FEBS Letters 589 (2015) 2050–2057

journal homepage: www.FEBSLetters.org

An electrogenic nitric oxide reductase Sinan Al-Attar, Simon de Vries ⇑ Department of Biotechnology, Delft University of Technology, Julianalaan 67, 2628 BC, The Netherlands

a r t i c l e

i n f o

Article history: Received 20 April 2015 Revised 24 June 2015 Accepted 24 June 2015 Available online 3 July 2015 Edited by Stuart Ferguson Keywords: Nitric oxide reductase Nitric oxide Denitrification Electrogenic Membrane potential

a b s t r a c t Nitric oxide reductases (Nors) are members of the heme-copper oxidase superfamily that reduce nitric oxide (NO) to nitrous oxide (N2O). In contrast to the proton-pumping cytochrome oxidases, Nors studied so far have neither been implicated in proton pumping nor have they been experimentally established as electrogenic. The copper-A-dependent Nor from Bacillus azotoformans uses cytochrome c551 as electron donor but lacks menaquinol activity, in contrast to our earlier report (Suharti et al., 2001). Employing reduced phenazine ethosulfate (PESH) as electron donor, the main NO reduction pathway catalyzed by CuANor reconstituted in liposomes involves transmembrane cycling of the PES radical. We show that CuANor reconstituted in liposomes generates a proton electrochemical gradient across the membrane similar in magnitude to cytochrome aa3, highlighting that bacilli using CuANor can exploit NO reduction for increased cellular ATP production compared to organisms using cNor. Ó 2015 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

1. Introduction Biological and chemical N2 fixation are pivotal for the supply of terrestrial life with the indispensible element nitrogen. In denitrification, the nitrogen oxides are successively reduced back to molecular nitrogen (Eq. (1)) employing four specific reductases for the reduction of nitrate (Nar), nitrite (Nir), NO (Nor) and N2O (N2Or), respectively [1,2]:

Nitrate ðNO3 Þ ! Nitrite ðNO2 Þ ! Nitric Oxide ðNOÞ ! Nitrous Oxide ðN2 OÞ ! dinitrogen ðN2 Þ

ð1Þ

Prokaryotic Nors are integral membrane enzymes (distinct from the soluble P450 eukaryotic Nors [3,4]), which catalyze the N–N bond formation between two NO molecules consuming two protons and two electrons and producing one N2O and one water (Eq. (2))[1,2,5–7]:

2NO þ 2Hþ þ 2e ! N2 O þ H2 O

ð2Þ

Author contributions: The study was conceived and supervised by SdV; SdV and SA designed the experiments; SA performed and analyzed the experiments; SA and SdV wrote the manuscript. All the authors that contributed to this manuscript have approved for submission. The manuscript describes original unpublished work which is not considered for publication elsewhere.The Organization for Scientific Research (NWO) from the Netherlands financed this research. It had no influence on the experiments conducted, data interpretation or manuscript preparation and submission. ⇑ Corresponding author. Fax: +31 152782355. E-mail address: [email protected] (S. de Vries).

Sequence analysis classifies Nors as members of the heme-copper oxidase (HCO) superfamily [3,8–11]. The proton pumping quinol and cytochrome c oxidases (Coxs) are the other members of the HCO superfamily. Both enzyme families share a similar catalytic core, which consists of a low-spin heme (b in Nors and a or b in Coxs) and a binuclear metal center (BNC) where the NO or O2 reduction chemistries take place [2,6,7,12–15]. The BNC in Nors contains a high-spin heme b3 [6,13–15] and a non-heme iron designated FeB. In Coxs the BNC consists of CuB and a high-spin heme of type a3, o3 or b3 [12,16]. The members of the HCO superfamily show great overall structural similarity in their membrane-buried catalytic subunit that contains 12–14 transmembrane a-helices (TMHs) and six conserved histidine residues that coordinate the redox centers [10]. Coxs and Nors show cross-reactivity with their respective electron acceptors (O2 and NO, resp.) [17–20] A notable kinetic property of all Nors is that the activity is quadratic with respect to NO concentration yielding non-Michaelis–Menten behavior [6,12,21]. In addition, Nors are substrate-inhibited at low micromolar NO concentrations resulting in sigmoidal reaction progress curves [6,12,21]. Three prokaryotic classes of Nor have been identified: cytochrome c oxidizing cNor [13,22,23]; quinol-dependent qNor [24–26] and the Nor from Bacillus azotoformans (qCuANor) reported to receive electrons from menaquinol or cytochrome c551 [3,27]. However, here we report that the Bacillus enzyme in fact lacks menaquinol activity (see Section 3) and have changed its name from qCuANor to CuANor.

http://dx.doi.org/10.1016/j.febslet.2015.06.033 0014-5793/Ó 2015 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

2051

S. Al-Attar, S. de Vries / FEBS Letters 589 (2015) 2050–2057

The structure of the heterodimeric cNor from the Gram-negative Pseudomonas aeruginosa has recently been solved [28]. Computational, mutational and structural analyses of Pseudomonas and Paracoccus cNors (52% sequence identity [29]), suggested that the enzyme takes both electrons and protons from the periplasm rendering it non-electrogenic [28–31], which was borne out experimentally [22,29,31]. In contrast, the crystal structure of the quinol-dependent qNor from the Gram-positive Geobacillus stearothermophilus [26] revealed the presence of a putative proton transfer pathway between the cytoplasmic aqueous phase and the BNC suggesting that NO reduction by qNor could be electrogenic. CuANor has so far only been found in bacilli [11,21,27]. Biochemical and protein N-terminal and genomic sequence analyses show that the enzyme from B. azotoformans is hetero-trimeric [8,21,27]. CuANor accepts electrons from the endogenous cytochrome c551 [27], which is negatively charged (pI  3, Em = 140 mV). CuANor contains a di-copper site (CuA) [21] similar to cytochrome c dependent Coxs [32]. Like cNor and qNor the enzyme has one heme b and a BNC consisting of heme b3 and FeB [21]. Among the HCO superfamily, Coxs are established as proton pumps conserving energy in the form of a proton electrochemical gradient by employing two processes [32–34]. In the first process, the four protons needed for water formation (chemical protons) are taken up from the negative side of the membrane (e.g. cytoplasm) while electrons enter the enzyme from the positive side of the membrane (periplasm). The second process involves protons being pumped across the membrane from the negative side to the positive side (Eq. (3)) [32,35]:

O2 þ ð4 þ nÞHþcyto ! 2H2 O þ nHþperi

ð3Þ

Here n is the number of protons pumped (n = 4 for Type A and n = 2–4 for Types B and C Cox) [36,37]. The subscripts cyto and peri indicate the cytoplasmic (negative) and periplasmic (positive) sides of the membrane, respectively. In this study the ability of CuANor from B. azotoformans to couple NO reduction to the formation of a membrane potential was investigated. To this end we used protein-reconstituted liposomes. Our results provide the first experimental evidence that NO reduction by CuANor in closed liposomes creates a proton electrochemical gradient indicating the energy conserving capability of the enzyme. 2. Materials and methods

liposomes were extruded 15 times over a membrane with 0.2 lm pore size (Nuclepore Track-Etch Membrane, Whatman). The vesicle suspension was diluted to 5 mg/mL and incubated for 30 min with 6 mM LM to destabilize the vesicles before the addition of 0.1 lM or 0.5 lM of cytochrome aa3 or CuANor or cNor. In the case of the mixed aa3/CuANor-PLs or aa3/cNor-PLs, 0.5 lM of each enzyme was added. The enzyme-containing liposome suspension was incubated for 1 h. Methanol-washed Bio-Beads (Bio-Rad SM-2) were used to adsorb the detergent in two successive steps (300 and 150 mg Bio-Beads/mL liposomes for 1, 5 h each step, respectively). The phospholipid and enzyme contents were determined in the final PL preparation from the phosphate content [39,40] and the Soret absorbance [6,27], respectively. Using cytochrome c the orientation of cNor was determined at 40 ± 5% right-side out a value similar to CuANor (see Section 4). 2.3. Activity measurements The activity of CuANor-PLs and cNor-PLs was monitored optically from the formation of oxidized PES (e387 nm = 26 mM1 cm1). At pH 7.6, reduced PES (PESH, estimated pK 3.7 [41]) is a two-electron plus one proton donor, which we confirmed experimentally. Thus the NO consumption rate was calculated according to Eq. (4):

PESH þ Hþ þ 2NO ! PESþ þ N2 O þ H2 O

ð4Þ

2.4. Determination of the respiratory control ratio (RCR) The Vmax values were determined by simulation of the time course of the reaction according to the model (Eq. (5)) proposed earlier [6]:

v ½E

¼

1 þ K2 



V max

1 ½NO

 K1 þ ½NO þ ½NO 2 K

ð5Þ

i

where K1 and K2 are the apparent binding constants for first en second NO molecules. The Ki term is the inhibition constant proposed to represent the binding of NO to the oxidized form of the enzyme [6]. The respiratory control ratio (RCR) was calculated by dividing the Vmax values of the uncoupled PLs by that of the coupled PLs. The PLs were assayed in 20 mM HEPES + 45 mM KCl + 45 mM sucrose, pH 7.6. For complete uncoupling, 40 lM CCCP was added with or without 50 lM valinomycin. The RCR for cytochrome aa3 PLs (aa3-PLs) was determined based on the decrease in absorbance of reduced cytochrome c at 550 nm.

2.1. Chemicals 2.5. Measurement of the membrane potential (Dw) Soybean asolectin was purchased from Fluka BioChemika, N-Dodecyl b-D-maltoside (LM) from Affymetrix and equine heart cytochrome c, Carbonyl cyanide 3-chlorophenylhydrazone (CCCP), valinomycin, phenazine ethosulfate (PES), safranine O and phenol red from Sigma–Aldrich. Saturated NO solutions (2 mM) were prepared by flushing anaerobic MilliQ water with 100% NO gas at room temperature. 2.2. Preparation of proteoliposomes (PLs) Expression and purification of CuANor, cytochrome aa3 and cNor from Paracoccus denitrificans were performed as described in refs. [6,21,38], respectively. Soybean asolectin was dissolved in chloroform and dried with nitrogen gas to form a thin phospholipid film. The film was hydrated with 1 mL of 0.1 M K+-HEPES pH 7.4 per 10 mg phospholipids. The resulting liposomes were fractured using 4 cycles of freezing in liquid nitrogen and thawing (75 °C). To obtain a uniform suspension of small unilamellar vesicles, the

Fluorescence quenching of safranine O [42,43] was used to monitor the formation of Dw by the PLs. The assay with CuANor-PLs and cNor-PLs contained 5 lM safranine O, 40 mM sucrose, 40 mM KCl, 60 lM PES+ and 20 mM ascorbate, pH 7.6. CuANor-PLs were also assayed without ascorbate in the presence of 60 lM PESH (with or without addition of 60 lM or 1 mM PES+). The reaction was started with 10, 20 or 40 lM NO. The Dw formation by aa3-PLs was monitored in the same buffer without PES using 10 lM equine heart cytochrome c as electron donor and 260 lM O2 as electron acceptor in the presence of 20 mM ascorbate to prevent inner-filter quenching effects on the fluorescence by cytochrome c. The measurements were carried out in a Cary Eclipse (Varian) fluorescence spectrophotometer at room temperature using excitation and emission wavelengths of 495 and 585 nm, respectively. Additions were made after removing the cuvette from the fluorescence spectrophotometer. The reaction buffer was mixed thoroughly using a magnetic stirrer prior to

2052

S. Al-Attar, S. de Vries / FEBS Letters 589 (2015) 2050–2057

Fig. 1. The crystal structure of T. thermophilus cytochrome ba3 (left) and the homology model of CuANor (right). Subunits I are colored gold, subunits II titanium and proton transfer pathways light blue. The histidines (black sticks) that coordinate the hemes (red lines), CuB (cyan sphere), FeB (orange sphere) and CuA (purple spheres) are all conserved in both structures. The ligands in CuANor of the low-spin heme b are His77 and His394; the high-spin heme b3 are His92; the non-heme FeB are His241, His290 and His291 and the CuA ligating residues are SUII: His100, Cys135, Cys139 and His143. The K-pathway analog (white arrow) residues of cytochrome ba3 (Thr315, Tyr248, Tyr244, Thr312, Ser309, Tyr237 and Glu15(SUII)) and CuANor (Tyr256, Tyr248, Thr320, Ser317 and Asn245) are shown as sticks. The active-site residues (Tyr237 and Asn245 in cytochrome ba3 and CuANor, respectively) are viewed as red sticks. Orange sticks indicate non-homologous amino acid substitutions (Thr315 and Ala323). Three putative residues involved in the proton exit pathway (black dotted arrow) in cytochrome ba3 (Asp372, His376 and Asp287) were also found in the CuANor homology model (Asp380, His384 and Asp295).

