Polyaminoquinoline iron chelators for vectorization of antiproliferative agents: Design, synthesis, and validation

Article abstract of DOI:10.1021/bc300324c

Iron chelation in tumoral cells has been reported as potentially useful during antitumoral treatment. Our aim was to develop new polyaminoquinoline iron chelators targeting tumoral cells. For this purpose, we designed, synthesized, and evaluated the biological activity of a new generation of iron chelators, which we named Quilamines, based on an 8-hydroxyquinoline (8-HQ) scaffold linked to linear polyamine vectors. These were designed to target tumor cells expressing an overactive polyamine transport system (PTS). A set of Quilamines bearing variable polyamine chains was designed and assessed for their ability to interact with iron. Quilamines were also screened for their cytostatic/cytotoxic effects and their selective uptake by the PTS in the CHO cell line. Our results show that both the 8-HQ moiety and the polyamine part participate in the iron coordination. HQ1-44, the most promising Quilamine identified, presents a homospermidine moiety and was shown to be highly taken up by the PTS and to display an efficient antiproliferative activity that occurred in the micromolar range. In addition, cytotoxicity was only observed at concentrations higher than 100 μ. We also demonstrated the high complexation capacity of HQ1-44 with iron while much weaker complexes were formed with other cations, indicative of a high selectivity. We applied the density functional theory to study the binding energy and the electronic structure of prototypical iron(III)-Quilamine complexes. On the basis of these calculations, Quilamine HQ1-44 is a strong tridentate ligand for iron(III) especially in the form of a 1:2 complex.

Full text of DOI:10.1021/bc300324c

Article
pubs.acs.org/bc
Polyaminoquinoline Iron Chelators for Vectorization of
Antiproliferative Agents: Design, Synthesis, and Validation
Vincent Corce,†,‡ Emmanuelle Morin,† Solene Guiheneuf,† Eric Renault,† Stephanie Renaud,‡
́
̀
́
́
Isabelle Cannie,‡ Raphael Tripier,§ Luís M. P. Lima,§ Karine Julienne,† Sebastien G. Gouin,†
́
̈
Olivier Loreal,‡ David Deniaud,*,† and Francois Gaboriau*,‡
̧
́
†LUNAM Universite,
et des Techniques, 2, rue de la Houssinier
‡INSERM, UMR991, CHRU Pontchaillou, 35033 Rennes, France; Universite
§CNRS, UMR 6521, Universite
de Brest, Laboratoire de Chimie, Electrochimie Molec
́
CEISAM, Chimie Et Interdisciplinarite,
́
Synthes
̀
e, Analyse, Model
́
isation, UMR CNRS 6230, UFR des Sciences
̀
e, BP 92208, 44322 NANTES Cedex 3, France
de Rennes1, 35043 Rennes, France
ulaires et Chimie Analytique, 6 Avenue Victor
́
́
́
Le Gorgeu, 29200 Brest, France
S
* Supporting Information
ABSTRACT: Iron chelation in tumoral cells has been reported as potentially
useful during antitumoral treatment. Our aim was to develop new polyaminoquino-
line iron chelators targeting tumoral cells. For this purpose, we designed,
synthesized, and evaluated the biological activity of a new generation of iron
chelators, which we named Quilamines, based on an 8-hydroxyquinoline (8-HQ)
scaffold linked to linear polyamine vectors. These were designed to target tumor
cells expressing an overactive polyamine transport system (PTS). A set of
Quilamines bearing variable polyamine chains was designed and assessed for their ability to interact with iron. Quilamines were
also screened for their cytostatic/cytotoxic effects and their selective uptake by the PTS in the CHO cell line. Our results show
that both the 8-HQ moiety and the polyamine part participate in the iron coordination. HQ1−44, the most promising Quilamine
identified, presents a homospermidine moiety and was shown to be highly taken up by the PTS and to display an efficient
antiproliferative activity that occurred in the micromolar range. In addition, cytotoxicity was only observed at concentrations
higher than 100 μM. We also demonstrated the high complexation capacity of HQ1−44 with iron while much weaker complexes
were formed with other cations, indicative of a high selectivity. We applied the density functional theory to study the binding
energy and the electronic structure of prototypical iron(III)-Quilamine complexes. On the basis of these calculations, Quilamine
HQ1−44 is a strong tridentate ligand for iron(III) especially in the form of a 1:2 complex.
and Deferasirox,14 more recently developed for the treatment of
INTRODUCTION
■
secondary overloads (thalassemias), inhibit the proliferation of
tumor cells in culture. Their antiproliferative effects are
modulated by their iron-chelating capacity and their cellular
uptake. We recently showed that the antiproliferative effect of
ICL670, a tridentate iron chelator with a hydroxyphenyltriazole
skeleton (Deferasirox, Novartis Pharma), and O-Trensox
resulted from the iron depletion induced in cells and the
inhibitory action on the polyamine metabolism.15,16 Deferox-
amine inhibits the growth of melanoma and hepatoma cells in
vitro and in vivo.17 This chelator was effective in the treatment
of leukemias and neuroblastomas in preliminary clinical trials.18
The high antiproliferative activity observed with iron chelators
derived from pyridoxal isonicotinoyl hydrazone, including
aroylhydrazones and thiosemicarbazones, was assumed to
involve both their iron-chelating efficiency and their ability to
generate reactive oxygen species.1 More recently, iron depletion
induced by the iron chelators Deferoxamine and 2-hydroxyl-1-
Iron is an essential metal in many biological functions and an
interesting and innovative target for anticancer therapy.1
Growing tumor cells require large quantities of iron, expressed
by an increase in the import proteins and a decrease in the
export proteins of iron.2,3 The activity and expression of
ribonucleotide reductase (RR), the iron-dependent key enzyme
regulating DNA synthesis and cell proliferation, are increased in
tumor cells.4 Previous studies have reported the relationship
between body iron stores and the risk of cancers.5,6 The iron
load stimulates the proliferation of tumor cells in vitro and in
vivo.7 The need for iron during the cell cycle is illustrated by the
overexpression of proliferating transferrin receptor cells on the
plasma membrane, which allows transferrin-chelated iron to
enter the cell.8 In contrast, the underexpression of the cellular
iron exporter, ferroportin, leads to an intracellular accumulation
of iron, and is a biomarker for tumor growth in breast
cancer.9,10 Decreasing the exchangeable intracellular iron level
with synthetic metal chelators is therefore an attractive method
for limiting the proliferation of cancer cells.11
Received: June 19, 2012
Revised: July 24, 2012
Published: August 8, 2012
Iron chelators, like Deferoxamine initially developed for the
treatment of primary iron overload, as well as Deferiprone12,13
© 2012 American Chemical Society
1952
dx.doi.org/10.1021/bc300324c | Bioconjugate Chem. 2012, 23, 1952−1968

Bioconjugate Chemistry
Article
naphtylaldehyde isonicotinoyl hydrazone was shown to increase
the expression of growth arrest and DNA damage genes.19
All these results suggest that, due to their antiproliferative
and apoptotic properties, iron chelators could be used as an
adjuvant in the treatment of cancers. The mechanism of action
of cytotoxic antitumor agents, like Doxorubicin and Bleomycin,
involves their iron-chelating ability to some extent. However,
such therapy must target tumoral cells. In fact, conversely to
what is observed in iron overload diseases, cells are not
overloaded and the iron stores are not excessive. Therefore,
uncontrolled iron depletion of normal cells may generate
strong adverse effects. Despite the fact that tumoral cells
require more iron, an antiproliferative therapeutic strategy
based on iron deprivation of tumoral cells will benefit from a
specific vectorization of the iron chelators to these cells.
Polyamine metabolism (biosynthesis and uptake), detrimen-
tal to cell proliferation, is also particularly amplified in tumor
cells.20 The Polyamine Transport System (PTS), which can
deal with natural and synthetic polyamines, is increased in
tumor cells.21 This property is of potential interest to target
antitumor agents.22−24 Thus, various polyamine analogues have
been grafted onto cytotoxic compounds like nitroimidazole,25
chlorambucil,26 acridine,27 anthracene,28 camptothecine,29
naphthoquinone,30 azepine, and benzazepine.31 This concept
of tumor vectorization by polyamines has also been developed
to inhibit their own transporter,32,33 to vectorize cytotoxic
compounds such as dimethylsilane polyamine derivatives34 or
molecules carrying boron atoms for boronotherapy,35,36 and
finally to monitor the intracellular traffic of polyamines toward
the cells by using fluorescent polyamine analogues.28,37−39
More recently, F14512, a novel spermine epipodophyllotoxin
conjugate, and inhibitor of DNA-topoisomerase II,40 was
selected for clinical development in AML phase I.
line moiety (HQ) linked by a single carbon atom (HQ1) to a
linear polyamine chain of various types including the
spermidine (HQ1-XY) and spermine (HQ1-XYZ) series (X,
Y, and Z represent the alkyl chain length between the amino
groups). In a cellular model, we screened the antiproliferative
efficiency of these polyaminoquinolines (Quilamines) HQ1-
XYZ as well as their specificity for the polyamine transport
system. Finally, we evaluated the complexation properties of the
most biologically efficient Quilamine HQ1−44 toward iron-
(III) and competitive cations.
EXPERIMENTAL PROCEDURES
■
Chemistry. NMR spectra were recorded at room temper-
ature with a Bruker Avance 300 Ultra Shield or Bruker Avance
III 400 spectrometers. Chemical shifts are reported in parts per
million (ppm); coupling constant are reported in units of Hertz
[Hz]. Infrared (IR) spectra were recorded with a Bruker Vector
22 FTIR using KBr films or KBr pellets. Low-resolution mass
spectra (MS) were recorded with a Thermo electron DSQ
spectrometer. High-resolution mass spectra (HRMS) were
recorded with a Thermofisher hybrid LTQ-orbitrap spectrom-
eter (ESI+) and with a Bruker Autoflex III SmartBeam
spectrometer (MALDI). Microanalyses were performed on a
Thermo Scientific FLASH 2000 Series CHNS/O Analyzers.
Melting points were determined in open capillary tubes and are
uncorrected. All reagents were purchased from Acros Organics
or Aldrich and were used without further purification. Column
chromatographies were conducted on silica gel Kieselgel SI60
(40−63 μm) from Merck. Reactions requiring anhydrous
conditions were performed under argon. Dichloromethane was
distilled from calcium hydride under nitrogen prior to use.
Toluene and tetrahydrofuran were distilled from sodium/
benzophenone under argon prior to use.
In order to generate iron chelator molecules targeting tumor
cells, we planned to associate polyamine cargos with iron-
chelating units to promote the uptake of the chelating motif by
the overactive polyamine transport system of tumor cells.
For this purpose, we synthesized and evaluated a new
generation of chelator-polyamine hybrids (Quilamines) which
were designed to decrease the iron content in the iron
complexes in cancerous cells, thus reducing tumor cell
proliferation. The difference in PTS activation between healthy
cells and tumor cells enables tumor cells to be targeted, whereas
the strong dependence of these cells on iron ensures a
secondary targeting.
The compounds 1−5, 8−20, 24, 25, 28, and 29 already have
been described in the literature.
Compound (21). To solution of N1,N4-di-tert-butyloxycar-
bonylspermidine 3 (1 g, 2.9 mmol) in anhydrous acetonitrile
(30 mL) was added potassium carbonate (1.41 g, 10.15 mmol),
and the suspension was stirred at room temperature for 10 min
under Ar. 4-Bromobutyronitrile (290 μL, 2.9 mmol) was added,
and the resulting mixture was then stirred for 20 h at 50 °C.
The mixture was filtered, the solvent was removed under
reduced pressure, and the residue was purified by chromatog-
raphy on silica gel (CHCl3/MeOH/NH4OH, 94:5:1) to give
compound 21 as a colorless oil (850 mg, 71%).
Such a concept has been reported in previous studies on the
synthesis and the biological properties of a chelating
Deferiprone-type moiety linked to spermine.41 In addition,
other studies revealed that homospermidine−anthracene
conjugates were more selectively recognized by the PTS than
spermine derivatives.38 Instead, we chose the hydroxyquinoline
moiety, found in the iron chelator O-Trensox, which has a
higher affinity for iron than Deferiprone.42 The potential value
of this structure in an anticancer strategy is illustrated by the
example of Clioquinol (5-chloro-7-iodo-8-hydroxyquinoline).
This antibiotic is also able to chelate metals, which results in an
antiproliferative activity on tumor cells in culture and on grafted
tumors in animals.43,44 The structural requirement of the PTS
for polyamine analogues has previously been demonstrated,
showing its selectivity for a single CH2 spacer between the
heterocyclic part linked to a homospermidine moiety.39,45
In order to meet these objectives, we report here the
synthesis of the first generation of Quilamines: a hydroxyquino-
Compound (21). Colorless oil (850 mg, 71%). IR (KBr): ν
= 3349, 2974, 2932, 2866, 1693, 1515, 1478, 1454, 1418, 1390,
1
1365, 1273, 1250, 1168, 868, 774 cm−1. H NMR (400 MHz,
[D4]MeOD): δ = 1.39−1.64 (m, 22H, 2CH2 3CH3), 1.69 (q,
2H, CH2, J = 7 Hz), 1.82 (q, 2H, CH2, J = 7.24 Hz), 2.51 (t,
2H, CH2, J = 7.16 Hz), 2.61 (dd, 2H, CH2, J = 7.14 Hz, J = 7.44
Hz), 2.69 (dd, 2H, CH2, J = 7.14 Hz, J = 7.44 Hz), 3.03 (t, 2H,
CH2, J = 6.78 Hz), 3.17−3.26 (m, 4H, 2CH2) ppm. 13C NMR
(100 MHz, [D4]MeOD): δ = 15.3, 26.3, 26.9, 27.7, 28.7 (tBu),
28.8 (tBu), 30.2, 38.9, 45.6, 48.1, 49.1, 50.3, 79.9, 80.9, 120.9,
137.5, 158.4. HRMS (MALDI): calcd. for C21H40N4O4 [M
+H]+ 413.3122; found 413.3118.
General Procedure for the Preparation of Cyanopoly-
amines 22 and 23. To solution of di-tert-butyloxycarbonylho-
mospermidine 1 (0.2 g, 1.06 mmol) for 22 or di-tert-
butyloxycarbonylspermidine 3 (0.2 g, 1.06 mmol) for 23 in
methanol was added acrylonitrile (70.3 μL, 1.06 mmol) under
Ar at 0 °C and stirred for 6 h, and the reaction mixture was
1953
dx.doi.org/10.1021/bc300324c | Bioconjugate Chem. 2012, 23, 1952−1968

