Báo cáo khoa học: Cytochrome P460 of Nitrosomonas europaea Formation of the heme-lysine cross-link in a heterologous host and mutagenic conversion to a non-cross-linked cytochrome c ¢ pot

Cytochrome P460 of Nitrosomonas europaea Formation of the heme-lysine cross-link in a heterologous host and mutagenic conversion to a non-cross-linked cytochrome c ¢ David J. Bergmann 1 and Alan B. Hooper 2 1 Department of Biology, Black Hills State University, Spearfish, SD, USA; 2 Department of Biochemistry, Molecular Biology, and Biophysics, University of Minnesota, MN, USA The heme of cytochrome P460 of Nitrosomonas europaea, which is covalently crosslinked to two cysteines of the polypeptide as with all c-type cytochromes, has an additional novel covalent crosslink to lysine 70 of the polypeptide [Arciero, D.M. & Hooper, A.B. (1997) FEBS Lett. 410, 457– 460]. The protein can catalyze the oxidation of hydroxyl- amine. The gene for this protein, cyp,wasexpressedin Pseudomonas aeruginosa strain PAO lacI, resulting in for- mation of a holo-cytochrome P460 which closely resembled native cytochrome P460 purified from N. europaea in its UV-visible spectroscopic, ligand binding and catalytic properties. Mutant versions of cytochrome P460 of N. europaea in which Lys70 70 was replaced by Arg, Ala, or Tyr, retained ligand-binding ability but lost catalytic ability and differed in optical spectra which, instead, closely resembled those of cytochromes c¢. Tryptic fragments con- taining the c-heme joined only by two thioether linkages were observed by MALDI-TOF for the mutant cyto- chromes P460 K70R and K70A but not in wild-type cyto- chrome P460, consistent with the structural modification of the c-heme only in the wild-type cytochrome. The present observations support the hypothesized evolutionary relationship between cytochromes P460 and cytochromes c¢ in N. europaea and M. capsulatus [Bergmann, D.J., Zahn, J.A., & DiSpirito, A.A. (2000) Arch. Microbiol. 173, 29–34], confirm the importance of a heme-crosslink to the spectro- scopic properties and catalysis and suggest that the crosslink might form auto-catalytically. Keywords: cytochrome c¢; cytochrome P460; hydroxyl- amine; nitric oxide, Nitrosomonas. Cytochromes P460 are mono-heme cytochromes character- ized by Soret absorption maxima at approximately 435, 460 and 450 nm in the ferric, ferrous and ferrous-CO forms, respectively [1–4]. Although the protein catalyzes the oxidation of hydroxylamine, its physiological role has not been clearly elucidated. Cytochrome P460 of Nitrosomonas is a homo-trimer [2,3] or possibly -dimer [1] of 18 kDa subunits. Several unique features of the optical spectra of cytochrome P460 are shared by Ôheme-P460Õ, the active site heme of hydroxyl- amine oxidoreductase (HAO) of N. europaea. HAO is a homo-trimer of octa-heme subunits which catalyzes high rates of dehydrogenation of hydroxylamine [5–8]. In addition to two thioether linkages to cysteine residues, active site hemes of cytochrome P460 or HAO have a third covalent linkage from a heme to Tyr467 in the adjacent subunit of HAO [7,9] or from the Ôheme P460Õ to Lys70 in cytochrome P460 [10], respectively. However, the two enzymes have no similarity in amino acid sequence [11,12]. Cytochromes P460 have been characterized from the autotrophic ammonia oxidizing bacterium Nitrosomonas europaea of the b-sub- division proteobacteria [1] and from the methane oxidizing bacterium Methylococcus capsulatus Bath of the c-subdivi- sion [14]. The amino acid sequences of cytochromes P460 from both N. europaea and M. capsulatus Bath have signi- ficant similarity to that of cytochrome c¢ from M. capsulatus Bath, suggesting a possible evolutionary link between the three cytochromes [13]. Cytochromes c¢, which are found in a wide variety of photosynthetic, denitrifying, and methano- trophic bacteria, are homo-dimers of 16 kDa to 18 kDa subunits which have one penta-coordinate c-type heme capable of binding small ligands such as NO or CO [14,15]. Although cytochrome c¢ of M. capsulatus has spectroscopic properties that are unique to the other cytochromes c¢ it exhibits only weak homology in primary structure with the majority of cytochromes c¢ [13]. We hypothesized that if the ancestral form of