SubstrateâProtein Interactions of Type II NADH:Quinone

Subscriber access provided by BOSTON UNIV

Article

Substrate-Protein Interactions of Type II NADH:quinone Oxidoreductase from Escherichia coli Johannes Salewski, Ana P Batista, Filipa V. Sena, Diego Millo, Ingo Zebger, Manuela M Pereira, and Peter Hildebrandt Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.6b00070 • Publication Date (Web): 25 Apr 2016 Downloaded from http://pubs.acs.org on May 2, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Biochemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Substrate-Protein Interactions of Type II NADH:quinone Oxidoreductase from Escherichia coli Johannes Salewski§1, Ana P. Batista§2,3, Filipa V. Sena2, Diego Millo4, Ingo Zebger1, Manuela M. Pereira2*, Peter Hildebrandt1* 1

Technische Universität Berlin, Institut für Chemie, Sekr. PC14, Straße des 17. Juni 135, D-

10623 Berlin, Germany 2

Instituto de Tecnologia Química e Biológica – António Xavier, Universidade Nova de

Lisboa, Av. da República EAN, P-2780-157 Oeiras, Portugal 3

Current address: iBET, Instituto de Biologia Experimental e Tecnológica, Apartado 12, P-

2780-901 Oeiras, Portugal 4

Biomolecular Spectroscopy/LaserLaB Amsterdam, Vrije Universiteit Amsterdam, De

Boelelaan 1083, 1081 HV Amsterdam, The Netherlands

§

These authors contributed equally to the work

* To whom correspondence should be addressed: Manuela M. Pereira, [email protected]; Peter Hildebrandt, [email protected]

FUNDING J. S. and P. H. acknowledge the financial support by the Leibniz Graduate School (LGS) of the Leibniz Institute for Molecular Pharmacology, the Deutsche Akademische Austauschdienst (DAAD), and the Cluster of Excellence “UniCat”, funded by the Deutsche Forschungsgemeinschaft. D. M. acknowledges the Alexander von Humboldt Foundation and the Netherlands Organisation for Scientific Research NWO (Veni grant 722.011.003) for funding. A.P.B. acknowledges the FEBS Short-Term Fellowship awarded in 2010. F.V.S. is recipient of a fellowship by Fundação para a Ciência e a Tecnologia (PD/BD/113985/2015). The work was funded by Fundação para a Ciência e a Tecnologia (PTDC/BBBBQB/2294/2012 to M.M.P.). ITQB is supported by Fundação para a Ciência e a Tecnologia through R&D Unit, UID/CBQ/04612/2013.

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 38

ABBREVIATIONS

ATR, attenuated total reflexion; BCA, bicinchoninic acid; BSA, bovine serum albumin; ∆F, change in emission at 330 nm; CV, cyclic voltammetry; DDB, 2,3-dimethoxy-5,6-dimethyl1,4-benzoquinone; DUQ, 2,3-dimethoxy-5-methyl-6-decyl-1,4-benzoquinone or decylubiquinone; KD, dissociation constant; E. coli, Escherichia coli; FAD, flavin adenine dinucleotide; FMN, flavin mononucleotide; FT-IR, Fourier transform infrared spectroscopy; Au, gold; IR, infrared spectroscopy; EcNDH-2, NDH-2 from Escherichia coli; OD600, optical density at 600nm; PMSF, phenylmethylsulfonyl fluoride; S. cerevisiae, Saccharomyces cerevisiae; SAM, self-assembled monolayer; Ag/AgCl, silver chloride electrode; SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis; SHE, standard hydrogen electrode; SEIRA, surface enhanced infrared absorption spectroscopy; NDH-2s, Type II NADH:quinone oxidoreductases; Vmax, maximal velocity.

2 ACS Paragon Plus Environment

Page 3 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

ABSTRACT

Type II NADH:quinone oxidoreductases (NDH-2s) are membrane proteins involved in respiratory chains and responsible for the maintenance of NADH/NAD+ balance in cells. NDH-2s are the only enzymes with NADH dehydrogenase activity present in the respiratory chain of many pathogens and thus they were proposed as suitable targets for antimicrobial therapies. In addition NDH-2s were also considered key players for the treatment of complex I-related neurodegenerative disorders. In this work we explored substrate-protein interaction in NDH-2 from Escherichia coli (EcNDH-2) combining surface enhanced infrared absorption spectroscopic studies with electrochemical experiments, fluorescence spectroscopy assays, and quantum chemical calculations. Due to the specific stabilization of substrate complexes of EcNDH-2 immobilized on electrodes, it was possible to demonstrate the presence of two distinct substrate binding sites for NADH and the quinone and to identify a bound semi-protonated quinol as a catalytic intermediate.


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 38

Type II NADH:quinone oxidoreductases (NDH-2s) are membrane proteins involved in respiratory chains and responsible for the regeneration of NAD+, contributing in this way to the maintenance of the NADH/NAD+ balance. These enzymes are homodimeric proteins, localized at the surface of the lipid bilayer through electrostatic interactions, and catalyze the transfer of two electrons from NADH to the quinone (Figure 1). Each monomer with approximately 45 kDa contains a non-covalently bound flavin adenine dinucleotide (FAD) molecule as the only prosthetic group 1. NDH-2 performs the same catalytic function as the respiratory complex I and its expression was observed to restore the mitochondrial activity in yeast and animals with complex I deficiency 2, 3. Therefore, NDH-2 gained interest as a key enzyme for the treatment of complex I-related neurodegenerative disorders like Parkinson’s disease 4. In addition, NDH-2 is the only enzyme with NADH:quinone oxidoreductase activity expressed in many pathogenic organisms and for that, it has been proposed as a possible new drug target for the rational design of antibiotics 5-7. In the last years a wealth of data has been collected on the characterization of NDH-2s, including the crystal structures of the proteins from yeast and from two bacteria. All these data stimulated several debates on the number and location of the substrates binding sites and on the type of the involved catalytic mechanism 8-11. The structural analysis of NDH-2 from Saccharomyces cerevisiae by Iwata and co-workers showed overlapping binding sites for NADH and quinone, whereas in the crystal structure obtained by Feng et al separated binding sites are observed 8, 10. In the case of the catalytic mechanism various kinetic studies favored a sequential substrate binding (“ping-pong”) mechanism 12, 13 but spectroscopic and fast kinetic analyses suggested a mechanism involving a ternary complex (enzyme-NADH-quinone) 11, 14. The present study on NDH-2 from Escherichia coli (EcNDH-2) uses an approach that provides structural information about the interaction of the individual substrates with the protein. Here, we have focused on the enzyme immobilized on a biocompatible coated gold (Au) electrode which mimics physiological reaction conditions of the membrane-associated enzyme more closely than solution studies. Using a nanostructured Au surface allows employing surface enhanced infrared absorption (SEIRA) spectroscopy which probes solely the molecular species in close proximity to the surface. Operating in the difference mode, the spectra selectively reflect the structural changes induced by specific reaction steps. These studies which are complemented by electrochemical experiments, fluorescence spectroscopy


Page 5 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

studies and quantum chemical calculations provide new insights into the catalytic mechanism of EcNDH-2.

EXPERIMENTAL PROCEDURES Materials. NADH, NAD+ and 2,3-dimethoxy-5-methyl-6-decyl-1,4-benzoquinone (decylubiquinone, DUQ) were purchased from Sigma Aldrich. 2,3-dimethoxy-5,6-dimethyl1,4-benzoquinone (DDB) was acquired from Sequoia Research Products. BCA Protein Assay Reagent was purchased from Pierce. Flavin adenine dinucleotide (FAD) and flavin mononucleotide (FMN) were acquired from Merck and Sigma Aldrich, respectively.

EcNDH-2 expression and purification. The gene coding for the NDH-2 from Escherichia coli K12 (b1109) was cloned into the vector pET-28a(+) (Novagen). E. coli Rosetta 2 (DE3) pLysS cells were transformed with this vector and grown in 2x YT rich medium supplemented with kanamycin (100 µg/mL) and chloramphenicol (34 µg/mL) at 37 °C and 180 rpm. Gene expression was induced by the addition of 1 mM isopropyl ß-D-1thiogalactopyranoside when cells reached an OD600 of 0.6. The cells were harvested 4 hours after induction by centrifugation at 8 000 rpm for 10 min and were then disrupted (French Press at 6 000 psi). Cellular debris and undisrupted cells were separated by centrifugation at 10 000 rpm for 10 min. The soluble and membrane fractions were separated by ultracentrifugation (42 000 rpm for 2h). EcNDH-2 was purified from the membrane fraction, washed with 2M NaCl in two purification steps, including a Q-Sepharose HP and a Superdex 200 columns (GE Healthcare). The purified protein was stored in 20 mM TRIS-HCl pH 7.5, 1 mM phenylmethylsulfonyl fluoride (PMSF), 150 mM NaCl at -80 ºC. Protein concentration was determined by the BCA protein assay reagent using BSA as standard. Protein purity was evaluated by SDS-PAGE. The flavin prosthetic group was identified by reverse phase chromatography using commercial FAD and FMN as standards as described elsewhere 11. Flavin content of EcNDH-2 was determined spectroscopically with a Shimadzu UV-1603 spectrophotometer, using the extinction coefficient of 11.3 mM−1 cm−1 at 450 nm for the free oxidized flavin. Extraction of the flavin was done in the presence of 10 % trichloroacetic acid (v/v) for 10 min followed by centrifugation for 15 min.


