Cold adaptation of the mononuclear molybdoenzyme periplasmic nitrate reductase from the Antarctic bacterium Shewanella gelidimarina

Biochemical and Biophysical Research Communications 414 (2011) 783–788

Contents lists available at SciVerse ScienceDirect

Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc

Cold adaptation of the mononuclear molybdoenzyme periplasmic nitrate reductase from the Antarctic bacterium Shewanella gelidimarina Philippa J.L. Simpson a, Rachel Codd a,b,⇑ a b

School of Chemistry, University of Sydney, New South Wales 2006, Australia School of Medical Sciences (Pharmacology) and Bosch Institute, University of New South Wales, New South Wales 2006, Australia

a r t i c l e

i n f o

Article history: Received 1 September 2011 Available online 7 October 2011 Keywords: Periplasmic nitrate reductase Shewanella gelidimarina Enzyme cold adaptation

a b s t r a c t The reduction of nitrate to nitrite is catalysed in bacteria by periplasmic nitrate reductase (Nap) which describes a system of variable protein subunits encoded by the nap operon. Nitrate reduction occurs in the NapA subunit, which contains a bis-molybdopterin guanine dinucleotide (Mo–MGD) cofactor and one [4Fe–4S] iron–sulfur cluster. The activity of periplasmic nitrate reductase (Nap) isolated as native protein from the cold-adapted (psychrophilic) Antarctic bacterium Shewanella gelidimarina (NapSgel) and middle-temperature adapted (mesophilic) Shewanella putrefaciens (NapSput) was examined at varied temperature. Irreversible deactivation of NapSgel and NapSput occurred at 54.5 and 65 °C, respectively. When NapSgel was preincubated at 21–70 °C for 30 min, the room-temperature nitrate reductase activity was maximal and invariant between 21 and 54 °C, which suggested that NapSgel was poised for optimal catalysis at modest temperatures and, unlike NapSput, did not benefit from thermally-induced refolding. At 20 °C, NapSgel reduced selenate at 16% of the rate of nitrate reduction. NapSput did not reduce selenate. Sequence alignment showed 46 amino acid residue substitutions in NapSgel that were conserved in NapA from mesophilic Shewanella, Rhodobacter and Escherichia species and could be associated with the NapSgel cold-adapted phenotype. Protein homology modeling of NapSgel using a mesophilic template with 66% amino acid identity showed the majority of substitutions occurred at the protein surface distal to the Mo–MGD cofactor. Two mesophilic M psychrophilic substitutions (Asn M His, Val M Trp) occurred in a region close to the surface of the NapA substrate funnel resulting in potential interdomain p–p and/or cation–p interactions. Three mesophilic M psychrophilic substitutions occurred within 4.5 Å of the Mo–MGD cofactor (Phe M Met, Ala M Ser, Ser M Thr) resulting in local regions that varied in hydrophobicity and hydrogen bonding networks. These results contribute to the understanding of thermal protein adaptation in a redox-active mononuclear molybdenum enzyme and have implications in optimizing the design of low-temperature environmental biosensors. Ó 2011 Elsevier Inc. All rights reserved.

1. Introduction Periplasmic nitrate reductase (Nap) is a member of the dimethylsulfoxide (DMSO) reductase family of mononuclear Mo enzymes, which catalyse the reduction of oxyanions (nitrate, selenate, arsenate) or marine osmoregulating agents (DMSO, trimethylamine Noxide (TMAO)) in two-electron redox reactions key to the metabolism of Bacteria and Archaea [1–4]. In Nap, the reduction of nitrate to nitrite takes place in the NapA catalytic subunit which contains a bis-molybdopterin guanine dinucleotide (Mo–MGD) cofactor and one [4Fe–4S] iron–sulfur cluster [5,6]. While the majority of mononuclear Mo enzymes characterised by X-ray crystallography have been isolated from organisms adapted to middle-range (mesophilic) temperatures, the first member of this enzyme superfamily, ⇑ Corresponding author at: School of Chemistry, University of Sydney, New South Wales 2006, Australia. E-mail address: [email protected] (R. Codd). 0006-291X/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2011.10.003

