Long-Distance Singlet Energy Transfer along α-Helical Polypeptide Chains

Article abstract of DOI:10.1021/jp9613798

α-Helical polypeptides carrying L-4-biphenylalanine and L-1-naphthylalanine that are separated by up to 11 amino acid units or 19.5 Angstroem were prepared by stepwise peptide synthesis.Singlet energy transfer from the biphenyl group to the napthyl group

Full text of DOI:10.1021/jp9613798

J. Phys. Chem. 1996, 100, 16019-16025
16019
Long-Distance Singlet Energy Transfer along r-Helical Polypeptide Chains
Masahiro Kuragaki and Masahiko Sisido*
Department of Bioscience and Biotechnology, Faculty of Engineering, Okayama UniVersity,
3-1-1 Tsushimanaka, Okayama 700, Japan
ReceiVed: May 13, 1996; In Final Form: July 10, 1996X
R-Helical polypeptides carrying L-4-biphenylalanine and L-1-naphthylalanine that are separated by up to 11
amino acid units or 19.5 Å were prepared by stepwise peptide synthesis. Singlet energy transfer from the
biphenyl group to the naphthyl group was investigated with steady-state fluorescence spectroscopy, and the
efficiency of energy transfer was evaluated as a function of the number of spacer units between the two
nonnatural amino acids. The observed efficiency showed qualitative agreement with theoretical values
calculated from interchromophore distances that were predicted from the helix geometry. When the biphenyl-
naphthyl pair is separated by a single amino acid, higher transfer efficiency than that predicted by theory was
observed. The high efficiency was interpreted in terms of contributions from higher multipole interactions.
An effect of a phenyl group inserted between the donor and acceptor was also studied. The intervening
phenyl group slightly enhanced the energy transfer when the aromatic rings of the donor, mediator, and
acceptor groups were nearly in a face-to-face configuration.
Singlet excitation energy transfer (EnT) along a polypeptide
chain has been studied by a number of workers.1,2 The primary
goal of these studies is to confirm the distance dependence of
EnT that has been predicted to follow the Fo¨rster’s r-6 rule.
The rigid and well-defined helical structure of synthetic
polypeptides is advantageous for the study of the distance
dependence. Most workers have been using synthetic polypep-
tides carrying energy donors and energy acceptors covalently
attached to the both ends. A common problem involved in these
samples is that the termini of oligopeptides are more or less
fluctuated and the end-to-end distance is distributed over some
range that is difficult to evaluate experimentally and theoreti-
cally. Oligoproline chains have been often employed as the
molecular framework, probably because of their synthetic
versatility.1 But oligoproline chains are not as stable as R-helix
and take several different conformations depending on chain
length, solvents, and other conditions.3 The flexibility of the
oligoproline chains leaves some uncertainty on the observed
distance dependence.
In some cases, a γ-benzyl L-glutamate (G) unit was replaced
with an alanine (A) unit, to minimize possible steric conflicts
caused by racemization associated with a fragment condensation.
By putting different numbers of amino acids between B and N
units, the interchromophore distance can be varied periodically
according to the geometry of the helix conformation.
It must be noted that the four G units at the C-terminal and
the Gn units (n ≈ 30-40) at the N-terminal play an essential
role in stabilizing R-helical conformation of the sequence
containing B and N units. Indeed, CD spectroscopy showed
nonhelical conformations for oligopeptides such as Boc-BGm-
NG4 (Boc ) tert-butyloxycarbonyl).
Another advantage of R-helical polypeptide chain as the
molecular framework is that a third component may be inserted
between the donor and acceptor, without changing the donor-
acceptor distances and their orientations substantially. In this
study, two types of polypeptides that carry a phenylalanine (F)
unit between the B and N units were prepared. One of the
polypeptide has a sequence of NFB (II), and the other has a
sequence of NG2FGAB (III). As will be described later, the
former polypeptide has the triad chromophore sequence that is
almost perpendicular to the helix and the latter has the triad
sequence that is almost parallel to the helix.
These samples were used to examine possible roles of
intervening aromatic groups on the EnT. Although EnT and
its distance dependence have been used to measure biological
distances, little has been known as to the possible roles of
aromatic groups located on the pathway of EnT. The study of
EnT on the above model for polypeptides may clarify the
possible mediation effects of aromatic residues.
