Glassy behavior of a protein

Excitation energy transfer and charge separation in the isolated Photosystem II reaction center

The nature of excitation energy transfer and charge separation in isolated Photosystem II reaction centers is an area of considerable interest and controversy. Excitation energy transfer from accessory chlorophyll a to the primary electron donor P680 takes place in tens of picoseconds, although there is some evidence that thermal equilibration of the excitation between P680 and a subset of the accessory chlorophyll a occurs on a 100-fs timescale. The intrinsic rate for charge separation at low temperature is accepted to be ca. (2 ps)(-1), and is based on several measurements using different experimental techniques. This rate is in good agreement with estimates based on larger sized particles, and is similar to the rate observed with bacterial reaction centers. However, near room temperature there is considerable disagreement as to the observed rate for charge separation, with several experiments pointing to a ca. (3 ps)(-1) rate, and others to a ca. (20 ps)(-1) rate. These processes and the experiments used to measure them will be reviewed.

Thermal access to amplified chemical potential and the determination of equilibrium constants in protein solutions at subfreezing temperatures

During rapid cooling of ferric heme protein solutions containing fluoride, locally concentrated ligand cannot fully equilibrate with heme before the temperature drops below 200 K and into the range where energy is insufficient for exchange with iron-bound water. When temperature is then jumped above 200 K, exchange of fluoride for bound water is activated. Between 200 and 240 K, further fluoride complex formation takes place over several minutes; its extent is measured along the kinetic curve by reimmersing the sample into liquid nitrogen and taking EPR spectra. Kinetic curves for replacement of iron-bound water by fluoride in horse aquo-ferrimyoglobin and human aquo-ferrihemoglobin, and corresponding equilibrium constants have been obtained at temperatures between 200 and 240 K. The reaction rates are affected by sucrose. Results indicate that the kinetics of exchange of fluoride for heme-bound water at subfreezing temperatures is protein specific and not diffusion-controlled, and is not affected by the phase transition of ice which takes place at subfreezing temperature. Free energy changes accompanying these reactions are largely continuous as the systems pass from above to below freezing.

Protein reaction kinetics in a room-temperature glass

Protein reaction kinetics in aqueous solution at room temperature are often simplified by the thermal averaging of conformational substates. These substates exhibit widely varying reaction rates that are usually exposed by trapping in a glass at low temperature. Here, it is shown that the solvent viscosity, rather than the low temperature, is primarily responsible for the trapping. This was demonstrated by placement of myoglobin in a glass at room temperature and subsequent observation of inhomogeneous reaction kinetics. The high solvent viscosity slowed the rate of crossing the energy barriers that separated the substates and also suppressed any change in the average protein conformation after ligand dissociation.

The old problems of glass and the glass transition, and the many new twists

In this paper I review the ways in which the glassy state is obtained both in nature and in materials science and highlight a "new twist"--the recent recognition of polymorphism within the glassy state. The formation of glass by continuous cooling (viscous slowdown) is then examined, the strong/fragile liquids classification is reviewed, and a new twist-the possibility that the slowdown is a result of an avoided critical point-is noted. The three canonical characteristics of relaxing liquids are correlated through the fragility. As a further new twist, the conversion of strong liquids to fragile liquids by pressure-induced coordination number increases is demonstrated. It is then shown that, for comparable systems, it is possible to have the same conversion accomplished via a first-order transition within the liquid state during quenching. This occurs in the systems in which "polyamorphism" (polymorphism in the glassy state) is observed, and the whole phenomenology is accounted for by Poole's bond-modified van der Waals model. The sudden loss of some liquid degrees of freedom through such weak first-order transitions is then related to the polyamorphic transition between native and denatured hydrated proteins, since the latter are also glass-forming systems--water-plasticized, hydrogen bond-cross-linked chain polymers (and single molecule glass formers). The circle is closed with a final new twist by noting that a short time scale phenomenon much studied by protein physicists-namely, the onset of a sharp change in d<r2>/dT (<r2> is the Debye-Waller factor)--is general for glass-forming liquids, including computer-simulated strong and fragile ionic liquids, and is closely correlated with the experimental glass transition temperature. The latter thus originates in strong anharmonicity in certain components of the vibrational density of states, which permits the system to access the multiple minima of its configuration space. The connection between the anharmonicity in these modes, vibrational localization, the Kauzmann temperature, and the fragility of the liquid is proposed as the key problem in glass science.

