A novel preparative method of silica nanotubes by Utilizing Self-assembly and disassembly of peptide amphiphiles

Article abstract of DOI:10.1246/cl.2012.95

The addition of 2,2,2-trifluoroethanol (TFE) induces both the transition from β-sheet to α-helix structure of peptides and disassembly of wormlike micelles of peptide amphiphiles. The hierarchical structural changes were utilized for the preparation of silica nanotubes from silica-micelle complexes with avoiding structural deterioration that is normally caused by thermal treatment. This method is advantageous for the preparation of silica nanotubes because of reusability of peptide templates, possible replication of the surface of β-sheet structure, and very mild conditions.

Full text of DOI:10.1246/cl.2012.95

doi:10.1246/cl.2012.95
Published on the web December 30, 2011
95
A Novel Preparative Method of Silica Nanotubes by Utilizing Self-assembly
and Disassembly of Peptide Amphiphiles
Tomoko Shimada,*1,3 Yasuhiro Tamura, Matthew Tirrell, and Kazuyuki Kuroda*
Department of Applied Chemistry, Faculty of Science and Engineering, Waseda University,
-4-1 Ohkubo, Shinjuku-ku, Tokyo 169-8555
Kagami Memorial Research Institute for Materials Science and Technology, Waseda University,
-8-26 Nishiwaseda, Shinjuku-ku, Tokyo 169-0051
Asahi Kasei Co., 1-105 Kanda Jinbocho, Chiyoda-ku, Tokyo 101-8101
Institute for Molecular Engineering, University of Chicago, 5735 South Ellis Avenue, Chicago, Illinois 60637, USA
1
4
1,2
1
3
2
2
3
4
(
Received October 22, 2011; CL-111040; E-mail: kuroda@waseda.jp)
The addition of 2,2,2-trifluoroethanol (TFE) induces both
the transition from ¢-sheet to ¡-helix structure of peptides and
disassembly of wormlike micelles of peptide amphiphiles. The
hierarchical structural changes were utilized for the preparation
of silica nanotubes from silica-micelle complexes with avoiding
structural deterioration that is normally caused by thermal
treatment. This method is advantageous for the preparation of
silica nanotubes because of reusability of peptide templates,
possible replication of the surface of ¢-sheet structure, and very
mild conditions.
Generally, wormlike PA micelles are structurally stable
compared to wormlike micelles of conventional surfactants
because the ¢-sheet structure, based on intermolecular hydrogen
bonding, stabilizes micelle structures. Thus, it is difficult to
dissociate the wormlike micelles by increasing temperature of
micelle solutions or adding alkyl alcohols to remove templates.
We have recently found that 2,2,2-trifluoroethanol (TFE)
disassemble the wormlike micelles of peptide amphiphile C16-
W3K by changing the peptide secondary structure from ¢-sheet
1
4
to ¡-helix. C16-W3K, consisting of a C16 alkyl tail and a
peptide W3K [WAAAAKAAAAKAAAAKA] (W: tryptophan,
1
5
A: alanine, and K: lysine) that has a high ¡-helical propensity,
Silica nanotubes have received broad attention due to the
potential applications not only in catalysis but also in medical
and biological fields, such as DNA sensing and drug carriers,
has been reported to form wormlike micelles in an aqueous
buffer solution with formation of a ¢-sheet structure, as reported
by us. Because C16-W3K has three lysines, that can act
1
2
16
by modifying the outer surface and by encapsulating functional
molecules in the inner pore. For these applications, control of the
microstructure and biomodification of the surfaces of nanotubes
is important. Silica nanotubes are normally prepared from
catalytically, and forms wormlike micelles at physiological pH
(pH 7.4), it is expected to be an ideal template for preparation
of silica nanotubes. In addition, the slow speed of the micelle
formation, which is characteristic of C16-W3K, facilitates the
control of the wormlike micelles. If the templates can be removed
by disassembly of wormlike micelles by TFE, preparation of
silica nanotubes without calcination should be achieved.
Here we propose a new method for the preparation of silica
nanotubes by utilizing self-assembly and disassembly of C16-
W3K as shown in Scheme 1.
Step (i) is the formation process of wormlike C16-W3K
micelles, and step (ii) depicts the preparation of wormlike
micelles coated with silica by hydrolysis and condensation of
silica precursors. Step (iii) shows the process for removing the
wormlike C16-W3K micelle template.
3
soluble silica species by templating using collagen fibers,
organic gel filaments,4 ¢-sheet peptides, etc., as rod-like
templates. However, because calcination around 500­600 °C is
required to remove such organic templates, all the organic
components burn off and the microstructures of silica tubes are
varied. Therefore, milder methods for removing templates are
required in order to make good use of the organic components
and silica microstructures.
