Polymer conjugates of acridine-type anticancer drugs with pH-controlled activation

Article abstract of DOI:10.1016/j.bmc.2012.05.007

Acridines are potent DNA-intercalating anticancer agents with high in vivo anticancer effectiveness, but also severe side effects. We synthesized five 9-anilinoacridine-type drugs and their conjugates with biocompatible water-soluble hydrazide polymer carrier. All of the synthesized acridine drugs retained their in vitro antiproliferative properties. Their polymer conjugates were sufficiently stable at pH 7.4 (model of pH in blood plasma) while releasing free drugs at pH 5.0 (model of pH in endosomes). After internalization of the conjugates, the free drugs were released and are visible in cell nuclei by fluorescence microscopy. Their intercalation ability was proven using a competitive ethidium bromide displacement assay.

Full text of DOI:10.1016/j.bmc.2012.05.007

Bioorganic & Medicinal Chemistry 20 (2012) 4056–4063
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Polymer conjugates of acridine-type anticancer drugs
with pH-controlled activation
a
a,
⇑
a
Ondrej Sedlácek , Martin Hruby , Martin Studenovsky , David Vetvicka b,c, Jan Svoboda d,
ˇ
ˇ
´
´
ˇ
ˇ
a
a
ˇ
ˇ e
Dana Kanková , Jan Kovár , Karel Ulbrich
a Institute of Macromolecular Chemistry AS CR, v.v.i., Heyrovsky Sq. 2, 162 06 Prague 6, Czech Republic
b Institute of Biophysics and Informatics, First Faculty of Medicine, Charles University in Prague, Salmovska 1, 120 00 Prague 2, Czech Republic
c CancerTech s.r.o., Pruhonek 1457, 155 00 Praha 5, Czech Republic
d Institute of Microbiology AS CR, v.v.i., Videnska 1083, 142 20 Prague 4, Czech Republic
e Institute for Clinical and Experimental Medicine, Videnska 1958/9, 140 21 Prague 4, Czech Republic
a r t i c l e i n f o
a b s t r a c t
Article history:
Acridines are potent DNA-intercalating anticancer agents with high in vivo anticancer effectiveness, but
also severe side effects. We synthesized five 9-anilinoacridine-type drugs and their conjugates with
biocompatible water-soluble hydrazide polymer carrier. All of the synthesized acridine drugs retained their
in vitro antiproliferative properties. Their polymer conjugates were sufficiently stable at pH 7.4 (model of
pH in blood plasma) while releasing free drugs at pH 5.0 (model of pH in endosomes). After internalization
of the conjugates, the free drugs were released and are visible in cell nuclei by fluorescence microscopy.
Their intercalation ability was proven using a competitive ethidium bromide displacement assay.
Ó 2012 Elsevier Ltd. All rights reserved.
Received 8 February 2012
Revised 27 April 2012
Accepted 4 May 2012
Available online 14 May 2012
Keywords:
Drug delivery systems
Cancer therapy
Controlled release
Acridines
Hydrazone
NH
2
H
N
1. Introduction
O
S
O
9-Anilinoacridines are potent DNA intercalators and strong topo-
isomerase II inhibitors used as anticancer agents.1–3 Amsacrine
(m-AMSA) was the first 9-anilinoacridine clinically employed for
the treatment of leukemia and lymphoma (see Fig. 1).2,4 Since that,
an enormous effort was made to investigate their structure–activity
relationship.5–7 Consequently, several 9-anilinoacridine derivatives
have shown superior antileukemic activity and a broader spectrum
of antitumor activity as compared to amsacrine. Among them, 3-(9-
acridinylamino)-5-hydroxymethylaniline8 (AHMA, 1, Fig. 1) has
greater antitumor efficacy against murine leukemia and solid
tumors than amsacrine, and its blood circulation time is prolonged
because it is less sensitive to deactivation by oxidation to the
corresponding quinone.8–10 Furthermore, it was proven that func-
tionalization of AHMA at certain positions, especially at positions
4 and 5 of the acridine ring, can even potentiate its activity.11,12
Despite their remarkable antitumor effects and preferable accumu-
lation in tumor tissue, 9-anilinoacridines suffer from severe
systemic toxicity and side effects.13–15
O
OH
HN
HN
N
N
m
-AMSA
AHMA (1)
Figure 1. Structures of m-AMSA and AHMA (1).
cer drugs into the tumor tissue while lowering their concentration
elsewhere in the organism.16–18 Furthermore, the optimal polymer
carrier delivers the drug in the inactive polymer-bound form into
the solid tumor tissue, where the free active drug needs to be re-
leased. Solid tumors spontaneously accumulate biocompatible
polymers, polymer micelles, liposomes and nanoparticles with
sizes <200 nm because of the leaky nature of the newly formed tu-
mor vasculature and the poor or missing lymphatic drainage sys-
tem in the solid tumor tissue.19 This so-called Enhanced
Permeation and Retention (EPR) effect is relatively universal for
many solid tumors and allows increased accumulation of the poly-
mers when compared to the surrounding tissue by an order of
magnitude or higher. In addition, this accumulation may be further
improved, for example, by ligand-mediated targeting.19,20
Polymer conjugates of anticancer drugs with tumor-specific
activation are widely studied due to their ability to deliver antican-
⇑
Corresponding author. Tel.: +420 296 809 212; fax: +420 296 809 410.
E-mail address: mhruby@centrum.cz (M. Hruby´).
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2012.05.007

ˇ
O. Sedlácek et al. / Bioorg. Med. Chem. 20 (2012) 4056–4063
4057
A hydrolytically labile hydrazone bond is a popular option for
binding anticancer drugs, such as doxorubicin,21–23 to carriers to
achieve pH-controlled drug release.24 The hydrazone bond is sus-
ceptible to hydrolysis under mildly acidic conditions (e.g., at pH
ca 6.5, typical for tumor interstitial space, or even at pH typical
for endosomes after internalization, which may be as low as ca
5.0), but it is relatively stable towards hydrolysis at neutral pH
(e.g., in blood plasma at pH 7.4). The rate of free active drug release
by hydrolysis of the hydrazone bond at different pH values can be
fine-tuned by changing the steric hindrance, electron density and
82–90% (Scheme 1). This activation was chosen due to the good
reactivity of thiazolidine-2-thione amides with amines, while
being rather stable towards alcoholysis.31 The respective
oxo-derivatives 2a–e were linked to pHPMA-MAAcap hydrazide
to produce conjugates 4a–e. This polymer precursor was chosen
because of its good biological behavior when used as a doxorubicin
carrier.32 Using this method, one can attach 0.54–6.2 wt.% drug to
the polymer, depending on the steric availability of the oxo-group
(Table 1).
However, part of the acridine was also bound in a non-cleavable
manner as the product of a side reaction. In this case, the hydrazide
group of the polymer substituted the aniline group at position 9 of
the acridine ring. When the oxo-derivative was more sterically hin-
dered, the hydrazone bond formed slower and more of the
uncleavable acridine-polymer product was produced (Table 1).
