Binding analysis of peptides that recognize preferentially cis-azobenzene groups of synthetic polymers

Article abstract of DOI:10.1002/psc.1299

Peptides identified by the PD method against a film surface composed of azobenzene-containing synthetic polymers were characterized by QCM measurements. Among the four peptides analyzed, the c16 peptide with the sequence Trp-His-Thr-Leu-Pro-Asn-Ala showed

Full text of DOI:10.1002/psc.1299

Research Article
Received: 15 July 2010
Revised: 18 August 2010
Accepted: 24 August 2010
Published online in Wiley Online Library: 26 October 2010
(wileyonlinelibrary.com) DOI 10.1002/psc.1299
Binding analysis of peptides that recognize
preferentially cis-azobenzene groups of
synthetic polymers‡
J. Chen,a T. Serizawab∗ and M. Komiyamab
Peptides identified by the PD method against a film surface composed of azobenzene-containing synthetic polymers were
characterized by QCM measurements. Among the four peptides analyzed, the c16 peptide with the sequence Trp-His-Thr-
Leu-Pro-Asn-Ala showed the highest binding affinity to the films rich in cis-azobenzene groups. SPR served to determine the
association/dissociation constants of the c16 peptide against the trans- and cis-azobenzene groups. The binding constants
were estimated to be 1.3 × 105 and 1.4 × 106/M, respectively, indicating a high specificity of the c16 peptide for cis-azobenzene
conformer. Then mutants of the c16 peptide were synthesized and characterized to gain information on the structural
requirements for the cis-form specificity. An Ala-scan clearly revealed that all the amino acid residues of the c16 peptide were
essential for the specificity. More importantly, the results with peptides of inverted sequence and containing Gly insertions
c
confirmed that the specificity is strictly derived from the primary sequence of the c16 peptide. Copyright ꢀ 2010 European
Peptide Society and John Wiley & Sons, Ltd.
Keywords: azobenzene; peptide ligands; recognition; cis/trans conformers; PD
Introduction
slight differences in polymer surfaces with an extremely simple
chemical structure. Based on these findings, our interest moved
to the recognition of small organic molecules introduced into
synthetic polymers as pendant groups.
The development of combinatorial biotechnologies such as
genetically engineered PD and CSD provides a highly efficacious
method to identify biological ligands such as peptides and
proteinsthatrecognizetargetmoleculesbasedonaffinityselection
processes [1]. The target could be material surfaces composed
of inorganic compounds such as metals [2,3], semiconductors
[4,5], metal oxides [6,7], as well as organic compounds such
as carbon nanotubes [8,9], carbon nanohorns [10], fullerenes
[11], and synthetic polymers [12–19]. The identified peptides
recognize their target specifically in a non-covalent manner based
on electrostatic, hydrogen bonding, π –π stacking, van der Waals,
and/or hydrophobic interactions. In general, the specific binding
of peptides to the target contributes to lower the Gibbs free
energy of adsorption, and provides a thermodynamic driving
force that is utilized in diverse applications such as catalysts for
the preparation of inorganic nanoparticles [20–23], adsorbents
for patterning [3], surface modifiers [13,21,24–26], and modifiers
of phages [5]/proteins [27] used for their assemblies.
