JB Accepts, published online ahead of print on 30 July 2010 J. Bacteriol. doi:10.1128/JB.00235-10 Copyright © 2010, American Society for Microbiology and/or the Listed Authors/Institutions. All Rights Reserved.

Revised to: J. Bacteriol. MS No. JB00235-10

1 2 3 4 5 6 7

Characterization of NADH Oxidase/NADPH Polysulfide Oxidoreductase and Its Unexpected Participation in Oxygen Sensitivity in an Anaerobic Hyperthermophilic Archaeon

8 9 10 11

Hiroki Kobori1, Masayuki Ogino2, Izumi Orita1, Satoshi Nakamura1, Tadayuki Imanaka3, and Toshiaki Fukui1*

12 13 14 15 16 17

Department of Bioengineering, Graduate School of Bioscience and Biotechnology, Tokyo Institute of Technology, B-37 4259 Nagatsuta, Midori-ku, Yokohama 226-85011, and School of Materials Science, Japan Advanced Institute of Science and Technology, 1-1 Asahidai, Nomi, Ishikawa 923-12922, Department of Biotechnology, College of Life Sciences, Ritsumeikan University, 1-1-1 Nojihigasi, Kusatsu, Shiga 525-85773, Japan

18 19 20 21 22

*Corresponding author. Toshiaki Fukui Department of Bioengineering Graduate School of Bioscience and Biotechnology

25

Tokyo Institute of Technology B-37 4259 Nagatsuta, Midori-ku, Yokohama 226-8501, Japan Tel/Fax: +81-45-924-5766

26

e-mail: [email protected]

23 24

27 28 29 30 31 32 33

Running title NADH Oxidase Homolog in Anaerobic Hyperthermophile

34 35 36 37

1

38

ABSTRACT

39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61

Many genomes of anaerobic hyperthermophiles encode multiple homologs of NAD(P)H oxidase which are thought to function in response to oxidative stresses. We investigated one of the seven NAD(P)H oxidase homologs (TK1481) in the sulfur-reducing hyperthermophilic archaeon Thermococcus kodakarensis, focusing on the catalytic properties and roles in oxidative stress defense and sulfur-dependent energy conservation. The recombinant form of TK1481 exhibited both NAD(P)H oxidase and NAD(P)H: polysulfide oxidoreductase activities. The enzyme also possessed low NAD(P)H peroxidase and NAD(P)H: elemental sulfur oxidoreductase activities under anaerobic conditions. A mutant form of the enzyme, in which the putative redox active residue Cys43 was replaced by Ala, still showed NADH-dependent FAD-reduction activity. Although it also retained successive oxidase and anaerobic peroxidase activities, the ability to reduce polysulfide and sulfur was completely lost, suggesting the specific reactivity of the Cys43 residue for sulfur. To evaluate the physiological function of TK1481, a gene deletant ∆TK1481 and mutant KUTK1481C43A, in which two-base mutations corresponding to Cys43Ala mutation of TK1481 were introduced, were constructed. ∆TK1481 exhibited nearly identical growth properties as parent strain KU216 in sulfur-containing media. Interestingly, in the absence of elemental sulfur, the growth of ∆TK1481 was not affected by dissolved oxygen, whereas those of KU216 and KUTK1481C43A were significantly impaired. These results indicate that although TK1481 does not play a critical role in either sulfur reduction or response against oxidative stress, the NAD(P)H oxidase activity of TK1481 unexpectedly participates in the oxygen sensitivity of the hyperthermophilic archaeon T. kodakarensis in the absence of sulfur.

62 63 64 65 66 67 68 69 70 71 72 73 74

2

75

INTRODUCTION

76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111

Recent genetic analyses have determined that hyperthermophiles, which are capable of growth at 90˚C and above, occupy the deep branches closest to the root within phylogenetic trees. This suggests that the study of hyperthermophiles will provide valuable perspectives on the evolution of biological systems essential for living organisms. Interestingly, although many hyperthermophiles are strict anaerobes, they are equipped with several systems for responding to oxidative stress. They often employ superoxide reductase for removal of the highly toxic superoxide anion, instead of the more universal superoxide dismutase (13). In Pyrococcus furiosus, it has been proposed that the reduction of superoxide anion to H2O2 by superoxide reductase is linked to the oxidoreduction of rubredoxin by NAD(P)H: rubredoxin oxidoreductase (10). Although the mechanism used for the removal of moderately toxic H2O2 by hyperthermophiles is unclear, Dps-like protein (27) and rubrerythrin (39), which both contain non-heme iron, exhibit peroxidase activity and may serve this role. Osmotically inducible protein C (OsmC) identified in Thermococcus kodakarensis also shows peroxidase activity in the presence of reducing agents in vitro (25). Detailed genome survey have also revealed the presence of multiple NAD(P)H oxidases (NOXs) in a wide range of anaerobic hyperthermophiles, including Thermococcus (14) and Pyrococcus spp. (38), Archaeoglobus fulgidus (17, 28), Methanocaldococcus jannaschii (4), and Thermotoga spp (41, 42). NOXs catalyze the two-electron reduction of O2 to H2O2 or four-electron reduction of O2 to H2O or four-electron reduction of O2 to H2O and contain a highly conserved domain containing FAD-binding and NAD(P)H-binding sites (7). They are members of the pyridine nucleotide disulfide oxidoreductase (PNDOR) class of enzymes, which are in turn included in the GR1 subfamily of the glutathione reductase structural family (1). PNDORs are divided into three groups, with NOXs belonging to Group 3 together with NADH peroxidases and CoA-disulfide reductases (CoADRs). The reaction mechanisms of a number of Group 3 PNDORs have been investigated in detail. For example, the H2O-forming NOX from mesophilic Lactobacillus sanfranciscensis uses the first equivalent of NAD(P)H to form an enzyme-FADH2 complex that reacts with O2 to generate H2O2 (18). The H2O2 molecule further reacts with a redox-active cysteine via a peroxyflavin intermediate to form cysteine-sulfenic acid (Cys-S-OH) and one water molecule. The second equivalent of NAD(P)H is then used to reduce the sulfenic acid to thiolate via FADH2 which is accompanied by the production of a second water molecule. The cysteine-sulfenic acid redox center is also important in reactions catalyzed by NADH peroxidases (26). Likewise, the active Cys43 residue in 3

112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141

CoADR from Staphylococcus aureus forms a Cys43-S-SCoA redox center during turnover (23). The NOX homologs in hyperthermophiles are estimated to play important roles in defense against oxidative stress (4, 14, 17, 28, 38, 41, 42), nevertheless, the actual physiological functions of NOXs in anaerobic hyperthermophiles have not yet been completely elucidated mainly due to a lack of useful genetic manipulation systems for these organisms. Members of the genera Pyrococcus and Thermococcus, which both belong to the hyperthermophilic archaeal order Thermococcales, are obligate heterotrophs which generally prefer proteinous compounds as carbon and energy sources and whose growth is strongly associated with the reduction of elemental sulfur (S0) to H2S. Interestingly, unlike mesophilic sulfur-reducing bacteria such as Wolinella succinogenes (6), the available genomes of these hyperthermophiles do not contain any homologs related to S0-reducing respiratory systems, such as molybdenum-containing sulfur reductases, quinones, and cytochromes. In P. furiosus, cytosolic hydrogenases (21) and a sulfide dehydrogenase (19) were initially reported to exhibit S0-reducing activity in vitro. However, the former enzymes are now thought to function in the recycling of H2 evolved by membrane-bound hydrogenase, while the latter has been re-characterized as ferredoxin: NADP oxidoreductase (20). Recently, Schut et al. shed light on the mysterious mechanism of sulfur reduction in the order Thermococcales (34). They proposed that a unique S0-dependent energy conservation system operates in P. furiosus, in which membrane-bound hydrogenase-related membrane-bound oxidoreductase first generates a proton-motive force by proton pumping via electron transfer from reduced ferredoxin to NAD(P)+. CoA-dependent NAD(P)H: elemental sulfur oxidoreductase then reoxidizes NAD(P)H with S0 as a terminal electron acceptor. It should be noted that NAD(P)H: elemental sulfur oxidoreductase, which is encoded by PF1186, is one of the NOX homologs in P. furiosus, suggesting that the energetic reduction of sulfur and defensive reduction of oxygen may share similar catalytic mechanisms mediated by NOX homologs in this order of hyperthermophiles. T. kodakarensis KOD1 (2) is a useful model hyperthermophile as its whole genome has been sequenced and annotated (8) and a practical genetic manipulation system, the first for

147

hyperthermophiles, has been developed for this archaeon (24, 32, 33). T. kodakarensis exhibits similar metabolic and energy-conserving mechanisms to P. furiosus. For example, although it prefers S0 as a terminal electron acceptor, it is able to couple growth with the concomitant reduction of protons to hydrogen when either pyruvate or starch is available in the absence of S0. One of the major distinctions between T. kodakarensis and P. furiosus is the preferred growth temperature, as the former displays a lower and wider growth

148

temperature range (60-100˚C) than the latter (70-103˚C). Previous comparative genomics

142 143 144 145 146

4

149 150 151 152 153 154 155 156

of T. kodakarensis and three Pyrococcus spp. have revealed that they share 1,204 proteins, including those involved in information processing and basic metabolism. However, 689 proteins were unique to T. kodakarensis which are likely responsible for the specific traits displayed by this species and other members of the genus Thermococcus (8). For example, the unique presence of methionine sulfoxide reductase, an enzyme which repairs oxidized proteins, has been suggested to facilitate growth in low-temperature environments which often contain increased dissolved oxygen concentrations (9). In this study, we focused on a NOX homolog of T. kodakarensis, TK1481, for which a

159

homologous counterpart was not present in Pyrococcus spp. The recombinant protein was purified and characterized in terms of its ability to reduce oxygen and sulfur. In addition, we examined its physiological role by constructing a site-directed mutant of the protein

160

and gene disruption mutant of T. kodakarensis.

