| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Geochemistry, Geological Survey of Denmark and Greenland (GEUS), Øster Voldgade 10, DK-1350 Copenhagen, Denmark
b Univ. of Aarhus, Faculty of Agricultural Sciences, Inst. of Agroecology and Environment, DK-8830 Tjele, Denmark
c Biotech Research and Innovation Centre (BRIC), Fruebjergvej 3, DK-2100 Copenhagen, Denmark
d Dep. of Natural Sciences, Faculty of Life Sciences, Univ. of Copenhagen, Thorvaldsensvej 40, DK-1871 Frederiksberg, Denmark
* Corresponding author (lf{at}geus.dk).
Received for publication May 30, 2006.
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMaterials and MethodsResultsDiscussionConclusionREFERENCES |
|---|
Abbreviations: ASA, arylsulfatase activity • CFU, colony-forming units • Ct, cycle threshold • 2,4-D, 2,4-dichlorophenoxyacetic acid • FDA, fluorescein diacetate hydrolysis • ISR, in situ soil respiration • MCP, 2-methyl-4-chorophenol • MCPA, 2-methyl-4-chlorophenoxyacetic acid • Corg, organic carbon • PCR, polymerase chain reaction • SIR, substrate-induced respiration • TSA, tryptic soy broth agar
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and MethodsResultsDiscussionConclusionREFERENCES |
|---|
The degradation of most pesticides in soil is enzymatically catalyzed by microorganisms (Topp et al., 1997) and generally follows zero-order or first-order kinetics. This indicates that they are degraded by co-metabolism along with the general metabolic activities of the soil community and hence provide little or no energy to the organisms involved. Repeated application of a pesticide to a soil for a period of several years may result in enhanced biodegradation, thereby indicating adaptation of the soil microbial community (Jensen and Petersen, 1952) and proliferation of organisms able to use the compound as an energy source (Bending et al., 2001).
As the indigenous soil microbial community adapts to repeated exposure to a pesticide, the presence and capacity of the degrading microorganisms in a soil are mainly determined by the pesticide application history at the site (Sorensen et al., 2003; Sorensen and Aamand, 2003). It does not explain the intra-field spatial variation in degradation rate, however. The latter has been attributed to parameters such as general metabolic diversity and microbial biomass, growth-linked versus co-metabolic degradation mechanisms, and soil pH and to intrinsic soil properties such as Ctotal/Ntotal ratio and water-extractable potassium (Bending et al., 2001; Rasmussen et al., 2005; Walker et al., 2001; Walker et al., 2002).
The phenoxyacetic acid herbicide MCPA (2-methyl-4-chlorophenoxyacetic acid) has been widely used to control a range of broadleaf weeds in agricultural fields and uncultivated areas over the past 50 years (Caux et al., 1995). The substance is highly water soluble with low retention in soil (Kd, 0.3–1.0 L kg–1), thus posing the risk that it may leach to and contaminate the groundwater (Helweg, 1987; Socias-Viciana, 1999). Under aerobic conditions, MCPA degrades rapidly (half-life, 3–16 d) (Muller, 1997; Thorstensen and Lode, 2001). Even though the MCPA mineralization potential of agricultural soils generally seems to be high when studied in the laboratory, MCPA is nevertheless detected in Danish groundwater beneath agricultural fields and in creeks. During the period 1993 to 2003, MCPA was detected in 9.3% of 118 groundwater wells monitored in Denmark (GEUS, 2004).
The degradation of phenoxyacetic acid herbicides such as 2,4-dichlorophenoxyacetic acid (2,4-D), MCPA, and related compounds has been studied intensively, and detailed knowledge is available about the catabolic pathways that these compounds follow (Bollag et al., 1967; Gaunt and Evans, 1971; Pieper et al., 1988). The first step in the degradation of MCPA is mediated by the TfdA enzyme (a ketoglutarate dioxygenase). This converts MCPA to 2-methyl-4-chorophenol (MCP) by oxidation of 
-ketoglutarate to succinate. The gene encoding for the TfdA enzyme is known to be carried within the tfdA-F gene cluster, which was first identified on the conjugative plasmid pJP4 in Alcaligenes eutrophus (Ralstonia eutropha) JMP134 (Don and Pemberton, 1981; Don and Pemberton, 1985). Based on sequence similarity analysis, it has recently been proposed to subdivide the tfdA genes into three classes: Class I contains those genes that are closely related to the pJP4 plasmid, and Classes II and III cover all other groups (Itoh et al., 2004). Class I, II, and III tfdA genes encode enzymes that convert 2,4-D to 2,4-dichlorophenol.
