Conversion of magnetic impulses into cellular responses by self-assembled nanoparticle-vesicle hydrogels

Article abstract of DOI:10.1002/anie.201103469

Crosslinking vesicles with magnetic nanoparticles produced MNPVs, self-assembled "nanopills" that can be "unlocked" by an alternating magnetic field (AMF), releasing chemical messengers stored within the vesicles. When MNPVs are co-immobilized with cells

Full text of DOI:10.1002/anie.201103469

Communications
DOI: 10.1002/anie.201103469
Magnetoresponsive Biomaterials
Conversion of Magnetic Impulses into Cellular Responses by
Self-Assembled Nanoparticle–Vesicle Hydrogels**
Felicity de Cogan, Andrew Booth, Julie E. Gough,* and Simon J. Webb*
Phospholipid vesicles are widely used as nano-sized drug
delivery vehicles[1] and as biomimetic model systems,[2] where
the bilayer allows fundamental biomembrane processes like
ion transport, signaling, and multivalent recognition to be
copied.[3] In particular, the remotely triggered transit of stored
chemicals across bilayer membranes is a key goal as it will
allow vesicles to communicate with cells either in vivo or in
vitro during cell culture.[4] The latter approach should produce
exciting new “smart” biomaterials, although non-invasive and
non-chemical control over drug release from vesicles remains
challenging.
Most mammalian cells are unaffected by oscillating or
permanent magnetic fields. To sensitize cells to magnetic
fields, they can be labeled with magnetic nanoparticles
(MNPs),[5] an approach used to effect gene transfection with
static magnetic fields[6] or cause hyperthermia with alternat-
ing magnetic fields (AMFs).[7] Alternatively MNPs can be
used to label vesicles, which allows magnetic manipulation
and AMF-triggered contents release; magnetic release is
attractive as nearby cells would only be affected by the
released biochemicals and not the AMF. Recently we used
10 nm Fe3O4 MNPs to crosslink 800 nm diameter phospho-
lipid vesicles and form magnetic nanoparticle–vesicle assem-
blies (MNPVs).[8] Embedding MNPVs within a hydrogel
matrix added a further level of assembly, with the hydrogel
fibrils acting as an artificial extracellular matrix that rein-
forced the vesicles, providing robust materials[9] that
responded to AMFs by releasing stored dyes. Such nano-
structured and responsive self-assembled biomaterials have
enormous potential in cell culture,[10] and replacing these dyes
with bioactive species should produce a new type of “smart”
cell culture scaffold[11] responsive to magnetic impulses.
Herein we describe a self-assembled bionanotechnological
system able to act as a “smart” biomaterial that translates
non-invasive magnetic signals into cellular responses
(Figure 1).
Figure 1. a) Cells and nanoparticle-vesicle assemblies (MNPVs) are co-
immobilized within a calcium alginate hydrogel (yellow). MNPVs are
self-assembled nanocarriers composed of magnetic nanoparticles
coated with N-biotinoyl dopamine 1 (1-MNP) and DPPC vesicles
containing biotin-DHPE 2 (2-DPPC), which are linked together by
avidin. b) Chemical messengers, such as drugs (blue), can be non-
invasively released by an alternating magnetic field (AMF), and these
released chemicals in turn induce responses from cultured cells.
An important design feature was the self-assembly of
Fe3O4 nanoparticles with gel-phase vesicles, an alternative to
physical incorporation[12] that was designed to allow heat
generated in the MNPs by the AMF to be efficiently
transferred to the bilayers.[13] When heated, gel-phase vesicles
“melt” at a triggering temperature (Tm), an all-or-nothing
event that allows complete and rapid escape of encapsulated
compounds. The biotin–avidin interaction was used to link
vesicles and MNPs, which improved compatibility across cell
types, including myoblasts (Figure 2) and chondrocytes. It
also allowed commercially available biotin lipids like N-
(biotinoyl)-1,2-dihexadecanoyl
phosphatidylethanolamine
(biotin-DHPE) to be used as vesicle crosslinkers.[14]
N-Biotinylated dopamine (1)[15] was used to give Fe3O4
MNPs an adhesive coating. MNPs were formed by co-
precipitation[16] then sonicated with 1 in deoxygenated
methanol (0.7 mm) to give 1-coated Fe3O4 nanoparticles ([1-
MNP]), with a coating efficiency of 50 Æ 20% (Figure 2a).