Bioconjugate Chemistry
Article
stirred at room temperature over a period of 12 h. The solvent
was removed under reduced pressure and the residue was
purified by chromatography on silica gel (chloroform/
methanol/ammonia, 94:5:1).
119.5, 157.1, 157.4, 158.5. HRMS (MALDI): calcd. for
C26H48N4O6 [M+Na]+ 535.3466; found 535.3479.
General Procedure for the Preparation of Homospermines
6 and 7. To a suspension of Raney Ni (2 g) in ethanol (10
mL) was added cyanopolyamines 26 or 27 (1 equiv, 1.4 mmol)
and slowly NH4OH (4 mL). The mixture was stirred under
hydrogen (20 bar) at room temperature for 24 h. The Raney
nickel was eliminated by filtration over a pad of Celite, and the
filtrate was evaporated under reduce pressure. The resulting
solid was dissolved in dichloromethane, washed with NaOH 2.5
M (2 times) and water, dried over anhydrous Na2SO4, filtered,
and concentrated under reduced pressure to yield homo-
spermine 6 or 7 as a colorless oil.
Compound (6). Colorless oil (99%). IR (KBr): ν = 3363,
2973, 2930, 2865, 1689, 1520, 1479, 1418, 1390, 1365, 1275,
1250, 1169, 875, 773 cm−1. 1H NMR (400 MHz, [D4]MeOD):
δ = 1.43 (s, 9H, 3CH3), 1.45 (s, 9H, 3CH3), 1.46 (s, 9H,
3CH3), 1.48−1.63 (m, 8H, 4CH2), 1.68 (q, 2H, CH2, J = 6.8
Hz), 2.64 (dd, 2H, CH2, J = 7.1 Hz), 3.03 (t, 2H, CH2, J = 6.8
Hz), 3.16−3.27 (m, 8H, 4CH2) ppm. 13C NMR (100 MHz,
[D4]MeOD): δ = 26.7, 27.1, 28.8 (9C), 29.6, 30.2, 31.1, 38.9,
42.3, 48.1 (4C), 79.9, 80.8, 80.9, 157.4 (2C), 158.4. HRMS
(MALDI): calcd. for C26H52N4O6 [M+Na]+ 539.3779; found
539.3760.
Compound (7). Colorless oil (99%). IR (KBr): ν = 3362,
2974, 2931, 2866, 1683, 1520, 1479, 1455, 1418, 1390, 1365,
1251, 1170, 875, 773 cm−1. 1H NMR (400 MHz, [D4]MeOD):
δ = 1.43 (s, 9H, 3CH3), 1.45 (s, 9H, 3CH3), 1.46 (s, 9H,
3CH3), 1.48−1.61 (m, 8H, 4CH2), 1.68 (q, 2H, CH2, J = 6.9
Hz), 2.62 (t, 2H, CH2, J = 6.9 Hz), 3.04 (t, 2H, CH2, J = 6.8
Hz), 3.15−3.29 (m, 8H, 4CH2) ppm. 13C NMR (100 MHz,
[D4]MeOD): δ = 26.5 (2C), 27.1 (2C), 28.3 (3C), 28.8 (6C),
32.1, 39.6, and 40.0 (rotamer), 41.0, 45.1, and 45.8 (rotamer),
47.9 (3C), 79.8, 80.8, 80.9, 157.4 (2C), 158.5. HRMS
(MALDI): calcd. for C26H52N4O6 [M+Na]+ 539.3779; found
539.3781.
General Procedure for the Reductive Amination (31−39).
To a stirred solution of polyamines 1−9 (1.2 equiv, 0.72 mmol)
in 1,2-dichloroethane (6 mL) at room temperature under argon
was added 8-hydroxyquinoline-2-carbaldehyde 30 (1 equiv, 0.6
mmol). After 15 min, sodium triacetoxyborohydride (3 equiv,
1.8 mmol) was added and the mixture was stirred 2 to 4 h until
aldehyde was consumed (monitored by TLC). The reaction
mixture was quenched by adding 1 M NaOH, stirred 20 min,
and the product was extracted three times with CH2Cl2, dried
over anhydrous Na2SO4, filtered, and concentrated under
reduced pressure. The residue was purified by chromatography
on silica gel (chloroform/methanol/ammonia, 94:5:1 for 31−
34; and 95:4:1 for 35−39).
Compound (22). Colorless oil (730 mg, 91%). IR (KBr): ν
= 3347, 2974, 2932, 2863, 1688, 1521, 1478, 1454, 1418, 1390,
1365, 1251, 1171 cm−1. 1H NMR (400 MHz, [D4]MeOD): δ =
1.43 (s, 9H, tBu), 1.46 (s, 9H, tBu), 1.46−1.63 (m, 8H, 4CH2),
2.60 (t, 2H, CH2, J = 6.85 Hz), 2.63 (dd, 2H, CH2, J = 7.1 Hz),
2.87 (dd, 2H, CH2, J = 6.76 Hz), 3.05 (t, 2H, CH2, J = 6.8 Hz),
3.17−3.26 (m, 4H, 2CH2) ppm. 13C NMR (100 MHz,
[D4]MeOD): δ = 18.3, 26.5, 26.9, 27.8, 28.3, 28.7 (6C),
41.0, 45.9, 47.9 (2C), 49.7, 79.8, 80.8, 119.9, 157.4, 158.5.
HRMS (MALDI): calcd. for C21H40N4O4 [M+Na]+ 435.2942;
found 435.2942.
Compound (23). Colorless oil (83%). IR (KBr): ν = 3582,
3341, 2976, 2932, 2863, 1683, 1520, 1479, 1418, 1390, 1365,
1
1251, 1164, 1081, 1040, 993, 869, 775 cm−1. H NMR (400
MHz, [D4]MeOD): δ = 1.43 (s, 9H, 3CH3), 1.46 (s, 9H,
3CH3), 1.47−1.63 (m, 4H, 2CH2), 1.69 (q, 2H, CH2, J = 7 Hz),
2.60 (t, 2H, CH2, J = 6.8 Hz), 2.63 (dd, 2H, CH2, J = 7.1 Hz, J
= 7.3 Hz), 2.86 (t, 2H, CH2, J = 6.9 Hz), 3.03 (t, 2H, CH2, J =
6.8 Hz), 3.17−3.27 (m, 4H, 2CH2) ppm. 13C NMR (100 MHz,
[D4]MeOD): δ = 18.3, 26.9, 27.8, 28.3, 28.7 (3C), 28.8 (3C),
41.0, 45.9, 47.9 (2C), 49.7, 79.9, 80.9, 119.9, 157.4, 158.4.
HRMS (MALDI): calcd. for C20H38N4O4 [M+Na]+ 421.2785;
found 421.2774.
General Procedure for the Preparation of Cyanopoly-
amines 26 and 27. A solution of di-tert-butyldicarbonate (2.5
equiv, 3.3 mmol) in anhydrous methanol was added dropwise
to a stirring solution of 21 (1 equiv, 1.32 mmol) or 22 (1 equiv,
1.32 mmol) in solution of methanol and 10% of triethylamine
(10 mL). After stirring for 24 h at room temperature, solvents
were removed under reduced pressure to give a colorless oil,
which was dissolved in dichloromethane, washed with saturated
NaHCO3 and then water. The organic layer was separated,
dried over anhydrous Na2SO4, filtered, and concentrated under
reduced pressure. The residue was purified by chromatography
on silica gel (dichloromethane then dichloromethane/ethyl
acetate: (8:2)).
Compound (26). Colorless oil (670 mg, 98%). IR (KBr): ν
= 3360, 2975, 2932, 2868, 1698, 1519, 1479, 1455, 1417, 1390,
1
1366, 1249, 1163, 1077, 1040, 872, 773 cm−1. H NMR (400
MHz, [D4]MeOD): δ = 1.43 (s, 9H, 3CH3), 1.46 (s, 9H,
3CH3), 1.47 (s, 9H, 3CH3), 1.49−1.58 (m, 4H, 2CH2), 1.69 (q,
2H, CH2, J = 6.9 Hz), 1.87 (q, 2H, CH2, J = 7 Hz), 2.44 (t, 2H,
CH2, J = 7 Hz), 3.03 (t, 2H, CH2, J = 6.8 Hz), 3.17−3.27 (m,
6H, 3CH2), 3.32 (t, 2H, CH2, J = 7.1 Hz) ppm. 13C NMR (100
MHz, [D4]MeOD): δ = 14.9, 25.6, 26.6, 27.0, 28.75 (3C),
28.78 (3C), 28.8 (3C), 29.5, 30.3, 38.9, 45.9, 47.1, 47.8 (2C),
79.9, 80.9, 81.3, 120.8, 157.4, 158.4. HRMS (MALDI): calcd.
for C26H48N4O6 [M+Na]+ 535.3466; found 535.3463.
Compound (27). Colorless oil (395 mg, 99%). IR (KBr): ν
= 3360, 2975, 2932, 2867, 1682, 1519, 1479, 1417, 1391, 1366,
1251, 1163, 1074, 871, 774 cm−1. 1H NMR (400 MHz,
[D4]MeOD): δ = 1.43 (s, 9H, 3CH3), 1.46 (s, 9H, 3CH3), 1.48
(s, 9H, 3CH3), 1.50−1.61 (m, 8H, 4CH2), 2.69 (t, 2H, CH2, J =
6.6 Hz), 3.05 (t, 2H, CH2, J = 6.8 Hz), 3.16−3.26 (m, 4H,
2CH2), 3.29 (t, 2H, CH2, J = 6.6 Hz), 3.50 (t, 2H, CH2, J = 6.6
Hz) ppm. 13C NMR (100 MHz, [D4]MeOD): δ = 17.3 and
17.8 (rotamer), 26.8 (4C), 28.3 (3C), 28.6 (3C), 28.8 (3C),
41.0, 44.3, and 44.6 (rotamer), 48.1 (3C), 79.8, 80.8, 81.7,
Compound (31). Viscous yellow oil (705 mg, 91%). IR
(KBr): ν = 3053, 2982, 2934, 1706, 1683, 1575, 1507, 1475,
1455, 1420, 1391, 1366, 1265, 1168, 1047 cm−1. 1H NMR (400
MHz, [D4]MeOD): δ = 1.35−1.69 (m, 26H, 8CH2 6CH3),
2.77 (bs, 2H, CH2), 3.03 (t, 2H, CH2, J = 6.8 Hz), 3.13−3.26
(m, 4H, 2CH2), 4.14 (s, 2H, CH2), 7.10 (dd, H, J = 7.5 Hz, J =
1.3 Hz), 7.34 (dd, H, J = 8.3 Hz, J = 1.3 Hz), 7.41 (dd, H, J =
8.1 Hz, J = 7.6 Hz), 7.45 (d, H, J = 8.5 Hz), 8.20 (d, H, J = 8.5
Hz) ppm. 13C NMR (100 MHz, [D4]MeOD): δ = 26.5, 27.0,
27.2, 28.4, 28.7 (3C), 28.8 (3C), 40.9, 48.1, 49.9, 54.9, 79.8,
80.8, 112.2, 118.8, 122.1, 128.3, 129.4, 137.9, 139.2, 154.2,
157.4 (2C), 158.5. HRMS (MALDI): calcd. for C28H44N4O5
[M+H]+ 517.3384; found 517.3386.
1954
dx.doi.org/10.1021/bc300324c | Bioconjugate Chem. 2012, 23, 1952−1968

