Characterization of Arthrobacter nicotinovorans HIM, an atrazine-degrading bacterium, from agricultural soil New Zealand

Authors


*Corresponding author. Tel.: +64 7 858 3713; fax: +64 7 858 4964. aislabiej@landcareresearch.co.nz

Abstract

Arthrobacter nicotinovorans HIM was isolated directly from an agricultural sandy dune soil 6 months after a single application of atrazine. It grew in minimal medium with atrazine as sole nitrogen source but was unable to mineralize 14C-ring-labelled atrazine. Atrazine was degraded to cyanuric acid. In addition to atrazine the bacterium degraded simazine, terbuthylazine, propazine, cyanazine and prometryn but was unable to grow on terbumeton. When added to soil, A. nicotinovorans HIM did enhance mineralization of 14C-ring-labelled atrazine and simazine, in combination with naturally occurring cyanuric acid degrading microbes resident in the soil. Using PCR, the atrazine-degradation genes atzABC were identified in A. nicotinovorans HIM. Cloning of the atzABC genes revealed significant homology (>99%) with the atrazine degradation genes of Pseudomonas sp. strain ADP. The atrazine degradation genes were held on a 96 kbp plasmid.

1Introduction

There is widespread use of pesticides in the environment. In New Zealand, triazine herbicides are used to control annual grasses and broad leaf weeds in arable croplands and in forests [1]. A national survey of pesticides in groundwater in New Zealand has revealed that triazines, including atrazine, simazine and terbuthylazine, are detected most often, albeit at trace levels [2]. Once in aquifers, the triazines are persistent [3], hence there is a need to develop pesticide management practices that minimise triazine contamination of groundwater.

Degradation of atrazine and its movement through soils are key factors that influence its potential to contaminate groundwater. A nationwide study was therefore initiated in New Zealand to assess pesticide behaviour in key soils. Experimental sites were established on a range of agricultural soils including a silt loam, a fine sandy loam, a sandy dune soil, and a silty clay derived from basalt. A number of papers were subsequently published that describe pesticide mineralization and leaching activity in these soils [4–6].

There are numerous reports of bacteria and fungi isolated from soil that degrade atrazine (reviewed in [7]). However, despite attempts to isolate atrazine-degrading microbes from New Zealand agricultural soils exhibiting significant levels of atrazine mineralization activity, only one bacterium was obtained. The atrazine-degrading bacterium, Arthrobacter nicotinovorans HIM was isolated directly from an experimental plot on a sandy dune soil, without the need for enrichment, 6 months after a single application of atrazine [6]. Isolation of the bacterium coincided with detection of an increase in mineralization activity in the soil. Furthermore, no atrazine degraders were isolated from the soil prior to this time, from either outside the treatment plot at time 0, or from within the plot at 1 and 3 months after atrazine application [6].

The aim of this paper is to describe A. nicotinovorans HIM, its ability to degrade atrazine and other triazine herbicides used in New Zealand, the genes involved in atrazine degradation, and the effect of the bacterium on atrazine and simazine mineralization activity in soil. To our knowledge this is the first atrazine-degrading bacterium to be described from a New Zealand soil.

2Materials and methods

2.1Source of bacteria

Arthrobacter nicotinovorans HIM was isolated from a sandy dune soil in New Zealand [6]. It is routinely maintained at 25 °C on nitrogen-free minimal agar plates supplemented with sodium citrate (0.1% w/v) as carbon source and 1000 ppm atrazine as nitrogen source [8]. Following growth on R2A agar (Difco) a spontaneous mutant of A. nicotinovorans HIM that does not degrade atrazine was isolated. The derivative, A. nicotinovorans HIM (Atr−), exhibited no characteristic clearing zones on atrazine agar plates. Pseudomonas sp. strain ADP was provided by R. T. Mandelbaum, Volcani Research Center, Bet Dagan, Israel.