aldehyde ferredoxin oxidoreductase, was isolated from the hyperthermophilic Archaeon, Pyrococcus furiosus [7]. Our interest in better understanding the mechanisms of temperature-dependent biomolecule adaption [8] and in the applications of cold-adapted redox-active enzymes in biotechnology and environmental sensing [9–15] prompted us in recent work to isolate the mononuclear Mo enzyme Nap from a cold-adapted (psychrophilic) bacterium Shewanella gelidimarina that was isolated from Antarctic sea ice [16]. We found that S. gelidimarina expressed two native 90 kDa isoforms of the NapA subunit [17] and supported this unexpected experimental finding using bioinformatics analysis which showed that the genomes of most Shewanella species encode two distinct Nap loci [18]. In this work, we describe the temperature-dependent catalytic competency and substrate specificity of Nap from psychophilic S. gelidimarina (NapSgel) together with the homologue from mesophilic Shewanella putrefaciens (NapSput). In concert, in order to better understand the cold-adaptive phenotype of NapSgel established

784

P.J.L. Simpson, R. Codd / Biochemical and Biophysical Research Communications 414 (2011) 783–788

from experiment, we have used a top-down approach and constructed a protein homology model of NapSgel using NapA from the mesophilic organism Rhodobacter sphaeroides as a template [19]. Our systems biology approach has revealed six amino acid residues that may govern the cold-adaptive phenotype of NapA from S. gelidimarina. This work contributes to understanding the mechanism of the cold-adaptation of mononuclear Mo enzymes and has implications in the design of devices for in situ sensing of nitrate and alternative oxyanion species and contaminants in rivers and lakes at environmental temperatures. 2. Materials and methods 2.1. Isolation of Nap from Shewanella and activity assays Shewanella frigidimarina ACAM 591T, S. gelidimarina ACAM 456T, and S. putrefaciens ATCC 8071T were grown in liquid cultures (medium: 0.5% w/v bactopeptone (Oxoid, Bacto), 0.2% w/v yeast extract (Oxoid, Bacto), 3.5% w/v sea salts (Sigma), 5 mM KNO3, 0.5 lM Na2MoO4, 0.5 lM FeCl3) which were aerated on an orbital shaker at 100 rpm and maintained at 4, 17 or 20 °C in temperature-controlled rooms or at room temperature (20–22 °C). Temperaturedependent growth was established in 200-mL cultures. Cells from S. gelidimarina or S. putrefaciens from large-scale cultures (10–12 L) were subjected to a total-cell-lysis procedure and the Nap purified according to a published method [17]. The enzymes studied were the 90 kDa NapA monomer from S. gelidimarina and the 106 kDa NapAB heterodimer from S. putrefaciens; both from the Nap-a (NapEDABC) locus. NapSgel and NapSput were >90% pure as determined by the limits of electrophoresis and had specific nitrate reductase activities (benzyl viologen) of 15 and 3 U mg1, respectively. Nitrate reductase activity was detected using the Griess reaction [20] which was scaled to a final assay volume of 900 lL. The enzyme sample (100 lL) was added to a reaction solution (175 lL) containing KNO3, methyl viologen (MV) and dithiothreitol (DTT) in K2HPO4/KH2PO4 (pH 7.1) and the reaction initiated by the addition of a solution (25 lL) of Na2S2O4/NaHCO3. Final assay conditions were: 50 mM KNO3, 1.3 mM MV, 50 lM DTT, 0.04% w/v Na2S2O4 and 0.04% w/v NaHCO3 in 0.25 M K2HPO4/KH2PO4 (pH 7.1). The reaction was stopped after 10 min by vortexing in air and equal-volume aliquots (300 lL) of 1% w/v sulfanilic acid in 20% w/v HCl and 0.129% w/v N-(1-naphthyl)ethylenediamine (NED) were added with vortexing between each addition. The pink color was developed for 10 min and the absorbance of the solution at 540 nm measured in a 1-mL cuvette. Alternatively, the assay was scaled down further by one-third to a final volume of 300 lL and the absorbance of the solutions measured at 492 nm in a 96well plate using a plate reader (Biotrak II Plate Reader) with a 492 nm filter. A standard curve for KNO2 solutions (6–60 lM) was measured. Nitrate reductase activity was also measured using an anaerobic benzyl viologen depletion assay [21,22]. In selected reactions, the KNO3 in the reaction solution was replaced with selenate (Na2SeO4, 60 mM), tellurite (K2TeO3, 60 mM) or chlorate (KClO3, 6 mM). Solutions of NapSgel or NapSput in 10 mM K2HPO4/ KH2PO4 buffer (pH 7.0) were pre-incubated at room temperature (21 °C) or at 30, 40, 45, 50, 54, 60, 65, 70, 76, 80, 85 or 90 °C in a heated water bath for 30 min. All solutions were then equilibrated to room temperature and the nitrate reductase activity measured over 10 min using the Griess assay.