cytochrome P460 was a cytochrome c¢ which evolved by the acquisition of a third covalent crosslink to the heme (causing it to gain the ability to catalyze hydroxylamine oxidation), then mutants of cytochrome P460, in which the Lys70 is replaced byanaminoacidresidueunabletocrosslinktotheactive site heme, might well have spectroscopic, ligand-binding and catalytic properties similar to cytochromes c¢.Inthis paper, we report the expression in Pseudomonas aeruginosa PAO lacI of wild-type cytochrome P460 and site-directed mutants in which Lys70 was replaced by arginine, alanine or tyrosine. The wild-type cytochrome P460 expressed in Correspondence to A.B.Hooper,DepartmentofBiochemistry, Molecular Biology and Biophysics, University of Minnesota, St. Paul, MN 55108, USA. Fax: + 1 612 625 5780, Tel.: + 1 612 624 4930, E-mail: hooper@cbs.umn.edu Abbreviations: HAO, hydroxylamine oxidoreductase. (Received 27 December 2002, revised 19 February 2003, accepted 3 March 2003) Eur. J. Biochem. 270, 1935–1941 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03550.x P. aeruginosa had properties very similar to that from N. euroapea. However, the mutant cytochromes P460 had properties very different from wild-type cytochrome P460 and similar to cytochromes c¢. Materials and methods DNA techniques The gene encoding cytochrome P460 of N. europaea, cyp, was amplified by PCR from a 7.8-kb BamHI fragment of genomic DNA which had previously been subcloned into the plasmid vector pUC119 [16]. The forward primer, 5¢-GCTACCATATGAAAACAGCTTGGTAGGT-3¢,en- compassed the ATG start codon of cyp and contained a 5¢ extension with an NdeI site. The reverse PCR primer, 5¢-CCTGATTCGTTCTGCTACCT-3¢, bound to a region just downstream of a cyp and a native SmaI–XmaIsite. PCR reaction mixture (100 lL) was used (PCR Super Mix, Life Technologies, Inc., Gaithersburg, MD, USA) contain- ing 0.2 nmol of each primer, 2 ng template, 0.2 m M dNTPs, 50 m M Tris/HCl (pH 8.4), 1.5 m M MgCl, and 1.0 U Taq DNA polymerase. After denaturation of the PCR mixture at 94 °C for 5 min, 30 cycles of 94 °Cfor30s,60°Cfor 30 s, and 72 °C for 30 s were performed, and the reaction incubated for 7 min at 72 °C and stored at 4 °C. The PCR product was purified using a spin-column kit (Qiagen, Inc., Valencia, CA, USA). The PCR product and the expression vector pUCPNde [17] were digested with NdeIandXmaI as directed by the manufacturer (Promega, Inc., Madison, WI, USA). The digested PCR product and expression vector were ligated and transformed into frozen competant Escherichia coli strain DH5aF¢IQ cells (Life Technologies) using standard methods [18]. Transformed colonies were grown on LB (Luria–Bertani) media with 100 lgÆmL )1 ampicillin, and plasmid DNA harvested by the alkaline lysis technique [18]. The orientation of cyp subcloned in the expression vector was confirmed by dideoxy dye-primer cycle sequencing using an ABI Model 377 DNA Sequencer at the University of Minnesota AGAC sequencing facility. The resulting plasmid, pUCYP2, con- tained the wild-type cyp gene downstream of the lac promoter and ribosome-binding site of the vector. Threesitedirectedmutantsofcyp, in which the codon, AAA, encoding Lys70 of cytochrome P460 is converted to AGA (Arg), GCA (Ala), or TAT (Tyr) were produced from pUCYP2 with the Transformer site directed mutagenesis kit (Clontech, Inc., Palo Alto, CA, USA) using the method of Deng and Nicloff [19], as directed by the manufacturer. The selection oligonucleotide, 5¢-AAATGCTTCAATGATAT CGAAAAAGGAAG-3¢, converted a unique SspIsiteon the vector to an EcoRV site. The mutagenesis oligonucleo- tides were 5¢-GTAACTGTAAGAGAACTGGTCAC-3¢ (Lys70 to Arg), 5¢-GTAACTGTAGCAGAACTGGTCA G-3¢ (Lys70 to Ala), and 5¢-GGTAACTGTATATGAA CTGGTCAG-3¢ (Lys70 to Tyr). The resulting plasmids, pUCYPKR, pUCYPKA, and pUCYPKY, were trans- formed into cells [20]. The plasmids pUCYP2, pUCYPKR, pUCYPKA, and pUCYPKY were each introduced into Pseudomonas aeruginosa strain PAO-LacI by electropora- tion using an Electroporator 2510 (Eppendorf Co., West- bury, NY, USA) as described by Cronin and McIntire [17]. Growth of cells and purification of cytochrome P460 Cells of P. aeruginosa PAO lacI-containing plasmids pUCYP2, pUCYPKR, pUCYPKA, or pUCYPKY were grown and periplasmic