NADH oxidation and NAD + reduction catalysed by tightly coupled inside-out vesicles from Paracoccus denitrificans Alexander B. Kotlyar and Natalia Borovok Department of Biochemistry, George S. Wise Faculty of Life Sciences, Tel Aviv University, Ramat Aviv, Israel Tightly coupled inside-out vesicles were prepared from Paracoccus denitrificans cells (SPP, sub-Paracoccus particles) and characterized kinetically. The rate of NADH oxidation, catalysed by SPP, increases 6–8 times on addition of gram- icidin. The vesicles are capable of catalysing DlH + -dependent reverse electron transfer from quinol to NAD + . The kinetic parameters of the NADH-oxidase and the reverse electron transfer carried out by membrane-bound P. denitrificans complex I were estimated and compared with those of the mitochondrial enzyme. The data demon- strate that catalytic properties of the dinucleotide-binding site of the bacterial and mitochondrial complex I are almost identical, pointing out similar organization of the site in mammals and P. denitrificans. Inhibition of the bacterial complex I by a specific inhibitor of Q reduction, rotenone, is very different from that of the mitochondrial enzyme. The inhibitor is capable of suppressing the NADH oxidation reaction only at micromolar concentrations, while the activity of mitochondrial enzyme is suppressed by nano- molar concentrations of rotenone. In contrast to the mito- chondrial enzyme, rotenone, even at concentrations as high as 10 l M , does not inhibit the reverse, DlH + -dependent NAD + -reductase reaction on SPP. Keywords: NADH:Q oxidoreductase; complex I; reverse electron transfer; Paracoccus denitrificans; rotenone. NADH-ubiquinone reductase (EC 1.6.5.3), commonly known as complex I, catalyzes electron transfer from NADH to ubiquinone and couples this process to proton translocation across the inner mitochondrial membrane. The isolated mitochondrial enzyme is composed of more than 40 individual subunits [1,2] and contains at least five iron–sulfur centers, a flavine mononucleotide moiety, and tightly bound ubiquinone molecules, which participate in electron transfer from NADH to ubiquinone. The mito- chondrial enzyme is capable of transferring electrons in the opposite direction, from quinol to NAD + . The electron transfer from high potential electron donor (quinol) through complex I to low potential electron acceptor (NAD + ) requires the input of free energy in the form of DlH + .The latter can be produced by mitochondria or coupled submitochondrial particles (SMP) either during ATP hydrolysis or oxidation of succinate. The former, ATP- supported electron transfer reaction is sensitive to uncou- plers and to excess of oligomycin. The latter, succinate energy-supported reaction is sensitive to inhibitors of succinate oxidation, i.e. antimycin, cyanide and malonate, and is insensitive to oligomycin (reviewed in [3]). Both the forward and reverse electron transfer reactions are inhibited by rotenone, a classical inhibitor of complex I [4–8]. The inhibitor blocks the electron transfer between the com- plex I-associated iron–sulfur clusters and the ubiquinone pool [8]. The affinity of rotenone to mitochondrial complex I is extremely high and the inhibitor is capable of suppressing NADH-Q reductase activity in nanomolar concentrations [9–11]. Complex I from P. denitrificans was shown to be almost identical to its mitochondrial counterpart in terms of composition and thermodynamic properties of redox active groups and sensitivity to specific inhibitors [12–15]. The electron transfer within complex I from P. denitrificans is coupled to proton translocation across the bacterial mem- brane. The latter was confirmed by the following experi- mental observations. The rate of NADH oxidation, catalysed by the inside-out vesicles prepared from P. deni- trificans cells increased up to 10 times on addition