157 158

161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185

5

186

MATERIALS AND METHODS

187 188 189 190 191 192 193 194 195 196

Bacterial strains, plasmid, and media. T. kodakarensis strains were grown anaerobically at 85°C in a rich growth medium composed of 0.8-fold dilution of artificial seawater (Senju Seiyaku, Osaka, Japan), 10 g/l yeast extract, and 5 g/l tryptone. Into this medium, 2 g/l elemental sulfur (ASW-YT-S0) or 5.0 g/l pyruvate (ASW-YT-Pyr) was added when appropriate. Escherichia coli DH5α and BL21-CodonPlus(DE3)-RIL (Stratagene, La Jolla, CA) were used as hosts for the expression plasmid derived from pET-21b(+) (Novagen, Madison, WI). E. coli transformants were cultivated at 37°C in Luria-Bertani (LB) medium containing 50-100 µg/ml ampicillin.

197 198 199 200

DNA manipulation and sequencing. DNA manipulations were carried out by standard methods, as described by Sambrook and Russell (30). Restriction enzymes and other modifying enzymes were purchased from

202

Takara Bio (Otsu, Shiga, Japan) or Toyobo (Osaka, Japan). DNA sequencing was carried out with a model ABI 3100 capillary DNA sequencer (Applied Biosystems, Foster City,

203

CA) with BigDye Terminators v1.1 Cycle Sequencing Kit (Applied Biosystems).

201

204

207

Construction of the expression vector. The structural gene encoding an NAD(P)H oxidase homolog, tk1481, was amplified from T. kodakarensis genomic DNA with a primer pair of TK1481-F

208

(5’-AGGCATATGAAGTACGATGTGGTTGTTATAG-3’)

209

(5’-AAGGTCGACTCAGGCCCCCGCTCTAGCTTC-3’) [underlined sequences indicate NdeI in TK1481-F and SalI in TK1481-R, respectively] by using KOD-Plus DNA

205 206

210 211 212 213 214

and

TK1481-R

polymerase (Toyobo). The amplified fragment of 1,344 bp was digested by NdeI and SalI, and then ligated with pET-21b(+) at the corresponding sites. The absence of unintended mutations in the insert was confirmed by DNA sequencing. The resulting plasmid, pETTK1481 was used to transform E. coli BL21-CodonPlus(DE3)-RIL.

215 216 217 218 219 220 221 222

Expression and purification of recombinant TK1481 (rTK1481). The recombinant E. coli harboring pETTK1481 was induced by Overnight Express Autoinduction System 1 (Novagen, Madison, WI) in a 50 ml LB medium with 100 µg/ml ampicillin at 37°C. The cells were harvested by centrifugation (5,000 x g for 10 min at 4˚C), washed and resuspended within buffur A (50 mM Tris-HCl (pH 8.0)), and then disrupted by sonication for 5 min. The supernatant after centrifugation (15,000 x g for 20 min at 4˚C) was incubated at 85°C for 10 min and centrifuged (15,000 x g for 20 min at 6

223 224 225 226 227 228 229 230 231 232 233

4˚C) to obtain a heat-stable supernatant. Further purification of the recombinant protein in the heat-stable supernatant was performed with an AKTA prime chromatography system (GE Healthcare, Uppsala, Sweden). First, the heat-treated cell extract was applied to an anion-exchange HiTrap Q HP column (5 ml) (GE Healthcare) that was pre-equilibrated using buffer A. rTK1481 was eluted with a linear gradient of 0 - 1.0 M NaCl, and the peak fractions at 0.4 M NaCl were collected. The fractions were further applied to an anion-exchange Resource Q column (1 ml) (GE Healthcare) that was pre-equilibrated using buffer A. The protein fractions eluted at 0.4 M NaCl through linear gradient of 0 0.6 M NaCl were concentrated using a VIVASPIN 6 (Sartorius Stedim Biotech, Gottingen, Germany). The concentrated protein solution was then applied to a gel-filtration Superdex 200 10/300 GL column (GE Healthcare) pre-equilibrated with buffer A containing 0.15 M

236

NaCl. The fractions of rTK1481 were combined and stored at -80°C until the use. The protein solutions were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The protein concentration was determined by Bradford

237

method with bovine serum albumin as the standard.

234 235

238 239 240 241 242 243 244 245 246 247 248 249 250

Biophysical properties of the rTK1481. The subunit molecular mass of rTK1481 was determined by SDS-PAGE, and the native molecular mass was determined by gel-filtration Superdex 200 10/300 GL column calibrated with standard proteins of glutamate dehydrogenase (290 kDa), lactate dehydrogenase (142 kDa), enolase (67 kDa), myokinase (32 kDa), cytochrome C (12.4 kDa). The native molecular mass was also determined by ultracentrifugation performed with an Optima XL-I analytical ultracentrifuge (Beckman, California, USA). The sedimentation equilibrium experiment was carried out with rTK1481 solution of which absorbance at 280 nm was 1.0 with rotor speeds of 6,000, 8,000, and 10,000 rpm, and scans were recorded at every 2 h. Partial specific volume of rTK1481 based on the amino acid composition, solvent viscosity, and solvent density were calculated by SEDNTERP program (5).

251 252 253 254 255 256 257 258 259

Enzyme assay. NAD(P)H oxidase activity was assayed by determining oxidation of NAD(P)H and production of H2O2. The reaction mixture (1.0 ml) was composed of 100 µM FAD, 100 µM NAD(P)H, and the enzyme in 50 mM HEPPS (3-[4-(2-hydroxyethyl)-1-piperazinyl]propanesulfonic acid) buffer (pH 7.0). After the reaction at 70˚C for 1 min under air atmosphere, the mixture was rapidly cooled within an ice-water bath to stop the reaction. The decrease of NAD(P)H was spectrophotometrically determined from absorbance at 365 nm. H2O2 was quantified by using horseradish 7

260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289

peroxidase (Oriental Yeast, Tokyo, Japan). After the reaction for NAD(P)H oxidase, 0.15 mM 4-aminoantipyrine, 0.15 mM N-ethyl-N-(2-hydroxy-3-sulfopropyl)-3-methylaniline sodium salt dehydrate, and 3 U/ml peroxidase were added to the mixture, and then reacted at room temperature for 10 min. H2O2 concentration was determined by measurement of absorbance at 555 nm (40). One unit of activity corresponds to oxidation of 1 µmol of NAD(P)H or production of 1 µmol H2O2 per min. NADP(H) peroxidase activity was assayed anaerobically in 1.0 ml of 50 mM HEPPS buffer (pH 8.0) containing 50 µM H2O2, 100 µM NAD(P)H, and the enzyme. The mixture in a rubber-capped test tube was prepared in an anaerobic chamber. The reaction was started by addition of NAD(P)H using a syringe, and incubated at 70˚C for 5 min. After the reaction followed by rapid cooling within an ice-water bath, the decrease of NAD(P)H and H2O2 were determined as described above. NAD(P)H: polysulfide oxidoreductase activity was anaerobically determined by continuous monitoring of NAD(P)H oxidation at 365 nm. The reaction mixture (500 µl) was composed of 100 µM FAD, 100 µM NAD(P)H, 0.5 mM polysulfide and the enzyme in 50 mM HEPPS buffer (pH 7.0). The polysulfide concentration was adjusted by addition of 500 mM polysulfide solution (0.8 g of sulfur and 6.0 g of Na2S dissolved in 50 ml of water), and 1.0 mg/L resazurin was added as an indicator for removal of dissolved oxygen by the polysulfide solution. The reaction was carried out at 70˚C, and one unit of activity was defined as oxidation of 1 µmol of NAD(P)H per min. NAD(P)H: elemental sulfur oxidoreductase activity was determined by quantification of H2S and NAD(P)H with a previously reported double-vial system (34) with slight modifications. The inner reaction vial contained a reaction mixture (1.0 ml) composed of 100 µM FAD, 300 µM NAD(P)H and 1% (wt/vol) colloidal sulfur (Wako) in 50 mM HEPPS buffer (pH 8.0), and the outside vial contained 0.1 N NaOH (1.0 ml). The mixtures in each vial were prepared in an anaerobic chamber, and the outside vial was tightly capped with a rubber stopper and an aluminum seal. The reaction was started by addition of NAD(P)H using a syringe, and incubated at 70˚C for 10 min. For determination of NAD(P)H oxidation, the reaction vial was immediately cooled within an ice-water bath, and absorbance at 365 nm of the inner mixture was measured. Alternatively, the reaction

