Dioxygen and glucose force motion of the electron-transfer switch in the iron(III) flavohemoglobin-type nitric oxide dioxygenase

When coupled to reductases, globins can function as nitric oxide dioxygenases (NODs) (EC 1.14.12.17) that attenuate NO toxicity and signaling in a variety of organisms and tissues [[1], [2], [3], [4], [5], [6], [7]]. Greater knowledge of the NOD mechanism promises insights for the development of therapeutic NOD effectors, clues to understanding the diverse functions of globins, better criteria for the annotation of globin genes, and a better understanding of hemoglobin (Hb) evolution.

The ability of O2 to bind the Fe(III)-heme iron is integral to the recently proposed allosteric NOD mechanism of flavohemoglobin (flavoHb) and other globins with tightly coupled reductases [7,8]. Moreover, CO and NO are thought to inhibit the NOD mechanism by competing with O2 for the Fe(III)-heme. O2, CO, and NO equilibrium dissociation constants are estimated to be 120 μM, 1 μM, and 1.2 μM, respectively, from steady-state kinetic analysis of the Escherichia coli flavoHb-NOD [8,9]. However, in the resting state, the Fe(III)-flavoHb binds NO with a large equilibrium dissociation constant estimated at 90 to 330 μM [[8], [9], [10]]. The lower than expected NO affinity measured in the resting state is explained by the opening of the NO gate and short tunnel allowing greater NO access, association, and a higher NO affinity during turnover [7]. The opening NO gate may also explain the potent, reversible, and competitive CO inhibition observed with various NODs [7,9,[11], [12], [13]].

While the idea of O2 and CO binding to ferric heme is unorthodox, binding can be reasonably expected. Shafizadeh, Soep, and coworkers recently reported O2 and CO affinities and binding energies for bare ferric heme using a gas phase cryogenic ion trap and electronic spectroscopy and modeled the bonding with hybrid density functional theory (DFT) [[14], [15], [16], [17]]. Dillinger et al. confirmed O2 and CO binding to ferric heme and further characterized the complexes using gas phase cryo infrared photon dissociation spectroscopy in conjunction with hybrid DFT [18]. Moreover, there has long been suggestive indirect evidence for transient Fe(III)CO formation in CO inhibition of the non-ferrous catalase mechanism [19,20] and in CO reduction of the ferric HbA and myoglobin (Mb) [21]. Indeed, the underlying concepts describing weak binding of CO to thiolate and imidazole-ligated ferric heme in proteins were recently expounded [22]. The binding of NO to Fe(III)-porphyrins [23], the ferric flavoHb-NOD [[8], [9], [10]], Fe(III)-globins, and other hemoproteins is well-understood and was recently reviewed [24,25].

The essential role of Fe(III)O2 complex formation in the proposed allosteric flavoHb-NOD mechanism is illustrated in Fig. 1 and briefly described here. The mechanism and supporting data has been elaborated upon in recent publications [7,8]. Fundamentals of the mechanism, including Fe(III)O2 complex formation, have been suggested to apply to all globins that have NOD roles with coupled reductases including Cygb, Ngb, Mb, and the HbA α and β chains [7].

In the Escherichia coli flavoHb X-ray crystal structure PDB:1GVH [26], the distal E11 leucine CD1 methyl group carbon atom shows a >6% overlap of its van der Waals radius with that of the ferric iron (Fig. 1, left side). In the allosteric NOD mechanism, O2 enters through a 20 Å long hydrophobic tunnel, displaces LeuE11, hydrogen bonds to the TyrB10/GlnE7 pair, and electrostatically and magnetically interacts with Fe(III). In so doing, O2 leverages upon the Leu E11 side chain, shifts and torsions the E-helix causing 1) the G8-E15 gate forming a short tunnel for NO migration to open, 2) the CD loop to furl, and 3) the F helix to acquire torsion and to strain the HisF8 imidazole N bond to the heme iron. CD loop furling and D helix formation concomitantly 4) releases the force of the D5 glutamine acting directly on FAD and indirectly on the electron-transfer (ET) switch structure and 5) expels the NO3− anion docked at the positive pole of the E-helix dipole. The substate conformation is ‘trigger-ready’. ET is then triggered by NO collision, and an NO and O2 spin chain interaction, that causes ferric heme spin crossover, heme doming, and accompanying movement of the iron, F8 histidine, F-helix, and F7 lysine in the ET switch or gate, and ET. With ET, the transient Fe(III)O2 complex is reduced to Fe(III)O2− which then reacts rapidly with NO through an exergonic radical-radical coupling reaction forming a transient Fe(III)OONO− intermediate which isomerizes to NO3− and simultaneously displaces LeuE11 and GlnE7, unfurls the CD loop, and unwinds the spring-like D helix to energize subsequent O2-triggered motions. NO3− migrates to and binds to the anion hole at the E-D junction stabilizing the unfurled off state, and LeuE11 returns to the ferric iron. In the model, NADH transfers a hydride at the FAD isoalloxazine ring face in the more accessible furled and open state, and the reduced FADH2, and the semiquinone FADH, transfer an electron when ET is switched on. Elementary rate constants have been measured or derived for the individual steps encompassed by the two reaction arrows connecting the off (left side) and on (right side) ET switch conformational substates [7,8].

