Scholarly article on N,O-diphenylcarbamoyl-2-hydroxyethylamine 103791-67-9 from Chemische Berichte p. 1280

T H E C H E M I C A L R E C O R D

Ultrafast Dynamics of Myoglobin

Probed by Time-Resolved Resonance

Raman Spectroscopy

T H E

C H E M I C A L

R E C O R D

YASUHISA MIZUTANI,¹TEIZO KITAGAWA²

¹Institute for Molecular Science, Okazaki National Research Institutes, Myodaiji,

Okazaki 444-8585, Japan

²Center for Integrative Bioscience, Okazaki National Research Institutes, Myodaiji,

Okazaki 444-8585, Japan

Received 12 October 2000; accepted 20 November 2000

ABSTRACT: Recent experimental work carried out in this laboratory on the ultrafast dynamics of

myoglobin (Mb) is summarized with a stress on structural and vibrational energy relaxation. Studies

on the structural relaxation of Mb following CO photolysis revealed that the structural change of

heme itself, caused by CO photodissociation, is completed within the instrumental response time of

the time-resolved resonance Raman apparatus used (~2 ps). In contrast, changes in the intensity and

frequency of the iron-histidine (Fe-His) stretching mode upon dissociation of the trans ligand were

found to occur in the picosecond regime. The Fe-His band is absent for the CO-bound form, and its

appearance upon photodissociation was not instantaneous, in contrast with that observed in the

vibrational modes of heme, suggesting appreciable time evolution of the Fe displacement from the

heme plane. The band position of the Fe-His stretching mode changed with a time constant of

about 100 ps, indicating that tertiary structural changes of the protein occurred in a 100-ps range.

Temporal changes of the anti-Stokes Raman intensity of the ν₄and ν₇bands demonstrated immedi-

ate generation of vibrationally excited heme upon the photodissociation and decay of the excited

populations, whose time constants were 1.1 ± 0.6 and 1.9 ± 0.6 ps, respectively. In addition, the

development of the time-resolved resonance Raman apparatus and prospects in this research field

1:258–275, 2001

Key words: protein dynamics; vibrational energy relaxation; time-resolved vibrational

spectroscopy; ultrafast spectroscopy; myoglobin

Introduction

Biological Relevance of Ultrafast Protein Dynamics

Protein dynamics spans over a wide range of time scales.^1,2To

answer questions on protein dynamics, we need the concat-

enation of experimental results recorded over many orders of

magnitude of time. In this regard, it is important that a single

experimental technique can examine protein structures evolv-

ing from the earliest moments, such as the picosecond regime,

toward time scales that are highly relevant to biological func-

v Correspondence to: T. Kitagawa (E-mail: teizo@ims.ac.jp).

Contract grant sponsor: Ministry of Education, Science, Sports and

Culture, Japan; contract grant number: 12045264 (T.K.).

Contract grant sponsor: Japan Society for the Promotion of Science;

contract grant number: 11740339 (Y.M.).

The Chemical Record, Vol. 1, 258–275 (2001)

258

U l t r a f a s t D y n a m i c s o f My o g l o b i n

tions, such as the microsecond or millisecond regimes. One

dissipation occurs in the sub-picosecond to picosecond regime

in the condensed phases.^4–6Ultrafast spectroscopy is required

in order to study how the deposited energy dissipates in pro-

teins. In addition to such relaxation processes, there are con-

formational fluctuations around the equilibrium points. The

tertiary structure is undergoing fluctuations on a time scale of

hindered molecular rotations. These motions can also be stud-

ied by various ultrafast spectroscopic techniques, including

photon echo^7–9and hole-burning spectroscopy.^8,10

might ask whether ultrafast spectroscopy can contribute any-

thing to the understanding of protein systems. There are some

good reasons for us to believe that the answer is positive. One

method for studying the dynamic motions of proteins is to

pursue the ensuing relaxation after shifting a protein from equi-

librium. If we use short pulses to initiate a very rapid photore-

action at an active site immediately after the start of a reaction,

the result is a metastable structure with an active site that is

different, but the tertiary and quaternary structures of the pro-

tein are similar to the original state. The surrounding “protein

solvent” is now the structural variable. A short probe pulse fol-

lowing the initiating pulse provides direct information on the

relaxation of this metastable form into the equilibrium protein

structure. In this way, the quickest responses of proteins can be

monitored by ultrafast spectroscopy.³Another feature to be

noted about ultrafast spectroscopy is the possibility of study-

ing energy dissipation in proteins, which is very important in

order to understand reactions in proteins. In general, energy

Myoglobin Corresponds to “Hydrogen Molecule in

Chemistry”

Protein dynamics are intimately connected to the structure/

function relationship of biological systems. In numerous bio-

logical processes, the ensuing protein structural changes ac-

companying a reaction at a specific site must spatially extend

to the mesoscopic dimensions of the protein to achieve a bio-

logical function. The molecular mechanism of cooperativity in

v Yasuhisa Mizutani was born in Mie, Japan, in 1964. He earned his Bachelor of Engineering

and Master of Engineering degrees in 1987 and 1989, respectively, from Kyoto University,

working with Professor Koichiro Nakanishi. He received his PhD degree from the Graduate

University for Advanced Studies in 1992 under the supervision of Professor Teizo Kitagawa.

After that, he did postdoctoral work at the Institute for Molecular Science and at the University

of Pennsylvania, where he worked as a postdoctoral fellow for Professor Robin M. Hochstrasser.

In 1994, he joined the faculty at the Institute for Molecular Science, as a research associate. His

research interests include dynamical aspects of chemical reaction in liquids and proteins and

spectroscopic approaches to them.

r

v Teizo Kitagawa was born in Kyoto, Japan in 1940. He graduated from the engineering de-

partment of Osaka University in 1963, and received his Master of Science and PhD degrees in

1965 and 1969, respectively, from Osaka University, under the supervision of Professor Tatsuo

Miyazawa. He became a research associate of the Institute for Protein Research, Osaka Univer-

sity, in 1966, an associate professor at the Medical School of Osaka University in 1980, and a

full professor at the Institute for Molecular Science in 1983. During his research associate pe-

riod, he studied vibrational spectra of polymer crystals and related solid state properties, includ-

ing neutron inelastic scattering, and he worked as a postdoctoral fellow for Professor Bryce L.

Crawford at the University of Minnesota, studying liquid dynamics using ATR spectroscopy.

Since coming back from the USA, he has studied the dynamic structure-function relationship of

proteins using vibrational spectroscopy.

r

259

T H E C H E M I C A L R E C O R D

oxygen binding of hemoglobin (Hb) is one of the classical prob-

lems in this aspect.¹¹The large-amplitude motions at the qua-

ternary level, which form a communication link at the subunit

interface, are driven by changes in the tertiary structure upon

ligation. In this respect, myoglobin (Mb) serves as a model

system for the tertiary relaxation processes. Structurally very

similar to a subunit of Hb, Mb is involved in oxygen storage in

muscles. Twenty years after a pioneering microsecond flash

photolysis study by Gibson,¹²the first ultrafast optical spectra

of hemeproteins were reported.¹³Ultrafast optical absorption

studies^13,14revealed that the excited-state surface of the ligated

heme is dissociative, so it is possible to photoinitiate a struc-

tural change from the ligated form to the deligated form within

50 fs with quantum yield of nearly unity. By using CO as the

ligand rather than O₂, it is possible to avoid the geminate re-

combination processes complicating the reaction dynamics in

the picosecond regime.¹⁵Thus, the carbonmonoxy myoglobin

(MbCO) system is unique in that it gives access to the full

range of protein relaxation processes to a single perturbation

(ligand dissociation) without complicated side reactions and

associated population dynamics. These features of MbCO make

it possible to trigger optically the structural changes of a pro-

tein and follow the structural relaxation using optical probes

that are sensitive to different aspects of the protein’s motion.

For this reason, an enormous body of work has been accumu-

lated about Mb relaxation processes, helping to establish gen-

eral principles related to protein dynamics.

The three-dimensional structure of the deligated form of

Mb (deoxyMb)¹⁶is shown schematically in Figure 1(A), in which

the arrow points to the heme prosthetic group. The heme is an

iron-protoporphyrin IX (Fe^IIPPIX), in which the Fe^IIion is bound

to the proximal histidine (His) as the only covalent link to the

protein in Mb. The heme iron binds diatomic molecules, such

as NO, CO, and O₂, at the opposite side of the proximal His. X-

ray crystallography provided extensive information about the end-

point equilibrium structures, as schematically shown in Figure

1(B). The magnitude of atomic displacements at the heme site

that must occur on ligand binding and release is estimated by

examining the structures containing the heme with and without

bound ligands. The ligated heme has a planar structure in which

the low-spin iron atom is in the porphyrin plane. For MbCO,

the core size, that is, the distance between the pyrrole nitrogen

atoms and the heme center, is 2.005Å.¹⁶Upon ligand dissocia-

tion, the iron atom is converted from low to high spin (S=0→S=2)

and moves out of the porphyrin plane by about 0.3Å. Concomi-

tantly, the porphyrin core size expands to 2.057Å.

Fig. 1. (A) Three-dimensional structure of the deoxy form of Mb based on

the PDB structure (identification code, 1BZP).¹⁶(B) Heme structure of the

ligated form (left) and of the deoxy form (right). The doming of the heme

ring, out-of-plane displacement of the iron, and tilting of the proximal histi-

dine are key motions. These motions ultimately couple to the helical sections

of globin and induce a change in the tertiary structure.

used to examine them must be sensitive to subtle changes in the

protein structure. For this reason, we decided to use the vibra-

tional spectroscopic method, which is sufficiently structure sen-

sitive at chemical-bond resolution so as to allow identification of

any ultrafast steps in protein dynamics. Among several techniques

to detect vibrational spectra, we used Raman spectroscopy for

the following reasons. Firstly, water interferes less with Raman

spectra compared to infrared spectra. This is an important ad-

vantage for studies of protein solutions, which have water as a

Choice of the Method to Monitor Ultrafast

Protein Dynamics

Some of the faster processes in protein motions involve relatively

small alterations in the overall structure. Therefore, the probes

260

U l t r a f a s t D y n a m i c s o f My o g l o b i n

solvent. Secondly, resonance Raman effects offer the advantages

aspects of proteins, including structural and energy relaxation,

and conformational fluctuations, can be understood from the

point of view of how the protein acts as a medium for chemical

reactions at the active site (i.e., the heme site).

