Electronic configuration assignment and the importance of low-lying excited states in high-spin imidazole-ligated iron(II) porphyrinates.

The synthesis and characterization of six new high-spin deoxymyoglobin models (imidazole(tetraarylporphyrinato)iron(II)) are described. These have been intensively studied by temperature-dependent Mossbauer spectroscopy from 295 to 4.2 K. All complexes show a strong temperature dependence for the quadrupole splitting consistent with low-lying excited states of the same or lower multiplicity. An analysis of the data obtained in applied magnetic fields leads to the assignment of the sign of the quadrupole splitting. All model compounds as well as those of deoxymyoglobin and deoxyhemoglobin, previously studied, have a negative sign for the quadrupole splitting. Although not previously predicted, this experimental observation leads to the assignment of the ground-state electronic configuration for all high-spin imidazole-ligated iron(II) porphyrinates as (d(xz)())(2)(d(yz)())(1)(d(xy)())(1)(d(z)()()2)(1)(d(x)()()2(-)(y)()()2)(1). This is a distinctly different ground-state electronic configuration from other high-spin iron(II) porphyrinates; differences in structural details for the two classes of high-spin complexes are also discussed. The apparent anomaly of differing signs for the zero-field splitting constant between previously studied model complexes and the heme proteins is addressed; the difference appears to result from the fact that the assumptions used in the spin Hamiltonian approach that has been applied to these complexes are not adequately satisfied. Structures of four of the new five-coordinate species have been determined. Core conformations in these derivatives show variation, but these and previously studied compounds reveal a limited number of conformational patterns. The bond lengths and other geometrical parameters such as porphyrin core size and iron out-of-plane displacement support a high-spin state assignment for the iron(II).

Leg ischemia: assessment with MR angiography and spectroscopy.

PURPOSE: To prospectively determine reproducibility of magnetic resonance (MR) angiography and MR spectroscopy of deoxymyoglobin in assessment of collateral vessels and tissue perfusion in patients with critical limb ischemia (CLI) and to follow changes in patients undergoing intramuscular vascular endothelial growth factor (pVEGF)-C gene therapy, percutaneous transluminal angioplasty, supervised exercise training, or no therapy. MATERIALS AND METHODS: Study and gene therapy protocols were approved, and all patients gave written informed consent. To determine repeatability and reproducibility, seven patients underwent MR angiography and five underwent MR spectroscopy. The techniques were used to judge disease progress in 12 other patients with or without therapy: MR angiography to help determine change in visualization of collateral vessels and MR spectroscopy to help assess change in perfusion at proximal and distal calf levels. MR angiographic results were subjectively analyzed by three blinded readers. Intraobserver variability was expressed as 95% confidence interval (CI) (n=7); interobserver variability, as kappa statistic (n=15). Reexamination variability of MR spectroscopy was given as 95% CI for subsequent recovery times, and correlation with disease extent was calculated with Kendall taub rank correlation. Fisher-Yates test was used to correlate changes with pressure measurements and clinical course. RESULTS: Intraobserver and interobserver concordance was sensitive for detection of collateral vessels. Intraobserver agreement was 85.7% (95% CI: 42.1%, 99.6%). Interobserver agreement was high for small collateral vessels (kappa=0.74, P <.001) and fair for large collateral vessels (kappa=0.36, P=.002). MR spectroscopy was reproducible (95% CI: +/-26 seconds for proximal, +/-21 seconds for distal) and showed a correlation with disease extent (proximal calf, taub=0.84, P <.001; distal calf, taub=0.68, P=.04). Small collateral vessels increased over time (P=.04) but did not correlate with pressure measurements and clinical course. Recovery time correlated with clinical course (proximal calf, P=.03; distal calf, P=.005). CONCLUSION: MR angiography and MR spectroscopy of deoxymyoglobin can help document changes in visualization of collateral vessels and tissue perfusion in patients with CLI.

