Abstract
Purified ArsH from Sinorhizobium meliloti exhibits NADPH:FMN-dependent reduction of molecular O2 to hydrogen peroxide and catalyzes reduction of azo dyes. The structure of ArsH was determined at 1.8 Å resolution. ArsH crystallizes with eight molecules in the asymmetric unit forming two tetramers. Each monomer has a core domain with a central five-stranded parallel β-sheet and two monomers interact to form a classical flavodoxin-like dimer. The N- and C-terminal extensions of ArsH are involved in interactions between subunits and tetramer formation. The structure may provide insight in how ArsH participates in arsenic detoxification.
1. Introduction
Arsenic resistance (ars) genes are widespread [1]. ArsH encoded by the S. meliloti ars operon has conserved domains related to the NADPH-dependent flavin mononucleotide reductases [2]. ArsH is widely distributed in bacteria (123 entries), sparsely in fungi (13 entries) and purple sea urchin, Strongylocentrotus purpuratus (3 entries), and, to date, none in archaea or other eukaryotes. ArsH from Yersinia enterocolitica confers resistance to both arsenite and arsenate [3]. Deletion of arsH from Serratia marcesens or S. meliloti results in loss of resistance [2,4]. In contrast, arsH from Acidothiobacillus ferrooxidans or Synechocystis do not appear to confer arsenic resistance [5,6]. Thus, the role of ArsH in arsenic resistance remains unclear. Here we report initial characterization and 1.8 Å resolution crystal structure of ArsH. ArsH is a H2O2-forming NADPH:FMN1 oxidoreductase that also reduces azo dyes, frequent industrial pollutants in waste water [7].
2. Materials and methods
2.1 Cloning, expression and purification
The arsH gene was amplified by polymerase chain reaction from genomic DNA of S. meliloti strain Rm1021, cloned into pET28a (Novagen) and expressed with a six-histidine tag in E. coli strain BL21 (DE3) pLysS cells (Novagen). Histagged ArsH was purified on a ProBond column (Invitrogen). ArsH concentrations were determined from A280 nm (ε = 30,940 M−1 cm−1).
2.2 Identification of prosthetic group
Purified ArsH was boiled for 5 min, centrifuged, and the supernatant analyzed in the dark on a cellulose thin layer plate in 1-butanol:acetic acid:acetone:water (5:1:2:3) [9]. The spots were visualized with ultraviolet light. The amount of FMN in purified ArsH was quantified by absorption at 445 nm using a molar extinction coefficient of 12,020 M−1cm−1. The concentration of FMN in a solution of 2.9 μM ArsH was determined to be 0.3 μM, indicating that only 10% of the ArsH molecules contained FMN.
2.3 Enzyme Assays
ArsH was assayed for NADPH:FMN oxidoreductase activity in 1 ml of 25 mM Tris, 25 mM Bis-Tris Propane (pH 7.0), 0.1 mg/ml bovine serum albumin, 0.2 mM NADPH, 50 μM FMN, 1 mM EDTA and 10 nM ArsH at 37 °C. NADPH oxidation was followed at 340 nm (ε = 6,220 M−1 cm−1). An Amplex® Red Hydrogen Peroxide/Peroxidase Assay Kit (Invitrogen) was used to determine H2O2 formation, as described. [10]. Reduction of the azo dye Ponceau S was performed as described [7].
2.4 Crystallization and solution of the ArsH structure
Crystals of ArsH in 10 mM HEPES (pH 7.2) and 50 mM NaCl were grown at 23 ± 2°C, in hanging drops, from a 1:1 mixture (1 μl of each) of protein (15 mg/ml) with 17% (w/v) PEG 6000, 0.1 M calcium acetate, and 0.1 M sodium cacodylate, pH 6.5. Micro-seeding was necessary for growing large single crystals. Crystals with a maximal size of 0.2–0.4 mm appeared within 3 days. The crystals were frozen in liquid nitrogen after soaking in cryoprotectant buffer containing 35% (w/v) PEG 6000, 0.1 M calcium acetate, 5% glycerol, and 0.1 M sodium cacodylate, pH 6.5, and a native data set was collected under cryogenic conditions at Argonne National Laboratory Advanced Photon Source (ANL-APS) beamline 32-ID-B. The data were indexed and integrated using MOSFLM [11] and scaled and merged using SCALA [12]. Heavy-atom derivatives were prepared by soaking crystals for 40 min in cryoprotectant buffer containing 5 mM ethyl mercuric phosphate. Hg-derivative crystals were flash-cooled and used for data collection with a Rigaku/MSC FR-D rotating-anode X-ray source equipped with an R-AXIS HTC image-plate detector. Data were processed with MOSFLM and SCALA.
