The Pseudomonas aeruginosa nirE gene encodes the S-adenosyl- L-methionine-dependent uroporphyrinogen III methyltransferase required for heme d 1 biosynthesis Sonja Storbeck 1 , Johannes Walther 1 , Judith Mu ¨ ller 1 , Vina Parmar 2 , Hans Martin Schiebel 3 , Dorit Kemken 4 , Thomas Du ¨ lcks 4 , Martin J. Warren 2 and Gunhild Layer 1 1 Institute of Microbiology, Technische Universita ¨ t Braunschweig, Germany 2 Department of Biosciences, University of Kent, Canterbury, UK 3 Institute of Organic Chemistry, Technische Universita ¨ t Braunschweig, Germany 4 Institute of Organic Chemistry, University of Bremen, Germany Introduction Some bacteria, such as Pseudomonas aeruginosa, use denitrification as an alternative form of respiration under conditions of low oxygen tension in the presence of nitrogen oxides (e.g. nitrate or nitrite) [1]. During denitrification, the dissimilatory nitrite reductase cata- lyzes the reduction of nitrite to nitric oxide. In denitri- Keywords heme d 1 biosynthesis; precorrin-2; Pseudomonas aeruginosa; SAM-dependent uroporphyrinogen III methyltransferase; uroporphyrinogen III Correspondence G. Layer, Institute of Microbiology, Technische Universita ¨ t, Braunschweig, Spielmannstr. 7, 38106 Braunschweig, Germany Fax: +49 531 391 5854 Tel: +49 531 391 5813 E-mail: g.layer@tu-bs.de Website: http://www.tu-braunschweig.de/ ifm (Received 23 June 2009, revised 10 August 2009, accepted 14 August 2009) doi:10.1111/j.1742-4658.2009.07306.x Biosynthesis of heme d 1 , the essential prosthetic group of the dissimilatory nitrite reductase cytochrome cd 1 , requires the methylation of the tetrapyr- role precursor uroporphyrinogen III at positions C-2 and C-7. We pro- duced Pseudomonas aeruginosa NirE, a putative S-adenosyl-l-methionine (SAM)-dependent uroporphyrinogen III methyltransferase, as a recombi- nant protein in Escherichia coli and purified it to apparent homogeneity by metal chelate and gel filtration chromatography. Analytical gel filtration of purified NirE indicated that the recombinant protein is a homodimer. NirE was shown to be a SAM-dependent uroporphyrinogen III methyltransfer- ase that catalyzes the conversion of uroporphyrinogen III into precorrin-2 in vivo and in vitro. A specific activity of 316.8 nmol of precorrin- 2h )1 Æmg )1 of NirE was found for the conversion of uroporphyrinogen III to precorrin-2. At high enzyme concentrations NirE catalyzed an overme- thylation of uroporphyrinogen III, resulting in the formation of trimethylpyrrocorphin. Substrate inhibition was observed at uroporphyri- nogen III concentrations above 17 lm. The protein did bind SAM, although not with the same avidity as reported for other SAM-dependent uroporphyrinogen III methyltransferases involved in siroheme and cobala- min biosynthesis. A P. aeruginosa nirE transposon mutant was not comple- mented by native cobA encoding the SAM-dependent uroporphyrinogen III methyltransferase involved in cobalamin formation. However, bacterial growth of the nirE mutant was observed when cobA was constitutively expressed by a complementing plasmid, underscoring the special require- ment of NirE for heme d 1 biosynthesis. Abbreviations HR-ESI-MS, high-resolution electrospray mass spectrometry; NirS, cytochrome cd 1 nitrite reductase; SAH, S-adenosyl-L-homocysteine; SAM, S-adenosyl- L-methionine; SUMT, SAM-dependent uroporphyrinogen III methyltransferase; Trx, thioredoxin. FEBS Journal 276 (2009) 5973–5982 ª 2009 The Authors Journal compilation ª 2009 FEBS 5973 fying bacteria there are two different types of dissimi- latory nitrite reductases. One is a copper-containing enzyme (NirK) and the other is the tetrapyrrole- containing cytochrome cd 1 nitrite reductase (NirS) [2]. P. aeruginosa possesses the latter enzyme [3]. NirS con- tains the tetrapyrroles heme c and heme d 1 as essential prosthetic groups [4]. Heme d 1 is a dioxo-isobacterio- chlorin, which is structurally related to siroheme and cofactor F 430 and is not a real heme [5]. The unique structural features of heme d 1 are the oxo groups on C-3 and C-8, the acrylate substituent on C-17 and the combination of acetate groups and methyl groups on C-2 and C-7, leading to partially saturated pyrrole rings A and B (Fig. 1A). The multistep biosynthesis of heme d 1 is not under- stood. All cyclic tetrapyrroles share the common precursor uroporphyrinogen III, which is converted into either hemes and (bacterio)chlorophylls via proto- porphyrin IX, or into siroheme, cofactor F 430 and cobalamin via precorrin-2 [6]. For the biosynthesis of heme d 1 , a pathway via precorrin-2 was suggested because the formation of heme d 1 requires methylation of tetrapyrrole rings A and B at positions C-2 and C-7. Indeed, it was found that the heme d 1 methyl groups attached to C-2 and C-7 are derived from methionine, probably via S-adenosyl-l-methionine (SAM) [7]. During the biosyntheses of siroheme and cobalamin, SAM-dependent uroporphyrinogen III methyltransferases (SUMTs) catalyze the methylation of uroporphyrinogen III to precorrin-2 (Fig. 1B). Several SUMTs from diverse organisms involved in siroheme (CysG, SirA, UPM1) and cobalamin (CobA) biosynthesis have been purified and bio- chemically characterized [8–13]. Most SUMTs are homodimeric proteins, except for the enzyme from Bacillus megaterium that was shown to be a monomer [9]. Some SUMTs show inhibition by the substrate uroporphyrinogen III and by the product S-adenosyl- A B C Fig. 1. Structure of