Thermal fluctuations between conformational substates of the Fe(2+)-HisF8 linkage in deoxymyoglobin probed by the Raman active Fe-N epsilon (HisF8) stretching vibration

We have measured the VFe-His Raman band of horse heart deoxymyoglobin dissolved in an aqueous solution as a function of temperature between 10 and 300 K. The minimal model to which these data can be fitted in a statistically significant and physically meaningful way comprises four different Lorentzian bands with frequencies at 197, 209, 218, and 226 cm-1, and a Gaussian band at 240 cm-1, which exhibit halfwidths between 10 and 12.5 cm-1. All these parameters were assumed to be independent of temperature. The temperature dependence of the apparent total band shape's frequency is attributed to an intensity redistribution of the subbands at omega 1 = 209 cm-1, omega 2 = 218 cm-1, and omega 3 = 226 cm-1, which are assigned to Fe-N epsilon (HisF8) stretching modes in different conformational substrates of the Fe-HisF8 linkage. They comprise different out-of-plane displacements of the heme iron. The two remaining bands at 197 and 240 cm-1 result from porphyrin modes. Their intensity ratio is nearly temperature independent. The intensity ratio I3/I2 of the vFe-His subbands exhibits a van't Hoff behavior between 150 and 300 K, bending over in a region between 150 and 80 K, and remains constant between 80 and 10 K, whereas I2/I1 shows a maximum at 170 K and approaches a constant value at 80 K. These data can be fitted by a modified van't Hoff expression, which accounts for the freezing into a non-equilibrium distribution of substates below a distinct temperature Tf and also for the linear temperature dependence of the specific heat of proteins. The latter leads to a temperature dependence of the entropic and enthalpic differences between conformational substates. The fits to the intensity ratios of the vFe-His subbands yield a freezing temperature of Tf = 117 K and a transition region of delta T = 55 K. In comparison we have utilized the above thermodynamic model to reanalyze earlier data on the temperature dependence of the ratio Ao/A1 of two subbands underlying the infrared absorption band of the CO stretching vibration in CO-ligated myoglobin (A. Ansari, J. Berendzen, D. Braunstein, B. R. Cowen, H. Frauenfelder, M. K. Kong, I. E. T. Iben, J. Johnson, P. Ormos, T. B. Sauke, R. Scholl, A. Schulte, P. J. Steinbach, R. D. Vittitow, and R. D. Young, 1987, Biophys. Chem. 26:237-335). This yields thermodynamic parameters, in particular the freezing temperature (Tf = 231 K) and the width of the transition region (AT =8 K), which are significantly different from the corresponding parameters obtained from the above vFe-His data, but very close to values describing the transition of protein bound water from a liquid into an amorphous state. These findings and earlier reported data on the temperature dependence exhibited by the Soret absorption bands of various deoxy and carbonmonoxymyoglobins led us to the conclusion that the fluctuations between conformational substates of the heme environment in carbonmonoxymyoglobin are strongly coupled to motions within the hydration shell, whereas the thermal motions between the substates of the Fe-HisF8 linkage in deoxymyoglobin proceed on an energy landscape that is mainly determined by the intrinsic properties of the protein. The latter differ from protein fluctuations monitored by Mossbauer experiments ondeoxymyoglobin crystals which exhibit a strong coupling to the protein bound water and most probably reflect a higher tier in the hierarchical arrangement of substates and equilibrium fluctuations.

Ligand binding to heme proteins. V. Light-induced relaxation in proximal mutants L89I and H97F of carbonmonoxymyoglobin

We have studied the proximal mutants L89I and H97F of MbCO with FTIR and temperature-derivative spectroscopy at temperatures between 10 and 160 K. The mutations give rise only to minor alterations of the stretch spectra of the bound and photodissociated CO ligand. The most pronounced difference is a larger population in the A3 substate at approximately 1930 cm-1 in the mutants. The barrier distributions, as determined by temperature-derivative spectroscopy, are very similar to native MbCO after short illumination. Extended illumination leads to substantial increases of the rebinding barriers in native MbCO and the proximal mutants. A larger fraction of light-relaxed states is found in the proximal mutants, implying that the conformational energy landscape has been modified to more easily allow light-induced transitions. These and other spectroscopic data imply that the large changes in the binding properties are brought about by a light-induced conformational relaxation involving the structure at the heme iron. Similarities with spectral hole-burning studies and physical models are discussed.