In this study we have incorporated an energy donor and
acceptor along an R-helical part of a long poly(γ-benzyl
L-glutamate) chain. An R-helical main chain is advantageous
because it is commonly found in protein structures and is stable
in solution due to intramolecular hydrogen bonds. The chro-
mophores were incorporated along the R-helix in the form of
non-natural amino acids with aromatic side groups, L-4-
biphenylalanine (B) and L-1-naphthylalanine (N). A common
form of the present polypeptide samples is written as GnBGm-
NG4 (I-m, m ) 0, 1, 2, 3, 4, 5, 7, 11).
Experimental Section
Synthesis of Polypeptides. Polypeptides were prepared by
a two-step procedure. First Boc-oligopeptides that contains the
sequence of the B-N pair and a G4 unit at the C-terminal (Boc-
BGmNG4) were prepared by stepwise synthesis through a liquid-
phase or solid-phase method. For long Boc-oligopeptides such
as Boc-BG5AG5NG4, fragment condensations of short peptides
such as Boc-BG5A-OH and H-G5NG4 were carried out. Ra-
cemization of up to a few percent has been known to occur at
X Abstract published in AdVance ACS Abstracts, September 1, 1996.
S0022-3654(96)01379-2 CCC: $12.00 © 1996 American Chemical Society

16020 J. Phys. Chem., Vol. 100, No. 39, 1996
Kuragaki and Sisido
the C-terminal unit of the N-fragment during the fragment
condensation. To reduce the effect of the racemization, an
alanine unit was chosen as the C-terminal of the N-fragment
because the small side group of the alanine unit will minimize
the steric perturbation on the helical conformation.
The Boc-oligopeptides were purified with preparative HPLC
until single peaks were obtained and characterized by 1H NMR.
The Boc group was then removed with 3 N HCl in dioxane
and neutralized by washing with sodium bicarbonate solution.
The oligopeptides with free amino terminals were used as
initiators for polymerization of γ-benzyl L-glutamate N-car-
boxyanhydride (G-NCA) in dimethylformamide (DMF). The
resulting polypeptides were fractionated with a Sephadex LH-
60 column in DMF to exclude low-molecular weight-fractions,
if they were found.
group was removed by a 25% trifluoroacetic acid/dichloro-
methane (TFA/DCM) mixture for 20 min. The Boc derivative
of the next amino acid (3-fold excess) was mixed in DMF
containing BOP reagent, HOBt, and diisopropylethylamine
(DIEA). The DMF solution was gently shaken with the resin
for 30 min at 25 °C and washed with DMF and DCM. The
Kaiser test was carried out to detect amino groups remaining
in the resin, and if the result was negative, the next Boc-amino
acid was attached.
Cleaving Peptides from Oxim Resins. The peptides on
oxim resins were cleaved off by two different procedures. When
Boc-peptide benzyl ester was a desired product, the peptide was
cleaved off by putting a 3-fold excess of TosOH‚G-OBzl, DIEA,
and acetic acid into the resin suspended in DMF. The mixture
was gently stirred at 25 °C for 24 h and filtered. The filtrate
was concentrated, and the peptide was precipitated by adding
10% citric acid. After the peptide was dried under vacuum, it
was redissolved in ethyl acetate and precipitated in ether. The
peptide was fractionated with preparative HPLC.
In the following section, procedures for preparing Boc-
oligopeptides are described. More detailed procedures have
been described before on the synthesis of polypeptides that
contain electron donors and acceptors.4
Liquid-Phase Synthesis of Peptides. Boc-NG4-OBzl and
Boc-BG4-OBzl. Boc-G4-OBzl4 was deprotected with 4 N HCl/
dioxane and coupled with Boc-N-OH or Boc-B-OH with
1-ethyl-3-[3-(dimethylamino)propyl]carbodiimide hydrochloride/
1-hydroxybenzotriazole hydrate/triethylamine (EDC‚HCl/HOBt/
TEA) in DMF. The reaction mixture was evaporated, and the
remaining oil (or solid) was redissolved in ethyl acetate. The
latter solution was washed with 10% citric acid, 10% NaCl,
4% NaHCO3, and 10% NaCl solution and evaporated under
reduced pressure. The product was recrystallized from ethyl
acetate/ether and isolated as colorless crystals. 1H NMR (500
MHz) in CDCl3 showed 5 different CRH protons.
