Synthesis, biological activity, and DNA-damage profile of platinum-threading intercalator conjugates designed to target adenine

Article abstract of DOI:10.1021/jm060035v

PT-ACRAMTU {[PtCl(en)(ACRAMTU)](NO₃)₂, 2; ACRAMTU = 1-[2-(acridin-9-ylamino)ethyl]-1,3-dimethylthiourea, 1, en = ethane- 1,2-diamine} is the prototype of a series of DNA-targeted adenine-affinic dual intercalating/platinating agents.

Full text of DOI:10.1021/jm060035v

3204
J. Med. Chem. 2006, 49, 3204-3214
Synthesis, Biological Activity, and DNA-Damage Profile of Platinum-Threading Intercalator
Conjugates Designed To Target Adenine
Rajsekhar Guddneppanavar,† Gilda Saluta,‡ Gregory L. Kucera,‡,§ and Ulrich Bierbach*,†,§
Department of Chemistry, Wake Forest UniVersity, Winston-Salem, North Carolina 27109, and Department of Internal Medicine,
Hematology-Oncology Section, and ComprehensiVe Cancer Center of Wake Forest UniVersity, Wake Forest UniVersity School of Medicine,
Winston-Salem, North Carolina 27157
ReceiVed January 11, 2006
PT-ACRAMTU {[PtCl(en)(ACRAMTU)](NO3)2, 2; ACRAMTU ) 1-[2-(acridin-9-ylamino)ethyl]-1,3-
dimethylthiourea, 1, en ) ethane-1,2-diamine} is the prototype of a series of DNA-targeted adenine-affinic
dual intercalating/platinating agents. Several novel 4,9-disubstituted acridines and the corresponding platinum-
acridine conjugates were synthesized. The newly introduced 4-carboxamide side chains contain H-bond
donor/acceptor functions designed to promote groove- and sequence-specific platinum binding. In HL-60
(leukemia) and H460 (lung) cancer cells, IC50 values in the micromolar to millimolar range were observed.
Several of the intercalators show enhanced cytotoxicity compared to prototype 1, but conjugate 2 appears
to be the most potent hybrid agent. Enzymatic digestion assays in conjunction with liquid chromatography-
electrospray mass spectrometry analysis indicate that the new conjugates produce PT-ACRAMTU-type DNA
damage. Platinum-modified 2′-deoxyguanosine, dG*, and several dinucleotide fragments, d(NpN)*, were
detected. One of the conjugates showed significantly higher levels of binding to A-containing sites than
conjugate 2 (35 ( 3% vs 24 ( 3%). Possible structure-activity relationships are discussed.
Introduction
The development of small molecules capable of interacting
with nuclear DNA in a sequence- and groove-specific manner
plays an important role in cancer chemotherapy.1 As a conse-
quence of our increased knowledge of the human genome and
its roles in cell-signaling pathways, anticancer drug design is
witnessing a paradigm shift away from chemically promiscuous
electrophiles toward agents targetable to specific DNA-associ-
ated processes. Recent developments in the field of DNA-
Figure 1. Structures of the drug prototypes.
directed chemotherapy include agents that bind the promoter
sequences of specific genes or sequences of specific nucleobase
thiourea (ACRAMTU (1), Figure 1). PT-ACRAMTU damages
content, as well as molecules that target nonclassical DNA
DNA by a dual mechanism involving intercalation and mono-
structures and DNA-enzyme complexes.2,3 All of the above
functional platination, but it does not induce cross-links.10 The
damage mechanisms interfere with the structural and functional
hybrid agent produces adducts with guanine (G) and adenine
integrity of the genome to alter or disrupt processes vital to
(A), specifically in the sequences 5′-CG*A, 5′-CGA*, and 5′-
cell-cycle progression, ultimately promoting cancer cell death.
Platinum drugs are among the most successful anticancer
agents, and cisplatin [cis-diamminedichloroplatinum(II)] has
TA*, where the asterisks denote the platinated bases.11,12 This
contrasts the DNA damage profile of cisplatin, which predomi-
nantly targets 5′-GG and 5′-AG base steps.13 NMR studies
become a mainstay in combination therapies, despite its serious
indicate that the sequence and groove specificity of the
clinical limitations.4 Several of these limitations are thought to
ACRAMTU moiety is the source of the unique binding profile
be linked to cisplatin’s reactivity and lack of target specificity.5
of PT-ACRAMTU.14 In essence, the intercalator “hijacks”
In a search for structurally novel non-cisplatin-type metallo-
platinum away from G-N7 in the major groove, the thermody-
namically and kinetically preferred binding site of divalent
platinum, ultimately resulting in a DNA-damage profile comple-
pharmacophores aimed at overcoming cisplatin’s drawbacks,
we have discovered a new class of platinum-intercalator
conjugates that show promising in vitro activity at (sub)-
mentary to that of the clinical drug. Most significantly, PT-
micromolar concentrations in various solid tumors.6-9 The
ACRAMTU proves to be the first case of a platinum-based agent
prototype, PT-ACRAMTU (2, Figure 1), was generated by
that targets the minor groove of B-form DNA. The distinct
simple chemical linkage between the fragment [PtCl(en)]+ (en
groove specificity of the 9-substituent in ACRAMTU leads to
) ethane-1,2-diamine) and the sulfur atom of the 9-aminoacri-
dine derivative, 1-[2-(acridin-9-ylamino)ethyl]-1,3-dimethyl-
platination of A-N3 in the minor groove in ∼5-10% of the
adducts.15 Furthermore, we demonstrated using electrophoretic
mobility shift assays (EMSA) that these adducts, when localized
* Address correspondence to this author. Telephone: 336-758-3507.
to 5′-TA sequences, inhibit the association of human TATA
binding protein (hTBP) with its recognition sequence, a mech-
anism previously known for covalent and noncovalent interca-
lators and groove binders.16 The potential biological implications
of this type of damage are paramount: A-N3 adducts, if
Fax: 336-758-4656. E-mail: bierbau@wfu.edu.
† Department of Chemistry, Wake Forest University.
‡ Department of Internal Medicine, Wake Forest University School of
Medicine.
§ Comprehensive Cancer Center of Wake Forest University, Wake Forest
University School of Medicine.
10.1021/jm060035v CCC: $33.50 © 2006 American Chemical Society
Published on Web 05/03/2006

Platinum-Threading Intercalator Conjugates
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11 3205
Figure 2. Design rationale for novel 4,9-disubstituted ACRAMTU derivatives. (a) Regioselective threading intercalation and dual mechanism of
groove recognition and platination (the red sphere represents a residue capable of H-bond formation, highlighted in red in b and c). (b) A-recognizing
groups. (c) G-recognizing group.
frequently formed in genomic DNA, can be expected to
contribute to the cytotoxic effect of PT-ACRAMTU by interfer-
ing with transcription initiation.17
predicted to be the preferred binding site in the minor groove.
Platination of minor-groove G-N3 should be disfavored due to
the steric hindrance produced by the exocyclic N(2)H2 group.
Consequently, the targeting of the minor groove at A-containing
base-pair steps can be expected to direct platinum away from
its “natural” target, G-N7, and result in increased levels of A
adducts.
The specific approach taken to tune the DNA binding of PT-
ACRAMTU derivatives using 4,9-disubstituted acridines is
summarized in Figure 2.
As part of our efforts to delineate structure-activity relation-
ships (SAR) within PT-ACRAMTU-type conjugates, we have
now begun to extend our studies to analogues containing 4,9-
disubstituted acridines derived from ACRAMTU. We hypo-
thesize that modification of the planar chromophore with
suitably functionalized side chains in the 4-position might
provide a mechanism for tuning the sequence specificity of
platination through additional noncovalent drug-nucleobase
interactions in the DNA major groove. Here, we report the
synthesis and biological activity of several of these “second-
generation” ACRAMTU derivatives and their corresponding
platinum conjugates. The DNA adduct profiles of the new
conjugates were also studied using an enzymatic digestion assay
in conjunction with in-line high-performance liquid chroma-
tography-electropray mass spectrometry (LC-ESMS) analysis.
Groove Selectivity. In the proposed pre-intercalation complex
the acridine chromophore is aligned with the base pairs’ long
dimensions at the intercalation pocket. In this geometry, the
9-substituent carrying platinum is directed toward the minor
groove and the side chain in the 4-position protrudes into the
major groove (Figure 2a), in accordance with the directionality
of intercalation usually observed in the DNA complexes of
acridine-based drugs, such as the acridine-4-carboxamides.18,19
It is noteworthy to mention that ACRAMTU itself has been
unequivocally demonstrated to intercalate DNA with the thiourea
side chain residing in the minor grooVe.14 (In a similar fashion,
Denny et al. have synthesized major-groove and minor-groove
targeted alkylating agents by tethering nitrogen mustard moieties
to the 4- and 9-position of the acridine chromophore, respec-
Results
Design. PT-ACRAMTU (2) induces damage at multiple
nucleophilic sites in double-stranded DNA. Using a combination
of enzymatic and newly developed acidic digestion assays we
were able to demonstrate that in calf thymus DNA, for instance,
complex 2 forms monoadducts with G (∼80%) and A (∼20%).11
While N7 was the only targeted site in G, platination of A occurs
at the N7 (∼11%), N3 (∼7%), and N1 (∼2%) positions.15 All
of these non-cisplatin-type adducts have to be considered
potential cytotoxic lesions. (Reactions with native DNA were
performed at drug-to-nucleotide incubation ratios (ri) as high
as 0.4.15 It should be noted that while these conditions were
appropriate for the detection and semipreparative chromato-
graphic isolation of the major adducts, they may not reflect the
true binding selectivity at physiological drug concentrations.)
To investigate the inter-relationship between chemical structure,
biological activity, and DNA damage profile, we are pursuing
various strategies of modulating the nucleobase specificity of
platination in PT-ACRAMTU-based conjugates. In particular,
we are interested in chemical modifications of the prototype
that might lead to enhanced targeting of adenine-containing
sequences. The design applied in this study using threading
intercalators can be rationalized as follows: Platination of
nucleobase nitrogen occurs from a pre-intercalated state. The
dynamic properties (“on/off” rates) and the base-pair-step and
regiospecificity of this initial drug-DNA complex determine
the base and site-selectivity of platinum binding. Of the two
purine bases, G will be the preferred target in the major groove
with Pt binding occurring at N7, whereas N3 of A can be
tively.20,21
)
Base Recognition. The design was inspired by the hydrogen-
bonding interactions observed between purine bases and amino
acid side chains in DNA-protein complexes. Crystal structures
of proteins bound to the DNA major groove show that adenine
has a high propensity for forming specific contacts with mixed
donor-acceptor groups, such as the hydroxyl group in serine/
threonine/tyrosine and the amide group in arginine/glutamine.22
In contrast, hydrogen-bond donor groups, such as amine in
lysine and guanidinium in asparagine, recognize guanine.22 In
this study, various functional groups were incorporated into the
4-carboxamide residues of the new ACRAMTU derivatives to
mimic base-specific amino acid interactions with the ultimate
goal of modulating the base selectivity of intercalation (and,
ultimately, platination) via major-groove read-out by the drug
molecule. Hydroxyl, urea, and methyl ester were chosen to
specifically recognize A. A derivative containing a G-affinic
dimethylammonium group was also included in the study
(Figure 2b,c). There is precedent for the latter binding mode in
the crystal structures of the DNA complexes formed by acridine-
4-carboxamide anticancer agents (DACA), which preferentially
intercalate into the 5′-CG/CG base-pair step.18,19
Molecular Modeling. To investigate the feasibility of the
proposed mode of combined threading intercalation and major-

3206 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11
GuddneppanaVar et al.