Progress in high resolution atomic force microscopy in biology

The atomic force microscope (AFM) was invented by Binnig, Quate and Gerber less than 10 years ago (Binniget al. 1986). In their first prototype, a piece of goldfoil was used as the cantilever, with a crushed diamond tip mounted at the end. On the back of the cantilever, a tunnelling junction was used to monitor the deflection of the cantilever (the gold-foil) when the specimen was scanned with the tip in contact with the surface. Thus, the surface topography of the specimen was obtained with a resolution critically dependent on the sharpness of the tip provided the deformation of the specimen was not serious. Even with such a crude set-up, they managed to obtain a lateral resolution of ˜ 30 Å and a vertical resolution of better than 1 Å on an amorphous A12O3surface. The operating principle of such an instrument is deceptively simple. However, such an arrangement was inconvenient for routine operations and unsuitable for imaging hydrated specimens, because the tunnelling junction is easily contaminated in air and works poorly in aqueous solutions (Alexanderet al. 1989). As a result, the application of this type of AFM to biological samples was rare (Engel, 1991).

Formation of Glasses from Liquids and Biopolymers

Glasses can be formed by many routes. In some cases, distinct polyamorphic forms are found. The normal mode of glass formation is cooling of a viscous liquid. Liquid behavior during cooling is classified between "strong" and "fragile," and the three canonical characteristics of relaxing liquids are correlated through the fragility. Strong liquids become fragile liquids on compression. In some cases, such conversions occur during cooling by a weak first-order transition. This behavior can be related to the polymorphism in a glass state through a recent simple modification of the van der Waals model for tetrahedrally bonded liquids. The sudden loss of some liquid degrees of freedom through such first-order transitions is suggestive of the polyamorphic transition between native and denatured hydrated proteins, which can be interpreted as single-chain glass-forming polymers plasticized by water and cross-linked by hydrogen bonds. The onset of a sharp change in d<r(2)>dT(<r(2)> is the Debye-Waller factor and T is temperature) in proteins, which is controversially indentified with the glass transition in liquids, is shown to be general for glass formers and observable in computer simulations of strong and fragile ionic liquids, where it proves to be close to the experimental glass transition temperature. The latter may originate in strong anharmonicity in modes ("bosons"), which permits the system to access multiple minima of its configuration space. These modes, the Kauzmann temperature T(K), and the fragility of the liquid, may thus be connected.

Stark-effect experiments on photochemical holes in chromoproteins: protoporphyrin IX-substituted myoglobin

We performed comparative Stark-effect experiments on spectral holes in a protein and a glass sample. The protein was protoporphyrin IX-substituted myoglobin in a glycerol/water solvent. The glass sample was a protoporphyrin IX-doped mixture of dimethylformamide/glycerol. As expected, in both cases the spectral holes varied linearly with the electric field. Yet, whereas in the protein the holes showed a clear splitting, they showed no splitting in the glass sample, irrespective of the chosen polarization of the laser. In both samples the hole broadened in the applied field. The magnitude of the broadening was about the same in both cases. The following conclusions were drawn. The absence of a splitting in the glass signals an effective global inversion symmetry of the chromophore, despite its low symmetry group. The dipole moment changes are random. In the protein the inversion symmetry is broken through the spatial correlation of the protein building blocks, leading to a molecular frame-fixed dipole moment difference and, hence, to the observed splitting. Despite these symmetry-breaking properties, the local structural randomness is of the same magnitude in the glass and in the protein, as is obvious from the broadening. The distinct difference in the Stark pattern shows that the range of the relevant chromophore interactions is confined to typical dimensions of the protein.

A fractional calculus approach to self-similar protein dynamics

Relaxation processes and reaction kinetics of proteins deviate from exponential behavior because of their large amount of conformational substrates. The dynamics are governed by many time scales and, therefore, the decay of the relaxation function or reactant concentration is slower than exponential. Applying the idea of self-similar dynamics, we derive a fractal scaling model that results in an equation in which the time derivative is replaced by a differentiation (d/dt)beta of non-integer order beta. The fractional order differential equation is solved by a Mittag-Leffler function. It depends on two parameters, a fundamental time scale tau 0 and a fractional order beta that can be interpreted as a self-similarity dimension of the dynamics. Application of the fractal model to ligand rebinding and pressure release measurements of myoglobin is demonstrated, and the connection of the model to considerations of energy barrier height distributions is shown.

Recent advances in biological atomic force microscopy

Recent developments in biological atomic force microscopy are reviewed. In addition to the advances in methodology, new structural information of different biological systems revealed by the atomic force microscopy is also presented. A discussion regarding the contrast, resolution and specimen deformation is provided based on a theoretical model.