model of NapSgel was generated using the SWISS-MODEL workspace [23] with NapA from R. sphaeroides (PDB: 1OGY Chain A) as a template [19]. Graphics were rendered using Pov-Ray 3.0 for Windows.

3. Results and discussion 3.1. Temperature-dependent phenotype of Shewanella species and native Nap In accordance with the literature that first described both S. gelidimarina and S. frigidimarina [16], temperature-dependent growth experiments in our laboratory showed that S. gelidimarina was a true psychrophile that thrived at 4 °C but was not viable at 20 °C (Fig. 1A). S. gelidimarina grew more rapidly at the reported optimal growth temperature of 17 °C than at 4 °C. Growth of the psychrotolerant bacterium S. frigidimarina was equally viable at 20 and 4 °C (Fig. 1B). The growth of S. putrefaciens exhibited the mesophilic phenotype and thrived at 20 °C but was not viable at 4 °C (Fig. 1C). Enzymes native to psychrophilic S. gelidimarina and mesophilic S. putrefaciens would be expected to have evolved to function at the respective environmental temperatures.

2.2. Bioinformatics analysis and protein homology modeling Fig. 1. Optical density (k = 580 nm) of cultures of S. gelidimarina (A), S. frigidimarina (B) or S. putrefaciens (C) grown in marine broth at 4 °C (d, j, N), 17 °C ( ) or 20 °C (s, h, D). f

Sequence alignments were generated using ClustalW hosted by Pôle BioInformatique Lyonnaise. The sequence of NapSgel (GI:296142301) has been previously reported [17]. The homology

P.J.L. Simpson, R. Codd / Biochemical and Biophysical Research Communications 414 (2011) 783–788

The catalytic behavior of native Nap proteins isolated from psychrophilic S. gelidimarina (NapSgel) or mesophilic S. putrefaciens (NapSput) was measured as a function of temperature. The irreversible deactivation of NapSgel and NapSput towards nitrate reduction (Fig. 2A) occurred at 54.5 and 65 °C, respectively. This 10 degree decrease in thermostability is characteristic of cold-adapted enzymes, relative to the mesophilic counterpart [10,11,13]. Following pre-incubation at temperatures between 21 and 70 °C, the catalytic competence of NapSgel towards nitrate reduction measured at room temperature, was maximal and invariant between 21 and 54 °C, which suggested that the enzyme was poised for maximum catalytic capacity at modest temperatures and did not benefit from any thermally-induced refolding. This was in contrast to NapSput, which showed increased room-temperature catalytic capacity when the enzyme had been preincubated at temperatures from 21 to 60 °C, suggesting that NapSput underwent thermally-induced refolding to conformational states that had increased catalytic competence. The catalytic activity of NapSgel or NapSput towards nitrate or selenate reduction was examined at 20 °C using the benzyl viologen depletion assay. While there was no measurable reduction of selenate by NapSput, selenate was reduced by NapSgel at a rate of 16% of the nitrate reductase activity (Fig. 2B). The Vmax for the reduction of selenate by Nap from R. sphaeroides has been reported as 140-fold less than the Vmax value for the reduction of nitrate [24]. The activity of NapSgel towards the reduction of chlorate and tellurite occurred at 22% and 7% of the nitrate reductase activity. Compared to mesophilic homologues, cold-adapted enzymes have been shown to have broader substrate specificities which have been attributed to increased conformational flexibility and plasticity at the active site.