Scheme 1. Synthesis of Threading Intercalators and Platinum-Threading Intercalator Conjugatesa
a Reagents and conditions: (i) I-IV (hydrochloride salts) or NH2(CH2)2OH or NH2(CH2)2N(CH3)2, Et3N/CH2Cl2/0-5 °C; (ii) (1) phenol/V/∆; (2) 2 M
NaOH; (3) AcOH/HCl; (iii) 2 M NH4OH; (iv) K2CO3/MeOH/H2O; (v) CH3NCS/EtOH/∆; (vi) 1 M HNO3; (vii) (1) [PtCl2(en)], 1 equiv. AgNO3/DMF/dark,
-
rt; (2) 8c-g, rt. The counterion in 5a-e, g is Cl-, while the counterion in 8a-g and 9c-g is NO3
.
Figure 3. Views along the helical axis of AMBER-minimized models of duplexes containing threading intercalators. Only the intercalated drug
molecules and the sandwiching base pairs (green) are shown. The nucleobase involved in hydrogen bonding with the drug molecule is highlighted
in orange. The yellow solid lines indicate drug-DNA hydrogen bonding. (a) Derivative 8f bound to the 5′-TA/TA step. Donor-acceptor distances:
Odrug‚‚‚N7A, 2.93 Å; N6A‚‚‚Odrug, 3.03 Å. (b) Derivative 8b bound to the 5′-TA/TA step. Donor-acceptor distances: N6A‚‚‚Odrug, 2.95 Å; N7A‚‚
‚Ndrug, 2.87 Å. (c) Derivative 8e bound to the 5′-CG/CG step. Donor-acceptor distance: Ndrug‚‚‚O6G, 2.81 Å.
groove hydrogen bonding, molecular mechanics calculations
were performed on DNA complexes containing several of the
proposed ACRAMTU derivatives (for compound numbering
used in this discussion, see Scheme 1). As starting structures,
AMBER models of ACRAMTU bound to octamer duplexes
previously derived from NMR data14 were used. Specifically,
intercalation of the hydroxyl- and urea-based derivatives (8f and
8b) into the 5′-TA/TA base-pair step and a dimethylammonium-
functionalized derivative (8e) into the 5′-CG/CG step was
studied. During the early stages of the calculations the proposed
hydrogen bonds were enforced by introducing suitable drug-
DNA distance constraints. After removal of the constraints, the
freely minimized structures relaxed into the geometries shown
in Figure 3. In all of the models, simultaneous threading
intercalation and drug-nucleobase hydrogen bonding proves
feasible without causing major conformational changes in the
host duplex. Watson-Crick hydrogen bonding persists in the
base pairs on both sides of the acridine chromophores. In all
cases, the conformation of the 4-carboxamide side chains is
stabilized by an intramolecular hydrogen bond between the
amide oxygen and the protonated endocyclic acridine nitrogen
(N10), a feature previously established for DNA-bound DACA-

Platinum-Threading Intercalator Conjugates
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11 3207
Table 1. Cytotoxicity Data for 1, 2, 8a-g, and 9c-g (IC50, µM)a
platinum-
type intercalators.18,19 This arrangement positions the hydroxyl
group in 8f and the ammonium group in 8e favorably for donor-
acceptor interactions (Figure 3a,c) with minimal rearrangement
of the chromophores relative to the geometry observed for
ACRAMTU. This is in contrast to 8b, whose planar unit has to
twist significantly relative to the helical axis to allow dual
hydrogen bonding between the urea group and adenine (Figure
3b; see figure caption for structural details). Another critical
difference between the urea derivatives and 8e,f is the require-
ment for an extended polymethylene chain. While an ethylene
group produces the ideal geometry in 8e and 8f, a minimum of
four carbon atoms are required to allow strainless hydrogen
bonding of the urea moiety in 8b.
acridines
HL-60b
H460b
acridines
HL-60
H460
1
11.5
>100
>100
>50
0.39
1.4
9.5
2
2.8
-
-
>50
1.5
10.6
39.7
30.2
0.26
-
c
8a
8b
8c
8d
8e
8f
>100
>100
31.9
3.1
2.2
41.1
17.6
c
-
9c
9d
9e
9f
26.7
3.7
6.1
3.8
8.4
27.4
29.5
8g
9g
a MTS colorimetric cell proliferation assay; IC50 values are given for
72-h drug exposure and are average values of at least three experiments.
b HL-60, leukemia; H460, lung cancer. c Platinum conjugates of compounds
8a and 8b were not synthesized.
Chemistry. The synthetic strategy for the preparation of the
threading intercalators and the corresponding conjugates is
outlined in Scheme 1. The first step involved reaction of the
reactive 4-C(O)X group in 9-chloroacridine-4-carbonyl chloride
(3′) or the analogous 9-chloro-p-nitrophenol ester (3′)23 with
the suitably functionalized primary amines to produce the
9-chloroacridine-4-carboxamide derivatives 4a-e in 40-60%
yield. Reactions performed with the p-nitrophenolato leaving
group usually gave better yields (∼80% for 4f). The urea-
functionalized amines24 I-IV were synthesized from the
monoprotected 1,2- and 1,4-diamines by reaction of the
unprotected amino group with ethyl isocyanate or phenyl
isocyanate and isolated as the hydrochloride salts after removal
of the BOC groups. Reaction of 4a-f with (2-aminoethyl)-
methylcarbamic acid tert-butyl ester (V) in the presence of
excess phenol and base and subsequent deprotection of the
secondary amino group resulted in intermediates 5a-e and 5g
in form of their hydrochloride salts (35-45% yield for two
steps). Under the esterifying conditions of the BOC group
removal, 5g is formed from 4f rather than 5f. The hydrochloride
salts 5a-e and 5g were converted directly to the corresponding
free bases, 6a-e and 6g, using dilute ammonium hydroxide
solution, while hydrolysis of 5g under strongly basic conditions
afforded the desired 4-(2-hydroxoethyl)carboxamide derivative,
6f. The secondary amino group in intermediates 6 was reacted
with methylisothiocyanate to give the thioureas 7a-g, which
were converted to the corresponding hydronitrate salts 8a-g
(40-55% yield for two steps). Finally, the acridinium salts 8c-g
were reacted with the monoactivated form of [PtCl2(en)], [PtCl-
(DMF-O)(en)]+, to generate the platinum-acridine conjugates
9c-g.9 Due to the high affinity of platinum for thiourea sulfur,
a substantial amount of the corresponding platinum-bis-
(acridine) complexes25 were present in the reaction mixtures.
Size-exclusion chromatography using nonaqueous eluents to
avoid hydrolysis of the chloro ligands proved to be the only
viable method for purifying several of the conjugates, which
were obtained in 30-50% yield. Analogues 9a and 9b were
not synthesized because of the marginal biological activity of
the free acridine precursors (vide infra). The compounds
included in the cell viability assays were fully characterized by
1H and 13C NMR spectroscopy, elemental analyses, and in-line
LC-ESMS analysis (see Experimental Section and Supporting
Information for details). The urea-based derivatives 8a-d and
9c-d, especially those containing extended linker chains (n )
4) in the 4-position, showed limited solubility in aqueous buffers
compared to the prototypes 1 and 2 and derivatives 8/9e-g.
ACRAMTU (1) was moderately active in both cell lines. A
significant enhancement in cytotoxicity is observed for PT-
ACRAMTU (2), especially in the lung cancer line, where the
conjugate is approximately 35-fold more active than the free
acridine, a trend previously observed in clonogenic survival
assays9 performed with this cell line. The new derivatives 8a-g
show a broad range of activities with low micromolar to low
millimolar IC50 values. While analogues 8a-c, 8f, and 8g show
no cytotoxic response at drug concentrations as high as 100
µM or prove to be significantly less active than 1, compounds
8d and 8e are more active than the prototype. In particular,
derivative 8d showed a cytotoxic effect in HL-60 cells in the
low micromolar range (IC50 ) 0.39 µM), previously observed
only for platinum conjugates such as 2 and a few of its direct
derivatives. Two interesting effects are observed within the
subgroup of urea analogues, 8a-d, where both the presence of
an n ) 4 and a phenyl group on the terminal urea nitrogen
seem to be a prerequisite for appreciable cytotoxicity. Charac-
teristically, both extension of the 4-carboxamide side chain in
8c by two carbon atoms and replacement of the ethyl with the
phenyl group in derivative 8b to yield compound 8d lead to a
dramatic increase in cytotoxicity by >100-fold on the basis of
the inhibitory concentrations determined in both cell lines. From
the molecular modeling study it seems plausible that the inferior
activity of the more constrained derivatives, 8a and 8c, may be
a consequence of their inability to form stable drug-DNA
complexes via threading intercalation. It is unclear what causes
the increase in activity by the simple ethyl/phenyl substitution.
In the platinum conjugates 9c and 9d, an increase in potency is
observed with an increase in 4-carboxamide chain length, in
agreement with the trend observed for the corresponding free
acridines. The majority of the conjugates generated, however,
show activity similar to that established for the platinum-free
intercalators. A significant effect of the metal on cytotoxicity
levels comparable to that established for the prototypical
complex is observed for conjugate 9f; however, only the H460
cell line appears to be sensitive to this type of structural
modification.
DNA Damage. To gain insight into the effects of the
structural modifications on the DNA-damage profiles of the
newly synthesized conjugates, enzymatic digests of native DNA
treated with selected derivatives were qualitatively and quan-
titatively analyzed by in-line LC-ESMS using an experimental
setup described previously.11 Briefly, calf thymus DNA was
incubated with 2 and 9d-f, unbound drug removed from the
reaction mixtures by microdialysis, and the platinated DNA
samples enzymatically digested and dephosphorylated using an
approved “cocktail” of endonucleases and alkaline phosphatase.