In our previous study [18], the PD method of 7-mer random
peptide libraries was applied to the film surfaces composed of
azobenzene-containing synthetic polymers under visible light, to
indentify peptides that recognize in differentiated manner the
cis- and trans-azobenzene group. Although under visible light
the azobenzene groups are preferentially in the trans-form in
the bulk films, ELISA experiments with the isolated phage clones
suggested that all the identified peptides preferentially recognize
the cis-azobenzene on the film surfaces. Considering the fact that
phage clones were eluted from the film surfaces under acidic
∗
Correspondence to: T. Serizawa, Research Center for Advanced Science and
Technology, The University of Tokyo, 4-6-1 Komaba, Meguro-ku, Tokyo 153-
8904, Japan. E-mail: t-serizawa@bionano.rcast.u-tokyo.ac.jp
After the successful targeting of chlorine-doped polypyrrole
[13] by the PD method, there is increasing interest in the selection
of polymer-binding peptides and their utilization in surface-
functionalizationofpolymericscaffolds.Inourpreviousstudies,we
have identified by the PD method 7-mer peptides that specifically
bind to the film surfaces composed of isotactic [14] or syndiotactic
poly(methyl methacrylate) [15], syndiotactic polystyrene [16], and
poly(L-lactide) [19]. In these studies, quantitative binding analysis
was performed to better understand the binding specificity of the
peptides,intermsoftheaffinitystrengthandkinetics.Thepeptides
successfully recognized the nanoscale arrangements of functional
groups of the polymers, which were dependent on polymeric
stereoregularties, amphiphilicities, and porosities. These results
revealed that short peptides have the potential to discriminate
a
b
‡
Department of Chemistry and Biotechnology, The University of Tokyo, 4-6-1
Komaba, Meguro-ku, Tokyo 153-8904, Japan
ResearchCenterforAdvancedScienceandTechnology,TheUniversityofTokyo,
4-6-1 Komaba, Meguro-ku, Tokyo 153-8904, Japan
Special issue devoted to contributions presented at the E-MRS Symposium
C ‘‘Peptide-based materials: from nanostructures to applications’’, 7-11 June
2010, Strasbourg, France.
Abbreviations used: CSD, cell-surface display; EtOAc, ethyl acetate;
HBTU, O-(benzotriazol-1-yl)-N,N,Nꢁ,Nꢁ-tetramethyluronium hexafluorophos-
phate; HEMA, 2-hydroxyethyl methacrylate; Ka, the binding constant; k1,
the association rate constant; k−1, the dissociation rate constant; NMP,
N-methylpyrrolidone; PBS, phosphate buffered saline; PD, phage display;
QCM, quartz crystal microbalance; RU, resonance unit; SPR, surface plasmon
resonance.
c
J. Pept. Sci. 2011; 17: 163–168
Copyright ꢀ 2010 European Peptide Society and John Wiley & Sons, Ltd.

CHEN, SERIZAWA AND KOMIYAMA
conditions for successful biopanning processes [28], the clones
with affinity for the cis-forms might be more readily eluted than
those with affinity for the trans-forms. Moreover, although it is
difficult to evaluate the exact trans/cis ratio of azobenzene groups,
the cis conformer content on the outermost surfaces of the film
might be increased when coated with aqueous phage solution.
Among the sequences identified, the binding property of the
c16 peptide (H-Trp-His-Thr-Leu-Pro-Asn-Ala-NH2), which consists
of the proposed motif for the cis-form specificity, was analyzed
by QCM measurements. The results clearly confirmed that the
binding affinity of the c16 peptide to films rich in cis-azobenzene
was greater than that to the trans-azobenzene rich films, and that
binding could reversibly be changed by alternate irradiation with
UV and visible light.
In this work, the binding analysis was extended to other
identified peptides and the binding of the c16 peptide was
characterized in more details by QCM and SPR measurements
of related peptide analogs (Figure 1). The QCM results indicated
that the c16 peptide contained the essential sequence responsible
for the specific binding affinity to the cis-azobenzene rich films.
The SPR measurements indicated that the Ka value of the c16
peptide for the cis-azobenzene was tenfold that for the films rich
in trans-azobenzene.
were purchased from Wako Pure Chemical Industries, Ltd. (Tokyo,
Japan), and were used without further purification.
Synthesis of 2-(4-Phenylazophenoxy)ethanol
4-Hydroxyazobenzene (19.8 g, 0.1 mol), potassium carbonate
(13.8 g, 0.1 mol), and potassium iodide (0.6 g, 3.6 mmol) were
dissolved by stirring in n-butanol (60 ml) in a flask under N2
protection. After adding 2-chloroethanol (6.75 ml, 0.1 mol), the
flask was heated in an oil bath (110 ◦C) for 7 h. The product
was filtered and washed twice by water, and then the aqueous
phase was extracted twice with chloroform. Solvent was removed
in vacuum and the resulting crude product was purified by
chromatography on silica gel column (hexane/EtOAc = 1/1, v/v).
Yield: 8.4 g, 35%; ESI-MS (m/z): 242.5 [M] (calcd. 242.27), 243.6
[M+H]+, 265.4 [M+Na]+, 369.4 [M+I]−; 1H NMR (500 MHz, CDCl3)
δ: 8.34–6.79 (m, 9H, ArH), 4.20 (t, 2H) 4.07 (t, 2H), 2.14 (s, 1H).
