Synthesis of DNA-directed pyrrolidinyl and piperidinyl confined alkylating chloroalkylaminoanthraquinones: Potential for development of tumor-selective N-oxides

Article abstract of DOI:10.1021/jm0608154

A novel series of 1,4-disubstituted chloroethylaminoanthraquinones, containing alkylating chloroethylamino functionalities as part of a rigid piperidinyl or pyrrolidinyl ring-system, have been prepared. The target compounds were prepared by ipso-displacement of halides of various anthraquinone chromophores by either hydroxylated or chlorinated piperidinyl- or pyrrolidinyl-alkylamino side chains. The chloroethylaminoanthraquinones were shown to alkylate guanine residues of linearized pBR322 (1-20 μM), and two symmetrically 1,4-disubstituted anthraquinones (compounds 14 and 15) were shown to interstrand cross-link DNA in the low nM range. Several 1,4-disubstituted chloroethylaminoanthraquinones were potently cytotoxic (IC₅₀ values: ≤40 nM) in human ovarian cancer A2780 cells. Two agents (compounds 18 and 19) exhibited mean GI₅₀ values of 96 nM and 182 nM, respectively, in the NCI human tumor cell line panel. Derivatization of the potent DNA cross-linking agent 15 to an N-oxide resulted in loss of the DNA unwinding, DNA interstrand cross-linking and cytotoxic activity of the parent molecule.

Full text of DOI:10.1021/jm0608154

J. Med. Chem. 2006, 49, 7013-7023
7013
Synthesis of DNA-Directed Pyrrolidinyl and Piperidinyl Confined Alkylating
Chloroalkylaminoanthraquinones: Potential for Development of Tumor-Selective N-Oxides
Klaus Pors,† Steven D. Shnyder,† Paul H. Teesdale-Spittle,‡ John A. Hartley,§ Mire Zloh,| Mark Searcey,| and
Laurence H. Patterson*,†
Institute of Cancer Therapeutics, UniVersity of Bradford, West Yorkshire, BD7 1DP, United Kingdom, School of Biological Sciences, Victoria
UniVersity of Wellington, Post Office Box 600, Wellington, New Zealand, and Cancer Research UK Drug-DNA Interactions Research Group,
Department of Oncology, UniVersity College London, School of Pharmacy, 29-39 Brunswick Square, London WC1N 1AX, United Kingdom
ReceiVed July 11, 2006
A novel series of 1,4-disubstituted chloroethylaminoanthraquinones, containing alkylating chloroethylamino
functionalities as part of a rigid piperidinyl or pyrrolidinyl ring-system, have been prepared. The target
compounds were prepared by ipso-displacement of halides of various anthraquinone chromophores by either
hydroxylated or chlorinated piperidinyl- or pyrrolidinyl-alkylamino side chains. The chloroethylamino-
anthraquinones were shown to alkylate guanine residues of linearized pBR322 (1-20 µM), and two
symmetrically 1,4-disubstituted anthraquinones (compounds 14 and 15) were shown to interstrand cross-
link DNA in the low nM range. Several 1,4-disubstituted chloroethylaminoanthraquinones were potently
cytotoxic (IC50 values: e40 nM) in human ovarian cancer A2780 cells. Two agents (compounds 18 and
19) exhibited mean GI50 values of 96 nM and 182 nM, respectively, in the NCI human tumor cell line
panel. Derivatization of the potent DNA cross-linking agent 15 to an N-oxide resulted in loss of the DNA
unwinding, DNA interstrand cross-linking and cytotoxic activity of the parent molecule.
Introduction
DNA intercalating topoisomerase inhibitors are an important
class of clinically useful antitumor agents. The anthraquinone
pharmacophore is an essential structural feature in this class of
therapeutics identified to date, notably in daunorubicin, doxo-
rubicin, epirubicin, and mitoxantrone.1 Substantial work has
identified a well-defined relationship between the basic amino-
alkylamino side chains and the configuration of functional
groups attached to the 1,4-disubstituted pharmacophore of
mitoxantrone.2-7 The clinical success of the anthraquinone-based
anticancer drugs is tempered by their failure in resistant tumors
expressing the ABCB1 (MDR1) gene8,9 or downregulation/
mutation10 or phosphorylation11 of topoisomerase II (topo II).
In an attempt to overcome these resistance mechanisms, we have
reported on 1,4-disubstituted chloroethylaminoanthraquinones
as a novel class of intercalators with alkylating functionalities,12
of which one agent, alchemix (ZP281M), possesses substantial
anticancer activity against doxorubicin (A2780AD) and cisplatin
resistant (A2780/cp70) xenografted tumors in mice.13 Alchemix
(1) is a nonsymmetrical (mixed side chains) 1,4-disubstituted
anthraquinone with a bisalkylating functionality confined to one
sidearm.
We suggest that N-chloroethylamino- moieties tethered to
anthraquinones irreversibly bind in a drug-DNA-topo II ternary
complex thereby preventing topo II dissociation and cell efflux.
Paradoxically, the high DNA affinity of the anthraquinone based
intercalators, while essential to their efficacy, also promotes
normal tissue sequestration limiting their intratumoral distribu-
tion.14 Inevitably, slow release from normal tissue will have a
major negative effect on an available drug reaching the tumor
as well as promoting organ specific toxicity. Sequestration in
tumor tissue close to the vasculature and its subsequent removal
will limit the duration and extent of tumor exposure. This could
substantially diminish the antitumor efficacy despite the known
potency of both noncovalent and covalent DNA binding
anthraquinones. In recognition of the improved pharmacology
of anthraquinones that covalently complex DNA/topo II, but
potential limitations of the nonspecific chemical reactivity of
the N-chlorethylamino moieties, we are exploring more stable
DNA-affinic alkylating agents.
In a novel series of compounds, we have prepared the
N-chloroethylamino moiety with alkylating potential as part of
a rigid piperidinyl or pyrrolidinyl ring-system. We wanted to
explore if the ring-constraint would curtail the reactivity and
lead to a more controlled and specific interaction with DNA. A
further feature of these agents is their potential for N-oxide
formation. We have pioneered the N-oxides of DNA binding
agents as prodrugs that can be bio-activated in the hypoxic
fraction of tumors to potent DNA intercalating topo II inhibitors.
As a result of this, AQ4N (banoxantrone, 1,4-bis-[[2-(dimethyl-
amino-N-oxide)ethyl]amino]5,8-dihydroxyanthracene-9,10-di-
* To whom correspondence should be addressed. Phone: +44 (0)1274
233 225. Fax: +44 (0)1274 233 234. E-mail: l.h.patterson@bradford.ac.uk.
† University of Bradford.
‡ Victoria University of Wellington.
§ University College London.
| School of Pharmacy.
10.1021/jm0608154 CCC: $33.50 © 2006 American Chemical Society
Published on Web 10/27/2006

7014 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 24
Pors et al.
Scheme 1
Reactants and conditions: (a) aminoalkylamino side chain, pyridine, 90 °C, 1/2-1 h; (b) (Ph)3P, CCl4, CH2Cl2, reflux, 3-10 h, ethereal HCl; (c)
aminoalkylamino side chain, reflux, EtOH; (d) chlorinated aminoalkylamino side chain (see Scheme 2), pyridine, 30-60 °C, 2-5 h, ethereal HCl; (e)
m-CPBA, -15 °C, 6 h (see Figure 5 for structure).
Scheme 2
one, 2a) is in advanced phase I trials against a range of solid
tumors. No normal tissue toxicity has been demonstrated in the
patients treated, but significantly, AQ4N has been shown to
release the active agent AQ4 (1,4-bis-[[2-(dimethylamino)ethyl]-
amino]5,8-dihydroxyanthracene-9,10-dione, 2b) at cytotoxic
concentrations within tumor tissue.15
Here we report on the synthesis, DNA interstrand cross-
linking, DNA sequence selectivity, and cytotoxicity of novel
symmetrical and nonsymmetrical 1,4-disubstituted piperidinyl
and pyrrolidinyl ring-constrained chloroalkylaminoanthra-
quinones. We describe also the influence of N-oxide formation
on this biological activity.
Reactants and conditions: aminoalkylamino side chain, R ) OH, n )
1, 2, synthesized as previously described;16 (a) Boc2O, Et3N, 45 °C, 16-
20 h, R ) OH, n ) 1, 2; (b) MsCl, Et3N, dry CH2Cl2, 0 °C, 1 h, R′ )
OMs, n ) 1, 2; (c) tetra-n-butylammonium chloride, dry DMF, 90 °C, 30
min, R′′ ) Cl, n ) 1, 2; (d) 4 M HCl in EtOAc, 1 h, R′′ ) Cl, n ) 1, 2.
Results and Discussion
of the hydroxyl group, (iii) conversion to the chloride with tetra-
n-butylammonium chloride, and (iv) deprotection of the Boc
group without required purification of any of the intermediates.
The deprotected side chain was then reacted with either 6 or
10 (Scheme 1) to obtain the symmetrical compound 14 (yield
39%, 1 step) and the nonsymmetrical compounds 18-21 (yield
6-28%, 2 steps). By employing route 2, the overall yield from
6 (apart from 20) was increased by approximately 10%.
The di-N-oxide derivative 15-NO was synthesized from 15
using m-CPBA (yield 69%).
DNA Sequence Selectivity. The design and synthesis of the
novel chloroalkylaminoanthraquinones was based on using
pyrrolidinyl and piperidinyl building blocks racemic in nature.
However, due to the commercial availability of the (S)-(+)-2-
pyrrolidinylmethanol moiety, we also decided to include the
enantiomeric compounds 13, 14, and 18 in our structure-activity
relationship studies.
Chemistry. The synthetic route to the target compounds
involved preparation of the heterocyclic (piperidinyl or pyrroli-
dinyl) alkylamino side chains by alkylating the heterocyclic
moiety with bromoacetonitrile, followed by reduction of the
nitrile to the respective primary amine, as previously described.16
The heterocyclic alkylamino side chains were then substituted
onto the anthraquinone moiety by ipso displacement of the
halide groups of the chromophores 3, 6, and 10 (Scheme 1).
The exception to this was compound 7, which was prepared by
condensation of 5,8-dihydroxyleucoquinizarine17 5 with an
excess amount of [1-(2-aminoethyl)-piperidin-2-yl]methanol.