heme d 1 (A), SAM-dependent methylation of uroporphyrinogen III (B) and amino acid sequence alignment of NirE, CysG and CobA from Pseudomonas aeruginosa (C). (A) The unique structural features of heme d 1 are the methyl-group ⁄ acetate-group combina- tions and the oxo-groups on rings A and B and the acrylate side chain on ring D. (B) SUMT proteins catalyze the SAM-dependent methyla- tion of uroporphyrinogen III, at positions C-2 and C-7, to precorrin-2. (C) Amino acid sequence alignment of P. aeruginosa NirE with P. aeruginosa CysG (SUMT domain = amino acid residues 219-461) and CobA shows that NirE exhibits 30% identity and 47% homology with the two other SUMTs. Identical residues are highlighted with black boxes. Uroporphyrinogen III methyltransferase NirE S. Storbeck et al. 5974 FEBS Journal 276 (2009) 5973–5982 ª 2009 The Authors Journal compilation ª 2009 FEBS l-homocysteine (SAH) [9,12]. Based on amino acid sequence analysis, two different types of SUMT have been found to exist. Members of the first type are usu- ally proteins of around 30 kDa and possess SUMT activity only (SirA, CobA). By contrast, members of the second type are usually proteins of larger size and possess, in addition to their SUMT activity, other cat- alytic activities such as siroheme synthase activity (CysG) or uroporphyrinogen III synthase activity (CobA+HemD) [14–16]. Some of the SUMTs show an overmethylation activity that catalyzes a third methyl transfer at position C-12, which results in the formation of trimethylpyrrocorphin [10,11,17–19]. Crystal structures are available for the monofunctional SUMT CobA and the multifunctional SUMT CysG [20,21]. So far, there are no reports about the enzyme which catalyzes the methylation of tetrapyrrole C-atoms C-2 and C-7 during heme d 1 biosynthesis. However, in the late 1990s a gene cluster was identified in P. aeruginosa (the so-called nir operon), of which several genes encode proteins potentially involved in heme d 1 bio- synthesis [22]. Based on amino acid sequence analysis, one of these genes, nirE, was proposed to encode a SUMT. NirE shares around 30% amino acid sequence identity and 47% homology with the two other SUMTs from P. aeruginosa (CysG, CobA; Fig. 1C) that are involved in siroheme and cobalamin bio- synthesis. Therefore, it was proposed that the NirE protein could be the SUMT required for heme d 1 formation [22]. However, so far, this has not been demonstrated experimentally. Here we report the pro- duction, purification and characterization of recombi- nant NirE from P. aeruginosa. We show that NirE is indeed a SUMT which catalyzes the methylation of uroporphyrinogen III to precorrin-2 in vivo and in vitro. Results and Discussion Production of recombinant P. aeruginosa NirE The recombinant P. aeruginosa NirE protein was pro- duced either as a fusion protein carrying a C-terminal His-tag (NirE-His) or as a fusion protein carrying both N-terminal thioredoxin (Trx)- and S-tags and a C-ter- minal His-tag (Trx-S-NirE-His). In both cases the recombinant protein was purified to apparent homoge- neity in a single chromatographic step on Ni Sepha- rose TM 6 Fast Flow (Fig. 2A). Initial experiments were performed using NirE-His; however, this protein showed a tendency to precipitate at concentrations above 3 mgÆmL )1 . By contrast, the Trx-S-NirE-His protein was soluble at high enzyme concentrations. A B C Fig. 2. Production and purification of recombinant NirE from Pseu- domonas aeruginosa and characterization of in vivo accumulated tetrapyrroles. (A) SDS ⁄ PAGE analysis of the production and purifi- cation of recombinant NirE. Lane 1, proteins within a cell-free extract prepared from Escherichia coli BL21(DE3) carrying pET32a- nirE-Trx after induction with isopropyl thio-b- D-galactoside (IPTG); lane 2, recombinant Trx-S-NirE-His after chromatography on Ni Sepharose TM Fast Flow; lane 3, S-NirE-His after thrombin cleavage and gel filtration chromatography; lane 4, purified recombinant NirE- His; lanes M, marker proteins with M r values indicated. (B) Tetra- pyrroles accumulated during NirE production in E. coli were extracted from the soluble protein fraction using C 18 reversed- phase material and were characterized using UV-visible absorption spectroscopy and mass spectrometry. The UV-visible absorption spectrum of the extracted tetrapyrroles is characteristic of tri- methylpyrrocorphin [10,11,17–19]. (C) HR-ESI-MS results; the experimental mass and isotopic pattern of the [M-H] ) ion of compound 872 are shown. HR-ESI-MS revealed an exact mass of 871.2687 for the [M-H] ) ion, which corresponds to the chemical formula C 43 H 44 N 4 O 16 , in accordance with the mass and chemical formula of trimethylpyrrocorphin in its trilactone form. S. Storbeck et al. Uroporphyrinogen III methyltransferase NirE FEBS Journal 276 (2009) 5973–5982 ª 2009 The Authors Journal compilation ª 2009 FEBS 5975 After thrombin cleavage and removal of the Trx-tag by gel filtration, the remaining S-NirE-His could be concentrated up to 20 mgÆmL )1 . Both NirE constructs (NirE-His and S-NirE-His) exhibited a slight reddish colour after concentration of the protein. UV-visible absorption spectroscopy indicated the presence of trimethylpyrrocorphin (data not shown), the NirE reaction product produced during protein production in Escherichia coli (see below). NirE thus seems to bind this overmethylation reaction product rather tightly because it remained bound to the protein, at least