Synthesis of 2-(4-Phenylazophenoxy)ethyl Acrylate
Acryloyl chloride (2.24 g, 25 mmol) in chloroform (6.6 ml) was
dropped into a chloroform (66 ml) solution containing 2-(4-
phenylazophenoxy)ethanol (6 g, 25 mmol) and triethylamine (3 g,
25 mmol) at 0 ◦C under N2 protection. After 2 h, the mixture was
warmed to room temperature and stirred overnight. The product
wasfilteredoffandwashedtwicebywater. Theaqueousphasewas
then extracted twice with chloroform and dried over magnesium
sulfate. After evaporation of the solvent, the product was purified
by chromatography on silica gel column (hexane/EtOAc = 4/1,
v/v) and recrystallized twice from hexane. Yield: 5.3 g, 72%; ESI-MS
(m/z): 296.5 [M] (calcd. 296.32), 319.5 [M+Na]+; 1H NMR (500 MHz,
CDCl3) δ: 7.95–6.80 (m, 9H, ArH), 6.49, 6.21 and 5.92 (m, 3H), 4.59
(t, 2H), 4.33 (t, 2H).
Materials and Methods
Chemicals
4-Hydroxyazobenzene, 2-chloroethanol, acryloyl chloride, and
HEMA were purchased from Tokyo Chemical Industry Co., Ltd
(Tokyo, Japan). Nα-Fmoc-Nε-biotin-L-lysine (Fmoc-Lys(biotin)-OH)
and NovaSyn TGR resin were obtained from Nova Biochem
(La¨ufelfingen, Switzerland). HBTU, HOBt, DIEA and the other
amino acids such as Fmoc-Ala-OH, Fmoc-Asn(Trt)-OH, Fmoc-
Asp(OtBu)-OH, Fmoc-Gln(Trt)-OH, Fmoc-Gly-OH, Fmoc-His(Trt)OH,
Fmoc-Leu-OH,Fmoc-Met-OH,Fmoc-Phe-OH,Fmoc-Pro-OH,Fmoc-
Ser(tBu)-OH, Fmoc-Trp(Boc)-OH, Fmoc-Thr(tBu)-OH, and Fmoc-
Tyr(tBu)-OH were purchased from Watanabe Chemical Industries
(Hiroshima, Japan). Solvents and other reagents of synthesis grade
Synthesis of Poly[2-(4-phenylazophenoxy)ethyl Acrylate-co-
HEMA]
2-(4-Phenylazophenoxy)ethyl acrylate (1.05 g, 3.6 mmol) and
HEMA (0.38 ml, 3.6 mmol) were copolymerized in the presence
of a free radical initiator 2,2ꢁ-azobisisobutyronitrile (11.7 mg,
71.0 µmol) in DMF (14.7 ml) at 60 ◦C for 24 h. The resulting
copolymer was purified twice by reprecipitation with diethyl ether
and dried under vacuum overnight. The Mn (36 000) and Mw/Mn
(1.7) values were determined by gel permeation chromatography
using a DMF (containing 10 mM LiBr) as solvent (Mn is the number-
averagemolecularweightandMw istheweight-averagemolecular
weight). 1H NMR (500 MHz, CDCl3) δ: 7.85 (4H), 7.48 (3H), 6.95 (2H),
4.50–3.68 (12H), and 0.88–1.45 (6H).
Preparation of Polymer Films
The films of poly[2-(4-phenylazophenoxy)ethyl acrylate-co-HEMA]
(∼20 nm thickness) were prepared by spin-coating the copolymer
in chloroform (17 mg/ml) at 2000 rpm for 1 min on the 27 MHz
QCM sensor tips (Affinix Q, Initiam) for QCM measurements or on
gold-coatedglassslides(SIAKitAu,Biacore)forSPRmeasurements.
The gold surface of QCM and SPR sensor chips were cleaned with
30% H2O2/H2SO4 (1/3, v/v) ‘piranha solution’ for 30 s, followed
by rinsing with Milli-Q water several times before spin-coating.
