9). Compounds 1, 2, and 4 were prepared
using Method A and compounds 3 and 5 were prepared using Method B as described in Section 4.
cessing enzymes. In principle, bisintercalators also should have en-
hanced DNA sequence specificity compared to monointercalators
because monointercalators have the potential of binding to only 10
distinguishable sites, but for a binding site of 6 base pairs this in-
creases to 2080 distinguishable sites.6 However, it should be noted,
because the majority of intercalators, at best, show only a slight pref-
erence for GCbase-pairs, sequencespecificitymaynot necessarilybe
achieved by increasing the number of distinguishable sites. Bisinter-
calatingcompounds, asexemplifiedbyechinomycin, occurnaturally
and it and several purely synthetic compounds have progressed into
clinical trials as antitumor agents.5,7,8 In order to direct the synthesis
of new bisanthrapyrazoles, molecular modeling and DNA docking
studies were carried out on the amide-linked bisanthrapyrazoles
containing different numbers of methylene linkers. Compound 6
(Fig. 1), which we previously had shown to have significant cell
ature, of sonicated calf thymus DNA (5
1. Doxorubicin (2 M), which is a well known DNA intercalating
drug, was used as a control and was observed to increase the Tm
by 17.8 °C from 70.2 °C. Compound 2 had the largest Tm, and thus
bound the strongest to DNA (Table 1 and Fig. 2). Compounds 2 and
3 both bound to DNA as or more strongly than doxorubicin, a well
known DNA intercalator.9 The bisanthrapyrazoles 1, 4, and 5 all in-
lg/mL) are shown in Table
l
D
creased
on the analogous ester-linked bisanthrapyrazoles4 we showed that
the slope of a plot of Tm versus drug concentration for a bisinter-
DTm about the same as the parent 6. In our previous study
D
calator is approximately twice that of the monomeric monointer-
calator 6. This is due to the bisintercalator occupying twice the
number of intercalation sites on the DNA as had been previously
shown from a comparison of the bisintercalator dye YOYO and
its monomeric form YO-PRO.10 This is based on a simple model
in which (1) the different intercalation sites contribute equally
and independently to the free energy difference between the
hybridized and melted states and (2) the presence of an intercala-
tor at a site lowers the free energy of the hybridized state by a
fixed, site-independent amount. Thus, the melting temperature of
a DNA duplex should increase in proportion to the number of inter-
growth and topoisomerase IIa
inhibitory activity,3,4 was chosen as
the scaffold for the synthesis of a series of the bisanthrapyrazole
congeners. In addition to measuring the cell growth inhibitory ef-
fectsof thebisanthrapyrazoleswealsomeasuredtheirabilitytobind
toDNA, to inhibittopoisomeraseII
mediated DNA cleavage.
a, andtocausetopoisomeraseIIa-
calated groups. The concentration dependence of
DTm for mono-
meric 6 (AP9) and the bisanthrapyrazoles 1–5 are shown in
Figure 2A and the slopes of these plots are given in Figure 2B. A
t-test comparison of the slopes showed that compounds 1, 2, and
3 had slopes that were significantly different (p <0.01, <0.001,
and <0.001, respectively) than monomer 6. Compounds 4 and 5
did not yield slopes that were significantly different than monomer
6. The ratio of the slopes of compounds 1–5 relative to that for 6
gave values of 1.7 0.3, 4.0 0.8, 3.4 0.8, 1.2 0.2, and 0.9 0.2
2. Results
2.1. Effect of the bisanthrapyrazoles on the thermal
denaturation of DNA
The effects of 2
tercalator 6 on
l
M of the bisintercalators 1–5 and the monoin-
DTm, the increase in the DNA thermal melt temper-
Table 1
DNA
D
Tm, cell growth inhibition and topoisomerase II
a
inhibitory effects of the amide-linked bisanthrapyrazoles and their precursors
Resistance factorb
Topoisomerase IIa inhibition
a
Compound
D
Tm (°C)
MTS cell growth inhibition
K562 IC50
(lM)
K/VP.5 IC50
(lM)
IC50 (lM)
1
2
3
4
9.6
21.8
17.8
6.8
4.8
5.4
1.1
10.2
17.8
2.6
3.5
5.7
6.7
4.6
3.1
3.6
3.6
0.3
2.0
1.6
2.3
3.3
2.5
1.9
ND
1.5
0.5
0.77
0.44
0.39
0.48
0.55
0.63
ND
42
7
25
48
22
10
25
18
ND
5
6, AP9
8
9
0.43
1.81
Doxorubicin
ND is not determined.
a
Experimental value at 2
l
M.
b
The resistance factor was calculated from the ratio of the IC50 value for the K/VP.5 cell line divided by that for the K562 cell line.

B. B. Hasinoff et al. / Bioorg. Med. Chem. 17 (2009) 4575–4582
4577
Figure 2. Concentration dependence of
D
Tm for the bisanthrapyrazoles binding to DNA. (A) Concentration dependence of
D
Tm for the bisanthrapyrazoles 1, 2, 3, 4, and 5
(n = 1–5) and parent monomer 6 for comparison. The solid straight lines are linear least squares calculated fits to the data. (B) Concentration dependence of the slopes SEM
*
**
calculated from the data in (A) (NS, not significant). The slopes of the plots in (A) for 1, 2, and 3 were all significantly different ( , ; p <0.01, <0.001, respectively) than the
parent 6. Compounds with slopes approximately twice that of the parent monomer 6 indicate that they formed bisintercalation complexes with DNA.
where the SEMs were calculated from a propagation of errors anal-
ysis. The data were consistent with compounds 2 and 3 forming
bisintercalating complexes. For compound 1 the data suggest a
possible formation of a mixture of mono and bisintercalating com-
plexes. This result differed from our previous study on the ester-
linked bisanthrapyrazoles4 which showed that the compound with
a single methylene linker did not bisintercalate into DNA. Com-
linear DNA comparable to that induced by etoposide (lane 10). If
the bisanthrapyrazoles poison topoisomerase IIa as we found with
some of the anthrapyrazoles of our previous study,3 they do so at a
level much lower than that of etoposide. As indicated by the ab-
sence of a relaxed DNA band, etoposide and all 5 bisanthrapyraz-
oles inhibited the relaxation of supercoiled pBR322 DNA by
topoisomerase II
both inhibited topoisomerase II
DNA cleavage, the bisanthrapyrazoles acted primarily by inhibition
of topoisomerase II strand passage reactions. The ester-linked bis-
a. These results indicate that, while etoposide
pounds
4
and
5
gave ratios close to
1
indicating mono
a
catalytic activity and induced
intercalation.