The synthesis of 1,4-disubstituted chloroalkylaminoanthra-
quinones from 6 was carried out by two different routes. The
first route involved treatment of the previously16 synthesized
anthraquinone alcohols 9, 11, and 12 with triphenylphosphine-
carbon tetrachloride complex (PPh3-CCl4) to obtain 15 (sym-
metrical congener), 16, and 17 (nonsymmetrical congeners) in
an overall yield of 20-30% (from 6). In an attempt to improve
the overall yield, a second route was devised that involved
preparation of chlorinated side chains in a 4-step procedure
(Scheme 2). This was comprised of (i) Boc-protection of the
hydroxylated heterocyclic alkylamino side chains, (ii) mesylation
The chloroalkylaminoanthraquinones were investigated for
their ability to alkylate double-stranded DNA, using the Taq
polymerase stop assay.18 A synthetic 20-base oligonucleotide
primer (SRM) that binds to the sequence 3303-3284 of pBR322
was used to analyze alkylation sites on the complementary
strand, following linear amplification. To select the lowest

Synthesis of Chloroalkylaminoanthraquinones
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 24 7015
This suggests that the planar aromatic anthraquinone chro-
mophore, common to all the compounds, is directing the site
of alkylation. Agents with a primary chloride leaving group
(13-18) were shown to be significantly more potent alkylating
agents than secondary chlorides (19 and 20).
DNA Interstrand Cross-Linking. Three chloroethylamino-
anthraquinones (14-16) and one chloropropyl analogue (22)
were investigated for their ability to interstrand cross-link DNA
at a range of concentrations (0.01-10 µM) using an agarose
gel based assay.20 The chloropropyl analogue 22 did not
interstrand cross-link DNA either in concentration- (up to 100
µM) or in time-dependent (up to 24 h) experiments (results not
shown). The nonsymmetrically disubstituted compound 16, with
a bis-N-chloroethyl functionality confined to one sidearm,
alkylated but did not interstrand cross-link DNA, indicating that
this was the only compound of the series where the intra-
molecular distance between the alkylating functionalities re-
quired to cross-link adjacent strands was not met. This
phenomenon was previously shown for Alchemix, an alkylating
agent also with a bis-N-chloroethylamino functionality confined
to one sidearm.12
A concentration- (Figures 2 and 3) and time-dependent
(Figure 4) increase in formation of DNA interstrand cross-links
was observed for both 14 and 15, although the latter was a more
potent cross-linker. Compound 15 contains a racemic piperidinyl
N-chloroethyl moiety, while 14 is specifically the S-pyrrolidinyl
enantiomer. Whether the ring size or the spatial conformation
of the ring-constrained alkylating group is contributing to this
difference in potency is not clear, but modeling suggests that
the R-configuration may have an important role to play in DNA
cross-linking, as described further below. Previously, individual
enatiomers of racemic bispiperidine alkylating agents21 with no
intrinsic DNA affinity did not reveal any difference in reactivity
in regards to DNA interstrand cross-linking. Hence, it would
appear that the planar aromatic chromophore influences the
orientation toward and, hence, the reactivity with DNA of these
conformationally restricted alkylating moieties.
Figure 1. DNA strand alkylation pattern of chloroethylanthraquinones.
Alkylation is denoted by an underscore. See Experimental Section for
experimental details.
concentration that afforded alkylation, each N-chloroethyl and
N-chloropropyl compound was first examined in a wide
concentration range (1 nM-10 µM). Subsequently, all chloro-
ethylaminoanthraquinones were compared on the same gel
alongside unmodified DNA and mechlorethamine. The unmodi-
fied DNA (pBR322, not treated with agent) showed elongation
with only a few background sites of early termination (Figure
1).
It was evident that the N-chloropropyl analogues 21 and 22
did not alkylate DNA after a 24 h incubation at 10 µM (data
not shown), consistent with their inability to form an azirdinium
intermediate.19 In contrast, the N-chloroethyl analogues were
shown to be equipotent with mechlorethamine at alkylating
guanine, although some difference in sequence selectivity was
observed. There were no marked differences in selectivity of
DNA alkylation in the series of chloroethylaminoanthraquinones
examined, despite the positional variation in their reactive
moieties. The molecules showed a marked preference for 5′-
AGGAT-3′ and ACGTT-3′ compared with mechlorethamine.
Significantly, we have shown that the alkylation capability
can be totally inhibited by oxidation of the piperidinyl nitrogen,
as in 15-NO (Figure 5A). Furthermore, using an agarose gel
based supercoiled plasmid pBR322 DNA mobility assay, it was
also shown that 15-NO did not induce DNA unwinding (cf.
15, Figure 5B), consistent with the inability of N-oxides of DNA
intercalators to bind to DNA.22
Figure 2. DNA cross-linking by 14. (A) Autoradiograph of an agarose gel showing a concentration-dependent cross-link formation for 14 (1 h
drug treatment). (B) Concentration-response curve: the cross-linked DNA was determined by densitometry and is expressed as the mean of three
replicates ((s.d.). See Experimental Section for experimental detail.

7016 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 24
Pors et al.
Figure 3. DNA cross-linking by 15. (A) Autoradiograph of an agarose gel showing a concentration-dependent cross-link formation for 15 (1 h
drug treatment). (B) Concentration-response curve: the cross-linked DNA was determined by densitometry and is expressed as the mean of three
replicates ((s.d.). See Experimental Section for experimental detail.
Figure 4. Time course of 14 and 15 cross-linking of pBR322 plasmid DNA. (A) Autoradiograph of an agarose gel showing time course of DNA
cross-link formation following treatment with 14 or 15, both at 50 nM. (B) Time course of DNA cross-linking was quantitated by densitometry and
is expressed as the mean of three replicates ((s.d.). See Experimental Sectionfor experimental detail.
DNA Modeling Studies. The studies on the DNA sequence
specificity of the chloroethylaminoanthraquinones did not reveal
significant differences in terms of the potency and the pattern
of alkylation. However, the DNA interstrand cross-linking
studies demonstrated a considerable difference in reactivity
between the racemate 15 and the S-enantiomer 14. Molecular
modeling of the R and S enantiomers of the cross-linking
compounds 14 and 15 and the monoalkylating compounds 17-
19 was carried out to investigate whether their relative spatial
geometries were predictive of DNA reactivity.
A set of five lowest-energy conformations of 14, 15, and 17-
19 were chosen to examine their preference for N7-guanine
alkylation, as it has been demonstrated that the principal sites
of attack by mechloroethamine on DNA are N7-guanine and
N3-adenine in a ratio of about 86:16.23 Docking was performed
using short sequences containing guanine as a part of the
intercalation site shown to be alkylated, as shown in Figure 7.
In every simulation, the ligand was shown to dock into the
intercalation site and with energies of interaction that differed
according to the base sequence investigated. Modeling also

Synthesis of Chloroalkylaminoanthraquinones
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 24 7017
However, the majority of compounds were significantly less-
active, and some (4, 8, 13, and 22) had IC50 values greater than
1 µM. Compound 13 shows that conjugation of a monosubsti-
tuted alkylating functionality to an anthraquinone chromophore
is not sufficient to create a potent cytotoxin, despite its DNA
binding and alkylating properties. However, the addition of a
N,N′-dimethylethylenediamine side arm to produce the bis-
substituted compound 19 results in a significant increase in
cytotoxicity, This confirms the importance of two basic amino-
alkylamino side chains, irrespective of alkylating potential, in
determining the potency of this class of agent.
The effect of these agents on doxorubicin- (A2780AD) and
cisplatin-resistant (A2780/cp70) cell lines was also explored.
The A2780AD cell line is shown to have elevated P-gp24 and
a diminution of topo IIR,25 while A2780/cp70 cells exhibit
elevated levels of glutathione,26 alterations in drug uptake/
efflux,27 and diminished DNA repair mechanisms, including
mismatch repair (hMLH1) deficiency.28-30 Generally, these
anthraquinones were less-active in the resistant cell line variants,
although 17 and 19 were equipotent with doxorubicin in the
A2780AD cells. The compounds with alkylating potential (16-
21) were also more affected by the resistance mechanism present
in the A2780/cp70 cell line, but notably were still shown to be
at least 100-fold more cytotoxic than cisplatin.
Figure 5. DNA binding by 15-NO. (A) Autoradiograph of time-
dependent DNA interstrand cross-linking experiment by 15-NO (100
nM). (B) Concentration-dependent effect of 15 and 15-NO on super-
coiled plasmid pBR322 DNA. Results are expressed as the mean of
three replicates ((s.d.).
Compound 18, which potentially can form the bicyclic
aziridinium ion intermediate common to 19 (Figure 6), was less
cytotoxic in these ovarian cancer cell lines (Table 3). However,
when screened against the NCI-60 human tumor cell line panel
(data not shown), 18 (mean GI50 ) 96 nM) was shown to be
approximately twice as cytotoxic as 19 (mean GI50 ) 182 nM).
It is an interesting observation that 18, possessing a primary
chloride in the alkylating subunit, exhibits no significant
advantage over 19 in DNA reactivity and cytotoxic activity, in
spite of the latter only comprising a secondary chloride as a
leaving group in the alkylating subunit. This may be important
in the future design of alkylating chloroethylaminoanthra-
quinones.
Figure 6. Compounds 18 and 19 will form the same bicyclic
aziridinium ion intermediate prior to DNA alkylation (A ) anthra-
quinone moiety). Due to the less sterically hindered site (b), the favored
alkylation product is assumed to be the DNA-pyrrolidine adduct.
The preparation of stable N-oxides of N,N′-bischloroethyl-
aminoanthraquinones, including 1, was not possible (Patterson
and Pors, unpublished observations), most likely due to their
potential to rearrange to a cyclic hydroxylamine, as has been
observed with chlorambucil N-oxide.31 However, these sterically
constrained tertiary nitrogen-containing ring systems were
rationalized to form stable N-oxides, and the successful synthesis
and characterization of 15-NO confirms this. Table 3 shows
that 15-NO was inactive (IC50 >1 µM) against the ovarian
cancer cell lines, which was consistent with its inability to
intercalate and alkylate DNA (Figure 5).
revealed that each ligand had a range of potential orientations
with respect to the intercalation site. The criteria used to
determine the probability of alkylation were based on the
proximity of respective pyrrolidinyl or piperidinyl â-carbons
from the nitrogen to N7-guanine The data in Table 1 show that
the energies of interaction with DNA are very similar for all
three sets of enantiomers, although there appears to be slight
differences in the probability of alkylation. A possible explana-
tion for the similarity in DNA-binding, irrespective of the nature
of the piperidinyl or pyrollidinyl side arm, may be due to the
free rotation of the tertiary nitrogen of these 5- and 6-membered
ring structures. The energy-minimized structures of the enan-
tiomeric pairs of 18 and 19, capable of forming the same bicyclic
aziridinium intermediate (1-azoniabicyclo[2.1.0]hexane, Figure
6), are shown covalently adducted to N7-guanine in Figure 7.