295

was stopped by addition of 2 N H2SO4 (200 µl) into the inner vial and incubated for 10 min at room temperature. H2S captured in the outside NaOH phase was quantified by the methylene blue method (3). To determine actual rates catalyzed by the enzyme, rates of spontaneous thermal degradation of NAD(P)H and H2O2 were determined by control reactions and subtracted from the observed reaction rates. Kinetic analyses of rTK1481 were performed with various concentrations of dissolved

296

oxygen (35-132 µM), H2O2 (10-100 µM), or polysulfide (50-500 µM) under the presence

290 291 292 293 294

8

297 298 299 300 301 302 303 304

of 100 µM NAD(P)H in the standard assay mixture. Dissolved oxygen concentrations were determined by sensION6 Dissolved Oxygen Meter (Hach Company, Colorado, USA) at 30°C. The buffers with different dissolved oxygen concentrations were fully poured in glass vials without headspace, and then tightly capped with a PTFE/silicone septum (GL Sciences, Tokyo, Japan) and screw cap. The vials were heated to 70°C, and the buffers were taken by syringe and quickly used for the assay. The dissolved oxygen concentrations at 70°C were estimated by Weiss’s equation. The kinetic parameters were calculated with Hanes-Woolf plot.

305 306 307 308 309 310 311 312 313 314 315 316 317 318

Site-directed Mutagenesis. Mutagenesis of putative active residue, Cys43, in TK1481 was carried out based on the procedure reported by Sugimoto et al (36) by using primers TK1481Cys43Ala-F (5’-CGTTTCCATGATTCCTGCCGGGATTCCGTACATC-3’) and TK1481Cys43Ala-R (5’-GATGTACGGAATCCCGGCAGGAATCATGGAAACG-3’) [underlined sequences indicate the mutation point for replacement of Cys43 to Ala] and pETTK1481 as a mutagenic primer pair and template, respectively. The PCR reaction was performed by using KOD-Plus DNA polymerase with the following temperature cycles: 16 cycles for denaturation at 98°C for 20 sec, annealing at 58°C for 1 min, and extension at 68°C for 7 min. The amplified fragment was used to transform E. coli DH5α. Sequence of the resulting plasmid pETTK1481C43A was confirmed, and then introduced into the expression strain E. coli BL21-CodonPlus(DE3)-RIL. Induction and purification of the mutant protein, rTK1481C43A, was done with the same procedures as those for rTK1481.

319 320 321 322 323 324 325 326 327 328 329 330 331 332 333

Targeted modification of tk1481 on T. kodakarensis chromosome. Disruption of tk1481 on the chromosome was performed using the pop-in-pop-out recombination technique developed for T. kodakarensis (33). The plasmid used for disruption of tk1481 was constructed as follows. A plasmid pUD2 was digested by PvuII, and a 763 bp fragment of pyrF marker cassette was inserted into pUC118 at the SspI site (outside of multi-cloning sites) to obtain pUD3. A DNA fragment including tk1481 along with its flanking regions (997 bp and 994 bp for upstream and downstream regions, respectively) was amplified from the genomic DNA of T. kodakarensis KOD1 with primers Dest-TK1481-F (5’-TATAGAGGAGCGGGCAGAGG-3’) and Dest-TK1481-R (5’-AGCTCTTTGAGCTGGGCAGG-3’), and then inserted into pUD3 at the HincII site. Inverse PCR with primers Inv-TK1481-F (5’-ATTCCCCTTTCATTTTCTAG-3’) and Inv-TK1481-R (5’-ACACTTTCCTTAACCCGAGG-3’) was carried out to amplify the flanking regions and plasmid backbone, thereby removing the coding region. The amplified DNA fragment was self-ligated after 5’-phosphorylation, resulting in the 9

334 335 336 337 338 339 340 341 342 343 344 345 346 347

construction of pUDTK1481. For substitution of 129-TGC-131 by GCC in chromosomal tk1481 (corresponding to Cys43Ala mutation in the translated product), a plasmid pUDTK1481C43A was constructed as the same procedure as that for pETTK1481C43A with the exception of using pUDTK1481 as a template. Transformation of T. kodakarensis uracil auxotroph strain KU216 (∆pyrF) with pUDTK1481 was performed according to the procedure described previously (32). A pop-in strain (pyrF+) exhibiting uracil prototrophy was concentrated by cultivation in uracil-free synthetic medium ASW-AA-S0 (32) for three times, followed by cultivation on ASW-YT plate medium containing 0.75% (w/v) 5-fluoroorotic acid monohydrate (5-FOA). Among several pop-out strains with 5-FOA-resistance (pyrF-), a tk1481-deleted strain (∆TK1481) was selected by PCR analysis using primers of Dest-TK1481-F and Dest-TK1481-R. Selection of tk1481-mutated strain (KUTK1481C43A) was performed by direct sequence analysis of the fragments amplified by using primers TK1481-F and TK1481-R.

348

351

Determination of growth properties. The deletant ∆TK1481, mutant strain KUTK1481C43A, host strain KU216 and wild strain KOD1 were pre-cultivated in 100 ml of ASW-YT-S0 or ASW-YT-Pyr at 85°C for 12

352

h, and the cells were then cultivated at 85˚C in 8 ml of the same medium containing 1

353

mg/L of resazulin within a test tube (φ16 mm). For general anaerobic inoculation, the pre-cultured broth was inoculated into the medium which had been anaerobically cooled after autoclaving for 15 min without addition of reducing agent, and all operations were performed in an anaerobic chamber. Otherwise, cooling of a culture medium after

349 350

354 355 356 357 358 359 360 361 362 363 364 365 366 367 368

autoclaving and inoculation were performed under air atmosphere for aerobic inoculation. To make intermediary aerobic condition, sodium sulfide was added as a reducing agent with a final concentration of 20 mg/ml. The test tube containing the medium was tightly capped with a rubber stopper and a screw cap after the inoculation, and then cultivated at 85˚C. Optical density of the culture in the test tube at 600 nm was directly measured with an S1200 Diode Array Spectrophotometer (Biochrom, Berlin, Germany) with appropriate intervals. No formation of precipitates was observed in the media during the cultivation examined (besides elemental sulfur in ASW-YT-S0), and it has been confirmed that the increase of OD600 was linearly correlated with the increase of protein concentrations in extracts of the harvested cells. The growth under each condition was examined with 10 test tubes (ASW-YT-S0) or 20 test tubes (ASW-YT-Pyr), and mean data were calculated after excluding outliers (p < 0.05).

369 370

10

371 372 373 374 375 376 377 378 379 380 381

Western blot analysis. T. kodakarensis KU216 cells grown in 50 ml of ASW-YT-S0 or ASW-YT-Pyr were harvested at early-exponential, mid-exponential, and early-stationary phases when OD600 were 0.15, 0.3, and 0.45 in ASW-YT- S0 or 0.3, 0.8, and 1.2 in ASW-YT-Pyr, respectively. The cells were washed with 0.8-fold dilution of artificial seawater, disrupted by sonication in 50 mM Tris-HCl (pH 8.0) for 1 min, and then centrifuged (15,000 x g for 20 min at 4˚C) to obtain a soluble protein fraction. The fractions were subjected to SDS-PAGE and successive Western blot analysis using specific antiserum (rabbit) against the rTK1481 protein. A goat anti-rabbit IgG–alkaline phosphatase conjugate was used to visualize the specific protein along with nitroblue tetrazolium chloride and 5-bromo-4-chloro-3’-indolylphosphate p-toluidine salt.

