Previous Article | Next Article ![]()
Journal of Bacteriology, August 2005, p. 5090-5096, Vol. 187, No. 15
0021-9193/05/$08.00+0 doi:10.1128/JB.187.15.5090-5096.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Ching Shang,2
Romy Chakraborty,2
Sara M. Belchik,1
John D. Coates,2 and
Laurie A. Achenbach1*
Department of Microbiology, Southern Illinois University, Carbondale, Illinois 62901,1 Department of Plant and Microbial Biology, University of California Berkeley, Berkeley, California 945982
Received 3 March 2005/ Accepted 25 April 2005
|
|
|---|
- and ß-subunits of microbial nitrate reductase, selenate reductase, dimethyl sulfide dehydrogenase, ethylbenzene dehydrogenase, and chlorate reductase, all of which are type II members of the microbial dimethyl sulfoxide (DMSO) reductase family. The pcrC gene product was similar to a c-type cytochrome, while the pcrD gene product exhibited similarity to molybdenum chaperone proteins of the DMSO reductase family members mentioned above. Expression analysis of the pcrA gene from Dechloromonas agitata indicated that transcription occurred only under anaerobic (per)chlorate-reducing conditions. The presence of oxygen completely inhibited pcrA expression regardless of the presence of perchlorate, chlorate, or nitrate. Deletion of the pcrA gene in Dechloromonas aromatica abolished growth in both perchlorate and chlorate but not growth in nitrate, indicating that the pcrABCD genes play a functional role in perchlorate reduction separate from nitrate reduction. Phylogenetic analysis of PcrA and other
-subunits of the DMSO reductase family indicated that perchlorate reductase forms a monophyletic group separate from chlorate reductase of Ideonella dechloratans. The separation of perchlorate reductase as an activity distinct from chlorate reductase was further supported by DNA hybridization analysis of (per)chlorate- and chlorate-reducing strains using the pcrA gene as a probe. |
|
|---|
Chlorite dismutase and perchlorate reductase are the only enzymes in the perchlorate reduction pathway that have been isolated and characterized (10, 16, 23, 32), and molecular data are available only for chlorite dismutase (2, 11). The first step in microbial perchlorate reduction is the reduction of perchlorate (ClO4) to chlorite (ClO2) by the perchlorate reductase enzyme. To date, data are available for purified perchlorate reductase from two perchlorate-reducing bacteria, strains GR-1 and perc1ace (16, 23). The GR-1 analysis revealed an oxygen-sensitive periplasmic enzyme that resembled known nitrate and selenate reductases in both subunit and metal composition. Iron, molybdenum, and selenium were the metal constituents of this heterodimeric (
3ß3) perchlorate reductase that was capable of reducing both perchlorate and chlorate to chlorite (16). Similarly, the perchlorate reductase from perc1ace (23) was composed of two subunits, and tryptic peptides obtained from the small subunit exhibited amino acid similarities to reductases, dehydrogenases, and heme proteins. Although the perc1ace tryptic peptide sequences and the N-terminal amino acid sequence of the ß-subunit of the GR-1 perchlorate reductase have been reported (14, 20), the genes encoding this enzyme were not identified for either organism.
While genes encoding a chlorate reductase operon were recently reported for the chlorate-reducing organism Ideonella dechloratans (12, 35), this enzyme was unable to reduce environmentally significant perchlorate. Furthermore, although microbial nitrate reductases also recognize chlorate as a substrate, it is not known whether perchlorate is similarly recognized (30). Even so, growth of dissimilatory nitrate-reducing bacteria cannot be sustained from the gratuitous reduction of chlorate or perchlorate unless some biochemical mechanism, such as chlorite dismutation (10), is present to alleviate the accumulation of toxic chlorite (30). Here we report the first identification and characterization of the genes encoding perchlorate reductase, the distribution of this enzyme among phylogenetically diverse perchlorate-reducing bacteria, and classification of perchlorate reductase as a member of the microbial dimethyl sulfoxide (DMSO) reductase family of molybdenum enzymes.
|
|
|---|
Nucleic acid extraction and mutant construction. Both genomic DNA and RNA were extracted as previously described using a PUREGENE DNA isolation kit (Gentra Systems Inc., Minneapolis, MN) and the RNAwiz reagent (Ambion, Austin, TX), respectively (2). A pcrA mutant of D. aromatica was constructed by replacement of the pcrA gene with a tetracycline resistance cassette as previously described (29). Briefly, a region upstream of the pcrA start codon and a region downstream of the pcrA stop codon were PCR amplified and inserted on either side of a 1.6-kb pBR322 tetracycline resistance cassette that had been cloned into a suicide vector. This construct was used to transform D. aromatica cells; double-recombination mutants in which the pcrA gene on the chromosome had been replaced with the resistance cassette were verified by PCR amplification.