In contrast, the enantiomeric configuration of the interstrand
DNA cross-linking agents 14 and 15 appeared to play a more
important role on their interaction with DNA, presumably
because two alkylation events are necessary for the interstrand
DNA cross-linking observed in Figures 2-4. The modeling
showed that the R-enantiomer is favored, in particular at 5′-
CTG_ACG sequences, which perhaps explains why 15, com-
posed of 50% R-enantiomer, is shown to be more reactive than
the pure S,S-enantiomer 14 (Table 2).
Conclusion
This series of ring-constrained chloroalkylaminoanthra-
quinones were prepared in part to explore whether such agents
would retain potent cytotoxicity, as observed previously with
intrinsically less-stable N,N′-bis-chloroethylaminoanthra-
quinones, and also to explore their properties as N-oxides. The
constraint on the alkylating subunits by the pyrrolidine and
piperidine ring systems led to less DNA reactive and cytotoxic
agents. However, several agents remained potently cytotoxic
in the doxorubicin- and cisplatin-resistant ovarian cancer cell
lines. In addition, we successfully derivatized the DNA inter-
strand cross-linking agent 15 to the corresponding di-N-oxide
15-NO. The lack of reactivity of the latter suggests that such
agents might have potential as molecular delivery devices,
transporting otherwise reactive alkylating agents to tumors
Cytotoxicity. Several agents (16-19) were shown to be
potent cytotoxins (IC50 e 40 nM) in wild-type A2780 cells.

7018 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 24
Pors et al.
Figure 7. DNA modeling studies of the enantiomeric pairs of 18 and 19. The images show the most favorable positions of the respective enantiomers
docked into the intercalation sites TTG_ATC with the ligands covalently bound to N7-guanine (indicated by yellow arrow).
Table 1. Ligand Interaction Energies (kcal/mol) and Alkylation Probabilities of Three Monoalkylating Chloroethylaminoanthraquinones
DNA
17(R)
Pa
17(S)
Pa
18(R)
Pa
18(S)
Pa
19(R)
Pa
19(S)
Pa
ACG_GGG
ACG_TTA
AGG_ATC
CTG_ACG
GGG_TCT
TGG_AAC
TTG_ATC
-14.926
-11.131
-13.422
-24.948
-7.3
-5.695
-22.118
m
l
l
m
l
l
-12.264
-10.01
-14.517
-22.984
-8.19
-6.909
-24.542
l
l
l
m
l
-14.241
-10.830
-15.308
-21.332
-8.268
m
l
m
h
l
m
h
-16.555
-10.666
-14.748
-25.776
-9.23
m
l
m
l
m
l
m
-15.17
-11.384
-16.569
-26.107
-9.888
-6.111
-24.601
l
l
l
l
l
l
l
-13.546
-8.524
-15.305
-26.293
-8.692
-6.833
-25.999
m
m
l
l
l
l
h
-5.682
-5.9
l
l
l
-23.501
-25.12
a P ) probability; l ) low (if the orientation of the ring is not in correct orientation to alkylate N7-G), m ) medium (the ring is in suitable orientation
to interact with N7-G, but the Cl atom was more than 3 Å away from N7-G, h ) high (the ring is in the correct orientation in respect to the N7-G and distances
between Cl and N7-G were smaller than 3 Å.
Table 2. Ligand Interaction Energies (kcal/mol) and Interstrand DNA
Cross-Linking Probabilities of Two Bisalkylating
Chloroethylaminoanthraquinones
Table 3. Growth Inhibition (IC50a) of Aminoanthraquinones and
Chloroalkylaminoanthraquinones against Ovarian Cancer Cell Lines
A2780b
nM
2780AD
nM
A2780/cp70
DNA
ACG_GGG -13.117
ACG_TTA
AGG_ATC -10.467
CTG_ACG -17.794
14(R,R) Pa 14(S,S) Pa 15(R,R) Pa 15(S,S) Pa
compd
RFc
nM
RFc
l
l
l
m
l
-11.999
-10.433
-15.38
-24.41
-8.332
-5.649
-19.749
l
l
l
l
l
l
l
-12.013
-8.845
-13.822
-24.064
-8.492
ndb
l
l
l
m
l
-16.621
-7.868
-12.131
-25.582
-6.327
l
l
l
l
l
l
l
4
7
8
9
>1000
70
>1000
740
>1000
428
178
>1000
24
>1000
>1000
>1000
>1000
>1000
782
>1000
>1000
106
>1000
70
>1000
>590
>1000
442
>1000
>1000
357
-6.216
1
0.9
1
GGG_TCT
TGG_AAC -5.244
TTG_ATC -18.212
-6.116
13
14
15
15-NO
16
l
l
ndb -6.741
1.8
-24.455
l
-19.579
a P ) probability; l ) low (if the orientation of the ring is not in correct
orientation to alkylate N7-G), m ) medium (the ring is in suitable orientation
to interact with N7-G, but the Cl atom was more than 3 Å away from N7-G,
h ) high (the ring is in the correct orientation in respect to the N7-G and
distances between Cl and N7-G were smaller than 3 Å. b nd ) interaction
energy was not detected.
4.4
2.3
14.9
1.7
2.7
7
1.4
0.9
17
40
91
69
18
19
29
9
292
75
10.1
8.3
79
63
20
68
342
5.0
96
21
576
426
0.7
535
22
>1000
>1000
2500
61.0
>1000
6600
3.0
cisplatin
doxorubicin
900
2.8
8.1
2.6
0.4
without the associated systemic toxicity common to all con-
ventional alkylating agents. Previously, we have identified
N-oxides of DNA intercalator-based topo II inhibitors as
bioreductively activated agents.32 A lead compound, AQ4N, is
currently in clinical trial as a tumor hypoxia-activated cytotoxin.
AQ4N shows no systemic toxicity, and tumor tissue shows a
significantly higher burden of reduction product AQ4 than
normal tissue (for reviews, see refs 15 and 33). By analogy,
N-oxides such as 15-NO may remain inactive until reduced in
7.5
a IC50 is the concentration of drug (nM) required to inhibit cell growth
by 50% ( standard error of the mean (n ) 3). b A2780 is the wild type
ovarian cell line; 2780AD and A2780/cp70 are doxorubicin and cisplatin
resistant variants. c RF ) resistance factor (IC50 in resistant cell line/IC50
in parent cell line).
the hypoxic regions of tumors to generate activated DNA
alkylating agents. We are currently exploring this potential.

Synthesis of Chloroalkylaminoanthraquinones
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 24 7019
15). The chromatographed solid was then crystallized from CHCl3,
affording the title compound as a dark blue powder (32 mg, 35%).
Mp 219-221 °C; 1H NMR (CDCl3) δ 13.65 (s, 2H, C(5)OH,
C(8)OH), 10.51 (t, 2H, C(1)NH and C(4)NH), 7.15 (s, 2H, C(6)H
and C(7)H), 7.05 (s, 2H, C(2)H and C(3)H), 3.55 (q, 4H, 2 ×
HNCH2CH2N), 2.65 (t, 4H, 2 × HNCH2CH2N), 2.41-2.68 (m,
8H, 2 × NCH2H2), 1.55-1.72 (m, 8H, 8 × ring-H), 1.45-1.55
(m, 4H, 4 × ring-H); 13C NMR (CDCl3) δ 185.35, 155.46, 146.31,
124.63, 123.63, 115.48, 109.21, 57.63, 54.64, 40.72, 26.09, 24.37;
FAB MS m/z (M + H)+ 493.
Experimental Section
The UV/vis absorbance of the compounds investigated biologi-
cally was recorded on a Beckman DU70 UV/vis spectrophotometer
fitted with deuterium and tungsten lamps. The infrared absorbance
was recorded on a Nicolet 205 FT-IR spectrometer. 1H NMR (250
MHz) and 13C NMR (69 MHz) spectra were obtained on a Bruker
AC-250 spectrometer. Fast atom bombardment (FAB+) mass
spectra sample identification was obtained on a V.G 70 SEQ mass
spectrometer. Microelemental analysis was obtained using a Carlo-
Erba EA 1108 instrument with a PC-based data system, Eager 200
for Windows, and Sartorious ultra microbalance 4504 MP8. Thin-
layer chromatography was carried out on aluminum-backed silica
plates (Merck, 60 F254), and column chromatography was carried
out on silica gel (particle size 35-70 µm and 20-35 µm).
Chemistry. The preparation of the chromophores, 5,8-dihydroxy-
leucoquinizarin (leuco-1,4,5,8,-tetrahydroxyanthraquinone 5,34 1,4-
difluoro-5,8-dihydroxyanthraquinone 6,35 and 1-[[2-(di-methyl-
amino)ethyl]amino]-4-fluoro-5,8-dihydroxyanthracene-9,10-dione
(10),16 was carried out as previously described. The aminoalkyl-
amino side chains and intermediate hydroxylaminoanthraquinones
9, 11, and 12 were prepared as described previously.16
1-{[2-(2-Hydroxyethylpyrrolidine)ethyl]amino}-anthracene-
9,10-dione (4). [1-(2-Aminoethyl)-pyrrolidin-2-yl-]methanol (1.63
g, 11.30 mmol) in pyridine (10 mL) was added to 1-chloro-
anthraquinone (1.37 g, 5.65 mmol), and the mixture was stirred at
65 °C for 24 h. Pyridine was removed in vacuo, and the resulting
mixture of oil and solid was dissolved in CH2Cl2 and washed with
H2O (3 × 50 mL) to remove any unreacted amine. The separated
organic layer was removed in vacuo, and the crude product was
chromatographed using CH2Cl2/CH3OH (97:3). The desired product
1-{[2-(2-Chloroethylpyrrolidine)ethyl]amino}-anthracene-
9,10-dione Hydrochloride (13). Ph3P (260 mg, 0.99 mmol) and
CCl4 (100 µL, 9.86 mmol) were stirred for 15 min before the
mixture was added dropwise to a stirred solution of 4 (115 mg,
0.329 mmol) in dry CH2Cl2 (5 mL) under N2 at reflux temperature.