extracts prepared as described by Cronin and McIntire [17]. The periplasmic extract was dialyzed in Union Carbide 5 cm-wide dialysis bags (m cut- off < 25 kDa) overnight at 5 °C against 6 L of KP i buffer (50 m M potassium phosphate, pH 7.5). Ammonium sulfate was then added to the periplasmic extract to 75% saturation andstirredfor45minat5°C before centrifugation at 15 000 g for 15 min at 5 °C. The pellet was discarded, and the supernatant brought to 100% saturation in ammonium sulfate and stirred for 45 min at 5 °C. After centrifugation at 15 000 g for 15 min at 5 °C, the pellet was resuspended in 20 mL KP i buffer and dialyzed overnight against 4 L of KP i buffer. KCl was then added to 200 m M and the sample concentrated to 1 mL on an Amicon stirred filtration cell with a YM 10 membrane and Centricon 10 microconcen- tration devices (Amicon-Grace Co., Danvers, MA, USA). The sample was added to a 2.5-cm diameter · 110 cm long column of Sephadex G-100 (Sigma Chemical Co., St. Louis, MO, USA) equilibrated with 200 m M KCl in KP i buffer. Cytochrome P460 eluted as a greenish-yellow band, free of endogenous c-type cytochromes from P. aeruginosa, but still containing other, nonheme, protein contaminants (Fig. 1) Fig. 1. SDS/PAGE (15% acrylamide/bisacrylamide) of wild-type cytochrome P460 from N . europaea and P. aeruginosa and mutant cytochromes P460 expressed in P. aeruginosa. (A) Stained with Coo- massie Blue R-250. (B) Stained for heme by the method of Goodhew et al. [30]. Lanes 1 and 7, high range prestained molecular mass markers (Life Technologies); lane 2, wild-type cytochrome P460 purified from N. europaea; lanes 3–6, cytochromes P460 expressed in and partially purified from the periplasm of P. aeruginosa;lane3, wild-type cytochrome P460; lane 4, cytochrome P460 K70R; lane 5, cytochrome P460 K70A; lane 6, cytochrome P460 K70Y. 1936 D. J. Bergmann and A. B. Hooper (Eur. J. Biochem. 270) Ó FEBS 2003 and had a Soret : 280 nm absorbance ratio of approxi- mately 2.0. Optical absorption spectroscopy was performed with a Hewlett-Packard 8452 diode-array spectrophotometer (Agilent Technologies). Cytochrome P460 from N. euro- paea was prepared by D. M. Arciero, University of Minnesota, St Paul, MN, USA, as described previously [21]. Cytochrome c552 was obtained from N. europaea by D. M. Arciero as described earlier [22] for use as an electron acceptor in assays of hydroxylamine and hydrazine oxida- tion by cytochrome P460. In these assays, the absorbance of 9 l M cytochrome c552 in 1 mL of KP i buffer at pH 7.5 was monitored at 552 nm in the presence of substrate and cytochrome P460 at 22 °C. SDS/PAGE, tryptic digests of cytochrome P460 and mass spectrometry Approximately 0.5–1.0 nmole of cytochrome P460 was denatured in 1· loading buffer [60 m M Tris/HCl (pH 6.8), 10% v/v glycerol, 1% w/v SDS] for 30 min at room temperature and was loaded onto a (15 : 0.4%) acrylamide- bisacrylamide gel for electrophoresis at room temperature using the Laemmli buffer system [23]. The yellow or orange cytochrome P460 band was excised from the gel and the cytochrome digested within the gel slice with porcine sequencing grade trypsin (Promega Corp., Madison WI, USA) and then eluted from the slice as described by Shevchenko et al.[24].PriortoMALDI-TOFanalysis,the sample was desalted using C18 ZipTips using the protocol described by the manufacturer (Millipore Corp., Bedford, MA, USA), with the following modification: the elution buffer was 75 : 25, acetonitrile/water, 0.1% trifluoroacetic acid. The instrument used for the collection of Matrix- Assisted Laser Desorption/Ionization Time-of-Flight (MALDI-TOF) mass spectrometric data was a Bruker Biflex III, equipped with an N 2 -laser (337 nm, 3 nanosec- ond pulse length) and a microchannel plate detector. The data was collected in the reflectron mode, positive polarity, with an accelerating potential of 19 kV. Each spectrum is the accumulation of 200 laser shots. External calibration was performed using human angiotensin II (monoisotopic mass [MH+] 1046.5) and adrenocorticotropin hormone (ACTH) fragment 18–39 (monoisotopic mass [MH+] 2465.2; Sigma Chemical Co., St Louis, MO, USA). The matrix used for samples and standards was cyano-4- hydroxycinnamic acid (4-HCCA; Hewlett-Packard, sold in solution, in methanol) diluted 1 : 1 with 50 : 50, acetonit- rile/nanopure water, 0.1% trifluoroacetic acid. HPLC grade acetonitrile was purchased from Fisher Scientific and 99+% spectrophotometric grade trifluoroacetic acid was purchased from Aldrich, (Milwaukee, WI, USA). The presence of heme in fragments was confirmed by the pattern from the 54/56 Fe isotope-distribution: a small secondary peak was seen, which had a mass two units below the peak in consideration. The secondary peak was present only in peaks reported in Table 2 to contain heme. Results Expression of wild-type and mutant cytochrome P460 in P. aeruginosa PAO lacI, containing the plasmid pUCYP2 (which encodes the gene for wild-type cytochrome P460 from N. europaea) expressed a cytochrome whose subunit molecular size, UV/visible spectral properties and reactivity with ligands and substrates were identical with those of cytochrome P460 purified from N. europaea. The yield of the cytochrome was low, however, ranging from 0.2 to 0.4 mg of cytochrome P460 per litre of cell culture based on the intensity of the Soret peak [3]. The migration pattern on SDS/PAGE gels for samples of cytochromes P460 are shown in Fig. 1. All samples exhibited a heme-containing band of the same mobility as the subunit of purified wild-type cytochrome P460 produced in Nitrosomonas. The wild-type and K70R, K70A, and K70 mutant cytochromes P460 expressed in P. aeruginosa eluted similarly on a Sephadex G100 column; they eluted later and well-separated from the P. aeruginosa cyt. c551 (M r 12 000) thus are likely to have the same quaternary structure as the wild-type expressed in N. europaea. Optical spectroscopy Wild-type cytochrome P460 expressed in Pseudomonas was indistinguishable from wild-type cytochrome P460 expressed in Nitrosomonas [1,3] with respect to the following spectroscopic, ligand binding and catalytic properties. The optical spectrum of ferric wild-type cytochrome P460 expressed in P. aeruginosa (Fig. 2A, Table 1) had a broad Soret peak at 434 nm, and shoulders at 510 and 540 nm. The dithionite-reduced cytochrome had Soret absorbance at 462 nm, the 510/540 nm shoulder was lost and small peaks at 660 and 688 nm appeared. When CO was bubbled though the dithionite-reduced cytochrome P460, the Soret peak shifted to 448 nm and the 660/688 nm peaks were lost. The latter may have been replaced by weak bands at 620 and 670 nm. Optical changes associated with the reduction of cytochrome P460 by dithionite occurred in two phases, which are not understood mechanistically but are com- monly observed with cytochrome P460 from Nitrosomonas. A rapid reduction in absorbance at 434 nm and 500 nm and increase at 688 nm preceded a slow increase at 462 nm, which was completed in 20 min. A few preparations of cytochrome P460 from P. aeruginosa lacked the broad shoulder at 510–540 nm in the oxidized state. This has also been observed with some preparations of cytochrome P460 from N. europaea and can occur during storage (D. M. Arciero, unpublished observation). The addition of several grains of potassium cyanide to ferric-cytochrome P460 expressed in P. aeruginosa caused the Soret band to decrease slightly and shift to 442 nm (data not shown). Incubation of ferric-cytochrome P460 with 100 l M hydroxylamine in 50 m M phosphate solution, pH 7.5, caused the Soret band to increase slightly while absorbance at 500 nm decreased and absorbance at approximately 620 nm increased (data not shown). Incuba- tion of ferric-cytochrome P460 with 100 l M hydrazine caused the Soret absorbance peak to initially shift to 438 nm and increase in intensity then greatly diminish over several minutes, while absorbance at 500 nm decreased and absorbance at 620 increased slightly (data not shown). Cytochrome P460 expressed in P. aeruginosa catalyzed the oxidation of either hydroxylamine or hydrazine, using cytochrome c552 from N. europaea as an electron acceptor. Ó FEBS 2003 Cytochrome c¢-like traits of cytochrome P460 (Eur. J. Biochem. 