Ligand binding to heme proteins: the effect of light on ligand binding in myoglobin

Extended illumination slows the rebinding of CO to myoglobin after photodissociation at cryogenic temperatures. Two types of models have been put forward to explain the effect: motions of the CO within the heme pocket or conformational transitions of the protein. To resolve this ambiguity, we have studied the effect of extended illumination on ligand binding to horse and sperm whale myoglobin (hMb and swMb) with temperature-derivative spectroscopy, monitoring the reaction in the CO stretch bands in the infrared and the conformation-sensitive band III near 760 nm. The experiments show that the stretch frequency of the photodissociated CO does not change upon illumination, implying that the slowing of the CO rebinding is caused by conformational relaxation of Mb from the bound state toward the deoxy structure. The light-induced relaxation (LIR) depends on the number of photons absorbed but not on the light intensity or duration separately. LIR occurs on photon absorption in either the bound or photodissociated state and depends on the temperature at which the MbCO is illuminated. The LIR proceeds in jumps through a small number of conformational substates. The effective barrier for rebinding increases with each step. The substates populated are similar to those found in the thermally-induced relaxation (TIR) that is observed above 160 K. LIR depends markedly on the structural details; it differs for swMbCO and hMbCO and even for the three A substates of swMbCO. Pronounced differences exist between the effects in MbCO and MbO2. The similarity of LIR and TIR leads to a revised model for ligand binding to swMbCO and hMbCO, in which the relaxation is crucial for the escape of the ligand from the pocket, as was first suggested by Friedman [Friedman, J. M. (1985) Science 228, 1273-1280].

The carbon monoxide stretching modes in camphor-bound cytochrome P-450cam. The effect of solvent conditions, temperature, and pressure

The effect of pH, monovalent cations, glycerol, temperature, and pressure on the carbonmonoxy (CO) stretching mode of camphor-bound cytochrome P-450cam (CYP 101) was studied. Two effects, band overlap and frequency shift, have been observed. The CO stretch infrared band located at about 1940 cm-1 is asymmetric because of the overlap of three bands at about 1931 cm-1, 1939 cm-1, and 1942 cm-1 with strongly different populations. Reducing the temperature or increasing the pressure leads to splitting the band or switching the asymmetry from the lower energy side to the higher energy side of the infrared band. The overlap of several CO stretch bands indicates conformational substates within the heme pocket. A frequency shift of the predominantly populated band is observed by changing all the parameters mentioned. The pH-induced frequency shift follows an S-shape with the pK at 6.2, which matches the pK observed for the pH-induced high-spin/low-spin transition. Conformational changes on the proximal heme side are suggested to be the origin. Monovalent cations at saturating concentration induce a small frequency shift depending on the ion radius. The potassium ion is the one that induces a CO stretch frequency with the highest wave-number while sodium and lithium (smaller radii) and rubidium and caesium ion (larger radii) have diminished values, which is supporting evidence for the special function of the potassium ion within the structure. Glycerol and hydrostatic pressure induce a red shift of the CO stretching frequency. Forced contact of the polar hydroxyl group of Thr252 of the I helix induced by pressure and indirectly by glycerol is suggested to change the CO dipole moment, reflecting in the decreased CO stretching frequency.

FTIR spectroscopic study of the dynamics of conformational substates in hydrated carbonyl-myoglobin films via temperature dependence of the CO stretching band parameters

Two hydrated carbonyl myoglobin (MbCO) films, one containing (0.30 g water)/(g MbCO) from MbCO solution in water at pH 5.5 and the other (0.32 g water)/(gMbCO) from 0.1 M potassium phosphate buffer solution at pH 6.8, were studied by FTIR spectroscopy from 293 K to 78 K at selected temperatures on cooling and reheating. Above approximately 180 K the general trend in temperature dependence of half-bandwidths, peak maxima, and band area ratios of the A1 and A3 conformer bands is similar to those reported by Ansari et al. (1987. Biophys. J. 26:337) for MbCO in 75% glycerol/water solution, but abrupt changes in slopes at approximately 180-200 K and freezing-in of conformer populations, which could be taken as indicator for glass transition of the solvent or the protein, are absent for the hydrated MbCO films. This is interpreted in terms of an exceptionally broad distribution of relaxation times, and is in accord with conclusions from recent calorimetric annealing studies of hydrated protein powders (Sartor et al. 1994. Biophys. J. 66:249). Exchange between the three A conformers does not stop at approximately 180-200 K but occurs over the whole temperature region studied. These results are then discussed with respect to MbCO's behavior in the glass-->liquid transition region of glass-forming solvents, and it is concluded that, in analogy to the behavior of low-molecular-weight compounds with a distribution of rapidly interconverting conformers, freezing-in of MbCO's A conformer populations by the solvent should not be mistaken for a glass transition of MbCO.