partially, during protein purification. This tight binding of trimethylpyrrocorphin seems to be a feature unique to NirE because it has not been reported for other SUMT proteins. However, a physiological role of this binding phenomenon can be excluded because tri- methylpyrrocorphin does not represent a physiological intermediate during heme d 1 biosynthesis. The native molecular mass of NirE was determined by gel filtration chromatography of NirE-His. A native relative molecular mass of 60 000 ± 3000 was deduced from this experiment, suggesting a dimeric structure for the NirE protein (with a subunit molecular mass of 30 kDa). Other SUMT proteins involved in siroheme (CysG) and cobalamin (CobA) biosynthesis were also reported to be dimeric proteins [20,21]. NirE carries SUMT activity in vivo During the production of both NirE-His and Trx-S- NirE-His in E. coli a red compound accumulated in the cells. This compound was extracted from the solu- ble protein fraction using C 18 -reversed phase material and analysed by UV-visible absorption spectroscopy and mass spectrometry. In both cases (NirE-His and Trx-S-NirE-His production) the UV-visible absorption spectrum of the extracted compound exhibited an absorption maximum at 354 nm and was very similar to the previously reported absorption spectrum of trimethylpyrrocorphin, the overmethylation product of SUMT proteins (Fig. 2B) [10,11,17–19]. Analysis of the extracted compound by high-resolution electro- spray mass spectrometry (HR-ESI-MS) in the negative ion mode revealed an exact mass of 871.2687 for the [M-H] ) ion, which corresponds to an elemental com- position of C 43 H 44 N 4 O 16 with a deviation of 0.8 p.p.m. This elemental composition is in accordance with the mass and chemical formula of trimethylpyrrocorphin in its trilactone form (Fig. 2C). Isolation of lactone derivatives of isobacteriochlorins has been reported previously [23,24] and lactone formation probably occurs during the tetrapyrrole extraction procedure. However, our results showed that the production of recombinant NirE in E. coli leads to the accumulation of trimethylpyrrocorphin in vivo, an observation that has been reported for most of the known SUMT proteins [10,11,17–19]. NirE carries SUMT activity in vitro Next, we tested the NirE protein for SUMT activity in vitro. The standard NirE activity assay was per- formed as described in the Materials and methods sec- tion with enzymatically produced uroporphyrinogen III. The formation of the NirE reaction product, pre- corrin-2, was followed using UV-visible absorption spectroscopy. Figure 3 shows the absorption spectra obtained from enzyme reactions after overnight incu- bation. The spectrum obtained from a reaction mix- ture containing the uroporphyrinogen III producing enzymes HemB, HemC and HemD did not show any characteristic features in the 300–700 nm region, as was expected for a solution containing colourless uro- porphyrinogen III under anaerobic conditions. Upon the addition of recombinant purified NirE and SAM to the reaction mixture, an absorption spectrum was observed that exhibited very broad absorption features (between 500–400 nm and 400–350 nm), in agreement with the yellow colour of the corresponding reaction mixture. This spectrum is characteristic for precorrin-2 [17,25]. When the precorrin-2 dehydrogenase SirC and NAD + were also included in the above reaction Fig. 3. UV-visible absorption spectra of NirE activity assays. The substrate uroporphyrinogen III (urogen III in the figure) was pro- duced from 5-aminolevulinic acid by an enzyme cocktail containing purified, recombinant Pseudomonas aeruginosa HemB and Bacil- lus megaterium HemC and HemD (dashed line). Precorrin-2 forma- tion was observed (dotted line) upon the addition of purified NirE and SAM. Precorrin-2 formed by NirE was converted into sirohydro- chlorin by SirC in the presence of NAD + (solid line). Uroporphyrinogen III methyltransferase NirE S. Storbeck et al. 5976 FEBS Journal 276 (2009) 5973–5982 ª 2009 The Authors Journal compilation ª 2009 FEBS mixture, the solution turned red instead of yellow, indicating the conversion of precorrin-2 (produced by NirE) into sirohydrochlorin. The UV-visible absorp- tion spectrum obtained from such a reaction mixture corresponds indeed to the typical absorption spectrum of sirohydrochlorin (Fig. 3) [25]. When SAM was omitted from the NirE activity assay no precorrin-2 formation was observed (data not shown). These results clearly show that NirE is able to catalyze the two SAM-dependent methylation reactions to convert uroporphyrinogen III into precorrin-2 in vitro. Initially we performed enzyme assays with all three NirE proteins – NirE-His, Trx-S-NirE-His and S-NirE- His – and compared their catalytic activities. We observed that Trx-S-NirE-His and S-NirE-His showed similar activities. By contrast, NirE-His showed only half of the activity of the other two proteins. Therefore, S-NirE-His was used for all subsequent enzyme assays and experiments. We observed the highest catalytic rates with chemically produced substrate uroporphyrinogen III at a concentration of 17 lm, a SAM concentration of 200 lm and at NirE concentrations of 1.5 lm. Under these assay conditions a specific activity of 316.8 nmol of precorrin-2 h )1 Æmg )1 of NirE was observed. Previously, it was reported that SUMT proteins catalyze a third methylation of uroporphyrinogen III to generate a trimethylpyrrocorphin, not only