Dependence of atrazine degradation on C and N availability in
adapted and non-adapted soils
Rahima Abdelha®d, Sabine Houot*, Enrique Barriuso
I.N.R.A., Unite de Science du Sol, BP 01, 78850 Thiverval-Grignon, France Accepted 23 August 1999
Abstract
Atrazine degradation in soil was aected by microbial adaptation and C and N availability. Accelerated atrazine mineralization was observed after repeated applications to a soil under continuous maize (adapted soil) while atrazine mineralization remained slow in an adjacent soil under wheat that had never received atrazine (non-adapted soil). Carbon-14 atrazine degradation and formation of unextractable `bound' residues were determined during laboratory incubations in soil alone or amended with dierent organic amendments (OAs) and N sources. The OAs varied from readily biodegradable (glucose) to more slowly mineralizable (cellulose and straw) and humi®ed organic matter (compost). The N forms included mineral (NH4NO3, (NH4)2SO4and Ca(NO3)2) and organic forms (adenine, arginine, albumin, biuret and pyrazine) which varied
in N availability. In the adapted soil, OA addition had little eects on atrazine degradation, whereas in the non-adapted soil, it stimulated atrazine dealkylation more than triazine ring mineralization which always remained lower than in the adapted soil. In both soils, mineral N decreased triazine ring mineralization. The depressive eect of the organic N forms on atrazine mineralization increased with their N mineralization rate. Despite its slow N mineralization rate, the addition of biuret greatly decreased atrazine mineralization, possibly because it is one of the last intermediates in atrazine degradation. The proportion of bound residues increased with the total microbial activity after addition of OAs or organic N forms. In conclusion, rapid triazine ring mineralization was dependent on micro¯ora adaptation after repeated atrazine application and was mainly regulated by N availability in soil.#2000 Elsevier Science Ltd. All rights reserved.
1. Introduction
Microbial degradation is the principal mechanism of atrazine [6-chloro-N2-ethyl-N4 -isopropyl-1,3,5-triazine-2,4-diamine] dissipation from the environment (Esser et al., 1975). Although this herbicide has been termed as recalcitrant (Kaufman and Kearney, 1970), a large variety of soil microorganisms are known to degrade atrazine partially by N-dealkylation or dehalogenation (Kaufman and Kearney, 1970; Behki and Khan, 1986; Mougin et al., 1994; Bouquard et al., 1997). Complete and rapid mineralization of the triazine ring has been
reported (Mandelbaum et al., 1993; Assaf and Turco, 1994; Yanze-Kontchou and Gschwind, 1994; Mandel-baum et al., 1995; Radosevich et al., 1995) and the possible adaptation of soil micro¯ora to atrazine degradation after repeated ®eld applications has been demonstrated (Barriuso and Houot, 1996; Vanderhey-den et al., 1997). Microbial growth has been observed with atrazine as sole C source (Behki and Khan, 1986; Yanze-Kontchou and Gschwind, 1994; Stucki et al., 1995). Rapid triazine ring mineralization however seems to imply the development of microorganisms using triazine nitrogen as a N source (Cook and HuÈt-ter, 1981; Mandelbaum et al., 1995; Radosevich et al., 1995).
The addition of organic amendments (OAs) to soils can modify the rate and pathways of pesticide degra-dation. Pesticides generally sorb readily to organic
Soil Biology & Biochemistry 32 (2000) 389±401
0038-0717/00/$ - see front matter#2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 3 8 - 0 7 1 7 ( 9 9 ) 0 0 1 6 7 - 4
www.elsevier.com/locate/soilbio
* Corresponding author. Tel.: 5401; fax: +33-1-3081-5396.
matter. Thus OAs promoting sorption reduce pesticide bioavailability and slow their biodegradation (Barriuso et al., 1997). On the other hand, OAs may accelerate pesticide degradation through a general stimulation of the microbial biomass (Hance, 1973). Nitrogen avail-ability in¯uences atrazine behaviour in soils also and the eect varies with the form and amount of N pre-sent. A large concentration of mineral N greatly decreased atrazine mineralization in soil when added alone or with organic amendments (Entry et al., 1993; Alvey and Crowley, 1995). On the other hand, dairy manure including a large proportion of organic N stimulated atrazine mineralization (Topp et al., 1996). In this latter case, the greater availability of nutrients allowed a larger number of atrazine degraders to mul-tiply in the amended soil.
Our purpose was to test the eect of C and N avail-ability on atrazine degradation in two soils: in one soil, enhanced mineralization of atrazine was observed and related to repeated ®eld applications of atrazine; the other soil had similar physicochemical character-istics, but had never received atrazine under ®eld con-ditions and atrazine mineralization remained slow in this soil (Barriuso and Houot, 1996). The dissipation of14C-atrazine via degradation and formation of unex-tractable, so-called `bound' residues was followed during laboratory incubations in soils alone or amended with dierent OAs or N forms. The OAs and the N compounds were selected according to their C or N availability, respectively. Their connection to the
atrazine degradation pathway was another selecting criteria for the N compounds.
2. Material and methods
2.1. Soils
Two adjacent experimental plots located in Grignon (Yvelines, France) were sampled in the upper 20 cm in April 96 and February 97, respectively for the two sets of incubation. Both soils were Typic Eutrochrept (Table 1). One plot has been cultivated with continu-ous maize since 1962 and has been treated yearly with atrazine at recommended agronomic doses. The other plot has been under continuous wheat since 1965 and has never received atrazine. Soils were immediately sieved (5 mm) after sampling and used for experiments without storage.