returning the cuvette to the fluorescence spectrophotometer, which took 10 s. The fluorescence spikes due to these manipulations are omitted from Fig. 4 for clarity. 2.6. Homology modeling Homology modeling of CuANor subunit I (cbaA1) and subunit II (cbaB1) from B. azotoformans [8] was carried out on the SWISS-Model server [44,45] using default modeling parameters and as templates the structures of subunits I and II of Thermus thermophilus ba3 cytochrome c oxidase SUI (PDB: 3eh4.1, 34% sequence identity) and SUII (PDB: 3eh4.1, 41% sequence identity), respectively. We have incorporated the cofactors into the finished homology model at exactly the same cofactor-ligand distances as in cytochrome ba3. The models were viewed using PyMOL Molecular Graphics System, Schrödinger, LLC. 2.7. Graphing and analysis Igor Pro version 6.1 (Wavemetrics) was used for data fitting and creating graphs. 3. Results 3.1. Homology modeling In order to obtain insight into the structure of CuANor, we created a homology model of the enzyme based on the crystal

structure of T. thermophilus cytochrome ba3 (T. thermophilus cytochrome ba3 numbering is used unless otherwise noted). The model of SUI (Fig. 1) shows 12 trans-membrane a-helices (TMHs) and one TMH in SUII, which is identical to cytochrome ba3 and cNor [28] and one TMH less than qNor [26]. In Type A Cox, the protons are transported by the D- and K-pathways. The former pathway is used solely for the pumped protons whereas the K-pathway is exploited for the transfer of the chemical protons [36]. Cytochrome ba3 (Type B) uses a single proton transfer pathway analogous to the K-pathway to deliver both chemical and pumped protons [37,46,47]. The cytochrome ba3 K-pathway analog (Tyr237, Tyr244, Tyr248, Ser309, Thr312, Thr315 in SUI and Glu15 in SUII) and that in CuANor (Fig. 1) share five out of seven amino-acid residues (Tyr248, Tyr256, Ser317, Thr320 and SUII-Glu5, CuANor numbering). Tyr244 in cytochrome ba3 is substituted by Val252 in CuANor. However Tyr248 in CuANor, which is located one turn higher in the same a-helix as Val252 is located at the same position as Tyr244 in cytochrome ba3. Ala323 was found in CuANor instead of Thr315. Although alanine lacks a polar side-chain its backbone carbonyl could fulfill a role in proton transfer. The strictly conserved ‘‘active-site’’ tyrosine (Tyr237) in Coxs, which is cross-linked to one of the histidine residues (His233) that coordinate CuB [48] is substituted for Asn245 in CuANor which according to the homology model is 6.7 Å away from FeB. In the crystal structures of cNor and qNor a Glu residue is found at this position and is established as a FeB ligand in cNor [28] but not in qNor [26]. Whether Asn245 is a (weak) ligand to FeB cannot be asserted from the homology modeling. A putative

S. Al-Attar, S. de Vries / FEBS Letters 589 (2015) 2050–2057

periplasmic proton exit pathway for pumped protons suggested for cytochrome ba3 [47,49] is also found in the homology model of CuANor (Asp380, His384 and Asp295) (Fig. 1). Structural inspection and simulation of menaquinol and HQNO binding to the CuANor homology model (SwissDock, [50]) did not reveal a putative quinol binding domain. We therefore reassessed the menaquinol activity of CuANor [27] and found that neither menaquinol nor ascorbic acid could serve as direct electron donors to CuANor. Hence, we propose to change the previous name of the enzyme, qCuANor [21] to CuANor. The activity observed previously [21] is artifactual and caused by traces of PES in the measuring chamber used for polarographic NO determination [51]. A very low KM for PESH (23 ± 3 nM, Supplementary Fig. S1) was determined and rapid reduction by ascorbate or menaquinol of the nanomolar concentrations of PES were found to produce significant CuANor dependent NO reductase activity. The considerable structural similarity between CuANor and Type B oxidases established by homology modeling strongly suggests that CuANor is functionally capable of electrogenic NO reduction employing the K-channel analog for proton uptake to the BNC. 3.2. CuANor generates a proton-dependent electrochemical gradient The endogenous B. azotoformans membrane-bound cytochrome c551 is the natural substrate for CuANor [27]. A soluble construct of cytochrome c551 that lacks the membrane-anchor was expressed in Escherichia coli but proved inactive as substrate. Ideally a membrane-impermeable electron donor is used to select only for the proteoliposomes with the desired right-side out enzyme orientation. Impermeant electron donors such as sodium dithionite, indigo carmine and anthraquinone-2,6-disulfonic acid proved not suitable due to their high reactivity with NO. We were therefore limited to the use of PESH as electron donor in this study despite its membrane-permeability [52,53]. To investigate the ability of CuANor in generating a membrane potential, CuANor-PLs, aa3-PLs, and mixed aa3/CuANor-PLs and aa3/cNor-PLs were prepared. NO and O2 reductase activities with and without the uncoupler CCCP showed RCR values of 2.4–5.5 (Table 1 and Fig. 2) indicating the tightness of the liposomes. Valinomycin when used with CCCP did not significantly increase the RCR value of the CuANor-PLs or aa3-PLs (Table 1). The RCR value of the non-electrogenic cNor in aa3/cNOR-PLs was close to 1, in agreement with [22], whilst the PLs were coupled indicated by the RCR value of 4.3 for O2 reduction by the co-reconstituted cytochrome aa3 (Table 1). Simulation of the ‘‘coupled’’ and ‘‘uncoupled’’ activity traces (Fig. 2) returned the same values for the binding constants (K1 = 2 ± 1 lM, K2 = 0.3 ± 0.1 lM and Ki = 5 ± 1 lM). When the time axis of the coupled reaction is divided by the RCR, the coupled and uncoupled CuANor traces become identical inferring that only Vmax (494 NO s1) and not the NO binding parameters are affected by coupling (Fig. 2, inset). The kinetic constants for lipid embedded cNor (Vmax = 165 NO s1) were found the

Table 1 Respiratory control ratios of various types of PLs.