To clarify this process, we treated the sterically hindered acridine
derivative 2e with the hydrazide monomer 5. Using this reaction,
we isolated almost entirely product of the unwanted acridine
nucleophilic substitution 6 and no hydrazone derivative 7 (Scheme
2). A similar reaction (nucleophilic substitution of the aniline part
of the 9-anilinoacridines) was described in the stability study of
cytostatic 9-anilinoacridines with various amines and thiols in
aqueous solution modeling the physiological milieu (the typical
half-lives of 9-anilinoacridines were ca 1 h with thiols and several
hours with various amines at 37 °C).33–35 Although hydrazides are
not strong nucleophiles as compared to thiols, we assume that the
non-cleavable fraction of the acridine is bound to the polymer in
the same way. Therefore, during further drug release experiments,
we only took into account the hydrolytically cleavable part of the
bound drug (i.e., the amount cleaved by 0.1 M HCl was equal to
100%). This strategy is relevant from a biological point of view be-
cause the polymer bearing the drug bound via a non-cleavable
linkage is most likely to be biologically inert, as described for
doxorubicin.36
conjugation of the substituent with the
p-system on the –C@N–
part of the hydrazone bond.25 The structure of the acyl substituent
on the hydrazone bond also influences the release kinetics but sig-
nificantly less.21 However, because the original ketone is a part of
the drug after hydrolysis, one must be cautious to not disrupt the
drug’s biological activity. The conversion of a 9-anilinoacridine’s
side phenyl ring amino group to an amide with various acids,
including the oxo-acids (e.g., levulinic acid), does not limit its bio-
logical activity.8 This property makes 9-anilinoacridines ideal for
fine-tuning of their release rate to achieve an optimal release pro-
file (slow release at pH 7.4 and fast release at pH 5.0).
In this paper, the results of the study of the relationship be-
tween the structure of 9-anilinoacridine substituents and the
hydrolytic stability of the hydrazone bond between the drug and
the polymer carrier are given. We synthesized conjugates of five
oxo-group-containing 9-anilinoacridine-type amides with poly[N-
(2-hydroxypropyl)methacrylamide-co-N1-(6-hydrazino-6-
oxohexyl)methacrylamide] with drugs bound to the polymer car-
rier via hydrolytically labile hydrazone bonds, and we tested the
drugs’ in vitro release profiles and in vitro biological properties.
To the best of our knowledge, no such polymer derivatives of acri-
dine drugs have been described to date.
2. Results and discussion
2.1. Chemistry
For the biological studies, we synthesized the conjugate with
the best release profile (see below) bearing the acridine drug
bound only via hydrazone bond using a different synthetic strategy
(Scheme 2). We obtained conjugate 8 with Mw = 21.5 kDa and acri-
dine 2e content of 6.3% wt. In this case, all of the acridine drug was
bound via a hydrazone bond, as proven by cleavage with 0.1 M HCl
The oxo moiety-containing 9-anilinoacridines 2a–e were syn-
thesized by selective acylation of AHMA with thiazolidine-2-thi-
one-activated carboxylic acids 3a–e in outstanding yields,
m
n
m
a
n
H N
O
O
NH
O
NH
O
NH
HN
O
O
Z
OH
OH
HN
O
O
O
NH
NH
NH
O
Z
S
HN
H N
OH
2
N
1
HN
N
Z
S
H+/MeOH
DMF
HN
N
N
ii
OH
HN
N
3a-e
2a-e
4a-e
i
O
O
O
O
O
O
O
O
O
O
Z
=
=
O
d
a
c
e
b
Scheme 1. Scheme of synthesis of 4a–e.

ˇ
O. Sedlácek et al. / Bioorg. Med. Chem. 20 (2012) 4056–4063
4058
Table 1
Yields of conjugation of 9-anilinoacridines to polymer
Polymer
Method of synthesisa
Mw (kDa)
Mw/Mn
AHMA-a. (%wt)b
4 Isomer ratio (i/ii)c
4a
4b
4c
4d
4e
8
A
A
A
A
A
B
21.9
20.7
20.0
17.8
17.6
21.5
2.91
2.24
2.19
2.14
2.25
1.75
5.0
4.8
6.2
2.1
0.54
6.3
95/5
96/4
91/9
85/15
43/57
d
—
a
b
c
A, reaction of hydrazide polymer with oxo-acyl acridine derivatives; B, radical copolymerization.
Total acridine content.
Molar ratio of hydrolytically cleavable/stable bound acridine derivative (Scheme 2).
All the drug bound via a hydrazone bond.
d
O
O
O
O
OH
HN
NH
HN
MeOH/H+
MeOH/H+
6
+
NH
N
HN
NH
O
HN
NH
O
2
N
m
HN
2e
n
5
O
NH
O
OH
O
O
O
O
NH
O
HN
NH
N
HN
HPMA,AIBN
60°C DMSO
AHMA
DCC, DMF
COOH
MeOH/H+
O
OH
OH
NH
O
O
NH
N
HN
N
O
HN
NH
O
NH
HN
N
N
H
2
N
O
OH
8
7
Scheme 2. Scheme of alternative synthesis of polymer conjugate 8.
after which the polymer did not absorb light at wavelengths over
280 nm.
2.2. In vitro drug release profiles of the conjugates
The drug release from conjugates 4a–e at pH 7.4 (a model of the
pH in blood where the conjugate should be as stable as possible)
and at pH 5.0 (a model of the pH in late endosomes where the drug
should be quickly released) was followed. As shown in Figs. 2 and
3, the release rate of the anilinoacridine derivatives is strongly
pH-dependent. A relatively fast rate was observed at pH 5.0, and
a slower to negligible rate was observed at pH 7.4. Therefore, in
some cases, the basic stability requirements for polymer–drug con-
jugates are fulfilled.
There are basically two main effects in the structure-release
rate relationship: conjugation with the aromatic ring and steric ef-
fects. Figure 2 displays the effects of conjugation of the hydrazone
bond with the aromatic ring. The conjugates 4a and 4b, where the
hydrazone bond is not conjugated with an aromatic ring, behave
similarly. They release their 9-anilinoacridine derivatives very
quickly at pH 5.0 (quantitatively within 2 h) and at a moderate rate
Figure 2. The release of 9-anilinoacridine derivatives 2a–c from conjugates 4a–c at
different pH (calculated as hydrolytically cleavable amount of drug = 100%).
at pH 7.4 (conjugate 4a: 64% anilinoacridine released within 24 h,
conjugate 4b: 55% anilinoacridine released within 24 h). The

ˇ
O. Sedlácek et al. / Bioorg. Med. Chem. 20 (2012) 4056–4063
4059
Figure. 3. The release of 9-anilinoacridine derivatives 2a and 2d–e from conjugates
4a and, 4d–e at different pH (calculated as hydrolytically cleavable amount of
drug = 100%).
