Preparation of anthryl group-tagged helical poly(γ-benzyl L-glutamate) self-assembled film on gold surface and its interaction with DNA

Article abstract of DOI:10.1021/ma0116093

A novel poly(γ-benzyl-L-glutamate) (PBLG, 1), which has an anthryl group at the C-terminus of PBLG segment, has been synthesized, and its immobilization onto gold surface based on a specific interhelix interaction and DNA binding property at a self-assembled monolayer surface have been studied. A PBLG carrying a S-S bond at the C-terminus, 3, formed self-assembled monolayers on a gold surface. Within a resulting monolayer of 3, helical rods of the PBLG segment were oriented in the antiparallel manner, allowing an attractive interaction between helix macrodipoles. When a helix-breaking solvent such as DMSO was added to the solution, about a half amount of the adsorbate was released from the monolayer, followed by weakening the interhelix interaction, and a porous monolayer including molecular cavities was formed. Such a cavity size should fit the occupied volume of the helices present in the original assembly. Guest helices of 1 and 2, which has a similar molecular structure to that of I except for the lack of the anthryl moiety, could interact with the host porous monolayer of 1 on gold substrates. The binding properties of DNA onto 1, 2, and 3 mixed monolayers, in which anthryl groups must be located at the monolayer surface, were examined by means of quartz crystal microbalance and X-ray photoelectron spectroscopy. The results indicated that DNA bound to the PBLG mixed monolayers when the anthryl groups at the monolayer surface were in an optimized distribution.

Full text of DOI:10.1021/ma0116093

Macromolecules 2002, 35, 2769-2775
2769
Preparation of Anthryl Group-Tagged Helical Poly(γ-benzyl
L-glutamate) Self-Assembled Film on Gold Surface and Its Interaction
with DNA
Ma sa zo Niw a ,* Ma sa -a k i Mor ik a w a , Tom oh ir o Na beta , a n d Nobu yu k i Higa sh i*
Department of Molecular Science & Technology, Faculty of Engineering, Doshisha University,
Kyotanabe, Kyoto 610-0321, J apan
Received September 10, 2001; Revised Manuscript Received December 18, 2001
ABSTRACT: A novel poly(γ-benzyl-L-glutamate) (PBLG, 1), which has an anthryl group at the C-terminus
of PBLG segment, has been synthesized, and its immobilization onto gold surface based on a specific
interhelix interaction and DNA binding property at a self-assembled monolayer surface have been studied.
A PBLG carrying a S-S bond at the C-terminus, 3, formed self-assembled monolayers on a gold surface.
Within a resulting monolayer of 3, helical rods of the PBLG segment were oriented in the antiparallel
manner, allowing an attractive interaction between helix macrodipoles. When a helix-breaking solvent
such as DMSO was added to the solution, about a half amount of the adsorbate was released from the
monolayer, followed by weakening the interhelix interaction, and a porous monolayer including molecular
cavities was formed. Such a cavity size should fit the occupied volume of the helices present in the original
assembly. Guest helices of 1 and 2, which has a similar molecular structure to that of 1 except for the
lack of the anthryl moiety, could interact with the host porous monolayer of 1 on gold substrates. The
binding properties of DNA onto 1, 2, and 3 mixed monolayers, in which anthryl groups must be located
at the monolayer surface, were examined by means of quartz crystal microbalance and X-ray photoelectron
spectroscopy. The results indicated that DNA bound to the PBLG mixed monolayers when the anthryl
groups at the monolayer surface were in an optimized distribution.
In tr od u ction
DNA is accomplished onto a cationic SAM film on gold
surface on the basis of the electrostatic interaction19 and
onto the acridine-tethered SAM film through intercala-
tion.20 In this paper, we propose a new approach to
immobilization of such functional groups at the mono-
layer surface by using helix-macrodipole interaction.
The DNA-protein hybridized structure is interested
not only in the biological information but also in mo-
lecular materials such as electron conducting media21-23
since the DNA itself has π-electron clouds of stacked
base, and the helical polypeptide generates the electro-
static potential directed from the N-terminus to the
C-terminus by a macrodipole.24,25 Thus, if the helical
polypeptide assemblies are hybridized with DNAs, it is
expected that novel functions such as the current
rectified molecular wire would be realized. We describe
herein the organization of “peptide intercalator”, which
has a DNA intercalatable moiety at one terminal end
of the polypeptide segment, by incorporating into the
polypeptide underlayer through a specific interhelix
interaction and then the binding property of DNA onto
this intercalator-tethered peptide film. The stepwise
preparation of anthryl group-tagged helical peptide self-
assemblies and their interaction with DNA is schemati-
cally represented in Chart 1. PBLGs with and without
anthryl group at the C-terminus of PBLG segment 1
and 2, respectively, are newly prepared and used as
guest peptide intercalators. PBLG having an S-S bond
at the C-terminus, 3, is also synthesized and used as a
monolayer former on a gold surface.