Bioconjugate Chemistry
Article
Compound (32). Viscous yellow oil (300 mg, 82%). IR
(KBr): ν = 3341, 2974, 2933, 2824, 1689, 1595, 1573, 1508,
1476, 1420, 1391, 1366, 1274, 1230, 1169, 1088, 1034 cm−1. 1H
NMR (400 MHz, [D4]MeOD): δ = 1.36−1.48 (m, 18H,
9CH3), 1.54−1.61 (m, 4H, 2CH2), 1.66 (qt, 2H, CH2, J = 6.8
Hz), 2.71 (t, 2H, CH2, J = 6.8 Hz), 3.02 (t, 2H, CH2, J = 6.8
Hz), 3.14−3.26 (m, 4H, 2CH2), 4.07 (s, 2H, CH2), 7.09 (dd,
H, J = 7.5 Hz, J = 1.3 Hz), 7.32 (dd, H, J = 8.3 Hz, J = 1.3 Hz),
7.38 (dd, H, J = 7.6 Hz, J = 7.5 Hz), 7.42 (d, H, J = 8.5 Hz),
8.17 (d, H, J = 8.5 Hz) ppm. 13C NMR (100 MHz,
[D4]MeOD): δ = 27.5 (2C), 28.7 (3C), 28.8 (3C), 29.4 and
30.2 (rotamer), 38.9, 45.6, and 46.1 (rotamer), 48.1, 50.1, 55.3,
79.9, 80.9, 112.1, 118.8, 122.1, 128.2, 129.4, 137.8, 139.3, 154.2,
157.4, 158.2, 158.4. HRMS (MALDI): calcd. for C27H42N4O5
[M+H]+ 503.3228; found 503.3207.
Hz), 3.13−3.25 (m, 6H, 3CH2), 3.31 (m, 2H, CH2), 4.08 (s,
2H, CH2), 7.09 (dd, H, J = 7.4 Hz, J = 1.35 Hz), 7.34 (dd, H, J
= 8.2 Hz, J = 1.35 Hz), 7.40 (dd, H, J = 8.2 Hz, J = 7.4 Hz),
7.45 (d, H, J = 8.5 Hz), 8.19 (d, H, J = 8.5 Hz) ppm. 13C NMR
(100 MHz, [D4]MeOD): δ = 26.6 (2C), 27.0 (2C), 28.3, 28.7
(3C), 28.8 (6C), 41.0, 47.9 (4C), 49.8, 55.1, 79.8, 80.8, 80.9,
112.0, 118.8, 122.1, 128.3, 129.4, 137.8, 139.3, 154.2, 157.4
(2C), 157.5, 158.5. HRMS (MALDI): calcd. for C36H59N5O7
[M+H]+ 674.4487; found 674.4492.
Compound (37). Viscous yellow oil (485 mg, 96%). IR
(KBr): ν = 3357, 3052, 2975, 2931, 1683, 1600, 1573, 1505,
1471, 1418, 1390, 1366, 1248, 1164, 1086, 878, 835, 754, 703
1
cm−1. H NMR (400 MHz, [D4]MeOD): δ = 1.35−1.53 (m,
33H, 3CH2 9CH3), 1.53−1.62 (m, 4H, 2CH2), 2.70 (bt, 2H,
CH2), 3.02 (t, 2H, CH2, J = 6.7 Hz), 3.21−3.26 (m, 8H, 4CH2),
4.07 (s, 2H, CH2), 7.08 (dd, H, J = 7.5 Hz, J = 1.3 Hz), 7.32
(dd, H, J = 8.2 Hz, J = 1.3 Hz), 7.39 (dd, H, J = 8.2 Hz, J = 7.5
Hz), 7.43 (d, H, J = 8.5 Hz), 8.17 (d, H, J = 8.5 Hz) ppm. 13C
NMR (100 MHz, [D4] MeOD): δ = 26.6, 27.1, 27.7 (2C), 28.8
(9C), 29.5 and 30.2 (rotamer), 38.9, 45.7, 48.1 (3C), 50.1, 55.3,
79.9, 80.8, 80.9, 112.1, 118.7, 122.1, 128.2, 129.3, 137.8, 139.3,
154.2, 157.3 (2C), 158.3 (2C). HRMS (MALDI): calcd. for
C36H59N5O7 [M+H]+ 674.4487; found 674.4475.
Compound (33). Viscous yellow oil (280 mg, 90%). IR
(KBr): ν = 3350.4, 2975, 2935, 2826, 1688, 1575−1250, 1171,
1
1033 cm−1. H NMR (400 MHz, [D4]MeOD): δ = 1.32−1.61
(m, 22H, 2CH2 6CH3), 1.83 (qt, 2H, CH2, J = 7.1 Hz), 2.72
(bt, 2H, CH2), 3.03 (t, 2H, CH2, J = 6.8 Hz), 3.17 (t, 2H, CH2,
J = 7.2 Hz), 3.22−3.38 (m, 2H, CH2), 4.09 (s, 2H, CH2), 7.10
(dd, H, J = 7.5 Hz, J = 1.3 Hz), 7.33 (dd, H, J = 8.3 Hz, J = 1.3
Hz), 7.39 (dd, H, J = 8.3 Hz, J = 7.6 Hz), 7.43 (d, H, J = 8.5
Hz), 8.18 (d, H, J = 8.5 Hz) ppm. 13C NMR (100 MHz,
[D4]MeOD): δ = 26.9, 28.3, 28.7 (3C), 28.8 (3C), 29.6, 40.9,
45.3, 46.4, and 46.9 (rotamer), 47.8, 54.8, 79.8, 80.9, 112.1,
118.8, 122.1, 128.3, 129.4, 137.9, 139.2, 154.2, 157.5, 158.5
(2C). HRMS (MALDI): calcd. for C27H42N4O5 [M+H]+
503.3228; found 503.3215.
Compound (38). Viscous yellow oil (290 mg, 96%). IR
(KBr): ν = 3364, 3052, 2974, 2930, 1683, 1600, 1574, 1505,
1
1471, 1418, 1390, 1366, 1248, 1161, 1086, 871, 835 cm−1. H
NMR (400 MHz, [D4]MeOD): δ = 1.34−1.56 (m, 31H, 2CH2
9CH3), 1.68 (qt, 2H, CH2, J = 6.8 Hz), 1.83 (qt, 2H, CH2, J =
7.1 Hz), 2.70 (t, 2H, CH2, J = 6.8 Hz), 3.03 (t, 2H, CH2, J = 6.8
Hz), 3.14−3.25 (m, 6H, 3CH2), 3.31 (m, 2H, CH2), 4.08 (s,
2H, CH2), 7.09 (dd, H, J = 7.5 Hz, J = 1.3 Hz), 7.34 (dd, H, J =
8.3 Hz, J = 1.3 Hz), 7.40 (dd, H, J = 8.3 Hz, J = 7.5 Hz), 7.45
(d, H, J = 8.5 Hz), 8.19 (d, H, J = 8.5 Hz) ppm. 13C NMR (100
MHz, [D4]MeOD): δ = 27.1 (2C), 28.6 (3C), 28.7 (3C), 28.8
(3C), 29.7 (2C), 38.9 and 45.4 (rotamer), 45.8, 47.1, 47.9
(2C), 49.8, 55.1, 79.9, 80.9, 81.0, 112.1, 118.8, 122.1, 128.3,
129.4, 137.9, 139.2, 154.2, 157.4 (2C), 158.1, 158.4. HRMS
(MALDI): calcd. for C35H57N5O7 [M+Na]+ 682.4150; found
682.4164.
Compound (34). Viscous yellow oil (325 mg, 88%). IR
(KBr): ν = 3357, 2976, 2935, 2827, 1689, 1600, 1574, 1509,
1
1477, 1422, 1392, 1367, 1251, 1170, 1089, 1033 cm−1. H
NMR (400 MHz, [D4]MeOD): δ = 1.42 (s, 18H, 6CH3), 1.68
(qt, 2H, CH2, J = 6.3 Hz), 1.83 (qt, 2H, CH2, J = 7.1 Hz), 2.69
(t, 2H, CH2, J = 7 Hz), 3.02 (t, 2H, CH2, J = 6.8 Hz), 3.21 (t,
2H, CH2, J = 7.2 Hz), 3.25−3.38 (m, 2H, CH2), 4.08 (s, 2H,
CH2), 7.09 (dd, H, J = 7.5 Hz, J = 1.3 Hz), 7.33 (dd, H, J = 8.2
Hz, J = 1.2 Hz), 7.39 (dd, H, J = 8.1 Hz, J = 7.6 Hz), 7.44 (d, H,
J = 8.5 Hz), 8.18 (d, H, J = 8.5 Hz) ppm. 13C NMR (100 MHz,
[D4]MeOD): δ = 28.7 (3C), 28.8 (3C), 29.5, 30.1, 38.9, 45.8,
46.5, 49.8, 55.1, 79.9, 81.1, 112.1, 118.8, 122.1, 128.3, 129.4,
137.8, 139.2, 154.2, 157.5, 158.4 (2C). HRMS (MALDI): calcd.
for C26H40N4O5 [M+H]+ 489.3071; found 489.3048.
Compound (39). Viscous yellow oil (240 mg, 89%). IR
(KBr): ν = 3363, 3051, 2973, 2929, 1682, 1600, 1573, 1505,
1
1471, 1418, 1365, 1250, 1163, 1091, 982, 866, 835 cm−1. H
NMR (400 MHz, [D4]MeOD): δ = 1.42 (bs, 18H, 6CH3), 1.44
(s, 9H, 3CH3), 1.68 (qt, 2H, CH2, J = 6.95 Hz), 1.75 (qt, 2H,
CH2, J = 7.1 Hz), 1.83 (qt, 2H, CH2, J = 7.1 Hz), 2.69 (t, 2H,
CH2, J = 7.1 Hz), 3.03 (t, 2H, CH2, J = 6.8 Hz), 3.13−3.25 (m,
6H, 3CH2), 3.31 (m, 2H, CH2), 4.07 (s, 2H, CH2), 7.09 (dd, H,
J = 7.5 Hz, J = 1.3 Hz), 7.33 (dd, H, J = 8.3 Hz, J = 1.3 Hz),
7.40 (dd, H, J = 8.3 Hz, J = 7.5 Hz), 7.44 (d, H, J = 8.5 Hz),
8.18 (d, H, J = 8.5 Hz) ppm. 13C NMR (100 MHz,
[D4]MeOD): δ = 28.7 (6C), 28.8 (3C), 29.5 (3C), 38.9,
45.9 (3C), 47.1, 49.8, 55.1, 79.4, 79.9, 81.1, 112.0, 118.8, 122.1,
128.3, 129.3, 137.8, 139.2, 154.2, 157.3 (2C), 158.1, 158.40.
HRMS (MALDI): calcd. for C34H55N5O7 [M+H]+ 646.4174;
found 646.4158.
General Procedure for Boc Removal (HQ1-XYZ). To a
stirred solution of polyamines 31−39 (1 equiv, 0.5 mmol)
dissolved in ethanol (3 mL) at 0 °C was added 6 M HCl (3
mL). After 18 h at room temperature, the solution was
concentrated in vacuo, and the resulting solid was washed with
cold ethanol, filtered, and dried in vacuo to afford Quilamine
40−48 as yellow solid, its hydrochloride salt.
Compound (35). Viscous yellow oil (200 mg, 95%). IR
(KBr): ν = 3351, 2974, 2932, 2865, 1699, 1601, 1573, 1506,
1
1472, 1418, 1366, 1250, 1173, 879, 836, 754 cm−1. H NMR
(400 MHz, [D4]MeOD): δ = 1.39−1.65 (m, 35H, 4CH2
9CH3), 2.72 (bt, 2H, CH2), 3.03 (t, 2H, CH2, J = 6.7 Hz),
3.13−3.24 (m, 8H, 4CH2), 4.09 (s, 2H, CH2), 7.09 (dd, H, J =
7.5 Hz, J = 1.3 Hz), 7.33 (dd, H, J = 8.2 Hz, J = 1.3 Hz), 7.39
(dd, H, J = 8.2 Hz, J = 7.5 Hz), 7.44 (d, H, J = 8.5 Hz), 8.19 (d,
H, J = 8.5 Hz) ppm. 13C NMR (100 MHz, [D4]MeOD): δ =
26.6 (2C), 27.1 (2C), 27.6, 28.3, 28.8 (9C), 41.0, 48.0 (4C),
50.1, 55.3, 79.8, 80.8 (2C), 112.1, 118.8, 122.2, 128.3, 129.4,
137.8, 139.3, 154.2, 157.4 (2C), 158.2, 158.5. HRMS
(MALDI): calcd. for C37H61N5O7 [M+H]+ 688.4644; found
688.4672.
Compound (36). Viscous yellow oil (310 mg, 94%). IR
(KBr): ν = 3360, 3051, 2972, 2929, 2864, 1695, 1616, 1558,
1520, 1506, 1495, 1472, 1464, 1456, 1363, 1248, 1164, 1090,
1044, 1005, 877, 835 cm−1. 1H NMR (400 MHz, [D4]MeOD):
δ = 1.34−1.61 (m, 35H, 4CH2 9CH3), 1.83 (qt, 2H, CH2, J =
7.1 Hz), 2.70 (t, 2H, CH2, J = 6.9 Hz), 3.04 (t, 2H, CH2, J = 6.8
1955
dx.doi.org/10.1021/bc300324c | Bioconjugate Chem. 2012, 23, 1952−1968

Bioconjugate Chemistry
Article
Compound HQ1-44. Yellow solid (73%). Mp sup 250 °C.
IR (KBr): ν = 3384, 2957, 2798, 1642, 1606, 1551, 1513, 1461,
1420, 1398, 1344, 1302, 1252−1055, 841 cm−1. UV−vis: λmax
(nm) ((ε M−1 cm−1 for pH 7.4), 272 (36300), 307 (2400). 1H
NMR (400 MHz, D2O): δ = 1.76−2.04 (m, 8H, 4CH2), 3.06−
3.22 (m, 6H, 3CH2), 3.38 (dd, 2H, J = 7.75 Hz, J = 7.45 Hz,
CH2), 4.78 (s, 2H, CH2), 7.36 (dd, 1H, J = 7.45 Hz, J = 1.49
Hz, H7ar), 7.60 (dd, 1H, J = 8.30 Hz, J = 1.49 Hz, H5ar), 7.65
(dd, 1H, J = 8.30 Hz, J = 7.45 Hz, H6ar), 7.76 (d, 1H, J = 8.60
Hz, H3ar), 8.64 (d, 1H, J = 8.60 Hz, H4ar) ppm. 13C NMR (100
MHz, D2O): δ = 22.7, 22.8 (2C), 24.0, 38.9, 46.9, 47.0, 47.3,
49.7, 114.1, 119.4, 121.1, 128.9, 129.3, 134.5, 141.8, 148.5,
149.5 ppm. HRMS (MALDI): calcd. for C18H29N4O [M+H-
4HCl]+ 317.2336; found 317.2348. Anal. Calcd for
C18H32Cl4N4O·1.2 H2O: C, 44.68; H, 7.17; N, 11.58. Found:
C, 44.52; H, 6.97; N, 11.57.
Compound HQ1-444. Yellow solid (95%). Mp sup 250 °C.
IR (KBr): ν = 3426, 3038, 2956, 2874, 2799, 2758, 1642, 1606,
1598, 1449, 1302 cm−1. UV−vis: λmax (nm) ((ε M−1 cm−1 for
pH 7.4) 272 (34500), 307 (2000). 1H NMR (400 MHz, D2O):
δ = 1.75−2.03 (m, 12H, 6CH2), 3.06−3.22 (m, 10H, 5CH2),
3.34−3.41 (m, 2H, CH2), 4.78 (s, 2H, CH2), 7.39 (dd, 1H, J =
7.15 Hz, J = 1.79 Hz, H7ar), 7.61−7.71 (m, 2H, H5ar and H6ar),
7.77 (d, 1H, J = 8.54 Hz, H3ar), 8.65 (d, 1H, J = 8.54 Hz, H4ar)
ppm. 13C NMR (100 MHz, D2O): δ = 22.7, 22.8 (4C), 24.0,
38.9, 46.9 (3C), 47.0, 47.3, 49.8, 114.0, 119.4, 121.1, 128.9,
129.2, 134.8, 141.5, 148.7, 149.7 ppm. HRMS (MALDI): calcd.
for C22H38N5O [M+H-5HCl]+ 388.3071; found 388.3059.
Anal. Calcd for C22H42Cl5N5O 1.7 H2O: C, 44; H, 7.62; N,
11.66. Found: C, 43.64; H, 7.45; N, 12.05.
Compound HQ1-443. Yellow solid (91%). Mp sup 250 °C.
IR (KBr): ν = 3441, 2951, 2777, 2577, 2421, 1641, 1597, 1515,
1477, 1451, 1418, 1401, 1352, 1301, 1247, 1232, 1209, 1168,
1092, 1056, 841 cm−1. UV−vis: λmax (nm) ((ε M−1 cm−1 for
pH 7.4) 272 (35300), 307 (2000). 1H NMR (400 MHz, D2O):
δ = 1.81−2.02 (m, 8H, 4CH2), 2.12−2.21 (m, 2H, CH2), 3.12−
3.27 (m, 10H, 5CH2), 3.35−3.39 (m, 2H, CH2), 4.77 (s, 2H,
CH2), 7.39 (dd, 1H, J = 6.9 Hz, J = 2 Hz, H7ar), 7.63−7.71 (m,
2H, H5ar and H6ar), 7.75 (d, 1H, J = 8.6 Hz, H3ar), 8.63 (d, 1H, J
= 8.6 Hz, H4ar) ppm. 13C NMR (100 MHz, D2O): δ = 22.8
(2C), 22.9 (2C), 23.8, 36.6, 44.6, 46.9 (2C), 47.1, 47.2, 50.1,
113.7, 119.4, 121.1, 128.9, 129.1, 135.2, 141.0, 148.9, 149.9
ppm. HRMS (MALDI): calcd. for C21H36N5O [M+H-5HCl]+
374.2914; found 374.2927. Anal. Calcd for C21H40Cl5N5O·1.35
H2O: C, 43.48; H, 7.42; N, 12.07. Found: C, 43.32; H, 7.25; N,
12.22.
Compound HQ1-34. Yellow solid (80%). Mp sup 250 °C.
IR (KBr): ν = 3493, 2957, 2770, 2528, 2403, 1599, 1515, 1480,
1458, 1420, 1398, 1304, 859 cm−1. UV−vis: λmax (nm) ((ε M−1
1
cm−1 for pH 7.4) 272 (32500), 307 (2100). H NMR (400
MHz, D2O): δ = 1.66−1.84 (m, 4H, 2CH2), 2.17−2.31 (m, 2H,
CH2), 2.97−3.06 (m, 2H, CH2), 3.07−3.14 (m, 2H, CH2),
3.15−3.24 (m, 2H, CH2), 3.41 (m, 2H, CH2), 4.68 (s, 2H,
CH2), 7.27 (dd, 1H, J = 6.61 Hz, J = 2.43 Hz, H7ar), 7.49−7.57
(m, 2H, H5ar and H6ar), 7.61 (d, 1H, J = 8.50 Hz, H3ar), 8.46 (d,
1H, J = 8.50 Hz, H4ar) ppm. 13C NMR (100 MHz, D2O): δ =
22.7, 22.8, 23.9, 38.8, 44.5, 44.7, 47.1, 50.5, 113.2, 119.4, 120.8,
128.6, 128.7, 136.1, 139.9, 149.2, 105.4 ppm. HRMS (MALDI):
calcd. for C17H27N4O [M+H-4HCl]+ 303.2179; found
303.2183. Anal. Calcd for C17H30Cl4N4O·0.4H2O: C, 41.89;
H, 6.37; N, 11.47. Found: C, 42.15; H, 6.36; N, 11.16.
Compound HQ1-43. Yellow solid (88%). Mp sup 250 °C.
IR (KBr): ν = 3039, 2984, 2825, 1641, 1606, 1538, 1461, 1419,
1402, 1352, 1302, 842 cm−1. UV−vis: λma1x (nm) ((ε M−1 cm−1
for pH 7.4) 272 (34500), 307 (2200). H NMR (400 MHz,
D2O): δ = 1.84−2.05 (m, 4H, 2CH2), 2.11−2.22 (m, 2H,
CH2), 3.14−3.27 (m, 6H, 3CH2), 3.34−3.41 (m, 2H, CH2),
4.75 (s, 2H, CH2), 7.39 (dd, 1H, J = 6.5 Hz, J = 2.4 Hz, H7ar),
7.62−7.70 (m, 2H, H5ar and H6ar), 7.71 (d, 1H, J = 8.5 Hz,
H3ar), 8.59 (d, 1H, J = 8.5 Hz, H4ar) ppm. 13C NMR (100 MHz,
D2O): δ = 22.8, 22.9, 23.8, 36.6, 44.6, 47.0, 47.1, 50.3, 113.3,
119.4, 120.9, 128.8 (2C), 135.8, 140.2, 149.3, 150.3 ppm.
HRMS (MALDI): calcd. for C17H27N4O [M+H-4HCl]+
303.2179; found 303.2190. Anal. Calcd for C17H30Cl4N4O·1.1
H2O: C, 43.62; H, 6.93; N, 11.97. Found: C, 43.4; H, 6.71; N,
12.19.
Compound HQ1-344. Yellow solid (92%). Mp sup 250 °C.
IR (KBr): ν = 3416, 2954, 2770, 2441, 1631, 1602, 1512, 1456,
1412, 1351, 1312, 1249, 1168, 1095, 1056 cm−1. UV−vis: λmax
1
(nm) ((ε M−1 cm−1 for pH 7.4) 272 (36900), 307 (2200). H
NMR (400 MHz, D2O): δ = 1.77−1.92 (m, 8H, 4CH2), 2.28−
2.39 (m, 2H, CH2), 3.07−3.23 (m, 8H, 4CH2), 3.25−3.32 (m,
2H, CH2), 3.41−3.47 (m, 2H, CH2), 4.74 (s, 2H, CH2), 7.35
(dd, 1H, J = 6.6 Hz, J = 2.3 Hz H3ar), 7.60−7.66 (m, 3H,), 8.49
(d, 1H, J = 8.6 Hz) ppm. 13C NMR (100 MHz, D2O): δ = 22.7,
22.8 (3C), 23.9, 38.8, 44.5, 44.6, 46.8, 46.9, 47.1, 50.8, 112.7,
119.4, 120.6, 128.4, 128.6, 136.6, 139.1, 149.5, 150.7 ppm.
HRMS (MALDI): calcd. for C21H36N5O [M+H-5HCl]+
374.2914; found 374.2905. Anal. Calcd for C21H40Cl5N5O·0.3
H2O: C, 44.94; H, 7.29; N, 12.48. Found: C, 44.7; H, 7.15; N,
12.72.
Compound HQ1-343. Yellow solid (92%). Mp sup 250 °C.
IR (KBr): ν = 3386, 2957, 2776, 2425, 1642, 1606, 1312, 1461,
1413, 1348, 1303, 1252, 1111, 1093, 1058 cm−1. UV−vis: λmax
Compound HQ1-33. Yellow solid (76%). Mp sup 250 °C.
IR (KBr): ν = 3423, 2962, 2760, 2692, 1636, 1607, 1595, 1517,
1458, 1397, 1303, 1272, 1094, 842 cm−1. UV−vis: λmax (nm)
1
(nm) ((ε M−1 cm−1 for pH 7.4) 272 (34500), 307 (2100). H
NMR (400 MHz, D2O): δ = 1.81−1.91 (m, 4H, 2CH2), 2.12−
2.22 (m, 2H, CH2), 2.28−2.38 (m, 2H, CH2), 3.13−3.32 (m,
10H, 5CH2), 3.41−3.48 (m, 2H, CH2), 4.79 (s, 2H, CH2), 7.38
(dd, 1H, J = 7 Hz, J = 1.93 Hz), 7.61−7.69 (m, 2H, CH2), 7.74
(d, 1H, J = 8.6 Hz, H3ar), 8.61 (d, 1H, J = 8.6 Hz, H4ar) ppm.
13C NMR (100 MHz, D2O): δ = 22.7, 22.8 (2C), 23.8, 36.6,
44.5, 44.6 (2C), 44.8, 47.1, 50.2, 113.7, 119.4, 121.0, 128.9,
129.1, 135.3, 140.5, 148.8, 150.0 ppm. HRMS (MALDI): calcd.
for C20H34N5O [M+H-5HCl]+ 360.2758; found 360.2764.
Anal. Calcd for C20H38Cl5N5O·0.7 H2O: C, 43.33; H, 7.16; N,
12.63. Found: C, 43.05; H, 7.14; N, 12.97.
1
((ε M−1cm−1 for pH 7.4) 272 (36200), 307 (2200). H NMR
(400 MHz, D2O): δ = 2.12−2.22 (m, 2H, CH2), 2.28−2.39 (m,
2H, CH2), 3.14−3.21 (m, 2H, CH2), 3.22−3.34 (m, 4H,
2CH2), 3.42−3.49 (m, 2H, CH2), 4.80 (s, 2H, CH2), 7.37 (dd,
1H, J = 7.28 Hz, J = 1.46 Hz, H7ar), 7.61 (dd, 1H, J = 8.37 Hz, J
= 1.64 Hz, H5ar), 7.66 (dd, 1H, J = 7.46 Hz, J = 7.28 Hz, H6ar),
7.75 (d, 1H, J = 8.55 Hz, H3ar), 8.63 (d, 1H, J = 8.55 Hz, H4ar)
ppm. 13C NMR (100 MHz, D2O): δ = 22.7, 23.7, 36.6, 44.7,
44.8, 44.9, 49.9, 114.0, 119.4, 121.1, 128.9, 129.2, 134.7, 141.6,
148.5, 149.7 ppm. HRMS (MALDI): calcd. for C16H25N4O [M
+H-4HCl]+ 289.2023; found 289.2034. Anal. Calcd for
C16H28Cl4N4O·1.45 H2O: C, 41.74; H, 6.77; N, 12.17.
Found: C, 41.59; H, 6.58; N, 12.17.
Compound HQ1-333. Yellow solid (96%). Mp sup 250 °C.
IR (KBr): ν = 3424, 2963, 2756, 2696, 2585, 2420, 1636, 1607,
1594, 1550, 1519, 1458, 1402, 1303, 1265, 1094 cm−1. UV−vis:
1956
dx.doi.org/10.1021/bc300324c | Bioconjugate Chem. 2012, 23, 1952−1968