2.2. Chemicals, organic amendments and N forms
Analytical standards of atrazine and its metabolites (hydroxyatrazine, deethylatrazine, deisopropylatrazine, deethyl-deisopropylatrazine) were purchased from ChemService (West Chester, PA, USA). The [U-ring- 14-C]atrazine (speci®c activity 659 MBq mmolÿ1; radio-purity greater than 97%) was purchased from Amersham (Buckinghamshire, UK).
Four OAs were compared (Table 2): glucose,
cellu-Table 1
Main physicochemical characteristics of the soils
Clay (g kgÿ1) Silt (g kgÿ1) Sand (g kgÿ1) Organic C (g kgÿ1) Total N (g kgÿ1) CaCO3(g kg
ÿ1
) pH (water)
Continuous maize plot 253 623 91 11.5 1.2 33 8.2
Continuous wheat plot 241 445 171 17.4 1.7 61 8.0
Table 2
Organic amendments (OAs) and N compounds: C and N contents and amount added gÿ1of soil, calculated to approximately double the C con-tent of the soil and to bring 2.5 mg of N gÿ1of soil in all the treatments, respectively
OAs and N sources C (mg kgÿ1) N (mg kgÿ1) Added C (mg kgÿ1)
Glucose 400.0 0.0 16.0
Cellulose 440.0 0.0 17.6
Straw 416.0 4.2 16.6
Compost 187.0 9.8 9.4
Ammonium nitrate 0.0 350.0 0.0
Calcium nitrate 0.0 170.6 0.0
Ammonium sulfate 0.0 211.9 0.0
Arginine 413.3 321.5 3.2
Albumin 483.8 163.9 8.0
Adenine 444.1 518.1 2.13
Biuret 232.8 407.5 1.40
Pyrazine 599.3 349.6 4.31
lose, wheat straw and a municipal solid waste com-post. The cellulose was purchased from Sigma (St Louis, MO, USA). The wheat straw came from the ex-perimental farm of Grignon. It was ground in a plant-blender to a 2 mm maximum particle size. The com-post was obtained after 12 d of accelerated fermenta-tion of municipal solid wastes and 6 weeks of maturation. It was air dried and sieved (2 mm); the smaller fraction was used.
Eight sources of N were used (Table 2): three min-eral forms (NH4NO3, Ca(NO3)2 and (NH4)2SO4), an
amino-acid (arginine), a protein (albumin), the biuret and two N-heterocycles, a purine (adenine) and a dia-zine (pyradia-zine).
2.3. Incubation experiments
Two sets of laboratory incubations of 14C-atrazine with fresh soil samples equivalent to 10 g of dry soil were conducted in triplicate and in hermetically stop-pered jars at 28218C for 50 d.
In the ®rst incubation experiment, 14C-atrazine degradation was followed in soil with or without ad-dition of OA and with or without adad-dition of mineral N. Cellulose, straw and compost were added in pow-der and glucose and mineral N in solution. The amount of OA was calculated in order to approxi-mately double the organic C content of the soil in the mixtures: 400 mg of straw and cellulose, 400 mg of glucose (0.8 ml of a solution at 500 g lÿ1) and 500 mg of compost (Table 2). Mineral N was added to half of the jars corresponding to 25 mg of N (0.8 ml of a sol-ution of NH4NO3at 90 g lÿ1). Atrazine was added to
each jar to a concentration of 0.5 mg kgÿ1 of soil (0.8 ml of a solution of 14C-atrazine prepared in water at 6.25 mg lÿ1 and 9.94 MBq lÿ1), corresponding to the dose applied in ®eld conditions. The water content of the mixtures was adjusted to 85% of the soil ®eld ca-pacity with MilliQ water (Millipore, Milford, MA, USA) taking into account the water added with the solutions of atrazine, glucose or mineral N. Similar in-cubations were carried out with 400 mg of cellulose or straw and 500 mg of compost alone receiving the same amount of atrazine and at a humidity equivalent to 85% of their water holding capacity.
In the second incubation experiment, 14C-atrazine degradation was followed in soil with or without ad-dition of the dierent N-forms. In each treated jar, 0.6 ml of water solutions of either Ca(NO3)2at 250 g lÿ1,
(NH4)2SO4at 200 g lÿ1, arginine at 128 g lÿ1, albumin
at 277 g lÿ1or pyrazine at 120 g lÿ1was added, corre-sponding to 25 mg of N (Table 2). Adenine and biuret were not soluble enough in water and 60 mg of biuret and 48 mg of adenine were added as powder per jar and mixed to soil. As previously, 14C-atrazine was added to each jar in order to reach a concentration of
0.5 mg kgÿ1 of soil (0.5 ml of a solution of 14 C-atra-zine prepared in water at 10 mg lÿ1 and 19.4 MBq lÿ1). The soil water content was adjusted to 85% of the soil ®eld capacity with the atrazine and nitrogen solutions supplemented with MilliQ water when necessary.
The evolved 14C±CO2 and total C±CO2 were
trapped in 5 ml of 1 M NaOH placed into the incu-bation jars. The traps were sampled periodically and replaced during the incubation. Total C±CO2was
ana-lyzed by colorimetry on a continuous ¯ow analyser (Skalar, Breda, the Netherlands) and 14C±CO2 was
determined by scintillation counting (Kontron Beta-matic V; Kontron Ins., Montigny le Bretonneux, France) using Pico¯uor scintillation cocktail (Packard, Meriden, CT, USA). During the incubations with the dierent N forms, mineral N was extracted after 7, 14 and 50 days with 50 ml of 1 M KCl and analyzed by colorimetry on the continuous ¯ow analyser.