2.2Growth of A. nicotinovorans HIM on triazine herbicides

Growth of A. nicotinovorans HIM on trazine compounds was determined in nitrogen-free minimal medium supplied with 1000 ppm of atrazine, simazine or terbuthylazine, or 500 ppm propazine, cyanazine, prometryn or terbumeton, as sole source of nitrogen, with sodium citrate (0.1% w/v) supplied as carbon source. Cells grown on atrazine were inoculated into growth media to give a final concentration of 1.8 × 106 colony forming units ml−1. As the triazine compounds formed a fine insoluble precipitate in culture medium, growth was measured as absorbance at 600 nm in a Shimadzu UV-160 Spectrophotometer once the triazine compounds were degraded and the media had cleared. Cultures that reached an absorbance of >0.2 and showed clearing in the medium were scored as positive. Cultures were incubated in the dark at 25 °C with shaking for 2 weeks. Metabolites were isolated from the cultures grown on atrazine, simazine and terbuthylazine after 14 days incubation as described below.

Mineralization of atrazine by A. nicotinovorans HIM was determined in triplicate in biometer flasks containing 30 ml nitrogen-free medium supplemented with 1000 ppm atrazine as sole nitrogen source, with citrate supplied as carbon source. The flasks were spiked with 3.3 kBq 14C-ring-labelled atrazine (specific activity 910 MBq mmol−1, purity > 95%, Sigma Chemical, St Louis, Mo). Each flask was inoculated with 108 cells grown on atrazine and incubated at 25 °C in the dark with shaking at 200 rpm. 1M KOH was used as CO2 trapping agent and the accumulation of 14CO2 determined using standard methods. Sterile control flasks were also prepared.

2.3Chemical analysis of metabolites

Triazine metabolites were isolated from cultures of A. nicotinovorans HIM grown on atrazine, terbuthylazine or simazine as nitrogen source. Sterile control flasks were also analysed. To extract the metabolites, water was removed from 20 ml of growth medium by rotary evaporation and then the samples were placed under vacuum for 30 min to remove residual water. The samples were derivatized with diazomethane and left overnight at 4 °C to ensure complete methylation. Solvent was removed by rotary evaporation and vacuum, and 5 ml dichloromethane added. The extracts were analysed by gas chromatography-mass spectrometry (GC-MS) on a Hewlett Packard HP 6890 series GC system connected to a HP5973 mass selective detector, and fitted with a Zebron ZB5 column with helium carrier gas. 2 μl of sample was injected. The inlet temperature was 250 °C and the pressure was 12 psi. The oven was held at 60 °C for 30 s and then raised to 295 °C at 15 °C per min, where it was held for 12 min. The mass spectrometer was turned on after a solvent delay of 2 min. Cyanuric acid (Merck-Schuchardt, purity > 98%) was also analysed as a reference compound.

2.4Effect of inoculation with A. nictotinovorans HIM on 14C-atrazine and 14C-simazine mineralization activity in soil

The effect of A. nicotinovorans HIM on atrazine and simazine mineralization in soil was determined by comparing mineralization activity in inoculated soil with uninoculated soils. Himatangi dune sand (0–10 cm) and Waikiwi loamy soil (0–20 cm) both with no history of triazine application, were used for this experiment. The rates of atrazine and simazine mineralization in the soils were measured using methods similar to those described by Sparling et al. [5] for atrazine. Ten grams of soil (dry weight equivalent) was mixed with approximately 18.5 kBq 14C-ring-labelled atrazine or 14C-ring-labelled simazine (573 MBq mmol−1, purity > 95%, Sigma) dissolved in approximately 500 μl of water, to give a final concentration of 0.44 μg radiolabelled atrazine g−1 soil and 0.65 μg radiolabelled simazine g−1 soil. Soil moisture was adjusted to 70% of maximum water-holding capacity and maintained at that level for all mineralization assays. Three soil replicates were inoculated with 5 × 107A. nicotinovorans HIM, grown on atrazine, and mineralization activity compared with untreated soil. Sterile control soil samples in beakers (plus and minus inoculation) were autoclaved twice for 1 h at 121 °C on consecutive days before triazine addition. The beakers of soils were placed in 1 l glass mason jars alongside 10 ml 1 M potassium hydroxide in a small beaker as a CO2 trap. Humidity was maintained by adding 3–4 ml of water into the base of the jar. The jars were incubated statically in the dark at 25 °C for 6 weeks. At regular intervals, CO2 traps were removed, 0.5 ml of the KOH mixed with scintillation cocktail (10 ml Ultima Gold) (Packard), and the radioactivity determined by liquid scintillation counting (LSC). The KOH was replaced at each sampling event. The amount of 14CO2 trapped from the soil, corrected for background radiation levels, was taken as the measure of mineralization of 14C atrazine or 14C simazine.

