Multifunctionalization of magnetic nanoparticles for controlled drug release: A general approach

Article abstract of DOI:10.1016/j.ejmech.2014.05.078

In this study, a general approach for the multifunctionalization of magnetic nanoparticles (MNPs) with drugs (Doxorubicin and Gemcitabine) and targeting moieties (Nucant pseudopeptide) for controlled and selective release is described. The functionalizati

Full text of DOI:10.1016/j.ejmech.2014.05.078

European Journal of Medicinal Chemistry 82 (2014) 355e362
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Multifunctionalization of magnetic nanoparticles for controlled drug
release: A general approach
Alfonso Latorre 1, Pierre Couleaud 1, Antonio Aires, Aitziber L. Cortajarena*,
*
ꢀ
Alvaro Somoza
~
Instituto Madrileno de Estudios Avanzados en Nanociencia & CNB-CSIC-IMDEA Nanociencia Associated Unit, Cantoblanco, Madrid, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
In this study, a general approach for the multifunctionalization of magnetic nanoparticles (MNPs) with
drugs (Doxorubicin and Gemcitabine) and targeting moieties (Nucant pseudopeptide) for controlled and
selective release is described. The functionalization is achieved by the formation of disulfide bonds
between MNPs and drugs derivatives synthesized in this work. Our strategy consists in the introduction
of a pyridyldisulfide moiety to the drugs that react efficiently with sulfhydryl groups of pre-activated
MNPs. This approach also allows the quantification of the covalently immobilized drug by measuring
the amount of the 2-pyridinethione released during the process. The linkers developed here allow the
release of drugs without any chemical modification. This process is triggered under highly reducing
environment, such as that present inside the cells.
Complete release of drugs is achieved within 5e8 h under intracellular conditions whereas negligible
percentage of release is observed in extracellular conditions.
We propose here a modular general approach for the functionalization of nanoparticles that can be
used for different types of drugs and targeting agents.
Received 13 March 2014
Received in revised form
30 May 2014
Accepted 31 May 2014
Available online 2 June 2014
Keywords:
Magnetic nanoparticles
Multifunctionalization
Controlled drug release
Anticancer therapy
Disulfide bond
Nanomedicine
© 2014 Elsevier Masson SAS. All rights reserved.
1. Introduction
colloidal stability in physiological conditions; (3) easy to load with
known amounts of therapeutic agents and targeting molecules and
The use of nanoparticles as carrier systems for therapeutic
molecules has been explored during the last 20 years [1] aiming to
improve the therapeutic effect of the drugs and their administra-
tion, as well as, to reduce their side effects. Different types of
nanostructures (metallic nanoparticles, polymeric coreeshell
nanoparticles or micelles) have been evaluated particularly in
cancer therapy, whose vast side effects compromise the health of
patients. In this regard, nanoparticles can be designed as multi-
functional platforms that can be loaded with several drugs and also
modified with targeting molecules to direct them to cancer cells [2].
Targeted therapies have been developed to diminish side effects of
current approaches [3] where specific cell membrane receptors
overexpressed by cancer cells are used as targets, improving the
efficacy of classical treatments [4,5].
(4) able to release the cargo efficiently inside the cells. Therefore,
the development of strategies for the functionalization of nano-
particles is a crucial point for the future clinical use to improve
anticancer therapies.
In this study, immobilization strategies for the functionalization
of magnetic nanoparticles (MNP) using tailored linkers have been
developed. The systems described have the required properties to
be successfully employed in biomedical applications.
In particular, dimercaptosuccinic acid coated magnetic nano-
particles [6,7] (DMSA-MNPs) have been functionalized with two
chemotherapeutic drugs and a targeting molecule. The controlled
release of the drugs has been evaluated. One of the drugs employed
is Doxorubicin (DOX), which is widely used to treat a broad spec-
trum of cancers (breast, stomach, non-Hodgkin's lymphoma, and
bladder cancer) [8]. One advantage of DOX is its strong visible ab-
sorption and fluorescence emission that makes it easy to monitor
during the different steps of the functionalization strategy. The
other chemotherapeutic drug used is Gemcitabine (GEM) that is
employed in several cancers such as pancreatic cancer [9e11], non-
small cell lung cancer, bladder cancer, soft-tissue sarcoma, meta-
static breast cancer and ovarian cancer and acts as an antineoplastic
Nanoparticles for medical applications, and particularly for
targeted cancer therapy, must be (1) non-toxic (2) with a good
* Corresponding authors.
E-mail addresses: aitziber.lopezcortajarena@imdea.org (A.L. Cortajarena), alvaro.
ꢀ
somoza@imdea.org (A. Somoza).
1
Contributed equally to this work.
http://dx.doi.org/10.1016/j.ejmech.2014.05.078
0223-5234/© 2014 Elsevier Masson SAS. All rights reserved.

356
A. Latorre et al. / European Journal of Medicinal Chemistry 82 (2014) 355e362
agent. Both drugs are cell-cycle specific therapeutic agents and
therefore need to reach the nucleus. DOX acts by intercalating DNA
whereas GEM replaces cytidine during DNA replication. Despite
positive results with these drugs in clinics, classical chemotherapy
still presents several problems. For example, doxorubicin has
shown great efficacy in both solid and liquid tumors, but the
emergence of drug resistance and several side effects such as heart
muscle damage are important limitations for successful cancer
treatment [12]. Gemcitabine undergoes rapid deamination into the
inactive uracil derivative, resulting in a short half-life. Also in the
case of Gemcitabine, drug resistance has been observed in in vitro
and in vivo preclinical models, which is critical for future clinical
usage [13]. The use of nanoparticle-based delivery systems has
been shown to overcome multidrug resistance in tumors and to
increase the stability immobilized molecules [14,15]. Therefore,
more efficient nanoparticle-based formulations that additionally
incorporate targeting for local administration are needed. In this
sense, to introduce targeting capabilities in the multifunctional
formulations the pseudopeptide Nucant (N6L) has been used
together with DOX and GEM. The Nucant pseudopeptide, which
acts both as an anticancer drug and as a targeting agent is nowa-
days in clinical trials. N6L binds nucleolin, which is a protein
overexpressed in the membrane of cancer cells, and nucleo-
phosmin [16,17], and can enter the cell nucleus to induce apoptosis
[18].
molecules linked to the nanoparticles by following the UVeVisible
absorption of the byproduct released during the functionalization
reaction. Finally, the linkers have been designed in order to release
the molecule without any chemical modification, after an internal
rearrangement [32,33]. Therefore, the initial molecule will be
released within the cell and its activity will not be affected.