Figure 4. The release of the 9-anilinoacridine derivative 2e from its conjugate 8 at
different pH.
slightly slower release rate observed with conjugate 4b compared
to conjugate 4a at pH 7.4 is probably caused by the steric hin-
drance caused by the benzyl group in the proximity of the hydra-
zone bond.25 On the contrary, conjugate 4c contains a hydrazone
bond conjugated with the phenyl aromatic ring and released the
drug slowly even at pH 5.0 (17% after 24 h), and it did not release
any drug at pH 7.4. The most plausible explanation is that the con-
jugation stabilizes the hydrazone bond by delocalization of its elec-
trons into the aromatic ring, which dramatically slows down the
rate of hydrolysis.37
strength of the drug into the double helix of DNA was measured
using a DNA–ethidium bromide (EtBr) displacement test.28 In this
experiment, the mixture of calf-thymus (CT) DNA and EtBr in HEPES
buffer was titrated with a solution of AHMA (1) or drug derivatives
2a, 2d and 2e. The concentration of drug that produces a 50% drop in
the fluorescence of the DNA–EtBr complex (CC50, see Table 2) is
approximately inversely proportional to its DNA affinity. Our results
show that the intercalation strength of the synthesized oxoacri-
dines, 2a, 2d and 2e, (22.8–36.6
lmol/L) is only slightly lower than
The effect of steric hindrance next to the hydrazone group on its
release rate was studied using conjugates 4a, 4d and 4e, differing
only by their substituents adjacent to the original oxo-group,
which were methyl (4a), i-propyl (4d) and t-butyl (4e), respec-
tively. Steric hindrance of the hydrazone increases in the order
4a < 4d < 4e. One can clearly see fromFigure 3 that steric hindrance
does not influence the drug release rate at pH 5.0 (it is nearly quan-
titative within the initial 2 h in all cases) but has a dramatic effect
at pH 7.4. A greater steric hindrance induces a slower release rate
at pH 7.4. This observation allows us to fine-tune the release rate
and create a conjugate that is almost entirely stable at the pH of
blood (7.4) but can quickly release the drug after internalization
into the target cell at pH 5.0. As we have shown from an in vitro
cytotoxicity assay (see below), the acyl substituent has only a small
effect on cytotoxicity; therefore, one can use it to modify the re-
lease rate. The t-butyl group-containing conjugate 4e has optimal
release rate in this study.
However, strong steric hindrance of the oxo-group by the
t-butyl group also slows the formation of conjugate 4e, which leads
to side reactions, namely, the nucleophilic substitution at the 9 po-
sition of the acridine ring with hydrazide. To overcome this prob-
lem, we synthesized the HPMA copolymer conjugate of 2e (with
the best release profile) using a procedure that avoided these side
reactions (Scheme 2). The product formed, conjugate 8, contained
all acridine drug bound via an acid-cleavable hydrazone bond. As
shown in Fig. 4, this conjugate is fairly stable at pH 7.4 (7% drug
released during 24 h) and should not quickly release the drug dur-
ing its transport to the target tumor tissue. In a slightly acidic envi-
ronment at pH 6.5 (a model of pH in tumor interstitial space), the
drug was released considerably faster (57% drug released during
24 h), whereas at pH 5.0 (a model of pH in the endosomal environ-
ment), the drug was released very quickly and quantitatively.
that of the parent compound, AHMA (1, 21.3
l
mol/L).
Furthermore, we tested the antiproliferative activity of the syn-
thesized derivatives in several human cell lines (MCF-7—breast
cancer, HepG-2–hepatocellular liver carcinoma and PC-3—prostate
adenocarcinoma; see Table 2 for the IC50 values). The cytotoxicities
of the 9-anilinoacridines, expressed as IC50 values, were in the
range of 0.6–6.2 lmol/L, which is in accordance with reported lit-
erature data for other AHMA derivatives.8,11,12 The cell lines differ
in their sensitivity towards AHMA (1), 2a, 2d and 2e. The cell line
HepG-2 is considerably more sensitive to the tested compounds
than the remaining two cell lines. Compound 2a is the most
cytotoxic of the compounds tested. Acyl substituent on the AHMA
amino group thus possesses a moderate effect on cytotoxicity. The
differences in IC50 values are within one order of magnitude but are
statistically significant (analysis of variance, ANOVA, on level
a
= 0.01). Our most important observation is that changing the acyl
group to optimize the release profile does not compromise antipro-
liferative activity. Compound 2e was chosen to optimize the poly-
mer conjugate synthesis because it fulfills both of the basic
requirements; it is sufficiently cytotoxic and has a favorable re-
lease profile from the polymer hydrazone at different pH values
(minimal release at pH 7.4 and relatively fast release at pH 5.0).
The cytotoxicity of the conjugates was not tested because we
Table 2
DNA binding and antiproliferative activity of 9-anilinoacridines
a
b
Compound
CC50
,
lmol/L
IC50 , lmol/L
MCF-7
HepG-2
PC-3
1 (AHMA)
21.3 0.6
22.8 0.4
36.6 0.4
30.7 0.8
4.5 0.6
2.9 0.4
4.8 0.4
6.2 1.0
0.6 0.4
1.4 0.6
2.1 0.6
1.6 0.6
3.3 0.5
2.3 0.7
2.3 0.7
2.2 0.6
2a
2d
2e
2.3. In vitro biological studies
a
Acridine drug concentration in
lmol/L that reduces fluorescence of DNA–
ethidium bromide complex (cDNA:cEtBr = 1.26:1) to 50% (CC50 standard deviation).
b
Concentrations causing 50% inhibition in the MTT test (IC50 standard devia-
The affinity of the acridine drugs to DNA often correlates with
their antiproliferative activity.38 Therefore, the relative intercalation
tion) in lmol/L.

ˇ
O. Sedlácek et al. / Bioorg. Med. Chem. 20 (2012) 4056–4063
4060
Figure 5a. The cellular internalization of 9-anilinoacridine 2e (c = 5 lg/mL). The images are sequentially shown according to the channels used (left to right): brightfield,
Hoechst 33342 dye (cell nuclei stained), acridine fluorescence, merged channels of Hoechst 33342 dye and acridine fluorescence.
showed by HPLC that free drug in its original form is released from
the conjugates after incubation in media. Therefore, the data on
in vitro cytotoxicity of the conjugates may be misleading due to
significantly different concentrations of the drug released into
the media during incubation with the cells (pH 7.4) compared to
the in vivo situation. This is because in an in vivo situation, the sys-
tem is opened, i.e. the released drug is being continuously removed
by internalization into cells or diffusion out of the tumor tissue. In
addition, pH in tumor tissue is generally slightly acidic, but varies
according to the exact site in the tumor within 1–1.5 pH units,
which has dramatic effect on the drug release rate (see above)
Figure 5b. The cellular internalization of polymer conjugate
anilinoacridine 2e equivalent)/mL). The images are sequentially shown according
to the channels used (left to right): brightfield, acridine fluorescence.
8 (c = 5 lg (9-
and therefore the published IC50 values of hydrazone conjugates,
typically one order of magnitude higher than the IC50 values of free
drugs,21–23 generally do not correspond with the in vivo antitumor
effectivity.
The intracellular fate of the 9-anilinoacridines and their conju-
gates was studied by confocal microscopy. The free 9-anilinoacri-
dine derivative 2e, which is significantly cytotoxic due to
intercalation with DNA, penetrates into cells and subsequently into
cell nuclei (Fig. 5a). A portion of the acridine-related fluorescence
was observed in vesicles inside the cells. The fluorescence of 2e
belonging, with high probability, to its polymer conjugate 8, was
clearly visible within intracellular vesicles of the cells (Fig. 5b).
However, its fluorescence in cell nuclei was less than that observed
from cells incubated with the same amount of free drug. This result
is probably due to a portion of 2e that was released from the con-
jugate and intercalated in DNA as well as portion of conjugate
internalized into cells and subsequently released in endosomes.
