Atrazine degradation previously has been shown to be carried out by individual bacterial species or by relatively simple consortia that have been isolated using enrichment cultures. Here, the degradative pathway for atrazine was examined for a complex 8-membered enrichment culture. The species composition of the culture was determined by PCR-DGGE. The bacterial species included Agrobacterium tumefaciens, Caulobacter crescentus, Pseudomonas putida, Sphingomonas yaniokuyae, Nocardia sp., Rhizobium sp., Flavobacterium oryzihabitans, and Variovorax paradoxus. All of the isolates were screened for the presence of known genes that function for atrazine degradation including atzA,-B,-C,-D,-E,-F and trzD,-N. Dechlorination of atrazine, which was obligatory for complete mineralization, was carried out exclusively by Nocardia sp., which contained the trzN gene. Following dechlorination, the resulting product, hydroxyatrazine was further degraded via two separate pathways. In one pathway Nocardia converted hydroxyatrazine to N-ethylammelide via an unidentified gene product. In the second pathway, hydroxyatrazine generated by Nocardia sp. was hydrolyzed to N-isopropylammelide by Rhizobium sp., which contained the atzB gene. Each member of the enrichment culture contained atzC, which is responsible for ring cleavage, but none of the isolates carried the atzD,-E, or -F genes. Each member further contained either trzD or exhibited urease activity. The enrichment culture was destabilized by loss of Nocardia sp. when grown on ethylamine, ethylammelide, and cyanuric acid, after which the consortium was no longer able to degrade atrazine. The analysis of this enrichment culture highlights the broad level bacterial community interactions that may be involved in atrazine degradation in nature.
Atrazine (2-chloro-4-ethylamino-6-isopropylamino-s-triazine) is the most widely used herbicide in North America. In the United States, approximately 35 million kg are applied annually to control broad leaf weeds in resistant crops such as corn, sorghum, pineapple, and sugarcane . The persistence of atrazine in the environment varies for different soil types, and the half-life can range from 60 days to over a year , with an average half life of approximately 150 days . The persistence of atrazine has led to its becoming the second most common ground and surface water contaminant in the United States , and atrazine is often found in drinking water at levels well above the United States Environmental Protection Agency limit of 3 ppb, especially in the midwestern USA. Toxicological studies have suggested that atrazine is an endocrine disruptor in humans and animals . The persistence and toxicity of the transformation products of atrazine is not as well understood, but is also of concern. Many of the chlorinated metabolites such as deethylatrazine (2-chloro-4-amino-6-isopropylamino-s-triazine) have been detected in ground water and may accumulate at concentrations that are greater than that of the residual parent compound [6,7].
Atrazine is degraded by both abiotic and biotic processes. Abiotic processes include photocatalytic reactions  and oxidation via reactions with metal oxides, particularly with manganese oxides . The biotic degradation of atrazine can follow several metabolic pathways that involve stepwise transformations by either single species or by microbial consortia. The most simple pathway involves three genes atzA,-B,-C that encode for hydrolases that dehalogenate and dealkylate atrazine in a stepwise fashion. These genes may be partly or totally plasmid-borne, and can occur within a single organism such as in Pseudomonas sp. strain ADP. In a previously described atrazine degrading bacterial consortium, the chlorohydrolase gene encoded by atzA was reported to occur in a strain of Clavibacter michiganese, and was also capable of N-dealkylating the N-isopropylamine moiety to generate the metabolite N-ethylammelide. Subsequent N-dealklyation of N-ethylammelide and cleavage of the triazine ring was thereafter accomplished by another bacterium Pseudomonas sp. CN1 . Recently, a novel pathway that is similar to the upper pathway identified in Pseudomonas sp. strain ADP has been described in a Nocardia sp. Both of these bacteria carry genes that encodes atrazine chlorohydrolases: AtzA in Pseudomonas sp. strain ADP and TrzN in Nocardia sp. . These enzymes are both amidohydrolases that share some common amino acid motifs, but which have no significant similarity at the nucleotide level. Following dechlorination, the subsequent steps in the atrazine degradation pathways in these bacteria differ in that Pseudomonas sp. strain ADP dealkylates the ethyl side chain of hydroxyatrazine to form N-isopropylammelide; whereas, Nocardia sp. dealkylates hydroxyatrazine to form N-ethylammelide. Recent sequence analysis of the entire atrazine catabolic plasmid from Pseudomonas sp. ADP has revealed the presence of three previously undiscovered genes that are designated atzD, atzE, and atzF. AtzD has 58% amino acid sequence identity to TrzD from Pseudomonas sp. strain NRRLB-12227. Crude extracts of E. coli expressing atzD cleave the cyanuric acid ring. The genes atzE and atzF apparently encode biuret hydrolase and allophanate hydrolase, respectively.