Bonding in HNO-myoglobin as characterized by X-ray absorption and resonance raman spectroscopies.

The EXAFS and resonance Raman spectra on the HNO-myoglobin adduct, 1, are consistent with the presence of HNO bound to a heme center. The three-dimensional structure about the heme center of 1 obtained from multiple-scattering (MS) analysis of the EXAFS of the heme protein yielded an Fe-N-O bond angle of 131 degrees and an Fe-N bond length of 1.82 A, which compare well with published values for model complexes containing RNO ligands. Resonance Raman spectra identified the nu(N=O) stretch at 1385 cm-1 (confirmed by 15N labeling), which corresponds well with those reported for small molecule HNO complexes. The wavelength of the nu(Fe-N) at 636 cm-1 of 1 is significantly higher than those of MbIINO and MbIIINO (554 and 595 cm-1, respectively). The XAFS, XANES, and resonance Raman data are all consistent with the structure deduced from the NMR experiments, providing more detail on the bonding between HNO and the metal center.

Design of a five-coordinate heme protein maquette: a spectroscopic model of deoxymyoglobin.

The substitution of 1-methyl-l-histidine for the histidine heme ligands in a de novo designed four-alpha-helix bundle scaffold results in conversion of a six-coordinate cytochrome maquette into a self-assembled five-coordinate mono-(1-methyl-histidine)-ligated heme as an initial maquette for the dioxygen carrier protein myoglobin. UV-vis, magnetic circular dichroism, and resonance Raman spectroscopies demonstrate the presence of five-coordinate mono-(1-methyl-histidine) ligated ferrous heme spectroscopically similar to deoxymyoglobin. Thermodynamic analysis of the ferric and ferrous heme dissociation constants indicates greater destabilization of the ferric state than the ferrous state. The ferrous heme protein reacts with carbon monoxide to form a (1-methyl-histidine)-Fe(II)(heme)-CO complex; however, reaction with dioxygen leads to autoxidation and ferric heme dissociation. These results indicate that negative protein design can be used to generate a five-coordinate heme within a maquette scaffold.

Ligand binding properties of myoglobin reconstituted with iron porphycene: unusual O2 binding selectivity against CO binding.

Sperm whale myoglobin, an oxygen storage hemoprotein, was successfully reconstituted with the iron porphycene having two propionates, 2,7-diethyl-3,6,12,17-tetramethyl-13,16-bis(carboxyethyl)porphycenatoiron. The physicochemical properties and ligand bindings of the reconstituted myoglobin were investigated. The ferric reconstituted myoglobin shows the remarkable stability against acid denaturation and only a low-spin characteristic in its EPR spectrum. The Fe(III)/Fe(II) redox potential (-190 mV vs NHE) determined by the spectroelectrochemical measurements was much lower than that of the wild-type. These results can be attributed to the strong coordination of His93 to the porphycene iron, which is induced by the nature of the porphycene ring symmetry. The O2 affinity of the ferrous reconstituted myoglobin is 2600-fold higher than that of the wild-type, mainly due to the decrease in the O2 dissociation rate, whereas the CO affinity is not so significantly enhanced. As a result, the O2 affinity of the reconstituted myoglobin exceeds its CO affinity (M' = K(CO)/K(O2) < 1). The ligand binding studies on H64A mutants support the fact that the slow O2 dissociation of the reconstituted myoglobin is primarily caused by the stabilization of the Fe-O2 sigma-bonding. The IR spectra for the carbon monoxide (CO) complex of the reconstituted myoglobin suggest several structural and/or electrostatic conformations of the Fe-C-O bond, but this is not directly correlated with the CO dissociation rate. The high O2 affinity and the unique characteristics of the myoglobin with the iron porphycene indicate that reconstitution with a synthesized heme is a useful method not only to understand the physiological function of myoglobin but also to create a tailor-made function on the protein.

Effects of systematic peripheral group deuteration on the low-frequency resonance Raman spectra of myoglobin derivatives.

