Langmuir 2010, 26(12), 10315–10324 Published on Web 03/31/2010
©2010 American Chemical Society
Acidic pH-Responsive Nanogels as Smart Cargo Systems for the
Simultaneous Loading and Release of Short Oligonucleotides and
Smriti R. Deka,†Alessandra Quarta,†Riccardo Di Corato,‡Andrea Falqui,†Liberato Manna,†
Roberto Cingolani,†and Teresa Pellegrino*,†,‡
†Istituto Italiano di Tecnologia, via Morego 30, 16163, Genova, Italy, and
Laboratory of CNR-INFM and IIT Research Unit, via per Arnesano km 5, 73100, Lecce, Italy
Received February 1, 2010. Revised Manuscript Received March 1, 2010
Smart materials able to sense environmental stimuli can be exploited as intelligent carrier systems. Acidic pH-
responsive polymers, for instance, exhibit a variation in the ionization state upon lowering the pH, which leads to their
swelling. The different permeability of these polymers as a function of the pH could be exploited for the incorporation
and subsequentrelease ofpreviously trappedpayloadmolecules/nanoparticles. We provide herea proofofconcept ofa
below 200 nm, as cargo system for magnetic nanoparticles, for oligonucleotide sequences, as well as for their simul-
taneous loading and controlled release mediated by the pH.
Research on nanocomposites aims at developing and minia-
turizing structures made of different functional entities, each of
them able to carry out specific tasks. In order to design multi-
functional nanostructures that might serve as new medical de-
vices,itiscrucial to identify“smart materials”thatarecapableof
responding to defined stimuli. Hydrogels are an interesting class
of functional materials that have been exploited as intelligent
cargo systems for the encapsulation and the delivery of different
and that can be useful for the capture and for the controlled
release of inorganic nanoparticles. These polymers, whose struc-
ture is composed of a three-dimensional network of cross-linked
units, can undergo substantial modifications of some of their
properties (such as their total charge or their hydrophobicity/
local environment, like a change in pH or in the temperature.5,6
Hydrogels in their bulk form have been applied so far in im-
plants, contact lenses, dental materials, and vascular grafts.1,7In
past years, there has been increasing research activity focused on
the miniaturization of hydrogel particles (henceforward referred
to as “nanogels”) and on the study of their potential as drug
nanogels to have reliable structure-properties relationships one
needs to finely control both their size and their purity.6Size
control is particularly critical because a nanogel designed for in
vivo delivery of drugs, genes, or nanoparticles should be smaller
than 200 nm.8-13Once injected intravenously, a nanogel smaller
than this size will remain colloidally stable, it will not be seq-
uestered immediately by the reticulo-endothelial system,11and
the tumor regions and pass through the pore vessels at these
While nanogelsbasedontemperature-responsive polymersare
generally designed to be altered by external stimuli, those based
on pH-responsive polymers can respond to variations in the
intracellular or tissue environment.1It is known, for example,
extracellular pH around 6.5,15while the pH of some intracellular
compartments, like lysosomes, is around 4.5.16The pH-dependent
swelling behavior of a nanogel can be useful not only for
the release of the cargo, but also for its loading. Indeed, when
the nanogel swells, its permeability increases, allowing either
for the incorporation of molecules/nanoparticles or alternatively
for the release of previously trapped payloads. So far, several
pH-responsive polymers have been widely used as a controlled
drug delivery system.17-21In some studies, pH-responsive poly-
mers have been exploited as templates for the in situ synthesis of
catalytic applications.22,23Only in a few works were pH-responsive
*Corresponding author. firstname.lastname@example.org; tel. þ39 0832 298 214;
fax þ39 0832 298 230.
(1) Schmaljohann, D. Adv. Drug Delivery Rev. 2006, 58, 1655.
(2) Nayak, S.; Lyon, L. A. Angew. Chem.-Int. Ed. 2005, 44, 7686.
(3) Gupta, P.; Vermani, K.; Garg, S. Drug Discovery Today 2002, 7, 569.
(4) Vinogradov, S. V.; Bronich, T. K.; Kabanov, A. V. Adv. Drug Delivery Rev.
2002, 54, 135.
(5) Ganta, S.; Devalapally, H.; Shahiwala, A.; Amiji, M. J. Controlled Release
2008, 126, 187.
(6) Raemdonck, K.; Demeester, J.; De Smedt, S. Soft Matter 2009, 5, 707.
(7) Liu, S.; Maheshwari, R.; Kiick, K. L. Macromolecules 2009, 42, 3.
(8) Takakura, Y.; Mahato, R. I.; Hashida, M. Adv. Drug Delivery Rev. 1998,
(9) Mitragotri, S.; Lahann, J. Nat. Mater. 2009, 8, 15.
(10) Decuzzi, P.; Pasqualini, R.; Arap, W.; Ferrari, M. Pharm. Res. 2009, 26,
(11) Cairns, R.; Papandreou, I.; Denko, N. Mol. Cancer Res. 2006, 4, 61.
(12) Torchilin, V. P. Adv. Drug Delivery Rev. 2006, 58, 1532.
(13) Torchilin, V. P. J. Controlled Release 2001, 73, 137.
(14) Prokop, A.; Davidson, J. M. J. Pharm. Sci. 2008, 97, 3518.
(15) Gerweck, L. E.; Seetharaman, K. Cancer Res. 1996, 56, 1194.
(16) Grabe, M.; Oster, G. J. Gen. Physiol. 2001, 117, 329.
(17) Dai, S.; Ravi, P.; Tam, K. C. Soft Matter 2008, 4, 435.
(18) Wu, D. Q.; Sun, Y. X.; Xu, X. D.; Cheng, S. X.; Zhang, X. Z.; Zhuo, R. X.