Recent reports suggest that the genes involved in atrazine biodegradation can be distributed in various combinations among different bacterial species. For example, a Nocardia sp. has been shown to contain the trzN, atzB, and atzC genes . The goal of this research was to examine the full complexity of an atrazine degrading bacterial community in a soil with a long term history of exposure to atrazine. Our results demonstrate the existence of two competing pathways for atrazine degradation, and that the production of atrazine metabolites supports considerable functional redundancy. We suggest that this complex degradation pathway is representative of the several different routes by which this chemical is degraded in nature and may help to explain the accumulation of atrazine metabolites in soils.
Materials and methods
Atrazine (2-chloro-4-ethylamino-6-isopropylamino-1,3,5-s-triazine) was purchased from Chem Service Inc. (West Chester, PA). Hydroxyatrazine (2,6-dihydroxy-4-ethylamino-1,3,5-s-triazine) was a gift from Ciba-Geigy Corp. (Greensboro, NC). N-ethylammelide (2-chloro-4-ethylamino-6-isopropylamino-1,3,5-s-triazine) was a gift from Novartis Crop Protection (Greensboro, NC). Ethylamine and isopropylamine were purchased from Sigma Chemical Co., St. Louis, Mo., and cyanuric acid (1,3,5-triazine-2,4,6-triol) was purchased from Aldrich Chemical Co. (Milwaukee, Wis).
Bacterial cultures and growth conditions
The enrichment culture was isolated from soil with a 15-year history of exposure to atrazine and was the same soil from which an earlier 3-membered bacterial consortium was reported . The enrichment culture was isolated by continuous enrichment culture. A 10 g soil sample was placed in a chemostat (Bio Flow III, New Brunswick Scientific, Edison NJ) containing 500 ml of a nitrogen- and chloride-free minimal salts medium (MS) containing 10 mM K2HPO4, 3 mM NaH2PO4, 1 mM MgSO4 and trace minerals. The bacterial enrichment culture was selected using atrazine (33 μg g−1) as a sole nitrogen source. Carbon was supplied as 1 g/l each of glucose and glycerol. The chemostat was operated as a batch culture for 48 h, after which a flow rate of 250 ml per day was established. The total capacity of the chemostat was 2 l. After 3 dilution volumes, 10 ml samples were removed and placed into 90 ml of sterile medium for maintenance in batch cultures. The culture medium used for monitoring degradation of atrazine and its metabolites contained either atrazine (250 μg g−1), hydroxyatrazine (50 μg g−1), ethylamine (250 μg g−1), isopropylamine (250 μg g−1), or cyanuric acid (250 μg g−1) and glucose (1000 μg g−1) where indicated. Atrazine, hydroxyatrazine and glucose were added prior to autoclaving; whereas, ethylamine, isopropylamine and cyanuric acid were added at room temperature after sterile filtration. Solid media containing noble agar (Difco) were prepared similarly except isopropylamine and ethylamine were added at 50 °C to minimize volatilization. All other metabolites were dissolved in minimal amounts of methanol and added immediately after autoclaving. Control plates with the same amount of methanol added after autoclaving failed to support growth. Cultures were incubated at 28 °C and liquid cultures were aerated at 200 rpm. Transfers to fresh media were performed at 6-day intervals. Atrazine degradation was monitored by HPLC.