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 38

UV-visible Absorption Spectroscopy. Activity assays were performed on Shimadzu UV-1800 spectrophotometer (installed in an anaerobic chamber) monitoring the change in absorbance of the electron donor NADH at 340 nm. The reaction mixture contained 50 nM EcNDH-2 in 100 mM potassium phosphate pH 7.0 (250 mM NaCl). The activity assays were made at 30 °C using 100 µM of NADH and 100 µM DUQ or 100 µM DDB. The NADH extinction coefficient of 6.22 mM-1 cm-1 was used to calculate the specific activity (µmol-1 min-1 mg-1 protein) of EcNDH-2. UV-Visible absorption spectra of EcNDH-2, reduced by the addition of NADH and oxidized by the addition of quinone, were also recorded. Redox titration of EcNDH-2 (20 µM) monitored by visible spectroscopy at room temperature and pH 7, was performed in a cuvette continuously flushed with argon, by stepwise addition of buffered sodium dithionite (reductive way) or hexachloroiridate (oxidative way). The following redox mediators, each at a final concentration of 2 µM, were used: methyl viologen, benzyl viologen, phenosafranin, anthraquinone-2-sulphonate, 2-hydroxy-1,4-naphtoquinone, indigo trisulphonate, plumbagin, 1,4-naphtoquinone, trimethylhydroquinone and 1,2naphtoquinone-4-sulphonic acid. Spectra from 350 to 700 nm were obtained at each solution redox potential, after attaining equilibrium. An Ag/AgCl electrode was used, calibrated against a saturated quinhydrone solution. The Nernst equation for a two-electron redox process was fitted to the experimental data.

Fluorescence Spectroscopy. Fluorescence spectra were recorded on a Varian Cary Eclipse spectrofluorimeter. The reaction mixture (500 µL) contained 6 µM of EcNDH-2 in 20 mM TRIS-HCl pH 7.5 and 250 mM NaCl. Conformational changes in EcNDH-2 due to interactions with NAD+ (100 µM), DDB (100 µM) and DUQ (100 µM) were monitored. NAD+ was used instead of NADH in order to determine fluorescence changes due to substrate binding by avoiding changes due to the reduction reaction. Three independent titrations were performed with each substrate in the range of 0 to 100 µM. Tryptophan and flavin fluorescence emission spectra (at 25 °C) were recorded with excitation wavelength of 280 nm and 450 nm, respectively. The related change in emission at 330 nm (∆F) was normalized and plotted versus the substrate concentration. As control experiments, fluorescence spectra of the substrates (NAD+, DDB and DUQ) without protein were also performed. To evaluate protein stability, thermal denaturation assays (20 − 90 °C) were carried out using a Peltier temperature controller with a rate of 0.5 °C/min, and the data were


Page 7 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

recorded in intervals of 0.5 °C with an acquisition time of 0.1 min. Denaturation was monitored using excitation at 450 nm and emission at 530 nm.

SEIRA Spectroscopy. A silicon prism surface typically used for infrared (IR) spectroscopy in the attenuated total reflexion (ATR) mode was coated with a nanostructured Au film via electroless deposition from an Au solution following the protocol described elsewhere 15. The prism was mounted in a home-built spectro-electrochemical glass cell using a Ag/AgCl (3.0 M KCl) and a Pt mesh as reference and counter electrode, respectively. Electrochemical cleaning of the Au surface was done in 0.1 M sulphuric or perchloric acid by applying six oxidation-reduction cycles with a scan rate of 0.05 V/s between +0.2 V and +1.5 V (Metrohm µAUTOLABIII) after purging the solution in the cell with argon for ca. 15 min to remove oxygen. The film was then incubated in a 1 mM solution of 6-amino-hexanethiol (Dojindo, purity >90%) in ethanol/water (4:1) at 4 °C for ca. 16 h to form a homogeneous selfassembled monolayer (SAM). After inserting the cell into the FT-IR spectrometer (Bruker IFS66v/s or Bruker Tensor27, liquid nitrogen-cooled HgCdTe detector, operated at 4 cm−1 resolution) the protein was adsorbed to the surface from a 2 µM solution in 20 mM TRIS-HCl buffer at pH 7.5. The glass cell was kept at constant temperature (27 °C) and a continuous flow of argon was applied in all experiments to remove oxygen from solution and atmosphere. The SEIRA spectroscopic experiments were carried out in the difference mode to selectively probe the spectral changes induced by individual reaction steps. Thus, the difference spectrum for a reaction step i of a sequential series of reaction steps was constructed by subtracting the spectrum of the preceding reaction step i-1. In such multi-step measurements, the subsequent step was made only after complete equilibration of the system, i.e. when no further changes in the absorption difference spectrum and no charge flow was observed. Substrates were added in excess by taking 1 mL of buffer solution from the cell to dilute the powder (NADH or NAD+) or quinone solution (DUQ or DDB, 20 mM in ethanol) and returning it into the cell. The final concentration was 400 µM and 200 µM for NADH and the quinone, respectively. After equilibration the excess substrate was removed by washing the cell several times with buffer and again spectra were recorded until equilibration. Typically, 400 interferograms were co-added for one spectrum.

Electrochemistry measurements. Cyclic voltammetry (CV) experiments were done in two different experimental configurations: inside the SEIRA spectroelectrochemical cell (see above), and in a home-made electrochemical glass cell equipped with the same reference and 7 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 38

counter electrodes of the SEIRA spectroelectrochemical cell but with a different working electrode: a 2 mm diameter disk electrode (Chambria Scientific). This electrode was immersed in Piranha solution (H2SO4:H2O2 mixed 3:1 (v/v)) at room temperature for 5 10 min and subsequently polished with 50 nm grain size alumina suspension (Buehler MasterPrep Polishing Suspension) until visual inspection revealed a mirror-like surface appearance. After removing the alumina in excess with the ultrasonic bath, the electrode was subjected to the same electrochemical treatment in H2SO4 or HClO4 followed by SAM formation as described for the SEIRA Au electrode. The electroactive surface area of the SEIRA and the disk electrodes was determined measuring the charge passed during the stripping of the Au oxide layer formed during the electrode treatment in acid solution 16. The surface roughness factor was obtained as described by others 17, and was found to be 3.0±0.5 for both the disk and the SEIRA Au electrode. Both electrodes afforded the same results as demonstrated for the CV studies on the EcNDH2DUQ couple (see below Results section). All potentials reported in this work were converted to the SHE reference electrode system (-210 mV vs. Ag/AgCl (3.0M KCl)).

Quantum chemical calculations. The IR spectra of DUQ in the oxidized and reduced state with different protonation patterns were calculated using the software package Gaussian09 with the bp86 functional and 6-31g* basis set applied on all atoms. We implemented a polarizable continuum model using the relative permittivity of water εr(H2O) = 78.36 to mimic the solvent-exposed DUQ binding site of the enzyme 18, 19.

RESULTS

Biochemical characterization. NDH-2 from Escherichia coli was expressed and purified from the membranes (Figure S1). EcNDH-2 presented an absorption spectrum typical of a flavoprotein, exhibiting bands with maxima at 375 nm and 450 nm (Figure S2). Full EcNDH2 reduction by NADH is characterized by the loss of absorption in this region. The flavin cofactor, identified as FAD by HPLC analysis, is non-covalently bound as it could be removed by acidic precipitation or thermal denaturation. The presence of other cofactors such as quinones was not observed. One FAD molecule was present per protein. EcNDH-2 maintains its FAD up to 60 °C as monitored by thermal denaturation assays. The reduction potential of FAD, determined by UV-visible-spectroscopic titration, was determined to be 8 ACS Paragon Plus Environment

Page 9 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

220 ± 10 mV, which is in the range typical for flavin cofactors 20 (Figure 2). The enzyme revealed a maximal velocity (Vmax) of 8.1±0.4 and 17.5±1.5 µmol NADH/min·mg protein for DUQ or DDB, respectively.

