Identification of the Missing Component in the Mitochondrial Benzamidoxime Prodrug-converting System as a Novel Molybdenum Enzyme*

Amidoximes can be used as prodrugs for amidines and related functional groups to enhance their intestinal absorption. These prodrugs are reduced to their active amidines. Other N-hydroxylated structures are mutagenic or responsible for toxic effects of drugs and are detoxified by reduction. In this study, a N-reductive enzyme system of pig liver mitochondria using benzamidoxime as a model substrate was identified. A protein fraction free from cytochrome b5 and cytochrome b5 reductase was purified, enhancing 250-fold the minor benzamidoxime-reductase activity catalyzed by the membrane-bound cytochrome b5/NADH cytochrome b5 reductase system. This fraction contained a 35-kDa protein with homologies to the C-terminal domain of the human molybdenum cofactor sulfurase. Here it was demonstrated that this 35-kDa protein contains molybdenum cofactor and forms the hitherto ill defined third component of the N-reductive complex in the outer mitochondrial membrane. Thus, the 35-kDa protein represents a novel group of molybdenum proteins in eukaryotes as it forms the catalytic part of a three-component enzyme complex consisting of separate proteins. Supporting these findings, recombinant C-terminal domain of the human molybdenum cofactor sulfurase exhibited N-reductive activity in vitro, which was strictly dependent on molybdenum cofactor.

Here the identification of this third N-reductive component in the outer mitochondrial membrane of pig liver is described. It turned out to be a novel molybdenum-containing protein hitherto not described in eukaryotes or bacteria sharing significant homology with the C-terminal domain of the molybdenum cofactor sulfurase (HMCS-CT).

EXPERIMENTAL PROCEDURES
Purification of the N-Reductive Mitochondrial Component-Pig liver mitochondria were obtained by differential centrifugation and an isotonic Percoll gradient (15) with modifications (4). The outer membrane vesicle (OMV) fraction was purified using the swell disruption method followed by two steps of sucrose density gradient centrifugation (16) with modifications (4). Frozen OMV fraction was thawed, pooled, and adjusted to 20 mM Tris-base, 0.1 mM EDTA, 0.1 mM dithiothreitol, 20% (m/v) glycerol, 0.9 mM Zwittergent 3-14, pH 7.4. The detergent/protein ratio was 0.5. This solubilization mixture was stirred for 60 min on ice. The solubilized membrane fraction (ϳ13 mg of protein) was applied to a DEAE-52 column (2.5 ϫ 10 cm) (DEAE 52-Cellulose Servacel; Serva Electrophoresis, Heidelberg, Germany) and equilibrated with buffer (20 mM Tris-base, 0.1 mM EDTA, 0.1 mM dithiothreitol, 20% (m/v) glycerol, 0.9 mM Zwittergent 3-14, pH 7.4). After sample application, the column was washed at a flow rate of 0.8 ml/min with two column volumes of the equilibrating buffer and was developed with a linear concentration gradient (250 ml) of 0 -1.0 sodium chloride in equilibrating buffer. Fractions of 4 ml were collected and assayed. Active fractions were pooled and stored at Ϫ80°C. All operations were performed at 0 -4°C. Protein was assayed using bicinchonic acid (17), according to the manufacturer's directions (BCA protein assay kit; Pierce), using bovine serum albumin as a standard.
Purification of NADH Cyt b 5 Reductase-NADH Cyt b 5 reductase was purified from pig liver microsomes by affinity chromatography on 5Ј-AMP-Sepharose 4 B (Amersham Biosciences), similar to the procedure described for the purification of NADH-P450 reductase (18) with modifications (8). Recombinant purified human NADH cyt b 5 reductase was obtained from Abnova Corp. (Taipei City, Taiwan).
SDS-PAGE-SDS-PAGE analyses were carried out by the method of Laemmli (19), with a 5% stacking gel. Staining was performed with Coomassie Brilliant Blue R250 (Serva, Heidelberg, Germany). Standards and samples were pretreated with ␤-mercaptoethanol for 5 min at 90°C.
Mass Spectrometry-Proteins were identified by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS) peptide mass finger printing or nanocapillary liquid chromatography electrospray ionization tandem mass spectrometry (MS/MS). The analysis were performed by Planton GmbH (Kiel, Germany) and Richard Jones (National Center for Toxicological Research).