Bioconjugate Chemistry
Article
λmax (nm) ((ε M−1 cm−1 for pH 7.4) 272 (35400), 307 (2100).
1H NMR (400 MHz, D2O): δ = 2.12−2.26 (m, 4H, 2CH2),
2.29−2.39 (m, 2H, CH2), 3.14−3.36 (m, 10H, 2CH2NHCH2
and CH2NH2), 3.42−3.49 (m, 2H, NH CH2), 4.79 (s, 2H,
CH2), 7.38 (dd, 1H, J = 6.8 Hz, J = 2 Hz, H7ar), 7.61−7.69 (m,
2H, CH2, H5ar and H6ar), 7.73 (d, 1H, J = 8.6 Hz, H3ar), 8.61 (d,
1H, J = 8.6 Hz, H4ar) ppm. 13C NMR (100 MHz, D2O): δ =
22.6, 22.7, 23.8, 36.6, 44.6, 44.7, 44.8 (3C), 50.3, 113.5, 119.4,
120.9, 128.8, 128.9, 135.5, 140.6, 148.6, 150.1 ppm. HRMS
(MALDI): calcd. for C19H32N5O [M+H-5HCl]+ 346.2601;
found 346.2614. Anal. Calcd for C19H36Cl5N5O·0.85 H2O: C,
42.02; H, 7; N, 12.9. Found: C, 42.13; H, 6.82; N, 12.73.
Study of the Acid−Base and Metal Complexation
Properties of Quilamine HQ1-44. Potentiometric titrations
were performed in a glass-jacketed titration cell thermostated at
25 0.1 °C and closed under an inert gas atmosphere, using a
Metrohm 702 SM Titrino titration stand connected to a
Metrohm 6.0233.100 combined glass electrode and controlled
by the Metrohm Tinet 2.4 software. The titrant was a carbonate-
free KOH solution at approximately 0.1 M prepared from a
commercial ampule of analytical grade, and its exact
concentration was obtained by titration of a standard HNO3
solution. A stock solution of HQ1−44 was prepared at ∼1.5 ×
10−3 M. Stock solutions of iron(III), copper(II), zinc(II), and
magnesium(II) cations were prepared at ∼0.05 M from the
analytical grade salts FeCl3, CuCl2, Zn(NO3)2, and MgSO4, and
were standardized by complexometry against a standard EDTA
(ethylenediaminetetraacetic acid) solution. Sample solutions
contained ∼0.03 mmol of ligand in a total volume of 30 mL,
with the ionic strength kept at 0.10 M using KNO3 as
background electrolyte. Complexation titrations were run by
addition of approximately 0.5 or 1 equiv for iron(III) and
copper(II), or 1 equiv for zinc(II) and magnesium(II). The
[H+] of the solutions was determined by measurement of the
electromotive force of the cell, E = E°′ + Q log[H+] + Ej. The
solutions of each chelator (10 mM) were prepared in water.
The high solubility of each derivative in culture medium in the
concentration range 0−400 μM was preliminary verified by
turbidimetry measurement. Three controls have been done for
each experiment: one with the standard culture medium, and
the other with culture medium. When chelator treatment was
performed in the presence of exogenous iron, they have been
iron saturated with 20 μM ferric iron (FeCl3). Unbound iron
was not removed from the medium. NTBI is classically used to
induce iron overload in various cells because it represents the
plasma iron species which is mainly responsible for creating
parenchymal (especially hepatocyte) iron deposition in human
chronic iron overload.
Comparison of Chelator Efficiency in Aqueous Phase,
Competition with Calcein. Fluorescence emission (λExc
=
485 nm, λEm = 520 nm) of calcein (100 nM) in an HEPES
buffer (20 mM HEPES, 150 mM NaCl, pH 7.3) was measured
at room temperature in a microplate fluorescence reader
(Packard, Fusion), equipped with an orbital stirring. Iron(III)
(1 μM) slowly reacted with calcein and maximal quenching of
its fluorescence was observed for time longer than 6 h. The
fluorescence recovery was monitored after 4 h incubation in the
presence of various chelator concentrations. In the absence of
iron, various chelators slightly modulated the calcein
fluorescence intensity. Results were expressed as a percentage
of fluorescence recovery with respect to the free calcein
fluorescence intensity measured in the absence of iron.
Cells Studies. The involvement of the PTS in the selective
uptake of Quilamines was investigated by using Chinese
hamster ovary cells (CHO) exhibiting high PTS activity46 and
the mutated derived-cell line CHO-MG cells, devoided of PTS
(provided by Dr. W. Flintoff, University of Western Ontario).
The CHO-MG cell was selected for growth resistance to
methylglyoxalbis-(guanylhydrazone), using a single-step selec-
tion after mutagenesis with ethyl methanesulfonate. Therefore,
highly selective PTS ligands should give high (CHO-MG/
CHO) IC50 ratios.
term pH is defined as −log [H+], and a value of Kw
=
[H+][OH−] = 10−13.778 was taken from the literature for our
experimental conditions. E°′ and Q were determined by
titrating a standard HNO3 solution at the same ionic strength.
The liquid-junction potential, Ej, was found to be negligible
under the experimental conditions used. Each titration
consisted of 150−200 equilibrium points in the range of pH
2.0−11.5, and at least two replicate titrations were performed
for each individual system. The potentiometric data were
refined with the Hyperquad software, and speciation diagrams
were plotted using the Hyss software. The overall equilibrium
(formation) constants βiH and βMmHhLl are defined by
βMmHhLl=[MmHhLl]/[M]m[H]h[L]l and βMH−1 L= βML(OH) × Kw, while
stepwise equilibrium constants are given by
The two cell lines were grown in RPMI 1640 medium,
supplemented with 10% fetal calf serum (Eurobio, Les Ulis,
France), 2 mM Glutamine (Bio Media, Boussens, France), 100
units/mL penicillin, and 50 μg/mL streptomycin, at 37 °C in
5% humidified CO2. L-Proline (2 μg/mL) was added to the
culture medium for CHO-MG cells.
Cell Treatment. For experiments, cells were harvested with
trypsin and seeded 24 h before the polyamine treatment in 96-
well microplates (Becton Dickinson, Oxnard, CA) at a density
of 2000 cells per well. In these conditions, cells reached
confluency in 5−6 days. Chelator exposure was performed one
day after cell seeding in proliferating CHO and CHO-MG cells.
The effect of Quilamines on cell viability was also tested in the
presence of DFMO (1 mM), a selective inhibitor of ODC
which induced a depletion of intracellular putrescine and
spermidine and leads to an activation of the PTS. In this assay,
the cells were seeded in the presence of the inhibitor. In other
experiments, 50 μM spermidine was added to the culture
medium to inhibit by a competitive inhibition, the active uptake
of Quilamines from PTS. To prevent the oxidation of the
exogenous polyamine by the serum amine oxidase present in
the fetal calf serum, aminoguanidine (1 mM), an inhibitor of
this enzyme, was added to the medium. The various derivatives
were compared to the tridentate iron chelator hydroxyphenyl-
triazole ICL670 (Deferasirox, Exjade) and to the bidentate 8-
hydroxyquinoline.
K
MmHhLl=[MmHhLl]/[MmHh−1Ll][H] and correspond to the difference
in log units between overall constants of sequentially
protonated (or hydroxide) species. The pM values for each
metal complex were calculated from the full set of stability
constants of each system at pH = 7.4 with cL = 1.0 × 10−5
and cM = 1.0 × 10−6 M.
M
BIOLOGY
■
Chelator Solutions. All reagents, which were of the highest
available grade, were obtained from Sigma-Aldrich Chimie
(Saint Quentin Fallavier, France). All synthesized Quilamines
were compared to the tridentate iron-chelator hydroxyphenyl-
triazole ICL670 as a reference (Deferasirox, Exjade from
Novartis Pharma) as well as to the 8-hydroxyquinoline. Stock
1957
dx.doi.org/10.1021/bc300324c | Bioconjugate Chem. 2012, 23, 1952−1968