Electronic Stark-effect spectroscopy utilizing high energy external electric fields [27,28] provides a sensitive and simple method for measuring and characterizing transition dipole moments in molecules including heme [29] and FAD [30,31]. Transition dipole moments can also detect changes in internal electric fields in proteins [27,32], for example due to motions or ligand binding. In the current study, high energy external electric fields are not applied. We use internal Stark-effect spectroscopy to interrogate the proposed flavoHb-NOD mechanism and O2 binding by ferric globins while applying knowledge of heme and FAD transition dipoles. Effects of O2 and other agents on the flavoHb ferric heme [33] and FAD transition dipole moments and spectra can be predicted, interpreted, and calculated, given high-resolution structures and the sum of the vectorial interactions of neighboring dipole moments and the internally-generated electric field. In theory, if we know electric the field tensor (F→) produced internally through interacting transition and permanent dipole moments, μ→t and μ→p, and the change in transition dipole polarizability (∆α), we can calculate the change in the energy potential (U) and corresponding shift in the absorbance frequency (∆ν) (Eq. (1)) and wavelength. We can also calculate the change in absorbance at a given frequency (∆Aν) by applying Liptay's formalism for electrochromism and solvatochromism, aka Stark effects [34](Eq. (2)). Eq. (2) contains factors Ax, Bx, and Cx which provide information about transition moment polarizability and hyperpolarizability, ∆α and μ→t, and the difference transition dipole moment, ∆μ→t, respectively, as elaborated elsewhere [27,30]. Planck's constant (h) and the speed of light (c) are also part of the equation. To calculate the electric field tensors (F→), we also need to determine electrostatic field potential differences (∆V→) generated by the transition dipole moment and permanent dipole moments at their distances (R) (Eq. (3)).∆UΔν=−∆μ→t∙F→–0.5F→∙∆α∙F→∆Aν=AxAν+Bx15hcνdAννdν+Cx30h2c2νd2Aννdν2·F→2F→=∆V→/R

Many of the values required for quantitating Stark effects in flavoHb, and other globins, using Eqs. (1), (2) are presently unknown. Nevertheless, we can initiate a quantitative analysis of observed Stark effects by identifying and characterizing μ→t and μ→p interactions and by estimating changes in the internal matrix F→ using high-resolution flavoHb and Mb structures. With heme, interpretations of electronic spectra can be complicated by structure- and ligand-induced spin crossovers [35]. Reduction states of FAD and heme also largely determine transition dipoles, but those are more readily controlled for in measurements of the oxidized state of flavoHb and other globins with no reducing agent added. Notwithstanding analytical challenges and limitations, Stark-effect theory and methods provide a powerful and fitting tool for examining low affinity ligand binding and motions in flavoHb and other globins. In addition, the development of methods for calculating matrix F→ values in proteins, such as the globins during ligand binding, may help inform the correction factor, f, required for external field Stark-effect spectroscopy [36] and may find other applications.

We now report on our analysis of the Stark effects of O2 on the ferric heme and FAD spectra of flavoHb. FlavoHb variants previously shown to affect O2 binding and the enzyme mechanism alter the O2-dependent Stark effects. O2 also showed Stark effects on metMb and metHbA in apparent competition with H2O binding. We compare the O2 binding energy reported for free ferric heme with that calculated for flavoHb while also considering the contributions of the ferric heme spin state and hydrogen bonding by TyrB10/GlnE7. Comparative binding energies for CO and NO are also calculated. We also discuss a potentially unique mechanism for O2 binding by metMb and metHb.

In the course of our investigations, we observed glucose to cause smaller Stark effects on FAD and cause shifts in the heme spin state. Glucose was found to also similarly affect the heme spectra of metMb and metHb. We describe a putative effector site for glucose or glucose-6-phosphate (G6P) binding in flavoHb and Mb and discuss the potential physiological relevance.

In summary, we discuss the O2-induced movements controlling the proposed ET switch formed by LysF7 and a water molecule (Fig. 1), the dampening of those movements by glucose or G6P binding, and the possible regulatory role in the NOD mechanism.

留言 (0)

沒有登入
gif