Search

Significance The ubiquitous enzymes called cytochromes P450 catalyze a broad range of chemical reactions using molecular oxygen. For example, in humans, these enzymes are involved in breakdown of foreign compounds, including drugs. The bacterial cytochrome P450cam is thought to open up to allow substrate to enter the active site, and then to close during catalysis to keep reactive intermediates inside. Surprisingly, recent crystal structures suggested that the enzyme is open during the reaction. We have studied the enzyme in solution using paramagnetic NMR spectroscopy, demonstrating that, in fact, the enzyme is closed. This finding indicates that the subtle balance between open and closed is affected by crystallization, which can lead to the wrong conclusions about the protein dynamics.

NMR relaxation dispersion techniques provide a powerful method to study protein dynamics by characterizing lowly populated conformations that are in dynamic exchange with the major state. Paramagnetic NMR is a versatile tool for investigating the structures and dynamics of proteins. These two techniques were combined here to measure accurate and precise pseudocontact shifts of a lowly populated conformation. This method delivers valuable long-range structural restraints for higher energy conformations of macromolecules in solution. Another advantage of combining pseudocontact shifts with relaxation dispersion is the increase in the amplitude of dispersion profiles. Lowly populated states are often involved in functional processes, such as enzyme catalysis, signaling, and protein/protein interactions. The presented results also unveil a critical problem with the lanthanide tag used to generate paramagnetic relaxation dispersion effects in proteins, namely that the motions of the tag can interfere severely with the observation of protein dynamics. The two-point attached CLaNP-5 lanthanide tag was linked to adenylate kinase. From the paramagnetic relaxation dispersion only motion of the tag is observed. The data can be described accurately by a two-state model in which the protein-attached tag undergoes a 23° tilting motion on a timescale of milliseconds. The work demonstrates the large potential of paramagnetic relaxation dispersion and the challenge to improve current tags to minimize relaxation dispersion from tag movements.

Cytochrome P450cam (P450cam) is a heme-containing monooxygenase that catalyzes the hydroxylation of D-camphor to produce 5-exo-hydroxycamphor. The catalytic cycle of P450cam requires two electrons, both of which are donated by putidaredoxin (Pdx), a ferredoxin containing a [2 Fe-2 S] cluster. Atomic-resolution structures of the Pdx-P450cam complex have recently been solved by X-ray crystallography and paramagnetic NMR spectroscopy. The binding interface showed the potential electron transfer pathways and interactions between Pdx Asp38 and P450cam Arg112, as well as hydrophobic contacts between the Pdx Trp106 and P450cam residues. Several polar residues not previously recognized as relevant for binding were found in the interface. In this study, site-directed mutagenesis, kinetic measurements, and NMR studies were employed to probe the energetic importance and role of the polar residues in the Pdx-P450cam interaction. A double mutant cycle (DMC) analysis of kinetic data shows that favorable interactions exist between Pdx Tyr33 and P450cam Asp125, as well as between Pdx Ser42 and P450cam His352. The results show that alanine substitutions of these residues and several others do not influence the rates of electron transfer. It is concluded that these polar interactions contribute to partner recognition rather than to electronic coupling of the redox centers.

The use of 13C NMR relaxation dispersion experiments to monitor micro-millisecond fluctuations in the protonation states of histidine residues in proteins is investigated. To illustrate the approach, measurements on three specifically 13C labeled histidine residues in plastocyanin (PCu) from Anabaena variabilis (A.v.) are presented. Significant Carr-Purcell-Meiboom-Gill (CPMG) relaxation dispersion is observed for 13C(epsilon1) nuclei in the histidine imidazole rings of A.v. PCu. The chemical shift changes obtained from the CPMG dispersion data are in good agreement with those obtained from the chemical shift titration experiments, and the CPMG derived exchange rates agree with those obtained previously from 15N backbone relaxation measurements. Compared to measurements of backbone nuclei, 13C(epsilon1) dispersion provides a more direct method to monitor interchanging protonation states or other kinds of conformational changes of histidine side chains or their environment. Advantages and shortcomings of using the 13C(epsilon1) dispersion experiments in combination with chemical shift titration experiments to obtain information on exchange dynamics of the histidine side chains are discussed.

