Observer-selective double electron-electron-spin resonance, a pulse sequence to improve orientation selection

By pulsed double electron-electron resonance (DEER), distances between spin labels in disordered systems up to 8 nm can be measured. In addition, the relative orientation of the interacting radicals can be determined, provided that the bandwidth of the pulses is sufficiently small. On the other hand, the bandwidth has to exceed the dipolar interaction considerably, because otherwise the DEER modulations become distorted and the modulation depth decreases, making distance determination impossible. Therefore, small bandwidths, i.e. long pulses, place a lower limit on the distance that can be determined. Two new pulse sequences, observer-selective DEER (os-DEER) and dead-time free os-DEER, are introduced that make it possible to use long observer pulses with bandwidths that are smaller than the dipolar interaction. The new pulse sequences do not suffer from the distortions caused by the limited bandwidth of the observer pulses, as demonstrated by measurements on a nitroxide biradical. With observer pulses of 140 ns, i.e., significantly longer than the 32 ns used in the conventional DEER sequence, a dipolar interaction of 7.8 MHz has been measured.     

The determination of pair distance distributions by pulsed ESR using Tikhonov regularization

Pulsed ESR techniques with the aid of site-directed spin labeling have proven useful in providing unique structural information about proteins. The determination of distance distributions in electron spin pairs directly from the dipolar time evolution of the pulsed ESR signals by means of the Tikhonov regularization method is reported. The difficulties connected with numerically inverting this ill-posed mathematical problem are clearly illustrated. The Tikhonov regularization with the regularization parameter determined by the L-curve criterion is then described and tested to confirm its accuracy and reliability. The method is applied to recent experimental results on doubly labeled proteins that have been studied using two pulsed ESR techniques, double quantum coherence (DQC) ESR and double electron-electron resonance (DEER). The extracted distance distributions are able to provide valuable information about the conformational constraints in various partially folded states of proteins. This study supplies a mathematically reliable method for extracting pair distributions from pulsed ESR experimental data and has extended the use of pulsed ESR to provide results of greater value for structural biology.     

Distance determination in human ubiquitin by pulsed double electron-electron resonance and double quantum coherence ESR methods

Recently, distance measurements by pulsed ESR (electron spin resonance) have been obtained using pulsed DEER (double electron-electron resonance) and DQC (double quantum coherence) in SDSL (site directed spin labeling) proteins. These methods can observe long range dipole interactions (15-80A). We applied these methods to human ubiquitin proteins. The distance between the 20th and the 35th cysteine was estimated in doubly spin labeled human ubiquitin. Pulsed DEER requires two microwave sources. However, a phase cycle is not usually required in this method. On the other hand, DQC-ESR at X-band ( approximately 9GHz) can acquire a large echo signal by using pulses of short duration and high power, but this method has an ESEEM (electron spin echo envelope modulation) problem. We used a commercial pulsed ESR spectrometer and compared these two methods.     

Two-Dimensional Distance Correlation Maps from Pulsed Triple Electron Resonance (TRIER) on Proteins with Three Paramagnetic Centers

Recently, we introduced the pulsed triple electron resonance (TRIER) experiment, which correlates dipolar frequencies of molecules with three electron spins. These correlation patterns contain valuable information: in combination with double electron–electron resonance (DEER) they allow to interpret distance distributions of biological systems that exist in more than one conformation. Together with an improved sequence and new data processing, we were now for the first time able to obtain two-dimensional distance correlation maps of the previously investigated model compounds as well as of spin-labeled proteins. For this we applied two-dimensional approximate Pake transformation to TRIER data. This enabled us to get distance correlation plots from two triple-labeled protein samples that were in good agreement with DEER data and simulations. With such information it should then be possible to assign peaks in DEER distance distributions for macromolecules that can exist in more than one conformation.     

Distance measurements in the borderline region of applicability of CW EPR and DEER: a model study on a homologous series of spin-labelled peptides

Inter-spin distances between 1 nm and 4.5 nm are measured by continuous wave (CW) and pulsed electron paramagnetic resonance (EPR) methods for a series of nitroxide-spin-labelled peptides. The upper distance limit for measuring dipolar coupling by the broadening of the CW spectrum and the lower distance limit for the present optimally-adjusted double electron electron resonance (DEER) set-up are determined and found to be both around 1.6-1.9 nm. The methods for determining distances and corresponding distributions from CW spectral line broadening are reviewed and further developed. Also, the work shows that a correction factor is required for the analysis of inter-spin distances below approximately 2 nm for DEER measurements and this is calculated using the density matrix formalism.     