Fluorescence Spectroscopy Studies. EcNDH-2 has two tryptophan residues, W272, located at the re side of FAD and W47 present at the si side of FAD close to its isoalloxazine ring of FAD (data not shown). The fluorescence emission spectrum of EcNDH-2, obtained with excitation at 280 nm, exhibits the characteristic features of a tryptophan fluorescence spectrum with a broad maximum at around 320 - 330 nm. The flavin fluorescence emission spectrum, measured with excitation at 450 nm, displays maximum intensity at 530 nm. Fluorescence emission spectra of EcNDH-2 in the presence of each substrate (NAD+, DDB and DUQ) were obtained. As it can be seen in Figure 3, the intensity in the region of the tryptophan fluorescence changed significantly and differently upon addition of each substrate. These changes are more pronounced in the case of quinone as compared to NAD+ (Figure 3A and 3B). These findings may be rationalized by different interactions of the two substrates with same binding site or, most likely, by the existence of two distinct substrate binding sites for NADH and quinone. A maximum decrease of ~75 % was obtained when EcNDH-2 was titrated with DUQ or DDB (Figure 3D and 3E), whereas for NAD+ the maximum was ~ 45 % (Figure 3C). The titration experiments afforded dissociation constants (KD) of 18.9 µM, 6.1 µM and 8.7 µM for NAD+, DUQ and DDB respectively, indicating similar affinity of both quinones to EcNDH-2. The flavin fluorescence did not change significantly in the presence of each substrate implying that substrate binding does not induce protein conformational changes in the immediate vicinity of the flavin cofactor.

SEIRA spectroscopy studies - NADH – DUQ substrate pair. Starting with the electrostatic adsorption of EcNDH-2 on the SAM-coated Au surface, the spectrum of the SAM-coated Au prior to enzyme addition was used as the reference. The corresponding difference spectra thus display the amide I and amide II bands of the protein at 1660 and 1550 cm−1, respectively. The absorbance at these positions increases by time in a nearly monoexponential manner reaching a plateau after ca. 45 min that corresponds to the maximum surface coverage (Figure S3). No changes in the amide I / amide II intensity ratio and thus no reorientation of the immobilized protein occurred over time. There were also no spectral changes when removing 9 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 38

the excess of protein by washing several times with fresh buffer, indicating a firmly bound enzyme. Figure 4A shows a SEIRA difference spectrum of surface-immobilized EcNDH-2 after NADH binding at open circuit, in contact with a 20 mM TRIS-HCl (pH 7.5) buffer solution. The absorption spectrum before addition of NADH to the buffer solution in the cell (final concentration was 400 µM) was subtracted from the spectrum after addition to show the spectral changes induced by the substrate. The positive signals are characteristic of NADH as demonstrated by comparison with the solution IR spectrum and literature data (Table 1). Compared to the solution spectrum we note slight frequency shifts of 1 - 3 cm−1 for most of the bands, pointing to specific intermolecular interactions of the bound NADH. Only the bands at 1654 and 1537 cm−1 show larger deviations also between individual experiments (2 5 cm−1) which may be due to the overlap of positive and negative signals of the amide I and amide II modes of the protein, respectively. In fact, small changes of these latter modes have been previously observed upon binding of NADH to transhydrogenase 21. When the excess of NADH was removed from the spectro-electrochemical cell by washing with fresh buffer solution ca. 60% of its signal was lost. This raises the question if the molecule is weakly bound to the protein or loosely attached to the SAM. In a control experiment with NADH in the absence of enzyme (Figure 5B) a difference spectrum was obtained that was very similar to the spectrum in presence of immobilized enzyme. Only minor (≤ 2cm−1) frequency shifts for a few modes are noted between the two spectra (Figures 5B, 4A). Thus, the two spectra do not allow for an unambiguous conclusion about the binding mode of NADH (EcNDH-2 vs. SAM). Although the IR spectra of NADH and NAD+ are similar, we can, however, rule out an oxidation of the bound NADH since three features are different in the IR spectra of NADH and NAD+. In solution 21, 22 as well as in the SEIRA spectra (Figure 5), the bands at 1689 and 1542 cm−1 of NADH are replaced by bands at 1697 and 1511 cm−1 in NAD+, respectively, and the small peak at 1183 cm−1 of NADH is absent in the spectrum of NAD+. Thus, in the SEIRA difference spectrum (Figure 4A) the bands at 1689 and 1537 cm−1 are distinctly closer to those of NADH than NAD+. This observation implies that NADH remains in the reduced state upon binding to the immobilized enzyme or to the SAM. When DUQ was used as the electron acceptor and added to the buffer solution in the cell (200 µM), distinct positive signals were observed in the 1800 - 1000 cm−1 region of the difference spectrum (Figure 4B) using the spectrum of the enzyme treated with NADH as the reference. These bands are readily attributed to the modes of the oxidized DUQ (Figure 4B, Table 2). Peaks located at 1663, 1652, and 1611 cm−1 are assigned to C=O and C=C 10 ACS Paragon Plus Environment

Page 11 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

stretching modes of the quinone ring, while two bands at 1269 and 1205 cm−1 involve the bending coordinates of the DUQ methoxy side chains 23-25. Peaks at 1458, 1436, and 1379 cm−1 are attributed to CH2 and CH3 bending modes of the decyl chain and 5-methyl group 26. The spectrum was not affected by washing with fresh buffer. Furthermore, in a control experiment with DUQ in the absence of immobilized EcNDH-2, none of the typical quinone bands were detected (data not shown). These findings demonstrate that the positive SEIRA signals in Figure 4B result from DUQ tightly bound to the immobilized EcNDH-2. Remarkably, no peaks were observed in the SEIRA difference spectrum that could either be related to the release or oxidation of the remaining 40% bound NADH or to the reduction of the protein-bound DUQ. The same SEIRA difference spectrum for DUQ binding was also obtained without adding NADH to EcNDH-2 as the first step (using the NADH-free EcNDH2 spectrum as the reference; Figure S4B). Following DUQ binding to EcNDH-2, NADH was added to the buffer solution in the cell (400 µM) and SEIRA spectra were recorded (Figure 4C). All bands of the oxidized DUQ were now observed as negative signals whereas new bands at 1493, 1468, 1431, 1389, 1119, 1094, and 1053 cm−1 appeared as positive signals. Among them, the distinct four bands in the 1500 - 1350 cm−1 region can readily be related to C-C stretching modes of the reduced quinone molecule (Table 2) 26, 27. None of these peaks is observed for NADH and DUQ in the absence of immobilized EcNDH-2 (data not shown) indicating that the reduction of DUQ is catalyzed by the enzyme and thus both substrates are bound to EcNDH-2 during electron transfer process. However, in contrast to the first addition of NADH (Figure 4A), no bands from NADH are found in this step, pointing to a transient binding and the release of NAD+ from the enzyme after oxidation. Interestingly, upon washing with fresh buffer the characteristic bands of reduced DUQ were observed as negative signals. This difference spectrum (Figure 4D) is a mirror image of the difference spectrum induced by NADH addition and demonstrates the re-oxidation of protein-bound quinol, possibly due to the introduction of traces of oxygen dissolved in the fresh buffer solution that could not be removed with argon purging. DUQ substrate molecules are fully redox-active but stuck in the binding site which might be an effect of protein immobilization. This interpretation is in fact supported by the observation that for a minority of experiments a different orientation of the immobilized enzyme, as indicated by a distinctly lower amide I/amide II intensity ratio (0.98 vs. 1.26), allowed the removal of the bound quinol by washing with buffer solution, as reflected by the disappearance of all positive bands in Figure 4C. This finding suggests that even if EcNDH-2-


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 38

SAM interactions promote a strong and stable binding, they do not provide a unique preferential orientation of the enzyme with respect to the electrode surface (see below). After DUQ was bound to EcNDH-2, the steps of (i) washing with fresh buffer and (ii) adding NADH could be repeated several times, inducing the very same spectral changes as described above. Note that this reproducibility refers to the behavior of the two enzyme orientations showing either quinol reoxidation or quinol release in the last step. When instead of NADH the oxidized dinucleotide NAD+ was added to the protein, the first spectrum measured after 3 minutes (Figure 6A) is reminiscent to that of the NADH-enzyme complex in Figure 4A. A closer inspection, however, reveals a broad band at 1235 cm−1 and in addition, two clearly detectable bands at 1510 cm−1 and 1700 cm−1 instead of the NADH bands at 1537 cm−1 and 1689 cm−1, respectively. Nearly the same difference spectrum is obtained for NAD+ in the absence of enzyme (Figure 5A), peak positions are shifted 1 - 3 cm−1. This finding does not only indicate that NAD+ retains the oxidation state but also confirms, a posteriori, the assignment of the positive signals in Figure 4A to NADH. In contrast to NADH, the signals in the difference spectrum of Figure 6A vary with time leading to broad and poorly resolved features after 60 minutes. Interestingly, the same temporal variation is observed for spectra of NAD+ without EcNDH-2. These findings might point to the fact that, similar to NADH, (a fraction of) the oxidized dinucleotide is attached to the SAM-coated Au surface rather than to the protein when added as the first step after enzyme immobilization. The decrease in peak intensity might be due to a repelling interaction of the positive charges of the nicotinamide moiety and the amine head groups, respectively. However, the subsequent addition of DUQ leads to essentially the same spectrum as observed after binding DUQ to EcNDH-2 with or without pretreatment with NADH (Figures 6B, 4B, S3B). If in the subsequent step NAD+ is added instead of NADH, no further spectral changes are observed (Figure 6C) and it requires the addition of NADH to the immobilized complex to reduce the bound DUQ as reflected by the positive and negative signals of its reduced and oxidized form, respectively. Note that this spectrum (Figure 6D) agrees very well with the corresponding spectrum Figure 4C obtained directly upon NADH addition. The reduction of the protein-bound DUQ is also possible via electrons supplied by the Au electrode, instead of NADH. The SEIRA difference spectra (Figure 7) were recorded under the same conditions as those in Figure 4 except that now the electrode potential is used as a variable. Starting point is the spectrum of DUQ bound to EcNDH-2 at open circuit (Figure 7A) displaying the characteristic signals of the oxidized DUQ. Upon applying an electrode potential of −90 mV, DUQ is reduced as indicated by the four positive bands between 1350 12 ACS Paragon Plus Environment

