Atrazine mineralisation rates in New Zealand soils are affected by time since atrazine exposure.
The herbicide atrazine is used in New Zealand to control annual grasses and broad-leaf weeds in arable croplands and forests. Atrazine (6-chloro-[N.sup.2]-ethyl-[N.sup.4]isopropyl-1,3,5-triazine-2,4-diamine) and its degradation products deethylatrazine (6-chloro-N-(1-methylethyl)-1,3, 5-triazine- 2,4-diamine) and deisopropylatrazine (6-chloro-N-(i-ethyl)-l,3,5-triazine-2,4-diamine) have been detected in the groundwaters of many countries including New Zealand (Close and Rosen 2001). The persistence of atrazine in soil is a key factor influencing its potential to contaminate groundwater.
One of the critical factors affecting the persistence of atrazine is the rate at which it is degraded in soil (Erickson and Lee 1989). The rate of atrazine degradation is dependent on prior exposure of the soils to atrazine (Barriuso and Houot 1996; Radosevich et al. 1996; Ostrofsky et al. 1997; Houot et al. 2000; Rousseaux et al. 2001), and on soil conditions. Generally, the highest rates of atrazine degradation occur in topsoil and decrease with depth (Radosevich et al. 1996). Elevated levels of atrazine degradation have occasionally been detected in subsoils (Vanderheyden et al. 1997; Sparling et al. 1998; Di et al. 2001). Soil environmental factors that influence atrazine degradation include soil temperature and moisture (Radosevich et al. 1996; Sparling et al. 1998; Dinnelli et al. 2000), pH (Vanderheyden et al. 1997; Houot et al. 2000) levels of soil nitrogen and carbon (Assaf and Turco 1994; Alvey and Crowley 1995; Topp et al. 1996; Di et al. 1998), and bioavailability of atrazine (Radosevich et al. 1997).
Bacteria and fungi that degrade atrazine have been isolated from soil. Early reports described microbes that dealkylated atrazine to deethylatrazine and deisopropylatrazine (Erickson and Lee 1989). In the last 10 years, a variety of bacteria that utilise atrazine as a nitrogen source have been reported (Wackett et al. 2002). These bacteria degrade atrazine via hydroxyatrazine; some also mineralise atrazine. Pathways for microbial degradation of atrazine have been determined (Erickson and Lee 1989; de Souza et al. 1998) and genes for bacterial degradation of atrazine described (de Souza et al. 1998). The detection of metabolites of atrazine in soils or groundwaters established microbial degradation as a route for atrazine breakdown (Wackett et al. 2002).
While several studies have reported that atrazine mineralisation is more rapid in soils previously exposed to atrazine (Barriuso and Houot 1996; Radosevich et al. 1996; Ostrofsky et al. 1997; Houot et al. 2000; Rousseaux et al. 2001), the time for establishment and subsequent maintenance of an atrazine-mineralising microbial population in soil has rarely been investigated. The objectives of this study were to monitor temporal changes in atrazine mineralisation activity in 2 soils following atrazine exposure and to enumerate and isolate atrazine degraders from the soils. Experimental plots were established on 2 New Zealand soils: a sandy dune soil (Himatangi) and a silty clay soil (Kiripaka) derived from basalt with no known history of atrazine application. Following application of triazine pesticides to the plots, field samples (topsoil and subsoil) collected at 1-3-monthly intervals over 2 years were analysed in the laboratory for atrazine mineralisation activity and numbers of atrazine degraders. The influence of time since atrazine application on subsequent atrazine mineralisation activity in topsoil and subsoil is related to residual levels of atrazine in the soil and atrazine adsorption isotherms.
Materials and methods
The 2 soils were selected for their contrasting chemical, physical, and mineralogical properties (Table 1) and were analysed using standard methods (Blakemore et al. 1987). The soils used in the study were: Himatangi sand (Grid ref. NZMS 260 $24 049848) (Typic Sandy Recent soil) (Hewitt 1992), an imperfectly drained soil developed in dune sand (Cowie et al. 1967) with low clay content contributing to the low cation exchange capacity; Kiripaka bouldery silty clay (Grid ref. NZMS 260 Q07 176051) (Typic Orthie Allophanic soil) (Hewitt 1992), a well-drained soil developed in strongly weathered basalt (Cox et al. 1983). The Kiripaka soil contains the amorphous clay mineral allophane, commonly a weathering product of volcanic material. Allophane has a very large surface area (700 000-900 000 [m.sup.2]/kg) and variable charge depending on pH. Consequently, the soil has significant anion retention properties.
Establishment of experimental plots, field soil sampling and analysis of residual atrazine
On each soil we established a single experimental plot 15 by 15m, and a mix of chemicals was applied in November (spring) 1999 via a pressurised sprayer at a rate of 2L/[m.sup.2]. The chemical mix included the triazine herbicides atrazine, terbuthylazine ([N.sup.2]_tert_butyl_6_chloro_[N.sup.4]_ethyl- 1,3,5-triazine-2,4-diyldiamine), and hexazinone (3-cyclohexyl-6-dimethylamino- 1 -methyl- 1,3,5-triazine-2, 4(1H,3H)-dione). An additional 2mm of water was applied following herbicide application to wash the chemicals into the soil and minimise evaporation. The target application rate for each of the triazine herbicides was 10 kg/ha of active ingredient, which is approximately 7 times the typical application rates. This rate was used to ensure collection of sufficient data for a companion study on triazine mobilisation into groundwaters. Both plots were established on grazing land with no known history of triazine application. The plots were fenced to exclude stock for the duration of the experiment.
Soil sampling from within the experimental plots was carried out 1 day after herbicide application, and 1,3, 6, 9, 12, 15, 18, and 24 months after application. Additionally, soil samples were collected from outside the plot at the time of herbicide application (time 0) and 12 months after application to the experiment plot. On each sampling date, soil cores (5 cm in diameter) were taken from the plots at 5 random locations, in 10-cm increments, to a depth of 1 m. Topsoil and subsoil fractions, 0-10 and 50-90cm depths, respectively, for Himatangi soil, and 0-20 and 40-80 cm for Kiripaka soil, from each core were bulked, thoroughly mixed, and stored at 4C until analysis. In the Himatangi plot, 5 individual soil cores taken randomly from the topsoil of the plot 18 months after pesticide application were also collected to study the spatial variability of atrazine mineralisation in the laboratory.