The phenoxyacetic acid herbicide 2,4-D, which is closely related to MCPA, is degraded by specific bacteria able to use it as a source of C and energy and by other microorganisms in co-metabolic pathways (Robertson and Alexander, 1994). Spatial variation in 2,4-D mineralization has been found to be inversely proportional to scale over the range field scale to microhabitat scale (<1 cm3) due to the uneven distribution of degrading bacteria and the C necessary for co-metabolism (Gonod et al., 2003).
Little information is available about spatial variation in the microbial degradation of MCPA. In microcosm studies of MCPA degradation in topsoil and subsoil from a sandy agricultural field, Baelum et al. (2006) found a shift in catabolic tfdA genes from the Class I tfdA genes that were initially dominant in the soil to the Class III tfdA genes that were responsible for the observed mineralization of MCPA.
The objective of the present study was to examine variation in microbial MCPA mineralization potential at field level and to relate this to relevant biotic and abiotic parameters, including soil sorption of MCPA, and to the initial presence of tfdA genes as measured by quantitative real-time polymerase chain reaction (PCR).
|
Materials and Methods |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and MethodsResultsDiscussionConclusionREFERENCES |
|---|
Soil Sampling and Storage
At each grid point, approximately 10 kg of soil was sampled from the topsoil layer (Ap horizon; 5–25 cm soil depth interval) and from the subsoil (Bs horizon; 40–60 cm soil depth interval). The samples were brought to the laboratory in closed plastic containers, and each container was mixed in a large end-over-end shaker for 10 min. Each soil sample was sieved through a 4-mm mesh, with the mesh being carefully cleaned between samples to avoid cross-contamination. For analysis of microbial activity, 200-g aliquots were kept in closed inert Rilsan plastic bags (Rotek A/S, Sdr. Felding, Denmark) at 4°C for a maximum of 10 d. For analysis of pesticide sorption, the soils were air-dried at room temperature and stored under dry conditions. For all other uses (mineralization analysis, colony-forming unit [CFU] counts, and DNA extraction), the sieved soil was stored at –20°C for up to several months before being processed consecutively. Seven days before use, the soil samples were acclimatized at 10°C (Mortensen and Jacobsen, 2004). The results of the pedologic and geochemical analyses performed on the soil samples are summarized in Table 1
.
|
CFU Counts
Frozen 15-g soil samples in glass vials were thawed at 10°C in the dark for 7 d, corresponding to the time used for acclimatization of soil samples before the mineralization experiments. Soil suspensions were prepared by blending 10.0 g soil with 100 mL 0.9% NaCl for 60 s. Large soil particles were allowed to settle for 15 min before sampling of 1.0 mL soil suspension for a 10-fold dilution series in 0.9% NaCl.
Total viable counts of bacteria in the samples were made by drop-plating on 1/300 tryptic soy broth agar (TSA) plates. The bacterial pseudomonad population in each sample was estimated from triplicate plate counts on Gould S1 agar plates (Gould et al., 1985). The plates were incubated at 20°C for 72 h.
Microbial Activity Measures
In situ soil respiration (ISR) was measured using a dynamic chamber method as described by Vinther et al. (2007). Arylsulfatase activity (ASA) was measured on 2-g soil samples as described by Elsgaard et al. (2002) and Vinther et al. (2007). Fluorescein diacetate (FDA) hydrolysis was measured according to the method of Schnurer and Rosswall (1982). Substrate-induced respiration (SIR) was measured using the method of West and Sparling (West and Sparling, 1986) as described by Vinther et al. (2007). All microbial activity measurements where made on nonfrozen soil samples in connection with or shortly after sample collection (10 d max).