2.5Purification of cellular DNA

Total cellular DNA from A. nicotinovorans HIM Atr+ and Atr− strains was purified [9]. Briefly, freshly grown cultures were suspended in 567 μl of Tris–EDTA (TE) buffer (10 mM of Tris–Cl (pH 8.0), 1 mM of EDTA] and 30 μl 10% (w/v) sodium dodecyl sulfate and 3 μl of proteinase K (20 mg/ml) (Sigma) and were lysed for 1 h at 37 °C. Next, 100 μl of 5 M NaCl and 80 μl of cetyltrimethylammonium bromide (CTAB)/NaCl were added and incubated for 10 min at 65 °C. DNA was purified by extraction with chloroform–isoamyl alcohol (24:1, v/v), followed by extraction with phenol–chloroform–isoamyl alcohol (25:24:1, v/v). DNA was then precipitated with isopropanol, centrifuged for 5 min at 10,000g, washed with cold 70% (v/v) ethanol, and dried in a DNA SpeedVac (Savant). The dried DNA was resuspended in 25 μl of TE buffer (pH 7.2), and the DNA concentration and purity were measured with a Lambda II spectrophotometer (Perkin–Elmer) at a wavelength of 260 nm, and a ratio of 260/280 nm wavelength readings, respectively.

2.6PCR amplification, cloning and nucleotide sequence analysis

PCR amplification of atzABCD and trzD in both Atr+ and Atr− strains of A. nicotinovorans HIM was conducted using gene-specific oligonucleotide primers described elsewhere [10,11] (Table 1). All primers were custom synthesized by Integrated DNA Technology, Inc., Coralville, IA. Purified genomic DNA (0.2 μg) from A. nicotinovorans was subjected to PCR amplification using the following PCR reaction parameters: 200 μM of each dNTP (Sigma), 1 μM of each of the oligonucleotide primers, 1.5 U of Taq DNA polymerase (Promega) and 1× PCR reaction buffer (50 mM Tris–Cl (pH 8.9), 50 mM KCl and 4 mM of MgCl2]. All PCR amplification reactions were performed in a DNA thermal cycler 2400 (Perkin Elmer, Shelton, CT) with the following PCR cycling parameters: initial denaturation at 98 °C for 3 min, followed by 30 cycles of amplification. Each amplification cycle consisted of denaturation at 98 °C for 30 sec, primer annealing at 58 °C for 40 sec, and primer extension at 72 °C for 40 sec. After amplification, a final extension step was done at 72 °C for 5 min. The amplicons were separated in an agarose gel (1% (w/v)]; stained with ethidium bromide and visualized using a Photoprep I UV transilluminator (Fotodyne) [9]. The gel images were documented using a Kodak digital photo-documentation system DC120 with Kodak 1D 3.5 software (Kodak, New Haven, CT).

Table 1.  List of oligonucleotide primers and sources, amplicon sizes, plasmid profile and nucleotide sequence analysis in A. nicotinovorans HIM and Pseudomonas ADP
Target genePrimer sequencesExpected amplicon size (bp)SourceA. nicotinovorans with 96 kbp plasmid (Atr+)A. nicotinovorans without 96 kbp plasmid (Atr−)Pseudomonas ADP with 96 kbp plasmidAmplicon nucleotide sequence homology with Pseudomonas ADP (%)
  1. +, present; −, absent.