The adducts were then separated by reverse-phase chromatog-
raphy and identified by ESMS run in negative ion mode. In all
Biological Activity. The new acridines and platinum-
acridines along with the drug prototypes were tested in HL-60
leukemia and NCI-H460 lung cancer cell lines using a cell
viability assay. The results are presented in Table 1. In the MTS-
based colorimetric assay, the prototypical acridine derivative

3208 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11
GuddneppanaVar et al.
Table 2. Summary of Platinum-Acridine-Modified DNA Fragments
tide fragments, but no (undigested) trinucleotides and higher
modified oligonucleotides were detected in the mixtures. The
results of this assay are summarized in Table 2 and Figure 4.
Previously, three adducts of conjugate 2 were detected: dG*
(dG ) 2′-deoxyguanosine), 5′-GpA*, and 5′-TpA*, where the
asterisk represents the platinum-acridine fragment [Pt(en)-
Observed in Enzymatic Digests
conjugates
fragments
dG* a
(GpA/ApG)*
(TpG/GpT)*
(TpA/ApT)*
(ApA)*
molecular ion
% abundance
m/z
2
[M - 4H]-
[M - 3H]-
[M - 3H]-
[M - 3H]-
[M - 3H]-
[M - 4H]-
[M - 3H]-
[M - 3H]-
[M - 3H]-
[M - 3H]-
[M - 3H]-
[M - 5H]-
[M - 4H]-
[M - 4H]-
[M - 4H]-
[M - 4H]-
[M - 3H]-
[M - 3H]-
[M - 3H]-
[M - 3H]-
[M - 3H]-
66(8)
13(2)
10(3)
7(2)
843.3
1156.3
1147.3
1131.3
1140.3
1076.5
1389.6
1380.5
1364.6
1373.5
1405.5
957.4
1270.5
1261.5
1245.5
930.4
1243.5
1234.6
1218.4
1227.7
1259.4
(1)]3+ 11
The sequence specificity (e.g., 5′-TpA* vs 5′-A*pT)
.
4(2)
9d
dG*
45(4)
15(1)
13(0)
11(1)
9(3)
and base specificity (e. g., in the 5′-GpA sequence) of the
damage in the latter dinucleotide fragments could be fortuitously
determined from tandem MS data,11 which were confirmed by
high-resolution structural and footprinting techniques.12,15,26
While the data acquired in the present study did not provide
this structural detail, the nucleobase content of the modified
dinucleotide fragments could be unambiguously established on
the basis of their expected molecular ions, [M - nH]-, observed
in negative ion mode mass spectra (Table 2; for HPLC traces
and mass spectra, see Supporting Information). The current study
confirms the DNA damage produced by conjugate 2. In addition
to the three modified fragments detected in previous assays,
two new dinucleotide adducts are observed, (TpG/GpT)* (9%)
and (ApA)* (4%). While the former damage site is quite
unexpected, the latter minor adduct confirms the damage within
(GpA/ApG)*
(TpG/GpT)*
(TpA/ApT)*
(ApA)*
(GpG)*
dG*
(GpA/ApG)*
(TpG/GpT)*
(TpA/ApT)*
dG*
(GpA/ApG)*
(TpG/GpT)*
(TpA/ApT)*
(ApA)*
7(0)
9e
9f
46(12)
13(4)
32(2)
9(1)
58(8)
17(2)
11(2)
8(1)
3(1)
3(0)
(GpG)*
a The asterisks denote the moieties [Pt(en)(1/8d-f)]3+/4+
.
the cases, enzymatic degradation of the drug-treated DNA
resulted in platinum-modified guanosine and multiple dinucleo-
Figure 4. (a-d) Relative abundance of platinum-acridine-modified DNA fragments observed in enzymatic digests of calf thymus DNA treated
with conjugates 2 and 9d-f. Drug-modified mononucleoside and dinucleotide fragments were separated and characterized by in-line LC-ESMS.
Relative abundances were determined by integrating the intensities of HPLC peaks monitored at acridine-specific wavelengths using diode-array
detection. Plotted values are the average of three individual incubations/digestions, and error bars represent (1 standard deviation. Assignments for
dinucleotide fragments are non-sequence- and non-base-specific. The notation (GpA/ApG)*, for instance, indicates a fragment in which either G
or A is platinated. The asterisks represent the platinum moieties [Pt(en)(1/8d-f)]3+/4+. (e) Percent targeted A-containing dinucleotide sites by each
drug. (f) Plot of relative total degree of modification by each drug (prototype 2 ) 100), based on all platinum-containing fragments.

Platinum-Threading Intercalator Conjugates
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11 3209
Figure 5. (a) Reverse-phase HPLC profile monitored at 414-430 nm of an acidic digest of a double-stranded decamer treated with conjugate 9d.
The arrow indicates the major drug-modified fragment, which elutes at a retention time of 52.7 min. (b) Positive-ion electrospray mass spectrum
of the major adduct. The inset shows the structure of the adduct and fragments resulting from CAD. “A” stands for adenine nucleobase.
the sequence 5′-TAA detected by transcriptional footprinting.12
In this sequence a stop site was observed at the 3′-A base,
suggesting platination of this nucleobase and intercalation of
the acridine chromophore between the purine bases, which
apparently protects the phosphodiester linkage in the ApA
sequence from enzymatic cleavage. (This mode of sequence-
specific intercalation has been discussed to explain the previ-
ously detected modified fragments.11) In 24% of the adducts
identified, prototype 2 targets A-containing dinucleotide steps.
Upon first inspection, the type of damage induced by conjugates
9d-f is similar to that of 2 based on the fragments identified
(Figure 4b-d). In the case of 9d and 9f, a small amount of
undigested (GpG)* is observed with abundances of 7% and 3%,
respectively, and the adduct distribution differs significantly
from that of the parent complex. Most significantly, the amount
of dG* is notably reduced. This effect is most pronounced for
9d and 9e, for which this species accounts for 45% and 46% of
the total array of adducts, respectively. In the case of 9d, a
significant increase in modified A-containing fragments is
observed relative to complex 2, accounting for 35% of the total
adducts. In contrast, conjugate 9e containing the G-directed
dimethylaminoethyl residue shows the lowest affinity for A
(22% of adducts; Figure 4e) but, curiously, produces a large
amount of (TpG/GpT)* (32% of adducts). In 9e, simultaneous
platination of G-N7 and H-bond formation by the NHMe2 group
with the same G in the major groove are incompatible. Thus, it
may be speculated that in the TpG/GpT sequence platination
of G occurs at N3 in the minor groove. Additional analytical
and high-resolution structural work, however, is required to
explore this possibility. Complex 9f shows an adduct profile
qualitatively and quantitatively similar to that of 2. Finally, the
overall DNA binding levels produced by derivatives 9d and 9e
are significantly reduced compared to those observed for
analogues 2 and 9f, based on the total amount of adducts
detected for each agent (Figure 4f).
incubated at limiting concentrations of drug with the model
duplex d(GCGATATCGC)2, which contains several of the
predicted high-affinity sites of this agent. The modified sequence
was then digested at 60 °C/pH 2 and the mixture analyzed by
LC-ESMS as described previously. The major adduct observed
in the HPLC traces (Figure 5a) was indeed identified as the
monofunctional A adduct [Pt(en)(8d)(A)]3+ ([M]3+). Electro-
spray mass spectra of this fraction acquired in positive-ion mode
in conjunction with fragmentation by in-source collisionally
activated dissociation (CAD) (Figure 5b) was used to establish
the nature of this species. Two peaks are observed for the intact
adduct, [M - 2H]+ (947 m/z) and [M - 3H,Na+]+ (968 m/z).
Fragment ions are also observed resulting from loss of A base,
[M - A - 2H]+ (811 m/z), and loss of both A and en ligand,
[M - A - en - 2H]+ (753 m/z). All of these ions show the
characteristic isotope pattern of platinum. In addition, a peak is
observed at 558 m/z, which is assigned to the protonated free
acridine, [8d]+, resulting from dissociation of the Pt-S bond
in conjugate 9d. On the basis of the ESMS data, it is not possible
to discern the platinated nitrogen in A (N1, N3, or N7).
Discussion
The major goal of this study was to examine the possibility
of tuning the DNA interactions of a novel platinum-acridi-
nylthiourea pharmacophore through chemical modification of
the drug prototype, PT-ACRAMTU (2). The targeting of DNA-
binding molecules to A-rich sequences of the genome that are
critical to cell function, such as TATA-based gene promoters27-30
and AT-islands,31 is an unexplored opportunity in platinum
antitumor chemistry. The divalent metal has a high binding
preference for G, and the N7 position of this base in the major
groove is the primary binding site of virtually all platinum-
based drugs in clinical use or clinical trials.32 Inspired by the
discovery of conjugate 2, which breaks this long-standing
paradigm by producing a high percentage of monofunctional
A adducts, structural changes were made to the prototype that
would result in more selective binding to A-containing se-
quences. The strategy employed here is based on the supposition
that the sites of irreversible platinum damage can be controlled
by the sequence and groove specificity of the threading
intercalators. While this binding mode is unknown in platinum-
DNA chemistry, several examples exist of bioactive intercala-
tor-alkylator conjugates that act through this mechanism: the
pluramycins33-37 and psorospermins38,39 are cytotoxic com-
pounds in preclinical development that intercalate double-
While the results of the enzymatic digestion convincingly
demonstrated that all of the new conjugates target A-containing
sites, this assay did not provide direct evidence for coordinative
attachment of platinum to A in the dinucleotide fragments
isolated, such as (TpA/ApT)* and (GpA/ApG)*. To demonstrate
that A residues are platinated, a previously designed depurination
assay15 was employed. The assay takes advantage of the fact
that platinum-modified A can be easily liberated from drug-
treated DNA samples under mildly acidic conditions. In this
experiment, conjugate 9d, the derivative exhibiting the highest
affinity for A-containing sequences (see Figure 4e), was

3210 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11
GuddneppanaVar et al.
stranded DNA sequence- and groove-specifically to produce
epoxide-mediated covalent adducts with G-N7 in the major
groove.
IC50 data determined for 8d and 9d in the H460 cell line). In
summary, while there appear to exist no obvious correlations
between DNA-damage profile and biological activity, several
interesting structure-activity relationships are observed that may
prove useful in our continued search for the DNA-processing
enzymes involved in the mechanism of these novel hybrid
agents. In particular, the ultimate role of transcription inhibition
in the mechanism of these conjugates has yet to be established.
Furthermore, it will be of interest to assess if the new hybrid
agents might act through a topoisomerase-dependent mechanism
similar to the mixed alkylating-intercalating psorospermin.38,39
In addition to the differences in the DNA-damage profiles, other
factors may exist, such as differential drug uptake and efflux,
which may limit the cytotoxic effect of the conjugates. Future
studies will address these issues.
