Photophysical Properties of Tyrosine and Its Simple Derivatives Studied by Time-Resolved Fluorescence Spectroscopy, Global Analysis, and Theoretical Calculations

Article abstract of DOI:10.1021/jp036721c

The photophysical properties of tyrosine and its derivatives with free and blocked functional groups in water were studied by steady-state and time-resolved fluorescence spectroscopy and global analysis. Tyrosine fluorescence intensity decays in water at

Full text of DOI:10.1021/jp036721c

J. Phys. Chem. B 2004, 108, 3879-3889
3879
Photophysical Properties of Tyrosine and Its Simple Derivatives Studied by Time-Resolved
Fluorescence Spectroscopy, Global Analysis, and Theoretical Calculations
†
†
†
†
Katarzyna Guzow, Robert Ganzynkowicz, Alicja Rzeska, Justyna Mrozek,
†
‡
†
,†
Mariusz Szabelski, Jerzy Karolczak, Adam Liwo, and Wiesław Wiczk*
Faculty of Chemistry, UniVersity of Gda n´ sk, Sobieskiego 18, 80-952 Gda n´ sk, Poland, and Quantum Electronics
Laboratory, Faculty of Physics, Adam Mickiewicz UniVersity, Umultowska 85, 61-614 Pozna n´ , Poland
ReceiVed: September 12, 2003; In Final Form: January 9, 2004
The photophysical properties of tyrosine and its derivatives with free and blocked functional groups in water
were studied by steady-state and time-resolved fluorescence spectroscopy and global analysis. Tyrosine
fluorescence intensity decays in water at pH ) 5.5 in the short-wavelength region (290-320 nm) are
monoexponential, whereas, at longer wavelengths, they are biexponential. The monoexponential fluorescence
intensity decay of O-methyl tyrosine across the fluorescence band is observed. The fluorescence lifetimes of
Tyr calculated using a global analysis are equal to 3.37 ( 0.04 ns at the short-wavelength region and 0.98 (
0.12 ns at the longer-wavelength region. This observation, together with the decay-associated spectra, indicate
that the short-lifetime component can be attributed to tyrosine with phenol hydroxyl groups hydrogen-bonded
with water molecules. The rotamer populations calculated from potentials of mean forces, as well as those
1
obtained from H NMR spectroscopy, do not correspond to the pre-exponential factors obtained from
R
â
fluorescence spectroscopy. The calculated energy barriers of rotations about the C -C bond indicate that
the interconversion rate constant for tyrosine and N-acetyl-tyrosinamide are much greater than the fluorescence
rate constant. Monoexponential fluorescence intensity decay of tyrosine derivatives in acetonitrile solution is
observed for all derivatives studied and, contrary to the aqueous solution, the amide group does not quench
the fluorescence. Thus, specific conformation(s) stabilized by the hydrogen-bond network seem to be responsible
for the heterogeneous fluorescence intensity decay of tyrosine derivatives in aqueous solution.
1
. Introduction
acid backbone (the carbonyl of the amide group), acting as an
acceptor. Such a mechanism of the fluorescence quenching of
The fluorescence of aromatic amino acids (phenylalanine,
tyrosine by amide groups was further supported by the
dependence of quenching efficiency on the distance between
the phenol group of tyrosine residue and the amide group, and
tyrosine, and tryptophan) and their residues, incorporated into
a peptide or protein chain, is a subject of extensive studies,
because of the use of such materials as internal probes in
conformational analysis.1-3 In the case of the tyrosine zwitterion
and tyrosine derivatives with an ionized R-carboxyl group,
1
3,14
the type of the substituent on the N amide atom,
as well as
by the parabolic relation of the logarithm of electron-transfer
rate constant from the aromatic portion to the amide group to
3
-8
monoexponential fluorescence decays were observed.
The
7
the ionization potential of aromatic amino acids.
conversion of the R-carboxyl group into the corresponding
amide or its protonation results in a complex fluorescence
decay.1 Some explanation of this behavior was offered by
the Gauduchon and Whal ground-state rotamer model. This
model assumes the existence of well-defined ground-state C -
C bond rotamers, where interconversion time is considerably
longer than the excited-state lifetimes of the rotamers. This
rotamer model has been further extended to the C -C bond
conformers.
from the interaction between the phenol fluorophore and the
quenching groups. The quenching efficiency is controlled by
the orientation of both groups and is dependent on distance.
In the case of tyrosine analogues and derivatives, the rotamer
model showed that the rotamer in which the phenol ring can
come into the closest contact with the carbonyl group had the
shortest fluorescence lifetime. Because of slow rotamer inter-
conversion, the rotamer model predicts that the fluorescence
intensity decay should be described by a sum of three expo-
nentials terms. For those tyrosine derivatives that exhibit a
monoexponential or biexponential decay, it is assumed that the
three or two rotamers have similar unresolved fluorescence
lifetimes or rotamer interconversion is fast, averaging the
-8
6
R
â
â
γ
9
-12
The different lifetimes of the rotamers result
13,14
3-5
emission. However, based on the ground-state rotamer model,
Cowgill,1 Tournon et al., and Feitelson suggested that the
fluorescence quenching of an aromatic amino acid side chain
by the peptide (amide) group occurred because of a charge
transfer between the excited aromatic chromophore (phenol
ring), acting as a donor, and electrophilic units in the amino
5,16
17
18
the photophysical properties of simple tyrosine derivatives or
19
20
21
22
analogues (â-Tyr(Me), â-Hty, Phg(OH), and Tic(OH) )
could not be fully explained. The acetylation of the amino group
of tyrosine or its analogues causes moderate decrease of the
fluorescence quantum yield and the fluorescence lifetime;
however, the fluorescence intensity decay is still monoexpo-
*
Author to whom correspondence should be addressed. E-mail: ww@
3
nential. The methylation of the hydroxyl group of the phenolic
chem.univ.gda.pl.
†
ring of tyrosine increases both the fluorescence quantum yield
and the fluorescence lifetime, in comparison to those of tyrosine,
University of Gda n´ sk.
Adam Mickiewicz University.
‡
1
0.1021/jp036721c CCC: $27.50 © 2004 American Chemical Society
Published on Web 03/02/2004

3880 J. Phys. Chem. B, Vol. 108, No. 12, 2004
Guzow et al.
but it does not change the character of the fluorescence intensity
decay of tyrosine itself, its amide, or its N-acetyl derivatives.
correlated single-photon counting apparatus at the Centre for
Ultrafast Laser Spectroscopy, Adam Mickiewicz University
(Pozna n´ , Poland). The excitation source (λex ) 277 nm) was a
picosecond/femtosecond laser system (Ti:Sapphire “Tsunami”
laser, pumped with an Ar ion laser “BeamLok” 2060 and
equipped with a cooled photomultiplier tube (model 3809U-05
3,14
The photophysical properties of buried tyrosine residues in
proteins are different than those of tyrosine that has been
exposed to the solvent.23 They also undergo substantial changes
while the protein folds from a random coil to a globular
structure.24 To explain the photophysical properties of tyrosine
and the influence of substituents on the amino and carboxyl
group on the photophysical properties in water and acetonitrile,
the obtained experimental results are discussed based on
theoretical calculations of the rotamer populations and the rate
of interconversion between them. The rate constant of rotamer
interconversion, as well as the rotamer population, were
estimated from the respective potential of mean force (PMF),
which is determined by the umbrella-sampling molecular
dynamics (MD) method.25-29
+
MCP)). The half-width of the instrument response function was
32
∼
35 ps. All fluorescence measurements were performed with
the emission polarizer set at 54.7°, with respect to the excitation
polarization. The channel width (δt) was set to 48.8 ps/ch in
all measurements, except for tyrosine amide derivatives in water
for which δt was 24.4 ps/ch. The desired emission wavelength
was selected using a monochromator (∆λ ) 4.8 nm). A Ludox
solution was used to collect the instrument response. Fluores-
cence decays for the sample and reference were measured to 1
4
4
×
10 and 4 × 10 counts in the channel, respectively. All
measurements were performed at 20 °C at pH ) 5.5 adjusted
using a HCl solution. The fluorescence decay data were fitted
by the iterative convolution to the sum of the exponents:
2. Materials and Methods
Tyrosine (Tyr), N-acetyl-tyrosine (AcTyr), N-acetyl-tyrosi-
namide (AcTyrNH2), and O-methyl-tyrosine (Tyr(Me)) were
obtained from Aldrich. N-acetyl-O-methyl-tyrosine methyl ester
t
(AcTyr(Me)OMe) and N-tert-butyloxycarbonyl-O-methyl ty-
I(λ,t) ) R (λ) exp -
(1)
∑i
i
(
)
rosine (BocTyr(Me)) were purchased from Bachem. Water was
purified using the Milli-Q sytem (Millipore). Acetonitrile for
high-performance liquid chromatography (HPLC) was obtained
from LabScan, whereas HCl (“Suprapur”, 30% solution) was
obtained from Merck. All compounds were used without
additional purification, after their purity was checked using
analytical high-performance reversed-phase liquid chromatog-
raphy (RP-HPLC).