of uncouplers [12]. The electron flow within the bacterial complex I can be reversed by DlH + tightly coupled sub- Paracoccus particles (SPP) were shown to catalyse an efficient DlH + -dependent reverse electron transfer from quinol to NAD + [16]. Energization of tightly coupled membrane vesicles from P. denitrificans results in changes of EPR characteristics of iron–sulfur cluster 2 of complex I [16]. All signs of energization of complex I detected by EPR in SPP [16] were also observed with SMP [17], indicating a similar mechanism of energy conservation in the bacterial and mitochondrial enzymes. The bacterial enzyme can serve as a useful model for studies of the mechanism of complex I. The aerobic respiratory chain of P. denitrificans is evolutionarily related to the mitochondrial one [18]. The functional properties of bacterial complex I (NDH-1) are almost identical to those of the mitochondrial enzyme; however, the bacterial enzyme is structurally simpler [19,20]. An additional advantage of using SPP for the study of the coupling mechanism in complex I stems from the ability to genetically manipulate bacteria using molecular biology techniques. Understanding the molecular mechanism of complex I requires knowledge about kinetics of DlH + -dependent Correspondence to A. Kotlyar, Department of Biochemistry, George S. Wise Faculty of Life Sciences, Tel Aviv University, Ramat Aviv, 69978, Israel. Fax: + 972 (3) 640 68 34, E-mail: s2shak@post.tau.ac.il Abbreviations: SMP, submitochondrial particles; SPP, sub-Paracoccus particles. (Received 19 March 2002, revised 11 June 2002, accepted 3 July 2002) Eur. J. Biochem. 269, 4020–4024 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.03091.x reactions catalysed by the enzyme. Unfortunately this information is available only for mitochondrial complex I. In this work the kinetic parameters of the direct and reverse reactions carried out by the membrane-bound P. denitrifi- cans complex I are estimated and compared with those of the mitochondrial enzyme. The data on inhibition of NADH-oxidase and DlH + -dependent NAD + -reductase reactions by specific inhibitors of NADH- and ubiquinone- binding sites of complex I are presented. MATERIALS AND METHODS All chemicals were obtained from the Sigma Chemical Company. The P. denitrificans strain Pd1222 was kindly supplied by R. van Spanning (Free University of Amsterdam, the Netherlands). Bacteria were grown anaerobically with succinate as the substrate and nitrate as the added terminal electron acceptor under growth conditions described by John and Watley [21]. Inside-out vesicles from P. denitrifi- cans were prepared as described by John & Hamilton [22] except that 1 mgÆmL )1 BSA (fatty acid free) was added to the buffer in which the lysozyme-treated cells were suspen- ded. The vesicles were stored at 4 °C for up to 2 weeks without noticeable reduction of either NADH-oxidase or reverse electron transfer activities. The protein content in SPP was determined with Biuret reagent. NADH-oxidase and NAD + -reductase activities were measured at 25 °C in 0.7 mL of assay solution containing: 5m M Hepes buffer, pH 7.0 and 1 m M magnesium acetate. NADH-oxidase reaction was initiated by addition of 10–50 lg of SPP to the assay solution supplemented with 100 l M NADH, 1 lg gramicidin and 15 m M ammonium acetate. The succinate-supported NAD + -reductase reaction was initiated by addition of 50–100 lgofSPPtotheassay solution supplemented with 2 m M NAD + and 2.5 m M succinate-K. The initial rates of NADH oxidation or NAD + reduction were followed at 340 nm (e ¼ 6.2 m M )1 Æcm )1 ). Other details of the assays are indicated in the legends to figures. RESULTS The SPP used in this work are tightly coupled. Addition of gramicidin to SPP, respiring on NADH, results in an % sevenfold increase of the NADH oxidation rate (data not presented). The SPP are