(Caution! Piranha solution should be handled with extreme care,
and only small volumes should be prepared at one time.) Then,
the films were irradiated by UV light using a hand-held UV lamp
Figure 1. Schematic representation of the peptides specific for cis-form
azobenzene groups on the polymer-film surfaces. The ratio of m/n was
determined to be 0.49 by 1H NMR.
c
wileyonlinelibrary.com/journal/psc Copyright ꢀ 2010 European Peptide Society and John Wiley & Sons, Ltd. J. Pept. Sci. 2011; 17: 163–168

AZOBENZENE-RECOGNIZING PEPTIDES
(SLUV-4, K.K. Iuchi Seieido) at 365 nm for 4 min in a darkroom Results and Discussion
to prepare the cis-rich films. On the other hand, the trans-rich
Binding Analysis of Identified Peptides
films were obtained by keeping the films in a bright room after
spin-coating.
In our previous study [18], 12 phage clones that displayed the
7-mer peptides possibly specific for cis-azobenzene groups were
isolated by the PD method. Among the peptides, in the first
instancethec16peptidewiththesequenceH-Trp-His-Thr-Leu-Pro-
Asn-Ala-NH2 was analyzed as it was consistent with the sequence
motif, Trp-His-Thr-(hydrophobic)-Pro-Asn-Ala, proposed for the
cis-azobenzene specificity. The binding affinity of the biotinylated
c16 peptide was analyzed by QCM measurements monitoring the
subsequent binding of streptavidin. In this work, the c05 (H-Trp-
Pro-Thr-Pro-Pro-Tyr-Ala-NH2, fouraminoacidresiduesareidentical
to the c16 peptide), c04 (H-Met-His-Gln-Gly-Ser-Asn-Thr-NH2 with
two amino acid residues identical to the c16 peptide), and c11
(H-His-Leu-His-Tyr-Ala-Leu-Pro-NH2 with no sequence similarity to
the c16 peptide; negative control) peptides were analyzed for the
binding affinities of their biotinylated derivatives to the polymer
films by QCM measurements.
Figure 2showsthefrequencychangesoftheQCMchipsafterthe
binding of streptavidin to the polymer films precoated with each
biotinylated peptide under visible and UV light to prepare trans-
and cis-azobenzene rich films, respectively. The ratios of −ꢀF
values under UV and visible light (−ꢀFUV/−ꢀFVis) were estimated
asindexofthecis-formspecificity, asshownintheinsetofFigure 2.
As expected, the binding affinity and specificity of the c16 peptide
for the cis-azobenzene rich film were the highest among the
four peptides analyzed. Although ELISA experiments of all phage
clones suggested a specificity for the cis-form [18], the c05 peptide
considerably lost specificity. Although the c04 peptide almost
maintained its specificity for the cis-form, the binding affinity was
Peptide Synthesis
Peptides were synthesized using standard Fmoc chemistry with
NovaSynTGRresin(300 mg, aminogroup0.25 mmol/g)togivethe
peptides with free N- and amidated C-termini. Fmoc amino acids
were activated with 0.45 M HBTU/HOBt in DMF and 0.9 M DIEA in
NMP, and were coupled in threefold molar excess to the resin. To
remove the Fmoc groups, 20% piperidine in NMP was used. The
coupling reactions were repeated sequentially. After the synthesis,
the resins were dried for 4 h and treated with a cleavage solution
containing 2.5% Milli-Q water and 2.5% Triisopropylsilane in TFA
at room temperature for 1 h. After filtration, the crude peptide
products were obtained by precipitation with cooled diethyl
ether and the subsequent evaporation in vacuum overnight, and
were then purified by high-performance liquid chromatography
(ELITE LaChrom, Hitachi Hitechnologies) with a Cosmosil 5C18-
AR-300 column (Nacalai Tesque) at a flow rate of 6 ml/min using
a linear gradient from 1 to 80% acetonitrile in water containing
0.1% TFA. After freeze-drying, the peptides were characterized
by matrix-assisted laser desorption/ionization time-of-flight mass
spectroscopy (Bruker AutoFLEX mass spectrometer).