Page 13 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

and 1500 cm−1 and the positive two-banded signal at ca. 1100 cm−1 on the one hand, and the negative bands of the oxidized DUQ on the other hand (Figure 7B). No signals were observed at potentials of +110 or +10 mV (data not shown). Electrochemical reduction is fully reversible, i.e. switching the potential back to open circuit affords the mirror image of this spectrum (Figure 7C). This can readily be rationalized in view of an open circuit potential of +100 mV which is above the reduction potential of ca. +10 mV for protein-bound DUQ as extracted from the CV data (see the cyclic voltammetry section below). Thus, electrochemical reduction and chemical reduction via NADH are equivalent as demonstrated by the spectrum in Figure 7D that is obtained after addition of NADH to the electrochemically re-oxidized enzyme-bound DUQ. Finally we refer to the band at 1743 cm−1 which is observed as negative and positive difference signal both upon electrochemical or chemical reduction and oxidation of the protein-bound DUQ. The position of this band is indicative for a C=O stretching of a protonated carboxyl side chain or an ester function. Thus, on the first sight, it is tempting to assign this band to a glutamate or an aspartate side chain of the enzyme 28 and the alternating sign of the signal to a protonation and deprotonation of these residues coupled to the catalytic process. In fact, amino acid residues of this type are found in the active site of NDH-2 8-11. An alternative interpretation is based on the observation of a band at this position in potentialdependent difference spectra of lipid bilayers on the electrodes, reflecting re-orientations of the C=O ester function in response to the potential 29. Additionally, the rise and shift of a very similar peak around 1740 cm−1 was reported for potential- and pH-dependent IR experiments on complex I after adding polar lipids to the sample 30. Furthermore, the vibrational pattern of CH2/CH3 stretching modes in the 2800 - 3000 cm−1 region, which are characteristic for lipids, and the C=O stretching at 1743 cm−1 were observed already during the adsorption of the enzyme with varying intensities among different enzyme preparations. Thus, this band may be assigned to the C=O stretching modes of protein-bound lipids. The variations of this band as indicated by positive and negative signals in the difference spectra may then reflect reorientations of the lipids or the lipid-protein complex under oxidative and reductive conditions 31. Hellwig et al. observed a band at the same position for potential-dependent FTIR difference spectra of quinone in the absence of protein 26. The authors assigned this band to the C=O stretching of the ester function of sucrose monolaurate, present in the buffer solution.


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 38

SEIRA spectroscopy studies - NADH – DDB substrate pair. The SEIRA spectroscopic analysis of the NADH-DDB substrate pair displays a different picture. When DDB (200 µM), instead of DUQ, is added to the protein, irreproducible and poorly structured spectral changes were obtained (Figure S4A), although electron acceptor activity of DDB can be deduced from steady-state kinetics and CV measurements. The lack of any distinct difference signals suggests much weaker binding affinities of the immobilized enzyme for the oxidized and reduced DDB and faster cycling through the different states of the catalytic process such that no appreciable and spectroscopically detectable concentrations of the bound substrate and product or any other intermediate could be obtained. This interpretation is in line with the CV data (see the cyclic voltammetry section below). The only common feature to the spectra of the EcNDH-2-NADH-DUQ complexes is the weak signal at 1743 cm−1.

Quantum mechanical calculations of substrate vibrational spectra. The IR spectra of DUQ were calculated for all four possible protonation states, i.e., fully protonated (O1-H/O4-H), partially protonated in either 1- or 4-position (O1-H/O4; O1/O4-H), and fully deprotonated (O1/O4). For the oxidized DUQ, the calculated spectrum shows a good agreement for the most prominent bands in the experimental spectrum (Figure S5), implying that the calculated IR intensities provide a satisfactory description for the SEIRA (difference) intensities. Thus, it is justified the use of the calculated frequencies and IR intensities for determining the specific protonation pattern of the reduced DUQ formed during the enzymatic process (Figure 8). In fact, among the four calculated spectra (Figure 8A, B, D, E) that of the partially protonated quinol with the proton in 4-position provides the best description of the SEIRA difference spectrum obtained after addition of NADH to the enzyme-bound DUQ (Figure 8C). Specifically, the four intense experimental bands in the region 1500 - 1350 cm−1 are well reproduced by the calculated modes of O1/O4-H which involve bending coordinates of the ring carbon atoms and the two methoxy groups as well as small contributions of the C-OH or C-O− stretchings. The calculated spectra of the other protonation states fail to reproduce this intensity pattern. In particular, for the fully protonated quinol only one prominent band at 1428 cm−1 is predicted in this region. A good match with the calculated spectrum of O1-/O4H is also observed in the region of the ring and alkyl bending modes which give rise to a complex band pattern between 1130 and 1040 cm−1 in the experimental SEIRA spectrum, involving at least four overlapping bands. These bands can as well be correlated with the calculated modes of the O(4)-protonated species with frequency deviations of less than 15 cm−1. The only discrepancy seems to refer to the C-OH bending mode that is calculated at 14 ACS Paragon Plus Environment

Page 15 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

1204 cm−1. However, the lack of such a band in the experimental SEIRA difference spectra can readily be rationalized by the interference with the negative signal at 1205 cm−1 that reflects the depletion of the oxidized DUQ. Note that for fully protonated quinol two modes of considerable IR intensity are calculated at distinctly higher frequencies (1234 and 1217 cm−1) which cannot account for the cancellation of positive and negative signals in the experimental difference spectrum. Interestingly, the redox-linked IR spectra of ubiquinone (UQ2) at a similar pH reported by Hellwig et al. agree very well with the SEIRA-spectroscopic data presented here 26. On the other hand, the calculated spectra of the other (de)protonated states of the reduced DUQ, which all display one or two strong bands in the region between 1200 and 1250 cm−1 as well as in the region between 1300 and 1500 cm−1 differ significantly from the experimental SEIRA spectra. Thus, we conclude that the two-electron reduction of DUQ in EcNDH-2 is accompanied by the transfer of one proton. The incomplete charge compensation of the reduced DUQ possibly destabilizes the interaction with the protein such that it could be released to the solution phase more easily. Here, it would readily take up a second proton 32.

Cyclic voltammetry. CV experiments with DUQ and DDB at SAM-coated electrodes in the absence of EcNDH-2, display an apparent half-wave potential E1/2 app of +120 mV and +80 mV, respectively (Figure 9, blue traces). These values are close to E1/2 = +100 mV reported for ubiquinone at pH 7 in protic polar solvent 33. The peak separation in the CV trace of DUQ was found to be 420 mV. These data and the overall CV shape are consistent with those reported by Quan et al. for quinones in buffered solutions 33. In the presence of the immobilized EcNDH-2, the midpoint potential shifted to +90 mV and the peak separation decreased down to 140 mV (Figure 9A, black trace). The CV did not display a contribution from the characteristic redox peak pair of SAM-bound DUQ, indicating preferential binding to the immobilized enzyme. Only small changes were observed in the CV when NADH was added to the immobilized enzyme prior to the electrochemical measurements. The CV of DUQ displays a slightly more negative midpoint potential of +70 mV whereas the peak separation remains unchanged (Figure 9A, red trace). Both in the presence and in the absence of EcNDH-2, the peak current was found to be linearly dependent on the scan rate in the range from 5 to 500 mV/s (Figure S6A), which is typical for diffusionless electron transfer of adsorbed redox species 34. The decrease of E1/2 app of DUQ upon binding to the EcNDH-2 indicates a preferential stabilization of the oxidized state of DUQ. On the one hand, stabilizing the substrate would increase the energy of a transition 15 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 38

state being thus detrimental for the enzymatic activity. On the other hand, the concomitant decrease in peak separation in the CV traces shows that DUQ binding leads to faster reduction kinetics. Therefore, our CV measurements indicate that (immobilized) EcNDH-2 compensates somehow the disadvantageous thermodynamic stabilization of the substrate. DDB showed a completely different redox behavior in the absence and presence of the immobilized enzyme (Figure 9B). In the former case, a well-defined pair of anodic and cathodic peaks was found for the DDB redox activity (blue trace) with a E1/2 app of +80 mV and a peak separation of 55 mV. The peak current was linearly dependent on the square root of scan rate in a 1 to 500 mV/s range indicating a diffusion-controlled oxidation and reduction process at the electrode surface 35, 36 (Figure S6B). In the presence of EcNDH-2 (red trace), a strong reductive electrocatalytic current was observed which is consistent with the electrocatalytic reduction of quinone catalyzed by the protein. As inferred from the first derivative, the inflection point of this CV wave is at −80 mV, thus indicating that the active site and/or the redox mediator involved in the electrocatalytic process have such an E1/2 37. For the immobilized enzyme alone, no oxidation and reduction peaks could be detected in the CV (Figure 9B, grey trace). The lack of a non-turnover CV trace may be tentatively explained considering that EcNDH-2-SAM interactions do not provide a unique preferential orientation of the enzyme with respect to the electrode. As such, the population of EcNDH-2 molecules having an orientation competent for electron transfer that is fast enough to be detected in a non-turnover CV experiment, is probably too small to give a detectable FAD signal in the CV traces shown in Figure 9B (grey trace). However, in the presence of NADH (400 µM), an increase in current was observed above +110 mV compared to the enzyme alone or to NADH in the absence of protein (Figure S7). These findings indicate that the immobilized enzyme is able to catalyze NADH oxidation.