Biomacromolecules 2008, 9, 1155.
(19) Zhao, C.; Zhuang, X.; He, P.; Xiao, C.; He, C.; Sun, J.; Chen, X.; Jing, X.
Polymer 2009, 50, 4308.
(20) Qu, T.; Wang, A.;Yuan, J.; Gao,Q. J.Colloid Interface Sci. 2009, 336, 865.
(21) Peppas, N. A.; Hilt, J. Z.; Khademhosseini, A.; Langer, R. Adv. Mater.
2006, 18, 1345.
(22) Zhang, J. G.; Xu, S. Q.; Kumacheva, E. J. Am. Chem. Soc. 2004, 126, 7908.
(23) Palioura, D.; Armes, S. P.; Anastasiadis, S. H.; Vamvakaki, M. Langmuir
2007, 23, 5761.
DOI: 10.1021/la1004819 Langmuir 2010, 26(12), 10315–10324
Article Deka et al.
polymerscombined withmagneticnanoparticles,24,25and insuch
the nanogel networks. In most of those works, the magnetic
nanoparticles were actually used as templates on which the
polymer was grown around,25,26or vice versa, the polymer was
used as template on which the magnetic nanoparticles were
nucleated (and remained bound to it). To our knowledge, there
for the controlled release of magnetic nanoparticles.
(IONPs, both maghemite and magnetite) are superparamagnetic
nanocrystals that have been widely investigated as drug delivery,
diagnosis, and therapeutic agents.27Due to their intrinsic mag-
netic properties, IONPs are ideal candidates as delivery agents:
they are able to accumulate to the site where the magnet is posi-
tioned, while upon removal of the magnet, they do not undergo
aggregation, as they do not exhibit any residual magnetization.
Furthermore, IONPs are valuable contrast agents in magnetic
resonance imaging (MRI) because their magnetic moments can
affect the relaxivity of the water molecule protons present in the
tissues, resulting in a negative contrast in the area where the
nanoparticles are localized.28IONPs can also serve as colloidal
mediators for generating heat for hyperthermia treatment in
cancer therapy, under the application of appropriate alternating
magnetic fields.29The inclusion of IONPs in the nanogel confers
to it all the additional advantages of IONPs as described above.
In the present work, we employ acidic pH-responsive nanogels
and short oligonucleotide sequences, and combinations of them.
We have modified a previously reported synthetic procedure for
preparing pH-responsive nanogels30in order to obtain nanogels
with sizes tunable from 40 to 200 nm, and we have tested those
materials as carrier systems. A full characterization based on
transmission electron microscopy (TEM), photoluminescence
spectroscopy, confocal microscopy, and dynamic light scattering
(DLS) was carried out in order to elucidate the loading and the
release processes of short DNA sequences, of IONPs, and the
combined loading and release of both payloads at the same time.
Our pH and of salt concentration results show that full control
over the loading and the release of IONPs and DNA is clearly
2. Experimental Section
able and were used as received. 2-Vinyl pyridine (2-VP, 97%),
sodium tetraborate decahydrate, boric acid, as well as all the
disposable materials were purchased from Sigma-Aldrich.
Divinylbenzene (DVB), 2,20-azobis(2-methylpropionamidine)-
dihydrochloride (AIBA), and Diamine-PEG (MW 897) were
purchased from Fluka. The HPLC purified oligonucleotide seq-
uence modified at the 50end with Cy3 (50-CAC CAC ACG GTC
was purchased from Thermo Electron Corporation. Doubly
(PC 14) was purchased from Sigma-Aldrich, although at pre-
can refer to a new polymer coating procedure implemented by us
which employs a similar polymer, which is still commercially
2.2. Synthesis of Nanogels via Emulsion Polymerization.
A series of polyvinyl pyridine nanogels were synthesized with a
control over the size diameter of the nanogel below 200 nm,
us. As an example, we describe here the experimental conditions
for the synthesis of nanogels of about 110 nm in diameter (as
determined by transmission electron microscopy, TEM). A mix-
ture of 2-vinyl pyridine (2-VP, 0.25 g) and divinylbenzene (DVB,
The flask was sealed with a rubber septum, and the aqueous
solution was degassed at ambient temperature by five vacuum/
a magnetic stirrer and heated at 60 ( 1 ?C. After 20 min, the
solution of the AIBA initiator (0.022 g in 1 mL water) was added
left to polymerize for a further 2 h under stirring conditions at
nitrogen atmosphere and to stopthe reaction. Inorder toremove
the residual monomers in solution, the 2-VP nanogel particles
were washed at least 10 times with a Millipore Dialysis System
(MWCO 100.000) on centricone tubes, and the reaction mixture
was centrifuged at 4000 rpm for 30 min. Fresh water was added
each time before any centrifugation. All dispersions were diluted
filter) prior to use. The solution pH was adjusted by using a
with a pH-meter equipped with a microelectrode (Crison pH-
Meter Basic 20þ). In order to tune the size of the nanogels below
200 nmindiameter,wehavevariedthe molarratio of2-VP/DVB
the other reaction conditions, as described above (Table 1).
2.3. Preparation of Diamino-PEG Conjugated Iron Oxide
Nanocrystals. Iron oxide nanocrystals (diameter of 7 nm) were
synthesized according to the Sun method.32The “as synthesized”
soluble in organic solvents. They were transferred into water by
usingapolymercoating proceduredeveloped byus.33Briefly,the
of poly(maleic anhydride alt-1-tetradecene), and such shell was
then cross-linked using a triamine. The nanocrystals were there-
fore soluble in water and were negatively charged, as determined
by zeta potential measurements (Table 1, Supporting Infor-
mation), due to the outstretched carboxylate moieties of the
polymer molecules. In order to remove the excess free polymer,
an ultracentrifugation step was performed at 150000 rcf on a
continuous sucrose gradient.31Then, diamino-PEG molecules
(molecular weight 897 Da) were bound to the carboxy groups at
the nanoparticle surface via EDC chemistry. The amino-PEG
molecules were introduced in order to make the nanoparticles
more stable at different conditions of pH and ionic strength.34In
detail, to 500 μL of a nanocrystal solution 6 μM, 500 μL of a
solution containing a molar ratio of diamino-PEG/NP equal to
an excess molar ratio of EDC/NP (equal to 75000) was also
(24) Bhattacharya, S.; Eckert, F.; Boyko, V.; Pich, A. Small 2007, 3, 650.