While the bioanalytical data suggest that the DNA-damage
profile of the new conjugates is altered compared to that of
complex 2, at least for one of the derivatives synthesized, 9d,
more drastic modifications seem to be necessary to produce a
truly A- and minor groove-specific platinum agent. Given the
distinct G-N7 specificity of this metal, this is not a trivial task.
Characteristically, previous attempts to direct platinum into the
minor groove by tethering the metal to distamycin were
unsuccessful.40,41 This contrasts the situation for minor-groove
alkylation, which can be readily induced by tethering an
appropriate electrophile to a minor-groove-specific agent.
9-Anilinoacridine-nitrogen mustard conjugates20,42 and N-
methylpyrrolecarboxamide dipeptide modified with a methyl
sulfonate ester (“Me-lex”)43 are examples of agents that have
been successfully designed to produce covalent adducts in the
minor groove, especially with A-N3. In the case of PT-
ACRAMTU-type conjugates, a combination of (threading)
intercalation and more efficient sequence recognition in the
minor groove by polyamide-type structures might provide a
means of producing less promiscuous agents with more predict-
able groove and long-range sequence specificity. These studies
are currently underway.
Experimental Section
Synthetic Chemistry and Product Characterization. 1H NMR
spectra for the target compounds and intermediates were recorded
on Bruker Avance 300 and DRX-500 instruments operating at 500
and 300 MHz, respectively. 13C NMR spectra were recorded on a
Bruker Avance 300 instrument operating at 75.5 MHz. Chemical
shifts (δ) are given in parts per million (ppm) relative to internal
standard trimethylsilane (TMS) or 3-(trimethylsilyl)-1-propane-
sulfonic acid sodium salt (DSS) for samples in D2O. Coupling
constants (J) are in Hz. 1H chemical shift assignments for the target
compounds 8a-f and 9c-f were assisted by gradient COSY and
1H-detected gradient HSQC and HMBC spectra recorded on a
Bruker DRX-500 MHz spectrometer. Elemental analyses were
performed by Quantitative Technologies Inc., Madison, NJ. All
reagents were used as obtained from commercial sources without
further purification unless indicated otherwise. Solvents were dried
and distilled prior to use. Sephadex LH-20 resin was purchased
from Amersham Biosciences. The HPLC/MS data of compounds
8a and 9c-f were collected on an Agilent Technologies 1100LS/
MSD Trap instrument equipped with an atmospheric pressure
electrospray ionization system and a multiwavelength diode array
detector.
Several interesting structure-activity relationships are ob-
served in compounds 8a-d, but no direct relationship seems
to exist between the absolute cytotoxicity levels and the DNA
damage produced by the conjugates. Compounds 2 and 9f show
similar adduct profiles (Figure 4a,d) and overall levels of
platinum adducts (Figure 4f). Among the conjugates tested, the
two derivatives show the most pronounced differential inhibitory
effects in HL-60 and H460 cells: both compounds are ∼10-
fold more active in the solid tumor cell line compared to the
leukemia cell line. In addition, 2 and 9f show greatly enhanced
activity compared to the free acridines in H460 cells, suggesting
that both compounds may act by a common mechanism in this
cell line. In contrast, no metal-enhanced activity is observed
for 9c-e. This raises the question as to the role of platinum in
the mechanism of DNA damage. Clearly, the hybrid binding
mode involving monofunctional platination differs from cis-
platin-induced cross-links. The 1,2 intrastrand cross-link, cis-
platin’s major adduct, bends DNA significantly, a structural
distortion that is thought to trigger cell death mediated by high-
mobility group (HMG) proteins, which tightly associate with
this lesion.44-46 The NMR solution structure of the monofunc-
tional G adduct formed by 2 suggests that the combination of
platination and intercalation significantly unwinds DNA locally
but does not bend the duplex.26 It is possible that platinum in
our hybrid agents primarily acts as a “covalent anchor” that
turns reversible intercalation into permanent damage and,
ultimately, prevents dissociation of the acridine derivative from
DNA. Because long-lived (here, irreversible) intercalator-DNA
complexes often interfere with RNA polymerases to disrupt
transcription,47 platinum might turn the acridines into more
efficient transcription inhibitors. This would explain why
modification with platinum of ACRAMTU (1), the simplest
acridine derivative in the series, whose DNA binding is
characterized by rapid on/off rates,14 results in greatly enhanced
cytotoxicity levels. On the other hand, platinum in conjugate
9d, for instance, containing a potentially threading intercalator
that can be expected to form longer-lived intercalative com-
plexes with DNA, apparently does not contribute at all to the
cytotoxicity of the conjugate (see Table 1 for a comparison of
The following compounds were prepared according to previously
described procedures: 1-[2-(acridin-9-ylamino)ethyl]-1,3-dimeth-
ylthiourea HNO3 salt (ACRAMTU, 1), [PtCl(en)(ACRAMTU)]-
(NO3)2 (PT-ACRAMTU, 2), and (2-aminoethyl)methylcarbamic
acid tert-butyl ester (V),9 and 9-chloroacridine-4-carbonyl chloride
(3′), 9-chloroacridine-4-carboxylic acid 4-nitrophenyl ester (3′′), and
9-chloroacridine-4-carboxylic acid (2-dimethylaminoethyl)amide
(4e).23
Synthesis of Urea Derivatives I-IV. All compounds were
synthesized according to a procedure reported previously for N-(2-
aminoethyl)-N′-phenylurea (III)24 and isolated as hydrochloride
salts.
1-(2-Aminoethyl)-3-ethylurea Hydrochloride, I. 1H NMR
(D2O): δ 3.41 (2H, t, J ) 5.8), 3.14-3.07 (4H, m), 1.07 (3H, t, J
) 7.2).
1-(4-Aminobutyl)-3-ethylurea Hydrochloride, II. 1H NMR
(DMSO-d6): δ 8.08 (br s), 6.78 (br s), 3.04-2.97 (4H, m), 2.80-
2.69 (2H, m), 1.44-1.35 (2H, m), 1.59-1.49 (2H, m), 0.98 (3H,
t, J ) 7.2).
1-(4-Aminobutyl)-3-phenylurea Hydrochloride, IV. 1H NMR
(DMSO-d6): δ 8.73 (1H, br s), 7.84 (2H, br s), 7.38 (2H, d, J )
8.1), 7.20 (2H, t, J ) 7.9), 6.88 (1H, t, J ) 7.2), 3.09 (2H, J )
6.4), 2.85-2.74 (2H, m), 1.62-1.44 (4H, m).
Compounds 4a-d were synthesized using a common procedure.
A typical synthetic procedure is described for 4c below.
9-Chloroacridine-4-carboxylic Acid [2-(3-Phenylureido)ethyl]-
amide, 4c. A mixture of 4.296 g (19.92 mmol) of 1-(2-aminoethyl)-
3-phenylurea hydrochloride (III) and 15.14 mL (108.7 mmol) of
triethylamine in anhydrous CH2Cl2 (50 mL) was stirred for 1 h at
room temperature. The reaction mixture was cooled to 0 °C and
5.0 g (18 mmol) of 9-chloroacridine-4-carbonyl chloride (3′) was

Platinum-Threading Intercalator Conjugates
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11 3211
added in one portion. The reaction mixture was stirred for another
4-5 h and then evaporated to dryness. The residue was washed
thoroughly with water and redissolved in CHCl3. The solution was
dried with anhydrous Na2SO4, filtered, and evaporated to dryness.
The reddish oil obtained was treated with diethyl ether to yield 4c
NMR (DMSO-d6): δ 12.5 (1H, br s), 8.63 (1H, d, J ) 6.8), 8.55
(1H, d, J ) 8.6), 8.41 (1H, d, J ) 8.8), 8.0 (1H, d, J ) 8.4), 7.7
(1H, t, J ) 7.6), 7.48-7.40 (2H, m), 4.99 (1H, br s), 3.93 (2H, t,
J ) 6.2), 3.73-3.68 (2H, m), 3.60-3.55 (2H, m), 2.87 (2H, t, J )
6.3), 2.30 (3H, s).
1
as a yellow microcrystalline solid (yield 4.0 g, 53%). H NMR
Compounds 7a-g were synthesized using a common procedure.
A typical synthetic procedure is described below for 7c.
9-[2-(1,3-Dimethylthioureido)ethylamino]acridine-4-carboxyl-
ic Acid [2-(3-Phenylureido)ethyl]amide, 7c. To 1.0 g (2.19 mmol)
of 6c in 25 mL of dry ethanol was added 0.203 g (2.78 mmol) of
CH3NCS in 2 mL of dry ethanol and the mixture was refluxed for
8 h. The solvent was evaporated to dryness, and the residue was
recrystallized from MeOH to give golden-yellow needles of
(DMSO-d6): δ 11.18 (1H, br s), 8.78 (1H, d, J ) 7.0), 8.66 (1H,
s), 8.62 (1H, d, J ) 8.6), 8.44-8.36 (2H, m), 7.89 (1H, t, J ) 7.8),
7.83-7.79 (2H, m), 7.39 (2H, d, J ) 7.7), 7.19 (2H, t, J ) 7.8),
6.88 (1H, t, J ) 7.3), 6.44 (1H, br s), 3.72-3.66 (2H, m), 3.54-
3.48 (2H, m).
9-Chloroacridine-4-carboxylic Acid (2-Hydroxyethyl)amide,
4f. Compound 3′′ (7.5 g, 19.8 mmol) was added to an ice-cooled
stirred solution of 2-aminoethanol (1.36 mL, 22.5 mmol) and
triethylamine (6.27 mL, 45.0 mmol) in anhydrous CH2Cl2 (50 mL).
After stirring at 0-5 °C for 4 h, the solvent was evaporated, and
the residue was washed thoroughly with cold absolute ethanol,
filtered, and dried in a vacuum, giving 4f as a yellow powder (4.646
g, 78%), which was used in the next step without further
1
compound 7c (0.700 g, 60%). H NMR (MeOH-d4): δ 8.78 (1H,
d, J ) 8.3), 8.61 (1H, d, J ) 8.5), 8.37 (1H, d, J ) 7.3), 7.92 (1H,
t, J ) 7.5), 7.76 (1H, d, J ) 8.5), 7.55-7.52 (2H, m), 7.21 (2H, d,
J ) 8.0), 7.09 (2H, t, J ) 7.7), 6.87 (1H, t, J ) 7.3), 4.52 (2H, t,
J ) 4.7), 4.43 (2H, t, J ) 4.7), 3.64 (2H, t, J ) 5.6), 3.54 (2H, t,
J ) 6.0), 3.10 (3H, s), 3.04 (3H, s).