capable of catalyzing the DlH + - dependent reverse electron transfer from quinol to NAD + , driven by succinate oxidation. The rates of the direct and reverse reactions depend hyperbolically on concentrations of NADH and NAD + , respectively (see Fig. 1A,B). The kinetic parameters of the reactions were estimated from the analysis of dependencies in Lineweaver–Burk plots (see Fig. 1). The K m and V max values are equal to 5.1 l M and 1.2 lmolÆmin )1 Æmg protein )1 and 19.6 l M and 0.1 lmolÆmin )1 Æmg protein )1 for NADH-oxidase and NAD + -reductase reactions, respectively. The above values are not significantly different from those estimated earlier [23] for SMP-catalyzed reactions (see Table 1). The bacterial complex I is strongly inhibited by ADP- ribose, a competitive inhibitor of the mitochondrial enzyme [24]. The K i value for competitive inhibition of the bacterial enzyme by ADP-ribose estimated from the analysis of the data in Dixon plots (Fig. 2) is equal to 45 l M . This value is similar to that estimated recently for ADP-ribose induced inhibition of mitochondrial complex I [24]. ADP-ribose selectively inhibits the direct, NADH-oxidase but not the reverse NAD + -reductase reaction, catalyzed by SPP. The data presented in Fig. 3 demonstrate that addition of ADP- ribose to the assay has no effect on the initial rate of the reverse electron transfer; furthermore, ADP-ribose stimu- lates accumulation of NADH in time. A similar effect of ADP-ribose on the reverse electron transfer reaction catalyzed by SMP has been demonstrated recently by Vinogradov and coworkers [24]. Comparison of the kinetic data obtained in the present study with those obtained previously for mitochondrial complex I (Table 1) shows resemblance of the NAD(H) binding sites of P. denitrificans and mitochondrial complex I. The bacterial complex I has much lower affinity to rotenone, a specific inhibitor of Q reduction, than the mitochondrial enzyme. As seen in Fig. 4, the inhibitor is capable of suppressing the rate of NADH oxidation of SPP Fig. 1. Kinetics of NADH oxidation and NAD + reductionbySPP. Lineweaver–Burk plots of the initial rates at different concentrations of NADH (A) and NAD + (B). The initial rates (V 0 )ofNADHoxidation or NAD + reduction were measured at different dinucleotide concen- trations as described in Materials and methods. V 0 is expressed in lmolÆmin )1 Æmg protein )1 . Ó FEBS 2002 NADH oxidation and NAD reduction catalysed by SPP (Eur. J. Biochem. 269) 4021 in micromolar concentrations, while the mitochondrial enzyme is inhibited by nanomolar concentrations of rote- none (see Table 1). Rotenone selectively suppresses direct, NADH-oxidase but not the reverse, NAD + -reductase reaction. As seen in Fig. 5, rotenone at 5 l M strongly inhibits NADH-oxidase activity of the enzyme; however, the inhibitor does not affect the initial rate of the reverse electron transfer reaction. Moreover, rotenone stimulates the process by increasing the extent of NAD + reduction in the reverse electron transfer reaction. As seen in Fig. 5B, the rate of NADH accumu- lation is reduced in time (curve 1). The reason for that is the dinucleotide oxidation in the NADH-oxidase reaction. At a certain NADH concentration the rate of NAD + reduction becomes equal to that of NADH oxidation and the steady state is achieved. The ability of rotenone to selectively suppress the direct reaction results in an increase of steady-state level NADH and in straightening up the curve (curve 2). DISCUSSION The results of this work clearly show that the affinity of the NADH-binding site of the bacterial complex I to substrates of the direct and the reverse reactions is not greatly different from that estimated for the mitochondrial Table 1. Catalytic properties of membrane particles from mitochondria (SMP) and P. denitrificans (SPP). Reaction Preparation V max (lmolÆmin )1 