Superparamagnetic core-shell nanoparticles as solid supports for peptide synthesis

Article abstract of DOI:10.1039/c2cc33492e

Functional superparamagnetic core-shell nanoparticles are synthesized by a microwave assisted route and can be used as colloidal supports for peptide synthesis in quasi solution .

Full text of DOI:10.1039/c2cc33492e

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 7176–7178
www.rsc.org/chemcomm
COMMUNICATION
Superparamagnetic core–shell nanoparticles as solid supports for peptide
synthesisw
Christian Stutz,a Idalia Bilecka,b Andreas F. Thunemann,c Markus Niederbergerb and
¨
a
¨
Hans G. Borner*
Received 14th May 2012, Accepted 28th May 2012
DOI: 10.1039/c2cc33492e
Functional superparamagnetic core–shell nanoparticles are
synthesized by a microwave assisted route and can be used as
colloidal supports for peptide synthesis in ‘‘quasi solution’’.
In 1963 Merrifield et al. introduced the method of solid-phase
supported synthesis and thus revolutionized peptide synthesis.1
Nowadays, solid-phase supported peptide synthesis (SPPS) is a
fully automated process with commercially available reagents
and a-amino acid derivatives.2 SPPS provided models to
deepen insight into protein function and paved the way to
peptide drugs.3 Even material scientists used this platform
to synthesize sequence-defined polymers for bioinspired soft-
matter science.4
Fig. 1 Synthesis of core–shell nanoparticles with amino functionalities
at the surface. (i) TEOS, NH3/H2O/ethanol, 60 1C, 3 min, microwave;
(ii) APTMS, NH3/H2O/ethanol, 60 1C, 3 min, microwave; (iii) Fmoc-
b-Ala OH, PyBOP/DIPEA/NMP, 3 h; (iv) piperidine/NMP, 5 min;
(v) Fmoc-Rink-Linker OH, PyBOP/DIPEA/NMP, 3 h.
Practically all reagents and components used in SPPS
were optimized. For instance, the solid supports underwent
adaptation from the first modified ion-exchange resins to
modern hybrid microparticles e.g. TentaGels resins.3b,5
However, conceptually not much has been changed compared
to the first proof-of-principle study. Still lightly cross linked
poly(styrene) resins of 70–120 mm dominate the used supports.
Certainly, established microgels have high capacity and allow
ease of purification by filtration. However, the accessibility of
functionalities requires particle swelling and the diffusion of
reagents within the microgel is limited. Solid 100 mm glass
beads exhibit improved accessibility to surface functionalities,
but suffer from low capacity. Nanoparticles overcome this
problem as e.g. B70 nm particles span a B1.1 ꢀ 105-times
higher surface area. Routes to nanoparticles of various
materials are established, offering large scale access.6 However,
filtration procedures are not practical if nanoparticles were applied
as supports. Within recent decades, magnetic sedimentation of
colloids by external magnetic fields has become a useful tool to
selectively pull-down compounds out of complex mixtures.6a,7
Only few reports on the peptide synthesis with magnetic supports
appeared, but the fundamental concept of SPPS supports was
not evolved.8
Here, we report on surface amino functionalized, super-
paramagnetic nanoparticles with a protective silica shell to
be applicable as colloidal supports for peptide synthesis in
‘‘quasi solution’’ (Fig. 1). The absence of diffusion limited
reactions in microgel environments and permanent surfaces
enable direct access to loci of synthesis. Convenient magnetic
sedimentation proved to ensure ease of purification after each
reaction step.
Sequential assembly of peptides requires repetitive coupling
and deprotection cycles to be performed on the support and
purification of the intermediate products in-between each
cycle is necessary. As magnetic sedimentation of magnetite
nanoparticles works in various solvents, superparamagnetic
magnetite particles were synthesized. A microwave facilitated
route enabled rapid synthesis of monodisperse magnetite
nanoparticles from iron(III)acetylacetonate.9 Well-defined,
monodomain magnetite nanoparticles could be isolated
from the resulting homogeneous suspension by magnetic
sedimentation. Transmission electron microscopy (TEM)
images showed well-dispersed, uniform monodomain nano-
particles with defined crystal structure and widths of 6 ꢁ 1 nm
(Fig. 2a, and Fig. S1, ESIw). Atomic force microscopy
(AFM) confirmed an average particle diameter of 9 ꢁ 3 nm
(Fig. S2, ESIw). Powder X-ray diffraction (XRD) patterns
suggested the presence of phase-pure magnetite with crystallite
sizes of B9 nm as estimated from the line widths of the
reflections (Fig. S3, ESIw).
a Humboldt-Universitat zu Berlin, Department of Chemistry,
¨
Brook-Taylor-Str. 2, Berlin 12489, Germany.
