Complete surface control of peptide nanospheres with detachable and attachable polymer brush layers

Article abstract of DOI:10.1039/c0cc01051k

The surfaces of biodegradable peptide nanospheres with density-controllable poly(ethylene glycol) (PEG) brush layers were amenable to high levels of control, from hydrophilic 'stealth' properties to hydrophobic adsorptive properties depending on the PEG density in response to environmental conditions - 'intelligent' properties that are expected to be useful for novel drug delivery systems.

Full text of DOI:10.1039/c0cc01051k

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Complete surface control of peptide nanospheres with detachable and
attachable polymer brush layersw
Tomonori Waku,ad Masahiro Matsumoto, Michiya Matsusaki and Mitsuru Akashi*ac
a
ab
Received 21st April 2010, Accepted 28th July 2010
DOI: 10.1039/c0cc01051k
The surfaces of biodegradable peptide nanospheres with density-
controllable poly(ethylene glycol) (PEG) brush layers were
amenable to high levels of control, from hydrophilic ‘stealth’
properties to hydrophobic adsorptive properties depending on the
PEG density in response to environmental conditions – ‘intelligent’
properties that are expected to be useful for novel drug delivery
systems.
(PPhe) core and a hydrophilic high-density poly(ethylene glycol)
(PEG) brush layer by the polymerization of N-carboxy
L-phenylalanine anhydride (Phe-NCA) with two initiators, a
hydrophilic amine-monoterminated PEG (NH
a–c
2
-PEG) and a
5
hydrophobic alkyl amine.
The obtained peptide nano-
spheres had extreme bioinert ‘stealth’ properties due to the
2
high density of PEG brushes (0.21–1.8 chains/nm). Further-
more, the target protein could be immobilized onto their
surface functional groups, which was introduced by using a
carboxyl-terminated PEG macroinitiator. Developing this
idea further, if a stimuli-sensitive group is introduced into
the PEG initiators, then peptide nanospheres with environ-
mentally sensitive PEG brush layers are expected to be
obtained.
Nanospheres have been widely studied for biomedical and
biochemical applications such as drug delivery, diagnosis and
1
imaging. The control of the surface sorbability towards
proteins or cell membranes is significant, since it controls the
2
biodistribution of the nanospheres. In particular, for success-
ful intravenous delivery, an ‘intelligent’ surface is required in
order to satisfy different requirements at different stages before
reaching to the target cell – a bioinert surface is desirable for
stability during long circulation times in the blood, whereas
Here, we report the synthesis and unique properties of
PEG-SS-peptide nanospheres possessing disulfide-conjugated
high-density PEG brush layers (Scheme 1). The surface PEG
chains were easily and completely detached by incubation with
a reductant; thus, the surface properties of the nanospheres
can be precisely controlled from hydrophilic to hydrophobic
properties. Moreover, thiol-monoterminated PEG (SH-PEG)
was conjugated onto the surface of ‘hairless’ nanospheres by
disulfide bond formation under oxidative conditions, completely
recovering the surface PEG density. These PEG-SS-peptide
nanospheres had bioinert and hemolytic activities depending
on their PEG densities. These surface-controllable peptide
nanospheres are expected to be a new class of nanocarriers
for biomedical applications.
2b,7a
the same property is unfavorable for intracellular translocation.
The fabrication of environmentally responsive polymer brushes
on the nanosphere surface is one of the most effective approaches
to control the surface, because their conformational change
3
,4
enables the switching of the surface properties. However, a
complete switch of the surface characteristics on the nano-
spheres from hydrophilic ‘stealth’ to hydrophobic adsorptive
properties cannot be obtained by conventional conformational
change methods. Moreover, atom transfer radical polymeri-
zation (ATRP) has been employed to produce a polymer brush
3
layer on particle surfaces, but the obtained nanospheres
have limitations in the biomedical field because of the non-
biodegradability of the vinyl polymer brushes. Therefore if
biodegradable nanospheres with complete surface controllability
are developed, they will have wider use for drug delivery systems
or biomedical applications.
PEG-SS-peptide nanospheres were prepared by the one-step
polymerization of Phe-NCA with two initiators, 2-(ethylthio)-
ethylamine and a PEG-SS-NH
2
macroinitiator with a disulfide
O (1:9 v/v) mixed solvent
group in its terminus, in DMSO–H
2
for 24 h at 0 1C, according to a previously reported procedure
Recently, we reported the one-step synthesis of peptide nano-
spheres composed of a hydrophobic poly(L-phenylalanine)
a
Department of Applied Chemistry, Graduate School of Engineering,
Osaka University, 2-1 Yamada-oka, Suita 565-0871, Japan.