Æmg )1 ) K m (l M ) K ADPÀribose i ðlM) K rotenone i ðlM) NADH-oxidase SMP 1.17 a 1.0 a 25 c 0.001 e NADH-oxidase SPP 1.20 b 5.1 b 45 d 1.0 f NAD + -reductase SMP 0.29 a 37.0 a No inhibition 0.03 g NAD + -reductase SPP 0.11 b 19.6 b No inhibition No inhibition a Data taken from [23]; b Estimated from Fig. 1; c Data taken from [24]; d Estimated from Fig. 2; e Data taken from [10]; f Estimated from Fig. 4; g Data taken from [9]. Fig. 2. Competitive inhibition of the NADH-oxidase by ADP-ribose. The initial rates of NADH oxidation (V 0 ) were measured as described in Materials and methods in the presence of: 2 (curve 1), 4 (curve 2), 6(curve3),and8l M NADH (curve 4). The dependencies of initial rates of the reaction on concentration of ADP-ribose are presented in Dixon coordinates. V 0 is expressed in lmol of NADH oxidized per min per mg of protein. Fig. 3. The effect of ADP-ribose on the time-course of the succinate- supported NAD + reduction. Traces depict the change of absorbency at 340 nm associated with succinate supported NAD + -reductase reac- tion. The reaction was assayed as described in Materials and methods in the solution, containing 50 l M NAD + ,2.5m M succinate-K (curve 1) and 1 m M ADP-ribose (curve 2). Fig. 4. Dependence of NADH-oxidase activity of SPP on the concen- tration of rotenone. SPP were preincubated in assay solution, con- taining 5 m M Hepes, pH 7.0, 1 m M magnesium acetate, 2 l M NADH, and rotenone (concentrations are indicated in the figure) for 2 min prior to simultaneous addition of: 100 l M NADH, 15 m M ammonium acetate and 1 lg of gramicidin to the assay mixture. The initial rates (V 0 ) were measured as described in Materials and methods and are expressed in lmol of NADH oxidized per min per mg of protein. Solid curve corresponds to a single hyperbolic best fit with the following parameters: K i ¼ 1.0 l M , V max ¼ 1.2 lmol of NADH oxidized per min per mg of protein. 4022 A. B. Kotlyar and N. Borovok (Eur. J. Biochem. 269) Ó FEBS 2002 enzyme (see Table 1). The NADH-oxidase activity of the bacterial enzyme is strongly suppressed by ADP-ribose, a competitive inhibitor of the dinucleotide-binding site of the mitochondrial enzyme [24]. As in the case of mitochondrial complex I, ADP-ribose is capable of select- ive suppression only of the NADH-oxidase reaction catalysed by highly coupled SPP. The initial rate of the energy-dependent NAD + reduction by succinate is insen- sitive to ADP-ribose (Fig. 3). The ability of ADP-ribose to selectively inhibit only the NADH-oxidase reaction results in an increase in the steady-state level of NADH, which was established during aerobic succinate-supported reverse electron transfer catalysed by tightly coupled SPP (Fig. 3). A simulative effect of ADP-ribose on the reverse electron transfer activity, similar to that shown in this work, has been demonstrated by Vinogradov and coworkers on mitochondrial complex I [24]. Comparison of the data presented in this work with those obtained previously on mitochondrial complex I (Table 1) clearly shows that the functional properties of the dinucleotide- binding site of P. denitrificans complex I are almost identical to those of the mitochondrial enzyme. Bacterial complex I is much less sensitive to rotenone than the mitochondrial one. The NADH-oxidase activity of SPP can be strongly suppressed only at micromolar rotenone concentrations. This result is in good agreement with the observation of Mejer and coworkers [25], demon- strating relatively low affinity of the whole cells and membrane particles of P. denitrificans to rotenone and complete reversibility of rotenone-induced inhibition by BSA. Rotenone is known to specifically block the electron flow within complex I at the Q-reductase region [8]. The different sensitivities of SMP and SPP to rotenone indicate a difference in the organization of the Q-reductase segment