E-mail: h.boerner@hu-berlin.de; Fax: +49 30 2093 7215
b ETH Zurich, Department of Materials, Wolfgang-Pauli-Str. 10,
¨
Zurich 8093, Switzerland
¨
c BAM Federal Institute for Materials Research and Testing,
Unter den Eichen 87, Berlin 12205, Germany
w Electronic supplementary information (ESI) available. See DOI:
10.1039/c2cc33492e
c
7176 Chem. Commun., 2012, 48, 7176–7178
This journal is The Royal Society of Chemistry 2012

View Article Online
The required functionalization of the Fe3O4@SiO2 particle
surface with amino groups was realized in a secondary Stober
¨
process with (3-aminopropyl)trimethoxysilane (APTMS) as a
precursor. In contrast to many time-consuming literature
procedures, microwave assisted APTMS hydrolysis and
condensation lead in B3 min at 60 1C to the functional
coating on the surface of Fe3O4@SiO2 particles. z-Potential
analysis gave a qualitative verification of the successful surface
functionalization by showing a change from z = ꢂ38.08 mV
for Fe3O4@SiO2 to z = +41.14 mV for Fe3O4@SiO2@NH2
nanoparticles. TEM micrographs showed the preservation of
the core–shell structure and the absence of obvious particle
agglomeration during coating (Fig. S7, ESIw). Analysis of the
TEM images revealed rather defined particles with a slightly
increased size of 69 ꢁ 8 nm (Fig. 2c). Small-angle X-ray
scattering (SAXS) confirmed these dimensions and the
core–shell structure by excellent curve fits based on the
dimensions derived from HRTEM (Fig. 2d, Fig. S13, ESIw).
EDX proved functionalization by indicating an increase of
carbon and nitrogen content (Table S1, ESIw). To accurately
determine the concentration of available amino groups on
Fig. 2 Analysis of seed and core–shell particles: TEM of
(a) Fe3O4 (inset HRTEM), (b) Fe3O4@SiO2 and (c) Fe3O4@SiO2@NH2
particles. SAXS of Fe3O4@SiO2@NH-(b-Ala)2-Rink particles (d).
the Fe3O4@SiO2@NH2 particle surface,
a quantitative
9H-fluoren-9-yl-methoxycarbonyl (Fmoc)-test was performed.11
Coupling Fmoc-protected glycine to the particles and subsequent
cleavage of the protective groups enabled a read-out of
the liberated dibenzofulvene-piperidine by UV-spectroscopy.
A concentration of 0.11 mmol amino groups per gram of
particles could be calculated, which is in the range of common
Tentagels peptide synthesis resins (0.14–0.27 mmol gꢂ1).
Probably the loading could be easily increased by successive
coupling of multiple Fmoc Lys(Fmoc) OH derivatives.
N2-sorption revealed an active surface area of 30 m2 gꢂ1 for
the functionalized particles. Assuming compact spheres with a
density of bulk silica allows for the calculation of a particle
diameter of 76 nm, which agrees well with the observed
dTEM = 70 nm. An average of two amino groups per 1 nm2 on
the particle surface could be calculated, assuming 70 nm particles.
The obtained loading appears to be a suitable compromise
between optimal peptide synthesis and high support capacity.
A higher loading will probably increase the risk of peptide–
peptide interactions, which might negatively interfere with
the peptide synthesis. To reduce unfavourable contacts between
peptides and the silica surface that also might prevent effective
peptide synthesis a spacer in form of two b-alanine units was
attached to the amino functionalities of the Fe3O4@SiO2@NH2
particles by coupling Fmoc-b-alanines (Fig. 1). Subsequent
attachment of a Rink-amide linker moiety introduced the semi-
permanent 4-(20,40-dimethoxyphenylhydroxymethyl-phenoxy)-
linker (Fig. 1 and Fig. S11, ESIw) that can be used to liberate
the final peptide from the support. The coupling of the
b-alanines and the Rink-linker took already the advantage
of magnetic sedimentation to purify the intermediates after
each reaction step. FTIR spectroscopy showed the occurrence
of characteristic amide bands at n = 1672 cmꢂ1 and 1548 cmꢂ1
(Fig. S12, ESIw). Wide angle X-ray scattering (WAXS) data
correspond to the calculated magnetite pattern and indicated
that the multistep process was not changing the crystal structure
of the magnetic core particles (Fig. S8, ESIw). To study the
sedimentation behaviour of the colloidal supports, gravitational
The sensitivity of magnetite particles against both acidic
and basic conditions, which will be utilized during peptide
synthesis, makes protection essential. A sufficiently thick silica
shell was constructed around the magnetite nanoparticles
(Fig. 1), using a rapid microwave assisted Stober method with
¨
tetraethoxysilane (TEOS) as a precursor.10 The magnetite seed
particles were readily dispersed in ethanol–water with TEOS
and ammonia. After 3 min microwave irradiation at 60 1C,
monodisperse core–shell Fe3O4@SiO2 particles were isolated
by magnetic sedimentation. Zeta-potential measurements
indicated qualitatively the successful coating by a change
from z = +44.3 mV of the magnetite seed particles to
z = ꢂ38.1 mV representative of silica. The chemical identity
of the Fe3O4@SiO2 particles was confirmed by Fourier
transform infrared (FTIR) spectroscopy, revealing an
absorption at n = 1063 cmꢂ1 that is typical for SiO2
(Fig. S6 and S12, ESIw). TEM proved the formation of defined
core–shell particles, with a size of 67 ꢁ 9 nm (Fig. 2 and
Fig. S5, ESIw). Energy-dispersive X-ray spectroscopy (EDX)
indicated an increased percentage of Si and O after coating
(Table S1, ESIw).