a
2.2. Effect of bisanthrapyrazoles on the decatenation activity of
anthrapyrazoles of our previous study also did not induce signifi-
topoisomerase II
a
a
and on the stabilization of the covalent
–DNA cleavable complex
cant DNA cleavage.4
topoisomerase II
2.3. Cell growth inhibitory effects of the bisanthrapyrazoles on
human leukemia K562 cells and K/VP.5 cells with a decreased
Topoisomerase II alters DNA topology by catalyzing the passing
of an intact DNA double helix through a transient double-stranded
break made in a second helix and is critical for relieving torsional
stress that occurs during replication and transcription and for
daughter strand separation during mitosis.11,12 As shown in Table
1 all of the bisanthrapyrazoles inhibited the decatenation activity
level of topoisomerase II
a
As shown in Table 1 all of the bisanthrapyrazoles potently
inhibited the growth of K562 and K/VP.5 cells in the low micromo-
lar range. These results suggest that the bisanthrapyrazoles were
not too large or too charged to enter cells. In general, changing
of human topoisomerase IIa in the low micromolar concentration
range. However, only 2, which was the compound that bound the
strongest to DNA, achieved a cytotoxic potency greater than that
of the parent 6. The decatenation assay is a measure of the ability
of these compounds to inhibit the catalytic activity only, and is not
a measure of whether these compounds acted as topoisomerase II
poisons as do some widely used anticancer drugs.11,12 These in-
clude doxorubicin and the other anthracyclines, mitoxantrone,
and etoposide.11,12 These and the anthrapyrazoles losoxantrone
and piroxantrone,13,14 are thought to be cytotoxic by virtue of their
ability to stabilize a covalent topoisomerase II–DNA intermediate
(the cleavable complex). Stabilization of the covalent complex
leads to double strand DNA breaks that are toxic to the cell. Thus,
DNA cleavage assay experiments15 as we previously described3,4,16
Figure 3. Effect of bisanthrapyrazoles 1, 2, 3, 4, and
precursors 6 and 9 on the topoisomerase II -mediated relaxation and cleavage of
50 ng of supercoiled pBR322 plasmid DNA. This fluorescent image of the ethidium
bromide-stained gel shows that topoisomerase II (Topo II ) relaxed supercoiled
(SC) pBR322 DNA (lane 1) to relaxed (RLX) DNA (lane 2). Topoisomerase II was
5 and their monomeric
were carried out using 50
whether 50 M of the test bisanthrapyrazoles stabilized the cleav-
able complex. As shown in Figure 3 the addition of etoposide (lane
10) to the reaction mixture containing topoisomerase II and
lM etoposide as a positive control to see
a
l
a
a
a
a
present in the reaction mixture for all lanes but lane 1. As shown in lane 10,
etoposide treatment produced significant amounts of linear DNA (LIN). A small
supercoiled pBR322 DNA induced formation of linear pBR322
DNA. Linear DNA was identified by comparison with linear
pBR322 DNA produced by action of the restriction enzyme HindIII
acting on a single site on pBR322 DNA (not shown). Based on inte-
grated band intensities of linear DNA in Figure 3 none of the bis-
anthrapyrazoles (lanes 3–7) appeared to induce formation of
amount of nicked circular (NC) is normally present in the pBR322 DNA.
A
densitometric analysis of the linear DNA bands showed that all of the bisanthra-
pyrazoles produced much less linear DNA than the etoposide positive control (lane
10) did. The amount of linear DNA produced by the bisanthrapyrazoles was only
slightly above control levels (lane 2). The concentration of all test compounds in the
assay mixture was 50 lM.

4578
B. B. Hasinoff et al. / Bioorg. Med. Chem. 17 (2009) 4575–4582
the number of methylene linkers did not produce a large effect on
the inhibition of the growth of K562 cells as the IC50 values for
compounds 1–5 varied only 2.6-fold.
Cancer cells can acquire resistance to topoisomerase II poisons
by lowering their level and/or activity of topoisomerase II.11,17 Cells
containing less topoisomerase II produce fewer DNA strand breaks
in the presence of topoisomerase II poisons and are less cytotoxic.
Thus, a resistant cell line may be used to test whether a drug that
catalytically inhibits topoisomerase II might also act as a topoiso-
merase II poison.18 Conversely, a lack of change in sensitivity of a
putative topoisomerase II poison to a cell line with a lowered topo-
isomerase II level can be taken to indicate that poisoning of topo-
isomerase II is not an important mechanism for a particular agent.
We previously reported that topoisomerase IIa and topoisomerase
IIb protein levels were reduced sixfold and threefold, respectively,
in the K/VP.5 cell line with acquired resistance to etoposide com-
pared with the parental K562 cells.19,20 As shown in Table 1 none
of the bisanthrapyrazoles were cross resistant in K/VP.5 cells. The
topoisomerase II poisons doxorubicin and mitoxantrone were
1.8–4.2-fold cross resistant, respectively. Thus both these results
and the results of the cleavage assay (Fig. 3) suggest that the bis-
anthrapyrazoles do not inhibit cell growth by acting as topoiso-
merase II poisons.
2.4. Docking of the bisanthrapyrazoles into DNA
All five bisanthrapyrazoles docked into the DNA minor groove
with similar configurations. All five also docked with both anthra-
pyrazole rings fully inserted into the doxorubicin intercalation
sites with their aromatic rings highly coplanar with the DNA bases.
An X-ray structure (PDB ID: 1DA9)9 of two molecules of doxorubi-
cin separated by 4 base pairs bound to duplex DNA was used for
the docking (Fig. 4). The structure of 2, the bisanthrapyrazole that
bound to DNA the strongest, docked into DNA is shown in Figure 4.