Conformational relaxation and ligand binding in myoglobin

Absorption spectroscopy with nanosecond time resolution shows that myoglobin undergoes conformational relaxation on the same time scale as geminate rebinding of carbon monoxide. Ligand rebinding following photodissociation of the heme-CO complex was measured from the amplitude of the average difference spectrum, while conformational changes were measured from changes in the detailed shape of the Soret spectra of the deoxyhemes. Experiments in which the solvent viscosity was varied between 1 and 300 cP and the temperature between 268 and 308 K were analyzed by fitting the multiwavelength kinetic data with both empirical and molecular models. Novel numerical techniques were employed in fitting the data, including the use of singular value decomposition to remove the effects of temperature and solvent on the spectra and of a Monte Carlo method to overcome the multiple minimum problem in searching parameter space. The molecular model is the minimal model that incorporates all of the major features of myoglobin kinetics at ambient temperatures, including a fast and slow rebinding conformation and two geminate states for each conformation. The results of fitting the kinetic data with this model indicate that the geminate-rebinding rates for the two conformations differ by at least a factor of 100. The differences between the spectra of the two conformations generated from the fits are similar to the differences between those of the R and T conformations of hemoglobin. In modeling the data, the dependence of the rates on temperature and viscosity was parametrized using a modification of Kramers theory which includes the contributions of both protein and solvent to the friction. The rate of the transition from the fast to the slow rebinding conformation is found to be inversely proportional to the viscosity when the viscosity exceeds about 30 cP and nearly viscosity independent at low viscosity. The viscosity dependence at high viscosities suggests that the two conformations differ by the global displacement of protein atoms on the proximal side of the heme observed by X-ray crystallography. We suggest that the conformational change observed in our experiments corresponds to the final portion of the nonexponential conformational relaxation recently observed by Anfinrud and co-workers, which begins on a picosecond time scale. Furthermore, extrapolation of our data to temperatures near that of the solvent glass transition suggests that this conformational relaxation may very well be the one postulated by Frauenfelder and co-workers to explain the decrease in the rate of geminate rebinding with increasing temperature above 180 K.

The proton activity at cryogenic temperatures--a possible influence on the spin state of the heme iron of cytochrome P-450cam in supercooled buffered solutions

The electronic absorption spectra for camphor-bound cytochrome P-450cam have been analysed in the temperature range between 78 K and 298 K. The well-known high-spin/low-spin equilibrium has been detected between 298 K and 220 K. Depending on the cooling rate, below 220 K a new species was found in the absorption spectra. In contrast, the electronic absorption spectra for camphor-free cytochrome P-450cam between 78 K and 295 K show no significant spectral changes. The conversion between the spin states of camphor-bound cytochrome P-450cam and the appearance of the new species do not correspond to the temperature-induced change in the paH value of the aqueous glycerol mixture containing phosphate or cacodylate buffer (paH 7.0). For this study a spectroscopic procedure for the determination of the temperature dependence of the paH value of the solvent for the range 78-295 K is presented using dyes as pH-indicators. It is shown that the state of the acid-base equilibrium frozen in is strongly dependent on the cooling rate of the mixture.

Calorimetric studies of the kinetic unfreezing of molecular motions in hydrated lysozyme, hemoglobin, and myoglobin

Differential scanning calorimetric (DSC) studies of the glassy states of as-received and hydrated lysozyme, hemoglobin, and myoglobin powders, with water contents of < or = 0.25, < or = 0.30, and < or = 0.29 g/g of protein, show that their heat capacity slowly increases with increasing temperature, without showing an abrupt increase characteristic of glass-->liquid transition. Annealing (also referred to as physical aging) of the hydrated proteins causes their DSC scans to show an endothermic region, similar to an overshoot, immediately above the annealing temperature. This annealing effect appears at all temperatures between approximately 150 and 300 K. The area under these peaks increases with increasing annealing time at a fixed temperature. The effects are attributed to the presence of a large number of local structures in which macromolecular segments diffuse at different time scales over a broad range. The lowest time scale corresponds to the > N-H and -O-H group motions which become kinetically unfrozen at approximately 150-170 K on heating at a rate of 30 K min-1 and which have a relaxation time of 5-10 s in this temperature range. The annealing effects confirm that the individual glass transition of the relaxing local regions is spread over a temperature range up to the denaturation temperature region of the proteins. The interpretation is supported by simulation of DSC scans in which the distribution of relaxation times is assumed to be exceptionally broad and in which annealing done at several temperatures over a wide range produces endothermic effects (or regions of DSC scans) qualitatively similar to those observed for the hydrated proteins.