DISCUSSION

Enzymatic activities and protein-substrate interaction studies of EcNDH-2 were performed using two different quinones, DUQ and DDB. The structural difference between DUQ and DDB refers to the substituent at position C(6) which is a methyl group in DDB but a decyl chain in DUQ (Figure 1). This difference can account for different binding interactions of DUQ and DDB with EcNDH-2.


Page 17 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Protein-substrate interaction of EcNDH-2 in solution. The purified EcNDH-2 in the absence of detergent showed all the spectroscopic characteristics features of a flavoprotein, the reduction potential of the FAD was in the range typical for flavins cofactors (−220 ± 10 mV) and the enzyme was redox active interacting with both its substrates (NADH and quinone). The dissociation constants of DUQ and DDB were quite similar (6.1 µM and 8.7 µM, respectively). Also the catalytic activities of EcNDH-2 towards both quinones differ only by a factor of 2 with 8.1 and 17.5 µmol NADH/min⋅mg protein determined for DUQ and DDB, respectively. The rather similar dissociation constants and enzymatic activities in solution indicate that quinone binding affinities are comparable for both DUQ and DDB. The determined Vmax values are lower compared to the high activity of 106-190 mmol NADH/min⋅mg that was determined previously for EcNDH-2 38. This discrepancy may be attributed to the effect of a different environment provided by the phospholipid bilayer that allows for different molecular interactions of the bound enzyme accelerating the catalytic process. Nevertheless, the obtained values are similar to those previously calculated for Ec NDH-2 in the absence of lipids 39 and to those measured for many other reported NDH-2s, such as those from S. cerevisiae 13, 40, Yarrowia lipolytica 12, Caldalkalibacillus thermarum 9 and S. aureus 11.

Protein-substrate interaction in the immobilized EcNDH-2. In the protein-substrate interaction studies with immobilized enzyme, DUQ displays a quite different behavior as compared to the solution phase as it may form a tight enzyme-substrate complex. The specific orientation of the adsorbed EcNDH-2 may allow the long flexible alkyl chain of DUQ to interact with the protein-bound lipids which thus further stabilize the bound quinone and quinol. Alternatively, the flexible DUQ side chain may interact with the largely hydrophobic helices R392-H404 and L408-V426 of EcNDH-2, which also provide the most probable binding sites for lipid molecules. Both scenarios may also account for the changes of the 1743 cm−1 band that may be attributed to orientation changes of the C=O function of the proteinattached lipids. Also in the absence of immobilized EcNDH-2, the aliphatic side chain of DUQ may stabilize the adsorbed state, in this case via intercalation into the hydrophobic core of the SAM. Additional attractive forces are expected to arise from hydrogen bond formation between the carbonyl functions of DUQ and the amino head groups of the SAM which may lead to a more planar orientation of the ring with respect to surface. Such a molecular picture can account for the lack of any SEIRA signal whereas the CV indicates a surface-confined redox process. 17 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 38

Although the binding affinities of DUQ to the SAM and the immobilized EcNDH-2 cannot be quantified, DUQ evidently binds preferentially to the enzyme since no contributions from SAM-bound DUQ are detectable in the CV. Binding to the enzyme is particularly strong since removal of DUQ from the bulk solution (washing step) does not cause dissociation of the substrate-enzyme complex as the SEIRA intensity remains unchanged. Consistent with the above interpretation, DDB that lacks a hydrophobic tail only transiently binds to the immobilized EcNDH-2, corresponding to a low steady state concentration of the enzyme-substrate and enzyme-product complexes such that no well-defined difference signals can be obtained from the SEIRA measurements. This is also true for the interaction with the SAM in the absence of the enzyme as the CV indicates at diffusion-controlled process. The binding behavior of NADH seems to be in between that of DUQ and DDB. NADH binds more strongly than DDB since SEIRA signals were obtained both in the presence and in the absence of the immobilized enzyme. However, binding interactions are evidently weaker than in the case of DUQ since the washing step causes a distinct depletion of the SEIRA intensity. However, the differences in the SEIRA band positions and of the CV shapes are relatively small in the presence and in the absence of adsorbed enzyme. Thus, we conclude that in the presence of EcNDH-2 NADH may bind to both the enzyme and the SAM, in contrast to the preferential binding to the enzyme in the case of DUQ.

Electrocatalytic processes of the immobilized EcNDH-2. Catalytic reduction of quinones by the immobilized EcNDH-2 is not only observed using NADH as a reductant but also in electrochemical and potential-dependent SEIRA experiments, where the electrode replaces NADH as the electron donor. Since DUQ forms a stable and durable complex with the immobilized enzyme, reduction of this quinone could only be monitored under non-turnover conditions, i.e. in the absence of excess DUQ in the bulk solution. A potential change from +90 to −90 mV leads to the complete reduction of DUQ to its semi-protonated quinol form as demonstrated by the SEIRA spectra and the quantum chemical calculations. These spectra do not provide any indications for IR bands of the FAD cofactor which is presumably due to the unfavorable orientation of the isoalloxazine ring system with respect to the electrode. Compared to the CV in the absence of immobilized EcNDH-2, in the enzymatic reduction of DUQ the midpoint potential is slightly shifted by −30 mV to +90 mV and the peak separation is distinctly reduced by 280 mV to 140 mV, thus closely resembling the CVs reported for quinones in unbuffered solutions 33. This similarity may point to a somewhat restricted


Page 19 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

solvent accessibility of the DUQ binding site in the immobilized enzyme, which may reflect the tight binding of DUQ. In contrast, DDB only loosely binds to the immobilized EcNDH-2 such that in this case the enzymatic reduction could be monitored under turnover conditions (excess of DBB in the bulk solution). The catalytic current trace could be analyzed in terms of the redox potential of the FAD cofactor. The value of −80 mV derived from the inflection point, albeit still within the range observed for flavoproteins 20, is distinctly higher by about +140 mV than that determined here for the enzyme in solution (-220 mV). The upshift of the redox potential upon immobilization may be due to stabilization of the reduced flavin cofactor caused by the interfacial electric field at the electrode. Indeed, a similar large increase of the redox potential has been previously observed for another redox protein upon immobilization on SAM-coated electrodes 41. Also the oxidation of NADH itself is catalyzed by the immobilized EcNDH-2 as indicated by the increase of the oxidative current (above +310 mV) in the CV compared to NADH in the absence of the enzyme. While these results demonstrate that the EcNDH-2 is immobilized in an electrocatalytically competent manner, we were not able to detect oxidation and reduction of the FAD cofactor in CV experiments of the enzyme alone, possibly due to the interference by residual traces of oxygen that could not be avoided in the electrochemical cell.

General considerations. Two different catalytic mechanisms for NDH-2 have been proposed, a ping-pong mechanism and a mechanism involving the formation of a ternary complex. In the ping-pong mechanism the release of the first product is observed before binding of the second substrate; in the case of NDH-2 this would imply of NAD+ release prior to quinone binding. Usually the substrates intervening in a ping-pong mechanism share the same binding site, but a two-site ping-pong mechanism has also been described 42. In contrast, a ternary complex mechanism requires, by definition, two different substrate binding sites. Support for the “ping-pong” mechanism was derived from some kinetic analyses 12, 13 and from the first crystal structures (3.0 Å resolution) of Ndi1 with either NAD+ or UQ2 bound, which suggested the substrates occupied the same site in the protein 10. Feng et al. 8 succeeded to solve the crystal structure of Ndi1 with both substrates NADH and quinone bound at the same time (Ndi1-NADH-UQ4) providing strong support for the “ternary complex” mechanism, but which could not exclude the possibility for a two-site ping-pong mechanism. In fact, such a mechanism was proposed for the NDH-2 from M. tuberculosis 5, 43. Recently 19 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 38

the observation of the establishment of a charge transfer complex between the FAD and NAD+, which was only dissociated by the quinone strongly favored the “ternary complex” mechanism 11. The present results provide a new contribution to the debate on the catalytic mechanism of NDH-2. Thanks to the specific stabilization of the substrate-enzyme and product-enzyme complexes in the reaction of DUQ with the immobilized EcNDH-2, we were able to monitor the individual steps of the catalytic process spectroscopically. The results unambiguously demonstrate the existence of two different binding sites for DUQ and NADH and are only compatible with the “ternary complex” mechanism. This mechanism, albeit derived from the experiments on the immobilized enzyme-DUQ complexes, may hold also for other quinone substrates and for the enzymatic processes in solution where DUQ and DDB were found to show much weaker binding affinities, thereby ensuring a rapid turnover. In fact, the present fluorescence data of substrate binding to the enzyme in solution are consistent with two distinct binding sites. Additionally, a catalytic intermediate species could be identified (semiprotonated quinol) for the enzymatic reduction of DUQ in the immobilized state which gives a new view of how NDH-2 enzymatic reaction proceeds. Specifically, this finding suggests that the catalytic reduction of quinones by the immobilized EcNDH-2 involves the transfer of two electrons and one proton whereas the second proton is taken up from the solution phase. Thus, the electrode coated by an amino-terminated SAM is a useful platform for analyzing the catalytic mechanism but does not represent an adequate model for the natural enzymemembrane complex. In future studies, one should employ different immobilization strategies to mimic the physiological reaction conditions more closely. Thus, one may explore, whether or not, specific membrane interactions can modulate the enzymatic processes of NDH-2s.