Bioconjugate Chemistry
Article
Cytostatic and Cytotoxic Effects Measurements.
Chelator exposure was performed one day after cell seeding
in proliferating CHO and CHO-MG cells. After 72 h
incubation at 37 °C, cell supernatants were collected for
cytotoxicity measurement by measuring extracellular LDH
activity (cytotoxicity detection kit − LDH, Roche, Penzberg,
Germany). Extracellular LDH activity was measured as
described by the manufacturer on a 20 μL aliquot of cell free
medium obtained by centrifugation (2500 rpm/min during 5
min). LDH activities were detected by reading absorbance at
485 nm. They were reported as a percentage of extracellular
LDH activity with respect to the control value.
Cell viability was determined by counting the cell nuclei after
staining with the fluorescent DNA intercalating dye Hoechst
33342. The treated CHO and CHO-MG cells were washed
twice with PBS (50 mM pH 7) and fixed with ethanol/acetic
acid during 20 min at 4 °C. Cells were counterstained 10 min
with Hoechst 33342 dye diluted by 1/1000 (5 μg/mL) in PBS
and images were taken with an upright microscope
(AxioImager M1, Zeiss, France). Images analysis and cell
nuclei counting was performed with the simple PCI software (C
Imaging Image Analysis System, Compix, USA). The number
of cell nuclei was reported as a percentage of the control value.
Parameters of the dose−response curves were deduced from a
four-parameter curve fit according to Rodbard47
starting points. These starting points have led to the most
stable geometries presented below. The existence of local
minima was checked by inspection of the Hessian matrix
eigenvalues. For all iron complexes, the stability of the wave
function was checked. Unscaled values were used for zero-point
vibrational energy corrections for energy values. Thermal
corrections were calculated for the evaluation of reaction
enthalpies ΔH and Gibbs energies ΔG at 298.15 K, using
standard statistical mechanic formulas in the independent
mode, harmonic oscillator, and rigid rotor approximations.
RESULTS AND DISCUSSION
■
Chemistry. We first developed a synthetic protocol to
tether 8-hydroxyquinoline-2-carbaldehyde to a set of linear
polyamines by reductive amination (Figure 1).
(Acmin − Acmax
)
y =
+ Acmax
⎛
⎞
C
⎜
⎟
1 +
P
ip
C
⎝
⎠
ip
Figure 1. Quilamine HQ1-XYZ synthesis.
in which y is the percentage of cell nuclei number with respect
to the control, ACmin and ACmax are the y values observed for
minimal and maximal chelator concentrations respectively, C is
the chelator concentration (Cip at the inflection point), and Pip
is the slope at inflection point of the sigmoid curve. Due to
their biphasic feature, the dose−response curves were fitted as
the sum of two sigmoids (double 4-parameter fit). Percentages
of CHO and CHO-MG cells involved in each viability
responses were deduced from the ACmin, ACmax, and IC50 values
of each sigmoid, which were obtained from these fits.
To optimize the efficiency of this novel family of Quilamines
in terms of cytotoxicity and PTS selectivity, we studied the
influence of the number of nitrogen atoms present in the
polyamine chain and the distance between them. The synthesis
of nine protected spermidine and spermine analogues described
in Figure 2 was carried out. Diverse methodologies for the
Three independent replicates were performed for each
experiment, which was repeated three times.
Computational Details. As its computational cost is in the
same order of the Hartree−Fock (HF) method, DFT is
applicable to much larger systems than the correlated ab initio
methods. DFT theory has been used to obtain the structure and
energy of all the compounds under study. We have selected for
our calculations OPBE, one of the best performing exchange-
correlation functional for the iron complexes,48 which is the
combination of Handy’s optimized exchange (OPTX)49 with
the PBE correlation.50 The 6-31G(d,p) basis set, a double-ζ
Pople-type basis set for H,C N,O, and Cl atoms and the
Lanl2DZ (Los Alamos National Laboratory 2 double-ζ) with
effective core potentials (ECP) for Fe atom are employed in the
present investigation.
The DFT calculations were carried out with the Gaussian 09
software package.51 All open-shell structures were calculated
with unrestricted DFT method, whereas for closed-shell
molecules, the restricted DFT method was employed.
Calculations of iron complexes were performed for spin
multiplicities 2, 4, and 6. The Fe(III) complexes were
optimized, often using distinct geometries by placing the
ligands in pseudo-octahedral orientation about the metal as
Figure 2. Structure of protected spermidine and spermine analogues
1−9.
synthesis of polyamine have recently been reported in the
literature.52−54 We selected an iterative approach involving
cyano-alkylation followed by nitrile reduction.45,55,56
The spermidine analogues 1−4 were prepared according to a
modified four-step procedure providing good yields (Scheme
1).
The first step consisted of reacting putrescine or 1,3-
diaminopropane with di-tert-butyl dicarbonate (Boc2O) leading
to monoprotected diamines 10 and 11, respectively. Amine 10
was reacted with 4-bromobutyronitrile to give secondary amine
12, followed by quantitative Boc-protection giving rise to
compound 13. The nitrile was reduced by hydrogenation with
Raney Ni to give masked homospermidine 1 with an overall
yield of 63%. This procedure represents an improved yield
compared to a previously described protocol (44% yield).31,57
1958
dx.doi.org/10.1021/bc300324c | Bioconjugate Chem. 2012, 23, 1952−1968

Bioconjugate Chemistry
Article
a
Scheme 1. Synthesis of Spermidine Analogues 1−4
a
Reagents: (a) (Boc)2O (0.33 equiv), Et3N, MeOH, 0 °C to r.t., 20 h; (b) 4-bromobutyronitrile (1 equiv), K2CO3, CH3CN, 50 °C, 12 h; (c)
(Boc)2O (2.5 equiv), Et3N, MeOH, r.t., 20 h; (d) acrylonitrile (1 equiv), MeOH, 0 °C, 14 h; (e) H2, Raney Ni, EtOH, NH4OH, 20 atm, r.t., 12 h.
a
Scheme 2. Synthesis of Spermine Analogues 5−9
a
Reagents: (a) 4-bromobutyronitrile (1 equiv), K2CO3, CH3CN, 50 °C, 12 h; (b) acrylonitrile (1 equiv), MeOH, 0 °C, 14 h; (c) (Boc)2O (2.5
equiv), Et3N, MeOH, r.t., 20 h; (d) H2, Raney Ni, EtOH, NH4OH, 20 atm, r.t., 12 h.
N4,N8-Di-tert-butyloxycarbonylspermidine 2 was prepared, still
from amine 10, which was alkylated this time via a Michael-
addition on acrylonitrile with a yield of 96%. Boc protection of
the resultant secondary amine 14 provided nitrile 15.
Treatment with Raney Ni under 20 atm of hydrogen induced
the expected nitrile reduction in quantitative yield with no
further purification required. This method proved to be more
suitable than that described by Paul M. O’Neill in 2010, which
used LiAlH4 as reagent and provided yields of no more than
24%.58 With this protocol, di-Boc-spermidine 2 was obtained
with a significantly higher overall yield than previously
described (74% vs 32%).59 N1,N4-Di-tert-butyloxycarbonylsper-
midine 3 and norspermidine 4 were prepared in the same
manner from Boc-protected diamine 11 with an overall yield of
49% and 78%, respectively.
Dihomospermine 5, homospermines 6 and 7, spermine 8,
and norspermine 9 were designed as homologues of 1−4,
where the N1-tether was sequentially lengthened. Compounds
5−9 were prepared similarly to the di-Boc spermidine
analogues in three steps (Scheme 2). Spermine skeletons 1−
4 were first alkylated with bromobutyronitrile or acrylonitrile to
give compounds 20−24. After protection of the secondary
amino group with di-tert-butyl dicarbonate, compounds 25−29
were converted into the expected spermines 5−9 with excellent
yields (84−95%).
8-Hydroxyquinaldine was protected as the silylether by
treatment with tert-butyldimethylsilyl chloride in the presence
of imidazole, followed by an oxidation of the methyl group by
selenium dioxide in 1,4-dioxane to yield the corresponding
aldehyde in 90% yield. After deprotection of phenol by
treatment with an excess of tetra-n-butylammonium fluoride in
THF, the 8-hydroxyquinoline-2-carbaldehyde 30 was isolated
with an overall yield of 66%. The reductive amination of this
aldehyde with a range of polyamines 1−9 was achieved in one
pot by using sodium triacetoxyborohydride for reduction of the
imine generated in situ.61 Protected intermediates 31−39 were
isolated in good yields (82−96%). Removal of the Boc-
protecting groups under acidic conditions at room temperature
gave the polyaminoquinoline HQ1-XYZ as hydrochloride salts.
These nine new Quilamines were fully characterized by a
combination of the usual techniques (IR, UV−vis, MS,
1
elemental analysis, 13C and H NMR) and fulfilled the purity
required for the biological evaluations.
Biological Evaluations. Cytostatic (Antiproliferative) and
Cytotoxic Effects of Chelators. Quilamines HQ1-XYZ were
screened for their antiproliferative activity and cytotoxicity in
CHO (Chinese Hamster Ovary) cells which exhibit high PTS
activity62 and the mutated derived-cell line CHO-MG cells,
devoid of the PTS (provided by Dr. W. Flintoff, University of
Western Ontario). CHO cells and the CHO-MG cell line were
chosen to evaluate the potential role of the PTS as a selective
transport system. We assessed the antiproliferative activity of
the various compounds by measuring DNA content after
Hoescht staining. The release of the cytoplasmic protein lactate
dehydrogenase (LDH) into cell supernatants was used as a
marker of cell membrane disruption correlated to cytotoxicity.
With protected polyamines 1−9 in hand, we decided to
evaluate a reductive-amination protocol with 8-hydroxyquino-
line-2-carbaldehyde to form the Quilamines HQ1-XYZ. 8-
Hydroxyquinoline-2-carbaldehyde 30 was obtained in three
steps according to methods previously described in the
literature (Scheme 3).60
1959
dx.doi.org/10.1021/bc300324c | Bioconjugate Chem. 2012, 23, 1952−1968

Bioconjugate Chemistry
Article
a
Scheme 3. Synthesis of Quilamines HQ1-XYZ
a
Reagents: (a) TBDMSCl (1.1 equiv), imidazole (1.1 equiv), CH2Cl2, r.t., 12 h; (b) selenium dioxide (1.2 equiv), 1,4-dioxane, 90 °C, 90 min; (c)
TBAF (1.5 equiv), THF, r.t., 2 h; (d) polyamines 1−9, NaBH(OAc)3 (3 equiv), C2H4Cl2, r.t., 12 h; (e) HCl 6 M, EtOH, r.t., 20 h.
The dose−response curves deduced from the counting of cell
nuclei by Hoescht fluorescence in proliferating CHO cells were
biphasic for the various Quilamines, including HQ1−44 shown
as an example in Figure 3A, except HQ1−43, HQ1−343,
HQ1−333, and the reference chelator 8-hydroxyquinoline (8-
HQ). They were fitted as a double four-parameter fit to
calculate the percentage of cells involved in each viability
component, the cytostatic (C1) and cytotoxic ones (C2), as
well as their corresponding IC50 (Table 1). Dose−effect curves
of the extracellular LDH released into the supernatant,
expressed in relation to the untreated CHO and CHO-MG
control cells, are reported in Figure 3B for HQ1−44 and the
two reference chelators (8-HQ and ICL670). The decrease in
cell viability observed with 8-HQ (Figure 3A) was associated
with a cytotoxic effect leading to LDH leakage (Figure 3B),
while that observed with ICL670 and HQ1−44 at concen-
trations lower than 10 μM corresponded to a cytostatic effect
(no LDH release). LDH values for compound concentrations
in the range of the cytostatic component 1 (10 μM) and the
cytotoxic component 2 (100 μM) are also reported in Table 1.
The low concentration component 1 involved 80%, 55%,
35%, 60%, 30%, and 18% of CHO cells treated with HQ1−44
(IC50−1 = 1.2 μM), HQ1−444 (IC50−1 = 1.8 μM), HQ1−443
(IC50−1 = 1.0 μM), HQ1−33 (IC50−1 = 2.0 μM), HQ1−34
(IC50−1 = 20.0 μM), and HQ1−344 (IC50−1 = 2.0 μM),
respectively; and only 20% of CHO cells treated by ICL670
(IC50−1 = 3.0 μM). As deduced from the absence of LDH
leakage into supernatants, the chelator treatments at 10 μM
were not cytotoxic and probably reflected the antiproliferative
activity of the various compounds. In contrast, the second
component of the dose−effect curves was associated with
membrane damage and LDH release for a concentration of 100
μM and corresponded to the cytotoxicity of the various
Quilamines and the reference chelators (8-HQ and ICL670). In
nonproliferative CHO cells, treated with the various com-
pounds at confluency (data not shown), the dose−response
curves of cell viability were monophasic, with IC50 higher than
50−100 μM, and reflected only their cytotoxicity (LDH
release).
The antiproliferative efficiency of the various Quilamines,
deduced from the percentage of the low concentration
component 1 (IC50−1 < 10 μM) was in the following order:
HQ1−44 (80%) > HQ1−33 (60%) ∼ HQ1−444 (55%) >
HQ1−443 (35%) > HQ1−34 (30%) > HQ1−344 (18%) >
HQ1−333 ∼ HQ1−43 ∼ HQ1−343. The Quilamine HQ1−
44 exhibited the highest antiproliferative activity and the lowest
cytotoxicity with respect to all the other compounds, including
the reference chelator ICL670 (20%, IC50 = 3.0 μM).
Selectivity for the Polyamine Transport System (PTS).
When cell viability was analyzed without multicomponent
analysis, the dose−effect curves obtained with Quilamine
HQ1−44 showed overall IC50 values of 1.4 μM and 344 μM
in the CHO and mutated PTS deficient CHO-MG cell line,
respectively (Figure 3A and Table 2). The antiproliferative
effect of this compound HQ1−44 was observed in the
micromolar range, while its cytotoxicity, which was detected
by membrane damage (LDH release, Figure 3B), was observed
only for concentrations higher than 100 μM in both CHO and
CHO-MG cell lines.
1960
dx.doi.org/10.1021/bc300324c | Bioconjugate Chem. 2012, 23, 1952−1968

Bioconjugate Chemistry
Article
on CHO-MG cells (data not shown). These results strongly
suggest that the antiproliferative capacity of the various
Quilamines depends mainly on their cell uptake by the PTS.
Quilamines bearing a 44 polyamine chain (HQ1−44, HQ1−
444) were more selectively recognized by the PTS than those
bearing a 33 chain (HQ1−33, HQ1−333). The influence of
the polyamine sequence on the selective uptake by the PTS in
CHO cells was previously reported by using a collection (a set)
of polyamine−anthracene conjugates.28 Our results were in
good agreement with this study showing the higher PTS
selectivity of the 4,4-triamine analogues, its decrease after
terminal amino-butylation such as in the 4,4,4-tetraamine and
the lower PTS recognition of polyamine derivatives bearing a
3,3-triamine chain. In agreement with this previous study, 34
chains (HQ1−34, HQ1−344, HQ1−343) appeared to be
poorly, or not, recognized by the PTS (Table 2). These
similarities in the PTS targeting of our Quilamines and the
polyamine−anthracene conjugates suggested the same cell
uptake pathways for these two families of polyamine-analogues
even if they could hit different cellular targets (DNA
intercalation or iron depletion). The greater selectivity of the
Quilamine HQ1−44 for the PTS is in close agreement with
previous works showing the higher cellular uptake of azepine−
polyamine31 and anthryl−polyamine38 conjugates bearing a
homospermidine chain.44
Complexation Properties of Quilamines toward Iron-
(III) and Competitive Cations. Comparison of the
Chelation Efficiency of Quilamines HQ1-XYZ, 8-HQ,
ICL670, and O-Trensox. The cellular labile iron pool (LIP)
consists of chelatable and redox-active iron, which plays a key
role as a crossroads of cell iron metabolism. The ability of iron
chelators to mobilize this temporary iron pool bound to low-
molecular-weight and low-iron-affinity chelators (citrate,
ascorbate, phosphate, and adenosine triphosphate) is an
essential factor influencing their biological efficiency. According
to previous work, the calcein test can predict the iron-chelating
efficiency of chelators.63,64 Thus, an acellular calcein test was
performed to compare the potential ability of Quilamines
HQ1-XYZ, ICL670, O-Trensox, and 8-HQ, iron chelators used
as references, to chelate this labile iron pool at physiological
pH. In solution, calcein, a fluoresceinated analogue of EDTA,
binds iron(II) and, more slowly, iron(III).65 The fluorescence
of this metallosensor dye is quenched during its interaction
with iron and, conversely, is restored if iron is removed from
the [calcein-Iron] complex by a most efficient competitive iron
chelator. The rate and extent of fluorescence recovery depend
on the chelator concentration, the kinetics and stoichiometry of
iron binding and the relative binding affinity. The calcein test is
qualitative, but the concentration of the chelator for which 50%
of the fluorescence is restored is indicative of the chelator
affinity (RC50).
Recovered calcein fluorescence was measured in aqueous
medium 30 min after the addition of the competitive chelators
and is reported as a function of their concentrations in Figure 4.
As expected from their relative affinity for iron(III), the three
reference chelators, 8-HQ, ICL670, and O-Trensox, exhibited
RC50 values equal to 4, 1.4, and 0.9 μM, respectively (Table 3).
Curves of calcein fluorescence recovery were monophasic for
the three reference chelators, while they showed two distinct
components for all Quilamines. The first component (C1) was
observed in the micromolar range of concentrations and the
second one for Quilamine concentrations higher than 10 μM
(C2). These biphasic curves were fitted as a double four-
Figure 3. Cytostatic and cytotoxic effects of HQ1−44, 8-HQ, and
ICL670 in the CHO (solid lines) and CHO-MG cell lines (dotted
lines). The cytostatic effect was determined by measuring DNA
content after Hoescht staining (A). The cytotoxic effect was evaluated
◆
by measuring LDH leakage into the culture medium (B). HQ1−44
◇
■
□
(CHO),
HQ1−44 (CHO-MG),
ICL670 (CHO), ICL670
▲
(CHO-MG),
8-HQ 5CHO), Δ 8-HQ (CHO-MG) for A and B.
Due to the absence of a polyamine chain, the two reference
chelators, 8-HQ and ICL670, were not imported by the PTS
and showed an equivalent efficiency in the two cell lines with a
CHO-MG/CHO IC50 ratio equal to 1. In contrast, a high value
of this IC50 ratio was characteristic of a high PTS selectivity
(Table 2). The Quilamine HQ1−44, which presents an
exceptionally high CHO-MG/CHO IC50 ratio (249), was the
one most efficiently imported into the CHO cells due to its
strong selectivity of recognition by the PTS. Exogenous
spermidine (50 μM), which induced a competitive inhibition
of the HQ1−44 uptake by the PTS, decreased the
antiproliferative efficiency of this Quilamine in CHO cells
(Figure S1, ESI). Conversely, the activation of the PTS induced
by a DFMO pretreatment increased the cytostatic effect of
HQ1−44. Both DFMO and exogenous spermidine treatments
were ineffective at modulating the cytotoxic effect of HQ1−44
1961
dx.doi.org/10.1021/bc300324c | Bioconjugate Chem. 2012, 23, 1952−1968