A great deal of research interest has been focused on
the construction of organized polypeptide monolayer on
a solid support by the self-assembling technique.1-4 To
reveal the self-organization mechanism of polypeptides
must be a key element for developing high performed,
peptide-based materials, since protein functions would
derive from such highly organized structures. In this
context, we have devised a strategy in which a purely
synthetic polypeptide such as poly(L-glutamic acid) is
aligned on two-dimensional media.5-9 In our recent
work, it has been found that a self-assembled monolayer
(SAM) of poly(L-glutamic acid) carrying an S-S bond
at the N-terminus has the ability to discriminate
between N- and C-termini of guest polypeptides in water
on the basis of helix-macrodipole interactions.6 In
addition, the spontaneous formation of a double-layered
poly(γ-benzyl-L-glutamate) (PBLG) having an S-S bond
at the chain end on gold surface has been observed, and
such a unique structure was rationalized in terms of
an antiparallel association property of helical rods,
allowing an attractive interaction between macro-
dipoles.10
SAMs of n-alkanethiols deposited on gold have been
widely studied and used for a range of applications
because of their powerful potentials and ease to fabri-
cate highly ordered, stable structures.11-15 Functional-
ized monolayers assembled onto gold surfaces are
currently examined as sensor interfaces16 and active
surfaces for patterning17 and chemical architecture18 of
solid supports. The construction of supramolecular
systems at surfaces, where various functional groups
are tagged at a proposed position, must result in an
intelligent molecular-scale device. Recently, we have
demonstrated that the successful immobilization of
Exp er im en ta l Section
Ma ter ia ls. PBLG having an anthryl group and an S-S
bond at the C-terminus 1 and 3, respectively, and the control
compound, 2, were synthesized as follows. The products were
1
identified by visible, IR, and H NMR spectoscopies, and their
10.1021/ma0116093 CCC: $22.00 © 2002 American Chemical Society
Published on Web 02/16/2002

2770 Niwa et al.
Macromolecules, Vol. 35, No. 7, 2002
Ch a r t 1
Ch a r t 2
Sch em e 1
Sch em e 2
purity and molecular weight distribution (MWD) were checked
by thin-layer chromatography (TLC) and matrix-assisted laser
desorption ionization time-of-flight (MALDI-TOF) mass spec-
trometry, respectively.
P BLG (1). Anthracene-9-carboxylic acid was dissolved in
CH2Cl2 and was reacted with N-(tert-butoxycarbonyl)-1,6-
diaminohexane in the presence of 1-ethyl-3-(3-(dimethyl-
amino)propyl)carbodiimide hydrochloride with 4-(dimethyl-
amino)pyridine at room temperature for 24 h. After removal
of CH2Cl2, the residue was purified by recrystallization from
THF to give a pale-yellow solid and was added to trifluoroacetic
acid. The solution was stirred at room temperature for 2 h,
and trifluoroacetic acid was then removed. The crude product
was washed with 0.1 M potassium carbonate aqueous solution
several times and dried. N-(Anthracene-9-carbonyl)hexanedi-
amine (4) thus obtained was immediately used as an initiator
for the polymerization of γ-benzyl-L-glutamate-N-carboxylic
anhydride (BLG-NCA) in CHCl3. After removal of CHCl3, the
residue was purified by washing with diethyl ether several
times and dried to give the desired compound, 1.
100 °C for 16 h, and the resulting precipitate at room
temperature was recrystallized from acetone to give a pale-
brown solid. 11-Bromo-1-undecylamine hydrobromide thus
obtained and thiourea were dissolved in ethanol, and the
solution was refluxed for 8 h. Hydrolysis of reaction product
was then performed, and the corresponding thiol derivative
formed was reacted with ethylethanethiosulfinate to give the
desired disulfide group-terminated primary amine derivative.
11-(Ethyldithio)-1-undecylamine (5) thus obtained was im-
mediately used as an initiator for the polymerization described
above, and the desired compound, 3, was obtained.
The Mw/Mn values, which denote the MWD of these PBLGs,
evaluated by MALDI-TOF mass spectroscopy were reasonably
narrow (Mw/Mn ) 1.07, 1.06, and 1.08 for 1, 2, and 3,
respectively).
P BLG (2). n-Propylamine was used as an initiator for the
polymerization of BLG-NCA in CHCl3 to give the control
compound, 2.