Resonance Raman spectra are reported for a series of systematically deuterated analogues of myoglobin in its deoxy state as well as for its CO and O(2) adducts. Specifically, the myoglobin samples studied are those that have been reconstituted with deuterated protoheme analogues. These include the methine deuterated, protoheme-d4; analogue bearing C(2)H(3) groups at the 1, 3, 5, and 8 positions, protoheme-d12; the species bearing C(2)H(3) groups at the 1 and 3 positions only, 1,3-protoheme-d6; and the species bearing C(2)H(3) groups at the 5 and 8 positions only, 5,8-protoheme-d6. While the results are generally consistent with previously reported data for synthetic metalloporphyrin models and previous studies of labeled heme proteins, the high-quality low-frequency RR data reported here reveal several important aspects of these low-frequency modes. Of special interest is the fact that, using the two d6-protoheme analogues, it is shown that certain modes are apparently localized on particular pyrrole rings, while others are localized on different rings; i.e., several of these low-frequency modes are localized on one side of the heme.

Different relaxations in myoglobin after photolysis.

To clarify the interplay of kinetic hole-burning (KHB), structural relaxation, and ligand migration in myoglobin (Mb), we measured time-resolved absorption spectra in the Soret region after photolysis of carbon monoxide Mb (MbCO) in the temperature interval 120-260 K and in the time window 350 ns to 200 ms. The spectral contributions of both photolyzed (Mb*) and liganded Mb (MbCO) have been analyzed by taking into account homogeneous bandwidth, coupling to vibrational modes, and static conformational heterogeneity. We succeeded in separating the "time-dependent" spectral changes, and this work provides possibilities to identify the events in the process of ligand rebinding. KHB is dominant at T <190 K in both the Mb* and the MbCO components. For MbCO, conformational substates interconversion at higher temperatures tends to average out the KHB effect. At 230-260 K, whereas almost no shift is observed in the MbCO spectrum, a shift of the order of approximately 80 cm(-1) is observed in Mb*. We attribute this shift to protein relaxation coupled to ligand migration. The time dependence of the Mb* spectral shift is interpreted with a model that enables us to calculate the highly nonexponential relaxation kinetics. Fits of stretched exponentials to this kinetics yield Kohlrausch parameter values of 0.25, confirming the analogy between proteins and glasses.

Human heme oxygenase oxidation of 5- and 15-phenylhemes.

Human heme oxygenase-1 (hHO-1) catalyzes the O2-dependent oxidation of heme to biliverdin, CO, and free iron. Previous work indicated that electrophilic addition of the terminal oxygen of the ferric hydroperoxo complex to the alpha-meso-carbon gives 5-hydroxyheme. Earlier efforts to block this reaction with a 5-methyl substituent failed, as the reaction still gave biliverdin IXalpha. Surprisingly, a 15-methyl substituent caused exclusive cleavage at the gamma-meso-rather than at the normal, unsubstituted alpha-meso-carbon. No CO was formed in these reactions, but the fragment cleaved from the porphyrin eluded identification. We report here that hHO-1 cleaves 5-phenylheme to biliverdin IXalpha and oxidizes 15-phenylheme at the alpha-meso position to give 10-phenylbiliverdin IXalpha. The fragment extruded in the oxidation of 5-phenylheme is benzoic acid, one oxygen of which comes from O2 and the other from water. The 2.29- and 2.11-A crystal structures of the hHO-1 complexes with 1- and 15-phenylheme, respectively, show clear electron density for both the 5- and 15-phenyl rings in both molecules of the asymmetric unit. The overall structure of 15-phenylheme-hHO-1 is similar to that of heme-hHO-1 except for small changes in distal residues 141-150 and in the proximal Lys18 and Lys22. In the 5-phenylheme-hHO-1 structure, the phenyl-substituted heme occupies the same position as heme in the heme-HO-1 complex but the 5-phenyl substituent disrupts the rigid hydrophobic wall of residues Met34, Phe214, and residues 26-42 near the alpha-meso carbon. The results provide independent support for an electrophilic oxidation mechanism and support a role for stereochemical control of the reaction regiospecificity.