Mineralization Experiments
A stock solution of ring-labeled 14C-MCPA was prepared by dissolving analytical grade MCPA (99.5% purity) purchased from Dr. Ehrenstorfer GmbH (Augsburg, Germany) in Milli-Q water (Millipore, Billerica, MA) and adding trace amounts of [Ring-14C]-MCPA (159.7 µCi mmol–1) purchased from IZOTOP (Budapest, Hungary) to achieve an initial radioactivity in each microcosm of approximately 50,000 dpm and a final MCPA concentration of 20 mg L–1. To ensure the complete dissolution of the MCPA and to avoid disturbance of the soil pH, the pH of the stock solution was adjusted to 6.5 with NaOH. The radiochemical purity of [Ring-14C]-MCPA was >95% according to the manufacturer.
The mineralization of MCPA was estimated in sieved (>4 mm) soil in 100-mL bottles with airtight glass lids. Microcosms were prepared in triplicate for each soil sample using 10.0 g soil (dry weight). The soil moisture content was adjusted to 90% of the soil water-holding capacity with sterile Milli-Q water (corresponding to approximately 9% H2O after MCPA spiking). The microcosms were spiked with 500 µL of the MCPA stock solution to obtain a final concentration of 1.0 mg MCPA kg–1 soil (dry weight). A sterile 3-mL glass vial containing 2.0 mL 0.5 M NaOH was placed in each microcosm to capture the CO2 produced during mineralization of the MCPA as described by Larsen et al. (2000). The microcosms were incubated at 10°C in the dark.
The NaOH in the microcosm CO2 traps was sampled at Days 8, 15, 22, 29, 39, 43, 52, 57, and 67. The 14CO2 content was determined using a Wallac 1409 liquid scintillation counter after mixing with 10 mL of OptiPhase Hisafe 3 scintillation cocktail (Wallac, Turku, Finland). Radioactivity was converted to percentage mineralization of the MCPA in the microcosms.
Sorption Experiments
Sorption of [Ring-14C]-MCPA was determined on all 102 soil samples in triplicate. Air-dried samples were sieved to <2 mm before initiation of the sorption experiments. The sorption experiments were performed in 13-mL Pyrex vials with Teflon screw-caps. A sample to solution ratio of 1:1 was used with a liquid phase consisting of 0.01 M CaCl2, as recommended in the OECD guidelines for sorption studies (OECD, 2000). The MCPA concentration was 250 µg kg–1, and each vial contained 20,000 dpm 14C-MCPA mL–1. The vials were incubated on an orbital shaker at 10°C for 96 h. This was followed by centrifugation (30 min at 2700 x g) and removal of the supernatant. The 14C in the supernatant was measured by liquid scintillation counting as described for the mineralization experiments, and the distribution coefficient (Kd) was calculated. The pH was measured on one of each set of triplicate soil subsamples after dilution in 0.01 M CaCl2. The spatial variation in MCPA sorption and mineralization was visualized on contour maps produced by ordinary kriging with the Surfer version 7.00 surface mapping system (Golden Software, Inc., Golden, CO).
DNA Extraction
Frozen 15-g soil samples in glass vials were thawed at 10°C in the dark for 7 d to acclimatize the soil microorganisms before DNA extraction. Whole-community DNA was extracted from 0.5-g (wet weight) soil samples using the FastDNA SPIN kit for soil (Bio101 Inc., Carlsbad, CA). The protocol recommended by the manufacturer was followed with two modifications. The beat-beating step was modified to four 30-s pulses at speed 4 instead of one 30-s pulse at speed 5.5 in the FastPrep FP 120 instrument (Bio101 Inc.). A freeze-thaw step was included with freezing of the samples for 1 h at –80°C and thawing for 30 min at 30°C. Each DNA sample was eluted in 100 µL RNase/DNase-free water (Bio101 Inc.) and stored at –80°C.
Real-Time Quantitative PCR
The forward primer 5'-GAGCACTACGC(G/A)CTGAA(T/C)TCCCG-3' and the reverse primer 5'-GTCGCGTGCTCGAGAAG-3' were used to yield a 210–base-pair DNA fragment. The primers were purchased from MWG Biotech (Ebersberg, Germany). The primer set was designed on the basis of 22 Class I and Class III tfdA genes retrieved from GenBank (Baelum et al., 2006).