Boc-NBG4-OBzl, Boc-FBG4-OBzl, and Boc-ABG4-OBzl. The
pentapeptide Boc-BG4-OBzl was deprotected with 4 N HCl/
dioxane and coupled with the Boc derivative of the N-terminal
amino acid (Boc-N-OH, Boc-F-OH, or Boc-A-OH) through the
same procedure as above. The product was recrystallized from
ethyl acetate/ether and isolated as colorless solid.
Boc-NABG4-OBzl and Boc-NFBG4-OBzl. The corresponding
Boc-hexapeptide (Boc-ABG4-OBzl or Boc-FBG4-OBzl) was
deprotected and coupled with the corresponding Boc-amino acid
through the same procedure as above. The product was
recrystallized from ethyl acetate/ether. 1H NMR (500 MHz)
of Boc-NABG4-OBzl in CDCl3 showed seven different CRH
protons.
Solid-Phase Synthesis Using Oxim Resins. Oxim resins
were prepared by a standard procedure.5 The resin was first
charged with Boc-G-OH (0.71 mmol/g of resin). The extension
of peptides was carried out as follows. First the Boc protecting
When a Boc-peptide with a free C-terminal was needed for
the N-terminal peptide of fragment condensation, the peptide
was cleaved from the resin by mixing a DMF solution of a 3-fold
excess of hydroxypiperidine. The mixture was gently stirred
at 25 °C for 24 h and filtered. The filtrate was concentrated
and redissolved in acetic acid. To the latter solution was added
a 5-fold excess of Na2S2O4 in H2O, and the mixture was stirred
at 25 °C for 1 h. The solution was concentrated, and the peptide
was precipitated by adding water. The peptide was fractionated
with preparative HPLC.
Preparative HPLC. Reverse-phase preparative HPLC was
carried out with a Millipore µ Bondasphere 5µ C4 100 Å (L 19
× 150 mm) column eluted with 0.1% TFA/water with a gradual
addition of acetonitrile from 50% to 100%.
Fragment Condensation. Long Boc-oligopeptides such as
Boc-BG5AG5NG4-OBzl that are denoted as Boc-peptide1-
peptide2-OBzl were synthesized by condensation of Boc-
peptide1-OH with H-peptide2-OBzl. Each fragment was pre-
pared by a liquid-phase or solid-phase method followed by
preparative HPLC. The Boc-peptide1-OH was designed to have
a C-terminal alanine (A) unit in order to minimize conforma-
tional conflicts due to partial racemization at the C-terminal
unit during the fragment condensation. The fragment condensa-
tion was carried out for equimolar mixture of Boc-peptide1-
OH and H-peptide2-OBzl in the presence of EDC‚HCl and
HOBt in DMF. The mixture was stirred at 0 °C for 2 h and
subsequently at 25 °C for 22 h. The resulting peptide was
treated as usual and fractionated with preparative HPLC until a
single peak was obtained.

Energy Transfer along Helical Polypeptides
J. Phys. Chem., Vol. 100, No. 39, 1996 16021
TABLE 1: Characterization of Polypeptide Samples,
Glu(OBzl)n-(chromophore sequence)-Glu(OBzl)4
chromophore
sequencea
nb
[B]/[N]c
B
N
NB
BG1N
NAB
NFB
BG2N
BG3N
BG4N
BG5N
BG4AG2N
BG3FAG2N
BG2AG5N
BG5AG5N
25
48
29
43
41
36
33
34
47
22
31
44
42
31
1.02
1.04
1.01
1.04
1.00
1.00
1.06
1.01
0.99
1.00
1.04
1.07
a B ) L-biphenylalanine, N ) L-1-naphthylalanine, F ) L-phenyl-
alanine, A ) L-alanine, G ) γ-benzyl L-glutamate. b Number of
Glu(OBzl) units on the left-hand side of polypeptides. The estimated
error is (10%. c Molar ratio of biphenyl to naphthyl groups determined
from least-squares resolution of the absorption spectrum.