ACKNOWLEDGMENTS

We would like to thank Lara Paulo for the first purification trials and potentiometric redox titrations.

SUPPORTING INFORMATION


Page 21 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

The supporting information is available free of charge on the ACS Publication website (http://pubs.acs.org).

REFERENCES

(1) (2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

Kerscher, S. J. (2000) Diversity and origin of alternative NADH:ubiquinone oxidoreductases. Biochim Biophys Acta 1459, 274-283. Kerscher, S. J., Eschemann, A., Okun, P. M., and Brandt, U. (2001) External alternative NADH:ubiquinone oxidoreductase redirected to the internal face of the mitochondrial inner membrane rescues complex I deficiency in Yarrowia lipolytica. J Cell Sci 114, 3915-3921. Seo, B. B., Kitajima-Ihara, T., Chan, E. K., Scheffler, I. E., Matsuno-Yagi, A., and Yagi, T. (1998) Molecular remedy of complex I defects: rotenone-insensitive internal NADH-quinone oxidoreductase of Saccharomyces cerevisiae mitochondria restores the NADH oxidase activity of complex I-deficient mammalian cells. Proc Natl Acad Sci U S A 95, 9167-9171. Marella, M., Seo, B. B., Nakamaru-Ogiso, E., Greenamyre, J. T., Matsuno-Yagi, A., and Yagi, T. (2008) Protection by the NDI1 gene against neurodegeneration in a rotenone rat model of Parkinson's disease. PLoS One 3, e1433. Yano, T., Li, L. S., Weinstein, E., Teh, J. S., and Rubin, H. (2006) Steady-state kinetics and inhibitory action of antitubercular phenothiazines on mycobacterium tuberculosis type-II NADH-menaquinone oxidoreductase (NDH-2). J Biol Chem 281, 11456-11463. Biagini, G. A., Fisher, N., Shone, A. E., Mubaraki, M. A., Srivastava, A., Hill, A., Antoine, T., Warman, A. J., Davies, J., Pidathala, C., Amewu, R. K., Leung, S. C., Sharma, R., Gibbons, P., Hong, D. W., Pacorel, B., Lawrenson, A. S., Charoensutthivarakul, S., Taylor, L., Berger, O., Mbekeani, A., Stocks, P. A., Nixon, G. L., Chadwick, J., Hemingway, J., Delves, M. J., Sinden, R. E., Zeeman, A. M., Kocken, C. H., Berry, N. G., O'Neill, P. M., and Ward, S. A. (2012) Generation of quinolone antimalarials targeting the Plasmodium falciparum mitochondrial respiratory chain for the treatment and prophylaxis of malaria. Proc Natl Acad Sci U S A 109, 8298-8303. Dong, C. K., Patel, V., Yang, J. C., Dvorin, J. D., Duraisingh, M. T., Clardy, J., and Wirth, D. F. (2009) Type II NADH dehydrogenase of the respiratory chain of Plasmodium falciparum and its inhibitors. Bioorg Med Chem Lett 19, 972-975. Feng, Y., Li, W., Li, J., Wang, J., Ge, J., Xu, D., Liu, Y., Wu, K., Zeng, Q., Wu, J. W., Tian, C., Zhou, B., and Yang, M. (2012) Structural insight into the type-II mitochondrial NADH dehydrogenases. Nature 491, 478-482. Heikal, A., Nakatani, Y., Dunn, E., Weimar, M. R., Day, C. L., Baker, E. N., Lott, J. S., Sazanov, L. A., and Cook, G. M. (2014) Structure of the bacterial type II NADH dehydrogenase: a monotopic membrane protein with an essential role in energy generation. Mol Microbiol 91, 950-964. Iwata, M., Lee, Y., Yamashita, T., Yagi, T., Iwata, S., Cameron, A. D., and Maher, M. J. (2012) The structure of the yeast NADH dehydrogenase (Ndi1) reveals overlapping binding sites for water- and lipid-soluble substrates. Proc Natl Acad Sci U S A 109, 15247-15252. 21 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(11)

(12)

(13)

(14)

(15)

(16)

(17) (18)

(19)

(20)

(21)

(22)

(23)

(24)

(25)

Page 22 of 38

Sena, F. V., Batista, A. P., Catarino, T., Brito, J. A., Archer, M., Viertler, M., Madl, T., Cabrita, E. J., and Pereira, M. M. (2015) Type-II NADH:quinone oxidoreductase from Staphylococcus aureus has two distinct binding sites and is rate limited by quinone reduction. Mol Microbiol 98, 272-288. Eschemann, A., Galkin, A., Oettmeier, W., Brandt, U., and Kerscher, S. (2005) HDQ (1-hydroxy-2-dodecyl-4(1H)quinolone), a high affinity inhibitor for mitochondrial alternative NADH dehydrogenase: evidence for a ping-pong mechanism. J Biol Chem 280, 3138-3142. Velazquez, I., and Pardo, J. P. (2001) Kinetic characterization of the rotenoneinsensitive internal NADH: ubiquinone oxidoreductase of mitochondria from Saccharomyces cerevisiae. Arch Biochem Biophys 389, 7-14. Yang, Y., Yamashita, T., Nakamaru-Ogiso, E., Hashimoto, T., Murai, M., Igarashi, J., Miyoshi, H., Mori, N., Matsuno-Yagi, A., Yagi, T., and Kosaka, H. (2011) Reaction mechanism of single subunit NADH-ubiquinone oxidoreductase (Ndi1) from Saccharomyces cerevisiae: evidence for a ternary complex mechanism. J Biol Chem 286, 9287-9297. Miyake, H., Ye, S., and Osawa, M. (2002) Electroless deposition of gold thin films on silicon for surface-enhanced infrared spectroelectrochemistry. Electrochem. Comm. 4, 973-977. Oesch, U., and Janata, J. (1983) Electrochemical study of gold electrodes with anodic oxide films- II. Inhibition of electrochemical redox reactions by monolayers of surface oxides. Electrochimica Acta 28, 1237-1253. Leopold, M. C., Black, J. A., and Bowden, E. F. (2002) Influence of Gold Topography on Carboxylic Acid Terminated Self-Assembled Monolayers. Langmuir. 18, 978–980. Pitera, J. W., Falta, M., and van Gunsteren, W. F. (2001) Dielectric properties of proteins from simulation: the effects of solvent, ligands, pH, and temperature. Biophys J 80, 2546-2555. Li, L., Li, C., Zhang, Z., and Alexov, E. (2013) On the Dielectric "Constant" of Proteins: Smooth Dielectric Function for Macromolecular Modeling and Its Implementation in DelPhi. J Chem Theory Comput 9, 2126-2136. Leger, C., Elliott, S. J., Hoke, K. R., Jeuken, L. J., Jones, A. K., and Armstrong, F. A. (2003) Enzyme electrokinetics: using protein film voltammetry to investigate redox enzymes and their mechanisms. Biochemistry 42, 8653-8662. Iwaki, M., Cotton, N. P., Quirk, P. G., Rich, P. R., and Jackson, J. B. (2006) Molecular recognition between protein and nicotinamide dinucleotide in intact, proton-translocating transhydrogenase studied by ATR-FTIR Spectroscopy. J Am Chem Soc 128, 2621-2629. Nadolny, C., and Zundel, G. (1997) Fourier transform infrared spectroscopic studies of proton transfer processes and the dissociation of Zn2+-bound water in alcohol dehydrogenases. Eur J Biochem 247, 914-919. Lamichhane, H., Wang, R., and Hastings, G. (2011) Comparison of calculated and experimental FTIR spectra of specifically labeled ubiquinones. Vib Spectrosc. 55, 279–286. Balakrishnan, G., Mohandas, P., and Umapathy, U. (1997) Ab initio studies on structure and vibrational spectra of ubiquinone and its radical anion. Spectrochimica Acta A 53, 1553–1561. Burie, J. R., Boussac, A., Boullais, C., Berger, G., Mattioli, T., Mioskowski, C., Nabedryk, E., and Breton, J. (1995) FTIR Spectroscopy of UV-Generated Quinone Radicals: Evidence for an Intramolecular Hydrogen Atom Transfer in Ubiquinone, Naphthoquinone, and Plastoquinone. J. Phys. Chem. 99, 4059–4070. 22 ACS Paragon Plus Environment