Bioconjugate Chemistry
Article
Table 1. Cytostatic/Cytotoxic Effects and IC50 of Quilamines HQ1-XYZ, 8-HQ, and ICL670 in the CHO Cell Line
a
b
% component 1 (Cytostatic) IC50-1 (μM) %LDH/control (10 μM) % component 2 (Cytotoxic) IC50-2 (μM) %LDH/control (100 μM)
HQ1−44
HQ1−444
HQ1−443
HQ1−33
HQ1−333
HQ1−43
HQ1−34
HQ1−344
HQ1−343
8-HQ
80
55
35
60
0
1.2
1.8
1
102
105
100
102
111
97
20
11
65
45
45
15
30
110
70
27
4
125
204
203
378
124
228
200
160
199
245
402
45
65
2
40
ND
ND
20
2
100
100
70
0
30
18
0
111
101
93
82
c
c
ND
ND
3
100
100
80
0
259
108
ICL670
20
11
a
Cytostatic effect was determined by measuring DNA content after Hoescht staining. Dose−effect curves were fitted as a double four-parameter fit
b
to calculate the percentage of cells involved in each viability component (cytostatic and cytotoxic) as well as their corresponding IC50. Cytotoxic
effect was evaluated by measuring LDH leakage into the culture medium. ND, not detectable.
c
Table 2. Selectivity of Quilamines for the PTS. Effect of
Quilamines HQ1-XYZ, 8-HQ, and ICL670 in CHO and
CHO-MG Cells
Table 3. Percentage of Each Component and RC50 of HQ1-
XYZ, 8-HQ, ICL670, and O-Trensox
a
% component 1
RC50-1
(μM)
% component 2
RC50-2
(μM)
a
compound
(C1)
(C2)
IC50 (μM)
CHO-MG
IC50 ratio
HQ1−44
HQ1−43
58
55
53
1
1
1
42
45
47
57
15
18
compounds
CHO
CHO-MG/CHO
HQ1−44
HQ1−444
HQ1−443
HQ1−33
HQ1−333
HQ1−43
HQ1−34
HQ1−344
HQ1−343
8-HQ
1.4
6.3
344.5
239.5
257.5
117.9
277.0
163.5
152.0
207.1
105.7
6.5
249
38
13
21
18
5
HQ1−
444
20.2
5.7
HQ1−
53
1
47
25
443
HQ1−34
HQ1−33
78
76
74
0.7
22
24
26
10
10
20
15.1
31.3
93.1
55.9
27.0
5.0
0.8
HQ1−
0.65
2
344
4
HQ1−
74
72
0.65
0.6
26
28
10
12
4
343
1
HQ1−
ICL670
9.2
8.0
1
333
a
8-HQ
100
100
100
4
0
0
0
0
0
0
This denotes the (CHO/CHO-MG) IC50 ratio, a measurement of
PTS selectivity.
ICL670
1.4
0.9
O-
Trensox
a
Biphasic curves (Figure 4) were fitted as a double four-parameter fit
to calculate the percentage of each component as well as their
corresponding concentration inducing half-fluorescence recovery
(RC50).
parameter fit to calculate the percentage of each component, as
well as their corresponding concentration inducing half-
fluorescence recovery (RC50, Table 3). The first component
C1 was greater with Quilamines HQ1−3YZ (72−78% with
RC50 = 0.6−0.8 μM) than with Quilamines HQ1−4YZ (53−
58% with RC50 = 1 μM). The ability of Quilamines to
transchelate iron from calcein was between those of 8-
hydroxyquinoline and O-Trensox. The shape of these curves
(Figure 4) allowed two distinct groups of Quilamines to be
discriminated: HQ1−3YZ (HQ1−34, HQ1−33, HQ1−344,
HQ1−343, HQ1−333) and HQ1−4YZ (HQ1−44, HQ1−43,
HQ1−444, HQ1−443) and suggested different modes of
interaction with iron, probably modulated by the polyamine
chain structure. Due to their amino groups, polyamines are able
to chelate metallic cations.66 The stability constants of these
chelates increase with the number of amino groups and with
the chain length of the polyamine.67 Our results suggest the
involvement of the amine group of the polyamine chain in the
formation of iron complexes, as expected during the design and
Figure 4. Comparison of the iron-chelating efficiency of Quilamines
HQ1-XYZ with 8-HQ, ICL670, and O-Trensox, by calcein
fluorescence measurements in a cell-free system.
1962
dx.doi.org/10.1021/bc300324c | Bioconjugate Chem. 2012, 23, 1952−1968

Bioconjugate Chemistry
Article
synthesis of this Quilamine concept. The first secondary amine
in Quilamines HQ1 could be involved in iron coordination and
probably the second one too, as deduced from the better iron-
chelating efficiency of Quilamines HQ1−3YZ compared to
Quilamines HQ1−4YZ.
There was an increasing tendency to form insoluble species
above neutral pH, most likely hydroxocomplex species, with
high amounts of iron(III). By refining together all the titrations
performed, thermodynamic stability constants could be
obtained accurately for only a limited set of complex species,
mostly of 1:1 stoichiometry. This is due to some precipitation
occurring at basic pH, which forced us to exclude the
corresponding experimental results. However, as the amount
of precipitate was small below pH = 9, we used the
experimental data up to this pH in a separate refinement to
obtain a rough approximation of the stability constants of the
remaining species of 1:2 stoichiometry in order to gain an
insight into the overall thermodynamic equilibria. A selection of
calculated overall (log β) and stepwise (log K) stability
constants is presented in Table 5 (a complete table showing the
overall stability constants is available in ESI, Table S1).
Complexation Properties of Quilamine HQ1−44 toward
Iron(III) and Competitive Cations. To understand better the
high iron affinity observed for the most promising chelator
Quilamine HQ1−44, we assessed its complexation behavior
toward iron(III) as well as other relevant cations such as
copper(II), zinc(II), and magnesium(II), which may act as
endogenous competitors of iron(III). This study was performed
by potentiometric titrations in aqueous solutions at 25 °C. The
ionic strength was adjusted to 0.10 M with potassium nitrate.
As a starting point, the acid−base properties of HQ1−44
were first determined in the form of the protonation constants
corresponding to its five basic centers located at the four
amines and the single hydroxyl function. The calculated overall
(log β) and stepwise (log K or pKa) protonation constants are
presented in Table 4. The four most basic centers were easily
Table 5. Overall and Stepwise Stability Constants for the
Iron(III) Complexes of HQ1-44 (25.0 °C, I = 0.10 M in
KNO3)
a
species
log β
log K
Table 4. Overall and Stepwise Protonation Constants for
Compound HQ1-44 (25.0 °C, I = 0.10 M in KNO3)
FeL
25.62
32.13
34.94
33.00
42.20
51.00
59.10
25.62
6.51
2.81
7.4
FeHL
FeH2L
FeL2
a
species
log β
log K (pKa)
HL
10.83
20.64
29.69
37.67
10.83
9.81
9.05
7.98
H2L
H3L
H4L
H5L
FeHL2
FeH2L2
FeH3L2
9.2
8.8
8.1
b
b
(39.0)
(1.3)
a
HL = HQ1−44, and charges are omitted for the sake of clarity.
a
HL = HQ1−44, and charges are omitted for the sake of clarity.
Values in brackets are approximate.
b
Overall, we found that compound HQ1−44 formed very
strong iron(III) complexes. The thermodynamic stability of the
complexes with 1:1 stoichiometry was very high, while that of
complexes with 1:2 stoichiometry was more modest, as shown
by the log K values of almost 26 for the former and below 8 for
the latter. This clearly illustrates the preference of the ligand to
form 1:1 complexes with adequate iron concentrations.
However, the calculated overall log β values indicated that
both 1:1 and 1:2 species were present successively in large
amounts all along the pH scale: 1:1 species were formed
between pH 2−6, while 1:2 species appeared above pH 6 with
the largest amount occurring at around pH 7.5. This is
demonstrated by the abundant presence of the three
protonated species (log β around 60). This high thermody-
namic stability also implies that a high complexation degree is
attained at low pH values, with almost all of the iron(III) being
chelated above pH = 3. This corresponds to a very small
amount of free iron(III) present in solution at physiological pH
(7.4), as indicated by a pFe3+ value of 19.4. It is thus clear that
HQ1−44 is a powerful chelator of iron(III).
We also studied the complexation of HQ1−44 with
competitive cations found in vivo. Titrations were performed
in the presence of 0.5 equiv or 1 equiv of copper(II), and in the
presence of 1 equiv of zinc(II) or magnesium(II). The
calculated overall (log β) and stepwise (log K) stability
constants are presented in Table S2 of the Supporting
Information. The constant values show that strong complexes
were formed with copper(II), with species of both 1:1 and 1:2
stoichiometries (pCu2+ = 12.6). The complexes of zinc(II) were
only moderately stable, and also contained 1:1 and 1:2 species
(pZn2+ = 7.8). With magnesium(II), only a few complex species
of 1:1 stoichiometry were formed with an extremely low
determined and have pKa’s in the range of 8−11, while the
remaining center shows a much lower basicity with a pKa that
can just be estimated at around 1.3 as it is outside the accurate
experimental range of the potentiometric technique used. The
constants obtained can be understood in light of the known
acid−base properties of the two functional moieties that
constitute the ligand, 8-hydroxyquinoline68 and homospermi-
dine.69 The highest pKa of 10.83 must correspond to the
protonation of the primary amine. The pKa at 9.81 can easily be
assigned to protonation of the hydroxyl group, as the yellow
color of the ligand solution intensifies visibly around this pH
value. The two remaining basic pKa’s of 9.05 and 7.98 belong to
the two secondary amines, but it is not possible to assign the
values individually. The acidic pKa at 1.3 must be assigned to
the protonation of the pyridine, as it is the least basic group in
the chelator due to the close vicinity of the electron-
withdrawing hydroxyl and secondary amine groups.
Next, we turned our attention to the complexation of
iron(III). Since HQ1−44 presents a total of 5 potential donor
atoms for metal coordination, we performed potentiometric
titrations with different amounts of metal in order to obtain a
complete insight into the potential 1:2 and 1:1 metal-to-ligand
complexation modes and the corresponding complex species.
Thus, metal ratios of approximately 0.5 equiv and 1 equiv of
iron(III) (relative to the ligand amount) were used during
complexation titrations, and the titration curves obtained were
used jointly to refine the complete model of stability constants
for all the complex species. For both iron(III) ratios, species of
1:2 and 1:1 stoichiometry were present in equilibrium, with a
predominance of 1:1 species for the higher amount of iron(III).
1963
dx.doi.org/10.1021/bc300324c | Bioconjugate Chem. 2012, 23, 1952−1968