The work presented here is a successful example of efficient
drug's covalent functionalization of nanoparticles for an intracel-
lular controlled release. The method herein presented could be
applied to other drugs (charged or neutral, fluorescent or not) and
targeting agents (peptide, antibodies, etc.) for multiple biomedical
applications.
2. Experimental
2.1. Materials
All reagents were purchased from Aldrich and used without
ꢀ
further purification. Jose Courty's group from CRRET-CNRS labora-
tory provided cysteine modified Nucant pseudopeptide (N6L-Cys).
Ultrapure reagent grade water (18.2 MU, Wasserlab) was used in all
experiments. DMSA-MNPs have been provided by Dr. Gorka Salas'
group at IMDEA Nanociencia.
2.2. Measurements
As a first approach to improve cancer therapy non-covalent
functionalization of nanoparticles, mainly through electrostatic or
hydrophobic interactions [19,20] has been explored intensively in
the past due to the ease of application, but has some concerns and
drawbacks. The most important limitation is the poor control on
the release of the drug immobilized onto such nanostructures. The
drug release occurs as a passive process based on the high con-
centration of salts and biomolecules in vivo or on pH changes. This
strategy is suitable for in vitro assays or in vivo assays using intra-
tumoral injection but it should not be applied intravenously.
Another drawback is the fact that neutral drugs under physiological
conditions, such as Gemcitabine cannot be immobilized
electrostatically.
Taking into account all these parameters, the covalent func-
tionalization of anticancer drugs onto targeted nanoparticles seems
to be a suitable way to tackle these problems. Ideally, only cancer
cells should be targeted by drug-loaded magnetic nanoparticles,
where anti-tumoral agents are inactivated until they are released
inside the cell. In this regard, different linkers sensitive to certain
intracellular triggering stimulus such as pH [21,22] and the pres-
ence of some enzymes [23e25], or external stimuli such as tem-
perature [26,27] have been employed to connect and release drugs
from magnetic nanoparticles in a controlled manner. On the other
hand, disulfide bond based linkers have excellent properties for this
application [28] because allow the formation of a covalent bond
between the nanoparticle and the required molecule. Then, the
disulfide bond can be broken by specific reducing agent such as
endogenous glutathione (GSH). It is well known that the intracel-
lular level of GSH in cells is in the millimolar range (0.5e10 mM),
whereas just micromolar concentrations are typically found in
blood plasma and the extracellular medium [29]. Moreover, it has
been shown by clinical studies that tumor tissue is often higher in
glutathione content than normal tissue [30,31]. By designing the
functionalization with a disulfide linker, the attached molecule will
be released only under highly reducing environment such as the
tumor cells' intracellular environment. Another advantage of the
strategy described here relies on the design of a modification
attached to the molecules in order to quantify the immobilization.
Thin layer chromatography (TLC) was carried out using Silica Gel
60 F254 plates. Column chromatography was performed using
Silica Gel (60 Å, 230 ꢀ 400 mesh). All NMR spectra were recorded
on a Bruker instrument (MHz indicated in brackets) as solutions in
CDCl3, and the chemical shifts are reported in parts per million
(ppm). Coupling constants are reported in hertz (Hz). MALDI-TOF
mass spectrometer analysis was performed using a Voyager DE
Pro (AB Applied Biosystems). UVeVis and fluorescence spectra were
recorded on a Synergy H4 microplate reader (BioTek) using 96-well
plates. Hydrodynamic diameter and zeta potential measurements
were determined using a Zetasizer Nano-ZS device (Malvern In-
struments). Hydrodynamic diameter and zeta potential were
measured from dilute sample suspensions in water at pH 7.4 using a
zeta potential cell. HPLC: Agilent Technologies, 1260 Infinity. Col-
umn ZORBAX 300SB-C18, 5
mm, 9.4 ꢀ 250 mm.
2.3. Synthesis and characterization
2.3.1. Synthesis of intermediates and drugs derivatives
2.3.1.1. Synthesis of 2-(pyridin-2-yldisulfanyl)ethanol (1) [32].
To a solution of aldrithiol (300 mg, 1.36 mmol) in MeOH (1.5 mL)
under Ar, 2-mercaptoethanol (53 mL, 0.75 mmol) was added slowly
and stirred for 16 h. Then, the solvent was evaporated in vacuum
and the residue purified by flash chromatography (CH2Cl2/AcOEt
5:1) to obtain compound 1 (Fig. 1) as a colorless oil in 86% yield; 1H
NMR (300 MHz, CDCl3)
d
8.51 (d, J ¼ 4.3 Hz, 1H), 7.58 (td, J ¼ 8.0,
1.7 Hz, 1H), 7.40 (d, J ¼ 8.0 Hz, 1H), 7.17e7.13 (m, 1H), 5.72 (bs, 1H),
3.80 (dd, J ¼ 10.4, 6.5 Hz, 2H), 2.97e2.94 (m, 2H).