2053

same as for CuANor (except for Ki = 2 ± 1 lM) and differ from those determined for the solubilized cNor (K1 = 6.0 lM, K2 = 0.55 lM and Ki = 13.5 lM [6]). Due to the membrane-permeability of PESH, CuANor in the right-side out orientation (Fig. 3A & C) and in the opposite orientation (Fig. 3B, D, E & F) can both participate in NO turnover. Fig. 3B shows that even a non-electrogenic enzyme could show respiratory control and/or generate a Dw due to the different permeabilities of H+ and PES+ (see Section 4). Therefore the RCR displayed by CuANor cannot be taken as direct proof for its electrogenicity in spite of the RCR 1 observed for cNor. However, in case the reaction is mediated solely by the neutral PES radical (PES⁄), the overall intravesicular reaction is proton and electroneutral for a non-electrogenic CuANor (Fig. 3E), and observation of a RCR directly indicates that CuANor is electrogenic (Fig. 3F and Section 4). In order to assess the Dw formed by CuANor, the Dw-induced fluorescence quenching of safranine O by CuANor and cNor was determined (Fig. 4 and Table 2). The stacking of safranine O and hence quenching of its fluorescence (Fig. 4) are independent of the orientation of Dw [54] thus both orientations of the Nors can contribute to fluorescence quenching (Fig. 3). Trace A in Fig. 4 shows fluorescence quenching (5.3 ± 0.3%) during NO reduction by 1 nM aa3/CuANor-PLs, inferring the generation of a Dw. Importantly, the fluorescence quenching by CuANor increases over time and is maximal just before all NO is consumed (at 2 lM NO) and thus similar to the activity traces of CuANor (Fig. 2) supporting the hypothesis that NO reduction by the enzyme is responsible for the gradient formed. When NO is depleted (Fig. 4, [NO]  0), the fluorescence intensity returns with a half-life of 10–15 s to the initial level due to H+ equilibration across the membrane (proton leakage) that renders Dw zero. After addition of CCCP fluorescence quenching was not observed (Fig. 4, trace A). The NO reductase activity of cNor in aa3/cNor-PLs at a 3 times higher concentration (3 nM) than used for CuANor does produce a small decrease in fluorescence 0.40 ± 0.15% (Fig. 4, trace B). The amount of fluorescence quenching by 1 nM CuANor averages to 5.0 ± 0.3% (Table 2) and is independent of co-reconstitution with cytochrome aa3, the presence of ascorbate or the concentrations of NO (10–40 lM) and externally added PES+ (0–1 mM). The inset in Fig. 4 shows fluorescence quenching for O2 reduction by 1 nM aa3-PLs, which equals 4.4 ± 0.20%. These results indicate that CuANor is electrogenic (Fig. 3C, D and F, and see Section 4) and can generate a Dw similar in magnitude to that formed by cytochrome aa3, the latter estimated at 150 ± 20 mV [55]. 4. Discussion Sequence alignment shows 34% identity between CuANor and the cytochrome ba3 oxidase from T. thermophilus. Homology modeling indicates a significantly similar spatial arrangement of the redox centers and the amino acid residues that constitute the K-pathway analog in both enzymes (Fig. 1). This similarity suggests that NO reduction by CuANor is potentially electrogenic. 4.1. Electrogenicity of CuANor

PLs

RCRCCCP

RCRCCCP + val

Substrates

aa3 aa3/CuANor aa3/CuANor CuANor CuANora aa3/cNor aa3/cNor

4.9 ± 0.70 – – 2.3 ± 0.23 2.4 ± 0.30 1.03 ± 0.1 4.3 ± 0.75

5.5 ± 0.71 2.6 ± 0.24 3.2 ± 0.80 2.4 ± 0.20 – – –

Cytochrome c/O2 PESH/NO Cytochrome c/O2 PESH/NO PESH/NO PESH/NO Cytochrome c/O2

a CuANor-PLs prepared using 0.1 lM Nor instead of 0.5 lM Nor/1.8 mg phospholipids for the other PL preparations, which did not appear to affect the RCR.

Using the membrane-permeable PESH and NO, the fluorescence quenching of safranine O by CuANor-PLs was determined nearly 20 times greater than observed with the non-electrogenic cNor in aa3/cNor-PLs (Fig. 4, Table 2). The permeability coefficient we determined for PES+ (1.5  109 cm s1) is relatively small; PES+ is approximately 4 times more permeable than H+ (see Supplementary Material). This difference in permeability causes a small and nearly constant Dw (0.27% fluorescence quenching, Table 2 and Fig. 4 trace B) seen with cNor according to Fig. 3B.

2054

S. Al-Attar, S. de Vries / FEBS Letters 589 (2015) 2050–2057

Fig. 2. Reduction of NO by CuANor-PLs monitored in the absence (coupled) and presence of CCCP (uncoupled). The data were fitted (solid lines) according to Eq. (5). Both kinetic traces show the same sigmoidal kinetics [6,27]. The inset shows PESH oxidation by coupled and uncoupled PLs with the time axis for the coupled PLs divided by 2.4 (the RCR). The similar time dependence of the traces directly shows that the NO binding parameters are the same and the ratio of the rates is equal to 2.4.

Fig. 3. The effects of different PES mediated NO reduction pathways (A–D vs E, F) and different orientations of non-electrogenic (e.g. cNor) (A, B and E) and electrogenic Nor (C, D and F) on the observed RCR and Dw. PESH and PES⁄ are rapidly membrane permeable whereas membrane permeation by PES+ is slow indicated by the dotted arrow. The scheme shows that although a non-electrogenic Nor could display a RCR and generate a small Dw (B) when the PESH/PES+ redox couple serves as the reductant, the reaction is proton and electroneutral in the case of PES⁄ cycling (E). Respiratory control is observed and a Dw is generated when the enzyme is electrogenic regardless of its orientation or the NO reduction pathway (C, D and F). Nor is depicted as a trapezoid where the narrow end harbors the CuA center (or the c-cytochrome in case of cNor), which is the site of electron entry. In the right-side out orientation (A and C), CuA faces the extravesicular phase.