2.4. Analysis of the 14C-atrazine residues
At the end of the incubations, all the samples trea-ted with 14C-atrazine were extracted by shaking in 50 ml of 10 mM CaCl2 in water for 16 h at 20228C in
the dark. After centrifugation at 4000 g for 10 min, the supernatant was removed and the residues extracted again for 16 h by shaking with 50 ml of methanol, three successive times. The extracted radio-activity was measured in the water and the methanol extracts. The non-extractable radioactivity, corre-sponding to the `bound' residues was measured by scintillation counting of the 14C±CO2 evolved after
combustion of the solid residues after methanol extrac-tion (Sample Oxidizer 307, Packard, Meriden, CT, USA). The water extracts were concentrated by solid± liquid extraction with Lichrolut EN (200 mg) car-tridges (Merck, Darmstadt, Germany). The carcar-tridges were eluted with 20 ml of methanol then evaporated until dryness under vacuum with a Rotavapor RE 111 (BuÈchi, Flawil, Switzerland). The residue was then dis-solved in 1 ml of the ®rst solvent used for the HPLC analysis, 40/60 methanol/water (v/v) buered with 50 mM ammonium acetate with pH adjusted to 7.4. The three methanol extracts of the three replications were pooled, then concentrated until dryness by evaporation under vacuum; the residue was then dissolved in 2 ml of the ®rst HPLC solvent. Samples for HPLC analyses were ®ltered through a Cameo 13N syringe nylon ®lter of 0.45 mm (MSI, Westboro, MA, USA). 14C-atrazine and 14C-metabolites were analyzed by HPLC on a Novapak C18 column (5 mm, 250 mm 4.6 mm; Waters, Milford, MA, USA) with a Waters instrument (600E Multisolvent Delivery System, 717 Autosampler) equipped with a UV detector at 222 nm, coupled on-line with a radioactive ¯ow detector
matic Flo-one A550). The mobile phase was methanol/ water buered with 50 mM ammonium acetate with pH adjusted to 7.4. The chromatography started with 40/60 methanol/water (v/v) for 15 min, then 80/20 methanol/water for 20 min. The mobile phase ¯ow was 1.0 ml minÿ1and the injected sample volume
var-ied from 300 to 800 ml, function of the total radioac-tivity detected in the extracts.
2.5. Sorption experiments
Ten millilitres of a solution of 14C-atrazine at 10 mg
Fig. 1. Eects of organic amendments and mineral N on total C±CO2mineralization during incubation in soil from the continuous maize plots. Results are expressed in g C±CO2kgÿ1soil. Standard deviations are indicated only where larger than symbols.
Fig. 2. Kinetics of total C±CO2 mineralization and total mineral N evolution during incubation with soil from the continuous maize, sup-plemented with dierent N sources. Results are expressed in mg C±CO2kgÿ1soil and in mg N kgÿ1 soil, respectively. Standard deviations are indicated where larger than symbols.
lÿ1 in 10 mM CaCl2and 0.11 MBq lÿ1were added to
5 g of air-dried soil alone or soil-OA mixtures (in the same proportions as in the incubation experiments: 200 mg straw or cellulose, 250 mg compost) into 25 ml Corex glass centrifuge tubes with Te¯on caps. Sorption experiments were also conducted with the OAs alone (same amounts as previously). Triplicate samples were carried out. After shaking for 24 h at 20218C, the samples were centrifuged at 7000 g for 20 min and atrazine concentration in solutions (Ce, mg lÿ1) was
calculated from the supernatant radioactivity measure-ment. The amount of sorbed atrazine (S, mg kgÿ1) was calculated from the dierence of atrazine concen-tration in solution before and after sorption. The sorp-tion coecient Kd (l kgÿ1) was calculated as
KdS=Ce:The sorption coecient on an organic
car-bon unit basis, Koc, was calculated as
Koc100Kd=C, where C is the organic C content
(%).
3. Results and discussion
3.1. Total heterotrophic microbial activity
Total microbial activity was evaluated from the kin-etics of total C mineralization. Both soils exhibited similar heterotrophic activity and only the results of the continuous maize plot are presented (Figs. 1 and 2). For both soils, 3 to 5% of the total organic C was mineralized after 50 d of incubation (soil alone). During incubations of the soil-OA mixtures, the
Fig. 3. Impact of organic amendments and mineral N on 14C-atrazine mineralization during incubation in soil from the continuous maize (adapted soil) or wheat plot (non-adapted soil). Results are expressed as percent of the initial radioactivity. Standard deviations are indicated only where larger than symbols.