An examination of the structure of 5 docked into DNA showed that
the linker chain was longer than required which resulted in a slight
bulge in the chain. The GOLDScores obtained from the docking are
plotted as a function of n, the number of methylene linkers in
Figure 5A. The bell-shaped plot showed a broad maximum with
an n value in the 2–4 range suggesting that bisintercalation bind-
ing to DNA is predicted to be optimal in this range. The maximum
value obtained from a quadratic fit (solid line) to the data indicated
that an n value of 3.7 was optimal for binding to DNA. The exper-
imentally determined strength of binding to DNA determined
Figure 4. Docking of the protonated bisanthrapyrazole 2 into DNA. The highest
scoring structure of the strongest DNA binding bisanthrapyrazole 2 (CPK structure)
is shown docked into DNA (stick structure). The H-atoms of the DNA are not shown
for clarity. The DNA structure is 1DA9 from the Protein Data Bank and is a DNA-
(doxorubicin)2 X-ray structure in which two doxorubicin molecules are bound to a
6-base pair piece of DNA.9 Both doxorubicin molecules were removed and 2 was
docked into its place with the genetic algorithm docking program GOLD
.
or any of its component terms. The logarithm of the K562 IC50 val-
ues versus Tm were also not correlated (p of 0.9). In our previous
study of a series of anthrapyrazoles that we synthesized, the loga-
D
rithm of the K562 IC50 values and
DTm were highly correlated (p
<0.0001).3 The logarithm of the K562 IC50 data was also poorly cor-
related with the logarithm of the IC50 for the catalytic inhibition of
topoisomerase II
tion of topoisomerase II
a
(p = 0.7). The lack of correlation with the inhibi-
activity IC50 does not necessarily indicate
a
that these compounds lack activity as topoisomerase II poisons,
rather they do not act through their inhibition of the catalytic
using
DTm as a measure is also shown plotted in Figure 5B as a
function of n, the number of methylene linkers. The maximum in
the quadratic fit (solid line) to the data in Figure 5B indicated that
an n value of 2.6 was optimal for binding to DNA. Thus, the optimal
number of methylene linkers calculated from the GOLD docking and
the experimental number were in reasonably good agreement
indicating that bisanthrapyrazole GOLD docking scores were a rea-
sonable predictor of optimal linker length for strength of binding
to DNA.
activity of topoisomerase II
a. Similarly, the logarithm of the IC50
for the catalytic inhibition of topoisomerase II
a
was also poorly
(p of 0.14) correlated with
DTm. The lack of correlation is generally
similar to what we found for our ester-linked bisanthrapyrazoles.4
3. Discussion
A series of amide-linked bisanthrapyrazoles based on the
monomeric scaffold 6 with 1–5 methylene linkers, was synthesized
in order to improve the potency and the stability of an earlier anal-
ogous series of ester-linked bisanthrapyrazoles we synthesized and
characterized.4 In general, both neither K562 growth inhibitory ef-
2.5. QSAR correlation analyses of GOLDScore, DNA binding
affinity, K562 cell growth inhibition, and topoisomerase II
inhibition
a
A correlation analysis was carried out on the logarithm of the
K562 IC50 values versus the GOLDScores and each of its component
terms: (DNA–ligand hydrogen bond energy (external H-bond);
DNA–ligand van der Waals (vdw) energy (external vdw); and the
sum of the two ligand internal (vdw and torsion) terms as we pre-
viously described for a series of anthrapyrazoles.3 No significant
correlation was found (all p values greater than 0.17). Likewise
fects nor inhibition of topoisomerase IIa of the amide-linked bis-
anthrapyrazoles were significantly changed compared to the
previous ester-linked bisanthrapyrazoles.4 As before4 molecular
modeling and docking into DNA was used in order to determine
the optimal linker length for optimal binding to DNA and selection
for their subsequent synthesis. The concentration dependence of
D
Tm (Fig. 2B) for 2 and 3 were highly significantly increased over
D
Tm values were not significantly correlated with the GOLDScores
that of the parent 6, a result which suggests that these two bis-

B. B. Hasinoff et al. / Bioorg. Med. Chem. 17 (2009) 4575–4582
4579
Figure 5. Calculated and experimental measures of binding of the bisanthrapyrazoles 1–5 binding to DNA. (A) GOLDScores obtained from the docking of the
bisanthrapyrazoles into DNA as a function of n, the number of CH2 linker groups in the bisanthrapyrazoles. The solid line is a quadratic fit of the GOLDScore values for the
bisanthrapyrazoles with n = 1–5. (B) Experimentally measured
DTm values for the bisanthrapyrazoles with n = 1–5 and parent 6 (AP9) for comparison. The solid line is a
quadratic fit of Tm (solid line) for the bisanthrapyrazoles with n = 1–5.
D
anthrapyrazoles exclusively formed bisintercalation complexes
with DNA. However, our docking results predicted that all of the
compounds 1–5 should be able form bisintercalation complexes
spanning 4 DNA base pairs. GOLDScores obtained from docking
compounds 1–5 was not correlated with inhibition of topoisomer-
ase II -mediated catalytic decatenation activity. This result indi-
cates that the catalytic inhibition of topoisomerase II was not the
primary mechanism by which these compounds inhibited cell
growth. Compounds 1–5 did not significantly induce formation of
a
and the experimentally determined
DTm values both displayed
broad maxima (Fig. 5) in the range of 2–4 methylene linkers. Thus,
the calculated and experimental DNA binding affinities were in
good agreement indicating that the docking was useful for select-
ing a range of optimal methylene linker lengths for synthesis.