The results are also influenced by the fact that the fluorescence
intensity of acridines greatly decreases after intercalation.14
calate into DNA and readily penetrate into the cell, as shown by
confocal microscopy.
4. Experimental part
The 3-(9-acridinylamino)-5-hydroxymethylaniline (AHMA, 1)
was prepared from 9-chloroacridine and 3-amino-5-(hydroxy-
methyl) aniline in 92% yield, according to the literature procedure.8
The p-oxopropylbenzoic acid (9b) was obtained from Rieke Metal.
The 5-methyl-4-oxohexanoic acid (9d) was synthesized in 68%
yield, according to the literature procedure.26 The poly[N-(2-
hydroxypropyl)methacrylamide-co-1-N-(6-hydrazino-6-oxohex-
yl)-2-methylacrylamide] (pHPMA-MAAcap hydrazide, weight
average molecular weight Mw = 17.5 kDa, polydispersity I = Mw/
Mn = 1.87, where Mn is number-average molecular weight, 7.5 mol%
hydrazide monomer unit) was synthesized by radical copolymeriza-
tion of N-(2-hydroxypropyl)methacrylamide with N1-(6-hydrazino-
6-oxohexyl)-2-methylacrylamide (5) using azobis(isobutyronitrile)
as an initiator according to reference.27 All other chemicals were pur-
chased from Sigma–Aldrich Ltd. (Prague, Czech Republic) and were
used without further purification. Analyses were performed on a
HPLC chromatograph (Shimadzu, Japan) using a reverse-phase col-
umn (Chromolith Performance RP-18e 100 ꢀ 4.6 mm) and UV detec-
tion. A mixture of water and acetonitrile was used as the eluent with
a gradient of 0–100 vol.% and a flow rate of 2 mL/min. The melting
point temperatures were determined with a Kofler’s block (VEB Anal-
ytik Dresden, Germany). NMR spectra were measured with a Bruker
Avance MSL 300 MHz NMR spectrometer (Bruker Daltonik, Ger-
many). The molecular weights of the polymers were determined by
gel permeation chromatography (GPC) using an HPLC Shimadzu sys-
tem equipped with a GPC column (TSKgel G3000SWxl 300 ꢀ 7.8 mm,
3. Conclusions
New types of biodegradable polymer-conjugates of anilinoacri-
dines were synthesized. In these conjugates, the acridine drug
derivatives were connected to the water-soluble polymer
(poly[N-(2-hydroxypropyl)methacrylamide-co-N1-(6-hydrazino-6-
oxohexyl)methacrylamide]) via a pH-labile hydrazone bond. The
dependence of the drug release on the linker structure was exam-
ined. The best release profile, where the conjugate was almost
completely stable at physiological pH (7.4) but sufficiently labile
at the pH of late endosomes (pH 5.0), showed the conjugate con-
taining a linker with the strongest steric hindrance of the original
oxo-group. However, this conjugate could not be easily synthe-
sized by reaction of the ketone linker-bearing acridine drug with
the hydrazide group-containing polymer because of unwanted side
reactions. Therefore, the acridine drug-containing monomer was
synthesized first, followed by its radical copolymerization. All of
the synthesized acridine drugs retain their in vitro antiproliferative
activity (the free intercalators have IC50 values in the range of
5 l
m), UV/VIS, refractive index (RI) OptilabÒ-rEX and multiangle light
scattering (MALS) DAWN EOS (Wyatt Technology Co., USA) detectors
using a methanol and sodium acetate buffer (0.3 M, pH 6.5) mixture
(80:20 vol.%, flow rate of 0.5 mL/min). UV/VIS spectra were measured
with a SPECORD 205 Spectrometer (Analytik Jena AG, Germany).
0.6–6.8 lmol/L in MCF-7, HepG-2 and PC3 tumor cell lines), inter-

ˇ
O. Sedlácek et al. / Bioorg. Med. Chem. 20 (2012) 4056–4063
4061
4.1. Synthesis of dimethyl pivalylmethylmalonate
(300 MHz, CDCl3)
d
ppm 7.63 (d, J = 8.34 Hz, 2H), 7.19 (d,
J = 8.43 Hz, 2H), 4.47 (t, J = 7.24 Hz, 2H), 3.69 (s, 2H), 3.41 (t,
J = 7.24 Hz, 2H), 2.11 (s, 3H); 13C NMR (75 MHz, CDCl3) d ppm
205.2, 202.0, 171.0, 139.4, 132.4, 130.0, 129.5, 56.6, 50.8, 29.8,
29.6; MS (ESI+) m/z 279.38 (calcd): 280.25 [M+H]+; Anal. calcd
for C13H13NO2S2: 55.89% C, 4.69% H, 5.01% N; found: 55.79% C,
4.64% H, 4.96% N.
Sodium methoxide (615 mg, 11.4 mmol) in methanol (5 mL)
was added dropwise to the solution of dimethyl malonate
(1.3 mL, 11.4 mmol) in methanol (50 mL). After stirring for
30 min, bromopinacolone (2 g, 11.4 mmol) was added dropwise
and the mixture was refluxed for 2 h. The solution was partitioned
between water and diethyl ether. The organic phase was separated,
dried with anhydrous magnesium sulfate and evaporated under re-
duced pressure. The crude product was purified by column chroma-
tography (silica, hexane/ethyl acetate 4:1) to obtain 1.28 g (49%) of
the title compound as colorless oil. Rf (hexane/ethyl acetate
4:1) = 0.56; 1H NMR (300 MHz, CDCl3) d ppm 3.88 (t, J = 7.18, 1H),
3.74 (s, 6H), 3.13 (d, J = 7.18 Hz, 2H), 1.16 (s, 9H); 13C NMR
(75 MHz, CDCl3) d ppm 212.6, 169.6, 52.9, 46.8, 44.0, 36.3, 26.5;
MS (ESI+) m/z 230.26 (calcd): 253.24 [M+Na]+, Anal. calcd for
C11H18O5: 57.38% C, 7.88% H; found: 57.36% C, 7.90% H.
4.3.3. 1-[4-(2-Thioxo-thiazolidine-3-carbonyl)-phenyl]-
ethanone (3c)
Synthesized from p-acetylbenzoic acid (9c) as yellow solid in
70% yield. Rf (hexane-ethyl acetate 4:1) = 0.54; 1H NMR
(300 MHz, CDCl3)
d ppm 7.90 (d, J = 8.60 Hz, 2H), 7.68 (d,
J = 8.61 Hz, 2H), 4.51 (t, J = 7.24 Hz, 2H), 3.45 (t, J = 7.24 Hz, 2H),
2.57 (s, 3H); 13C NMR (75 MHz, CDCl3) d ppm 202.0, 197.3, 170.4,
139.5, 137.9, 129.4, 128.2, 56.3, 29.9, 26.9; MS (ESI+) m/z 265.35
(calcd): 266.26 [M+H]+; Anal. calcd for C12H11NO2S2: 54.32% C,
4.18% H, 5.28% N; found: 54.17% C, 4.21% H, 5.25% N.