Bioconjugate Chemistry
Article
stability (pMg2+ is insignificant). Distribution diagrams of the
participating species for each metal are presented in ESI (SI
Figures S2−S5). A competition diagram between Fe(III) and
Zn(II) has been obtained from the values of all the stability
constants with both metals; it clearly indicates the formation of
a major amount of ternary species with Fe(III) all along the pH
scale (in ESI SI Figure S6).
Taken together, these complexation studies prove that
Quilamine HQ1−44 forms very strong complexes with
iron(III) and much weaker complexes with other endogenously
relevant metal cations, indicative of a high selectivity for
iron(III).
Ab Initio Calculations. Density functional theory (DFT)
calculations were carried out using the simplified model
compound (HQ1−1) in which the polyamine moieties were
replaced by a methylamine chain to render the calculations
computationally tractable. All geometries reported in this paper
are available in the Supporting Information. The ferric iron has
3d electrons and can thus adopt different spin configurations.
The complexes were calculated in different spin states, but in all
cases, the most stable Fe(III) complexes considered in this
paper are high spin.
Despite the existence of a strong Nqui−HO hydrogen bond
for the free ligand, which has been extensively studied by
Shchavlev et al.,70 iron(III) complexation is possible. The
bonding in these complexes is expected to be dominated by
ion−dipole electrostatic interactions although contributions
arising from charge transfer, polarization, dispersion inter-
actions, and exchange repulsive interactions at the equilibrium
configuration of the complex are also likely to be present.
The geometries were compared with the bidentate ligand 8-
HQ. The complexes were constructed by placing the ligands in
pseudo-octahedral, but random, orientation about the metal
under the C1 point group because we did not wish to
presuppose any particular symmetry.
The preferential 1:2 [Fe/HQ1−44] stoichiometry of the
ferric complex at pH = 7.4 was confirmed by the Job method of
continuous variation (based on the MLCT (metal−ligand
charge transfer) band) (Figure 5). The iron complex presents
We modeled a highly stable complex like [Fe(8-
3+
HQ)n(H2O)6−2n
]
(n = 1 or 2) to use as an energy reference
and to consider reactions that correspond to an exchange of
ligands (8-HQ and HQ1−1) as in eq 1.
Figure 5. Job plot of HQ1−44 at pH = 7.4 (λmax = 560 nm). Solutions
containing different ligand/Fe(III) ratios were prepared so that
[ligand] + [Fe(III)] = 0.5 mM. The theoretical mole fraction
maximum for a 2/1 ligand/Fe complex is 0.667; a linear intercept
maximum of 0.65 was found.
The equilibrium constant of the exchange reactions can be
calculated from the computed standard free energy change of
reaction in vacuo. All free energy values were negative, so the
colorimetric features (ε560 = 900 M−1 cm−1 for pH 7.4) and
good stability properties (absorbance remains stable after
several weeks). This clearly indicates that the homospermidine
fragment of HQ1−44 at pH = 7.4 alters the stoichiometry of
the ferric complex. In fact, in the same conditions,
hydroxyquinoline forms a 3:1 ligand/iron complex.
In view of this stoichiometry, it could be assumed that the
iron complex [Fe/HQ1−44] forms three five-membered
chelate rings in which the iron is probably coordinated by
two vicinal N-atoms and a hydroxyl group (Figure 6). This
1
chelation reaction was highly exothermic (Table 6). On the
3+
basis of the log Knexc obtained, the [Fe(HQ1−1)n(H2O)6−3n
]
were the most stable for the 1:1 and 1:2 species.
The [Fe(HQ1−1H)2]5+ complex is not presented here
because it leads to the formation of a tetrahedral complex, due
to the repulsion of the protonated amines. Figure 7 shows the
minimum energy structures obtained for [Fe(8-
HQ)1(H2O)4]3+, [Fe(8-HQ)2(H2O)2]3+, [Fe(HQ1−
1)1(H2O)3]3+, and [Fe(HQ1−1)2]3+ complexes. All are
distorted octahedrons. The successive replacement of H2O by
the methylamine chain for HQ1−1 results in an increase in
both the average metal−ligand bond distance and the
stabilization energy of the complex. The most stable structure
for the [Fe(HQ1−1)2]3+ complex is a distorted octahedron in
which oxygen and nitrogen atoms from both HQ1−1 neutral
ligands are bound to the metal. The symmetry of this complex
is very close to the C2 symmetry. The two 8-hydroxyquinoline
groups are planar and form a five-membered chelate ring with
iron, leading to the facial monomer. The bond lengths with iron
are in narrow ranges: between 2.084 and 2.190 Å for Fe(III)−N
bonds and 2.474 Å for Fe(III)−O bonds. The Fe(III)−O
distances are much longer than observed with [Fe(H2O)6]3+:
1.995 Å.71 The Fe(III)−N distances are also significantly longer
than those observed in triazamacrocyclic complexes (i.e., 2.23−
2.33 Å).72 The average five-membered chelate ring (NquiFeO
Figure 6. Hypothetical structure of HQ1−44/Fe complex with five-
membered chelate rings.
stoichiometry was confirmed by elemental analysis, but
unfortunately, repeated attempts to isolate X-ray quality crystals
of the iron complex have so far been unsuccessful.
Due to the lack of the experimental electronic structure of
metal complexes, the OPBE calculation was used to determine
the energy and the structure of [Fe-Ln]3+ (n = 1 or 2)
complexes with a prototypical ligand HQ1−1.
1964
dx.doi.org/10.1021/bc300324c | Bioconjugate Chem. 2012, 23, 1952−1968

Bioconjugate Chemistry
Article
Table 6. Log βn, Log Knexc, Selected Bond Distances (Å) and Angles (deg) of Octahedral Complexes Determined by Quantum
Chemical UOPBE/6-31G(d,p) and (Fe Lanl2DZ) Calculations
complex
Logβn
[Fe(8-HQ)1(H2O)4]3+
[Fe(HQ1−1)1(H2O)3]3+
75
[Fe(8-HQ)2(H2O)2]3+
[Fe(HQ1−1)2]3+
58
82
111
LogKnexc
17
29
a
FeOqui
2.498
2.547
2.360 and 2.360
2.474 and 2.474
a
FeOw
2.145 and 2.176 and 2.176
2.172
2.131 and 2.200 and 2.243 2.200 and 2.200
a
FeNqui
2.117
2.161
2.117 and 2.117
2.084 and 2.084
2.190 and 2.190
a
FeNma
a
FeOw
2.146
68.85
2.200 and 2.200
72.18 and 72.18
a
NqiFeOqui
68.37
78.82
70.34 and 70.34
78.21 and 78.21
a
NqiFeNam
a
NqiFeOw
93.92 and 93.92 and 102.65 and 150.77
94.16 and 97.77 and 157.95 97.29 and 97.29 and 99.97 and
99.97
a
NqiFeOqui
OqiFeOqui
NqiFeNqui
146.10 and 146.10
82.40
a
a
92.79
155.85
163.19
a
N
maFeNma
114.42
a
OwFeOw
88.01 and 88.01 and 87.14 and 87.14 and 106.57 and
171.82
83.62 and 84.31 and 167.77 88.38
a
Nqui: N of quinoline, Nma: N of methylamine, Oqui: O of quinoline, Ow: O of water.
(PTS). Nine Quilamines HQ1-XYZ, differing in the nature of
their polyamine chain, were synthesized in six or nine steps
with good yields. The influence of the polyamine chain on the
selective recognition of different Quilamines by the PTS was
evidenced by comparing the antiproliferative activity obtained
on a cell line displaying an efficient PTS (CHO) with that on a
PTS-deficient mutant cell line (CHO-MG). This screening led
us to choose the molecule HQ1−44 from the synthesized
Quilamines. The antiproliferative effect of HQ1−44 occurred
in the micromolar range, while a significant cytotoxicity was
only observed at concentrations higher than 100 μM. We also
demonstrated the high complexation capacity of HQ1−44 with
iron, while much weaker complexes were formed with other
competitive cations. The ferric complex was shown to adopt
preferentially a 1:2 [Fe/HQ1−44] stoichiometry at pH = 7.4.
The binding energies and the equilibrium geometries of [Fe-
Ln]3+ (n = 1 or 2) complexes were assessed in vacuo from ab
initio calculations. The stronger equilibrium constant of the
exchange reactions between iron(III) and HQ1−1 compared to
8-HQ was deduced from these calculations. The Quilamine is a
tridentate ligand for iron(III) where both nitrogen and oxygen
atoms form five-membered chelate rings.
In summary, Quilamines, and particularly HQ1−44, may be
promising compounds for iron depletion in cancer therapies,
due to their high affinity and selectivity for iron, their high
antiproliferative activity associated with a low cytotoxicity, and
their selective uptake by the PTS.
Further studies will evaluate the following: (i) the influence
of modifying the length and the nature of the spacer between
the chelating moiety and the homospermidine chain; (ii) the
capacity to redox cycle and the generation of reactive oxygen
species (ROS) (like the objective is an antitumor effect, the
induction of oxidative stress may actually be a beneficial feature
of an iron-chelator complex); (iii) the biological mechanisms of
action of Quilamines, especially of HQ1−44; (iv) the
antiproliferative efficiency of the Quilamine HQ1−44 in vitro
in various human tumor cell lines and in vivo in animal models.
Figure 7. Minimum energy structures for [Fe(8-HQ)1(H2O)4]3+,
[Fe(HQ1−1)1(H2O)3]3+, [Fe(8-HQ)2(H2O)2]3+, and [Fe(HQ1−
1)2]3+ determined by quantum chemical UOPBE/6-31G(d,p) and
(Fe Lanl2DZ) calculations.
and NquiFeNma) is 74° (compared to 90° for an ideal
octahedron). The apical angles (NquiFeO and NmaFeNma) are
146.10°, 146.10°, and 163.19° (compared to 180° for an ideal
octahedron). However, this distorted geometry does not impair
the affinity of HQ1−1 for iron as evidenced in the
potentiometric assay.
CONCLUSION
■
We have developed a new type of iron chelator where an 8-
hydroxyquinoline chelating subunit is grafted to polyamine
vectors, in order to vectorize the iron chelator inside the
cancerous cell with an overactive polyamine transport system
1965
dx.doi.org/10.1021/bc300324c | Bioconjugate Chem. 2012, 23, 1952−1968

Bioconjugate Chemistry
Article
(11) Richardson, D. R. (2002) Iron chelators as therapeutic agents
for the treatment of cancer. Crit. Rev. Oncol. Hematol. 42, 267−281.
(12) Hoffbrand, A. V., Cohen, A., and Hershko, C. (2003) Role of
deferiprone in chelation therapy for transfusional iron overload. Blood,
17−24.
(13) Maggio, A., D’Amico, G., Morabito, A., Capra, M., Ciaccio, C.,
Cianciulli, P., Di Gregorio, F., Garozzo, G., Malizia, R., Magnano, C.,
Mangiagli, A., Quarta, G., Rizzo, M., D’Ascola, D. G., Rizzo, A., and
Midiri, M. (2002) Deferiprone versus deferoxamine in patients with
thalassemia major: a randomized clinical trial. Blood Cells, Molecules,
and Diseases 28, 196−208.
(14) Cappellini, M. D., Cohen, A., Piga, A., Bejaoui, M., Perrotta, S.,
Agaoglu, L., Aydinok, Y., Kattamis, A., Kilinc, Y., Porter, J., Capra, M.,
Galanello, R., Fattoum, S., Drelichman, G., Magnano, C., Verissimo,
M., Athanassiou-Metaxa, M., Giardina, P., Kourakli-Symeonidis, A.,
Janka-Schaub, G., Coates, T., Vermylen, C., Olivieri, N., Thuret, I.,
Opitz, H., Ressayre-Djaffer, C., Marks, P., and Alberti, D. (2006) A
phase 3 study of deferasirox (ICL670), a once-daily oral iron chelator,
in patients with beta-thalassemia. Blood 107, 3455−3462.
(15) Chantrel-Groussard, K., Gaboriau, F., Pasdeloup, N., Havouis,
R., Nick, H., Pierre, J. L., Brissot, P., and Lescoat, G. (2006) The new
orally active iron chelator ICL670A exhibits a higher antiproliferative
effect in human hepatocyte cultures than O-trensox. Eur. J. Pharmacol.
541, 129−137.
(16) Gaboriau, F., Laupen-Chassay, C., Pasdeloup, N., Pierre, J.-L.,
Brissot, P., and Lescoat, G. (2006) Modulation of cell proliferation and
polyamine metabolism in rat liver cell cultures by the iron chelator O-
trensox. BioMetals 19, 623−632.
(17) Richardson, D., Ponka, P., and Baker, E. (1994) The effect of
the iron(III) chelator, desferrioxamine, on iron and transferrin uptake
by the human malignant melanoma cell. Cancer Res. 54, 685−9.
(18) Donfrancesco, A., Deb, G., De Sio, L., Cozza, R., and Castellano,
A. (1996) Role of deferoxamine in tumor therapy. Acta Haematol. 95,
66−69.
(19) Saletta, F., Rahmanto, Y. S., Siafakas, A. R., and Richardson, D.
R. (2011) Cellular iron depletion and the mechanisms involved in the
iron-dependent regulation of the growth arrest and DNA damage
family of genes. J. Biol. Chem. 286, 35396−35406.
(20) Wallace, H. M., Fraser, A. V., and Hughes, A. (2003) A
perspective of polyamine metabolism. Biochem. J. 376, 1−14.
(21) Grillo, M. A., and Colombatto, S. (1994) Polyamine transport in
cells. Biochem. Soc. Trans. 22, 894−898.
(22) Palmer, A. J., and Wallace, H. M. (2010) The polyamine
transport system as a target for anticancer drug development. Amino
Acids 38, 415−422.
(23) Seiler, N. (2003) Thirty years of polyamine-related approaches
to cancer therapy. Retrospect and prospect. Part 2. Structural
analogues and derivatives. Curr. Drug Targets 4, 565−585.
(24) Seiler, N. (2003) Thirty years of polyamine-related approaches
to cancer therapy. Retrospect and prospect. Part 1. Selective enzyme
inhibitors. Curr. Drug Targets 4, 537−564.
(25) Holley, J., Mather, A., Cullis, P., Symons, M. R., Wardman, P.,
Watt, R. A., and Cohen, G. M. (1992) Uptake and cytotoxicity of
novel nitroimidazole-polyamine conjugates in Ehrlich ascites tumour
cells. Biochem. Pharmacol. 43, 763−769.
(26) Verschoyle, R. D., Carthew, P., Holley, J. L., Cullis, P., and
Cohen, G. M. (1994) The comparative toxicity of chlorambucil and
chlorambucil-spermidine conjugate to BALB/c mice. Cancer Lett. 85,
217−222.
(27) Delcros, J. G., Tomasi, S., Carrington, S., Martin, B., Renault, J.,
Blagbrough, I. S., and Uriac, P. (2002) Effect of spermine conjugation
on the cytotoxicity and cellular transport of acridine. J. Med. Chem. 45,
5098−5111.
(28) Phanstiel, O., IV, Kaur, N., and Delcros, J. G. (2007) Structure-
activity investigations of polyamine-anthracene conjugates and their
uptake via the polyamine transporter. Amino Acids 33, 305−313.
(29) Dallavalle, S., Giannini, G., Alloatti, D., Casati, A., Marastoni, E.,
Musso, L., Merlini, L., Morini, G., Penco, S., Pisano, C., Tinelli, S., De
Cesare, M., Beretta, G. L., and Zunino, F. (2006) Synthesis and
ASSOCIATED CONTENT
* Supporting Information
DFMO and exogenous spermidine treatments, stability
constant data, and Cartesian coordinates of optimized
structures. This material is available free of charge via the
Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
■
* David Deniaud, Ph.D.: Phone 00-33-2-51125406, Fax 00-33-
2-51125402, E-mail: david.deniaud@univ-nantes.fr. Franco̧ is
Gaboriau, Ph.D.: Phone 00-33-2-23233861, Fax 00-33-2-
99540137, E-mail: francois.gaboriau@univ-rennes1.fr.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
The authors are grateful to the Conseil Reg
■
́
ional Pays de la
Loire, to the French Ministry of Education, to the Ligue
Nationale contre le Cancer (LNCC, Ille et Vilaine/Loire
Atlantique), to the Association pour la Recherche sur le Cancer
(ARC), and to the Centre National de la Recherche
Scientifique (CNRS) for financial support. The geometry
optimizations and the vibrational calculations reported in this
paper were performed at CCIPL (Centre de Calcul Intensif des
Pays de la Loire, Nantes, France) regional supercomputing
center thanks to a dedicated CPU grant.
REFERENCES
■
(1) Richardson, D. R., Kalinowski, D. S., Lau, S., Jansson, P. J., and
Lovejoy, D. B. (2009) Cancer cell iron metabolism and the
development of potent iron chelators as anti-tumour agents. Biochim.
Biophys. Acta 1790, 702−717.
(2) Kwok, J. C., and Richardson, D. R. (2002) The iron metabolism
of neoplastic cells: alterations that facilitate proliferation? Crit. Rev.
Oncol. Hematol. 42, 65−78.
(3) Brookes, M. J., Hughes, S., Turner, F. E., Reynolds, G., Sharma,
N., Ismail, T., Berx, G., McKie, A. T., Hotchin, N., Anderson, G. J.,
Iqbal, T., and Tselepis, C. (2006) Modulation of iron transport
proteins in human colorectal carcinogenesis. Gut 55, 1449−1460.
(4) Kolberg, M., Strand, K. R., Graff, P., and Andersson, K. K. (2004)
Structure, function, and mechanism of ribonucleotide reductases.
Biochem. Biophys. Acta 1699, 1−34.
(5) Ali, M. A., Akhmedkhanov, A., Zeleniuch-Jaquotte, A., Toniolo,
P., Frenkel, K., and Huang, X. (2003) Reliability of serum iron, ferritin,
nitrite, and association with risk of renal cancer in women. Cancer
Detect. Prev. 27, 116−121.
(6) Huang, X. (2003) Iron overload and its association with cancer
risk in humans: evidence for iron as a carcinogenic metal. Mutat. Res.
533, 153−171.
(7) Thompson, H. J., Kennedy, K., Witt, M., and Juzefyk, J. (1991)
Effect of dietary iron deficiency or excess on the induction of
mammary carcinogenenis by 1-methyl-1-nitrosourea. Carcinogenesis 12,
111−114.
(8) Daniels, T. R., Delgado, T., Rodriguez, J. A., Helguera, G., and
Penichet, M. L. (2006) The transferrin receptor part I: Biology and
targeting with cytotoxic antibodies for the treatment of cancer. Clin.
Immunol. 121, 144−158.
(9) Pinnix, Z. K., Miller, L. D., Wang, W., D’Agostino, R., Jr., Kute,
T., Willingham, M. C., Hatcher, H., Tesfay, L., Sui, G., Di, X., Torti, S.
V., and Torti, F. M. (2010) Ferroportin and iron regulation in breast
cancer progression and prognosis. Sci. Transl. Med. 2, 43−56.
(10) Pogribny, I. P. (2010) Ferroportin and hepcidin: a new hope in
diagnosis, prognosis, and therapy for breast cancer. Breast Cancer Res.
12, 314−315.
1966
dx.doi.org/10.1021/bc300324c | Bioconjugate Chem. 2012, 23, 1952−1968