785

3.2. Sequence analysis of NapA from S. gelidimarina The characteristics of NapSgel at the experimental level (reduced temperature of irreversible thermal deactivation, broader substrate specificity with respect to selenate) were consistent with the general understanding of enzyme cold adaptation. With an increasing focus upon adopting a systems biology approach in biochemistry and genetics programs, in concert with our bottom-up experimental studies, we undertook a top-down approach using NapSgel sequence analysis and homology modeling to further explore NapA cold adaptation. Other studies have shown a difference in the frequency of amino acid residues in functionally equivalent enzymes isolated from organisms that permanently reside at coldor middle-temperature environments. Compared to warm-adapted homologues, cold-adapted enzymes have been shown to have a higher frequency of Gly residues, a lower frequency of Pro residues and a lower Arg/(Arg + Lys) ratio, which manifests as increased conformational flexibility and low-temperature catalytic competence [25,26]. Analysis of NapSgel and NapSput as whole proteins (Table S1) showed no significant difference between the frequency of Gly and Pro residues or between the Arg/(Arg + Lys) ratio. The sequence of NapSgel was aligned with NapA from psychrotrophic S. frigidimarina and NapA from mesophilic S. putrefaciens, Rhodobacter sphaeroides and Escherichia coli (Fig. 3 and S1). The latter two mesophilic constructs were selected based upon the availability of X-ray crystal structures of NapAB from R. sphaeroides and NapA from E. coli [19,27]. The overall identity across the 5 protein sequences was 51.87%. Sequence alignment showed 46 amino acid residues in NapSgel that were different from residues conserved across NapA proteins from the two species from the same genus, S. frigidimarina, S. putrefaciens; and NapA proteins from two mesophilic organisms, R. sphaeroides and E. coli. A contiguous two amino acid deletion in NapSgel between Pro585 and Ala588 (as referenced to the R. sphaeroides NapA coordinates) was also evident. The number of amino acid substitutions and deletions in NapSgel comprised approximately 5% of the protein. The absolute conservation across two moderate-temperature adapted Shewanella species (S. frigidimarina, S. putrefaciens) and two mesophilic species from different genera (Rhodobacter, Escherichia) suggested the substituted residues in NapSgel could be associated with the cold-adapted phenotype. In the 46 residues that were ascribed to cold adaptation (Table 1 and S2), there was a higher number of Gly residues (2 vs. 0), a lower number of Pro residues (1 vs. 3) and a decreased Arg/(Arg + Lys) ratio (0 vs. 0.25) in NapSgel compared to NapSput, which indicated that trends in the frequencies of amino acid residues that might govern a cold-adaptive phenotype are more appropriately considered at a local rather than global level. 3.3. Homology modeling of NapA from S. gelidimarina

Fig. 2. Normalised nitrate reductase activity of NapSgel (circles) and NapSput (triangles) as measured by the Griess assay at room temperature after a 30-min period of enzyme pre-incubation at temperature (A); and activity of NapSgel towards the reduction of nitrate (black), chlorate (dark grey), selenate (grey), tellurite (light grey) and no substrate (black, open) (B).

To explore the implications of these amino acid residue substitutions and deletions, a homology model of NapSgel using the SWISS-MODEL workspace [23] using R. sphaeroides NapA (PDB: 1OGY Chain A) as a template [19] was generated (Fig. 4A and B). The veracity of the homology model was supported by the 66.3% sequence identity between NapSgel and R. sphaeroides NapA and the QMEAN4 global score of 0.62 [28]. Of the 46 substituted residues in NapSgel, 34 were located at the protein surface and of these, 12 involved the exchange of very similar residues (Lys M Arg, Asp M Glu, Ala M Gly, Leu M Val, Ile M Leu). Ten substitutions occurred in the protein interior and two at the surface/interior interface. The two amino acid deletion occurred in a loop region in domain 2 (Table 1 and S2). Each of the mesophilic to psychrophilic substitutions Met77 M Gln77, Leu332 M Gln331 and Pro278 M Asp278 replaced hydrophobic residues with hydrophilic residues at the protein

786

P.J.L. Simpson, R. Codd / Biochemical and Biophysical Research Communications 414 (2011) 783–788