Though the bisanthrapyrazoles differed significantly in their ability
to bind to DNA, this difference was not reflected in their ability to
inhibit the growth of K562 cells. Thus, it can be concluded that
while the design goal of increasing strength of DNA binding com-
pared to monomer 6 through bisintercalation was achieved, the
goal of increasing cell growth inhibition by changing the ester link-
age to an amide linkage did not result in a significant improve-
ment. Though the in vitro stability of the amide-linked
bisanthrapyrazoles was not determined, it is likely that they are
much more stable than the ester-linked bisanthrapyrazoles we
previously described which should make the amide-linked bis-
anthrapyrazoles more efficacious as anticancer drugs.4 Future
work will evaluate the stability, pharmacokinetics and antitumor
efficacy of ester-linked bisanthrapyrazoles compared to the
amide-linked bisanthrapyrazoles using a xenograft rodent model
for human tumors.
linear DNA through stabilization of a covalent topoisomerase IIa–
DNA cleavable complex in the DNA cleavage assay. These results
are in contrast to our previously described4 results that showed
the ester-linked bisanthrapyrazoles did have some topoisomerase
II poisoning activity. Also, several of the monomeric anthrapyraz-
oles we previously studied3 did induce formation of linear DNA.
The bisanthrapyrazoles may also target other DNA processing en-
zymes due to their ability to bisintercalate into DNA.
In summary, a series of amide-linked bisanthrapyrazoles were
designed to bisintercalate into DNA using molecular modeling
and docking. As predicted from the docking results, bisanthrapy-
razoles with more than one methylene linker formed bisintercala-
tion complexes with DNA and significantly inhibited K562 cell
growth.
4. Experimental
4.1. Biological assays
4.1.1. Materials, cell culture, and growth inhibition assays
The plasmid DNA and the absorbance-based cell proliferation
assay and other materials were as previously described.4 The hu-
man leukemia K562 cells, obtained from the American Type Cul-
ture Collection and K/VP.5 cells (a 26-fold etoposide-resistant
The bisanthrapyrazoles 1–5 have a large number of rotatable
bonds. The high degree of conformational flexibility of these mol-
ecules makes it computationally difficult to obtain the best possi-
ble GOLDScore. This likely explains the lack of a smooth change
in GOLDScore as the number of linker groups was varied
(Fig. 5A). Even so, the docking produced structurally reasonable
bisintercalation complexes as shown by the example in Figure 4.
Most of the driving force for binding is likely due to the high affin-
ity that the anthrapyrazole rings have for the sites into which the
bisanthrapyrazoles intercalate, thus anchoring the two ends of the
bisanthrapyrazoles. DNA base pair specificity in bisanthrapyrazole
binding to DNA may be another factor that contributes to GOLD-
Scores lacking the power to accurately predict the strength of
DNA binding. The bisanthrapyrazoles were docked into a single
X-ray structure (PDB ID: 1DA9) which does not necessarily have
the base pair sets which would lead to optimum binding to DNA.
Our previous study3,4 showed that 6 and the ester-linked bis-
K562-derived sub-line with decreased levels of topoisomerase II
a
mRNA and protein)17 were maintained as suspension cultures as
described.4 The IC50 values for growth inhibition in the cell growth
inhibition and decatenation assays were measured by fitting the
absorbance- or fluorescence-drug concentration data, respectively,
to a four-parameter logistic equation as described.4 Errors, where
quoted, are SEMs.
4.1.2. Topoisomerase IIa kDNA decatenation and pBR322 DNA
relaxation and cleavage assays
The fluorescence plate reader based spectrofluorometric decat-
enation assay we developed was used to determine the catalytic
inhibition of topoisomerase IIa 4,18
.
kDNA which consists of highly
anthrapyrazoles4 were strong topoisomerase II
the ester-linked bisanthrapyrazoles4 cell growth inhibition by
a inhibitors. As with
catenated networks of circular DNA is decatenated by topoisomer-

4580
B. B. Hasinoff et al. / Bioorg. Med. Chem. 17 (2009) 4575–4582
ase II
a
in an ATP-dependent reaction to yield individual minicircles
75.5 MHz for 13C NMR, respectively, in chloroform-d, unless other-
wise indicated. Chemical shifts are given in parts per million
(ppm) (+/ꢀ 0.01 ppm) relative to tetramethylsilane (0.00 ppm) in
the case of the 1H NMR spectra, and to the central line of CDCl3 (d
77.0) for the 13C NMR spectra. Melting points were taken on an Elec-
trothermal (England) melting point apparatus and were uncor-
rected. The high resolution mass spectra were run on an Agilent
6210 Accurate-Mass Time-of-Flight (TOF) LC/MS system (Agilent
Technologies, Mississauga, Canada) using electrospray ionization.
The samples were dissolved in methanol and were infused into the
mass spectrometer with an Agilent 1200 HPLC using acetonitrile/
water containing 0.1% (v/v) formic acid as the mobile phase. The sil-
ica gel (SiGel, 230–400 mesh ultrapure) was obtained from Silicycle
(Quebec, Canada). The alumina (aluminum oxide, activated, basic,
Brockmann I, standard grade, ꢁ150 mesh) was obtained from Fisher
(Ottawa, Canada). TLC was performed on plastic-backed plates bear-
of DNA. The assay conditions and the expression, extraction and
purification of recombinant full-length human topoisomerase II
a
were previously described.4,18 Topoisomerase II-cleaved DNA com-
plexes produced by anticancer drugs may be trapped by rapidly
denaturing the complexed enzyme with sodium dodecyl sulfate
(SDS).3,15 The drug-induced cleavage of double-stranded closed cir-
cular pBR322 DNA to form linear DNA at 37 °C was followed by
separating the SDS-treated reaction products using ethidium bro-
mide gel electrophoresis essentially as described, except that all
components of the assay mixture were assembled and mixed on
ice prior to addition of the drug.4,15 This greatly increased the
amount of linear DNA produced, and thus increased the sensitivity
of the assay.
4.1.3. Thermal denaturation of DNA assay
Compounds that intercalate into DNA stabilize the DNA double
helix and increase the temperature at which the DNA denatures or
ing 200 lm Silica Gel 60 F254 (Silicycle). Compounds were visualized
by quenching of fluorescence by UV light (254 nm) where applica-
ble. The reaction conditions were not optimized for reaction yields.