To verify the desired shielding effect and indirectly to
confirm the integrity of the silica shell around the magnetite
core, the Fe3O4@SiO2 particles were exposed to harsh
conditions used during SPPS. A treatment with piperidine in
N-methyl-2-pyrrolidone (20 vol%) for several days could
not harm the core–shell particles, while uncoated magnetite
dissolves within minutes. The stability under acidic conditions
was proved by treating the particles with trifluoroacetic acid in
dichloromethane (50 vol%). UV/vis-spectroscopy was used to
follow the development of soluble iron compounds that could
not be magnetically sedimented (Fig. S10, ESIw). While the
core–shell particles showed excellent stability for at least 3 h,
the pure magnetite particles instantly started to bleed-out
soluble iron compounds.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 7176–7178 7177

View Article Online
In summary, functional core–shell nanoparticles were
synthesized by a microwave assisted route, enabling rapid
and reproducible synthesis of well-defined Fe3O4@SiO2@NH2
nanoparticles. They could be decorated on the surfaces with
Rink-amide linkers to make the Fe3O4@SiO2@NH-(bAla)2-
Rink core–shell particles applicable as colloidal supports for
peptide synthesis. A tetrapeptide model and a potentially
bioactive undecapeptide were accessed, both exhibiting excellent
purities as isolated crude products. As the magnetic sedimenta-
tion proved to be an effective isolation step that might be easily
automatized and microwave assisted particle synthesis can be
combined with continuous flow reactors, large scale automated
peptide synthesis on colloidal supports can be foreseen.
We acknowledge Prof. Rademann, R. Wendt and F. Fischer
(HU) for their contribution, M. Suess (ETH) for TEM and
¨
CEM for helpful assistance. HGB acknowledges support by
German Research Council DFG (CORE BO1762/4-1).
Fig. 3 HPLC-trace and ESI-MS spectrum of synthesized peptide
H2N–Phe-Lys-Leu-Gly–CONH2 (gradient: 3–50% MeCN, RP-C18-
column, l = 210 nm). Mass of peptide found with 95% purity.
Notes and references
1 (a) R. B. Merrifield, J. Am. Chem. Soc., 1963, 85, 2149;
(b) R. B. Merrifield, Angew. Chem., 1985, 97, 801.
2 (a) T. Kimmerlin and D. Seebach, J. Pept. Res., 2005, 65, 229;
(b) A. R. Mitchell, Biopolymers, 2008, 90, 175; (c) B. L. Nilsson,
M. B. Soellner and R. T. Raines, Annu. Rev. Biophys. Biomol.
Struct., 2005, 91, Palo Alto.
3 (a) P. Vlieghe, V. Lisowski, J. Martinez and M. Khrestchatisky,
Drug Discovery Today, 2010, 15, 40; (b) E. Bayer, Angew. Chem.,
Int. Ed., 1991, 30, 113; (c) E. M. Gordon, R. W. Barrett,
W. J. Dower, S. P. A. Fodor and M. A. Gallop, J. Med. Chem.,
1994, 37, 1385.
and magnetic sedimentation was investigated by turbidity
measurements (Fig. S9, ESIw). While the supports sediment in
the absence of external magnetic fields within B4 hours, a strong
magnetic field induced rapid sedimentation within B2 minutes.
Redispersion of the sedimented particles occurs upon gentle
shaking.
Ultimate proof for the applicability of colloidal supports,
however, requires the synthesis of peptides on Fe3O4@SiO2@NH-
(b-Ala)2-Rink supports. Two peptides were synthesized using
sequential assembly of Fmoc-a-amino acids, implementing
colloidal supports to standard SPPS procedures. A magnetic
sedimentation step was employed to isolate the intermediate
products after each reaction step. According to standard
SPPS, coupling reactions were facilitated by benzotriazol-
1-yl-oxytripyrrolidinophosphonium hexafluorophosphate with
N,N-diisopropyl-ethylamine in NMP. The deprotection of the
Fmoc protected a-amino groups was achieved by piperidine in
NMP. As a first model a tetrapeptide with a Phe-Lys-Leu-Gly
sequence was synthesized (Fig. 3). Liberation from the support
after peptide synthesis was performed with TFA–DCM
(50 vol%). Magnetic sedimentation of the empty supports
resulted rapidly in a colourless cleavage solution. The peptide
could be isolated by ether precipitation with B70% yields.