Inoculum was prepared from each isolate by growing the cells from a single colony in sterile 50 ml Falcon tubes containing 10 ml of sterile filtered TS broth. Tubes were capped and incubated on a platform shaker at 30 °C and aerated at 200 rpm for 18 h. The cells were then harvested by centrifugation at 6000 rpm for 8 min, washed, and were resuspended in 30 ml of phosphate buffer solution at pH 7.4. Washing was repeated 3 times, after which the cells were resuspended into 10 ml of buffer. 25 μl of each suspension was used to inoculate sterile MSM (nitrogen free) containing 33 ppm atrazine.
The enrichment culture was inoculated into flasks containing atrazine, cyanuric acid, ethylamine, or isopropylamine as sole sources of carbon and/or nitrogen. Dilutions from each medium were streaked onto plates containing the same metabolite. Colonies were repetitively streaked through a series of plates to isolate the individual members. Isolates were tentatively identified by fatty acid methyl ester analysis (Sherlock MIDI, Newark, Del.) to eliminate redundant isolates .
Cells harvested from liquid media (consortium) or from isolation plates were processed using a FastPrep bead beater (Bio 101, Vista, Calif). The total DNA was isolated with a Fast DNA Kit for Soil using the manufacturer’ s protocols (Bio 101). The eluted DNA product was stored at −20 °C. Plasmid extractions were carried out using the MaxiPrep kit from Quiagen.
Total genomic DNA (40 ng) was used as a PCR template. PCR amplification of the Eubacterial V3 region of 16S rRNA gene was performed using primers PRBA338f with a 5′ GC clamp attached and PRUN518r as previously described . PCR products were separated on an acrylamide gel containing a 20–60% linear denaturing gradient . The gels were run using a DCode TM universal mutation detection system (Bio-Rad Laboratories, Hercules, CA), after which they were stained with ethidium bromide and imaged using a computer digital image capture and analysis system (Bio-Rad Quantity One).
Genomic isolate DNA (40 ng) extracts were used as PCR templates. Primers PRA46f and PRUN518r were used to amplify 500 bp regions of the16S rRNA genes . PCR was carried out using the following method: 95 °C for 5 min, 35 cycles at 92 °C for 1 min, 55 °C for 30 s, 72 °C for 1 min, and a final extension step for 6 min at 72 °C. The PCR products were run on 1% agarose gels, after which the bands were cut and the fragments were extracted using a gel extraction kit (Quiagen). The DNA fragments were cloned into plasmids (Promega P-Gem), purified with a miniprep plasmid kit (Quiagen), and sequenced at the Arizona State University DNA Sequencing Facility. The species were tentatively identified using the GenBank database by comparison of the partial sequences with those of previously named accessions .
Detection of atrazine genes
Isolate genomic and plasmid DNA were screened for the presence of atrazine genes using PCR. The PCR primer sets that were used to amplify conserved regions of these genes were from previously published studies or were designed here. The gene sequences included: atzA,-B,-C, atzD (5′ GGTAATGGGCAAGACCGAG and 5′ ACGCCAGTGACGATGAGAG), atzE, (5′ GCCTTCGTTATCACTGCC and 5′ GTGTCTTGTAATACCTTGCCTG), atzF (5′ CAGCAATCTGGGCTTCTAC and 5′ ACTTACAAACGCACCGAAC), trzD (5′ATCCGATGTCCACTTCGTTC and 5′GAATCGTCCAGCATCGTGT), and trzN. The PCR method for trzA and trzD was as follows: 95 °C for 5 min, 35 cycles of 92 °C for 1 min, 55 °C for 30 s, 72 °C for 1 min, and a final extension for 6 min at 72 °C. The same method was used for atzD,-E,-F except the annealing temperature was 58 °C. The atzD,-E,-F primers were generated using GCG primer design software (Accelrys, Burlington, MA). The specificities of the primer sets were checked by amplifying the respective sequences in Pseudomonas sp. strain ADP. The PCR products were separated on a 1% agarose gel. All of the resulting DNA bands were cut from the gel, eluted, sequenced and checked for similarity to published atzA,-B,-C,-D,-E,-F and trzD,-N sequences using the GenBank database .