2.3.1.2. Synthesis of 4-nitrophenyl 2-(pyridin-2-yldisulfanyl)ethyl
carbonate (2) [32]. To
0.53 mmol), and bis(4-nitrophenyl) carbonate (241 mg, 0.79 mmol)
in CH2Cl2 (2 mL) under Ar, DIPEA (158 L, 0.79 mmol) was added
a solution of compound 1 (100 mg,
m
and stirred for 5 h. The mixture was washed with water, and the
organic phase dried with MgSO4. After solvent evaporation, the
residue was purified by flash chromatography (Hexane/AcOEt 4:1
and then 2:1) to obtain compound 2 (Fig. 1) as a colorless oil 67%
By anchoring
a linker ending with a thiol activated by 2-
mercaptopyridyl group, it is possible to quantify the amount of
yield; 1H NMR (300 MHz, CDCl3)
d
8.50 (d, J ¼ 4.8 Hz, 1H), 8.28 (d,

A. Latorre et al. / European Journal of Medicinal Chemistry 82 (2014) 355e362
357
Fig. 1. General scheme of synthesis of drugs derivatives.
J ¼ 9.1 Hz, 2H), 7.72e7.59 (m, 2H), 7.38 (d, J ¼ 9.1 Hz, 2H), 7.15e7.10
(m, 1H), 4.57 (t, J ¼ 6.4 Hz, 2H), 3.16 (t, J ¼ 6.4 Hz, 2H).
After the reaction was finished, a NAP-10 column was used to
separate 5 from the excess of 2,20-dipyridyldisulfide and DMSO.
Nucant-S-S-Pyr (5) was collected in the first fractions, obtained
with a yield of 70% and stored at ꢁ20 ꢂC. Quantification of activated
thiol moieties of the Nucant-S-S-Pyr was achieved by adding DDT
on an aliquot and measuring the band at 343 nm due to 2-
pyridinthione released (ε343nm ¼ 8080 L molꢁ1 cmꢁ1). MS (ESI):
m/z (%), (Mþ) found 4237.56 (100), calculated for C190H354N71O34S2
(Mþ) 4238.8.
2.3.1.3. Synthesis of doxorubicin derivative, DOX-S-S-Pyr (3) [33].
To a solution of compound 2 (10 mg, 0.028 mmol) and doxorubicin
hydrochloride (12 mg, 0.020 mmol) in DMF (1 mL) under N2, DIPEA
(8 mL, 0.028 mmol) was added at room temperature and stirred for
16 h. Then, the solvent was evaporated and the residue was purified
by flash chromatography (eluent: CH2Cl2/MeOH 20:1) to obtain
compound 3 (Fig. 1) as red solid in 90% yield; 1H NMR (500 MHz,
2.3.2. Covalent attachment of drug derivatives to thiolated DMSA-
MNPs
2.3.2.1. Pre-activation of DMSA-MNPs. DMSA-MNPs were first
modified with cysteamine hydrochloride to introduce thiol moi-
eties (thiolated DMSA-MNPs). To 1 mL of DMSA-MNPs at 2.4 mg Fe/
CDCl3)
d
13.94 (s, 1H), 13.18 (s, 1H), 8.40 (d, J ¼ 3.8 Hz, 1H), 8.00 (d,
J ¼ 7.6 Hz, 1H), 7.76 (t, J ¼ 8.0 Hz, 1H), 7.70 (d, J ¼ 7.5 Hz, 1H), 7.63 (t,
J ¼ 7.4 Hz, 1H), 7.37 (d, J ¼ 8.5 Hz, 1H), 7.07 (t, J ¼ 1H), 5.50 b (s, 1H),
5.26 (dd, J ¼ 3.7, 2.0 Hz, 1H), 5.15 (d, J ¼ 8.8 Hz, 1H), 4.74 b (s,
J ¼ 21.4 Hz, 2H), 4.56 (bs,1H), 4.42e4.34 (m, 1H), 4.21e4.08 (m, 2H),
4.06 (s, 3H), 3.81 (m,1H), 3.62 (bs,1H), 3.22 (dd, J ¼ 18.8,1.5 Hz,1H),
3.14e2.86 (m, 5H), 2.33 (d, J ¼ 14.6 Hz,1H), 2.15 (dd, J ¼ 15.0, 3.6 Hz,
1H), 1.81 (m, 2H), 1.48e1.42 (m, 2H), 1.29 (d, J ¼ 6.5 Hz, 3H); 13C
mL were added 50
ously neutralized by 1 equivalent of NaOH, 150
and 75 mol of NHS/g Fe. After 16 h, the sample was washed by
mmol of cysteamine hydrochloride/g Fe, previ-
mmol of EDC/g Fe
m
cycles of centrifugation and redispersion in milliQ water at least 3
times.
Thiolated DMSA-MNPs were used for the functionalization with
the different molecules (DOX, GEM and N6L).
NMR (126 MHz, CDCl3) d 213.9,187.0,186.6,161.0,160.1,156.2,155.6,
155.2, 149.8, 149.4, 137.3, 135.7, 135.4, 133.6, 133.6, 120.9, 120.8,
119.9, 119.8, 118.4, 111.5, 111.4, 100.9, 69.7, 69.0, 67.4, 65.6, 63.5, 56.7,
53.4, 47.0, 37.5, 35.6, 33.9, 30.0, 29.7, 28.3, 17.0; MS (ESI): m/z (%)757
(100).
2.3.2.2. Reaction between thiolated DMSA-MNPs and Doxorubicin
derivative (3). 1 mL of aqueous suspension of thiolated DMSA-
2.3.1.4. Synthesis of Gemcitabine derivative, GEM-S-S-Pyr (4).
To a solution of compound 2 (25 mg, 0.071 mmol) and Gemcitabine
chlorhydrate (24 mg, 0.09 mmol) in DMF (1.5 mL) under Ar, DIPEA
MNPs at 2.4 mg Fe/mL was mixed with 240
solution at 500 M in DMF (0.012
mol) during 16 h at 37 ꢂC. After
this time, 20 L of brine were added and the sample centrifuged
mL of DOX-S-S-Pyr (3)
m
m
m
(18 mL, 0.09 mmol) and DMAP (catalytic amount) were added and
10 min at 5000 ꢀ g. From the collected supernatants, the covalently
immobilized DOX onto thiolated DMSA-MNPs was determined by
quantification of the 2-pyridinethione released (lmax ¼ 343 nm,
stirred for 16 h. Then, the solvent was evaporated in vacuum and
the residue purified by flash chromatography (CH2Cl2/MeOH 15:1)
to obtain compound 4 (Fig.1) in 38% yield as a colorless oil; 1H NMR
ε343
¼ 8080 L molꢁ1 cmꢁ1, Fig. S.1). Finally the sample was
nm
(500 MHz, CDCl3)
d 8.44e8.43 (m, 1H), 7.70e7.64 (m, 2H), 7.52 (d,
redispersed in 1 mL of MilliQ water.