S. Al-Attar, S. de Vries / FEBS Letters 589 (2015) 2050–2057

2055

Fig. 4. Formation of Dw measured by fluorescence quenching of safranine O by aa3/CuANor-PLs (A), aa3/cNOR-PLs (B) and aa3-PLs (inset). (A) The CuANor reaction was started by addition of 40 lM NO to 1 nM of CuANor in the presence of 60 lM PES+ and 20 mM ascorbate. Subsequent addition of NO in the presence of CCCP does not result in fluorescence quenching. The time taken to obtain [NO]  0 (dashed line) for the uncoupled reaction was estimated from the RCR value of 2.4. (B) Same NO reduction conditions as in A, but with 3 nM cNor. At the end of the reaction 1 nM CuANor and 40 lM NO were added resulting in similar fluorescence quenching as in trace A. The turnover of cNor is approximately one-third of that of CuANor. The inset shows the fluorescence quenching for aa3-PLs (1 nM) in aerobic buffer. Addition of CCCP abolishes the fluorescence quenching.

Table 2 The relative change in fluorescence (DF/F) of safranine O during NO and O2 reduction.a PLs

DF/F (%)

Substrates

CuANor CuANor CuANor CuANor aa3/CuANor aa3/CuANor aa3/cNor aa3/cNorb aa3 Average CuANor

4.9 ± 0.10 5.1 ± 0.20 5.5 ± 0.40 4.9 ± 0.20 5.3 ± 0.30 4.5 ± 0.30 0.27 ± 0.15 0.40 ± 0.15 4.4 ± 0.20 5.0 ± 0.3

asc, 60 lM PES+, 40 lM NO 60 lM PESH, 20 lM NO 60 lM PESH, 10 lM NO 60 lM PESH, 60 lM PES+, 20 lM NO asc, 60 lM PES+, 20 lM NO 60 lM PESH, 1 mM PES+, 40 lM NO asc, 60 lM PES+, 40 lM NO asc, 60 lM PES+, 40 lM NO asc, 30 lM cyt. c, 260 lM O2

a Data obtained for the same phospholipid concentration (0.018 mg/mL) and enzyme (1 nM) concentrations. b Three times higher concentrations than under a.

Importantly the average CuANor and cNor PESH oxidation activities, 49.4 nM s1 and 16.5 nM s1, respectively, are much larger than the PES+ (1.65 nM s1) and H+ (0.44 nM s1) fluxes calculated for the extreme case of 25 mM intravesicular PES+ (see Supplementary Material). Thus, H+ and PES+ can be regarded impermeable relative to the rate of intravesicular PES+ production validating the use of PESH in our assays. The great difference between the amounts and time dependences of the fluorescence quenching observed with CuANor and cNor – the former relating to enzyme activity, the latter only to the difference between PES+ and H+ permeabilities – can only be due to a different charge balance of the overall reactions catalyzed by the two enzymes. Therefore we conclude that the reaction catalyzed by CuANor is electrogenic as depicted in Fig. 3C and D. The observation that CCCP completely prevents the formation of a Dw proves that ~þ CuANor generates a proton electrochemical gradient ðDl H Þ. 4.2. The PES radical as intermediate in the reaction In the literature, data on phenazine methosulfate (PMS) greatly outnumber those on PES. Given their similar permeability coefficients (P+PMS = 1.2  109 cm s1) and redox properties

[56,57], the results obtained with PMS can be used to support the PES radical (PES⁄) cycling mechanism sketched in Fig. 3F as the main mechanism in our assays (see below). At low pH, PMS or PES can be stoichiometrically converted to PMS⁄ or PES⁄ [58], whereas at pH 7.5 approximately 15% is maximally in the radical form [58]. Experiments with intravesicularly trapped ferricyanide and extravesicular ascorbate/PMS show that the transport of PMSH across liposomal membranes is fast judged from the initial rapid reduction of ferricyanide and the associated intravesicular acidification [59,60]. The ferricyanide reduction dramatically slows down almost immediately owing to the slow export of PMS+, consistent with its small permeability constant determined in the present study. However, the rate of PMS+ transport in these liposomes is not limiting in the presence of tetraphenylboron (TPB-) [61–63], which accelerates PMS+ transmembrane transport. On the other hand, when an equimolar mixture of ferri- and ferrocyanide is trapped inside the liposomes, the rate of ferricyanide reduction by external ascorbate/PMS in the absence of TPB is fast and remains fast. At the redox potential set by the ferri- and ferrocyanide couple, sufficient PMS⁄ is formed [60] sustaining a rapid rate of ferricyanide reduction via PMS⁄ transmembrane cycling (cf. Fig. 3E). Similar observations were made using (inverted) submitochondrial particles that contain cytochrome c oxidase (cf. orientation in Fig. 3F) and internal cytochrome c. The rate of PMSH mediated O2 reduction was not increased by TPB [62], because cytochrome c is an oxidant not as strong as ferricyanide, and consequently the former oxidizes a significant portion of PMSH only to the PMS⁄ form that cycles across the membrane [62]. In our experiments, CuANor-PLs with the orientation as in Fig. 3F would generate 60 lM intravesicular PES+ within one second, matching the [PESH] and so maximizing the [PES⁄]. The midpoint potential of the Nor redox centers (0.2–0.35 V) is in the same range as that of cytochrome c (see example above), which suggests that during the reaction, CuANor-PLs will generate enough PES⁄ to make the overall reaction shown in Fig. 3F the dominant steady-state reaction. As a consequence, the intravesicular pH change and PES+ concentration will be small and hence the PES+ and H+ diffusion rates are both much smaller than assessed for the extreme case

2056

S. Al-Attar, S. de Vries / FEBS Letters 589 (2015) 2050–2057

2e-

5NADH þ 37HþN þ 2NO3 ! 5NADþ þ 30HþP þ 6H2 O þ N2

H2O + N2O 2NO

CuA periplasm (+)

Heme b

Thus denitrification in bacilli yields potentially 6.7% more ATP (Table S1B) than organisms employing the cNor. The yield of ATP can be 13% higher in case CuANor would pump, in addition, 2 protons per 2 NO (Table S1C). In order to assess proton pumping by CuANor, experiments with an active soluble form of cytochrome c551 are scheduled.