Dead-time free measurement of dipole-dipole interactions between electron spins

A four-pulse version of the pulse double electron-electron resonance (DEER) experiment is presented, which is designed for the determination of interradical distances on a nanoscopic length-scale. With the new pulse sequence electron-electron couplings can be studied without dead-time artifacts, so that even broad distributions of electron-electron distances can be characterized. A version of the experiment that uses a pulse train in the detection period exhibits improved signal-to-noise ratio. Tests on two nitroxide biradicals with known length indicate that the accessible range of distances extends from about 1.5 to 8 nm. The four-pulse DEER spectra of an ionic spin probe in an ionomer exhibit features due to probe molecules situated both on the same and on different ion clusters. The former feature provides information on the cluster size and is inaccessible with previous methods.     

Measuring Cu2+-Nitroxide Distances Using Double Electron–Electron Resonance and Saturation Recovery

Distance measurements were obtained between a bound Cu²⁺ and a spin label on two polypeptides of differing length using the double electron–electron resonance (DEER) and saturation recovery experiments. Distance distributions obtained from the DEER results resolved differences between the average distance and distribution of distances for each peptide. An average distance was also obtained for each peptide using the relaxation-based saturation recovery experiment. Predicted average distances for the relaxation-based method, <r_ESR>, were calculated using the distance distributions from the DEER experiment. The predicted <r_ESR> values were similar to those measured by saturation recovery; both were biased to shorter values compared with the DEER results. The breadth of the distance distributions had a significant effect on the average distance measured by saturation recovery. This work highlights the advantage of using DEER to measure metal-nitroxide distances in that the average distances measured are less biased than in relaxation-based techniques.     

DEER in biological multispin-systems: a case study on the fatty acid binding to human serum albumin

In this study, self-assembled systems of human serum albumin (HSA) and spin-labeled fatty acids are characterized by double electron-electron resonance (DEER). HSA, being the most important transport protein of the human blood, is capable to host up to seven paramagnetic fatty acid derivatives. DEER measurements of these self-assembled multispin clusters are strongly affected by correlations of more than two spins, the evaluation of the latter constituting the central topic of this paper. While the DEER modulation depth can be used to obtain qualitative information of the number of coupled spins, the quantitative analysis is hampered by the occurrence of cluster mixtures with different numbers of coupled spins and contributions from unbound spin-labeled material. Applying flip angle dependent DEER measurements, unwanted multispin correlations were found to lead not only to a broadening of the distance peaks but also to cause small distances to be overestimated and large distances to be suppressed. It is thus favorable to use spin-diluted systems with an average of two paramagnetic molecules per spin cluster when a quantitative analysis of the distance distribution is sought.     

Spin pair geometry revealed by high-field DEER in the presence of conformational distributions

Orientation selection on two nitroxide-labelled shape-persistent molecules is demonstrated by high-field pulsed electron-electron double resonance experiments at a frequency of 95 GHz with a commercial spectrometer. The experiments are performed with fixed observer and pump frequencies by variation of the magnetic field, so that the variation of both the dipolar frequencies and the modulation depths can be analyzed. By applying the deadtime-free four-pulse double electron-electron resonance (DEER) sequence, the lineshapes of the dipolar spectra are obtained. In the investigated linear biradical and equilateral triradical the nitroxide labels undergo restricted dynamics, so that their relative orientations are not fixed, but are correlated to some extent. In this situation, the general dependence of the dipolar spectra on the observer field can be satisfyingly modelled by simple geometrical models that involve only one rotational degree of freedom for the biradical and two rotational degrees of freedom for the triradical. A somewhat better agreement of the dipolar lineshapes for the biradical is obtained by simulations based on a molecular dynamics trajectory. For the triradical, small but significant deviations of the lineshape are observed with both models, indicating that the technique can reveal deficiencies in modelling of the conformational ensemble of a macromolecule.     