τ
i
where τi is the decay time of the ith component and Ri(λ) is its
pre-exponential factor at the emission wavelength λ.
This model assumes that τi is independent of λ; therefore,
the decay curves, which have been taken at different emission
wavelengths, can be analyzed simultaneously.
analyzed by a global least-squares iterative convolution program
that was supplied by Edinburgh Analytical Instruments (FLA-
900 Level 2 software). The adequacy of the exponential decay
fitting was judged by the inspection of the plots of weighted
residuals and by the statistical parameters ø and the shape of
the autocorrelation function of the weighted residuals and serial
variance ratio (SVR). In all cases, the fitting was started at the
peak channel of the response function. The fluorescence lifetime
errors presented in this work were calculated as asymptotic
errors by the software used.
3
3,34
The data were
2
.1. Synthesis. 2.1.1. Synthesis of N-Acetyl-O-methyl-
tyrosine (AcTyr(Me)). N-Acetyl-O-methyl-tyrosine was obtained
from AcTyr(Me)OMe by saponification, according to the
literature procedure.30
2
R
2
.1.2. Synthesis of N-Acetyl-O-methyl-tyrosinamide (AcTyr-
Me)NH2). N-Acetyl-O-methyl-tyrosinamide was obtained from
AcTyr(Me)OMe by ammonolysis, according to the literature
(
3
0
procedure.
2
.1.3. Synthesis of O-Methyl-tyrosinamide (Tyr(Me)NH2).
O-Methyl-tyrosinamide (Tyr(Me)NH2) was synthesized using
BocTyr(Me), by applying a mixed anhydride method, followed
by the removal of the Boc-protecting group, using trifluoroacetic
acid/methylene chloride (50% v/v).30
Decay-associated spectra (DAS), which are the emission
spectra associated with each individual decay component,3
were calculated from eq 2:
5,36
All prepared derivatives were purified by RP-HPLC, and
the chemical purity of all compounds studied was assessed by
analytical RP-HPLC and mass spectrometry (field desorption
i
R (λ)τi
I (λ) ) I (λ)
(2)
(FD) or fast atom bombardment (FAB)). The structure of all
i
SS
synthesized compounds was confirmed by Fourier transform
infrared (FTIR) and H nuclear magnetic resonance (NMR)
analyses.
R (λ)τi
i
(
∑i
)
1
2
.2. Spectroscopic Measurements. The absorption spectra
where Ii(λ) is the emission spectrum associated with the ith
component and ISS is the total steady-state spectrum.
of tyrosine and its studied derivatives were recorded on a
spectrophotometer (Perkin-Elmer, model Lambda-18).
The emission spectra were recorded using a spectrofluorimeter
1
1
2
.4. H NMR Measurements. H NMR spectra of tyrosine
and its derivatives were recorded using a Varian Unity Plus
00 MHz instrument in D2O. All H NMR spectra were obtained
at 20 °C in the pulse and Fourier transform mode, and the
coupling constants were obtained with an estimated accuracy
of (0.2 Hz. For AcTyrNH2, H NMR spectra were recorded in
the temperature range of 20-80 °C, with a step of 10 °C. The
fractional populations pI, pII, and pIII of rotamers I, II, and III,
respectively, that were associated with the rotation about the
(Perkin-Elmer, model LS-50B). The quantum yields of tyrosine
1
5
and its derivatives were obtained by comparing the integral
intensity of the steady-state emission spectra that were excited
at 275 nm (corrected for absorbance) with that of tyrosine in
water, using a value of 0.14 for the latter.3 The optical density
of the sample at this excitation wavelength did not exceed 0.1,
and the monochromator band-pass was set to 2.5 nm in both
the excitation and emission sides.
1
1
R
â
2
.3. Time-Resolved Fluorescence Measurements. The
C -C bond (the ø1 rotamers assigned according to Laws et
al. ) were calculated from the coupling constants between vicinal
5
fluorescence intensity decays were measured using the time-

Photophysical Properties of Tyrosine
J. Phys. Chem. B, Vol. 108, No. 12, 2004 3881
where V0(θ) is the energy function of the system that consists
of the solute and the solvent, Ki is the force constant of the
0
i
restraining potential, and θ is the equilibrium angle of the
restraining potential; we assumed a value of K ) 2.0 kcal/(mol
2
-1
Å ) and we assumed that 13 windows with θ varying from
-
180° to 180°, in increments of 30°, were used. To calculate
the PMF from umbrella-sampling simulation data, we used the
26-28
weighted histogram analysis method (WHAM).
This method
processes the results of simulations from all windows simulta-
neously and removes the contribution due to the restraining
potentials Vi. MD simulations were performed to sample the
configurations corresponding to each window.
The simulations were conducted in the NVT scheme with
39
40,41
the AMBER 5.0 force field and the TIP3P model of water.
Figure 1. Designation of dihedral angles in tyrosine derivatives (top)
and Newman projection about the C -C bond (bottom).
Calculations were performed using a cubic periodic box with
sides 20 Å long. A 7-Å cutoff and an 8-Å second cutoff were
R
â
applied, to control the nonbonded interaction, except electrostatic
R
âR
âS
H and H (H ) using following equations:
40,41
interactions that follows on the parametrization.
Conse-
quently, the minimum-image convention, according to which
the cutoff on nonbonded interactions should not exceed half of
the box size, which prevents the molecule studied from
interacting with its images, was applied. The duration of
simulation in each window was 1 ns and the sampling frequency
was 2 fs. The bin size in counting the occurrence of the dihedral
angle of interest was 5°.
3
R
âR
3
J(H - H ) - J
g
pI )
(3a)
3
∆
J
3
R
âS
3
J(H - H ) - J
g
pII )
(3b)
(3c)
3
∆
J
p ) 1 - p - p
II
Electrostatic energy was evaluated using the particle-mesh
III
I
41
Ewald summation. The force-field parameters of tyrosine were
3
3
3
3
3
39
where ∆ J ) Jt - Jg, and Jt ) 13.56 Hz and Jg ) 2.6 Hz
are the nominal values of the coupling constants for vicinal
protons in trans and gauche conformations, respectively.
taken from the AMBER database. According to the rotamer
3-6
theory, the heterogeneity of the fluorescence intensity decay
of tyrosine derivatives is determined by the population of
ground-state rotamers, which do not interconvert during the
excited-state lifetime of fluorophores. Therefore, we did not
derive additional force fields for the tyrosine derivatives studied
in the excited state, nor did we use them any further. To check
the accuracy of the derived PMF curves, we evaluated the PMF
without symmetry constrains, even for those dihedral angles
for which the PMF surface should be symmetric (the ø2, ø3, æ,
and ψ angles of tyrosine). The deviation of the resulting PMF
curves from 2-fold or 3-fold symmetry, respectively, were <0.1
kcal/mol, which demonstrates that the sampling was sufficient
(see Table 5 later in this paper).
5
2
.5. Theoretical Calculations. One fundamental measure of
the effective interactions between solute molecules in solution
is the potential of mean force (PMF), i.e., the free energy as a
function of the configuration of the solute. In this work, the
PMF profiles were expressed as functions of single dihedral
angles (ø1, ø2, ø3, æ, or ψ), which are defined in Figure 1. Each
PMF is defined by eq 4:
W(θ) ) -RT ln P(θ,T,V)
(4)
where P(θ) is the probability density of the angle of interest
θ) for a given pressure p, temperature T, and volume V of the
system, as expressed by eq 5:
(
The PMF was used to calculate the relative rotamer popula-
tion (p ) by applying eqs 7 and 8:
i
θ2i
PMF(θ)
∫
Ωθexp[(-H(x))/(RT)] dΩ
θ
zi )
∫
exp -
) dθ
(7)
(8)
i
(
θ1
P(θ,T,V) )
(5)
RT
∫
Ωexp[(-H(x))/(RT)] dΩ
zi
pi )
where H is the Hamiltonian of the system (solute plus solvent),
x denotes all degrees of freedom of the solute and the solvent,
V denotes the entire configuration space of the system, and Ωθ
is the configurational space of the system restricted to a given
angle θ.