Figure 1. Absorption spectrum of the polypeptide GnBG2NG4 in
trimethyl phosphate and its resolution into components of naphthyl
(- - -), biphenyl (---), and phenyl (- - -) chromophores. The
component spectra were taken from spectrum of Boc-L-1-naphthyl-
alanine, Boc-L-p-biphenylalanine, and Boc-L-phenylalanine, respec-
tively. The relative weights of the component spectra are 1:1:40 for
naphthyl:biphenyl:phenyl chromophores, respectively.
squares fitting of the observed spectrum to the sum of the
component spectra. The results are listed in Table 1. In most
cases, the B/N ratios are close to unity, indicating that the two
chromophores are correctly incorporated in the polypeptides.
The number average degrees of polymerization of the Gn units
are large enough to hold stable R-helical conformations of the
polypeptides.
Attachment of the Long Glu(OBzl) Chain to the Oli-
gopeptide. Each Boc-oligopeptide was treated with 4 N HCl/
dioxane or 25% TFA/DCM to remove the Boc group and
neutralized by shaking the peptide in chloroform with 4%
NaHCO3 solution. After drying in vacuum, the peptide was
weighed and dissolved in DMF. A 40-fold amount of G-NCA
was added and polymerized at 25 °C for 48-72 h. The
completion of the polymerization was checked by the disap-
pearance of the characteristic IR peaks of NCA. The polypep-
tide was precipitated in ether, washed with ether, and fraction-
ated on a Sephadex LH60/DMF column to remove low-
molecular-weight fractions, if they are present.
Methods
UV, CD, and fluorescence spectra were measured in trimethyl
phosphate that is transparent down to 190 nm. Fluorescence
spectra were measured under an argon atmosphere. The
following spectrometers were used: absorption, Jasco Ubest 55;
CD, Jasco J500; steady-state fluorescence, Jasco FP777. The
output of each spectrometer was interfaced to a NEC PC9801
computer.
Characterization of Polypeptides. Polypeptides were char-
acterized by absorption spectroscopy. As a typical example,
the spectrum of GnBG2NG4 is shown in Figure 1 (solid line).
The spectrum was resolved into contributions from N, B, and
benzyl group, as follows. First, the amount of the N group was
determined by the absorbance at 295 nm and the spectrum
contribution of the N group was taken from the absorption
spectrum of Boc-N (small dash line). The contribution of the
B group was taken from the spectrum of Boc-B (dot line). The
amount of the B unit was, in this case, assumed to be the same
as that of N unit. Finally the amount of the benzyl group was
determined to give the best fit between the combined spectrum
of the three components (dash-dot line) and the observed
spectrum of the polypeptide. In the case of BG2NG4, the
number of benzyl groups that gives the best fit to the observed
spectrum was 40 ( 4. The relatively large uncertainty is caused
by the small absorption coefficient of the phenyl group in this
region. The combined spectrum is shown by a dash-dot line
in Figure 1. Although the observed spectrum of the polypeptide
showed a small red shift compared to the combined spectrum,
the total profile indicates that the polypeptide contains an equal
number of N and B groups and about 40 benzyl groups. Since
seven benzyl groups are already included in the oligopeptide
BG2NG4, the average number of G units at the N side must be
n ) 33 ( 4.
Molecular mechanics calculations of the synthetic polypep-
tides containing several non-natural amino acids were made on
a software PEPCON.6 The structure and energy parameters of
PEPCON were taken from the ECEPP system.7 Molecular
models were drawn with a PC version of NAMOD (version
3).8
Results and Discussion
Conformation and Chromophore Arrangement of Polypep-
tides. UV absorption spectra of the polypeptides were es-
sentially the sum of B and N chromophores, indicating that the
chromophores are correctly incorporated in the polypeptides and
that they are electronically isolated from each other in the ground
state.
The CD spectrum of each polypeptide was measured in TMP
solution. The CD spectrum in the amide absorption region
(190-240 nm) showed a typical pattern of an R-helix for all
the polypeptides examined. As a typical example, the CD
spectrum of GnBG2NG4 is shown in Figure 2. The ∆ꢀ value at
222 nm was in the range -12 ( 2 for all polypeptides,
indicating stable R-helical conformations. The relatively large
scattering of ∆ꢀ values is again caused by the difficulty in
determining the number of G units in Gn.