  2. aNot applicable.

atzA5′-CCATGTGAACCAGATCCT-3′500de Souza et al. [10]++99.6
 5′-TGAAGCGTCCACATTACC-3′      
atZB5′-TCACCGGGGATGTCGCGGGC-3′500de Souza et al. [10]++99.5
 5′-CTCTCCCGCATGGCATCGGG-3′      
atzC5′-GCTCACATGCAGGTACTCCA-3′600de Souza et al. [10]++99.8
 GTACCATATCACCGTTTGCCA-3′      
atzD5′-ACGCTCAGATAACGGAGA-3′558Fruchey et al. [11]+NAa
 5′-TGTCGGAGTCACTTAGCA-3′      
trzD5′-CACTGCACCATCTTCACC-3′663Fruchey et al. [11]NAa
 5′-GTTACGAACCTCACCGTC-3′      

The PCR amplified atzABC gene fragments were cloned into a pCR4 plasmid vector using the TOPO™ TA cloning kit (Invitrogen, Inc., Carlsbad, CA) and transformed into competent cells of Escherichia coli (Invitrogen, Inc.). The positive clones were selected by the blue-green/white screening method on an LB agar medium supplemented with kanamycin (50 μg ml−1) and isopropyl-β-d-galactopyranoside (X-gal) (40 μg ml−1) (Sigma). Four white colonies from each of the three cloned genes were selected at random; plasmid DNA was purified using a Qiagen plasmid miniprep kit (Qiagen, Inc., Valencia, CA) and restricted with EcoRI restriction endonuclease [9] to confirm the cloned gene fragments.

Nucleotide sequence of the cloned atzABC genes were performed using an ABI Prizm automated DNA sequencer and a Cycle Sequencing kit with BigDye reagent (Applied Biosystems, Foster City, CA). The partial nucleotide sequence of these three genes were subjected to a NCBI Blast® search and aligned with the atrazine gene sequences that have been reported from other bacteria including Pseudomonas sp. strain ADP.

2.7Analysis of plasmid DNA

Plasmid DNA was purified using a modification of a procedure described previously [12]. Approximately 0.2 g of cell biomass from freshly grown cultures of A. nicotinovorans HIM Atr+ and Atr− and Pseudomonas sp. strain ADP was resuspended in 0.5 ml resuspension buffer (0.3 M sucrose, 25 mM Tris, 25 mM Na2EDTA (pH 8); sterile) supplemented with 2 mg ml−1 lysozyme (Sigma). The cells in the lysis buffer were incubated for 30 min at 37 °C and then three freeze–thaw cycles were performed to weaken the cell wall. The samples were treated with 0.3 ml freshly made lysis solution (0.3 M NaOH, 2% SDS), mixed thoroughly by gentle inversion immediately upon addition, and then incubated at 55 °C for 20 min. The samples were cooled to room temperature and 0.25 ml phenol/chloroform (1:1) added and mixed by inversion. The samples were subjected to centrifugation (10,500g, 5 min) and the supernatant (ca 700 μl of upper layer) transferred to a new tube. The samples were treated with 0.25 ml chloroform and mixed by inverting the tube several times. After centrifugation (10,500g, 5 min) the top aqueous phase was transferred into a new tube. The DNA was then treated with 0.11 volume 3 M Na acetate (pH 4.5), an equal volume of 95% ethanol from the freezer (−20 °C) was added and mixed well by gently inverting the tubes several times. The DNA was precipitated at −80 °C for 30 min. The samples were centrifuged (10,500g, 10 min) and the supernatant discarded. The DNA pellets were washed once with 70% ethanol, centrifuged and dried in a Savant DNA Speedvac. Purified plasmid DNA was resuspended in 50 μl of TE buffer (pH 7.2) and stored at 4 °C. The plasmid DNA was separated on a 0.8% (w/v) horizontal agarose gel using Tris–acetate-EDTA (pH 8.3) buffer [9] at 3 V/cm. The gels were stained, visualized and documented as described previously.