A model describing conformational exchange of His 61 in plastocyanin from Anabaena variabilis is presented. A detailed picture of the exchange dynamics has been obtained using solution NMR relaxation measurements, chemical shift titrations, and structural information provided by a high-resolution crystal structure of the protein. A three-site model for chemical exchange that involves interconversion among the tautomeric and protonated forms of the histidine side chain with rates that are fast on the NMR chemical shift time scale can account for all of the data. In general, in the limit of fast exchange, it is not possible to obtain separate measures of chemical shift differences and populations of the participating states using NMR. However, we show here that when the data mentioned above are combined, it is possible to extract values of all of the parameters that characterize the exchange process, including rates, populations, and chemical shift changes, and to provide cross-validations that establish their accuracy. The methodology is generally applicable to the study of histidine side chain dynamics, which can play an important functional role in many protein systems.

Electric fields generated in native proteins affect almost every aspect of protein function. We present a method that probes changes in the electric field at specific locations within a protein. The method utilizes the dependence of the amide 1H and 15N NMR chemical shifts on electric charges in proteins. Charges were introduced at different positions in the blue copper protein plastocyanin, by protonation of side chains or by substitution of the metal ion. It is found that the associated chemical shift perturbations (CSPs) stem mainly from long-range electric field effects caused by the change in the electric charge. It is demonstrated that the CSPs can be used to estimate the dielectric constant at different locations in the protein, estimate the nuclear shielding polarizability, or position charges in proteins. Proteins 2008. © 2008 Wiley-Liss, Inc.

Metal ions play a key role for the function of many proteins. The interaction of the metal ion with the protein and its involvement in the function of the protein vary widely. In some proteins, the metal ion is bound tightly to the ligand residues and may be the key player in the function of the protein, as in the case of blue copper proteins. In other proteins, the metal ion is bound only temporarily and loosely to the protein, as in the case of some metalloenzymes and other proteins where the metal ion acts as a cofactor necessary for the function of the protein. Such proteins are often known as metal ion-activated proteins. The review focuses on recent nuclear magnetic resonance (NMR) studies of a series of metal-dependent proteins and the characterization of the metal-binding sites. In particular, we focus on NMR techniques for studying metal binding to proteins such as chemical shift mapping, paramagnetic NMR and changes in backbone dynamics upon metal binding.

Electrostatics play an important role in many aspects of protein chemistry. However, the accurate determination of side chain proton affinity in proteins by experiment and theory remains challenging. In recent years the field of nuclear magnetic resonance spectroscopy has advanced the way that protonation states are measured, allowing researchers to examine electrostatic interactions at an unprecedented level of detail and accuracy. Experiments are now in place that follow pH-dependent 13C and 15N chemical shifts as spatially close as possible to the sites of protonation, allowing all titratable amino acid side chains to be probed sequence specifically. The strong and telling response of carefully selected reporter nuclei allows individual titration events to be monitored. At the same time, improved frameworks allow researchers to model multiple coupled protonation equilibria and to identify the underlying pH-dependent contributions to the chemical shifts.

Paramagnetic NMR probes provide valuable long-range structural information on proteins and protein complexes. A new, stable, two-armed lanthanoid probe is reported that can be attached to a protein site-specifically via chemically inert thioether linkages.

Paramagnetic NMR spectroscopy has evolved rapidly in the last decade, and has shown to be a very useful tool for solving structures of protein–protein complexes. A major breakthrough has been the development of paramagnetic metal binding tags that can be attached specifically to the protein. These tags have greatly facilitated the use of anisotropic paramagnetic restraints such as pseudocontact shifts and residual dipolar couplings arising from paramagnetic self-alignment. Such restraints are particularly useful for the study of large protein complexes. This review focuses on the recent developments in structural characterization of protein–protein complexes using anisotropic paramagnetic NMR restraints.