Page 23 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

(26)

(27)

(28) (29)

(30)

(31)

(32) (33)

(34) (35) (36)

(37)

(38)

(39)

(40)

(41)

Hellwig, P., Mogi, T., Tomson, F. L., Gennis, R. B., Iwata, J., Miyoshi, H., and Mantele, W. (1999) Vibrational modes of ubiquinone in cytochrome bo(3) from Escherichia coli identified by Fourier transform infrared difference spectroscopy and specific (13)C labeling. Biochemistry 38, 14683-14689. Ritter, M., Anderka, O., Ludwig, B., Mantele, W., and Hellwig, P. (2003) Electrochemical and FTIR spectroscopic characterization of the cytochrome bc1 complex from Paracoccus denitrificans: evidence for protonation reactions coupled to quinone binding. Biochemistry 42, 12391-12399. Barth, A. (2000) The infrared absorption of amino acid side chains. Prog Biophys Mol Biol 74, 141-173. Kozuch, J., Weichbrodt, C., Millo, D., Giller, K., Becker, S., Hildebrandt, P., and Steinem, C. (2014) Voltage-dependent structural changes of the membrane-bound anion channel hVDAC1 probed by SEIRA and electrochemical impedance spectroscopy. Phys Chem Chem Phys 16, 9546-9555. Hielscher, R., Wenz, T., Stolpe, S., Hunte, C., Friedrich, T., and Hellwig, P. (2006) Monitoring redox-dependent contribution of lipids in Fourier transform infrared difference spectra of complex I from Escherichia coli. Biopolymers 82, 291-294. Zhu, Z., and Gunner, M. R. (2005) Energetics of quinone-dependent electron and proton transfers in Rhodobacter sphaeroides photosynthetic reaction centers. Biochemistry 44, 82-96. Urban, P. F., and Klingenberg, M. (1969) On the redox potentials of ubiquinone and cytochrome b in the respiratory chain. Eur J Biochem 9, 519-525. Quan, M., Sanchez, D., Wasylkiw, M. F., and Smith, D. K. (2007) Voltammetry of quinones in unbuffered aqueous solution: reassessing the roles of proton transfer and hydrogen bonding in the aqueous electrochemistry of quinones. J Am Chem Soc 129, 12847-12856. Laviron, E. (1979) General expression of the linear potential sweep voltammogram in the case of diffusionless electrochemical systems. J. Electroanal. Chem. 101, 19-28. Randles , J. E. B. (1948) A cathode ray polarograph. Part II.—The current-voltage curves. Trans. Faraday Soc. 44, 327-338. Millo, D., Bonifacio, A., Ranieri, A., Borsari, M., Gooijer, C., and van der Zwan, G. (2007) Voltammetric and surface-enhanced resonance Raman spectroscopic characterization of cytochrome C adsorbed on a 4-mercaptopyridine monolayer on silver electrodes. Langmuir 23, 4340-4345. Heering, H. A., Hirst, J., and Armstrong, F. A. (1998) Interpreting the Catalytic Voltammetry of Electroactive Enzymes Adsorbed on Electrodes. J. Phys. Chem. B 102, 6889–6902. Bjorklof, K., Zickermann, V., and Finel, M. (2000) Purification of the 45 kDa, membrane bound NADH dehydrogenase of Escherichia coli (NDH-2) and analysis of its interaction with ubiquinone analogues. FEBS Lett 467, 105-110. Villegas, J. M., Volentini, S. I., Rintoul, M. R., and Rapisarda, V. A. (2011) Amphipathic C-terminal region of Escherichia coli NADH dehydrogenase-2 mediates membrane localization. Arch Biochem Biophys 505, 155-159. de Vries, S., and Grivell, L. A. (1988) Purification and characterization of a rotenoneinsensitive NADH:Q6 oxidoreductase from mitochondria of Saccharomyces cerevisiae. Eur J Biochem 176, 377-384. Rivas, L., Soares, C. M., Baptista, A. M., Simaan, J., Di Paolo, R. E., Murgida, D. H., and Hildebrandt, P. (2005) Electric-field-induced redox potential shifts of tetraheme cytochromes c3 immobilized on self-assembled monolayers: surface-enhanced resonance Raman spectroscopy and simulation studies. Biophys J 88, 4188-4199. 23 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(42) (43)

(44)

(45)

Page 24 of 38

Cook, P. F., and Cleland, W. W. (2007) Enzyme kinetics and mechanism, 1st ed., Garland Science, New York. Yano, T., Rahimian, M., Aneja, K. K., Schechter, N. M., Rubin, H., and Scott, C. P. (2014) Mycobacterium tuberculosis type II NADH-menaquinone oxidoreductase catalyzes electron transfer through a two-site ping-pong mechanism and has two quinone-binding sites. Biochemistry 53, 1179-1190. Hielscher, R. (2009) The role of lipids and nucleotides on the catalytic mechanism of proteins from the respiratory chain: an electrochemical and infrared spectroscopic approach, Albert-Ludwigs-Universität Freiburg im Breisgau. Bauscher, M., and Maentele, W. (1992) Electrochemical and infrared-spectroscopic characterization of redox reactions of p-quinones. J. Phys. Chem. 96, 11101–11108.


Page 25 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

TABLES

Table 1. Assignment of IR bands of NADH measured in the presence and absence of EcNDH-2 without NDH-2 / cm-1 1076 1114 1183 1219 1248 1303 1337 1421 1480 1542 1580 1606 1654 1689 a

with NDH-2 / cm−1 1075 1114 1183 1221 1250 1303 1337 1423 1480 1537 1580 1606 1654 1689

in solution / cm−1 1075a 1114a 1183a 1228b / 1236a 1252b / 1236a 1304a 1336a 1421a 1481a 1546a 1580a 1606a 1649a,b 1688a

modec

moiety

O-P-O

pyrophosphate

ν(P=O)

pyrophosphate

ν(C-N)

nicotinamide adenine ring

δ(NH2) ν(C=O)

adenine ring adenine ring 3-carboxamide

Taken from ref. 21; b Taken from ref. 44; c v: stretching, δ: bending


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 38

Table 2. Assignment of the IR bands of the protein-bound DUQ peak position / cm−1 oxidized 1027 1065 1100 1158 1205 1269 1379 1436 1458 1611 1652 1663 reduced 1051 1094

in solution / cm−1

moded

moiety

1030a 1065a 1102a 1167a 1210a 1264a 1379a 1433a / 1436c 1454c / 1473a 1610b / 1614a 1649a / 1650b 1659a

δ(OCH3) δ(OCH3) δ(CH2) / δ(CH3) δ(CH2) / δ(CH3)

methoxy groups methoxy groups alkyl groups alkyl groups

ν(C=C) v(C=O) ν(C=O)

quinone ring quinone ring quinone ring

ω(CH2)e / ω(CH3)e ω(CH2)e +δ(C-C-C)e δ(C-C-C)e

1117 1389 1431 1468

1388c 1432c 1470c

1493

1490c

δ(OCH3)e

alkyl groups alkyl groups quinol ring quinol ring methoxy groups quinol ring quinol ring

+ δ(C-C-C)e a

decyl Q0, taken from ref. 25; b UQ10, taken from ref. 45; c UQ2 taken from ref. 26; d ν: stretching, δ: bending, ω: twisting or wagging; e this work


Page 27 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

FIGURE LEGENDS Figure 1. (A) Two-electron transfer reaction from NADH to a quinone (UQ) catalyzed by EcNDH-2. R: adenosine dinucleotide moiety of NADH. UQH2 is the corresponding quinol formed from UQ. (B) Structure of 2,3-dimethoxy-5-methyl-6-decyl-1,4-benzoquinone (decylubiquinone, DUQ) and 2,3-dimethoxy-5,6-dimethyl-1,4-benzoquinone (DDB) that function as electron acceptors for EcNDH-2.

Figure 2. Redox titration curve for Type II NADH: quinone oxidoreductase from E. coli. The data were obtained at 445 nm which corresponds to the maximum absorption of FAD. EcNDH-2 reduction is represented by closed circles and oxidation by open circles. The solid line was obtained by fitting a two-electron Nernst function to the data.

Figure 3. Fluorescence spectra of EcNDH-2 with excitation at 280 nm, in the absence (solid line) and in the presence of 100 µM NAD+ (dashed line) and/or 100 DDB µM (dotted line) (A and B). Change in the fluorescence at 330 nm with excitation at 280 nm of EcNDH-2 by sequential addition of NAD+ (C), DUQ (D) and DDB (E).