J ¼ 6.8 Hz, 1H), 7.11 (ddd, J ¼ 6.7, 4.9, 1.5 Hz, 1H), 6.23 (bs, 2H), 5.80
(d, J ¼ 7.5 Hz, 1H), 5.33 (bs, 1H), 4.50e4.38 (m, 3H), 4.13 (d,
J ¼ 7.5 Hz, 1H), 4.04 (d, J ¼ 12.0 Hz, 1H), 3.84 (d, J ¼ 12.0 Hz, 1H),
2.3.2.3. Reaction between thiolated DMSA-MNPs and Gemcitabine
derivative (4). 1 mL of aqueous suspension of thiolated DMSA-
3.13e3.02 (m, 2H); 13C NMR (126 MHz, CDCl3)
d 165.8, 159.2, 155.6,
MNPs at 2.4 mg Fe/mL was mixed with 240
(4) solution at 500 M in DMF (0.012
mol) during 16 h at 37 ꢂC.
After reaction, 20 L of brine were added and the sample centri-
mL of GEM-S-S-Pyr
153.5, 149.8, 149.6, 137.4, 121.1, 120.0, 95.7, 78.7, 78.7, 72.7, 66.8,
65.4, 59.6, 36.5; MS (ESI): m/z (%)239 (23), 477 (MþþH, 100), 499
(MþþNa, 2); HRMS (ESI) calculated for C17H19F2N4O6S2 (MþþH)
477.0708, found 477.0708; HRMS (ESI) calculated for
m
m
m
fuged 10 min at 5000 ꢀ g. From the collected supernatants, the
covalently immobilized GEM onto thiolated DMSA-MNPs was
determined by quantification of the 2-pyridinethione released
C
17H18N4O6F2NaS2 (MþþNa) 499.0547, found 499.0528.
(
lmax ¼ 343 nm, ε343 nm ¼ 8080 L molꢁ1 cmꢁ1, Fig. S.2). Finally, the
sample was redispersed in 1 mL of MilliQ water.
2.3.1.5. Synthesis of Nucant-S-S-Pyr (5). To 1.2 mL of Nucant-Cys
(600 M) 400
L of 2,20-dipyridyldisulfide (30 mM in DMSO)
m
m
were added and incubated during 4 h at room temperature (RT).
The progress of the reaction was followed measuring the formation
2.3.2.4. Synthesis of MNP-N6L conjugate. 1 mL of aqueous suspen-
sion of thiolated DMSA-MNPs at 2.4 mg Fe/mL was mixed with
of 2-pyridinethione by UV absorption at 343 nm (ε343
¼
84
mL of Nucant-S-S-Pyr (5) at 200 mM in water (0,0168 mmol)
nm
8080 L molꢁ1 cmꢁ1).
during 16 h at RT. The reaction mixture was centrifuged and

358
A. Latorre et al. / European Journal of Medicinal Chemistry 82 (2014) 355e362
Fig. 2. General scheme of functionalization of DMSA-MNPs.
washed with brine to eliminate electrostatically immobilized
Nucant and then with water. From the collected supernatant the
covalently immobilized Nucant onto DMSA-MNPs was determined
by quantification of the 2-pyridinethione released (lmax ¼ 343 nm,
ε343 nm ¼ 8080 L molꢁ1 cmꢁ1, Fig. S.3).
analyzed by HPLC using a C-18 column, mobile phase water/
acetonitrile 80/20, at flow rate of 0.3 mL/min, measuring the
absorbance at 270 nm (Fig. S.6). The amount of released N6L was
analyzed using Bradford's method [34] by measuring the absor-
bance of the sample at 595 nm with the UVeVisible spectropho-
tometer (Fig. S.7).
The yield of immobilization was 70% (5 mmol N6L/g Fe). The zeta
potential of the sample was ꢁ41.9 3.3 mV at pH 7.4 with a hy-
drodynamic diameter of 102.2 0.4 nm.
3. Results and discussion
2.3.2.5. Bi-functionalization of thiolated DMSA-MNPs (MNP-DOX-
N6L and MNP-GEM-N6L). 1 mL of aqueous suspension of MNP-DOX
The use of disulfide bonds for covalent attachment and triggered
intracellular delivery of the drugs relies on the following strategy.
First, a modification of the drug to introduce pyridyldisulfide group
is required to promote the reaction with pre-thiolated DMSA-
MNPs. At the same time, the pyridyl group in the activated mole-
cules (Fig. 1) will allow to quantify the covalently linked drug by
measuring the pyridine-2-thione released during the coupling re-
action with MNPs (Fig. 2). In addition, the presence of disulfide
bonds between DMSA-MNPs and drugs will permit the controlled
intracellular release of the drug under intracellular reducing con-
ditions. What is more, the linker design allows to release the drug
without any chemical modification, since the free thiol generated
upon disulfide cleavage is able to attack the carbonate moiety
favored by the formation of a five member ring, leaving the drug
unaltered (Fig. 3). This kind of autoimmolative-linker has been used
previously conjugated with targeting molecules [24], but as far as
we know it has not been used to immobilize drugs onto the MNPs
surface.
at 2.4 mg Fe/mL was mixed with 84
at 200 M in water (0.0168 mol) during 16 h at RT. After reaction,
20 L of brine were added and the sample was centrifuged 10 min
mL of N6L-S-S-Pyr (5) solution
m
m
m
at 5000 ꢀ g three times to eliminate any electrostatically bound
N6L. UVeVis absorption of supernatants was checked to quantify
the N6L covalently bound as described in 2.3.2.4. The same protocol
was employed for the synthesis of MNP-GEM-N6L.