Simultaneous acquisition of pulse EPR orientation selective spectra

High resolution pulse EPR methods are usually applied to resolve weak magnetic electron-nuclear or electron-electron interactions that are otherwise unresolved in the EPR spectrum. Complete information regarding different magnetic interactions, namely, principal components and orientation of principal axis system with respect to the molecular frame, can be derived from orientation selective pulsed EPR measurements that are performed at different magnetic field positions within the inhomogeneously broadened EPR spectrum. These experiments are usually carried out consecutively, namely a particular field position is chosen, data are accumulated until the signal to noise ratio is satisfactory, and then the next field position is chosen and data are accumulated. Here we present a new approach for data acquisition of pulsed EPR experiments referred to as parallel acquisition. It is applicable when the spectral width is much broader than the excitation bandwidth of the applied pulse sequence and it is particularly useful for orientation selective pulse EPR experiments. In this approach several pulse EPR measurements are performed within the waiting (repetition) time between consecutive pulse sequences during which spin lattice relaxation takes place. This is achieved by rapidly changing the main magnetic field, B(0), to different values within the EPR spectrum, performing the same experiment on the otherwise idle spins. This scheme represents an efficient utilization of the spectrometer and provides the same spectral information in a shorter time. This approach is demonstrated on W-band orientation selective electron-nuclear double resonance (ENDOR), electron spin echo envelope modulation (ESEEM), electron-electron double resonance (ELDOR)--detected NMR and double electron-electron resonance (DEER) measurements on frozen solutions of nitroxides. We show that a factors of 3-6 reduction in total acquisition time can be obtained, depending on the experiment applied.     

Double Electron–Electron Resonance Between Trapped Electron and Hole in a Semiconductor

We demonstrate the spin interactions between dispersedly trapped electrons and holes in a semiconductor using the double electron–electron resonance (DEER) method of the pulsed electron paramagnetic resonance (EPR) techniques. An aluminum-doped titanium dioxide crystal is adopted as a spin system, in which optically generated electrons and holes are trapped, to reveal EPR signals that appear close to each other at a selected crystal orientation under an external magnetic field. We used the four-pulse DEER method by applying two microwave frequencies to a microwave cavity for pumping electrons and probing holes at the optimum temperature of 32 K. The dipolar modulation in the probed signal by pumping interacting spins was successfully detected. The observed non-oscillating decay shape indicates that the detected interaction is caused by widely distributed trapped electron and hole spins over long distances. We were able to extract a spin-pair distribution function by the first derivative of a background-corrected curve, referring to a previously reported method.     

High-frequency 180 GHz PELDOR

For aromatic organic radicals, pulsed electron-electron double resonance (PELDOR) experiments at high magnetic fields offer the possibility to achieve orientation-selective pumping and detection that could allow one not only to determine the distance between paramagnetic species but also their relative orientation with respect to the interconnecting dipolar axis. We present a PELDOR two-frequency setup that was introduced into our homebuilt 180 GHz pulsed electron paramagnetic resonance (EPR) spectrometer and we discuss its technical and experimental features. The capability of 180 GHz PELDOR has been tested using the three-pulse ELDOR sequence on the protein RNR-R2 (ribonucleotide reductase) fromEscherichia coli, which contains two tyrosyl radicals at a distance of 3.3 nm. At 180 GHz, orientation selectivity is observed and the modulation frequency was found in good agreement with theoretical predictions, which take into account the relative orientation of the radicals from X-ray data.     

Distance between a native cofactor and a spin label in the reaction centre of Rhodobacter sphaeroides by a two-frequency pulsed electron paramagnetic resonance method and molecular dynamics simulations

The distance between the paramagnetic state of a native cofactor and a spin label is measured in the photosynthetic reaction centre from the bacterium Rhodobacter sphaeroides R26. A two-frequency pulsed electron paramagnetic resonance method [double-electron-electron spin resonance (DEER)] is used. A distance of 3.05 nm between the semiquinone anion state of the primary acceptor (Q(A)) and the spin label at the native cysteine at position 156 in the H-subunit is found. Molecular-dynamics (MD) simulations are performed to interpret the distance. A 6 ns run comprising the entire RC protein yields a distance distribution that is close to the experimental one. The average distance found by the MD simulation is smaller than the distance obtained by DEER by at least 0.2 nm. To better represent the experiments performed at low temperature (60K), a MD method to mimic the freezing-in of the room-temperature conformations is introduced. Both MD methods yield similar distances, but the second method has a trend towards a wider distance distribution.     

