Induction of structure and function in a designed peptide upon adsorption on a silica nanoparticle

Article abstract of DOI:10.1002/anie.200600965

(Chemical Equation Presented) Small change: Nanoparticles can induce a functional helix from an unstructured peptide (see picture). The ability to generate stable, well-defined structures on surfaces opens up the possibility of creating nanosystems with a variety of functionalities, which is demonstrated by the introduction of a catalytic site for ester hydrolysis.

Full text of DOI:10.1002/anie.200600965

Angewandte
Chemie
Helical Peptides
DOI: 10.1002/anie.200600965
Induction of Structure and Function in a Designed
Peptide upon Adsorption on a Silica
Nanoparticle**
Martin Lundqvist, Patrik Nygren, Bengt-
Harald Jonsson,* and Klas Broo*
The ability to regulate biological processes at the molecular
level is often achieved by controlling the function of proteins.
Efficient use of proteins and peptides in industrial processes
and diagnostic tools also requires means to induce and
maintain their functional conformations. Interestingly, the
current discussion[1] on the origin of life has highlighted the
fact that reactions such as vesicle creation[2] and peptide
formation[3,4] are promoted by clay particles. Also in this
context, it is interesting to investigate whether silica (which is
a main constituent of most clay surfaces) could have a role in
inducing functional conformations of peptides. Herein, we
present a novel approach in which adsorption on silica
nanoparticles is used to induce a well-defined structure in a
designed peptide. The ability to generate stable structures on
surfaces opens up the possibility of creating closely regulated
[*] Dr. M. Lundqvist,[+] P. Nygren,[+] Prof. B.-H. Jonsson
Division of Molecular Biotechnology, IFM
Linköping University
58183 Linköping (Sweden)
Fax: (+46)1312-2587
E-mail: nalle@ifm.liu.se
Dr. K. Broo
Division of Applied Optics, IFM
Linköping University
58183 Linköping (Sweden)
Fax: (+46)705-121-674
E-mail: klabr@amm.gu.se
Dr. K. Broo
Department of Occupational and Environmental Medicine
Sahlgrenska Academy at Göteborg University
Göteborg (Sweden)
[+] These authors contributed equally to this work.
[**] The silica particles were kindly provided by EKA Chemicals,
Stenungsund, Sweden. This work was supported by a grant from the
Swedish National Science Research Council to B.-H.J. (K5104-5999)
and by a grant from the Knut and Alice Wallenberg Foundation to
K.B.
Angew. Chem. Int. Ed. 2006, 45, 8169 –8173
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8169