Dipalmitoyl phosphatidylcholine (DPPC) vesicles (800 nm
diameter) were chosen as the nanocontainers as these bilayers
have Tm ꢀ 428C,[17] a triggering temperature above cell
culture conditions (378C). Vesicles with stored chemical
payloads were created by extrusion of 0.2% mol/mol biotin-
DHPE 2 in DPPC in a solution of the compound to be
encapsulated. The thermal release of encapsulated 5/6-
carboxyfluorescein (5/6-CF) showed these [2-DPPC] vesicles
had Tm ꢀ 408C. Addition of the avidin “glue” to a mixture of
[1-MNP] and [2-DPPC] vesicles produced large magnetic
[*] F. de Cogan, A. Booth, Dr. S. J. Webb
School of Chemistry and Manchester Interdisciplinary Biocentre
The University of Manchester
131 Princess St, Manchester M1 7DN (UK)
E-mail: s.webb@manchester.ac.uk
Homepage: http://www.webblab.org
F. de Cogan, Dr. J. E. Gough
School of Materials, Materials Science Centre
The University of Manchester, Manchester, M13 9PL (UK)
[**] This work was supported by the award of a BBSRC Doctoral Training
Grant to F.C. We thank Dr. R. Collins and Dr. A. Harvey for TEM, and
Dr. L. Carney for cell culture.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201103469.
12290
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 12290 –12293

the hydrogel after magnetic release but prior to diffusion out
of the gel block was calculated as ca. 100 mm, which should
allow biochemicals with micromolar Kd values to induce
cellular responses.
Biomaterials composed of chondrocytes encased within
calcium alginate have shown promise as engineered scaffolds
to replace articular cartilage,[21] with the cells producing
cartilage markers such as glycosaminoglycans and collagen
II.[22] For the production of extracellular collagen by chon-
drocytes and osteoblasts, which normally occurs over several
days, ascorbic acid is an essential additive.[23] Therefore either
ascorbate (150–300 mm)[24] or ascorbic acid-2-phosphate
(AAP, a polar and air-insensitive precursor that is readily
converted to ascorbate by cells) are typically added to
chondrocyte cell culture media.
The response of chondrocytes to AAP was the ideal
model system to illustrate magnetically-induced changes in
cellular behavior. Chondrocytes show good tolerance for
calcium alginate matrices, with the inert 3D alginate matrix
maintaining their rounded morphology and therefore their
phenotype.[22] Furthermore, adding MNPVs ( ꢀ 20 mm
DPPC) did not significantly affect the growth of chondrocytes
in the gel (Figure 3a).[19b]
Figure 2. a) TEM image of [1-MNP] (scale bar 100 nm). b) Fluores-
cence microscopy images of MNPVs formed from [1-MNP], rhod-
amine-labeled [2-DPPC] (0.1% rhodamine–dipalmitoyl phosphatidyle-
thanolamine) and fluorescein isothiocyanate (FITC)–avidin (left: rhod-
amine channel, right: fluorescein channel, scale bar 20 mm). c) Con-
focal microscopy image of myoblasts and rhodamine-labeled MNPVs
co-immobilized in alginate (scale bar 25 mm). Cells stained with 4’,6-
diamidino-2-phenylindole (DAPI, nuclei) and FITC-conjugated phalloi-
*
din (f-actin). d) 5/6-CF release at ( ) 208C (&) at 208C after
*
incubation at 508C (4 min) ( ) at 208C after exposure to an AMF
(4 min).
nanoparticle–vesicle assemblies (MNPVs); using fluorescein-
labeled avidin with rhodamine-labeled vesicles revealed these
components were co-localized in the MNPVs (Figure 2b).[18]
These MNPVs were susceptible to external magnetic fields, a
useful feature that allowed magnetic sedimentation and
separation from non-encapsulated material. Six-fold repeti-
tion of magnetic sedimentation, supernatant removal (60–
80% of the total volume), and re-suspension in buffer could
remove 99.5% of non-encapsulated material.
To obtain robust biomaterials, these MNPVs were immo-
bilized in a calcium alginate matrix. This hydrogel allows
these smart biomaterials to be compatible with many cell
types, particularly chondrocytes,[19] while being easy to form
and manipulate at physiological temperatures. Purified
MNPVs were magnetically sedimented then re-suspended in
2% wt/vol sodium alginate in phosphate buffered saline
(PBS). The mixture was then gelled by infusion of CaCl2
(0.1m) through a polycarbonate membrane; adding cells prior
to gelation produced materials with cells and MNPVs co-
immobilized. Confocal microscopy showed cells surrounded
by the smaller MNPVs (Figure 2c); adhesive interactions
between cells and MNPVs occur if they are co-incubated for
> 4 h before gelation.[20]
The efficiency of magnetic release was quantified using
encapsulated 5/6-CF. Fluorescence spectroscopy indicated
little release of 5/6-CF (0.05m inside the vesicles) occurred
during gelation and the MNPVs largely remained intact.
