Previous Article | Next Article ![]()
Applied and Environmental Microbiology, May 2009, p. 3258-3262, Vol. 75, No. 10
0099-2240/09/$08.00+0 doi:10.1128/AEM.02396-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.
,
Department of Civil and Environmental Engineering, University of Washington, 201 More Hall, Seattle, Washington 98195-2700,1 Department of Microbiology, University of Washington, Seattle, Washington 98195-7242,2 Centre for Novel Agricultural Products, Department of Biology, University of York, York YO10 5YW, United Kingdom,3 College of Forest Resources, University of Washington, Seattle, Washington 98195-21004
Received 17 October 2008/ Accepted 28 February 2009
|
|
|---|
|
|
|---|
As yet there is relatively limited information concerning natural rates of microbial RDX degradation or degradation mechanisms; such information is needed to predict or control rates of degradation in the environment. Of the three general pathways for RDX degradation or transformation based on metabolite analysis outlined in the review by Crocker and associates (6), aerobic degradation initiated by XplA is among the better-characterized systems. This enzyme, a novel cytochrome P450 with a fused flavodoxin reductive domain (18, 33), was first identified by Seth-Smith et al. in Rhodococcus rhodochrous 11Y as being encoded by xplA (33). This gene has been identified in 24 bacterial isolates of the Corynebacterineae capable of utilizing RDX as a sole nitrogen source (4, 25, 32-34). While mammalian nitric oxide synthase family enzymes are known to be P450-like enzymes with fused flavodoxin domains, there are very few identified examples of this type of protein fusion among characterized microbial species (2, 15, 18, 24). Subsequent studies by Jackson and associates (18) demonstrated that XplA, in association with an electron-transferring flavodoxin reductase (XplB), functions to efficiently denitrate RDX aerobically to the aliphatic nitramine 4-nitro-2,4-diazabutanal (NDAB) (18). NDAB has been shown to serve as a viable nitrogen source for Methylobacterium sp. strain JS178 (13) and to be degraded by Phanerochaete chrysosporium (11). Thus, complete mineralization often appears to be mediated by multiple microbial populations.
The capacity for microbial degradation of recalcitrant organics, many of which are apparently new to the biosphere as a result of chemical manufacture, is often determined by plasmids and associated mobile genetic elements (36, 39). Plasmids both serve as a reservoir of genetic information and promote metabolic innovation, since their replication is independent of the chromosome and they do not generally encode essential functions. Although it was earlier suggested that genes in Rhodococcus sp. strain DN22 associated with initial steps of RDX degradation are carried by plasmids (5), no direct evidence for an extrachromosomal location was provided. We now show that nearly identical genes for XplA and XplB are carried on plasmids in two phylogenetically and geographically distinct bacterial isolates: Microbacterium sp. strain MA1, isolated from North America (Milan, TN), and Rhodococcus rhodochrous 11Y, isolated from England (33). Thus, these genes are more broadly distributed within the Actinomycetales than previously recognized, and the near identity of gene sequence (6,710 of 6,721 bp) in these divergent genera is indicative of recent plasmid-mediated transfer. Analysis of approximately 52 kbp of sequence near xplA and xplB in strain MA1 revealed closely linked genes for transport and degradation that are flanked by transposable elements, suggesting that plasmid-carried xplA and xplB are part of a larger class I transposable element encoding both transport and degradation of RDX.
|
|
|---|
200 rpm (28°C) before being added to the growth medium (1:100, vol/vol). RDX degradation was monitored by high-pressure liquid chromatography (HPLC), and RDX-degrading bacteria were recovered by repeated colony isolation on 1.5% agar plates containing either the enrichment medium or the complex medium R2A (29). RDX degradation of individual colonies was confirmed by clearing RDX overlay plates (33) and by monitoring RDX loss from broth cultures.
HPLC quantification of RDX.
RDX concentrations in cultures were analyzed with a modular Waters HPLC system consisting of a Waters 717+ autosampler, two Waters 515 HPLC pumps, and a Waters 9926 photodiode array detector. A 4.6- by 250-mm Waters C18 column was used for separation under run conditions similar to those outlined previously (33), with concentration determined based on absorbance at 240 nm. Peak integration and analysis were conducted using the Millennium32 software (Waters, Milford, MA).
Growth of Microbacterium sp. strain MA1 using RDX as a sole nitrogen source.
Growth studies using RDX (approximately 190 µM) as a sole nitrogen source were conducted in triplicate, along with a control flask that was not inoculated, under conditions described previously (33). Cultures were regularly sampled to monitor turbidity (600 nm) and RDX concentration. Samples taken for RDX determination (800 µl) were processed by first removing cells by centrifugation (20,000 x g for 15 min in microcentrifuge) and amending 250 µl of supernatant with 10% (wt/vol) sodium azide to a final concentration of 0.1% (wt/vol) and stored at 4°C until analyzed by HPLC.
DNA extraction.