The use of paramagnetic NMR data for the refinement of structures of proteins and protein complexes is widespread. However, the power of paramagnetism for protein assignment has not yet been fully exploited. PARAssign is software that uses pseudocontact shift data derived from several paramagnetic centers attached to the protein to obtain amide and methyl assignments. The ability of PARAssign to perform assignment when the positions of the paramagnetic centers are known and unknown is demonstrated. PARAssign has been tested using synthetic data for methyl assignment of a 47 kDa protein, and using both synthetic and experimental data for amide assignment of a 14 kDa protein. The complex fitting space involved in such an assignment procedure necessitates that good starting conditions are found, both regarding placement and strength of paramagnetic centers. These starting conditions are obtained through automated tensor placement and user-defined tensor parameters. The results presented herein demonstrate that PARAssign is able to successfully perform resonance assignment in large systems with a high degree of reliability. This software provides a method for obtaining the assignments of large systems, which may previously have been unassignable, by using 2D NMR spectral data and a known protein structure.

An efficient way to rapidly generate protein-ligand costructures based on solution-NMR using sparse NOE data combined with selective isotope labeling is presented. A docked model of the 27 kDa N-terminal ATPase domain of Hsp90 bound to a small molecule ligand was generated using only 21 intermolecular NOEs, which uniquely defined both the binding site and the orientation of the ligand. The approach can prove valuable for the early stages of fragment-based drug discovery.

Paramagnetic lanthanides ions are broadly used in NMR spectroscopy. The effects of unpaired electrons on NMR spectral parameters provide a powerful tool for the characterization of macromolecular structures and dynamics. Here, a new lanthanide-chelating NMR probe, Caged Lanthanide NMR Probe-7 (CLaNP-7), is presented. It can be attached to protein surfaces via two disulfide bridges, yielding a probe that is rigid relative to the protein backbone. CLaNP-7 extends the application range of available probes. It has a yellow color, which is helpful for sample preparation. Its effects are comparable to those of CLaNP-5, but its charge is two units lower (+1) than that of CLaNP-5 (+3), reducing the change in surface potential after probe attachment. It also has a different magnetic susceptibility tensor, so by using both tags, two sets of structural restraints can be obtained per engineered cysteine pair. Moreover, it was found that the orientation of the magnetic susceptibility tensor is pH dependent (pK(a) ≈ 7) when a histidine residue is located in the neighborhood of the probe attachment site. The results show that the His imidazole group interacts with the CLaNP-7 tag. It is proposed that the histidine residue forms a hydrogen bond to a water/hydroxyl molecule that occupies the ninth coordination position on the lanthanide, thus breaking the two-fold symmetry of the CLaNP tag in a pH-dependent way.

Paramagnetic lanthanide tags potentially can enhance the effects of microsecond to millisecond dynamics in proteins on NMR signals and provide structural information on lowly populated states encoded in the pseudocontact shifts. We have investigated the microsecond to millisecond mobility of a two-point attached lanthanide tag, CLaNP-5, using paramagnetic (1)H CPMG relaxation dispersion methods. CLaNP-5 loaded with Lu(3+), Yb(3+), or Tm(3+) was attached to three sites on the surface of two proteins, pseudoazurin and cytochrome c. The paramagnetic center causes large relaxation dispersion effects for two attachment sites, suggesting that local dynamics of the protein at the attachment site causes mobility of the paramagnetic center. At one site the relaxation dispersions are small and limited to the immediate environment of the tag. It is concluded that paramagnetic relaxation dispersion could represent a sensitive method to probe protein dynamics. However, the selection of a rigid attachment site is of critical importance.

In oxygenic photosynthetic cells, carbon metabolism is regulated by a light-dependent redox signaling pathway through which the light signal is transmitted in the form of electrons via a redox chain comprising ferredoxin (Fd), ferredoxin:thioredoxin reductase (FTR), and thioredoxin (Trx). Trx affects the activity of a variety of enzymes via dithiol oxidation and reduction reactions. FTR reduces an intramolecular disulfide bridge of Trx, and Trx reduction involves a transient cross-link with FTR. NMR spectroscopy was used to investigate the interaction of Fd, FTR, and an m-type Trx. NMR titration experiments indicate that FTR uses distinct sites to bind Fd and Trx simultaneously to form a noncovalent ternary complex. The orientation of Trx-m relative to FTR was determined from the intermolecular paramagnetic broadening caused by the [4Fe-4S] cluster of FTR. Two models of the noncovalent binary complex of FTR/Trx-m based on the paramagnetic distance restraints were obtained. The models suggest that either a modest or major rotational movement of Trx must take place when the noncovalent binary complex proceeds to the covalent complex. This study demonstrates the complementarity of paramagnetic NMR and X-ray diffraction of crystals in the elucidation of dynamics in a transient protein complex.