2.8Southern-blot DNA–DNA hybridization

Purified plasmid DNA from A. nicotinovorans (HIM) Atr+ and Atr− strains was separated in three separate agarose gels as described previously. The gels were treated with 0.25 N HCl for 10 min for depurination of DNA followed by 30 min neutralization using a neutralization buffer consisting of 0.5 N NaOH and 1 M NaCl. The DNA was then capillary transferred onto a Zetaprobe™ nylon membrane (Biorad) for 16 h using 0.4 N NaOH as the transfer buffer (BioRad). Following transfer, the membrane was subjected to a pre-hybridization step in a hybridization chamber (Hybaid, Franklin, MA) by treatment with a hybridization buffer (7% (w/v) SDS, 1 mM (w/v) EDTA, 0.5 M (w/v) NaH2PO4 (pH 7.2)] (BioRad) at 65 °C for 15 min.

The probes were prepared from EcoRI (New England BioLab) restriction endonuclease-treated cloned atzA, atzB and atzC gene fragments from the TOPO TA™ cloning plasmid vector [9]. The restricted gene fragments were purified from a low-melt agarose gel (FMC Bioproducts) [9] and radiolabelled using α (32P]dATP (800 Ci mmol) (NEN DuPont) and a nick-translation kit (Amersham) using the procedures described by the manufacturer. The radiolabelled probes were denatured in a boiling water bath for 10 min and quickly transferred to ice prior to use for hybridization.

For hybridization, the prehybridized membranes with the immobilized DNA in hybridization buffer were exposed to each gene probe (300 ng) in three separate reactions and incubated at 65 °C for 16 h in a rotational hybridization chamber. Following hybridization, the membranes were washed using washing buffer (BioRad) and the autoradiography was done for 4–6 h using either a Kodak or a Fuji X-OMAT X-ray film.

3Results and discussion

The isolation of A. nicotinovorans HIM coincided with the detection of an increased rate of atrazine mineralization in soil from an experimental plot on sandy dune soil (Himatangi) 6 months after atrazine application. Three months after application, atrazine mineralization activity in the soil was low, with <5% of 14C-ring-labelled atrazine mineralized to 14CO2, whereas 45.9% was mineralized at 6 months [6]. Furthermore, we were unable to isolate atrazine degraders from the site at time 0, 1 and 3 months after atrazine application. The soil had received a single application of atrazine and had no history of previous application. As A. nicotinovorans HIM was isolated directly from the soil, without the need for enrichment, it is likely that this bacterium plays a role in the degradation of atrazine in situ in the experimental plot from which it was isolated.

3.1Triazine degradation by A. nicotinovorans HIM

A. nicotinovorans HIM grew with atrazine, simazine, terbuthylazine, cyanazine, propazine, and prometryn as nitrogen source but was unable to utilize terbumeton. The herbicides were substituted with Cl except for prometryn and terbumeton where the Cl was replaced with methylthio (–SCH3) and methyoxy (–OCH3) groups, respectively. The bacterium did not mineralize atrazine (data not shown). Following growth on atrazine, simazine or terbuthylazine a sole metabolite accumulated in the broth cultures. When methylated the metabolite had a retention time of 8.5 min and mass spectral analysis revealed a molecular ion (M+, base peak) at m/z 171 and major peaks at m/z 58 and 143. The metabolite was identified as 1,3,5-trimethyl-1,3,5-triazone-2,4,6(1H,3H,5H)-trione, the methylated derivative of cyanuric acid, by comparison with a known standard. Triazines, but no metabolites were detected in sterile control flasks.

Arthrobacter spp. capable of degrading atrazine have been isolated previously. Arthrobacter crystallopoietes was isolated from French agricultural soils using enrichment cultures [13], Arthrobacter aurescens was isolated directly from a spill site in South Dakota with atrazine levels of 29,000 mg kg−1 of soil [14], and Arthrobacter AD1 was enriched from wastewater from a herbicide production facility [15]. Similar to A. nicotinovorans HIM, all these Arthrobacter strains degraded atrazine to cyanuric acid and none of the strains mineralized 14C-ring-labelled atrazine.