BNC

cytoplasm (-)

2H+cyto

Acknowledgment

Fig. 5. A simplified scheme outlining the mechanism for the formation of a proton ~þ ~þ electrochemical gradient ðDl H Þ by CuANor without actual proton pumping. The DlH is formed because protons are taken from the cytoplasm via the K-pathway analog causing a decrease in cytoplasmic charge (Dw) and proton concentration (DpH). In case the reaction is coupled to proton pumping (as in Coxs), additional protons would be taken up from the cytoplasm.

where 25 mM intravesicular PES+ accumulates (see also Supporting Material). That the PES⁄ radical mechanism must be dominant (Fig. 3F) is further supported by the finding that much more NO can be converted than needed to titrate out the intravesicular buffer (see Supplementary Material for the calculation). Hence, the finding of respiratory control by CuANor and the similarity between Dw formation and NO reduction kinetics (Figs. 2 and 4) directly support electrogenic NO reduction by the enzyme. In conclusion, two reaction mechanisms have been discussed above: (i) PESH oxidation to PES+ where PES+ permeability is negligible (Fig. 3D) and (ii) PESH oxidation to the PES⁄ radical, which results in transmembrane cycling of PES⁄ without significant accumulation of PES+ inside the PLs and yielding a proton neutral and electroneutral overall reaction within the liposome (Fig.3F). Both mechanisms occur under the assay conditions. Interpretation of the data leads for either mechanism to the conclusion that NO reduction by CuANor is electrogenic. 4.3. Bioenergetic implications of electrogenic NO reduction ~þ The simplest way to generate ðDl H Þ is according to:

2NO þ 2Hþcyto þ 2eperi ! N2 O þ H2 O

ð7Þ

ð6Þ

Herein the protons needed for the formation of H2O are taken up from the cytoplasm, whereas reduction of the redox centers within the membrane dielectric occurs via the CuA center that faces the periplasm (Fig. 5). Proton transport from the cytoplasm to the BNC is proposed to occur via the K-pathway analog (Fig. 1). The homology model predicts a proton exit pathway for CuANor equivalent to that in the proton pumping cytochrome ba3 oxidase (Fig. 1, [37]). Thus apart from taking the substrate protons from the cytoplasm (Eq. (6)) and electrons from the periplasm rendering the enzyme electrogenic, CuANor might be a proton pump as well. To quantitatively investigate whether CuANor is a proton pump, the pH change at the exterior of CuANor-PLs was determined with the membrane-impermeable pH dye phenol red (Supplementary Fig. S2). The results show similar alkalization ( -1H+/PESH) for coupled and uncoupled reactions without indication of proton-burst kinetics. This latter observation and the permeability of PESH prevent, at present, assessment as to whether or not CuANor is a proton pump. The experimentally determined alkalization enables an estimation of the orientation of CuANor in the PLs. In case CuANor does not pump protons, 33% of the enzyme would be in the right-side out orientation whereas this is 43% in case the enzyme pumps 2 protons per turnover (see Fig. S2). The finding that CuANor is electrogenic but cNor is not, implies that the q+/e ratio of 3.0 for NADH driven nitrate reduction to N2 by e.g. bacteria using cNor [2] is increased to 3.2 for bacilli according to Eq. (7):

This work was supported by an ECHO Grant (711.011.004) from the Netherlands Organization for Scientific Research (NWO) to SdV. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.febslet.2015.06. 033. References [1] van Spanning, R.J., Richardson, D.J. and Ferguson, S.J. (2007) Introduction to the biochemistry and molecular biology of denitrification. Biol. Nitrogen Cycle, 382–395. [2] Wasser, I.M., de Vries, S., Moenne-Loccoz, P., Schroder, I. and Karlin, K.D. (2002) Nitric oxide in biological denitrification: Fe/Cu metalloenzyme and metal complex NO(x) redox chemistry. Chem. Rev. 102, 1201–1234. [3] Hendriks, J., Oubrie, A., Castresana, J., Urbani, A., Gemeinhardt, S. and Saraste, M. (2000) Nitric oxide reductases in bacteria. Biochim. Biophys. Acta 1459, 266–273. [4] Shoun, H., Fushinobu, S., Jiang, L., Kim, S.W. and Wakagi, T. (2012) Fungal denitrification and nitric oxide reductase cytochrome P450nor. Philos. Trans. R. Soc. Lond. B Biol. Sci. 367, 1186–1194. [5] Tavares, P., Pereira, A.S., Moura, J.J. and Moura, I. (2006) Metalloenzymes of the denitrification pathway. J. Inorg. Biochem. 100, 2087–2100. [6] Girsch, P. and de Vries, S. (1997) Purification and initial kinetic and spectroscopic characterization of NO reductase from Paracoccus denitrificans. Biochim. Biophys. Acta 1318, 202–216. [7] Moenne-Loccoz, P. (2007) Spectroscopic characterization of heme ironnitrosyl species and their role in NO reductase mechanisms in diiron proteins. Nat. Prod. Rep. 24, 610–620. [8] Heylen, K. and Keltjens, J. (2012) Redundancy and modularity in membraneassociated dissimilatory nitrate reduction in Bacillus. Front. Microbiol. 3, 371. [9] van der Oost, J., de Boer, A.P., de Gier, J.W., Zumft, W.G., Stouthamer, A.H. and van Spanning, R.J. (1994) The heme-copper oxidase family consists of three distinct types of terminal oxidases and is related to nitric oxide reductase. FEMS Microbiol. Lett. 121, 1–9. [10] Sousa, F.L., Alves, R.J., Ribeiro, M.A., Pereira-Leal, J.B., Teixeira, M. and Pereira, M.M. (1817) The superfamily of heme-copper oxygen reductases: types and evolutionary considerations. Biochim. Biophys. Acta 2012, 629–637. [11] Hemp, J. and Gennis, R.B. (2008) Diversity of the heme-copper superfamily in archaea: insights from genomics and structural modeling. Results Probl. Cell Differ. 45, 1–31. [12] de Vries, S. and Schroder, I. (2002) Comparison between the nitric oxide reductase family and its aerobic relatives, the cytochrome oxidases. Biochem. Soc. Trans. 30, 662–667. [13] Hendriks, J., Warne, A., Gohlke, U., Haltia, T., Ludovici, C., Lubben, M. and Saraste, M. (1998) The active site of the bacterial nitric oxide reductase is a dinuclear iron center. Biochemistry 37, 13102–13109. [14] Moenne-Loccoz, P. and de Vries, S. (1998) Structural characterization of the catalytic high-spin heme b of nitric oxide reductase: a resonance Raman study. J. Am. Chem. Soc. 120, 5147–5152. [15] Moënne-Loccoz, P., Richter, O.-M.H., Huang, H.-W., Wasser, I.M., Ghiladi, R.A., Karlin, K.D. and de Vries, S. (2000) Nitric oxide reductase from Paracoccus denitrificans contains an oxo-bridged heme/non-heme diiron center. J. Am. Chem. Soc. 122, 9344–9345. [16] Zumft, W.G. (2005) Nitric oxide reductases of prokaryotes with emphasis on the respiratory, heme-copper oxidase type. J. Inorg. Biochem. 99, 194–215. [17] Butler, C., Forte, E., Maria Scandurra, F., Arese, M., Giuffre, A., Greenwood, C. and Sarti, P. (2002) Cytochrome bo(3) from Escherichia coli: the binding and turnover of nitric oxide. Biochem. Biophys. Res. Commun. 296, 1272–1278. [18] Flock, U., Lachmann, P., Reimann, J., Watmough, N.J. and Adelroth, P. (2009) Exploring the terminal region of the proton pathway in the bacterial nitric oxide reductase. J. Inorg. Biochem. 103, 845–850. [19] Huang, Y., Reimann, J., Lepp, H., Drici, N. and Adelroth, P. (2008) Vectorial proton transfer coupled to reduction of O2 and NO by a heme-copper oxidase. Proc. Natl. Acad. Sci. U.S.A. 105, 20257–20262. [20] Salomonsson, L., Reimann, J., Tosha, T., Krause, N., Gonska, N., Shiro, Y. and Adelroth, P. (1817) Proton transfer in the quinol-dependent nitric oxide

