Commun., 2013, 49, 5393 Red blood cells decorated with functionalized core–shell magnetic nanoparticles: elucidation of the adsorption mechanism† Thanh Duc Mai,abcFanny d’Orlye´,bChristi
Trang 1This journal is c The Royal Society of Chemistry 2013 Chem Commun., 2013, 49, 5393 5395 5393
Cite this: Chem Commun., 2013,
49, 5393
Red blood cells decorated with functionalized core–shell magnetic nanoparticles: elucidation
of the adsorption mechanism†
Thanh Duc Mai,abcFanny d’Orlye´,bChristine Me´nager,aAnne Varenne*band
The decoration of red blood cells (RBCs) with aminated and
carboxylated core–shell magnetic nanoparticles (CSMNs) was studied
and elucidated It was demonstrated that only aminated CSMNs could
decorate the RBCs and their adsorption interaction is mainly ruled by
electrostatic attraction between the positively charged amino groups on
CSMNs and the abundant sialic acid groups on the outer surface of RBCs.
The use of ferric oxide nanoparticles for theranostic applications has
gained much interest in recent years, notably as contrast agents for
magnetic resonance imaging, colloidal mediators for heat
genera-tion and magnetic drug cargos with controllable release in time and
space.1 Magnetic nanoparticles, nevertheless, still suffer from the
major problem of limited bio-distribution Some notable strategies
to achieve these nanoparticles with a longer-circulating attribute are
PEG functionalization and/or conjugation with polymersomes,
microgels and liposomes (consult ref 2 for a typical example)
A more natural and biological approach to prolong the in vivo
circulation half-life of nanoparticles, inspired from life sciences, is to
employ RBCs as theranostic vectors Recently, Kolesnikova et al gave
an account of the preparation and application of RBC-based drug
delivery vehicles in comparison with those of their synthetic polymeric
counterparts.3Applications of RBC-inspired delivery systems can be
gleaned from some selected reviews4–7 whereas some recent
bio-medical uses of magnetized erythrocytes can be referred to in
ref 8–10 To the best of our knowledge, magnetized red blood cells
so far have been produced by the trapping of magnetic nanoparticles
inside RBCs, a process in which the cell membrane is forced to distort
by osmotic stress (see ref 11 for example) The modification of the
cell’s nature after this encapsulation, notably the release of hemo-globin and the uptake of unwanted compounds into its inner compartment, as a result, is inevitable and undesirable On the other hand, the decoration of RBCs with magnetic nanoparticles seems to
be a more gentle technique With the ultimate goal of constructing
a novel biocompatible, magnetically controllable platform for diag-nostic and therapeutic applications, our group has very recently laid the groundwork for this technique by establishing some preliminary multimodal imaging demonstrations of CSMN-decorated RBCs.12 Following this pioneering work, important insights into the mecha-nism of grafting RBCs with CSMNs are reported herein
Based on the work describing the interaction of functionalized nanoparticles with giant unilamellar vesicles13and with supported lipid bilayers,14two mechanisms of adsorption of CSMNs onto RBCs are addressed The first one relies on the well-known strong affinity
of the silanol groups to the phosphatidylcholine-rich cell membrane.15,16 The second direction, which has recently been evidenced by the decoration of RBCs with hydroxyapatite,17 chito-san18or gold19nanoparticles, focuses on the general electrostatic interaction between the oppositely-charged functional groups on the concerned objects To unveil this mechanism for RBCs–CSMNs,
a series of interrelated experiments, in which the charge of CSMNs was modulated, were carried out The detailed procedure
to synthesize cationic and anionic CSMNs, as well as their charge and size characterisation, were described previously.12,13,20 Briefly, maghemite nanoparticles (g-Fe2O3, 7 nm mean physical diameter) were embedded in a fluorescent silica shell by co-condensation of tetraethylorthosilicate (TEOS)† and 3-aminopropyltriethoxysilane (APTS)† reacted with fluorescein isothiocyanate (FITC).† The silica shell functionalization was then implemented by co-condensation of 2-(methoxy(polyethyleneoxy)propyl)-trimethoxysilane (PEOS)† and APTS.† The positive charge of cationic CSMNs can be tuned by varying the APTS to PEOS molar ratios (A/P) up to 3.5 The exclusion
of APTS (A/P = 0) results in the formation of CSMNs containing only PEG with silanol functional groups on the surface Carboxylated nano-particles were produced by conversion of amino to carboxylic functions (denoted by a negative value of the A/P ratio) The mean physical and hydrodynamic diameters of CSMNs are 40 nm and 75 nm, respectively In MOPS ((ionic strength I = 100 mM)† and sucrose
a
UPMC University of Paris 06-CNRS-ESPCI Laboratoire Physicochimie des
Electrolytes, Colloı¨des et Sciences Analytiques PECSA UMR 7195, 4 place Jussieu,
75252 Paris, France E-mail: christine.menager@courriel.upmc.fr
b Chimie ParisTech, Ecole Nationale Supe´rieure de Chimie de Paris, Imagery,
Chemical and Genetic Pharmacology Unit (UPCGI), UMR CNRS 8151 – U INSERM
1022, 11, rue Pierre et Marie Curie, 75231 Paris Cedex 05, France.