270) 1937 In the presence of 5 m M hydroxylamine or hydrazine, respectively, turnover numbers as high as 10.8 or 0.3 mol cytochrome c552 reduced per minute per mole of cyto- chrome P460 were obtained. These values are comparable to values measured with cytochrome P460 from Nitroso- monas [3]. In mutant versions of cytochrome P460 expressed in P. aeruginosa, Lys70 was replaced by arginine, alanine, or tyrosine to form cytochromes P460 K70R, P460 K70A, or P460 K70Y, respectively. None of the mutant cytochromes P460 catalyzed the oxidation of hydroxylamine using N. europaea cytochrome c552 as an electron acceptor nor did the optical spectra of any of the three mutant ferric- cytochromes change in the presence of hydroxylamine, hydrazine, or potassium cyanide. The UV-visible spectra of the P460 mutants K70R, K70A, and K70Y were similar to each other but were strikingly different from those of wild-type cytochrome P460 produced in N. europaea or P. aeruginosa (Fig. 2B–D, Table 2). Mutant ferric-cyto- chromes P460 had Soret peaks in the range 392–404 nm, smaller broad peaks at 498 and 540 nm and an even smaller peak in the range 622–638 nm. Mutant cytochromes P460 were rapidly reduced by dithionite and the spectra of the ferrous forms displayed Soret peaks in the range 432– 434 nm and small peaks at 552–554 nm and 590 nm. In common with cytochromes c¢ they lacked distinct a and b peaks in the ferrous form. After bubbling CO into solutions of the dithionite-reduced mutant cytochromes P460, the Soret peak shifted to 416–418 nm and smaller peaks shifted to 532–534 nm and 562–564 nm. These spectral features are strikingly similar to those observed in cytochromes c¢ from Methylococcus capsulatus [25] and Paracoccus denitrificans [26]. Proteolysis and MALDI-TOF MS Tryptic fragments of cytochromes P460 were prepared and analyzed by MALDI-TOF mass spectrometry (Table 2). MALDI-TOF spectra of tryptic fragments of wild-type cytochrome P460 produced by either N. europaea or P. aeruginosa (Table 2) were nearly identical. This sugges- ted, again, that the wild-type cytochrome P460 is expressed in its native crosslinked form in the heterologous host. Heme-containing tryptic peptide fragments representing the cysteine-containing residues #130–145, NLPTAECA ACHKENAK (M r 2314.5), or residues #130–141, NLPTA ECAACHK (M r 1872.4) were not observed in the MALDI-TOF spectrum in digests of either of the wild-type Table 1. Absorption maxima of cytochromes P460 produced in P. aeruginosa. Cytochromes P460 include K70 (wild-type), K70R, K70A and K70Y. Data are from Fig. 2. Type of cytochrome P460 Absorption maxima (nm) Ferric cytochrome Ferrous cytochrome Ferrous + CO cytochrome K70 (wt) 434, 510, 540 462, 660, 688 448 (620, 670?) K70R 392, 498, 540, 638 434, 554, 590 418, 534, 566 K70A 402, 498, 540, 622 432, 552, 590 416. 532, 562 K70Y 404, 498, 540, 628 432, 554, 590 416, 534, 564 Fig. 2. UV/visible spectra of cytochromes P460 expressed in wild-type P. aeruginosa. The purified cytochromes are in 50 m M potassium phosphate solution, pH 7.5. Spectra for resting state of cytochrome, as isolated, (solid line); dithionite-reduced (dotted-and-dashed line); dithionite-reduced cytochrome + CO (dotted line). (A) Wild-type cytochrome P460 K70, (B) cytochrome P460 K70R, (C) cytochrome P460 K70A, (D) cytochrome P460 K70Y. Wavelength of peaks are labeled for ferric-, ferrous- and ferrous + CO-cytochrome P460 in D, C and B, respectively. 1938 D. J. Bergmann and A. B. Hooper (Eur. J. Biochem. 270) Ó FEBS 2003 cytochromes. In addition, no detectable tryptic fragment from a wild-type cytochrome P460 contained either of these two heme-containing polypeptides cross-linked to another peptide containing Lys70. It is not known why these tryptic/ MALDI-TOF fragments, which are predicted from the structure of wild-type cytochrome P460, are absent. The chromophoric lysyl-heme-di-cysteinyl cross-linked dipep- tide is known to be very labile and to require extreme care for its isolation [10]. Hence it is likely to have been degraded to a family of fragments at some step in the analysis. Alternatively it may have been lost during elution from the gel or desalting of the eluate or was not desorbed/ionized during MS analysis. Free heme was not observed in the MALDI-TOF spectrum of tryptic fragments of the wild- type cytochromes. The appearance of some of the Lys70- containing peptide of residues #64–78 (lacking heme) in each of the spectra of the wild-type proteins