S. Al-Attar, S. de Vries / FEBS Letters 589 (2015) 2050–2057

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36] [37]

[38]

[39] [40] [41]

reductase from Geobacillus stearothermophilus during reduction of oxygen. Biochim. Biophys. Acta 2012, 1914–1920. Suharti, Strampraad, M.J., Schroder, I. and de Vries, S. (2001) A novel copper A containing menaquinol NO reductase from Bacillus azotoformans. Biochemistry 40, 2632–2639. Hendriks, J.H., Jasaitis, A., Saraste, M. and Verkhovsky, M.I. (2002) Proton and electron pathways in the bacterial nitric oxide reductase. Biochemistry 41, 2331–2340. Bell, L.C., Richardson, D.J. and Ferguson, S.J. (1992) Identification of nitric oxide reductase activity in Rhodobacter capsulatus: the electron transport pathway can either use or bypass both cytochrome c2 and the cytochrome bc1 complex. J. Gen. Microbiol. 138, 437–443. de Vries, S., Strampraad, M.J., Lu, S., Moenne-Loccoz, P. and Schroder, I. (2003) Purification and characterization of the MQH2:NO oxidoreductase from the hyperthermophilic archaeon Pyrobaculum aerophilum. J. Biol. Chem. 278, 35861–35868. Cramm, R., Pohlmann, A. and Friedrich, B. (1999) Purification and characterization of the single-component nitric oxide reductase from Ralstonia eutropha H16. FEBS Lett. 460, 6–10. Matsumoto, Y., Tosha, T., Pisliakov, A.V., Hino, T., Sugimoto, H., Nagano, S., Sugita, Y. and Shiro, Y. (2012) Crystal structure of quinol-dependent nitric oxide reductase from Geobacillus stearothermophilus. Nat. Struct. Mol. Biol. 19, 238–245. Suharti, Heering, H.A. and de Vries, S. (2004) NO reductase from Bacillus azotoformans is a bifunctional enzyme accepting electrons from menaquinol and a specific endogenous membrane-bound cytochrome c551. Biochemistry 43, 13487–13495. Hino, T., Matsumoto, Y., Nagano, S., Sugimoto, H., Fukumori, Y., Murata, T., Iwata, S. and Shiro, Y. (2010) Structural basis of biological N2O generation by bacterial nitric oxide reductase. Science 330, 1666–1670. ter Beek, J., Krause, N., Reimann, J., Lachmann, P. and Adelroth, P. (2013) The nitric-oxide reductase from Paracoccus denitrificans uses a single specific proton pathway. J. Biol. Chem. 288, 30626–30635. Pisliakov, A.V., Hino, T., Shiro, Y. and Sugita, Y. (2012) Molecular dynamics simulations reveal proton transfer pathways in cytochrome C-dependent nitric oxide reductase. PLoS Comput. Biol. 8, e1002674. Reimann, J., Flock, U., Lepp, H., Honigmann, A. and Adelroth, P. (2007) A pathway for protons in nitric oxide reductase from Paracoccus denitrificans. Biochim. Biophys. Acta 1767, 362–373. Al-Attar, S. and de Vries, S. (2013) Energy transduction by respiratory metalloenzymes: from molecular mechanism to cell physiology. Coord. Chem. Rev. 257, 64–80. Lee, H.J., Reimann, J., Huang, Y. and Adelroth, P. (1817) Functional proton transfer pathways in the heme-copper oxidase superfamily. Biochim. Biophys. Acta 2012, 537–544. Wikstrom, M. and Verkhovsky, M.I. (2007) Mechanism and energetics of proton translocation by the respiratory heme-copper oxidases. Biochim. Biophys. Acta 1767, 1200–1214. Hosler, J.P., Ferguson-Miller, S. and Mills, D.A. (2006) Energy transduction: proton transfer through the respiratory complexes. Annu. Rev. Biochem. 75, 165–187. Kaila, V.R., Verkhovsky, M.I. and Wikstrom, M. (2010) Proton-coupled electron transfer in cytochrome oxidase. Chem. Rev. 110, 7062–7081. von Ballmoos, C., Adelroth, P., Gennis, R.B. and Brzezinski, P. (1817) Proton transfer in ba(3) cytochrome c oxidase from Thermus thermophilus. Biochim. Biophys. Acta 2012, 650–657. Wiertz, F.G., Richter, O.M., Ludwig, B. and de Vries, S. (2007) Kinetic resolution of a tryptophan-radical intermediate in the reaction cycle of Paracoccus denitrificans cytochrome c oxidase. J. Biol. Chem. 282, 31580–31591. Chen, P.S., Toribara, T.Y. and Warner, H. (1956) Microdetermination of phosphorus. Anal. Chem. 28, 1756–1758. Fiske, C.H. and Subbarow, Y. (1925) The colorimetric determination of phosphorus. J. Biol. Chem. 66, 375–400. Ksenzhek, O.S., Petrova, S.A. and Kolodyazhny, M.V. (1977) Electrochemical properties of some redox indicators. Bioelectrochem. Bioenerg. 4, 346–357.