In this article, we summarize our recent experimental work

on the ultrafast dynamics of Mb. Our development of a time-

resolved resonance Raman apparatus is described in the next

section. Studies on the structural and vibrational energy relax-

ation of Mb following CO photolysis are also explained. A

summary and perspectives are given in the last section.

of high selectivity and increased sensitivity when the wavelength

of the probe light is tuned to an electronic transition of the in-

termediate of interest. At the same time, the selectivity can be

utilized to probe only the prosthetic groups of proteins, which

often serve as active sites. For example, we can monitor only the

structure of the heme in the protein resulting from the reso-

nance effect.¹⁷Both time-resolved resonance Raman (TR³) and

infrared spectroscopy provide inherently ultrafast time resolu-

tion and, thus, detailed structural information. In infrared spec-

troscopy, high resolutions in frequency and time can be obtained

simultaneously if the coherent interaction between an ultrashort

probing pulse and a sample is carefully considered,^18,19because

frequency selection by a spectrometer can be performed after

the light pulse interacts with the molecule. However, it is still a

hard task to measure femtosecond or picosecond time-resolved

IR spectra in a wide frequency range (300 to 3000 cm^–1) because

of interference by solvent bands and difficulties in the genera-

tion of broad infrared pulses and their multichannel detection,

although development of the transient femtosecond IR technique

is progressing.^20–24Comparatively, in TR³spectroscopy, sub-

picosecond time resolution is not compatible with a high resolu-

tion in frequency because the resolutions are achieved during

the same interaction between the light pulse and molecule. Nev-

ertheless, this technique has an advantage in that many vibra-

tional transitions in a wide frequency range can be simultaneously

observed with a single frequency of probe light. Moreover, time-

resolved anti-Stokes Raman spectroscopy is selective for

vibrationally excited modes, and it is therefore a powerful method

for measuring vibrational energy relaxation.^25–27The high rep-

etition rate of modern Ti:sapphire laser systems enables suffi-

ciently high signal-to-noise ratios, which in turn enables the

application of ultrafast TR³experiments in protein systems, as

described in the following section.

Picosecond Time-Resolved Resonance Raman

Spectrometer^30,31

Picosecond time-resolved resonance Raman (ps-TR³) spectros-

copy is a promising technique to investigate the ultrafast struc-

tural changes of molecules.³²Nevertheless, ps-TR³spectroscopy

is not implemented as widely as nanosecond TR³spectroscopy

because of practical limitations. The most crucial factor is a

lack of a light source that fulfills the requirements for small

timing jitters, appropriate repetition rates, and the wavelength

tunability of pulses applicable to ps-TR³spectroscopy, which

requires two beams: a pump beam to photoexcite molecules of

interest and a probe beam to monitor the subsequent changes

occurring in the excited or reacted molecules. To obtain the

TR³spectra of a wide variety of molecules with high S/N ratios

within a reasonable measuring time, it is quite important to

use two independently tunable light sources with a high rep-

etition rate for the pump and probe beams. Repetition rates in

the kilohertz range are desirable for practical purposes, but so

far, no one has succeeded in observing ps-TR³spectra with a

widely tunable pulse source at kilohertz repetitions. We con-

structed an apparatus consisting of widely tunable light sources

for ps-TR³spectroscopy using a 1-kHz picosecond Ti:sapphire

laser/regenerative amplifier system.³⁰

Mb as a Solution Model: Heme Solute in

Protein Solvent

Laser System

Although a heme and a protein constitute a single molecule, the

heme prosthetic group is relatively isolated from the protein and

approximates to a heme in a solution. The heme is linked co-

valently to the surrounding globin through the proximal His.

Motion along the Fe-His coordinate is orthogonal and, to a first

approximation, decoupled from the vibrational modes of the

porphyrin ring. The heme is maintained in a cavity by approxi-

mately 90 van der Waals contacts with the protein.²⁸Therefore,

the heme can be regarded as a solute molecule dissolved in a

“protein solvent.” For example, the protein structure is perturbed

by a structural change in the heme site upon CO photolysis.

The accompanied structural relaxation is analogous to the solva-

tion dynamics studied by dynamic Stokes shift of fluorescent

molecules in polar solvents.²⁹Therefore, for Mb, the dynamic

Figure 2 is a schematic of the apparatus. A picosecond mode-

locked Ti:sapphire oscillator (Spectra-Physics, Tsunami 3950),

pumped by an Ar⁺ion laser (Spectra-Physics, BeamLok 2060),

produced approximately 1.5-ps pulses with a repetition of 82

MHz and an average power of about 0.7 W. The seed pulse was

amplified by a regenerative amplifier (Positive Light, Spitfire)

operated at 1 kHz by pumping with the 527-nm output of an

intracavity frequency-doubled Nd:YLF laser. This amplifica-

tion unit provides 784-nm pulses, each with an energy of about

0.8 mJ, a duration of 2.5 ps, and a spectral width of 6 cm^–1in

a nearly TEM₀₀mode under operation at 1 kHz. The whole

laser system was covered with a plastic sheet, which was

equipped with dust cleaners containing high-efficiency particu-

261

T H E C H E M I C A L R E C O R D

Fig. 2. ps-TR³apparatus: BBO = β-barium borate, GLP = Glan laser linear polarizer, HWP = half-wave plate, L

= lens, LBO = lithium triborate, LPF = long wavelength pass filter, ND = neutral density filter, PC = personal

computer, SH = shutter. The inset shows a cross-correlation trace (dots) of the pump and probe pulses, which was

obtained using background-free sum frequency generation in a BBO crystal. The trace was best fit by a Gaussian

function (solid line) to yield a width of 2.3 ps.

late air (HEPA) filters to keep the laser system free of dust. In

that of the probe beam to minimize the effects of molecular

rotations on the observed kinetics. Both beams were always

monitored with photodiodes (Hamamatsu Photonics, S2387-

1010R) and were found to be stable within ±10%. The inset of

Figure 2 shows a cross correlation trace of the pump and probe

pulses measured with a 1-mm BBO crystal, indicating a width

of 2.3 ps. The 0.0 ps of delay time (uncertainty <0.2 ps) was

calibrated using sum frequency mixing in the same crystal.

the pump arm, a pump pulse of 540 nm was generated with a

home-built optical parametric generator (OPG) and amplifier

(OPA), which were pumped with the second harmonic of the

784-nm output. In the probe arm, a probe pulse of 442 nm

was generated as the first Stokes stimulated Raman scattering

from compressed methane gas (50 kg/cm²) excited by the sec-

ond harmonic of the 784-nm output. Components other than

the first Stokes scattering were removed spectrally with a glass

filter and dichroic mirrors and spatially with a Pellin-Broca

prism. The energy and bandwidth of the OPG-OPA output

was 30 µJ and about 3 nm, respectively. The pulses were used

for the pump beam for the ps-TR³measurements in this work

after they were attenuated to 10 µJ using a Cr-coated quartz

ND filter. The energy and bandwidth of the stimulated Raman

scattering was between 1.0 and 1.5 µJ and 14 cm^–1, respec-

tively. The first Stokes scattering was attenuated to 0.1 µJ us-

ing the Cr-coated quartz ND filter. The pump and probe beams

were made collinear and coaxial using a dichroic mirror. The

polarization of the pump beam was rotated by 55° relative to

Data Acquisition

The sample solution was placed in a 10-mmφ NMR tube and

spun with a spinning cell device that was designed to minimize

the off-center deviation during rotation.³³The sample was spun

at 3400 rpm in the spinning cell configured for 135° backscat-

tering illumination and collection. This configuration is im-

portant to minimize the effects of molecular rotations on the

observed kinetics.³⁴Spherical and cylindrical lenses were used

to focus the pump and probe beams on the sample. In ps-TR³

experiments, it is very important to use weak probe power. The

262

U l t r a f a s t D y n a m i c s o f My o g l o b i n

probe power was empirically selected so that the sample had

detector (Princeton Instruments, CCD-1100PB). Raman shifts

were calibrated with cyclohexane, benzene, or carbon tetrachlo-

ride. The peak positions of Raman bands are accurate within

±2 cm^–1.

no anti-Stokes ν₄band of the photoproduct in the probe-only

spectrum (this is discussed elsewhere). The spectral features of

photoproducts in the pump-probe spectra were also shown to

be invariant to a threefold change of the probe power. The

pump power was selected so that no saturation effect or spec-

tral changes occurred by a tripling of the pump power.

Structural Relaxation of Mb³⁵

The TR³data acquisitions were carried out as follows. In

the forward scan of delay time, delay times were changed by

increasing them. At each delay time, Raman signals were col-

lected for three 20-s exposures with both the pump and probe

beams present in the sample. This was followed by equivalent

exposures for pump-only, probe-only, and dark measurements.

Measurements in the backward scan of delay time were done

using the same procedure, except for the direction of change in

delay time. The transient Raman data were obtained by aver-

aging the data for 10 entire cycles for Stokes spectra and 90 en-

tire cycles for anti-Stokes spectra. This method enabled us to

avoid the errors caused by a slow drift of laser power and to

obtain quantitatively reproducible spectra from one day to the

next, which was possible because of the excellent long-term sta-

bility of this laser system. The pump-only spectrum was directly

subtracted from the pump-and-probe spectrum, yielding the

“probe-with-photolysis” spectrum. The dark spectrum was di-

rectly subtracted from the probe-only spectrum, yielding the

“probe-without-photolysis” spectrum (MbCO).The probe-with-

out-photolysis spectrum was subtracted from the probe-with-

photolysis spectrum to yield the photoproduct spectrum. The

subtraction parameter was determined by subtracting the probe-

without-photolysis spectrum from the probe-with-photolysis

spectrum until negative features were seen at the location of

prominent bands of MbCO.The subtraction parameter was then

reduced until these negative peaks were just eliminated, thereby

accounting for the depletion of MbCO caused by the pump

pulse. The scattering intensity for the change in the optical ab-

sorption of the sample at each time point was corrected by nor-

malizing the data to the intensity changes of the 982-cm^–1line

of sulfate ions dissolved into the sample.