The mixture used for the real-time PCR consisted of the QuantiTect SYBR green PCR kit (QIAGEN, Crawley, UK) containing deoxynucleoside triphosphate mix, HotStar Taq DNA polymerase, PCR buffer, Rox, and 2.5 mM MgCl2. The reaction mixtures contained 0.4 µM of each primer, 12.5 µL of the respective SYBR green mix, 25.0 µg bovine serum albumin (Amersham Bioscience, Buckinghamshire, United Kingdom), 1.0 µL of 1:10-diluted DNA extract, and RNase/DNase-free water to a final volume of 25 µL.
Duplicate PCR was performed on DNA extracted from each of the 51 topsoil samples and 27 of the subsoil samples. Standards for the quantitative PCR were prepared using the phenoxy acid degrader Ralstonia eutropha AEO106, which carries the pRO101 plasmid. After inoculation into 0.5 g topsoil in amounts of 8 x 106, 8 x 105, 8 x 104, 8 x 103, 8 x 102, 8 x 101, and 8 x 100 cells g soil–1, the DNA was extracted from the soil as described previously. The extracts were diluted 10-fold to reduce the effect of humic acid disturbances. The DNA extracted from Pseudomonas cepacia DBO1 (pRO101) (Harker et al., 1989) was used as a positive control for the real-time PCR reaction.
Real-time PCR was performed using an iCycler iQ (Bio-Rad, Hercules, CA) under the following conditions: 6 min at 95°C; 50 cycles of 45 s at 94°C, 30 s at 64°C, and 2 min at 72°C; 6 min at 72°C. To obtain a specific melting profile of the real-time PCR products, an 80-cycle protocol of 45 s at 58°C; 30 s at 58°C to 98°C, increasing 0.5°C every cycle; and 45 s at 58°C was performed after the real-time PCR. The melting profile was used to confirm the presence of the specific product and to determine whether the PCR product was composed of DNA sequences of more than one length. The size of the PCR products was confirmed by gel electrophoresis of 8 µL PCR product on a 1.5% agarose gel in 1x Tris-acetate-EDTA buffer. The gels were stained in ethidium bromide and visualized under UV light.
Principal Component Analysis
The correlation between real-time PCR cycle threshold (Ct) and soil properties and MCPA mineralization was investigated by partial-least-squares regression (Espensen, 2002) using MatLab (Mathworks Inc., 2006) with the PLS-toolbox software (Eigenvector Research Inc., 2006). The partial-least-squares regression models were validated using cross-validation with the leave-one-out procedure. Outliers were identified based on their influence on the model (leverage) together with their prediction residuals. The following variables were included in the principal component analysis: accumulated 14C mineralization on all sampling days, Corg, soil pH, clay content, total CFU, pseudomonad CFU, microbial activity parameters (ASA, FDA, and SIR), sorption Kd, and tfdA real-time PCR Ct.
|
Results |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and MethodsResultsDiscussionConclusionREFERENCES |
|---|
The microbial biomass (CFU and SIR) and activity (ISR, ASA, and FDA) decreased by one to two orders of magnitude from the topsoil to the subsoil (Vinther et al., 2007), and the CV of these parameters was considerably higher in the subsoil than in the topsoil (98–465% and 25–255%, respectively). In general, the spatial distribution of the above microbiological parameters (ISR, SIR, ASA, FDA, total CFU, and pseudomonad CFUs) corresponded to the spatial distribution of clay and Corg, as described in detail by Vinther et al. (2007).
MCPA Mineralization Experiments
The accumulation of 14CO2 in all topsoil samples followed a sigmoidal curve, comprising a lag phase with little or no mineralization, a steep segment with rapid MCPA mineralization, and a plateau after mineralization had ceased.