4.2. Synthesis of 5,5-dimethyl-4-oxohexanoic acid
(pivalylpropionic acid, 9e)
4.3.4. 5-Methyl-1-(2-thioxothiazolidin-3-yl)hexane-1,4-dione
(3d)
Dimethyl pivalylmalonate (1.2 g, 5.2 mmol) and KOH (1 g,
17.8 mmol) were refluxed in 50% aqueous methanol (10 mL) for
2 h. The reaction mixture was then neutralized with 1 M HCl and
the resulting malonic acid derivative was precipitated by excess
of aqueous calcium chloride as described by Hill.39 The white solid
was filtered off, dissolved in 1 M HCl, extracted in diethyl ether and
evaporated. The resulting malonic acid derivative was decarboxyl-
ated by heating at 145 °C for 2 h without solvent. After cooling, the
crude product was recrystalized from petroleum ether to give
640 mg (78%) of the title compound 9e as white needles.
Synthesized from 5-methyl-4-oxohexanoic acid (9d) as yellow
oil in 86% yield. Rf (hexane/ethyl acetate 4:1) = 0.52; 1H NMR
(300 MHz, CDCl3)
d ppm 4.54 (t, J = 7.53 Hz, 2H), 3.44 (t,
J = 5.88 Hz, 2H), 3.30 (t, J = 7.53 Hz, 2H), 2.85 (t, J = 5.91 Hz, 2H),
2.71–2.59 (m, J = 6.92 Hz, 1H), 1.12 (d, J = 6.96 Hz, 9H); 13C NMR
(75 MHz, CDCl3) d ppm 212.9, 201.6, 174.0, 56.0, 40.8, 34.8, 33.2,
28.5, 18.3; MS (ESI+) m/z 245.36 (calcd): 246.23 [M+H]+; Anal.
calcd for C10H15NO2S2: 48.95% C, 6.16% H, 5.71% N; found:
48.85% C, 6.20% H, 5.66% N.
Mp = 66–69 °C; 1H NMR (300 MHz, CDCl3)
d ppm 2.80 (t,
4.3.5. 5,5-Dimethyl-1-(2-thioxothiazolidin-3-yl)hexane-1,4-
dione (3e)
J = 6.54 Hz, 2H), 2.60 (t, J = 6.38 Hz, 2H), 1.15 (s, 9H); 13C NMR
(75 MHz, CDCl3) d ppm 213.9, 179.1, 43.8, 31.2, 28.0, 26.4; MS
(ESI+) m/z 158.20 (calcd): 181.08 [M+Na]+; Anal. calcd for
C8H14O3: 60.74% C, 8.92% H; found: 60.71% C, 8.95% H.
Synthesized from 5,5-dimethyl-4-oxohexanoic acid (9e) as yel-
low oil in 73% yield. Rf (hexane/ethyl acetate 4:1) = 0.60; 1H NMR
(300 MHz, CDCl3)
d ppm 4.49 (t, J = 7.53 Hz, 2H), 3.37 (t,
J = 5.88 Hz, 2H), 3.25 (t, J = 7.53 Hz, 2H), 2.85 (t, J = 5.78 Hz, 2H),
1.12 (s, 9H); 13C NMR (75 MHz, CDCl3) d ppm 214.3, 201.6, 174.1,
56.1, 44.0, 33.2, 31.6, 28.6; MS (ESI+) m/z 259.39 (calcd): 260.27
[M+H]+; Anal. calcd for C11H17NO2S2: 50.94% C, 6.61% H, 5.40% N;
found: 50.81% C, 6.63% H, 5.76% N.
4.3. General procedure for synthesis of thiazolidine-2- thione
(TT) activated oxoacids 3a–e
A mixture of the respective oxoacid 9a–e (2 mmol) and thiazol-
idine-2-thione (262 mg, 2.2 mmol) in dichloromethane (3 mL) was
cooled to 0 °C in an ice-water bath. Then, N,N0-dicycloehexylcarbo-
diimide (DCC, 515 mg, 2.5 mmol) in dichloromethane (2 mL) was
added. The reaction mixture was allowed to warm to ambient tem-
perature and stirred for 2 days. The precipitated N,N0-dicyclohexyl-
urea was filtered off and the filtrate was evaporated under reduced
pressure. The crude product was purified by column chromatogra-
phy (silica, hexane/ethyl acetate 4:1) yielding activated acid deriv-
atives 3a–e.
4.4. General procedure for the synthesis of AHMA-amides 2a–e
3-(9-Acridinylamino)-5-hydroxymethylaniline
(AHMA,
1,
158 mg, 0.5 mmol) and the respective TT-activated oxo-acid 3a–e
(0.6 mmol) were stirred in N,N-dimethylformamide (DMF) at room
temperature overnight. The solvent was evaporated under reduced
pressure, and the residue was purified by column chromatography
(silica, chloroform-methanol 4:1) followed by crystallization from
ethanol.
4.3.1. 1-(2-Thioxo-thiazolidin-3-yl)-pentane-1,4-dione (3a)
Synthesized from levulinic acid (9a) as yellow solid in 84% yield.
Rf (hexane-ethyl acetate 4:1) = 0.45; 1H NMR (300 MHz, CDCl3) d
ppm 4.50 (t, J = 7.54 Hz, 2H), 3.39 (t, J = 6.02 Hz, 2H), 3.26 (t,
J = 7.54 Hz, 2H), 2.77 (t, J = 5.92 Hz, 2H), 2.15 (s, 3H); 13C NMR
(75 MHz, CDCl3) d ppm 206.9, 201.6, 173.7, 56.0, 37.9, 33.2, 29.9,
28.5; MS (ESI+) m/z 217.31 (calcd): 218.22 [M+H]+; Anal. calcd
for C8H11NO2S2: 44.22% C, 5.10% H, 6.45% N; found: 44.24% C,
5.18% H, 6.40% N.
4.4.1. 3-(9-Acridinylamino)-5-hydroxymethylaniline N-
levulinate (2a)
Activated oxo-acid 3a was used. Yield 79% of orange crystals. Rf
(chloroform/methanol 4:1) = 0.38; mp (ethanol) = 198–201 °C; 1H
NMR (300 MHz, DMSO-d6) d ppm 9.75 (s, 1H), 8.06 (brs, 2H), 7.42
(s, 4H), 7.25 (s, 1H), 7.14 (s, 3H), 6.83 (1H), 5.07 (br s, 1H), 4.34
(s, 2H), 2.63 (t, J = 6.05 Hz, 2H), 2.44 (t, J = 6.13, 2H), 2.04 (s, 3H);
13C NMR (75 MHz, DMSO-d6) d ppm 207.4, 170.1, 154.6, 150.7,
144.4, 142.8, 140.3, 137.6, 131.2, 127.1, 124.4, 120.5, 116.7,
109.9, 106.3, 63.0, 37.6, 30.1, 29.7; MS (ESI+) m/z 413.48 (calcd):
414.37 [M+H]+; Anal. calcd for C25H23N3O3: 72.62% C, 5.61% H,
10.16% N; found: 72.59% C, 5.67% H, 10.28% N.