1
purification. H NMR (DMSO-d6): δ 11.50 (1H, br s), 8.79 (1H,
The nitrate salts 8a-g were generated by adding 1 M HNO3 (1
or 2 equiv) to the methanolic solutions of the corresponding free
bases, 7a-g. The compounds were precipitated with EtOAc and
recrystallized from a mixture of MeOH and EtOAc.
dd, J ) 7.1, 1.3), 8.60 (1H, dd, J ) 8.7, 1.3), 8.43 (1H, d, J )
8.6), 8.31 (1H, d, J ) 8.7), 8.05 (1H, t, J ) 7.7), 7.91-7.83 (2H,
m), 5.07 (1H, br s), 3.78-3.72 (2H, m), 3.65-3.60 (2H, m).
Compounds 5a-e and 5g were synthesized from 4a-e and 4f,
respectively, using a common procedure. A typical procedure is
described below for 5c.
9-[2-(1,3-Dimethylthioureido)ethylamino]acridine-4-carboxyl-
ic Acid [2-(3-Ethylureido)ethyl]amide Hydronitrate (8a). Yield:
0.290 g (51%). 1H NMR (MeOH-d4): δ 8.82 (1H, d, J ) 8.6, H1),
8.66 (1H, d, J ) 8.6, H8), 8.38 (1H, d, J ) 7.3, H3), 7.98 (1H, t,
J ) 7.6, H6), 7.88 (1H, d, J ) 8.5, H5), 7.60-7.56 (2H, m, H7/
H2), 4.53 (2H, t, J ) 5.1, CH2 in thiourea linkage), 4.46 (2H, t,
CH2 in thiourea linkage), 3.57 (2H, t, J ) 5.8, CH2 in amide
residue), 3.46 (2H, t, J ) 5.8, CH2 in amide residue), 3.12 (2H, q,
J ) 7.2, CH2CH3), 3.09 (3H, s, CH3), 3.03 (3H, s, CH3), 1.05 (3H,
t, J ) 7.2, CH2CH3). 13C{H} NMR (MeOH-d4): δ 184.3, 169.5,
161.5, 135.6, 53.5, 50.8, 42.3, 40.2, 36.7, 35.9, 33.5, 15.7. (The
LC-MS profile and 1H NMR spectrum of 8a and combustion data
for the free base precursor, 7a, are given as the Supporting
Information.)
9-(2-Methylaminoethylamino)acridine-4-carboxylic Acid [2-(3-
Phenylureido)ethyl]amide Trihydrochloride, 5c. A mixture of
4c (4.0 g, 9.55 mmol) and dry phenol (8.98 g, 95.5 mmol) was
heated at 60 °C for 30 min. To the clear solution, (2-aminoethyl)-
methylcarbamic acid tert-butyl ester (V) (1.90 g, 10.9 mmol) was
added in one portion, and stirring was continued at 120 °C for 2 h.
The reaction mixture was cooled to room temperature and poured
into 400 mL of 2 M NaOH solution. The mixture was extracted
with CHCl3, and the organic layer was dried over anhydrous Na2-
SO4, filtered, and evaporated to dryness to afford a bright orange
viscous oil. The residue was dissolved in a mixture of CH3COOH
(100 mL) and concentrated HCl (10 mL) and stirred at room
temperature for 1.5 h. After removal of the acid using a rotary
evaporator, the resulting reddish oil was treated with a 1:1 mixture
of ethanol and diethyl ether to crystallize compound 5c (2.25 g,
9-[2-(1,3-Dimethylthioureido)ethylamino]acridine-4-carboxyl-
ic Acid [4-(3-Ethylureido)butyl]amide Hydronitrate (8b).
1
Yield: 0.235 g (42%). H NMR (MeOH-d4): δ 8.82 (1H, d, J )
8.5, H1), 8.65 (1H, d, J ) 8.6, H8), 8.40 (1H, d, J ) 7.3, H3),
7.97 (1H, t, J ) 7.6, H6), 7.82 (1H, d, J ) 8.4, H5), 7.58-7.55
(2H, m, H2/H7), 4.53 (2H, t, J ) 4.8, CH2 in thiourea linkage),
4.46 (2H, t, J ) 4.7, CH2 in thiourea linkage), 3.52 (2H, t, J ) 7.0,
CH2 in amide residue), 3.20 (2H, t, J ) 7.0, CH2 in amide residue),
3.13 (2H, q, J ) 7.2, CH2CH3), 3.10 (3H, s, CH3), 3.03 (3H, s,
CH3), 1.76-1.70 (2H, m, CH2 in amide linkage), 1.64-1.59 (2H,
m, CH2 in amide linkage), 1.08 (3H, t, J ) 7.2, CH2CH3). 13C{H}
NMR (MeOH-d4): δ 184.3, 169.2, 161.3, 160.3, 137.0, 135.6,
120.4, 120.4, 53.5, 50.79, 40.8, 40.7, 36.8, 35.9, 33.5, 29.0, 27.6,
15.8. Anal. (C26H36N8O5S‚CH3OH) C, H. N: calcd, 18.53; found,
18.12.
1
42%). H NMR (D2O): δ 8.38 (1H, d, J ) 8.3), 8.18-8.17 (2H,
m), 7.86 (1H, t, J ) 7.4), 7.59 (1H, d, J ) 8.2), 7.54-7.48 (2H,
m), 6.93-6.88 (2H, m), 6.83-6.78 (3H, m), 4.42 (2H, t, J ) 6.3),
3.63-3.60 (4H, m), 3.54-3.50 (2H, m), 2.80 (3H, s).
Compounds 6a-e and 6g were synthesized from 5a-e and 5g
using a common procedure. Compound 5g was also used to
synthesize 6f. Representative synthetic procedures are given for
6c and 6f.
9-(2-Methylaminoethylamino)acridine-4-carboxylic Acid [2-(3-
Phenylureido)ethyl]amide, 6c. A mixture of 1.75 g (3.31 mmol)
of 5c and 100 mL of 2 M NH4OH was stirred for 15 min and then
extracted with 5 × 30 mL of CHCl3. The combined extracts were
washed with 2 × 50 mL of water and dried over anhydrous Na2-
SO4. Removal of the solvent under reduced pressure gave 6c as a
yellow solid, which was dried in a vacuum at 60 °C (yield 1.0 g,
9-[2-(1,3-Dimethylthioureido)ethylamino]acridine-4-carboxyl-
ic Acid [2-(3-Phenylureido)ethyl]amide Hydronitrate (8c).
1
Yield: 0.172 g (51%). H NMR (MeOH-d4): δ 8.79 (1H, d, J )
8.6, H1), 8.63 (1H, d, J ) 8.7, H8), 8.38 (1H, d, J ) 7.4, H3),
7.94 (1H, t, J ) 7.7, H6), 7.79 (1H, d, J ) 8.5, H5), 7.57-7.52
(2H, m, H2/H7), 7.21 (2H, d, J ) 8.1, Ph-o-H), 7.09 (2H, t, J )
7.9, Ph-m-H), 6.87 (1H, t, J ) 7.4, Ph-p-H), 4.55-4.51 (2H, m,
CH2 in thiourea linkage), 4.46-4.42 (2H, m, CH2 in thiourea
linkage), 3.66-3.62 (2H, m, CH2 in amide residue), 3.56-3.52 (2H,
m, CH2 in amide residue), 3.10 (3H, s, CH3), 3.03 (3H, s, CH3).
13C{H} NMR (MeOH-d4): δ 184.3, 169.8, 160.3, 158.8, 140.7,
136.9, 135.6, 129.7, 123.4, 120.8, 120.4, 120.1, 61.5, 53.5, 50.8,
42.0, 40.3, 36.7, 33.5. Anal. (C28H32N8O5S‚0.5H2O‚0.5CH3-
COOC2H5) C, N. H: calcd, 5.78; found, 5.34.
1
66%). H NMR (CDCl3): δ 12.86 (1H, br s), 8.67 (1H, d, J )
6.8), 8.16-8.11 (2H, m), 7.93 (1H, d, J ) 8.7), 7.64-7.59 (2H,
m), 7.43-7.41 (2H, m), 7.35 (1H, t, J ) 7.7), 7.24-7.16 (4H, m),
6.96 (1H, t, J ) 7.4), 6.28 (1H, br s), 3.87 (2H, t, J ) 5.6), 3.82-
3.76 (2H, m), 3.67-3.62 (2H, m), 2.93 (2H, t, J ) 5.6), 2.54 (3H,
s).
9-(2-Methylaminoethylamino)acridine-4-carboxylic Acid (2-
Hydroxyethyl)amide, 6f. A mixture of 2.650 g (5.845 mmol) of
acetic acid 2{[9-(2-methylaminoethylamino)acridine-4-carbonyl]-
amino}ethyl ester trihydrochloride (5g) and 4.8 g (35 mmol) of
K2CO3 in 50 mL of MeOH/H2O (1:25 v/v) was stirred at room
temperature for 2 h. The solvent was partially removed by rotary
evaporation, and the product was extracted into CHCl3. The organic
layer was dried over anhydrous Na2SO4, filtered, and evaporated
9-[2-(1,3-Dimethylthioureido)ethylamino]acridine-4-carboxyl-
ic Acid [4-(3-Phenylureido)butyl]amide Hydronitrate (8d).
1
Yield: 0.250 g (45%). H NMR (MeOH-d4): δ 8.82 (1H, d, J )
8.6, H1), 8.66 (1H, d, J ) 8.7, H8), 8.42 (1H, d, J ) 7.4, H3),
7.97 (1H, t, J ) 7.7, H6), 7.85 (1H, d, J ) 8.5, H5), 7.57 (2H, m,
1
to dryness to give bright orange crystals of 6f (1.90 g, 96%). H

3212 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11
GuddneppanaVar et al.
H2/H7), 7.28 (2H, d, J ) 8.5, Ph-o-H), 7.18 (2H, t, J ) 7.9, Ph-
m-H), 6.93 (1H, t, J ) 7.3, Ph-p-H), 4.53 (2H, t, J ) 4.8, CH2 in
thiourea linkage), 4.46 (2H, t, J ) 4.9, CH2 in thiourea linkage),
3.56 (2H, t, J ) 6.9, CH2 in amide residue), 3.09 (3H, s, CH3),
3.03 (3H, s, CH3), 1.81-1.76 (2H, m, CH2 in amide residue), 1.71-
1.65 (2H, m, CH2 in amide linkage). 13C{H} NMR (MeOH-d4): δ
184.3, 169.2, 160.4, 158.4, 140.9, 137.0, 135.6, 129.7, 123.3, 120.5,
120.4, 120.1, 61.5, 53.5, 50.8, 40.7, 40.4, 36.7, 33.5, 28.7, 27.4,
20.9, 14.5. Anal. (C30H36N8O5S‚0.7CH3COOC2H5) C, H, N.