2.3.3. In vitro drugs release studies
The cumulative drugs release experiments were carried out
using two different conditions in order to evaluate the stim-
ulieresponse behavior of functionalized DMSA-MNPs toward
reducing environment. The release of drugs, from the functional-
ized DMSA-MNPs was carried out under physiological conditions
(pH 7.4 and 37 ꢂC) using two different concentrations of reducing
agent (1 mM and 1 mM of 1,4-Dithiothreitol (DTT) to mimic the
extracellular and intracellular conditions, respectively). For each
experiment, 2.4 mg of functionalized MNPs (or 4.8 mg in the case of
the MNP-GEM and MNP-DOX) were dissolved in 1 mL of 10 mM
3.1. Synthesis of drugs derivatives
phosphate buffer at pH 7.4 containing 1
m
M of DTT, or 10 mM
The synthesis of drug derivatives was achieved in three steps.
Starting from commercially available aldrithiol, 2-mercaptoethanol
was added to give rise to compound 1 by a simple disulfide ex-
change reaction. Then, the primary alcohol was converted in the
electrophilic carbonate compound 2 by the reaction with bis(4-
nitrophenyl) carbonate (Fig. 1). Finally, the corresponding chemo-
therapeutic drugs were incorporated to this linker through the
carbonate moiety by the nucleophilic substitution of the 4-
phosphate buffer pH 7.4 containing 1 mM DTT and incubated at
37 ꢂC. The amount of each drug released was determined by
different methods at regular time intervals. The percentage of drug
released was calculated from a standard calibration curve of free
drug solution. The amount of DOX released was analyzed by
measuring the absorbance of the sample at 495 nm with UVevis
spectrophotometer (Fig. S.5). The amount of released GEM was

A. Latorre et al. / European Journal of Medicinal Chemistry 82 (2014) 355e362
359
Fig. 3. Scheme of drug release strategy in presence of reducing agent (DTT).
nitrophenol leaving group. Nucant derivative (5) was obtained by
reaction of cysteine modified Nucant with aldrithiol.
3.2.3. Covalent attachment of Nucant pseudopeptide on pre-
activated DMSA-MNPs
The functionalization was achieved by the formation of disulfide
bonds between the reactive thiols of the modified DMSA-MNPs and
the thiol-activated Nucant pseudopeptide. The analysis of the 2-
pyridinethione released during the reaction allowed the quantifi-
cation of the covalently bound N6L (Fig. S.3). The standard load
3.2. Functionalization of DMSA-MNPs
3.2.1. Pre-activation of DMSA-MNPs by cysteamine
DMSA coated nanoparticles have thiol groups available from the
DMSA. However, we have observed that the amount of these
groups on the particles surface decreases over time and at different
pHs [35]. Therefore, the DMSA-MNPs employed in this work have
been stabilized by basic treatment in order to favor disulfide bonds
formation between DMSA molecules of the coating [5]. Then, to
assure a fixed and controlled amount of free thiol functions on
DMSA-MNPs, an activation step was performed before the func-
tionalization with the desired molecules. The introduction of thiol
groups on the surface of DMSA-MNPs was achieved by reaction
between cysteamine hydrochloride and carboxylate groups of
DMSA-MNPs in presence of EDC/NHS in basic medium (Fig. 2).
obtained of covalently linked N6L was 5 mmol N6L/g Fe, which is
lower than in the cases of DOX and GEM. This difference could be
due to the size of Nucant, which is almost 10 times larger than DOX
and GEM as well as to the positive charge displayed at physiological
pH. Indeed, the addition of high amounts of activated N6L neu-
tralizes all the negative charges of the DMSA coating leading to the
aggregation of the MNPs. We have observed that particles with
5
mmol N6L/g Fe are stable at pH 7.4 with a zeta potential
of ꢁ41.9 3.3 mV and a hydrodynamic diameter of 102.2 0.4 nm.
3.2.4. Bi-functionalization of DMSA-MNPs with drug (DOX or GEM)
and N6L (MNP-DOX-N6L and MNP-GEM-N6L)
Particularly, the reaction of 50
mmol of cysteamine/g Fe led to a
The last step of this study is the preparation of bi-functionalized
DMSA-MNPs in order to provide a targeting moiety, Nucant pseu-
dopeptide, and an anticancer drug, GEM or DOX. The general
strategy began with the immobilization of the drug (DOX or GEM)
as described previously. Once the drugs have been conjugated there
are still enough thiol moieties available in the MNP to introduce the
N6L. This order of functionalization is also driven by the fact that
the N6L is larger than DOX and GEM and might cause steric hin-
drance if placed first on the MNPs surface. The conjugation yield of
maximum of 30 mol of reactive thiol groups/g Fe. The pre-
activated DMSA-MNPs are stable at physiological pH. DMSA-
m
MNPs thiolated present a zeta potential of ꢁ54.5 2.1 mV and a
hydrodynamic diameter of 59.0
1.8 nm whereas DMSA-MNPs
present a zeta potential of ꢁ60.1 2.1 mV and a hydrodynamic
diameter of 53.1 0.9 nm.
3.2.2. Covalent attachment of Doxorubicin and Gemcitabine on pre-
activated DMSA-MNPs
The reaction of modified DOX and GEM with pre-activated
DMSA-MNPs led to stable colloidal formulations of functionalized
N6L was 70% (5 mmol N6L/g Fe) in particles loaded with DOX or
GEM. Zeta potentials and hydrodynamic diameters obtained for
each formulation are showed in Table 2.