Immunoblot Analysis-Immunoblot analysis was performed by gel-blotting protein-fraction A and B subjected to 12% SDS-PAGE using a primary polyclonal antibody raised against recombinantly expressed C terminus of Arabidopsis thaliana molybdenum cofactor sulfurase (ABA3-CT) (1:7000 dilution). The secondary horseradish peroxidaseconjugated anti-rabbit Ig (Sigma) was used in a 1:10,000 dilution, and chemiluminescence was detected using the ECL system (Amersham Biosciences).
nit-1 Reconstitution-Neurospora crassa nit-1 extract was prepared as described previously (20) and stored in aliquots at Ϫ70°C. All reconstitutions were performed in nit-1 buffer (50 mM sodium phosphate, 200 mM NaCl, and 5 mM EDTA, pH 7.2) containing 2 mM reduced glutathione either in the absence or in the presence of 5 mM sodium molybdate. The reconstitution assay was performed in a 40-l reaction volume containing 20 l of gel-filtrated nit-1 extract. Complementation was carried out anaerobically for 2 h at room temperature. After the addition of 20 mM NADPH and incubation for 10 min in the dark, reconstituted NADPH-nitrate reductase activity was determined as described (20). Importantly, the addition of sodium molybdate to the reaction mix did not significantly enhance reconstitution of NADPH-nitrate reductase activity, indicating that all fractions tested contained molybdenum cofactor (Moco) rather than molybdenum-free molybdopterin.
Cloning of hmcs-CT-Polymerase chain reaction was performed with a human liver cDNA library (-Uni-ZAP XR, Stratagene) as template and by using primers HMCS-1429ϩ (5Ј-GGA TAC ATG TCG ACG CTG GAT GAT-3Ј) and HMCS-stop (5Ј-TTA GGA GGT AAC ATC CTG GTG TTT CTC-3Ј) derived from GenBank TM entry AK000740, which represents a full-length hmcs-cDNA. By this procedure, a partial hmcs cDNA fragment of ϳ1.3 kb was obtained, which contained the 3Ј-region of the hmcs open reading frame and whose correctness was confirmed by sequencing. Restriction sites for BamHI were introduced by polymerase chain reaction using primers HMCS-CT-start/BamHI (5Ј-ACC CAG GGA TCC ATG TCA GAG AAA GCT GCA GGA GTC CTG-3Ј) and HMCS-CT-stop/BamHI (5Ј-ACG GTG GAT CCT TAG GAG GTA ACA TCC TGG TGT TTC TC-3Ј), which amplified the last 966 bp of the open reading frame encoding for the putative C-terminal domain of HMCS. Simultaneous introduction of an ATG codon at the N-terminal end in frame with the vectorencoded His 6 tag enabled cloning into the pQE80-plasmid (Qiagen, Hilden, Germany), thereby allowing expression of a His 6 -HMCS-CT fusion protein with an estimated molecular mass of 37.4 kDa. Cloning and expression of ABA3-CT for generation of anti-ABA3-CT polyclonal antibodies was performed as described earlier by Heidenreich et al. (21).
Expression and Purification of HMCS-CT-Routine protein expression was performed in freshly transformed Escherichia coli TP1000 cells (22). Cells were grown aerobically in Luria broth medium in the presence of 100 g/ml ampicillin at 22°C to a A 600 ϭ 0.1 before induction with 15 M isopropyl-␤-Dthiogalactopyranoside and the addition of 1 mM sodium molyb-  NOVEMBER 17, 2006 • VOLUME 281 • NUMBER 46 date. After induction, cells were grown for a further 20 h at 22°C. Expression in E. coli strains RK5206 and RK5204 (23) was carried out likewise without the addition of sodium molybdate. Cells were harvested by centrifugation and stored at Ϫ70°C until use. Cell lysis was achieved by several passages through a French pressure cell followed by sonication for 5 min. After centrifugation, His 6 -tagged protein was purified on a nickel-nitrilotriacetic acid superflow matrix (Qiagen, Hilden, Germany) under native conditions at 4°C according to the manufacturer's manual. Eluted fractions were analyzed by SDS-PAGE. Molybdenum binding pterin bound to the purified proteins was detected and quantified by converting it to the stable oxidation product FormAdephospho, according to Johnson et al. (24). Oxidation, dephosphorylation, QAE chromatography, and high performance liquid chromatography (HPLC) analysis were performed as described previously (25). FormA-dephospho was quantified by comparison with a standard isolated from xanthine oxidase for which the absorptivity was ⑀ 380 ϭ 13,200 M Ϫ1 cm Ϫ1 (24). Thereby, average saturation of HMCS-CT with MPT/Moco, determined as FormA-dephospho, was found to be about 41% (412 Ϯ 96 pmol of FormA/nmol of protein; n ϭ 5).