Atrazine degraders appear to vary with respect to the range of triazine herbicides they utilize for growth. Pseudaminobacter sp. C147 for example grows on atrazine and simazine, but unlike A. nicotinovorans HIM does not grown on cyanazine, terbuthylazine or the methylthio-substituted prometryn [16]. Ralstonia M19-3 grows on simazine and cyanazine but not propazine, terbutryn and prometryn [17] whereas Agrobacterium radiobacter J14a degraded all triazines tested, specifically atrazine, simazine, propazine, cyanazine, prometon and ametryn [18]. Of the known atrazine degraders, A. aurescens TC1 is reported to be the most metabolically diverse. It degrades many triazine substrates as sole nitrogen source including the herbicides simazine, cyanazine, and ametryn [14], however, the ability of the bacterium to grow on the herbicides terbuthylazine, propazine, prometryn and terbumeton was not reported. In comparison to these strains A. nicotinovorans HIM appears able to grow on a more diverse range of triazine herbicides than Pseudaminobacter sp. C147 and Ralstonia M19-3 but not A. radiobacter J14a. Unlike A. nicotinovorans HIM, A. radiobacter J14a utilized a methoxy-substituted triazine, prometon, for growth.

3.2Effect of A. nicotinovorans HIM on atrazine and simazine mineralization activity in soils

To confirm that A. nicotinovorans HIM does contribute to triazine mineralization in soil we inoculated the bacterium into Himatangi sandy dune soil, from where it was isolated, and Waikiwi loamy soil. The soils were spiked with 14C-ring labelled atrazine or 14C-ring labelled simazine at typical agricultural levels. Both soil samples used for this study have no history of atrazine application.

Following inoculation into Himatangi soil, the maximum percent of atrazine mineralization was 21% after 42 days incubation for soils inoculated with A. nicotinovorans. The maximum extent of simazine mineralization was 12% for inoculated soil (Fig. 1(a)). In soil without inoculation <2.5% atrazine or simazine was mineralized. The extent of mineralization of atrazine and simazine in Waikiwi loamy soil inoculated with A. nicotinovorans HIM was higher than that detected in Himatangi dune sand (Fig. 1). In inoculated Waikiwi loamy soil, 50% and 35%14CO2 accumulated after 42 days incubation with atrazine and simazine, respectively (Fig. 1(b)). In Waikiwi soil, without inoculation, <5% atrazine or simazine was mineralized. Atrazine and simazine mineralization was negligible in sterile soil (with or without inoculation) (<0.05%14CO2) (Fig. 1).

Figure 1.

Effect of A. nicotinovorans HIM on mineralization of 14C-ring-labelled atrazine (closed symbols) and simazine (open symbols) in Himatangi sandy dune soil (a) and Waikiwi loamy soil (b). Lines represent inoculated soil (▪, □), uninoculated soil (▴, ▵) and sterile control soil (♦).

Clearly, inoculation of A. nicotinovorans HIM into soil increased atrazine and simazine mineralization activity. While A. nicotinovorans HIM contains genes that encode degradation of atrazine to cyanuric acid, the soils used in this study must contain indigenous microbes that metabolize cyanuric acid to carbon dioxide for mineralization of 14C-atrazine or 14C-simazine to 14CO2 to proceed. This is not surprising as there are a number of reports that indicate microbial degradation of cyanuric acid is relatively common in soil [19]. The difference in the extent of mineralization of atrazine in the soils may be due to lower numbers of cyanuric acid-degrading bacteria in Himatangi soil compared with Waikiwi soil. Reasons for the lower extent of simazine mineralization compared with atrazine were not clear, but may be due to the lower water solubility of simazine compared with atrazine [20]. In contrast, laboratory studies with Pseudaminobacter C147 demonstrated degradation rates were correlated with the molecular weight of the alkyl groups, hence the degradation rate for simazine was greater than that of atrazine [16]. Like A. nicotinovorans HIM, the atrazine degraders Nocardioides C190, Pseudaminobacter C147 [21] and Chelabacter heintzii Cit1 [22] enhanced atrazine mineralization rates in soil at agricultural levels without the addition of nutrients.