Raman scattering was collected by a doublet achromat [80-

mm focal length (FL), f/2] and was imaged onto the 200-µm

entrance slit of a single spectrometer (Spex, 500M) by a dou-

blet achromat (200-mm FL, f/5). A dichroic short-pass filter

was placed between the lenses to remove the scattered pump

beam. A holographic notch filter (Kaiser Optical Systems,

HSNF-441.6-1.0) was used to reject the unshifted scattering.

A polarization scrambler was placed at the entrance slit to re-

move the effects of polarization on the spectrograph through-

put. The spectrograph is equipped with a blazed-holographic

grating (2400 grooves/mm) that enables measurements of a

spectrum as wide as about 1000 cm^–1in the Soret region, and

with a spectral slit width of approximately 8 cm^–1. The dis-

persed light was detected by a liquid nitrogen cooled CCD

Raman Spectral Change of the Heme

Figure 3 shows the Stokes time-resolved resonance Raman dif-

ference spectra of photodissociated Mb for various values of

delay times of the probe pulse with respect to the pump pulse.

In these spectra, the contribution of unreacted species has been

subtracted. The fraction of photolyzed MbCO was estimated

to be 6% based on the intensity loss of the Raman bands of

MbCO. The TR³spectrum in the 1000 to 1700-cm^–1region

for a 1-ps delay (Fig. 3) contains only the bands arising from

the in-plane vibrations of heme at 1352 (ν₄), 1560 (ν₂), and

1617 (ν₁₀) cm^–1.^36,37These bands exhibited an appreciable nar-

rowing and frequency upshift in the first few picoseconds, but

no further changes occurred after that (as is discussed in the

section “Vibrational Energy Relaxation”). The TR³spectrum

for a 10-ps delay closely resembles the spectrum of deoxyMb,

indicating that the photodissociated heme has relaxed to the

equilibrium structure within the instrument response time (~2

ps). Most interestingly, the ν₂band, which has a frequency that

is well known to be sensitive to the core size of the porphyrin

ring, appeared at a position close to that of the ν₂band of

deoxyMb. This demonstrates that expansion of the core size is

instantaneous upon CO photodissociation. There is no Raman

band that can be assigned to the excited electronic state. The

involvement of the excited electronic state in the relaxation

pathway remains a possibility, and there are reports that sug-

gest this is likely.^{38 –40}However, the results in this work and in

other studies^41,42indicate that a majority of the population is

directly channeled into the vibrational manifolds of the ground

electronic state, and minor channels may lead to an appear-

ance of small amounts of long-lived (ps time scale) intermedi-

ate states.⁴¹The intensity invariance in the spectra for delay

times between 5 and 50 ps is consistent with the fact that re-

combination of CO to the heme takes place in the time regime

of micro- to milliseconds,⁴³in contrast with the cases for NO

and O_2.¹⁵Accordingly, these results show that the heme in Mb

is led to the ground electronic state and expanded to the equi-

librium core size within the instrument response time.

Figure 4 shows the TR³spectrum in the 150 to 850-cm^–1

region and delay of between –7 and 1000 ps. The bands at

220, 301, 369, and 671 cm^–1in the spectrum for a 1000-ps

delay are assigned to the vibrations of the heme: ν (Fe-His),⁴⁴

γ_7,⁴⁵δ(C_βC_cC_d),⁴⁵and ν₇, respectively. The large peak at about

220 cm^–1is caused by the stretching mode of the covalent bond

between the heme iron and the N_εof His93, ν (Fe-His). The

263

T H E C H E M I C A L R E C O R D

Fig. 4. TR³Stokes spectra of photodissociated MbCO in the 150 to 850-

cm^–1region. The Stokes spectra of the equilibrium deoxyMb and MbCO are

depicted at the bottom for comparison. Experimental conditions are the same

as those of Figure 3.

Fig. 3. TR³Stokes spectra of photodissociated MbCO in the 1100 to 1700-

cm^–1region. Stokes spectra of the equilibrium deoxyMb and MbCO are de-

picted at the bottom for comparison. Horse skeletal Mb (Sigma, M-0630) was

dissolved into a buffer (50 mM Tris·HCl, pH 8.0) and filtrated through a 0.45-

µm PTFE membrane (Millipore) to make a 300-µM solution. One milliliter of

the Mb solution was placed into an airtight NMR tube (10 mmφ). The re-

maining oxygen in the solution was removed by more than five cycles of degas-

sing and back-filling with CO. Finally the solution was equilibrated under 1

atm of CO. The samples were reduced with a fivefold stoichiometric amount of

sodium dithionite in a small amount of solution (approximately 50 µL). The

sample solution in the cell was continuously spun during the measurements to

prevent multiple probing of the same portion of sample. The sample was re-

placed with a new one every 3 h. Sample integrity was confirmed with ultravio-

let (UV)-visible absorption spectra after the TR³measurements. The samples of

deoxyMb were prepared by the same procedure as MbCO, except for back-

filling with N₂and the equilibration of the sample under 1 atm of N₂.

the first few picoseconds, but no further large changes occurred

after that.

Structural Change of the Fe-His Linkage

The iron-proximal histidine bond is the only linkage between

the heme and globin, and it thus functions as the focal point

for the forces driving the structural changes. In this regard,

TR³studies of the proximal His are key to elucidating the pro-

tein dynamics. The ν(Fe-His) mode is one of the most focused

Raman bands in Mb and Hb, and it is known to serve as a

good measure of the tertiary and quaternary structures of

globin.^46,47Recently, work on the Mb and Hb problem with

the use of resonance Raman spectroscopy has made significant

progress. The assignment of the Fe-His mode at around 216–

220 cm^–1by Kitagawa and his coworkers^44,56led to a variety of

investigations involving R →T conformational changes, as well

as histidine deprotonation, cryogenic, and comparative stud-

ies. To study protein relaxation, we closely examined the tem-

peak at 301 cm^–1is γ₇(out-of-plane methine wagging). The

peak at 369 cm^–1is assigned to δ (C_βC_cC_d), a mode involving

deformation of the propionate methylene groups. The peak at

671 cm^–1is ν₇(breathing-like mode of the porphyrin inner

ring). As was observed for the spectra in the high- frequency

region, RR bands exhibited appreciable frequency upshifts in

264

U l t r a f a s t D y n a m i c s o f My o g l o b i n

poral behavior of the ν(Fe-His) band in comparison with those

of other heme Raman bands.

The temporal behaviors of the Raman band intensities

observed for Mb are plotted in Figures 5 (ν₇and ν₄bands) and

6 [γ₇and ν(Fe-His) bands]. Results from an investigation of

the temporal behavior demonstrate that the band intensities of

heme in-plane (ν₇and ν₄bands) and out-of-plane vibrations

(γ₇band) develop within the instrument response time and

remain constant. In contrast, the band intensity of ν(Fe-His)

shows a gradual increase up to a 20-ps delay, in addition to an

instantaneous rise upon photodissociation. The fraction of the

slow rise component is 8 ± 1%. Because there is no intensity

Fig. 6. Temporal change of the Raman intensity of the Stokes ν(Fe-His)

(circles) and γ₇bands (triangles) of photodissociated MbCO. The solid line

for the ν(Fe-His) mode is a fit using an exponential function of the form

A[1–B exp(–t/τ_rise)] (a rise containing instantaneous and slow phases and very

slow decay) convoluted with the instrument response function, as shown in

the caption of Figure 5. The line for the ν(Fe-His) mode shown in this figure

was obtained with the parameters of τ_rise= 6.5 ± 2.5 ps and B = 0.08 ± 0.01.

The solid line for the γ₇mode is a fit using a step function convoluted with

the instrument response function. The lower panel shows a closeup of the

curve in the early time region.

change in the heme in-plane and out-of-plane vibrational

modes in a time range of 3 to 50 ps, the gradual intensity

increase in the ν(Fe-His) band is ascribed to the structural

changes in the Fe-His linkage. The same measurements were

carried out for a model system without a protein matrix, which

is a 2-methylimidazole complex of iron protoporphyrin-IX,

(Fe^IIPPIX)-2MeIm, in a 0.1% CTAB (hexadecyltrimethylam-

monium bromide) aqueous solution. The results are shown

in Figure 7. Temporal behavior similar to that for the corre-

Fig. 5. Temporal change of the Raman intensity of the Stokes ν₄(squares)

and ν₇bands (diamonds) of photodissociated MbCO. The lines are calcu-

lated for a step function (assuming an instantaneous rise and very slow decay)

convoluted with a Gaussian instrument response function for which the cross

correlation time is 2.3 ps. The lower panel shows a closeup of the curve in the

early time region.

265

T H E C H E M I C A L R E C O R D

To understand the intensity behavior of the Fe-His stretch-

ing mode, we followed the analysis of Bangcharoenpaurpong

et al.⁴⁸The Fe-His configuration is defined by: the out-of-plane

displacement caused by ligand photodissociation (defined as

the distance from the average plane of the porphyrin ring); the

tilt angle of the Fe-His bond (relative to heme normal); and

the azimuthal angle (about heme normal). Bangcharoenpaur-

pong et al. found that the Fe-His stretching vibration is rela-

tively strongly coupled to the Soret transition.⁴⁸Their hypothesis

is that the origin of the RR intensity of the Fe-His vibrational

mode results from the electronic overlap between the σ^*or-

bital of the Fe-His bond (σ^*_Fe-NHis) and the π^*orbital of the

porphyrin ring (π^*_por) created by these structural changes. For

nearly zero degree of the azimuthal angle, the orbital mixing of

σ^*_Fe-NHis-π^*_pormay occur. The electron-electron repulsive inter-

action between the single electron in σ^*_Fe-NHisand the filled π_por

electron density on the heme nitrogens can lead to σ^*_Fe-NHis→π^*_por

electron donation. The π_por-π^*_porexcitation can result in direct

harmonic coupling of the Fe-N_Hismode via the orbital mixing

of π^*_porand σ^*

(the π_por-π^*_portransition gives rise to the

Fe-NHis

Soret band). The stronger this coupling, the higher the reso-

nance Raman intensity of the Fe-His stretching mode upon

Raman excitation into the Soret band. The band intensity of

the ν(Fe-His) mode increases with increasing out-of-plane dis-

placement of iron because of the creation of an overlap be-

tween the σ^*_Fe-NHisand π^*_por(which is zero by symmetry for the

in-plane iron). Just as the azimuthal rotation affects the cou-

pling, the tilt angle also affects it: the greater the tilt angle, the

stronger the coupling. No (or only weak) coupling to the Soret

transition occurs when the tilt angle is zero.