382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407

11

408

RESULTS

409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437

Primary structure of TK1481. Homology and domain searches of T. kodakarensis genome identified 7 homologs of NOXs, and the counterparts in three Pyrococcus spp. (P. furiosus, P. abyssi, and P. horikoshii) and some anaerobic hyperthermophiles are listed in Table 1. As one homolog, TK1481, was found to be unique to T. kodakarensis and did not present in Pyrococcus spp., we examined the enzymatic characteristics and physiological functions of this NOX homolog, taking a possibility into consideration that it may be responsible for the specific traits of T. kodakarensis distinguished from Pyrococcus spp. The gene tk1481 consisted of 1,344 bp and encoded a protein of 448 amino acids with a predicted molecular mass of 48,780 Da. The deduced amino acid sequence shared 24-33% identity with the other six NOX paralogs in T. kodakarensis, and showed 31-35% identity with previously reported NOXs from hyperthermophiles, including H2O2-H2O-forming NOX1 from P. furiosus (PF1532) (33%) (38), H2O2-forming NoxA-2 from A. fulgidus (AF0395) (32%) (28), H2O-forming NOX from Thermococcus profundus (31%) (14), CoADR from P. horikoshii (PH0572) (31%) (11), and NAD(P)H: elemental sulfur oxidoreductase from P. furiosus (PF1186) (35%) (34). The homology of TK1481 with well-studied NOXs from mesophiles was lower, as demonstrated by the 27-28% identity with H2O-forming NOX from L. sanfranciscensis (18), Streptococcus mutans (12), and Enterococcus faecalis (29), and the 25% identity with H2O2-forming NOX from S. mutans (12). In TK1481, a putative redox-active cysteine residue was identified at position 43 in the N-terminal region of the Pyr_redox_2 domain. In the T. kodakarensis genome, TK1481 was annotated as NADH: polysulfide oxidoreductase, which was derived from the description of a NOX homolog from Thermotoga neapolitana (NpoTn) submitted to the nr database in 2002. However, as the specific properties of this NOX homolog have not yet been published, we therefore examined if TK1481 could function as an oxidoreductase of oxygen, sulfur, and their related compounds, and its possible involvement in defense against oxidative stress and/or sulfur-dependent energy conservation in T. kodakarensis.

438 439 440 441 442 443 444

Overexpression of tk1481 and purification of the recombinant protein. The recombinant protein of TK1481 (rTK1481) could be obtained in a soluble, heat-stable form from E. coli cells harboring pETTK1481, and was further purified using a series of column chromatography separation steps until apparent homogeneity was achieved. rTK1481 had a subunit molecular mass of 47 kDa by SDS-PAGE (data not shown), and the native molecular mass was determined to be 188.5 kDa and 184.2 kDa by 12

445 446 447 448 449

ultracentrifugation and gel filtration chromatography, respectively, which were consistent with the formation of a homotetramer. Generally, mesophilic NOXs form either homodimers or homotetrames, while a few NOXs from hyperthermophiles are also known to be homodimeric (AF0254 (17), PF1186 (34), PF1532 (38)) or homotetrameric (PH0572) (11). An NOX from T. profundus has been reported to form a hexameric ring structure (14).

450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474

NAD(P)H oxidase and NAD(P)H peroxidase activities of rTK1481. After its successful purification, the NOX activity of rTK1481 was examined. In this assay, both the initial rates of NAD(P)H oxidation and H2O2 formation were measured to determine the stoichiometry of the reaction catalyzed by this enzyme. As shown in Table 2, rTK1481 actually showed both NADH- and NADPH-dependent oxidase activities, with NADH acting as a superior electron donor to NADPH. The optimum pH and temperature of the NADH oxidase activity were determined to be 7.0 and 90˚C, respectively, indicating that this enzyme can suitably function under the typical growth environments of this hyperthermophile. The results also demonstrated that regardless of the electron donor, the specific activities determined from NAD(P)H oxidation were similar to those from H2O2 formation. Although there were slight differences between the two measurements, the amount of NAD(P)H consumed (35.1 µmol [NADH] and 37.8 µmol [NADPH]) was almost equivalent to that consumed during H2O2 formation (35.6 µmol [NADH] and 36.2 µmol [NADPH]) after a 10 min reaction at 70˚C. From this result, it therefore appeared that rTK1481 was a H2O2-forming NOX. The specific activity of rTK1481 (5.62 U/mg as NADH oxidase) was comparable to that of other NOXs from other hyperthermohiles, which range from 4–16 U/mg (4, 11, 14, 17, 38). In addition to these enzymes, a heterodimeric NADH oxidase from T. maritima of which α-subunitis NOX homolog, has been reported to have quite high activity (144 U/mg) (42). The solution of purified rTK1481 was yellowish, suggesting the presence of oxidized prosthetic FAD groups in this enzyme, which have also been observed in other NOXs from hyperthermophiles (4, 10, 14, 17, 28, 38, 41, 42). This was supported by the presence of absorption peaks at 450 nm in the spectrum of the purified enzyme (Fig. 1A). When NADH was added to the enzyme solution at 90˚C under anaerobic conditions, the

480

yellowish color and absorbance at 450 nm quickly disappeared, but could be restored by stirring the solution at 90˚C under an air atmosphere (Fig. 1B). These observations coincided with the reduction of FAD to FADH2 by NADH, followed by electron transfer from FADH2 to an oxygen molecule serving as the terminal acceptor. In the presence of free FAD, the reduced form of the protein could not be arrested, most likely due to turnover by catalytic electron transfer from NADH to free FAD. As determined from the

481

NAD(P)H oxidation rates, the specific activity of rTK1481 with free FAD was determined

475 476 477 478 479

13

482 483 484 485 486 487 488 489

to be 7.99 ± 0.30 U/mg [NADH] and 0.711 ±0.14 U/mg [NADPH], suggesting the preference of this enzyme for NADH as an electron donor during the reduction of FAD. When rTK1481 was incubated with H2O2 at 70˚C under anaerobic conditions, appreciably equivalent decreased rates of NAD(P)H and H2O2 were observed (Table 2), indicating that rTK1481 also possessed NADH peroxidase activity. For this activity, no preference for electron donor was observed. Although the specific NADH peroxidase activity (0.34-0.41 U/mg) was one magnitude lower than that of the NADH oxidase activity of rTK1481, it was similar to the NADPH oxidase activity.

490 491

NAD(P)H:

492

oxidoreductase activities of rTK1481. As TK1481 was considered to play a role in sulfur reduction by T. kodakarensis, the NAD(P)H: polysulfide oxidoreductase and NAD(P)H: elemental sulfur oxidoreductase activities of rTK1481 were next investigated (Table 2). When the oxidation of NAD(P)H was measured in the presence of polysulfide at pH 8.0 and 70˚C, rTK1481 showed a high NAD(P)H: polysulfide oxidoreductase activity. Interestingly, unlike the observed oxidase

493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518

polysulfide

oxidoreductase

and

NAD(P)H:

elemental

sulfur

activity, NADPH supported the reaction (56.8 U/mg) better than NADH (29.8 U/mg). For the NADPH-dependent activity, the optimum pH and temperature were 7.0 and 70˚C, respectively. The NAD(P)H: elemental sulfur oxidoreductase assay was performed anaerobically with suspended colloidal sulfur, and both NAD(P)H oxidation and H2S production were evaluated. Although rTK1481 also exhibited NAD(P)H: elemental sulfur oxidoreductase activity (approximately 1 U/mg), it was much lower than the NAD(P)H: polysulfide oxidoreductase activity. The consumption of NAD(P)H during the enzymatic reaction agreed with the stoichiometric formation of H2S, suggesting that exclusive electron transfer from NAD(P)H to elemental sulfur occurred under the conditions examined. To date, the sulfur-reducing activity of several enzymes within the order Thermococcales have been reported, including the NADPH-dependent reduction of polysulfide by cytosolic NiFe-hydrogenases I and II from P. furiosus (2.54 U/mg and 0.07 U/mg, respectively, calculated from the apparent kcat values (21)), NADPH-dependent reduction of polysulfide (7.0 U/mg) and elemental sulfur (2.9 U/mg) by heterodimeric sulfide dehydrogenase from P. furiosus (19), and NADPH-dependent reduction of polysulfide by recombinant NsoC (one subunit of a putative NADPH-dependent oxidoreductase) from T. litoralis (0.35 U/mg) (37). As a recently identified NAD(P)H: elemental sulfur oxidoreductase from P. furiosus (PF1186) was reported to exhibit a very high specific activity (112 U/mg) in a CoA-dependent manner (34), the effects of CoA on the NAD(P)H: elemental sulfur oxidoreductase activity of rTK1481 were examined, however, a significant dependency on CoA was not observed. The NAD(P)H: polysulfide 14

519 520 521 522 523 524 525 526 527 528

oxidoreductase activity of rTK1481 was higher than the activity of NAD(P)H: polysulfide oxidoreductases previously reported in hyperthermophile, and was comparable with the NAD(P)H: elemental sulfur oxidoreductase activity of PF1186. When the reduction of S0 by rTK1481 under aerobic conditions was performed, the enzyme exclusively produced H2O2 and the formation of only a small amount of H2S was detected. Although it is known that H2S can non-enzymatically react with H2O2 to form S0 and water, the appreciably stoichiometric formation of H2O2 (63.9 ±5.04 µmol) and H2S (1.73 ±0.56 µmol) to NADH oxidation (63.7 ±2.82 µmol) suggested that this non-enzymatic reaction did not occur or was negligible under the conditions examined here.