P BLG (3). 11-Bromo-1-undecanol and phthalimide potas-
sium salt were dissolved in DMF, the mixture was stirred at
50 °C for 6 h, and the insoluble potassium bromide formed
was then filtered off. After removal of DMF, the residue was
washed with 0.2 M KOH aqueous solution several times and
dried to give a white powder. 11-Phthalimido-1-undecanol thus
obtained and hydrazine monohydrate were dissolved in etha-
nol, and the solution was refluxed for 2 h. The resulting
precipitate was filtered off, and the solvent was then removed
under reduced pressure. The residue was washed with 2 M
KOH aqueous solution and water several times and was then
added to 48% hydrobromic acid. The solution was stirred at
P r ep a r a tion of Gold Su bstr a tes. A gold disk electrode
(BAS, surface area 0.02 cm-2) was cleaned by sonication in
0.1 M KOH aqueous solution for a few minutes and careful
polish with diamond and alumina paste (particle size 1 and
0.05 mm, respectively) and then rinsed with water (purified
with a Milli-Q purification system, Millipore). A gold-coated
electrode for quartz crystal microbarance (QCM) measurement
was cleaned by rinsing with CHCl3 and Milli-Q water, respec-
tively, until a frequency change in dry air was not found and
then immediately used. Gold substrates on clean glass plates
used for reflection-absorption FT-IR and X-ray photoelectron
spectroscopy (XPS) measurement were prepared by thermal

Macromolecules, Vol. 35, No. 7, 2002
Helical Poly(γ-benzyl L-glutamate) 2771
evaporation of gold at the pressure of about 10-6 Torr and
cleaned by rinsing with CHCl3 prior to use.
Mea su r em en ts. Cyclic voltammetry (CV) was performed
at 25 °C with a CV-1B cyclic voltammograph (BAS) connected
with an RW-21 X-Y recorder (Rikadenki, Tokyo). A standard
three-electrode configuration was used with the monolayer
sample on gold as the working electrode, Ag/AgCl (3 M NaCl)
as the reference electrode, and a platinum wire as the counter
electrode in a glass vessel. K4[Fe(CN)6] was obtained com-
mercially and used as the redox probe in the blocking experi-
ment. The electrolyte solution containing 1 M KCl as a
supporting electrolyte were deoxgeneted by purging with
nitrogen. The scan rate was kept constant at 100 mV/s for all
measurements.
Quartz crystal microbalance (QCM) measurements were
performed with gold-coated 9 MHz, AT-cut quartz crystals
(USI System, Fukuoka), in which only one side of the resonator
was in contact with the surface of the solution. The frequency
shifts were recorded continuously with a universal counter
(HP53131A, Hewlett-Packard Co., Ltd.) at 20 °C.
Reflection-absorption Fourier transform infrared (RA-
FTIR) spectroscopy was performed on a Nicolet System 800
spectrometer with an mercury-cadmium-tellurium (MCT)
detector. The measurements were carried out with the 1024
scans of interferogram accumulations using a bare gold
substrate as a reference. The optical path was purged with
dry air before and during measurements. A reflection attach-
ment at an incident angle of 80° was used together with a
polarizer.
X-ray photoelectron spectroscopy (XPS) was performed on
a Shimadzu ESCA-1000 system using a Mg KR source. The
takeoff angle of photoemitted electron was set at 90° from the
surface. The peak locations were corrected on the basis of the
C1s line emitted from neutral hydrocarbon.
Visible absorption spectra were recorded on a Shimadzu UV-
2100 spectrometer, using 1 cm path length quartz cells at room
temperature. Fluorescence studies were conducted on a J asco
FP-770 spectrofluorophotometer, using 1 cm path length
quartz cells at room temperature.
F igu r e 1. (A) Absorption spectra of 1(PLGA) in 50 mM
HEPES buffer/DMSO (95/5, v/v) at pH 7.2. The concentration
of 1(PLGA) is kept constant at 24 mM, and the concentration
of calf thymus DNA is 0 unit M (a), 1.2 unit mM (b), and 1.4
unit mM (c). (B) Fluorescence emission spectra of 1(PLGA) in
50 mM HEPES buffer/DMSO (95/5, v/v) at pH 7.2 with
excitation at 350 nm. The concentration of 1(PLGA) is kept
constant at 5.0 mM, and the concentration of calf thymus DNA
is 0 unit M (a), 2.2 × 10-4 unit M (b), and 2.7 × 10-4 unit M
(c).