Fig. 3. Excerpt of the sequence alignment of NapA from S. gelidimarina (SGEL), S. frigidimarina (SFRI), S. putrefaciens (SPUT), R. sphaeroides (RSPH) and E. coli (ECOL) with elements of secondary structure (SSST) and protein domains (DOMA) indicated as determined from X-ray crystal structures [19,27]. Residues within 4 Å of the NapA prosthetic groups (Mo–MGD; [4Fe–4S] iron–sulfur cluster) are marked with (⁄; RSPH and ECOL) or (.; ECOL); and residues within 4 Å of the NapA–NapB interface are marked with (I) or (_) (ASIF). Residues conserved in SFRI, SPUT, RSPH and ECOL and which are different from SGEL are marked with X (ACON). The lower numbering scheme is referenced to the X-ray crystal structure of NapA from RSPH and the NapSgel homology model. The upper numbering scheme is referenced to a study of NAP bioinformatics [18]. The complete alignment is given in Fig. S1.

Table 1 Amino acid residues in NapA from S. gelidimarina that differ from residues conserved across NapA proteins from S. frigidimarina, S. putrefaciens, R. sphaeroides and E coli.a Sgel

meso

Sb

Sgel

meso

Sb

Sgel

meso

Sb

Sgel

meso

Sb

Q77 F82 M127 M135 Y230 A241 H264 D278 A287

M Y W L I L N P K

L L H H S H S L L

S296 Q331 F332 M347  A407 C480 S499  V517 I544

P L Y F T V A F T

L H H H L S L S L

N563 M564 Y577 584c 585c Y594 Y618 G650 W666

T L F Zd Zd F F K V

L L L L L L L L S

A674 P676 T700  V712 Y718 L736à N739 I770 I776

P A S F F K/R S Q T

L L H L L S S L L

a

Substitutions between amino acid residues at the protein surface with very strong similarities have been omitted from this Table (complete data provided in Table S2). S = secondary structure: L, loop; S, sheet; H, helix. c Deleted residues. d Z = variable amino acid residue. Within 4.5 Å of active site. Within 4 Å of NapA–NapB interface. b

  à

surface, which is a hallmark of protein cold-adaptation. Other mesophilic to psychrophilic substitutions at the protein cell surface included Lys287 M Ala287 and Lys653 M Gly650, which occurred in loop regions and would increase local protein conformational flexibility and decrease the number of salt bridges. These are factors associated with cold-adaptation in other proteins that have been studied [10,11,13,14]. A strong hydrogen bonding interaction between Tyr230 and Thr318 (2.67 Å) in NapSgel would not be possible in the mesophilic analogue between the equivalent residues Ile230 and Ser319. The mesophilic to psychrophilic substitutions of Asn264 M His264 and Val669 M Trp666 in NapSgel resulted in the close positioning of the His264 imidazole ring in domain 3 to the Trp666 benzene ring in domain 4 with a centroid distance of 5.07 Å (marked with an arrow in Fig. 4A). This proximity could manifest as a p–p interaction and restricted flexibility at the opening of the substrate funnel. Since the pKa values of His residues in myoglobin have been shown to increase with decreased temperature (0.02 pKa unit per degree) [29], a His264–Trp666 cation–p interaction might also be possible at Antarctic temperatures. Three residues at the protein interior were identified that were within 4.5 Å of the NapA Mo–MGD cofactor that were conserved across S. putrefaciens, E. coli and R. sphaeroides, and were different

in S. gelidimarina (Fig. 4C and D). As referenced to the R. sphaeroides NapA coordinates [19], Phe348, Ala500 and Ser703 were conserved across the three mesophilic species and were substituted in NapSgel as Met347, Ser499 and Thr700, respectively. In psychrotrophic S. frigidimarina, the residues were conserved in accordance to the mesophilic system. As revealed from the homology model, the mesophilic to psychrophilic Ala500 M Ser499 substitution manifested as a distinct hydrogen bond pattern with a proximal Glu residue. The model indicated a strong hydrogen bond between the NapSgel Ser499 hydroxyl group and the Glu503 carbonyl oxygen atom in the amide chain (2.23 Å). Additional hydrogen bonding interactions could be formed between the Ser499 hydroxyl group and the Glu503 carboxylate group (Fig. 4C). The analogous hydrogen bonding interactions would not be possible between Ala500 and Glu504 in mesophilic NapA (Fig. 4D). The mesophilic to psychrophilic Ser703 M Thr700 substitution manifested in the homology model as stronger hydrogen bonding interactions between Thr700 and Thr698 in NapASgel relative to the interactions between Ser703 and Ser701 in the mesophilic analogue. There was a strong hydrogen bond (2.14 Å) between the 20 -OH group of the ribose ring of the P-MGD group and Thr700 in NapSgel. These substitutions suggested that increased conformational flexibility in regions close to the active site might be managed in the cold-adapted enzyme via