Table 3
Distribution of the initial radioactivity from14C-atrazine into the mineralized, extractable (water plus methanol extracts combined) and bound residue fractions after 50 d of incubation in the adapted soil from the continuous maize plot receiving the dierent OAs with or without mineral N. The percentages of initial radioactivity present as atrazine, hydroxyatrazine (HYA) and deal-kylated metabolites (deethyl-atrazine, deisopropyl-atrazine and deethyl-deisopropyl-atrazine) in the extractable fraction are given
Adapted soil Mineralized CO2 (% of initial14C-atrazine)
Extractable residues Bound residues
(% of initial14C-atrazine) total
(% of initial14C-atrazine)
atrazine
(% of initial14C-atrazine) HYA
(% of initial14C-atrazine)
dealkylated
(% of initial14C-atrazine)
Without mineral N
Soil 90.322.3 aa 3.820.3 a 0.0b 0.3 3.5 6.720.1 a
Soil+straw 90.424.8 a,b 4.020.4 a 1.3 0.0 2.7 7.020.5 a
Soil+compost 83.922.0 b 5.420.5 b 0.0 0.2 5.1 10.520.6 b
Soil+cellulose 87.721.6 a,b 5.320.7 a,b 0.0 0.5 4.7 6.920.2 a
Soil+glucose 74.222.7 d 7.821.6 b 0.0 2.8 4.9 9.520.1 b
With mineral N
Soil 53.522.1 e 28.322.6 d 0.0 2.5 25.8 17.720.4 c
Soil+straw 59.222.0 c 19.421.6 c 0.8 0.0 18.6 19.821.5 c
Soil+compost 63.522.2 c 18.821.8 c 0.2 0.2 18.3 20.020.7 c
Soil+cellulose 31.522.7 f 41.122.9 e 0.7 0.0 40.3 25.720.2 d
Soil+glucose 8.922.4 g 58.124.4 f 0.0 12.3 45.1 30.620.4 e
aMeans within a column followed by the same letter do not dier signi®cantly P0
:05according to Student's test. bVariance of HPLC analysis could not be given since the three replicates were combined before analysis.
R.
Abdelha®d
et
al.
/
Soil
Biology
&
Biochemistr
y
32
(2000)
389±401
Table 4
Distribution of the initial radioactivity from14C-atrazine into the mineralized, extractable (water plus methanol extracts combined) and bound residue fractions after 50 d of incubation in the non-adapted soil from the continuous wheat plot receiving the dierent OAs with or without mineral N. The percentages of initial radioactivity present as atrazine, hydroxyatrazine (HYA) and dealkylated metabolites (deethyl-atrazine, deisopropyl-atrazine and deethyl-deisopropyl-atrazine) in the extractable fraction are given
Non adapted soil Mineralized CO2 (% of initial14C-atrazine)
Extractable residues Bound residues
(% of initial14C-atrazine) total
(% of initial14C-atrazine)
atrazine
(% of initial14C-atrazine) HYA
(% of initial14C-atrazine)
dealkylated
(% of initial14C-atrazine)
Without mineral N
Soil 2.820.3 aa 65.321.9a 53.8b 1.6 10.0 32.621.3 a
Soil+straw 12.320.5 b 52.422.7b 15.0 1.7 35.6 38.421.5 b
Soil+compost 6.220.7 c 49.523.1b 11.9 2.7 34.9 47.921.0 c,d
Soil+cellulose 12.321.8 b 52.025.6 b 14.1 1.2 36.7 36.820.6 b
Soil+glucose 7.320.3 c 52.927.7 a,b 34.2 1.0 17.6 38.321.4 b
With mineral N
Soil 0.420.1 d 69.524.6 a,c 60.7 2.8 5.7 33.521.5 a
Soil+straw 3.320.2 a 49.422.4 b 1.4 0.5 47.2 50.220.9 c
Soil+compost 0.920.1 e 56.626.3 a,b,c 15.7 2.5 38.2 47.721.6 c,d
Soil+cellulose 1.520.1 f 56.723.0 b 1.7 1.0 53.9 45.821.0 d
Soil+glucose 0.220.01 d 54.024.8 b 12.1 2.8 39.2 45.720.6 d
a
Means within a column followed by the same letter do not dier signi®cantly P0:05according to Student's test. b
Variance of HPLC analysis could not be given since the three replicates were combined before analysis.
R.
Abdelha®d
et
al.
/
Soil
Biology
&
Biochemistr
y
32
(2000)
389±401
increased C mineralization compared to the soil alone was attributed to the OA mineralization, assuming that no interactions occurred between indigenous and exogenous organic matter and corresponded to 2, 14, 22 and 44% of the initially added C for cellulose, straw, compost and glucose, respectively. Mineral N addition did not modify C mineralization in the soils alone (Fig. 1). In contrast, the addition of mineral N increased C mineralization for cellulose, straw and glu-cose, which reached 27, 32 and 61% of the initially added C, respectively. Compost C mineralization was not stimulated by mineral N addition.
Total C mineralization was not aected by Ca(NO3)2 addition but was doubled when (NH4)2SO4
was added (Fig. 2). All the organic forms of N enhanced C mineralization with dierent intensities. The mineralization of the added organic materials was estimated from the increase of C±CO2 evolved during
the incubation as previously. At the end of the incu-bations, 8, 12, 49, 73 and 74% respectively of pyrazine, biuret, adenine, arginine and albumin organic C were mineralized.