Resu lts a n d Discu ssion
DNA Bin d in g P r op er ties of 1 in Solu tion s. To
elucidate the DNA intercalation behavior of 1, which
has an anthryl group at the C-terminus of PBLG
segment, the binding experiment was performed in an
aqueous solution. Unfortunately, 1 did not dissolve in
water, and therefore the benzyl groups were removed
to give a water-soluble form 1(PLGA). The binding
experiment was carried out by first dissolving 1(PLGA)
in DMSO and then mixing with calf thymus DNA in an
aqueous buffer (50 mM HEPES; buffer/DMSO ) 95/5
v/v) at pH 7.2, and the binding process was monitored
by means of visible absorption spectroscopy. Figure 1A
shows the spectral change of 1(PLGA) by addition of
DNA at various molar ratios of 1(PLGA) against the
phosphate unit of DNA.
The spectra exhibit remarkable hypochromism and
a slight red shift of absorption maxima. These observa-
tions are explained by a stacking interaction of the
intercalated anthryl moiety of 1(PLGA) with nucleic acid
base pairs in DNA without electrostatic contribu-
tions.26-29 The fluorescence studies were also carried out
to clarify the intercalation of 1(PLGA) with DNA. Figure
1B shows the fluorescence emission spectra of 1(PLGA)
in the presence and absence of DNA at various molar
ratios in buffer solutions. The addition of DNA led to a
drastic fluorescence decrease due to quenching derived
from the formation of an anthracene-nucleic acid base
complex.26-29 When the molar ratio of 1(PLGA) against
DNA phosphate unit reaches to be about 50, further
spectral changes are not observed in both the absorption
and fluorescence spectra. These spectral variation im-
plies that 1(PLGA) surely works as an intercalator to
DNA despite the attachment of a sterically large PLGA
segment to the anthryl group.
P r ep a r a t ion of Host P BLG Mon ola yer of 3 on
Gold Su bstr a tes. The adsorption behavior of 3 onto
gold electrode was first examined by means of two inde-
pendent methods. One of them is the previously estab-
lished electrochemical method,30 in which [Fe(CN)6]4-
has been used as a redox couple probing the blocking
ability due to adsorption of 3 onto gold surface. Another
is to use a quartz crystal microbalance (QCM) that can
monitor in situ the adsorption process of 3, allowing the
quantitative estimation of mass increase on gold at
nanogram level. Figure 2 shows the change in the anodic
peak current (ipa) of [Fe(CN)6]4- by immersing the gold
electrode into CHCl3 solution of 3 at a concentration of
1 mM.
Upon immersion, the peak current apparently de-
creases and the separation of oxidative and reductive
peak potentials increases, indicating that contact of
[Fe(CN)6]4- with electrode surface is depressed due to
the adsorption of 3 onto gold surface. Two minutes

2772 Niwa et al.
Macromolecules, Vol. 35, No. 7, 2002
F igu r e 4. Typical change in cyclic voltammogram of 3-mono-
layer-covered electrode by treating with DMSO: (a) as pre-
pared from CHCl3 solution; (b) as treated with DMSO.
Experimental conditions are the same in Figure 3.
Ta ble 1. F r equ en cy Ch a n ges of QCM Electr od e Du e to
Tr ea tm en t of 3-Mon ola yer w ith DMSO
F igu r e 2. Change in the peak current (ip) of [Fe(CN)6]4- due
to adsorption of 3 onto gold electrode from CHCl3 solution.
Inset shows the typical change in cyclic voltammogram by
immersing clean gold electrode into CHCl3 solution of 3: (a)
bare electrode; (b) 90 s immersion. The electrolyte solution is
1 M KCl containing 3 mM [Fe(CN)6].4-
∆F
∆m × 107
Aa
treatment
as prepared
(Hz)
(g/cm2)
(nm2/molecule)
-44 ( 1 2.3 ( 0.1
2.9 ( 0.1
5.5 ( 0.1
as treated with DMSO -23 ( 2 1.3 ( 0.12
a
A ) apparent occupied areas of 3 on gold surface.
interaction and eventually be in side-by-side, anti-
parallel orientation. The changes both in the blocking
ability toward [Fe(CN)6]4- and in the mass on gold by
addition of helix-breaking solvent such as DMSO are
then investigated to confirm such a structural model.
Figure 4 shows cyclic voltammograms of 3 assembly
covered electrode before and after treatment of DMSO.
Considerable decrease in the blocking ability toward
[Fe(CN)6]4- is observed after treatment of DMSO,
indicating that the packing density of the SAM became
loose. The QCM data for 3 assembly after treatment of
DMSO are summarized in Table 1. It can be seen that
the values of ∆m and apparent occupied area per
molecule became about 1/2- and 2-fold, respectively, of
those before treatment. This means that half of the
molecules were released from the original assembly; i.e.,
they had not directly attached to the gold surface. One
of the possible explanations for this phenomenon is as
follows. The 3 molecule takes R-helix conformation and
exists mainly in a dimeric form due to helix-helix
interaction in CHCl3, and the resultant assembly on
gold surface must be in a double-layered structure. If
this structural model would be resulted from a macro-
dipole interaction working between helices, the double-
layered structure might be converted into the monolayer
structure by weakening such a macrodipole interaction.