Cell cultures from each of the isolates were inoculated into a chloride free, nitrogen free minimal salts media containing 3 mM atrazine with or without glucose (1000 ppm). Cultures were incubated under the conditions described above for 10 days. Chloride release was determined by AgNO3 precipitation as previously described .
The presence of urease activity in each isolate was determined by growth of bacterial cells in sterile filtered Urease Broth (Difco). Cell suspensions were placed into flasks containing reconstituted broth medium and were incubated for up to 7 days at 28 °C, during which time the rate of formation of a colored product indicating urease activity was recorded for each isolate.
HPLC. Culture samples were centrifuged at 14,000 rpm for 10 min and the supernatants were collected for HPLC analysis on a Hewlett Packard HP1050 liquid chromatography system. Atrazine was separated from its metabolites on a C18 reversed phase column (Discovery C18 silica based 5 μm, 180 Å 4.6 mm × 15 cm, Supelco, Belfonte, PA) eluted with an acetonitrile and water (60:40) mixture at a flow rate of 1.0 ml min−1. Spectral data was obtained at 215 and 230 nm, and was referenced against a 550 nm signal. Under these conditions, atrazine and its metabolites eluted at 4.1 and 2.6 min, respectively. Separation of the individual metabolites of atrazine employed a previously published HPLC method .
Identification of unknown metabolites was accomplished by LC/MS/MS analysis at the University of California, Riverside Analytical Chemistry Laboratory. An electrospray QTOF mass spectrometer (QTOF Ultima-Global, Micromass, UK) was coupled on-line with a capillary HPLC (Agilent 1100, Hewlett Packard) to perform the analysis. An Agilent 0.5 × 150 mm ZORBAX SB-C18 column (5-μm particle diameter, 80 Å pore size) with a gradient mobile phase of A (0.1% formic acid in water) and B (0.1% formic acid in methanol/isopropanol (2/5)) was used, with a linear gradient of 2–65% of mobile phase B over 65 min at a flow rate of 6 l/min. The flow was directly introduced into the electrospray source of the mass spectrometer. 1 μl of sample at a concentration of 30μg/mL in water was injected for each run. QTOF experiments were performed at a capillary voltage of 3.0 V and a cone voltage of 77 V. The source block and desolvation gas (N2) temperatures were 90 and 120 °C, respectively. Collisionally activated dissociation (CAD) experiments were performed in a hexapole collision cell with Ar (10 psi) as the collision gas and the collision energy was optimized at 18 eV. The quadrupole mass filter before the TOF analyzer was set with LM and HM resolution of 10 (arbitrary units), which is approximately equivalent to a 2.0-Da mass window for transmission of precursor ions. The QTOF mass spectrometer was operated in a survey-scan mode for the entire LC run, and the CAD spectra of the top-three abundant ions were acquired approximately every 30 s.
Isolation and characterization of bacterial species
Streak plates yielded eight unique pure cultures. DGGE was used to determine whether the resulting isolates accounted for the complete species composition of the atrazine degrader community. A DGGE gel was generated with one lane containing a 16S rRNA gene fingerprint of the complete enrichment culture with the flanking 8 lanes displaying the 16S rRNA gene fingerprint of each isolate (Fig. 1). The results indicated that the isolates accounted for all members of the enrichment culture. Species identifications were determined by comparison of the partial 16S rRNA sequences with those for accessions in GenBank. The 8 isolates were tentatively identified as Agrobacterium tumefaciens (100% similarity to accession D12784), Caulobacter crescentus (99% similarity to AF125194), Pseudomonas putida (98% similarity to D85999), Sphingomonas yaniokuyae (99% similarity to D16145), Nocardiodes sp. (100% similarity to AF005024), Rhizobium sp. (99% similarity to AF195069), Flavobacterium oryzihabitans (97% similarity to AF288732), and Variovorax paradoxus (97% similarity to D30793).