Communications
systems with a variety of potential functionalities, which we
demonstrate by introduction of a catalytic site.
should strongly favor the formation of helical structure upon
binding to the nanoparticles and strongly disfavor an ordered
structure in solution. For this reason, we decided to use
electrostatic interactions as the structure-determining factor.
Herein, we focus on the interactions between a silica nano-
particle and a peptide (peptide 2) with two rows of positively
charged side chains and residues that can form a reactive site
(Figure 1).
The introduction of functionality into designed peptides
was achieved earlier by Benner and co-workers,[5] who
constructed an efficient oxaloacetate decarboxylase, and by
Ghadiri and co-workers,[6,7] who constructed a self-replicating
peptide and an efficient catalytic synthetic peptide ligase.
Chmielewski and co-workers designed a replicating peptide
system in which the reactivity is under the control of pH or
salt.[8,9] DNA recognition modules, such as the basic region
leucine zipper, are unfolded in solution but adopt a stable,
ordered conformation upon interaction with DNA.[10–12]
DeGrado and co-workers[13] analyzed the leucine zipper
proteins, and found fundamental principles for the design of
peptides that adopt a helical structure upon interaction with
recognition sites in DNA. One important factor for induction
of a helical structure is charge–charge interaction between
basic residues and negative charges (phosphate groups) on
DNA. Thus, it should be possible to induce a helical structure
in a designed peptide that contains a motif of basic residues by
providing a surface with appropriately spaced negative charges.
From these considerations, we designed a peptide (pep-
tide 1; Figure 1) that would be unstructured in solution but
“forced” to adopt a well-defined helical structure upon
adsorption onto silica nanoparticles. The design of peptide 1
was further developed in peptide 2 to include precisely placed
amino acids, intended to form a catalytic site[14] upon folding
of the helix. To ensure that the induction of structure by
interaction with silica nanoparticles will have switchlike
properties, it is essential to incorporate a combination of
positive and negative design elements; that is, the design
Silica nanoparticles were chosen as the solid material for
two main reasons: their high negative surface charge density
and their size, which allows structural investigations of the
peptide/nanoparticle complex in solution with standard
spectroscopic techniques[15,16] and analytical ultracentrifuga-
tion.[17] Arginine residues were incorporated in the design
rather than lysine because of their guanidinium group, which
has a delocalized positive charge, high pKa value, and low
reactivity. The arginine residues were positioned so that two
neighboring rows (Figure 1) would give rise to a large
interaction area when a helix was formed upon contact with
the negatively charged nanoparticle surface. The positive
charge density of the helix interaction surface is of the same
magnitude as the negative charge density of the silica
nanoparticles at the ion concentration used in this study.[18]
The majority of the remaining amino acids in the peptide
were selected for their high helix-forming propensities.[19] The
requirement for the peptide to be unstructured in solution
was also considered in the design. The introduction of
arginine side chains in rows C and F should efficiently
prevent helix formation in solution, because the electrostatic
repulsion between them tends to elongate the peptide. His15
and Lys19 (Figure 1) were also included, to introduce a
reactive site for ester hydrolysis upon induction of the helical
structure.[14]
Sedimentation equilibrium experiments were conducted
to determine the aggregation state of the free peptide in
solution and its binding to the nanoparticles. The data in
Figure 2b show that a mixture of peptide 2 and particles
sediments between 2500 and 8000 rpm, that is, all peptide
molecules bind to the particles. The nanoparticles used in this
study have a particle weight of ꢀ 425 kDa and sediment
completely at 8000 rpm as previously shown.[17] In the absence
of particles the peptide does not sediment at these rotor
speeds, that is, the peptide does not form large aggregates in
solution (Figure 2a), and experiments at higher rotor speeds
indicated that the peptide behaves as a monomer in solution
(data not shown). Similar results were observed for peptide 1
(data not shown).
To determine the extent of formation of the secondary
structure, samples of free peptide and peptide bound to the
nanoparticles were analyzed by far-UV CD. The small
diameter (9 nm) of the silica nanoparticles allows UV light
to penetrate the sample without scattering, and therefore
conventional light spectroscopy can be used to characterize
these systems.[20–22] As illustrated in Figure 3a, the CD signals
indicate that the free peptide in solution has no defined
secondary structure, irrespective of the pH used in this study
(pH 7.9–9.8). Calculations carried out with ContinLL[23]
indicate ꢀ 9% a helix for free peptide 1 and ꢀ 8% a helix
for free peptide 2. However, addition of nanoparticles to the
Figure 1. Peptide 2 displayed as a) a helical wheel and b) a helix
viewed from the side showing the side chains of Arg and the catalytic
His–Lys pair. c) The amino acid sequences of the two peptides.
8170
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 8169 –8173

Angewandte
Chemie
Figure 2. Sedimentation equilibrium analysis of peptide 2 by ultra-
centrifugation monitored at A280 a) without and b) with 9-nm silica
particles at pH 9 and 228C. The colors represent different rotor speeds
(rpm).
solution gave rise to a CD spectrum that is typical for the
helical conformation; that is, the nanoparticles induced
formation of a helical structure upon interaction with the
peptide. Figure 3a also shows that the deletion of the His–Lys
pair results in an increase of adopted a helix upon binding of
the peptide to the nanoparticle, and the calculations indicate
that peptide 1 has 39% a helix compared with 29% for
peptide 2. There is a clear correlation between the amount of
induced secondary structure in the peptide and the pH of the
solution, as demonstrated for peptide 2 in Figure 3b. The
calculated amount of a helix is 25, 29, and 35% at pH 7.9, 9.0,
and 9.8, respectively. The nanoparticles induce more secon-
dary structure at higher pH values, probably because the
density of negative charges on the surface is increased by
raising the pH.[18] The results of melting experiments (Fig-
ure 3c) clearly suggest that the helical structure content
decreases gradually (noncooperatively) upon increasing the
temperature from 25 to 858C. The data also reveal that the
temperature-dependent structural change is fully reversible,
that is, when the sample is cooled to 258C the original amount
of helix is observed in the sample (Figure 3c, inset).
Figure 3. a) CD data for peptides 1 and 2 at pH 9.0 and 228C.
Black=peptide 2 in solution, blue=peptide 1 with 9-nm silica
particles, and red=peptide 2 with 9-nm silica particles. b) CD data for
peptide 2 at pH 7.9 (green), 9.0 (red), and 9.8 (blue) with 9-nm silica
&
particles. c) Melting of peptide 2 adsorbed on 9-nm particles ( ). The
Peptide 2 includes a His–Lys pair that forms a catalytic
unit for ester hydrolysis, provided that peptide 2 becomes
helical. Thus, strong proof of a successful design would be a
significant rate enhancement of ester hydrolysis in samples
containing both peptide 2 and nanoparticles. Experiments in
which peptide 2 or the nanoparticles were dissolved in buffer
showed no detectable rate enhancement over the background
reaction in buffer alone. Notably, samples containing the
peptide 2/nanoparticle complex showed a dramatically higher
catalytic efficiency (Figure 4). A control experiment with the
peptide 1/nanoparticle complex showed no rate enhancement
over the background reaction in buffer alone. Thus, the
enhancement of catalysis requires both the His–Lys pair and
the formation of structure upon binding to the nanoparticles.
The design of the catalytic His–Lys site was inspired by the
study of Lundh et al.,[14] in which the imidazole moiety of a
^
( ) represents the observed CD value after the sample was cooled to
258C. The inset shows the CD spectrum of the sample at 258C before
heating (red) and after heating to 858C and subsequent cooling (blue).
His residue acts as a nucleophile and the positive charge of a
nearby ornithine (Orn) residue may stabilize the transition
state of the reaction. A comparison of second-order rate
constants shows that the peptide 2/nanoparticle complex
(k2 = 0.25 mmÀ1 minÀ1) is 54times more efficient in catalysis
than peptide 2 (k2 = 0.0046 mmÀ1 minÀ1) alone. Clearly, the
nanoparticles can be used as an efficient switch to turn on
catalysis. Notably, catalysis with the peptide 2/nanoparticle
complex is 40 times more efficient than the catalysis reported
for the peptide that was designed by Lundh et al., and 410
times more efficient than catalysis by imidazole.
Angew. Chem. Int. Ed. 2006, 45, 8169 –8173
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8171