4.3.2. 1-[4-(2-Thioxo-thiazolidine-3-carbonyl)-phenyl]-propan-
2-one (3b)
Synthesized from p-(2-oxopropyl)-benzoic acid (9b) as yellow
solid in 67% yield. Rf (hexane/ethyl acetate 4:1) = 0.52; 1H NMR

ˇ
O. Sedlácek et al. / Bioorg. Med. Chem. 20 (2012) 4056–4063
4062
4.4.2. 3-(9-Acridinylamino)-5-hydroxymethylaniline N-(p-
oxopropylbenzoate) (2b)
(t, J = 7.24 Hz, 2H), 1.79 (s, 3H), 1.58–1.16 (m, 6H); 13C NMR
(75 MHz, DMSO-d6) d ppm 169.1, 167.3, 145.9, 140.8, 140.1, 138.5,
133.4, 126.0, 120.9, 120.4, 118.6, 38.8, 34.3, 29.0, 26.3, 25.1, 18.7;
MS (ESI+) m/z 390.49 (calcd): 391.30 [M+H]+; Anal. calcd for
Activated oxo-acid 3b was used. Yield 81% of orange crystals. Rf
(chloroform/methanol 4:1) = 0.48; mp (ethanol) = 218–221 °C; 1H
NMR (300 MHz, DMSO-d6) d ppm 10.11 (s, 1H), 8.01 (s, 2H), 7.81
(d, J = 8.28 Hz, 2H), 7.63 (s, 4H), 7.48 (s, 1H), 7.35 (s, 1H), 7.25 (d,
J = 8.29 Hz, 2H), 7.08 (s, 2H), 6.64 (s, 1H), 5.16 (br s, 1H), 4.40 (s,
2H), 3.81 (s, 2H), 2.10 (s, 3H); 13C NMR (75 MHz, DMSO-d6) d
ppm 205.5, 165.3, 151.4, 147.2, 144.5, 142.9, 140.3, 138.6, 135.0,
133.7, 133.1, 132.3, 129.6, 127.6, 120.5, 119.1, 113.8, 111.6,
107.2, 62.7, 49.2, 29.6; MS (ESI+) m/z 475.55 (calcd): 476.36
[M+H]+; Anal. calcd for C30H25N3O3: 75.77% C, 5.30% H, 8.84% N;
found: 75.72% C, 5.31% H, 8.80% N.
C23H26N4O2: 70.75% C, 6.71% H, 14.35% N; found: 70.53% C, 6.75%
H, 14.29% N.
4.6. 5,5-Dimethyl-4-{[6-(2-methyl-acryloylamino)-hexanoyl]-
hydrazono}-hexanoic acid [3-(acridin-9-ylamino)-5-hydroxy
methyl-phenyl]-amide (7)
Pivalylpropionic acid (9e, 158 mg, 1 mmol), 6-methacry-
lamidohexanohydrazide (5, 214 mg, 1 mmol) and 2,5-di-tert-butyl-
hydroquinone (5 mg) were stirred with 50 lL of acetic acid in
4.4.3. 3-(9-Acridinylamino)-5-hydroxymethylaniline N-(p-acety
lbenzoate) (2c)
methanol (2 mL) at 60 °C overnight. The mixture was evaporated
under reduced pressure, dissolved in DMF (3 mL) and cooled to
0 °C. N,N’-Dicyclohexyl carbodiimide (DCC, 309 mg, 1.5 mmol)
and N,N-dimethylaminopyridine (10 mg) were added, and the mix-
ture was stirred while cooling. After 2 h, 3-(9-acridinylamino)-5-
hydroxymethylaniline (AHMA, 1, 316 mg, 1 mmol) was added.
The reaction mixture was allowed to warm to ambient tempera-
ture and stirred overnight. The solvents were evaporated under re-
duced pressure, the residue was extracted with dichloromethane
and the suspension was filtered. The filtrate was evaporated and
purified on preparative TLC (silica, chloroform/methanol/triethyl-
amine 9:1:0.2) to give the title compound 7 as an orange solid in
26% yield. Rf (chloroform/methanol/triethylamine 9:1:0.2) = 0.40;
1H NMR (300 MHz, DMSO-d6) d ppm 10.01 (br s, 1H), 8.03–7.31
(m, 11H), 6.63 (s, 1H), 5.56 (s, 1H), 5.26 (s, 1H), 5.22 (t, 1H), 4.34
(s, 2H), 3.25 (t, J = 6.84 Hz, 2H), 2.87 (t, J = 6.50 Hz, 2H), 2.52 (t,
J = 6.54 Hz, 2H), 2.29 (t, J = 6.43 Hz, 2H), 2.07 (s, 3H), 1.65–1.41
(m, 6H), 1.08(s, 9H); 13C NMR (75 MHz, DMSO-d6) d ppm 177.4,
171.2, 169.8, 166.7, 150.0, 149.1, 147.6, 143.5, 140.6, 140.2,
132.6, 125.9, 121.7, 119.3, 117.4, 116.8, 113.5, 112.2, 108.7, 62.4,
40.6, 39.8, 35.8, 32.5, 27.0, 26.3, 24.4, 23.9, 22.7, 18.5; MS (ESI+)
m/z 650.83 (calcd): 651.69 [M+H]+; Anal. calcd for C38H46N6O4:
70.13% C, 7.12% H, 12.91% N; found: 70.040% C, 7.09% H, 12.87% N.
Activated oxo-acid 3c was used. Yield 74% of orange crystals. Rf
(chloroform/methanol 4:1) = 0.49; mp (ethanol) = 236–239 °C; 1H
NMR (300 MHz, DMSO-d6) d ppm 10.25 (s, 1H), 8.00 (br s, 6H),
7.39 (s, 4H), 7.28 (s, 2H), 7.08 (s, 1H), 6.44 (s, 1H), 5.14 (br s, 1H),
4.43 (s, 2H), 2.58 (s, 3H); 13C NMR (75 MHz, DMSO-d6) d ppm
197.6, 164.5, 153.8, 150.2, 149.4, 144.5, 143.9, 139.9, 138.8, 131.3,
130.2, 129.5, 128.1, 127.9, 124.4, 120.7, 111.4, 110.5, 107.8, 62.9,
26.9; MS (ESI+) m/z 461.52 (calcd): 462.36 [M+H]+; Anal. calcd for
C29H23N3O3: 75.47% C, 5.02% H, 9.10% N; found: 75.33% C, 5.06%
H, 9.04% N.
4.4.4. 3-(9-Acridinylamino)-5-hydroxymethylaniline N-(5-methyl
-4-oxohexanoate) (2d)
Activated oxo-acid 3d was used. Yield 82% of orange crystals. Rf
(chloroform/methanol 4:1) = 0.43; mp (ethanol) = 166–169 °C; 1H
NMR (300 MHz, DMSO-d6) d ppm 9.91 (s, 1H), 7.99 (s, 2H), 7.65 (s,
4H), 7.27 (s, 1H), 7.15 (s, 3H), 6.58 (s, 1H), 5.15 (s, 1H), 4.36 (s, 2H),
2.69 (t, J = 6.41 Hz, 2H), 2.57 (m, J = 6.92 Hz, 1H), 2.45 (t,
J = 6.40 Hz, 2H), 0.95 (d, J = 6.92 Hz, 6H); 13C NMR (75 MHz,
DMSO-d6) d ppm 213.5, 170.9, 151.9, 145.3, 142.5, 141.7, 141.0,
133.4, 126.9, 122.7, 121.4, 119.8, 117.2, 112.7, 109.4, 63.3, 40.9,
35.0, 30.6, 18.7; MS (ESI+) m/z 441.53 (calcd): 442.38 [M+H]+; Anal.
calcd for C27H27N3O3: 73.45% C, 6.16% H, 9.52% N; found: 73.29% C,
6.16% H, 9.47% N.