Exchange on the microsecond time scale between the protonated and deprotonated forms of His92 in the copper site of reduced plastocyanin from the cyanobacteria Anabaena variabilis was monitored using 15N NMR relaxation measurements. On the basis of the dependence of the kinetics on pH and phosphate buffer concentration, we propose a two-step model for the protonation of the copper site in agreement with previous crystallographic studies. It is shown that the proton transfer is the rate-limiting step in the reaction at low buffer concentrations, whereas at high buffer concentrations, another step becomes rate-limiting. We suggest that the latter step is a concerted dissociation of His92 from the Cu(I) ion and a 180 degrees rotation of the imidazole ring, which precede the protonation. The first-order rate constant for the dissociation of His92 from the Cu(I) ion is estimated to be 2.4 x 10(4) s(-1). Also, a cooperative effect of the protonation of the remote His61 on the protonation of His92 and the redox properties of the protein was investigated by substituting His61 with asparagine. The mutation causes a modest change in both the pKa value of His92 and the redox potential of the protein.

Two methods for estimating the microsecond–millisecond dynamics in proteins from only two 15N relaxation parameters at one magnetic field strength are investigated. Thus, the chemical exchange contribution, Rex, to the transversal relaxation rate, which contains the dynamics information, is evaluated by two methods: (i) one in which the Rex term is derived from the 15N R1 and R2 relaxation rates alone, and (ii) one in which it is obtained from the transversal dipole-chemical shift anisotropy (CSA) cross-correlation rate, ηxy, and the R2 rate. Since the R1, R2, and ηxy experiments are fast and sensitive, both methods are attractive in studies where large amounts of dynamical information are required. However, both methods are liable to effects that can compromise the estimation of the Rex terms. In the R2/R1 method, internal ps-ns dynamics and rotational anisotropy can interfere with the determination of Rex, while in the R2/ηxy method it can be affected by variations in the 15N chemical shift anisotropy. Here, the applicability of the two methods is investigated using plastocyanin from Anabaena variabilis as an example, and the quality of the obtained Rex terms is evaluated both theoretically and experimentally. It is found that the R2/R1 method gives reliable Rex terms if the protein is relatively rigid and tumbles fast and nearly isotropically in solution, as for instance plastocyanin, and is preferable in such cases. In contrast, the R2/ηxy method gives better results if the protein is flexible or highly non-spherical and can be used for such proteins, if the sequential variation in the 15N chemical shift anisotropy is negligible. For exchange terms