The evolution of mineral N (N±NH4++N±NO3ÿ)
was followed in the treatments with the dierent N-forms. Again, the results were similar in the two soils and only the results corresponding to the continuous maize plot are presented (Fig. 2). At the end of the control incubations, 1% of the soil organic N was mineralized in both soils. All the N compounds increased the mineral N concentration in soils, except
pyrazine and biuret which were the least biodegradable organic materials. Mineral N remained constant when Ca(NO3)2or (NH4)2SO4 was added. As for C, N
min-eralization of the organic N compounds was estimated from the increase in mineral N during the incubations of amended soils as compared to the soils alone, assuming that the addition of exogenous organic mat-ter did not modify the indigenous N cycle. With ade-nine, N mineralization reached 26% of the added amount and 45% with albumin and arginine and min-eral N was mainly present as ammonium. With pyra-zine and biuret, only 5% of the initial organic N was mineralized at the end of the incubations, including as much ammonium as nitrate. The availability of C and N in the various organic N compounds increased in the same order: pyrazine<biuret<adenineRalbumin arginine:
3.2. Atrazine degradation in the control soils
During incubation with soil from the continuous maize plot, atrazine mineralization increased rapidly and reached a plateau corresponding to 90% of the in-itially added radioactivity after only 14 d (Fig. 3). At the end of the incubation, 3.8% of the initially added
14
C remained extractable and 6.7% formed non-extrac-table bound residues (Table 3). No atrazine was found in the extractable fraction which mainly included deal-kylated metabolites (Table 3).
During incubation with soil from the continuous
Fig. 4. Kinetics of14C-atrazine mineralization during incubation with adapted and non-adapted soil, supplemented with dierent N sources. Results are expressed as percent of the initial radioactivity. Standard deviations are indicated where larger than symbols.
Table 5
Distribution of the initial radioactivity from14C-atrazine into the mineralized (14C±CO2), extractable (water plus methanol extracts combined) and bound residue fractions after 50 d of incu-bation in the adapted and non-adapted soil receiving the dierent N forms. The percentages of initial radioactivity present as atrazine, unidenti®ed metabolite (M16), hydroxyatrazine (HYA) and dealkylated metabolites (deethyl-atrazine, deisopropyl-atrazine and deethyl-deisopropyl-atrazine) in the extractable fraction are given
14C±CO2(% of initial 14C-atrazine)
Extractable residues Bound residues (% of
initial14C-atrazine) total (% of initial
14C-atrazine)
atrazine (% of initial 14C-atrazine)
M16 (% of initial 14C-atrazine)
HYA (% of initial 14C-atrazine)
dealkylated (% of initial 14C-atrazine)
Adapted soil
Control soil 93.020.6 aa 2.320.1 a 2.0b 0.0 0.0 1.3 6.620.1 a
Soil+pyrazine 88.022.2 b 6.420.5 b 2.2 0.3 0.1 3.8 9.120.4 b
Soil+albumin 22.922.9 c 44.124.5 c 28.0 1.7 0.7 16.7 29.520.1 c
Soil+adenine 24.521.6 c 48.222.0 c 38.3 1.4 3.9 4.1 29.320.4 c
Soil+arginine 9.320.4 d 54.929.4 c,d 43.6 1.8 1.3 3.6 35.620.6 d
Soil+(NH4)2SO4 7.620.5 e 61.022.8 d,f 54.0 1.6 0.7 4.5 31.220.1 e
Soil+biuret 7.320.7 e,f 72.122.3 e 48.8 0.1 6.4 16.8 21.820.2 f
Soil+Ca(NO3)2 6.120.4 f 67.221.9 e,f 57.2 0.0 4.3 5.6 26.420.1 g
Non adapted soil
Control soil 44.222.8 aa 25.623.5 a 9.7 0.0 4.9 11.0 29.220.6 a
Soil+pyrazine 19.421.7 b 34.9213.4 a,b 24.9 0.9 4.7 4.4 29.320.1 a
Soil+albumin 0.320.04 c 56.724.2 b,c 38.1 2.1 6.1 9.8 41.820.1 b
Soil+adenine 0.720.2 d 57.424.9 b,d 43.8 2.6 4.8 5.9 37.721.3 c,d
Soil+arginine 0.120.0 e 60.920.7 b,e 48.5 3.3 3.5 4.9 40.420.6 c
Soil+(NH4)2SO4 0.220.0 f 62.823.2 c,d,e 54.8 1.5 3.2 3.2 37.520.5 d
Soil+biuret 0.620.1 d 67.722.3 d 52.2 0.2 6.4 8.7 31.020.1 e
Soil+Ca(NO3)2 0.120.0 e 67.723.2 d 56.4 0.0 5.5 5.7 32.621.0 e
a
Means within a column followed by the same letter do not dier signi®cantly P0:05according to Student's test. b
Variance of HPLC analysis could not be given since the three replicates were combined before analysis.
R.
Abdelha®d
et
al.
/
Soil
Biology
&
Biochemistr
y
32
(2000)
389±401
wheat plot, atrazine mineralization increased progress-ively throughout all the incubation but remained very low (Fig. 3). At the end of the incubation, 65.3% of the initially added14C remained extractable and 32.6% formed bound residues (Table 4). Atrazine was the dominant compound in the extractable fraction and mainly dealkylated metabolites were detected with very few hydroxyatrazine (Table 4).
These results con®rmed the presence of a microbial community with the capacity to mineralize the triazine ring of atrazine in the continuous maize plot (adapted soil) but not in the continuous wheat plot (non-adapted soil) (Barriuso and Houot, 1996). Dealkyla-tion of the atrazine amino-substituants has been shown to be the major mechanism involved in atrazine microbial degradation and was associated with low rates of triazine ring degradation (Kaufman and Kear-ney, 1970). More recently, rapid atrazine degradation and triazine ring mineralization have been demon-strated and dehalogenation has been shown to be the ®rst step of atrazine degradation in many studies with a consortium of microorganisms or isolated strains (Mandelbaum et al., 1993; Assaf and Turco, 1994; Mandelbaum et al., 1995; Radosevich et al., 1995; de Souza et al., 1998). Alternatively, dealkylation could be the ®rst step of a rapid atrazine degradation and dealkylation and dehalogenation can occur simul-taneously (Yanze-Kontchou and Gschwind, 1994; Shao et al., 1995; Stucki et al., 1995). Both hydroxya-trazine and dealkylated metabolites were detected in the extracts after incubation in the adapted and non-adapted soils. However, microorganisms responsible for triazine ring mineralization were active only in the adapted soil. In the non-adapted soil, bound residue formation constituted an alternative pathway of atra-zine dissipation (Barriuso and Houot, 1996).