The structure was determined by single isomorphous replacement. SnB [13] was used to find the 10 Hg sites. Their positions were entered into PHENIX [14,15] for phasing, density modification. Initial model building led to tracing of about 80% of the residues. Additional residues were built manually using Xfit [16]. Refinement was carried out using Refmac5 [17]. Refinement statistics are shown in Table 1. All figures were created with PyMol [18].
Table 1.
Data collection and refinement statistics
| Crystal data | Native | Hg derivative |
|---|---|---|
| Data collection | ||
| Resolution (Å) | 39.6–1.80 (1.90–1.80) | 36.1–2.50 (2.64–2.50) |
| Space group | P42 | P42 |
| Unit cell parameters (Å) | a = b = 158.53; c = 87.95 | a = b = 158.28; c = 88.21 |
| Total number of reflections | 2,104,522 (227,608) | 458,434 (52,229) |
| Total number of unique reflections | 185,319 (24,428) | 75,309 (10,782) |
| Rmerge (%)a | 8.8 (53.2) | 8.7 (35.3) |
| Completeness (%) | 92.1 (83.5) | 99.5 (97.9) |
| I/σ □(I) | 18.3 (3.1) | 18.0 (4.0) |
| Phasing | ||
| Resolution (Å) | 36.1–2.50 | |
| Figure of merit (after density modification) | 0.25 (0.60) | |
| Refinement | ||
| Resolution (Å) | 39.6–1.80 (1.85–1.80) | |
| I/σ cutoff | 0 | |
| Number of protein atoms | 14,077 | |
| Number of sulfate ions | 7 | |
| Number of water molecules | 1,682 | |
| R/Rfree (%) | 17.3/21.2 (25.3/32.4) | |
| R.m.s.d. bonds (Å) | 0.016 | |
| R.m.s.d. angle (°) | 1.7 | |
| Mean B factor (Å2), protein/ion/water | 22.2/24.0/29.4 | |
Values in parentheses are for highest resolution shell.
Rmerge = Σ|I − <I>|/ΣI, where I is the observed intensity, □ <I> is the statistically weighted average intensity of multiple observations of symmetry-related reflections.
3. Results
3.1 Expression and purification of ArsH
The 241-residue S. meliloti ArsH (NP_385180) was purified as a C-terminally six histidine tagged construct with a mass of 27,890 Da. ArsH was soluble in the cytosol and purified to >95% homogeneity by nickel-affinity chromatography. A mass of 28 kDa, that of an ArsH monomer, was observed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (data not shown). ArsH eluted as a 112 kDa tetramer by size-exclusion chromatography (data not shown).
Purified ArsH is yellow in appearance and exhibits typical flavoprotein absorption spectrum, with major peaks at 372 nm and 453 nm. On thin layer chromatography of the yellow supernatant from a boiled ArsH preparation, a single fluorescent spot was found that migrated at the position of an FMN standard, indicating that FMN is the cofactor. However, only 10% of purified ArsH has bound FMN.
3.2 Enzymatic properties of recombinant S. meliloti ArsH
ArsH exhibited NADPH:FMN oxidoreductase activity with exogenously added FMN, with a Vmax of 70 μmol of NADPH oxidized/min/mg protein. Optimal activity was observed between pH 6.5 and 7.5. The Km for NADPH was 0.1 mM, and the concentration of FMN required for half-maximal activation was 7.5 μM. No activity was detected with NADH, and no requirement for arsenite or arsenate was observed, nor was there any conversion of arsenite into other species. It may be that the conditions for coupling the reaction to arsenicals in vitro have not yet been identified.
NADPH:flavin oxidoreductase from Entamoeba histolytica reduces O2 to H2O2 [19]. Under similar conditions, ArsH produced 20 μmol of H2O2 min−1mg−1 during NADPH oxidation (data not shown). ArsH has weak homology with azoreductases from Bacillus and Staphylococcus species [7,20]/ The ability of ArsH to reduced the azo dye Ponceau S with NADPH and FMN was assayed Reduction of Ponceau S was assayed as the decrease in dye absorbance at 514 nm using an extinction coefficient of 33,470 M−1 cm−1. The rate of reduction by purified ArsH was determined to be 363 nmol of Ponceau S reduced/min/mg protein.