D. Z. J. Magn. Magn. Mater. 2009, 321, 2799.
X. H.; Shen, D. Z. Mater. Lett. 2009, 63, 1567.
(27) Laurent, S.; Forge, D.; Port, M.; Roch, A.; Robic, C.; Elst, L. V.; Muller,
R. N. Chem. Rev. 2008, 108, 2064.
(28) Na, H. B.; Song, I. C.; Hyeon, T. Adv. Mater. 2009, 21, 2133.
(29) Gazeau, F.; Levy, M.; Wilhelm, C. Nanomedicine 2008, 3, 831.
(30) Dupin, D.; Fujii, S.; Armes, S. P.; Reeve, P.; Baxter, S. M. Langmuir 2006,
(31) Di Corato, R.; Quarta, A.; Piacenza, P.; Ragusa, A.; Figuerola, A.;
Buonsanti, R.; Cingolani, R.; Manna, L.; Pellegrino, T. J. Mater. Chem. 2008,
(32) Sun, S. H.; Zeng, H. J. Am. Chem. Soc. 2002, 124, 8204.
(33) Pellegrino, T.; Manna, L.; Kudera, S.; Liedl, T.; Koktysh, D.; Rogach,
A. L.; Keller, S.; Radler, J.; Natile, G.; Parak, W. J. Nano Lett. 2004, 4, 703.
(34) Sperling, R. A.; Pellegrino, T.; Li, J. K.; Chang, W. H.; Parak, W. J. Adv.
Funct. Mater. 2006, 16, 943.
Langmuir 2010, 26(12), 10315–10324
Deka et al.Article
added. After a reaction time of 3 h at room temperature under
vigorous stirring, the unbound diamino-PEG molecules were
removed byperformingseveral washingsteps oncentrifuge tubes
having a MWCO of 30000.
2.4. Loading and Release Experiments of Diamino-PEG
Conjugated Iron Oxide Nanocrystals in the Nanogel. The
of nanoparticles in this solution was equal to 14.5 μM) and the
resulting mixture was stirred for 24 h at pH 3.5 at room tempera-
ture. Under these conditions, the swollen nanogels started incor-
porating the IONPs. The pH of the solution was then increased
slowly to 7 by dropwise addition of a solution of NaOH 0.1 M
(a slight turbidity appeared as soon as the pH reached 5.25, indi-
with IONPs were separated from free excess of IONPs using a
magnet: the solution was placed close to the magnet, and within
1 h, the nanogels loaded with nanoparticles were attracted
toward it. The nanogels were characterized by TEM and by
DLS measurements. For the release experiments, the pH of the
of a solution of HCl (0.1 M), in order for the nanogels to swell
again. For the quantification of the average number of IONPs
loaded within the nanogel, the determination of iron concentra-
tion was carried out via elemental analysis on the loaded IONPs-
2.5. Loading Experiment of Cy3-DNA in Nanogel and
Subsequent Release. For the loading experiments of oligonu-
cleotide sequences of 25 bases, 3 mL of nanogel solution (0.053
after, the pH was increased again to 7 by dropwise addition of a
solution of NaOH 0.1 M, after which it was left to stir for addi-
tional 3 h. To remove the excess of free Cy3-DNA, the final solu-
The process was repeateduntilall the freeCy3-DNA waswashed
away, as monitored by PL spectra on the filtered solution (5 to
centriconetube,while theCy3-DNA/nanogelswere recovered on
the upper side of the filter and were redispersed in 1 mL of water.
we have recorded the PL of loaded nanogel samples (both Cy3-
DNA/nanogel and Cy3-DNA-IONP/nanogel). We have then
extrapolated the Cy3-DNA concentration of those samples on
tion (PL/Cy3-DNA concentration). These were obtained by
preparing standard solutions at known Cy3-DNA concentra-
tions, in which we have simulated the matrix. In more detail, for
building the calibration curve for the Cy3-DNA/nanogel sample,
added the same amount of nanogel that we had in our sample.
Likewise, in order to build the calibration curve for the Cy3-
DNA-IONP/nanogel sample, to each of the standards we have
that we had in our sample (Figure 8S, Supporting Information).
For the release experiment of the Cy3-DNA, 50 μL of a
solution loaded with Cy3-DNA, and the pH was adjusted to 3.5.
The sample was left under stirring for 72 h, and soon after the
solution was centrifuged on an amicon tube of MWCO 100000.
part and on the lower part of the membrane.
and IONPs and Subsequent Release Experiments. To load
IONPs and Cy3-DNA simultaneously within the nanogels, the
same procedure as described above (to load IONPs and Cy3-
DNA separately) was applied. The only difference in the present
case was that 3 mL of nanogel in water (0.053 w/v (g/mL %) was
μL) and with 9 μL of the IONPs solution (14.5 μM), after which
the pHwasadjustedto3.5 using HCl 0.1M.Also, inthiscasethe
loading and the release were monitored by TEM and by PL.
light scattering measurements (DLS) were performed on a Zeta-
sizer Nano ZS90 (Malvern, USA) equipped with a 4.0 mW
He-Ne laser operating at 633 nm and with an avalanche photo-
diode detector. Measurements were made at 25 ?C in water. All
the samples were filtered before analysis. 0.2 μm filters were used
for the nanogel alone, while for nanogel samples loaded with
nanoparticles and Cy3-DNA solution, 0.5 μm filters were pre-
2.8. UV-vis Absorption, Photoluminescence (PL) Spec-
troscopy. UV-visible absorption spectra were measured using a
To record the PL spectra of Cy3-DNA alone and in nanogel, the
samples were excited at 500 nm.