The reaction mixture was kept at reflux temperature for 3 h before
it was cooled down to rt. Ethereal HCl was added to the solution,
and after 1 h of stirring, the precipitated solid was filtered off. To
remove excess of Ph3P and Ph3PO, the precipitated solid was
dissolved in warm CH3OH (10 mL). While stirring the dark blue
solution at reflux, a mixture of EtOAc and EtOH (1:1) was added
until precipitation of solid was observed. The solution was set aside
for 1 h before the precipitated product was isolated by filtration;
the excess Ph3P and Ph3PO remained in the EtOAc/EtOH solution.
The product was afforded as an orange powder (92 mg, 69%). Mp
1
250-253 °C; H NMR (CDCl3) δ 9.95 (s, 1H, C(1)NH), 8.14-
8.32 (m, 2H), 7.75-7.95 (m, 2H), 7.45-7.65 (m, 2H), 7.22 (2 ×
d, 1H), 3.8 (2 × d, 1H), 3.42-3.61 (m, 4H), 3.05-3.15 (m, 2H),
2.65-2.75 (m, 2H), 2.15-2.35 (m, 1H), 1.61-2.13 (m, 4H), HCl
salt was not observable in the spectrum; 13C NMR (CDCl3) δ
184.38, 183.65, 151.29, 135.26, 133.91, 132.81, 126.47, 115.51,
113.35, 65.26, 62.61, 55.82, 50.99, 41.74, 27.45, 24.31; FAB MS
m/z (M + H)+ 351. Anal. (C21H21N2O2Cl‚HCl) C, H, N.
1
was yielded as a red powder (0.21 g, 11%). Mp 113-115 °C; H
NMR (CDCl3) δ 10.05 (s, 1H, C(1)NH), 8.20 (2 × d, H), 8.45 (2
× d, H), 7.65-7.81 (m, 2H), 7.45-7.62 (m, 2H), 7.01 (2 × d,
1H), 3.75-3.85 (2 × d, 1H), 3.35-3.55 (m, 4H), 3.15-3.30 (m,
2H), 2.60-2.83 (m, 2H), 2.25-2.35 (m, 1H), 1.75-2.02 (m, 4H);
13C NMR (CDCl3) δ 184.89, 183.77, 151.49, 135.32, 133.81,
132.86, 126.57, 115.60, 113.15, 65.26, 62.61, 53.82, 52.99, 41.64,
27.15, 24.11; FAB MS m/z (M + H)+ 351.
1,4-Bis-{[2-(2-chloroethylpyrrolidine)ethyl]amino}-5,8-dihy-
droxyanthracene-9,10-dione Hydrochloride (14). Preparation of
14 involved 5 steps, where none of the intermediate products
(i-iv) were isolated and purified: (i) Boc protection of monohy-
droxylated diamine side chain, (ii) mesylation of Boc-protected
monohydroxylated diamine side chain, (iii) chlorination of mono-
mesylated diamine side chain, (iv) deprotection of the Boc group
of chlorinated diamine side chain, and (v) ipso substitution of the
fluoride of 1-[[2-(dimethylamino)ethyl]amino]-4-fluoro-5,8-dihy-
droxyanthraquinone by chlorinated diamine.
1,4-Bis-{[2-(3-piperidinemethanol)ethyl]amino}-5,8-dihydroxy-
anthracene-9,10-dione (7). 1-(2-Aminoethyl)-3-piperidin-3-ol (1.0
g, 6.33 mmol) in EtOH (5 mL) was added to a suspension of
tetrahydroxyleucoquinizarin (0.14 g, 0.53 mmol) in EtOH (25 mL)
under N2. After 8 h of stirring at reflux temperature, the reaction
mixture was stirred at rt another 14 h. The EtOH was removed in
vacuo, and the remaining residue was added to ice-cold brine. The
precipitated solid was isolated by filtration and lyophilised. The
dark blue solid was flash chromatographed using CH2Cl2 to remove
nonpolar side-products, then by increasing the polarity gradually
to CH2Cl2/CH3OH (4:1) to obtain the desired product as a dark-
Preparation of Side Chain: tert-Butyl 2-(2-(Hydroxymethyl)-
pyrollidin-1-yl)ethylcarbamate (i). [1-(2-Aminoethyl)-pyrrolidin-
2-yl-]methanol (7 g, 48.61 mmol) and Et3N (8.12 mL, 58.33 mmol)
were stirred together with CH3OH (50 mL) for 5 min before Boc2O
(12.73 g, 58.33 mmol) in dissolved in CH3OH (15 mL) was added
dropwise over 15-20 min. The reaction mixture was then stirred
20 h at 45 °C before being concentrated in vacuo. The oil was
diluted in EtOAc (40 mL) and washed with 2 × H2O (20 mL) and
brine (20 mL). The organic phase was dried with magnesium sulfate
(MgSO4) and, after filtration, was concentrated in vacuo, yielding
a straw-colored oil that needed no further purification (9.36 g, 79%).
FAB MS m/z (M + H)+ 245. (1-(2-(tert-Butoxycarbonylamino)-
ethyl)pyrrolidin-2-yl)methyl Methanesulfonate (ii). MsCl (3.81
mL, 49.20 mmol) was added dropwise to an ice-cold solution of
the Boc-protected amine (8 g, 32.8 mmol) and Et3N (6.85 mL, 49.2
mmol) in dry CH2Cl2 (50 mL) under N2. After the reaction mixture
was stirred for 1 h at 0 °C, the solution was diluted with cold CH2Cl2
and washed with ice-cold NaHCO3 and ice-cold brine. The organic
phase was dried with MgSO4, filtered, and concentrated in vacuo
at room temperature. The mesylated product was afforded as a crude
straw-colored oil (9.06 g, 86%). FAB MS m/z (M + H)+ 323. tert-
Butyl 2-(2-(Chloromethyl)pyrrolidin-1-yl)ethylcarbamate (iii).
Tetra-n-butylammonium chloride (2.0 g, 7.20 mmol) was added to
a stirred solution of the crude mesylate (9 g, 28.01 mmol) in dry
DMF (50 mL). The reaction mixture was heated at 90 °C for 30
min before DMF was removed in vacuo. The residual oil was taken
up in CH2Cl2 and washed with ice-cold NaHCO3 and ice-cold brine.
The organic phase was dried (MgSO4) and filtered, and the solvent
1
blue solid (64 mg, 22%). Mp 180-183 °C; H NMR (CDCl3) δ
13.64 (s, 2H, C(5)OH and C(8)OH), 10.54 (t, 2H, C(1)NH and
C(4)NH), 7.10 (s, 2H, C(2)H and C(3)H), 7.23 (s, 2H, C(6)H and
C(7)H), 3.63-3.66 (m, 4H, 2 × CHCH2OH), 3.45-3.58 (m, 4H,
2 × HNCH2CH2), 2.71 (t, 6H, 2 × HNCH2CH2N and 2 × ring-
H), 2.95 (2 × d, 2H, 2 × ring-H), 2.15-2.35 (m, 4H, 4 × ring-H),
1.10-1.95 (m, 12H, 2 × OH and 10 × ring-H); 13C NMR (CDCl3)
δ 182.36, 156.36, 146.60, 134.46, 132.09, 126.13, 123.87, 109.89,
65.70, 57.75, 56.95, 54.43, 40.52, 37.90, 26.92, 24.08; FAB MS,
m/z (M + H)+ 553; IR υmax/cm-1; 3400 (OH), 3225 (NH), 3100
(Ar-CH), 2960-2800 (CH2, CH3), 1650, 1625, 1575, 1480, 1370,
1225.
1,4-Bis-{[2-(piperidine)ethyl]amino}-5,8-dihydroxy-anthracene-
9,10-dione (8). 1,4-Difluoro-5,8-hydroxyanthraquinone (50 mg,
0.18 mmol) and 1-(2-aminoethyl)-piperidine (232 mg, 1.81 mmol)
were stirred in pyridine (1 mL) at 90 °C for 30 min. The reaction
mixture was added to ice-cold brine and set aside at 4 °C overnight.
The precipitated solid was isolated by filtration and lyophilised.
The desired product was purified by flash chromatography, initially
eluting with CH2Cl2/CH3OH (95:5) to remove nonpolar impurities,
followed by a gradual increase of CH3OH to CH2Cl2/CH3OH (85:

7020 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 24
Pors et al.
was concentrated in vacuo at room temperature. The crude product
was yielded as a straw-/yellowish-colored oil (6.78 g, 61%). FAB
MS m/z (M + H)+ 263. 2-(2-(Chloromethyl)pyrrolidin-1-yl)-
ethanamine (iv). The crude chloride (6.78 g, 25.70 mmol) was
stirred in 4 M HCl in EtOAc for an hour to remove the Boc group.
To the acidic EtOAc solution, cooled in an ice-bath, was slowly
added a solution of brine and NH3 (pH ) 12) until the aqueous
phase was pH ∼11. The chlorinated diamine was then extracted
into the organic phase, which was dried with MgSO4. The solvent
was removed in vacuo, and the crude product was yielded as a
brownish oil (1.98 g, 47%) that was used directly in the next step.
FAB MS m/z (M + H)+ 163.
h of reflux. The product was afforded as a dark blue powder (12
mg, 81%). H NMR (CDCl3) δ 13.65 (s, 2H, C(5)OH, C(8)OH),
1
10.55 (t, 2H, C(1)NH and C(4)NH), 7.12-7.18 (m, 2H, C(6)H and
C(7)H), 7.10 (s, 2H, C(2)H and C(3)H), 3.81-3.94 (d, 4H, 2 ×
CHCH2Cl), 3.35-3.45 (q, 4H, 2 × HNCH2CH2N), 3.02 (t, 2H, 1
× HNCH2CH2N), 2.81-2.93 (m, 2H, 2 × NCHCH2), 2.72 (t, 2H,
1 × HNCH2CH2N), 2.35 (s, 6H, 2 × NCH3), 1.30-1.65 (m, 6H,
6 × ring-H), HCl salt was not observable in the spectrum; FAB
MS m/z (M + H)+ 549. Anal. (C28H34N4O4Cl2‚2HCl) C, H, N.