in vivo but also in vitro at high enzyme concentrations in the assay [17,18]. As the production of recombinant NirE in E. coli leads to the accumulation of trimethylpyrro- corphin in vivo, we tested if high concentrations of NirE in our enzyme assay also formed this compound in vitro. We observed the formation of trimethylpyrro- corphin in our activity assays at NirE concentrations above 10 lm in the presence of 500 lm SAM (data not shown). Thus, NirE is indeed a SUMT that shows the same catalytic behaviour in vivo and in vitro as the SUMTs for siroheme and cobalamin biosynthesis. NirE exhibits substrate inhibition by uroporphyrinogen III and product inhibition by SAH SUMT proteins were reported to exhibit inhibition by their substrate uroporphyrinogen III, as well as by the reaction by-product SAH. Therefore, we tested NirE for such inhibition phenomena. In our enzyme assay using chemically produced uroporphyrinogen III we observed substrate inhibition at uroporphyrinogen III concentrations above 17 lm, as shown in Figure 4A. In order to test NirE for inhibition by SAH we added increasing amounts of SAH to our activity assay. We observed inhibition of the NirE reaction at SAH concentrations above 2 lm (Fig. 4B). NirE thus displays the same inhibition phenomena as those previously reported for other SUMT proteins [9,12]. However, the question of whether these substrate and product-inhibition characteristics are physiologically relevant, or if they only represent in vitro assay artefacts, requires further investigation. SAM binding In previous studies, rapid SAM-binding assays were performed in order to characterize SUMT proteins [10,26,27]. Therefore, we also tested the NirE protein for its ability to bind SAM. After incubation of NirE with radioactively labelled SAM, the mixture was A B Precorrin-2 (pmol)·min –1 × NirE (nmol) Precorrin-2 (pmol)·min –1 × NirE (nmol) Fig. 4. Inhibition of NirE activity by the substrate uroporphyrinogen III and by the product SAH. (A) NirE activity assays (1.5 l M NirE, 200 l M SAM) were performed with increasing amounts of chemi- cally synthesized uroporphyrinogen III. Initial rates of precorrin-2 formation were plotted against the uroporphyrinogen III concentra- tion. (B) Increasing amounts of S-adenosyl- L-homocysteine (SAH) were added to the NirE activity assay (1.5 l M NirE, 200 lM SAM, 17 l M uroporphyrinogen III). Initial rates of precorrin-2 formation were plotted against the SAH concentration. S. Storbeck et al. Uroporphyrinogen III methyltransferase NirE FEBS Journal 276 (2009) 5973–5982 ª 2009 The Authors Journal compilation ª 2009 FEBS 5977 passed over a desalting column and the elution frac- tions were analyzed for radioactivity using a liquid scintillation counter. As a control, the same experiment was carried out with BSA. In the BSA control experi- ment all radioactivity eluted in the small-molecules fractions. By contrast, when SAM was mixed with NirE, the label was found to co-elute with the protein- containing fractions (data not shown). We also tested whether SAM remained bound to NirE during dena- turing electrophoresis, as observed for other SUMT proteins [10,26,27]. After incubation of NirE with radioactively labelled SAM, the protein was subjected to SDS ⁄ PAGE and fluorography. No radioactivity was found to be associated with the protein after denatur- ing electrophoresis (data not shown). These experi- ments show that NirE binds SAM; however, the binding seems to be weaker than for other SUMT proteins because SAM did not remain bound to the protein under denaturing conditions. P. aeruginosa cobA is able to complement a P. aeruginosa nirE mutant We have unambiguously shown, from the results described in the previous section, that the NirE protein is a SAM-dependent uroporphyrinogen III methyl- transferase. Although P. aeruginosa also possesses the genes encoding the SUMTs involved in cobalamin (cobA) and siroheme (cysG) biosynthesis, the nirE gene product was found to be essential for heme d 1 biosyn- thesis. A P. aeruginosa nirE knockout mutant was unable to synthesize heme d 1 and produced only heme d 1 -lacking, inactive cytochrome cd 1 [22]. This absolute requirement for NirE during heme d 1 biosynthesis is surprising considering the identical catalytic abilities of NirE and CobA. CobA catalyzes the SAM-dependent methylation of uroporphyrinogen III to form precor- rin-2 during cobalamin biosynthesis. The third SUMT in P. aeruginosa, the trifunctional siroheme synthase CysG, probably does not release precorrin-2 during siroheme formation and therefore cannot provide this precursor for heme d 1 biosynthesis in the absence of NirE. The observation that CobA is apparently not able to replace NirE during heme d 1 formation may have several explanations. One possibility could be that specific protein–protein interactions between NirE and the subsequent heme d 1 biosynthesis protein are required to allow substrate channelling of the highly labile precorrin-2. Another explanation may be that CobA, although probably present under anaerobic denitrifying conditions in order to sustain cobalamin biosynthesis for cobalamin-dependent enzymes, such as class II ribonucleotide reductase [28], is produced in amounts too low to sustain efficient heme d 1 biosynthesis. In order to investigate these possibilities, we tested whether P. aeruginosa cobA, when constitutively expressed from a plasmid, was able to complement a P. aeruginosa PAO1 