Dichloromethane and triethylamine were refluxed over calcium hy-
dride and distilled. Compound 6 (AP9) (7-chloro-2-[2-[(2-hydroxy-
ethyl)methylamino]ethyl]anthra[1,9-cd]pyrazol-6(2H)-one) was
prepared from 1,5-dichloroanthraquinone as we previously
described.3
unwinds.21–23 The effect of 0.1–2
l
M anthrapyrazole or bisanthra-
Tm of sonicated calf thymus DNA
g/mL) was measured in 10 mM Tris–HCl buffer (pH 7.4) in a
pyrazole on the increase in the
(5
D
l
Cary 1 (Varian, Mississauga, Canada) double beam spectrophotom-
eter by measuring the absorbance increase at 260 nm upon the
application of a temperature ramp of 1 °C/min as we previously de-
scribed.3,4 Doxorubicin (2
lM), which is a strong DNA intercalator,
was used as a positive control.3,4 The slopes of the plots for mono-
mer 6 were compared to those of the bisanthrapyrazoles 1–5 using
a t-test comparison of the slopes.24
4.3.2. General procedure for the synthesis of the
bisanthrapyrazoles
The amide-linked bisanthrapyrazoles 1, 2, and 4 were prepared
by the reaction of 9 with the corresponding acid chlorides in
dichloromethane. Compounds 3 and 5 were prepared by the reac-
tion of 9 with the corresponding bis-N-hydroxysuccinimide esters
in dichloromethane. Method A for the synthesis of 1, 2, and 4 was
as follows. The acid chlorides (0.28 mmol, 0.5 equiv) were added
slowly to a clear orange-red solution of 9 (200 mg, 0.564 mmol)
in dichloromethane (3 mL). A light orange precipitate formed
immediately. Ether (3 mL) was added with overnight stirring. The
mixture was centrifuged and the solid was twice washed with
ether and dried. The solid was extracted into chloroform/aqueous
NaOH and dried over Na2SO4. The solvent was removed with a vac-
uum pump. We found that the reaction of 9 with the acid chlorides
gave low yields because the HCl produced in the reaction caused 9
to precipitate as the HCl salt. It was for this reason that 3 and 5
were prepared using Method B. Method B for the synthesis of 3
and 5 was as follows. N-Hydroxysuccinimide ester 10 or 11
(0.14 mmol, 0.5 equiv) was treated with 9 (100 mg, 0.282 mmol)
in dichloromethane (5 mL). The mixture was stirred overnight
and the solvent was removed under reduced pressure. In general,
the free base bisanthrapyrazoles formed sticky films. These could
be precipitated into easily handled powders by dissolving the oily
amides in dichloromethane and then adding hexane.
4.2. Molecular modeling and docking
4.2.1. Docking of the bisanthrapyrazoles into an X-ray structure
of DNA
All molecular modeling was done using SYBYL 7.325 on a PC work-
station with a Redhat Enterprise 4 Linux operating system. All mol-
ecules except the DNA were built using SYBYL and geometry
optimized with the Tripos force field. The protonated bisanthrapy-
razoles were docked into the doxorubicin binding site of a 6-base
pair X-ray crystal structure of 2 molecules of doxorubicin bound
to double stranded DNA, d(TGGCCA)/doxorubicin, (http://
www.rcsb.org/pdb/; PDB ID: 1DA9)9 using the genetic algorithm
docking program GOLD version 3.226 with default GOLD parameters
and atom types and with 500 starting runs27 as we described.4
GOLDScore was used as the fitness function with flipping options
of planar and pyramidal nitrogens being allowed. The top 10 struc-
tures were then re-scored using a local optimization (simplexing)
to obtain the final GOLDScore. Rescoring did not affect the initial
relative scoring order. The two doxorubicin molecules in the
1DA9 DNA X-ray structure are separated by four intervening base
pairs. The first and second base pairs buckle out to accommodate
the bound doxorubicin in this structure.9 The DNA structure was
prepared by removing both of the bound doxorubicin and water
molecules to avoid potential interference with the docking. Hydro-
gens were added to the DNA with the SYBYL Biopolymer module. The
binding site (6 Å) was defined using a previously docked ester-
linked bisanthrapyrazole in the minor grove.4 Doxorubicin docked
back into the DNA structure with a heavy atom root-mean-squared
distance of 1.3 Å compared to the X-ray structure.9 The graphic
was prepared with DS Visualizer 2.0 (Accelrys, San Diego, CA).
4.3.2.1. 7-Chloro-2-[2-[(2-chloroethyl)methyl-
amino]ethyl]anthra[1,9-cd]pyrazol-6(2H)-one (7). Mesyl chlo-
ride (1.74 mL, 22.5 mmol) and triethylamine (3.13 mL,
22.5 mmol) were added to a solution of 6 (AP9, 2.0 g, 5.62 mmol)
in chloroform (30 mL) on ice. The mixture was stirred overnight
under nitrogen. The product was extracted into chloroform/aque-
ous sodium hydroxide, and the chloroform phase was dried over
Na2SO4. The solvent was removed, affording crude 7, 2.07 g as yel-
low crystals. This reaction afforded the chloroethyl derivative
rather than the expected mesylate, presumably via a reactive aziri-
dine intermediate. The crude product was purified by silica gel col-
umn chromatography (ethyl acetate as eluant), affording 1.82 g
(0.487 mmol) in 86% yield; mp: 144–146 °C; 1H NMR (CDCl3) d
8.20 (dd, 1H); 7.99 (d, 1H); 7.73–7.53 (m, 4H); 4.61 (t, 2H); 3.43
(t, 2H); 3.09 (t, 2H); 2.79 (t, 2H); 2.39 (s, 3H). 13C NMR (CDCl3) d
4.3. Chemistry
4.3.1. General
1H and 13C nuclear magnetic resonance (NMR) spectra were re-
corded at 300 K in 5 mm NMR tubes on a Bruker Avance 300 MHz
NMR spectrometer operating at 300.13 MHz for 1H NMR and

B. B. Hasinoff et al. / Bioorg. Med. Chem. 17 (2009) 4575–4582
4581
ꢁ183; 139.3; 137.2; 134.7; 132.9; 132.6; 129.1; 128.7; 126.9;
121.8; 121.0; 114.9; 59.2; 57.1; 48.4; 42.4; 41.2; HRMS (ESI), m/z
(M+H)+: calcd 374.0827, obsd 374.0819. The chloride 7 was quite
reactive and reacted smoothly with sodium azide affording the
azide 8 in high yield.