1-{[(2-Dimethylamino)ethyl]amino}-4-{[2-(2-chloroethylpip-
eridine)ethyl]-amino}-5,8-dihydroxyanthracene-9,10-dione Hy-
drochloride (17). The method follows that of 13, using 12 (105
mg, 0.21 mmol), Ph3P (165 mg, 0.63 mmol), CCl4 (500 µL, 5.25
mmol), and dry CH2Cl2 (10 mL). The reaction was stopped after 3
h of reflux. The product was afforded as a dark blue powder (89
Preparation of Title Compound 14 (v): A mixture of 10 (125
mg, 0.363 mmol) and the crude 2-(2-(chloromethyl)pyrrolidin-1-
yl)ethanamine (198 mg, 12.1 mmol) was reacted in pyridine (5 mL)
at 30 °C for 4 h. The reaction mixture was concentrated in vacuo,
and the crude product was purified by initially eluting with CH2Cl2
to remove nonpolar impurities, followed by a gradual increase of
CH3OH to CH2Cl2/CH3OH (97:3). The product was dissolved in
CH2Cl2 (2 mL), and ethereal HCl was added to generate the
hydrochloride salt 19, which was isolated by filtration and dried
under vacuum to afford the title compound as a dark blue solid
1
mg, 73%). Mp 233-235 °C; H NMR (CDCl3) δ 13.35 (s, 2H,
C(5)OH, C(8)OH), 10.45 (t, 2H, C(1)NH and C(4)NH), 7.64 (s,
2H, C(6)H and C(7)H), 7.15 (s, 2H, C(2)H and C(3)H), 4.20-
4.24 (m, 1H), 3.92-4.15 (m, 4H), 3.61-3.75 (m, 2H), 3.24-3.35
(m, 4H), 2.81 (s, 6H, 2 × NCH3), 1.55-2.10 (m, 8H, 8 × ring-H),
HCl salt was not observable in the spectrum; 13C NMR (CDCl3) δ
184.20, 154.65, 145.90, 145.83, 124.93, 124.81, 114.23, 108.34,
62.25, 54.85, 51.73, 42.21, 36.70, 26.03, 21.60; FAB MS m/z (M
+ H)+ 501. Anal. (C26H33N4O4Cl‚2HCl‚3H2O) C, H, N.
1
(89 mg, 39%). Mp > 300 °C dec; H NMR (DMSO) δ 13.45 (s,
2H, C(5)OH, C(8)OH), 10.45 (t, 2H, C(1)NH and C(4)NH), 7.15
(s, 2H, C(6)H and C(7)H), 7.05 (s, 2H, C(2)H and C(3)H), 4.10-
4.19 (m, 2H, 2 × NCHCH2Cl), 3.45-3.55 (q, 4H, 2 × HNCH2-
CH2N), 3.15 (d, 2H, 2 × NCHCH2Cl), 2.65-2.85 (m, 6H, 2 ×
HNCH2CH2N and 2 × ring-H), 2.15-2.35 (m, 6H, 6 × ring-H),
1.45-1.85 (m, 6H, 6 × ring-H), HCl salt was not observable in
the spectrum; 13C NMR (DMSO) δ 183.28, 156.69, 144.33, 127.56,
124.33, 114.93, 107.91, 65.26, 60.52, 55.43, 50.83, 41.78, 35.05,
28.11; FAB MS m/z (M + H)+ 561. Anal. (C28H34N4O4Cl2‚2HCl)
C, H, N.
1-{[(2-Dimethylamino)ethyl]amino}-4-{[2-(2-chloroethylpyr-
rolidine)ethyl]-amino}-5,8-dihydroxyanthracene-9,10-dione Hy-
drochloride (18). The method follows that of 14.
Preparation of Side Chain: tert-Butyl 2-(2-(Hydroxymethyl)-
pyrollidin-1-yl)ethylcarbamate (i). [1-(2-Aminoethyl)-pyrrolidin-
2-yl-]methanol (5 g, 34.72 mmol), Et3N (5.8 mL, 41.67 mmol),
CH3OH (40 mL), and Boc2O (9.10 g, 41.67 mmol) were dissolved
in CH3OH (10 mL). The reaction mixture was stirred 18 h. The
product was afforded as a straw-colored oil that needed no further
purification (6.9 g, 82%). (1-(2-(tert-Butoxycarbonylamino)ethyl)-
pyrrolidin-2-yl)methyl Methanesulfonate (ii). Boc-protected amine
(5.1 g, 20.90 mmol), MsCl (2.43 mL, 31.35 mmol), Et3N (4.32
mL, 31.35 mmol), and dry CH2Cl2 (50 mL) were used. The
mesylated product was afforded as a crude straw-colored oil (5.63
g, 84%). FAB MS m/z (M + H)+ 245. tert-Butyl 2-(2-(Chloro-
methyl)pyrrolidin-1-yl)ethylcarbamate (iii). Crude mesylate (5.63
g, 17.48 mmol), tetra-n-butylammonium chloride (9.72 g, 11.26
mmol), and dry DMF (30 mL) at 90 °C for 30 min were the reaction
conditions. The crude chloride was afforded as a straw/yellowish-
colored oil (2.2 g, 48%). FAB MS m/z (M + H)+ 323. 2-(2-
(Chloromethyl)pyrrolidin-1-yl)ethanamine (iv). Crude chloride
(2.0 g, 7.58 mmol), 4 M HCl, and EtOAc for 1 h were the reaction
conditions. The crude Boc-deprotected amine was afforded as a
brownish oil (675 mg, 55%). FAB MS m/z (M + H)+ 263.
Preparation of Title Compound 18: Compound 10 (95 mg,
0.28 mmol), crude 2-(2-(chloromethyl)pyrrolidin-1-yl)ethanamine
(675 mg, 4.15 mmol), and pyridine (2 mL) for 2 h at 30 °C were
the reaction conditions. The title compound was afforded as a dark
blue solid (64 mg, 41%). FAB MS m/z (M + H)+ 163. Mp 253-
255 °C; 1H NMR (CDCl3) δ 13.55 (s, 2H, C(5)OH, C(8)OH), 10.45
(t, 2H, C(1)NH and C(4)NH), 7.12 (s, 2H, C(6)H and C(7)H), 7.05
(s, 2H, C(2)H and C(3)H), 3.38-4.47 (m, 5H, 2 × HNCH2CH2N
and 1 × NCHCH2Cl), 3.15 (d, 1H, 1 × NCHCH2Cl), 2.62-2.83
(2 × t, 4H, 2 × HNCH2CH2N), 2.35 (s, 6H, 2 × NCH3), 2-21.25
(m, 3H, 3 × ring-H), 1.51-1.85 (m, 4H, 4 × ring-H), HCl salt
was not observable in the spectrum; 13C NMR (CDCl3) δ 185.11,
155.32, 146.2, 146.01, 124.49, 123.69, 123.55, 115.40, 109.02,
61.43, 58.35, 56.27, 56.03, 55.78, 52.82, 45.59, 41.19, 40.33, 34.86,
24.93; FAB MS m/z (M + H)+ 487. Anal. (C25H31N4O4Cl‚2HCl‚
2H2O) C, H, N.
1,4-Bis-{[2-(2-chloroethylpiperidine)ethyl]amino}-5,8-dihy-
droxyanthracene-9,10-dione Hydrochloride (15). The method
follows that of 13, using 9 (60 mg, 0.11 mmol), Ph3P (172 mg,
0.65 mmol), CCl4 (190 µL, 1.96 mmol), and dry CH2Cl2 (5 mL).
The reaction was stopped after 5 h of reflux. The product was
afforded as a dark blue powder (50 mg, 78%). Mp > 300 °C dec;
1H NMR (DMSO) δ 13.45 (s, 2H, C(5)OH, C(8)OH), 10.45 (t,
2H, C(1)NH and C(4)NH), 7.65 (s, 2H, C(6)H and C(7)H), 7.24
(s, 2H, C(2)H and C(3)H), 4.15 (2 × d, 4H, 2 × NCHCH2Cl),
4.03-4.12 (q, 4H, 2 × HNCH2CH2N), 3.44-3.83 (m, 8H, 2 ×
HNCH2CH2N and 4 × ring-H), 1.75-2.15 (m, 10H, 10 × ring-
H), 1.51-1.65 (m, 4H, 4 × ring-H), HCl salt was not observable
in the spectrum; 13C NMR (DMSO) δ 184.28, 154.69, 145.93,
126.56, 125.00, 114.33, 108.41, 62.26, 60.32, 51.83, 50.78, 42.78,
37.05, 26.11; FAB MS m/z (M + H)+ 589. Anal. (C30H38N4O4Cl2‚
2HCl) C, H, N.
1,4-Bis-{[2-(2-chloroethylpiperidine-N-oxide)ethyl]amino}-
5,8-dihydroxyanthracene-9,10-dione (15-NO). m-CPBA (25 mg,
0.15 mmol) dissolved in dry DCM (1 mL) was added dropwise to
a stirred solution of 15 (33 mg, 0.06 mmol) in dry CH2Cl2 (5 mL)
under N2. After 15 min of stirring at -10 °C (acetone ice bath),
the reaction was stirred 3 h at 4 °C. The crude product was
chromatographed by initially eluting with CH2Cl2/CH3OH (1:10),
then followed by a gradual increase of polarity to CH2Cl2/CH3OH/
NH3 (29:70:1). The desired product was afforded as a crude dark
blue solid (24 mg, 69%). 1H NMR (DMSO) δ 13.52 (s, 2H,
C(5)OH, C(8)OH), 10.54 (t, 2H, C(1)NH and C(4)NH), 7.71 (s,
2H, C(6)H and C(7)H), 7.32 (s, 2H, C(2)H and C(3)H), 4.55 (2 ×
d, 4H, 2 × NCHCH2Cl), 3.95-4.44 (m, 12H, 2 × HNCH2CH2N,
2 × HNCH2CH2N, and 4 × ring-H), 2.95-3.23 (m, 6H, 6 × ring-
H), 1.82-2.02 (m, 6H, 6 × ring-H), 1.64-1.71 (m, 4H, 4 × ring-
H), HCl salt was not observable in the spectrum; FAB MS m/z (M
+ H)+ 623. Anal. (C30H38N4O6Cl2) C, H, N.
1-{[(2-Dimethylamino)ethyl]amino}-4-{[2-(3-chloropiperidine)-
ethyl]amino}-5,8-dihydroxyanthracene-9,10-dione Hydrochlo-
ride (19). The method follows that of 14.