2.9. Transmission Electron Microscopy. TEM images
were recorded on a JEOL jem 1011 microscope operated at an
accelerating voltage of 100 kV. TEM samples were prepared by
dropping a dilute solution of nanogel in water on carbon-coated
diameters were measured on an average of 100 particles.
images were acquired with an Olympus FV-1000 microscope,
was set at 565 ( 25 nm.
2.11. Elemental Analysis. An inductively coupled plasma
atomic emission spectrometer (ICP-AES) Varian Vista AX was
used to measure the concentration of Fe and thus the concentra-
tion of IONPs. The samples were digested in the following way:
they were dissolved in a concentrated acid solution (HCl/HNO3
(3/1 v/v) and were left for 24 h, before performing elemental
analysis. The Fe concentration was converted into nanoparticle
concentration using a procedure described by us in a previously
published paper.35In detail, the average diameter of the nano-
average number of Fe atoms per nanoparticle was determined by
Table 1. Experimental Conditions for the Synthesis of Nanogels of Different Diametersa
sample name2-VP (g)DVB (g) [2-VP]/[DVB] Molar ratioTEM diameter (nm)DLS diameter (nm) polydispersity index
the sizes of the nanogels from 41 nm to 197, as determined by statistical TEM measurements (column 5) on an average of 100 nanogel particles. The
hydrodynamic diameters of the same samples, as measured by dynamic light scattering, (column 6) were clearly bigger. The low polydispersity index
indicates uniform size distribution (column 7) (all measurements were conducted at pH 7.5).
(35) Deka, S.; Quarta, A.; Lupo, M. G.; Falqui, A.; Boninelli, S.; Giannini, C.;
Morello, G.; De Giorgi, M.; Lanzani, G.; Spinella, C.; Cingolani, R.; Pellegrino,
T.; Manna, L. J. Am. Chem. Soc. 2009, 131, 2948.
DOI: 10.1021/la1004819Langmuir 2010, 26(12), 10315–10324
Article Deka et al.
building a structural model of the nanoparticle, with the same
geometrical parameters of the nanoparticles as determined by
TEM. Then, by knowing the average number of Fe atoms per
nanoparticle and the total concentration of this species in solu-
In order to elucidate Fe3þleakage in the condition of the loading
experiment, we kept the IONPs at pH 3.5 overnight and we
collected the supernatant solution, i.e., the solution separated
from the IONPs by filtration on centricone filter. Finally, we
measured the Fe concentration in both fractions.
3. Results and Discussion
3.1. Preparation of pH-Responsive Nanogels and Char-
acterization of Their Swelling Behavior. The pH-responsive
nanogels employed in this study are based on copolymers of di-
vinylbenzene(DVB) andvinyl pyridine(VP) (sketchofFigure1).
minor modifications.30,36This type of surfactant-free emulsion
polymerization procedure was first described by Loxley and
Vincent,36who synthesized monodisperse cationic nanogels of
2-vinylpyridine by varying the amount of styrene (the hydropho-
bic monomer) and that of DVB (the cross-linker agent). The
authors demonstrated a tight control over the particle size in the
range between 160 and 200 nm. According to a modified version
the synthesis of sterically stabilized PVP latexes at much higher
solid density, and with control over the diameter from 300 to
1000 nm. They used suitable stabilizer molecules, namely, mono-
methoxy-capped poly(ethylene glycol) methacrylate (PEGMA),
and surfactant molecules named “336”.
Our interest in the present study was to control the size of the
potential use of such nanogels as cargo system, as highlighted
above. We were able to synthesize a series of nanogels in the size
range between 40 and 200 nm, by reducing the monomer con-
centration of 2-VP, while keeping all the other parameters
constant (Table 1 and Figure 1). Reducing the concentration of
2-VP corresponds to a decrease in the 2-VP/DVB molar ratio, or
the same to an increase in the amount of DVB (the cross-linker
diameter of about 40 nm (Figure 1S, Supporting Information).
by DLS (Table 1, column 5), was slightly higher than that
measured by TEM. This was expected, since the DLS measure-
Figure 1. Sketch of the structure of the vinyl pyridine (VP) and divinylbenzene (DVB) units, which were employed for the synthesis of
(d) 197 nm (the TEM diameters reported were estimated on an average of 100 nanogel particles; see Table 1, column 5).
(36) Loxley, A.; Vincent, B. Colloid Polym. Sci. 1997, 275, 1108.
Langmuir 2010, 26(12), 10315–10324
Deka et al.Article
distribution of the nanogels (Table 1, column 7).
It is also worth highlighting that in our preparation we ran the
reaction for 2 h, while in previously reported methods the
reaction time was 24 h.30We additionally observed that by inc-
reasing the reaction time from 2 to 24 h the size of the nanogel
To investigate the swelling behavior of the various nanogel
samples in water, the pH of each nanogel solution was lowered
from an initial value of 7.5 by dropwise additions of an HCl
diameters increased sharply at pH below 4.3, due to progressive
protonation of the nitrogen of the 2-vinylpyridine residues37
(Figure 2). Regardless of the starting size of the nanogels, the
swelling occurred always at pH below 4.3. The various samples
3.9, depending on the ratio between 2-VP and DVB employed in
The majority of the pyridine groups became protonated below
pH 3.8, and the average diameters of the nanogels reached a
the absence of inner charges, and they behaved like conventional
polymer latex particles.37The swelling of the nanogel particles
was also confirmed by visual inspection, as the solution turned
from turbid, milky-white to clear when the pH was decreased
measurements indicated a strong cationic character of the nano-
gels at pH 3.5, which is the pH at which the payload was usually
loaded. However, even at pH 7 the nanogels retained a slightly
positive charge (Table 2, Supporting Information).