might suggest that the K70 crosslinks had not ever formed in some molecules of cytochrome P460 or were broken during the analysis. The tryptic peptides of two mutant cytochromes P460, P460 K70R and P460 K70A, had nearly identical MALDI- TOF spectra, however, their spectra were substantially different from corresponding MALDI-TOF spectra of wild-type cytochromes P460 (Table 2). The spectra of tryptic peptides of mutant cytochromes P460 K70R and P460 K70A contained a 617.2-Da fragment representing free heme and fragments of M r 2314.5 (residues #130–145, NLPTAECA ACHKENAK) or M r 1872.4 (residues #130–141, NLPTA ECAACHK) representing heme bound to tryptic peptides through thioether linkages to cysteine residues. The presence of the latter two fragments from the mutant cytochromes P460 in contrast to the absence of heme-cysteinyl-linked peptides in the fragments from the wild-type protein is in keeping with an apparent heme lability imparted by the crosslink to lysine in the wild-type protein [10]. Fragments representing the heme crosslinked to two peptides through thioether linkages to two cysteines and also to a third residue (such as Lys70) were not found by MALDI-TOF of tryptic peptides of P460 K70R or P460 K70A. Discussion The product of expression of the gene encoding wild-type cytochrome P460 from N. europaea in P. aeruginosa was a cytochrome with catalytic and ligand-binding capabilities and UV-visible spectroscopic properties identical to cyto- chrome p460 expressed in N. europaea.Thisimpliesthat formation of the covalent crosslink between Lys70 and the heme of cytochrome P460 is facilitated by an enzyme present in both Nitrosomonas and the heterologous host or is, in fact, auto-catalytic. The presence, in addition to the lysyl-heme crosslink, of the two thioether bonds from heme vinyl groups to peptide cysteines means that cytochrome P460 can be thought of as a modified c-cytochrome. By comparison with typical c-type cytochromes, the spectra of ferric cytochrome P460 has a greatly red-shifted Soret maximum, low and broad 510 and 540 nm shoulders and a small maximum at 688 nm. On reduction, the Soret shifts to 460 nm and the 500 and 688 nm peaks disappear but a and b maxima typical of c-type cytochromes do not appear. The addition of CO causes the Soret maximum to shift to 450 nm. The UV/ visible spectra of the ferric, ferrous or ferrous-CO forms of the mutant cytochromes P460, K70R, K70A, and K70Y, all lack the characteristic red-shifted 435, 460 or 450 nm absorption maxima of the wild-type cytochrome P460 thus confirming the role of the crosslink in determining the Soret spectrum. Although catalytic activity was lost in the mutants, the ligand-binding capability and thus the penta- coordinate nature of the heme was conserved. Significantly, typical c-cytochrome a and b maxima do not appear in the Table 2. Relative intensities, as detected by MALDI, of identifiable tryptic fragments of cytochromes P460 from N. europaea (NE) and P. aeruginosa (PA). Cytochrome P460 70K (wild-type) isolated from NE or heterologously expressed in and purified from PA and mutants K70R and K70A heterologously expressed in and purified from PA. Data were obtained as described in Materials and methods and analyzed by the methods of Wilkins et al. [29]. Residue number is based on the sequence of the mature protein. Fragment M r P460–70K (NE) P460–70K (PA) P460-K70R (PA) P460-K70A (PA) Cytochrome P460 residues 2831.4 0.023 0.0 0.025 0.020 20–44 2314.5 0.0 0.0 0.193 0.125 130–145 + Heme 2247.6 0.105 0.104 0.093 0.079 79–100 2119.5 0.030 0.052 0.016 0.016 80–100 2016.6 0.059 0.058 0.017 0.019 20–37 1872.4 0.0 0.0 0.172 0.154 130–141 + Heme 1722.6 0.019 0.014 0.0 0.0 45–58 1700.5 0.130 0.067 0.031 0.019 1–16 1632.5 0.183 0.133 0.031 0.032 unknown 1616.5 1.000 1.000 1.000 1.000 146–158 1594.3 0.413 0.362 0.073 0.065 45–57 1575.5 0.045 0.031 0.0 0.0 64–78 874.4 0.026 0.020 0.013 0.0 71–78 833.4 0.433 0.683 0.072 0.068 38–44 769.4 0.0 0.013 0.0 0.0 10–16 737.4 0.040 0.017 0.072 0.0 58–63 617.2 0.0 0.0 0.129 0.121 Heme 609.4 0.0 0.015 0.0 0.0 59–63 Ó FEBS 2003 Cytochrome c¢-like traits of cytochrome P460 (Eur. J. Biochem. 