Preparation of Side Chain: tert-Butyl 2-(2-(Hydroxymethyl)-
piperidin-1-yl)ethylcarbamate (i). 1-(2-Aminoethyl)-piperidin-3-
ol (1 g, 6.94 mmol), Et3N (1.16 mL, 8.33 mmol), CH3OH (10 mL),
and Boc2O (1.82 g, 8.33 mmol) dissolved in CH3OH (5 mL) were
1-{[(2-Dimethylamino)ethyl]amino}-4-{[2-(2,6-dichloroethyl-
piperidine)ethyl]-amino}-5,8-dihydroxyanthracene-9,10-dione Hy-
drochloride (16). The method follows that of 13, using 11 (14
mg, 0.03 mmol), Ph3P (43 mg, 0.164 mmol), CCl4 (79 µL, 0.82
mmol), and dry CH2Cl2 (5 mL). The reaction was stopped after 5

Synthesis of Chloroalkylaminoanthraquinones
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 24 7021
stirred 20 h at 45 °C. The product, a straw-colored oil, needed no
further purification (1.35 g, 80%). FAB MS m/z (M + H)+ 245.
(1-(2-(tert-Butoxycarbonylamino)ethyl)piperidin-2-yl)methyl Meth-
anesulfonate (ii). MsCl (420 µL, 5.41 mmol), Boc-protected amine
(880 mg, 3.61 mmol), Et3N (755 µL, 5.41 mmol), and dry CH2Cl2
(10 mL) were stirred for 1 h at 0 °C. The mesylated product was
afforded as a crude straw-colored oil (965 mg, 83%). FAB MS
m/z (M + H)+ 323. tert-Butyl 2-(2-(Chloromethyl)piperidin-1-
yl)ethylcarbamate (iii). Tetra-n-butylammonium chloride (2.0 g,
7.20 mmol), crude mesylate (965 mg, 3.01 mmol), and dry DMF
(5 mL) were heated at 90 °C for 30 min. The crude product was
yielded as a straw-/yellowish-colored oil (624 mg, 79%). FAB MS
m/z (M + H)+ 263. 2-(2-(Chloromethyl)piperidin-1-yl)ethan-
amine (iv). The crude chloride (624 mg, 2.38 mmol) was stirred
in 4 M HCl in EtOAc for an hour to remove the Boc group. The
crude product was yielded as a brownish oil (175 mg, 45%) that
was used directly in the next step. FAB MS m/z (M + H)+ 163.
1-{[(2-Dimethylamino)ethyl]amino}-4-{[2-(4-chloropiperidine)-
ethyl]amino}-5,8-dihydroxyanthracene-9,10-dione Hydrochlo-
ride (21). The method follows that of 14.
Preparation of Side Chain: tert-Butyl 2-(4-Hydroxypiperidin-
1-yl)ethylcarbamate (i). 1-(2-Aminoethyl)-piperidin-4-ol (2.0 g,
13.89 mmol), Et3N (2.32 mL, 16.65 mmol), CH3OH (20 mL), and
Boc2O (3.63 g, 16.65 mmol) were dissolved in CH3OH (5 mL).
The reaction mixture was stirred 18 h. The product was afforded
as a straw-colored oil that needed no further purification (2.35 g,
69%). FAB MS m/z (M + H)+ 245. 1-(2-(tert-Butoxycarbonyl-
amino)ethyl)piperidin-4-yl Methanesulfonate (ii). Boc-protected
amine (1.70 g, 6.94 mmol), MsCl (810 µL, 10.41 mmol), Et3N
(1.45 mL, 10.41 mmol), and dry CH2Cl2 (20 mL) were the reactants.
The mesylated product was afforded as a crude straw-colored oil
(1.39 g, 62%). FAB MS m/z (M + H)+ 323. tert-Butyl 2-(4-
chloropiperidin-1-yl)ethylcarbamate (iii). Crude mesylate (1.39
g, 4.29 mmol), tetra-n-butylammonium chloride (2.38 g, 8.58
mmol), and dry DMF (10 mL) at 120 °C for 30 min were the
reaction conditions. The crude chloride was afforded as a straw/
yellowish-colored oil (0.73 g, 65%). FAB MS m/z (M + H)+ 263.
2-(4-Chloropiperidin-1-yl)ethanamine (iv). Crude chloride (0.73
g, 2.78 mmol), 4 M HCl, and EtOAc for 1 h were the reaction
conditions. The crude Boc-deprotected amine was afforded as a
brownish oil (160 mg, 35%). FAB MS m/z (M + H)+ 163.
Preparation of Title Compound 21: Compound 10 (36 mg,
0.11 mmol), crude 2-(4-chloropiperidin-1-yl)ethanamine (160 mg,
0.98 mmol), and pyridine (2 mL) for 5 h at 45 °C were the reaction
conditions. The title compound 21 was afforded as a dark blue solid
Preparation of Title Compound 19: A mixture of 10 (36 mg,
0.11 mmol) and the crude 2-(2-(chloromethyl)piperidin-1-yl)-
ethanamine (175 mg, 1.08 mmol) was reacted in pyridine (2 mL)
at rt for 2 h. The title compound was afforded as a dark blue solid
1
(35 mg, 60%). Mp > 300 °C dec; H NMR (CDCl3) δ 13.51 (s,
2H, C(5)OH, C(8)OH), 10.45 (t, 2H, C(1)NH and C(4)NH), 7.15
(s, 2H, C(6)H and C(7)H), 7.05 (s, 2H, C(2)H and C(3)H), 4.10-
4.14 (m, 1H, CH2CHCl), 3.45 (t, 4H, 2 × HNCH2CH2N), 3.15-
3.24 (2 × d, 1H, ring-H), 2.65-2.81 (2 × t, 4H, 2 × HNCH2CH2N),
2.35 (s, 6H, 2 × NCH3), 2.15-2.25 (m, 3H, 3 × ring-H), 1.52-
1.93 (m, 4H, 4 × ring-H), HCl salt was not observable in the
spectrum; 13C NMR (CDCl3) δ 185.35, 185.28, 155.41, 146.24,
146.05, 124.66, 123.78, 123.64, 115.39, 109.23, 61.43, 58.39, 56.28,
55.78, 52.85, 45.63, 41.26, 40.37, 34.88, 29.65, 24.95; FAB MS
m/z (M + H)+ 487. Anal. (C25H31N4O4Cl‚2HCl) C, H, N.
1
(31 mg, 54%). Mp 200-202 °C; H NMR (CDCl3) δ 13.55 (s,
2H, C(5)OH, C(8)OH), 10.45 (t, 2H, C(1)NH and C(4)NH), 7.14
(s, 2H, C(6)H and C(7)H), 7.05 (s, 2H, C(2)H and C(3)H), 3.42-
3.47 (m, 1H, CH2CHCl), 3.31 (q, 4H, 2 × HNCH2CH2N), 2.62-
2.84 (t, 4H, 2 × HNCH2CH2N), 2.35 (s, 6H, 2 × NCH3), 1.95-
2.12 (m, 2H, 2 × ring-H), 1.32-1.75 (m, 6H, 6 × ring-H), HCl
salt was not observable in the spectrum; 13C NMR (CDCl3) δ
185.29, 161.33, 155.34, 146.22, 146.12, 124.61, 123.95, 123.77,
115.31, 109.09, 58.96, 56.25, 50.79, 45.38, 34.69, 24.06; FAB MS
m/z (M + H)+ 487. Anal. (C25H31N4O4Cl‚2HCl‚2H2O) C, H, N.
1,4-Bis-{[2-(3-chloroethylpiperidine)ethyl]amino}-5,8-dihy-
droxyanthracene-9,10-dione Hydrochloride (22). The method
follows that of 13, using Ph3P (404 mg, 1.54 mmol), CCl4 (447
µL, 4.63 mmol), and 7 (142 mg, 0.26 mmol) with stirring in dry
CHCl3/CH3CN (4:1; 5 mL) under N2 at reflux temperature for 5 h.
The title compound was afforded as a dark blue solid (117 mg,
68%). Mp > 300 °C dec; 1H NMR (CDCl3/D2O (10:1)) δ 7.01 (s,
2H, C(6)H and C(7)H), 6.96 (s, 2H, C(2)H and C(3)H), 3.43-
3.83 (m, 16H, 2 × HNCH2CH2N, 2 × HNCH2CH2, and 8 × ring-
H), 3.04 (m, 4H, ring-H), 1.85-2.31 (m, 8H, 8 × ring-H), 1.38-
1.43 (m, 2H, 2 × CH2CHCH2OH), HCl salt was not observable in
the spectrum; 13C NMR (CDCl3/D2O(10:1)) δ 187.18, 156.32,
148.32, 127.37, 126.78, 117.38, 111.43, 58.40, 56.36, 48.97, 40.03,
38.45, 27.99; FAB MS m/z (M + H)+ 589. Anal. (C30H38N4O4-
Cl2‚2HCl‚2H2O) C, H, N.
1-{[(2-Dimethylamino)ethyl]amino}-4-{[2-(3-chloropyrrolidine)-
ethyl]amino}-5,8-dihydroxyanthracene-9,10-dione Hydrochlo-
ride (20). The method follows that of 14.
Preparation of Side Chain: tert-Butyl 2-(3-Hydroxypyrroli-
din-1-yl)ethylcarbamate (i). 1-(2-Aminoethyl)-pyrrolidin-3-ol (1
g, 7.94 mmol), Et3N (1.32 mL, 9.52 mmol), CH3OH (10 mL), and
Boc2O (2.08 g, 9.52 mmol) dissolved in CH3OH (5 mL) were used.
The reaction mixture was stirred 16 h. The product was afforded
as a straw-colored oil that needed no further purification (1.54 g,
86%). FAB MS m/z (M + H)+ 230. 1-(2-(tert-Butoxycarbonyl-
amino)ethyl)pyrrolidin-3-yl Methanesulfonate (ii). Boc-protected
amine (960 mg, 4.16 mmol), MsCl (483 µL, 6.24 mmol), Et3N
(868 µL, 6.24 mmol), and dry CH2Cl2 (10 mL) were used. The
crude product was afforded as a straw-colored oil (1.0 g, 78%).
FAB MS m/z (M + H)+ 308. tert-Butyl 2-(3-Chloropyrrolidin-
1-yl)ethylcarbamate (iii). Crude mesylate (1 g, 3.24 mmol), tetra-
n-butylammonium chloride (1.35 g, 4.86 mmol), and dry DMF (10
mL) at 100 °C for 30 min were the reaction conditions. The product
was afforded as a straw-/yellowish-colored oil (705 mg, 81%). FAB
MS m/z (M + H)+ 248. 2-(3-Chloropyrrolidin-1-yl)ethanamine
(iv). Crude chloride (705 mg, 2.88 mmol), 4 M HCl, and EtOAc
for 1 h were the reaction conditions. The crude Boc-deprotected
amine was afforded as a brownish oil (124 mg, 30%). FAB MS
m/z (M + H)+ 148.