9-[2-(1,3-Dimethylthioureido)ethylamino]acridine-4-carboxyl-
ic Acid (2-Dimethylaminoethyl)amide Hydronitrate (8e). Yield:
0.150 g (47%). 1H NMR (D2O): δ 8.40 (1H, br s, H1), 8.21-8.20
(2H, m, H8/H3), 7.86 (1H, t, J ) 7.6, H6), 7.60 (1H, d, J ) 8.4,
H5), 7.47-7.41 (2H, m, H7/H2), 4.30-4.28 (2H, m, CH2 in
thiourea linkage), 4.22-4.20 (2H, m, CH2 in thiourea linkage), 3.91
(2H, t, J ) 6.2, CH2 in amide residue), 3.52 (2H, t, J ) 6.2, CH2
in amide residue), 3.05 (6H, s, N(CH3)2), 2.94 (3H, s, CH3), 2.91
(3H, s, CH3). 13C{H} NMR (MeOH-d4): δ 184.3, 170.6, 160.4,
137.0, 136.0, 120.4, 119.6, 58.5, 53.5, 50.81, 44.0, 36.7, 36.4, 33.5.
Anal. (C23H32N8O7S‚H2O‚0.3 CH3COOC2H5) C, H, N.
9-[2-(1,3-Dimethylthioureido)ethylamino]acridine-4-carboxyl-
ic Acid (2-Hydroxyethyl)amide Hydronitrate (8f). Yield: 0.280
g (49%). 1H NMR (MeOH-d4): δ 8.82 (1H, d, J ) 8.6, H1), 8.65
(1H, d, J ) 8.6, H8), 8.41 (1H, d, J ) 7.4, H3), 7.97 (1H, t, J )
7.7, H6), 7.84 (1H, d, J ) 8.5, H5), 7.59-7.55 (2H, m, H7/H2),
4.53 (2H, t, J ) 5.0, CH2 in thiourea linkage), 4.46 (2H, t, J ) 4.8,
CH2 in thiourea linkage), 3.81 (2H, t, J ) 5.7, CH2 in amide
residue), 3.64 (2H, t, J ) 5.7, CH2 in amide residue), 3.10 (3H, s,
CH3), 3.03 (3H, s, CH3). 13C{H} NMR (MeOH-d4): δ 184.3, 169.5,
160.3, 137.0, 135.7, 120.7, 120.33, 61.4, 53.5, 50.8, 43.5, 36.7,
33.5. Anal. (C21H26N6O5S‚0.5 CH3COOC2H5) H, N. C: calcd,
53.27; found, 52.81.
[PtCl(en)(C30H36N7O2S)](NO3)2, 9d. This compound was re-
crystallized from hot ethanol. Yield: 0.06 g (33%). 1H NMR
(D2O): δ 8.45 (1H, d, J ) 8.3), 8.32 (1H, d, J ) 8.7), 8.29 (1H,
d, J ) 7.4), 7.97 (1H, t, J ) 7.8), 7.80 (1H, d, J ) 8.6), 7.63-7.57
(2H, m), 7.04 (1H, t, J ) 7.7), 6.93 (1H, t, J ) 7.4), 6.87 (1H, d,
J ) 8.3), 5.28 (br s, NH2, slow H,D exchange), 5.19 (br s, NH2,
slow H,D exchange), 4.47 (4H, br m), 3.55 (2H, t, J ) 6.2), 3.23
(2H, t, J ) 6.2), 3.13 (3H, s), 2.91 (3H, s), 2.59 (4H, s, broad base
due to unresolved Pt-satellites), 1.82-1.69 (4H, m). ESI-MS
(MeOH, +ve mode) m/z: 848.2 [M - H]+. Anal. (C32H44N11O8-
SClPt‚C2H5OH‚H2O) C, H, N.
1
[PtCl(en)(C23H32N6OS)](NO3)3, 9e. Yield: 0.110 g (43%). H
NMR (D2O): δ 8.55 (1H, d, J ) 8.5), 8.40 (1H, d, J ) 8.6), 8.35
(1H, d, J ) 7.6), 8.02 (1H, t, J ) 7.4), 7.88 (1H, d, J ) 8.1),
7.67-7.65 (2H, m), 5.19 (br s, NH2, slow H,D exchange), 5.01
(br, NH2, slow H,D exchange), 4.53 (4H, br m), 3.94 (2H, t, J )
6.1), 3.53 (2H, t, J ) 6.1), 3.13 (3H, s), 3.05 (6H, s), 2.93 (3H, s),
2.61 (4H, s, broad base due to unresolved Pt-satellites). ESI-MS
(MeOH, +ve mode) m/z: 729.3 [M-2H]+. Anal. (C25H40N11O10-
SClPt‚2H2O‚0.2 CH3COOC2H5) C, H, N.
[PtCl(en)(C21H26N5O2S)](NO3)2, 9f. Yield: 0.130 g (46%). 1H
NMR (D2O): δ 8.46 (1H, d, J ) 8.5), 8.33 (1H, d, J ) 8.6), 8.30
(1H, d, J ) 7.4), 7.99 (1H, t, J ) 7.8), 7.80 (1H, d, J ) 8.6),
7.64-7.59 (2H, m), 5.18 (br s, NH2, slow H,D exchange), 5.01 (br
s, NH2, slow H,D exchange), 4.48 (4H, br m), 3.88 (2H, t, J )
5.5), 3.68 (2H, t, J ) 5.4), 3.10 (3H, s), 2.87 (3H, s), 2.61 (4H, s,
broad base due to unresolved Pt-satellites). ESI-MS (MeOH, +ve
mode) m/z: 702.2 [M - H]+. Anal. (C23H34N9O8SClPt) C, H. N:
calcd, 15.24; found, 14.78.
[PtCl(en)(C23H28N5O3S)](NO3)2, 9g. Yield: 0.110 g (51%). 1H
NMR (D2O): δ 8.44 (1H, d, J ) 8.5), 8.30 (1H, d, J ) 8.4), 8.22
(1H, d, J ) 7.5), 7.98 (1H, t, J ) 7.8), 7.77 (1H, d, J ) 8.5),
7.63-7.58 (2H, m), 5.18 (br s, NH2, slow H,D exchange), 5.01 (br
s, NH2, slow H,D exchange), 4.46 (4H, br m), 4.42 (2H, t, J )
5.2), 3.79 (2H, t, J ) 5.2), 3.10 (3H, s), 2.88 (3H, s), 2.61 (4H, s,
broad base due to unresolved Pt-satellites), 2.15 (3H, s). ESI-MS
(MeOH, +ve mode) m/z: 744.2 [M - H]+. Anal. (C25H36N9O9-
SClPt‚3H2O) C, H, N.
Acetic Acid 2-({9-[2-(1,3-Dimethylthioureido)ethylamino]-
acridine-4-carbonyl}amino)ethyl Ester Hydronitrate (8g).
1
Yield: 0.300 g (53%). H NMR (MeOH-d4): δ 8.82 (1H, d, J )
8.6, H1), 8.64 (1H, d, J ) 8.6, H8), 8.37 (1H, d, J ) 7.4, H3),
7.97 (1H, t, J ) 7.7, H6), 7.84 (1H, d, J ) 8.5, H5), 7.59-7.55
(2H, m, H2/H7), 4.53 (2H, t, J ) 5.1, CH2 in thiourea linkage),
4.46 (2H, t, J ) 4.8, CH2 in thiourea linkage), 4.35 (2H, t, J ) 5.5,
CH2 in amide residue), 3.75 (2H, t, J ) 5.5, CH2 in amide residue),
3.09 (3H, s, CH3), 3.03 (3H, s, CH3), 2.07 (3H, s, OCH3). 13C{H}
NMR (MeOH-d4): δ 184.3, 172.9, 169.6, 160.4, 137.0, 135.7,
125.4, 123.4, 120.3, 63.8, 53.5, 50.8, 40.3, 36.7, 33.5, 20.8. Anal.
(C23H28N6O6S) C, H, N.
Molecular Modeling. Molecular mechanics calculations were
performed using the AMBER force field in the InsightII/Discover
software (version 2000, Accelrys, San Diego, CA). NMR-based
energy-minimized models of octamer duplexes containing ACRA-
MTU (derivative 1) were used as starting structures.14 The structures
of derivatives 8b, 8e, and 8f were built with the Biopolymer module
and energy-minimized. To produce the drug-DNA complexes, the
ACRAMTU molecule was deleted from the starting complexes and
replaced with the new threading intercalators by manual docking
with the vacant intercalation pocket. The derivatives 8 were inserted
regioselectively with the thiourea and 4-carboxamide residues
protruding into the minor and major groove, respectively. To induce
the proposed hydrogen bonds in the major groove, distance
constraints were introduced between the appropriate donor and
acceptor atoms using a quadratic harmonic. Initially, these distance
constraints were included in the minimizations with force constants
of 100 kcal mol-1 Å-2 and the donor-acceptor distances fixed at
d ) 2.5 Å. During the minimizations, Na+ counterions were used
to neutralize the negative charge of the phosphodiester backbone,
and solvent was simulated with a distance-dependent dielectric (ꢀ
) 4rij). Coulombic and 1-4 parameters were scaled by the common
factor of 0.5. The drug-DNA complexes were subjected to steepest
descent and conjugate gradient minimization to a final ∆rms of
0.01 kcal mol-1Å-1 with constraints “on” and to 0.001 kcal
mol-1Å-1 with contraints “off”. Molecular views of the model were
generated from car files using the DS ViewerPro software (version
6.0, Accelrys, San Diego, CA).
Conjugates 9c-g were synthesized using a common procedure.
A typical procedure is described for 9c below.