4.7. Conjugation of 2a-e with pHPMA-MAAcap hydrazide
A solution of pHPMA-MAAcap hydrazide (40 mg), AHMA deriv-
4.4.5. 3-(9-Acridinylamino)-5-hydroxymethylaniline N-(5,5-
dimethyl-4-oxohexanoate) (2e)
ative 2a–e (10 mg) and glacial acetic acid (80
lL) in methanol
(800 L) was stirred at room temperature overnight. The polymer
l
Activated oxo-acid 3e was used. Yield 90% of orange crystals. Rf
(chloroform/methanol 4:1) = 0.49; mp (ethanol) = 145–148 °C; 1H
NMR (300 MHz, DMSO-d6) d ppm 9.91 (s, 1H), 7.97 (s, 2H), 7.64
(s, 4H), 7.27 (s, 1H), 7.14 (s, 3H), 6.56 (s, 1H), 5.16 (s, 1H), 4.36
(s, 2H), 2.75 (t, J = 6.39 Hz, 2H), 2.45 (t, J = 6.41 Hz, 2H), 1.03 (s,
9H); 13C NMR (75 MHz, DMSO-d6) d ppm 214.1, 170.4, 151.2,
144.7, 141.3, 140.5, 132.6, 126.4, 122.0, 119.3, 119.2, 116.7,
112.4, 111.8, 108.6, 62.7, 43.3, 31.1, 30.1, 26.3; MS (ESI+) m/z
455.56 (calcd): 456.39 [M+H]+; Anal. calcd for C28H29N3O3:
73.82% C, 6.42% H, 9.22% N; found: 73.75% C, 6.40% H, 9.21% N.
was then isolated by gel filtration on a Sephadex LH-20 column
(60 mL bed volume) with methanol as the eluent, and the poly-
mer-containing fractions were collected and concentrated under
reduced pressure. The solution was poured into an excess of
diethyl ether, and the precipitated polymer conjugate was filtered
off and dried under vacuum. A yield of 40-45 mg (80–90%) of 4a–e
was obtained.
4.8. Synthesis of polymer 8 by radical copolymerization
Monomer 7 (12 mg) was mixed with N-(2-hydroxypropyl)-
methacrylamide (88 mg) and azobis(isobutyronitrile) (AIBN,
20 mg) in dry dimethylsulfoxide (0.6 mL), introduced under argon
into a polymerization ampoule and sealed. The reaction mixture
was stirred under argon at 60 °C overnight. The polymer was then
precipitated with diethyl ether and purified by gel filtration on a
Sephadex LH-20 column (60 mL bed volume) using methanol as
the eluent. A yield of 55 mg (55%) of polymer 8 was obtained.
4.5. N-[5-(N0-Acridin-9-yl-hydrazinocarbonyl)-pentyl]-2-methyl-
acrylamide (6)
9-Anilinoacridine 2e (227 mg, 0.5 mmol) and 6-methacry-
lamidohexanohydrazide (5, 107 mg, 0.5 mmol) were stirred with
50 lL of acetic acid in methanol (2 mL) at room temperature over-
night. The solvent was evaporated under reduced pressure, and the
residue was purified by column chromatography (silica, chloro-
form/methanol 9:1) to give 72 mg (37% yield) of 6 as an orange
solid. Rf (chloroform-methanol 9:1) = 0.56; 1H NMR (300 MHz,
DMSO-d6) d ppm 10.52 (br s, 1H), 8.26–6.97 (m, 10H), 5.58 (s, 1H),
5.23 (t, J = 1.50, 1H), 3.07 (dd, J = 12.90, 6.74 Hz, 2H), 2.23
4.9. Determination of drug content
The molar ratio of the cleavable (i) versus non-cleavable (ii) acri-
dine drug bound to the polymer was determined by size exclusion

ˇ
O. Sedlácek et al. / Bioorg. Med. Chem. 20 (2012) 4056–4063
4063
chromatography (SEC) on a TSK3000 column with UV detection at
408 nm after hydrolysis with a 0.1 M HCl solution. It was calcu-
lated as a wagered ratio of polymer peak to low molecular weight
peak area at 408 nm using the equation (1), where Ip is the area un-
der the polymeric peak representing the non-cleavably bound acri-
dine content, Ia is the area under the low molecular weight peak
representing the hydrazone-bound anilinoacridine, en is the molar
absorption coefficient of the non-cleavable monomer derivative 6 at
408 nm (8761 L molꢁ1 cmꢁ1), eh is the molar absorption coefficient
of the hydrazone monomer 7 at 408 nm (8490 L molꢁ1 cmꢁ1).
the intensive acridine fluorophore caused fluorescence in analogy
as described for the doxorubicin conjugates.30 The cell nuclei were
also visualized using the Hoechst 33342 nucleic acid dye. Statisti-
cal evaluation of the differences (analysis of variance, ANOVA) was
performed on level
a = 0.01 using Microcal Origin program, version
5.0 (Microcal Software Inc., Northampton, MA, U.S.A.).
Acknowledgement
Financial support from the Grant Agency of the Academy of
Sciences of the Czech Republic (Grant #IAAX00500803) and from
the Grant Agency of the Czech Republic (Grant #P207/10/P054)
is gratefully acknowledged.
Ip Ia
i : ii ¼
:
en eh
Because the UV–VIS absorption spectra and molar absorptivities
of the acridine derivatives 2a–e do not substantially differ from
that of 7, the same eh value was used for all acridine conjugates.
The total amount of acridine drug in the conjugate was determined
using the sum of the wagered peak areas (i and ii in (1)).
References and notes
1. Atwell, G. J.; Cain, B. F.; Seelye, R. N. J. Med. Chem. 1972, 15, 611.
2. Baguley, B. C.; Nash, R. Eur. J. Cancer 1965, 1981(17), 671.
3. Ferguson, L. R.; Macphee, D. G.; Baguley, B. C. Chem. Biol. Interact. 1983, 44, 53.
4. Legha, S. S.; Keating, M. J.; Zander, A. R.; McCredie, K. B.; Bodey, G. P.; Freireich,
E. J. Ann. Intern. Med. 1980, 93, 17.
5. Baguley, B. C.; Denny, W. A.; Atwell, G. J.; Cain, B. F. J. Med. Chem. 1981, 24, 170.
6. Denny, W. A.; Cain, B. F.; Atwell, G. J.; Hansch, C.; Panthananickal, A.; Leo, A. J.
Med. Chem. 1982, 25, 276.
7. Ferguson, L. R.; Denny, W. A. J. Med. Chem. 1979, 22, 251.
8. Su, T.-L.; Chou, T.-C.; Kim, J. Y.; Huang, J.-T.; Ciszewska, G.; Ren, W.-Y.; Otter, G.
M.; Sirotnak, F. M.; Watanabe, K. A. J. Med. Chem. 1995, 38, 3226.
9. Jurlina, J. L.; Lindsay, A.; Packer, J. E.; Baguley, B. C.; Denny, W. A. J. Med. Chem.
1987, 30, 473.
10. Robertson, I. G. C.; Palmer, B. D.; Paxton, J. W.; Shaw, G. J. Xenobiotica 1992, 22,
657.