Mechanism of nitrosylmyoglobin autoxidation: temperature and oxygen pressure effects on the two consecutive reactions.

As shown by singular value decomposition and global analysis of the absorption spectra, oxidation of nitrosylmyoglobin, MbFe(II)NO, by oxygen occurs in two consecutive (pseudo) first-order reactions in aqueous air- saturated solutions at physiological conditions (pH 7.0, I=0.16 m (NaCl)). Both reaction steps have a large temperature dependence with the following activation parameters: DeltaS++(1) = 121+/-7 and DeltaS++(1) = 23+/-29; and DeltaS++(2) = 88+/-14 kJ mol(-1) and DeltaS++(2)-63+/-51 J(-1) K(-1) mol(-1) at 25 degrees C for the first and second step, respectively. At physiological temperature, the initial reaction is faster, while at lower temperatures, the first reaction is slower and rate-determining. The rate of the first reaction is linearly dependent on oxygen pressure at lower pressures, while for oxygen pressures above atmospheric, the rate exhibits saturation behaviour. The second reaction is independent of oxygen pressure. The rate of the second reaction increases when oxymyoglobin is added. In contrast, the rate of the first reaction is independent of the presence of oxymyoglobin. The observed kinetics are in agreement with a reaction mechanism in which the nitric oxide that is initially bound to the Fe(II) centre of myoglobin is displaced by oxygen in a reversible ligand-exchange reaction prior to an irreversible electron transfer. The ligand-exchange process is dissociative in nature and depends bond breaking, and nitric oxide is suggested to be trapped in a protein cavity. The absorption spectrum of the intermediate, as resolved from the global analysis, is in agreement with a peroxynitrite complex, and the initial process must involve partial electron transfer.

NO-bound myoglobin: structural diversity and dynamics of the NO ligand.

We used femtosecond infrared polarization spectroscopy and density functional theory in a study on the key signaling molecule nitric oxide (NO) bound to myoglobin. Our results show that after photolysis, a substantial fraction of NO recombines within the first few picoseconds. We discovered that the diatomic ligand is severely tilted in the protein and present evidence that the Fe-NO moiety can sample a wide range of off-axis tilting and bending conformations.

Efficient trapping of HNO by deoxymyoglobin.

Nitrosyl hydride, HNO, also commonly termed nitroxyl, is a transient species that has been implicated in the biological activity of nitric oxide, NO. Herein, we report the first generation of a stable HNO-metal complex by direct trapping of free HNO. Deoxymyoglobin (Mb-Fe(II)) rapidly reacts with HNO produced from the decomposition of methylsulfonylhydroxylamine (MSHA) or Angeli's salt (AS) in aqueous solutions from pH 7 to pH 10, forming an adduct, Mb-HNO. The unique 1H NMR signal of the Fe-bound HNO at 14.8 ppm allows definitive proof of its formation. The generation of Mb-HNO and quantification of various myoglobin byproducts were accomplished by correlation of 1H NMR, UV-vis, and EPR spectroscopies. Typically, the maximum Mb-HNO yield obtained is 60-80%; competitive side reactions with byproducts as well as the further reactivity of the Mb-HNO decrease the overall yield. At pH 10, the observed rate of Mb-HNO generation by trapping HNO from MSHA is close to that for MSHA decomposition; kinetic simulations give a lower limit to the bimolecular rate of trapping as 1.4 x 10(4) M(-1) s(-1). The binding of HNO to deoxymyoglobin is rapid and essentially irreversible, which suggests that the biological activity of nitroxyl may be mediated by its reactivity with ferrous heme proteins such as myoglobin and hemoglobin.

Hybridization of modified-heme reconstitution and distal histidine mutation to functionalize sperm whale myoglobin.