E-mail: anne-varenne@chimie-paristech.fr
c Centre for Environmental Technology and Sustainable Development (CETASD),
Hanoi University of Science, Nguyen Trai Street 334, Hanoi, Viet Nam.
E-mail: maithanhduc83@gmail.com
† Electronic supplementary information (ESI) available: More experimental
details and supporting figures See DOI: 10.1039/c3cc41513a
Received 27th February 2013,
Accepted 18th April 2013
DOI: 10.1039/c3cc41513a
www.rsc.org/chemcomm
ChemComm
COMMUNICATION
View Article Online View Journal | View Issue
Trang 25394 Chem Commun., 2013, 49, 5393 5395 This journal is c The Royal Society of Chemistry 2013
(concentration C = 50 mM)† (pH 7.4)), CSMNs are stable for months
The average concentration of CSMNs is 6 1017particles per L
CSMNs bearing different functional groups exhibit a
varia-tion in z potentials from 14 to +13 mV as shown in Fig 1 A
higher A/P ratio leads to a more positive charge,20which results
in a higher z potential The z potential of intact RBCs is around
10 mV The decoration of RBCs with cationic CSMNs leads to
a higher z potential A level-off of the RBCs–CSMNs’ z potential
curve was observed at A/P ratios higher than 1.75, leading to a
value of +18 mV The presence of CSMNs bearing PEG–silanol
or carboxylic groups (negative charges) does not affect the
charge status of the surface of the erythrocytes
The hybrid objects were then visualized using fluorescent
micro-scopy The inclusion of FITC† into the shell layer of nanoparticles
allows the tracking and localization of CSMNs on the surface of
erythrocytes by observing the green fluorescence The RBCs after having
been in contact with CSMNs for about 15 min were pictured as in
Fig 2 The absence of interaction between anionic CSMNs and the cell
membrane was evidenced by the lack of illumination (Fig 2A and B)
On the other hand, the decoration of RBCs with cationic nanoparticles
results in an observable green coating over the cell surface (Fig 2C and
D) The obtained light intensity is correlated to the number of CSMNs
on RBCs It thus indicates that the RBCs–CSMNs interactivity is
proportional to the density of amino groups on the CSMNs The
exterior of grafted RBCs was then zoomed in using transmission
electron microscopy (TEM) The more CSMNs are adsorbed on the
membrane façade, the better electronic contrast can be achieved As
can be seen in Fig 3, the distribution of CSMNs of smaller A/P is more
scattered, reflected by a less contrast imaging capture
Clearly, the RBC–CSMN interaction occurs only with nano-particles possessing a positive charge due to the amino groups whereas those with negative charge induce no adsorption The well-studied high affinity of silanol groups to phosphatidylcholine on the cell membrane does not trigger any decoration in this case It is very possibly because such affinity is suppressed due to the presence of PEG on the outer layer of CSMNs The addition of PEG to the shell of nanoparticles, albeit limiting the hemolytic activity thanks to its biocompatibility,21–23 and preventing these tiny particles from translocation into the bilayer membrane,13 diminishes the accessibility of surface silica-moieties to the outer layer of the erythrocytes, as already described elsewhere.16,24 Although phosphatidylcholine (PC), sphingomyelin (SPH) and cholesterol are the main components of the outer lipid membrane of erythrocytes, their surface charge is mainly due to the carboxyl groups of N-acetylneuraminic (sialic) acid residues
in glycoproteins of the external surface.25,26 This negatively charged surface of RBCs facilitates the immobilization of cationic nanoparticles, and at the same time hinders any approach of carboxylic-functionalized counterparts
The interaction was then carried out with both intact erythrocytes and those with a diminished density of sialic groups The diminu-tion of sialic groups was implemented by enzymatic treatment of RBCs with neuraminidase – a selective enzyme for breakdown of the sialic group binding.26,27After enzymatic treatment, an increase in z
interaction of these enzyme-treated RBCs with cationic CSMNs, referenced to that of intact RBCs, is interpreted in terms of z potential as shown in Fig 4 The enzymatic treatment leads to a less decoration degree, as observed with confocal microscopy in Fig 5 The fluorescence intensity was higher intact RBC–CSMN
Fig 1 z Potentials of cationic (positive A/P ratio) and anionic (negative A/P
ratio) CSMNs, as well as intact and CSMN-grafted RBCs according to the A/P ratio.