Sequence analysis. DNA sequences obtained from D. agitata lambda library screening (2), as well as the complete genome sequence of D. aromatica, obtained courtesy of the Joint Genome Institute (http://www.jgi.doe.gov), were subjected to BLAST analysis (1). DNA sequence manipulations were performed using the MacVector sequence analysis software for the Macintosh (version 7.0; Oxford Molecular) and the Se-Al sequence alignment editor, v. 1.0 (A. Rambaut, University of Oxford).
Hybridization analyses. Northern blotting was performed using the NorthernMax-Gly glyoxal-based system (Ambion) as previously described (2). For all growth conditions, 5 µg of total RNA was loaded onto a 1% (wt/vol) glyoxal agarose gel. Following RNA transfer, the blot was hybridized at 50°C in Easyhyb hybridiztion solution (Roche Applied Science, Indianapolis, IN) with a digoxigenin-labeled probe corresponding to 436 bp in the 5' half of the D. agitata pcrA gene. This probe was generated via PCR at an annealing temperature of 55°C with the following primers: PR-750F (5'-CGCGAAGGTAGTCAGCATCT-3') and PR-1185R (5'-TCCATCCTGCAACTTGACCT-3'). For DNA slot blotting, genomic DNAs from known DPRB and non-perchlorate-reducing close relatives were blotted as previously described (2). The blot was hybridized at 45°C with the same perchlorate reductase probe used in the Northern blot analysis.
Phylogenetic analysis.
Protein sequences from the
-subunits of known DMSO reductase enzymes were obtained from the GenBank database (3) and aligned with the
-subunit of perchlorate reductase using the CLUSTALW 1.82 program (31). A phylogenetic tree was constructed with the PAUP* v 4.0 program (D. L. Swofford, Sinauer Associates) using distance as the criterion and neighbor joining as the drawing method.
GenBank accession numbers. The GenBank accession numbers for the D. agitata perchlorate reductase genes are as follows: pcrA, AY180108; pcrB, AY953269; pcrC, AY953270; and pcrD, AY953271. The accession numbers for the protein sequences shown in Fig. 2 are as follows: SerB, Q9S1G9; ClrB, P60069; EbdB, CAD58340; DdhB, AAN46633; and NarH, CAD22070. The accession number for the Nitrosomonas europaea cytochrome c554 shown in Fig. 3 is NP_842334. The GenBank accession numbers for the 16S rRNA gene sequences of the organisms shown in Fig. 7 are as follows: D. agitata, AF047462; Rhodocyclus tenuis, D16209; D. aromatica, AY032610; Dechloromonas sp. strain JJ, AY032611; Dechlorospirillum anomolous strain WD, AF170352; Magnetospirillum magnetotacticum, Y10110; Pseudomonas sp. strain PK, AF170358; Pseudomonas stutzeri, U26415; Dechloromarinus chlorophilus strain NSS, AF170359; Azospira suillum, AF170348; Dechloromonas sp. strain LT-1, AY124797; and I. dechloratans, X72724.
![]() View larger version (86K): [in a new window] |
FIG. 2. Amino acid alignment of the ß-subunit of perchlorate reductases (PcrB) from strain GR-1 (St.GR1) (N-terminal sequence only), D. agitata (D.agit), and D. aromatica (D.arom), selenate reductase B (SerB) from T. selenatis (T.sele), chlorate reductase (ClrB) from I. dechloratans (I.dech), ethylbenzene dehydrogenase B (EbdB) from Azoarcus sp. strain EbN1 (A.EbN1), dimethyl sulfide dehydrogenase B (DdhB) from R. sulfidophilum (R.sulf), and nitrate reductase H (NarH) from H. marismortui (H.mars). Light shading indicates amino acids identical to amino acids in both perchlorate reductases. Dark shading indicates conserved cysteine clusters for Fe-S center binding. The numbers below the cysteine residues indicate the associated Fe-S centers.
|
|
View larger version (18K): [in a new window] |
FIG. 3. Amino acid alignment of the -subunit of perchlorate reductase (PcrC) from D. aromatica (D.arom) and cytochrome c554 from N. europaea (N.euro). Shading indicates identical residues, while boldface and underlining indicate residues shown to bind heme in cytochrome c554 (14).