E-mail: akashi@chem.eng.osaka-u.ac.jp; Fax: +81-6-6879-7359;
Tel: +81-6-6879-7356
b
Precursory Research for Embryonic Science and Technology
(
4
PRESTO), Japan Science and Technology Agency (JST),
-1-8 Honcho, Kawaguchi-shi 332-0012, Japan
c
Core Research for Evolutional Science and Technology (CREST),
Japan Science and Technology Agency (JST), 4-1-8 Honcho,
Kawaguchi-shi 332-0012, Japan
Current affiliation: Department of Bio-molecular Engineering,
Kyoto Institute of Technology, Matsugasaki, Sakyo, Kyoto 606-
d
8
585, Japan
Scheme 1 Schematic illustration of the surface control of peptide
w Electronic supplementary information (ESI) available: Detailed
experimental procedures, structural analysis of PEG-SS-peptide nano-
spheres after reduction and oxidation, and detailed definition of
surface PEG percentage. See DOI: 10.1039/c0cc01051k
nanospheres, from being covered with high-density PEG brushes to a
‘hairless’ surface, by detaching and attaching the brush layer based on
dithiol–disulfide interconversions.
This journal is c The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 7025–7027 7025

View Article Online
and after reduction, respectively, measured in d-TFA–
d-chloroform to evaluate the amount of detached PEG. The
relative intensity of the PEG methylene proton peak at
4
.02–3.62 ppm compared with the methine proton peak of
the Phe segment at 4.90–4.35 ppm decreased to 28 Æ 6%
n = 6) after reduction (see ESIw). This result clearly shows
(
that about 72% of the PEG chains on the nanospheres were
successfully detached, assuming that the amount of Phe
segments was the same before and after reduction. Further-
1
more, the H NMR spectrum of the nanospheres after reduction
6
in d -ethanol shows no peaks at 4.3–2.7 ppm corresponding to
the PEG methylene proton, in sharp contrast to the nano-
spheres before reduction (Fig. S6). This result indicates that
there were no PEG chains on the nanosphere surface after
1
reduction, on the basis that H NMR measurements in
d6-ethanol provide the composition of the entire surface of
the nanospheres. Therefore, it is concluded that almost all of
the PEG on the surface of the nanospheres was detached upon
cleavage of the disulfide bonds with a reductant, to yield
1
Fig. 1 (a–c) Photographs, (d–f) SEM images, and (g–i) H NMR
spectra in d-TFA–d-chloroform of (a, d, g) PEG brush peptide nano-
spheres with disulfide linkages between the core and brush layer as
prepared, (b, e, h) brush layer-detached ‘hairless’ peptide nanospheres
after reduction with DTT, and (c, f, i) the regenerated PEG brush
peptide nanospheres after oxidation with thiol-monoterminated PEG
‘hairless’ nanospheres. The reason for the remaining 28%
PEG even after reduction is probably that some of the
disulfide groups of the PEG-b-PPhe copolymers were buried
inside the core, and thus DTT could not react with them.
Next, we defined ‘surface PEG percentage’ as the ratio of
the amount of PEG on the nanosphere surface after reduction
versus that before (see ESIw). As shown in Fig. 2a and Fig. S7,
the PEG surface percentage decreased with increasing DTT
concentration and reaction time. Thus, the surface structure of
the PEG-SS-peptide nanospheres can be easily controlled over
a wide range from a high-density brush to a ‘hairless’ surface
by tuning the PEG detachment (Scheme 1).
(
M
w
= 2000) in an iodine solution.
5
(
see ESIw). Scanning electron microscopy (SEM) and dynamic
laser scattering (DLS) measurements showed that nanospheres
with a diameter of 430 nm were formed (Fig. 1d and Fig. S5a).
1
H-NMR and MALDI TOF mass measurements confirmed
that the nanospheres were composed of two polymers, a PPhe
homopolymer and a PEG-b-PPhe copolymer with a disulfide
linkage between the PEG and PPhe (see ESIw). Furthermore,
the PEG graft density of the PEG-SS-peptide nanospheres was
We also addressed the surface remodification of the ‘hairless’
nanospheres by reforming the disulfide bonds between SH-PEG
(Mw = 2000) and the surface thiol groups. The ‘hairless’
nanospheres (4 mg/mL) were stirred with 0375 mM SH-PEG
(Mw = 2000) (about a 1.4-fold molar equivalent of the thiol
groups on the nanospheres), in 2.5–10 mM iodine solution at
room temperature for 24 h. They were then washed with water
to remove any unattached PEG and oxidants. Interestingly,
the obtained sample showed high water-dispersibility, similar
to that before reduction (Fig. 1c). SEM and DLS measure-
ments confirmed the consistency of the morphology and size
distribution before and after oxidation (Fig. 1f and Fig. S4c).