Dispersion medium: MOPS (I = 100 mM) and sucrose (C = 50 mM).† Each
experimental point is the mean of 4 replicates, and the error bars stand for
one standard deviation.
Fig 2 Optical fluorescence images of RBCs after being in contact for 15 min with
(A) carboxylic-functionalized CSMNs, (B) PEG-functionalized CSMNs (A/P = 0),
(C and D) amino-functionalized CSMNs (A/P = 1 and 1.75, respectively).
Fig 3 TEM micrographs of RBCs after being in contact for 15 min with amino-functionalised CSMNs of (A) 1.00 A/P; (B) 1.75 A/P.
Fig 4 z Potentials of intact and enzyme-treated erythrocytes before and after adsorption of cationic CSMNs.
Trang 3This journal is c The Royal Society of Chemistry 2013 Chem Commun., 2013, 49, 5393 5395 5395
(Fig 5A and B) whereas weak illumination was observed from
enzyme-treated RBC–CSMN (Fig 5C and D) If the brightness
from intact RBC–CSMN of 1.75 A/P is assigned as 100%, then those
from intact RBC–CSMN of 1 A/P, enzyme-treated RBC–CSMN of
1.75 A/P, and enzyme-treated RBC–CSMN of 1 A/P are 50%, 56%
and 3%, respectively (data obtained from the image-processing
package ImageJ) The reduced interaction of cationic CSMNs
with neuraminidase-treated RBCs confirms the theory that
such decoration is mainly ruled by electrostatic attraction
between sialic acid and amino functional groups The
inter-action between these functional groups was also evidenced by
Delcea et al., using Raman spectroscopy.19
The in vitro stability of the hybrid object, in terms of
desorption of CSMNs from RBCs and hemolytic activity, was
then evaluated Our experimental data showed that within
2 hours in the dispersion medium, CSMNs were still tightly
attached to the erythrocyte membrane (see Fig S1, ESI†) A
drastic reduction of the hybrid object’s z potential was observed
after 5 hours due to self-stripping of CSMNs from RBCs On the
other hand, when the hybrid object was stored with a surfeit of
CSMNs, no significant change in z potential was observed over
24 hours (data not shown) These results indicate that the
interaction equilibrium is shifted to the decoration process in
the presence of an excess of CSMNs in the medium The
hemolysis of these magnetic RBCs was accordingly tested over
8 hours during which the decoration is still observable Our
results (see Fig S2, ESI†) showed that no hemolysis was
induced by CSMNs of A/P = 1 during this period, which is
relevant to the results reported by Laurencin et al.12However,
CSMNs of higher A/P ratios were found to cause hemolysis after
4 hours of incubation Cationic nanoparticles with a higher
amino functional density provoked a more significant lysis of
cells It seems that the more the adsorption process occurs, the
more the rupture of the RBC membrane is induced Indeed, a
similar phenomenon was already observed for the interaction
of silica-based moieties with RBCs (see ref 28 and some
references listed therein) Thus the subsequent design of the
cationic-functionalized CSMNs for decoration should be
improved according to the following objectives: (1) high
ima-ging contrast by employing large magnetic cores and a high
FITC concentration in the shell, (2) high magnetization degree
by using large magnetic cores, and (3) high stability and low
hemolysis of the magnetic RBCs by tuning the CSMNs amino
group density and nature
It was demonstrated that the interaction between RBCs and CSMNs is mainly ruled by electrostatic attraction The facile decoration opens the floor for some possible applications at hand, such as quick detection of haemorrhage and monitoring of the healing processes Quantification of this adsorption interaction, i.e determination of the binding constant, as the ground work for any further optimization, will be soon carried out
This work was supported by the fellowship for prospective researchers (Grant No PBBSP2_141401) from the Swiss National Science Foundation The authors thank Aude Michel for technical assistance and electron microscopy operation Notes and references
1 S Laurent, D Forge, M Port, A Roch, C Robic, L V Elst and
R N Muller, Chem Rev., 2008, 108, 2064–2110.