The Mitochondrial Benzamidoxime-converting System
NADH Cyt b 5 Reductase Activity-Enzyme activity was determined by modification of the ferricyanide reduction assay (26).
Heme Content-Heme content was estimated by recording the sodium dithionite-reduced minus the oxidized spectrum (27).
Assay for the Reduction of Benzamidoxime-For determination of benzamidoxime reduction, incubations were carried out under aerobic conditions at 37°C in a shaking water bath. Incubation mixtures of the subcellular fractions contained 56 g (mitochondria) or 6 g of protein (OMV fraction), 0.5 mM benzamidoxime (synthesized from benzonitrile and hydroxylamine as described in Ref. 28), and 1.0 mM NADH (or 0.4 mM NADH in the case of the OMV fraction) in a total volume of 150 l of 100 mM potassium phosphate buffer, pH 6.0. After preincubation for 3 min at 37°C, the reaction was initiated by the addition of NADH and terminated after 15 min (OMV fraction) or 20 min (mitochondria) by adding aliquots of methanol. If not otherwise stated, standard incubation mixtures of the reconstituted system contained 240 ng of purified mitochondrial enzyme fraction, 0.3 units of cyt b 5 reductase, 100 pmol of recombinant purified cyt b 5 (MoBiTec GmbH, Göttingen, Germany), 0.5 mM benzamidoxime, and 1.0 mM NADH in a total volume of 150 l of 100 mM potassium phosphate buffer, pH 6.0. After preincubation for 3 min at 37°C, the reaction was started by NADH and terminated after 15 or 30 min by adding aliquots of methanol. If not otherwise stated, incubations with HMCS-CT contained 135 g of recombinant protein, 0.5 mM benzamidoxime, and 1.0 mM NADH in a total volume of 150 l of 100 mM potassium phosphate buffer, pH 6.0. After preincubation for 3 min at 37°C, the reaction was started by benzamidoxime and terminated after 10 min by adding aliquots of methanol. The precipitated proteins were sedimented by centrifugation, and the supernatant was analyzed by HPLC.
Inhibition Studies-Incubations with the OMV fraction were performed as described above with slight modifications. After 5 min of preincubation of protein, inhibitor, and NADH, the reaction was started by the addition of benzamidoxime. Studies were performed with 0 -100 M sodium vanadate.
HPLC Method for Benzamidine Quantification-The separation was carried out isocratically by 10 mM 1-octylsulfonate sodium salt and 17% acetonitrile (v/v) (pH not adjusted) by a LiChroCART 250-4 HPLC-Cartridge with LiChrospher 60 RP-select B (5 m) and a LiChroCART 4-4 guard column (Merck) at a flow rate of 1.0 ml/min. The effluent was monitored at 229 nm. For the determination of the recovery rate, reaction mixtures with defined concentrations of synthetic reference substance (1-300 M) were incubated and worked up under the same conditions as the experimental samples but without adding cofactor. The standard curves were linear over this range with correlation coefficients of 0.999 (n ϭ 48). The signals (peak areas) obtained were compared with those of the same amount of benzamidine dissolved in the mobile phase. The recovery rate from pig liver mitochondria amounted to 105% (r 2 ϭ 0.9993). Similar values were obtained from the OMV fraction. The retention times were 7.9 Ϯ 0.1 min (benzamidoxime) and 26.7 Ϯ 0.1 min (benzamidine).
Mass Spectrometry-The formation of the reductive metabolite benzamidine was examined by mass spectrometry according to the method described by Froehlich et al. (29).