Bioconjugate Chemistry
Article
cytotoxic activity of polyamine analogues of camptothecin. J. Med.
Chem. 49, 5177−86.
(30) Esteves-Souza, A., Lucio, K. A., Da Cunha, A. S., Da Cunha
Pinto, A., Da Silva Lima, E. L., Camara, C. A., Vargas, M. D., and
Gattass, C. R. (2008) Antitumoral activity of new polyamine-
naphthoquinone conjugates. Oncol. Rep. 20, 225−231.
(31) Tomasi, S., Renault, J., Martin, B., Duhieu, S., Cerec, V., Le
Roch, M., Uriac, P., and Delcros, J. G. (2010) Targeting the polyamine
transport system with benzazepine- and azepine-polyamine conjugates.
J. Med. Chem. 53, 7647−7663.
(46) Byers, T. L., and Pegg, A. E. (1989) Properties and physiological
function of the polyamine transport system. Am. J. Physiol. 257 (Cell.
physiol. 26), C545−C553.
(47) Rodbard, D., and McClean, S. W. (1977) Automated computer
analysis for enzyme-multiplied immunological techniques. Clin. Chem.
23, 112−115.
(48) Swart, M., Groenhof, A. R., Ehlers, A. W., and Lammertsma, K.
(2004) Validation of exchange-correlation functionals for spin states of
iron complexes. J. Phys. Chem. A 25, 5479−5483.
(49) Handy, A. J., and Cohen, N. C. (2001) Left-right correlation
energy. Mol. Phys., 403−412.
(32) Burns, M. R., Graminski, G. F., Weeks, R. S., Chen, Y., and
O’Brien, T. G. (2009) Lipophilic lysine-spermine conjugates are
potent polyamine transport inhibitors for use in combination with a
polyamine biosynthesis inhibitor. J. Med. Chem. 52, 1983−1993.
(33) Covassin, L., Desjardins, M., Soulet, D., Charest-Gaudreault, R.,
Audette, M., and Poulin, R. (2003) Xylylated dimers of putrescine and
polyamines: influence of the polyamine backbone on spermidine
transport inhibition. Bioorg. Med. Chem. Lett. 13, 3267−3271.
(34) Delcros, J. G., Vaultier, M., Roch, N. L., Havouis, R., Moulinoux,
J. P., and Seiler, N. (1997) Bis(7-amino-4-azaheptyl)dimethylsilane: a
new tetramine with polyamine-like features. Effects on cell growth.
Anti-cancer Drug Design 12, 35−48.
(35) Martin, B., Posseme, F., Le Barbier, C., Carreaux, F., Carboni, B.,
Seiler, N., Moulinoux, J. P., and Delcros, J. G. (2002) Z)-1,4-diamino-
2-butene as a vector of boron, fluorine, or iodine for cancer therapy
and imaging: synthesis and biological evaluation. Bioorg. Med. Chem.
10, 2863−2871.
(36) Zhuo, J. C., Cai, J., Soloway, A. H., Barth, R. F., Adams, D. M., Ji,
W., and Tjarks, W. (1999) Synthesis and biological evaluation of
boron-containing polyamines as potential agents for neutron capture
therapy of brian tumors. J. Med. Chem. 42, 1282−1292.
(37) Annereau, J. P., Brel, V., Dumontet, C., Guminski, Y., Imbert, T.,
Broussas, M., Vispe, S., Breand, S., Guilbaud, N., Barret, J. M., and
Bailly, C. (2010) A fluorescent biomarker of the polyamine transport
system to select patients with AML for F14512 treatment. Leuk. Res.
34, 1383−1389.
(38) Wang, C., Delcros, J. G., Biggerstaff, J., and Phanstiel, O., IV
(2003) Synthesis and biological evaluation of N1-(anthracen-9-
ylmethyl)triamines as molecular recognition elements for the poly-
amine transporter. J. Med. Chem. 46, 2663−2671.
(50) Perdew, J. P., Burke, K., and M., E. (1997) Generalized gradient
approximation made simple [phys. rev. lett. 77, 3865 (1996)]. Phys.
Rev. Lett. 7, 1396−1398.
(51) Frisch, M. J., Trucks, G. W., Schlegel, H. B., Scuseria, G. E.,
Robb, M. A., Cheeseman, J. R., Scalmani, G., Barone, V., Mennucci, B.,
Petersson, G. A., Nakatsuji, H., Caricato, M., Li, X., Hratchian, H. P.,
Izmaylov, A. F., Bloino, J., Zheng, G., Sonnenberg, J. L., Hada, M.,
Ehara, M., Toyota, K., Fukuda, R., Hasegawa, J., Ishida, M., Nakajima,
T., Honda, Y., Kitao, O., Nakai, H., Vreven, T., Montgomery, J., J. A.,
Peralta, J. E., Ogliaro, F., Bearpark, M., Heyd, J. J., Brothers, E., Kudin,
K. N., Staroverov, V. N., Kobayashi, R., Normand, J., Raghavachari, K.,
Rendell, A., Burant, J. C., Iyengar, S. S., Tomasi, J., Cossi, M., Rega, N.,
Millam, J. M., Klene, M., Knox, J. E., Cross, J. B., Bakken, V., Adamo,
C., Jaramillo, J., Gomperts, R., Stratmann, R. E., Yazyev, O., Austin, A.
J., Cammi, R., Pomelli, C., Ochterski, J. W., Martin, R. L., Morokuma,
K., Zakrzewski, V. G., Voth, G. A., Salvador, P., Dannenberg, J. J.,
Dapprich, S., Daniels, A. D., Farkas, O., Foresman, J. B., Ortiz, J. V.,
Cioslowski, J., and J., F. D. (2009) Gaussian 09 Revision A.02, Gaussian
Inc., Wallingford, CT.
(52) Kuksa, V., Buchan, R., and Kong Thoo Lin, P. (2000) Synthesis
of polyamines, their derivatives, analogues and conjugates. Synthesis,
1189−1207.
(53) Tian, Z.-y., Xie, S.-q., Mei, Z.-h., Zhao, J., Gao, W.-y., and Wang,
C.-j. (2009) Conjugation of substituted naphthalimides to polyamines
as cytotoxic agents targeting the Akt/mTOR signal pathway. Org.
Biomol. Chem. 7, 4651−4660.
(54) Bergeron, R. J., Bharti, N., Wiegand, J., McManis, J. S., Yao, H.,
and Prokai, L. (2005) Polyamine-vectored iron chelators: the role of
charge. J. Med. Chem. 48, 4120−4137.
(55) Kaur, N., Delcros, J.-G., Archer, J., Weagraff, N. Z., Martin, B. n.
d., and Phanstiel, O., Iv (2008) Designing the polyamine
pharmacophore: influence of N-substituents on the transport behavior
of polyamine conjugates. J. Med. Chem. 51, 2551−2560.
(56) Gardner, R. A., Belting, M., Svensson, K., and Phanstiel. (2007)
Synthesis and transfection efficiencies of new lipophilic polyamines. J.
Med. Chem. 50, 308−318.
(57) Xie, S., Cheng, P., Liu, G., Ma, Y., Zhao, J., Chehtane, M.,
Khaled, A. R., Phanstiel, Iv, O., and Wang, C. (2007) Synthesis and
bioevaluation of N-(arylalkyl)-homospermidine conjugates. Bioorg.
Med. Chem. Lett. 17, 4471−4475.
(58) Chadwick, J., Jones, M., Mercer, A. E., Stocks, P. A., Ward, S. A.,
Park, B. K., and O’Neill, P. M. (2010) Design, synthesis and
antimalarial/anticancer evaluation of spermidine linked artemisinin
conjugates designed to exploit polyamine transporters in Plasmodium
falciparum and HL-60 cancer cell lines. Bioorg. Med. Chem. 18, 2586−
2597.
(39) Wang, C., Delcros, J.-G., Biggerstaff, J., and Phanstiel, O., IV
(2003) Molecular requirements for targeting the polyamine transport
system. synthesis and biological evaluation of polyamine anthracene
conjugates. J. Med. Chem. 46, 2672−2682.
(40) Kruczynski, A., Vandenberghe, I., Pillon, A., Pesnel, S., Goetsch,
L., Barret, J. M., Guminski, Y., Le Pape, A., Imbert, T., Bailly, C., and
Guilbaud, N. (2011) Preclinical activity of F14512, designed to target
tumors expressing an active polyamine transport system. Invest. New
Drugs 29, 9−21.
(41) Bergeron, R. J., Singh, S., Bharti, N., and Jiang, Y. (2010)
Design, synthesis, and testing of polyamine vectored iron chelators.
Synthesis, 3631−3636.
́
(42) Baret, P., Beguin, C. G., Boukhalfa, H., Caris, C., Laulhere, J. P.,
Pierre, J. L., and Serratrice, G. (1995) O-TRENSOX: A promizing
water-soluble iron chelator (both FeIII and FeII) potentially suitable
for plant nutrition and iron chelation therapy. J. Am. Chem. Soc. 117,
9760−9761.
(59) de Medina, P., Paillasse, M. R., Payre, B., Silvente-Poirot, S., and
Poirot, M. (2009) Synthesis of new alkylaminooxysterols with potent
cell differentiating activities: identification of leads for the treatment of
cancer and neurodegenerative diseases. J. Med. Chem. 52, 7765−7777.
(60) Jotterand, N., Pearce, D. A., and Imperiali, B. (2001)
Asymmetric synthesis of a new 8-hydroxyquinoline-derived alpha-
amino acid and its incorporation in a peptidylsensor for divalent zinc.
J. Org. Chem. 66, 3224−3228.
(61) Abdel-Magid, A. F., Carson, K. G., Harris, B. D., Maryanoff, C.
A., and Shah, R. D. (1996) Reductive amination of aldehydes and
ketones with sodium triacetoxyborohydride. studies on direct and
indirect reductive amination procedures. J. Org. Chem. 61, 3849−3862.
(43) Du, T., Filiz, G., Caragounis, A., Crouch, P. J., and White, A. R.
(2008) Clioquinol promotes cancer cell toxicity through tumor
necrosis factor alpha release from macrophages. J. Pharmacol. Exp.
Ther. 324, 360−367.
(44) Ding, W. Q., Liu, B., Vaught, J. L., Palmiter, R. D., and Lind, S.
E. (2006) Clioquinol and docosahexaenoic acid act synergistically to
kill tumor cells. Mol. Cancer Ther. 5, 1864−1872.
(45) Wang, J., Xie, S., Li, Y., Guo, Y., Ma, Y., Zhao, J., Phanstiel, O.,
IV, and Wang, C. (2008) Synthesis and evaluation of unsymmetrical
polyamine derivatives as antitumor agents. Bioorg. Med. Chem. 16,
7005−7012.
1967
dx.doi.org/10.1021/bc300324c | Bioconjugate Chem. 2012, 23, 1952−1968

Bioconjugate Chemistry
Article
(62) Byers, T. L., Wechter, R., Nuttall, M. E., and Pegg, A. E. (1989)
Expression of a human gene for polyamine transport in Chinese-
hamster ovary cells. Biochem. J. 263, 745−752.
́ ́
(63) Lescoat, G., Leonce, S., Pierre, A., Gouffier, L., and Gaboriau, F.
(2012) Antiproliferative and iron chelating efficiency of the new bis-8-
hydroxyquinoline benzylamine chelator S1 in hepatocyte cultures.
Chem.-Biol. Interact. 195, 165−172.
(64) Rodriguez-Lucena, D., Gaboriau, F., Rivault, F., Schalk, I. J.,
Lescoat, G., and Mislin, G. L. A. (2010) Synthesis and biological
properties of iron chelators based on a bis-2-(2-hydroxy-phenyl)-
thiazole-4-carboxamide or -thiocarboxamide (BHPTC) scaffold.
Bioorg. Med. Chem. 18, 689−695.
(65) Thomas, F., Serratrice, G., Beg
́
uin, C., Saint Aman, E., Pierre, J.
L., Fontecave, M., and Laulhere, J. P. (1999) Calcein as a fluorescent
̀
probe for ferric iron. J. Biol. Chem. 274, 13375−13383.
(66) Lovaas, E. (1997) Antioxidative and metal-chelating effects of
polyamines. Adv. Pharmacol., 119−149.
(67) Palmer, B. N., and Powell, H. K. J. (1974) Polyamine complexes
with seven-membered chelate rings: Complexe formation of 3-
azaheptane-1,7-diamine, 4-azaoctane-1,8-diamine (spermidine), and
4,9-diazadodecane-1,12-diamine (spermine) with copper(II) and
hydrogen ions in aqueous solution. J. Chem. Soc., Dalton Trans.,
2089−2092.
(68) Bardez, E., Devol, I., Larrey, B., and Valeur, B. (1997) Excited-
state processes in 8-hydroxyquinoline: photoinduced tautomerization
and solvation effects. J. Phys. Chem. B 101, 7786−7793.
(69) Kimberly, M. M., and Goldstein, J. H. (1981) Determination of
pKa values and total proton distribution pattern of spermidine by
carbon-13 nuclear magnetic resonance titrations. Anal. Chem. 53, 789−
793.
(70) Shchavlev, A. E., Pankratov, A. N., and Shalabay, A. V. (2006)
Dft computational studies on rotation barriers, tautomerism, intra-
molecular hydrogen bond, and solvent effects in 8-hydroxyquinoline.
Int. J. Quantum Chem. 4, 876−886.
(71) Beattie, J. K., Best, S. P., Skelton, B. W., and White, A. H. (1981)
Structural studies on the caesium alums, CsMIII[SO4]2•12H2O. J.
Chem. Soc., Dalton Trans., 2105−2111.
(72) Notni, J., Pohle, K., Peters, J. A., Gorls, H., and Platas-Iglesias, C.
̈
(2009) Structural study of Ga(III), In(III), and Fe(III) complexes of
triaza-macrocycle based ligands with N3S3 donor set. Inorg. Chem. 7,
3257−3267.
1968
dx.doi.org/10.1021/bc300324c | Bioconjugate Chem. 2012, 23, 1952−1968