2
1
6
estimated to be 1.8 chains/nm from H NMR measurements.
This value is consistent with that of the peptide nanospheres,
which were prepared by the polymerization of Phe-NCA with
2 w
the dual initiators NH -PEG (M = 2000) and n-butylamine
5b
in our previous report. These results indicate that disulfide
groups strongly accumulate at the interface between the core
and brush layer of the nanospheres, as was expected.
We investigated PEG detachment from the surface of the
PEG-SS-peptide nanospheres upon cleavage of the disulfide
bonds using dithiothreitol (DTT) as a reductant. A 5 mg/mL
PEG-SS-peptide nanosphere dispersion was stirred with
4
2 mM DTT, which is about 100-fold molar equivalent of
the PEG within the PEG-SS-peptide nanospheres, in 25 mM
Tris buffer (pH 8.0) for 6 h at room temperature. The addition
of DTT rapidly induced the aggregation of the nanospheres,
whereas the nanospheres formed a stable dispersion in the
buffer without DTT for more than 24 h, supporting the idea
that successful detachment of the PEG chains from the nanosphere
surface occurs with cleavage of the disulfide bonds (Fig. 1a
and b). The aggregates obtained were washed thoroughly with
water to remove the detached PEG chains and DTT, and were
Fig. 2 (a) Relationship between the concentration of DTT and the
surface PEG percentage after detaching the PEG layer by reduction
for 24 h. The inset is the expanded figure from 0 to 3.0 mM DTT.
1
then analyzed by SEM, DLS, and H NMR. Interestingly,
SEM and DLS measurements revealed that the nano-
spheres maintained their morphology and size after reduction
(
2
b) Relationship between the concentration of KI/I and the surface
(
Fig. 1d and e, and Fig. S4a and S4b). Fig. 1g and h show the
percentage after re-modification with SH-PEG (Mw = 2000) by
1
H NMR spectra of the PEG-SS-peptide nanospheres before
026 Chem. Commun., 2010, 46, 7025–7027
oxidation for 24 h.
7
This journal is c The Royal Society of Chemistry 2010

View Article Online
of the above results, it would be expected that the disulfide
cleavage within the endosome would trigger protein release
and surface-switching to the endosome-disruptive property,
which would then enhance cytoplasmic delivery.
In conclusion, high-density PEG brush peptide nanospheres
with disulfide linkages between the hydrophobic core and the
brush layers were successfully synthesized. Their surface proper-
ties were precisely controlled over a wide range, from being
bioinert to bio-adsorptive, by detaching and attaching surface
PEG brushes based on the dithiol–disulfide interconversion. The
intracellular distribution of the target protein-immobilized
PEG-SS-peptide nanospheres will now be investigated. These
surface-controllable peptide nanospheres are expected to be a
new class of nanocarriers, especially for the cytoplasmic delivery
of therapeutic protein.
Fig. 3 (a) Effect of the surface PEG percentage of peptide nano-
spheres on hemolytic activity using RBCs. The results are expressed as
means Æ S.D. (n = 3). (b) Photographs of the supernatant of mixtures
of the nanospheres and RBCs after hemolysis activity test.
This research was supported mainly by CREST from JST
and partly by the Center of Excellence (COE) Program for the
2
1st Century, Osaka University.
Surprisingly, the surface PEG percentage recovered almost
completely to 100 Æ 6% (n = 6) after oxidation, when the
KI/I
2
concentration was 10 mM (Fig. 1i and 2b). In addition,
Notes and references
w w
SH-PEG with a higher M (M = 10 000) was successfully
1
(a) Y. Bae and K. Kataoka, Adv. Drug Delivery Rev., 2009, 61, 768;
(b) M. Matsusaki and M. Akashi, Expert Opin. Drug Delivery, 2009,
immobilized on the ‘hairless’ nanospheres (Fig. S7). These
results indicate that we can fabricate the desired surface on the
peptide nanospheres by attaching and detaching brush layers
based on dithiol–disulfide interconversions.