3.3 Overall structure of S. meliloti ArsH
The S. meliloti ArsH structure was solved by single isomorphous replacement and refined to 1.8 Å (PDB ID code 2Q62). Most residues in the final model are well defined in the 2Fo-Fc electron density map. From Ramachandran plots [21], 92.9% of the residues are in the most favored regions, 7.0% in additional allowed regions, and none in disallowed regions.
ArsH crystals belong to space group P42 with eight molecules in a unit cell (molecules A, B, C, D, E, F, G, and H) in two tetramers of approximately 75 × 60 × 70 Å3 each. The tetramers are identical, with RMSD of 0.43 Å between the Cα atoms and only weak interactions between tetramers. A model of the tetramer structure is shown (Fig. 1A). The final model contains residues 8–227 (monomer A), residues 8–229 (monomer B), residues 9–229 (monomer C, E, and H), residues 8–242 (monomer D), residues 9–227 (monomer F), residues 8–226 (monomer G), seven sulfate ions, and 1,682 water molecules.
Fig. 1. X-ray crystal structure of S. meliloti ArsH.

(A) Tetramer form of ArsH. The monomers A, B, C, and D are colored red, green, blue, and yellow, respectively. (B) Ribbon representation of the ArsH monomer with secondary structural units. Helices, strands, and loops are colored cyan, magenta, and yellow, respectively.
Each monomer is a global α/β protein with a single core domain of flavodoxin-like architecture with a central five-stranded parallel β-sheet (strand order β2, β1, β3, β4, and β5) flanked by helices α1 and α5 on one side and three helices (α2–α4) on the other (Fig. 1B). There is a short 310 helix (residues 178 to 180) between α5 and β5. The ArsH secondary structure is consistent with that of other flavoproteins. ArsH has an N-terminal extension with two 310 helices (residues 15 to 17 and residues 36 to 41) and a C-terminal extension with two helices: α6 is almost perpendicular to α5, while α7 is perpendicular to α6. Together these helices form a small domain involved in subunit interactions (see below). There are interactions between the N-terminal and C-terminal extensions of the same monomer. The amide nitrogens of Asn-14 and Asp-22 make hydrogen bonds with the hydroxyls of Tyr-222 and Tyr-217, respectively.
3.4 Interactions between subunits
ArsH is likely a functional tetramer: it is a tetramer in solution, and there are two tetramers in the asymmetric unit. Monomers A and B form a classical flavodoxin-like dimer, while monomers C and D form another dimer. Upon dimer formation, a surface area of 2,300 Å2 is buried, comparable with other flavoproteins such as T1501, a dimeric 20-kDa FMN reductase from Pseudomonas aeruginosa [22]. The ArsH tetramer is a dimer of dimers. The monomers in the tetramer show non-crystallographic 222 symmetry. The buried surface area of the tetramer is 12,000 Å2, or 3,000 Å2 per monomer, compared with 1,150 Å2 per monomer for classical dimers, suggesting that this tetrameric structure is biologically relevant. The N-terminus helices, α1, α4, α5, and the C-terminal helices α6 and α7 are involved in dimer-dimer interaction. The N- and C-terminal extensions of ArsH are extensively involved in interactions between subunits and are essential for tetramer formation. If 30 residues from both the N- and C-termini of ArsH are excluded, it would resemble a classical flavodoxin, and the buried surface area would decrease to 3,600 Å2 or 900 Å2 per monomer. Each monomer interacts with two others to form a tetramer. For example, monomer A interacts with B and D. Molecules A and B interact with each other mainly through helices α3 and α4 to form a classical flavodoxin-like dimer (Fig. 2A). Monomer A also interacts with monomer D to form an interface that not found in the classical flavodoxin dimer. The N-terminal extensions, helices α4, α5, α6, and α7, are involved in interactions (Fig. 2B). Helix α5 is rich in polar residues that form salt bridges or hydrogen bonds with residues from the terminal extensions of a different subunit (Fig. 2C). For example, Glu-202 and Arg-225 from different subunits form a salt bridge.
Fig. 2. Interactions between ArsH subunits.

(A) A flavodoxin-like dimer interface is formed by interactions of helices α3 and α4 from both monomer A and monomer B. Only the helices from one monomer are labeled. (B) A non-flavodoxin-like dimer interface is formed by interactions between monomers A and D through the extensions of helices α4, α5, α6 and α7 of each subunit. (c) Polar interactions (hydrogen bonds or salt bridges) between helix α5 of subunit A and the C-terminal extensions from subunit D. For clarity, only residues involved in polar interactions are shown as sticks. Dashed lines represent polar interactions.