2057

[42] Åkerman, K.E.O. and Wikström, M.K.F. (1976) Safranine as a probe of the mitochondrial membrane potential. FEBS Lett. 68, 191–197. [43] Komary, Z., Tretter, L. and Adam-Vizi, V. (2010) Membrane potential-related effect of calcium on reactive oxygen species generation in isolated brain mitochondria. Biochim. Biophys. Acta 1797, 922–928. [44] Arnold, K., Bordoli, L., Kopp, J. and Schwede, T. (2006) The SWISS-MODEL workspace: a web-based environment for protein structure homology modelling. Bioinformatics 22, 195–201. [45] Bordoli, L., Kiefer, F., Arnold, K., Benkert, P., Battey, J. and Schwede, T. (2008) Protein structure homology modeling using SWISS-MODEL workspace. Nat. Protoc. 4, 1–13. [46] Soulimane, T., Buse, G., Bourenkov, G.P., Bartunik, H.D., Huber, R. and Than, M.E. (2000) Structure and mechanism of the aberrant ba(3)-cytochrome c oxidase from Thermus thermophilus. EMBO J. 19, 1766–1776. [47] Chang, H.Y., Hemp, J., Chen, Y., Fee, J.A. and Gennis, R.B. (2009) The cytochrome ba3 oxygen reductase from Thermus thermophilus uses a single input channel for proton delivery to the active site and for proton pumping. Proc. Natl. Acad. Sci. U.S.A. 106, 16169–16173. [48] Hemp, J., Robinson, D.E., Ganesan, K.B., Martinez, T.J., Kelleher, N.L. and Gennis, R.B. (2006) Evolutionary migration of a post-translationally modified activesite residue in the proton-pumping heme-copper oxygen reductases. Biochemistry 45, 15405–15410. [49] Fee, J.A., Case, D.A. and Noodleman, L. (2008) Toward a chemical mechanism of proton pumping by the B-type cytochrome c oxidases: application of density functional theory to cytochrome ba3 of Thermus thermophilus. J. Am. Chem. Soc. 130, 15002–15021. [50] Grosdidier, A., Zoete, V. and Michielin, O. (2011) SwissDock, a protein-small molecule docking web service based on EADock DSS. Nucleic Acids Res. 39, W270–277. [51] Pouvreau, L.A., Strampraad, M.J., Van Berloo, S., Kattenberg, J.H. and de Vries, S. (2008) NO, N2O, and O2 reaction kinetics: scope and limitations of the Clark electrode. Methods Enzymol. 436, 97–112. [52] Kimelberg, H.K. and Lee, C.P. (1970) Interactions of cytochrome c with phospholipid membranes: reactivity of cytochromec bound to phospholipid liquid crystals. J. Membr. Biol. 2, 252–262. [53] Schneider, D.L., Kagawa, Y. and Racker, E. (1972) Chemical modification of the inner mitochondrial membrane. J. Biol. Chem. 247, 4074–4079. [54] Figueira, T.R., Melo, D.R., Vercesi, A.E. and Castilho, R.F. (2012) Safranine as a fluorescent probe for the evaluation of mitochondrial membrane potential in isolated organelles and permeabilized cells. Methods Mol. Biol. 810, 103–117. [55] Singh, A.P. and Nicholls, P. (1985) Cyanine and safranine dyes as membrane potential probes in cytochrome c oxidase reconstituted proteoliposomes. J. Biochem. Biophys. Methods 11, 95–108. [56] Michaelis, L. (1935) Semiquinones, the intermediate steps of reversible organic oxidation-reduction. Chem. Rev. 16, 243–286. [57] Prince, R.C., Linkletter, S.J. and Dutton, P.L. (1981) The thermodynamic properties of some commonly used oxidation-reduction mediators, inhibitors and dyes, as determined by polarography. Biochim. Biophys. Acta 635, 132–148. [58] Zaugg, W.S. (1964) Spectroscopic characteristics and some chemical properties of N-methylphenazinium methyl sulfate (phenazine methosulfate) and pyocyanine at the semiquidnoid oxidation level. J. Biol. Chem. 239, 3964–3970. [59] Deamer, D.W., Prince, R.C. and Crofts, A.R. (1972) The response of fluorescent amines to pH gradients across liposome membranes. Biochim. Biophys. Acta 274, 323–335. [60] Hauska, G.A. and Prince, R.C. (1974) Lipophilicity and catalysis of photophosphorylation. Artificial proton translocation by lipophilic, quinoid hydrogen carriers in chloroplasts and liposomes. FEBS Lett. 41, 35–39. [61] Hinkle, P.C. (1973) Electron transfer across membranes and energy coupling. Fed. Proc. 32, 1988–1992. [62] Hinkle, P.C. (1979) A model system for transmembrane redox reactions. Methods Enzymol. 55, 541–544. [63] Shinbo, T., Yamaguchi, T., Sugiura, M., Kamo, N. and Kobatake, Y. (1984) Kinetic-study of transmembrane electron-transport by the Hinkle process. J. Membr. Sci. 20, 285–296.

An electrogenic nitric oxide reductase.

Nitric oxide reductases (Nors) are members of the heme-copper oxidase superfamily that reduce nitric oxide (NO) to nitrous oxide (N₂O). In contrast to...
1MB Sizes 1 Downloads 12 Views