Binding of carbon monoxide to isolated hemoglobin chains

Azurin is a small blue copper protein in the electron transfer chain of denitrifying bacteria. It forms a photolabile complex with nitric oxide (NO) at low temperatures. We studied the temperature dependence of the ligand binding equilibrium and the kinetics of the association reaction after photodissociation over a wide range of temperature (80-280 K) and time (10-6-102 s). The nonexponential rebinding below 200 K is independent of the NO concentration and is interpreted as internal recombination. The rebinding can be modeled with the Arrhenius law by using a single preexponential factor of 6.3 x 108 s-' and a Gaussian distribution of enthalpy barriers centered at 23 kJ/mol with a width of 11 kJ/mol. Above 200 K, a slower, exponential rebinding process appears. The dependence of the kinetics on the NO concentration characterizes this reaction as bimolecular rebinding. The binding kinetics of NO to azurin show impressive analogies to the binding of carbon monoxide to myoglobin. We conclude that conformational substates occur not only in heme proteins but also in proteins with different active sites and secondary structures.Experiments measuring the binding of small ligands to heme proteins have contributed considerably to our understanding of the relation between the structure, dynamics, and function of proteins (1, 2). Sperm whale myoglobin (Mb) has served as a model system in these studies. Mb, a small, globular protein of 17.8 kDa, consists of 153 amino acids folded into a globular structure with eight a-helices. Diatomic ligands (02, CO) bind at the heme iron. The rebinding after photodissociation is nonexponential in time at low temperatures. An important concept results from the kinetic studies-namely, that proteins do not exist in a unique, well-defined structure, but in a large number of slightly different structures, the conformational substates (CS). Evidence for CS also comes from analysis of the Debye-Waller factor in protein crystals by x-ray diffraction (3, 4) and from the inhomogeneity of spectral lines (5, 6). The CS appear to be grouped into different tiers with distinctly different free energy barriers between the substates, leading to a hierarchical model of the conformational energy landscape in MbCO (7,8).We believe that these concepts apply not only to MbCO but to proteins in general. Strong covalent bonds establish the primary sequence, while only relatively weak forces like hydrogen bonds and hydrophobic effects determine the folding into the three-dimensional structure. Consequently, the positions of the amino acids are not unique, and competition among neighboring amino acids for the energetically optimal configuration arises from the dense packing. These two effects, disorder and frustration, lead to a complex conformational energy landscape with many local minima, the CS. The hierarchical arrangement of the CS may reflect the hierarchy of structural features in proteins.To study the generality of these concepts, we chose azurin (Az) (from Pseudomonas aeruginosa), a protein of ...

Section: Methodsmentioning

confidence: 99%

Conformational substates in azurin.

Ehrenstein

Nienhaus

1992

“…[3] ar Eq. 2 corresponds to zero net flux across the outer boundary, which can describe two different physical situations-either the ligand cannot escape into the solvent surrounding the protein or the outflux from the protein equals the influx of ligands from the solvent.…”

Section: Ligand Migrationmentioning

confidence: 99%

“…Our attention was drawn to this problem while investigating ligand migration through protein matter. Consider, for example, the most thoroughly researched case: ligand migration through myoglobin (Mb) (2)(3)(4)(5)(6)(7)(8)(9). Because no channel of sufficient size is available in the averaged static conformation of the protein (10), diffusion of ligand molecules (CO or 02) through the protein matrix cannot occur in the absence of conformational fluctuations (4,11,12).…”

mentioning

confidence: 99%

Biological transport processes and space dimension.

Nadler

Stein

1991

We discuss the generic time behavior of reaction-diffusion processes capable of modeling various types of biological transport processes, such as ligand migration in proteins and gating fluctuations in ion channel proteins. The main observable in these two cases, the fraction of unbound ligands and the probability of finding the channel in the closed state, respectively, exhibits an algebraic t-F/2 decay at intermediate times, followed by an exponential cutoff. We provide a simple framework for understanding these observations and explain their ubiquity by showing that these qualitative results are independent of space dimension. We also derive an experimental criterion to dis h between a one-dimensional process and one whose effective dimension is higher.Transport processes in biological systems can often be described as reaction-diffusion processes either in real space or in a configuration space of some appropriate dimension. A prominent example of the former is the diffusive transport of biological molecules through the cell protoplasm from the site where they are produced or incorporated to the location where they are needed and absorbed. Adam and Delbruck (1) pointed out the importance of dimensionality in such processes, by observing that an effective reduction of the dimension of the diffusion space-e.g., by restricting the motion of those molecules to certain network structures that pervade the cell-can be utilized to speed up this transport. In this paper we will be concerned with a different question: How does the time behavior of such transport processes depend on the space dimension involved, and, consequently, can experimental observation give any information on it? Our attention was drawn to this problem while investigating ligand migration through protein matter. Consider, for example, the most thoroughly researched case: ligand migration through myoglobin (Mb) (2-9). Because no channel of sufficient size is available in the averaged static conformation of the protein (10), diffusion of ligand molecules (CO or 02) through the protein matrix cannot occur in the absence of conformational fluctuations (4,11,12). The question then arises whether the ligand motion through the protein matrix is effectively three-dimensional or whether the necessary protein fluctuations occur only within a lower-dimensional subspace (the single-channel hypothesis) (4, 13).Usually, the ligand diffusion process can be experimentally studied only indirectly: bound ligands are separated from the active site by flash photolysis and the fraction of unbound ligands after a time t is monitored (2). The nonexponential rebinding kinetics then observed at low temperatures (process I, below 170 K) is commonly ascribed to a distribution of reaction barriers (2). At higher temperatures the ligand wanders off into the protein matrix (processes II and higher), and this diffusion process is also reflected in the rebinding curve. The time dependence of the curve that results from this diffusion is typically seen to exhibit a t-...

“…One is that the ATP(out)-ADP(in) exchange catalyzed by the well-known ATP-ADP translocase may actually be electroneutral, but this is contrary to a great deal of direct and indirect evidence (3,9,10,(13)(14)(15)(16) Communicated by Britton Chance, August 11, 1978 ABSTRACT (7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20). Most of this work was directed toward studying the rebinding of CO or 02 to Hb or myoglobin on time scales of microseconds or longer (8, 9, 11, 14-16, 18, 20).…”

mentioning

confidence: 99%

Spectroscopic studies of oxy- and carbonmonoxyhemoglobin after pulsed optical excitation.

Greene¹,

Hochstrasser²,

Weisman³

et al. 1978

161

107

The photolysis of HbO2 and HbCO has been investigated with picosecond laser techniques. Transient absorption spectra were measured in the Soret and visible regions after excitation with 353- or 530-nm pulses. The photoproducts appeared within 8 psec and exhibited considerably broadened deoxyhemoglobin-like spectra, which persisted to 680 psec. The altered spectra are attributed to the production of deoxyheme conformational and spin states that might result from the intense excitation.