The plots were bare of vegetation for approximately 12 months after herbicide application. At this time, weeds became established on the Kiripaka plot, whereas vegetation on the Himatangi plot remained sparse for the duration of the experiment.
Analysis of triazine compounds
Residual levels of triazine in the topsoils and subsoils were determined by solvent extraction and gas chromatography (Roos et al. 1987). Field-wet soil (25-50 g) spiked with an internal standard solution (tetrachlorvinphos in ethylacetate) was extracted with 100 mL ethylacetate. After extraction the samples were filtered through a GF/A microfilter, filtrate concentrated by nitrogen blowdown, and then cleaned up by gel permeation chromatography.
The sample extracts were analysed by gas chromatography with electron capture and thermionic specific detection using a Varian 3500 Gas Chromatogram and a Hewlett Packard Ultra-2 capillary GC column (30m by 0.25mm i.d. by 0.33 [micro]m). Analyte peaks were identified and quantified by relative retention time against the internal standard and absolute responses to external calibration standards for both detectors. Calibration curves for the analysed compounds were prepared from 6 calibration standards within the range 0.02-2.0 [micro]g/mL. The quality assurance acceptance criteria procedure recommended for adoption in the European Union (European Commission 1997) was used to assess the internal standard recovery data for each sample and provide validation for reporting detected residues.
Atrazine adsorption isotherms
The short-term adsorption of atrazine by topsoil and subsoil from both sites was measured using methods similar to those described by Baskaran et al. (1996). Solutions containing 2, 4, 8, 12, and 16 mg/L of atrazine were prepared in a background solution of 0.01 M Ca[Cl.sub.2] and spiked with [sup.14]C-ring labelled atrazine to give approximately 5.8 x [10.sub.4] Bq/L. Soil samples in duplicate (1 g), with soil moisture adjusted to 60% of their maximum water-holding capacity (MWHC), were shaken with 10mL of these solutions for 4 h in an end-over-end shaker. The soil solution was centrifuged at 3000G for 5 min to collect the supernatant. To determine the [sup.14]C activity of the supernatant, 1 mL was mixed with 10 mL scintillation cocktail (Ultima Gold) (Packard), and the radioactivity determined by liquid scintillation counting. The difference between the atrazine solution concentration before adding soil samples and the supernatant solution concentrations after 4 h shaking with soil was attributed to adsorption. Atrazine adsorption was described by the Freundlich equation:
S = [K.sub.f][C.sup.n]
where S is the adsorbed concentration of atrazine at equilibrium (mg/kg soil), C is the concentration of atrazine in solution at equilibrium (mg/L), [K.sub.f] is the sorption coefficient ([micro][g.sup.1-n] [mL.sup.n]/g), and n is the Freundlich exponent (dimensionless).
Mineralisation of [sup.14]C-ring labelled atrazine
The rate of atrazine mineralisation in the soil samples collected from the field plots was measured using methods similar to those described by Spading et al. (1998). Soil (10g) was mixed with approximately 37 kBq [sup.14]C-ring-labelled atrazine (910 MBq/mmol) (Sigma) dissolved in approximately 500 [micro]L of water, to give a final concentration of 0.876 [micro]g radiolabelled atrazine/g soil. Soil moisture was adjusted to 60% of MWHC and maintained at that level for all mineralisation assays. For sterile controls, soil samples in beakers were autoclaved for 3 h at 121[degrees]C before atrazine addition. The beakers of soils were placed in 1-L glass mason jars alongside 10mL 1 M potassium hydroxide in a small beaker as a C[O.sub.2] trap. Five subsamples were prepared for each experiment. 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[degrees]C for 6-13 weeks depending on the time required for mineralisation activity to level off. At regular intervals, C[O.sub.2] traps were removed and 0.5 mL of the KOH was mixed with scintillation cocktail (10mL Ultima Gold) (Packard) and the radioactivity determined by liquid scintillation counting. The KOH was replaced at each sampling event. The amount of [sup.14]C[O.sub.2] trapped from the soil, corrected for background radiation levels, was taken as the measure of mineralisation of [sup.14]C-atrazine. The composite samples bulked from soil cores from the plots were used for both topsoil and subsoil of the Himatangi and Kiripaka soils.
To assess the spatial variability of atrazine mineralisation, 5 additional soil cores (not bulked) from the Himatangi plot, collected at 18 months after herbicide application, were also used in the study. All soil samples were stored at 4 C and analysed within 1 month of collection.
Microbial analyses of field samples
Numbers of atrazine degraders in bulked field samples were determined by a most probable number technique (MPN), using a 5-tube method and [sup.14]C-ring-labetled atrazine (Sparling et al. 1998). Soil or soil diluted in 0.1% (w/v) sterile sodium pyrophosphate solution was inoculated into 10mL minimal medium amended with 50 [micro]g/mL of atrazine as sole carbon source in a serum bottle. Each serum bottle was spiked with 500 Bq-carrier free [sup.14]C-atrazine and sealed with a subaseal suspending a small plastic cup (Kontes well) containing 300 [micro]L 1 M NaOH as the C[O.sub.2] trap. Control bottles were sterilised by autoclaving on 2 consecutive days before addition of [sup.14]C-atrazine. The serum bottles were incubated at 25[degrees]C for 5 weeks. After incubation, the radioactivity from [sup.14]C[O.sub.2] in the trapping solution was determined by liquid scintillation counting as described above. Those traps containing at least 3 times the radioactivity of the autoclaved controls were regarded as positive. This method enumerates microbes that mineralise atrazine to release C[O.sub.2], not those microbes that may partially oxidise atrazine.