We now consider the origin of the slow phase of the ν(Fe-

His) intensity change that was observed both for Mb and the

model compound. A high-resolution X-ray study¹⁶indicated

that the iron out-of-plane displacement is about 0.3Å in the

deoxy structure and decreases to around 0Å upon CO bind-

ing. The difference in the tilt angle is slightly altered by CO

binding: 10° in the deoxy form⁴⁹and 0° in the CO-bound

form.⁵⁰However, the azimuthal angle remains unaltered at 19°

for both the deoxy and CO-bound forms.^49,50Accordingly,

among the three factors mentioned above, the iron out-of-plane

displacement and the tilt angle are possible structural param-

eters from which the slow phase of the ν(Fe-His) intensity

change originated.

Fig. 7. Temporal change of the Raman intensity of the Stokes ν(Fe-Im)

(circles) and γ₇bands (triangles) of the photoproduct of the CO-bound

(Fe^IIPPIX)-2MeIm complex. The solid line for the ν(Fe-Im) mode is a fit

using an exponential function of the form A[1–B exp(–t/τ_rise)] (a rise contain-

ing instantaneous and slow phases and very slow decay) convoluted with the

instrument response function, as shown in the caption of Figure 5. The line

for the ν(Fe-Im) mode shown in this figure was obtained with the parameters

of τ_rise= 7.8 ± 2.7 ps and B = 0.11 ± 0.02. The solid line for the γ₇mode is a

fit using a step function convoluted with the instrument response function.

Hemin (Sigma, H-2250) was dissolved into buffer (50 mM Tris·HCl, pH

8.0) containing 0.1% CTAB and 15 mM 2-methylimidazole, and then fil-

trated through a 0.45-µm PTFE membrane (Millipore) to make a 300-µM

solution. The lower panel shows a closeup of the curve in the early time re-

gion. The procedure for the preparation of the CO-bound form of the com-

plex was the same as that for MbCO.

A molecular dynamics (MD) simulation by Kuczera et al.⁵¹

pursued the time-dependent out-of-plane displacement of iron

after ligand dissociation. The trajectory indicated that 80% of

the out-of-plane displacement occurred within 50 fs, corre-

sponding to the motion of the iron along the repulsive part of

the iron-porphyrin potential energy surface. The results also

predicted that 20% of the out-of-plane displacement of iron

occurs over time scales ranging from a picosecond to tens of

picoseconds. The bandshift of band III in the near IR region

sponding modes of Mb were observed for the ν(Fe-Im) and

the heme vibrational bands of the complex, indicating that

the structural change is not associated with protein relaxation

and that it is a characteristic of the heme-histidine (or imida-

zole) unit.

266

U l t r a f a s t D y n a m i c s o f My o g l o b i n

observed in the ultrafast transient absorption spectra of Mb at

an ambient temperature was interpreted successfully along this

line.⁵²A comparison of the band III shift with the trajectory of

the MD simulations by Kuczera et al.⁵¹indicated that the tem-

poral behaviors correlate with one another over the first 100 ps

of relaxation. The temporal behaviors of ν(Fe-His) intensity

resemble those of the position of band III and the out-of-plane

displacement of Fe in the MD simulation. According to this,

we concluded that the slow rise of ν(Fe-His) intensity prima-

rily originates from the slow phase of the iron out-of-plane

displacement. However, there are contrasting MD simula-

tions^53,54that indicate that complete heme doming occurs on a

time scale of less than 1 picosecond. These results suggest that

the slow rise of ν(Fe-His) Raman intensity must result from a

change in the tilt angle. At the present stage, it is difficult to

attribute the slow rise of ν(Fe-His) intensity exclusively to ei-

ther structural parameter. In either case, the structural change

responsible for the slow rise does not involve global conforma-

tional changes of the protein, as was similarly observed for Mb

and the (Fe^IIPPIX)-2MeIm complex.

The peak positions of the observed five bands of Mb and

the (Fe^IIPPIX)-2MeIm complex are plotted against time in Fig-

ure 8. For Mb, the ν(Fe-His) band showed a small downshift

in the 100-ps time range, but the other two bands showed no

shift in this range [Fig. 8(A,B)]. In contrast, neither the ν(Fe-

Im) nor the γ₇band of the (Fe^IIPPIX)-2MeIm complex showed

a frequency shift as the delay time increased [Fig. 8(C)]. There-

fore, the downshift of the ν(Fe-His) band of Mb should be

attributed to some change in the ligation state of the proximal

His caused by a structural change of globin. The downshift of

the ν(Fe-His) band of Mb can be well described by single-ex-

ponential kinetics, providing a time constant of around 100

ps, although the kinetics would involve several steps of relax-

ation. The protein relaxation of Mb upon CO dissociation was

previously thought to complete within 30 ps because previous

TR³measurements with a 30-ps wide pulse could not detect

any frequency difference between the initial 30-ps spectrum

and the equilibrium spectra.⁵⁵However, this high-precision

study revealed the presence of the relaxation process of Mb in

the 100-ps time range.

Fig. 8. Temporal changes of the Raman frequencies of some Stokes Raman

bands in the low-frequency region. (A) The time dependence of the position

of the Stokes ν(Fe-His) (circles) and γ₇(triangles) bands of photodissociated

MbCO. The temporal shift of the ν(Fe-His) band was fitted using a single

exponential function, yielding time constants of 106 ± 14 ps (solid line). (B)

The time dependence of the position of the Stokes δ(C_βC_cC_d) band (dia-

monds) of photodissociated MbCO. (C) The time dependence of the posi-

tion of the Stokes ν(Fe-Im) (circles) and γ₇(triangles) bands of the

photoproduct of the CO-bound (Fe^IIPPIX)-2MeIm complex. The band posi-

tions were obtained as the first moment of the observed bands. The broken

lines show the band positions of the equilibrium deligated form (i.e., the

deoxy form).

Previous Raman studies on equilibrium deoxyHbs^56–58and

photoinduced ligand-free transient Hbs^59–63have demonstrated

that the frequency of the Fe-His stretching mode is responsive

to a quaternary structure. In addition, studies on the transients^59–

with the equilibrium species, each of the T and R states has a

frequency range: 216 to 222 cm^–1and 222 to 232 cm^–1for the

photodissociated T and R preparations (≤10 ns), respectively.

For Hb, the difference in the ν(Fe-His) frequencies of the T

and R forms was attributed to a difference in the azimuthal

angle.⁴⁸Recently, Peterson et al. measured the ν(Fe-His) fre-

quencies of several mutants of sperm whale Mb and examined

a correlation between the ν(Fe-His) frequency and heme proxi-

mal-His geometry.⁶⁴They attributed changes in the ν(Fe-His)

frequency to changes in the azimuthal angle of the His imida-

zole ring driven by a change in steric factors. However, a

61

have revealed that this frequency is sensitive to ligand-in-

duced changes in the tertiary structure within both the R and

T quaternary states. The general trend is that the frequency of

this mode increases in going from the T to the R states, and

that the changes within a given quaternary structure caused by

ligand binding push the frequency still higher. Picosecond stud-

ies show that the frequency in a nanosecond time scale is al-

ready fully developed within 25 ps following photolysis.⁵⁵As

267

T H E C H E M I C A L R E C O R D

deoxyMb/MbCO comparison of the crystallographic data in-

shift [Fig. 8(B)], although the δ(C_βC_cC_d) band for the CO-

bound form appears at a position 9 cm^–1higher than that for

the deoxy form. This result suggests that the change in a hy-

drogen bond of the propionates upon CO dissociation is small

and/or takes place within the instrument response time.

dicated that the azimuthal angle does not change in Mb as

described above.^49,50If the frequency shift originated from a

change in the iron out-of-plane displacement and/or the histi-

dine tilt angle, the shift would accompany an appreciable in-

tensity change in the ν(Fe-His) RR band. Figure 6 shows,

however, that there is no change in ν(Fe-His) intensity in the

100-ps time range. Therefore, it is unlikely that the observed

frequency shift of the ν(Fe-His) mode is caused by a change of

these structural parameters.

Structural Relaxations Probed with

Different Techniques

Several experiments have been performed to study the protein

response upon CO dissociation, and their results indicated that

the protein motion of Mb spans on different time scales. The

differences among all of the measurements may be due to the

differences in what was probed. Miller and his coworkers found,

with femtosecond heterodyne-detected four-wave-mixing mea-

surements, that protein effectively changed its global shape within

500 fs.⁶⁷The observed fast response is consistent with a collec-

tive motion mechanism, in which a protein vibration along the

reaction coordinate is directly excited by photodissociation of

CO. Picosecond time-resolved IR (TRIR) spectroscopy on the

amide I band of MbCO indicated that the conformation change

of a polypeptide skeleton occurs in a 6 to 8-ps time scale.⁶⁸There-

fore, changes in the tertiary structure probed by transient grat-

ing or picosecond TRIR spectroscopy do not affect the

ν(Fe-His) mode. In contrast, Xie and Simon reported that re-

laxation of the transient circular dichroism (CD) signal of the

N band (probed at 355 nm) of the photodissociated MbCO to

the equilibrium deoxy spectrum requires 300 ps.⁶⁹They attrib-

uted the CD change to a change in the orientations of the

transition dipole moments of the heme relative to those of the

aromatic side chains located closely to the heme. The observed

temporal behavior of the ν(Fe-His) frequency resembles that

of the transient CD signal. The transient absorption measure-

ments of band III showed a highly nonexponential relaxation

to its equilibrium position with time constants of 3.5, 83, and

3300 ps.⁷⁰The time constant of the second phase is close to

that of the ν(Fe-His) downshift. The latter two experiments

probed relatively local changes of protein, that is, changes in

the proximal heme pocket. Therefore, there may be relatively

slow structural rearrangements around the heme, including a

change in hydrogen bonding and the orientational changes of

aromatic side chains in the proximal heme pocket.