RESULTS
Purification of a N-Reductive Protein Fraction-Identification of the third N-reductive component in the outer mitochondrial membrane remained elusive in previous studies. Therefore, a rigorously improved purification protocol was elaborated (for details, see "Experimental Procedures") consisting of three main steps. (i) The OMV fraction was prepared from pig liver mitochondria (4). (ii) The membrane-bound N-reductive enzyme complex was solubilized in a stable and highly active form. Here, Zwittergent 3-14 turned out to be superior to all other detergents (CHAPS, cholate, digitonin, thesit, and Triton X-100) tested, and full N-reductive activity was retained up to a detergent/protein ratio of 5. The activity was stable over a period of 6 days at 4°C. Vanadate as inhibitor of benzamidoxime reductase almost abolished the N-reductive activity of the outer mitochondrial membrane, exerting an IC 50 value of 8 M. (iii) ion exchange chromatography of detergentsolubilized OMV fractions gave two pools (designated as fractions A and B) with benzamidoxime reductase activity (Fig. 2). Fraction A was devoid of any heme content (as tested by dithionite-reduced difference spectra) and of cyt b 5 reductase activity, whereas fraction B contained minor contents of cyt b 5 and its reductase. Therefore, fraction A was used for further analysis. Purification parameters are given in Table 1, and the results of SDS-PAGE analysis are shown in Fig. 3. The purified enzyme fraction A was stored at Ϫ80°C, and the same activity in the reconstituted system was observed at least after one freezethaw cycle (data not shown).
Enzymatic Characterization of Fraction A-The reduction of benzamidoxime by NADH requires cyt b 5 , its reductase, and an additional N-reductive protein contained in the purified mitochondrial protein fraction A. Fraction A alone was not able to catalyze the reduction of benzamidoxime by NADH ( Table 2).
The low rate of benzamidoxime reduction obtained with the mitochondrial fraction A and purified NADH cyt b 5 reductase (Table 2) is caused by the contamination of the reductase with small amounts of cyt b 5 . Similar results were obtained when replacing purified microsomal NADH cyt b 5 reductase by purified recombinant NADH cyt b 5 reductase (not shown). The N-reductive activity of the standard incubation mixture followed Michaelis-Menten kinetics (apparent K m 0.18 mM, V max was 12.25 mol of benzamidine/min/mg of fraction A).
The specific activity of fraction B in the reconstituted system was slightly lower than the specific activity of fraction A (7.10 Ϯ 0.08 mol of benzamidine/min/mg (n ϭ 4)). However, in contrast to fraction A, fraction B also reduced benzamidoxime without adding cyt b 5 and its reductase with a specific activity of 30 Ϯ 1 nmol of benzamidine/min/mg (n ϭ 4), demonstrating that it is contaminated with cyt b 5 and its reductase.
Composition of Fraction A-SDS-PAGE analysis of fraction A showed an almost electrophoretically pure protein with two dominant spots (35 and 66 kDa) in a Coomassie-stained SDS gel (Fig. 3B). Whereas the 66-kDa protein was identified as monoamine oxidase-B by MALDI-TOF, tryptic digestion and electrospray ionization-MS/MS sequencing of the 35-kDa spot led to identification of four peptides, LWIYPVK, TEAYR, CILTTVDPDTGVIDRK, and VGD-PVYR, which all displayed 100% identity to a putative protein from Homo sapiens hitherto referred to as "Moco sul-furase C-terminal domain-containing protein 2" (protein accession entry NP_060368). This result was confirmed by identifying the peptides LWIYPVKSCK, CILTTVDPDT-GVIDRK, and VGDPVYRMV in a second independent experiment. Sequence comparisons showed that the protein identified shares 48% homology with the C-terminal domain of HMCS, which catalyzes the post-translational activation of the Moco-containing enzymes xanthine dehydrogenase and aldehyde oxidase (30). Antibodies raised against the heterologously expressed C-terminal domain of the Moco sulfurase ABA3 from A. thaliana, the plant counterpart of HMCS (31), showed specific cross-reaction with the C-terminal domain of HMCS (HMCS-CT) (Fig. 4A). Also in fractions A and B from the outer membrane of pig liver mitochondria, the antibody specifically recognized the 35-kDa protein (Fig. 4B), which gives evidence for structural relation of the 35-kDa protein and the C termini of HMCS and ABA3. The upper band in Fig. 4B may derive from the 35-kDa protein with post-translational modifications, which influence the migration pattern.