An approach is presented that allows a detailed, quantitative characterization of conformational exchange processes in proteins on the micros-ms time scale. The approach relies on a combined analysis of NMR relaxation rates and chemical shift changes and requires that the chemical shift of the exchanging species can be determined independently of the relaxation rates. The applicability of the approach is demonstrated by a detailed analysis of the conformational exchange processes previously observed in the reduced form of the blue copper protein, plastocyanin from the cyanobacteria Anabaena variabilis (A.v. PCu) (Ma, L.; Hass, M. A. S.; Vierick, N.; Kristensen, S. M.; Ulstrup, J.; Led, J. J. Biochemistry 2003, 42, 320-330). The R1 and R2 relaxation rates of the backbone 15N nuclei were measured at a series of pH and temperatures on an 15N labeled sample of A.v. PCu, and the 15N chemical shifts were obtained from a series of HSQC spectra recorded in the pH range from 4 to 8. From the R1 and R2 relaxation rates, the contribution, Rex, to the transverse relaxation caused by the exchanges between the different allo-states of the protein were determined. Specifically, it is demonstrated that accurate Rex terms can be obtained from the R1 and R2 rates alone in the case of relatively rigid proteins with a small rotational anisotropy. The Rex terms belonging to the same exchange process were identified on the basis of their pH dependences. Subsequently the identifications were confirmed quantitatively by the correlation between the Rex terms and the corresponding chemical shift differences of the exchanging species. By this approach, the Rex terms of 15N nuclei belonging to contiguous regions in the protein could be assigned to the same exchange process. Furthermore, the analysis of the exchange terms shows that the observed micros-ms dynamics in A.v. PCu are caused primarily by the protonation/deprotonation of two histidine residues, His92 and His61, His92 being ligated to the Cu(I) ion. Also the exchange rate of the protonation/deprotonation process of His92 and its pH and temperature dependences were determined, revealing a reaction pathway that is more complex than a simple specific-acid/base catalysis. Finally, the approach allows a differentiation between two-site and multiple-site exchange processes, thus revealing that the protonation/deprotonation of His61 is at least a three-site exchange process. Overall, the approach makes it feasible to obtain exchange rates that are sufficiently accurate and versatile for studies of the kinetics and the mechanisms of local protein dynamics on the sub-millisecond time scale.

The dynamics of the backbone of the electron-transfer protein plastocyanin from the cyanobacterium Anabaena variabilis were determined from the (15)N and (13)C(alpha) R(1) and R(2) relaxation rates and steady-state [(1)H]-(15)N and [(1)H]-(13)C nuclear Overhauser effects (NOEs) using the model-free approach. The (13)C relaxation studies were performed using (13)C in natural abundance. Overall, it is found that the protein backbone is rigid. However, the regions that are important for the function of the protein show moderate mobility primarily on the microsecond to millisecond time scale. These regions are the "northern" hydrophobic site close to the metal site, the metal site itself, and the "eastern" face of the molecule. In particular, the mobility of the latter region is interesting in light of recent findings indicating that residues also on the eastern face of plastocyanins from prokaryotes are important for the function of the protein. The study also demonstrates that relaxation rates and NOEs of the (13)C(alpha) nuclei of proteins are valuable supplements to the conventional (15)N relaxation measurements in studies of protein backbone dynamics.

An NMR approach is presented that provides detailed information about short-lived, transient interactions between protein molecules in solution. The approach is based on the longitudinal paramagnetic relaxation rates of the protein nuclei and requires that at least one of the interacting molecules is paramagnetic. The specific interactions are monitored by the intermolecular paramagnetic contribution to the relaxation of protons at or close to the interaction surface. By applying the approach to plastocyanin from Anabaena variabilis, specific regions of interaction that may be involved in the electron self-exchange process of this plastocyanin were identified. This is in accord with recent 15N NMR relaxation studies of the backbone dynamics of Anabaena variabilis plastocyanin, with site-directed mutagenesis studies of the functional importance of the corresponding regions in Phormidium laminosum plastocyanin and with the crystal packing surface of P. laminosum plastocyanin.

Cytochrome P450cam catalyzes the hydroxylation of camphor in a complex process involving two electron transfers (ETs) from the iron-sulfur protein putidaredoxin. The enzymatic control of the successive steps of catalysis is critical for a highly efficient reaction. The injection of the successive electrons is part of the control system. To understand the molecular interactions between putidaredoxin and cytochrome P450cam, we determined the structure of the complex both in solution and in the crystal state. Paramagnetic NMR spectroscopy using lanthanide tags yielded 446 structural restraints that were used to determine the solution structure. An ensemble of 10 structures with an RMSD of 1.3Å was obtained. The crystal structure of the complex was solved, showing a position of putidaredoxin that is identical with the one in the solution structure. The NMR data further demonstrate the presence of a minor state or set of states of the complex in solution, which is attributed to the presence of an encounter complex. The structure of the major state shows a small binding interface and a metal-to-metal distance of 16Å, with two pathways that provide strong electronic coupling of the redox centers. The interpretation of these results is discussed in the context of ET. The structure indicates that the ET rate can be much faster than the reported value, suggesting that the process may be gated.