270) 1939 mutant proteins; rather, the ferric, ferrous and ferrous-CO absorption spectra strongly resemble those of cyto- chromes c¢ [25,26]. It appears that, in the absence of the lysine crosslink, the resulting heme environment of mutant cytochrome P460 is typical of cytochromes c¢ rather than the more common c-cytochromes. A degree of sequence similarity between the cytochromes P460 and c¢ has lead to the hypothesis that cytochromes P460 may have evolved from cytochromes c¢ [13]. We note that the ancestor of cytochromes c¢ and P460 could also have been a catalyti- cally active and heme-crosslinked cytochrome P460-like protein. The present observations further document an evolutionary relationship between the two proteins. In HAO the catalytic heme is crosslinked to a tyrosine of the adjacent subunit [7] and forms the catalytic Ôheme P460Õ with a ferrous absorption maximum of 460 nm. As shown here by their elution behavior in Sephadex G-100, the wild- type or mutant cytochromes P460 appear also to be oligomeric. Thus intersubunit crosslinking would have been at least theoretically possible. The K to Y mutant of cytochrome P460 was constructed for the present report to test the remote possibility that a heme P460 derivative homologous to heme P460 of HAO would be formed. That the chromophore was not formed may result from differ- ences in the chemical nature of the lysyl- or tyrosyl- linkage or the relative orientation of K70 and Y70 in the cytochromes P460. In nature, the lysyl- or tyrosyl- to heme crosslink appears to be of key importance to catalysis of electron and/or proton removal from substrate [27]. This is illustrated by the difference between the dimeric nitrite reductase, NrfA, the monomer of which is a penta-heme c-cytochrome lacking a covalent crosslink to the catalytic heme [28] and hydroxyl- amine oxidoreductase (a dehydrogenase), in which the monomer is a octa-heme c-cytochrome possessing a tyrosine crosslink to the catalytic heme. Cytochromes c¢, lacking the lysyl-heme crosslink, can reversibly bind but not transform NO whereas cytochrome P460, having the lysine crosslink, can bind and catalyze the transformation of NH 2 OH. The relevance of the lysyl-heme crosslink to catalysis is suppor- ted here by the lack of catalytic ability in mutant cytochromes P460 lacking the lysine crosslink. Acknowledgements We thank Bradley Elmore, Ciarran Cronin, David Arciero, Mark Whittaker, Michelle Wagner, and Leann Higgins for their help with this project. This research was funded by the National Science Foundation (MCB-9723608). References 1. Erickson, R.H. & Hooper, A.B. (1970) Preliminary characteriza- tion of a variant CO-binding heme protein from Nitrosomonas. Biochim. Biophys. Acta 275, 231–244. 2. Miller, D.J., Wood, P.M. & Nichols, D.J.D. (1984) Further characterization of cytochrome P-460 in Nitrosonomas europaea. J. Gen. Microbiol. 130, 3049–3054. 3. Numata, M., Saito, T., Yamakazi, T., Fukumori, Y. & Yama- naka, T. (1990) Cytochrome p460 of Nitrosomonas europaea: further purification and further characterization. J. Biochem. 108, 1016–1021. 4. Zahn, J.A., Duncan, C. & DiSpirito, A.A. (1994) Oxidation of hydroxylamine by cytochrome P-460 of the obligate methano- troph, Methylococcus capsulatus Bath. J. Bacteriol. 176, 5879– 5887. 5. Hooper, A. & Nason, A. (1965) Characterization of hydro- xylamine-cytochrome c reductase from the chemoautotrophs Nitrosomonas europaea and Nitrosocystis oceanus. J. Biol. Chem. 240, 4044–4057. 6. Arciero, D.M. & Hooper, A.B. (1993) Hydroxylamine oxido- reductase is a multimer of an octaheme subunit. J. Biol. Chem. 268, 14645–14654. 7. Igarashi, N., Moriyama, H., Fujiwara, T., Fukumori, Y. & Tanaka, N. (1997) The 2.8 A structure of hydroxylamine oxidoreductase from a nitrifying chemolithotrophic bacterium, Nitrosomonas europaea. Nat. Struct. Biol. 4, 276–284. 8. Hendrich, M.P., Petasis, D., Arciero, D.M. & Hooper, A.B. (2001) Correlations of structure and electronic properties from EPR spectroscopy of hydroxylamine oxidoreductase. J. Am. Chem. Soc. 123, 2997–3005. 9. Arciero, D.M., Hooper, A.B., Cai, M. & Timkovitch, R. (1993) Evidence for the structure of the active site heme P460 in hydro- xylamine oxidoreductase of Nitrosomonas europaea. Biochemistry 32, 9370–9378. 