,5
6,7
Here we report a new soft preparative method of silica
nanotubes by utilizing self-assembly and disassembly of peptide
amphiphiles (PAs). PAs are surfactants that have peptide chains
in the hydrophilic regions, and some of the PAs are known to
self-assemble into wormlike micelles.8 Wormlike micelles of
PAs with homogeneous and tunable diameter have peptides on
the surface that can form secondary structures (¡-helix, ¢-sheet,
etc.) and exhibit bioactivity. Consequently, the micelles are
well-suited as templates for preparing silica nanotubes. More
importantly, self-assembled templates can be softly removed by
disassembly.
As the first step of the process, the templates of the
wormlike micelles were prepared. The peptide amphiphile C16-
­12
Recently, Yuwano et al. reported the preparation of silica
1
3
nanotubes with PA templates having lysines or histidines. In
the study, they found that no catalyst is required in the system
because the amine moieties in lysines or histidines work as a
catalyst for hydrolysis and condensation of silicon alkoxides.
Though their study has exhibited advantages of PA micelles as
templates, calcination was inevitably conducted in the process to
remove wormlike PA micelles.
Scheme 1. Schematic of the formation of silica nanotubes.
Chem. Lett. 2012, 41, 95­97
© 2012 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/

9
6
Figure 1. TEM images (without staining) of (a) silica-coated C16-
W3K incubated for 10 min and the TMOS/PA molar ratio is one and
(
b) incubated for 10 min and the TMOS/PA molar ratio is three. Scale
bar, 100 nm.
Figure 2. IR spectrum of the precipitates after the step (iii).
W3K was synthesized by Scrum Inc. by covalently bonding of
the peptide W3K to the alkyl tail of C16 and verified by
MALDI-TOF mass spectrometry. The template samples were
prepared by adding C16-W3K to a buffer solution (pH 7.4) at a
peptide concentration of about 125 ¯M and subsequently by
incubating the solution at 50 °C for 1, 2, 5, 10, and 20 min to get
short wormlike micelles. Networking of long wormlike micelles
causes the gelation of the PA solution, which makes it difficult to
mix PAs and silica precursors homogeneously. Short micelles
are useful for preparation of detached silica nanotubes which are
appropriate for the characterization. Because wormlike micelles
of C16-W3K elongate with incubation time, the length of the
wormlike micelle can be controlled by changing the incubation
time.
Tetramethoxysilane (TMOS) was added to the solutions
immediately after the incubation at 50 °C, and the mixtures were
stirred at room temperature for two days. TMOS was chosen as a
silica precursor because of its high hydrolysis rate. Neither acid
nor base catalyst was added, and the molar ratio of TMOS to
C16-W3K was 1, 2, 3, 5, and 20.
200 nm. This suggests that the wormlike PA micelles were
elongated during the condensation of hydrolyzed species of
TMOS due to the neutralization of the negatively charged
hydrolysed TMOS and positively charged lysines in the W3K at
the ends of the micelles. Though the incubation at 50 °C was
applied here to make wormlike micelles, it has been found that
C16-W3K also forms wormlike micelles at room temperature,
1
6
though requiring longer than 10 days. Thus, if the appropriate
incubation time at room temperature can be found, the entire
process should be performed at room temperature, which may
facilitate establishment of a preparation system for silica
nanotubes.
As the final step to prepare silica nanotubes, TFE was added
to the solution to remove the templates. The concentration of
TFE employed in solutions was 25%, because the value is most
effective to dissociate wormlike C16-W3K micelles, as reported
1
4
previously. The solution was stirred for one day at room
temperature. When the solution was left for a while after
centrifugation, glittering precipitates were obtained in the
solution. The IR spectrum (KBr disc) of the precipitates is
shown in Figure 2. The IR spectrum shows two characteristic
The transmission electron microscopy (TEM) image (with-
out staining) of only the sample, prepared with C16-W3K
incubated for 10 min and at the TMOS/PA molar ratio of one,
shows mainly fibrous structures with a diameter of ca. 40 nm
¹
1
absorption bands at 1105 and 1165 cm , assignable to the
Si­O­Si stretching vibration. The broad bands at 3400 and
¹
1
(Figure 1a). The fibrous materials have no branch and have
950 cm are assigned to SiOH. Though very small sharp bands
¹
1
uniform diameter. A small amount of membrane-like materials,
presumed to be extra C16-W3K, were observed among the
fibrils. The fibrous materials were presumed to be wormlike
micelles coated with silica judging from its diameter. The
images of the samples prepared with C16-W3K incubated for 1,
at 1630 and 2930 cm are assigned to C=O and CH2 in C16-
W3K, respectively, the bands should be due to a small amount
of C16-W3K that remained on the washed nanotubes. Because
the silica-coated wormlike micelles are dispersed and cannot be
isolated from the solution before adding TFE, it is inappropriate
to directly compare the IR spectra of the samples before and
after adding TFE. The washed sample virtually does not contain
C16-W3K inside and outside of the nanotubes because the IR
bands due to C16-W3K are so small.