∑i zi
where R is the gas constant and T is the absolute temperature.
The rate of rotamer interconversion (krot1frot2) was estimated
To determine the PMF by molecular simulations, we used
22
from transition-state theory:
37,38
the umbrella-sampling method.
We imposed a series of
restraining potentials V on the reaction coordinate to ensure that
all regions are sampled sufficiently. Each restraint defined a
sampling window. The total Hamiltonian of the system (includ-
ing the restraining potential) for the ith window was defined
by
kT
h
∆PMF
RT
krot1frot2 ) κ
exp -
(9)
(
)
(
)
where κ is the transition coefficient, k is the Boltzmann constant,
h is Planck’s constant, and ∆PMF is the difference between a
minimal and maximal values of PMF (i.e., the free energy of
activations). It is difficult to assign a value to the transition
coefficient (κ), which incorporates all correction factors and
H (θ) ) V (θ) + V (θ) )
i
0
i
1
2
0
2
V (θ) + K [mod(θ - θ ,2π)] (6)
0
i
42
uncertainties. Using stochastic MD simulations, Kuharski et al.

3882 J. Phys. Chem. B, Vol. 108, No. 12, 2004
Guzow et al.
2
R
a
TABLE 1: Fluorescence Lifetime (τ), Pre-exponential Factor (r), and Quality of Fit (ø ) for Tyrosine (Tyr) and O-Methyl
Tyrosine (Tyr(Me)) in Water at Different Observation Wavelengths
Tyr
Tyr(Me)
2
R
2
R
λ
em (nm)
τ
i
(ns)
R
i
ø
τ (ns)
4.82
4.83
4.82
4.84
4.82
4.83
4.81
R
ø
285
290
300
310
315
320
1
1
1
1
1
1
1
1.12
1.14
1.13
1.07
1.14
1.02
1.14
3.39
3.4
3.4
1
1
1
1.12
1.11
1.14
3.37
1
1.13
2.59
1.14
7.46
1.10
14.2
1.19
1.04
12.7
1.18
1.06
10.2
1.18
1.00
3
3
3
3
3
.31
1.000
3
30
.55; 1.43
.90
.83; 1.38
.05
4.17; 1.27
.99; 0.85; 6.01
.75
4.36; 1.35
.84; 0.81; 6.67
.54
4.55; 1.30
.966; 0.77; 5.47
0.756; 0.244
1.000
0.500; 0.500
1.000
0.607; 0.393
0.443; 0.473; 0.084
1.000
0.507; 0.493
0.407; 0.43; 0.160
1.000
4.85
1
1.11
3
40
4.82
4.95; 1.75
1.000
0.889; 0.112
1.41
1.07
3
3
3
50
60
70
2
3
4.85
4.91; 1.67
1.000
0.863; 0.137
4.72
1.10
2
3
4.79
5.04; 1.97
1.000
0.820; 0.180
6.03
1.12
0.431; 0.569
0.398; 0.391; 0.211
2
a
For one, two, and three components, respectively.
computed a value of 0.4 for cyclohexane isomerization in a low
viscosity; therefore, we chose κ ) 1 for a single bond rotation.
Note that eq 9 provides only a crude estimate of the intercon-
version rate, because it neglects the friction and stochastic forces.
These could be accounted only in direct MD simulations of the
interconversion process, which are, however, infeasible, because
the time scale is too large.
3
. Results and Discussion
.1. Tyrosine and O-Methyl-tyrosine Emission in Water
3
at pH ) 5.5. 3.1.1. Steady-State Fluorescence Spectroscopy.
The absorption and emission spectra of Tyr and Tyr(Me) in
water at pH ) 5.5 are very similar to those of AcTyrNH2 and
AcTyr(Me)NH2, as well as to the published spectra. Methylation
of the hydroxyl group of tyrosine causes a small (∼1-2 nm)
blue-shift of both the absorption and fluorescence spectra,
compared to that of Tyr, and substantially increases the
fluorescence quantum yield (0.201 for Tyr(Me),7
,14,31
compared
to 0.14 for Tyr; see Table 3 later in this paper).
3
.1.2. Time-Resolved Fluorescence Spectroscopy. The fluo-
rescence intensity decays in water at pH ) 5.5 for Tyr, as well
as the fluorescence intensity decay of impurities in the water,
were measured at 9 different observation wavelengths in the
range of 290-370 nm, whereas for Tyr(Me), the decays were
measured at 11 wavelengths (285-370 nm). In the short-
wavelength region (290-320 nm), the fluorescence intensity
decays of Tyr are monoexponential, whereas they are hetero-
geneous for the longer observation wavelengths (see Table 1
and Figure 2). For an adequate description of the Tyr fluores-
cence intensity decays recorded at 330 and 340 nm, two
exponents were needed, whereas for longer observation wave-
Figure 2. Fluorescence intensity decays of O-methyl tyrosine (Tyr-
Me), top) and tyrosine (Tyr, bottom) in water at pH ) 5.5.
(
lengths (350, 360, and 370 nm), functions in the form of a sum
2
models, with either one, two, or three fluorescence lifetimes
used as free or global parameters. The global analysis of the
fluorescence decay curves of Tyr, with an assumption of the
monoexponential decay function and the fluorescence lifetime
of three exponents gave acceptable values of ø . In the case of
R
Tyr(Me), the monoexponential fluorescence intensity decays
were observed up to 340 nm. For longer observation wave-
lengths (350, 360, and 370 nm), the fluorescence intensity
decays were heterogeneous and the heterogeneity, measured as
2
as a free parameter, gave an unacceptably high value of øR
2
(
5.55); however, when the fluorescence lifetime was a global
a deviation from the monoexponential decay (øR values),
2
increased as the observation wavelength increased (see Table
parameter (the same for all data sets), the ø value increased to
R
1
).
The obtained data for the Tyr fluorescence intensity decays
were globally analyzed, assuming different fluorescence decay
6.59. The fit of a biexponential function with two nonglobal
2
lifetimes gave a value of ø ) 1.16. The global fit of the
R
biexponential function with two fluorescence decay timessone

Photophysical Properties of Tyrosine
J. Phys. Chem. B, Vol. 108, No. 12, 2004 3883
TABLE 2: Pre-exponential Factor (r
i
) and the Fractional
with the shorter fluorescence lifetime (1 ns) is red-shifted,
compared to the previous spectrum, and possesses a maximum
at ∼350 nm and is similar to the fluorescence spectrum of
a
Contribution (f
i
) Obtained from Global Analysis (τ
1
and τ
2
Global Parameters) of Tyrosine Fluorescence Decays in
Water at pH ) 5.5b
46-52
tyrosine hydrogen-bonded to the hydrogen acceptor
or the
G
1
G
2
τ ) 3.40 ns τ ) 0.98 ns
47
fluorescence spectrum of tyrosinate. The biexponential fluo-
rescence intensity decay for Tyr and tyrosine residue in a
dipeptide at a single observation wavelength (340 nm) was
2
λem (nm)
R
1
f
1
R
2
f
2
τ
3
R
3
f
3
ø
R
2
3
3
3
3
3
3
3
3
90
00
10
20
30
40
50
60
70
0.994 0.998 0.01
0.995 0.998 0.01
0.988 0.996 0.012
0
0
0
0.88
1.01
1.02
0.97
1.03
5
0
53
observed by Willis and Szabo and Harms et al., as well as
5
4
by Ferreira et al. in a water-glycerol mixture. Taking into
account that the fluorescence lifetime of tyrosinate is much
) than that observed, and that the ionization
of the Tyr hydroxyl in water is thermodynamically allowed but
0.965 0.99 0.035 0.01
0.775 0.917 0.221 0.08 5.28
0.586 0.806 0.401 0.159 6.63
0.409 0.688 0.583 0.283 7.39
0.392 0.603 0.562 0.25 7.08
0
0
47,50,55
shorter (∼30 ps
0
0
0
0.04 1.06
0.03 1.07
0.147 1.06
3
,5,45,47
kinetically forbidden,
it can be assumed that the species
0.386 0.58 0.546 0.237 6.08 0.07 0.183 1.17
with a fluorescence lifetime of ∼1 ns (the spectrum of which
is red-shifted, compared to that of Tyr) is tyrosine that has been
hydrogen-bonded to the water molecule(s) by its hydrogen
a
). b Global ø ) 1.10.
2
R
f
i
) R
i
τ
i
/(∑
i
R
i
τ
i
50,53
group.