All members were shown to carry at least one gene from known atrazine degradation pathways. Nocardia sp. contained atrazine chlorohydrolase, trzN, and was the only isolate that contained a chlorohydrolase gene and was able to generate chloride ions during incubation with atrazine in pure culture. Rhizobium sp. was the only bacterium that contained atzB. All of the isolates contained genes showing 99–100% sequence identity to the partial sequence region of atzC that was amplified by PCR. Several isolates were shown to carry trzD (Table 1). Using PCR with nondegenerate primers, we did not detect atzA,-D,-E,-F in any of the isolates or in DNA extracted from the enrichment culture grown on atrazine. These data indicate that TrzN was the only chlorohydrolase present in the bacterial community, and that the genes encoding the cyanuric acid amidohydrolase (atzD), biuret hydrolase (atzE) and allophanate hydrolase (atzF) described in Pseudomonas sp. ATZ were not present in this degrader community.
|trzN (100%)||atzB (100%)||atzC (100%)||trzD (100%)||Ureaseb|
|trzN (100%)||atzB (100%)||atzC (100%)||trzD (100%)||Ureaseb|
aValues shown are sequence similarities based on comparisons of partial gene sequences contained in the PCR amplified regions of these genes; (–) indicates not detected. trzN encodes atrazine chlorohydrolase which converts atrazine to hydroxyatrazine , atzB encodes hydroxyatrazine ethylaminohydrolase which converts hydroxyatrazine to N-isopropylammelide ; atzC encodes N-isopropylammelide isopropyl-aminohydrolase which converts n-ethylammelide to cyanuric acid , and trzD encodes cyanuric acid amidohydrolase which cleaves the triazine ring .
bRelative differences in urease activity was estimated visually by formation of a colored product by cells grown in Difco urea broth for 72 h.
Substrate utilization patterns and metabolite identification
Isolates were plated on to agar media containing either atrazine or the metabolites of atrazine as sources of carbon or nitrogen. Nocardia sp. was the only isolate to grow on atrazine media, which was observed as clearing zones of microcrystalline atrazine particles suspended in the agar medium. Nocardia sp. also could metabolize both atrazine and hydroxyatrazine as sole sources of carbon and nitrogen in liquid medium. When Nocardia sp. was grown on atrazine in liquid medium, it produced an atrazine metabolite that accumulated as a dead end product. This metabolite was analyzed by LC/MS/MS and had a mass ion at m/z 157 (M + 1), and was identified as N-ethylammelide. Rhizobium sp. could grow on both hydroxyatrazine and N-ethylammelide as sole nitrogen sources. LC/MS/MS analysis of media sampled while the enrichment culture was actively metabolizing atrazine revealed the presence of two metabolites that accumulated temporarily in the culture medium. One showed a mass ion at m/z 157 (M + 1) and was identified as N-ethylammelide. The other showed a mass ion at m/z 171 (M + 1) and was identified as N-isopropylammelide. Both metabolites showed mass ions and fragmentation patterns identical to their respective standards and similar to results published in similar studies [11,21]. Lastly, all of the isolates were able to grow on isopropylamine and, with the exception of Nocardia sp., could also metabolize ethylamine and cyanuric acid as sole nitrogen sources.
Of the eight isolates inoculated into liquid media containing atrazine, only Nocardia sp. tested positive for chloride release (data not shown). It was determined that Nocardia sp. released chloride from atrazine in near stoichiometric amounts with the production of hydroxyatrazine. Approximately 2.79 mM Cl− was released from the initial 3 mM concentration of atrazine in solution, resulting in a 93% yield. There was no significant difference in terms of chloride release between the glucose plus and minus treatments, indicating that induction of the chlorohydrolase enzyme was not regulated by use of atrazine as a carbon or nitrogen source for growth.