P.J.L. Simpson, R. Codd / Biochemical and Biophysical Research Communications 414 (2011) 783–788

787

Fig. 4. Homology model of NapA from S. gelidimarina (A) generated with the SWISS-MODEL workspace using R. sphaeroides NapA (PDB: 1OGY Chain A) (B) as a template (identity 66%) [23]. The ribbon diagram includes NapA amino acid residues that are conserved across psychrotrophic S. frigidimarina, mesophilic S. putrefaciens, R. sphaeroides and E. coli and are different from psychrophilic S. gelidimarina. The His264. . .W666 double substitution in NapA from S. gelidimarina is marked by an arrow. Substituted residues in NapA within 4.5 Å of the bis-molybdopterin guanine dinucleotide (Mo–MGD) unit in the psychrophilic (Met – and proximal Ala; Thr, Ser) or psychrotrophic and mesophilic (Phe – and proximal Leu; Ser, Ala) NapA proteins are shown in (C) and (D), respectively, together with a conserved Glu residue.

restrictions induced by additional hydrogen-bonding interactions. The mesophilic to psychrophilic Phe348 M Met347 substitution decreased hydrophobicity in the region close to the guanidine group of the Q-MGD unit. In mesophilic NapA, the side chain of Leu241 is positioned close to the phenyl group of Phe348 (distance 4.2 Å). In NapSgel these residues have been substituted with Met347 and Ala241, respectively, which would attenuate local hydrophobic interactions. In summary, relative to NapSput, NapSgel was shown to be coldadapted, according to the enzyme exhibiting the following characteristics: (i) a lower temperature of irreversible thermal deactivation (NapSgel = 54.5 °C, NapSput = 65 °C); (ii) constant maximal catalytic capacity (measured at room temperature) after pre-incubation at temperatures 21–54 °C, indicating that NapSgel is poised for low-temperature activity and, unlike NapSput, gained no catalytic benefit from thermally-induced refolding; and (iii) broader substrate specificity with the oxyanion, selenate. Sequence analysis and homology modeling of NapSgel revealed from the 777 amino acid residue protein, 46 residues as candidates for governing the cold-adaptive phenotype. Of the 10 residues that were present in the protein interior, six residues were located close to the Mo–MGD cofactor: M347, S499, T700, H264, W666, A241; and comprise a tractable shortlist for future site-directed mutagenesis studies. At the forefront of mononuclear Mo enzyme research is the design of robust biosensing devices for the detection of oxyanion contamination in the environment [30]. An arsenite biosensor using immobilized arsenite oxidase has recently been described [31,32]. Further, both amperometric- [33] and conductometricbased [34] nitrate biosensors have been developed to measure nitrate contamination in surface and groundwater. Elevated levels of nitrate in the aquatic ecosystem, which result from agricultural runoff, industrial waste and sewage processing, presents a significant environmental problem. Each of these biosensors used a mesophilic nitrate reductase [33,34]. Using the cold-adapted Nap from