Fraction A Contains Molybdenum Cofactor-Since fraction A contains a protein related to the Moco-sulfurating enzyme HMCS, it was tested for the presence of Moco. The most sensitive assay for the detection of biologically active Moco is the so-called nit-1 assay (20), which is based on the ability of any eukaryotic Moco source to complement the apo-NADPH nitrate reductase in crude extracts of the N. crassa nit-1 mutant, which is deficient in Moco biosynthesis, thus reconstituting apo-NADPH nitrate reductase activity. Both fraction A and fraction B were able to reconstitute apo-NADPH nitrate reductase activity, indicating that both fractions contain Moco-binding proteins (Table 3). Although only a moderate enrichment of reconstituted apo-NADPH nitrate reductase activity was obtained in fraction B (3-fold), fraction A showed a 37-fold enrichment of this activity by purification from the OMV fraction. Chemical detection of FormA-dephospho, the oxidation product of Moco (24), con-

TABLE 1 Purification of mitochondrial N-reductive components
Benzamidoxime reductase activity was determined by HPLC as described under "Experimental Procedures." Benzamidoxime reductase activities are means of four purifications. Protein content was determined by bicinchonic acid (17). Data are means of two determinations. Benzamidoxime reductase activity can only be determined in the reconstituted assay system consisting of the N-reductive component as purified and cyt b 5 /cyt b 5 reductase added. Therefore, all purification calculations are based on the activities using this assay system.

HMCS C-terminal Domain Is Able to Reduce
Benzamidoxime-Since both the 35-kDa protein from fraction A and HMCS-CT share not only sequence homologies but also bind Moco, the question raised whether HMCS-CT could replace the 35-kDa protein enzymatically (i.e. whether it would be able to reduce benzamidoxime). As shown in Table  4, HMCS-CT was able to catalyze the reduction of benzamidoxime to benzamidine. The formation of benzamidine was also proven by liquid chromatography/MS. Interestingly, only HMCS-CT protein that was expressed in E. coli TP1000 and that therefore had Moco bound was capable of reducing benzamidoxime. Control incubations with HMCS-CT protein that was either binding molybdopterin, a molybdenumfree precursor of Moco, or that had no pterin bound due to expression in E. coli RK5206 and RK5204, respectively, showed no reductive activity, indicating that the molybdenum center is essential for the catalytic activity. In contrast to the mitochondrial protein, HMCS-CT is able to catalyze the benzamidoxime reduction without cyt b 5 and its reductase. A minor N-reductive activity was also detectable without adding NADH (Table 4). The reaction was linear with protein content and time over ϳ10 min (data not shown) and followed Michaelis-Menten kinetics. The apparent K m for the reaction was 15 M, and V max was 0.24 nmol of benzamidine/min/mg.

DISCUSSION
Prodrug principles were developed to enhance oral bioavailability of those drugs that are charged under physiological conditions and therefore impaired in intestinal absorption (1). For the basic amidines, N-hydroxylation converts them to the less basic N-hydroxyamidines that are more easily absorbed (2). Two cellular compartments have been identified to catalyze this reaction, the microsomes (5,8) and the mitochondrial outer membrane (3,4), harboring a multicomponent enzyme system consisting of cyt b 5 , its reductase, and a third protein (Fig. 1). In this study, the missing protein was isolated from a mitochondrial OMV fraction free from cyt b 5 and cyt b 5 reductase, since these enzymes were previously claimed to be solely responsible for the reduction of amidoximes (32). This was accomplished by an improved purification protocol for mitochondrial outer membranes and a gentle solubilization process using Zwittergent 3-14. The N-reductive activity of this protein fraction can only be determined in a reconstitution assay containing all three components of the N-reductive enzyme complex (i.e. cyt b 5 and its reductase have to be added). Here the well defined purified microsomal proteins cyt b 5 (recombinantly expressed) and its reductase (biochemically purified from pig liver) were used. Microsomal cyt b 5 has a protein fold similar to cyt b 5 from the outer mitochondrial membrane (33) and was previously shown to enhance the N-reductive activity of the enzyme system located in the outer mitochondrial membrane fraction (4). The in vitro reconstitution assays required   ND a a ND, not detectable (below the limit of detection); limit of detection ϭ 17 nmol/min/mg of fraction A. b ND, not detectable (below the limit of detection); limit of detection ϭ 1 nmol/min/mg of total protein. c ND, not detectable (below the limit of detection); limit of detection ϭ 1 nmol/min/mg protein of fraction A and cyt b 5 .
no additional phospholipids, probably due to Zwittergent 3-14 present in fraction A, which is known to simulate the hydrophobic environment of native membranes (34). The protein fraction free from cyt b 5 and cyt b 5 reductase contained monoamine oxidase-B and a 35-kDa protein with significant similarities to the C-terminal domain of the Moco sulfurase HMCS. Recombinant monoamine oxidase-B failed to reduce benzamidoxime in the reconstituted system (data not shown), so this enzyme was excluded from being the N-reductive component in fraction A.