Communications
instead bring positively charged side chains closer to the silica
surface.[23,24] These observations, together with the results of
the present study, point toward the possibility that peptides
with a high fraction of positively charged amino acid residues
have a high probability of forming well-defined structures on
clay. As the amino acid content in peptides can be highly
diverse, the adsorption on clay offers rich possibilities that
side chains are brought together into functional (catalytic)
units. In fact, the possibility for diversity is much larger on
peptides than on RNA, which has been suggested to form
functional structures on clay.[25] Hence, the results add a new
twist to the hypothesis that inorganic clays had an important
role in the prebiotic chemistry that led to the origin of life.
Figure 4. Ester bond hydrolysis by: the peptide 2/nanoparticle complex
(turquoise); peptide 2 (pink); nanoparticles (yellow); and buffer (dark
blue). The reaction product m-nitrophenol was monitored at A357.6
.
Experimental Section
CD spectra were recorded as described by Lundqvist et al.[17] The
samples consisted of peptide ( ꢀ 0.1 mm) with 9-nm silica particles
added to give a peptide/nanoparticle ratio of 1:0.25. The helical
content was calculated by use of the program ContinLL.[26] Ultra-
centrifugation experiments were performed and analyzed as de-
scribed by Lundqvist et al.[17] The samples were the same as those
used in the CD analysis. The kinetics of esterase activity were
measured at pH 8.2 in Tris buffer (20 mm), with 4-sulfamoyl(benzoy-
lamino)acetic acid 3-nitrophenyl ester (0.04m m) as substrate. The
reaction was monitored at the isosbestic point for m-nitrophenol of
357.6 nm. The peptide concentration was 0.1 mm and the concen-
tration of silica nanoparticles was approximately 0.05 mm. Under
these conditions, the rate with buffer solution alone was indistin-
guishable from those measured on samples with nanoparticles or on
samples with free peptide in solution. To determine the (very low)
rate enhancement from free peptide in solution, experiments with
2 mm peptide were performed under otherwise identical conditions.
Catalysis by 10 and 50 mm imidazole was also measured. Second-
order rate constants k2 were calculated after subtraction of the
background rate in buffer alone. The peptides were synthesized by
standard 9-fluorenylmethoxycarbonyl (fmoc) protection group
chemistry.
It is difficult to distinguish between the alternatives that
either all peptide molecules adopt a partial helical structure
or an equilibrium situation occurs, in which a fraction of
peptides is completely helical while another fraction is
without helical conformation. The latter alternative would
indicate that the temperature dependence of the catalytic
activity is directly correlated to the temperature-dependent
change in CD signal shown in Figure 3c. However, measure-
ments of ester hydrolysis show that the peptide 2/nanoparticle
complex is catalytically active at 128C (k2 = 0.17 mmÀ1 minÀ1)
and 228C (k2 = 0.25 mmÀ1 minÀ1), while there is no observable
increase over the background buffer catalysis at 32 and 528C.
Apparently, the helical conformation at the catalytic site is
critically affected in the 22–328C temperature range. Thus,
the peptides on the silica surface probably all adopt a partial
helical conformation, which gradually melts at higher temper-
atures. Notably, the original esterase activity is completely
restored (98%) after heating to 808C and subsequent cooling
to 228C, which further shows that the catalytic activity is well-
correlated to the formation of a helix, that is, the system can
be regulated by temperature.
Received: March 11, 2006
Revised: September 21, 2006
The results of the experiments collectively show that the
designed peptides adsorb efficiently and bind strongly to silica
nanoparticles, and that they adopt a defined helical structure
upon binding. Moreover, the design strategy also successfully
incorporated the planned functional properties, as the
observed catalysis of ester hydrolysis shows that the forma-
tion of a helix leads to a catalytic unit that is oriented away
from the surface of the nanoparticles so that catalysis can
proceed unhindered.
The described method has potential use in the creation of
novel recognition elements and catalysts that can be switched
on by the introduction of nanoparticles. By adjusting the
design it might also be possible to construct a two-state system
that would allow the activity to be switched on and off by
small changes in temperature. The ability to create surfaces
with well-defined properties (such as reactive groups
arranged at predetermined distances on the nanometer
scale) would have several important applications in areas
such as biocatalysis, biosensing, and nanotechnology. More-
over, it is well-known that interaction with inorganic surfaces
tends to induce conformational and functional changes in
proteins.[15–17,21,23] Interestingly, it has recently been shown
that these rearrangements do not occur at random, but
Keywords: amino acids · catalysis · helical structures ·
.
nanoparticles · peptides
[1] M. J. Russell, Science 2003, 302, 580.
[2] M. M. Hanczyc, S. M. Fujikawa, J. W. Szostak, Science 2003, 302,
618.
[3] C. Huber, G. Wachtershauser, Science 1998, 281, 670.
[4] C. Huber, W. Eisenreich, S. Hecht, G. Wachtershauser, Science
2003, 301, 938.
[5] K. Johnsson, R. K. Allemann, H. Widmer, S. A. Benner, Nature
1993, 365, 530.
[6] K. Severin, D. H. Lee, A. J. Kennan, M. R. Ghadiri, Nature 1997,
389, 706.
[7] D. H. Lee, J. R. Granja, J. A. Martinez, K. Severin, M. R.
Ghadiri, Nature 1996, 382, 525.
[8] S. Yao, I. Ghosh, R. Zutshi, J. Chmielewski, Angew. Chem. 1998,
37, 478; Angew. Chem. Int. Ed. 1998, 37, 478.
[9] S. Yao, I. Ghosh, R. Zutshi, J. Chmielewski, Nature 1998, 396,
447.
[10] L. Patel, C. Abate, T. Curran, Nature 1990, 347, 572.
[11] R. V. Talanian, C. J. McKnight, P. S. Kim, Science 1990, 249, 769.
[12] M. A. Weiss, T. Ellenberger, C. R. Wobbe, J. P. Lee, S. C.
Harrison, K. Struhl, Nature 1990, 347, 575.
8172
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 8169 –8173