To isolate microbes that degrade atrazine, soil or soil dilutions demonstrating atrazine mineralisation activity were inoculated onto agar plates containing atrazine as the sole nitrogen source (Mandelbaum et al. 1995). Because atrazine forms an opaque precipitate in the plates, those microbes able to metabolise atrazine were identified by the formation of zones of clearing around colonies in the atrazine layer.
An atrazine-degrading bacterium isolated from Himatangi topsoil was characterised by cell morphology, Gram-staining, production of endospores, and motility tests. For 16S rDNA isolation and sequencing, genomic DNA was isolated from bacterial colonies using the method of Woo et al. (1992). PCR and DNA sequencing were performed as described by Saul et al. (1993) and a Perkin-Elmer/Applied Biosystems 377 automated DNA sequencer. Full-length sequences were assembled from overlapping fragments covering both strands. The similarity between the 16S rDNA sequence of this strain and those in GenBank was determined.
Kinetics analysis of atrazine mineralisation rates
A logistic model was used to quantitatively compare laboratory studies of atrazine mineralisation activity in soils samples collected from the experimental plots:
x = a/(1 + [be.sub.-kt])
where x is the percentage of mineralisation as measured by the labelled C[O.sub.2] fluxes, a is the maximum percentage of mineralisation, b is a regression coefficient, and k is the mineralisation rate constant. These parameters were used to calculate the maximum mineralisation rate ([v.sub.m] = ak/4) (% per day).
Residual levels of triazine pesticides in soil
Residual levels of atrazine, terbuthylazine, and hexazinone in field samples from the experimental plots collected at various times after application showed similar trends (Table 2), although hexazinone was more persistent in Himatangi subsoil. The amount of triazines in topsoil collected 1 day after herbicide application was much less than that calculated.
For example, atrazine concentrations in topsoil were 4.56 and 3.20mg/kg, respectively, for Himatangi and Kiripaka soils. These levels were less than the 7.63 and 10.75 mg/kg estimated with an application rate of approximately 10 kg a.i. atrazine/ha. The difference between applied triazines and the amount recovered in the soil profile on day 1 was attributed to interception or absorption by plant material (lsensee and Sadeghi 1994).
Following application of the herbicides to Himatangi soil, triazine concentrations in the topsoil decreased from 1 month after application, and atrazine and terbuthylazine were below detectable levels (<0.01 mg/kg) by 18 months. In Himatangi subsoil, levels of the triazines were low and atrazine and terbuthylazine were only detected in the sample collected 1 month after application, whereas hexazinone was detected at all sample times. In Kiripaka topsoil and subsoil samples, atrazine, terbuthylazine, and hexazinone were detected on most sampling occasions and the levels declined over time. Levels of triazines detected in Kiripaka subsoil were always lower than the topsoil levels. By 18 months, levels of all 3 triazine herbicides in topsoil were low and in subsoil below detection limits. Levels of atrazine, terbuthylazine, and hexazinone in soils from outside the plots were always below detectable levels (data not shown).
Solvent and control soil blanks and spiked control soil samples (control soil spiked with target analytes) were analysed periodically. Solvent blank and control soil samples gave no detectable residues. The recovery of the target spiked compounds was 80-120% when spiked at either 0.04 or 4.0 mg/kg (ppm).
For all of the analysed samples the mean internal standard recovery was 76% with a 95% confidence limit of 73.0-79.1% and median value of 77.2%. This represents excellent recovery over all the analysed samples when consideration is made for the difference in soil type between the 2 experimental sites, and the significant change in soil properties with depth in the soil profile.
Atrazine adsorption isotherms
Atrazine adsorption isotherms for both soils are shown in Fig. 1. The sorption coefficient ([K.sub.f]) varied with soil type and decreased with soil depth. More atrazine was adsorbed in Himatangi topsoil ([K.sub.f] =4.69 [micro][g.sup.1-n] [mL.sup.n]/g) compared with the Kiripaka ([K.sub.f] = 3.80 ([micro][g.sup.1-n] [mL.sup.n]/g). In contrast, significantly more atrazine was adsorbed by Kiripaka subsoil ([K.sub.f] = 1.70 ([micro][g.sup.1-n] [mL.sup.n]/g) than Himatangi subsoil ([K.sub.f] = 0.53 ([micro][g.sup.1-n] [mL.sup.n]/g).
[FIGURE 1 OMITTED]
Atrazine mineralisation in soil samples from experimental plots
Soil atrazine mineralisation activity measured in the laboratory was well described by the logistics model with regression coefficients typically between 0.93 and 0.99 (Table3). The maximum percentage of atrazine mineralisation (a) and the maximum rates of atrazine mineralisation ([v.sub.m]), measured in soil samples from the experimental plots, show changes with time since exposure to atrazine in the field (Table 3), particularly in the surface soil.
Low levels of atrazine mineralisation activity were detected in samples collected from within the Himatangi plot at 1 and 3 months after atrazine application. With 3 months prior exposure of the soil to atrazine in the field, the maximum percentage of mineralisation was 3.2%, with the maximum rate of mineralisation reaching 0.16% per day. However, after 6 months the maximum extent of mineralisation reached a peak of 45.9% in topsoil from the experimental plot (Table 3) and the maximum rate reached 1.79% per day. While only low mineralisation activity (< 5% [sup.14]C[O.sub.2]) was detected at 9 months, in winter, activity was higher in subsequent samples collected from within the plot at 12, 15, 18, and 24 months. At these sampling times the maximum extent of mineralisation in the topsoil ranged from 16.3-27.6% (Table 3). These results were obtained from composite samples bulked from 5 soil cores. In individual soil cores collected from the treated plot at 18 months, mineralisation activity varied widely (Fig. 2). The maximum extent of atrazine mineralisation in the individual soil cores ranged from 1 to 15.6%, suggesting that plot heterogeneity could influence results from this site.
[FIGURE 2 OMITTED]
Atrazine mineralisation activity in the subsoil remained low (<7.5% [sup.14]C[O.sub.2]). A slight increase in the maximum extent and rate of mineralisation was recorded with time of exposure, reaching 7.3% and 0.12% per day, after 12 months prior exposure.