ArH), 6.85 (t, J = 4.6 Hz, 2H, amide NH), 4.54 (t, 4H, J = 6.3 Hz, NNCH2),
3.18 (q, 4H, J = 5.5 Hz), 2.98 (t, J = 6.3 Hz, 4H, NNCH2CH2), 2.68 (s, 2H
COCH2CO), 2.52 (t, J = 5.5 Hz, 4H), 2.30 (s, 6H, 2 ꢂ NCH3); 13C NMR d
182.4, 166.8, 139.1, 138.1, 137.1, 134.5, 132.9, 132.5, 128.9, 128.6,
126.8, 122.9, 121.8, 120.9, 114.7, 56.9, 56.2, 48.3, 42.7, 42.2, 36.9;
HRMS (ESI), m/z (M+H)+: calcd 777.2471, obsd 777.2468.
4.3.2.2. 7-Chloro-[2-[(2-azidoethyl)methylamino]ethyl]anthra-
[1,9-cd]pyrazol-6(2H)-one (8). Compound 7 (1.10 g, 2.54 mmol)
was added to sodium azide (1.65 g, 25.4 mmol) in dimethylform-
amide (10 mL) and the mixture was heated at 60 °C under nitrogen
for 24 h. The solution was cooled to room temperature, added to
water (100 mL), and stirred for 30 min. The precipitate was filtered
affording 0.88 g (2.31 mmol) in 91% yield; mp: 101–102 °C; 1H
NMR (CDCl3) d 8.20 (dd, 1H); 8.00 (d, 1H); 7.74–7.52 (m, 4H);
4.62 (t, 2H); 3.23 (t, 2H); 3.08 (t, 2H); 2.66 (t, 2H); 2.39 (s, 3H).
13C NMR (CDCl3) d 182.6; 139.3; 137.2; 134.7; 132.9; 132.5;
128.6; 126.9; 122.7; 121.8; 120.9; 114.7; 57.2; 56.8; 48.9; 48.5;
42.4; HRMS (ESI), m/z (M+H)+: calcd 381.1230, obsd 381.1220.
The product was pure enough to be used directly in the next step.
4.3.2.7. N,N0-Bis(2-((2-(7-chloro-6-oxo-6H-dibenzo[cd,g]inda-
zol-2-yl)ethyl)methylamino)ethyl) succinamide (2). Compound
9
(200 mg, 0.564 mmol) was treated with succinyl chloride
(31.04 L, 43.7 mg, 0.282 mmol) according to Method A affording
l
2 as an orange solid. The product was separated on a short silica
gel column. The crude product was purified by silica gel column
chromatography using chloroform/methanol 9:1 as eluant. Yield
25 mg (11%); mp: 138–140 °C; 1H NMR d 8.16 (dd, 2H); 7.95 (d,
2H); 7.60 (m, 8H); 5.86 (t, 2H); 4.58 (t, 4H); 3.16 (q, 4H); 3.00 (t,
4H); 2.51 (t, 4H); 2.33 (s, 6H); 2.00 (s, 4H); 13C NMR d 182.6,
171.7, 139.3, 138.3, 137.2, 134.6, 133.0, 132.6, 129.1, 128.8,
127.0, 122.6, 121.7, 121.0, 114.7, 57.1, 56.4, 48.4, 42.2, 36.7, 31.1;
HRMS (ESI), m/z (M+H)+: calcd 791.2627, obsd 791.2593.
4.3.2.3. 7-Chloro-[2-[(2-aminoethyl)methylamino]ethyl]anthra-
[1,9-cd]pyrazol-6(2H)-one (9). Compound 8 (300 mg, 0.789 mmol)
was dissolved in methanol (80 mL) containing HCl (1.80 mmol). Pal-
ladium on charcoal (60 mg, 10% Pd, 0.056 mmol) was added and the
mixture was placed under a positive hydrogen atmosphere for 2 h.
The mixture was then filtered through a Celite pad. The methanol fil-
trate was condensed, extracted into chloroform/aqueous sodium
hydroxide, the chloroform phase dried over Na2SO4, condensed, and
dried on a vacuum pump, affording 238 mg (0.672 mmol) of crude 9
as a yellow solid in 85% yield; 1H NMR (CDCl3) d 8.202 (dd, 1H);
7.99 (d, 1H); 7.70–7.54 (m, 4H); 4.61 (t, 2H); 3.01 (t, 2H); 2.65 (t,
2H); 2.48 (t, 2H); 2.34 (s, 3H). 13C NMR (CDCl3) d 182.8; 139.4;
138.4; 138.2; 137.3; 134.7; 133.0; 132.6; 129.1; 128.7; 126.8;
122.7; 121.8; 121.0; 114.8; 60.68; 57.5; 48.5; 42.4; HRMS (ESI), m/z
(M+H)+: calcd 355.1325, obsd 355.1317. The product was pure en-
ough to be used directly in the next step.
4.3.2.8. N,N0-Bis(2-((2-(7-chloro-6-oxo-6H-dibenzo[cd,g]inda-
zol-2-yl)ethyl)methylamino)ethyl) glutaramide (3). Compound
9 (100 mg, 0.282 mmol) was treated with 10 (45.6 mg, 0.14 mmol)
according to Method B affording 3 as an orange solid. The crude
product was purified as for 1. Yield 60 mg (53%); mp: 130–
133 °C; 1H NMR (300 MHz, CDCl3)
d 8.12 (dd, J1 = 7.3 Hz,
J2 = 1.8 Hz, 2H, ArH), 7.90 (d, J = 7.3 Hz, 2H, ArH), 7.70–7.46 (m,
8H, ArH), 5.72 (s, br, 2H, amide NH), 4.59 (t, 4H, J = 6.0 Hz, NNCH2),
3.18 (q, 4H, J = 5.4 Hz), 3.02 (t, J = 6.0 Hz, 4H, NNCH2CH2), 2.52 (t,
J = 5.4 Hz, 4H), 2.35 (s, 6H, 2 ꢂ NCH3), 2.01 (t, J = 6.4 Hz, 4H,
2 ꢂ COCH2), 1.56–1.50 (m, 2H, COCH2CH2); 13C NMR (75.5 MHz,
CDCl3) d 182.5, 172.1, 139.3, 138.1, 137.1, 134.6, 133.0, 132.6,
129.0, 128.7, 126.9, 122.5, 121.8, 120.9, 114.8, 56.9, 56.5, 48.2,
42.0, 36.5, 35.2, 21.5; HRMS (ESI), m/z (M+H)+: calcd 805.2784,
obsd 805.2784.