nirE transposon mutant (strain PAO1 ID35553). For these experiments, wild-type P. aeruginosa PAO1 and P. aeruginosa PAO1 ID35553 carrying diverse complementation plasmids were grown as described in the Materials and methods. When anaerobic growth conditions were reached (after about 4 h), strain PAO1 ID35553 as well as this strain carry- ing the basic plasmid pUCP20T showed greatly impaired growth when compared with the wild-type strain (Fig. 5). By contrast, strain PAO1 ID35553, car- rying the complementation plasmid pUCP20T-nirE, grew almost as well as the wild-type strain, as expected. Interestingly, strain PAO1 ID35553, carrying plasmid pUCP20T-cobA, showed growth behaviour similar to that of the wild-type strain and the same growth behaviour as the nirE-complemented strain PAO1 ID35553 (Fig. 5). Therefore, cobA was able to complement the P. aeruginosa nirE mutant strain when constitutively expressed from a complementation plas- mid. By contrast, the concentration of native CobA produced under anaerobic denitrifying growth condi- tions was apparently not sufficient to restore efficient heme d 1 biosynthesis in the nirE ) background. Indeed, cobA transcript levels were found to be absent in Affymetrix microarray analyses of anaerobically grown Fig. 5. Growth curves of wild-type Pseudomonas aeruginosa PAO1 and of P. aeruginosa strain PAO1 ID35553. P. aeruginosa was grown under anaerobic growth conditions in the presence of nitrate, as described in the Materials and methods. Strain PAO1 ID35553 (D) and this strain carrying plasmid pUCP20T (.) showed impaired growth under these growth conditions compared with the wild-type strain ( ). Bacterial growth of strain PAO1 ID35553 was restored by plasmids pUCP20T-nirE ( • ) and pUCP20T-cobA (s). Uroporphyrinogen III methyltransferase NirE S. Storbeck et al. 5978 FEBS Journal 276 (2009) 5973–5982 ª 2009 The Authors Journal compilation ª 2009 FEBS P. aeruginosa (M. Schobert, personal communication). These results are in agreement with the fact that the genes for nitrite reductase NirS and the proposed heme d 1 biosynthesis proteins, including NirE, are organized in one large operon [22]. The transcription of the nir operon genes was found to be highly up-regulated in P. aeruginosa under anaerobic conditions in the presence of nitrate [28]. Under such conditions the co-transcription of heme d 1 biosynthesis genes and the co-production of both NirE and the other heme d 1 biosynthesis proteins, in high amounts ensures the efficient and highly concerted action of the proteins. In order to cope with the high demand for heme d 1 under denitrifying conditions, such a concerted action of heme d 1 biosynthesis proteins is required and therefore native cobA, which is not co-transcribed with the nir genes, is not able to replace nirE. Thus, the NirE protein is a SAM-dependent uroporphyrinogen III methyltransferase which is specifically required for heme d 1 biosynthesis. Materials and methods Chemicals Unless stated otherwise, all chemicals, reagents and antibi- otics were obtained from Sigma-Aldrich (Taufkirchen, Germany) or Merck (Darmstadt, Germany). DNA polymerase, restriction endonucleases and PCR requisites were purchased from New England Biolabs (Frankfurt a.M., Germany). Oligonucleotide primers were purchased from metabion international AG (Martinsried, Germany). Kits for PCR purification and gel extraction were pur- chased from Qiagen GmbH (Hilden, Germany). Ni Sepha- rose TM 6 Fast Flow was purchased from GE Healthcare (Mu ¨ nchen, Germany). [Methyl- 14 C]-S-adenosyl-l-methio- nine was obtained from Hartmann Analytic (Braunschweig, Germany). Uroporphyrin III was obtained from Frontier Scientific Europe (Carnforth, UK). Plasmids, bacteria and growth conditions The E. coli strain DH10B was used as the host for cloning, and E. coli BL21(DE3) was used as the host for protein production. For complementation studies a P. aeruginosa PAO1 mutant was used, which carries a transposon inser- tion in the nirE gene (strain PAO1 ID35553) [29]. This mutant was transformed with plasmids pUCP20T-nirE, pUCP20T-cobA or pUCP20T. For anaerobic growth condi- tions [30], LB (Luria–Bertani) medium was supplemented with 50 mm nitrate and carbenicillin at a final concentra- tion of 250 lgÆmL )1 . P. aeruginosa precultures were grown aerobically overnight and the anaerobic cultures (140-mL bottles filled with 135 mL of LB medium) were inoculated with appropriate volumes of these precultures to obtain a final A of 0.05 at 578 nm. Culture was carried out at 37 °C. The plasmids used for the production of recombinant P. aeruginosa NirE were pET32a-nirE-Trx and pET22b- nirE (see below). The plasmids used for the production of recombinant P. aeruginosa HemB [31] and B. megaterium HemC, HemD and SirC [32] were described previously. For the production of recombinant proteins, E. coli BL21(DE3) cells, carrying the respective plasmid, were grown at 37 °C in 500 mL of LB medium supplemented with ampicillin at a final concentration of 100 lgÆ mL )1 . At a A of 0.6 at 578 nm, protein production was induced by the addition of 50 lm isopropyl thio-b-d-galactoside (IPTG). The cells were then cultured further at 17 °C. After 18 h of culture the cells were harvested by centrifugation and stored at )20 °C until required. Construction of vectors For the construction of nirE expression vectors, the nirE gene from P. aeruginosa PAO1 was PCR amplified using the primers nirE_Pa_BamHI_for (GCCG GGATCCAT GAACACTACCGTGATTC) and nirE_Pa_XhoI_rev (GA CTCGAGGGCGCATGCGAC) containing BamHI and XhoI restriction sites (underlined), respectively, for cloning the nirE gene into pET32a (Novagen, Darmstadt, Germany). For cloning the nirE gene into pET22b (Nov- agen), the PCR primers NirE_NdeI (GT CATATGACA CTACCGTGATTCC) and NirE_HindIII (GT AAGCTT GCATGCGACGGCCTCG), containing NdeI and HindIII restriction sites (underlined), respectively, were used. The plasmid pHAE2 [22], containing a fragment of the P. aeru- ginosa PAO1 nir operon, was used as the DNA template. The resulting PCR fragments were digested with BamHI and XhoI, or with NdeI and HindIII, and ligated into the appropriately digested vectors pET32a or pET22b to gener- ate pET32a-nirE-Trx and pET22b-nirE, respectively. For construction of complementation vectors the nirE gene and the cobA gene, including a 50-bp upstream region bearing the ribosome-binding sites, were PCR amplified using the primers NirE_Compl_for (GA GAATTCGGA AATCGGCCTCG) and NirE_Compl_rev (CT AAGCTTT CAGGCGCATGCG) for the nirE gene and CobA_ Compl_for (GA GAATTCACTGCTGGCGGCC) and CobA_Compl_rev (CT AAGCTTTCAGGCGCTCAGGG) for the cobA gene, respectively, containing EcoRI and HindIII restriction sites (underlined). A colony of P. aeruginosa PAO1 was used as a template. The resulting PCR fragments were digested with EcoRI and HindIII and ligated into the appropriately digested vector pUCP20T to generate pUCP20T-nirE and pUCP20T-cobA, respectively. The vectors were transferred into strain PAO1 ID35553 by diparental mating using E. coli ST18 [33] as a donor. Stan- dard procedures were used for PCR amplification, agarose S. Storbeck et al. Uroporphyrinogen III methyltransferase NirE FEBS Journal 276 (2009) 5973–5982 ª 2009 The Authors Journal compilation ª 2009 FEBS 5979 gel electrophoresis, dephosphorylation, ligation and trans- formation of chemocompetent E. coli cells [34]. Restriction enzymes were used as recommended by the manufacturer. Purification of enzymes All protein purification steps were carried out at 4 °C. Harvested E. coli cells, harbouring recombinantly pro- duced NirE protein, were resuspended in buffer A [50 mm Tris ⁄ HCl (pH 7.5), 200 mm KCl, 10% (w ⁄ v) glycerol] containing 1 mm phenylmethanesulfonyl fluoride. The cells were disrupted by a single passage through a French press at 1000 p.s.i. (68947.57 hPa) and then centrifuged for 60 min at 175 000 g. The supernatant was applied to 1 mL of silica gel 100 C 18 -reversed phase material, acti- vated first with methanol then equilibrated with buffer A, to extract accumulated tetrapyrroles. The flow-through, containing the NirE protein, was applied to 1.5 mL of Ni Sepharose TM 6 Fast Flow equilibrated with buffer A. After extensive washing with buffer A, the recombinant NirE protein was eluted with 2.5 mL of buffer A contain- ing 200 mm imidazole. After elution of NirE, buffer exchange was performed in an anaerobic chamber (Coy Laboratories, Grass Lake, MI, USA) by passing the pro- tein solution through a NAP-25 column (GE Healthcare) that had been equilibrated with buffer A containing 5 mm dithiothreitol. When NirE was produced as a fusion pro- tein with an N-terminal Trx-tag, the Trx-tag was cleaved off with thrombin using the Thrombin Cleavage Capture Kit (Novagen) according to the manufacturer’s instruc- tions. NirE was separated from the Trx-tag by gel filtra- tion chromatography on a HiLoad 16 ⁄ 60 Superdex 200 column (GE Healthcare) equilibrated with buffer A con- taining 5 mm dithiothreitol. Protein solutions were con- centrated by ultrafiltration (Amicon, Millipore GmbH, Eschborn, Germany). The purified NirE protein was stored at )20 °C. The N-terminal amino acid sequences of the purified proteins were determined by Edman degrada- tion and were found to be identical to those expected from the cloning strategy (MNTTVIP for NirE-His and GSGMKET for S-NirE-His). Recombinant P. aeruginosa HemB and B. megaterium HemC, HemD and SirC were purified as previously described [31,32]. Determination of protein concentration The Bradford Reagent (Sigma-Aldrich) was used to deter- mine protein concentrations, according to the manufac- turer’s instructions, using BSA as a standard. Molecular mass determination The native molecular mass was estimated from gel filtration chromatography using a Superdex TM 200 10 ⁄ 300 GL column attached to an A ¨ KTA TM Purifier system (GE Healthcare). The column was equilibrated with buffer A containing 5 mm dithiothreitol. Protein samples of 150 l L were loaded onto the column and chromatographed at a flow rate of 0.5 mLÆmin )1 . Protein elution was monitored by determining the absorption of the eluate at 280 nm. The column was calibrated using the protein standards carbonic anhydrase (29 000 Da), BSA (66 000 Da), conalbumin (77 000 Da), alcohol dehydrogenase (150 000 Da) and b-amylase (200 000 Da). Extraction of tetrapyrrole compounds In vivo accumulated tetrapyrrole compounds were extracted from the soluble protein fraction by passing the cell-free extracts over a 1-mL silica gel 100 C 18 -reversed phase col- umn, activated first with methanol then equilibrated with buffer A. The silica gel was washed with water and the bound tetrapyrroles were eluted with methanol. The solvent was removed by evaporation and the dried tetrapyrroles were stored at )20 °C. UV-visible absorption spectroscopy UV-visible absorption spectra of extracted tetrapyrroles were recorded using