The mineralization of MCPA varied little between topsoil samples, all of which exhibited a lag phase of <7 d and a similar maximal mineralization rate of approximately 2% d–1 (Fig. 1A ). The plateau was reached approximately 20 d after addition of MCPA. By the end of the incubation period (Day 67), 49.4 to 62.2% of the MCPA had mineralized (Table 2 ).
|
|
Total accumulated mineralization of MCPA was >50% in all of the topsoil samples and most of the subsoil samples. 2-Methyl-4-chlorophenoxyacetic acid mineralization potential did not correlate with any of the soil parameters or microbiologic activity parameters tested (Table 3 ).
|
|
In the topsoil (Ap horizon), Kd for MCPA sorption to soil correlated significantly with clay content and Corg content (r = 0.60 and 0.93, respectively) (Table 3). In the subsoil (Bs horizon), in contrast, Kd for MCPA correlated significantly with Corg content (r = 0.83) but not with clay content. In the Ap and Bs horizons, Kd correlated negatively with soil pH (r = –0.54 and –0.57, respectively).
From the contour maps of the spatial distribution of the MCPA sorption (Fig. 2), it can be seen that Kd was highest in the northwestern corner of the field and lowest toward the central and southern part of the field in the Bs horizon. Because Kd correlated strongly with Corg content, the contour maps for Kd are similar to those for Corg presented in Vinther et al. (2007).
Real-Time PCR of tfdA Genes
Real-time PCR analysis revealed a low background level of tfdA genes in the subsoil (2.5 x 102 to 2.0 x 104 tfdA gene copies g soil–1) and an even lower level in the topsoil. In view of this discrepancy, it is suspected that the measurements of background level in the topsoil are biased by PCR-inhibiting substances in the soil DNA extracts. Thus, although the results show the intrafield variation in the natural tfdA gene content within each horizon, the results from the two horizons are not directly comparable. In view of this, we decided to present the real-time PCR results in terms of the Ct for PCR amplification of tfdA gene templates rather than converting these values to the absolute number of tfdA genes. The conversion of Ct to the absolute number of tfdA genes (assuming one copy per cell) is expressed by the equation y = –4.17x + 44.41, where y is Ct, –4.17 is the slope constant, and 44.41 is the y intercept. The formula is derived from the real-time PCR standard curve (Fig. 3
).
|
|
Principal Component Analysis
Principal component analysis did not reveal any parameters that significantly accounted for the variation in MCPA mineralization in the Ap horizon samples. With the Bs horizon samples, principal component analysis was able to explain 77% of the variation in Ct using 95% of the variation in the following six parameters (in descending order of impact): accumulated MCPA mineralization at Day 15, Corg, pH, accumulated mineralization at Day 39, CFU count on TSA, and pseumonad CFU count on Gould S1 agar. Three identified outliers were removed from the data matrix. The ability of the model to predict Ct is illustrated in Fig. 5
.
|
|
Discussion |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and MethodsResultsDiscussionConclusionREFERENCES |
|---|
MCPA Mineralization
Mineralization of MCPA in the topsoil (Ap horizon) was in good accordance with previous studies on mineralization of this phenoxyacetic acid herbicide in agricultural soils. The studies report a short lag phase of 1 to 10 d (Crespin et al., 2001; Jensen et al., 2004), a high maximal MCPA mineralization rate (Baelum et al., 2006), and high total accumulated mineralization when the plateau phase is reached (Helweg, 1987; Jensen et al., 2004; Thorstensen and Lode, 2001).
High intrafield spatial variation in pesticide degradation rate in topsoil has been reported for the insecticide carbofuran (Parkin and Shelton, 1992), the triazine herbicides metribuzin and simazine (Walker and Brown, 1983), and the phenylurea herbicides isoproturon (Beck et al., 1996; Bending et al., 2001; Bending et al., 2003; Parkin, 1993; Walker et al., 2001; Walker et al., 2002) and chlorotoluron (Walker et al., 2002).
Intrafield spatial variation in MCPA mineralization has not previously been studied in detail. Our finding that mineralization of MCPA exhibits very little spatial variation in topsoil is not unexpected because MCPA is considered to be an easily degradable herbicide.