this work in the biosensor design may improve the stability of the biosensor when stored at refrigerated temperatures and the reliability of on-site detection of nitrate levels at the cooler environmental temperatures of rivers and lakes. Acknowledgments Dr. D.S. Nichols (University of Tasmania) is kindly acknowledged for supplying S. gelidimarina ACAM 456T and S. putrefaciens ATCC 8071T and Drs. M.J. Maher and J.M. Santini are thanked for their contributions to the early stages of this work. Financial support is acknowledged from The Hermon Slade Foundation (HSF 06/7) and from the Australian Antarctic Division [AAS Grant 2547 (RC) and Top-Up Postgraduate Scholarship (PJLS)]. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.bbrc.2011.10.003. References [1] J.F. Stolz, P. Basu, J.M. Santini, R.S. Oremland, Arsenic and selenium in microbial metabolism, Annu. Rev. Microbiol. 60 (2006) 107–130. [2] J.J.G. Moura, C.D. Brondino, J. Trincão, M.J. Romão, Mo and W bis-MGD enzymes: nitrate reductases and formate dehydrogenases, J. Biol. Inorg. Chem. 9 (2004) 791–799. [3] R. Hille, The mononuclear molybdenum enzymes, Chem. Rev. 96 (1996) 2757– 2816. [4] A.G. McEwan, J. Pridge, C.A. McDevitt, The DMSO reductase family of microbial molybdenum enzymes. Molecular properties, and role in the dissimilatory reduction of toxic elements, Geomicrobiol. J. 19 (2002) 3–21. [5] P.J. González, C. Correia, I. Moura, C.D. Brondino, J.J.G. Moura, Bacterial nitrate reductases: molecular and biological aspects of nitrate reduction, J. Inorg. Biochem. 100 (2006) 1015–1023. [6] D.J. Richardson, B.C. Berks, D.A. Russell, S. Spiro, C.J. Taylor, Functional, biochemical and genetic diversity of prokaryotic nitrate reductases, Cell. Mol. Life Sci. 58 (2001) 165–178.

788

P.J.L. Simpson, R. Codd / Biochemical and Biophysical Research Communications 414 (2011) 783–788

[7] M.K. Chan, S. Mukund, A. Kletzin, M.W.W. Adams, D.C. Rees, Structure of a hyperthermophilic tungstopterin enzyme, aldehyde ferredoxin oxidoreductase, Science 267 (1995) 1463–1469. [8] A.A.H. Pakchung, C.Z. Soe, R. Codd, Studies of iron-uptake mechanisms in two bacterial species of the Shewanella genus adapted to middle-range (Shewanella putrefaciens) or Antarctic (Shewanella gelidimarina) temperatures, Chem. Biodivers. 5 (2008) 2113–2123. [9] A.A.H. Pakchung, P.J.L. Simpson, R. Codd, Life on Earth. Extremophiles continue to move the goal posts, Environ. Chem. 3 (2006) 77–93. [10] D. Georlette, V. Blaise, T. Collins, S. D’Amico, E. Gratia, A. Hoyoux, J.-C. Marx, G. Sonan, G. Feller, C. Gerday, Some like it cold: biocatalysis at low temperatures, FEMS Microbiol. Rev. 28 (2004) 25–42. [11] G. Feller, E. Narinx, J.L. Arpigny, M. Aittaleb, E. Baise, S. Genicot, C. Gerday, Enzymes from psychrophilic organisms, FEMS Microbiol. Rev. 18 (1996) 189– 202. [12] R. Cavicchioli, T. Charlton, H. Ertan, S. Mohd Omar, K.S. Siddiqui, T.J. Williams, Biotechnological uses of enzymes from psychrophiles, Microbial Biotechnol. 4 (2011) 449–460. [13] K.S. Siddiqui, R. Cavicchioli, Cold-adapted enzymes, Annu. Rev. Biochem. 75 (2006) 403–433. [14] L.J. Rothschild, R.L. Mancinelli, Life in extreme environments, Nature 409 (2001) 1092–1101. [15] H.A. Cristóbal, A.W. Schmidt, E. Kothe, J. Breccia, C.M. Abate, Characterization of inducible cold-active b-glucosidases from the psychrotolerant bacterium Shewanella sp. G5 isolated from a sub-Antarctic ecosystem, Enzyme Microb. Technol. 45 (2009) 498–506. [16] J.P. Bowman, S.A. McCammon, D.S. Nichols, J.S. Skerratt, S.M. Rea, P.D. Nichols, T.A. McMeekin, Shewanella gelidimarina sp. nov. and Shewanella frigidimarina sp. nov., novel species with the ability to produce eicosapenataenoic acid (20:5w3) and grow anaerobically with dissimilatory Fe(III) reduction, Int. J. Syst. Bacteriol. 47 (1997) 1040–1047. [17] P.J.L. Simpson, A.A. McKinzie, R. Codd, Resolution of two native monomeric 90 kDa nitrate reductase active proteins from Shewanella gelidimarina and the sequence of two napA genes, Biochem. Biophys. Res. Commun. 398 (2010) 13– 18. [18] P.J.L. Simpson, D.J. Richardson, R. Codd, The periplasmic nitrate reductase in Shewanella: the resolution, distribution and functional implications of two NAP isoforms, NapEDABC and NapDAGHB, Microbiology 156 (2010) 302–312. [19] P. Arnoux, M. Sabaty, J. Alric, B. Frangioni, B. Guigliarelli, J.-M. Adriano, D. Pignol, Structural and redox plasticity in the heterodimeric periplasmic nitrate reductase, Nat. Struct. Biol. 10 (2003) 928–934. [20] C.H. MacGregor, Isolation and characterisation of nitrate reductase from Escherichia coli, Methods Enzymol. 53 (1978) 347–355.