CD spectra at the aromatic absorption region showed complex
patterns depending on the chromophore sequences. In the case
of GnBG2NG4, for example, a broad positive peak was observed
for the biphenyl group and a small negative peak was detected
The observed spectrum profile sometimes deviated from the
sum of equimolar amounts of N and B units. In these cases,
the numbers of B and benzyl groups were determined by a least-

16022 J. Phys. Chem., Vol. 100, No. 39, 1996
Kuragaki and Sisido
TABLE 2: Center-to-Center Distances and Energy Transfer
Efficiencies for the Biphenyl-Naphthyl Pair on r-Helical
Polypeptides, Glu(OBzl)n-(chromophore
sequence)-Glu(OBzl)4
b
chromophore no. of spacer orientation rcc
sequence
Eent
Eent
amino acids of N unita (Å) (calcd)c (obsd)d
NB
0
I
II
I
II
I
II
I
II
I
II
I
II
I
II
I
II
I
II
I
II
I
II
I
9.9
11.8
12.8
13.6
13.7
13.5
13.7
13.6
6.4
0.92
0.80
0.71
0.64
0.63
0.64
0.63
0.64
0.99
0.96
0.99
1.0
0.60
0.65
0.68
0.54
0.60
0.72
0.64
0.74
0.27
0.25
0.14
0.16
0.96
0.90
0.86
0.84
0.96
0.95
0.61
0.51
0.61
0.66
0.37
0.25
BG1N
1
NAB
1
NFB
1
BG2N
2
8.8
6.7
4.7
BG3N
3
BG4N
4
14.0
13.4
13.1
14.5
14.0
12.8
13.5
12.5
17.6
17.9
20.2
19.6
BG5N
5
Figure 2. Circular dichroic spectrum of the polypeptide GnBG2NG4
in trimethyl phosphate.
BG4AG2N
BG3FAG2N
BG2AG5N
BG5AG5N
7
7
in the absorption region of the naphthyl group. The observation
of CD peaks suggests that the chromophores are not freely
rotating around the CR-Câ (ø1) and Câ-Cγ (ø2) bonds but are
restricted within a narrow range of the rotational angles.
8
11
II
Molecular mechanics calculations were carried out to predict
side-chain orientations, assuming an R-helical main chain. All
glutamate units were replaced by alanine units, for simplicity.
There were two possible orientations for a naphthyl group, one
being with (ø1,ø2) ) (180°,65°) (type I) and the other with
(180°,265°) (type II). The two energy minima have almost
equal energies and equal allowed areas in the (ø1,ø2) energy
contour map. Therefore, two equally possible center-to-center
biphenyl-naphthyl distances exist for each biphenyl-naphthyl
pair. The distances are listed in Table 2. The computer-
predicted conformations of the polypeptide that contains a B-N
pair separated by 11 amino acids are shown in Figure 3.
a Type I orientation: (ø1,ø2) ) (180°,65°). Type II orientation:
(180°,265°). b Center-to-center distance between B and N groups in
Å. c Theoretical efficiency using r0 ) 14.9 Å. d Observed efficiency
from steady-state fluorescence spectrum.
In the case of the polypeptide GnNFBG4 that contains an NFB
sequence, the N-B distance is predicted to be essentially the
same as that in the polypeptide GnNABG4. The computer-
predicted conformation of GnNFBG4 is shown in Figure 4. Since
there is an enough space between the N and B groups, the
insertion of an F group will not change the N-B distance very
much. Similarly, the B-N distance of the polypeptide GnBG3-
FAG2NG4 is predicted to be the same as that of the polypeptide
GnBG4AG2NG4. The computer-predicted conformation of the
former polypeptide is shown in Figure 5. For the two
polypeptides carrying an F unit between B and N groups, we
may discuss the effect of the phenyl group on the EnT under
the condition of the same interchromophore distances and
orientations of biphenyl and naphthyl groups.
Steady-State Fluorescence Spectra and the Efficiency of
Energy Transfer. Fluorescence spectra of the polypeptides
were measured in trimethyl phosphate solution. The excitation
wavelength (275 nm) was chosen so as to minimize excitation
of phenyl groups attached on the side chain. At this excitation
wavelength, 58% of the photoenergy is absorbed by the biphenyl
group and 42% by the naphthyl group. The fluorescence
spectrum of a polypeptide carrying a BG5N sequence is shown
in Figure 6. In the same figure, results of least-squares
decomposition of the spectrum into fluorescence spectra of B
and N groups are shown by dashed lines. Energy transfer
efficiencies were evaluated from the intensity of B fluorescence
(IB) and that of N fluorescence (IN). These intensities are
expressed in terms of the absorption coefficients of the B and
Figure 3. Computer-predicted conformation of the polypeptide
A16BA11NA4: (top) type I orientation of naphthyl group; (bottom) type
II orientation.