3.2. Loading and Release Experiments of Iron Oxide
gel sample having average “TEM” diameter equal to 110 nm at
pH 7 was employed (henceforward referred to as “NG110”), and
(Figure 3). After mixing the nanogels with IONPs (PEG-coated
nanoparticles,7 nmin diameter)32-34and uponswitchingthe pH
from 7 to 3.5, the nanogel was turned into a swollen state (under
TEM, the edge of the nanogel was not sharply defined anymore;
see Figure 3B). A gentle overnight shaking at room temperature
was thenfollowedbyrestorationofthesolution pHbackto7(by
inside which the IONPs remained entrapped (Figure 3C).
By application of a magnet, the nanogels loaded with IONPs
could be recovered and they were separated by the excess of free
IONPs (Figure 3D), as the former were attracted faster than the
latter to the magnet (Figure 3D). In order to achieve a complete
tionof the IONPs in the nanogel induced an appreciable increase
of the average nanogel diameter, as determined by TEM (in one
sample, for instance, it varied from 110 ( 8 nm to 117 ( 12 nm).
nanogels at each step of the procedure. Immediately after mixing
the IONPs with the nanogel, at pH 3.5 the DLS diameter of the
nanogel was around 480 ( 94 nm (Table 1, Supporting Infor-
mation), which was lower than that of the nanogels when they
were swollen at the same pH but in the absence of IONPs (713 (
ionic interactions in solution between the IONPs and the nano-
gels. After switching the pH of the same solution back to 7, the
DLS diameters of the nanogels in the presence of the IONPs was
for the empty nanogels (Table 1, Supporting Information). TEM
characterization confirmed the presence of IONPs within the
nanogel structure (Figure 3C). The nanogels loaded with IONPs
could release their payload by switching the pH again from 7 to
from nanogels, as confirmed by TEM (Figure 4).
Inorder torationalizeandunderstandthe drivingforcefor the
loading, we have characterized the system in more detail by
analyzing the surface charge of the individual units, namely, the
nanogels and the IONPs, and that of the nanogels loaded with
potentials were þ56 mV and þ8 mV, respectively), and at the
attempted to load the nanogels at pH 7 instead of pH 3.5. At this
pH, the surface charge of the nanogels was still moderately posi-
tive, whilethatoftheIONPs was negative.The negativechargeis
likely due to the charge balance at the surface of IONPs given by
the sum ofamino-PEG moieties and carboxyl-terminated groups
see Table 2, Supporting Information). Therefore, even if at this
pH value the nanogels were swollen, one should expect a higher
observedindeedthatalso afterincubationunder theseconditions
wecould loadIONPswithinthe nanogels(Figure 4S, Supporting
at which the surface charge of the nanogel was only slightly
positive (zeta potentials for the nanogels and for the IONPs were
þ15 and -42 mV); hence, the electrostatic interactions between
the nanogels and the IONPs were weaker than at pH 7. In this
case, we could still observe (by TEM) the adsorption of a few
nanoparticles on the surface of the nanogels, but most nanogels
had not beenable toincorporatetheIONPs(Figure4S, Support-
For the quantification of IONPs loaded within the nanogel
at the different pH values, the various samples were digested in
HCl/HNO3, and their iron content was estimated by means of
elemental analysis, which allowed us to estimate quantitatively
Figure 2. Swellingbehavior of the nanogels. The DLS diameter is
reported as a function of the pH (each measurement was carried
out three times.) All nanogel samples exhibited a sharp swelling
behavior below pH 4.3.
(37) Fernandez-Nieves, A.; Fernandez-Barbero, A.; Vincent, B.; de las Nieves,
F. J. Macromolecules 2000, 33, 2114.
DOI: 10.1021/la1004819Langmuir 2010, 26(12), 10315–10324
Article Deka et al.
the number of IONPs entrapped within the nanogels.35The
highest concentration of IONPs was found in the beads loaded
at pH 3.5, followed by those loaded at pH 7 and then by those at
reproducible and provided a clear indication of the average
loading efficiency of the nanogels. These data nicely correlate
the DLS), which is bigger for the gels loaded at pH 3.5, followed
again by those loaded at pH 7 and at pH 10, respectively. This
correlation suggests that the swelling behavior is the main driver
for the encapsulation of IONPs, although a contribution due to
electrostatic interactions between the nanogels and the IONP
surface cannot be excluded.
Once the nanogels were loaded with IONPs, their surface
charge became negative at pH 7. However, the trend in absolute
values of surface charges was reversed in this case: it was higher
for thenanogelsloadedatpH10, followedbythose loadedatpH
7 and pH 3.5, respectively. This might likely be attributed to a
much lower fraction of nanoparticles adsorbed at the nanogel
surface with respect to those trapped deeper in the network
structure of the nanogel at pH 3.5 (the nanoparticles contributed
with negatively charges).31,33
3.3. TEM Characterization of the Entrapment of the
IONPs within the Nanogels. In order to confirm the entrap-
ment ofthe IONPs, we carried out additionalTEM characteriza-
tion. Several brightfield electron microscopy (BF-EM) imagesof
a nanogel sample loaded with IONPs were taken at different tilts
on a large angular range (from -55? to 0?, to 60?). Two BF-EM
B. Figure 5A corresponds to the specimen tilted at 0? (i.e., the
plane of the sample is basically normal to the electron beam
60?. From the high-tilt image, two main considerations can be
at 0? tilt) is elliptical at high tilt, indicating that the polymer
behaves as a sphere pressed on the plane of the carbon grid in a
direction perpendicular to it. Second, at high tilt the spherical
nanoparticles are located inside the contours of the light-gray
zone that corresponds to the polymer. This suggests that the
IONPs were embedded within the first few polymer layers. If, on
the polymeric crushed sphere, they should have appeared also on
the external side of the light-gray zone’s contour.