During the second set of experiments, similar results were observed during atrazine incubation in the soil from the continuous maize plot (Fig. 4), but atrazine mineralization reached 44.2% in the control soil sampled in the permanent wheat plot. This large and unexpected atrazine percentage of mineralization was probably related to the contamination of the wheat plot by soil of the maize plot since both plots are adja-cent.
3.3. Nitrogen eect on atrazine degradation
All of the N compounds decreased atrazine mineral-ization signi®cantly in both soils. In the adapted soil, the intensity of the decrease varied with the form of N added and atrazine mineralization was negatively cor-related with the mineral N content of the soil in the dierent treatments at the end of the incubations
r0:797, P< 0.05). The largest decreases in atrazine mineralization were observed with the mineral N
forms. With the same concentration of Ca(NO3)2,
Alvey and Crowley (1995) also observed a large inhi-bition of atrazine mineralization. With the organic N compounds, atrazine mineralization decreased when N availability of the molecules increased. Pyrazine was the least available form of N; arginine, albumin and adenine were the most easily degraded molecules (Fig. 2). The only exception was observed with biuret. In spite of the low availability of biuret N, only 7.3% of the initial atrazine was mineralized at the end of the incubation. Biuret is one of the last metabolites in the atrazine degradation pathway (Cook et al., 1985) and large concentrations of this ®nal intermediate could have inhibited the enzymatic transformations occurring in atrazine degradation. Atrazine is used as an N source by atrazine-degrading microorganisms (Mandel-baum et al., 1995; Radosevich et al., 1995) and mineral N addition could inhibit atrazine mineralization by oering an alternative source of N. However, since large concentrations of mineral N are required to ob-serve a negative eect on atrazine mineralization (Alvey and Crowley, 1995), the synthesis or the ac-tivity of the enzymes responsible for triazine ring degradation could also be aected (Entry et al., 1993). Nitrogen has been found to regulate the synthesis or the activity of enzymes like peroxidases (Li et al., 1994; Kaal et al., 1995). The negative eect of N on atrazine degradation was not related to the increase of soil conductivity after addition of large salt concen-tration (Smith and Doran, 1996). Atrazine mineraliz-ation was not decreased when similar concentrmineraliz-ations (10 mM) of CaCl2or Na2SO4 were added (results not
shown).
In the non-adapted soil, less than 1% of the initially added atrazine was mineralized in all the N treatments, except with pyrazine which had little eect on atrazine mineralization as in the adapted soil.
In both soils, the largest residual extractable radio-activities were measured in the treatments which had the largest depressive eect on atrazine mineralization: the mineral N forms and the biuret (Table 5). The ad-dition of N decreased triazine ring mineralization and also atrazine partial degradation. Atrazine was always the dominant compound in the extracts. With all the N compounds, both hydroxyatrazine and dealkylated metabolites were detected. An unidenti®ed metabolite (M16) was detected with albumin, adenine, arginine and (NH4)2SO4. It was chromatographied after 16 min
between hydroxyatrazine (13 min) and atrazine (21 min).
In both soils, the bound residue proportion increased when atrazine mineralization decreased (Table 5) and when the total microbial activity as esti-mated with the total C±CO2 evolved during the
incu-bation, increased. This con®rmed the microbial
contribution to pesticide bound residue formation in soils (Benoit and Barriuso, 1997).
3.4. Atrazine degradation in organic amended soils
Less than 1% of the initial radioactivity was minera-lized when atrazine was incubated with compost, straw or cellulose alone. The proportion of bound residues was larger with compost (68.3% of the initial tivity) than with straw (28.0% of the initial radioac-tivity). Few bound residues were formed with cellulose (0.6% of the initial radioactivity). Atrazine sorption on straw was greater than on compost and was very low on cellulose (Table 6). The larger anity of com-post organic matter than straw organic matter for atrazine, revealed by larger Koc values may partly
explain the larger proportion of bound residues formed with compost. Sorption could be the initial process leading to formation of highly stabilized unextractable residues. With cellulose, most of the radioactivity remained extractable (95.1% of the initial radioac-tivity) and mainly as parent atrazine. With compost, 44.3% of the initial radioactivity remained extractable and atrazine was also the main component of the extract. With straw, 75.5% of the initial radioactivity was extracted after the incubation and only 23.8% remained as atrazine. Mainly dealkylated metabolites were detected (50.6% of the initial radioactivity).
The eect of adding OAs on atrazine degradation diered in the adapted and non-adapted soils. In the adapted soil, the addition of OAs had little eect on atrazine mineralization which decreased signi®cantly only when compost and glucose were added (Fig. 3). With these OAs, bound residue formation slightly increased (Table 3).