Changes in the basicity of an imidazole ring of His residue

are known to change the frequency of the ν(Fe-His) mode.⁴⁶

An extreme example is found in cytochrome c peroxidase

(CCP),⁶⁵in which the carboxyl group of Asp235 is hydrogen-

bonded to the proximal His at N_δH. The ν(Fe-His) band in

CCP contains two components: one at 233 cm^–1, assigned to a

conformation with a very strong hydrogen bond to the N_δH

on the imidazole ring, and the other at 246 cm^–1, assigned to a

conformation in which a complete proton transfer from the

imidazole to the Asp carboxylate group has occurred. The

changes in the iron out-of-plane displacement and/or the his-

tidine tilt and azimuthal angles result in a change in the cou-

pling of the Fe-N_Hismode to the π_por-π^*_portransition via the

orbital mixing of π^*_porwith σ^*_Fe-NHis, thereby appreciably chang-

ing ν(Fe-His) intensity. However, the change in the hydrogen

bond gives rise to a change in the extent of the σ^*_Fe-NHis→π^*_por

electron donation, and would hardly affect the orbital mixing

of π^*_porwith σ^*_Fe-NHis. Therefore, the change in the hydrogen

bond does not necessarily accompany a change in the ν(Fe-

His) Raman intensity. In fact, the N_δproton of His93 is hy-

drogen-bonded to the backbone carbonyl of Leu89 and also to

the O_γatom of Ser92. Accordingly, the most probable origin

of the ν(Fe-His) downshift of the photodissociated Mb is a

change in the hydrogen bond of the proximal His to its sur-

rounding residues.

Structural Changes of the Heme Pocket

The carboxylate group of the heme-7-propionate is hydrogen-

bonded to the His97 imidazole ring and the hydroxyl group of

Ser92. The frequency of the δ(C_βC_cC_d) band correlates with

the status of the hydrogen bond to the propionates in the proxi-

mal heme pocket. If the propionates are a part of hydrogen

bonds with other residues in the pocket, as in Mb and dimeric

Hb from the blood clam Scaphrca inaequivalvis,⁶⁶the δ(C_βC_cC_d)

mode tends to appear at higher frequencies; but if the propi-

onates are solvent-exposed, as in human Hb, this frequency

decreases. In addition, the frequency in a given protein increases

as the number of hydrogen bonds to the propionates increases,

although there is no fixed relationship between a given fre-

quency and the number of hydrogen bonds. Unlike the ν(Fe-

His) band, the δ(C_βC_cC_d) band did not exhibit any frequency

Vibrational Energy Relaxation⁷¹

Anti-Stokes Intensities of the Heme Modes

Anti-Stokes Raman intensity is expected to be the most direct

probe of the extent and dissipation rate of excess vibrational

energy. Figure 9 shows an example of the anti-Stokes TR³spec-

trum of the photodissociated Mb. Spectra A, B, and C are a

probe-with-photolysis spectrum at a 1-ps delay, probe-with-

268

U l t r a f a s t D y n a m i c s o f My o g l o b i n

Fig. 9. Anti-Stokes spectra obtained in measurements of an MbCO sample.

Spectra A, B, and C represent “probe-with-photolysis” spectrum at a 1-ps

delay, “probe-without-photolysis” spectrum, and difference spectrum between

A and B (see text for the subtraction procedure), respectively. Note that there

is no Raman intensity at the ν₄frequency of photodissociated Mb in spec-

trum B. Also note that there is no Raman intensity for the iron-CO stretch-

ing mode [ν(Fe-CO)] in spectrum C. The weak band around 490 cm^–1in C

is due to γ₂₂.

out-photolysis spectrum, and the difference spectrum between

A and B, respectively. Note that the ν₄band in spectrum B

gives the contribution from only the CO-bound form and no

intensity at the frequency of photodissociated species. This in-

dicates that the probe pulse is weak enough to prevent mul-

tiple interactions (i.e., absorption and Raman scattering) by a

single heme over the duration of the probe pulse. Because the

photon flux becomes high for picosecond pulses in a given av-

erage power, it is very important to use weak probe power. With

a high photon flux, the ν₄band appeared for only the photo-

dissociated form in a probe – without – photolysis spectrum,

indicating that there is a significant probability for the heme to

be photoexcited and provide Raman scattering within a single

probe pulse. Because the quantum efficiency of the CO photo-

dissociation is near unity, the complete absence of the Raman

band belonging to the photodissociated form in the spectrum

without photolysis corroborates the adoption of a proper ex-

perimental condition for the probe power. The anti-Stokes TR³

difference spectra between –5 and 50 ps are shown in Figure

10. Anti-Stokes intensities were highest at a 1 ps delay, when

the ν₃, ν₄, ν₅, ν₆, and ν₇bands were observed at 1469, 1353,

1118, 783, and 669 cm^–1, respectively. These bands lost inten-

sity as the delay time increased, and reached their equilibrium

intensities at a 10 ps delay.

Fig. 10. TR³anti-Stokes spectra of photodissociated MbCO in the 450 to

1600-cm^–1region.

Figure 11 shows the temporal changes of anti-Stokes ν₄

and ν₇band intensities. The anti-Stokes intensity develops

within the instrument response time. This is consistent with

the photodissociation of CO from the heme, which takes place

within 50 fs.¹⁴The instrument response was deconvoluted from

the decay of the anti-Stokes intensity using a Gaussian fit to

the cross correlation signal. This analysis obtained the decay

constants of 1.1 ± 0.6 ps for the ν₄band and 1.9 ± 0.6 ps for

the ν₇band. Because there is no intensity change in the Stokes

ν₄and ν₇bands in the 3 to 50-ps time range, the observed

intensity decay in the anti-Stokes ν₄and ν₇bands can be as-

cribed to vibrational energy relaxation.

Figure 12 delineates the detailed spectral change of the

Stokes ν₄band after photodissociation. At a delay of 1 ps, the

center of the ν₄band is located at 1352 cm^–1, and the band is

broadened with a full width at half maximum (FWHM) of

19.7 ± 0.6 cm^–1. The band shifts to 1355 cm^–1and becomes

269

T H E C H E M I C A L R E C O R D

Fig. 11. Temporal changes of the Raman intensity of the anti-Stokes ν₄(A)

and ν₇bands (B) of photodissociated MbCO. The solid lines are fit using an

exponential function of the form A[exp(–t/τ_decay) + B] (an instantaneous rise

and an exponential decay) convoluted with an instrument function as shown

in the caption of Figure 5. The lines shown in this figure were obtained with

the parameters of τ_decay= 1.1 ± 0.6 ps and B = 0.03 ± 0.01 for the ν₄band, and

τ_decay= 1.9 ± 0.6 ps and B = 0.31 ± 0.02 for the ν₇band.

Fig. 12. Closeup of TR³Stokes spectra in the ν₄region.

narrower (FWHM = 17.4 ± 0.6 cm^–1) at a 50-ps delay. Similar

trends are observed for the ν₇band, which is centered at 669

cm^–1and is broadened (FWHM = 22.7 ± 0.7 cm^–1) for a 1-ps

delay, but is shifted to 671 cm^–1and becomes narrower (FWHM

= 19.7 ± 0.7 cm^–1) for a 50-ps delay. Recently, similar features

were also observed for nickel octaethylporphyrin in benzene.⁷²

energy of between 50 and 100 kJ/mol immediately after pho-

todissociation.

Pioneering work on the energy flow from the heme into

the protein matrix was reported by Henry et al.,²⁸who carried

out molecular dynamics simulations for Mb and cytochrome c

with initial energy deposits corresponding to optical excitation

at both 532 and 355 nm. The simulation predicted a fast ini-

tial relaxation phase with decay constants of between 1 and 4

ps, followed by a slower relaxation phase with decay constants

of between 20 and 40 ps. To compare those predictions with

the results in this work, it is necessary to estimate the tempera-

ture decay rate of the heme. The anti-Stokes/Stokes Raman

intensity ratio is determined by the Boltzmann factor for the

vibrational mode in question, assuming a statistical distribu-

tion of vibrational energy. When the observed kinetics of popu-

lation decay in vibrationally excited states are converted to

temperatures by assuming a Boltzmann distribution, the time

constants of 1.1 ± 0.6 (ν₄) and 1.9 ± 0.6 ps (ν₇) for the popu-

lation decay give rise to time constants of about 2 ps for the

temperature decay. This value appears to be close to the value

of the fast component predicted by Henry et al.²⁸However, the

molecular dynamics simulations suggested a ratio of 50:50 for

Heme Cooling

A 540-nm photon excites the heme into the π_por-π^*_porstate [¹Q(1,

0)], thereby depositing 221 kJ/mol of energy. The π_por-π^*_por

-

excited heme is believed to relax to a level having the

antibonding character of the Fe-CO bond, although its energy

level is presently unclear. The absorbed photon energy is sig-

nificantly higher than the energy required to break the Fe-CO

bond, and therefore, excess energy is deposited on the photo-

lyzed heme. The enthalpy difference between Mb + CO (gas)

and MbCO is approximately 90 kJ/mol,⁷³so the heme and

CO share approximately 130 kJ/mol of excess energy. Parti-

tioning of this excess energy depends on its mechanism. If the

⁵T₂or ³T₁state of Fe^IIwere involved, as was suggested for he-

moglobin-CO (HbCO),¹³the heme would possess an excess

270

U l t r a f a s t D y n a m i c s o f My o g l o b i n

the magnitudes of the fast and the slow components, whereas

than that of water heating (7.5 ps). This time lag may corre-

spond to the time required for energy propagation via the col-

lective low-frequency modes of the protein. Very recently,

Sagnella et al. proposed that a possible doorway for energy re-

lease from the vibrationally excited heme involves the interac-

tion of its propionate groups with neighboring solvent

molecules.⁷⁹

we did not observe such a large contribution of the slow com-

ponent in Figure 10. It is noted that the simulation was con-

ducted in vacuo and neglected the effect of the water bath, which

works as an extensive thermal sink. In fact, fast heating of wa-

ter with a time constant of 7.5 ± 2.0 ps was observed for Mb in

D₂O.⁷⁴Genberg et al. studied the vibrational energy relaxation

of optically excited hemeproteins by using a transient phase

grating technique, and they found that vibrational energy was

transferred from the heme to the water interface through the

protein matrix in less than 20 ps.⁷⁵A molecular dynamics simu-

lation with the presence of water molecules indicated that the

inclusion of water increases the rate of cooling to a 10-ps time

scale.⁷⁶Disregarding the effect of the water bath may deceler-

ate the cooling rate of heme, resulting in a larger contribution

by the slow component.