However 5/6-CF rapidly escaped from a gel block (25 mm3)
after exposure to a 392 kHz AMF pulse (240 s) (Figure 2d).
The amount of 5/6-CF that escaped from the gel was
quantified after 2 h, which gave 1.3 mm 5/6-CF in the solution
covering the gel block (2 mL); without an AMF only 35%
escaped after 3 days. The corresponding concentration within
Figure 3. a) Confocal microscopy image of chondrocytes and unla-
belled MNPVs co-immobilized in alginate (scale bar 100 mm). b) Ascor-
bic acid-2-phosphate (AAP) release from MNPVs immobilized in
*
*
alginate at ( ) 208C (&) at 208C after incubation at 508C (4 min) ( )
at 208C after exposure to an AMF (4 min).
The encapsulation and release of AAP (8.6 mm) from
alginate-embedded MNPVs was demonstrated in a similar
manner to 5/6-CF. A pulse of AMF (240 s) released encapsu-
lated AAP from a 25 mm2 gel block, and the concentration of
the AAP in the surrounding PBS solution (2 mL) was
determined by phosphatase hydrolysis followed by titration
with triiodide (Figure 3b).[25] This showed that the concen-
tration in the gel block after release was ca. 400 mm, which
should be high enough to induce a strong cellular response.
The release rate of anionic AAP[26] at pH 7.4 was akin to that
observed for 5/6-CF, with 40% release after 0.5 h and 80%
release after 1.5 h (compared to incubation at 508C). This
suggests that the passage of bioactive anions through calcium
alginate is not significantly impeded by the hydrogel matrix.
To measure the response of chondrocytes cultured in
these AAP-loaded MNPV hydrogels to an AMF, alginate gels
containing chondrocytes (25000 cells per 500 mm3) and
MNPVs ( ꢀ 20 mm DPPC) were fabricated in glass vials,
while control hydrogels containing [2-DPPC] vesicles with
Angew. Chem. Int. Ed. 2011, 50, 12290 –12293
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 12291

Communications
stored AAP but no 1-MNP were also formed. Media (1 mL)
releasing the contents of phospholipid vesicles. Crosslinking
thermally-sensitive DPPC vesicles with magnetically-respon-
sive Fe3O4 nanoparticles allowed remote magnetic release of
the vesicle contents, which occurred at low loadings of Fe3O4
and did not otherwise affect cells within the material. An
alginate gel was the synthetic extracellular matrix around the
MNPVs, producing robust materials suitable for cell culture.
We found that encapsulating polar compounds at ca. 10 mm
within the vesicles gave sufficient concentrations after
magnetic release to generate cellular responses. The exciting
potential of these materials was exemplified using magnetic
release of AAP, which switched on collagen production by
chondrocytes and added this extracellular matrix protein to
the alginate scaffold. The transparency of tissue to magnetic
fields means that cellular responses could be initiated non-
invasively in vivo, as recently illustrated with in vivo magnetic
release of cells from ferrogel-based scaffolds.[29] Given
external magnetic fields can pattern MNPVs within these
materials, we hope in future to initiate cellular responses that
are both spatially and temporally controlled.
was added to cover the gel and the samples incubated at 378C.
Collagen production was measured at regular intervals over
14 days using the Sircol collagen assay. After an initial lag
period following application of the AMF, the chondrocytes in
the gel responded to the magnetic release of AAP by
producing collagen at levels comparable to literature studies
(up to 6 mgmLÀ1, Figure 4a).[27] Little cell response was
observed with uncoated MNP and [2-DPPC], showing an
adhesive link between both components of the MNPVs is
needed to produce a cellular response to the AMF.
Received: May 19, 2011
Revised: August 16, 2011
Published online: October 25, 2011
Keywords: biomaterials · magnetoresponsive materials ·
.
nanoparticles · self-assembly · vesicles
[1] V. P. Torchilin, Nat. Rev. Drug Discovery 2005, 4, 145 – 160.
[2] F. M. Menger, M. I. Angelova, Acc. Chem. Res. 1998, 31, 789 –
797.