Straw and compost increased atrazine retention in both soils, but cellulose did not (Table 6). Greater atrazine sorption can decrease its availability for mi-crobial degradation. This was observed in the adapted soil for compost, con®rming the results of Barriuso et al. (1997), while straw did not modify atrazine mineral-ization. The anity of straw organic C for atrazine was lower than that of soil C, indicated by the smaller
Koc(Table 6). Thus, even if adsorbed on straw organic
matter, atrazine remained available for microbial degradation in the adapted soil where the degrading micro¯ora was very active.
Organic amendments can also increase pesticide degradation by stimulating microbial activity (Entry and Emmingham, 1995; Topp et al., 1996). Indeed, in the non-adapted soil, all the OAs stimulated both total microbial activity and atrazine mineralization. How-ever both activities could not be directly related since the dierent OAs did not have the same eect on total CO2 and 14CO2 evolution, con®rming the results of
Alvey and Crowley (1995). OA addition also increased bound residue formation in the non-adapted soil and the largest proportion of bound residues was observed with compost (Table 4), probably because of its greater anity for atrazine than the other OAs (Table 6).
In the adapted soil, the addition of OAs little modi-®ed the nature of the extractable radioactivity with dealkylated metabolites as dominant compounds (Table 3). Atrazine was only detected after incubation with straw and represented only 1.3% of the initial radioactivity. The largest proportion of hydroxyatra-zine was observed with glucose (36.5% of the extracta-ble radioactivity). It could not be related to an increase in chemical hydrolysis accompanying acidi®ca-tion (Khan, 1978). The pH in the glucose-amended soil incubation remained above 8.0 and did not decrease as observed by Alvey and Crowley (1995).
In the non-adapted soil, OA addition increased the partial degradation of atrazine. The extractable atra-zine represented 53.8% of the initial radioactivity at the end of the control incubation and only 34.2% of the initial radioactivity when glucose was added and 12±15% with the other OAs (Table 4). Mainly dealky-lated metabolites were detected in the extracted radio-activity with all of the OAs. The largest increases in atrazine dealkylation and mineralization were observed when straw or cellulose were added which could mean that cellulolytic microorganisms could be active in atrazine degradation.
When mineral N was added simultaneously with the OAs, atrazine mineralization decreased signi®cantly as compared to the same treatment without mineral N (Fig. 3), con®rming the inhibiting eect of mineral N on atrazine mineralization previously described. In the adapted soil, the addition of OAs and mineral N dif-ferentiated the eects of the OAs (Table 3). Atrazine
Table 6
Distribution coecientsKdandKocfor atrazine sorption on organic amendments, soils and soil-organic amendment mixtures
Kd(l kgÿ1) Koc(l kgÿ1C) Soil+cellulose 0.5820.05 d 20.621.7 d Continuous wheat plot
Soil alone 0.7720.04 e 44.222.2 b Soil+straw 1.3320.03 c 40.821.0 a, b Soil+compost 1.1820.02 c 46.520.7 b Soil+cellulose 0.7820.04 e 23.121.1 d
a
Means within a column followed by the same letter do not dier signi®cantly P0:05according to Student's test.
mineralization was very low in the presence of glucose (8.9%), reached 31.5% with cellulose, 59.2% with straw and 63.5% with compost. Simultaneously, in all the treatments, the bound residues increased signi®-cantly with the total microbial activity as revealed by the kinetics of total CO2 mineralization. This again
con®rmed the relation between bound residue for-mation and microbial activity (Benoit and Barriuso, 1997). The direct inhibitory eect of mineral N, as pre-viously discussed, probably also decreased atrazine mineralization. Few atrazine and mainly dealkylated metabolites were detected in the extractable fractions (Table 3). With glucose, a large proportion of hydro-xyatrazine was observed.
In the non-adapted soil, the simultaneous addition of mineral N and OAs decreased atrazine mineraliz-ation (less than 1% of the initial radioactivity was mineralized with most of the OAs and 3% with straw) but increased the partial degradation of atrazine. Deal-kylated metabolites were the predominant compounds in the extractable radioactivity (Table 4). Atrazine stabilization as bound residues did not increase signi®-cantly.
In summary, the eect of C and N availability on atrazine degradation in soils varied in relation to the micro¯ora active in atrazine degradation. In a soil adapted to atrazine mineralization, the OAs alone had little eect on atrazine degradation. In a non-adapted soil, OA addition stimulated atrazine mineralization, but mineralization was always lower than in the adapted soil. However, large atrazine dealkylation was observed which increased when mineral N was added. In the non-adapted soil, the OAs promoted dealkyla-tion but not mineralizadealkyla-tion. In both soils, mineral N decreased atrazine mineralization. The negative eect of organic N compounds increased with their N miner-alization rate, thus their N availability. Both triazine ring mineralization and atrazine dealkylation or hy-droxylation were inhibited by N addition.
Acknowledgements
This work was ®nancially supported by the INRA program ``AIP- Etude et Gestion de l'EcosysteÁme Sol''. The authors thank J.N. Rampon and V. Ber-gheaud for their technical assistance.
References
Alvey, S., Crowley, D.E., 1995. In¯uence of organic amendments on biodegradation of atrazine as a nitrogen source. Journal of Environmental Quality 24, 1156±1162.
Assaf, N.A., Turco, R.F., 1994. Accelerated biodegradation of atra-zine by a microbial consortium is possible in culture and soil. Biodegradation 5, 29±35.
Barriuso, E., Houot, S., 1996. Rapid mineralization of thes-triazine ring of atrazine in soils in relation to soil management. Soil Biology & Biochemistry 28, 1341±1348.