To modulate the physiological function of a hemoprotein, most approaches have been demonstrated by site-directed mutagenesis. Replacement of the native heme with an artificial prosthetic group is another way to modify a hemoprotein. However, an alternate method, mutation or heme reconstitution, does not always demonstrate sufficient improvement compared with the native heme enzyme. In the present study, to convert a simple oxygen storage hemoprotein, myoglobin, into an active peroxidase, we applied both methods at the same time. The native heme of myoglobin was replaced with a chemically modified heme 2 having two aromatic rings at the heme-propionate termini. The constructed myoglobins were examined for 2-methoxyphenol (guaiacol) oxidation in the presence of H2O2. Compared with native myoglobin, rMb(H64D.2) showed a 430-fold higher kcat/Km value, which is significantly higher than that of cytochrome c peroxidase and only 3-fold less than that of horseradish peroxidase. In addition, myoglobin-catalyzed degradation of bisphenol A was examined by HPLC analysis. The rMb(H64D.2) showed drastic acceleration (>35-fold) of bisphenol A degradation compared with the native myoglobin. In this system, a highly oxidized heme reactive species is smoothly generated and a substrate is effectively bound in the heme pocket, while native myoglobin only reversibly binds dioxygen. The present results indicate that the combination of a modified-heme reconstitution and an amino acid mutation should offer interesting perspectives toward developing a useful biomolecule catalyst from a hemoprotein.

NMR investigation of the heme electronic structure in deoxymyoglobin possessing a fluorinated heme.

The heme electronic structures of deoxymyoglobins (deoxy-Mbs) reconstituted with 13,17-bis(2-carboxylatoethyl)-3,8-diethyl-2,12,18-trimethyl-7-(trifluoromethyl)porphyrinatoiron(III) (7-PF), 13,17-bis(2-carboxylatoethyl)-3,7-difluoro-2,8,12,18-tetramethylporphyrinatoiron(III) (3,7-DF), and 13,17-bis(2-carboxylatoethyl)-3,8-diethyl-2-fluoro-7,12,18-trimethylporphyrinatoiron(III) (2-MF) have been characterized by (1)H and (19)F NMR. The analysis of heme methyl proton shift patterns of the hemes in their bis-cyano forms demonstrated that, owing to the substitution of a strongly electron-withdrawing perfluoromethyl group, CF(3), to porphyrin, the porphyrin pi-system of 7-PF is more significantly distorted from four-fold symmetry than those of the ring-fluorinated hemes, 3,7-DF and 2-MF. The presence of the heme orientation disorder resulted in the observation of the two well-resolved (19)F signals in the spectra of deoxy-Mbs possessing 7-PF and 2-MF. The (19)F signals of deoxy-Mb possessing 7-PF exhibited a relatively large difference in paramagnetic shift (approximately 30 ppm), despite their small paramagnetic shifts (approximately 30 ppm), supporting the significant contribution of a pi spin delocalization mechanism in this Mb due to the d-electron configuration derived from the (5)E ground state. On the other hand, (19)F signals of deoxy-Mbs with 3,7-DF as well as 2-MF exhibited large paramagnetic shifts (approximately 250 ppm) with a relatively small difference in the paramagnetic shift (approximately 20 ppm), indicating the predominant contribution of spin delocalization, due to a d-electron configuration derived from the (5)B(2) ground state. These results demonstrate for the first time that the relative contributions of the orbital ground states derived from (5)E and (5)B(2) states to the heme electronic structure in deoxy-Mb are affected by the distortion of the porphyrin pi-system exerted by chemical properties of the heme peripheral side-chains.

Residues in the distal heme pocket of neuroglobin. Implications for the multiple ligand binding steps.