QCM Analysis
The peptides were biotinylated with Fmoc-Lys(biotin)-OH at the
C-termini and used for QCM measurements. The peptides (1 µM in
PBS,pH7.4)werefirstincubatedwiththepolymerfilmsontheQCM
sensor chips for 1 h at room temperature. After gentle rinsing the
film surfaces with Milli-Q, the chips were dried with nitrogen gas
and were setup into the QCM cell containing 7 ml PBS. The buffer
solution was maintained at 25 ◦C with stirring and the frequency
changes were recorded continuously over time by a 27 MHz QCM
apparatus (Affinix Q, Initiam). After the baseline became stable,
3.5 µl of streptavidin solution (100 µM in PBS) was added into the
cell to give a final concentration of streptavidin of 50 nM. Each
experiment was carried out 3–10 times. The frequency decreases
(−ꢀF, Hz)oftheQCMwascalculatedasfollows:−ꢀF = F0 s −F60 s
,
where F0 s and F60 s is the resonant frequency before and after the
binding of streptavidin for 60 s, respectively.
SPR Analysis
SPR measurements were performed with a dual channel BIAcore
X (BIAcore AB, Uppsala, Sweden) and the data were collected
using Biacore X Control Software ver. 2.3. The gold-coated glass
slides, which had been coated with the polymer films, were setup
accordingtotheBiaevaluationhandbook. Abufferof10 mM HEPES
containing 150 mM NaCl (HBS-N, pH 7.4, Biacore) was flowed at
a rate of 20 µl/min at 25 ◦C. After the baseline became stable,
the peptides dissolved in HBS-N at appropriate concentrations
were applied to the polymer film for 3 min (association), and
then dissociated into HBS-N buffer under the same conditions for
15 min (dissociation). The resulting sensorgrams at four different
concentrations were fitted by local fitting analysis for RUmax and
globalfitting analysisfork1 (1/Ms)andk−1 (1/s)usingBiaevaluation
ver. 4.1 (assuming a 1 : 1 Langmuir binding model). Finally, Ka was
Figure 2. Frequency changes of streptavidin at 50 nM for polymer films
precoated with the C-terminally amidated c16, c05, c04, and c11 peptides
at 1 µM for 1 h under visible (white) and UV light (black). The inset shows
the ratios of the frequency changes under UV light to that under visible
light.
estimated by the following equation: Ka = k1/k (1/M). The
−1
baseline drift was maintained at less than 0.5 (RU/min) during the
SPR measurements.
c
J. Pept. Sci. 2011; 17: 163–168 Copyright ꢀ 2010 European Peptide Society and John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/psc

CHEN, SERIZAWA AND KOMIYAMA
drastically decreased. This observation suggests that the Ka of
the c04 peptide for the cis-form rich films was relatively small
compared to that of the c16 peptide. The low specificity of the c11
peptide was consistent with the previous ELISA.
Table 1. SPR kinetic parameters for the c16 peptide
Polymer film
k1 (Ms)
k
−1 (10−4/s)
Ka (105/M)
Under visible light
Under UV light
37
2.8
5.5
1.3
14
The different results between the ELISA experiments and the
present QCM analysis are possibly derived from the different
molecular environment of the peptides. In the case of phage
clones, 3–5 copies of the peptides are fused onto the N-
terminus of the pIII coat proteins via a Gly-Gly-Gly-Ser linker.
Through this, the peptides on the phage surface might form an
adequate conformation for recognition of the cis-form rich films.
In other words, the c05 and c04 peptides freed from the phage
surfaces seem to lose their binding affinity and/or specificity. As
a consequence, it was found from the QCM measurements that
the c16 peptide exhibits the greatest affinity and specificity for
the cis-azobenzene rich films among the four peptides previously
identified by the PD method [18].
760
peptide for the cis-azobenzene-rich films is relatively large among
polymer-binding peptides.