2 M R Preiss and G D Bothun, Expert Opin Drug Delivery, 2011, 8, 1025–1040.
3 T A Kolesnikova, A G Skirtach and H Mo ¨hwald, Expert Opin Drug Delivery, 2013, 10, 47–58.
4 R H Fang, C M J Hu and L F Zhang, Expert Opin Biol Ther.,
2012, 12, 385–389.
5 C.-M J Hu, R H Fang and L Zhang, Adv Healthcare Mater., 2012, 1, 537–547.
6 S Biagiotti, M F Paoletti, A Fraternale, L Rossi and M Magnani, IUBMB Life, 2011, 63, 621–631.
7 V R Muzykantov, Expert Opin Drug Delivery, 2010, 7, 403–427.
8 C Cinti, M Taranta, I Naldi and S Grimaldi, PLoS One, 2011,
6, e17132.
9 D E Markov, H Boeve, B Gleich, J Borgert, A Antonelli, C Sfara and M Magnani, Phys Med Biol., 2010, 55, 6461–6473.
10 A Antonelli, C Sfara, E Manuali, I J Bruce and M Magnani, Nanomedicine, 2010, 6, 211–223.
11 M Brahler, R Georgieva, N Buske, A Muller, S Muller,
J Pinkernelle, U Teichgraber, A Voigt and H Baumler, Nano Lett.,
2006, 6, 2505–2509.
12 M Laurencin, T Georgelin, O Cle ´ment, G Autret, J.-M Siaugue and
C Me ´nager, Adv Healthcare Mater., 2013, DOI: 10.1002/ adhm.201200384.
13 M Laurencin, T Georgelin, B Malezieux, J.-M Siaugue and
C Me ´nager, Langmuir, 2010, 26, 16025–16030.
14 X F Zhang and S H Yang, Langmuir, 2011, 27, 2528–2535.
15 I I Slowing, C.-W Wu, J L Vivero-Escoto and V S Y Lin, Small,
2009, 5, 57–62.
16 Y N Zhao, X X Sun, G N Zhang, B G Trewyn, I I Slowing and
V S Y Lin, ACS Nano, 2011, 5, 1366–1375.
17 Y C Han, X Y Wang, H L Dai and S P Li, ACS Appl Mater Interfaces, 2012, 4, 4616–4622.
18 W Fan, W Yan, Z S Xu and H Ni, Colloids Surf., B, 2012, 95, 258–265.
19 M Delcea, N Sternberg, A M Yashchenok, R Georgieva,
H Baumler, H Mohwald and A G Skirtach, ACS Nano, 2012, 6, 4169–4180.
20 F d’Orlye, A Varenne, T Georgelin, J.-M Siaugue, B Teste,
S Descroix and P Gareil, Electrophoresis, 2009, 30, 2572–2582.
21 K Knop, R Hoogenboom, D Fischer and U S Schubert, Angew Chem., Int Ed., 2010, 49, 6288–6308.
22 J M Harris and R B Chess, Nat Rev Drug Discovery, 2003, 2, 214–221.
23 F M Veronese and G Pasut, Drug Discovery Today, 2005, 10, 1451–1458.
24 Q He, J Zhang, J Shi, Z Zhu, L Zhang, W Bu, L Guo and Y Chen, Biomaterials, 2010, 31, 1085–1092.
25 E H Eylar, M A Madoff, J L Oncley and O V Brody, J Biol Chem.,
1962, 237, 1992–2000.
26 Y Suzuki, N Tateishi and N Maeda, Biorheology, 1998, 35, 155–170.
27 F R Gaffar, F F J Franssen and E de Vries, Int J Parasitol., 2003,
33, 1595–1603.
28 J Shi, Y Hedberg, M Lundin, I O Wallinder, H L Karlsson and
L Moller, Acta Biomater., 2012, 8, 3478–3490.
Fig 5 Confocal micrographs of an enzyme-treated and an intact erythrocyte
after decoration with cationic CSMNs Intact RBCs–CSMNs of (A) 1.75 A/P and (B)
1.00 A/P; neuraminidase-treated RBCs–CSMNs of (C) 1.75 A/P and (D) 1.00 A/P.