of the bacterial and mitochondrial enzymes. The absence of active/inactive transition of P. denitrificans complex I [16], the phenomenon that is related to Q-reductase function of complex I [26], further supports the above suggestion. Perhaps the most unexpected finding of this work is the inability of rotenone to inhibit the DlH + -dependent NAD + -reductase reaction. The different sensitivities of NADH-oxidase and NAD + -reductase reactions catalyzed by coupled SMP to rotenone has been shown in our earlier studies [9]; however, both reactions were completely inhibited by submicromolar concentrations of the inhi- bitor. Inability of rotenone and ADP-ribose to inhibit DlH + -dependent reverse electron transfer catalyzed by the coupled SPP can be explained by assuming lower affinity of the energized complex I, compared to the affinity of the uncoupled enzyme, to both ligands. It has been shown previously [3,27,28] that the affinities for NADH and NAD + are significantly different for Ôcou- pledÕ and ÔuncoupledÕ complex I, supporting the above proposal. REFERENCES 1. Fearnley, I.M. & Walker, J.E. (1992) Conservation of sequences of subunits of mitochondrial complex-I and their relationships with other proteins. Biochim. Biophys. Acta 1140, 105–134. 2. Walker, J.E., Skehel, J.M. & Buchanan, S.K. (1995) Structural analysis of NADH: ubiquinone oxidoreductase from bovine heart mitochondria. Methods Enzymol. 260, 14–34. 3. Vinogradov, A.D. (1998) Catalytic properties of the mitochond- rial NADH-ubiquinone oxidoreductase (Complex I) and the pseudo-reversible active/inactive enzyme transition. Biochim. Biophys. Acta 1364, 169–185. 4. Lindahl, P.E. & O ¨ berg, K.E. (1961) The effect of rotenone on respiration and its point of attack. Exp. Cell Res. 23, 228–237. 5. Ernster,L.,Dallner,G.&Azzone,G.F.(1963)Differentialeffects of rotenone and amytal on mitochondrial electron and energy transfer. J. Biol. Chem. 238, 1124–1131. 6. Burgos, J. & Redfearn, E.R. (1965) The inhibition of mitochon- drial reduced nicotinamide-adenine dinucleotide oxidation by rotenoids. Biochim. Biophys. Acta 110, 475–483. 7. Horgan, D.J. & Singer, T.P. (1968) Studies on the respiratory chain-linked reduced nicotinamide adenine dinucleotide dehy- drogenase. XIII. Binding sites of rotenone, piericidin A, and amytal in the respiratory chain. J. Biol. Chem. 243, 834–843. Fig. 5. The effect of rotenone on the time-course of NADH oxidation (A) and succinate-supported NAD + reduction (B) catalyzed by tightly coupled SPP. (A) SPP (30 lg) were preincubated for 2 min in assay solution, containing 5 m M Hepes, pH 7.0, 1 m M magnesium acetate, 2 l M NADH (curve 1) and 5 l M rotenone(curve2).TheNADH- oxidase reaction was initiated by simultaneous addition of 50 l M NADH, 15 m M ammonium acetate and 1 lgofgramicidintothe solution. (B) SPP (100 lg) were preincubated as in (A) in the presence (curve 2) and the absence (curve 1) of 5 l M rotenone. The NAD + reduction was initiated by addition of 2 m M NAD + and 2.5 m M succinate to the assay mixture. Ó FEBS 2002 NADH oxidation and NAD reduction catalysed by SPP (Eur. J. Biochem. 269) 4023 8. Palmer, G., Horgan, D.J., Tisdale, H., Singer, T.P. & Beinert, H. (1968) Studies on the respiratory chain-linked reduced nicotin- amide adenine dinucleotide dehydrogenase. XIV. Location of the sites of inhibition of rotenone, barbiturates, and piericidin by means of electron paramagnetic resonance spectroscopy. J. Biol. Chem. 243, 844–847. 9. Kotlyar, A.B. & Gutman, M. (1992) The effect of delta-lH + on the interaction of rotenone with complex-I of submitochondrial particles. Biochim. Biophys. Acta 1140, 169–174. 10. Grivennikova, V.G., Maklashina, E.O., Gavrikova, E.V. & Vinogradov, A.D. (1997) Interaction of the mitochondrial NADH-ubiquinone reductase with rotenone as related to the enzyme active/inactive transition. Biochim. Biophys. Acta 1319, 223–232. 