The QCM data listed in Table 1 strongly support these
considerations. The monolayer of 3 thus obtained would
carry homogeneously distributed molecular cavities that
must be suitable size for welcoming a guest helical
PBLG.
Im m obilization of An th r acen e-Ter m in ated P BLG
th r ou gh Helix-Helix In ter a ction on Gold . The
interaction of helical guest PBLG 1 with the monolayer
of 3 was examined in view of blocking ability toward
[Fe(CN)6]4- and the mass increase on gold surface.
Figure 5A shows a change in the peak current of
[Fe(CN)6]4- by immersing the 3-monolayer-covered
electrode into a CHCl3 solution containing 1, and Figure
5B shows the resonance frequency change of quartz
crystal upon addition of 1 under the same conditions
for the electrochemical measurement.
F igu r e 3. Frequency shift of clean gold-deposited QCM by
addition of CHCl3 solution of 3 at the concentration of 8.3 mM.
immersion is found to provide the perfect blocking
against [Fe(CN)6],4- resulting from the formation of a
densely packed monolayer of 3 on the electrode. To
further confirm the formation of self-assembly of 3 on
gold, a QCM technique was also employed. A gold-
deposited QCM resonator was placed in pure CHCl3,
and the solution of 3 was then added (Figure 3).
The frequency is found to decrease steeply with
addition and then becomes unchanged after 2 h at a
frequency shift (∆F) of -44 ( 1 Hz in dry air. The ∆F
value can be converted to the mass (∆m) adsorbed on
gold surface by Sauerbrey’s equation.31 As a result, ∆m
value is calculated to be (2.4 ( 0.1) × 10-7 g/cm2, and
the apparent occupied area of 3 molecule on gold surface
can be evaluated to be 2.9 ( 0.1 nm2 per molecule. This
value is well consistent with the cross-sectional area of
helical PBLG (2.0-5.3 nm2),32 suggesting the formation
of a densely packed R-helix assembly of 3 on gold
surface. In fact, when the secondary structure of PBLG
segment on gold surface was examined by means of
FTIR spectroscopy, amide I and II bands were observed
at 1655 and 1549 cm-1, respectively, which are charac-
teristic absorptions based on an R-helix structure.30 In
our previous work, it was discovered that helical PBLG
existed mainly in a dimeric aggregate in CHCl3.10
Therefore, the self-assembly of 3 on gold thus obtained
would take a double-layered structure consisting of the
dimeric aggregate of helices (Chart 1). Such a helix
dimer formation must be due to helix-macrodipole

Macromolecules, Vol. 35, No. 7, 2002
Helical Poly(γ-benzyl L-glutamate) 2773
no anthryl group at the terminal end of PBLG segment
but has the same segment length with that for 1, was
incorporated into the host 3-monolayer on gold in the
same manner as described above. The formation of a
1:1 complex between the guest 2 and the 3-monolayer
was also observed. In addition, the ternary monolayers
composed of 1, 2, and 3 were prepared by mixing of 1
and 2 in CHCl3 solution at various compositions (f1:
molar fraction of 1) and then incorporated into the
3-monolayers. First, to obtain information about the
orientational state of helical PBLG rods on gold surface,
reflection-absorption (RA) FTIR spectra were meas-
ured. Enriquez et al. demonstrated that the nature of
the order and average orientation of the polypeptide
R-helix in the films can be inferred from RA-FTIR
data.34,35 In the amide I and amide II regions (1700-
1500 cm-1), all spectra of 3-monolayer before and after
treatment with DMSO, and mixed monolayer composed
of 1, 2, and 3 prepared at f1 ) 0.3 exhibit two bands at
1660 and 1550 cm-1, which are assigned to the R-helix
structure.33 The additional CdO stretching band due
to ester groups of the side chain of PBLG is observed at
1740 cm-1. By using the ratio of the integrated intensi-
ties of the amide I and amide II regions, the average
tilt of the helix axis from the surface normal was
evaluated to be 43 ( 3° for the 3-monolayer as prepared
from CHCl3 and 66 ( 3° for 3-monolayer after treatment
with DMSO and then with CHCl3 (Chart 1). Such a
difference in the tilting angle can be reasonably inter-
preted by accounting for the molecular packing density
and interhelix interaction in the monolayers. The rigid
PBLG helical rods, anchored to the gold surface through
a flexible methylene spacer, will lie as flat as possible
on the substrate. However, for the densely packed
3-monolayer as prepared from CHCl3, in which the
helical rods are aligned in the antiparallel manner based
on the attractive interaction between macrodipoles, the
helices are obliged to orient more perpendicularly to the
surface due to crowding with the helical rods. On the
other hand, the porous 3-monolayer, in which the helical
rods are loosely packed and aligned in the parallel
manner, horizontal orientation of helices to the surface
must be favored because of an energetically unfavorable
parallel packing of the macrodipoles and a release of
the crowding with helical rods in the monolayer.