Barriuso, E., Houot, S., Serra-Wittling, C., 1997. In¯uence of com-post addition to soil on the behaviour of herbicides. Pesticide Science 49, 65±75.
Behki, R.M., Khan, S.U., 1986. Degradation of atrazine by Pseudomonas: N-dealkylation and dehalogenation of atrazine and its metabolites. Journal of Agricultural and Food Chemistry 34, 746±749.
Benoit, P., Barriuso, E., 1997. Fate of 14C-ring-labeled D, 2,4-dichlorophenol and 4-chlorophenol during straw composting. Biology and Fertility of Soils 25, 53±59.
Bouquard, C., Ouazzani, J., Prome, J.-C., Michel-Briand, Y., Plesiat, P., 1997. Dechlorination of atrazine by a Rhizobium sp isolate. Applied and Environmental Microbiology 63, 862±866.
Cook, A.M., HuÈtter, R., 1981. s-Triazines as nitrogen sources for bacteria. Journal of Agricultural and Food Chemistry 29, 1135± 1143.
Cook, A.M., Beilstein, P., Grossenbacher, H., HuÈtter, R., 1985. Ring cleavage and degradative pathway of cyanuric acid in bacteria. The Biochemical Journal 231, 25±30.
de Souza, M.L., Newcombe, D., Alvey, S., Crowley, D.E., Hay, A., Sadowski, M.J., Wackett, L.P., 1998. Molecular basis of a bac-terial consortium: interspecies catabolism of atrazine. Applied and Environmental Microbiology 64, 178±184.
Entry, J.A., Mattson, K.G., Emmingham, W.H., 1993. The in¯uence of nitrogen on atrazine and 2,4-dichlorophenoxyacetic acid miner-alization in grassland soils. Biology and Fertility of Soils 16, 179± 182.
Entry, J.A., Emmingham, W.H., 1995. The in¯uence of dairy manure on atrazine and 2,4-dichlorophenoxyacetic acid mineralization in pasture soils. Canadian Journal of Soil Science 75, 379±383. Esser, H.O., Dupuis, G., Ebert, E., Marco, G.J., Vogel, C., 1975.s
-Triazines. In: Kearney, P.C., Kaufman, D.J. (Eds.), Herbicides, Chemistry, Degradation and Mode of Action, Vol 1. Marcel Dekker, New York, pp. 129±208.
Hance, R.J., 1973. The eect of nutrients on the decomposition of the herbicides atrazine and linuron incubated with soil. Pesticide Science 4, 817±822.
Kaal, E.E.J., Field, J.A., Joyce, T.W., 1995. Increasing ligninolytic enzyme activities in several white-rot basidiomycetes by nitrogen-sucient media. Bioresource Technology 53, 133±139.
Kaufman, D.D., Kearney, P.C., 1970. Microbial degradation of s-triazine herbicides. Residue Reviews 32, 235±265.
Khan, S.U., 1978. Kinetics of hydrolysis of atrazine in aqueous ful-vic acid solution. Pesticide Science 9, 39±43.
Li, D., Alic, M., Gold, M.H., 1994. Nitrogen regulation of lignin peroxidase gene transcription. Applied and Environmental Microbiology 60, 3447±3449.
Mandelbaum, R.T., Wackett, L.P., Allan, D.L., 1993. Mineralization of the s-triazine ring of atrazine by stable bacterial mixed cul-tures. Applied and Environmental Microbiology 59, 1695±1701. Mandelbaum, R.T., Allan, D.A., Wackett, L.P., 1995. Isolation and
characterization of aPseudomonassp. that mineralizes thes -tria-zine herbicide atra-tria-zine. Applied and Environmental Microbiology 61, 1451±1457.
Mougin, C., Laugero, C., Asther, M., Dubroca, J., Frasse, P., Asther, M., 1994. Biotransformation of the herbicide atrazine by the white rot fungus Phanerochaete Chrysosporium. Applied and Environmental Microbiology 60, 705±708.
Radosevich, M., Traina, S.J., Hao, Y.L., Tuovinen, O.H., 1995. Degradation and mineralization of atrazine by a soil bacterial iso-late. Applied and Environmental Microbiology 61, 297±302. Shao, Z.Q., Seens, W., Mulbry, W., Bekhi, R.M., 1995. Cloning
and expression of the s-triazine hydrolase gene (trzA) from Rhodoccocus corallinusand development ofRhodococcus recombi-R. Abdelha®d et al. / Soil Biology & Biochemistry 32 (2000) 389±401
nant strains capable of dealkylating and dechlorinating the herbi-cide atrazine. Journal of Bacteriology 177, 5748±5755.
Smith, J.L., Doran, J.W., 1996. Measurement and use of pH and electrical conductivity for soil quality analysis. In: Methods for Assessing Soil Quality. Soil Science Society of America, Madison, WI, pp. 169±185.
Stucki, G., Yu, C.H., Baumgartner, T., Gonzalez-Valero, J.F., 1995. Microbial atrazine mineralization under carbon limited and deni-trifying conditions. Water Research 29, 291±296.
Topp, E., Tessier, L., Gregorich, E.G., 1996. Dairy manure incorpor-ation stimulates rapid atrazine mineralizincorpor-ation in an agricultural soil. Canadian Journal of Soil Science 76, 403±409.
Vanderheyden, V., Debongnie, P., Pussemier, L., 1997. Accelerated degradation and mineralization of atrazine in surface and subsur-face soil materials. Pesticide Science 49, 237±242.