The fluorescence excitation spectra of Tyr measured
at 300 and 350 nm (spectra not shown) revealed that the former
is the excitation spectrum of tyrosine, whereas the latter is red-
51
shifted by ∼1-2 nm, as it was observed by Lee et al. for Tyr
hydrogen-bonded complexes. The third component was observed
at longer fluorescence wavelengths (on the tail of the fluores-
cence spectrum of tyrosine, where the fluorescence intensity of
tyrosine is weak), with a small contribution to the fluorescence
decay and the fluorescence lifetime was too long, as it was for
tyrosine (∼7 ns). The long-lifetime component in the sample
decay observed at the long-wavelength fluorescence could be
associated with the fluorescence of water impurities, because
of the similar fluorescence lifetime of the “solvent” blank (8-
1
0 ns). However, the contribution of photoproduct(s) fluores-
5
3,56,57
cence cannot be excluded.
The global analysis of the fluorescence intensity decays of
Tyr(Me) measured under the same conditions as the Tyr
fluorescence intensity decay (water at pH ) 5.5) reveals that
the fluorescence intensity decay is monoexponential. The fit of
the monoexponential function with all nonglobal parameters
2
R
gave a low ø value (1.24), which slightly increased to 1.29
when the fluorescence lifetime is a global parameter (τ ) 4.87
ns). The fit to the biexponential function with one fluorescence
lifetime as a global parameter and all remaining parameters as
2
independent parameters gave a low ø value (1.09) and a
R
Figure 3. Decay-associated spectra (DAS) of tyrosine in water at pH
5.5.
fluorescence lifetime of τ ) 4.82 ns. The second nonglobal
fluorescence lifetime is in the range from 1.2 ns to ∼8 ns with
a very low contribution (∼1%). The obtained fluorescence decay
time of Tyr(Me) is comparable to the published value obtained
)
common time for all analyzed data sets (global parameters) and
3
-5,7
a second independent (nonglobal) time and independent pre-
for single observation wavelength.
The monoexponential
2
R
exponential factors resulted in slightly higher ø values (1.28)
fluorescence intensity decay of Tyr(Me) additionally supports
the conclusion that the biexponential fluorescence intensity
decay of Tyr is caused by the interaction of phenolic hydroxyl
with water molecules. The shorter fluorescence lifetime of
tyrosine, compared to Tyr with an O-methylated hydroxyl group,
is caused by the quenching of Tyr fluorescence by water and
the quenching action can be attributed to a Tyr-H2O exciplex.58
3.2. Influence of a Substituent on Amino or/and Carboxyl
Group on Fluorescence Properties of Tyr and Tyr(Me) in
Water. The absorption spectra of Tyr or Tyr(Me) derivatives
that possess a substituent(s) on the amino group (AcTyr or
AcTyr(Me)), the carboxyl group (TyrNH2 or Tyr(Me)NH2), or
both groups (AcTyrNH2 or AcTyr(Me)NH2) are similar to those
of the parent molecules; however, the fluorescence intensities
of these derivatives are diversified and are dependent on the
type of substituent (Table 3). The stronger fluorescence quench-
ing effect is observed for the TyrNH2 and Tyr(Me)NH2
derivatives, compared to the parent molecules, whereas the
than those for the biexponential fit with nonglobal parameters.
2
R
The ø values substantially increased (to 2.26) when both
fluorescence lifetimes were the global parameters. The fit of
the three-exponential function, with two fluorescence lifetimes
as global parameters and a third fluorescence lifetime, and pre-
exponential factors as nonglobal parameters to the experimental
2
data set gave a low value of øR (1.10) and fluorescence
lifetimes equal to 3.37 and 0.98 ns with the pre-exponential
2
R
factors presented in Table 2. The same value of ø was
obtained using the three-exponential function with all nonglobal
parameters. The longer fluorescence lifetime is very similar to
that published previously3
-6,13,20,43-45
for Tyr at short observa-
tion wavelengths. Based on eq 2 and the parameters collected
in Table 2, the DAS were constructed (Figure 3). The
fluorescence spectrum associated with the longer fluorescence
lifetime (3.4 ns) conforms to the fluorescence spectrum of Tyr,
with a maximum at ∼302 nm, whereas the spectrum associated

3884 J. Phys. Chem. B, Vol. 108, No. 12, 2004
Guzow et al.
2
R
with a protected carboxyl group can change the polarization of
the surrounding water molecules and will lead to higher
polarization of the carbonyl group and an increasing electron-
acceptor property of the amide group that is facilitating electron
transfer. This hypothesis has been supported by the lower
quenching efficiency of the Tyr-Gly dipeptide with the ionized
TABLE 3: Fluorescence Lifetime (τ) and Quality of Fit (ø )
Obtained from Global Analysis, and the Pre-exponential
Factor (r) Obtained from the Single-File Analysis at 315
f
nm, the Fluorescence Rate Constant (k ), and the
Nonradiative Rate Constant (knr) for Tyrosine and Its
Derivatives in Water at pH ) 5.5
2
R
a
(× 10 s-1
-7
nr (× 10 s-1
-8
τ (ns)
.40 ( 0.04
.23
R
ø
〈τ〉 (ns)
k
f
)
k
)
5,6
carboxyl group (electron donor), compared to TyrNH2 and
b
14
Tyr, QY ) 0.14
the N-methyl-substituted amide group, and the weaker acidity
3
1
1.06
AcTyr, QY ) 0.135
1.000 1.31
4.17
2.56
of the carboxyl group of N-acetyl tyrosine, compared to tyrosine
6
4
with a protonated amino group.
3
3
0
3
.2.1. Time-Resolved Fluorescence Spectroscopy. The fluo-
.27 ( 0.01 0.881
rescence intensity decays of Tyr and Tyr(Me) derivatives in
water at pH ) 5.5 were measured at the same observation
wavelength ranges, and the global analysis of the fluorescence
intensity decays were performed in the same manner as those
for the parent molecules (Tyr and Tyr(Me); see Section 3.1.2,
.60 ( 0.08 0.119 1.04
3.22
4.2
2.69
7.88
TyrNH
1.000 8.67
2
, QY ) 0.05
1
1
0
.19
.39 ( 0.01 0.564
.43 ( 0.01 0.436 1.10
1.21
4.15
“
Time-Resolved Fluorescence Spectroscopy”). The obtained
AcTyrNH
1.000 2.82
.67 ( 0.01 0.798
.40 ( 0.07 0.202 1.07
2
, QY ) 0.07
1
1
0
.58
results are collected in Table 3 in which the pre-exponential
factors from the single-file analysis at 310 nm for the Tyr
derivatives and 315 nm for Tyr(Me) are also presented. For the
AcTyr derivative, the double exponential function was needed
to fit to the experimental fluorescence intensity decay. The
revealed fluorescence lifetimes are equal to 3.27 and 0.60 ns
1.6
4.07
5.85
1.66
1.82
Tyr(Me), QY ) 0.201
1.06 4.17
4
.82 ( 0.06
.48 ( 0.04
.49
1
1
AcTyr(Me), QY ) 0.185
1.06
4
4.13
(from the global analysis), with a contribution of R1 ) 0.881
Tyr(Me)NH
1.000 1.74
2
, QY ) 0.06
and R2 ) 0.119 (from single-file analysis), respectively (Figure
4). To the best of our knowledge, this is the first report of the
biexponential fluorescence intensity decay of AcTyr. According
to the literature data, the fluorescence intensity decay of AcTyr
was monoexponential, with fluorescence lifetimes of τ ) 3.2
1
1
0
.52 ( 0.06 0.750
.42 ( 0.14 0.250 1.10
1.43
4.06
6.59
3.72
AcTyr(Me)NH
1.26
2
, QY ) 0.1
2
2
0
.43
.46 ( 0.03 0.936
.66 ( 0.44 0.064 1.02
1
6
4,5
13
45
ns, τ ) 3.6 ns at pH ) 5.2, τ ) 3.27 ns, τ ) 3.32 ns.
The DAS, calculated based on the steady-state fluorescence
spectrum, and obtained fractional contributions have the same
shape and position as the Tyr fluorescence spectrum for both
lifetimes (figure not shown). Moreover, the fluorescence lifetime
decay of AcTyr(Me) is monoexponential, with a fluorescence
lifetime of τ ) 4.48 ns, which is shorter than that of the parent
molecule (4.82 ns).