DMSA-MNPs with 32
mmol DOX/g Fe and 30 mmol GEM/g Fe,
respectively. As discussed previously, a decisive parameter is the
colloidal stability in physiological medium. In this case, a slight
difference was observed in the zeta potential and the hydrody-
namic diameter of these particles compared with DMSA-MNPs. The
immobilization yields were 64% for DOX and 60% for GEM with
loads of 32 mmol DOX/g Fe and 30 mmol GEM/g Fe, respectively. Zeta
potentials and hydrodynamic diameters obtained for each formu-
lation are showed in Table 1.
3.3. In vitro drugs release studies
In this work, we studied the release of drugs in different media
and concentrations of reducing agent for the different types of
functionalized DMSA-MNPs (mono- and bi-functionalized). As
Nucant pseudopeptide acts as both targeting moiety and cytotoxic
agent, release studies have been carried out with DMSA-MNPs
Table 1
Characterization of mono-functionalized DMSA-MNPs.
Table 2
Characterization of bi-functionalized DMSA-MNPs.
Sample
Zeta potential (mV)
Hydrodynamic diameter (nm)
Sample
Zeta potential (mV)
Hydrodynamic diameter (nm)
DMSA-MNPs
MNP-DOX
MNP-GEM
ꢁ60.1 2.1 mV
ꢁ45.4 0.7 mV
ꢁ43.0 1.1 mV
53.1 0.9 nm
86.3 3.1 nm
82.6 1.5 nm
MNP-DOX-N6L
MNP-GEM-N6L
ꢁ38.5 1.4 mV
ꢁ36.5 4.2 mV
144.1 0.8 nm
118.1 0.7 nm

360
A. Latorre et al. / European Journal of Medicinal Chemistry 82 (2014) 355e362
Fig. 5. Release kinetics of N6L pseudopeptide from bi-functionalized DMSA-MNPs.
MNP-DOX-N6L (A) and MNP-GEM-N6L (B), (1
and 1 mM DTT, empty circles and solid line).
mM DTT, filled squares and dashed line
The release of the attached molecules from the functionalized
DMSA-MNPs was monitored during 12 h in PB buffer (10 mM
phosphate pH 7.4). In vitro drug release experiments were first done
with mono-functionalized DMSA-MNPs using intracellular condi-
tions (1 mM of DTT) and extracellular conditions (1
mM of DTT) at
37 ꢂC (Fig. 4).
The three mono-functionalized formulations showed a 95%
release when treated with 1 mM DTT after 8 h, although after 1 h
almost 70% of cargo was released. On the other hand, when the
particles were exposed to extracellular conditions (1 mM DTT), only
7% of the cargo was released after 8 h. In Fig. 5, in vitro release
experiments of N6L in the both bi-functionalized formulations are
shown, using the same experimental conditions.
The release experiments of N6L from bi-functionalized formu-
lations showed the same profiles than the ones observed for the
mono-functionalized formulation of Nucant (MNP-N6L).
Finally, release kinetics of DOX and GEM from the two bi-
functionalized formulations are presented in Fig. 6. The experi-
mental conditions were the same as in the previous example.
Similar values of release were observed on mono and bi-
functionalized formulations when exposed to intracellular condi-
tions, releasing 95% of the drugs after 8 h.
Fig. 4. Release kinetics of mono-functionalized DMSA-MNPs with DOX (A), GEM (B)
and N6L (C) (1
solid line).
mM DTT, filled squares and dashed line and 1 mM DTT, empty circles and
functionalized with DOX, GEM and N6L, and with bi-functionalized
MNPs. Two concentrations of DTT were employed in order to mimic
These results show that a reducing environment caused the
rapid dissociation of the drugs from MNP and that the release rate
extracellular and intracellular medium, DTT
DTT ¼ 1 mM, respectively.
¼
1 mM and

A. Latorre et al. / European Journal of Medicinal Chemistry 82 (2014) 355e362
361
developed strategy permits a controlled release of the drugs only
under a reducing environment that mimics the intracellular media.
Additionally, it is shown the bi-functionalization of DMSA-MNPs
with drugs and a peptide that is both a targeting and anticancer
agent. These multifunctional nanostructures still exhibit controlled
release properties of the two ligands.
The developed nanostructures described here have great po-
tential in directed controlled drug release strategies in nano-
medicine, and promise to overcome already present drug
resistance problems in cancer treatment. In vitro and in vivo ex-
periments are being carried out in order to explore the applicability
of the functionalized MNP presented in this work.
Acknowledgments
This work was supported by EU-FP7 MULTIFUN project (nꢂ
262943). The authors thank Dr. Gorka Salas, Leonor de La Cueva and
ꢀ
Rebeca Amaro for providing DMSA-MNPs, and Prof. Jose Courty for
providing the Nucant pseudopeptide.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2014.05.078.
References
[1] S.M. Moghimi, A.C. Hunter, J.C. Murray, Nanomedicine: current status and
future prospects, FASEB J. 19 (2005) 311e330.
[2] A.K. Gupta, M. Gupta, Synthesis and surface engineering of iron oxide nano-
particles for biomedical applications, Biomaterials 26 (2005) 3995e4021.
[3] O. Veiseh, J.W. Gunn, M. Zhang, Design and fabrication of magnetic nano-
particles for targeted drug delivery and imaging, Adv. Drug Deliv. Rev. 62
(2010) 284e304.
[4] J.D. Byrne, T. Betancourt, L. Brannon-Peppas, Active targeting schemes for
nanoparticle systems in cancer therapeutics, Adv. Drug Deliv. Rev. 60 (2008)
1615e1626.
[5] B. Book Newell, Y. Wang, J. Irudayaraj, Multifunctional gold nanorod ther-
agnostics probed by multi-photon imaging, Eur. J. Med. Chem. 48 (2012)
330e337.