For genomic DNA extractions, cultures were grown to late exponential phase before harvest. Cells were recovered by centrifugation (10 min at 10,000 x g) and resuspended to approximately 20 mg/ml of sucrose lysis solution (400 mM sucrose, 100 mM EDTA, 100 mM Tris [pH 8.0], 1 mg/ml lysozyme, 120 U/ml mutanolysin). Following an overnight incubation at 37°C with gentle shaking (100 rpm), cells were lysed using a sodium dodecyl sulfate-proteinase K lysis solution by following established protocols (14). This was followed by RNase A (0.5 µg/ml) incubation, phenol chloroform extraction, and DNA precipitation using standard protocols (31). DNA was suspended in Tris-EDTA buffer, and concentration was estimated by measuring A260 using a NanoDrop ND-1000 spectrophotometer (ThermoFisher Scientific, Wilmington, DE).
PCR amplification, cloning, and sequencing.
Sequences for xplA were amplified using xplAF (5'-CCGACGTAACTGTCCTGTTCGGAA-3') and xplAR (5'-CGGGTCCGTCCGCCGGCTGGAAGG-3') as PCR primers as previously described (30). A region of sequence for the R. rhodochrous 11Y flavin adenine dinucleotide/NADH binding domain protein was amplified using dapBF (5'-ATGACGAACATCAGAGCTGTCGT-3') and dapBR (5'-TTACAGTTCTTCGCGCACGATGTA-3') primers, designed for this study. Well-characterized primers for the bacterial 16S rRNA genes (27F and 1492R) were used to recover sequences for phylogenetic analysis (20). Correctly sized amplification products were ligated into the pCR4 vector (Invitrogen, Carlsbad, CA) and transformed with the TOPO-TA cloning kit (Invitrogen). Vector priming sites were used to determine 400 to 1,100 bp of sequence from each end of an insert by two University of Washington sequencing services. Recombinant colonies were submitted directly to High-Throughput Sequencing Solutions (www.htseq.org), or, alternatively, the BigDye, version 3.1, kit (Applied Biosystems, Foster City, CA) was first used to generate product from recombinant plasmid DNA for submission to the sequencing facility maintained by the Department of Biochemistry, University of Washington, Seattle.
PFGE and Southern analysis.
The Bio-Rad CHEF DRII system was used for pulse field gel electrophoresis (PFGE). Cultures of Microbacterium sp. strain MA1 and Rhodococcus rhodochrous 11Y were grown and harvested from late exponential phase growth Luria-Bertani broth. Cell plugs were molded according to the manufacturer's instructions and embedded in 1% SeaKem Gold agarose dissolved in 0.5x Tris-borate-EDTA and run for 24 h at 6 V/cm with a 10- to 100-s switch time ramp at a 120° angle with buffer recirculating at 14°C. The Saccharomyces cerevisiae YNN295 and Lambda ladder markers (Bio-Rad, Hercules, CA) were used as size standards. SYBR green I was used to stain the gel for visualization.
DNA from the pulse field gel was transferred to a Magnacharge membrane (Micron Separations Inc., Westborough, MA) by overnight capillary transfer by the alkaline transfer method (31). PCR-amplified DNA probe hybridization and detection were done with the Gene Images Alkphos Direct labeling and detection system kit (GE Healthcare, Piscataway, NJ) using the CDP-Star chemiluminescent detection reagent (GE Healthcare) by exposing it to Hyperfilm ECL (GE Healthcare).
Fosmid library construction and sequence analysis.
A fosmid library of the Microbacterium sp. strain MA1 DNA was constructed with the pCC1FOS vector from the CopyControl fosmid library production kit and the phage T1-resistant EPI300-T1 Escherichia coli plating strain (Epicentre Biotechnologies, Madison, WI) by following the instructions provided. Approximately 400 fosmid clones were screened for the xplA gene by PCR amplification using the previously described xplAF/xplAR primer set. A subset of the positive clones were selected for shotgun sequence analysis using the TOPO-TA shotgun sequencing kit with pCR4 vector (Invitrogen) by following the manufacturer's instructions.
Vector priming sites were used for initial end sequencing of the shotgun library (as previously described) and for subsequent sequencing of subclones. The Sequencher 4.6 software (Gene Codes Corp., Ann Arbor, MI) was used for initial assembly. Restriction mapping (NotI, KpnI, PvuII, MscI, BamHI, SacI, EcoRI, EcoRV, BsmI, MluI, HindIII, AscI, and DraI) was then used to order contigs and to direct subcloning (data not shown) into the TOPO-pCR4 Zero Blunt vector (Invitrogen) and subsequent sequencing.
The fosmid sequence was submitted to the JCVI Annotation Service for automated annotation using the JCVI prokaryotic annotation pipeline. This service includes gene finding using Glimmer, Blast-Extend-Repraze (BER) searches, hidden Markov model (HMM) searches, transmembrane HMM (TMHMM) searches, SignalP predictions, and AutoAnnotate. All of this information was stored in a MySQL database and associated files, which were downloaded for review and manual annotation using the Manatee manual annotation tool downloaded from SourceForge (manatee.sourceforge.net). Gene predictions were verified using GeneMark.hmm for prokaryotes (version 2.4), with Mycobacterium avium subsp. paratuberculosis as a model organism (22). Coding sequence start sites were subsequently changed as needed. Inverted repeats were queried using the Palindrome software (Institut Pasteur and Ressource Parisienne en Bioinformatique Structurale), distributed by Mobyle.