Angewandte
Chemie
[13] K. T. OꢀNeil, R. H. Hoess, W. F. DeGrado, Science 1990, 249,
774.
[14] A. C. Lundh, K. Broo, L. Baltzer, J. Chem. Soc. Perkin Trans. 2
1997, 209.
[15] P. Billsten, P. O. Freskgꢁrd, U. Carlsson, B. H. Jonsson, H.
Elwing, FEBS Lett. 1997, 402, 67.
[16] M. Karlsson, L. G. Mꢁrtensson, B. H. Jonsson, U. Carlsson,
Langmuir 2000, 16, 8470.
[17] M. Lundqvist, I. Sethson, B. H. Jonsson, Langmuir 2004, 20,
10639.
[18] J. Sonnefeld, J. Colloid Interface Sci. 1996, 183, 597.
[19] S. J. Anthony-Cahill, P. A. Benfield, R. Fairman, Z. R. Wasser-
man, S. L. Brenner, W. F. Stafford, C. Altenbach, W. L. Hubbell,
W. F. DeGrado, Science 1992, 255, 979.
[20] A. Kondo, S. Oku, K. Higashitani, J. Colloid Interface Sci. 1991,
143, 214.
[21] W. Norde, J. P. Favier, Colloids Surf. 1992, 64, 87.
[22] S. R. Clark, P. Billsten, H. Elwing, Colloids Surf. B 1994, 2, 457.
[23] M. Karlsson, U. Carlsson, Biophys. J. 2005, 88, 3536.
[24] M. Lundqvist, C. Andresen, S. Christensson, S. Johansson, M.
Karlsson, K. Broo, B. H. Jonsson, Langmuir 2005, 21, 11903.
[25] J. P. Ferris, G. Ertem, J. Am. Chem. Soc. 1993, 115, 12270.
[26] G. Delꢂage, C. Geourjon, Comput. Appl. Biosci. 1993, 9, 197.
Angew. Chem. Int. Ed. 2006, 45, 8169 –8173
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8173