3.3PCR amplification and nucleotide sequence analysis

Genes encoding atrazine degradation enzymes have been characterized including atzABCD in Pseudomonas sp. strain ADP [10,11] and trzN in Nocardioides strain C190 [23]. PCR amplification of atrazine degradation genes in A. nicotinovorans HIM resulted in 500 bp amplicons for atzA and atzB, and a 600 bp atzC. These results are consistent with atzABC in Pseudomonas sp. strain ADP [10] (Fig. 2). Furthermore, nucleotide sequence analysis of the cloned genes revealed significant homology with atrazine degradation genes in Pseudomonas sp. strain ADP with atzA (GenBank Accession No. AY650035) exhibiting 99.6%, atzB (GenBank Accession No. AY650036) 99.5% and atzC (GenBank Accession No. AY650037) 99.8% homology, respectively (Table 1).

Figure 2.

Agarose gel electrophoresis of PCR amplified atzABC and trzD genes in Atr+ (with 96 kbp plasmid) and Atr− (without 96 kbp plasmid) strains of A. nicotinovorans HIM. DNA size marker in Lanes SS. Lanes 1–10, 1, amplification of atzA in Atr+ strain; 2, amplification of atzA in Atr− strain; 3, amplification of atzB in Atr+ strain; 4, amplification of atzB in Atr− strain; 5, amplification of atzC in Atr+ strain; 6, amplification of atzC in Atr− strain; 7, amplification of atzD in Atr+ strain; 8, amplification of atzD in Atr− strain; 9, amplification of trzD in Atr+ strain; 10, amplification of trzD in Atr− strain.

PCR amplification using oligonucleotide primers for atzD or trzD encoding cyanuric acid hydrolase [11] was negative (Fig. 2; Table 1). The absence of atzD or trzD in A. nicotinovorans HIM is consistent with the inability of this strain to mineralize atrazine and the accumulation of cyanuric acid in broth cultures during growth on atrazine.

Similar to A. nicotinovorans HIM, the atzA gene of Arthrobacter AD1 had high nucleotide sequence homology with the atzA gene of Pseudomonas ADP [15]. In contrast, atzA was not detected in A. aurescens[14] and total DNA from A. crystallopoietes contained a gene that hybridized only weakly with an atzA gene probe [13]. These isolates do, however, contain trzN[23,24]. All the reported Arthrobacter atrazine-degrading strains contain atzBC genes with a high degree of homology with those from Pseudomonas sp. strain ADP.

Gene sequences homologous to atzABC from Pseudomonas sp. strain ADP are widespread; they have been reported in atrazine-degrading bacteria from the United States, Canada, France, China [10,13,15,16], and now New Zealand. Furthermore, the report in this paper that atzABC genes from A. nicotinovorans HIM are also highly conserved provides additional support for the suggestion of de Souza et al. [10] that these genes arose recently from a single origin.

3.4Plasmid analysis

The presence of nearly identical atzABC atrazine degradation genes in many genera of bacteria has lead to the suggestion that horizontal gene flow via plasmids and/or transposable elements is involved in the dissemination of atrazine degradation genes [7].

Plasmid analysis of A. nicotinovorans HIM revealed two plasmids, 96 and 53 kbp, similar in size to those reported in Pseudomonas sp. strain ADP [25] (Fig. 3). Plasmid containing cultures exhibited characteristic zones of clearing on nitrogen-free minimal agar supplemented with atrazine as sole carbon and nitrogen source. However, a single A. nicotinovorans HIM (Atr−) colony that failed to produce a clearing zone on atrazine agar plates, exhibited loss of the 96 kbp plasmid, but the 53 kbp plasmid was present (Fig. 3). PCR amplification of atzABC using gene-specific primers also exhibited negative results for A. nicotinovorans HIM (Atr−) (Fig. 2). Subsequent DNA–DNA hybridization exhibited positive detection of all three atrazine degradative genes on the single 96 kbp plasmid of A. nicotinovorans (HIM) (Atr+). No hybridization signal was evidenced on the 53 kbp plasmid present in A. nicotinovorans Atr+ or Atr− strain (Fig. 4).