Vibrational Relaxation of the CO Stretching Mode

One-⁸⁰and two-color⁸¹infrared pump-infrared probe spectros-

copy has been used to determine the vibrational relaxation time

(T₁) for the CO stretching vibrations in the unphotolyzed state

of MbCO in D₂O at 300 K. Both measurements obtained simi-

lar results of 17 ps for T₁time. The CO T₁time can be affected

by the protein structure⁸⁰: two different conformers of MbCO

have significantly different T₁times. It is interesting to com-

pare the time constants of the vibrational energy relaxations of

the heme modes with that of the internal mode of CO bound

to the heme. The T₁of the CO stretch is much longer than the

time constants observed for heme cooling in Mb, indicating

that there is not efficient intramolecular energy exchange be-

tween the bound CO and the porphyrin ring. Recent studies

demonstrated that the inter- and intramolecular processes of vi-

brational relaxation must be considered to be in competition

even for large molecules such as trans-stilbene^82–87and

metalloporphyrins^72,88–90. One reason for the less efficient energy

exchange is the relatively low frequencies of the other FeCO

modes (577 and 512 cm^–1for the FeCO bend and FeC stretch

of MbCO, respectively). These frequencies imply that at least

three quanta are required to come within a few hundred

wavenumbers of the energy of the CO stretching mode. A path-

way involving the transfer of the CO vibrational energy into

other FeCO modes therefore requires large anharmonic coupling.

There are studies on the vibrational energy relaxation of

CO in the photolyzed state, where CO is not covalently at-

tached but is trapped within a docking site. The structure and

dynamics of the photolyzed state have recently been probed

using X-ray diffraction,^91–94vibrational spectroscopy,^95,96and

computer simulation.^97–99Photolysis of MbCO produces a sig-

nificant population of CO in its excited vibrational state. Based

on time-resolved mid-IR absorbance measurements, T₁time

of 600 ± 150 ps was determined.¹⁰⁰Because the photodissoci-

ated CO has no covalent interactions, its T₁time is signifi-

cantly longer than that observed in the bound state.

Measurement of this relaxation time will clarify the role of the

“protein solvent” in the relaxation of CO. The Landau-Teller

theory predicted that the relaxation rate of 1/T₁is expressed

remarkably well by the Fourier component of time-dependent

friction evaluated at the frequency of the oscillator,¹⁰¹which is

connected to the time correlation function of the fluctuating

force along the vibrational coordinate by the second fluctua-

Hopkins and his coworkers measured the TR³spectra of

deoxyHb with a time resolution of 8 ps.⁷⁷Intensity changes of

both negative transients in Stokes and positive transients in

anti-Stokes showed population changes with a time constant

of between 2 and 5 ps, to which the 1 to 2-ps decay of Figure

10 obtained with a higher time resolution is comparable, de-

spite differences in the globin (Mb vs. Hb) and phenomena

(CO-photodissociation vs. electronic excitation). Petrich et al.

studied the vibrational relaxation of hemeproteins indirectly

using femtosecond Stokes resonance Raman spectroscopy, and

they estimated the temperature of heme in HbCO based on

the band shift of the ν₄mode.⁷⁸They concluded that the heme

in the HbCO photoproduct was substantially cooled within

10 ps. Li et al. qualitatively analyzed Stokes and anti-Stokes

Raman scattering of different hemeproteins to determine the

heme vibrational temperature as a function of nanosecond la-

ser flux. They obtained a time constant of about 4 ps for heme

cooling.⁴¹These experimental results on the time scale of vi-

brational energy relaxation are in agreement with the results in

this work.

The fast energy dissipation from the globin to the water

bath has also been noted in femtosecond TRIR studies that

monitor the heating of water caused by the photoexcitation of

deoxyMb.⁷⁴The observed kinetics was fitted with a model hav-

ing two time constants. The fast component was best fitted by

a Gaussian rise function with a time constant of 7.5 ± 1.5 ps,

and the slow component was described by a time constant of

about 20 ps with 40% of the total amplitude. A comparison of

the heme cooling (TR³) and water heating (TRIR) studies sug-

gests that there are two channels of energy dissipation from the

protein to the water bath. One is a classical diffusion process,

and it is responsible for the slow component; the observed time

constant is in reasonable agreement with that calculated using

the classical diffusion theory. The other is through the collec-

tive motions of the protein,⁷⁴and it is responsible for the fast

component. The time constant of heme cooling (3 ps) is shorter

271

T H E C H E M I C A L R E C O R D

tion-dissipation theorem. Sagnella et al. computed the influ-

Summary and Prospects

ence spectrum, which is the distribution of the square of the

coupling constants between the CO stretch and the bath modes

as a function of frequency.¹⁰⁰In their simulations, the relax-

ation was dominated by short-ranged van der Waals interac-

tions, with the residues forming the inner wall of the heme

pocket of Mb. The data suggest that the mechanism for the

vibrational relaxation of CO in Mb takes place through suc-

cessive, noncorrelated collisions with the pocket residues, act-

ing as a first solvation shell to the CO molecule.

Mb looks like a “hydrogen molecule” for physicochemical stud-

ies on protein dynamics. Many kinds of spectroscopic tech-

niques, including ultrafast spectroscopy, have been applied to

study the dynamic properties of this protein, as described above.

It became apparent from these studies on the heme that the

structural change of the heme ring is completed within 1 ps

after CO photodissociation. At the same time, large amounts

of excess vibrational energy are selectively deposited at the heme

site. As described in the section on “Vibrational Energy Relax-

ation,” the mechanism of vibrational energy transfer within

proteins has not been extensively studied thus far because of a

lack of an appropriate method. Raman bands of some aromatic

side chains, such as tryptophan, tyrosine, and phenylalanin

residues, are resonantly enhanced with UV probe light.¹⁰⁸The

anti-Stokes scattering of these residues serve as a good probe of

the extent of the vibrational excess energy at their locations in

proteins. If such measurements are accomplished, the vibra-

tional energy transfer from the heme to the surrounding pro-

tein, and from the protein to water, can be followed using

ultrafast spectroscopic methods to provide spatial mapping of

the vibrational energy relaxation pathways. Detailed structural

information on these systems facilitates the construction of

conceptual models on the vibrational energy exchange and also

a detailed theoretical modeling of the process. Thus,

hemeproteins provide ideal model systems for understanding

vibrational energy relaxation not only in protein but also in

general condensed phases.

The ultrafast structural change at the heme site perturbs

the protein structure in a step-function manner. Therefore,

photodissociation of CO in Mb provides a good opportunity

to study the protein response function for the reaction at the

heme. In this study, the protein response was observed as the

frequency shift of the ν(Fe-His) mode with the 100-ps time

constant. Although the response function is expected to be far

from simple, one can consider two limiting cases with respect

to the length scale and dynamics of the protein response func-

tion for the reaction at the heme site. On a long time scale for

the motions, atomic displacements should be considered dif-

fusive in nature. The protein motion between local minima in

its structure is averaged over a certain time interval, and the

net atomic displacements along the reaction coordinate appear

as a thermally activated Brownian process. The length of any

given relaxation component in this particular scenario can be

on the order of a single amino acid residue (i.e., localized mo-

tions). An interconversion between the so-called “open” and

“closed” forms is a typical example.¹⁰⁹A large number of nearly

degenerate conformational substates leads to a distribution of

barrier heights and pathways so that the structural relaxation is

highly nonexponential, as observed in the geminate recombi-

nation of CO^{110, 111}and other ligands.^112,113Because proteins

Vibrational Relaxation of the Amide I Mode

The exchange mechanism of vibrational energy within pro-

teins is little understood. Several time-resolved experiments were

performed with picosecond and femtosecond infrared pulses

to study the dissipation of locally deposited energy in the pro-

tein backbone (amide I mode).^102–104The amide I vibrational

mode involves mainly the C=O stretching motion of the pep-

tide backbone under a small amount of mixing with the CN

and NH motions.¹⁰⁵Because the amide I vibrational mode is a

strong infrared absorber, this band is ideal for mid-infrared

transient absorption (“pump-probe”) measurements. Mb is al-

most entirely α-helical in the secondary structure, and it shows

the amide I band centered near 1650 cm^–1in D₂O. Peterson

et al. carried out mid-infrared transient absorption measure-

ments on the amide I band of Mb in a wide temperature range

(6 to 310 K).¹⁰³The temperature dependence of the vibra-

tional relaxation in the solvent mixture is slight, changing

from 1.9 ± 0.2 ps below 100 K to 1.2 ± 0.2 ps at 310 K.

Tokmakoff et al.¹⁰⁶and Kenkre et al.¹⁰⁷gave detailed discus-

sions on the temperature dependencies of vibrational relax-

ation. In general, if two or more low-frequency bath modes

are involved, vibrational relaxation has a significant tempera-

ture dependence, because at higher temperatures, the ther-

mal population of the bath modes increase the relaxation rate

through a stimulated process. The lack of a strong tempera-

ture dependence is indicative of a low-order relaxation pro-

cess where energy is transferred into the high-energy modes

of the system rather than directly to the low-energy solvent

or protein “bath” modes. Femtosecond pump-probe measure-

ments by Hamm et al. on small proteins, including apamin,

scyllatoxin, and bovine pancreatic tripsin inhibitor, each of

which contains both of the α-helix and β-sheet structures in

different proportions, show an identical lifetime of 1.2 ps for

amide I.¹⁰²Despite the differences in their sizes and second-

ary structures, the amide I relaxation dynamics observed in

both of these studies are essentially the same and appear to be

fundamental features of the amide I mode in polypeptides

and proteins. It is interesting that the heme and the amide I

modes show similar vibrational lifetimes,¹⁰⁴although the co-

incidence may be accidental.