The high rate of MCPA mineralization and the sigmoidal nature of the mineralization curve indicate growth and multiplication of the bacteria initially present in the soil samples. Because the mineralization of MCPA follows approximately first-order kinetics after an initial lag phase in all topsoil samples, it is presumably linked to growth of the degrading organisms. The presence of the phenoxyacetic acid herbicide 2,4-D in concentrations as low as 50 µg L–1 has been shown to cause limited growth of a 2,4-D degrader in soil (Jacobsen and Pedersen, 1992). In the present study and in previous studies, we have rarely found total mineralization rates exceeding 70% (Baelum et al., 2006; Mortensen and Jacobsen, 2004), indicating that a high proportion of the 14C in the MCPA assimilated by the soil microorganisms is converted to new microbial biomass. This provided further evidence for growth-linked degradation of MCPA.
The mineralization potential of most samples from the Bs horizon was high. This indicates that the subsoil microbiota must have been exposed to phenoxyacetic acid herbicides previously because the ability to degrade them is acquired through repeated exposure and resultant adaptation of the bacterial community (Robertson and Alexander, 1994). The latter authors found that degradation of 2,4-D is growth linked, with approximately 10% of the mineralized carbon being incorporated into microbial biomass, thereby confirming the original finding of Jensen and Petersen (1952) that repeated application of the herbicide enhances the degradation rate due to acclimatization of the phenoxyacetic acid herbicide-degrading microbial population.
The number of MCPA-degrading bacteria in the subsoil samples is likely to be less than 10% of that in the topsoil samples due to the general oligotrophic conditions, as is reflected by the total soil bacteria CFU on TSA and the pseudomonad CFU. The longer lag phase of MCPA mineralization in the subsoil samples is probably attributable to the low bacterial density caused by the oligotrophic conditions because mineralization of MCPA is based on growth of the degrading community.
The spatial variation in MCPA mineralization potential in the Bs horizon was high, but no correlation was found to any of the parameters tested. The mineralization patterns seen with subgroups A and B of the subsoil samples suggest that mineralization is based on growth and multiplication of the degrading bacteria, whereas that seen with subgroup C indicates co-metabolic mineralization or limitation of the degrading organisms by lack of nutrients or enhanced predation. This is supported by the findings of Bending et al. (2001) that areas where isoproturon degrades rapidly are characterized by higher 14C-isoproturon mineralization and higher 14C microbial biomass (indicating growth-linked degradation) than areas where it degrades slowly.
In a field study, Gonod et al. (2003) found that 2,4-D mineralization potential was extremely heterogeneous among millimeter-size soil aggregates and that intrafield spatial variation was high at the microhabitat scale. In a later study of the variability of potential aerobic microbial 2,4-D mineralization at spatial scales ranging from field to microhabitat level (cm-scale) in a cultivated topsoil (Gonod et al., 2006), it was found that 2,4-D mineralization was spatially structured in hotspots at the microhabitat (cm) scale, with the variation increasing from field to meter scale, and further from meter to centimeter scale. The authors attributed this variation to the presence of organic compounds supporting growth and co-metabolism of the 2,4-D in the degrading microorganisms. Our results based on sampling distances of 5 or 35 m in the 3-ha sampling grid suggest that MCPA degradation potential is homogenously distributed in topsoil at the field level. The distribution was also homogenous at a smaller scale because there was little variation between triplicate 10-g subsamples of each topsoil sample.
Heterogeneous herbicide exposure of the underlying subsoil due to preferential streaming of drainage water could explain the intrafield variation in MCPA mineralization seen in the Bs horizon samples. The mineralization of MCPA has previously been shown to vary between triplicate soil samples from the same B horizon sampling point (Baelum et al., 2006). In the present study the soil samples were very carefully mixed on collection, and the variation between triplicate samples was found to be very low for all but 6 of the 51 Bs horizon samples.
Because no correlation was found between MCPA mineralization and sorption or soil pH, biodegradation of MCPA at the study site is not determined by these parameters.