Figure 3.

Photonegative of an agarose gel electrophoresis of plasmids in A. nicotinovorans HIM and Pseudomonas ADP. Lanes 1, A. nicotinovorans HIM Atr+ strain with both 96 and 53 kbp plasmids; 2, A. nicotinovorans HIM Atr− strain with only 53 kbp plasmid; 3, Pseudomonas ADPAtr+ strain with both 96 and 53 kbp plasmids.

Figure 4.

Southern-blot DNA–DNA hybridization of the atzABC genes on A. nicrotinovorans HIM plasmids. Lanes 1, purified 96 and 53 kbp plasmid DNA from A. nicotinovorans HIM Atr+ strain; 2, blank; 3, 53 kbp plasmid and the absence of the 96 kbp plasmid from A. nicotinovorans HIM Atr− strain. (a) Agarose gel electrophoresis of the A. nicotinovorans HIM Atr+ and Atr− strains; (b) hybridization with the atzA gene as a probe; (c) hybridization with the atzB gene as a probe; (d) hybridization with the atzC gene as a probe.

Plasmid-localized atrazine-degradation genes have been identified in different genera of atrazine-degrading bacteria including Pseudomonas ADP [25], Chelatobacter heintzii, A. crystallopoietes[23], A. aurescens[24] and now A. nicotinovorans (Fig. 4). An exception is Arthrobacter ADI; in this bacterium the atrazine-degradation genes are reported to be on the chromosome [15]. The atrazine-degradation genes are found on plasmids of different sizes with different combinations of atrazine-degradation genes. In A. crystallopoietes and A. aurescens TC1 atzB, and atzC are located on plasmids sized 117 and 350 kbp, respectively [23,24] whereas in Pseudomonas ADP and A. nicotinovorans atzABC are located on a 96 kbp plasmid. The discovery that atrazine degradation genes are flanked by sequences with high sequence identity to known transposons [7,16] has led to speculation that transposons play a role in mobilization of atrazine-degradation genes from one plasmid to another and from one host to another [7]. Furthermore, the plasmid pADP-1 of Pseudomonas sp. strain ADP is self-transmissible to gram-negative bacteria [25]. It is possible therefore that A. nicotinovorans HIM has acquired the atzABC genes through a self-transmissible plasmid. However, mechanisms for transmission of atrazine-degradation genes between gram-positive and gram-negative bacteria and among gram-positive bacteria in soil have not been investigated.

4Concluding statements

The isolation of A. nicotinovorans HIM from an agricultural soil coincided with the detection of enhanced atrazine mineralization activity in the soil 6 months after a single application of atrazine. The direct isolation of this bacterium from soil without the need for enrichment indicates that it may play an important role in atrazine degradation in situ. Enrichment of such bacteria in soil could result in not only enhanced rates of degradation of atrazine, but also enhanced rates of degradation of related triazine herbicides such as simazine and propazine, should they be subsequently applied to the soil, leading to a decrease in efficacy of the herbicides. The genes for atrazine degradation, atzABC, were held on a 96 kbp plasmid in A. nicotinovorans HIM. The atzABC genes were homologous to those isolated from Pseudomonas sp. ADP. The ease of transfer of the atrazine-degradation genes from gram-positive bacteria such as A. nicotinovorans HIM to other bacteria or vice versa, in agricultural soils with typical agricultural levels of triazine herbicides, has yet to be investigated.

Acknowledgements

This work was partially funded by the Foundation for Research, Science and Technology (C09X0017). All technical grade triazine compounds, except cyanazine, were a gift from Orion Crop Protection Ltd. (New Zealand). Cyanazine was provided by AgriSource (New Zealand). We thank Gitika Panicker (University of Alabama at Birmingham) for technical assistance.

Ancillary