11. Nakashima, Y., Shinzawa-Itoh, K., Watanabe, K., Naoki, K., Hano, N. & Yoshikawa, S. (2002) Steady-state kinetics of NADH: coenzyme Q oxidoreductase isolated from bovine heart mitochondria. J. Bioenergetics Biomembranes 34, 11–19. 12. John, P. & Whatley, F.R. (1977) The bioenergetics of Paracoccus denitrificans. Biochim. Biophys. Acta 463, 129–153. 13. Meijer, E.M., Wever, R. & Stouthamer, A.H. (1977) The role of iron–sulfur center 2 in electron transport and energy conservation in the NADH-ubiquinone segment of the respiratory chain in Paracoccus denitrificans. Eur. J. Biochem. 81, 267–275. 14. Albracht, S.P.J., Van Verseveld, H.W., Hagen, W.R. & Kalkman, M.L. (1980) Comparison of the respiratory chain in particles from Paracoccus denitrificans and bovine heart mitochondria by EPR spectroscopy. Biochim. Biophys. Acta 593, 173–186. 15. Meinhardt, S.W., Kula, T., Yagi, T., Lillich, T. & Ohnishi, T. (1987) EPR characterization of the iron-sulfur clusters in the NADH:ubiquinone oxidoreductase segment of the respiratory chain in Paracoccus denitrificans. J. Biol. Chem. 262, 9147–9153. 16. Kotlyar, A.B., Albracht, S.P.J. & van Spanning, R.J.M. (1998) Comparison of energization of complex I in membrane particles from Paracoccus denitrificans and bovine heart mitochondria. Biochim. Biophys. Acta 1365, 53–59. 17.VanBelzen,R.,Kotlyar,A.B.,Moon,N.,Dunham,W.& Albracht, S. (1997) The iron-sulfur clusters 2 and ubisemiquinone radicals of NADH:ubiquinone oxidoreductase are involved in energy coupling in submitochondrial particles. Biochemistry 36, 886–893. 18. John, P. & Whatley, F.R. (1975) Paracoccus denitrificans and the evolutionary origin of the mitochondrion. Nature 254, 495–498. 19. Yagi, T. (1986) Purification and characterization of NADH dehydrogenase complex from Paracoccus denitrificans. Arch. Biochem. Biophys. 250, 302–311. 20. Yagi, T., Yano, T., Di Bernardo, S. & Matsuno-Yagi, A. (1998) Procaryotic complex I (NDH-1), an overview. Biochim. Biophys. Acta 1364, 125–133. 21. John, P. & Whatley, F.R. (1970) Oxidative phosphorylation coupled to oxygen uptake and nitrate reduction in Micrococcus denitrificans. Biochim. Biophys. Acta 216, 342–352. 22. John, P. & Hamilton, W.A. (1971) Release of respiratory control in particles from Micrococcus denitrificans by ion-translocating antibiotics. Eur. J. Biochem. 23, 528–532. 23. Frenkin, M.V. & Kotlyar, A.B. (1999) Arylazido-alanine ADP- ribose, a novel irreversible competitive inhibitor of mitochondrial NADH-ubiquinone reductase. Biochim. Biophys. Acta 1413,139– 146. 24. Zharova, T.V. & Vinogradov, A.D. (1997) A competitive inhibi- tion of the mitochondrial NADH-ubiquinone oxidoreductase (complex I) by ADP-ribose. Biochim. Biophys. Acta 1320,256– 264. 25. Meijer,E.M.,Schuitenmaker,M.G.,Boogerd,F.C.,Wever,R.& Stouthamer, A.H. (1978) Effects induced by rotenone during aerobic growth of Paracoccus denitrificans in continuous culture. Changes in energy conservation and electron transport associated with NADH dehydrogenase. Arch. Microbiol. 119, 119–127. 26. Kotlyar, A.B. & Vinogradov, A.D. (1990) Slow active/inactive transition of the mitochondrial NADH-ubiquinone reductase. Biochim. Biophys. Acta 1019, 151–158. 27. Avraam, R. & Kotlyar, A.B. (1991) Kinetics of NADH oxidation and NAD + reduction by mitochondrial Complex I. Biochimia (Moscow) 56, 1676–1686. 28. Vinogradov, A.D. & Grivennikova, V.G. (2001) The mitochon- drial complex I. Prog. Understanding Catalytic Properties. IUBMB Life 52, 129–134. 4024 A. B. Kotlyar and N. Borovok (Eur. J. Biochem. 269) Ó FEBS 2002 . NADH oxidation and NAD + reduction catalysed by tightly coupled inside-out vesicles from Paracoccus denitrificans Alexander B. Kotlyar and Natalia. Ramat Aviv, Israel Tightly coupled inside-out vesicles were prepared from Paracoccus denitrificans cells (SPP, sub -Paracoccus particles) and characterized