272

U l t r a f a s t D y n a m i c s o f My o g l o b i n

have a highly associated nature, this part of the protein response

ary important protein motions and cooperative oxygen-bind-

ing in Hb. Once the dominant mechanism for the tertiary struc-

tural changes is clarified, the molecular mechanism behind the

R to T quaternary structural changes and general allosteric ef-

fects may become clear. Combined advances in theoretical and

experimental methods involving studies over a broad range of

time scales and length scales for the protein response are re-

quired to reach this ultimate goal.

function is often modeled as analogous to glass relaxation dy-

namics.¹¹⁴At the other extreme, in the short-time dynamics of

the protein response to the reaction forces, the dominant con-

tribution to the protein displacements can involve nondiffusive

motions. In this limit, the reaction forces are redistributed over

the entire protein structure. The nascent potential energy gra-

dients that develop during ligand dissociation would lead to

the collective displacement of a large number of atoms. In this

event, the propagation of the structural changes would best be

described as the displacement of atoms in a superposition of

the low-frequency collective modes of the protein. The validity

of the description is the same as the justification for the modal

treatment (e.g., instantaneous-normal mode approach) that

models the short-time behavior of liquids such as inertial mo-

tions.^115,116For proteins, the atoms are covalently bonded, and

the overall correlation length scale is determined by forces much

stronger than the intermolecular forces in liquids. The use of a

modal description for the short-time dynamics of protein mo-

tions is more rigorous than it is for liquids. Normal mode analy-

sis by Seno and Go found that 57% of the tertiary structural

changes can be accounted for by the displacements of six spa-

tially extended modes with frequencies ranging from 5 to 12

cm^–1.¹¹⁷Finally, it should be pointed out that the validity of the

modal description is connected to vibrational energy relaxation

in the protein. The time scale of energy relaxation forms the

guidelines for considering whether protein motions are collec-

tive or diffusive. For example, if the protein modes are strongly

anharmonically coupled, intramolecular vibrational redistribu-

tion within the protein can occur very rapidly, which can then

result in ultrafast exponential initial relaxation. This is a gen-

eral argument for stochastic processes in a condensed phase.^118,119

Thus, there are strong connections between the protein

dynamics and condensed-phase dynamics involving chemical

reactions. Moreover, proteins are the media of which static struc-

tures are often determined at atomic resolution with X-ray crys-

tallography and NMR spectroscopy. The structures in nature

can be modified artificially using site-directed mutation tech-

niques. These features are distinct advantages in physicochemi-

cal studies on proteins compared to such studies on liquids.

Therefore, biological questions, besides being important to

address in their own right, present a relatively untapped fertile

source stimulating studies of physical chemistry in condensed

phases. For this reason, the most detailed understanding of

condensed-phase reaction dynamics may come from studies of

biological systems. In oxygen binding by Hb, for example, the

central molecule (the heme) and the environment (the globin)

profoundly affect each other in a beautiful manner: the protein’s

structural changes form a strongly coupled system in which

tertiary relaxation is cascaded into the quaternary structural

transition. The biomechanics of the tertiary structural changes

in Mb are important for a detailed understanding of function-

The authors thank Dr. Yuki Uesugi for her

contribution to the development of our picosecond

time-resolved resonance Raman apparatus. Stimulating

discussions with Prof. Paul M. Champion of

Northeastern University and with Prof. Robin M.

Hochstrasser of the University of Pennsylvania are

gratefully acknowledged. This research was supported

in part by a Grant-in-Aid for Scientific Research on

Priority Area (No. 12045264) given to T.K. from the

Ministry of Education, Science, Sports and Culture,

Japan, and by a Grant-in-Aid for Encouragement

Young Scientists (No. 11740339) given to Y.M. from

the Japan Society for the Promotion of Science.

REFERENCES

[1] McCammon, J. A.; Harvey, S. C. Dynamics of Proteins and

Nucleic Acids; Cambridge University Press: Cambridge, 1987.

[2] Brooks, C. L., III; Karplus, M.; Pettit, B. M. Proteins: ATheo-

retical Perspective of Dynamics, Structure, and Thermody-

namics; Wiley: New York, 1988.

[3] Hochstrasser, R. M. J Chem Educ 1998, 75, 559.

[4] Elsaesser, T.; Kaiser, W. Ann Rev Phys Chem 1991, 42, 83.

[5] Seilmeier, A.; Kaiser, W. In Ultrashort Laser Pulses, Genera-

tion and Applications, Vol 60; Kaiser, W., Ed.; Springer-Verlag:

Berlin, 1993; p. 279.

[6] Owrutsky, J. C.; Raftery, D.; Hochstrasser, R. M. Ann Rev

Phys Chem 1994, 45, 519.

[7] Joo, T.; Jia, Y.; Yu, J.-Y.; Jonas, D. M.; Fleming, G. R. J Phys

Chem 1996, 100, 2399.

[8] Leeson, D. T.; Wiersma, D. A.; Fritsch, K.; Friedrich, J. J

Phys. Chem B 1997, 101, 6331.

[9] Lim, M.; Hamm, P.; Hochstrasser, R. M. Proc Natl Acad Sci

USA 1998, 95, 15315.

[10] Shibata, Y.; Kurita, A.; Kushida, T. Biochemistry 1999, 38,

1789.

[11] Dickerson, R. E.; Geiss, I. Hemoglobin: Structure, Function,

Evolution, and Pathology; Benjamin Cummings: Menlo Park,

1983.

[12] Gibson, Q. H. J Physiol 1956, 134, 112.

[13] Greene, B. I.; Hochstrasser, R. M.; Weisman, R. B.; Eaton,

W. A. Proc Natl Acad Sci USA 1978, 75, 5255.

273

T H E C H E M I C A L R E C O R D

[14] Petrich, J. W.; Poyart, C.; Martin, J. L. Biochemistry 1988,

[46] Kitagawa, T. In Biological Applications of Raman Spectros-

copy, Vol. III; Spiro, T. G., Ed.; Wiley: New York, 1987; p 97.

[47] Rousseau, D. L.; Friedman, J. M. In Biological Applications

of Raman Spectroscopy, Vol. III; Spiro, T. G., Ed.; Wiley:

New York, 1987; p 133.

[48] Bangcharoenpaurpong, O.; Schomacker, K. T.; Champion, P.

M. J Am Chem Soc 1984, 106, 5688.

[49] Takano, T. J Mol Biol 1977, 110, 569.

[50] Takano, T. J Mol Biol 1977, 110, 537.

[51] Kuczera, K.; Lambry, J.-C.; Martin, J.-L.; Karplus, M. Proc

Natl Acad Sci USA 1993, 90, 5805.

27, 4049.

[15] Gibson, Q. H.; Olson, J. S.; McKinnie, R. E.; Rohlfs, R. J. J

Biol Chem 1986, 261, 10228.

[16] Kachalova, G. S.; Pepov, A. N.; Bartunik, H. D. Science 1999,

284, 473.

[17] Spiro, T. G., Ed. Biological Applications of Raman Spectros-

copy; Wiley: New York, 1987.

[18] Wynne, K.; Hochstrasser, R. M. Chem Phys 1995, 193, 211.

[19] Hamm, P. Chem Phys 1995, 200, 415.

[20] Hamm, P.; Wiemann, S.; Zurek, M.; Zinth, W. Opt Lett 1994,

19, 1642.

[21] Hamm, P.; Lim, M.; Hochstrasser, R. M. J Chem Phys 1997,

107, 10523.

[22] Okamoto, H.; Tasumi, M. Chem Phys Lett 1996, 256, 502.

[23] Okamoto, H. Chem Phys Lett 1998, 283, 33.

[24] Akhremitchev, B. B.; Wang, C.; Walker, G. C. Rev Sci Inst

1996, 67, 3799.

[52] Lim, M.; Jackson, T. A.; Anfinrud, P. A. Proc Natl Acad Sci

USA 1993, 90, 5801.

[53] Schaad, O.; Zhou, H.-X.; Szabo, A.; Eaton, W. A.; Henry, E.

R. Proc Natl Acad Sci USA 1993, 90, 9547.

[54] Li, H.; Elber, R.; Straub, J. E. J Biol Chem 1993, 268, 17908.

[55] Findsen, E. W.; Scott, T. W.; Chance, M. R.; Friedman, J. M.

J Am Chem Soc 1985, 107, 3355.

[25] Schomacker, K. T.; Bangcharoenpaurpong, O.; Champion, P.

M. J Chem Phys 1984, 80, 4701.

[56] Nagai, K.; Kitagawa, T. Proc Natl Acad Sci USA 1980, 77,

2033.

[26] Okamoto, H.; Nakabayashi, T.; Tasumi, M. J Phys Chem A

1997, 101, 3488.

[57] Nagai, K.; Kitagawa, T.; Morimoto, H. J Mol Biol 1980, 136,

271.

[27] Shreve, A. P.; Mathies, R. A. J Phys Chem 1995, 99, 7285.

[28] Henry, E. R.; Eaton, W. A.; Hochstrasser, R. M. Proc Natl

Acad Sci USA 1986, 83, 8982.

[29] Horng, M. L.; Gardecki, J. A.; Papazyan, A.; Maroncelli, M. J

Phys Chem 1995, 99, 17311.

[30] Uesugi, Y.; Mizutani, Y.; Kitagawa, T. Rev Sci Instrum 1997,

68, 4001.

[31] Uesugi, Y.; Mizutani, Y.; Kruglik, S. G.; Shvedko, A. G.;

Orlovich, V. A.; Kitagawa, T. J Raman Spectrosc 2000, 31,

339.

[58] Ondrias, M. R.; Rousseau, D. L.; Shelnutt, J. A.; Simon, S.

R. Biochemistry 1985, 21, 3428.

[59] Friedman, J. M.; Rousseau, D. L.; Ondrias, M. R.; Stepnoski,

R. A. Science 1982, 218, 1244.

[60] Friedman, J. M.; Stepnoski, R. A.; Nobel, R. W. FEBS Lett

1982, 146, 278.

[61] Friedman, J. M.; Rousseau, D. L.; Ondrias, M. R. Ann Rev

Phys Chem 1982, 33, 471.

[62] Irwin, M. J.; Atkinson, G. H. Nature (London) 1981, 293,

317.