2.42
4.09
a
2
i
i i
R τ
). b Calculated from the fluorescence decays
〈
τ〉 ) R
i
τ /(∑
i
measured for the short-wave region at 310 nm.
introduction of a second substituent (acetylation of the amino
group) slightly increases the fluorescence quantum yield. The
small quenching effect can be observed for the derivatives that
contain only an acetyl group. For AcTyr and AcTyr(Me), the
fluorescence quantum yields are 96% and 92% of those of the
parent molecule, respectively (see Table 3). A substantial
decrease of the fluorescence quantum yield caused by the
conversion of the ionized carboxyl group into the amide group
The fluorescence intensity decays for the amide derivative
of both Tyr and Tyr(Me) are biexponential. Moreover, the
heterogeneity of the fluorescence intensity decay observed for
TyrNH2, expressed by the ø value obtained from the global
analysis for the fit of the monoexponential function to the
experimental data (ø ) 8.67) is greater than that for Tyr(Me)-
NH (ø ) 1.74). The obtained result for the fluorescence
decay of TyrNH is comparable to the data obtained from the
2
2
R
2
was also observed by other workers.3
-6,15,43,59,60
A weaker
R
2
3
-6,13,43
quenching ability of the fluorescence of the Tyr
and
2
R
1
9,21,22
tyrosine analogues
by the acetyl group, compared to the
3
,5-7,14
single-file analysis presented in the literature.
The DAS
amide group, was also observed and is consistent with the
statement that the N-terminal portion of a peptide is a weaker
quencher of the Tyr fluorescence than its C-terminal por-
for both individuals associated with the long and short decay
times obtained for TyrNH2, as well as Tyr(Me)NH2, as in the
case of AcTyr, are identical (the same shape and position) to
those of the parent molecules (figure not shown). A conversion
of the amino group of the tyrosine amide or O-methyl-
tyrosinamide to a N-acetyl derivative caused an increase of the
fluorescence lifetime, compared to the amide with a protonated
amino group, and simultaneously decreased the heterogeneity
of the decay (see Table 3). For AcTyrNH2, the ø
the global analysis for monoexponential fit to the fluorescence
intensity decay is 3 times lower than that for TyrNH2; moreover,
the decay still remains heterogeneous. The biexponential func-
tion with lifetimes of τ ) 1.67 ns and τ ) 0.40 ns is needed for
an acceptable fit to the experimental data. The fluorescence
13,60,61
tion,
probably because of the difference of redox potential
6
2
between the N- and C-terminal residues. The fluorescence
quantum yield of a tyrosine derivative that possesses two
quenching groups (acetyl and amide) is greater than that of a
tyrosine amide (one quenching group) with a protonated amino
group. The influence of the protonated amino group on the
quenching ability of the phenol chromophore by the amide group
is dependent on the distance between them.1 These parameters
explicitly indicate the important role of the amino group in the
neighborhood of the quencher (carbonyl) in the quenching
process of the fluorescence of aromatic amino acids. The
protonated amino group does not quench the tyramine fluores-
2
R
value from
3,61
cence5 and is not attributed directly to a straight quenching
,6,8
lifetimes obtained from the global analysis are comparable to
the published values.
process of the tyrosine fluorescence,3
,6,8,9,15,63
which has been
3-5,13,43
For AcTyr(Me)NH2 the reduction
discussed in terms of electrostatic effect. The electrostatic field
produced by the ammonium group in the derivatives of tyrosine
of heterogeneity leads to the almost-monoexponential fluores-
2
cence decay. The global ø value for the fit of the monoexpo-
R

Photophysical Properties of Tyrosine
J. Phys. Chem. B, Vol. 108, No. 12, 2004 3885
Figure 4. (- - -) Measured and (s) fitted fluorescence intensity decay
versus time (left set of points represents a laser profile) of AcTyr in
water at pH ) 5.5. Weighted residuals for biexponential and mono-
exponential fits are plotted in the lower portion of the figure.
Figure 5. PMF profile (top) and rotamer probability distribution
function (bottom) for AcTyrNH
2
, each expressed as a function of the
dihedral angle ø
1
. The dashed line represents the best fit of the Gaussian
2
function.
TABLE 4: Fluorescence Lifetime (τ) and Quality of Fit (ø )
R
Obtained from Global Analysis, Pre-exponential Factor (r)
Obtained from the Single-File Analysis at 315 mm,
7
-1
equal to ∼4.1 × 10 s and are the same for all compounds
studied. Much more diversified are the nonradiative rate
constants calculated from equation knr ) (1 - QY)/τ (see Table
f
Fluorescence Rate Constant (k ), and Nonradiative Rate
Constant (knr) for Tyrosine and Its Derivatives in
Acetonitrile
3
), which indicates that the quenching process by an amide
kf
knr
2
R
(× 10 s-1
-7
(× 10 s-1
-8
compound
QY
τ (ns)
R
ø
)
)
group is a dynamic process, the effectiveness of which is
dependent on the distance between the chromophore and the
quencher group, their mutual spatial location, and the hydration
of all molecules. This quenching process can be associated with
the electron (charge) transfer from the excited phenol chro-
mophore (donor) to the amide group (acceptor), as was
Tyra
AcTyr
TyrNH2
AcTyrNH2
Tyr(Me)
AcTyr(Me)
Tyr(Me)NH2
0.2 4.20 ( 0.05
0.2 4.19 ( 0.04
0.2 4.26 ( 0.06
0.18 4.62 ( 0.07
0.19 4.74 ( 0.04
0.19 4.83 ( 0.03
1
1
1
1
1
1
1
1.07
1.04
1.06
1.1
1.12
1.09
1.09
4.69
4.87
4.79
3.9
4
3.93
3.93
1.91
1.9
1.87
1.77
1.71
1.67
1.67
6
,7,13,17,22,60
suggested previously.
AcTyr(Me)NH2 0.19 4.83 ( 0.02
3
.3. Photophysical Properties of Tyrosine Derivatives in
a
Solubility was too low for fluorescence quantum yield and lifetime
Acetonitrile. To explain the influence of water on the tyrosine
photophysical properties and the role of hydrogen-bond network
in the modifications of photophysical properties of tyrosine and
its simple derivatives, we performed time-resolved and steady-
state fluorescence measurements in acetonitrile, which is a
solvent with very weak proton-accepting properties.
3.3.1. Steady-State Fluorescence Spectroscopy. The presence
of substituent(s) on the amino or carboxyl groups does not
change the shape and position of the absorption and emission
spectra of the Tyr derivatives in acetonitrile (spectra not shown).
Absorption spectra of Tyr and Tyr(Me) derivatives in acetonitrile
measurements.
nential function is 1.26 and decreases to a value of 1.02 for the
biexponential fit with lifetimes of τ ) 2.46 ns and τ ) 0.66 ns
with a low (∼6% in the decay) contribution of the second
component (see Table 3). Thus, the hydroxyl group of a phenol
ring devoid of the possibility of forming strong hydrogen-bond
network with the solvent molecule(s) substantially influences
the fluorescence intensity decay of the Tyr derivatives.
The fluorescence rate constants calculated based on the
equation kf ) QY/〈τ〉 are, within the range of experimental error,

3886 J. Phys. Chem. B, Vol. 108, No. 12, 2004
Guzow et al.
Figure 6. PMF profile (top) and rotamer probability distribution
function (bottom) for AcTyrNH , each expressed as a function of the
2
Figure 7. PMF profile (top) and rotamer probability distribution
dihedral angle ψ. The dashed line represents the best fit of the Gaussian
2
function (bottom) for AcTyrNH , each expressed as a function of the
dihedral angle æ. The dashed line represents the best fit of the Gaussian
function.
function.
are red-shifted for ∼4 nm, with a more-pronounced vibrational
structure than in water (the absorption and emission spectra of
Tyr itself could not be determined, because of its low solubility).
The influence of acetonitrile on the fluorescence spectra of Tyr
derivatives was less than that on the absorption spectra, causing
only a small blue-shift (∼1-2 nm) of fluorescence spectra. The
fluorescence quantum yields of the Tyr derivatives in acetonitrile
are much less diversified than those in water, except for Tyr-
the nonradiative rate constants are almost the same, regardless
of the form of the functional group(s), which indicates that the
possibility of hydrogen-bond network formation between the
solvent and the phenol chromophore and amino acid moiety is
very important in effective fluorescence quenching of tyrosine
by the amide group.