N groups (ꢀB ) 9180 and ꢀN ) 6650) at 275 nm, quantum yields
of B and N fluorescence (ηB and ηN), concentrations of B and
N groups ([B] and [N]), and the EnT efficiency (Eent):
IB ) (instrumental factor)ηBꢀB[B](1 - Eent)
(1)
IN ) (instrumental factor)ηN(ꢀN[N] + ꢀB[B]Eent) (2)
Setting [B] ) [N], the IN/IB ratio can be expressed as
IN/IB ) ηN(ꢀN + ꢀBEent)/ηBꢀB(1 - Eent)
(3)

Energy Transfer along Helical Polypeptides
J. Phys. Chem., Vol. 100, No. 39, 1996 16023
Figure 4. Computer-predicted conformation of the polypeptide A16NFBA4. Orientation of the naphthyl group is type I.
Figure 5. Computer-predicted conformation of the polypeptide A16BA3FA3NA4. Orientation of the naphthyl group is type I.
Figure 6. Fluorescence spectrum of the GnBG5NG4 polypeptide in
Figure 7. Energy transfer efficiency plotted against the center-to-center
trimethyl phosphate. The results of least-squares resolution into B and
N components are also shown. Fluorescence spectra of GnBG4 and Gn-
NG4 polypeptides are used as the component spectra for B and N
groups, respectively.
distance between B and N groups, rcc. Experimental data are plotted
against rcc’s that are predicted from molecular mechanics calculations.
Since there are two types of naphthyl orientations, the rcc values have
some uncertainties that are expressed by horizontal bars. The solid curve
represents theoretical values according to Fo¨rster’s r-6 rule with r0 )
14.9 Å.
The ηN/ηB ratio was obtained from the fluorescence intensity
of each polypeptide that contains a single B group or single N
group (ηN(325 nm)/ηB(302.5 nm) ) 0.687, λex ) 275 nm).
Using eq 3, the EnT efficiencies were calculated and the results
are listed in Table 2. The EnT efficiencies are plotted against
the center-to-center B-N distances in Figure 7. Since there
are two possible orientations for a naphthyl group, the possible
ranges of the inter-chromophore distance are shown by hori-
zontal bars.
biphenyl fluorescence and naphthyl absorbance and the fluo-
rescence quantum yield of biphenyl (0.143) to be 14.9 Å.
Most of the data points lie near the theoretical curve,
indicating that the EnT obeys the r-6 law on R-helical
polypeptides and that the chromophores are correctly spaced
up to 20 Å along the R-helix as predicted from conformational
calculations. The steep distance dependence clearly demon-
strates that thermal fluctuations of the R-helix are of minor
importance even at room temperature.
Figure 7 also shows a theoretical curve calculated from the
Fo¨rster’s r-6 theory:
It must be argued that the EnT rates should also depend on
the relative orientation of the two chromophores. The above
discussion presumes that the orientations are dynamically
6
6
Eent ) r0 /(r0 + r6)
(4)
2
averaged to give κ2 ) /3. For the evaluation of orientation
The critical distance of EnT from an excited biphenyl to a
naphthyl group, r0, was calculated from an overlap integral of
factors, the direction of transition moment for fluorescence
emission of a biphenyl group and that for absorption of a

16024 J. Phys. Chem., Vol. 100, No. 39, 1996
Kuragaki and Sisido
naphthyl group are needed. The biphenyl fluorescence is
originated from a vibronic mixing of an electronic transition
polarized along the long axis with a variety of vibrational modes.
Therefore, its fluorescence band consists of many differently
polarized vibronic peaks.9 Furthermore, the absorption band
of the naphthyl group in the 285 nm region is assigned to a 1La
band that is electronically polarized along the short axis.