[63] Stein, P.; Terner, J.; Spiro, T. G. J Phys Chem 1982, 86, 168.

[64] Peterson, E. S.; Friedman, J. M.; Chien, E. Y. T.; Sliger, S. G.

Biochemistry 1998, 37, 12301.

[65] Smulevich, G.; Mauro, J. M.; Fishel, L. A.; English, A. M.;

Kraut, J.; Spiro, T. G. Biochemistry 1988, 27, 5477.

[66] Royer, W. E., Jr. J Mol Biol 1994, 235, 657.

[67] Goodno, G. D.; Astinov, V.; Miller, R. J. D. J Phys Chem A

1999, 103, 10630.

[68] Causgrove, T. P.; Dyer, R. B. J Phys Chem 1996, 100, 3273.

[69] Xie, X.; Simon, J. D. Biochemistry 1991, 30, 3682.

[70] Jackson, T. A.; Lim, M.; Anfinrud, P. A. J Chem Phys 1994,

180, 131.

[71] Mizutani, Y.; Kitagawa, T. Science 1997, 278, 443.

[72] Mizutani, Y.; Uesugi, Y.; Kitagawa, T. J Chem Phys 1999,

111, 8950.

[73] Keyes, M. H.; Falley, M.; Lumry, R. J Am Chem Soc 1971,

93, 2035.

[74] Lian, T.; Locke, B.; Kholodenko, Y.; Hochstrasser, R. M. J

Phys Chem 1994, 98, 11648.

[75] Genberg, L.; Heisel, F.; McLendon, G.; Miller, R. J. D. J Phys

Chem 1987, 91, 5521.

[32] Hamaguchi, H.; Gustafson, T. L. Ann Rev Phys Chem 1994,

45, 593, and references therein.

[33] Kaminaka, S.; Kitagawa, T. Appl Spectrosc 1995, 49, 685.

[34] Mizutani, Y. J Chem Phys 1998, 109, 9197.

[35] Mizutani, Y.; Kitagawa, T. Manuscript in preparation.

[36] Abe, M.; Kitagawa, T.; Kyogoku, Y. J Chem Phys 1978, 69,

4526.

[37] Li, X.-Y.; Czernucszewicz, R. S.; Kincaid, J. R.; Stein, P.; Spiro,

T. G. J Phys Chem 1990, 94, 47.

[38] Lim, M.; Jackson, T. A.; Anfinrud, P. A. J Phys Chem 1996,

100, 12043.

[39] Petrich, J. W.; Martin, J. L. Chem Phys 1989, 131, 31.

[40] Cornelius, P. A.; Steele, A. W.; Chernoff, D. A.; Hochstrasser,

R.M. Proc Natl Acad Sci USA 1981, 78, 7526.

[41] Li, P.; Sage, J. T.; Champion, P. M. J Chem Phys 1992, 97,

3214.

[42] Rosca, F.; Kumar, A. T. N.; Ye, X.; Sjodin, T.; Demidov, A. A.;

Champion, P. M. J Phys Chem A 2000, 104, 4280.

[43] Gibson, Q. H.; Olson, J. S.; McKinnie, R. E.; Rohlfs, R. J. J

Biol Chem 1986, 261, 10228.

[44] Kitagawa, T.; Nagai, K.; Tsubaki, M. FEBS Lett 1979, 104,

376.

[45] Hu, S.; Smith, K. M.; Spiro, T. G. J Am Chem Soc 1996,

118, 12638.

[76] Miller, R. J. D. Ann Rev Phys Chem 1994, 42, 581.

[77] Lingle, R., Jr.; Xu, X.; Zhu, H.; Yu, S.-C.; Hopkins, J. B. J

Phys Chem 1991, 95, 9320.

274

U l t r a f a s t D y n a m i c s o f My o g l o b i n

[78] Petrich, J. W.; Martin, J. L.; Houde, D.; Poyart, C.; Orszag,

[100] Sagnella, D. E.; Straub, J. E.; Jackson, T. A.; Lim, M.;

Anfinrud, P. A. Proc Natl Acad Sci USA 1999, 96, 14324.

[101] Whitnell, R. M.; Wilson, K. R.; Hynes, J. T. J Phys Chem

1990, 94, 8625.

A. Biochemistry 1987, 26, 7914.

[79] Sagnella, D. E.; Straub, J. E.; Thirumalai, D. J Chem Phys

2000, 113, 7702.

[80] Hill, J. R.;Tokmakoff, A.; Peterson, K. A.; Sauter, B.; Zimdars,

D.; Dlott, D. D.; Fayer, M. D. J Phys Chem 1994, 98, 11213.

[81] Owrutsky, J. C.; Li, M.; Locke, B.; Hochstrasser, R. M. J Phys

Chem 1995, 99, 4842.

[102] Hamm, P.; Lim, M.; Hochstrasser, R. M. J Phys Chem B 1998,

102, 6123.

[103] Peterson, K. A.; Rella, C. W.; Engholm, J. R.; Schwettman,

H. A. J Phys Chem B 1999, 103, 557.

[82] Sension, R. J.; Szarka, A. Z.; Hochstrasser, R. M. J Chem

Phys 1992, 97, 5239.

[104] Xie, A.; van der Meer, L.; Hoff, W.; Austin, R. H. Phys Rev

Lett 2000, 84, 5435.

[83] Sension, R. J.; Repinec, S. T.; Szarka, A. Z.; Hochstrasser, R.

M. J Chem Phys 1993, 98, 6291.

[105] Miyazawa, T.; Shimanouchi, T.; Mizushima, S. J Chem Phys

1958, 29, 611.

[84] Qian, J.; Schultz, S. L.; Jean, J. M. Chem Phys Lett 1995, 233, 9.

[85] Schultz, S. L.; Qian, J.; Jean, J. M. J Phys Chem A 1997, 101,

1000.

[86] Nakabayashi, T.; Okamoto, H.; Tasumi, M. J Phys Chem A

1997, 101, 7189.

[106] Tokmakoff, A.; Sauter, B.; Feyer, M. D. J Chem Phys 1994,

100, 9035.

[107] Kenkre, V. M.; Tokmakoff, A.; Feyer, M. D. J Chem Phys

1994, 101, 10618.

[108] Harada, I.; Takeuchi, H. In Spectroscopy of Biological Sys-

tems, Vol. 13; Clark, R. J. H.; Hester, R. E., Eds.; Wiley: New

York, 1986; p 113.

[87] Nakabayashi, T.; Okamoto, H.; Tasumi, M. J Phys Chem A

1998, 102, 9686.

[88] Sato, S.; Kitagawa, T. Appl Phys B 1994, 59, 415.

[89] Schneebeck, M. C.; Vigil, L. E.; Ondrias, M. R. Chem Phys

Lett 1993, 215, 251.

[90] Simpson, M. C.; Peterson, E. S.; Shannon, C. F.; Eads, D. D.;

Friedman, J. M.; Cheatum, C. M.; Ondrias, M. R. J Am Chem

Soc 1997, 119, 5110.

[91] Schlichting, I.; Berendzen, J.; Phillips, G. N., Jr.; Sweet, R.

M. Nature 1994, 371, 808.

[92] Teng, T. Y.; Srajer, V.; Moffat, K. Nat Struct Biol 1994, 1, 701.

[93] Hartmann, H.; Zimser, S.; Komninos, P.; Scheinder, R. T.;

Nienhaus, G. U.; Parak, F. Proc Natl Acad Sci USA 1996, 93,

7013.

[94] Šrajer, V.; Teng, T.; Ursby, T.; Pradervand, C.; Ren, Z.; Adachi,

S.; Schildkamp, W.; Bourgeois, D.; Wulff, M.; Moffat, K. Sci-

ence 1996, 274, 1726.

[95] Lim, M.; Jackson, T. A.; Anfinrud, P. A. Science 1995, 269,

962.

[96] Lim, M.; Jackson, T. A.; Anfinrud, P. A. Nat Struct Biol 1997,

4, 209.

[97] Straub, J. E.; Karplus, M. Chem Phys 1991, 158, 221.

[98] Ma, J.; Huo, S.; Straub, J. E. J Am Chem Soc 1997, 119, 4355.

[99] Vitkup, D.; Petsko, G. A.; Karplus, M. Nat Struct Biol 1997,

4, 202.

[109] Tian, W. D.; Sage, J. T.; Champion, P. M.; Chien, E.; Sligar,

S. G. Biochemistry 1996, 35, 3487.

[110] Lambright, D. G.; Balasubramanian, S.; Boxer, S. G. Bio-

chemistry 1993, 32, 10116.

[111] Tian, W. D.; Sage, J. T.; Šrajer, V.; Champion, P. M. Phys Rev

Lett 1992, 68, 408.

[112] Chernoff, D. A.; Hochstrasser, R. M.; Steele, A. W. Proc Natl

Acad Sci USA 1980, 77, 5606.

[113] Petrich, J. W.; Lambry, J. C.; Kuczera, K.; Karplus, M.; Poyart,

C.; Martin, J. L. Biochemistry 1991, 30, 3975.

[114] Hagen, S. J.; Eaton, W. A. J Chem Phys 1996, 104, 3395.

[115] Bunchner, M.; Ladanyi, B. M.; Stratt, R. M. J Chem Phys

1992, 97, 8522.

[116] Cho, M.; Fleming, G. R.; Saito, S.; Ohmine, I.; Stratt, R. M.

J Chem Phys 1994, 100, 6672.

[117] Seno, Y.; Go, N. J Mol Biol 1990, 216, 111.

[118] Kubo, R.; Toda, M.; Hashitsume, N. Statistical Physics II:

Nonequilibrium Statistical Mechanics; Springer-Verlag: Ber-

lin, 1991; Chapter 2.

[119] Kubo, R. In Fluctuations, Relaxation, and Resonance in Mag-

netic Systems; ter Haar, D., Ed.; Plenum Press: New York,

1962; p 23.

275

Article Doi

DOI: 10.1002/tcr.1012

Source and publish data:

Authors:

Article abstract of DOI:10.1002/tcr.1012

Full text of DOI:10.1002/tcr.1012

Products guided by the article

R&D Labs maybe for 103791-67-9

Relevant to this article

Hot Product