3.5 The putative FMN binding site
T1501 from P. aeruginosa is an FMN-containing dimer that lacks the six-residue flavodoxin fingerprint motif (T/S)XTGXT [22]. ArsH also lacks the flavodoxin motif. The atypical T1501 FMN binding site (residues GSLRSGSYN) is superimposable on ArsH residues GSLRTVSYS, the putative FMN binding site. FMN was not observed in the ArsH structure, most likely due to the low FMN occupancy of the protein. However, in seven of the eight monomers, a sulfate was found in a deep pocket formed by Ser-43, Arg-45, Ser-48, and Tyr-49. The sulfate probably occupies a ribityl phosphate binding site, as observed in the T1501 apoprotein [22]. We predict that ArsH binds FMN similarly (Fig. 3). Most likely the FMN phosphate replaces the sulfate and participates in hydrogen bonding with the side chains of Ser-43, Arg-45, and Ser-48. The isoalloxazine ring of FMN is predicted to interact with residues 107 to 111.
Fig. 3. The FMN binding site.

FMN was modeled into the putative binding site of ArsH by superimposing the main chain atoms of the GSLRSGSYN loop from T1501 onto those of the GSLRTVSYS loop of ArsH (with an RMSD of 0.6 Å).
4. Discussion
We have reported that disruption of arsH in S. meliloti leads to As(III) sensitivity [2], indicating that ArsH is involved in As(III) detoxification. Possibly the ability to generate H2O2 is indirectly involved. H2O2 oxidizes trivalent forms of arsenic to pentavalent forms in vitro [23], so perhaps ArsH-generated H2O2 oxidizes the more toxic arsenite to the less toxic arsenate. Alternately, trivalent arsenicals might serve as a terminal electron acceptor for ArsH, becoming reduced to volatile arsines such as AsH3 or (CH3)3As that escape from cells. In conclusion, ArsH is an atypical flavodoxin with a noncanonical FMN binding site that catalyzes oxidation of NADPH, generating H2O2 and reducing azo dyes. It is a tetrameric dimer of dimers, each of which is similar to a flavodoxin dimer. The mechanism by which ArsH activity is coupled to arsenic detoxification is the subject of current study.
Acknowledgments
This work was supported by National Institutes of Health Grants GM52216 and GM55425. Use of the Advanced Photon Source was supported by the U.S. Department of Energy, Basic Energy Sciences, Office of Science, under Contract No. W-31-109-Eng-38. Use of the LS-CAT Sector 21 was supported by the Michigan Economic Development Corporation and the Michigan Technology Tri-Corridor (Grant 085P1000817).
Abbreviations
- NADPH
reduced nicotinamide adenine dinucleotide phosphate
- FMN
flavin mononucleotide
- NADH
reduced nicotinamide adenine dinucleotide
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Bhattacharjee H, Rosen BP. Arsenic metabolism in prokaryotic and eukaryotic microbes. In: Nies DHS, Simon, editors. Molecular Microbiology of Heavy Metals. Springer-Verlag; Heidelberg/New York: 2007. p. 390. [Google Scholar]
- 2.Yang HC, Cheng J, Finan TM, Rosen BP, Bhattacharjee H. Novel pathway for arsenic detoxification in the legume symbiont Sinorhizobium meliloti. J Bacteriol. 2005;187:6991–7. doi: 10.1128/JB.187.20.6991-6997.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Neyt C, Iriarte M, Thi VH, Cornelis GR. Virulence and arsenic resistance in Yersiniae. J Bacteriol. 1997;179:612–9. doi: 10.1128/jb.179.3.612-619.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Ryan D, Colleran E. Arsenical resistance in the IncHI2 plasmids. Plasmid. 2002;47:234–40. doi: 10.1016/s0147-619x(02)00012-4. [DOI] [PubMed] [Google Scholar]
- 5.Butcher BG, Deane SM, Rawlings DE. The chromosomal arsenic resistance genes of Thiobacillus ferrooxidans have an unusual arrangement and confer increased arsenic and antimony resistance to Escherichia coli. Appl Environ Microbiol. 2000;66:1826–33. doi: 10.1128/aem.66.5.1826-1833.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Lopez-Maury L, Florencio FJ, Reyes JC. Arsenic sensing and resistance system in the cyanobacterium Synechocystis sp. strain PCC 6803. J Bacteriol. 2003;185:5363–71. doi: 10.1128/JB.185.18.5363-5371.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Suzuki Y, Yoda T, Ruhul A, Sugiura W. Molecular cloning and characterization of the gene coding for azoreductase from Bacillus sp. OY1-2 isolated from soil. J Biol Chem. 2001;276:9059–65. doi: 10.1074/jbc.M008083200. [DOI] [PubMed] [Google Scholar]