Detailed Binding Analysis of the c16 Peptide
For a better insight into the importance of the primary amino
acid sequence of the c16 peptide, each amino acid was stepwise
replaced with alanine (Ala-scan). The affinities of the mutant
peptides were analyzed by the QCM measurements under the
same conditions. Figure 4 shows the frequency changes of the
QCM chips after the binding of streptavidin to the polymer
films precoated with each biotinylated mutant peptide under
visible and UV light. The inset shows the ratios of −ꢀF values
(−ꢀFUV/−ꢀFVis)asthespecificityindex.Anyreplacementresulted
in the extreme decrease in the amounts of streptavidin bound to
the film, particularly after the irradiation with UV light. Therefore,
the specificity was obviously decreased, except for W1A. These
observations strongly suggest that all amino acids of the c16
peptide are essential for the binding to cis-azobenzene. In the
case of W1A, the specificity seems to be retained as the amounts
Ka Analysis of the c16 Peptide
SPRmeasurementswereperformedtoestimatekineticparameters
of the c16 peptide. Figure 3 shows typical SPR sensorgrams for
the trans-azobenzene-rich films composed of both the association
and dissociation processes. The sensorgrams were suitably fitted
to determine the kinetic parameters. The Chi-squared values (an
index of fitting reliability) were 7.06 and 1.64 for the trans- and
cis-azobenzene rich films, respectively. The values of less than ten
are considered to be acceptable according to the Biaevaluation
handbook.
ThederivedKa valueofthec16peptideforthetrans-azobenzene
was 1.3 × 105/M and for the cis-azobenzene 1.4 × 106/M, which
is tenfold greater than the former value (Table 1). The significant
increase in the Ka value was mainly due to 20-fold greater k1 (37/Ms
and760/Ms, respectively), sincetheformerk−1 washalfofthelatter
(2.8×10−4/sand5.5×10−4/s,respectively).Inourpreviousstudies
[19,29], the Ka values of 7-mer peptides for each polymer target
have been estimated similarly by SPR measurements, and were in
the range between 104 and 105/M. Therefore, Ka value of the c16
Figure 4. Frequency changes of streptavidin at 50 nM for polymer films
precoated with the mutant peptides of W1A, H2A, T3A, L4A, P5A, and N6A
at 1 µM for 1 h under visible (white) and UV light (black). The inset shows
the ratios of the frequency changes under UV light to that under visible
light. The peptides are C-terminally amidated.
Figure 3. SPR sensorgrams of the c16 peptide with trans-azobenzene rich
polymer-film surfaces.
c
wileyonlinelibrary.com/journal/psc Copyright ꢀ 2010 European Peptide Society and John Wiley & Sons, Ltd. J. Pept. Sci. 2011; 17: 163–168

AZOBENZENE-RECOGNIZING PEPTIDES
Figure 5. EffectsofchangesofW1, T3, andN6ofthec16peptidetoF, S, and
D on the binding to the polymer films. The binding activity was assayed as
described in Figure 4. The inset shows the ratios of the frequency changes
under UV light to that under visible light. The peptides are C-terminally
amidated.
Figure 6. Effects of insertion of a G after H2-T3, T3-L4, and L4-P5 of the
c16 peptide on the binding to the polymer films. The binding activity
was assayed as described in Figure 4. The inset shows the ratios of the
frequency changes under UV light to that under visible light. The peptides
are C-terminally amidated.
of streptavidin bound to the trans-azobenzene rich films is also
drastically decreased.
confirming the importance of the intact sequence. To investigate
whether the essential motif of the c16 peptide consists of smaller
domains in spatial neighborhood, Gly was used as a linker and
inserted into the middle three positions of the c16 peptide. All the
peptides with these insertions (3G, 4G, and 5G) lost the cis-form
specificity (Figure 6). Therefore, the specificity has to derive from
the adequate conformation induced by the primary sequence of
the c16 peptide. It is worthy to note that the c16 peptide did
not form any secondary structure in aqueous solution (data not
shown).
To further investigate the importance of amino acid residues
involved in interactions with the azobenzene groups, the C-
terminal Trp was replaced with Phe (W1F). If the aromatic group
of Trp is essential for the binding to azobenzene groups based on
π –π interactions,W1Fissupposedtobehaveinamannersimilarto
thec16peptide.However,W1Fcompletelylostitsspecificityforthe
cis-azobenzene (Figure 5). This observation clearly suggests that
the indole ring of Trp rather than the benzene ring is essential for
binding. Next, we replaced His with Phe (H2F). Unfortunately, H2F
could not be dissolved in any buffer employed. We then replaced
Thr in position 3, which could interact with azobenzene groups
via hydrogen bonding interactions, with Ser (T3S) to investigate
the importance of its hydroxyl groups (Figure 5). In contrast to our
expectation, T3S showed binding to the trans-form rather than
to the cis-form. It is difficult to rationalize this result; however,
the lack of a methyl group in the middle position of the c16
peptide obviously changes the binding and specificity of the c16
peptide. Furthermore, Asn as additional candidate for hydrogen
bonding interactions with azobenzene groups, was replaced with
Asp (N6D) and the specificity was similarly lost (Figure 5). All
these observations strongly indicate the essential requirement of
each amino acid residue of the c16 peptide not only for efficient
interactions with cis-azobenzene groups but also for adopting the
suitable peptide conformation required for the interactions.