529 530 531 532 533 534 535 536 537 538 539 540 541

Kinetic analysis of rTK1481. Kinetic parameters of rTK1481 for oxygen, H2O2, and polysulfide were then determined as shown in Table 3. Apparent Km values of rTK1481 toward oxygen were 42.9-47.1 µM, which were almost the same as that of NOX from T. maritima (43 µM) (42) and slightly lower than those of a few NOXs from hyperthermophiles (60-110 µM) (17, 38, 41). It has been reported that a Km value of NOX from M. jannaschii (MJ0649) was quite high (1,900 µM) (4). Interestingly, apparent Km values toward H2O2 (4.31-15.9 µM) were much lower than those to oxygen. As described above, we observed nearly stoichiometric formation of H2O2 accompanied by oxidation of NAD(P)H under aerobic conditions, which indicated that reductive consumption of H2O2 by NADH peroxidase activity of rTK1481 was negligible despite the high affinity of the enzyme toward H2O2. This was accounted for by much higher reduction rate for oxygen than for H2O2 as a result of the

545

large Vmax toward oxygen and the presence of dissolved oxygen with saturated concentration in the assay buffer at 70˚C (approximately 132 µM). rTK1481 showed much smaller apparent Km values (53.5-64.6 µM) toward polysulfide when compared to 540-2000 µM of the enzymes exhibiting sulfur-reducing activity from the order

546

Thermococcales (19, 21, 37).

542 543 544

547

554

Role of the putative redox-active cysteine Cys43. Previous studies on the reaction mechanisms by Group 3 PNDORs have confirmed the formation of a Cys-S-OH (for NOX and NADH peroxidase) or Cys-S-S-CoA (for CoADR) intermediate at the redox-active cysteine during the catalytic cycle. The mutation of Cys42 to Ser in the NOX from E. faecalis resulted in the production of H2O2 as opposed to H2O (22). Likewise, the Cys42Ser mutant of the NADH peroxidase from E. faecalis showed substantially lower activity than the wild-type enzyme (26). We therefore constructed a

555

Cys43Ala mutant, rTK1481C43A, by site-directed mutagenesis and investigated the

548 549 550 551 552 553

15

556 557 558 559 560 561 562 563 564 565 566

enzymatic activities of the mutated enzyme to examine the role of this putative redox-active cysteine residue (Table 4). The observed NADH-dependent oxidase and peroxidase activities (9.90 U/mg and 4.02 U/mg, respectively, determined from NADH oxidation) were significantly higher than those of the wild-type enzyme (5.62 U/mg and 0.41 U/mg, respectively), whereas the NADPH-dependent activities did not change. Interestingly, the C43A mutant did not show either NAD(P)H: polysulfide oxidoreductase or NAD(P)H: elemental sulfur oxidoreductase activities with NADPH, and in addition, the formation of H2S could not be detected in the NAD(P)H: elemental sulfur oxidoreductase assay with either NADH or NADPH. Although low levels of NADH consumption were observed in the NAD(P)H: polysulfide oxidoreductase and NAD(P)H: elemental sulfur oxidoreductase assays, this was thought to be caused by the catalytic reduction of free FAD

569

added to the reaction mixture. Indeed, when a reaction mixture lacking free FAD was examined, the NADH consumption completely diminished (as shown in the parentheses in Table 4). These results indicated that Cys43 in TK1481 was irrelevant to electron transfer

570

from NAD(P)H to O2 and H2O2 via FADH2, but essential for that to S0 and polysulfide.

567 568

571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592

Growth properties of ∆TK1481 under various cultivation conditions. With the aim of elucidating the physiological functions of TK1481 in T. kodakarensis, a deletant strain ∆TK1481 was constructed as shown in Fig. 2A. At downstream of tk1481, there is a long non-coding region (1,993 bp between tk1481 and tRNA-Arg). The desired chromosomal modification at the tk1481 locus in ∆TK1481 was confirmed by PCR analysis (Fig. 2A), and a cellular deficiency in the corresponding protein was demonstrated by Western blot analysis (Fig. 2B). The growth properties of ∆TK1481 and the parent strain KU216 were then examined in ASW-YT-Pyr (terminal electron acceptor: H+) and ASW-YT-S0 (terminal electron acceptor: S0) media incubated at 85˚C. To investigate the effects of the gene deletion on oxygen resistance, the cells were inoculated into the media with different initial concentrations of dissolved oxygen. One inoculation was performed under anaerobic atmosphere by using transparent media in the presence of resazurin (“anaerobic inoculation”). We had confirmed that pink color of oxidized resazurin was disappeared at dissolved oxygen concentrations less than 1.6-3.1 µM at 25˚C. Another inoculation was done under aerobic atmosphere (“aerobic inoculation”). Dissolved oxygen concentrations of the aerobic media were 206-220 µM at 25˚C. Moreover, the cells were inoculated into the aerobic media after addition of 20 mg/ml of sodium sulfide (“intermediary aerobic inoculation”), in which dissolved oxygen concentrations were determined to be 42-49 µM at 25˚C. Although actual dissolved oxygen concentration of each medium during the cell growth could not be determined within the tightly capped test tubes at high temperature, saturated dissolved oxygen concentration at 85˚C in ASW was 16

593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622

estimated to be 110 µM under atmospheric pressure. Therefore, it can be estimated that the cells after aerobic inoculation were initially under oxygen saturation at the cultivation temperature. Initial dissolved oxygen concentration after intermediary aerobic inoculation at 85˚C was estimated to be roughly half of saturated concentration. As shown in the left panels of Fig. 3, no obvious change was observed between growth of the parent strain KU216 and the deletant ∆TK1481 in ASW-YT-S0 medium, even after aerobic inoculation (Fig. 3E). TK1481 appeared to have only little role in both sulfur reduction and resistance against oxygen during the sulfur-dependent growth of T. kodakarensis. In ASW-YT-Pyr, the cell growth of KU216 after intermediary aerobic inoculation (Fig. 3D) was also similar to that after anaerobic inoculation (Fig. 3B), whereas the cell growth was significantly inhibited after aerobic inoculation (Fig. 3E), most likely due to dissolved oxygen presented at high concentrations in the medium. A maximum specific growth rate of KU216 decreased to 0.23 h-1 after aerobic inoculation from 0.49 h-1 after anaerobic inoculation. T. kodakarensis wild strain KOD1 also showed the same growth properties, indicating that T. kodakarensis was sensitive to high concentration of dissolved oxygen when elemental sulfur was absent from the medium. ∆TK1481 exhibited the similar growth rate to the parent strain in ASW-YT-Pyr after anaerobic and intermediary aerobic inoculation. However, very interestingly, the growth inhibition in the aerobic medium was less significant for ∆TK1481 than for KU216 (Fig. 3F). The maximum specific growth rates of ∆TK1481 were constant within 0.45-0.51 h-1 during the sulfur-independent growth in the presence of various initial dissolved oxygen concentrations. The results suggested that TK1481 participated in oxygen sensitivity rather than defense to oxidative stress in the absence of elemental sulfur. It was confirmed by complementation analysis that the growth properties of ∆TK1481 was not caused by polar effects due to the lesion of tk1481: When ∆TK1481 was transformed by a shuttle vector pLC71 (31)-based recombinant plasmid harboring tk1481 along with the upstream region, the complemented strain showed lower growth ability in ASW-YT-Pyr after aerobic inoculation than that transformed by the empty vector (data not shown). We further constructed another recombinant strain KUTK1481C43A, in which 2-bases mutation corresponding to Cys43Ala was introduced into tk1481 gene on the chromosome.

628

Since the mutant enzyme rTK1481C43A retains oxygen-reducing activities but lost sulfur-reducing activities as described above, investigation of KUTK1481C43A was expected to make clear whether the sulfur-reducing activity was participated in the response to oxygen or not. As the results shown in Fig. 3, KUTK1481C43A showed similar growth properties to KU216 in all cultivation conditions examined; the growth was also inhibited when the cells were inoculated aerobically into ASW-YT-Pyr. This strongly

629

supported that the oxygen sensitivity of T. kodakarensis during the sulfur-independent

623 624 625 626 627

17

630

growth was likely caused by NOX activity of TK1481.

631 632 633 634 635 636 637 638 639 640

Endogenous expression of TK1481 protein in T. kodakarensis. Finally, the endogenous expression of TK1481 protein in T. kodakarensis was examined by Western blot analysis (Fig. 4). The cells of KU216 grown after inoculation under both anaerobic and aerobic conditions were harvested at early-exponential, mid-exponential, and early-stationary phases. Throughout the growth in ASW-YT-Pyr, the amount of TK1481 protein was constitutive and did not vary regardless of the inoculation condition. The constitutive expression of TK1481 was also observed in ASW-YT-S0, although the amount of protein in aerobically inoculated cells was higher when compared to cells inoculated anaerobically.