However, this band is only partially allowed [oscillator strength
≈ 0.05 (calculated10) and 0.13 (observed for 1-methylnaphtha-
lene)] and its polarization depends on the mode of coupling in
each vibronic peak. Therefore, the orientation factor of EnT
from an excited biphenyl to a naphthyl group is preaveraged
and may not sharply depend on the relative orientation of the
two chromophores. In the present study, as a first approxima-
tion, the orientational factor for dynamic average (2/3) was
assumed for all polypeptides that possess different relative
orientations of the two chromophores.
are arranged with designed distances and orientations. Polypep-
tides that contain sequences of NAB and NFB were prepared
for the study of the possible superexchange mediation of the F
group. As expected from Figure 4, the spatial arrangement of
the N-B pair may be kept unchanged by the insertion of a
phenyl group, but the phenyl group is outside the line that
connects the N-B pair. In this case, the EnT was not enhanced
by the phenyl group. Indeed, the efficiency of the NFB
polypeptide was slightly smaller than that of the NAB polypep-
tide.
To examine the effect of the phenyl group that is lying on
the B-N line, polypeptides that contain sequences of BG3-
FAG2N and BG4AG2N were synthesized. In both polypeptides,
the B-N pairs are separated by seven spacer amino acids, but
the former has a phenylalanine unit in the middle of the spacer.
The computer-predicted conformation of the BG3FAG2N polypep-
tide is shown in Figure 5. The center-to-center distances
between the B-N pair in the BA3FA3N sequence are predicted
to be 14.0 and 12.8 Å for type I and II orientations of the N
group, respectively. The corresponding distances in the BA7N
sequence are 13.5 and 12.5 Å. The center-to-center distances
between B and F groups in the BA3FA3N sequence are 5.9 Å
for the two naphthyl orientations, and the center-to-center
distances between F and N groups are 7.7 and 6.6 Å for types
I and II, respectively. The calculation suggested that the B-N
distance is very slightly enlarged by the insertion of the phenyl
group.
Highly Efficient Energy Transfer in BXN and NXB
Sequences. Fairly large deviations were found for three
polypeptides that contain BG1N, NAB, and NFB sequences.
All the three polypeptides have the B-N pair separated
by a single amino acid and show higher EnT efficiencies
than those expected from the Fo¨rster theory. The authors
believe that the high efficiencies are attributed neither to
any problem in polypeptide samples nor any experimental
error involved in fluorescence measurements, since the phe-
nomenon is common to the polypeptides with BXN or NXB
sequences and those polypeptides show normal absorption and
CD spectra.
The EnT efficiencies of the two polypeptides are compared
in Table 2 and in Figure 7. The efficiency increased by 5% by
the insertion of a phenyl group just on the line of the B-N
pair. The enhancement, although very small, may be attributed
to some electronic effect, since it is unlikely to expect that the
B-N distance becomes shorter by the replacement of a γ-benzyl
L-glutamate unit by a phenylalanine unit. Absorption spectra
of the two polypeptides were the same, and the positions of
fluorescence peaks were not shifted at all. These observations
suggest that the small enhancement may be caused by a
superexchange mechanism that involves virtual states of the
phenyl π system. Since the enhancement is very small,
however, a modest conclusion is that a very small mediation
of EnT, if it exists actually, may be observed when donor,
mediator, and acceptor groups are arranged in a face-to-face
configuration.
As illustrated in the case of the polypeptide GnNFBG4 in
Figure 4, a common feature of the BXN or NXB polypeptide
is that the edge-to-edge B-N distance is much shorter than the
center-to-center distance. In the case of GnNFBG4, for example,
the edge-to-edge distance is 9.4 Å for both type I and II
orientations of the naphthyl group, that is shorter than the center-
to-center distances, 13.7 and 13.6, for type I and II orientations,
respectively. The EnT efficiency calculated according to the
Fo¨rster theory is based on the interaction between point dipoles
that are positioned on the centers of the B and N groups. If the
B-N distance is shorter, contributions of higher multipole-
multipole interactions will become more important and will
increase the EnT efficiency. Furthermore, direct interactions
between π orbitals of B and N groups including the Dexter’s
mechanism of EnT may enhance the EnT. This may explain
the high EnT efficiencies of the three polypeptides. High
efficiency of the polypeptide that contains the NB sequence is
also explained in a similar manner. In the case of the
polypeptides that contain BG2N and BG3N sequences, the
center-to-center distances are already very small and the above
consideration makes little sense.