In order to localize the radial distribution of IONPs within the
nanogels, we have performed TEM on the cross sections of the
IONP-loaded nanogels, which had been embedded within a
paraffin resin. The sections analyzed had thickness of 70 nm
(Figure 5C) and 50 nm (inset of Figure 5C), respectively. As
the edge of the beads, within the first layers of the polymer, and
only few of them were found deep inside in the nanogels. It is
interesting to compare these results with the cross-sectional
images of the same type of nanogel used as template for the gold
synthesis reported by Nakamura.38In that case, as the gold
nucleation occurred only at the surface on the TEM cross
sections, no nanoparticles were found within the nanogel.
Figure 3. (A)SchemeoftheloadingofmagneticnanoparticleswithinthepH-responsivenanogels.CorrespondingTEMcharacterizationof
the different steps:(B) atacidicpHthe nanogels weremixedwiththe IONPs;(C) after 12h,the pHwas switchedbacktopH7,suchthatthe
IONPs were entrapped within the nanogel network. (D) The application of a small magnet helped to remove most of the free IONPs in
solution. A complete cleaning was achieved by performing an additional purification step on a Sephadex column.
Figure 4. Release experiment of the IONPs. Switching the pH of
the IONP-loaded nanogel solution from 7 to 3.5 induced the
entrapped within the polymer network.
(38) Kensuke Akamatsu, , Takaaki Tsuruoka, Hidemi Nawafune, Syuji Fujii
Yoshinobu Nakamura Langmuir 2009, [Online early access].
Langmuir 2010, 26(12), 10315–10324
Deka et al.Article
It is also worth noting that our IONP-loaded nanogels can be
kept for months at pH 7 and at room temperature without
observing leakage of IONPs. This is likely an indication of the
entrapment of the particles within the polymer networks. Taken
oftheIONPs within thefirstlayersofthepolymernetworkinthe
nanogel. This configuration rationalizes the loading and thus the
consequent release of the IONPs that we observe. Our results are
that case, however, the authors provided other indirect proofs
that pointed to the nanoparticle entrapment.39
nanoparticles wasobserved, and noleakageofFe3þwasdetected
in the acidic medium.
experiments to load the IONPs at pH 3.5, as in these conditions we
achieved the highest efficiency of nanoparticle loading.
We have applied the procedure described above to load and
release short oligonucleotide sequences of about 25 bases. In
a sequence bearing at the 50end the fluorophore molecule Cy3,
nanogel by photoluminescence (PL) spectroscopy (Figure 6) and
UV-visible absorption spectroscopy (Figure 5S, Supporting
Information). Figure 6 shows the PL spectra of free Cy3-DNA
(red line) and ofthe nanogelsloadedwith Cy3-DNA (black line),
after the solution was purified from the excess Cy3-DNA (see
section 1.5 of the Supporting Information). When loaded in the
gels, the Cy3-DNA exhibited a PL spectrum that was red-shifted
by about 3 nm with respect to that of free Cy3-DNA.
the nanogel and not to free Cy3-DNA. As proof, we have
recorded the PL spectra after each washing step (the solution
recovered from the lower part of the centricone filter used for the
purification). The signal of free Cy3-DNA in this solution was
progressively reduced, and after 6 washing steps there was no PL
of the Cy3-DNA within the nanogel. The Cy3-DNA loading was
at pH 3.5 and measured at pH 7 (the dye signal is quenched at
pH 3.5); see Table 1S, Supporting Information).
3.5. Simultaneous Loading and Release of Oligonucleo-
tides and IONPs. In a third series of experiments, we have
loaded simultaneously IONPs and the oligonucleotide sequences
in the nanogels, by mixing together solutions of IONPs, Cy3-
DNA, and nanogels, according to the protocols described above.
The simultaneous loading of IONPs and Cy3-DNA was con-
firmed by a combination of TEM measurements, by which we
could locate the IONPs in the nanogels, and by confocal micro-
scopy, by which we could identify the PL signal from the Cy3-
DNA within the nanogel (Figure 7).
Figure 5. (A,B) Inverted bright field electron microscopy images
ofa sample ofnanogel loadedwith IONPs. (A) corresponds tothe
edge, corresponding tothe polymer,coversthe brightspots,which
are the magneticnanoparticles.(C)Cross-sectionalTEMimage of
nanogel loaded with IONPs recorded on a section having a thick-
nessof70nm(while for the insetthe section thickness isof50nm).
Figure 6. PL spectrum of Cy3-DNA loaded within the nanogel
after the cleaning procedure had been applied (black curve); PL
spectrum of the free Cy3-DNA (red curve) and starting nanogel
aliquots collected at the different washing steps, as well as those of
the loaded nanogel solution, Cy3-DNA and nanogel only. After
6 washing steps, the free Cy3-DNA was removed completely from
the solution containing the loaded nanogels. The inset shows a
scheme of the loading of Cy3-DNA within the nanogel.
Figure 7. (A) TEM characterization of nanogels loaded simulta-
neously with IONPs and Cy3-DNA. (B) Confocal microscopy
characterization of the sample shown in A. The fluorescent signal
the spots seen in the phase contrast image of the nanogel (right
panel). The central panel is a merged image of both panels.