The maximum extent of atrazine mineralisation in soil from outside the plot was low, with 4.2% and 6.9% detected in topsoil and subsoil, respectively, 12 months after pesticide application. Atrazine mineralisation was negligible (<0.025% [sup.14]C[O.sub.2]) in sterile control soil (data not shown).
Numbers of atrazine degraders in Himatangi soil samples ranged from below detection (<1/g oven-dry soil) to 150 000/g oven-dry soil by the MPN method. Atrazine degraders were not detected for the first 3 months after pesticide application. Numbers in topsoil increased more than 1000-fold between 18 and 24 months after pesticide application (Table 4). Numbers in subsoil remained low, reaching 218/g oven-dry soil at 24 months.
A Gram-positive bacterium that metabolised atrazine as sole nitrogen source was isolated from Himatangi topsoil 6 months after application of atrazine. The bacterium was identified as Arthrobacter nicotinovorans by 16S rDNA sequence analysis.
Following a single application of pesticides to an experimental plot on Kiripaka soil, atrazine mineralisation activity in field samples was detected at elevated levels in topsoil at 3 months and in subsoil at 6 months after atrazine application in the field (Table 3). In the topsoil, the maximum extent of mineralisation varied from 2.1% after 1 month prior exposure to atrazine in the field to 65.2% after 9 months prior exposure (Table 3). The maximum rate of mineralisation reached a peak of 5.27% per day after 15 months exposure (Table 3). in contrast to the Himatangi subsoil, the Kiripaka subsoil showed significant increases in the maximum extent of mineralisation, reaching a peak of about 66% at about 9-12 months. The maximum rate of mineralisation in subsoil increased from 0.07% per day after 1 month prior field exposure to 3.55% per day after 12 months prior exposure in the field. The extent of mineralisation in the subsoil from 6 to 24 months was similar to, and sometimes higher than, that detected in topsoil for field samples from the experimental plot collected at the same time. As for Himatangi soils, the maximum extent of atrazine mineralisation in soil from outside the plot was low, with 4.1% and 1.4% detected in topsoil and subsoil, respectively, 12 months after pesticide application. Less than 0.025% [14.sup]C[O.sub.2] was accumulated from sterile soils spiked with [sup.14]C-ring-labelled atrazine (results not shown).
Numbers of atrazine degraders at the Kiripaka site remained below 100/g oven-dry soil except for topsoil analysed at 24 months post-pesticide application in which levels reached 35 000/g oven-dry soil (Table 4). No atrazine degraders were isolated from this site.
A limitation of this study is a lack of replication at each site; however, this experimental design has now been replicated on 5 different soils to allow comparison of trends throughout New Zealand, and some consistent effects have been detected. In this paper we report on temporal changes in atrazine mineralisation in soil from 2 experimental plots established on a sandy dune soil (Himatangi) and a silty clay (Kiripaka) following atrazine application.
Atrazine is reported to persist in soils lacking a history of atrazine application (Barriuso and Houot 1996; Larsen et al. 2000), but our results indicate that a single application of triazine pesticides, including atrazine, was sufficient to enhance the rate of atrazine mineralisation in topsoil from both plots and subsoil from the Kiripaka plot. There was, however, a lag period between field application and the detection of elevated atrazine mineralisation activity (1-6 months). The rate and extent of mineralisation detected in field samples from the experimental plots were generally higher (Table 3) than those reported previously for New Zealand soils (Spading et al. 1998) under comparable experimental conditions, but less than the maximum extent of mineralisation reported in other agricultural soils where >70% may be mineralised (Table 3) (Barriuso and Houot 1996; Radosevich et al. 1996; Ostrofsky et al. 1997; Houot et al. 2000; Loiseau and Barriuso 2002). The extent of mineralisation in the experimental plot on Kiripaka soil was, however, comparable to silty clay soils from France (Houot et al. 2000), some of which had received multiple applications of atrazine. Similar studies on sandy soils such as the Himatangi soil are not common, but the mineralisation activity we report here at 6 months is higher than that reported for a Danish sand (Jacobsen et al. 2001). Atrazine mineralisation activity was detected in samples from the experimental plots up to 24 months after atrazine application, although the rate of mineralisation had begun to decline. Furthermore, the mineralisation activity was sustained even though residual levels of triazines in the soil were low (<0.07mg/kg) or below detectable levels 18 months after application (Table 2).
Differences in the extent of mineralisation detected in soil samples from the Himatangi and Kiripaka experimental plots compared with other agricultural soils treated with atrazine could be attributed to low soil pH (Houot et al. 2000; Loiseau and Barriuso 2002). Whereas Houot et al. (2000) reported that in soils with pH <6.5, less than 25% of the initial [sup.14]C-atrazine was mineralised, results from this study and that of Loiseau and Barriuso (2002) do not support this observation. Both Himatangi and Kiripaka topsoils have a pH <6.5, yet more than 25% of the initial [sup.14]C-atrazine incubated with the field soils collected on more than one sampling occasion was mineralised in vitro. Loiseau and Barriuso (2002) reported that in soil with pH <6 the formation and stabilisation of hydroxylated derivatives of atrazine is favoured. Hence, in low-pH soils such as Himatangi and Kiripaka, the binding of atrazine microbial degradation products such as hydroxyatrazine to soil organic matter could result in decreased availability of the hydroxy products for microbial mineralisation. Another factor not evaluated in this study which could have limited atrazine mineralisation activity is high levels of mineral nitrogen and readily metabolised carbon (Alvey and Crowley 1995). The release of nutrients into the soil from decaying plant material killed by triazine application could have delayed the onset of detectable atrazine mineralisation in the field.