Results of the urease test showed strong urease activity in A. tumefaciens and F. oryzihabitans, which was indicated by a colorometric change from yellow to dark fushia in the assay medium (Table 1). There was weak urease activity in V. paradoxus, Nocardiodia sp., and Rhizobium sp. which took 7 d for these isolates as compared to a color change in less than 12 h by A. tumefaciens and F. oryzihabitans.
Substrate induced changes in community structure
DGGE results indicated changes in community structure, and differences in the band intensities between lanes were used to ascertain which species functioned in the degradation of different atrazine metabolites (Fig. 2). The degrader community was represented by 8 distinct bands, with several of the 8 species being represented by one or two additional low intensity 16S rRNA gene bands. No species diversity was lost upon transferring the enrichment culture from atrazine to isopropylamine. However, after transferring the enrichment culture from atrazine to ethylamine, the band representing Nocardia was lost from the profile. Similarly, the Nocardia sp. was not maintained when the reconstructed community was grown on cyanuric acid as a sole nitrogen source (Fig. 2). The ability to degrade atrazine was retained when the community was grown on isopropylamine and transferred back to atrazine (data not shown). However, the atrazine degrading activity of the culture could not be recovered after growth on ethylamine or cyanuric acid, emphasizing the crucial role of Nocardia for stable maintenance of the degrader community.
Artificially constructed communities were assembled by addition of each isolate minus one member to culture media containing atrazine to confirm the individual roles of each species in the enrichment culture (Fig. 3). There was no difference in the degradation rates between the enrichment culture and the construct containing all 8 members. The full consortium degraded 100% of the atrazine in approximately 4 days and the rate was more rapid than any of the community constructs missing 1 member. In agreement with the above experimental data in Fig. 2, in which the entire community was transferred between atrazine and various metabolites, the constructed community lacking Nocardia was unable to metabolize atrazine. This confirmed the essential role of Nocardia and requirement for dechlorination of atrazine prior to further use by the other members of the enrichment culture.
Atrazine degradation in soils can be accomplished both by individual bacterial species and by microbial consortia [20,22]. Although several bacterial isolates have been shown to contain all of the genes required for atrazine degradation, such bacteria are relatively rare and are thought to have evolved from consortia . In the atrazine degradation pathways that have been elucidated so far, there are three key enzymatic steps in the upper pathway for atrazine degradation, in which the xenobiotic is first dechlorinated, and then undergoes two dealkylations to yield cyanuric acid . There are several possible ways that these steps can be achieved, especially during removal of the side chains, which may be cleaved with or without the amine nitrogen [21,23,24]. In this enrichment culture, there appeared to be two key species or bottleneck organisms, Nocardia sp. and Rhizobium sp. Nocardia sp. was the only organism in the culture that was capable of dechorinating atrazine to hydroxyatrazine via TrzN, thus initiating the degradation process. Following dechlorination, Nocardia sp. further degraded hydroxyatrazine to N-ethylammelide; however, Nocardia sp. lacked atzB which is currently the only known gene to hydrolyze hydroxyatrazine . The primers that were used to detect atzB were not degenerate; thus it is possible that the specific primers failed to detect similar genes that had slightly varied nucleotide sequences in the region recognized by the PCR primers. Alternatively, Nocardia sp. may carry utilize a novel hydroxyatrazine isopropylamino hydrolase. In either event, more thorough molecular analysis will be necessary to characterize this gene.