3
.4. Potential of Mean Force Calculation Results. Un-
(Me) and AcTyr(Me) (Table 4).
restricted MD simulations usually lead to the sampling of an
energy hypersurface in the vicinity of energy minima, whereas
the PMF derived by the umbrella-sampling method gives more
complete information about the effective energy surface of the
solute molecule. Therefore, we used PMF to estimate the
changes of free energy that are associated with the rotation about
the dihedral angles ø1, ø2, ø3, æ, and ψ. Knowledge of the free-
energy changes allows not only calculation of the rotamer
populations that are associated with a given dihedral angle, but
also estimation of the rotamer interconversion rate constant. For
calculation, we chose tyrosine and the most satirically crowded
N-acetyl-tyrosinamide. Sample PMF profiles and the corre-
sponding probability distribution in ø1, ψ, and æ for AcTyrNH2
are shown in Figures 5-7, respectively, for illustration.
3
.3.2. Time-Resolved Fluorescence Spectroscopy. The fluo-
rescence intensity decays of Tyr derivatives in acetonitrile were
measured at the same observation wavelength ranges and the
global analysis of the fluorescence intensity decays were
performed in the same manner as for the parent molecules (Tyr
and its derivatives; see Section 3.1.2, “Time-Resolved Fluo-
rescence Spectroscopy”). The obtained results are collected in
Table 4, in which the pre-exponential factors from the single-
file analysis at 310 nm are also presented. For all derivatives
studied, the fluorescence intensity decay is adequately described
by a monoexponential function. The fluorescence lifetimes
calculated from global analysis are presented in Table 4.
Derivatives with free, unblocked hydroxyl groups have slightly
shorter lifetimes than those of O-methyl tyrosine and its
derivatives. This difference is probably caused by a slightly
higher fluorescence rate constant calculated for tyrosine deriva-
tives with free hydroxyl groups than the O-methylated ones.
Contrary to the water solution for all tyrosine derivatives studied,
The difference in the free energy of rotation about the
respective dihedral angles obtained from PMF calculations and
the forward and backward rate constant of barrier crossing,
calculated based on eqs 6 and 7, as well as the dihedral angle

Photophysical Properties of Tyrosine
J. Phys. Chem. B, Vol. 108, No. 12, 2004 3887
TABLE 5: Energy Barrier of Rotation around the Dihedral Angle Obtained from PMF Calculation and Forward and
Backward Rate Constants of Barrier Crossing, Calculated Based on eqs 6 and 7 for Tyrosine (Tyr) and N-Acetyl-tyrosinamide
(AcTyrN
2
) in Water
rate constant (× 10 s-1
-9
)
PMFmin1
PMFmax
PMFmin2
rotation on
dihedral angle value (kcal/mol) angle (deg) value (kcal/mol) angle (deg) value (kcal/mol) angle (deg)
forward
backward
Tyr
-115
4
0
0.0
.7
-167
-55
55
-86
99
-58
54
180
-31
150
4.0
183
3.05
3.42
6.42
3.59
3.72
3.01
3.07
3.03
2.01
2.01
3.44
3.44
0.0
0.9
0.7
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
-55
55
167
99
-86
54
180
-58
150
-31
183
111.2
19.3
0.76
14.5
11.7
38.7
34.9
37.4
2.09
2.09
18.7
18.7
36.1
88.5
0.37
14.5
11.7
38.7
34.9
37.4
2.09
2.09
18.7
18.7
ø
1
2
0.9
0.0
0.0
0.0
125
10
179
-2
ø
æ
ψ
0.0
116
-122
-121
60
0.0
0.0
0.0
0.0
0.0
-86
95
ø
3
4.0
AcTyrNH
2
-113
4
0
0.0
.08
-171
-58
63
-89
89
163
-72
96
150
-23
173
-2
3.02
4.51
6.55
3.64
3.64
5.92
8.66
1.00
3.30
4.96
3.25
3.25
0.0
-58
63
43.5
3.1
38.0
7.4
0.11
13.3
18.0
85.5
0.0052
1200
25.2
1.4
27.1
27.1
ø
1
2
0.52
0.08
0.0
0.17
3.38
0.35
0.0
1.4
0.0
0.0
0.0
0
0
0
0
.52
.17
.0
126
-6
180
-123
5
130
-125
50
-171
0.24
18.0
13.3
0.28
0.87
2100
23.7
15.3
27.1
27.1
89
ø
-89
-72
96
.0
æ
ψ
3.38
0
0
1
0
0
.35
.0
.4
163
-23
150
-2
.0
.0
-87
90
ø
3
173
values for appropriate rotamers for tyrosine and N-acetyl-
tyrosinamide in water, are summarized in Table 4.
TABLE 6: Relative Rotamer Population (pop) and Dihedral
2
Angle for Tyr and AcTyrNH in Water Obtained from the
The differences between minimal and maximal values in the
PMF profiles, expressed as a function of the dihedral angles ø2
and ø3, are ∼3.5-3.7 kcal/mol for both Tyr and AcTyrNH2;
thus, the rate constant of rotamer interconversion is on the order
Theoretical Calculations (PMF)
Tyr
AcTyrNH
2
pop (%)
3.6
21.0
5.4
angle (deg)
pop (%)
angle (deg)
of 101 s , i.e., much higher than the fluorescence rate constant.
0 -1
ø
1
, rotamer p
I
6
-55.3
43.3
-57.3
The free-energy barrier obtained for phenol hydroxyl group
rotation (∼3.3-3.4 kcal/mol; see Table 5) is comparable to that
ø
1
, rotamer pII
37.5
published by Hollas,65 which indicates that the barriers of
-166.2
-168.4
62.0
internal rotations obtained from our PMF profiles are reasonable.
The correctness of the PMF profile calculation additionally
supports a lack of temperature dependence, in the range of
experimental error, of the ø1 rotamer population of AcTyrNH2,
ø
1
, rotamer pIII
1
55.2
19.2
ø
2
53.3
46.7
-85.1
98.1
42.5
56.5
-92.4
88.4
1
calculated based on H NMR. For this derivative, the difference
in energy of the ø1 rotamers is in the range of 0.52 kcal/mol.
As a result, the dependency of rotamer population on temper-
ature is very weak. The barrier that is associated with the rotation
of the ammonium and carboxylate group of tyrosine are ∼3
and ∼2 kcal/mol, respectively (see Table 5). For the aforemen-
tioned angle, the PMF profiles are symmetrical and a periodical
function of the dihedral angle. A more-diversified profile
represents the PMF in the dihedral angle ø1 for Tyr and
AcTyrNH2, for which three distinct minima and three maxima
with different heights are present. The highest barriers calculated
for Tyr and AcTyrNH2, relative to the rotation about the dihedral
angle ø1, are ∼6.5 kcal/mol, and the calculated rate constants
for rotamer interconversion between rotamer populations pIII
æ
ψ
3
3
3
5.0
2.8
2.2
-58.2
53.7
180.2
0.2
36.0
63.8
-72.2
99.4
163.8
6
.3
-46.5
-18.3
148.5
4
5
9.8
0.2
-32.0
148.2
4.2
89.5
ø
3
5
49.4
0.6
4.4
184.5
51.0
49.0
-0.5
-180.8
The rotamer populations calculated based on the PMF profile
generally are comparable to those obtained from H NMR
spectroscopy (Table 6). The PMF calculation for tyrosine
overestimates pI and underestimates that of pIII, whereas, for
AcTyrNH2, the opposite is true. Taking into account that the
1
8
8 -1
and pII are ∼4 × 10 and ∼1 × 10 s for Tyr and AcTyrNH2,
respectively. These values are almost 1 order of magnitude
higher than the rate constant for radiative transition for these
compounds (see Table 3). The rate constant of interconversion
between rotamer populations pIII and pI, as well as between
rotamer populations pI and pII (and vice versa) are much higher
for both tyrosine and N-acetyl-tyrosinamide. This conclusion
agrees with the results obtained by Kungl6 for tyrosine and
tyrosine-containing peptides from MD calculations.
5
error associated with NMR rotamer estimation is ∼0.05, the
obtained results are reasonable. Table 7 also presents the ø1
rotamer populations for N-acetyl tyrosine and tyrosinamide and
their O-methylated derivatives. For all compounds studied, three
ø1 rotamers exist in the ground state. As can be observed, the
conversion of the hydroxyl group to the methoxy group does
6

3888 J. Phys. Chem. B, Vol. 108, No. 12, 2004
Guzow et al.