Atrazine mineralisation activity detected in topsoil from the Himatangi experimental plot was variable. Activity was high in samples collected 6 months and 12 months after pesticide application but was low in field samples collected at 9 months. Although the variability could be attributed to seasonal effects (i.e. winter conditions) slowing atrazine mineralisation, this was discounted, as rainfall and temperature at the 2 sites are similar and the highest levels of atrazine mineralisation activity in the laboratory were detected in Kiripaka soil in winter (9 months). To investigate the variability further, we measured atrazine mineralisation activity in individual field samples of Himatangi topsoil collected 18 months after atrazine treatment. Considerable variation in atrazine mineralisation activity was detected (Fig. 2). We postulate that the variability is due to uneven distribution of atrazine in the soil leading to a corresponding uneven distribution of atrazine-degrading microbes. The uneven distribution of atrazine is likely caused by the water-repellent nature of soils of this type (Wallis et al. 1991). As water does not move uniformly through sandy dune soil, atrazine is not distributed evenly, and atrazine-degrading microbes only become established and active in those areas exposed to atrazine. Spatial variability in the degradation of the herbicide isoproturon in topsoil of a sandy loam has recently been linked to activity of a degradative Sphingomonas sp. that only degrades isoproturon in soil with a pH >7 (Bending et al. 2003).
In subsoil samples, only a small increase in the extent and rate of atrazine mineralisation was detected in Himatangi soil.
This was probably related to the low concentration of residual atrazine that accumulated in the subsoil of the experimental plot (Table 2). In the absence of atrazine accumulation there was no selection for the establishment of an active atrazine-mineralising microbial population. In contrast to Himatangi, the mineralisation rate in field samples of Kiripaka subsoil was greater, probably because of the higher concentrations of atrazine retained in the soil (Table 2), which would have stimulated the establishment of an atrazine-mineralising microbial population in situ.
Low concentrations of atrazine in the Himatangi subsoil can be attributed to the lack of movement of atrazine into the subsoil or, alternatively, the lack of retention of atrazine in the subsoil. As atrazine was detected in the subsoil 1 month after pesticide application, it is possible atrazine leached through the subsoil and was not retained. Furthermore, the Himatangi subsoil has low adsorption for atrazine as indicated by the [K.sup.f] value of 0.53 (Fig. 1). Atrazine leaching through Himatangi subsoil could also have been influenced by preferential flow. Although sandy soils, given their uniform soil structure, are not normally expected to exhibit preferential flow, we have observed such behaviour in another Recent Soil, with very similar characteristics to Himatangi soil (McLeod et al. 2001).
The atrazine adsorption isotherm experiments showed atrazine to be more strongly adsorbed by the Himatangi topsoil than Kiripaka topsoil, with the reverse true for the corresponding subsoil (Fig. 1). The higher [K.sup.f] value for Himatangi topsoil was surprising as sorption of pesticides is generally higher for allophanic than for non-allophanic soils (Baskaran et al. 1996). In addition, Kiripaka topsoil has a higher total carbon content than the Himatangi topsoil (Table 1). The difference in atrazine adsorption isotherms for these topsoils may best be explained by soil pH. in acidic soils, atrazine, a weak base, is sorbed strongly as it is protonated and becomes cationic. Hence, atrazine is more strongly sorbed to Himatangi topsoil (pH 5.1) than Kiripaka topsoil (pH 6.2). Baskaran et al. (1996) have reported experimental data confirming that the adsorption of atrazine increases when soil pH decreases.
In soil, atrazine is retained against leaching loss primarily by sorption to organic matter. Detailed analysis of bound atrazine residues in soil (Loiseau and Barriuso 2002) demonstrated that most of the residues were in the finest soil fraction that contained humified organic matter. Loiseau and Barriuso (2002) postulated that atrazine and its residues were either entrapped within voids in the soil organic matter or strongly associated with soil humic material. The Kiripaka subsoil, with higher levels of carbon, has more adsorption sites for retaining atrazine than the Himatangi subsoil, and this is reflected in its higher sorption coefficient.
The lag period between application of atrazine in the field and detection of atrazine mineralisation activity in field samples from the experimental plots is due to the time required for establishment of an active atrazine-degrading microbial population under in situ conditions. Numbers of atrazine degraders detected in the soils do not seem to be high enough to account for the higher levels of atrazine mineralisation activity (>50%) measured (Ostrofsky et al. 2002). Alvey and Crowley (1996), for example, increased the extent of atrazine mineralisation in soils laboratory studies from <5% to approximately 70% following inoculation of the soil with 4.5 x [10.sup.6] bacteria/g soil. The bacteria were from a mixed culture grown on atrazine as sole source of carbon and nitrogen. In this study the numbers of atrazine degraders remained low (<250/g oven-dry soil) in both soils until 24 months after atrazine application, when > [10.sup.4] atrazine degraders/g oven-dry soil were detected in topsoil. Slightly higher numbers of atrazine-degrading bacteria may have been detected using MPN media with atrazine provided as the nitrogen source rather than the carbon source (Jayachandran et al. 1998).
The maximum mineralisation rates detected in the soil samples from the field plots did not coincide with highest numbers of atrazine degraders detected. For the Himatangi topsoil obtained from the experimental plot, the maximum mineralisation rate was detected following 6 months exposure to atrazine, yet numbers of atrazine degraders were <10/g oven-dry soil. Similarly for samples from the Kiripaka soil, the maximum rate of mineralisation was detected in topsoil at 15 months and subsoil at 12 months, whereas numbers of atrazine degraders were negligible at 33 and 50/g oven-dry soil, respectively. The detection of high numbers in topsoil collected at 24 months was surprising, as residual atrazine in the soil had fallen to low levels by 18 months.
It is possible that very few atrazine degraders (<250/g oven-dry soil) are responsible for the atrazine mineralisation activity detected and that M PN counts are not a good predictor of atrazine mineralisation activity in soil. Alternatively, under the conditions used in this study, the microbes mineralising atrazine may have been unculturable as reported by Ostrofsky et al. (2002), at least for the first 15 months of this study. Hence, molecular enumeration techniques independent of the need to culture atrazine-degrading microbes may be more appropriate for enumerating the atrazine-mineralising microbial populations within these soils. Competitive polymerase chain reaction targeting the atrazine degradation gene atzA, for example, has been used to quantitate Pseudomonas sp. strain ADP in sediment (Ciausen et al. 2002). However, this method would be unsuitable if microbes degrading atrazine in the sample of interest did not contain the target gene atzA (Ostrofsky et al. 2002).