The tilting angle of the helical rods for mixed mono-
layer composed of 1, 2, and 3 (f1 ) 0.3) reverted to that
(43 ( 3°) of the original double-layered 3-assembly
(Chart 1).
F igu r e 5. (A) Change in the peak current (ip) of [Fe(CN)6]4-
due to incorporation of 1 into porous 3-monolayer on gold
electrode. Inset shows the typical change in cyclic voltammo-
gram by immersing porous 3-monolayer covered electrode into
CHCl3 solution of 1: (a) 0 min and (b) 30 min immersion.
Experimental conditions are the same in Figure 3. (B) Fre-
quency shift of porous 3-monolayer-covered QCM by addition
of CHCl3 solution of 1. Experimental conditions are the same
in Figure 4. Arrow indicates the addition of 1.
The peak current of [Fe(CN)6]4- steeply decreases
with immersion because of incorporation of the guest 1
into the host 3-monolayer on gold. The complexation of
the guest 1 with 3-monolayer is found to be completed
after 30 min immersion, which leads the perfect block-
ing toward [Fe(CN)6].4- The result is also supported by
QCM data as shown in Figure 5B. The resonance
frequency decreases drastically with addition of the
guest 1 and becomes unchanged after about 30 min at
a frequency shift of -28 ( 2 Hz in dry air. Such a
frequency decrease must be due to a mass increase on
the QCM electrode, resulting from the adsorption of the
guest 1 onto the 3-monolayer. The total mass increase
(∆m) could be calculated to be (1.5 ( 0.1) × 10-7 g/cm2.
By using both this value and the values listed in Table
1, one can compute the molar ratio of the guest 1
molecule to 3 molecule of the host monolayer to be close
to unity, implying the formation of a 1:1 complex
through an interhelix interaction. If such an interaction
is based upon the helix-macrodipole moment and
allows antiparallel orientation of helical rods, the an-
thryl group of 1 is expected to be fixed at the monolyer
surface, and then the DNA duplex must be bound to
the monolayer surface through the intercalation of
anthryl groups. Consequently, a novel DNA-polypep-
tide hybridized ultrathin film would be constructed on
a gold surface. A reference compound of 2, which has
DNA Bin d in g P r op er ties of An th r yl Gr ou p s-
Ta gged P BLG F ilm on Gold Su r fa ce. The interaction
of the 2/3 binary monolayer with DNA was tested for
comparison with the monolayer containing 1. When the
resonance frequency became unchanged within ( 1 Hz
in 5 mM Tris/HCl buffer at pH 7.2, in which the 2/3
binary monolayer-deposited QCM resonator was placed,
a buffer solution of DNA was injected, and the frequency
shift was recorded as shown in Figure 6a.
Upon addition of DNA solution, the frequency de-
crease is observed by a slight shift, and the same
tendency of frequency shift is also observed in the blank
experiment, in which the buffer solution without DNA
was added. Therefore, the observed small shift must be
due to some physical changes around the QCM electrode
by addition of aqueous solutions. In fact, when the
resonance frequency was measured in dry air, no
frequency difference was observed within experimental

2774 Niwa et al.
Macromolecules, Vol. 35, No. 7, 2002
F igu r e 8. XPS Au4f spectra at a 90° takeoff angle for a bare
substrate (a) 1, 2, and 3-mixed monolayer prepared at f1
0.3 (b) and after immersing this monolayer into DNA solution
(c).
)
F igu r e 6. Frequency shifts of 1, 2, and 3-mixed monolayers
covered QCMs prepared from f1 ) 0 (a), 0.3 (b), 0.5 (c), 0.6 (d),
and 1.0 (e) upon addition of DNA solution at pH 7.2 ([DNA-
(phosphate unit)] ) 1.0 × 10-5 M.
F igu r e 7. Frequency shift due to adsorption of DNA onto 1,
2, and 3-mixed monolayers covered QCMs as a function of f1.
errors before and after addition of DNA solution. These
results suggest that this binary monolayer did not bind
DNA duplex. On the other hand, for the monolayers
composed of 1, 2, and 3, which were prepared by mixing
1 and 2 in CHCl3 solution at various compositions (f1 )
0.3, 0.5, 0.6, and 1.0) and then incorporated into the
3-monolayers (Figure 6b-e), obvious frequency de-
creases were observed, suggesting binding of DNA onto
mixed monolayer surfaces. The same experiment was
performed for the monolayers prepared from the other
compositions of f1. Figure 7 shows a total frequency shift
during adsorption of DNA as a function of the f1 value.
