KINETIC ANALYSIS OF E. COLI WrbA REDOX ACTIVITY

I.Kishko^{1, 2}, B. Harish³ , T. Gustavsson⁴, D. Beri³ , J.Carey ^1,3, R. Ettrich ^{1, 2}

¹Institute of Nanobiology and Structural Biology of ISBE, AS CR, 37333 Nove Hrady, Czech Republic, 37333 Nove Hrady, Czech Republic

²University of South Bohemia in České Budějovice, 37333 Nove Hrady, Czech Republic

³Department of Chemistry, Princeton University, Princeton, New Jersey 08544-1009, USA

⁴Biochemistry Department, Lund University, Chemical Center, S-221 00 Lund, Sweden

The flavoprotein WrbA from E. coli was identified as a founding member of a new family of multimeric flavodoxin-like proteins that are implicated in cell protection against oxidative stress. WrbA hereby bridges bacterial flavodoxins and eukaryotic NAD(P)H:quinone oxidoreductases with its three-dimensional structure clearly revealing a close relationship to mammalian NAD(P)H:quinone oxidoreductase, Nqo1 [1]. A closer analysis of apo and holo crystal structures, together with flavodoxin structures, rationalizes functional similarities and differences of the WrbAs relative to flavodoxins, suggesting that WrbAs are not a remote and unusual branch of the flavodoxin family as previously thought but rather a central member with unifying structural features [2].

In this study we try to elucidate the kinetic mechanism of WrbA and its range of substrates, activators, and inhibitors. Assays of WrbA NADH oxidation/quinone reduction activity were carried out. These reactions follow the decrease in NADH absorbance at 340 nm upon oxidation during reduction of ubiquinone-0 or benzoquinone. The reactions are carried out by manual mixing under ambient (i.e., aerobic) conditions since non-enzymatic oxidation is negligible. Standard kinetic approaches to evaluate alternative mechanisms are well-established in studies with the NQO1s, and the classical work of Hosoda et al. (3) provides a model for the comparison with NQO1s. Discrimination among various mechanisms is based on the patterns of substrate and inhibitor concentration dependence of reaction rates. Thus, the concentrations of NADH, NAD, and quinone are varied separately and together to determine patterns of competition that are used to infer which compounds can occupy the active site simultaneously. Ping-pong kinetics have been generally accepted as the mechanism for NQO1s (3, 4), indicating that FAD is reduced in a first step by oxidation of NADH, which then dissociates. In a second step, quinone binds to the site vacated by NAD and is reduced by FADH2. In addition to defining the basic kinetic mechanism, we test WrbA also for a number of the characteristic mechanistic features of NQO1s.

Comparison of the high-resolution E. coli WrbA structure with NQO1 implies significant differences in access to the enzyme active site by the cofactor and the NADH and quinone substrates. Despite these structural differences our results clearly demonstrate the unusual two-plateau behaviour on the substrate concentration-dependence plots for NADH or benzoquinone as described for NQO1 by Hosada et al. [3]. The experiments show that WrbA activity increases upon addition of membrane-mimicking detergents, and they demonstrate the ability of the protein to inactivate reversibly by shifting temperature from 5 to 25 ^oC. The similarity of these properties to NQOs imply a common structural and functional basis of the kinetic mechanism, however, even for the NQOs the reported two-plateau behaviour could not be explained in structural and molecular terms to date.

1. Carey J, Brynda J, Wolfová J, Grandori R, Gustavsson T, Ettrich R, Kutá Smatanová I Protein Science 16: 10. (2007) 2301-2305.

2. Wolfova J, Kuta Smatanova I, Brynda J, Mesters JR, Lapkouski M, Kuty M, Natalello A, Chatterjee N, Chern SY, Ebbel E, Ricci A, Grandori R, Ettrich R, Carey J Biochimica et Biophysica Acta (BBA) - Proteins and Proteomics 1794: 9. (2009) 1288-1298.

3. Hosoda S, Nakamura W, Hayashi K. J. Biol. Chem., 249, (1974), 6416-23.

4. Ernster, L.. Chem. Scripta 27A, (1987) 1- 13.

We gratefully acknowledge support from the Ministry of Education, Youth and Sports of the Czech Republic (MSM6007665808, LC06010), the Academy of Sciences of the Czech Republic (AVOZ60870520), the Grant Agency of the Czech Republic (Nos. P207/10/1934), and joint Czech - US National Science Foundation International Research Cooperation (ME09016 and INT03-09049), Additionally, I.K. was supported by the University of South Bohemia, grant GAJU 170/2010/P.