10. Arciero, D.M. & Hooper, A.B. (1997) Evidence for a crosslink between c-heme and a lysine residue in cytochrome P460 of Nitrosomonas europaea. FEBS Lett. 410, 457–460. 11. Sayavedra-Soto, L.A., Hommes, N.G. & Arp, D.A. (1994) Char- acterization of the gene encoding hydroxylamine oxidoreductase in Nitrosomonas europaea. J. Bacteriol. 176, 504–510. 12. Bergmann,D.J.&Hooper,A.B.(1994)Theprimarystructureof cytochrome P460 of Nitrosomonas europaea:presenceofac-heme binding motif. FEBS Lett. 352, 324–326. 13. Bergmann, D.J., Zahn, J.A. & DiSpirito, A.A. (2000) Primary structure of cytochrome c¢ of Methylococcus capsulatus Bath: evidence of a phylogenetic link between P460 and c¢-type cyto- chromes. Arch. Microbiol. 173, 29–34. 14. Moir, J.W.B. (1999) Cytochrome c¢ from Paracoccus denitrificans: spectroscopic studies consistent with a role for the protein in nitric oxide metabolism. Biochim. Biophys. Acta 1430, 65–72. 15. Tahirov, T.T., Misaki, S., Meyer, T.E., Cusanovitch, M.A., Higuchi, Y. & Yasuoka, N. (1996) High-resolution crystal struc- tures of two polymorphs of cytochrome c¢ from the purple phototrophic bacterium Rhodobacter capsulatus. J. Mol. Biol. 259, 467–479. 16. Bergmann, D.J., Zahn, J.A., Hooper, A.B. & DiSpirito, A.A. (1998) Cytochrome P460 genes from the methanotroph Methy- lococcus capsulatus Bath. J. Bacteriol. 180, 6440–6445. 17. Cronin, C.N. & McIntire, W.S. (2000) Heterologous expression in Pseudomonas aeruginosa and purification of the 9.2-kDa c-type cytochrome subunit of p-cresol methylhydroxylase. Protein Expr Purif. 19, 74–83. 18. Sambrook, J., Frittsch, E.F. & Maniatis, T. (1989) Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. 19. Deng, W.P. & Nicloff, J.A. (1992) Site-directed mutagenesis of virtually any plasmid by eliminating a unique site. Anal. Biochem. 200, 81–88. 20. Chung, C.T., Niemala, S.L. & Miller, R.H. (1989) One-step pre- paration of competent Escherichia coli: Transfomation and sto- rage of bacterial cells in the same solution. Proc.NatlAcad.Sci. USA 86, 2172–2175. 21. Collins, M., Arciero, D.M. & Hooper, A.B. (1993) Optical spec- trophotometric resolution of the hemes of hydroxylamine oxido- reductase. Heme quantitation and pH dependence of E m . J. Biol. Chem. 268, 14655–14662. 1940 D. J. Bergmann and A. B. Hooper (Eur. J. Biochem. 270) Ó FEBS 2003 22. Arciero, D.M., Peng, Q., Peterson, J. & Hooper, A.B. (1994) Identification of axial ligands of cytochrome c552 from Nitrosomonas europaea. FEBS Lett. 342, 217–220. 23. Laemmli, U.K. (1970) Cleavage of structural proteins during assembly of the head of bacteriophage T4. Nature 277, 680–685. 24. Schevchenko,A.,Wilm,M.,Vorn,O.&Mann,M.M.(1996) Mass spectroscopy sequencing of proteins silver-stained in poly- acrylamide gels. Anal. Chem. 68, 850–858. 25. Zahn, J.A., Arciero, D.M., Hooper, A.B. & DiSpirito, A.A. (1996) Cytochrome c¢ of Methylococcus capsulatus Bath. Eur. J. Biochem. 240, 684–691. 26. Gilmour,R.,Goodhew,C.F.&Pettigrew,G.W.(1991)Cyto- chrome c¢ of Paracoccus dentrificans. Biochim. Biophys. Acta 1059, 233–238. 27. Arciero, D.M. & Hooper, A.B. (1998) Consideration of a phlorin structure for haem P-460 of hydroxylamine oxidoreductase and its implications regarding reaction mechanism. Biochem. Soc. Trans. 26, 385–389. 28. Einsle, O., Messerschmidt, A., Stach, P., Bourenkov, G.P., Bartunik, H.D., Hubner, R. & Kroneck, M.H. (1999) Struc- ture of cytochrome c nitrite reductase. Nat. Struct. Biol. 400, 476–480. 29. Wilkins, M.R., Gasteiger, E., Bairoch, A., Sanchez, J C., Williams, K.L., Appe, I.R.D. & Hochstrasser, D.F. (1998) Protein Identification and Analysis Tools in the Expasy Server. In 2-D Proteome Analysis Protocols. Humana Press, New Jersey. (http://us.expasy.org/tools/peptide-mass.html) 30. Goodhew, C.F., Brown, K.R. & Pettigrew, G.W. (1986) Haem staining in gels, a useful tool in the study of bacterial c-type cytochromes. Biochim. Biophys. Acta 852, 288–294. Ó FEBS 2003 Cytochrome c¢-like traits of cytochrome P460 (Eur. J. Biochem. 270) 1941 . SspIsiteon the vector to an EcoRV site. The mutagenesis oligonucleo- tides were 5¢- GTAACTGTAAGAGAACTGGTCAC- 3¢ (Lys70 to Arg), 5¢- GTAACTGTAGCAGAACTGGTCA G- 3¢ (Lys70. Cytochrome P460 of Nitrosomonas europaea Formation of the heme-lysine cross-link in a heterologous host and mutagenic conversion to a non-cross-linked