A Gram-positive atrazine-degrading bacterium identified as Arthrobacter nicotinovorans was isolated from Himatangi soil. To our knowledge, this is the first report of a bacterium from New Zealand soils that degrades atrazine. Furthermore, the bacterium was isolated directly from agricultural soils, without the need for enrichment. Attempts to isolate atrazine degraders in pure culture, from Kiripaka soil, on agar plates with atrazine supplied as the sole source of nitrogen were unsuccessful, despite the high rates of mineralisation.
The influence of terbuthylazine and hexazinone on atrazine mineralisation activity warrants further investigation. As the chemical structure of terbuthylazine is very similar to atrazine, we anticipate mineralisation activity of atrazine and terbuthylazine in these soils could follow similar trends. Furthermore, bacteria that degrade atrazine are reported to degrade other triazine compounds including terbuthylazine (Strong et al. 2002), as does the atrazine-degrading bacterium we isolated from Himatangi soil (data not provided). It is possible that terbuthylazine could enhance atrazine mineralisation and vice versa, by providing a metabolisable co-substrate. As hexazinone has a very different chemical structure to atrazine or terbuthylazine, we anticipate that it will not greatly influence the mineralisation of atrazine or terbuthylazine in soil when applied at the rates used in this study.
In conclusion, a single application of atrazine was sufficient to enhance the rate of [sup.14]C-atrazine mineralisation in vitro by topsoil from both Himatangi and Kiripaka plots, and subsoil from the Kiripaka plot. Rates of mineralisation of atrazine in the soil increased 1-6 months after pesticide application and remained elevated for 18-24 months. Numbers of atrazine degraders detected, however, did not correlate with atrazine mineralisation rates.
Table 1. Classification and morphological, chemical, and physical properties of the 2 soils studied Soil property Himatangi soil Kiripaka soil Classification (A) Typic Sandy Recent Typic Orthic Allophanic Origin Dune sand Strongly weathered basalt Drainage class Imperfect Well Texture group Sandy Clayey Morphology Weak crumb and single Strong fine grain, buried polyhedral, topsoil at 40 cm gravels, very firm at 40 cm Mineralogy Mixed Allophanic Topsoil Subsoil Topsoil Subsoil (0-10cm) (50-90 cm) (0-20 cm) (40-80cm) pH 5.1 6.5 6.2 6.1 Total carbon (%) 3.29 0.13 6.97 1.85 Total nitrogen (%) 0.17 0.03 0.61 0.16 Mineral N (mg/kg) 6.5 3.9 23.1 0.7 MWHC (% v/v) 38 27 88 90 CEC (cmol/kg) 9.7 1.8 31.9 20.9 Sand (%) 88 88 19 15 Silt (%) 10 10 38 34 Clay (%) 2 2 43 51 (A) Hewitt (1992). Table 2. Residual concentrations (mg/kg) of triazine pesticides in soil in bulk samples from experimental plots at various time intervals post application AZ, Atrazine; TB, terbuthylazine; HZ, hexazinone. Detection limit for analysis was 0.01 mg/kg Himatangi Time since 0-10 cm 50-90 cm application (months) AZ TB HZ AZ TB HZ 0 4.56 4.44 7.06 NT NT NT 1 4.31 5.02 4.43 0.05 0.09 0.03 3 0.71 0.64 2.22 <0.01 <0.01 0.05 6 0.27 0.47 0.17 <0.01 <0.01 0.03 9 0.15 0.61 0.17 <0.01 <0.01 0.03 12 0.05 0.05 0.07 <0.01 <0.01 0.03 18 <0.01 <0.01 0.06 <0.01 <0.01 <0.01 Kiripaka Time since 0-10 cm 40-80 cm application (months) AZ TB HZ AZ TB HZ 0 3.20 3.3 4.7 NT NT NT 1 5.0 6.6 1.5 0.33 0.75 0.45 3 2.0 3.4 0.46 0.06 0.10 0.18 6 <0.01 0.03 0.40 0.08 0.14 0.10 9 0.29 0.49 0.19 0.01 0.02 0.02 12 0.27 0.44 0.12 0.03 0.04 0.02 18 0.02 0.06 0.02 <0.01 <0.01 <0.01 NT, not measured. Table 3. Modelled parameters of atrazine mineralisation (a, the maximum extent of mineralisation, [v.sub.m], maximum mineralisation rate) from laboratory studies with topsoil and subsoil samples from 2 experimental plots on Himatangi and Kiripaka soil at 0, 1, 3, 6, 9, 12, 15, 18, and 24 months after pesticide application Soil from outside the plot (not treated with pesticides) was assayed at 0 and 12 months Himatangi Time since applic. a [v.sub.m] (months) (%) (% per day) [R.sup.2] Inside the plot Topsoil 1 1.8 0.04 0.984 3 3.2 0.16 0.757 6 45.9 1.79 0.992 9 4.2 0.05 0.992 12 27.6 0.81 0.982 15 16.3 0.62 0.996 18 23.2 0.47 0.990 24 26.2 0.80 0.993 Subsoil 1 1.1 0.03 0.997 3 0.7 0.04 0.991 6 3.6 0.06 0.99 9 3.1 0.04 0.982 12 7.3 0.12 0.978 15 4.7 0.11 0.995 18 6.4 0.09 0.983 24 NF NF NF Outside the plot Topsoil 0 2.0 0.04 0.998 12 4.2 0.08 0.995 Subsoil 0 0.9 0.04 0.972 12 6.9 0.19 0.992 Kiripaka Time since applic. a [v.sub.m] (months) (%) (% per day) [R.sup.2] Inside the plot Topsoil 1 2.1 0.05 0.985 3 26.4 1.08 0.988 6 18.9 0.26 0.997 9 65.2 2.75 0.988 12 62.3 3.55 0.975 15 59.2 5.27 0.984 18 34.3 0.96 0.970 24 53.3 1.51 0.996 Subsoil 1 1.9 0.07 0.969 3 2.2 0.09 0.935 6 20.5 0.38 0.991 9 66.6 2.31 0.992 12 65.9 3.55 0.989 15 50.3 2.25 0.993 18 51.0 1.05 0.984 24 35.1 0.99 0.996 Outside the plot Topsoil 0 2.2 0.07 0.975 12 4.1 0.09 0.948 Subsoil 0 1.7 0.09 0.948 12 1.4 0.06 0.930 NF, Model cannot be fitted. Table 4. Most probable number (and 95% confidence intervals; n = 5) of atrazine degraders (per g oven-dry soil) in soil collected from pesticide plots on Himatangi and Kiripaka soil at various times after pesticide application Time since Himatangi applic. (months) Topsoil Subsoil 0 ND ND 1 ND ND 3 ND ND 6 9.6 0.3 (2-45) (0.06-1.3) 9 20 0.2 (6-66) (0.06-1) 12 4 0.3 (1-12) (0.1-1) 15 3 3 (0.7-8) (0.8-8) 18 38 0.2 (11-13) (0.06-1) 24 150 000 218 (44 466-484 800) (66-723) Time since Kiripaka applic. (months) Topsoil Subsoil 0 ND ND 1 2 ND (0.5-2.5) 3 1 ND (0.3-6) 6 4 ND (1-16) 9 75 4 (29-249) (1-12) 12 31 50 (9-104) (15-167) 15 33 12 (10-110) (3-40) 18 33 7 (10-108) (2-23) 24 35 000 15 (10 600-115 6070) (4-47) ND, Not detected.