[21] S.W.M. Kengen, G.B. Rikken, W.R. Hagen, C.G. Van Ginkel, A.J.M. Stams, Purification and characterization of (per)chlorate reductase from the chloraterespiring strain GR-1, J. Bacteriol. 181 (1999) 6706–6711. [22] I. Schröder, S. Rech, T. Krafft, J.M. Macy, Purification and characterisation of the selenate reductase from Thauera selenatis, J. Biol. Chem. 272 (1997) 23765– 23768. [23] K. Arnold, L. Bordoli, J. Kopp, T. Schwede, The SWISS-MODEL workspace. A web-based environment for protein structure homology modelling, Bioinformatics 22 (2006) 195–201. [24] M. Sabaty, C. Avazeri, D. Pignol, A. Verméglio, Characterization of the reduction of selenate and tellurite by nitrate reductases, Appl. Environ. Microbiol. 67 (2001) 5122–5126. [25] G. Feller, C. Gerday, Psychrophilic enzymes: hot topics in cold adaptation, Nat. Rev. Microbiol. 1 (2003) 200–208. [26] S. D0 Amico, P. Claverie, T. Collins, D. Georlette, E. Gratia, A. Hoyoux, M.-A. Meuwis, G. Feller, C. Gerday, Molecular basis of cold adaptation, Philos. Trans. R. Soc. Lond. B 357 (2002) 917–925. [27] B.J.N. Jepson, S. Mohan, T.A. Clarke, A.J. Gates, J.A. Cole, C.S. Butler, J.N. Butt, A.J. Hemmings, D.J. Richardson, Spectropotentiometric and structural analysis of the periplasmic nitrate reductase from Escherichia coli, J. Biol. Chem. 282 (2007) 6425–6437. [28] P. Benkert, M. Biasini, T. Schwede, Toward the estimation of the absolute quality of individual protein structure models, Bioinformatics 27 (2011) 343– 350. [29] S. Bhattacharya, J.T. Lecomte, Temperature dependence of histidine ionization constants in myoglobin, Biophys. J. 73 (1997) 3241–3256. [30] T.H. Osborne, H.E. Jamieson, K.A. Hudson-Edwards, D.K. Nordstrom, S.R. Walker, S.A. Ward, J.M. Santini, Microbial oxidation of arsenite in a subarctic environment: diversity of arsenite oxidase genes and identification of a psychrotolerant arsenite oxidiser, BMC Microbiol. 10 (2010) 205. [31] P.V. Bernhardt, J.M. Santini, Protein film voltammetry of arsenite oxidase from the chemolithoautotrophic arsenite-oxidizing bacterium NT-26, Biochemistry 45 (2006) 2804–2809. [32] K.B. Male, S. Hrapovic, J.M. Santini, J.H.T. Luong, Biosensor for arsenite using arsenite oxidase and multiwalled carbon nanotube modified electrodes, Anal. Chem. 79 (2007) 7831–7837. [33] S.A. Galzier, E.R. Campbell, W.H. Campbell, Construction and characterisation of nitrate-reductase-based amperometric electrode and nitrate assay of fertilizers and drinking water, Anal. Chem. 70 (1998) 1511–1515. [34] W. Xuejiang, S.V. Dzyadevych, J.-M. Chovelon, N. Jaffrezic Renault, C. Ling, X. Siqing, Z. Jianfu, Conductometric nitrate biosensor based on methyl viologen/ Nafion/nitrate reductase interdigitated electrodes, Talanta 69 (2006) 450–455.