641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666

18

667

DISCUSSION

668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703

In this study, we examined the catalytic and physiological functions of TK1481, one of the seven NOX homologs in the hyperthermophilic archaeon T. kodakarensis, and focused on its participation in the defense against oxidative stress and its role in sulfur reduction. The recombinant form of TK1481 exhibited both NAD(P)H: polysulfide oxidoreductase and NOX activities as well as low NADH peroxidase and NAD(P)H: elemental sulfur oxidoreductase activities under anaerobic conditions. Among the multiple activities, NAD(P)H: polysulfide oxidoreductase activity with NADPH showed the highest kcat/Km value. The role of the putative redox-active cysteine Cys43 in electron transfer within TK1481 was demonstrated by site-directed mutagenesis. It should be emphasized that the C43A mutant of TK1481 completely lost the reduction activity towards polysulfide and S0, whereas retained the oxidase and anaerobic peroxidase activities (Table 4). These results suggested the following three points. First, the Cys43 residue in TK1481 did not participate in the electron transfer from NAD(P)H to oxygen via FAD as previously demonstrated for H2O-forming NOXs (18, 22). FADH2 formed by the electron transfer from NAD(P)H likely reacted directly with oxygen to generate H2O2. Second, Cys43 did not participate in the reduction of H2O2 to H2O, probably due to the inability of the mutant protein to form Cys43-S-OH with H2O2. This result is inconsistent with the previously reported H2O-forming NOXs and NADH peroxidases in which H2O was formed by reduction of the sulfenic acid intermediate (1, 22, 26). The residual NADH peroxidase activity of rTK1481C43A strongly suggests that electron transfer occurs directly from FADH2 to H2O2 or via other redox-active residues, which would represent a distinct mechanism from conventional NOXs and NADH peroxidases. Finally, Cys43 played a critical role in the reduction of sulfur. During turnover, Cys43-SH is expected to react with sulfur to generate Cys43-S-SH as a redox intermediate, and would then be regenerated by FADH2-mediated reduction with the concomitant evolution of H2S. Therefore, it is estimated that the reactivity of the redox-active cysteine with individual substrates governs the catalytic properties of Group 3 PNDOR members. TK1481 enabled the generation of H2S by reduction of Cys-S-SH formed between Cys43 and sulfur, while the inability to form Cys-S-OH with H2O2 would lead to a deficiency in H2O-forming activity, thereby releasing H2O2 as a final product under aerobic conditions. The high reactivity of Cys43 towards polysulfide would reflect the higher NAD(P)H: polysulfide oxidoreductase activity of TK1481. A few NOXs from hyperthermophiles have been reported to produce both H2O2 and H2O (4, 38), and the generation of dual products may be a result of the partial formation of a sulfenic acid intermediate with H2O2 being formed by the first 19

704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733

reduction of O2. Based on the high NAD(P)H: polysulfide oxidoreductase activity of rTK1481, we initially assumed that the sulfur-reducing ability of TK1481 was essential for sulfur utilization by T. kodakarensis, and therefore expected poor growth of ∆TK1481 on elemental sulfur. However, nearly identical growth properties of ∆TK1481 and KU216 were observed in ASW-YT-S0 (Fig. 3A). The results strongly suggest that TK1481 does not contribute, or only in limited manner, to sulfur-dependent energy conservation. With regard to sulfur reduction in Thermococcales, one NOX homolog in P. furiosus, PF1186, has been recently proposed to be a probable candidate of a sulfur-reducing enzyme (34). When compared to PF1186, TK1481 showed much lower NAD(P)H: elemental sulfur oxidoreductase activity and a lack of CoA-dependency, and was not induced when T. kodakarensis cells were grown on sulfur (Fig. 4). These differing characteristics are consistent with the distinct roles of TK1481 and PF1186. Nevertheless, considering the high reactivity of the redox active Cys43 to polysulfide, we cannot rule out the possibility that TK1481 participates to some extent in the regeneration of NAD(P)+ by reduction of the polysulfide spontaneously formed in the reaction between sulfide and S0. As NADH oxidases are generally thought to function in defense against oxidative stress in both aerobic and anaerobic organisms including anaerobic hyperthermophiles, another functional relevance of TK1481 was predicted to be resistance to oxidative stress. However, no obvious effect of the deletion of TK1481 on growth was observed when the cells were aerobically inoculated into ASW-YT-S0. We observed that a pink color of resazurin in the aerobic medium was more rapidly disappeared in ASW-YT-S0 (30 min) than in ASW-YT-Pyr (2 hours) after the inoculation. That is, dissolved oxygen in the medium was supposed to be removed by sulfide generated from elemental sulfur by the inoculated cells, irrespective of NOX activity within the cells. Interestingly, when elemental sulfur was absent from the medium, the growth properties of ∆TK1481 demonstrated that TK1481 had a function in the cellular sensitivity, not resistance, to dissolved oxygen. Moreover, KUTK1481C43A, in which tk1481 allele encodes a mutant lacking sulfur-reducing activity, also showed the growth inhibition after aerobic inoculation (Fig. 3F). These results supported the idea that the NOX activity of TK1481 most likely participated in the

739

oxidative sensitivity. One possible reason for this phenomenon may be generation of excess H2O2 accompanied by removal of dissolved oxygen by the function of TK1481 in the sulfur-free medium. In several anaerobic hyperthermophiles, including the genera Thermococcus and Pyrococcus, superoxide reductase was proposed to convert the highly toxic superoxide to H2O2 (10), and many NOX homologs also produced H2O2 from dissolved oxygen (4, 17, 28, 38, 41, 42). As these hyperthermophiles did not encode

740

catalase in their genomes, the moderately toxic H2O2 is supposed to be reductively

734 735 736 737 738

20

741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760

removed by proteins with peroxidase activity (25, 27, 39). However, the precise mechanism for H2O2 removal, probably occurred after primary damage repair (35), has been still unclear. We predict that the true function of TK1481 would participate in oxidoreduction of a substance other than molecular oxygen, but during the sulfur-independent growth, the NOX activity of TK1481 may cause growth inhibition by generating excess H2O2 from dissolved oxygen. In deed, it has been reported that T. kodakarensis was more sensitive to oxidative shock induced by addition of H2O2 than P. furiosus (35). On the other hand, Jia et al. recently reported that oxidized hexameric form of NADH oxidase from T. profundus (NoxTP) accelerated aggregation of partially unfolded proteins and degradation of nucleic acids in vitro, so they proposed that the archaeal NADH oxidase may act as an ancestral cell death protein (15). Although TK1481 is not a closely related counterpart of NoxTP (31% identity), archaeal NOXs including TK1481 might generally cause damages to cellular components under oxidative stresses, which might explain oxygen insensitivity of ∆TK1481 when compared to the parent strain possessing the intact gene. The detailed mechanisms of sulfur and oxygen reduction by T. kodakarensis and other related hyperthermophiles are still interesting subjects. As the genome of T. kodakarensis harbors several uncharacterized NOX homologs, including a highly homologous counterpart for P. furiosus NAD(P)H: elemental sulfur oxidoreductase, the examination of the catalytic properties and physiological functions of these homologs is now underway.

761 762 763

ACKNOWLEDGEMENTS

764 765

767

We are grateful to Dr. F. Arisaka of the Department of Bioengineering, Graduate School Bioscience and Biotechnology, Tokyo Institute of Technology, for his kind help with the

768

ultracentrifugation analysis.

766

769 770 771 772 773 774 775 776 777

21

REFERENCES

778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800

1. Argyrou, A., and J. S. Blanchard. 2004. Flavoprotein disulfide reductases: Advances in chemistry and function. Prog. Nucleic Acid Res. Mol. Biol. 78:89-142. 2. Atomi, H., T. Fukui, T. Kanai, M. Morikawa, and T. Imanaka. 2004. Description of Thermococcus kodakaraensis sp. nov., a well studied hyperthermophilic archaeon previously reported as Pyrococcus sp. KOD1. Archaea 1:263-7. 3. Beinert, H. 1983. Semi-micro methods for analysis of labile sulfide and of labile sulfide plus sulfane sulfur in unusually stable iron-sulfur proteins. Anal. Biochem. 131:373-8. 4. Case, C. L., J. R. Rodriguez, and B. Mukhopadhyay. 2009. Characterization of an NADH oxidase of the flavin-dependent disulfide reductase family from Methanocaldococcus jannaschii. Microbiology 155:69-79. 5. Cole, J., and J. Hansen. 1999. Analytical ultracentrifugation as a contemporary biomolecular research tool. J. Biomol. Tech. 10:163-76. 6. Dietrich, W., and O. Klimmek. 2002. The function of methyl-menaquinone-6 and polysulfide reductase membrane anchor (PsrC) in polysulfide respiration of Wolinella succinogenes. Eur. J. Biochem. 269:1086-95. 7. Dym, O., and D. Eisenberg. 2001. Sequence-structure analysis of FAD-containing proteins. Protein Sci. 10:1712-28. 8. Fukui, T., H. Atomi, T. Kanai, R. Matsumi, S. Fujiwara, and T. Imanaka. 2005. Complete genome sequence of the hyperthermophilic archaeon Thermococcus kodakaraensis KOD1 and comparison with Pyrococcus genomes. Genome Res. 15:352-63.