Asalltheaminoacidresiduesofthec16peptidewerefoundtobe
important, we focused on the order of the residues. Interestingly,
the inverted sequence of the c16 peptide (ANPLTHW) also showed
binding affinity and specificity for the cis-azobenzene almost
identical to those of the parent c16 peptide. However, the
shuffled c16 peptide (NWAPHLT) does not show any specificity
Conclusion
The peptides identified by the PD method against the film surface
ofazobenzene-containingcopolymerswerecharacterizedbyQCM
and SPR measurements. Among the four peptides identified,
the c16 peptide, which was fully consistent with our previous
hypothesis of the essential motif specific for cis-azobenzene
groups [18], showed the highest binding affinity and specificity
for the cis-azobenzene-rich films. The association/dissociation
processes of the c16 peptide were quantified by SPR experiments
to estimate the kinetic parameters. The c16 peptide showed a
tenfold greater Ka for the cis-azobenzene than for the trans-
form, due to greater k1. Various mutants of the c16 peptide
were characterized to gain structural information for the cis-form
specificity. Each of the amino acid residues in the c16 peptide
was found to be essential, and could not be replaced even by
amino acids with a similar structure. Moreover, insertion of Gly
into the c16 peptide revealed the importance of the continuous
sequence of the c16 peptide for the specificity. These promising
results on the c16 peptide are likely to give important information
c
J. Pept. Sci. 2011; 17: 163–168 Copyright ꢀ 2010 European Peptide Society and John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/psc

CHEN, SERIZAWA AND KOMIYAMA
about the recognition of photoresponsive azobenzene polymers
for appropriate biological applications in the near future.
15 Serizawa T, Sawada T, Kitayama T. Peptide motifs that recognize
differences in polymer-film surfaces. Angew. Chem. Int. Ed. 2007; 46:
723–726.
16 Serizawa T, Techawanitchai P, Matsuno H. Isolation of peptides that
can recognize syndiotactic polystyrene. Chembiochem 2007; 8:
989–993.
17 Serizawa T, Iida K, Matsuno H, Kurita K. Cellulose-binding
heptapeptides identified by phage display methods. Chem. Lett.
2007; 36: 988–989.
18 Chen J, Serizawa T, Komiyama M. Peptides recognize photorespon-
sive targets. Angew. Chem. Int. Ed. 2009; 48: 2917–2920.
19 Matsuno H, Sekine J, Yajima H, Serizawa T. Biological selection
of peptides for poly(L-lactide) substrates. Langmuir 2008; 24:
6399–6403.
20 Sano KI, Sasaki H, Shiba K. Specificity and biomineralization activities
of Ti-binding peptide-1 (TBP-1). Langmuir 2005; 21: 3090–3095.
21 Slocik JM, Stone MO, Naik RR. Synthesis of gold nanoparticles using
multifunctional peptides. Small 2005; 1: 1048–1052.
References
1 Sarikaya M, Tamerler C, Schwartz DT, Baneyx FO. Materials assembly
and formation using engineered polypeptides. Annu. Rev. Mater. Res.
2004; 34: 373–408.
2 Brown S. Metal-recognition by repeating polypeptides. Nat.
Biotechnol. 1997; 15: 269–272.
3 Naik RR, Stringer SJ, Agarwal G, Jones SE, Stone MO. Biomimetic
synthesis and patterning of silver nanoparticles. Nat. Mater. 2002; 1:
169–172.
4 Whaley SR, English DS, Hu EL, Barbara PF, Belcher AM. Selection
of peptides with semiconductor binding specificity for directed
nanocrystal assembly. Nature 2000; 405: 665–668.
5 Lee SW, Mao CB, Flynn CE, Belcher AM. Ordering of quantum dots
using genetically engineered viruses. Science 2002; 296: 892–895.