Phylogenetic analysis.
The ARB software package was used for 16S rRNA gene sequence alignment and tree construction (21). 16S rRNA gene sequences for other RDX-degrading bacteria were downloaded from the NCBI database, and other bacteria used in the alignment and analysis were imported from the Silva database (28). The PHYLIP software package was used to determine bootstrap values by the neighbor-joining method using the Kimura two-parameter model (10, 19).
Nucleotide sequence accession numbers.
The GenBank accession numbers for the MA1 16S rRNA gene sequence and partial plasmid (pMA1) sequence are FJ357539 and FJ577793, respectively.
|
|
|---|
![]() View larger version (35K): [in a new window] |
FIG. 1. Phylogenetic tree of selected RDX-degrading bacteria inferred from 16S rRNA sequence relationships. Phylogenetic relationships of characterized RDX-degrading bacteria that carry xplA and close relatives were inferred by the neighbor-joining method using the Kimura two-parameter model (10, 19). RDX degraders are shown in boldface. GenBank accession numbers are in parentheses.
|
![]() View larger version (31K): [in a new window] |
FIG. 2. Hybridization of xplA gene probe to Microbacterium sp. strain MA1 and Rhodococcus rhodochrous 11Y DNA resolved by PFGE. (A) Sybr green I-stained gel. Lanes 3 to 7, Microbacterium sp. strain MA1; lanes 9 to 12, Rhodococcus rhodochrous 11Y; lanes 1 and 15, S. cerevisiae YNN295 marker; lanes 2 and 14, Lambda ladder. (B) Hybridization with a 403-bp fragment of the xplA gene.
|
![]() View larger version (43K): [in a new window] |
FIG. 3. Distribution of transposases and IS elements in pMA1. (A) The six ORFs associated with transposition in the 52-kbp sequence of pMA1 are shown in relation to xplB and xplA. The apparent metabolic module that xplB and xplA belong to is underlined in green. (B) Two identical ISL3 family elements (ISMA1), each encoding a single transposase (ORFs pMA1.028 and pMA1.040). Imperfect indirect repeat and direct repeat sequences characteristic of ISL3 elements are shown (23). (C) IS21 family element (ISMA2) encoding an ATP binding domain protein (pMA1.037) and an integrase (pMA1.038). Direct and indirect repeat sequences are displayed below. Repeat sequences found throughout the indirect repeats are highlighted in blue. (D) Three IS256 family elements. pMA1.015, pMA1.042, and 161 bp of sequence flanking each share 100% identity and encode a transposase. pMA1.035 is a truncated gene whose product has an incomplete DDE motif but that shares 100% nucleotide identity with portions of pMA1.015 and pMA1.042.
|
|
|
|---|
Our analysis of a 52-kbp region of the Microbacterium plasmid sequence also suggests that xplA and xplB may be part of a larger gene cluster (pMA1.029 to pMA1.034) associated with RDX degradation. In addition to xplA and xplB, the gene cluster includes a gene highly similar to E. coli genes encoding a general aromatic amino acid permease (aroP), an flavin adenine dinucleotide/NADP binding domain protein, an aldehyde dehydrogenase domain protein, and an acetyl-coenzyme A synthetase homolog. The proximity of the aroP gene to xplA suggests a potential role in the cellular uptake of RDX. The remaining genes in the cluster are less likely to be directly involved in RDX degradation, as in vitro experiments have shown XplB and XplA are capable of breaking down RDX (18). However, a formaldehyde dehydrogenase gene (pMA1.057) located on pMA1 outside the described gene cluster could aid the cell through removal of the toxin formaldehyde, an identified degradation product along with nitrite and NDAB in experiments conducted in vitro with XplA and XplB (18) and in the xplA-bearing isolates R. rhodochrous 11Y, Rhodococcus sp. strain DN22, Williamsia sp. strain KTR4, (18) and Gordonia sp. strain KTR9 (11, 12, 33, 34).
The xplA gene has been found in almost every bacterial isolate that aerobically uses RDX as a nitrogen source (18, 25, 32-34) and has been recovered from RDX-contaminated soils (P. Andeer et al., unpublished observations), suggesting that this gene should provide a useful monitoring tool in applications of bioremediation. Recognition that XlpA is plasmid encoded and that its gene is likely part of a larger metabolic module carried on a transposable element could provide a foundation for better process control, for example, by promoting environmental conditions that foster its transfer among resident microbial populations. The presence of several IS elements in the vicinity of the xplA gene cluster also suggests that these genes could be readily integrated into different broad-range plasmids for selective transfer to disparate microbial species (7).
We thank Jose de la Torre, Sergey Stolyar, and Nicolas Pinel for their advice and assistance and Helena Seth-Smith for providing the pHSX1 sequence. Also, we thank JCVI for providing the JCVI Annotation Service, which performed the initial automatic annotation and provided the Manatee tool for manual annotation.
Published ahead of print on 6 March 2009. ![]()
Supplemental material for this article may be found at http://aem.asm.org/. ![]()
|
|
|---|
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»