Elasticity of globular proteins. The relation between mechanics, thermodynamics and mobility

An analysis of elasticity of lysozyme and myoglobin crystals in terms of thermodynamics has revealed a direct relation between entropy and enthalpy of deformation and delta S* and delta H* terms in the standard free energy change in proteins, delta G(o), (K.P. Murphy, P.L. Privalov, S.J. Gill (1990) Science 247, 559-561), so that at any temperature (between the glass-transition and denaturation temperatures) free energy of deformation is proportional to the hydration independent part of delta G(o). Both energies are characterized with large enthalpy-entropy compensation and tend to zero at the same temperature, Tm = (delta H*/delta S*) = 353 +/- 20 K. Large positive entropy contribution to deformation energy causes large linear decrease in protein elasticity, and increase in thermal mobility of protein atoms with temperature. Being plotted in inverse coordinates, temperature dependence of the mean-square amplitudes, obtained in neutron and mossbauer experiments as well as in molecular dynamic simulations, gives the same 353 +/- 10 K for the temperature, where the amplitudes tend to infinity. Mechanism explaining large positive entropy contribution in deformation energy of native protein molecules presumably involves emergence of more room for motion of protein side-chain groups squeezed between alpha-helices and other rigid skeleton elements, when precise packing of atoms in native protein molecule is distorted as a result of deformation.

Ligand binding to heme proteins: II. Transitions in the heme pocket of myoglobin

Phenomena occurring in the heme pocket after photolysis of carbonmonoxymyoglobin (MbCO) below about 100 K are investigated using temperature-derivative spectroscopy of the infrared absorption bands of CO. MbCO exists in three conformations (A substrates) that are distinguished by the stretch bands of the bound CO. We establish connections among the A substates and the substates of the photoproduct (B substates) using Fourier-transform infrared spectroscopy together with kinetic experiments on MbCO solution samples at different pH and on orthorhombic crystals. There is no one-to-one mapping between the A and B substates; in some cases, more than one B substate corresponds to a particular A substate. Rebinding is not simply a reversal of dissociation; transitions between B substates occur before rebinding. We measure the nonequilibrium populations of the B substates after photolysis below 25 K and determine the kinetics of B substate transitions leading to equilibrium. Transitions between B substates occur even at 4 K, whereas those between A substates have only been observed above about 160 K. The transitions between the B substates are nonexponential in time, providing evidence for a distribution of substates. The temperature dependence of the B substate transitions implies that they occur mainly by quantum-mechanical tunneling below 10 K. Taken together, the observations suggest that the transitions between the B substates within the same A substate reflect motions of the CO in the heme pocket and not conformational changes. Geminate rebinding of CO to Mb, monitored in the Soret band, depends on pH. Observation of geminate rebinding to the A substates in the infrared indicates that the pH dependence results from a population shift among the substates and not from a change of the rebinding to an individual A substate.

Structural heterogeneity of the Fe(2+)-N epsilon (HisF8) bond in various hemoglobin and myoglobin derivatives probed by the Raman-active iron histidine stretching mode