(39) Kuang, M.; Wang, D. Y.; Bao, H. B.; Gao, M. Y.; Mohwald, H.; Jiang, M.
Adv. Mater. 2005, 17, 267.
DOI: 10.1021/la1004819Langmuir 2010, 26(12), 10315–10324
Article Deka et al.
Despite the confocal images were taken by working at the
resolution limit of the confocal setup (hence the nanogel parti-
cles could not be focalized), on dilute solutions these fluorescent
spots were colocalized with spots in the corresponding phase
contrast images, and which could be ascribed to the nanogels
Additionally, under the same experimental conditions the
TEM and DLS diameters of the nanogels simultaneously loaded
with IONPs and Cy3-DNA were bigger than those of the corres-
ponding nanogels loaded either with Cy3-DNA or with IONPs
alone (Supporting Information Table 1S). As an example, the
diameter of the loaded nanogel increased to 250 ( 50 nm (by
DLS) and the zeta potential became negative (-10.5 ( 1.5 mV).
Additional PL characterization of the nanogels loaded with Cy3-
DNA and IONPs was performed and confirmed the presence of
DNA (Supporting Information Figure 6S).
In order to release the multicargo, the nanogels were first
achieve complete release of the DNA from the nanogel, it was
necessary to keep the nanogel at pH 3.5 for 72 h. After this time,
we first separated the Cy3-DNA from the nanogel and IONP
portions by using centrifuge filters. By choosing an appropriate
pore size for the membrane, we could retain the IONPs and the
nanogel on the upper side of the membrane, while molecules like
Cy3-DNA (see inset Figure 8A) were able to pass through the
membrane. By recording the PL spectra on the fraction collected
we could still record the fluorescent signal, not only on the lower
part of the centrifuge tube, but also on the upper part of the
membrane (data not shown). Only after 72 h was a complete
release of the DNA achieved, since at this time no further PL
signal was detected on the upper side of the membrane.
These data were also supported by confocal microscopy
observations on the various aliquots that had been recovered
the IONPs were packed within the nanogels, in the confocal
fluorescence image the spots appeared point-like. In addition,
they were colocalized with spots in the phase contrast image
(Figure 8C1). After the complete release, on the upper side of the
nanogel, while no fluorescence could be recorded in the corres-
ponding channel (Figure 8C4 and C5). The portion recovered
from the lower part of the membrane still showed a fluorescent
signal. This signal, however, was not clumped any more in point-
like regions, but was rather distributed homogeneously in the
whole field of view. This could be interpreted as arising from the
In the corresponding phase contrast image, the nanogels could
Figure 8. (A)PLcharacterizationofthereleaseprocess.TheinsetisasketchshowingtheseparationonthecentrifugefilterbetweentheIONPsand
PL of Cy3-DNA recovered on the lower side of the centrifuge filter (at pH 3.5, the dye signal is quenched). (B) Corresponding TEM
characterization of the sample recovered on the upper side of the membrane. The IONPs released by the nanogels are retained on the upper
side of the filter. In addition, the nanogel structure appears damaged after the simultaneous release of both cargo elements. (C) Confocal
of the nanogels loaded with IONPs and Cy3-DNA, before the release. C2 and C3 correspond, respectively, to the phase contrast and fluorescent
visible (C5), no signal was recorded in the corresponding fluorescent channel (C4)), indicating the completed release of DNA by the nanogels.
Langmuir 2010, 26(12), 10315–10324
Deka et al.Article
the filter indicated the presence of both released IONPs and
nanogels, but the nanogels appeared disrupted in this case
(Figure 8B). These findings are somehow unique, since in all pre-
vious experiments involving either DNA or IONPs, unloading the
nanogels had retained their original shape. Apparently, the simul-
taneous release of both DNA and IONPs was responsible for this
Such irreversible swelling during unloading of both DNA and
IONPs deserved a deeper analysis. We tested therefore the effect
of the pH on the swelling of nanogels (both with and without the
cargo) by switching the pH of the medium from 8 to 3 and back.
Swelling of the empty nanogels was reversible, since the curve
describing their size dependence on the pH, when this was cycled
from3.5to8andback, didnotshowanyhysteresis(Figure7S A,
Supporting Information). These results are in agreement with
previously published data.36A similar behavior was also obser-
ved in the case of nanogels loaded with IONPs (Figure 7S C,
Supporting Information), while an appreciable hysteresis was
recorded on the nanogels loaded with Cy3-DNA (Figure 7S B,
The situation was drastically different when the nanogels were
filled with both IONPs and Cy3-DNA (Figure 7S D, Supporting
Information). This time the curve describing the size dependence
on the pH, when this was increased from 3.5 to 8 (the “forward
curve”), did not overlap with the corresponding curve when the
pH was decreased from 8 back to 3.5 (the “backward curve”).
Starting from pH 6, the nanogel size from the backward curve
was always higher than that from the forward curve, pointing
to a modification in the structure of the nanogel after it was
used as cargo. These data, together with the TEM charac-
terization, are indicative of the structural degradation of the
nanogel after the simultaneous release of Cy3-DNA and IONPs
For the quantification of DNA loaded within the nanogel,
calibration curves of PL/[DNA] (photoluminescence/DNA con-
centration) were used (Figure 8S, Supporting Information).