Amino acid residues in the ligand binding pocket of human neuroglobin have been identified by site-directed mutagenesis and their properties investigated by resonance Raman and flash photolysis methods. Wild-type neuroglobin has been shown to have six-coordinate heme in both ferric and ferrous states. Substitution of His96 by alanine leads to complete loss of heme, indicating that His96 is the proximal ligand. The resonance Raman spectra of M69L and K67T mutants were similar to those of wild-type (WT) neuroglobin in both ferric and ferrous states. By contrast, H64V was six-coordinate high-spin and five-coordinate high-spin in the ferric and ferrous states, respectively, at acidic pH. The spectra were pH-dependent and six-coordinate with the low-spin component dominating at alkaline pH. In a double mutant H64V/K67T, the high-spin component alone was detected in the both ferric and the ferrous states. This implies that His64 is the endogenous ligand and that Lys67 is situated nearby in the distal pocket. In the ferrous H64V and H64V/K67T mutants, the nu(Fe-His) stretching frequency appears at 221 cm(-1), which is similar to that of deoxymyoglobin. In the ferrous CO-bound state, the nu(Fe-CO) stretching frequency was detected at 521 and 494 cm(-1) in WT, M69L, and K67T, while only the 494 cm(-1) component was detected in the H64V and H64V/K67T mutants. Thus, the 521 cm(-1) component is attributed to the presence of polar His64. The CO binding kinetics were biphasic for WT, H64V, and K67T and monophasic for H64V/K67T. Thus, His64 and Lys67 comprise a unique distal heme pocket in neuroglobin.

Engineering peroxidase activity in myoglobin: the haem cavity structure and peroxide activation in the T67R/S92D mutant and its derivative reconstituted with protohaemin-l-histidine.

Atomic co-ordinates and structure factors for the T67R/S92D metMbCN mutant have been deposited with the Protein Data Bank, under accession codes 1h1x and r1h1xsf, respectively. Protein engineering and cofactor replacement have been employed as tools to introduce/modulate peroxidase activity in sperm whale Mb (myoglobin). Based on the rationale that haem peroxidase active sites are characterized by specific charged residues, the Mb haem crevice has been modified to host a haem-distalpropionate Arg residue and a proximal Asp, yielding the T67R/S92D Mb mutant. To code extra conformational mobility around the haem, and to increase the peroxidase catalytic efficiency, the T67R/S92D Mb mutant has been subsequently reconstituted with protohaem-L-histidine methyl ester, yielding a stable derivative, T67R/S92D Mb-H. The crystal structure of T67R/S92D cyano-metMb (1.4 A resolution; R factor, 0.12) highlights a regular haem-cyanide binding mode, and the role for the mutated residues in affecting the haem propionates as well as the neighbouring water structure. The conformational disorder of the haem propionate-7 is evidenced by the NMR spectrum of the mutant. Ligand-binding studies show that the iron(III) centres of T67R/S92D Mb, and especially of T67R/S92D Mb-H, exhibit higher affinity for azide and imidazole than wild-type Mb. In addition, both protein derivatives react faster than wild-type Mb with hydrogen peroxide, showing higher peroxidase-like activity towards phenolic substrates. The catalytic efficiency of T67R/S92D Mb-H in these reactions is the highest so far reported for Mb derivatives. A model for the protein-substrate interaction is deduced based on the crystal structure and on the NMR spectra of protein-phenol complexes.

Solution NMR characterization of the electronic structure and magnetic properties of high-spin ferrous heme in deoxy myoglobin from Aplysia limacina.