We have examined the Fe(2+)-N epsilon (HisF8) complex in hemoglobin A (HbA) by measuring the band profile of its Raman-active nu Fe-His stretching mode at pH 6.4, 7.0, and 8.0 using the 441-nm line of a HeCd laser. A line shape analysis revealed that the band can be decomposed into five different sublines at omega 1 = 195 cm-1, omega 2 = 203 cm-1, omega 3 = 212 cm-1, omega 4 = 218 cm-1, and omega 5 = 226 cm-1. To identify these to the contributions from the different subunits we have reanalyzed the nu Fe-His band of the HbA hybrids alpha(Fe)2 beta(Co)2 and alpha(Co)2 beta(Fe)2 reported earlier by Rousseau and Friedman (D. Rousseau and J. M. Friedman. 1988. In Biological Application on Raman Spectroscopy. T. G. Spiro, editor, 133-216). Moreover we have reanalyzed other Raman bands from the literature, namely the nu Fe-His band of the isolated hemoglobin subunits alpha SH- and beta SH-HbA, various hemoglobin mutants (i.e., Hb(TyrC7 alpha-->Phe), Hb(TyrC7 alpha-->His), Hb M-Boston and Hb M-Iwate), N-ethylmaleimide-des(Arg141 alpha) hemoglobin (NES-des(Arg141 alpha)HbA) and photolyzed carbonmonoxide hemoglobin (Hb*CO) measured 25 ps and 10 ns after photolysis. These molecules are known to exist in different quaternary states. All bands can be decomposed into a set of sublines exhibiting frequencies which are nearly identical to those found for deoxyhemoglobin A. Additional sublines were found to contribute to the nu Fe-His band of NES-des(Arg141 alpha) HbA and the Hb*CO species. The peak frequencies of the bands are determined by the most intensive sublines. Moreover we have measured the nu Fe-His band of deoxyHbA at 10 K in an aqueous solution and in a 80% glycerol/water mixture. Its subline composition at this temperature depends on the solvent and parallels that of more R-like hemoglobin derivatives. We have also measured the optical charge transfer band III of deoxyHbA at room temperature and found, that at least three subbands are required to fit its asymmetric band shape. This corroborates the findings on the nu Fe-His band in that it is indicative of a heterogeneity of the Fe(2+)-N epsilon(HisF8) bond. Finally we measured the nu Fe-His band of horse heart deoxyMb at different temperatures and decomposed it into three different sublines. In accordance with what was obtained for HbA their intensities rather than their frequencies are temperature-dependent. By comparison with VFe-His bands of some Mb mutants (i.e., Mb(His E7.->Gly) and Mb(HisE7__*Met) we suggest that these sublines may be attributed to different conformations of the heme pocket. Our data show, that the V Fe-His band is governed by at least two different coordinates x and y determining its frequency and intensity, respectively. While the former can be assigned to the tilt angle theta between the Fe2+-NJ(HisF8) bond and the heme normal and/or to the displacement delta of the iron from the heme plane, variations in the intensity may be caused by changes of the azimuthal angle phi formed by the projection of the proximal imidazole and the N(l)-Fe2+-N(III) line of the heme. The sublines are therefore interpreted as resulting from different conformational substates of the Fe2+-N(HisFa) complex which differ in terms of x (theta and/or delta). Each of them may further be subdivided in sub-substates with respect to the coordinate y (theta). Quaternary and tertiary transitions of the protein alter the population of these substates thus giving rise to a redistribution of the VFe-HiS sublines which shifts the corresponding peak frequency to higher values.

Steric constraints in the retinal binding pocket of sensory rhodopsin I

Steric constraints in the retinal binding pocket of sensory rhodopsin I (SR-I) are analyzed by studying effects of sample temperature and retinal analogs. The flash-induced yield of the earliest detected intermediate S610, which corresponds to the K intermediate in the bacteriorhodopsin (BR) photocycle, decreases below 220 K and reaches zero at 100 K, while K formation is independent of temperature. The reduced S610 formation at low temperatures indicates a more restricted retinal binding pocket in SR-I during primary photochemical events. Introduction of bulky substituents on the retinal polyene chain in four retinal analogs greatly retards or blocks the final step of chromophore binding to the apoprotein of SR-I. Except for the 14-methyl substitution, these modifications exhibit little or no effect on chromophore binding to BR apoprotein. These results corroborate that the retinal polyene chain binding domain in SR-I is more sterically constrained than that of the retinal pocket in BR. Deletion of the beta-ionone ring renders the analog SR-I pigments nonfunctional, as does deletion of the 13-methyl group, but the corresponding BR analogs are both photochemically and physiologically active. In contrast to the corresponding BR analog, photolysis of the analog SR-I reconstituted with 13-desmethylretinal does not produce an S610-like intermediate at room temperature. The above results and the previous findings that protein constraints inhibit the accommodation of a stable 13-cis-retinal configuration in SR-I suggest a model in which the 13-methyl group functions as a fulcrum to permit movement of one or both ends of retinal to overcome an energy barrier against isomerization.

Kinetics and thermodynamics of folding in model proteins

Monte Carlo simulations on a class of lattice models are used to probe the thermodynamics and kinetics of protein folding. We find two transition temperatures: one at T theta, when chains collapse from a coil to a compact phase, and the other at Tf (< T theta), when chains adopt a conformation corresponding to their native state. The kinetics are probed by several correlation functions and are interpreted in terms of the underlying energy landscape. The transition from the coil to the native state occurs in three distinct stages. The initial stage corresponds to a random collapse of the protein chain. At intermediate times tau c, during which much of the native structure is acquired, there are multiple pathways. For longer times tau r (>> tau c) the decay is exponential, suggestive of a late transition state. The folding time scale (approximately tau r) varies greatly depending on the model. Implications of our results for in vitro folding of proteins are discussed.