802

9. Fukushima, E., Y. Shinka, T. Fukui, H. Atomi, and T. Imanaka. 2007. Methionine sulfoxide reductase from the hyperthermophilic archaeon Thermococcus kodakaraensis,

803

an enzyme designed to function at suboptimal growth temperatures. J. Bacteriol.

804

189:7134-44. 10. Grunden, A. M., F. E. Jenney, Jr., K. Ma, M. Ji, M. V. Weinberg, and M. W.

801

805 806 807 808 809 810 811 812 813 814

Adams. 2005. In vitro reconstitution of an NADPH-dependent superoxide reduction pathway from Pyrococcus furiosus. Appl. Environ. Microbiol. 71:1522-30. 11. Harris, D. R., D. E. Ward, J. M. Feasel, K. M. Lancaster, R. D. Murphy, T. C. Mallet, and E. J. Crane, 3rd. 2005. Discovery and characterization of a Coenzyme A disulfide reductase from Pyrococcus horikoshii. Implications for this disulfide metabolism of anaerobic hyperthermophiles. FEBS J. 272:1189-200. 12. Higuchi, M., M. Shimada, Y. Yamamoto, T. Hayashi, T. Koga, and Y. Kamio. 1993. Identification of two distinct NADH oxidases corresponding to H2O2-forming oxidase and H2O-forming oxidase induced in Streptococcus mutans. J. Gen. Microbiol. 22

815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851

139:2343-51. 13. Jenney, F. E., Jr., M. F. Verhagen, X. Cui, and M. W. Adams. 1999. Anaerobic microbes: oxygen detoxification without superoxide dismutase. Science 286:306-9. 14. Jia, B., S. C. Park, S. Lee, B. P. Pham, R. Yu, T. L. Le, S. W. Han, J. K. Yang, M. S. Choi, W. Baumeister, and G. W. Cheong. 2008. Hexameric ring structure of a thermophilic archaeon NADH oxidase that produces predominantly H2O. FEBS J. 275:5355-66. 15. Jia, B., S. Lee, B. P. Pham, Y. S. Cho, J. K. Yang, H. S. Byeon, J. K. Kim, and G. W. Cheong. 2010. An archaeal NADH oxidase causes damage to both proteins and nucleic acids under oxidative stress. Mol. Cells 29:363-71. 16. Kawakami, R., H. Sakuraba, H. Tsuge, S. Goda, N. Katunuma, and T. Ohshima. 2005. A second novel dye-linked L-proline dehydrogenase complex is present in the hyperthermophilic archaeon Pyrococcus horikoshii OT-3. FEBS J. 272:4044-54. 17. Kengen, S. W., J. van der Oost, and W. M. de Vos. 2003. Molecular characterization of H2O2-forming NADH oxidases from Archaeoglobus fulgidus. Eur. J. Biochem. 270:2885-94. 18. Lountos, G. T., R. Jiang, W. B. Wellborn, T. L. Thaler, A. S. Bommarius, and A. M. Orville. 2006. The crystal structure of NAD(P)H oxidase from Lactobacillus sanfranciscensis: insights into the conversion of O2 into two water molecules by the flavoenzyme. Biochemistry 45:9648-59. 19. Ma, K., and M. W. Adams. 1994. Sulfide dehydrogenase from the hyperthermophilic archaeon Pyrococcus furiosus: a new multifunctional enzyme involved in the reduction of elemental sulfur. J. Bacteriol. 176:6509-17. 20. Ma, K., and M. W. Adams. 2001. Ferredoxin:NADP oxidoreductase from Pyrococcus furiosus. Methods Enzymol. 334:40-5. 21. Ma, K., R. Weiss, and M. W. Adams. 2000. Characterization of hydrogenase II from the hyperthermophilic archaeon Pyrococcus furiosus and assessment of its role in sulfur reduction. J. Bacteriol. 182:1864-71. 22. Mallett, T. C., and A. Claiborne. 1998. Oxygen reactivity of an NADH oxidase C42S mutant: evidence for a C(4a)-peroxyflavin intermediate and a rate-limiting conformational change. Biochemistry 37:8790-802. 23. Mallett, T. C., J. R. Wallen, P. A. Karplus, H. Sakai, T. Tsukihara, and A. Claiborne. 2006. Structure of coenzyme A-disulfide reductase from Staphylococcus aureus at 1.54 Å resolution. Biochemistry 45:11278-89. 24. Matsumi, R., K. Manabe, T. Fukui, H. Atomi, and T. Imanaka. 2007. Disruption of a sugar transporter gene cluster in a hyperthermophilic archaeon using a host-marker system based on antibiotic resistance. J. Bacteriol. 189:2683-91. 23

852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882

25. Park, S. C., B. P. Pham, L. Van Duyet, B. Jia, S. Lee, R. Yu, S. W. Han, J. K. Yang, K. S. Hahm, and G. W. Cheong. 2008. Structural and functional characterization of osmotically inducible protein C (OsmC) from Thermococcus kodakaraensis KOD1. Biochim. Biophys. Acta 1784:783-8. 26. Parsonage, D., and A. Claiborne. 1995. Analysis of the kinetic and redox properties of NADH peroxidase C42S and C42A mutants lacking the cysteine-sulfenic acid redox center. Biochemistry 34:435-41. 27. Ramsay, B., B. Wiedenheft, M. Allen, G. H. Gauss, C. M. Lawrence, M. Young, and T. Douglas. 2006. Dps-like protein from the hyperthermophilic archaeon Pyrococcus furiosus. J. Inorg. Biochem. 100:1061-8. 28. Reed, D. W., J. Millstein, and P. L. Hartzell. 2001. H2O2-forming NADH oxidase with diaphorase (cytochrome) activity from Archaeoglobus fulgidus. J. Bacteriol. 183:7007-16. 29. Ross, R. P., and A. Claiborne. 1992. Molecular cloning and analysis of the gene encoding the NADH oxidase from Streptococcus faecalis 10C1. Comparison with NADH peroxidase and the flavoprotein disulfide reductases. J. Mol. Biol. 227:658-71. 30. Sambrook, J., and D. W. Russell. 2001. Molecular cloning: a laboratory manual, 3rd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. 31. Santangelo, T. J., L. Cubonová, J. N. Reeve. 2008. Shuttle vector expression in Thermococcus kodakaraensis: contributions of cis elements to protein synthesis in a hyperthermophilic archaeon. Appl. Environ. Microbiol. 74:3099-3104. 32. Sato, T., T. Fukui, H. Atomi, and T. Imanaka. 2003. Targeted gene disruption by homologous recombination in the hyperthermophilic archaeon Thermococcus kodakaraensis KOD1. J. Bacteriol. 185:210-20. 33. Sato, T., T. Fukui, H. Atomi, and T. Imanaka. 2005. Improved and versatile transformation system allowing multiple genetic manipulations of the hyperthermophilic archaeon Thermococcus kodakaraensis. Appl. Environ. Microbiol. 71:3889-99. 34. Schut, G. J., S. L. Bridger, and M. W. Adams. 2007. Insights into the metabolism of elemental sulfur by the hyperthermophilic archaeon Pyrococcus furiosus: characterization of a coenzyme A-dependent NAD(P)H sulfur oxidoreductase. J. Bacteriol. 189:4431-41.

884

35. Strand K. R., C. Sun, T. Li, F. E. Jenney Jr., G. J. Schut, M. W. W. Adams. 2010. Oxidative stress protection and the repair response to hydrogen peroxide in the

885

hyperthermophilic archaeon Pyrococcus furiosus and in related species. Arch. Microbiol.

886

192:447-459. 36. Sugimoto, M., N. Esaki, H. Tanaka, and K. Soda. 1989. A simple and efficient method for the oligonucleotide-directed mutagenesis using plasmid DNA template and

883

887 888

24

889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907

phosphorothioate-modified nucleotide. Anal. Biochem. 179:309-11. 37. Toth, A., M. Takacs, G. Groma, G. Rakhely, and K. L. Kovacs. 2008. A novel NADPH-dependent oxidoreductase with a unique domain structure in the hyperthermophilic Archaeon, Thermococcus litoralis. FEMS Microbiol. Lett. 282:8-14. 38. Ward, D. E., C. J. Donnelly, M. E. Mullendore, J. van der Oost, W. M. de Vos, and E. J. Crane, 3rd. 2001. The NADH oxidase from Pyrococcus furiosus. Implications for the protection of anaerobic hyperthermophiles against oxidative stress. Eur. J. Biochem. 268:5816-23. 39. Weinberg, M. V., F. E. Jenney, Jr., X. Cui, and M. W. Adams. 2004. Rubrerythrin from the hyperthermophilic archaeon Pyrococcus furiosus is a rubredoxin-dependent, iron-containing peroxidase. J. Bacteriol. 186:7888-95. 40. Yang, X., and K. Ma. 2005. Determination of hydrogen peroxide generated by reduced nicotinamide adenine dinucleotide oxidase. Anal. Biochem. 344:130-4. 41. Yang, X., and K. Ma. 2005. Purification and characterization of an NADH oxidase from extremely thermophilic anaerobic bacterium Thermotoga hypogea. Arch. Microbiol. 183:331-7. 42. Yang, X., and K. Ma. 2007. Characterization of an exceedingly active NADH oxidase from the anaerobic hyperthermophilic bacterium Thermotoga maritima. J. Bacteriol. 189:3312-7.