Fig. 6. Release kinetics of DOX from MNP-N6L-DOX (A) and GEM from MNP-N6L-GEM
[6] G. Salas, C. Casado, F.J. Teran, R. Miranda, C.J. Serna, M.D.P. Morales, Controlled
synthesis of uniform magnetite nanocrystals with high-quality properties for
biomedical applications, J. Mater. Chem. 22 (2012) 21065.
(B) (DTT 1
line).
mM, filled squares and dashed line and DTT 1 mM, empty circles and solid
ꢀ
[7] M. Marciello, V. Connord, S. Veintemillas-Verdaguer, M.A. Verges, J. Carrey,
M. Respaud, et al., Large scale production of biocompatible magnetite nano-
crystals with high saturation magnetization values through green aqueous
synthesis, J. Mater. Chem. B 1 (2013) 5995e6004.
was strongly dependent on the reducing environment. We can also
conclude that the bi-functionalization does not affect the release of
the drug and the N6L. What is more, the loads of the described
nanoparticles are only released under reducing conditions as those
present in cytoplasm and lysosomal environments.
[8] F. Meyer-Losic, J. Quinonero, V. Dubois, B. Alluis, M. Dechambre, M. Michel, et
al., Improved therapeutic efficacy of doxorubicin through conjugation with a
novel peptide drug delivery technology (Vectocell), J. Med. Chem. 49 (2006)
6908e6916.
[9] H. Burris, M. Moore, J. Andersen, M.R. Green, M.L. Rothenberg, M.R. Modiano,
et al., Improvements in survival and clinical benefit with gemcitabine as first-
line therapy for patients with advanced pancreas cancer: a randomized trial,
J. Clin. Oncol. 15 (1997) 2403e2413.
4. Conclusions
[10] W. Plunkett, P. Huang, V. Gandhi, Preclinical characteristics of gemcitabine,
Anticancer Drugs 6 (1995) 7e13.
[11] L.W. Hertel, G.B. Boder, J.S. Kroin, S.M. Rinzel, G.A. Poore, G.C. Todd,
G.B. Grindey, Evaluation of the antitumor activity of gemcitabine (20,20-
difluoro-20-deoxycytidine), Cancer Res. 50 (1990) 4417e4422.
[12] C.F. Thorn, C. Oshiro, S. Marsh, T. Hernandez-Boussard, H. McLeod, T.E. Klein,
R.B. Altman, Doxorubicin pathways: pharmacodynamics and adverse effects,
Pharmacogenet. Genomics 21 (2011) 440e446.
This study shows optimized strategies for controlled and
quantitative functionalization of DMSA coated iron oxide nano-
particles with molecules such as commercially anticancer drugs
(Gemcitabine and Doxorubicin) and a targeting and cytotoxic agent
(Nucant pseudopeptide). This strategy is modular and generally
applicable thus, could be easily implemented for the immobiliza-
tion of other molecules of interest.
[13] L. Harivardhan Reddy, Catherine Dubernet, Sinda Lepetre Mouelhi, Pierre
Emmanuel Marque, Didier Desmaele, Patrick Couvreur, A new nanomedicine
of gemcitabine displays enhanced anticancer activity in sensitive and resistant
leukemia types, J. Control Release 124 (1e2) (4 December 2007) 20e27.
The colloidal stability of every formulation is preserved in
physiological conditions, with drug loads of 32
30 mol/g Fe for DOX and GEM, respectively. These nanostructures
can carry drugs to final concentrations of 16 M and 15 M of DOX
mmol/g Fe and
€
[14] X. Zeng, R. Morgenstern, A.M. Nystrom, Nanoparticle-directed sub-cellular
m
localization of doxorubicin and the sensitization breast cancer cells by cir-
cumventing GST-mediated drug resistance, Biomaterials 35 (2014)
1227e1239.
m
m
and GEM, respectively with iron doses of 0.5 mg Fe/mL, commonly
used in vitro for DMSA-MNPs [36,37]. Therefore, these nanocarriers
can deliver drugs at local concentrations much higher than the IC50
values reported for different cell lines (IC50 of DOX in MCF-7 cells is
[15] L.S. Jabr-Milane, L.E. van Vlerken, S. Yadav, M.M. Amiji, Multi-functional
nanocarriers to overcome tumor drug resistance, Cancer Treat. Rev. 34 (2008)
592e602.
[16] D. Destouches, N. Page, Y. Hamma-Kourbali, V. Machi, O. Chaloin, S. Frechault,
C. Birmpas, P. Katsoris, J. Beyrath, P. Albanese, M. Maurer, G. Carpentier, J.-
0.23 mM [38] and IC50 of GEM in Panc-1 cells is 0.11 mM [39]). The

362
A. Latorre et al. / European Journal of Medicinal Chemistry 82 (2014) 355e362
M. Strub, A. Van Dorsselaer, S. Muller, D. Bagnard, J.P. Briand, J. Courty,
A simple approach to cancer therapy afforded by multivalent pseudopeptides
that target cell-surface nucleoproteins, Cancer Res. 71 (2011) 3296e3305.
nanoparticles coated with polymers: advantages of Pluronic® F68-PEG
mixture, Nanotechnology 24 (2013) 395605.
[27] C. Dionigi, Y. Pineiro, A. Riminucci, M. Banobre, Regulating the thermal
response of PNIPAM hydrogels by controlling the adsorption of magnetite
nanoparticles, Appl. Phys. A 114 (2014) 585e590.
[28] G. Saito, J. Swanson, K. Lee, Drug delivery strategy utilizing conjugation via
reversible disulfide linkages: role and site of cellular reducing activities, Adv.
Drug Deliv. Rev. 55 (2003) 199e215.
~
~
[17] D. Destouches, E. Huet, M. Sader, S. Frechault, G. Carpentier, F. Ayoul, J.-
P. Briand, S. Menashi, J. Courty, Multivalent pseudopeptides targeting cell
surface nucleoproteins inhibit cancer cell invasion through tissue inhibitor of
metalloproteinases
3
(TIMP-3) release, J. Biol. Chem. 287 (2012)
43685e43693.