DNA Modeling. Initial models of these selected chloroethyl-
aminoanthraquinones were built using Maestro 6.05 graphic user
interface (Schrodinger, Inc.). The conformational searches for all
molecules were performed by Macromodel 8.536 using Monte Carlo
simulation,37 Amber force field,38 and generalized Born/surface area
(BD/SA) solvent representation.39 Various models of duplex DNA
with and without intercalation site were built using NAMOT
software.40
The optimized structures of the selected chloroethylamino-
anthraquinones were used as input files for GRID 22a software.41
Whole molecules of target DNA and ligand were considered during
calculations and all GRID parameters were kept at their default
values. Docking of ligands into intercalation site of different DNA
sequences were examined using the Glue method implemented in
GRID software. All reported interaction energies were determined
by GRID software.
Preparation of Title Compound 20: Compound 10 (50 mg,
0.18 mmol), crude 2-(3-chloropyrrolidin-1-yl)ethanamine (124 mg,
0.86 mmol), and pyridine (2 mL) for 2 h at 60 °C were the reactions
conditions. The product was afforded as a dark blue powder (15
1
mg, 15%). Mp > 300 °C dec; H NMR (DMSO/CDCl3 (1:1)) δ
13.55 (s, 2H, C(5)OH, C(8)OH), 10.45 (t, 2H, C(1)NH and
C(4)NH), 7.21 (s, 2H, C(6)H and C(7)H), 7.11 (s, 2H, C(2)H and
C(3)H), 3.81-4.12 (m, 5H), 2.85-3.05 (m, 7H), 2.35 (s, 6H, 2 ×
NCH3), 2.31-2.35 (m, 1H, ring-H), 1.81-2.15 (m, 2H, 2 × ring-
H), HCl salt was not observable in the spectrum; 13C NMR (CDCl3)
δ 184.44, 154.74, 146.09, 125.15, 124.91, 114.40, 107.65, 68.95,
62.05, 55.05, 52.01, 42.31, 37.19; FAB MS m/z (M + H)+ 473.
Anal. (C24H29N4O4Cl‚2HCl‚4H2O) C, H, N.
Most favorable positions of ligands docked into the intercalation
sites were used to model DNA-ligand conjugates, and in some

7022 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 24
Pors et al.
cases, the positions of ligands were adjusted manually to enable
the generation of a bond between the alkylating moiety and N7-
guanine. The latter was assumed to be the preferred site of alkylation
as the principal sites of attack by nitrogen mustard on DNA are
N7-guanine and N3-adenine in a ratio of about 86:16.42 Conjugates
were minimized and subjected to short stochastic dynamics simula-
tions of 10 ps at 300 K using Amber force field and GB/SA solvent
representations, and final structures were minimized until conver-
gence criteria was satisfied (0.5 × 10-2 kJ/mol/A). The hydrogen
bonding between base pairs was preserved during simulations by
using distance restraints between hydrogen acceptor and hydrogen
donor of bases. Single ligand and DNA molecules were subjected
to the same protocols.
Biology. Agarose Gel Cross-link Assay. The experiments
followed the method of Hartley et al.20 Briefly, linearized and 5′-
end 32P-radiolabelled plasmid pBR322 DNA (10 µL, ∼100 ng),
drug (25 µL to provide a range of final concentrations), and TEOA
buffer (15 µL), to give a total final volume of 50 µL, were incubated
at 37 °C for 1 h and subsequently terminated by the addition of
DNA stop solution (50 µL) to each sample. Samples were
precipitated, dried, and resuspended in strand separation buffer and
heated for 2 min at 95 °C before being chilled in ice. The samples
were electrophoresed in 0.8% agarose gel at 40 V overnight, and
the resultant gel (covered in film wrap) was dried for 2 h at 80 °C
onto one layer of Whatman 3MM paper and one layer of DE81
filter paper on a BIO-RAD 583 gel drier connected to a vacuum
pump. Autoradiography was performed using Kodak Hyperfilm for
5 h at -70 °C using a cassette with an intensifying screen.
Taq Polymerase Stop Assay. The following constituents were
added to a PCR tube: 0.2% gelatine (5 µL), 25 mM MgCl2 (10
µL), 10× buffer (10 µL), 2.5 mM dNTP mix (10 µL), drug-treated
linearized DNA (50 µL), and dH2O (8 µL). To the mixture (93
µL) was then added the labeled SRM primer (5 µL) and the Taq
polymerase (2 µL, 5 U/µL), and the final mixture was vortexed.
Each tube was covered with mineral oil, and the PCR was
programmed to with the following steps: (1) 95 °C for 5 min, (2)
95 °C for 1 min, (3) 58 °C for 1 min, (4) 74 °C for 1 min and an
additional 1 min per cycle. Steps 2-4 were repeated 29 times before
denaturing at 94 °C for 5 min, followed by 10 min at 25 °C. The
samples were then transferred into sterile eppendorfs, and the DNA
was precipitated with three volumes of ethanol (95%, 300 µL) and
NaOAc (3 M 2 µL), vortexed, and cooled in a dry ice bath for 10
min before being centrifuged for 10 min. The supernatant was
removed, and the samples were washed with ethanol (70%, 150
µL), vortexed, and spun for 10 min, and the supernatant was
removed. The wash was repeated before the samples were lyophi-
lised. Each dried sample was resuspended in formamide dye (4
µL), heated to 95 °C for 3 min, and cooled in an ice-bath to denature
the DNA. The samples were loaded into the wells of the gel, and
electrophoresis was performed in TBE buffer at 1600-2000 V
(approximately 2-3 h) at 55 °C using vertical glass electrophoresis
plates. The resulting gel was then transferred onto one Whatman
3MM filter paper and one layer of DE81, covered in film wrap,
and dried under pressure on a BIO-RAD 583 gel drier for
approximately 2 h. Once dry, the gel was exposed to a Kodak
Hyperfilm for 24 h before developing.
DNA Unwinding Assay. The 10× reaction buffer (includes 100
mM Tris-HCl, pH 7.9, 500 mM NaCl, 500 mM KCl, 50 mM
MgCl2, 1 mM EDTA, 150 µg/mL BSA, and 10 mM ATP; 2 µL),
ddH2O (15 µL), circular supercoiled pBR322 (250 ng, 1 µL), and
15 or 15-NO (2 µL) were added to an eppendorf, mixed, and kept
on ice. Each sample was carefully vortexed before incubation at
30 °C for 15 min. The reaction was stopped by addition of stop
solution (2 µL). A sucrose loading buffer (2 µL) was added, and
the mixture was vortexed and pulse spun. Electrophoresis was
performed on 0.8% submerged horizontal agarose gels in 1 × TAE
running buffer at 100 V for 3 h. Once the electrophoresis was
completed, the agarose gel was submerged in dH2O (enough to
cover the gel) containing EtBr (30 µL) for 30 min. The EtBr
containing water was discarded, and the gel was washed with water.
The gel was observed under UV fluorescence, and the result was
photographed.
Cytotoxicity Studies. The effect of the aminoanthraquinones on
viability of human ovarian carcinoma cell lines was used as a
measure of cytotoxicity. The parent cell line A2780 and the resistant
variants A2780AD and A2780/cp70 were used to study the
cytotoxic activity of the novel compounds. All cell lines were
maintained as previously described.43 The novel compounds were
dissolved in DMSO and cell culture medium such that the final
DMSO concentration was less than 0.1%. Aliquots of 20 µL were
added to the cells to provide a range of concentrations from 0.1-
1000 nM. The cell/drug combinations were incubated for 4 days
at 37 °C. Cells were incubated with 0.5 mg ml-1 MTT solution for
4 h at 37 °C in a humidified 5% CO2 atmosphere. The arrest of
cell growth was determined as previously reported by Mosmann.44
Acknowledgment. We thank Yorkshire Cancer Research
and Cancer Research U.K. for support.
Supporting Information Available: Data of elemental analysis
of aminoanthraquinones. This material is available free of charge
via the Internet at http://pubs.acs.org
References
(1) Lown J. W. Anthracycline and anthraquinone anticancer agents:
Current status and recent developments. Pharmacol. Ther. 1993, 60,
185-214.
(2) Stefanska, B.; Dzieduszycka, M.; Martelli, S.; Borowski, E. Synthesis
of nonsymmetrically substituted 1,4-bis[(aminoalkyl)amino]anthracene-
9,10-diones as potential antileukemic agents. J. Med. Chem. 1989,
32, 1724-1728.
(3) Krapcho, A. P.; Landi, J. J.; Shaw, K. J.; Phinney, D. G.; Hacker,
M. P.; McCormack, J. J. Synthesis and antitumor activities of
unsymmetrically substituted 1,4-bis[(aminoalkyl)amino]anthracene-
9,10-diones and related systems. J. Med. Chem. 1986, 29, 1370-
1373.
(4) Johnson, R. K.; Zee-Cheng, R. K.; Lee, W. W.; Acton, E. M.; Henry,
D. W.; Cheng, C. C. Experimental antitumor activity of amino-
anthraquinones. Cancer Treat. Rep. 1979, 63, 425-439.
(5) Zee-Cheng, R. K. Y.; Mathew, A. E.; Xu, P.; Northcutt, R. V.; Cheng
C. C. Structural modification study of mitoxantrone (dihydroxy-
ethylaminoanthraquinone). Chloro-substituted mono- and bis[(amino-
alkyl)amino]anthraquinones. J. Med. Chem. 1987, 30, 1682-1686.
(6) Murdock, K. C.; Child, R. G.; Fabio, P. F.; Angier, R. B.; Wallace,
R. E.; Durr, F. E.; Citarella, R. V. Antitumor agents. 1. 1,4-Bis-
[(aminoalkyl)amino]-9,10-anthracenediones. J. Med. Chem. 1979, 22,
51-60.
(7) Krapcho, A. P.; Zelleka, G.; Avery, K. L.; Vargas, K. J.; Hacker,
M. P. Synthesis and antitumor evaluations of symmetrically and non-
symmetrically substituted 1,4-bis[(aminoalkyl)amino]anthracene-9,10-
diones and 1,4-bis[(aminoalkyl)amino]-5,8-dihydroxyanthracene-
9,10-diones. J. Med. Chem. 1991, 34, 2373-2380.