Conclusions
A variety of R-helical polypeptides that contain a pair of B
and N groups were synthesized. Energy transfer efficiency from
steady-state fluorescence measurements followed the Fo¨rster
theory up to 20 Å separation of the B-N pair. The polypeptides
that contain BXN and NXB sequences showed higher EnT
efficiency than those expected from the theory, indicating a
contribution of higher multipole-multipole interactions. The
phenyl group in the NFB sequence showed no evidence of
enhancing EnT. The phenyl group in the BG3FAG2N sequence
showed a slight enhancement of EnT, suggesting the presence
of the superexchange mechanism.
Effect of Phenyl Group Lying between B and N Groups.
One of the unknown factors in the singlet EnT is a possible
effect of π orbitals that are inserted between the energy donor
and acceptor. Although the superexchange mechanism on
electron transfer has been discussed by a number of workers,
the superexchange mechanism on EnT has been attracting very
little attention. Recently, Ghiggino and co-workers11 discussed
theoretically the role of the superexchange mechanism in EnT.
Experimental evidence of the superexchange effect is usually
difficult to obtain, since we have to compare efficiencies of EnT
in the presence and in the absence of an intervening aromatic
group without changing other factors, such as donor-acceptor
distance and orientation.
References and Notes
(1) (a) Stryer, L.; Haugland, R. P. Proc. Natl. Acad. Sci. 1967, 58,
720. (b) Gabor, G. Biopolymers 1968, 6, 809. (c) Chiu, H. C.; Bersohn, B.
Biopolymers 1977, 16, 277. (d) Leclerc, M.; Pre´milat, S.; Englert, A.
Biopolymers 1978, 17, 2459.
R-Helical polypeptides provide an ideal molecular framework
on which energy donor, possible mediator, and acceptor groups

Energy Transfer along Helical Polypeptides
J. Phys. Chem., Vol. 100, No. 39, 1996 16025
(2) (a) Guillard, R.; Leclerc, M.; Loffet, A.; Leonis, J.; Wilmet, B.;
Englert, A. Macromolecules 1975, 8, 134. (b) Leclerc, M.; Pre´milat, S.;
Englert, A. Biopolymers 1978, 17, 2459. (c) Haas, E.; Wilchek, M.;
Katchalski-Katzir, E.; Steinberg, I. Z. Proc. Natl. Acad. Sci. 1975, 72, 1807.
(d) Haas, E.; Katchalski-Katzir, E.; Steinberg, I. Z. Biopolymers 1978, 17,
11.
(5) (a) Kaiser, E. T.; Mihara, H.; Laforet, G. A.; Kelly, J. W.; Walters,
L.; Findeis, M. A.; Sasaki, T. Science 1989, 243, 187. (b) Kaiser, E. T.
Acc. Chem. Res. 1989, 22, 47.
(6) Sisido, M. Peptide Chemistry 1991; Suzuki, A., Ed.; 1992; pp 105-
110.
(7) Momany, F. McGuire, R. F.; Burgess, A. W.; Scheraga, H. A. J.
Phys. Chem. 1975, 79, 2361.
(8) Beppu, Y. Comput. Chem. 1989, 13, 101.
(9) (a) Gustav, K. Z. Chem. 1988, 28, 405. (b) Bree, A.; Edelson, M.;
Zwarich, R. Chem. Phys. 1975, 8, 27.
(10) Skancke, P. N. Acta Chem. Scand. 1964, 18, 1671.
(11) Scholes, G. D.; Ghiggino, K. P. J. Chem. Phys. 1995, 103, 8873.
(3) (a) Okabayashi, H.; Isemura, T.; Sakakibara, S. Biopolymers 1968,
6, 323. (b) Isied, S. S.; Ogawa, M. Y.; Wishart, J. F. Chem. ReV. 1992, 92,
381.
(4) (a) Sisido, M.; Tanaka, R.; Inai, Y.; Imanishi, Y. J. Am. Chem.
Soc. 1989, 111, 6790. (b) Inai, Y.; Sisido, M.; Imanishi, Y. J. Phys. Chem.
1990, 94, 6237. (c) Inai, Y.; Sisido, M.; Imanishi, Y. J. Phys. Chem. 1991,
95, 3847.
JP9613798