[3] a) N. Sakai, J. Mareda, S. Matile, Acc. Chem. Res. 2005, 38, 79 –
87; b) P. Barton, C. A. Hunter, T. J. Potter, S. J. Webb, N. H.
Williams, Angew. Chem. 2002, 114, 4034 – 4037; Angew. Chem.
Int. Ed. 2002, 41, 3878 – 3881; c) C. C. Hayden, J. S. Hwang, E. A.
Abate, M. S. Kent, D. Y. Sasaki, J. Am. Chem. Soc. 2009, 131,
8728 – 8729.
[4] M. Biondi, F. Ungaro, F. Quaglia, P. A. Netti, Adv. Drug Delivery
Rev. 2008, 60, 229 – 242.
&
Figure 4. a) AMF-induced production of collagen by chondrocytes: ( )
&
calcium alginate only; ( ) AAP-containing [2-DPPC] vesicles in calcium
[5] a) J. Gao, H. Gu, B. Xu, Acc. Chem. Res. 2009, 42, 1097 – 1107;
b) J. Riegler, J. A. Wells, P. G. Kyrtatos, A. N. Price, Q. A.
Pankhurst, M. F. Lythgoe, Biomaterials 2010, 31, 5366 – 5371.
[6] J. Dobson, Nat. Nanotechnol. 2008, 3, 139 – 143.
[7] a) Q. A. Pankhurst, N. K. T. Thanh, S. K. Jones, J. Dobson,
J. Phys. D 2009, 42, 224001; b) J.-P. Fortin, C. Wilhelm, J. Servais,
C. Mꢀnager, J.-C. Bacri, F. Gazeau, J. Am. Chem. Soc. 2007, 129,
2628 – 2635; c) F. Sonvico, S. Mornet, S. Vasseur, C. Dubernet, D.
Jaillard, J. Degrouard, J. Hoebeke, E. Duguet, P. Colombo, P.
Couvreur, Bioconjugate Chem. 2005, 16, 1181 – 1188; d) A. Ito,
Y. Kuga, H. Honda, H. Kikkawa, A. Horiuchi, Y. Watanabe, T.
Kobayashi, Cancer Lett. 2004, 212, 167 – 175.
[8] a) K. P. Liem, R. J. Mart, S. J. Webb, J. Am. Chem. Soc. 2007, 129,
12080 – 12081; b) R. J. Mart, K. P. Liem, S. J. Webb, Pharm. Res.
2009, 26, 1701 – 1710.
[9] R. J. Mart, K. P. Liem, S. J. Webb, Chem. Commun. 2009, 2287 –
2289.
&
alginate; ( ) AAP-containing [2-DPPC] vesicles with uncoated MNP in
&
calcium alginate ( ) AAP-containing MNPVs in calcium alginate.
b,c) Confocal microscopy images of alginate gels containing chondro-
cytes and MNPVs, with fluorescent staining of cell nuclei (blue). Top:
stained for collagen II (stained red with tetramethyl rhodamine
isothiocyanate (TRITC)–secondary antibody). Bottom: stained for colla-
gen I (stained green with FITC–secondary antibody).
The production of collagen by chondrocytes in response to
the magnetic signal was visualized after 14 days using
fluorescence microscopy on gels containing AAP-loaded
MNPVs. Little collagen was produced in samples that had
not been exposed to an AMF (Figure 4b). However, large
amounts of extracellular collagen I and collagen II were
produced in samples exposed to an AMF, which formed a
mesh around the chondrocytes (Figure 4c). This response
shows the potential of these 3D cell culture scaffolds for non-
invasive production of cartilage-like extracellular matrices.[28]
In summary, we have developed a self-assembled and
biomimetic biomaterial that responds to a magnetic signal by
[10] a) J. H. Collier, Soft Matter 2008, 4, 2310 – 2315; b) S. I. Stupp,
Nano Lett. 2010, 10, 4783 – 4786; c) R. J. Mart, R. D. Osborne,
M. M. Stevens, R. V. Ulijn, Soft Matter 2006, 2, 822 – 835;
d) A. R. Hirst, B. Escuder, J. F. Miravet, D. K. Smith, Angew.
Chem. 2008, 120, 8122 – 8139; Angew. Chem. Int. Ed. 2008, 47,
8002 – 8018.
12292 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 12290 –12293

[11] a) D. G. Anderson, J. A. Burdick, R. Langer, Science 2004, 305,
1923 – 1924; b) M. P. Lutolf, F. E. Weber, H. G. Schmoekel, J. C.