2, and 5 min show inhomogeneous structures including aggre-
gated structures and very short fibrils (data not shown). The
sample with C16-W3K incubated for 20 min was too viscous to
mix the PA with TMOS homogeneously. Consequently, incu-
bation time of 10 min is appropriate to prepare the templates
of wormlike C16-W3K micelles. In the TEM images of the
samples that have higher TMOS/PA molar ratios than one under
the same incubation time (10 min), not only fibrous silica but
also aggregated silica was observed. Figure 1b shows the TEM
image of the sample (TMOS/PA = 3). Because it is difficult
to isolate the wormlike micelles coated with silica from the
mixture, the TMOS/PA molar ratio of one was found to be
suitable for the process.
The TEM image of the precipitates (Figure 3) shows the
fibrils with an outer diameter of about 30­40 nm, judging from
the TEM image. In the image at low magnification of the sample
(Figure 3a), the membrane-like materials of C16-W3K observed
in Figure 1a vanished, and the fibrous materials can be observed
more clearly. In the fibrils, the linear areas of a light color with
a diameter of ca. 10 nm were observed (Figure 3b). Because the
TEM image was taken without staining, it was difficult to
determine whether the PA templates were removed or not by the
comparison between Figure 1a and Figure 3. However, as the IR
result indicates that the precipitates consist of mainly silica, the
Even though C16-W3K was incubated only for 10 min
before adding TMOS, the length of the fibrils was longer than
Chem. Lett. 2012, 41, 95­97
© 2012 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/

9
7
drugs in the self-assembled templates can remain in the silica
nanotubes after removing the templates. It is known that
silicatein filaments and their subunits govern the enzymatic
and structurally controlled synthesis of silica in vivo at ambient
2
3,24
temperature and at near-neutral pH in a marine sponge,
which is similar to our system that the wormlike PA micelles
serve as catalytic templates for the formation of silica layers at
room temperature and at pH 7.4 in vitro. We believe that the
new preparative method reported here is also fascinating as a
model for understanding the mechanism of biomineralization.
In conclusion, structural changes of peptide amphiphiles
with TFE are utilizable for the preparation of silica nanotubes
while avoiding structural deterioration by thermal treatment.
Because of the peptide secondary structure in the outer surface
of the template and the mild condition of the preparation
processes, the resultant silica nanotubes may exhibit structural
uniqueness and open up new possibilities for these silica
nanotubes in bioapplications.
Figure 3. TEM images (without staining) of the precipitates (a) at
low magnification and (b) at high magnification after the step (iii).
fibrils can be regarded as silica nanotubes. Because the inner
pore diameter of ca. 10 nm, observed in the TEM image, roughly
corresponds to the diameter of wormlike C16-W3K micelles
measured by contrast variation SANS (small angle neutron
1
7
scattering), it is reasonable to conclude that silica layers are
constructed on the outer surface of PA micelles.
In the process, hydrolyzed TMOS was condensed exclu-
sively on the micellar surface due to the catalytic amino acids of
W3K. Because the peptide forms relatively stretched structure
The study was supported by the Global COE Program
“Practical Chemical Wisdom” and “Elements Science and
Technology Project,” both from the Ministry of Education,
Culture, Sports, Science and Technology (MEXT). Asahi Kasei
generously provided the PA.
(
¢-sheet) in the shell of the wormlike micelles, it is presumed
that the outermost lysine mainly serves as a catalyst. Even
though the wormlike micelles were covered with silica before
adding TFE, we surmise that TFE can reach the interior from the
ends or through the defects in silica layers and dissociate the
wormlike PA micelles. The remarkable effect of TFE on the
structural changes is explained by the presence of the hydro-
References
1
R. Fan, R. Karnik, M. Yue, D. Li, A. Majumdar, P. Yang, Nano Lett.
2005, 5, 1633.
2
3
C. R. Martin, P. Kohli, Nat. Rev. Drug Discovery 2003, 2, 29.
Y. Ono, Y. Kanekiyo, K. Inoue, J. Hojo, M. Nango, S. Shinkai, Chem.
Lett. 1999, 475.
1
8­20
phobic TFE clusters.