This work was supported by funding from the Foundation for Research, Science and Technology, New Zealand (C09X0017). We thank Malcolm McLeod (Landcare Research) for provision of soils information, Dr Graham Spading (Landcare Research) for helpful discussions during preparation of this manuscript, and Dr David Saul (University of Auckland) for 16S rDNA sequence analysis of the atrazine-degrading bacterium. Mr Greg Arnold, Biometrician, Landcare Research, provided advice on data presentation and analysis.
Alvey S, Crowley D (1995) Influence of organic amendments on biodegradation of atrazine as a nitrogen source. Journal of Environmental Quality 24, 1156-1162.
Alvey S, Crowley DE (1996) Survival and activity of an atrazine-mineralizing consortium in rhizosphere soil. Environmental Science and Technology 30, 1596-1603. doi: 10.1021/ES950575+
Assaf NA, Yurco RF (1994) Influence of carbon and nitrogen application on the mineralization of atrazine and its metabolites in soil. Pesticide Science 41, 41-47.
Barriuso E, Houot S (1996) Rapid mineralisation of the s-triazine ring of atrazine in soils in relation to soil management. Soil Biology and Biochemistry 28, 1341-1348. doi: 10.1016/S0038-0717(96) 00144-7
Baskaran S, Bolan NS, Rahman A, Tillman RW (1996) Pesticide sorption by allophanic and non-allophanic soils of New Zealand. New Zealand Journal of Agricultural Research 39, 297-310.
Bending GD, Lincoln SD, Sorensen SR, Morgan JAW, Aamand J, Walker A (2003) In-field spatial variability in the degradation of the phenyl-urea herbicide isoproturon is the result of interactions between degradative Sphingomonas spp. and soil pH. Applied and Environmental Microbiology 69, 827-834. doi: 10.1128/AEM.69.2.827-834.2003
Blakemore LC, Searle PL, Daly BK (1987) Methods for chemical analysis of soils. NZ Soil Bureau Scientific Report 80. Lower Hutt, New Zealand.
Clausen GB, Larsen L, Johnsen K, Radnoti de Lipthay J, Aamand J (2002) Quantification of the atrazine-degrading Pseudomonas sp. strain ADP in aquifer sediment by quantitative competitive polymerase chain reaction. FEMS Microbiology Ecology 41, 221-229. doi: 10.1016/S0168-6496(02)00303-3
Close ME, Rosen M (2001) 1998/89 National survey of pesticides in groundwater using GCMS and ELISA. New Zealand Journal of Marine and Freshwater Research 35, 205-219.
Cowie JD, Fitzgerald P, Owers W (1967) Soils of the Manawatu sand country. Soil Bureau Bulletin 29, Lower Hutt, New Zealand.
Cox JE, Taylor NH, Sutherland CF, Wright ACS (1983) Northland Peninsula soil legends. NZ Soil Bureau, DSIR, Lower Hutt, New Zealand.
Di HJ, Aylmore LAG, Kookana RS (1998) Degradation rates of eight pesticides in surface and subsurface soils under laboratory and field conditions. Soil Science 163, 404-411. doi: 10.1097/00010694-199805000-00008
Di HJ, Sparling GP, Lee R, Magesan GN (2001) The effect of mineralisation rates of atrazine in surface and subsurface soils on its groundwater contamination potential. Australian Journal of Soil Research 39, 175-183. doi: 10.1071/SR00004
Dinnelli G, Accinelli C, Vicari A, Catizone P (2000) Comparison of the presistence of atrazine and metolachlor under field and laboratory conditions. Journal of Agricultural and Food Chemistry 48, 3037-3043. doi: 10.1021/JF991057B
Erickson LE, Lee KH (1989) Degradation of atrazine and related s-triazines. Critical Review Environmental Control 19, 1-13.
European Commission (1997) Quality Control Procedures for pesticide residue analysis. Guidelines for Residues Monitoring in the European Union. Document 7826/VI/97. Prepared on behalf of the European Commission by Alan Hill, Central Science Laboratory, UK, and Organising Committee of the EUQCW 97.
Hewitt AE (1992) 'New Zealand Soil Classification.' DSIR Land Resources Scientific Report No. 19, Lower Hutt, New Zealand.
Houot S, Topp E, Yassir A, Soulas G (2000) Dependence of accelerated degradation of atrazine on soil pH in French and Canadian soils. Soil Biology and Biochemistry 32, 615-625. doi: 10.1016/S0038-0717(99)00188-1
Isensee AR, Sadeghi AM (1994) Effects of tillage and rainfall on atrazine residues in soil. Weed Science 42, 462-467.