DeerAnalysis2006—a comprehensive software package for analyzing pulsed ELDOR data

Pulsed electron-electron double resonance techniques such as the four-pulse double electron-electron resonance experiment measure a dipolar evolution function of the sample. For a sample consisting of spin-carrying nanoobjects, this function is the product of a form factor, corresponding to the internal structure of the nanoobject, and a background factor, corresponding to the distribution of nanoobjects in space. The form factor contains information on the spin-to-spin distance distribution within the nanoobject and on the average number of spins per nanoobject, while the background factor depends on constraints, such as a confinement of the nanoobjects to a two-dimensional layer. Separation of the dipolar evolution function into these two contributions and extraction of the spin-to-spin distance distribution require numerically stable mathematical algorithms that can handle data for different classes of samples, e.g., spin-labelled biomacromolecules and synthetic materials. Furthermore, experimental imperfections such as the limited excitation bandwidth of microwave pulses need to be considered. The software package DeerAnalysis2006 provides access to a comprehensive set of tools for such data analysis within a common user interface. This interface allows for several tests of the reliability and precision of the extracted information. User-supplied models for the spin-to-spin distance distribution within a certain class of nanoobjects can be added to an existing library and be fitted with a universal algorithm.     

A short note on orientation selection in the DEER experiments on a native cofactor and a spin label in the reaction center ofRhodobacter sphaeroides

The analysis of the two-frequency pulsed electron paramagnetic resonance (EPR) (double electron-electron spin resonance, DEER) investigation on the coupling between the semiquinone anion state of the primary acceptor (Q_A) and the spin label at the cysteine 156 in the H-subunit in the photosynthetic reaction center (RC) fromRhodobacter sphaerodes (R26) (I. V. Borovykh, S. Ceola, P. Gajula, P. Gast, H. J. Steinhoff, M. Huber: J. Magn. Reson. 180, 178–185, 2006) is reinvestigated to include orientation selection. The combination of the EPR properties of the two radicals and the pump and observer frequencies suggests that such an effect could play a role even at the X-band (9 GHz) EPR fields and frequencies employed. The magnitude of the effect is estimated from the structures obtained from the molecular-dynamics (MD) simulations from the previous study: the distance change is small (around 2%) and the distance of 3.05 nm obtained then is possibly underestimated by 0.06 nm. Thus, the difference of at least 0.2 nm between the measured distance and the average distance of 2.8 nm found by the MD simulation remains, suggesting a significant difference between the measurement and the X-ray structure of the RC, as discussed previously.     

A pulsed EPR method to determine distances between paramagnetic centers with strong spectral anisotropy and radicals: The dead-time free RIDME sequence

Methods to determine distances between paramagnetic metal centers and radicals are scarce. This is unfortunate because paramagnetic metal centers are frequent in biological systems and so far have not been employed much as distance markers. Successful pulse sequences that directly target the dipolar interactions cannot be applied to paramagnetic metal centers with fast relaxation rates and large g-anisotropy, if no echos can be detected and the excitation bandwidth is not sufficient to cover a sufficiently large part of the spectrum. The RIDME method Kulik et al. (2002) [20] circumvents this problem by making use of the T₁-induced spin-flip of the transition-metal ion. Designed to measure distance between such a fast relaxing metal center and a radical, it suffers from a dead time problem. We show that this is severe because the anisotropy of the metal center broadens the dipolar curves, which therefore, only can be analyzed if the full curve is known. Here, we introduce five-pulse RIDME (5p-RIDME) that is intrinsically dead-time free. Proper functioning of the sequence is demonstrated on a nitroxide biradical. The distance between a low-spin Fe(III) center and a spin label in spin-labeled cytochrome f shows the complete dipolar trace of a transition-metal ion center and a spin label, yielding the distance expected from the structure.     