HPLC-ESI-MS analysis of the crude peptide exhibits remarkable
product purity of ca. 95% (Fig. 3; Fig. S14 and S15, ESIw).
Moreover, a longer undecapeptide with biological relevance
was synthesized to demonstrate the feasibility of the strategy.
The sequence Gln-Thr-Thr-Thr-Trp-Gln-Asp-Pro-Arg-Lys-Gly
as part of the hYAP WW peptide domain12 could be obtained.
MALDI-TOF-MS proved the identity of the crude peptide and
confirmed that longer sequences can be synthesized on colloidal
supports, as well (Fig. S16, ESIw). The high purity of both
peptides was in the upper range of peptides that can be obtained
on common SPPS resins. Only minor amounts of side products
could be observed, which are assignable to either deletion
sequences or stop sequences, both of which are common side
reactions to standard SPPS (Fig. S14–S16, ESIw).
4 (a) R. V. Ulijn and A. M. Smith, Chem. Soc. Rev., 2008, 37, 664;
(b) H. G. Borner, Prog. Polym. Sci., 2009, 34, 811;
¨
(c) L. Hartmann, E. Krause, M. Antonietti and H. G. Borner,
¨
Biomacromolecules, 2006, 7, 1239; (d) G. W. M. Vandermeulen and
H.-A. Klok, Macromol. Biosci., 2004, 4, 383; (e) A. Verch, H. Hahn,
E. Krause, H. Colfen and H. G. Borner, Chem. Commun., 2010,
46, 8938.
¨
¨
5 (a) D. C. Sherrington, Chem. Commun., 1998, 2275; (b) H. Becker,
H. W. Lucas, J. Maul, V. N. R. Pillai, H. Anzinger and M. Mutter,
Makromol. Chem., Rapid Commun., 1982, 3, 217.
6 (a) L. An-Hui, E. L. Salabas and F. Schuth, Angew. Chem., Int.
¨
Ed., 2007, 46, 1222; (b) I. Bilecka and M. Niederberger, Nanoscale,
2010, 2, 1358; (c) N. Pinna and M. Niederberger, Angew. Chem.,
Int. Ed., 2008, 47, 5292; (d) B. L. Cushing, V. L. Kolesnichenko
and C. J. O’Connor, Chem. Rev., 2004, 104, 3893.
7 (a) I. K. Herrmann, M. Urner, F. M. Koehler, M. Hasler, B. Roth-
Z’Graggen, R. N. Grass, U. Ziegler, B. Beck-Schimmer and
W. J. Stark, Small, 2010, 6, 1388; (b) A. H. Latham and
M. E. Williams, Acc. Chem. Res., 2008, 41, 411.
8 (a) P. Sasikumar and M. Kempe, Int. J. Pept. Res. Ther., 2007,
13, 129; (b) J. Marık, D. H. Lau, A. Song, X. Wang, R. Liu and
K. S. Lam, J. Magn. Magn. Mater., 2003, 264, 153; (c) S. Rana,
P. White and M. Bradley, Tetrahedron Lett., 1999, 40, 8137;
(d) I. Sucholeiki and J. M. Perez, Tetrahedron Lett., 1999,
40, 3531; (e) M. J. Szymonifka and K. T. Chapman, Tetrahedron
Lett., 1995, 36, 1597; (f) A. Basso, P. Braiuca, C. Ebert,
L. Gardossi and P. Linda, J. Chem. Technol. Biotechnol., 2006,
81, 1626.
9 (a) I. Bilecka, I. Djerdj and M. Niederberger, Chem. Commun.,
2008, 886; (b) I. Bilecka, M. Kubli, E. Amstad and
M. Niederberger, J. Sol-Gel Sci. Technol., 2011, 57, 313.
10 N. M. Bahadur, T. Furusawa, M. Sato, F. Kurayama,
I. A. Siddiquey and N. Suzuki, J. Colloid Interface Sci., 2010,
355, 312.
11 L. A. Carpino, Acc. Chem. Res., 1987, 20, 401.
12 (a) S. Marius, Prog. Biophys. Mol. Biol., 1996, 65, 113; (b) P. Bork
and M. Sudol, Trends Biochem. Sci., 1994, 19, 531.
c
7178 Chem. Commun., 2012, 48, 7176–7178
This journal is The Royal Society of Chemistry 2012