Antibodies raised against the C-terminal domain of the Moco sulfurase ABA3 from A. thaliana revealed cross-reaction to this 35-kDa protein, indicating that they share structural similarities. Determination of the Moco-dependent nit-1 reconstitution activity showed an ϳ40-fold enrichment of nit-1 reconstitution activity and, simultaneously, of the N-reductive activity of fraction A. In common with the strong inhibitory effect of vanadate, an inhibitor of molybdenum enzymes (35), it is assumed that this 35-kDa Moco-containing protein represents the N-reductive component in fraction A. Since benzamidoxime was the model amidoxime used in this study, the 35-kDa Moco-containing protein was named mARC, which stands for mitochondrial benzamidoxime reducing component. However, other N-hydroxylated structures, such as N-hy-droxyguanidines, N-hydroxyamidinohydrazones, and hydroxylamines, will probably also be reduced by the enzyme system described here. Further studies to prove this assumption are in process. Preliminary results already demonstrated the mitochondrial reduction of other amidoximes and N-hydroxyamidinohydrazones (4), so in general this mARC is an N-hydroxy reducing system. One can expect that it is also of great importance in reducing mutagenic and toxic N-hydroxylated aromatic hydroxylamines as part of a detoxification system.
The discovery of a novel Moco-containing protein in the outer mitochondrial membrane is in agreement with the findings of Johnson et al. (36), who almost 3 decades ago detected considerable amounts of Moco associated with the outer membrane of rat liver mitochondria that did not derive from the known molybdenum enzymes sulfite oxidase or xanthine oxidase. At that time, the actual Moco source could not be identified. Hence, with the present work, it becomes likely that the Moco-containing fraction as described by Johnson et al. (36) corresponds to the mARC protein described here.
Since mARC is a homolog of the C-terminal domain of HMCS, the latter was taken for in vitro biotransformation studies. Although the physiological function of HMCS is different from N-reduction, HMCS-CT developed N-reductive catalytic activity. Most importantly, this activity was strictly dependent on Moco, since control incubations with HMCS-CT that was either binding molybdopterin or that had no Moco bound, showed no reductive activity. Thus, it was proven that the molybdenum center is essential for the catalytic activity. Such N-reductive potential has been described earlier for the molybdenum enzymes xanthine oxidase and aldehyde oxidase (37,38), thereby supporting the assumption that Moco-binding enzymes play a fundamental role in reducing N-hydroxylated substrates.
In summary, this study provided evidence that besides cyt b 5 and its reductase, a Moco-binding protein participates predominantly in the mitochondrial N-reductive enzyme system. This mARC protein represents a novel molybdenum enzyme hitherto not described in eukaryotes or bacteria. It is also unique, since it forms the catalytic part of a three-component enzyme complex. This is different from the other three molybdenum enzymes known in mammals, namely sulfite oxidase, xanthine oxidase, and aldehyde oxidase, where the single components of the electron transport chains are comprised within a single polypeptide chain. When comparing the arrangement of these electron transport chains, the N-reductive complex in the outer mitochondrial membrane is strikingly similar to eukary-

TABLE 3 Identification of molybdenum cofactor by reconstitution of nit-1 NADPH nitrate reductase activity
Moco-dependent reconstitution of NADPH nitrate reductase activity was measured by co-incubation of N. crassa nit-1 extracts and the respective mitochondrial subfractions. Data are means Ϯ S.D. of two determinations.

TABLE 4 N-Reduction of benzamidoxime by HMCS-CT
The incubation mixture consisted of 135 g of purified recombinant HMCS-CT as expressed in E. coli TP1000, RK5206, or RK5204, respectively, 0.5 mM benzamidoxime, and 1 mM NADH in 150 l of 100 mM phosphate buffer, pH 6.0. The specific activity of HMCS-CT from TP1000 without co-substrate was significantly (p Ͻ 0.05) lower. otic nitrate reductases, occurring in plants, algae, and fungi, that consist of a single polypeptide chain with a cyt b 5 domain, a cyt b 5 reductase domain, and Moco-binding domain (39). Bacterial nitrate reductases, however, consist of separate proteins forming a multicomponent complex (40). Hence, the N-reductive complex in the outer mitochondrial membrane with its mARC protein represents the first case for a eukaryotic molybdenum enzyme consisting of separate proteins. The isolation of this reductase complex will help to test new N-hydroxylated compounds developed as prodrugs. In vitro assays with this complex could predict if a new prodrug would be reduced in vivo.