|
![]() View larger version (64K): [in a new window] |
FIG. 7. Slot blot hybridization of genomic DNAs from DPRB and non-perchlorate-reducing close relatives of DPRB using the D. agitata pcrA probe. A total of 250 ng of genomic DNA was loaded for each strain. Row A (left to right): 1, D. agitata; 2, R. tenuis; 3, D. aromatica; 4, Dechloromonas sp. strain JJ. Row B: 1, D. anomolous strain WD; 2, M. magnetotacticum; 3, Pseudomonas sp. strain PK; 4, P. stutzeri. Row C: 1, D. chlorophilus strain NSS; 2, A. suillum; 3, Dechloromonas sp. strain LT-1; 4, I. dechloratans. D. agitata, D. aromatica, D. anomalous, A. suillum, and Dechloromonas sp. strain LT-1are capable of perchlorate reduction; Pseudomonas sp. strain PK, D. chlorophilus strain NSS, and I. dechloratans are capable of only chlorate reduction.
|
|
|
|---|
![]() View larger version (9K): [in a new window] |
FIG. 1. Diagram of the pcrABCD genes of D. agitata and D. aromatica. The N-terminal sequences of PcrA are indicated, and the twin-arginine motif is in boldface type.
|
|
View this table: [in a new window] |
TABLE 1. BLAST analysis of the pcrABCD translation products
|
-subunit of the perchlorate reductase. In addition, the calculated molecular mass of the PcrA subunit is 105 kDa, a value that corresponds well to the 95 kDa predicted for the
-subunit of the purified perchlorate reductase from GR-1 (16).
pcrB.
The inferred amino acid sequence of the 1,002-bp pcrB gene product indicated the presence of four cysteine-rich clusters for Fe-S center binding, a feature shared with ß-subunits of type II DMSO reductase enzymes (Fig. 2). This cysteine organization has been shown to bind one 3Fe-4S center and three 4Fe-4S centers in both dimethyl sulfide dehydrogenase (19) and nitrate reductase (15) ß-subunits. Based on data for the ß-subunit of the Escherichia coli nitrate reductase (13), these Fe-S centers may be responsible for electron transfer to the molybdopterin-containing
-subunit of perchlorate reductase.
The predicted N-terminal amino acid sequences of the D. agitata and D. aromatica PcrB proteins were aligned with the N-terminal sequence of purified PcrB from GR-1 (16). This alignment reinforced the identity of the pcrB gene (Fig. 2). The predicted D. agitata PcrB N terminus contained 10 of the 18 residues and the predicted D. aromatica PcrB sequence contained 16 of the 18 residues reported for the purified perchlorate reductase ß-subunit from strain GR-1. Since no signal sequence was detected, the ß-subunit of perchlorate reductase is likely translocated with the
-subunit in a manner similar to that proposed for selenate reductase (18), dimethyl sulfide dehydrogenase (20), and chlorate reductase (12). The calculated molecular mass of the PcrB subunit was 37 kDa, a value similar to the 40 kDa reported for the ß-subunit of the purified perchlorate reductase from GR-1 (16).
pcrC.
Although a
-subunit was not detected in the enzyme analysis of the perchlorate reductase from strain GR-1, a third cytochrome-type subunit responsible for connecting the reductase to the membrane was believed to have been lost during purification of the enzyme (16). This observation was borne out by our identification of a 711-bp open reading frame immediately downstream of the pcrB gene in both D. aromatica and D. agitata, whose product exhibited sequence similarity to cytochrome c554 from N. europaea (Table 1). Amino acid alignment indicated that PcrC also has the unique tetraheme organization of cytochrome c554 from N. europaea (14) (Fig. 3). The lack of amino acid sequence similarity between PcrC (ca. 25 kDa) and other type II DMSO reductase
-subunits was not surprising due to the overall sequence diversity noted in the SerC, EbdC, DhdC, and ClrC subunits (12).
The ProteinPredict server (http://cubic.bioc.columbia.edu/pp/) indicated that the pcrC translation product is not a membrane-bound protein and therefore cannot link the PcrAB complex to the membrane. However, further analysis of the D. aromatica genome revealed the presence of a NirT-type cytochrome gene downstream of the chlorite dismutase gene. The cytochrome may link the periplasmic PcrABC reductase to the membrane quinol pool (Fig. 4). Membrane-bound NirT-type cytochromes have been shown to shuttle electrons to the periplasmic nitric reductase of P. stutzeri, the Fe3+ and fumarate reductases of Shewanella putrefaciens, and the periplasmic nitrate reductase (21, 27, 28). The model predicted for perchlorate reduction (Fig. 4) differs from the model projected for selenate reduction (21) by replacement of the bc1 complex with a NirT-type cytochrome.