a V-550 spectrophotometer (Jasco, Gross-Umstadt, Germany). NirE activity assay In vitro NirE activity assays were performed under anaero- bic conditions in an anaerobic chamber (Coy Laboratories) in a final volume of 1 mL of thoroughly degassed buffer containing 50 mm Tris ⁄ HCl (pH 8.0), 100 mm KCl, 5 mm MgCl 2 and 50 mm NaCl. The final NirE concentration was 1.5 lm. The substrate uroporphyrinogen III was generated from 1 mm 5-aminolevulinic acid using a coupled assay system including HemB (0.14 lm), HemC (0.15 lm) and HemD (0.17 lm). Alternatively, uroporphyrin III was reduced chemically and used at a final concentration of 8 lm. The reaction was started by the addition of SAM to a final concentration of 200 lm and was incubated at 37 °C in the dark. The reaction was monitored using a Lambda 2 spectrophotometer (PerkinElmer Instruments, U ¨ berlingen, Germany). In order to quantify the precorrin-2, it was converted to sirohydrochlorin (e 376 = 2.4 · 10 5 m )1 Æcm )1 [35]) by the addition of SirC (1.5 lm) and 100 lm NAD + . Preparation of uroporphyrinogen III Uroporphyrinogen III was prepared by chemical reduction of uroporphyrin III with 3% sodium amalgam, as described previously for coproporphyrinogen III [36]. Uroporphyrinogen III methyltransferase NirE S. Storbeck et al. 5980 FEBS Journal 276 (2009) 5973–5982 ª 2009 The Authors Journal compilation ª 2009 FEBS SAM-binding assay The SAM-binding assay was performed as described previ- ously [26]. Briefly, 100 lm purified NirE protein was incu- bated with 0.5 l Ci of [methyl- 14 C]-S-adenosyl-l-methionine in a final volume of 250 lL of buffer A at 25 °C for 1 h. The protein solution was then passed over a NAP-5 column (GE Healthcare) and eluted with 3 mL of buffer A. Fractions of 100 lL were collected and analysed for radio- activity using a Liquid Scintillation Analyzer Tri-Carb 2900 TR (PerkinElmer Life Sciences). Analysis of SAM binding by fluorography was performed as described previously [26]. Mass spectrometry of tetrapyrroles HR-ESI-MS data were acquired using a Bruker microTOF- Q II equipped with an Apollo ESI ion source (Bruker Daltonik, Bremen, Germany). Samples were dissolved in methanol and introduced, via direct infusion, at a flow rate of 4 lLÆmin )1 . Acknowledgements We thank Professor H. Arai (University of Tokyo, Japan) for the gift of plasmid pHAE2, and Professor S. Ha ¨ ußler (Helmholtz-Centre for Infection Research, Braunschweig, Germany) for the gift of the P. aerugin- osa nirE transposon mutant (strain PAO1 ID35553). We thank Dr Jan Willmann (Bruker Daltonik, Bre- men, Germany) for HR-ESI-MS measurements. We also thank Prof. D. Jahn and Drs J. Moser and M. Schobert for helpful discussions. This work was sup- ported by the Emmy-Noether-Program of the Deut- sche Forschungsgemeinschaft and by funds from the Fonds der Chemischen Industrie to G.L. References 1 Zumft WG (1997) Cell biology and molecular basis of denitrification. Microbiol Mol Biol Rev 61, 533–616. 2 Tavares P, Pereira AS, Moura JJ & Moura I (2006) Metalloenzymes of the denitrification pathway. J Inorg Biochem 100, 2087–2100. 3 Philippot L (2002) Denitrifying genes in bacterial and archaeal genomes. Biochim Biophys Acta 1577, 355–376. 4 Cutruzzola F, Rinaldo S, Centola F & Brunori M (2003) NO production by Pseudomonas aeruginosa cd 1 nitrite reductase. IUBMB Life 55, 617–621. 5 Chang CK (1994) Haem d 1 and other haem cofactors from bacteria. Ciba Found Symp 180, 228–238. 6 Battersby AR (2000) Tetrapyrroles: the pigments of life. Nat Prod Rep 17, 507–526. 7 Yap-Bondoc F, Bondoc LL, Timkovich R, Baker DC & Hebbler A (1990) C-methylation occurs during the biosynthesis of heme d 1 . J Biol Chem 265, 13498–13500. 8 Warren MJ, Roessner CA, Santander PJ & Scott AI (1990) The Escherichia coli cysG gene encodes S-adeno- sylmethionine-dependent uroporphyrinogen III methyl- ase. Biochem J 265, 725–729. 9 Robin C, Blanche F, Cauchois L, Cameron B, Couder M & Crouzet J (1991) Primary structure, expression in Escherichia coli, and properties of S-adenosyl-L-methio- nine:uroporphyrinogen III methyltransferase from Bacillus megaterium. J Bacteriol 173, 4893–4896. 10 Leustek T, Smith M, Murillo M, Singh DP, Smith AG, Woodcock SC, Awan SJ & Warren MJ (1997) Siroheme biosynthesis in higher plants. Analysis of an S-adenosyl- L-methionine-dependent uroporphyrinogen III methyl- transferase from Arabidopsis thaliana. J Biol Chem 272, 2744–2752. 11 Fan J, Wang D, Liang Z, Guo M, Teng M & Niu L (2006) Maize uroporphyrinogen III methyltransferase: overexpression of the functional gene fragments in Escherichia coli and one-step purification. Protein Expr Purif 46, 40–46. 12 Blanche F, Debussche L, Thibaut D, Crouzet J & Cameron B (1989) Purification and characterization of S-adenosyl-L-methionine:uroporphyrinogen III methyl- transferase from Pseudomonas denitrificans. J Bacteriol 171, 4222–4231. 13 Blanche F, Robin C, Couder M, Faucher D, Cauchois L, Cameron B & Crouzet J (1991) Purification, charac- terization, and molecular cloning of S-adenosyl-L- methionine:uroporphyrinogen III methyltransferase from Methanobacterium ivanovii. J Bacteriol 173, 4637– 4645. 14 Warren MJ, Bolt EL, Roessner CA, Scott AI, Spencer JB & Woodcock SC (1994) Gene dissection demon- strates that the Escherichia coli cysG gene encodes a multifunctional protein. Biochem J 302, 837–844. 15 Anderson PJ, Entsch B & McKay DB (2001) A gene, cobA+hemD, from Selenomonas ruminantium encodes a bifunctional enzyme involved in the synthesis of vitamin B 12 . Gene 281, 63–70. 