F igu r e 9. XPS C 1s (A) and P 2p (B) spectra at a 90° takeoff
angle for 1, 2, and 3-mixed monolayer prepared at f1 ) 0.3 (a)
and after immersing this monolayer into DNA solution (b).
(f1 ) 0.3), XPS spectra were measured before and after
binding of DNA duplex (Figure 8).
Attenuation of the gold signal for the monolayer after
immersing into a DNA solution is observed, resulting
from the immobilization of DNA onto the monolayer
surface. Additionally, the film thickness could be esti-
mated from XPS data based on the exponential attenu-
ation of the Au 4f signal,36 wherein the mean free path
of Au photoelectron through an organic overlayer was
obtained using a monolayer of hexadecanethiol as a
standard.37 For the monolayers (f1 ) 0.3) before and
after binding of DNA, the film thicknesses were evalu-
ated to be 2.0 and 3.6 nm, respectively, were obtained;
i.e., the average thickness of DNA overlayer can be
calculated to be 1.6 nm. This value may be reasonable
and demonstrates that the DNA overlayer must not be
It is clearly seen from the figure that the amount of
the frequency shift depends on the monolayer composi-
tion, f1, and the mixed monolayers prepared at higher
f1 values (f1 ) 0.7-1.0) do not allow to interact with
DNA. This result suggests that the lateral density of
the anthryl group on the monolayer surface is too high
to intercalate with DNA. In the lower f1 region (f1 <
0.7), however, more effective binding of DNA is found
to take place, indicating the necessity to avoid the steric
crowdedness for optimized binding capacity. To confirm
the immobilization of DNA onto the mixed monolayer

Macromolecules, Vol. 35, No. 7, 2002
Helical Poly(γ-benzyl L-glutamate) 2775
Ack n ow led gm en t. This work was supported in part
by the project “Nano Structure Hybrid” of High Tech-
nology Center at RCAST of Doshisha University.
Refer en ces a n d Notes
(1) Whitesell, J . K.; Chang, H. K. Science 1993, 621, 73.
(2) Worley, C. G.; Linton, R. W.; Samulski, E. T. Langmuir 1995,
11, 3805.
(3) Miura, Y.; Kimura, S.; Imanishi, Y.; Umemura, J . Langmuir
1998, 14, 2761.
F igu r e 10. Possible illustration of DNA-PBLG hybridized
thin film on gold surface.
(4) Fujita, K.; Bunjes, N.; Nakajima, K.; Hara, M.; Sasabe, H.;
Knoll, W. Langmuir 1998, 14, 6167.
(5) Higashi, N.; Shimoguchi, M.; Niwa, M. Langmuir 1992, 8,
1509.
at least a multilayer structure, since the diameter of
DNA is known to be about 2 nm although it is sensitive
to its surroundings. Further direct evidence of im-
mobilization of DNA onto the monolayer surface was
obtained by P2p and C1s signals as shown in Figure 9.
In the P2p region, the signal at 133 eV assigned to
the phosphate unit of DNA was observed, which was
not appeared at all in the spectrum before binding. In
the C1s region, the characteristic signal of DNA assigned
to the nucleic acid base is observed at 284 eV besides
the other contributions. Therefore, it is evident that the
successful binding of DNA is performed at the two-
dimensional polypeptide monolayer surface as illus-
trated schematically in Figure 10.
(6) Higashi, N.; Sunada, M.; Niwa, M. Langmuir 1995, 11, 1864.
(7) Niwa, M.; Matsui, M.; Koide, K.; Higashi, N. J . Mater. Chem.
1997, 7, 2191.
(8) Niwa, M.; Morikawa, M.; Higashi, N. Langmuir 1999, 15,
5088.
(9) Niwa, M.; Morikawa, M.; Higashi, N. Angew. Chem., Int. Ed.
2000, 39, 960.
(10) Niwa, M.; Murata, T.; Kitamatsu, M.; Matsumoto, T.; Hi-
gashi, N. J . Mater. Chem. 1999, 9, 343.
(11) Nuzzo, R. G.; Allara, D. L. J . Am. Chem. Soc. 1983, 105, 4481.
(12) Katz, E.; Willner, I. Langmuir 1997, 13, 3364.
(13) Frey, B. L.; J ordan, M.; Schmitt, F.-J .; Guder, H.-J .; Anger-
maier, L.; Knoll, W. J . Chem. Phys. 1993, 99, 7012.
(14) Niwa, M.; Mori, T.; Higashi, N. Macromolecules 1993, 26,
1936.
