A multi-frequency pulse EPR and ENDOR approach to study strongly coupled nuclei in frozen solutions of high-spin ferric heme proteins.

In spite of the tremendous progress in the field of pulse electron paramagnetic resonance (EPR) in recent years, these techniques have been scarcely used to investigate high-spin (HS) ferric heme proteins. Several technical and spin-system-specific reasons can be identified for this. Additional problems arise when no single crystals of the heme protein are available. In this work, we use the example of a frozen solution of aquometmyoglobin (metMb) to show how a multi-frequency pulse EPR approach can overcome these problems. In particular, the performance of the following pulse EPR techniques are tested: Davies electron nuclear double resonance (ENDOR), hyperfine correlated ENDOR (HYEND), electron-electron double resonance (ELDOR)-detected NMR, and several variants of hyperfine sublevel correlation (HYSCORE) spectroscopy including matched and SMART HYSCORE. The pulse EPR experiments are performed at X-, Q- and W-band microwave frequencies. The advantages and drawbacks of the different methods are discussed in relation to the nuclear interaction that they intend to reveal. The analysis of the spectra is supported by several simulation procedures, which are discussed. This work focuses on the analysis of the hyperfine and nuclear-quadrupole tensors of the strongly coupled nuclei of the first coordination sphere, namely, the directly coordinating heme and histidine nitrogens and the 17O nucleus of the distal water ligand. For the latter, 17O-isotope labeling was used. The accuracy of our results and the spectral resolution are compared in detail to an earlier single-crystal continuous-wave ENDOR study on metMb, and it will be shown how additional information can be obtained from the multi-frequency approach. The current work is therefore prone to become a template for future EPR/ENDOR investigations of HS ferric heme proteins for which no single crystals are available.

Neuroglobin and cytoglobin as potential enzyme or substrate.

The possible enzymatic activities of neuro- and cytoglobin as well as their potential function as substrates in enzymatic reactions were studied. Neuro- and cytoglobin are found to show no appreciable superoxide dismutase, catalase, and peroxidase activities. However, the internal disulfide bond (CD7-D5) of human neuroglobin can be reduced by thioredoxin reductase. Furthermore, our in vivo and in vitro studies show that Escherichia coli cells contain an enzymatic reducing system that keeps the heme iron atom of neuroglobin in the Fe(2+) form in the presence of dioxygen despite the high autoxidation rate of the molecule. This reducing system needs a low-molecular-weight compound as co-factor. In vitro tests show that both NADH and NADPH can play this role. Furthermore, the reducing system is not specific for neuroglobin but allows the reduction of the ferric forms of other globins such as cytoglobin and myoglobin. A similar reducing system is present in eukaryotic tissue protein extracts.

Tracing the structure-function relationship of neuroglobin and cytoglobin using resonance Raman and electron paramagnetic resonance spectroscopy.

The physiological role of neuroglobin and cytoglobin, two vertebrate globins discovered in the last 5 years, is not yet clearly understood. In this work, we review the structural information on these globins and its implication on the possible protein function, obtained by electron paramagnetic resonance and resonance Raman spectroscopy. All studies reveal a high flexibility in the heme-pocket region of neuroglobin. Together with the observation that the distal ligand of the heme iron is the endogenous E7-histidine in both the ferric and ferrous form of neuroglobin and cytoglobin, the flexibility of the heme environment in neuroglobin will play a crucial role in the globins' ability to bind and stabilize exogenous ligands.