[PtCl(en)(C28H32N7O2S)](NO3)2, 9c. A mixture of 0.091 g (0.28
mmol) of [PtCl2(en)] and 0.047 g (0.28 mmol) of AgNO3 in 10
mL of anhydrous DMF was stirred at room temperature in the dark
for 14 h. The precipitated AgCl was filtered off through a Celite
pad, 0.155 g (0.26 mmol) of 8c was added to the filtrate, and the
solution was stirred for 5 h in the dark. The solvent was removed
in a vacuum at 30 °C, yielding an oily residue, which was
redissolved in hot methanol. The solution was treated with activated
carbon and filtered while hot, and the solvent was removed by rotary
evaporation. The crude material was redissolved in a minimum
amount of methanol and subjected to size exclusion chromatography
on a Sephadex LH-20 column using HPLC-grade methanol as the
eluent. The collected fractions were subjected to in-line HPLC/
mass spectrometry (LC-MS) analysis and the desired fractions were
pooled and evaporated to dryness to afford bright yellow micro-
1
crystalline complex 9c (0.120 g, 46%). H NMR (D2O): δ 8.41
(1H, d, J ) 8.1), 8.24 (1H, d, J ) 7.1), 8.21 (1H, d, J ) 7.4), 7.89
(1H, t, J ) 7.5), 7.64 (1H, d, J ) 8.6), 7.57 (2H, t, J ) 7.2), 6.96
(2H, br s), 6.86 (3H, br s), 5.20 (br s, NH2, slow H,D exchange),
5.0 (br s, NH2, slow H,D exchange), 4.46 (2H, br s), 4.41 (2H, br
s), 3.67 (2H, br s), 3.55 (2H, br s), 3.19 (3H, s), 2.97 (3H, s), 2.60
(4H, br s, broad base due to unresolved Pt-satellites). ESI-MS
(MeOH, +ve mode) m/z: 819.1 [M - H]+. Anal. (C30H40N11O8-
SClPt‚3H2O) C, H, N.
Bioanalytical Assays. (a) Enzymatic Digestion/LC-ESMS
Analysis. The stock solutions of 2, 9e, and 9f were prepared in 20
mM Tris buffer (pH 7.1), and the stock solution of 9d was prepared
in DMF. The amount of DMF was less than 1% for 9d in the final
incubation mixtures. The stock solutions were prepared immediately
prior to the incubations and if necessary stored at -20 °C. The

Platinum-Threading Intercalator Conjugates
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11 3213
calf thymus DNA was purchased from Sigma and its stock solution
was made in 20 mM Tris buffer (pH 7.1). The nucleic acid was
annealed by slow cooling of the buffered solution from 80 °C to
room temperature to ensure that the DNA was in its double-stranded
form. DNA concentrations (base pairs, bp) were determined from
absorbances at 260 nm using Beer’s law with ꢀ260 (calf thymus) )
12 824 M-1 cm-1 bp.48 Millipore water was used for the preparation
of all buffers. DNase I and calf intestinal alkaline phosphatase (CIP)
were obtained from New England BioLabs, Ipswich, MA, and
nuclease P1 (from penicillium citrinium) was from Sigma. Stock
solutions of nuclease P1 were prepared in enzyme buffer provided
by the vendor. All other chemicals and solvents were purchased
from common vendors and used as supplied. HPLC-grade solvents
were used in all chromatographic separations. Calf thymus DNA
(2.0 × 10-4 M) was incubated with the drugs at a platinum-to-
nucleotide ratio (ri) of 0.05 at 37 °C for 40 h in dark. Incubation
of the four drugs was performed in triplicate. The following protocol
was used to digest 500-µL samples of the platinum-modified DNA
(total incubation time 26 h at 37 °C): (i) 40 µL of 50 mM MnCl2
+ 40 units of DNase I (2 h), (ii) 26 units of DNase I (2 h), (iii) 16
units of nuclease P1 (2 h), (iv) 4 units of nuclease P1 (16 h), (v)
20 units of alkaline phosphatase + 20 µL of alkaline phosphatase
buffer (2 h), and (vi) 12 units of alkaline phosphatase (2 h). The
mixtures were centrifuged at 13 000 rpm for 5 min, and the
supernatant was collected. The digested samples were desalted
against water for 24 h at room temperature using 100 Da cutoff
dialysis membranes and stored at -20 °C. To account for an
increase in the sample volume during dialysis, all samples were
lyophilized and redissolved in 275 µL of water before (quantitative)
LC-MS analysis. The individual adducts were quantified by diode-
array UV-visible detection using molar absorptivities for ACRA-
MTU and the 4,9-substituted derivatives of 9450 and 9930 M-1
cm-1, respectively.
H460 and 27 500 cells/well for HL-60 cells. Stock solutions of 1
and 8a-d were prepared in DMSO while the stock solutions of
acridines 8e-g and conjugates 2, 9c and 9e-g were prepared in
phosphate-buffered saline (PBS). Conjugate 9d was dissolved in
DMF. Prior to the incubations, drug solutions were diluted with
media to a final concentration of less than 0.1% in DMSO or DMF.
The cells growing in log phase were incubated with appropriate
serial dilutions of the drugs in triplicate for 72 h. To each well was
added 20 µL of MTS/PMS solution, and the mixtures were allowed
to equilibrate for 3 h. The absorbance at 490 nm (none of the
derivatives tested absorbs at this wavelength) was recorded using
a Precision Microplate Reader (Molecular Devices, Sunnyvale, CA).
The reported IC50 data were calculated from nonlinear curve fits
using a sigmoidal dose-response equation in GraphPad Prism
(version 3.02, GraphPad Software Inc., San Diego, CA) and are
averages of a minimum of three individual experiments.
Acknowledgment. This research was supported by a grant
from the National Institutes of Health/National Cancer Institute
(CA101880). We thank Dr. Marcus W. Wright for assistance
with the NMR experimental setup and helpful discussions. A
generous loan of tetrachloroplatinate from Johnson Matthey PLC
(Reading, England) is also gratefully acknowledged.
Supporting Information Available: Analytical data and LC-
MS profiles for selected target compounds and LC-MS results for
enzymatic digestion experiments. This material is available free of
charge via the Internet at http://pubs.acs.org.
References
(1) Nelson, S. M.; Ferguson, L. R.; Denny, W. A. DNA and the
chromosomesVaried targets for chemotherapy. Cell Chromosome
2004, 3, 2.
Analytical separation of adducts in enzymatic digestion mixtures
was performed using the HPLC module of an Agilent Technologies
1100LS/MSD Trap instrument equipped with a multiwavelength
diode-array detector under the following conditions: column,
Agilent Zorbax SB-C18 reverse phase 150 × 4.6 mm/5 µm, T )
25 °C; eluents, solvent A, 0.1% formic acid in water, solvent B,
0.1% formic acid in acetonitrile; gradient, 95% A/5% Bf65%
A/35% B over 0-30 min; flow rate, 0.75 mL/min; injection volume,
20 µL. Elution of Pt-acridine-modified DNA fragments for 2 and
9d-f was monitored at λmax 413 and 422 nm, respectively. Mass
spectra were recorded on an Agilent Technologies 1100 LC/MSD
ion trap mass spectrometer equipped with an atmospheric pressure
electrospray ionization source. The ESMS acquisition parameters
were adopted from a published protocol.11
(b) Acidic Digestion/LC-ESMS Analysis. The stock solution
of 9d was prepared as described under section a. The oligomer
5′-GCGATATCGC-3′ was synthesized using phosphoramidite
chemistry and desalted by Integrated DNA Technologies Inc.
(Coralville, IA). The sequence was dissolved in 10 mM Tris buffer
(pH 7.2) containing 100 mM NaCl, quantified by UV-visible
spectroscopy using ꢀ260 ) 95 600 M-1 cm-1, and annealed by slow
cooling of the buffered solution from 80 °C to room temperature.
Thermal melting curves were recorded of the oligomer to confirm
that the sequence was in its double-stranded form at the incubation
temperature. All incubations were performed in duplicate at a
platinum-to-nucleotide ratio (ri) of 0.3 at 37 °C for 18 h in the
dark. The samples were dialyzed for 48 h against water, treated
with formic acid to lower the pH to 2.4, and heated at 60 °C in
dark for 12 h. Finally, the digested samples were subjected to in-
line LC-MS analysis using the parameters of a published protocol15
with one exception: the HPLC gradient was modified to 95% A/5%
Bf81% A/19% B over 0-60 min, at a flow rate of 0.5 mL/min.
(Eluents A and B are the same as under section a.)
(2) Hurley, L. H. DNA and its associated processes as targets for cancer
therapy. Nat. ReV. Cancer 2002, 2, 188-200.
(3) Neidle, S.; Thurston, D. E. Chemical approaches to the discovery
and development of cancer therapies. Nat. ReV. Cancer 2005, 5, 285-
296.
(4) Reedijk, J. New clues for platinum antitumor chemistry: Kinetically
controlled metal binding to DNA. Proc. Natl. Acad. Sci. U.S.A. 2003,
100, 3611-3616.
(5) van Zutphen, S.; Reedijk, J. Targeting platinum anti-tumour drugs:
Overview of strategies employed to reduce systemic toxicity. Coord.
Chem. ReV. 2005, 249, 2845-2853.
(6) Ackley, M. C.; Barry, C. G.; Mounce, A. M.; Farmer, M. C.; Springer,
B. E.; Day, C. S.; Wright, M. W.; Berners-Price, S. J.; Hess, S. M.;
Bierbach, U. Structure-activity relationships in platinum-acridi-
nylthiourea conjugates: Effect of the thiourea nonleaving group on
drug stability, nucleobase affinity, and in vitro cytotoxicity. J. Biol.
Inorg. Chem. 2004, 9, 453-461.
(7) Hess, S. M.; Anderson, J. G.; Bierbach, U. A non-crosslinking
platinum-acridine hybrid agent shows enhanced cytotoxicity com-
pared to clinical BCNU and cisplatin in glioblastoma cells. Bioorg.
Med. Chem. Lett. 2005, 15, 443-446.
(8) Hess, S. M.; Mounce, A. M.; Sequeira, R. C.; Augustus, T. M.;
Ackley, M. C.; Bierbach, U. Platinum-acridinylthiourea conjugates
show cell line-specific cytotoxic enhancement in H460 lung carci-
noma cells compared to cisplatin. Cancer Chemother. Pharmacol.
2005, 56, 337-343.
(9) Martins, E. T.; Baruah, H.; Kramarczyk, J.; Saluta, G.; Day, C. S.;
Kucera, G. L.; Bierbach, U. Design, synthesis, and biological activity
of a novel non-cisplatin-type platinum-acridine pharmacophore. J.
Med. Chem. 2001, 44, 4492-4496.
(10) Baruah, H.; Rector, C. L.; Monnier, S. M.; Bierbach, U. Mechanism
of action of non-cisplatin type DNA-targeted platinum anticancer
agents: DNA interactions of novel acridinylthioureas and their
platinum conjugates. Biochem. Pharmacol. 2002, 64, 191-200.
(11) Barry, C. G.; Baruah, H.; Bierbach, U. Unprecedented monofunctional
metalation of adenine nucleobase in guanine- and thymine-containing
dinucleotide sequences by a cytotoxic platinum-acridine hybrid
agent. J. Am. Chem. Soc. 2003, 125, 9629-9637.
(12) Budiman, M. E.; Alexander, R. W.; Bierbach, U. Unique base-step
recognition by a platinum-acridinylthiourea conjugate leads to a
DNA damage profile complementary to that of the anticancer drug
cisplatin. Biochemistry 2004, 43, 8560-8567.