(15) Spinke, J .; Liley, C. E.; Kornguth, S.; Corn, R. M. Anal. Chem.
Orientational variation of PBLG helical rods on gold
surface induced by hybridization with DNA was further
tried to examine by measuring the RA-FTIR spectrum.
The OdP-O- stretching vibration assigned to the
phosphate unit of DNA was clearly observed at 1060
cm-1 in addition to the CdO, CdC, and CdN vibration
1995, 67, 4452.
(16) Willner, I.; Lapidot, N.; Riklin, A.; Kasher, R.; Zahavy, E.;
Katz, E. J . Am. Chem. Soc. 1994, 66, 1535.
(17) Kumar, A.; Biebuyck, H. A.; Whitesides, G. M. Langmuir
1994, 10, 1498.
(18) Zhong, C.-J .; Porter, M. D. Anal. Chem. 1995, 67, 709A.
(19) Higashi, N.; Inoue, T.; Niwa, M. Chem. Commun. 1997, 1507.
(20) Higashi, N.; Takahashi, M.; Niwa, M. Langmuir 1999, 15,
111.
assigned to the nucleic acid base at 1700-1500 cm-1
.
The amide I and amide II bands of PBLG segment were
unfortunately superposed by remarkable absorptions of
nucleic acid base, and it was therefore difficult to
examine the change in the orientation of PBLG helical
rods on gold surface.
(21) Okahata, Y.; Kobayashi, T.; Tanaka, K.; Shimomura, M. J .
Am. Chem. Soc. 1998, 120, 6165.
(22) Arkin, M. R.; Stemp, E. D. A.; Holmlin, R. E.; Barton, J . K.;
Hormann, A.; Olson, E. J . C.; Barbara, P. F. Science 1996,
273, 475.
(23) Kelley, S. O.; J ackson, N. M.; Hill, M. G.; Barton, J . K. Angew.
Chem., Int. Ed. 1999, 38, 941.
(24) Wada, A. Adv. Biophys. 1976, 9, 1.
Con clu sion s
(25) Hol, W. G. J .; van Duijnen, P. T.; Berendsen, H. J . C. Nature
We have demonstrated a novel approach for organiza-
tion of functional groups such as DNA intercalator,
anthracene, by a host-guest helix-macrodipole inter-
action and a specific interaction with DNA. The SAM
film of 3 prepared from the CHCl3 solution would take
a double-layered structure consisting of the dimeric
aggregate of helices. When the helix-breaking solvent
such as DMSO was added to the solution, the SAM film
of 3 could provide homogeneously distributed molecular
cavities within the film. The porous monolayer thus
obtained could uptake the guest PBLG helices carrying
an anthryl group at the C-terminal end, allowing the
anthryl groups oriented to the solution phase due to
antiparallel macrodipole interaction. Effective binding
of DNA was accomplished when the surface density of
anthryl group was controlled to be appropriate. The
study on the electron transfer through oriented DNA-
helical polypeptide film on gold electrode is now in
progress.
(London) 1978, 273, 443.
(26) Porumb, H. Prog. Biophys. Mol. Biol. 1978, 34, 175.
(27) Fukuda, R.; Takenaka, S.; Takagi, M. J . Chem. Soc., Chem.
Commun. 1990, 1028.
(28) Asakawa, M.; Endo, K.; Kobayashi, K.; Toi, H.; Aoyama, Y.
Bull. Chem. Soc. J pn. 1992, 65, 2050.
(29) Tomosaka, H.; Omata, S.; Anzai, K. Biosci. Biotechnol.
Biochem. 1994, 58, 1420.
(30) Niwa, M.; Mori, T.; Nishio, E.; Nishimura, H.; Higashi, N. J .
Chem. Soc., Chem. Commun. 1992, 547.
(31) Sauerbrey, G. Z. Phys. 1959, 155, 206.
(32) Chang, Y.-C.; Frank, C. W. Langmuir 1996, 12, 5824.
(33) Miyazawa, T.; Blout, E. R. J . Am. Chem. Soc. 1961, 83, 712.
(34) Tsuboi, M. J . Polym. Sci. 1962, 59, 139.
(35) Enriquez, E. P.; Samulski, E. T. Mater. Res. Soc. Symp. Proc.
1992, 255, 423.
(36) Pressprich, K. A.; Maybury, S. G.; Thomas, R. E.; Linton, R.
W.; Irene, E. A.; Murray, R. W. J . Phys. Chem. 1989, 93, 5568.
(37) Bain, C. D.; Troughton, E. B.; Tao, Y.-T.; Evall, J .; Whitesides,
G. M.; Nuzzo, R. G. J . Am. Chem. Soc. 1989, 111, 321.
MA0116093