Using those curves, we found that, when only DNA was loaded
within the nanogel, the amount of DNA that could be actually
loaded corresponded to about 16% of the initial DNA added
(whichcorrespondedtoanamountofDNA equal to0.048pmol/
initial DNA added (0.0623 pmol/ μL of DNA for 0.053 g weight
3.6. Salt Effect on the Swelling of the Loaded Nanogels.
The swelling behavior of the nanogel was affected by the pre-
sence of salt in solution (Figure 9). We report here only the data
size of the loaded nanogels was not altered significantly by the
presence of salt. At pH 7.4, on the other hand, the loaded nano-
gels in 100 mM and 200 mM NaCl solutions were bigger than
those loaded in plain water, by about 50 and 70 nm, respecti-
vely. At pH 6.5, the loaded nanogels in 100 mM and 200 mM
NaCl solutions were affected significantly by the presence of salt
in solution, since an increase in size of 130 and 150 nm,
respectively, as compared to the sample of nanogel in water was
At pH 3.9, the differences in size were even more remarkable:
the nanogels loaded with IONP in 200 mM NaCl were again the
biggest (their diameter was around 850 nm, which corresponded
almost to their swelling limit), followed by those loaded at 100 mM
(529 nm), and was still 373 nm for the nanogel in water.
It has been reported by others40that on the vinyl pyridine-
divinyl benzene-based nanogels, the addition of salts reduces the
screening effect of the charges, resulting in a reduced swelling of
the nanogels. The higher the amount of salt added, the stronger
the screening effectis, and thus the lower the extentofswelling of
the nanogel. Also, in our case, when the nanogels were loaded
with IONPs, the swelling behavior in the presence of salt had a
trend in the pH range from 7 to 4.2 that was similar to that of
previous reports. Namely, the additionof 200 mM NaCl resulted
in a reduced swelling with respect to 100 mM. The difference
occurred for the swelling of the IONP-loaded nanogel below
4.2 in 200 mM, which is bigger than that in 100 mM. The IONPs
have charged groups at their surface, and those groups can
coordinate counterions in their surroundings. This results in a
high local ionic strength that can break the nanogel structure
and consequently increase the DLS diameter of the polymer.
under these conditions (data not shown).
The different diameters of the nanogels in a solution 100 mM
NaCl indicate that the leaking of the IONPs occurred already at
pH close to 6.5 (which is actually the pH of the extracellular
tumor environment), while no appreciable leakage was observed
at pH 7.4 (which is the pH of the blood) (Figure 9).
In this work, we have reported the fabrication of a multivalent
nanosystem based on a class of functional molecules known as
stimuli responsive polymers, which can work as cargo system for
gene (or drug) delivery, and which can entrap at the same time
inorganic magnetic nanoparticles. Differently from previously
reported studies, the magnetic nanoparticles in this work are not
covalently linked to the gel networks, and thus they can be loaded
and released by tuning the pH. The full characterization provided
when only DNA or IONPs or a combination of them were
employed has allowed us to understand both the mechanism by
which the different payloads are retained within the gel and the
The system developed in this work, especially in the case when
both DNA and IONPs havebeen loaded, has interesting features
and might find application as a therapeutic agent. It can act as a
heat mediator for performing hyperthermia, as gene delivery
system (for instance in si-RNA treatment), and at the same time
Figure 9. Salt effect on the swelling behavior of nanogels loaded
and 200 mM NaCl (violet curve).
(40) Fujii, S.; Dupin, D.; Araki, T.; Armes, S. P.; Ade, H. Langmuir 2009, 25,
10324 Download full-text
DOI: 10.1021/la1004819Langmuir 2010, 26(12), 10315–10324
Article Deka et al.
as an imaging contrast agent in MRI. The magnetic nanocarriers
developed appear to have the right geometry for performing
thosetasks. Preliminarystudiesbyother groups41-43haveshown
indeed that clustering of IONPs (like in our case when they are
loaded in the nanogels) improves the relaxivity signals recorded
with respect to individual magnetic nanoparticles. On the other
hand, hyperthermia seems more efficient when the magnetic
nanoparticles are not encapsulated within a matrix, but they are
freely delivered to a certain target site.29Our system appears to
have the right features for such purposes. When circulating in a
medium with pH below 7.4, like blood, the nanogels will be in a
packed configuration, allowing for a better enhancement of
the MRI signal form the IONPs. On the other hand, once the
like the extracellular tumor environment, they would begin to
swell, and thus they would release the IONPs. The hyperthermia
treatment could be therefore performed on the IONPs, once they
will be delivered into the extracellular tumor environment, where
the pH is around 6.5. The further uptake by tumor cells would
investigation in our laboratory.
Additionally,itisworthnoting that thenanogelsdevelopedby
spatial and temporal control. The presence of magnetic nanopar-
ticles allows spatially controlled delivery, since the nanogels feel
specific locations of the body, where the magnetic field will be
placed. Temporally controlled delivery will be ensured by the
variations in pH that the nanogels will sequentially experience
during their journey (hence by the response of the nanogels to
these variations) in the various body/cellular compartments. The
of the nanogel cargos.
Acknowledgment. This work was supported in part by the
We thank Mario Malerba for TEM sample preparation and
Sergio Marras for helpful discussion.
Supporting Information Available: Experimental details;
two tables summarizing the DLS and TEM diameters and
the zeta potential measurements; additional TEM characte-
rization images of the nanogels, additional PL and absorp-
of different loaded nanogel solution in the decreasing and
increasing pH profiles; PL/[Cy3-DNA] calibration curves.
This material is available free of charge via the Internet at
(41) Nasongkla, N.; Bey, E.; Ren, J. M.; Ai, H.; Khemtong, C.; Guthi, J. S.;
Chin, S. F.; Sherry, A. D.; Boothman, D. A.; Gao, J. M. Nano Lett. 2006, 6, 2427.
(42) Taboada, E.; Solanas, R.; Rodriguez, E.; Weissleder, R.; Roig, A. Adv.
Funct. Mater. 2009, 19, 2319.
(43) Fresnais, J.; Berret, J. F.; Petesic, B. F.; Sandre, O.; Perzynski, R. Adv. Mater.
2008, 20, 1.