(8) Ueda, K.; Yoshida, A.; Amachi, T. Recent progress in P-glycoprotein
research. Anticancer Drug Des. 1999, 14, 115-121.
(9) Leslie E. M.; Deeley R. G.; Cole S. P. Multidrug resistance
proteins: Role of P-glycoprotein, MRP1, MRP2, and BCRP
(ABCG2) in tissue defense. Toxicol. Appl. Pharmacol. 2005, 204,
216-237.
(10) Li, T. K.; Liu, L. F. Tumor cell death induced by topoisomerase-
targeting drugs. Annu. ReV. Pharmacol. Toxicol. 2001, 41, 53-77.
(11) Chikamori, K.; Grabowski, D. R.; Kinter, M.; Willard, B. B.; Yadav,
S.; Aebersold, R. H.; Bukowski, R. M.; Hickson, I. D.; Andersen,
A. H.; Ganapathi, R.; Ganapathi, M. K. Phosphorylation of serine
1106 in the catalytic domain of topoisomerase II alpha regulates
enzymatic activity and drug sensitivity. J. Biol. Chem. 2003, 278,
12696-12702.
(12) Pors, K.; Paniwnyk, Z.; Ruparelia, K. C.; Teesdale-Spittle, P. H.;
Hartley, J. A.; Kelland, L. R.; Patterson, L. H. Synthesis and
biological evaluation of novel chloroethylaminoanthraquinones with
potent cytotoxic activity against cisplatin-resistant tumor cells. J. Med.
Chem. 2004, 47, 1856-1859.
(13) Pors, K.; Paniwnyk, Z.; Teesdale-Spittle, P. H.; Plumb, J. A.;
Willmore, E.; Austin, C. A.; Patterson, L. H. Alchemix, A novel
alkylating anthraquinone with potent activity against anthracycline
and cisplatin resistant ovarian cancer. Mol. Cancer Ther. 2003, 2,
607-610.
(14) Lee, C. M.; Tannock, I. F. Inhibition of endosomal sequestration of
basic anticancer drugs: Influence on cytotoxicity and tissue penetra-
tion. Br. J. Cancer 2006, 94, 863-869.

Synthesis of Chloroalkylaminoanthraquinones
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 24 7023
(15) Patterson, L. H. Bioreductively activated antitumor N-oxides: The
case of AQ4N, a unique approach to hypoxia-activated cancer
chemotherapy. Drug Metab. ReV. 2002, 34, 581-592.
(16) Pors, K.; Plumb, J. A.; Brown, R.; Teesdale-Spittle, P. H.; Searcey,
M.; Smith, P. J.; Patterson, L. H. Development of nonsymmetrical
1,4-disubstituted anthraquinones that are potently active against
cisplatin-resistant ovarian cancer cells. J. Med. Chem. 2005, 48,
6690-6695.
(17) Greenhalgh, C. W.; Hughes, N. The reaction of leucoquinizarins with
alkylenediamines. J. Chem. Soc. 1968, 1284-1288.
(18) Ponti, M.; Forrow, S. M.; Souhami, R. L.; D’Incalci, M.; Hartley, J.
A. Measurement of the sequence specificity of covalent DNA
modification by anti-neoplastic mitoxantrone agents using Taq DNA
polymerase. Nucleic Acids Res. 1991, 19, 2929-2933.
(19) Anderson, F. M.; O’Hare, C.; Hartley, J. A.; Robins, D. J. Synthesis
of new homochiral bispyrrolidines as potential DNA cross-linking
antitumor agents. Anticancer Drug Des. 2000, 15, 119-126.
(20) Hartley, J. A.; Berardini, M. D.; Souhami, R. L. An agarose gel
method for the determination of DNA interstrand crosslinking
applicable to the measurement of the rate of total “second-arm”
crosslink reactions. Anal. Biochem. 1991, 193, 131-134.
(21) Henderson, N. D.; Lacy, S. M.; O’Hare, C. C.; Hartley, J. A.;
McClean, S.; Wakelin, L. P.; Kelland, L. R.; Robins, D. J. Synthesis
of new bifunctional compounds which selectively alkylate guanines
in DNA, Anticancer Drug Des. 1998, 13 (7), 749-768.
(22) Patterson, L. H.; Craven, M. R.; Fisher, G. R.; Teesdale-Spittle, P.
H. Aliphatic amine N-oxides of DNA binding agents as bioreductive
drugs. Oncol. Res. 1994, 6, 533-538.
(23) Osborne, M. R.; Wilman, D. E.; Lawley, P. D. Alkylation of DNA
by the nitrogen mustard bis(2-chloroethyl)methylamine). Chem. Res.
Toxicol. 1995, 8, 316-320.
(24) Bakker, M.; Renes, J.; Groenhuijzen, A.; Visser, P.; Timmer-Bosscha,
H.; Muller, M.; Groen, H. J.; Smith, E. F.; de Vries, E. G.
Mechanisms for high methoxymorpholino doxorubicin cytotoxicity
in doxorubicin-resistant tumor cell lines. Int. J. Cancer 1997, 73,
362-366.
(25) Cummings, J.; Macpherson, J. S.; Meikle, I.; Smyth, J. F. Develop-
ment of anthracenyl-amino acid conjugates as topoisomerase I and
II inhibitors that circumvent drug resistance. Biochem. Pharmacol.
1996, 52, 979-990.
(26) Jansen, B. A.; Brouwer, J.; Reedijk, J. Glutathione induces cellular
resistance against cationic dinuclear platinum anticancer drugs. J.
Inorg. Biochem. 2002, 89, 197-202.
(27) Parker, R. J.; Eastman, A.; Bostick-Bruton, F.; Reed, E. Acquired
cisplatin resistance in human ovarian cancer cells is associated with
enhanced repair of cisplatin-DNA lesions and reduced drug ac-
cumulation. J. Clin. InVest. 1991, 87, 772-777.
(28) Kelland, L. R.; Barnard, C. F.; Mellish, K. J.; Jones, M.; Goddard,
P. M.; Valenti, M.; Bryant, A.; Murrer, B. A.; Harrap, K. R. A novel
trans-platinum coordination complex possessing in Vitro and in ViVo
Antitumor Activity. Cancer Res. 1994, 54, 5618-5622.
(29) Strathdee, G.; MacKeen, M. J.; Illand, M.; Brown, R. A role for
methylation of the hMLH1 promoter in loss of hMLH1 expression
and drug resistance in ovarian cancer. Oncogene 1999, 18, 2335-
2341.
(30) Brown, R.; Hirst, G. L.; Gallagher, W. M.; McIlwrath, A. J.;
Margison, G. P.; van der Zee, A. G.; Anthoney, D. A. hMLH1
expression and cellular responses of ovarian tumor cells to treatment
with cytotoxic anticancer agents. Oncogene 1997, 15, 45-52.
(31) Tercel, M.; Wilson, W. R.; Denny, W. A. Hypoxia-selective antitumor
agents. 11. Chlorambucil N-oxide: A reappraisal of its synthesis,
stability, and selective toxicity for hypoxic cells. J. Med. Chem. 1995,
38, 1247-1252.
(32) Smith, P. J.; Blunt, N. J.; Desnoyers, R.; Giles, Y.; Patterson, L. H.
DNA topoisomerase II-dependent cytotoxicity of alkylamino-
anthraquinones and their N-oxides. Cancer Chemother. Pharmacol.
1997, 39, 455-461.
(33) Patterson, L. H.; McKeown, S. R. AQ4N: A new approach to
hypoxia-activated cancer chemotherapy. Br. J. Cancer 2000, 83,
1589-1593.
(34) Mori, T. A method for producing 1,5-diamino-4,8-dihydroxy-
anthraquinone and derivatives thereof. Japanese Patent 47-14352,
1972.
(35) Krapcho, P. A.; Getahun, Z.; Avery, K. J. The synthesis of
1,4-Difluoro-5,8-dihydroxyanthracene-9,10-dione and ipso substitu-
tions of the fluorides by diamines leading to 1,4-bis[(aminoalkyl)-
amino]-5,8-dihydroxyanthracene-9,10-diones. Synth. Commun. 1990,
20, 2139-2146.
(36) Mohamadi, F.; Richards, N. G. J.; Guida, W. C.; Liskamp, R.; Lipton,
M.; Caufield, C.; Chang, G.; Hendrickson, T.; Still, W. C.
Macromodelsan integrated software system for modelling organic
and bioorganic molecules using molecular mechanics. J. Comput.
Chem. 1990, 11, 440-467.
(37) Chang, G.; Guida, W. C.; Still, W. C. An internal-coordinate Monte
Carlo method for searching conformational space. J. Am. Chem. Soc.
1989, 111, 4379-4386.
(38) Weiner, S. J.; Singh, U. C.; Kollman, P. A. An all atom force-field
for simulations of proteins and nucleic acids. J. Comput. Chem. 1986,
7, 230-252.
(39) Still, W. C.; Tempczyk, A.; Hawley, R. C.; Hendrickson, T.
Semianalytical treatment of solvation for molecular mechanics and
dynamics. J. Am. Chem. Soc. 1990, 112, 6127-6129.
(40) Tung, C. S.; Carter, E. S. Nucleic-acid modelling tool (NAMOT)s
an interactive graphic tool for modelling nucleic-acid structures.
CABIOS 1994, 10, 427-433.
(41) Goodford P. J. A computational procedure for determining energeti-
cally favorable binding sites on biologically important macromol-
ecules J. Med. Chem. 1985, 28, 849-857.
(42) Osborne, M. R.; Wilman, D. E.; Lawley, P. D. Alkylation of DNA
by the nitrogen mustard bis(2-chloroethyl)methylamine. Chem. Res.
Toxicol. 1995, 8, 316-320.
(43) Shnyder, S. D.; Hasan, J.; Cooper, P. A.; Pilarinou, E.; Jubb, E.;
Jayson, G. C.; Bibby, M. C. Development of a modified hollow fibre
assay for studying agents targeting the tumor neovasculature.
Anticancer Res. 2005, 25, 1889-1894.
(44) Mosmann, T. Rapid colorimetric assay for cellular growth and
survival: Application to proliferation and cytotoxicity assays. J.
Immunol. Methods 1983, 65, 55-63.
JM0608154