- 8.Sambrook J, Fritsch EF, Maniatis T. Molecular cloning, a laboratory manual. Cold Spring Harbor Laboratory; New York: 1989. [Google Scholar]
- 9.Schottel JL. The mercuric and organomercurial detoxifying enzymes from a plasmid-bearing strain of Escherichia coli. J Biol Chem. 1978;253:4341–9. [PubMed] [Google Scholar]
- 10.Zhou M, Diwu Z, Panchuk-Voloshina N, Haugland RP. A stable nonfluorescent derivative of resorufin for the fluorometric determination of trace hydrogen peroxide: applications in detecting the activity of phagocyte NADPH oxidase and other oxidases. Anal Biochem. 1997;253:162–8. doi: 10.1006/abio.1997.2391. [DOI] [PubMed] [Google Scholar]
- 11.Powell HR. The Rossmann Fourier autoindexing algorithm in MOSFLM. Acta Crystallographica Section D-Biological Crystallography. 1999;55:1690–1695. doi: 10.1107/s0907444999009506. [DOI] [PubMed] [Google Scholar]
- 12.Potterton E, McNicholas S, Krissinel E, Cowtan K, Noble M. The CCP4 molecular-graphics project. Acta Crystallogr D Biol Crystallogr. 2002;58:1955–7. doi: 10.1107/s0907444902015391. [DOI] [PubMed] [Google Scholar]
- 13.Weeks CM, DeTitta GT, Hauptman HA, Thuman P, Miller R. Structure solution by minimal function phase refinement and Fourier filtering: II. implementation and applications. Acta Cryst. 1994;A50:210–220. doi: 10.1107/s0108767393008992. [DOI] [PubMed] [Google Scholar]
- 14.Adams PD, et al. Recent developments in the PHENIX software for automated crystallographic structure determination. J Synchrotron Radiat. 2004;11:53–5. doi: 10.1107/s0909049503024130. [DOI] [PubMed] [Google Scholar]
- 15.Adams PD, et al. PHENIX: building new software for automated crystallographic structure determination. Acta Crystallogr D Biol Crystallogr. 2002;58:1948–54. doi: 10.1107/s0907444902016657. [DOI] [PubMed] [Google Scholar]
- 16.McRee DE. XtalView/Xfit--A versatile program for manipulating atomic coordinates and electron density. J Struct Biol. 1999;125:156–65. doi: 10.1006/jsbi.1999.4094. [DOI] [PubMed] [Google Scholar]
- 17.Murshudov GN, Vagin AA, Dodson EJ. Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallographica Section D-Biological Crystallography. 1997;53:240–255. doi: 10.1107/S0907444996012255. [DOI] [PubMed] [Google Scholar]
- 18.DeLano WL. The PyMOL user’s manual. DeLano Scientific; San Carlos, CA: 2001. [Google Scholar]
- 19.Bruchhaus I, Tannich E. Purification and molecular characterization of the NAD(+)-dependent acetaldehyde/alcohol dehydrogenase from Entamoeba histolytica. Biochem J. 1994;303 (Pt 3):743–8. doi: 10.1042/bj3030743. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Chen H, Hopper SL, Cerniglia CE. Biochemical and molecular characterization of an azoreductase from Staphylococcus aureus, a tetrameric NADPH-dependent flavoprotein. Microbiology. 2005;151:1433–41. doi: 10.1099/mic.0.27805-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Laskowski RA, Macarthur MW, Moss DS, Thornton JM. Procheck - a program to check the stereochemical quality of protein structures. Journal of Applied Crystallography. 1993;26:283–291. [Google Scholar]
- 22.Agarwal R, Bonanno JB, Burley SK, Swaminathan S. Structure determination of an FMN reductase from Pseudomonas aeruginosa PA01 using sulfur anomalous signal. Acta Crystallogr D Biol Crystallogr. 2006;62:383–91. doi: 10.1107/S0907444906001600. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Aposhian HV, Zakharyan RA, Avram MD, Sampayo-Reyes A, Wollenberg ML. A review of the enzymology of arsenic metabolism and a new potential role of hydrogen peroxide in the detoxication of the trivalent arsenic species. Toxicol Appl Pharmacol. 2004;198:327–35. doi: 10.1016/j.taap.2003.10.027. [DOI] [PubMed] [Google Scholar]