1
TABLE 7: ø
compound
Tyr
1
Rotamer Populations (p
i
) for Tyrosine and Its Derivatives in Water, Calculated Based on H NMR Spectroscopy
p
I
p
II
p
III
a
b
c
a
b
c
a
b
c
0.43; 0.49; 0.49; 0.48
0.24; 0.26; 0.21; 0.25
0.33; 0.25; 0.30; 0.27
0.25; 0.27
0.16; 0.18
b
b
b
Tyr(Me)
AcTyr
0.43; 0.48
0.29; 0.25
b
b
b
0.60; 0.57
0.24; 0.25
AcTyr(Me)
0.57
0.45; 0.46
0.43
0.58; 0.56
0.25
0.18
b
0.38; 0.37b
0.39 b
0.17; 0.17b
0.18 b
TyrNH
Tyr(Me)NH
AcTyrNH
AcTyr(Me)NH
2
2
b
2
0.32; 0.32
0.32
0.10; 0.12
0.09
2
0.59
a
Data from ref 68. b Data from ref 5. c Data from ref 69.
not have any influence on the rotamer populations. The
conversion of the carboxylate into the amide group increases
the population pII at the expense of pIII, whereas the acetylation
of the ammonium group causes an increase of pI and a decrease
of that of pIII, as compared to Tyr, but the changes are not as
large.
causes a decrease of the fluorescence quantum yield, as observed
previously, whereas the fluorescence lifetime is biexponential,
with a small contribution from a subnanosecond component.
The acetyl group also quenches the N-acetyl-O-methyl-tyrosine
(AcTyr(Me)) fluorescence; however, monoexponential fluores-
cence intensity decay is observed for N-acetyl tyrosine with
blocked hydroxyl groups, beside stronger quenching than that
observed for AcTyr. This indicates that the heterogeneous
fluorescence intensity decay observed for AcTyr and other Tyr
derivatives is probably caused by specific hydration of the acetyl
(the entire amino acid moiety) and hydroxyl groups by solvent
molecules, forming a hydrogen-bond network, and solvent
reorientation is, possibly, also an important factor in the
The rate constant of rotamer interconversion associated with
the rotation about the dihedral angle ψ for AcTyrNH2 is higher
than the rate constant of radiative transition calculated based
on the fluorescence quantum yield and the fluorescence lifetime.
The PMF profile, expressed as a function of ψ (rotation of the
amide group), possesses three minima (at -47°, -18°, and 149°)
and two maxima (at -120° (3.30 kcal/mol) and ∼40° (5 kcal/
mol)) (see Figure 6 and Table 4). The calculated rotamer
probability distribution function reveals the presence of three
rotamers: one centered at 149°, with a population of ∼90%,
and two weakly resolved rotamers, at -47° and -18°, with
populations of 6% and 4%, respectively. The slowest rotamer
5
8
quenching. The global analysis of the fluorescence intensity
decays of tyrosinamide with blocked or free amino or/and
3-6,13,19,21,22,60,61
hydroxy groups conforms with the earlier finding,
that the amide group is a strong Tyr quencher, regardless of
the form of the amino or the hydroxy group (blocked or free).
Monoexponential fluorescence intensity decays of Tyr deriva-
tives and the lack of quenching by an amide group in acetonitrile
strongly support an important role of the hydrogen-bond network
in the Tyr fluorescence quenching process by electron transfer
from an excited phenol chromophore to an amide group. The
potential of mean force (PMF) calculations indicate that the
ground-state heterogeneous ø1 rotamer population in water,
assuming κ ) 1 in eq 9, is not responsible for the biexponential
fluorescence intensity decays of the Tyr derivatives, because
of the high estimated rate constant of rotamer interconversion,
which is higher than the radiative rate constant. However, the
exact value of κ is unknown, and a hydrogen-bond network or
other currently unknown factor can increase friction (or the
barrier) to slow the rotation, so it can affect the fluorescence.
Our results are in agreement with the findings of Lakowicz,67
that spectral relaxation rather than the rotamer seems to be a
significant, if not dominant, source of heterogeneous fluores-
cence intensity decay in single tryptophan proteins.
9
-1
interconversion rate constant is ∼10 s , i.e., much higher than
the radiative rate constant obtained for AcTyrNH2. Thus, during
the excited-state lifetime, the amide group can rotate signifi-
cantly. All presented results indicate that the entire amino acid
moiety (rotation by the dihedral angle ø1), as well as the amide
group in particular (rotation by the dihedral angle ψ), can rotate,
with the rate constant being higher than the fluorescence rate
constant. Taking into account that the ø1 rotamer populations
for all compounds studied are similar, it can be assumed that,
for AcTyr and TyrNH2, for which the PMF profile was not
studied, the rate constant of ø1 rotamer interconversion is at
least of the same order of magnitude as that for AcTyrNH2.
The exception is the rotation of the acetyl group in AcTyrNH2,
for which the highest energy barrier (8.6 kcal/mol) and the
smallest interconversion rate constant between two rotamers
6
-1
were obtained (5.2 × 10 s ; see Table 5 and Figure 7). Thus,
the acetyl group cannot freely rotate but only vibrate in the range
of ∼120°, which is associated with the presence of two
rotamers: pI ) 36% (centered at 99°) and pII ) 64% (centered
at 163.5°) (see Figure 7). However, the acetyl group exerts a
small quenching effect and only a biexponential fluorescence
intensity decay of AcTyr can be associated with hindered
rotation of this group. Thus, the heterogeneous ø1 rotamer
population in the ground state cannot be responsible for the
heterogeneous fluorescence intensity decay of tyrosinamide and
its derivatives.
Acknowledgment. This work was supported by the State
Committee for Scientific Research (KBN), under Grant No.
0369/T09/98/15.
References and Notes
(
1) Beechem, J. M.; Brand, L. Annu. ReV. Biochem. 1985, 54, 43-71.
(2) Eftink, M. R. In Protein Structure Determination; Schulter, C. H.,
Ed.; Methods of Biochemical Analysis, Vol. 35; Wiley: New York, 1991.
(
3) Ross, J. B. A.; Laws, W. R.; Rousslang, K. W.; Wyssbrod, H. R.
4
. Conclusions
In Biochemical Applications; Lakowicz, J. R., Ed.; Topics in Fluorescence
Spectroscopy, Vol. 3; Plenum Press: New York, 1992; pp 1-63.
The global analysis of the fluorescence intensity decays of
(4) Ross, J. B. A.; Laws, W. R.; Sutherland, J. C.; Buku, A.;
tyrosine (Tyr) and O-methyl tyrosine (Tyr(Me)) has revealed
that the hydroxyl group of the phenol chromophore interacts
with water molecules, resulting in very weak long-wave
fluorescence, similar to that observed for Tyr fluorescence in a
buffer solution containing a strong hydrogen acceptor. The
replacement of the amino group in tyrosine by the acetyl group
Katsoyannis, P. G.; Schwartz, I. L.; Wyssbrod, H. R. Photochem. Photobiol.
1986, 44, 365-370.
(5) Laws, W. R.; Ross, J. B. A.; Wyssbrod, H. R.; Beechem, J. M.;
Brand, L.; Sutherland, J. C. Biochemistry 1986, 25, 599-607.
(
6) Gauduchon, P.; Whal, P. Biophys. Chem. 1978, 8, 87-104.
(7) Wiczk, W.; Rzeska, A.; Łukomska, J.; Stachowiak, K.; Karolczak,
J.; Malicka, J.; Łankiewicz, L. Chem. Phys. Lett. 2001, 341, 99-106.

Photophysical Properties of Tyrosine
J. Phys. Chem. B, Vol. 108, No. 12, 2004 3889
(
8) Guzow, K.; Szabelski, M.; Rzeska, A.; Karolczak, J.; Sulowska,
H.; Wiczk, W. Chem. Phys. Lett. 2002, 362, 519-526.
9) Rayner, D. M.; Szabo, A. G. Can. J. Chem. 1978, 56, 743-745.
10) Rayner, D. M.; Szabo, A. G. J. Am. Chem. Soc. 1980, 102, 554-
63.
(39) Case, D. A.; Pearlman, D. A.; Caldwell, J. W.; Cheatham, T. E.,
III.; Ross, W. S.; Sim-Merling, C. L.; Darden, T. A.; Merz, K. M.; Stanton,
R. V.; Cheng, A. L.; Vincent, J. J.; Crowley, M.; Ferguson, D. M.; Radmer,
R. J.; Seibel, G. L.; Singh, U. C.; Weiner, P. K.; Kollman, P. A. AMBER
5; University of California: San Francisco, 1997.
(40) Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D.; Impey, R. W.;
Klein, M. L. J. Chem. Phys. 1983, 79, 926-935.
(41) Jorgensen, W. L.; Tirado-Rives, J. J. J. Am. Chem. Soc. 1988, 110,
1666-1671.
(
(
5
(
11) Engh, R. A.; Chen, L. X.-Q.; Fleming, G. R. Chem. Phys. Lett.
986, 126, 365-372.