908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925

25

926

LEGENDS TO FIGURES

927 928 929 930 931 932 933 934

Fig. 1. Spectrometric analyses of rTK1481. (A), UV-visible spectrum of the purified rTK1481 (0.7 mg) in 50 mM HEPPS buffer (pH 7.0). (B), change of spectrum of the purified rTK1481 (3.2 mg) by stepwise addition of NADH and oxygen. rTK1481 was incubated at 90°C for 1 min in 50 mM HEPPS buffer (pH 7.0) within a cuvette tightly screw-capped (solid line). NADH was added to the mixture at a final concentration of 100 µM, and incubated at 90°C for 10 min (broken line). Then, the mixture was vigorously stirred at 90°C for 1 min under air atmosphere (dashed line).

935 936

Fig. 2. Schematic drawing of the tk1481 locus on KU216 chromosomes and the

937

corresponding region in ∆TK1481, and confirmation of gene disruption by PCR analysis at the tk1481 loci (A) and Western blot analysis of the translated product (B). TK1480;

938 939 940 941 942 943 944 945

conserved hypothetical protein, TK1481; NAD(P)H oxidase, TK1482; 5'-methylthioadenosine phosphorylase, TK1483; conserved hypothetical protein. (A), Amplification of tk1481 loci in ∆TK1481 (lane 1) and KU216 (lane 2) using the primer set of Dest-TK1481-F and Dest-TK1481-R and the genomic DNA. M represents the DNA size marker. (B), Western blot analysis of cell extracts prepared from ∆TK1481 (lane 1) and KU216 (lane 2) by using the specific antiserum against rTK1481. The cells were grown in ASW-YT-S0 for 12 h. M represents the molecular mass marker of 46.6 kDa.

946 947

Fig. 3. Growth properties of T. kodakarensis strains KU216, ∆TK1481, and

948

KUTK1481C43A. The cells were cultivated at 85˚C after anaerobic inoculation (A and B), intermediary aerobic inoculation (C and D) and aerobic inoculation (E and F). Left (A, C,

949 950 951 952 953

and E) and right (B, D, and F) panels represent the growth in ASW-YT-S0 and ASW-YT-Pyr, respectively. Open circles, parent strain KU216; open squares, ∆TK1481; closed diamond, KUTK1481C43A. The data was represented by mean of multiple experiments (n=9~10 for ASW-YT-S0 or n=15~18 for ASW-YT-Pyr).

954 955

Fig. 4. Endogenous expression of TK1481 protein in T. kodakarensis KU216 grown on

956

960

various conditions. Western blot analysis was performed loading 5 µg protein extracted from cells grown in ASW-YT-Pyr (lanes 1-3) or ASW-YT-S0 (lanes 4-6) with inoculation under anaerobic (A) or aerobic (B) conditions. The cells of early-exponential (lanes 1 and 4), mid-exponential (lanes 2 and 5), and early-stationary phases (lanes 3 and 6) were taken at OD600 of 0.15, 0.3, and 0.45 in ASW-YT-S0, or 0.3, 0.8, and 1.2 in AST-YT-Pyr,

961

respectively. M represents molecular mass marker.

957 958 959

26

Table 1. NAD(P)H oxidase homologs in T. kodakarensis, and the counterparts in Pyrococcus spp. and other hyperthermophiles T. kodakarensis (% identity) TK0304 (100%)

P. furiosus (% identity) PF1532 (84%)

TK0828 (100%)

PF1197 (68%)

TK1299 (100%)

PF1186 (88%)

P. abyssi (% identity) PAB1931 (86%)

P. horikoshii (% identity) PH1509 (86%)

Others (% identity) MJ0649 (43%)

Reported function (Ref) PF1532 H2O2-H2O-forming NADH oxidase (38) MJ0649 H2O2-H2O-forming NADH oxidase from Methanocaldococcus jannaschii (4) PF1197 NADPH: rubredoxin oxidoreductase (10)

PAB0936 (85%)

PH0572 (83%)

TK1481 (100%)

NoxTP (90%) AF0395 (60%)

NpoTn (55%)

PF1186 NAD(P)H: elemental sulfur oxidoreductase (34) PH0572 CoA disulfide reductase (11) NoxTp H2O-forming NADH oxidase from Thermococcus profundus (14, 15) AF0395 H2O2-forming NADH oxidase from Archaeoglobus fulgidus (28) NpoTn

NADH: polysulfide oxidoreductase (putative) from Thermotoga neapolitana (YP_002533859)

PH0890 TK0116 (100%)

PF1245 (66%)

PAB1842 (69%)

PH1363 (70%)

PH1363 α-subunit of dye-linked L-proline dehydrogenase (α4β4) (16)

TK0119 (100%)

PF1795 (75%)

PAB0212 (76%)

PH1749 (75%)

PH1749 α-subunit of dye-linked L-proline dehydrogenase (αβγδ) (16)

TK1392 (100%)

PF2006 (79%)

PAB0184 (75%)

TM1433 (46%)

27

TM1433 α-subunit of H2O2-forming NADH oxidase from Thermotoga maritima (42)

Table 2. Enzymatic activities of rTK1481 NAD(P)H oxidase

NAD(P)H peroxidase a

NAD(P)H: polysulfide oxidoreductase a

NAD(P)H: elemental sulfur oxidoreductase a

Cofactor

NAD(P)H oxidation (U/mg)b

H2O2 formation (U/mg)c

NAD(P)H oxidation (U/mg)b

H2O2 consumption (U/mg)c

NAD(P)H oxidation (U/mg)b

NAD(P)H oxidation (U/mg)b

NADH

5.62 ±0.30 6.33 ±0.23

0.41 ±0.03

0.39 ±0.07

29.8 ±1.50

1.07 ±0.30 0.92 ±0.30

NADPH

0.91 ±0.07 1.01 ±0.12

0.34 ±0.17

0.41 ±0.02

56.8 ±5.52

1.21 ±0.12 0.85 ±0.11

H2S formation (U/mg)d

The procedures for each assay were described in Materials and Methods. The values are represented with standard deviation obtained from at least triplicate measurements. a The reaction was carried out under an anaerobic condition. b Determined by absorbance at 365 nm. c Determined by horseradish peroxidase with N-ethyl-N-(2-hydroxy-3-sulfopropyl)-3-methylaniline sodium salt dehydrate at 555 nm. d Determined by methylene blue method at 670 nm.

28

Table 3. Kinetic properties of rTK1481 Km App (µM)

Vmax App (U/mg)

kcat (min-1)

kcat/Km (min-1/µM)

NADH NADPH

47.1 42.9

10.8 1.38

1987 253.7

44.8 6.28

Polysulfide (50-500)b

NADH NADPH

53.5 64.6

13.7 59.8

H2O2 (10-100)b

NADH NADPH

4.31 15.9

0.722 0.358

Substrate (µM)

Cosubstratea

O2 (35-132)

2531 11017 132.9 65.9

The procedures for each assay were described in Materials and Methods. a NAD(P)H was added at a final concentration of 100 µM. b The reaction was carried out under anaerobic conditions.

29

47.3 170.5 30.8 4.14

Table 4. Enzymatic activities of rTK1481C43A mutant NAD(P)H oxidase Cofactor

NAD(P)H peroxidase a

NAD(P)H: polysulfide oxidoreductase a

NAD(P)H: elemental sulfur oxidoreductase a

H2O2 formation (U/mg)c

NAD(P)H oxidation (U/mg)b

H2O2 consumption (U/mg)c

NAD(P)H oxidation (U/mg)b

NAD(P)H oxidation (U/mg)b

H2S formation (U/mg)d

NADH

9.90 ±0.80 8.29 ±0.94

4.02 ±0.77

3.15 ±0.49

3.33 ±0.28 (N.D.)e

3.06 ±0.33 (N.D.)e

N.D. (N.D.)e

NADPH

0.92 ±0.07 1.06 ±0.13

0.50 ±0.15

0.57 ±0.10

N.D. (N.D.)e

N.D. (N.D.)e

N.D. (N.D.)e

NAD(P)H oxidation (U/mg)b

The procedures for each assay were described in Materials and Methods. The values are represented with standard deviation obtained from at least triplicate measurements. N.D: not detectable. The reaction was carried out under an anaerobic condition. b Determined by absorbance at 365 nm. c Determined by horseradish peroxidase with N-ethyl-N-(2-hydroxy-3-sulfopropyl)-3-methylaniline sodium salt dehydrate at 555 nm. d Determined by methylene blue method at 670 nm. e The values in the parentheses are reaction rates in the absence of free FAD in the reaction mixture. a

30