Aldehyde oxidoreductase from Desulfovibrio gigas: Characterization of the electron transfer chain between Mo and the proximal FeS center by EPR and DFT studies

MARÍA C. GÓMEZ; NICOLÁS I. NEUMAN; PABLO J. GONZÁLEZ; SERGIO D. DALOSTO; ALBERTO C. RIZZI; CARLOS D. BRONDINO

Chascomús

Encuentro; IV LABIC Fourth Latin American Meeting on Biological Inorganic Chemistry V WOQUIBIO Fifth Workshop on Bioinorganic Chemistry; 2014

UBA, UNR, CONICET, AGENCIA, Workshop Argentino de Química Bioinorgánica

Session I - Metalloproteins/Metalloenzymes P20.-Aldehyde oxidoreductase from Desulfovibrio gigas: Characterization of the electron transfer chain between Mo and the proximal FeS center by EPR and DFT studies María C. Gomez1,Nicolas I. Neuman1, Pablo J. Gonzalez1, Sergio D. Dalosto2, Alberto C.Rizzi1, and Carlos D. Brondino1 1Departamento de Física, Facultad de Bioquímica y Ciencias Biológicas, Universidad Nacional del Litoral, Santa Fe, Argentina. 2Instituto de Fisica del Litoral, UNL-Conicet, Santa Fe, Argentina mcgomez@fbcb.unl.edu.ar INTRODUCTION Aldehyde oxidoreductase from Desulfovibrio gigas (DgAOR) is a homodimeric protein belonging to the XO family of mononuclear molybdenum-containing proteins that catalyzes the two-electron oxidation of aldehyde to acid.1,2 The 3D structure of each monomer is organized into two major domains called Mo-domain and FeS-domain, which contains two Fe2-S2 clusters (FeS 1 and FeS 2). The Mo atom in its oxidized form (MoVI) is in a distorted square pyramidal coordination with two S atoms from one pyranopterin, an oxo ligand, and a OH/OH2 molecule (the catalytic labile site) in equatorial positions, and an oxo ligand in the apical position (Figure 1). FeS 1 is closer to the Mo site and is buried inside the protein in a helical domain inaccessible to solvent. FeS 2 and the corresponding domain is of the plant ferredoxin type and is exposed to solvent. FeS 1 and FeS 2 clusters are 15 Å and 24.4 Å away, respectively, from the Mo atom. The metal cofactors are along an electron transfer pathway that mediates electron transfer between substrate and an external electron acceptor, in which the pyranopterin moiety is essential for electron flow between Mo and FeS 1. The redox centers are paramagnetic in certain protein oxidation states and, despite the long both distances and chemical paths, they present weak magnetic couplings produced by dipolar interaction (D) and isotropic exchange J.3,4 We report here EPR studies complemented with first-principles computational simulations of reduced as-purified DgAOR and arsenite-, Figure 1: electron transfer chain between Mo and FeS 1. glycerol-(GOL), and ethylene glycol-(EDO) inhibited DgAOR. We evaluate the exchange coupling constants J associated with the pterin moiety in the different enzyme forms. These results are rationalized on the basis of the EPR properties of the Mo center and on theoretical calculations. EXPERIMENTAL METHODS Sample preparation DgAOR was purified as reported previously.5,6. Arsenite-, EDO-, and GOL- reacted samples for EPR spectroscopy were prepared according to procedures published elsewhere.7,8 EPR spectroscopy X-band CW-EPR spectra of DgAOR were obtained on a Bruker EMX-Plus spectrometer, equipped with an Oxford Instrument helium continuous-flow cryostat and a rectangular cavity with 100 kHz field modulation. EPR spectra were simulated with the EasySpin toolbox based on MATLAB®.9 Computational methods Spin-polarized ground state calculations were performed using the HSE functional,10 with 6- 31G basis set for all atoms except molybdenum, which was treated with an effective core potential as in the basis set LANL2DZ. The model system used for computations is a simplified version of the cofactor, where the pterin and Mo (O2OH) were conserved, while the phosphate moiety was removed and replaced by a methyl group. GOL, EDO, and arsenite were placed according to the crystallographic information available for each case (Protein Data Bank code 3FAH, 3F4C, 3L4P, respectively). Partial optimization of the models was performed, in order to avoid the ligand to move to a position not consistent with the crystallographic data. RESULTS AND DISCUSSION EPR simulations were performed assuming a pair of interacting S=1/2 spins with the Hamiltonian B ( 1 1 2 2 ) 1 ( ) 2 H = μ S ×g + S ×g ×B + S × J I + D ×S r r r r r where 1 = Mo and 2 = FeS 1, and all the symbols have their usual meaning in magnetic resonance. Dipolar interaction was considered under the point dipole approximation. This procedure yielded an exchange coupling constant J = 8 G for reduced as-purified DgAOR, which is increased ~2-3 times in all the inhibited forms of the enzyme. This result indicates that the MoV unpaired spin density is delocalized on the pterin ligand with different extent upon inhibition. Computational methods showed that the spin density is mainly localized on the Mo and on the equatorial O-ligands in as-purified DgAOR, whereas the inhibited samples show a fraction of the MoV spin density delocalized on the pterin moiety. CONCLUSION The increase of the coupling constants J in inhibited DgAOR samples is due to a larger MoV spin delocalization on the pterin ligand, thus allowing a larger overlap between the Mo and the FeS 1 magnetic orbitals. The delocalization of the MoV spin density on the pterin moiety is governed by the presence of the catalytic labile OHx(x=1,2) ligand at the Mo center. The present results suggest that the pterin moiety functions as a tunable ?wire? to facilitate the electron flow Mo ®FeS 1 REFERENCES 1. Brondino C. D. et al., Acc. Chem. Res. (2006) 39:788-796. 2. Brondino C. D. et al., Curr. Opin. Chem. Biol. (2006), 10:109-114. 3. NeumanN. I. et al., J. Physical. Chem. A (2010), 114:13069-13075. 4. González P. J.et al., J. Inorg. Biochem. (2009), 103:1342-1346. 5. Moura, J. J.et al., Biochem. Biophys. Res. Commun. (1976), 72:782-789. 6. Moura, J. J.et al., Biochem. J. (1978), 173:419-425. 7. Thapper, A.et al., J. Biol. Inorg. Chem (2007) 12:353-366. 8. Santos-Silva, T.et al., J. Am. Chem. Soc. (2009), 131:7990-7998. 9. Stoll, S.et al., J. Magn. Reson.(2006), 178:42- 55 10. Heyd, J.et al., J. Chem. Phys (2003), 118:8207?8215. ACKNOWLEDGMENTS We thank FONCYT, CONICET, and CAI+DUNL for financial support. MCG and NIN thank CONICET for a fellowship grant. PJG, SDD, and CDB are members of CONICET-Argentina.