(13) Jamieson, E. R.; Lippard, S. J. Structure, recognition, and processing
of cisplatin-DNA adducts. Chem. ReV. 1999, 99, 2467-2498.
Cytotoxicity Assay. The cytotoxicity studies were carried out
using the Celltiter 96 Aqueous Non-Radioactive Cell Proliferation
Assay kit. HL-60 leukemia cells and H460 lung cancer cells were
kept in a humidified atmosphere of 5% CO2/95% air at 37 °C. Cells
were plated on 96-well plates with a total of 700 cells/well for

3214 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11
GuddneppanaVar et al.
(14) Baruah, H.; Bierbach, U. Unusual intercalation of acridin-9-ylthiourea
into the 5′-GA/TC DNA base step from the minor groove: Implica-
tions for the covalent DNA adduct profile of a novel platinum-
intercalator conjugate. Nucleic Acids Res. 2003, 31, 4138-4146.
(15) Barry, C. G.; Day, C. S.; Bierbach, U. Duplex-promoted platination
of adenine-N3 in the minor groove of DNA: Challenging a
longstanding bioinorganic paradigm. J. Am. Chem. Soc. 2005, 127,
1160-1169.
(16) Budiman, M. E.; Bierbach, U.; Alexander, R. W. DNA minor groove
adducts formed by a platinum-acridine conjugate inhibit association
of TATA-binding protein with its cognate sequence. Biochemistry
2005, 44, 11262-11268.
(32) Baruah, H.; Barry, C. G.; Bierbach, U. Platinum-intercalator
conjugates: From DNA-targeted cisplatin derivatives to adenine
binding complexes as potential modulators of gene regulation. Curr.
Topics Med. Chem. 2004, 4, 1537-1549.
(33) Hansen, M.; Hurley, L. Altromycin-B threads the DNA helix
interacting with both the major and the minor grooves to position
itself for site-directed alkylation of guanine-N7. J. Am. Chem. Soc.
1995, 117, 2421-2429.
(34) Hansen, M.; Yun, S.; Hurley, L. Hedamycin intercalates the DNA
helix and, through carbohydrate-mediated recognition in the minor-
groove, directs N7-alkylation of guanine in the major groove in a
sequence-specific manner. Chem. Biol. 1995, 2, 229-240.
(35) Nakatani, K.; Okamoto, A.; Matsuno, T.; Saito, I. Highly selective
DNA alkylation at the 5′ side G of a 5′ GG3′ sequence by an aglycon
model of pluramycin antibiotics through preferential intercalation into
the GG step. J. Am. Chem. Soc. 1998, 120, 11219-11225.
(36) Sun, D.; Hansen, M.; Clement, J. J.; Hurley, L. H. Structure of the
altromycin B (N7-guanine)-DNA adduct. A proposed prototypic
DNA adduct structure for the pluramycin antitumor antibiotics.
Biochemistry 1993, 32, 8068-8074.
(37) Sun, D. Y.; Hansen, M.; Hurley, L. Molecular-basis for the DNA-
sequence specificity of the pluramycinssA novel mechanism involv-
ing groove interactions transmitted through the helix via intercalation
to achieve sequence selectivity at the covalent bonding step. J. Am.
Chem. Soc. 1995, 117, 2430-2440.
(17) Orphanides, G.; Lagrange, T.; Reinberg, D. The general transcription
factors of RNA polymerase II. Genes DeV. 1996, 10, 2657-2683.
(18) Adams, A.; Guss, J. M.; Collyer, C. A.; Denny, W. A.; Wakelin, L.
P. Crystal structure of the topoisomerase II poison 9-amino-[N-(2-
dimethylamino)ethyl]acridine-4-carboxamide bound to the DNA
hexanucleotide d(CGTACG)2. Biochemistry 1999, 38, 9221-9233.
(19) Adams, A.; Guss, J. M.; Denny, W. A.; Wakelin, L. P. Crystal
structure of 9-amino-N-[2-(4-morpholinyl)ethyl]-4-acridinecarboxa-
mide bound to d(CGTACG)2: Implications for structure-activity
relationships of acridinecarboxamide topoisomerase poisons. Nucleic
Acids Res. 2002, 30, 719-725.
(20) Fan, J. Y.; Ohms, S. J.; Boyd, M.; Denny, W. A. DNA adducts of
9-anilinoacridine mustards: Characterization by NMR. Chem. Res.
Toxicol. 1999, 12, 1166-1172.
(21) Fan, J. Y.; Tercel, M.; Denny, W. A. Synthesis, DNA binding and
cytotoxicity of 1-[[omega-(9-acridinyl)amino]alkyl]carbonyl-3-(chlo-
romethyl)-6-hydroxyindolines, a new class of DNA-targeted alkyl-
ating agents. Anticancer Drug Des. 1997, 12, 277-293.
(22) Cheng, A. C.; Chen, W. W.; Fuhrmann, C. N.; Frankel, A. D.
Recognition of nucleic acid bases and base-pairs by hydrogen bonding
to amino acid side-chains. J. Mol. Biol. 2003, 327, 781-796.
(23) Atwell, G. J.; Cain, B. F.; Baguley, B. C.; Finlay, G. J.; Denny, W.
A. Potential antitumor agents. 43. Synthesis and biological activity
of dibasic 9-aminoacridine-4-carboxamides, a new class of antitumor
agent. J. Med. Chem. 1984, 27, 1481-1485.
(24) Kopka, K.; Wagner, S.; Riemann, B.; Law, M. P.; Puke, C.; Luthra,
S. K.; Pike, V. W.; Wichter, T.; Schmitz, W.; Schober, O.; Schafers,
M. Design of new beta1-selective adrenoceptor ligands as potential
radioligands for in vivo imaging. Bioorg. Med. Chem. 2003, 11,
3513-3527.
(38) Hansen, M.; Lee, S.-J.; Cassady, J. M.; Hurley, L. W. Molecular
details of the structure of a psorospermin-DNA covalent/intercalation
complex and associated DNA sequence selectivity. J. Am. Chem.
Soc. 1996, 118, 5553-5561.
(39) Heald, R. A.; Dexheimer, T. S.; Vankayalapati, H.; Siddiqui-Jain,
A.; Szabo, L. Z.; Gleason-Guzman, M. C.; Hurley, L. H. Confor-
mationally restricted analogues of psorospermin: Design, synthesis,
and bioactivity of natural-product-related bisfuranoxanthones. J. Med.
Chem. 2005, 48, 2993-3004.
(40) Kostrhunova, H.; Brabec, V. Conformational analysis of site-specific
DNA cross-links of cisplatin-distamycin conjugates. Biochemistry
2000, 39, 12639-12649.
(41) Loskotova, H.; Brabec, V. DNA interactions of cisplatin tethered to
the DNA minor groove binder distamycin. Eur. J. Biochem. 1999,
266, 392-402.
(25) Augustus, T. M.; Anderson, J.; Hess, S. M.; Bierbach, U. Bis-
(acridinylthiourea)platinum(II) complexes: Synthesis, DNA affinity,
and biological activity in glioblastoma cells. Bioorg. Med. Chem.
Lett. 2003, 13, 855-858.
(26) Baruah, H.; Wright, M. W.; Bierbach, U. Solution structural study
of a DNA duplex containing the guanine-N7 adduct formed by a
cytotoxic platinum-acridine hybrid agent. Biochemistry 2005, 44,
6059-6070.
(27) Kim, J. S.; Kim, J.; Cepek, K. L.; Sharp, P. A.; Pabo, C. O. Design
of TATA box-binding protein/zinc finger fusions for targeted
regulation of gene expression. Proc. Natl. Acad. Sci. U.S.A. 1997,
94, 3616-3620.
(28) Dickinson, L. A.; Gulizia, R. J.; Trauger, J. W.; Baird, E. E.; Mosier,
D. E.; Gottesfeld, J. M.; Dervan, P. B. Inhibition of RNA polymerase
II transcription in human cells by synthetic DNA-binding ligands.
Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 12890-12895.
(29) Dickinson, L. A.; Trauger, J. W.; Baird, E. E.; Ghazal, P.; Dervan,
P. B.; Gottesfeld, J. M. Anti-repression of RNA polymerase II
transcription by pyrrole-imidazole polyamides. Biochemistry 1999,
38, 10801-10807.
(30) Chiang, S. Y.; Welch, J.; Rauscher, F. J., 3rd; Beerman, T. A. Effects
of minor groove binding drugs on the interaction of TATA box
binding protein and TFIIA with DNA. Biochemistry 1994, 33, 7033-
7040.
(31) Woynarowski, J. M.; Trevino, A. V.; Rodriguez, K. A.; Hardies, S.
C.; Benham, C. J. AT-rich islands in genomic DNA as a novel target
for AT-specific DNA-reactive antitumor drugs. J. Biol. Chem. 2001,
276, 40555-40566.
(42) Fan, J. Y.; Valu, K. K.; Woodgate, P. D.; Baguley, B. C.; Denny,
W. A. Aniline mustard analogues of the DNA-intercalating agent
amsacrine: DNA interaction and biological activity. Anticancer Drug
Des. 1997, 12, 181-203.
(43) Varadarajan, S.; Shah, D.; Dande, P.; Settles, S.; Chen, F. X.; Fronza,
G.; Gold, B. DNA damage and cytotoxicity induced by minor groove
binding methyl sulfonate esters. Biochemistry 2003, 42, 14318-
14327.
(44) Ohndorf, U. M.; Rould, M. A.; He, Q.; Pabo, C. O.; Lippard, S. J.
Basis for recognition of cisplatin-modified DNA by high-mobility-
group proteins. Nature 1999, 399, 708-712.
(45) Zamble, D. B.; Mikata, Y.; Eng, C. H.; Sandman, K. E.; Lippard, S.
J. Testis-specific HMG-domain protein alters the responses of cells
to cisplatin. J. Inorg. Biochem. 2002, 91, 451-462.
(46) Cohen, S. M.; Mikata, Y.; He, Q.; Lippard, S. J. HMG-domain protein
recognition of cisplatin 1,2-intrastrand d(GpG) cross-links in purine-
rich sequence contexts. Biochemistry 2000, 39, 11771-11776.
(47) Wakelin, L. P.; Bu, X.; Eleftheriou, A.; Parmar, A.; Hayek, C.;
Stewart, B. W. Bisintercalating threading diacridines: Relationships
between DNA binding, cytotoxicity, and cell cycle arrest. J. Med.
Chem. 2003, 46, 5790-5802.
(48) Jenkins, T. C. Optical absorbance and fluorescence techniques for
measuring DNA drug interactions. In Drug DNA Interaction Pro-
tocols; Fox, K. R, Ed.; Humana Press: Totowa, NJ, 1997; pp
195-218.
JM060035V