[18] D. Destouches, D. El Khoury, Y. Hamma-Kourbali, B. Krust, P. Albanese,
P. Katsoris, G. Guichard, J.P. Briand, J. Courty, A.G. Hovanessian, Suppression of
tumor growth and angiogenesis by a specific antagonist of the cell-surface
expressed nucleolin, PLoS One 3 (2008) e2518.
[29] A. Meister, M.E. Anderson, Glutathione, Annu. Rev. Biochem. 52 (1983)
711e760.
[30] W.G. Kirlin, J. Cai, S.A. Thompson, D. Diaz, T.J. Kavanagh, D.P. Jones, Gluta-
thione redox potential in response to differentiation and enzyme inducers,
Free Radic. Biol. Med. 27 (1999) 1208e1218.
[31] S.L. Blair, P. Heerdt, S. Sachar, A. Abolhoda, S. Hochwald, H. Cheng, M. Burt,
Glutathione metabolism in patients with non-small cell lung cancers, Cancer
Res. 57 (1997) 152e155.
[32] M. Lapeyre, J. Leprince, M. Massonneau, H. Oulyadi, P.-Y. Renard, A. Romieu,
G. Turcatti, H. Vaudry, Aryldithioethyloxycarbonyl (Ardec): a new family of
amine protecting groups removable under mild reducing conditions and their
applications to peptide synthesis, Chemistry 12 (2006) 3655e3671.
[33] A. Satyam, Design and synthesis of releasable folate-drug conjugates using a
novel heterobifunctional disulfide-containing linker, Bioorg. Med. Chem. Lett.
18 (2008) 3196e3199.
ꢀ
ꢀ
[19] R. Mejías, S. Perez-Yagüe, L. Gutierrez, L.I. Cabrera, R. Spada, P. Acedo,
ꢀ
C.J. Serna, F.J. Lazaro, A. Villanueva, M.D.P. Morales, D.F. Barber, Dimercapto-
succinic acid-coated magnetite nanoparticles for magnetically guided in vivo
delivery of interferon gamma for cancer immunotherapy, Biomaterials 32
(2011) 2938e2952.
[20] A.Z. Mirza, H. Shamshad, Preparation and characterization of doxorubicin
functionalized gold nanoparticles, Eur. J. Med. Chem. 46 (2011) 1857e1860.
[21] S.S. Banerjee, D.-H. Chen, Multifunctional pH-sensitive magnetic nanoparticles
for simultaneous imaging, sensing and targeted intracellular anticancer drug
delivery, Nanotechnology 19 (2008) 505104.
[22] C. Fang, F.M. Kievit, O. Veiseh, Z.R. Stephen, T. Wang, D. Lee, R.G. Ellenbogen,
M. Zhang, Fabrication of magnetic nanoparticles with controllable drug
loading and release through a simple assembly approach, J. Control Release
162 (2012) 233e241.
[34] M.M. Bradford, A rapid and sensitive method for the quantitation of micro-
gram quantities of protein utilizing the principle of protein-dye binding, Anal.
Biochem. 72 (1976) 248e254.
[23] H. Wang, T.B. Shrestha, M.T. Basel, R.K. Dani, G.-M. Seo, S. Balivada, M.M. Pyle,
H. Prock, O.B. Koper, P.S. Thapa, D. Moore, P. Li, V. Chikan, D.L. Troyer,
S.H. Bossmann, Magnetic-Fe/Fe(3)O(4)-nanoparticle-bound SN38 as
carboxylesterase-cleavable prodrug for the delivery to tumors within mono-
cytes/macrophages, Beilstein J. Nanotechnol. 3 (2012) 444e455.
[24] F. Cengelli, J.A. Grzyb, A. Montoro, H. Hofmann, S. Hanessian, L. Juillerat-
Jeanneret, Surface-functionalized ultrasmall superparamagnetic nanoparticles
as magnetic delivery vectors for camptothecin, ChemMedChem 4 (2009)
988e997.
[25] G.Y. Lee, W.P. Qian, L. Wang, Y.A. Wang, C.A. Staley, M. Satpathy, S. Nie,
H. Mao, L. Yang, Theranostic nanoparticles with controlled release of gemci-
tabine for targeted therapy and MRI of pancreatic cancer, ACS Nano 7 (2013)
2078e2089.
[35] N. Fauconnier, J. Pons, J. Roger, A. Bee, Thiolation of maghemite nanoparticles
by dimercaptosuccinic acid, J. Colloid Interface Sci. 194 (1997) 427e433.
[36] M. Calero, L. Gutierrrez, G. Salas, Y. Luengo, A. Lazaro, P. Acedo, M.P. Morales,
R. Miranda, A. Villanueva, Efficient and safe internalization of magnetic iron
oxide nanoparticles: two fundamental requirements for biomedical applica-
tions, Nanomedicine (2013) 1e11.
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
[37] R. Mejías, L. Gutierrez, G. Salas, S. Perez-Yagüe, T.M. Zotes, F.J. Lazaro,
M.P. Morales, D.F. Barber, Long term biotransformation and toxicity of
dimercaptosuccinic acid-coated magnetic nanoparticles support their use in
biomedical applications, J. Control Release 171 (2013) 225e233.
[38] A.-M.M. Osman, H.M. Bayoumi, S.E. Al-Harthi, Z.A. Damanhouri, M.F. Elshal,
Modulation of doxorubicin cytotoxicity by resveratrol in a human breast
cancer cell line, Cancer Cell Int. 12 (2012) 47.
[39] C. Ramachandran, A. Resek, Potentiation of gemcitabine by Turmeric Force™
in pancreatic cancer cell lines, Oncol. Rep. 23 (2010) 1529e1535.
ꢀ
ꢀ
ꢀ
[26] M. Chiper, K. Herve Aubert, A. Auge, J.-F. Fouquenet, M. Souce, I. Chourpa,
Colloidal stability and thermo-responsive properties of iron oxide