These IR and TEM results described above exhibit that the
wormlike PA micelles inside the nanotubes can be removed by
adding TFE, and silica nanotubes are prepared successfully
without calcination. The TEM image exhibits relatively thicker
silica walls if we consider the TMOS/PA ratio. A detailed
explanation is not possible, but we may propose the following
scheme. The PA solution is composed of monomeric PA, PA
spherical micelles, and wormlike micelles. The initial hydrolysis
and condensation of TMOS may occur randomly. Further
condensation of dissolved silica species should preferentially
occur on initially deposited silica which is thought to be mainly
located on wormlike micelles because the catalytic lysine sites
are denser on wormlike micelles. The control of wall thickness
of silica nanotubes prepared with PA templates by changing the
4
Y. Ono, K. Nakashima, M. Sano, Y. Kanekiyo, K. Inoue, S. Shinkai,
M. Sano, J. Hojo, Chem. Commun. 1998, 1477.
J. H. Jung, M. Park, S. Shinkai, Chem. Soc. Rev. 2010, 39, 4286.
J. E. Meegan, A. Aggeli, N. Boden, R. Brydson, A. P. Brown, L.
Carrick, A. R. Brough, A. Hussain, R. J. Ansell, Adv. Funct. Mater.
5
6
2
004, 14, 31.
7
8
9
H. Acar, R. Garifullin, M. O. Guler, Langmuir 2011, 27, 1079.
J. D. Hartgerink, E. Beniash, S. I. Stupp, Science 2001, 294, 1684.
G. A. Silva, C. Czeisler, K. L. Niece, E. Beniash, D. A. Harrington,
J. A. Kessler, S. I. Stupp, Science 2004, 303, 1352.
1
0
J. D. Tovar, R. C. Claussen, S. I. Stupp, J. Am. Chem. Soc. 2005, 127,
7
337.
11
S. E. Paramonov, H.-W. Jun, J. D. Hartgerink, J. Am. Chem. Soc. 2006,
128, 7291.
I. W. Hamley, Soft Matter 2011, 7, 4122.
V. M. Yuwono, J. D. Hartgerink, Langmuir 2007, 23, 5033.
T. Shimada, M. Tirrell, in Peptide Science 2006, ed. by H. Mihara,
12
13
14
molar ratio of PA and TEOS, and volume fraction of ethanol in
2
007, p. 382.
the solvent has been reported,2
1,22
though there are differences
15 S. Marqusee, V. H. Robbins, R. L. Baldwin, Proc. Natl. Acad. Sci.
U.S.A. 1989, 86, 5286.
between their study and ours in various points including PA
types, silica precursor, solvent, template removal method, etc.
The results of the same procedure with the spherical
micelles of C16-W3K in the sample without incubation showed
no fibrous or spherical silica formation, which may be caused by
the destabilization of the spherical micelles due to the interaction
between the peptides and TMOS. We presume that stabilization
of the micelles by ¢-sheet is important for the PA templates that
are formed by the balance of various interactions.
The new method described here has several advantages
including the use of a physiological pH of the solution, no need
of catalysts, and an absence of a high-temperature process.
Further, with this method through mild conditions, relatively
expensive peptide amphiphiles can be reused and encapsulated
1
6
T. Shimada, S. Lee, F. S. Bates, A. Hotta, M. Tirrell, J. Phys. Chem. B
009, 113, 13711.
2
1
7
N. Sakamoto, T. Shimada, Abstracts of Papers, Presented at the
Technical Seminar on Industrial Use of Neutrons, Miraikan, Tokyo,
February 13, 2008.
S. Kuprin, A. Graslund, A. Ehrenberg, M. H. J. Koch, Biochem.
Biophys. Res. Commun. 1995, 217, 1151.
1
8
9
1
D.-P. Hong, M. Hoshino, R. Kuboi, Y. Goto, J. Am. Chem. Soc. 1999,
1
21, 8427.
20 K. Shiraki, K. Nishikawa, Y. Goto, J. Mol. Biol. 1995, 245, 180.
21 Q. Ji, R. Iwaura, T. Shimizu, Chem. Lett. 2004, 33, 504.
2
2
2
3
Q. Ji, R. Iwaura, T. Shimizu, Chem. Mater. 2007, 19, 1329.
K. Shimizu, J. Cha, G. D. Stucky, D. E. Morse, Proc. Natl. Acad. Sci.
U.S.A. 1998, 95, 6234.
J. N. Cha, K. Shimizu, Y. Zhou, S. C. Christiansen, B. F. Chmelka,
G. D. Stucky, D. E. Morse, Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 361.
2
4
Chem. Lett. 2012, 41, 95­97
© 2012 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/