6 Sano KI, Shiba K. A hexapeptide motif that electrostatically binds to
the surface of titanium. J. Am. Chem. Soc. 2003; 125: 14234–14235.
7 Hayashi T, Sano K, Shiba K, Kumashiro Y, Iwahori K, Yamashita I,
Hara M. Mechanism underlying specificity of proteins targeting
inorganic materials. Nano Lett. 2006; 6: 515–519.
8 Wang SQ, Humphreys ES, Chung SY, Delduco DF, Lustig SR, Wang H,
Parker KN, Rizzo NW, Subramoney S, Chiang YM, Jagota A. Peptides
with selective affinity for carbon nanotubes. Nat. Mater. 2003; 2:
196–200.
9 Li XJ,Chen W,Zhan QW,Dai LM,Sowards L,Pender M,Naik RR.Direct
measurements of interactions between polypeptides and carbon
nanotubes. J. Phys. Chem. B 2006; 110: 12621–12625.
10 Kase D, Kulp JL, Yudasaka M, Evans JS, Iijima S, Shiba K. Affinity
selection of peptide phage libraries against single-wall carbon
nanohorns identifies a peptide aptamer with conformational
variability. Langmuir 2004; 20: 8939–8941.
11 Morita Y,Ohsugi T,Iwasa Y,Tamiya E.Ascreeningofphagedisplayed
peptides for the recognition of fullerene (C60). J.Mol.Catal.B:Enzym.
2004; 28: 185–190.
12 Adey NB, Mataragnon AH, Rider JE, Carter JM, Kay BK. Characteri-
zation of phage that bind plastic from phage-displayed random
peptide libraries. Gene 1995; 156: 27–31.
13 Sanghvi AB, Miller KPH, Belcher AM, Schmidt CE. Biomaterials
functionalization using a novel peptide that selectively binds to
a conducting polymer. Nat. Mater. 2005; 4: 496–502.
22 Ahmad G, Dickerson MB, Church BC, Cai Y, Jones SE, Naik RR, King JS,
Summers CJ, Kroger N, Sandhage KH. Rapid, room-temperature
formation of crystalline calcium molybdate phosphor microparticles
via peptide-induced precipitation. Adv. Mater. 2006; 18: 1759–1763.
23 Slocik JM, Naik RR. Biologically programmed synthesis of bimetallic
nanostructures. Adv. Mater. 2006; 18: 1988–1992.
24 Pender MJ, Sowards LA, Hartgerink JD, Stone MO, Naik RR. Peptide-
mediated formation of single-wall carbon nanotube composites.
Nano Lett. 2006; 6: 40–44.
25 Sano K, Sasaki H, Shiba K. Utilization of the pleiotropy of a peptidic
aptamer to fabricate heterogeneous nanodot-containing multilayer
nanostructures. J. Am. Chem. Soc. 2006; 128: 1717–1722.
26 Matsui T, Matsukawa N, Iwahori K, Sano KI, Shiba K, Yamashita I.
Realizing a two-dimensional ordered array of ferritin molecules
directly on a solid surface utilizing carbonaceous material affinity
peptides. Langmuir 2007; 23: 1615–1618.
27 Sano K,Ajima K,Iwahori K,Yudasaka M,Iijima S,Yamashita I,Shiba K.
Endowingaferritin-likecageproteinwithhighaffinityandselectivity
for certain inorganic materials. Small 2005; 1: 826–832.
28 Roberts BL, Markland W, Siranosian K, Saxena MJ, Guterman SK,
Ladner RC. Protease inhibitor display M13-phage-selection of high-
affinity neutrophil elastase inhibitors. Gene 1992; 121: 9–15.
29 Serizawa T, Sawada T, Matsuno H. Highly specific affinities of
short peptides against synthetic polymers. Langmuir 2007; 23:
11127–11133.
14 Serizawa T, Sawada T, Matsuno H, Matsubara T, Sato T. A peptide
motif recognizing a polymer stereoregularity. J.Am.Chem.Soc. 2005;
127: 13780–13781.
c
wileyonlinelibrary.com/journal/psc Copyright ꢀ 2010 European Peptide Society and John Wiley & Sons, Ltd. J. Pept. Sci. 2011; 17: 163–168