6
6
, 1207; (c) T. Akagi, M. Baba and M. Akashi, Polymer, 2007, 48,
729; (d) R. Hong, G. Han, J. M. Fernndez, B. J. Kim, N. S. Forbes
and V. M. Rotello, J. Am. Chem. Soc., 2006, 128, 1078;
e) E. G. Bellomo, M. D. Wyrsta, L. Pakstis, D. J. Pochan and
(
Many researchers have reported that disulfide bonds can be
7
cleaved within the endosome of cells. Therefore, if the peptide
T. J. Deming, Nat. Mater., 2004, 3, 244; (f) T. Serizawa, S. Takehara
and M. Akashi, Macromolecules, 2000, 33, 1759; (g) M. Matsusaki,
K.-i. Hiwatari, M. Higashi, T. Kaneko and M. Akashi, Chem. Lett.,
nanospheres with lower PEG graft density can disrupt an
endosomal membrane, PEG-SS-peptide nanospheres would
be expected to switch their surface properties from being
bioinert to endosome-disruptive upon disulfide cleavage with-
in the endosome. To investigate the effect of PEG graft density
on the membrane-disruptive activity, we performed a hemolysis
activity test for PEG-SS-peptide nanospheres with different
2
004, 33, 398; (h) C. Khemtong, C. W. Kessinger and J. Gao, Chem.
Commun., 2009, 3497; (i) M. Motornov, Y. Roiter, I. Tokarev and
S. Minko, Prog. Polym. Sci., 2010, 35, 174.
(a) A. Verma, O. Uzun, J. Y. Hu, H. S. Han, N. Watson, S. Chen,
D. J. Irvine and F. Stellacci, Nat. Mater., 2008, 7, 588;
(b) M. J. Joralemon, S. McRae and T. Emrick, Chem. Commun.,
2010, 46, 1377.
(a) Y. Tsujii, K. Ohno, S. Yamamoto, A. Goto and T. Fukuda, Adv.
Polym. Sci., 2006, 197, 1; (b) F. Albertorio, S. Daniel and
S. P. Creamer, J. Am. Chem. Soc., 2006, 128, 7168.
(a) D. Li, X. Sheng and B. Zhao, J. Am. Chem. Soc., 2005, 127,
6248; (b) J. N. Kizhakkedathu, N.-J. Raymond and D. E. Brooks,
Macromolecules, 2004, 37, 734; (c) M. Motornov, R. Sheparovych,
R. Lupitskyy, E. MacWilliam, O. Hoy, I. Luzinov and S. Minko,
Adv. Funct. Mater., 2007, 17, 2307.
2
3
4
surface PEG percentages using erythrocytes as a cell model
8
Fig. 3). Sheep red blood cells (RBCs) were incubated with
(
1
mg/mL PEG-SS-peptide nanospheres for 1 h at 37 1C at the
endosomal pH value of 5.5 in 25 mM MES buffer containing
.15 M NaCl. The PEG-SS-peptide nanospheres with a surface
0
PEG percentages of 100, 94 and 69% showed low activity,
supporting their bioinert properties. On the other hand,
PEG-SS-peptide nanospheres with surface PEG percentages
of 39, 15 and 0% showed high hemolytic activity, indicating
that the surface with a lower density PEG graft can interact
with cell membranes due to their hydrophobicity. These results
suggest that PEG-SS-peptide nanospheres can switch their
surface properties from being bioinert to endosome-disruptive
within the endosome. We are now preparing PEG-SS-peptide
nanospheres with functional carboxyl groups at the outer
chain-end of PEG brush layer and covalently conjugating a
target protein to the functional carboxyl group. On the basis
5 (a) M. Matsusaki, T. Waku, T. Kaneko, T. Kida and M. Akashi,
Langmuir, 2006, 22, 1396; (b) T. Waku, M. Matsusaki, T. Kaneko
and M. Akashi, Macromolecules, 2007, 40, 6385; (c) M. Matsusaki,
M. Matsumoto, T. Waku and M. Akashi, J. Biomater. Sci., Polym.
Edn., DOI: 10.1163/092050610X497890; (d) T. Waku, M. Matsusaki,
S. Chirachanchai and M. Akashi, Chem. Lett., 2008, 37, 1262.
6
S. Kawaguchi, M. A. Winnik and K. Ito, Macromolecules, 1996, 29,
465.
4
7
(a) S. Takae, K. Miyata, M. Oba, T. Ishii, N. Nishiyama, K. Itaka,
Y. Yamasaki, H. Koyama and K. Kataoka, J. Am. Chem. Soc.,
2008, 130, 6001; (b) J. Yang, H. Chen, I. R. Vlahov, J. -X.
Cheng and P. S. Low, Proc. Natl. Acad. Sci. U. S. A., 2006, 103,
13872.
C. Plank, B. Oberhauser, K. Mechtler, C. Koch and E. Wagner,
J. Biol. Chem., 1994, 269, 12918.
8
This journal is c The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 7025–7027 7027