The second keystone microorganism identified in this consortium is Rhizobium sp., which is required to catabolize N-ethylammelide. Interestingly, Rhizobium sp. carries atzB and is also able to participate with Nocardia in transformation of hydroxyatrazine. We are still not sure whether atzB also functions in Rhizobium to remove the ethylamine side chain from N-ethylammelide or if this bacterium carries a second unidentified gene encoding for an N-ethylammelide ethylaminohydrolase. Nonetheless, the data suggest that there are 2 upper pathways for the catabolism of atrazine to cyanuric acid that operate within this culture (Fig. 4). The first pathway proceeds in three steps with atrazine being converted to hydroxyatrazine, which is then transformed to N-ethylammelide, and finally to cyanuric acid. The first 2 steps are carried out by Nocardia sp. and the third is carried out by Rhizobium sp. The second pathway proceeds through N-isopropylammelide rather than N-ethylammelide. In this scenario, Nocardia sp. dechlorinates atrazine by TrzN, making hydroxyatrazine available to Rhizobium sp. The hydroxyatrazine is then converted to N-isopropylammelide via AtzB. At this point, all of the members of the enrichment culture members can remove the isopropyl side chain using atzC and subsequently, the pathways converge to cyanuric acid. The operation of these two pathways simultaneously accounts for the presence of both N-ethylammelide and N-isopropylammelide as atrazine degradation metabolites that accumulate temporarily during the degradation process.
With respect to the lower pathway in which cyanuric acid is mineralized into CO2 and NH3, only the trzD was detected in the enrichment culture. It therefore seems likely that cleavage of the triazine ring was carried out by TrzD. The presence of trzD and urease activity suggests that cyanuric acid mineralization proceeds through biuret rather than an allophanate intermediate [25,26]. This is further supported by the absence of atzD,-E,-F, which have been shown to function in the degradation of cyanuric acid through allophonate in Pseudomonas sp. strain ADP . The inability of any single bacterial species to carry out both ring cleavage and to hydrolyze urea at a significant rate means that at least two members of the enrichment culture are required to free nitrogen from the triazine ring, one to cleave the ring, and a second to release ammonia from the urea. Additional bacteria not present in the enrichment culture could also participate in the hydrolysis of biuret to urea .
This atrazine degrading community is one of the most complex degrader communities that has been described to date in that it contains multiple degradation pathways as well as considerable functional redundancy. This poses an interesting question since in enrichment cultures there is generally selection for bacterial consortia only when there is interdependence among the degrader organisms. Many different degrader consortia have been described recently for atrazine and other substrates. A similar example of carbon sharing was observed for 4-chlorosalicylate  in which 3 degrader organisms carried out two separate degradation pathways. In this case, the stability of the latter community was based on the provision of intermediate carbon metabolites and removal of toxic substances. In other cases, consortia may be stabilized by interdependencies based on cross-feeding of growth factors. In the research conducted here, there are still many questions concerning how this particular atrazine degrading community is stably maintained. Nonetheless, degradation rates for the community constructs appeared to vary depending on which members were present or excluded (data not shown). The fastest degradation rate was observed with the full community, with the greatest species diversity. Similar enhancement of degradation has been observed with a linuron degrading community, in which a single isolate was capable of degrading the primary substrate, but was stimulated by synergistic interactions with other members of the culture.
An interesting question that remains to be addressed is whether this enrichment culture is representative of the same bacterial species that carry out atrazine in the soil from which it was isolated or is simply the result of the enrichment process in which a particular group of bacteria formed a stable relationship. A similar enrichment culture from this same soil had earlier revealed a three membered bacterial consortium . In that case, there was a clear sequential degradation pathway that involved the atzA,B,C genes. Thus the discovery of competing pathways in this new atrazine degrading culture suggests that there may be several pathways that operate simultaneously. Soils contain tremendous species diversity, and it is possible that nonculturable bacteria beyond those isolated here by enrichment culture may also participate in different steps in the biodegradation process. A better understanding of the diversity of atrazine degraders in soils and the potential routes for biodegradation of this pesticide will eventually permit better prediction of the potential for mineralization of this compound and accumulation of atrazine metabolites in different soils where this herbicide is used.
This research was supported by a grant from the University of California Toxic Substances Training and Research Program.