16 Lobo SA, Brindley A, Warren MJ & Saraiva LM (2009) Functional characterization of the early steps of tetrapyrrole biosynthesis and modification in Desulfovib- rio vulgaris Hildenborough. Biochem J 420, 317–325. 17 Warren MJ, Stolowich NJ, Santander PJ, Roessner CA, Sowa BA & Scott AI (1990) Enzymatic synthesis of dihydrosirohydrochlorin (precorrin-2) and of a novel pyrrocorphin by uroporphyrinogen III methylase. FEBS Lett 261, 76–80. 18 Sattler I, Roessner CA, Stolowich NJ, Hardin SH, Har- ris-Haller LW, Yokubaitis NT, Murooka Y, Hashimoto Y & Scott AI (1995) Cloning, sequencing, and expres- S. Storbeck et al. Uroporphyrinogen III methyltransferase NirE FEBS Journal 276 (2009) 5973–5982 ª 2009 The Authors Journal compilation ª 2009 FEBS 5981 sion of the uroporphyrinogen III methyltransferase cobA gene of Propionibacterium freudenreichii (sherma- nii). J Bacteriol 177, 1564–1569. 19 Roessner CA & Scott AI (1995) Fluorescence-based method for selection of recombinant plasmids. BioTech- niques 19, 760–764. 20 Vevodova J, Graham RM, Raux E, Schubert HL, Roper DI, Brindley AA, Ian Scott A, Roessner CA, Stamford NP, Elizabeth Stroupe M et al. (2004) Struc- ture ⁄ function studies on a S-adenosyl-l-methionine- dependent uroporphyrinogen III C methyltransferase (SUMT), a key regulatory enzyme of tetrapyrrole biosynthesis. J Mol Biol 344, 419–433. 21 Stroupe ME, Leech HK, Daniels DS, Warren MJ & Getzoff ED (2003) CysG structure reveals tetrapyrrole- binding features and novel regulation of siroheme biosynthesis. Nat Struct Biol 10, 1064–1073. 22 Kawasaki S, Arai H, Kodama T & Igarashi Y (1997) Gene cluster for dissimilatory nitrite reductase (nir) from Pseudomonas aeruginosa: sequencing and identifi- cation of a locus for heme d 1 biosynthesis. J Bacteriol 179, 235–242. 23 Battersby AR, Jones K, McDonald E, Robinson JA & Morris HR (1977) The structures and chemistry of isobacteriochlorins from Desulfovibrio gigas. Tetrahe- dron Lett 25, 2213–2216. 24 Battersby AR, McDonald E, Morris HR, Thompson M & Williams DC (1977) Biosynthesis of vitamin B 12 : structural studies on the corriphyrins from Propionibac- terium shermanii and the link with sirohydrochlorin. Tetrahedron Letters 25, 2217–2220. 25 Schubert HL, Raux E, Brindley AA, Leech HK, Wilson KS, Hill CP & Warren MJ (2002) The structure of Saccharomyces cerevisiae Met8p, a bifunctional dehy- drogenase and ferrochelatase. EMBO J 21, 2068–2075. 26 Woodcock SC & Warren MJ (1996) Evidence for a covalent intermediate in the S-adenosyl-l-methionine- dependent transmethylation reaction catalysed by sirohaem synthase. Biochem J 313, 415–421. 27 Woodcock SC, Raux E, Levillayer F, Thermes C, Rambach A & Warren MJ (1998) Effect of mutations in the transmethylase and dehydrogenase ⁄ chelatase domains of sirohaem synthase (CysG) on sirohaem and cobalamin biosynthesis. Biochem J 330, 121–129. 28 Platt MD, Schurr MJ, Sauer K, Vazquez G, Kukavica- Ibrulj I, Potvin E, Levesque RC, Fedynak A, Brinkman FS, Schurr J et al. (2008) Proteomic, microarray, and signature-tagged mutagenesis analyses of anaerobic Pseudomonas aeruginosa at pH 6.5, likely representing chronic, late-stage cystic fibrosis airway conditions. J Bacteriol 190, 2739–2758. 29 Jacobs MA, Alwood A, Thaipisuttikul I, Spencer D, Haugen E, Ernst S, Will O, Kaul R, Raymond C, Levy R et al. (2003) Comprehensive transposon mutant library of Pseudomonas aeruginosa. Proc Natl Acad Sci USA100, 14339–14344. 30 Eschbach M, Schreiber K, Trunk K, Buer J, Jahn D & Schobert M (2004) Long-term anaerobic survival of the opportunistic pathogen Pseudomonas aeruginosa via pyruvate fermentation. J Bacteriol 186, 4596–4604. 31 Frankenberg N, Kittel T, Hungerer C, Ro ¨ mling U & Jahn D (1998) Cloning, mapping and functional charac- terization of the hemB gene of Pseudomonas aeruginosa, which encodes a magnesium-dependent 5-aminolevulinic acid dehydratase. Mol Gen Genet 257, 485–489. 32 Raux E, Leech HK, Beck R, Schubert HL, Santander PJ, Roessner CA, Scott AI, Martens JH, Jahn D, Thermes C et al. (2003) Identification and functional analysis of enzymes required for precorrin-2 dehydroge- nation and metal ion insertion in the biosynthesis of sirohaem and cobalamin in Bacillus megaterium. Biochem J 370, 505–516. 33 Thoma S & Schobert M (2009) An improved Escheri- chia coli donor strain for diparental mating. FEMS Microbiol Lett 294, 127–132. 34 Sambrock J, Fritsch EF & Maniatis T (1989) Molecular Cloning: A Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. 35 Leech HK, Raux E, McLean KJ, Munro AW, Robin- son NJ, Borrelly GP, Malten M, Jahn D, Rigby SE, Heathcote P et al. (2003) Characterization of the cobalt- ochelatase CbiXL: evidence for a 4Fe-4S center housed within an MXCXXC motif. J Biol Chem 278, 41900– 41907. 36 Layer G, Verfu ¨ rth K, Mahlitz E & Jahn D (2002) Oxygen-independent coproporphyrinogen-III oxidase HemN from Escherichia coli. J Biol Chem 277, 34136–34142. Uroporphyrinogen III methyltransferase NirE S. Storbeck et al. 5982 FEBS Journal 276 (2009) 5973–5982 ª 2009 The Authors Journal compilation ª 2009 FEBS . The Pseudomonas aeruginosa nirE gene encodes the S-adenosyl- L-methionine-dependent uroporphyrinogen III methyltransferase required for heme d 1 biosynthesis Sonja. Under such conditions the co-transcription of heme d 1 biosynthesis genes and the co-production of both NirE and the other heme d 1 biosynthesis proteins, in