Solution (1)H NMR has been used to elucidate the magnetic properties and electronic structure of the prosthetic group in high-spin, ferrous deoxy myoglobin from the sea hare Aplysia limacina. A sufficient number of dipolar shifted residue signals were assigned to allow the robust determination of the orientation and anisotropy of the paramagnetic susceptibility tensor, chi. The resulting quantitative description of dipolar shifts allows a determination of the contact shifts for the heme. Chi was found to be axial, with Deltachi(ax) = -2.07 x 10(-8) m(3)/mol, with the major axis tilted (approximately 76 degrees) almost into the heme plane and in the general direction of the orientation of the axial HisF8 imidazole plane which coincides approximately with the beta-,delta-meso axis. The factored contact shifts for the heme are shown to be consistent with the transfer of positive pi spin density into one of the two components of the highest filled pi molecular orbital, 3e(pi), and the transfer of negative pi-spin density, via spin-spin correlation, into the orthogonal excited-state component of the 3e(pi) molecular orbital. The thermal population of the excited state leads to strong deviation from the Curie law for the heme substituents experiencing primarily the negative pi-spin density. The much larger transfer of negative spin density via the spin-paired dpi orbital into the excited state 3e(pi) in high-spin iron(II) than in low-spin iron(III) hemoproteins is attributed to the much stronger correlation exerted by the four unpaired spin on the iron in the former, as compared to the single unpaired spins on iron in the latter.

Myocardial oxygenation and high-energy phosphate levels during KATP channel blockade.

Inhibition of ATP-sensitive K+ (KATP) channel activity has previously been demonstrated to result in coronary vasoconstriction with decreased myocardial blood flow and loss of phosphocreatine (PCr). This study was performed to determine whether the high-energy phosphate abnormality during KATP channel blockade can be ascribed to oxygen insufficiency. Myocardial blood flow and oxygen extraction were measured in open-chest dogs during KATP channel blockade with intracoronary glibenclamide, whereas high-energy phosphates were examined with 31P magnetic resonance spectroscopy (MRS), and myocardial deoxymyoglobin (Mb-delta) was determined with 1H MRS. Glibenclamide resulted in a 20 +/- 8% decrease of myocardial blood flow that was associated with a loss of phosphocreatine (PCr) and accumulation of inorganic phosphate. Mb-delta was undetectable during basal conditions but increased to 58 +/- 5% of total myoglobin during glibenclamide administration. This degree of myoglobin desaturation during glibenclamide was far greater than we previously observed during a similar reduction of blood flow produced by a coronary stenosis (22% of myoglobin deoxygenated during stenosis). The findings suggest that reduction of coronary blood flow with an arterial stenosis was associated with a decrease of myocardial energy demands and that this response to hypoperfusion was inhibited by KATP channel blockade.

The magnetic properties of myoglobin as studied by NMR spectroscopy.

Deoxymyoglobin has been investigated by NMR spectroscopy to determine the magnetic anisotropy through pseudocontact shifts and the total magnetic susceptibility through Evans measurements. The magnetic anisotropy values were found to be Deltachi(ax)=-2.03+/-0.08 x 10(-32) m(3) and Deltachi(rh)=-1.02+/-0.09 x 10(-32) m(3). The negative value of the axial susceptibility anisotropy originates from the z tensor axis lying in the heme plane, unlike all other heme systems investigated so far. This magnetic axis is almost exactly orthogonal to the axial histidine plane. The other two axes lie essentially in the histidine plane, the closest to the heme normal being tilted by about 36 degrees from it, towards pyrrole A on the side of the proximal histidine. From the comparison with cytochrome c' it clearly appears that the position of the one axis lying in the heme plane is related to the axial histidine orientation. Irrespective of the directions, the magnetic anisotropy is smaller than that of the analogous reduced cytochrome c' and of the order of that of low-spin iron(III). The magnetic anisotropy of the system permits the measurement of residual dipolar couplings, which, together with pseudocontact shifts, prove that the solution structure is very similar to that in the crystalline state. Magnetic measurements, at variance with previous data, demonstrate that there is an orbital contribution to the magnetic moment, micro(eff)=5.5 micro(B). Finally, from the magnetic anisotropy data, the hyperfine shifts of iron ligands could be separated in pseudocontact and contact components, and hints are provided to understand the spin-delocalisation mechanism in S=2 systems by keeping in mind the delocalisation patterns in low-spin S=1/2 and high-spin S= 5/2 iron(III) systems.