Schense, T. Kohler, R. Mꢁller, J. A. Hubbell, Nat. Biotechnol.
2003, 21, 513 – 518; c) M. W. Tibbitt, K. S. Anseth, Biotechnol.
Bioeng. 2009, 103, 655 – 663.
[12] a) G. Beaune, M. Levy, S. Neveu, F. Gazeau, C. Wilhelm, C.
Mꢀnager, Soft Matter 2011, 7, 6248 – 6254; b) S. Nappini, F. B.
Bombelli, M. Bonini, B. Nordꢂn, P. Baglioni, Soft Matter 2010, 6,
154 – 162.
[13] Hyperthermia assays suggest release occurs through MNP
heating by Nꢀel or Brown relaxation, but membrane disruption
due to MNP oscillations could also be possible.
[14] S. Chiruvolu, S. Walker, J. Israelachvili, F.-J. Schmitt, D.
Leckband, J. A. Zasadzinski, Science 1994, 264, 1753 – 1756.
[15] H. Gu, Z. Yang, J. Gao, C. K. Chang, B. Xu, J. Am. Chem. Soc.
2005, 127, 34 – 35.
Leng, J. E. Gough, S. J. Webb, Mater. Res. Soc. Symp. Proc. 2010,
1272, 1272-PP07-02.
[20] F. de Cogan, J. E. Gough, S. J. Webb, J. Mater. Sci. Mater. Med.
2011, 22, 1045 – 1051.
[21] T. A. E. Ahmed, M. T. Hincke, Tissue Eng. Part B 2010, 16, 305 –
329.
[22] S. Loty, J.-M. Sautier, C. Loty, H. Boulekbache, T. Kokubo, N.
Forest, J. Biomed. Mater. Res. 1998, 42, 213 – 222.
[23] H. Lodish, A. Berk, C. A. Kaiser, M. Krieger, M. P. Scott, A.
Bretscher, H. Ploegh, P. T. Matsudaira, in Molecular Cell
Biology, 6th ed., W. H. Freeman, New York, 2008, p. 826.
[24] a) P. S. Leboy, L. Vaias, B. Uschmann, E. Golub, S. L. Adams, M.
Pacifici, J. Biol. Chem. 1989, 264, 17281 – 17286; b) J. C. Daniel,
B. U. Pauli, K. E. Kuettner, J. Cell Biol. 1984, 99, 1960 – 1969.
ˇ
ˇ
ˇ
˙
˙
˙
[25] B. Kazakeviciene, G. Valincius, M. Kazemekaite, V. Razumas,
Electroanalysis 2008, 20, 2235 – 2240.
[16] S. Laurent, D. Forge, M. Port, A. Roch, C. Robic, L. Vander Elst,
R. N. Muller, Chem. Rev. 2008, 108, 2064 – 2110.
[17] J. Hjort Ipsen, G. Karlstrçm, O. G. Mouritsen, H. Wennerstrçm,
M. J. Zuckermann, Biochim. Biophys. Acta Biomembr. 1987,
905, 162 – 172.
[18] Several other assembly outcomes may also occur, including
crosslinking of biotin residues across the surface of MNPs or
vesicles, which could lower release efficiency.
[19] a) G. Klçck, A. Pfeffermann, C. Ryser, P. Grohn, B. Kuttler, H. J.
Hahn, U. Zimmermann, Biomaterials 1997, 18, 707 – 713; b) F.
[26] The first two pKa values of monoalkylphosphoric acids are ca. 1
and 6. See: W. D. Kumler, J. J. Eiler, J. Am. Chem. Soc. 1943, 65,
2355 – 2361.
[27] Chondrocytes have been reported as producing 5 mgmLÀ1
collagen in an alginate hydrogel. See: G. M. Williams, T. J.
Klein, R. L. Sah, Acta Biomater. 2005, 1, 625 – 633.
[28] J. Kisiday, M. Jin, B. Kurz, H. Hung, C. Semino, S. Zhang, A. J.
Grodzinsky, Proc. Natl. Acad. Sci. USA 2002, 99, 9996 – 10001.
[29] X. Zhao, J. Kim, C. A. Cezar, N. Huebsch, K. Lee, K. Bouhadir,
D. J. Mooney, Proc. Natl. Acad. Sci. USA 2011, 108, 67 – 72.
Angew. Chem. Int. Ed. 2011, 50, 12290 –12293
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 12293