In a parallel study by Vinther et al. (2007), mineralization and sorption of the pesticides glyphosate, metribuzin, and triazinamin were determined in the same field samples used in the present study. It was concluded that the spatial distribution of 14C mineralization in the Ap horizon differed for the three pesticides because the mineralization rate was high only for glyphosate. Mineralization of glyphosate correlated positively with all measured soil and microbiological parameters, whereas mineralization of metribuzin and triazinamin generally correlated negatively with these parameters. The positive correlation between mineralization of glyphosate and the measured soil and microbiological parameters was assumed to be due to the strong affinity of glyphosate to clay particles and humus components in the soil, which again would lead to close contact between the glyphosate and the degrading organisms present in the soil (which degrade glyphosate co-metabolically). Degradation of metribuzin and triazinamin is also thought to occur by co-metabolism, although degradation was very low, with a maximum of 3.1% (Vinther et al., 2007). In the present study, degradation of MCPA in the topsoil seemed to support growth of the degrading microorganisms in the topsoil, whereas that in the subsoil followed growth-linked kinetics or occurred co-metabolically, thereby resulting in slower degradation kinetics. The intersample variation in subsoil MCPA mineralization could be attributable to differences in exposure of the subsoil to phenoxyacetic acid herbicides because exposure has probably been limited to certain parts of the soil matrix.
CFU Counts and Microbial Activity
Mineralization of MCPA did not correlate with soil microbial activity or CFU counts. This finding is in line with the results of the parallel analysis of the spatial variation in metribuzin and triazinamin mineralization at the same site (Vinther et al., 2007); as in the present study, no significant correlation was found to any of the microbiological parameters tested. However, variation in glyphosate mineralization was found to correlate positively with the microbial activity parameters SIR, ASA, FDA, and ISR (Vinther et al., 2007). The sigmoidal nature of the MCPA mineralization curves for all topsoil samples (Fig. 1) indicates that MCPA is mineralized by a population of microorganisms that are able to use MCPA for growth. We have previously seen growth of microbial degraders during mineralization of MCPA in a sandy soil (Baelum et al., 2006).
Soil Sorption of MCPA
The MCPA sorption (expressed as Kd) found in this study lies within the range previously reported for MCPA (Helweg, 1987; Socias-Viciana, 1999). Variation at the field scale was slightly lower in the Ap horizon (CV 49%) than in the Bs horizon (75%). Variation in Corg content was also higher in the Bs horizon. In general, variation in all the parameters tested tended to be greatest in the undisturbed soil layers of the field. The sorption of MCPA was highly significantly correlated to Corg content in both horizons and to clay content in the Ap horizon.
Soil organic carbon and soil clay and mineral composition have been shown to influence adsorption of MCPA and 2,4-D (Ogram et al., 1985; Thorstensen et al., 2001). In a long-term field experiment on soils amended with different types and amounts of organic matter, Haberhauer et al. (2001) found that the origin of the soil organic matter seems to be crucial for the sorption behavior of MCPA and that sorption is inversely correlated to pH.
Examining 10 soils differing in clay and Corg content, Bolan and Baskaran (1996) found that Kd for 2,4-D increased with increasing soil Corg, whereas the rate of degradation decreased. At a soil Corg content exceeding 120 g kg–1, however, Kd and degradation rate increased, suggesting enhanced microbial activity in these soils. In a sandy soil, MCPA mineralization was found to correlate strongly with Kd after the addition of artificial sorbents (crushed peat and activated carbon) (Jensen et al., 2004). The authors concluded that it is important to take the bonding strength of MCPA into consideration when estimating pesticide degradation in soil.
Acidic pesticides sorb to organic soil colloids in a pH-dependent manner, with sorption being greatest under acidic conditions where the pesticides are sorbed in their neutral form (Weber, 1972). At pH of 5.4 to 6.2 (the mean values found in the present study), MCPA is anionic (pKa = 3.07), and electrostatic repulsion between the anionic herbicide and negatively charged soil particles could reduce sorption (Helweg, 1987).
Walker et al. (1989) measured sorption and degradation rates of chlorsulfuron and metsulfuron-methyl in soils taken from different depths and found sorption of both herbicides to be inversely correlated with soil pH and positively correlated with Corg content. They suggested that soil pH was the main determinant of sorption in most soils. Several authors have suggested that soil pH is a determinant of herbicide biodegradation because pH tends to be higher at sites where the degradation rate is particularly high than at sites where it is slower (Cox et al., 1996; Walker et al., 2001).