Figure 4. SEIRA difference spectra of the immobilized EcNDH-2 after successive addition of (A) 400 µM NADH, (B) 200 µM DUQ, (C) again 400 µM NADH, and (D) after washing the cell with buffer (20 mM TRIS-HCl, pH 7.5) to remove excess NADH from the solution. The intensity scale of trace A is reduced by a factor of 2. The washing step was carried out after each substrate addition but in most cases (see text for details) had no effect on the substrateinduced spectra. The assignment of the bands is discussed in the text and listed in Tables 1 and 2. The experiments were carried out in the presence of a 20 mM TRIS-HCl buffer at pH 7.5 and the cell was purged with argon during all experiments to avoid the presence of oxygen. Figure 5. SEIRA difference spectra of (A) NAD+ and (B) NADH in 20 mM TRIS-HCl buffer at pH 7.5, after addition to the buffer solution (400 µM) in contact with the SAM-coated electrode (open circuit) The intensity scale of trace A is reduced by a factor of 3. Asterisks mark vibrational peaks that indicate the redox state of the dinucleotide. The spectrum for NADH remained unchanged for at least 2 h, while spectrum B transformed to a broad featureless spectrum within 60 min, likely due to the loose attachment and release of the molecule to the surface. 27 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 38

Figure 6. SEIRA difference spectra of the immobilized EcNDH-2 after successive addition of (A) 400 µM NAD+, measured 3 minutes (grey trace) and 60 minutes (black trace) after addition, (B) 200 µM DUQ, (C) again 400 µM NAD+, and (D) 400 µM NADH. After each step, the cell was washed with buffer (20 mM TRIS-HCl, pH 7.5). The assignments of the bands are discussed in the text and listed in Tables 1 and 2. The experiments were carried out in the presence of a 20 mM TRIS-HCl buffer at pH 7.5 and the cell was purged with argon during all experiments to avoid the presence of oxygen.

Figure 7. SEIRA difference spectra of immobilized EcNDH-2 after subsequent addition of NADH and DUQ as described in Figure 4, (A) at open circuit potential (ca. +90 mV), (B) after switching to a potential of −90 mV, (C) after switching back to the open circuit potential, and (D) after subsequent addition of 400 µM NADH.

Figure 8. Comparison of the experimental NADH-induced SEIRA difference spectrum of EcNDH-2 with DUQ bound (black trace, see Figures 3C and 5C) with the calculated IR spectra of the reduced DUQ with different protonation patterns at the positions 1 and 4, protonated at both oxygens (O1-H,O4-H), only at one oxygen (O1,O4-H, and O1-H,O4) and no protonation (O1,O4) as depicted in the structural formula on the left side. R1 and R2 are methoxy- and decyl- substituents, respectively.

Figure 9. Cyclic voltammetry of (A) DUQ and (B) DDB in the presence and absence of EcNDH-2 bound to a SAM-coated Au electrode. (A) blue trace; DUQ (200 µM) in the absence of EcNDH-2 and NADH; black trace, DUQ (200 µM) in the presence of immobilized EcNDH-2 but in the absence of NADH; red trace, DUQ (200 µM) after treating the immobilized EcNDH-2 with NADH. (B) blue trace; DDB (200 µM) in the absence of EcNDH-2 and NADH; red trace, DDB (200 µM) in the presence of immobilized EcNDH-2; grey trace, immobilized EcNDH-2 in the absence of any substrate. The current was divided by the geometrical Au electrode surface area to obtain current density. Experimental conditions: buffer was 20 mM TRIS-HCl at pH 7.5 and scan rates were 50 and 10 mV/s for (A) and (B), respectively.


Page 29 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Figure 1. (A) Two-electron transfer reaction from NADH to a quinone (UQ) catalyzed by EcNDH-2. R: adenosine dinucleotide moiety of NADH. UQH2 is the corresponding quinol formed from UQ. (B) Structure of 2,3-dimethoxy-5-methyl-6-decyl-1,4-benzoquinone (decylubiquinone, DUQ) and 2,3-dimethoxy-5,6dimethyl-1,4-benzoquinone (DDB) that function as electron acceptors for EcNDH-2. 66x55mm (300 x 300 DPI)


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2. Redox titration curve for Type II NADH: quinone oxidoreductase from E. coli. The data were obtained at 445 nm which corresponds to the maximum absorption of FAD. EcNDH-2 reduction is represented by closed circles and oxidation by open circles. The solid line was obtained by fitting a twoelectron Nernst function to the data. 55x36mm (300 x 300 DPI)


Page 30 of 38

Page 31 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Figure 3. Fluorescence spectra of EcNDH-2 with excitation at 280 nm, in the absence (solid line) and in the presence of 100 µM NAD+ (dashed line) and/or 100 DDB µM (dotted line) (A and B). Change in the fluorescence at 330 nm with excitation at 280 nm of EcNDH-2 by sequential addition of NAD+ (C), DUQ (D) and DDB (E). 160x148mm (300 x 300 DPI)


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4. SEIRA difference spectra of the immobilized EcNDH-2 after successive addition of (A) 400 µM NADH, (B) 200 µM DUQ, (C) again 400 µM NADH, and (D) after washing the cell with buffer (20 mM TRISHCl, pH 7.5) to remove excess NADH from the solution. The intensity scale of trace A is reduced by a factor of 2. The washing step was carried out after each substrate addition but in most cases (see text for details) had no effect on the substrate-induced spectra. The assignment of the bands is discussed in the text and listed in Tables 1 and 2. The experiments were carried out in the presence of a 20 mM TRIS-HCl buffer at pH 7.5 and the cell was purged with argon during all experiments to avoid the presence of oxygen. 117x161mm (300 x 300 DPI)


Page 32 of 38

Page 33 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Figure 5. SEIRA difference spectra of (A) NAD+ and (B) NADH in 20 mM TRIS-HCl buffer at pH 7.5, after addition to the buffer solution (400 µM) in contact with the SAM-coated electrode (open circuit) The intensity scale of trace A is reduced by a factor of 3. Asterisks mark vibrational peaks that indicate the redox state of the dinucleotide. The spectrum for NADH remained unchanged for at least 2 h, while spectrum B transformed to a broad featureless spectrum within 60 min, likely due to the loose attachment and release of the molecule to the surface. 120x170mm (300 x 300 DPI)


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 6. SEIRA difference spectra of the immobilized EcNDH-2 after successive addition of (A) 400 µM NAD+, measured 3 minutes (grey trace) and 60 minutes (black trace) after addition, (B) 200 µM DUQ, (C) again 400 µM NAD+, and (D) 400 µM NADH. After each step, the cell was washed with buffer (20 mM TRISHCl, pH 7.5). The assignments of the bands are discussed in the text and listed in Tables 1 and 2. The experiments were carried out in the presence of a 20 mM TRIS-HCl buffer at pH 7.5 and the cell was purged with argon during all experiments to avoid the presence of oxygen. 117x162mm (300 x 300 DPI)


Page 34 of 38

Page 35 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Figure 7. SEIRA difference spectra of immobilized EcNDH-2 after subsequent addition of NADH and DUQ as described in Figure 4, (A) at open circuit potential (ca. +90 mV), (B) after switching to a potential of −90 mV, (C) after switching back to the open circuit potential, and (D) after subsequent addition of 400 µM NADH. 118x166mm (300 x 300 DPI)


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 8. Comparison of the experimental NADH-induced SEIRA difference spectrum of EcNDH-2 with DUQ bound (black trace, see Figures 3C and 5C) with the calculated IR spectra of the reduced DUQ with different protonation patterns at the positions 1 and 4, protonated at both oxygens (O1-H,O4-H), only at one oxygen (O1,O4-H, and O1-H,O4) and no protonation (O1,O4) as depicted in the structural formula on the left side. R1 and R2 are methoxy- and decyl- substituents, respectively. 107x135mm (300 x 300 DPI)


Page 36 of 38

Page 37 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Figure 9. Cyclic voltammetry of (A) DUQ and (B) DDB in the presence and absence of EcNDH-2 bound to a SAM-coated Au electrode. (A) blue trace; DUQ (200 µM) in the absence of EcNDH-2 and NADH; black trace, DUQ (200 µM) in the presence of immobilized EcNDH-2 but in the absence of NADH; red trace, DUQ (200 µM) after treating the immobilized EcNDH-2 with NADH. (B) blue trace; DDB (200 µM) in the absence of EcNDH-2 and NADH; red trace, DDB (200 µM) in the presence of immobilized EcNDH-2; grey trace, immobilized EcNDH-2 in the absence of any substrate. The current was divided by the geometrical Au electrode surface area to obtain current density. Experimental conditions: buffer was 20 mM TRIS-HCl at pH 7.5 and scan rates were 50 and 10 mV/s for (A) and (B), respectively. 103x153mm (300 x 300 DPI)


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

TOC 36x28mm (300 x 300 DPI)


Page 38 of 38

SubstrateâProtein Interactions of Type II NADH:Quinone

Recommend Documents

SubstrateâProtein Interactions of Type II NADH:Quinone