12) Gordon, H. R.; Jarrell, H. C.; Szabo, A. G.; Willis, K. J.; Somorjai,
R. L. J. Phys. Chem. 1992, 96, 1915-1921.
13) Seidel, C.; Orth, A.; Greulich, R. O. Photochem. Photobiol. 1993,
8, 178-184.
14) Łukomska, J.; Rzeska, A.; Malicka, J.; Wiczk, W. J. Photochem.
1
(
(
(42) Kuharski, R. A.; Chandler, D.; Montgomery J. A., Jr.; Rabii, F.;
Singer, S. J. J. Phys. Chem. 1988, 92, 3261-3267.
(43) Lakowicz, J. R.; Laczko, G.; Gryczynski, I. Biochemistry 1987,
26, 82-90.
5
(
Photobiol., A: Chem. 2001, 142, 135-139.
(
(
(
15) Cowgill, R. W. Biochim. Biophys. Acta 1967, 133, 6-18.
16) Cowgill, R. W. Arch. Biochem. Biophys. 1963, 100, 36-44.
17) Tournon, J. E.; Kuntz, E.; El Bayoumi, M. A. Photochem.
(44) Harms, G. S.; Paulus, S. W.; Hendstrom, J. F.; Johnson, C. K. J.
Fluoresc. 1997, 7, 283-292.
(45) Pal, H.; Palit, D. K.; Mukherjee, D.; Mittal, J. P. J. Photochem.
Photobiol., A: Chem. 1990, 52, 391-409.
Photobiol. 1972, 16, 425-433.
(
(
18) Feitelson, J. J. Phys. Chem. 1964, 68, 391-397.
(46) Bhatnagar, G. M.; Grunen, L. C.; Maclaren, J. A. Aust. J. Chem.
1968, 21, 3005-3013.
19) Wiczk, W.; Łankiewicz, L.; Czaplewski, C.; Ołdziej, S.; Stachowiak,
K.; Michniewicz, A.; Micewicz, B.; Liwo, A. J. Fluoresc. 1997, 7, 257-
66.
20) Wiczk, W.; Łankiewicz, L.; Czaplewski, C.; Ołdziej, S.; Stachowiak,
K.; Michniewicz, A.; Liwo, A. J. Photochem. Photobiol., A: Chem. 1996,
01, 171-181.
21) Wiczk, W.; Stachowiak, K.; Czaplewski, C.; Łankiewicz, L.;
Michniewicz, A. J. Photochem. Photobiol., A: Chem. 1997, 102, 189-
95.
22) Wiczk, W.; Stachowiak, K.; Skurski, P.; Łankiewicz, L.; Mich-
niewicz, A.; R o´ j, A. J. Am. Chem. Soc. 1996, 118, 8300-8307.
23) Giancotti, V.; Quadrifoglio, F.; Cowgill, R. W.; Crane-Robinson,
C. Biochim. Biophys. Acta 1980, 624, 60-65.
24) Giancotti, V.; Fonda, M.; Crane-Robinson, C. Biophys. Chem. 1977,
, 379-383.
25) Guardia, E.; Rey, R.; Padr o´ , J. A. J. Chem. Phys. 1991, 95, 2823-
831.
26) Kumar, S.; Bouzida, D.; Swendsen, R. H.; Kollman, P. A.;
Rosenberg, J. M. J. Comput. Chem. 1992, 13, 1011-1021.
27) Kumar, S.; Rosenberg, J. M.; Bouzida, D.; Swendsen, R. H.;
Kollman, P. A. J. Comput. Chem. 1995, 16, 1339-1350.
28) Czaplewski, C.; Rodziewicz-Motowidło, S.; Liwo, A.; Ripoll, D.
R.; Wawak, J. R.; Scheraga, H. A. Protein Sci. 2000, 9, 1235-1245.
29) Masunov, A.; Lazardis, T. J. Am. Chem. Soc. 2002, 125, 1722-
730.
30) Bodansky, M.; Bodansky, A. The Practice of Peptide Synthesis;
(47) Rayner, D. M.; Krajcarski, D. T.; Szabo, A. G. Can. J. Chem. 1977,
56, 1238-1245.
2
(
(48) Shimizu, O.; Imakubo, K. Photochem. Photobiol. 1977, 26, 541-
543.
1
(49) Schnarr, M.; Helene, C. Photochem. Photobiol. 1982, 36, 91-
93.
(50) Willis, J. K.; Szabo, A. G. J. Phys. Chem. 1991, 95, 1585-
1589.
(51) Lee, J. K.; Ross, R. T.; Thampi, S.; Leurgans, S. J. Phys. Chem.
1992, 96, 9158-9162.
(52) Lee, J. K.; Ross, R. T. J. Phys. Chem. B 1998, 102, 4612-
4618.
(
1
(
(
(
(53) Harms, G. S.; Paulus, S. W.; Hendstrom, J. F.; Johnson, C. K. J.
Fluoresc. 1997, 7, 273-282.
6
(
(54) Ferreira, S. T.; Stella, L.; Gratton, E. Biophys. J. 1994, 66, 1185-
2
1196.
(
(55) Willis, K. J.; Szabo, A. G.; Krajcarski, D. T. Photochem. Photobiol.
1990, 51, 375-377.
(
(56) Creed, D. Photochem. Photobiol. 1984, 39, 563-575.
(57) Mahomoud, S. F.; Bialkowski, S. E. Appl. Spectrosc. 1995, 49,
1669-1676.
(
(58) McCurie, R.; Feldman, I. Photochem. Photobiol. 1973, 18, 119-
(
124.
1
(59) Contino, P. B.; Laws, W. R. J. Fluoresc. 1991, 1, 5-13.
(60) Schwartzwald, R.; Seidel, C.; Goody, R. S.; Kuhn, P. K.; Greulich,
K. O. Ber. Bunsen-Ges. Phys. Chem. 1989, 93, 342-346.
(61) Edelhogh, H.; Perlman, R. L.; Wilchek, M. Biochemistry 1968, 7,
3839-3900.
(62) Fraggi, M.; DeFelippis, M. R.; Klapper, M. H. J. Am. Chem. Soc.
1989, 111, 5141-5145.
(63) Cowgill, R. In Biochemical Fluorescence; Chen, R. F., Edehoch,
H., Eds.; Marcel Dekker: New York and Basel, Switzerland, 1976; Chapter
9.
(64) Szabelski, M.; Guzow, K.; Rzeska, A.; Malicka, J.; Przyborowska,
M.; Wiczk, W. J. Photochem. Photobiol., A: Chem. 2002, 152, 73-78.
(65) Hollas, J. M. Modern Spectroscopy, 2nd ed.; Wiley: Chichester,
U.K., 1992; p 180.
(
Springer-Verlag: Berlin, Heidelberg, New York, Tokyo, 1984.
(
(
31) Chen, R. F. Anal. Lett. 1967, 1, 35-42.
32) Karolczak, J.; Komar, D.; Kubicki, J.; Wr o´ z˘ owa, T.; Dobek, K.;
Ciesielska, B.; Maciejewski, A. Chem. Phys. Lett. 2001, 344, 154-164.
(
33) Knutson, R. J.; Bechem, J. M.; Brand, L. Chem. Phys. Lett. 1983,
02, 501-507.
34) Beechem, J. M.; Gratton, E.; Ameloot, M.; Knutson, R. J.; Brand,
1
(
L. In Principles; Lakowicz, J. R., Ed.; Topics in Fluorescence Spectroscopy,
Vol. 2; Plenum Press: New York, 1991; pp 241-305.
(
35) Knutson, R. J.; Waldbridge, D. G.; Brand, L. Biochemistry 1982,
2
1, 4671-4679.
(
36) Willis, K. J.; Szabo, A. G.; Drew, J.; Zuker, M.; Ridgewey, J. M.
Biophys. J. 1990, 57, 183-189.
(66) Kungl, A. J. Biophys. Chem. 1992, 45, 41-50.
(67) Lakowicz, J. R. Photochem. Photobiol. 2000, 72, 421-437.
(68) Cavanaugh, J. R. J. Am. Chem. Soc. 1970, 93, 1488-1493.
(69) Kainosho, M.; Ajisaka, K. J. Am. Chem. Soc. 1975, 97, 5630-
5631.
(
37) Torrie, G. M.; Valleau, J. P. J. Comput. Phys. 1977, 23, 187-199.
(38) Frenkel, D.; Smit, B. Understanding Molecular SimulationsFrom
Algorithms to Applications; Academic Press: San Diego, CA, 1996; Chapter
, pp 176-181.
7