The role of solvent viscosity in the dynamics of protein conformational changes

Nanosecond lasers were used to measure the rate of conformational changes in myoglobin after ligand dissociation at ambient temperatures. At low solvent viscosities the rate is independent of viscosity, but at high viscosities it depends on approximately the inverse first power of the viscosity. Kramers theory for unimolecular rate processes can be used to explain this result if the friction term is modified to include protein as well as solvent friction. The theory and experiment suggest that the dominant factor in markedly reducing the rate of conformational changes in myoglobin at low temperatures (less than 200 K) is the very high viscosity (greater than 10(7) centipoise) of the glycerol-water solvent. That is, at low temperatures conformational substates may not be "frozen" so much as "stuck."

Solvent damping of internal processes in myoglobin studied by specific heat spectroscopy and flash photolysis

We address the question of dynamic coupling between protein and solvent by comparing the enthalpy relaxation of the solvent (75% v/v glycerol-water) to internal ligand binding in myoglobin. When the solvent relaxation is slow compared to intramolecular events we observe decoupling of protein motions from the solvent. In the opposite limit there is a significant contribution of the solvent to internal friction. The solvent enhances the apparent activation energy of transitions in myoglobin. This result is discussed in terms of a generalized Kramer's law involving a dynamic friction coefficient.

Structural fluctuations and conformational entropy in proteins: entropy balance in an intramolecular reaction in methemoglobin

The reversible intramolecular binding of the distal histidine side chain to the heme iron in methemoglobin is of special interest due to the very large negative reaction entropy which overcompensates the large reaction enthalpy. It may be considered as a prominent example of the ability of proteins (including enzymes) to provide global entropy in a local process. In this work new experiments and model calculations are reported which aim at finding the structural elements contributing to the reaction entropy. Geometrical studies prove the implication of the 20 residue E-helix being shifted by more than 2 A. Vibrational entropies are calculated by a procedure derived from the method of Karplus and Kushik. It turns out that neither the histidine alone nor the complete E-helix contribute more than 15 per cent of the required entropy.

Spectroscopic evidence for conformational relaxation in myoglobin

The time and temperature dependencies of the line area (M0) and position (M1) of band III at approximately 760 nm have been measured with Fourier-transform infrared spectroscopy in deoxymyoglobin (Mb) and continuously photolyzed carbon monoxide myoglobin (MbCO). Below 200 K, the area of band III in the photoproduct Mb* increases with time even on time scales of hours. This behavior indicates changes in the distribution of activation enthalpy barriers for ligand rebinding under extended illumination. The band position of Mb* shifts to higher wavenumbers with increasing temperature up to 100 K owing to kinetic hole burning; the same protein coordinate that controls the position of band III also determines the rebinding barrier height. The shift ceases above 100 K, implying that more than one protein coordinate affects the height of the rebinding barrier. Above 160 K, the line position in Mb* shifts again and coalesces with the value of Mb for temperatures above 200 K. The shift is accompanied by an increase of the line area, reflecting a slowing of rebinding kinetics. Both effects are explained in the framework of the model introduced by Steinbach et al. [(1991) Biochemistry 30, 3988-4001]. Above approximately 160 K, the conformational relaxation Mb*----Mb simultaneously shifts the line position of band III and increases the enthalpy barrier for ligand rebinding. Furthermore, equilibrium fluctuations lead to an averaging of the band position and the rebinding enthalpy.

Spectral hole burning study of protoporphyrin IX substituted myoglobin

Protoporphyrin IX substituted myoglobin reveals excellent hole burning properties. We investigated the frequency shift of persistent spectral holes under isotropic pressure conditions in a range from 0 to 2.4 MPa. In this range, the protein behaves like an elastic solid. The shift of the holes under pressure shows a remarkable frequency dependence from which the compressibility of the protein can be determined. The compressibility, in turn, allows for an estimation of the equilibrium volume fluctuations. Within the frame of the model used to interpret the pressure data, it is possible to determine the absorption frequency of the isolated chromophore and the associated solvent shift in the protein environment.

Determination of rate distributions from kinetic experiments

Rate processes in proteins are often not adequately described by simple exponential kinetics. Instead of modeling the kinetics in the time domain, it can be advantageous to perform a numerical inversion leading to a rate distribution function f(lambda). The features observed in f(lambda) (number, positions, and shapes of peaks) can then be interpreted. We discuss different numerical techniques for obtaining rate distribution functions, with special emphasis on the maximum entropy method. Examples are given for the application of these techniques to flash photolysis data of heme proteins.