![]() View larger version (49K): [in a new window] |
FIG. 4. Predicted model for electron transfer during (per)chlorate reduction. Electrons from a quinone pool are transferred from the membrane via a NirT-type cytochrome to the PcrABC reductase. While PcrD is absent from the functional enzyme, this protein is predicted to be involved in enzyme assembly.
|
Expression and mutagenesis of pcrA. Both expression analysis and mutagenesis of the pcrA gene verified the identity of the pcrABCD operon. Northern analysis of D. agitata RNA indicated that there was pcrA gene expression in anaerobic perchlorate- and chlorate-grown cultures (data not shown). However, the presence of perchlorate, chlorate, or nitrate was not enough to induce pcrA expression in aerobic cultures and, as such, indicates the ability of oxygen to completely inhibit pcrA expression, as suggested by the previously documented inhibitory effects of oxygen on perchlorate reduction (7, 22).
Functional proof that the pcrA gene is involved in perchlorate reduction was obtained by mutational knockout in D. aromatica, in which insertional inactivation of the pcrA gene with a tetracycline resistance cassette abolished both perchlorate and chlorate reduction (Fig. 5). However, as expected, the D. aromatica pcrA mutant was still able to grow aerobically (data not shown), as well as anaerobically via nitrate reduction, indicating that there are separate metabolic pathways for each electron acceptor (Fig. 5).
![]() View larger version (12K): [in a new window] |
FIG. 5. Anaerobic growth of wild-type D. aromatica and pcrA mutant with nitrate (a), perchlorate (b), or chlorate (c) as the sole electron acceptor. , wild-type growth; , pcrA mutant growth. The data are averages for duplicate incubations.
|
-subunit protein sequences from known microbial DMSO enzymes (20, 21) and from the PcrA sequences resulting from this study, a phylogenetic tree was constructed (Fig. 6), and this tree had a topology similar to that of a DMSO reductase family tree constructed by McEwan and coworkers (21).
![]() View larger version (26K): [in a new window] |
FIG. 6. Unrooted neighbor-joining tree indicating the evolutionary distances in the DMSO reductase family of molybdoenzymes. GenBank accession numbers are indicated after the names.
|
Our analysis indicated that PcrA forms its own monophyletic group in the type II DMSO enzymes and has a common ancestor with E. coli and Bacillus subtilis NarG, I. dechloratans ClrA, Thauera selenatis SerA, Rhodovulum sulfidophilum DdhA, and Azoarcus sp. strain EB1 EbdA. The alignment used for tree construction indicated that PcrA contains the type II DMSO signature motif [HX3CX2CX(n)C] for binding one 4Fe-4S center in domain I (data not shown) (15). Based on the NarG analysis of Jormakka and coworkers, type II DMSO enzymes have also been shown to contain a conserved Asp residue for Mo ion binding (15). This residue is present at position 212 in PcrA.
The tree topology also indicated that PcrA is more closely related to NarG from B. subtilis and E. coli than to ClrA from I. dechloratans, further emphasizing the differences between the perchlorate and chlorate reductases. The distance between the perchlorate and chlorate reductases indicates that they are distinct enzymes, which was supported by our molecular probing of genomic DNAs from perchlorate and chlorate reducers, as well as from close relatives unable to reduce either electron acceptor (Fig. 7). The slot blot analysis resulted in hybridization signals for the pcrA gene from perchlorate reducers alone. No signal was observed for the close relatives or for Pseudomonas sp. strain PK, D. chlorophilus, or I. dechloratans, organisms that are capable of chlorate reduction but not perchlorate reduction (Fig. 7). This finding supports the hypothesis that two distinct metabolic pathways involved in the reduction of these analogous compounds evolved.
From the current study, it is clear that a more complete understanding of perchlorate reduction and other environmentally significant pathways is pivotal for obtaining knowledge applicable to the design of future bioremediation strategies. Perchlorate reductase and other members of the type II DMSO reductase family play a role in a broad range of substrate reductions and oxidations, and differences in various active sites are the major differences between family members. The different active sites indicate that there was a common reductase ancestor which acquired mutations advantageous for utilization of specific substrates. Thus, it is possible that directed mutagenesis of the active sites of DMSO enzymes could lead to creation of novel enzymes useful for biotechnological as well as bioremediation applications.
This work was supported by grant DACA72-00-C-0016 from the U.S. Department of Defense to J.D.C. and L.A.A.
Present address: Department of Biochemistry, University of Missouri-Columbia, Columbia, MO 65211. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»