4.3.2.4. 1-({5-[(2,5-Dioxo-1-pyrrolidinyl)oxy]-5-oxopentanoyl}-
oxy)-2,5-pyrrolidine-dione (10). Compound 10 was prepared as
described.28 Mp: 128–130 °C (lit. 129–131 °C). The NMR matched
that reported.
4.3.2.9. N,N0-Bis(2-((2-(7-chloro-6-oxo-6H-dibenzo[cd,g]inda-
zol-2-yl)ethyl)methylamino)ethyl) adipamide (4). Adipoyl chlo-
ride (41.8 lL, 51.6 mg, 0.282 mmol) was added to a solution of 9
(200 mg, 0.564 mmol) in dichloromethane at 0 °C. A yellow precip-
itate formed. The reaction mixture was stirred overnight at room
temperature and was then was centrifuged. The supernatant was
removed and the solid was recrystallized twice in methanol/ether.
The solid was then extracted into chloroform/aqueous NaOH and
dried over Na2SO4. Compound 4 was afforded as an orange solid
after the solvent was removed by a vacuum pump. The crude prod-
uct was purified as for 1. Yield 15 mg (6%); mp: 110–114 °C; 1H
4.3.2.5. 1-({7-[(2,5-Dioxo-1-pyrrolidinyl)oxy]-7-oxoheptanoyl}-
oxy)-2,5-pyrrolidine-dione (11). N-Hydroxysuccinimide (1.27 g;
11.0 mmol), tetrahydrofuran (30 mL), and triethylamine (1.11 g,
1.52 mL; 11.0 mmol) were added into a 100 mL round bottom
flask. After the mixture was cooled to 0 °C, pimeloyl chloride
(0.8 mL, 0.985 g, 5.0 mmol) was slowly added. The mixture was
stirred for 2 h at room temperature. The solvent was removed
and the residue was dissolved in dichloromethane (100 mL) and
washed with water (3 ꢂ 50 mL) and dried over Na2SO4. An off-
white solid was obtained after filtration and evaporation of sol-
vent. Recrystallization from isopropyl alcohol yielded a white solid
(1.6 g, 90%); mp: 146–148 °C; 1H NMR (300 MHz, CDCl3) d 2.82 (s,
8H); 2.63 (t, J = 7.3 Hz, 4H); 1.74–1.84 (m, 4H); 1.47–1.60 (m, 4H);
13C NMR (75.5 MHz, CDCl3) d 169.1, 168.4, 30.7, 27.8, 25.6, 24.1;
HRMS (ESI), m/z (M+H)+: calcd 355.1141, obsd 355.1131.
NMR
d 8.11 (dd, J1 = 7.1 Hz, J2 = 2.0 Hz, 2H, ArH), 7.90 (d,
J = 7.1 Hz, 2H, ArH), 7.68–7.45 (m, 8H, ArH), 5.71 (t, J = 4.9, 2H,
amide NH), 4.59 (t, 4H, J = 6.0 Hz, NNCH2), 3.18 (q, 4H, J = 5.7 Hz),
3.02 (t, J = 6.0 Hz, 4H, NNCH2CH2), 2.53 (t, J = 5.7 Hz, 4H), 2.36 (s,
6H, 2 ꢂ NCH3), 1.54 (t, J = 6.5 Hz, 4H, 2 ꢂ COCH2), 1.25 (t,
J = 6.5 Hz, 4H, COCH2CH2); 13C NMR d 182.5, 172.5, 139.4 138.1,
137.1, 134.6, 133.1, 132.6, 128.8, 128.6, 126.7, 122.4, 121.8,
120.9, 114.8, 56.9, 56.6, 48.2, 42.0, 36.5, 35.8, 22.6; HRMS (ESI),
m/z (M+H)+: calcd 819.2940, obsd 819.2925.
4.3.2.6. N,N0-Bis(2-((2-(7-chloro-6-oxo-6H-dibenzo[cd,g]indazol-
4.3.2.10. N,N0-Bis(2-((2-(7-chloro-6-oxo-6H-dibenzo[cd,g]inda-
zol-2-yl)ethyl)methylamino)ethyl) pimelamide (5). Compound
9 (100 mg, 0.282 mmol) was treated with 11 (49.5 mg, 0.14 mmol)
according to Method B affording 5 as an orange solid. The crude
product was purified as for 1. Yield 81 mg (69%); mp: 110–
d 8.07 (dd, J1 = 7.0 Hz,
J2 = 2.0 Hz, 2H, ArH), 7.83 (d, J = 7.0 Hz, 2H, ArH), 7.65–7.44 (m,
8H, ArH), 5.60 (t, J = 4.7 Hz, 2H, amide NH), 4.56 (t, 4H, J = 6.0 Hz,
2-yl)ethyl)methylamino)ethyl) malonamide (1). Compound
9
(200 mg, 0.564 mmol) was treated with malonyl chloride (27.4
lL,
40.0 mg, 0.282 mmol) according to Method A affording 1 as an orange
solid. The crude product was purified by alumina column chromatog-
raphy using chloroform/methanol 20:1 and then 10:1 as eluants.