In the present study, Kd for MCPA correlated negatively with pH (CV, –0.54 and –0.57 in the Ap and Bs horizons, respectively). Our detailed study of one field supports the overall findings that Corg content and soil pH are the main determinants of MCPA sorption in sandy soils (Haberhauer et al., 2001; Thorstensen et al., 2001).
Real-Time PCR of tfdA Genes
Our initial hypothesis that high background tfdA levels would be reflected in a short lag-phase before onset of mineralization was not confirmed. Real-time PCR analysis of tfdA genes involved in MCPA catabolism revealed only a weak correlation between a high background level of tfdA genes and a high initial MCPA degradation rate in individual samples. Total MCPA degradation at the end of the 67-d incubation period was not linked to the initial presence of tfdA genes in the samples. Thus, although biodegradation of MCPA at field scale is high in samples with a high initial level of tfdA genes, the fact that the initial level of tfdA genes is low does not preclude mineralization potential being high.
The tfdA primers used to enumerate the background level of tfdA genes were designed on the basis of tfdA-I and tfdA-III type sequences retrieved from GenBank (Baelum et al., 2006). The lack of correlation between the tfdA gene level in a sample and its MCPA mineralization potential may be due to the initial presence of other classes of tfdA genes because the identified tfdA genes vary (Fulthorpe et al., 1995; Smejkal et al., 2001). Whether or not all the tfdA gene alleles that mediate MCPA catabolism are targeted in the present real-time PCR analysis is unclear.
The presence or absence of other genes involved in the metabolic pathway of MCPA probably explains the discrepancy between the tfdA gene level in a sample and its MCPA mineralization potential. Both 2,4-dichlorophenol and dichlorocatechol could have been produced and sorbed to the soil because the genes necessary for metabolism of these intermediary products were not present within the microorganisms in the samples. Thus, the background level of tfdA-I and tfdA-III genes may not reflect the full potential of a soil sample to mineralize MCPA.
Soil microcosm experiments on sandy soils from the Danish outwash plain in southern Jutland have recently shown that formation of the phenolic metabolite MCP in topsoil is associated with the presence of tfdA genes belonging to the tfdA Class I. After a lag phase, mineralization of the MCPA ring structure was initiated by the multiplication of organisms carrying the tfdA Class III genes (Baelum et al., 2006). From analysis of tfdA functional genes, Baelum et al. (2006) and de Lipthay et al. (2002) concluded that the number of copies of a catabolic gene in soil before contaminant exposure might not necessarily reflect the size of the specific degrader population. In the present study, the DNA was extracted from the soil before application of MCPA. Thus, the results do not reflect the likely increase in the signal caused by growth of microorganisms harboring tfdA genes.
It is known that a population of tfdA gene bearing bacteria can be maintained even if a soil is not exposed to phenoxyacetic acid herbicides for a period of many years (Baelum et al., 2006). The present study shows that tfdA genes are present in detectable numbers in topsoil and subsoil samples throughout an agricultural field.
|
Conclusion |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and MethodsResultsDiscussionConclusionREFERENCES |
|---|
Spatial variation in MCPA sorption was low in the topsoil and the subsoil, with Kd ranging from 0.36 to 4.16 L kg–1. Sorption correlated strongly with soil Corg in both horizons and negatively with soil pH. No significant correlation could be established between MCPA sorption and MCPA mineralization potential.
We conclude that MCPA mineralization potential is high and homogenously distributed in the topsoil (Ap horizon) but exhibits considerable variation in the subsoil (Bs horizon). This is the first study to report that MCPA mineralization potential varies considerably in subsoil. We suggest that this variation is attributable to differences in prior exposure of the subsoil to phenoxyacetic acid herbicides because this is limited to certain parts of the soil matrix.
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and MethodsResultsDiscussionConclusionREFERENCES |
|---|
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and MethodsResultsDiscussionConclusionREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Soil Science Society of America Journal | Journal of Plant Registrations | The Plant Genome | |||