11. Chang, J.-Y.; Lin, C.-F.; Pan, W.-Y.; Bacherikov, V.; Chou, T.-C.; Chen, C.-H.;
Dong, H.; Cheng, S.-Y.; Tasi, T.-J.; Lin, Y.-W.; Chen, K.-T.; Chen, L.-T.; Su, T.-L.
Bioorg. Med. Chem. 2003, 11, 4959.
12. Su, T.-L.; Chen, C.-H.; Huang, L.-F.; Chen, C.-H.; Basu, M. K.; Zhang, X.-G.; Chou,
T.-C. J. Med. Chem. 1999, 42, 4741.
13. Boos, G.; Stopper, H. Toxicol. Lett. 2000, 116, 7.
4.10. In vitro drug release study
The drug release profiles of the polymer conjugates were eval-
uated by size exclusion chromatography (SEC) at pH 5.0 and 7.4
(also, at pH 6.5 for conjugate 8) using 10 mM acetate buffer (pH
5.0 or 6.5) or phosphate buffer saline (pH 7.4). The conjugate
(0.1 mg) was dissolved in the appropriate buffer (1 mL) and placed
into a thermostatted autosampler of the SEC instrument (37 °C).
After 0, 2, 6, 12 and 24 h, 20 lL of the conjugate solution was in-
jected directly on the TSK3000 column. The drug release was deter-
mined from the integral areas of the low molecular weight (free
drug) and high molecular weight (polymer conjugate) peaks. After
the last measurement (24 h), the solutions were analyzed by re-
verse-phased HPLC chromatography to confirm that only the anili-
noacridine derivatives 2a–e were released.
14. Ferlin, M. G.; Marzano, C.; Chiarelotto, G.; Baccichetti, F.; Bordin, F. Eur. J. Med.
Chem. 2000, 35, 827.
15. Fyfe, D.; Price, C.; Langley, R. E.; Pagonis, C.; Houghton, J.; Osborne, L.; Woll, P.
J.; Gardner, C.; Baguley, B. C.; Carmichael, J. Cancer Chemother. Pharmacol. 2001,
47, 333.
4.11. In vitro ethidium bromide displacement assay
16. Allan, S. H. J. Controlled Release 2008, 132, 153.
17. Maeda, H. Bioconjugate Chem. 2010, 21, 797.
18. Knop, K.; Hoogenboom, R.; Fischer, D.; Schubert, U. S. Angew. Chem., Int. Ed.
2010, 49, 6288.
19. Talelli, M.; Rijcken, C. J. F.; van Nostrum, C. F.; Storm, G.; Hennink, W. E. Adv.
Drug Deliv. Rev. 2010, 62, 231.
20. Pike, D. B.; Ghandehari, H. Adv. Drug Deliv. Rev. 2010, 62, 167.
21. Etrych, T.; Chytil, P.; Jelinkova, M.; Rihova, B.; Ulbrich, K. Macromol. Biosci.
2002, 2, 43.
22. Kaneko, T.; Willner, D.; Monkovic, I.; Knipe, J. O.; Braslawsky, G. R.; Greenfield,
R. S.; Vyas, D. M. Bioconjugate Chem. 1991, 2, 133.
23. Ulbrich, K.; Etrych, T.; Chytil, P.; Jelinkova, M.; Rihova, B. J. Controlled Release
2003, 87, 33.
24. Ulbrich, K.; Subr, V. Adv. Drug Deliv. Rev. 2004, 56, 1023.
25. Krakovicova, H.; Etrych, T.; Ulbrich, K. Eur. J. Pharm. Sci. 2009, 37, 405.
26. Barberis, M.; Pérez-Prieto, J. Tetrahedron Lett. 2003, 44, 6683.
27. Etrych, T.; Mrkvan, T.; Chytil, P.; Konak, C.; Rihova, B.; Ulbrich, K. J. Appl. Polym.
Sci. 2008, 109, 3050.
The ethidium bromide displacement assay was performed using
a solution of 1.26
determined spectrophotometrically,
and 1 M ethidium bromide in HEPES buffer (10 mM HEPES,
l
M(bp) calf thymus (CT) DNA (concentration
e
260 = 12,824 M(bp)ꢁ1 cmꢁ1
)
l
76 mM NaCl, 0.1 mM ethylenediaminetetraacetic acid, pH = 7.0).27
This solution was titrated with a solution of the measured drug in
the same buffer to determine the drug concentration, CC50, whichre-
duced the fluorescence (excitation 546 nm, emission 595 nm) of the
DNA–ethidium bromide complex by half. None of the compounds
showed fluorescence emission at 595 nm. All experiments were
carried out in triplicate.
4.12. In vitro cytotoxicity assay and cellular internalization
28. In Drug–DNA Interaction Protocols, 2nd ed.; Fox, K. R., Ed., 2010; Vol. 613, p. 1.
29. Mosmann, T. J. Immunol. Methods 1983, 65, 55.
30. Hovorka, O.; Etrych, T.; Subr, V.; Strohalm, J.; Ulbrich, K.; Rihova, B. J. Drug
Target. 2006, 14, 391.
In vitro cytotoxicity of the 9-anilinoacridine derivatives was as-
sessed with human cancer cell lines. Ninety-six-well flat-bottomed
microplates (NUNC, Roskilde, Denmark) were seeded with MCF-7
(breast cancer), HepG-2 (hepatocellular liver carcinoma) and
PC-3 (prostate adenocarcinoma) cancer cells (5 ꢀ 104 cells/well).
The tested samples (in triplicate) were then added to the wells to
obtain the desired drug concentration (0.1–2660 ng mLꢁ1 free drug
equivalent). The plates were incubated at 37 °C in a humidified
atmosphere with 5% CO2 and 95% air for 72 h. The cell viability
was then evaluated by a standard MTT test according to a refer-
ence29 on a plate reader UT-2100C (MRC, Israel).
31. Šubr, V.; Ulbrich, K. React. Funct. Polym. 2006, 66, 1525.
32. Ulbrich, K.; Subr, V. Adv. Drug Deliv. Rev. 2010, 62, 150.
33. Khan, M. N.; Kuliyaumar, A. F. Bioorg. Med. Chem. 1995, 3, 881.
34. Khan, M. N.; Malspeis, L. J. Org. Chem. 1982, 47, 2731.
35. Paul, A.; Ladame, S. Org. Lett. 2009, 11, 4894.
36. Yeung, T. K.; Hopewell, J. W.; Simmonds, R. H.; Seymour, L. W.; Duncan, R.;
Bellini, O.; Grandi, M.; Spreafico, F.; Strohalm, J.; Ulbrich, K. Cancer Chemother.
Pharmacol. 1991, 29, 105.
37. Yang, J.; Chen, H. T.; Vlahov, I. R.; Cheng, J. X.; Low, P. S. J. Pharmacol. Exp. Ther.
2007, 321, 462.
38. Wakelin, L. P. G.; Atwell, G. J.; Rewcastle, G. W.; Denny, W. A. J. Med. Chem.
1987, 30, 855.
39. Hill, G. A.; Salvin, V.; O’Brien, W. T. M. J. Am. Chem. Soc. 1937, 59, 2385.
Internalization of the acridine derivatives and conjugates into
murine MCF-7 cells was assayed by confocal microscopy using