Yield 25 mg (11%); mp: 105–108 °C; 1H NMR d 8.07 (dd, J1 = 7.0 Hz,
J2 = 1.9 Hz, 2H, ArH), 7.87 (d, J = 7.0 Hz, 2H, ArH), 7.70–7.46 (m, 8H,
113 °C; 1H NMR (300 MHz, CDCl3)

4582
B. B. Hasinoff et al. / Bioorg. Med. Chem. 17 (2009) 4575–4582
6. Gago, F. Methods 1998, 14, 277.
7. Dawson, S.; Malkinson, J. P.; Paumier, D.; Searcey, M. Nat. Prod. Rep. 2007, 24,
109.
NNCH2), 3.16 (q, 4H, J = 5.7 Hz), 3.00 (t, J = 6.0 Hz, 4H, NNCH2CH2),
2.50 (t, J = 5.7 Hz, 4H), 2.34 (s, 6H, NCH3), 1.58 (t, J = 7.6 Hz, 4H,
COCH2), 1.24–1.14 (m, 4H, COCH2CH2), 0.91–0.81 (m, 2H,
COCH2CH2CH2); 13C NMR (75.5 MHz, CDCl3) d 183.2, 172.8, 139.3,
138.0, 137.0, 134.5, 133.0, 132.5, 128.8, 128.7, 126.7, 122.4,
121.8, 120.9, 114.8, 56.8, 56.6, 48.1, 41.9, 36.5, 36.1, 28.7, 25.1;
HRMS (ESI), m/z (M+H)+: calcd 833.3097, obsd 833.3077.
8. Martinez, R.; Chacon-Garcia, L. Curr. Med. Chem. 2005, 12, 127.
9. Leonard, G. A.; Hambley, T. W.; McAuley-Hecht, K.; Brown, T.; Hunter, W. N.
Acta Crystallogr. D. Biol. Crystallogr. 1993, 49, 458.
10. Bjorndal, M. T.; Fygenson, D. K. Biopolymers 2002, 65, 40.
11. Fortune, J. M.; Osheroff, N. Prog. Nucleic Acid Res. Mol. Biol. 2000, 64, 221.
12. Li, T. K.; Liu, L. F. Annu. Rev. Pharmacol. Toxicol. 2001, 41, 53.
13. Leteurtre, F.; Kohlhagen, G.; Paul, K. D.; Pommier, Y. J. Natl. Cancer Inst. 1994,
86, 1239.
Acknowledgments
14. Capranico, G.; Palumbo, M.; Tinelli, S.; Mabilia, M.; Pozzan, A.; Zunino, F. J. Mol.
Biol. 1994, 28, 1218.
15. Burden, D. A.; Froelich-Ammon, S. J.; Osheroff, N. Methods Mol. Biol. 2001, 95,
283.
16. Hasinoff, B. B.; Wu, X.; Begleiter, A.; Guziec, L.; Guziec, F. S., Jr.; Giorgianni, A.;
Yang, S.; Jiang, Y.; Yalowich, J. C. Cancer Chemother. Pharmacol. 2006, 57, 221.
17. Fattman, C.; Allan, W. P.; Hasinoff, B. B.; Yalowich, J. C. Biochem. Pharmacol.
1996, 52, 635.
18. Hasinoff, B. B.; Wu, X.; Krokhin, O. V.; Ens, W.; Standing, K. G.; Nitiss, J. L.;
Sivaram, T.; Giorgianni, A.; Yang, S.; Jiang, Y.; Yalowich, J. C. Mol. Pharmacol.
2005, 67, 937.
Supported by grants from the Canadian Institutes of Health Re-
search, the Canada Research Chairs Program, a Canada Research
Chair in Drug Development to Brian Hasinoff, the Robert A. Welch
Foundation (Grant AF-0005), the Herbert and Kate Dishman
Endowment at Southwestern University to Frank Guziec and NIH
grant CA90787 to Jack Yalowich.
19. Ritke, M. K.; Allan, W. P.; Fattman, C.; Gunduz, N. N.; Yalowich, J. C. Mol.
Pharmacol. 1994, 46, 58.
Supplementary data
20. Ritke, M. K.; Roberts, D.; Allan, W. P.; Raymond, J.; Bergoltz, V. V.; Yalowich, J. C.
Br. J. Cancer 1994, 69, 687.
21. Sissi, C.; Leo, E.; Moro, S.; Capranico, G.; Mancia, A.; Menta, E.; Krapcho, A. P.;
Palumbo, M. Biochem. Pharmacol. 2004, 67, 631.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2009.04.072.
22. Priebe, W.; Fokt, I.; Przewloka, T.; Chaires, J. B.; Portugal, J.; Trent, J. O. Methods
Enzymol. 2001, 340, 529.
23. McGhee, J. D. Biopolymers 1976, 15, 1345.
References and notes
24. Jones, D. Pharmaceutical Statistics; Pharmaceutical Press: London, 2002.
25. SYBYL 7.3, Tripos, 1699 South Hanley Rd., St. Louis, MO, 63144, USA, 2008.
26. GOLD 3.2, CCDC Software Ltd, Cambridge, UK, 2008.
27. Verdonk, M. L.; Cole, J. C.; Hartshorn, M. J.; Murray, C. W.; Taylor, R. D. Proteins
2003, 52, 609.
1. Corbett, A. H.; Osheroff, N. Chem. Res. Toxicol. 1993, 6, 585.
2. Malisza, K. L.; Hasinoff, B. B. Arch. Biochem. Biophys. 1995, 321, 51.
3. Liang, H.; Wu, X.; Guziec, L. J.; Guziec, F. S., Jr.; Larson, K. K.; Lang, J.; Yalowich, J.
C.; Hasinoff, B. B. J. Chem. Inf. Model. 2006, 46, 1827.
4. Hasinoff, B. B.; Liang, H.; Wu, X.; Guziec, L. J.; Guziec, F. S., Jr.; Yalowich, J. C.
Bioorg. Med. Chem. 2008, 16, 3959.
28. van den Berg, T. A.; Feringa, B. L.; Roelfes, G. Chem. Commun. 2007, 180.
5. Antonini, I.; Santoni, G.; Lucciarini, R.; Amantini, C.; Sparapani, S.; Magnano, A.
J. Med. Chem. 2006, 49, 7198.