Exchange and dipolar interactions in solid organic radicals

Investigations of the E.S.R. linewidths of exchange narrowed lines from solid organic free radicals allow one to obtain information about both the exchange interaction and the electron-electron dipolar interaction in some instances. In cases in which the dipolar interaction is incompletely averaged by spin exchange one can use the frequency dependence of the E.S.R. linewidth to determine the magnitudes of the exchange frequency and the dipolar interaction. This paper reports results from a study of the frequency and temperature dependence of the linewidth of four solid organic radicals. The dipole-dipole interaction of the two nitroxide radicals included in this study was found to be temperature dependent, indicating changes in intermolecular spin separation with temperature. The exchange energy of these two radicals appeared to vary exponentially with the separation of the spins through most of the temperature interval which was investigated.     

HYSCORE and DEER with an upgraded 95GHz pulse EPR spectrometer

The set-up of a new microwave bridge for a 95 GHz pulse EPR spectrometer is described. The virtues of the bridge are its simple and flexible design and its relatively high output power (0.7 W) that generates pi pulses of 25 ns and a microwave field, B(1)=0.71 mT. Such a high B(1) enhances considerably the sensitivity of high field double electron-electron resonance (DEER) measurements for distance determination, as we demonstrate on a nitroxide biradical with an interspin distance of 3.6 nm. Moreover, it allowed us to carry out HYSCORE (hyperfine sublevel-correlation) experiments at 95 GHz, observing nuclear modulation frequencies of 14N and 17O as high as 40 MHz. This opens a new window for the observation of relatively large hyperfine couplings, yet not resolved in the EPR spectrum, that are difficult to observe with HYSCORE carried out at conventional X-band frequencies. The correlations provided by the HYSCORE spectra are most important for signal assignment, and the improved resolution due to the two dimensional character of the experiment provides 14N quadrupolar splittings.     

Membrane-peptide interaction studied by PELDOR and CW ESR: Peptide conformations and cholesterol effect on the spatial peptide distribution in the membrane

Pulsed electron-electron double resonance (PELDOR) combined with continuous-wave electron paramagnetic resonance was used to study inter- and intramolecular dipole-dipole interactions between spin labels for spin-labeled analogs of trichogin GA IV bound to multilamellar membranes of egg L-α-phosphatidylcholine (ePC) and in ePC membranes containing cholesterol. All samples were frozen to 77 K. For mono-labeled peptide concentrations in lipid over the range between 0.5 to 2.2 mol%, it is shown that in these membranes trichogin molecules are distributed homogeneously and are likely to be located on or near the inner and outer membrane surfaces. Addition of cholesterol to a final concentration of 16.5 mol% leads to an increase of the local concentration of trichogin molecules in the membranes. For the double-labeled trichogin, a distribution of the intramolecular distance between the two spin labels was observed. The distribution function is characterized by two main maxima located at distances of 1.3 and 1.8 nm. The distance of 1.3 nm is close to that expected for the α-helix structure of the peptide chain. The distance of 1.8 nm corresponds to a mixed structure in which a 3₁₀-helix is combined with a set of even more elongated conformations.     

Nonselective excitation of pulsed ELDOR using multi-frequency microwaves

The use of a polychromatic microwave pulse to expand the pumping bandwidth in pulsed electron-electron double resonance (PELDOR) was investigated. The pumping pulse was applied in resonance with the broad (～100 mT) electron paramagnetic resonance (EPR) signal of the manganese cluster of photosystem II in the S2 state. The observation pulses were in resonance with the narrow EPR signal of the tyrosine radical, YD·. It was found that in the case of the polychromatic pumping pulse containing five harmonics with the microwave frequencies between 8.5 and 10.5 GHz the PELDOR effect corresponding to the dipole interaction between the Mn cluster and YD· was about 2.9 times larger than that achieved with a monochromatic pulse. In addition to the dipolar modulation, the nuclear modulation effects were observed. The effects could be suppressed by averaging the PELDOR trace over the time interval between the observation microwave pulses. The polychromatic excitation technique described will be useful for improving the PELDOR sensitivity in the measurements of long distances in biological samples, where the pair consists of a radical with a narrow EPR spectrum and slow phase relaxation, and a metal center that has a broad EPR spectrum and a short phase relaxation time.