Jayachandran K, Stolpe NB, Moorman TB, Shea PJ (1998) Application of [sup.14]C-most probable number technique to enumerate atrazine-degrading microorganisms in soil. Soil Biology and Biochemistry 30, 523-529. doi: 10.1016/S0038-0717197) 00137-5
Jacobsen CS, Shapir N, Jensen LO, Jensen EH, Juhler RK, Streibeg JC, Mandelbaum RT, Helweg A (2001) Bioavailability of traizine herbicides in a sandy soil profile. Biology and Fertility of Soils 33, 501-506. doi: 10.1007/S003740100359
Larsen L, Sorensen SR, Aamand J (2000) Mecoprop, isoproturon, and atrazine in and above a sandy aquifer: vertical distribution of mineralisation potential. Environmental Science and Technology 34, 2426-2430. doi: 10.1021/ES9911723
Loiseau L, Barriuso E (2002) Characterization of the Atrazine's bound (non extractable) residues using fractionation techniques for soil organic matter. Environmental Science and Technology 36, 683-689. doi: 10.1021/ES010146D
Mandelbaum RT, Allan DL, Wackett LP (1995) Isolation and characterization of a Pseudomonas sp. that mineralizes the s-triazine herbicide atrazine. Applied and Environmental Microbiology 61, 1451-1457.
McLeod M, Aislabie J, Smith J, Fraser R, Roberts A, Taylor M (2001) Viral and chemical movement through contrasting soils. Journal of Environmental Quality 30, 2134-2140.
Ostrofsky EB, Robinson JB, Traina SJ, Tuovinen OH (2002) Analysis of atrazine-degrading communities in soils using most-probable number enumeration, DNA hybridization, and inhibitors. Soil Biology and Biochemistry 34, 1449-1459. doi: 10.1016/S0038-0717(02)00089-5
Ostrofsky EB, Traina SJ, Tuovinen OH (1997) Variation in atrazine mineralization rates in relation to agricultural management practice. Journal of Environmental Quality 26, 647-657.
Radosevich M, Traina SJ, Tuovinen OH (1996) Biodegradation of atrazine in surface soils and surface sediments collected from an agricultural research farm. Biodegradation 7, 137-149.
Radosevich M, Traina S J, Tuovinen OH (1997) Atrazine mineralization in laboratory-aged soil microcosms inoculated with s-triazine-degrading bacteria. Journal of Environmental Quality 26, 206-214.
Roos AH, Van Munsteren AJ, McNab FM, Tuinstra LGMTh (1987) Universal extraction/clean-up procedure for extraction with ethylacetate and size exclusion chromatography. Analytica Chimica Acta 196, 95-102. doi: 10.1016/S0003-2670(00)83074-3
Rousseaux S, Hartmann A, Soulas G (2001) Isolation and charaeterisation of new Gram-negative and Gram-positive atrazine degrading bacteria from different French soils. FEMS Microbiology Ecology 36, 211-222. doi: 10.1016/S0168-6496(01)00135-0
Saul D J, Rodrigo A, Reeves RA, Williams LC, Borges KM, Morgan HW, Bergquist PL (1993) Phylogeny of twenty Thermus isolates constructed from 16S rRNA gene sequence data. International Journal of Systematic Bacteriology 43, 754-760.
de Souza ML, Seffernick J, Martinez B, Sadowsky MJ, Wackett LP (1998) The atrazine catabolism genes atzABC are widespread and highly conserved. Journal of Bacteriology 180, 1951-1954.
Sparling G, Dragten R, Aislabie J, Fraser R (1998) Atrazine mineralisation in New Zealand topsoils and subsoils: influence of edaphic factors and numbers of atrazine-degrading microbes. Australian Journal of Soil Research 36, 557-570. doi: 10.1071/S97065
Strong LC, Rosendahl C, Johnson G, Sadowsky MJ, Wackett LP (2002) Arthrobacter aurescens TC1 metabolizes diverse s-triazine ring compounds. Applied and Environmental Microbiology 68, 5973-5980. doi: 10.1128/AEM.68.12.5973-5980.2002
Topp E, Tessier L, Gregorich EG (1996) Dairy manure incorporation stimulates rapid atrazine mineralization in an agricultural soil. Canadian Journal of Soil Science 76, 403-409.
Vanderheyden V, Debongnie P, Pussemier L (1997) Accelerated degradation and mineralisation of atrazine in surface and subsurface soil materials. Pesticide Science 49, 237-242. doi: 10.1002/(SICI)1096-9063(199703)49:3 <237::AID-PS511> 3.3.CO;2-W
Wackett LP, Sadowsky MJ, Martinez B, Shapir N (2002) Biodegradation of atrazine and related s-triazine compounds: from enzymes to field studies. Applied Microbiology and Biotechnology 58, 39-45. doi: 10.1007/S00253-001-0862-Y
Wallis MG, Scotter DR, Home DJ (1991) An evaluation of the intrinsic sorptivity water repellency index on a range of New Zealand soils. Australian Journal of Soil Research 29, 353-362.
Woo HS, Cheng AF, Ling JM (1992) An application of a simple method for the preparation of bacterial DNA. BioTechniques 13, 696-697.
Manuscript received 30 May 2003, accepted 14 May 2004
J. Aislabie (A,D), D. Hunter (A), J Ryburn (A), R. Fraser (A), G.L. Northcott (B), and H.J. Di (A) Landcare Research, Private Bag 3127, Hamilton, New Zealand. (B) Food Sector, HortResearch, Private Bag 3123, Hamilton, New Zealand. (C) Center for Soil and Environmental Research, PO Box 84, Lincoln University, Canterbury, New Zealand. (D) Corresponding author. Email: firstname.lastname@example.org
|Printer friendly Cite/link Email Feedback|
|Author:||Aislabie, J.; Hunter, D.; Ryburn, J.; Fraser, R.; Northcott, G.L.; Di, H.J.|
|Publication:||Australian Journal of Soil Research|
|Date:||Dec 1, 2004|
|Previous Article:||Soil microbial biomass and microbial biodiversity in some soils from New South Wales, Australia.|
|Next Article:||Participatory soil quality assessment: the case of smallholder farmers in Ethiopian highlands.|