| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Environ. Sci. Graduate Program, The Ohio State Univ., Columbus, OH 43210-1085 (present address: Dep. of Civil and Environ. Engineering, Univ. of Illinois, Urbana, IL 61801) USA
b School of Natural Resour., The Ohio State Univ., Columbus, OH 43210-1085 USA
c Dep. of Microbiol., The Ohio State Univ., 484 West 12th Ave., Columbus, OH 43210-1292 USA
olli.tuovinen{at}osu.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
Abbreviations: Exp., Experiment • HPLC, high-pressure liquid chromatography
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
The widespread use of atrazine and its detection in aqueous, terrestrial, surface, and subsurface environments have resulted in increased research efforts concerning degradation of this herbicide. Biodegradation of atrazine has been studied under oxic and anoxic conditions, in pure and mixed bacterial or fungal cultures, and in soil and aquatic systems; and several degradation pathways have emerged. The dehalogenation of atrazine, involving the nucleophilic displacement of Cl- with OH-, occurs abiotically as well as enzymatically under both oxic and anoxic conditions (Armstrong et al., 1968; Mandelbaum et al., 1993b; Wackett et al., 1998). Dealkylation of atrazine or hydroxyatrazine can be coupled with microbial C and energy metabolism (Erickson and Lee, 1989), and the deamination of dealkylated metabolites can provide N (as NH+4) for assimilatory needs. The sequence of the dechlorination, dealkylation, and deamination reactions acting on the s-triazine ring substitutions has not been elucidated and may vary in different microorganisms. The C in the s-triazine ring does not serve as an energy source because of its valence of +4. Ring cleavage produces biuret (NH2·CO·NHCO·NH2), which is further degraded to urea (NH2·CO·NH2) and NH+4 (Erickson and Lee, 1989). Urea and ammonia are N sources that are readily used by soil microorganisms.
Mixed bacterial cultures have been described that are capable of aerobic mineralization of s-triazine herbicides (Hogrefe et al., 1985; Mandelbaum et al., 1993a). Several pure cultures, among them Rhodococcus and Pseudomonas spp. (Behki et al., 1993; Vandepitte et al., 1994), can aerobically dealkylate atrazine but are not capable of ring scission. Very few pure bacterial cultures have been characterized that are capable of atrazine mineralization (Yanze-Kontchou and Gschwind, 1994; Mandelbaum et al., 1995; Radosevich et al., 1995).
Anaerobic degradation of atrazine by bacteria derived from freshwater sediments has been reported (Jessee et al., 1983), but the mechanism of anaerobic biodegradation is not understood in as much detail as that of aerobic biodegradation. Acetate-dependent biodegradation of atrazine has also been reported in anaerobic biofilm column experiments under denitrifying, sulfate-reducing, and methanogenic conditions (Wilber and Parkin, 1995). Biodegradation does not seem to be ubiquitous in anaerobic sediments in agricultural watersheds (Topp et al., 1995). Wetland sediment studies suggested slow and only partial biodegradation under anoxic conditions (Chung et al., 1996). The formation of 14C-hydroxyatrazine and the evolution of 14CO2 from ring-labeled 14C-atrazine by an atrazine-mineralizing bacterium (M91-3) growing under denitrifying conditions have been reported (Crawford et al., 1998a). Denitrification-coupled growth of M91-3 with atrazine was much slower than under aerobic conditions.
Among factors that influence biodegradation and chemical stability in soil are bioavailability, nutrient flux, and redox conditions in vadose zones and saturated environments. Redox conditions are influenced by the availability of organic matter, electron acceptors, and microbial activities, which are in turn subject to oxic and anoxic fluctuations in poorly drained soils, drainage ditches, and stream beds. Aerobic and anaerobic microsites occur in close proximity at soil–water interfaces such as capillary fringes and in saturated systems; and interfacial conditions associated with these systems are characterized by chemical and biochemical gradients. Inevitably, redox potential gradients occur under these conditions. In general, redox potential gradients and their spatial and temporal changes are important elements in the microbial dynamics in soil.
The purpose of this work was to evaluate the biodegradation of atrazine in redox potential gradients typical of interfacial, oxic–anoxic zones in capillary fringes in subsurface and in other sediment environments. Experimental amendments included a labile, easily oxidizable organic amendment (glucose) and terminal electron acceptors (oxygen, NO-3) in a column system using sand as a solid matrix for bacterial colonization. The design and preparation of the column involved spatially distinct oxic, anoxic, and reduced zones that comprised redox potential gradients (Crawford et al., 1998b). An atrazine-mineralizing bacterium (M91-3) was chosen as the test organism because it can degrade this herbicide aerobically and under denitrifying conditions.
|
Materials and methods |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
The atrazine medium contained (L-1) 21.6 mg atrazine, 0.5 g K2HPO4, 1.5 g PIPES [piperazine-N,N-bis(2-ethane-sulfonic acid) dipotassium buffer], 0.5 g MgSO4·7H2O, and 10 mL trace metal stock solution. The pH was adjusted to 7.4 with NaOH before autoclaving (20 min, 121°C). The trace metal stock solution contained (L-1) 2.0 g nitrilotriacetic acid, 0.9 g CaSO4, 1.0 g MnSO4·H2O, 0.23 g CoSO4·7H2O, 0.8 g Fe(NH4)2 (SO4)2·6H2O, 0.2 g ZnSO4·7H2O, 0.03 g CuSO4·5H2O, 0.02 g NiCl2·6H2O, 0.02 g Na2MoO4·2H2O, 0.02 g Na2SeO4, and 0.02 g Na2WO4; pH was 6.0.
The Hungate protocol (Hungate, 1969) and its modifications for media preparation and inoculation in serum bottles (Miller and Wolin, 1974) were followed to remove dissolved O2 from all media solutions. In general, solutions were heated to the boiling point and transferred with N2–sparging, in 60-mL aliquots, to O2–free 100-mL serum bottles. The bottles were then capped with 2.5-cm (one-inch) thick butyl rubber stoppers and crimp sealed. The bottles were placed in an autoclavable pan and water was added to a depth of 2.5 cm before autoclaving for 20 min at 121°C. Solutions that could not be boiled, such as Na2S and glucose, were sparged with N2 for 30 min to remove O2 (Hungate, 1969) and were filter sterilized into O2–free serum bottles. This preparation to remove O2 from anaerobic media bottles is a firmly established laboratory protocol and has been successfully used for culturing anaerobic microorganisms (Dolfing and Tiedje, 1991; Kaiser and Bollag, 1992). Anaerobic media prepared in this manner were confirmed O2–free using gas chromatography as described by Kaspar and Tiedje (1994). Trace metals and other amendments were sterilized separately and added aseptically by syringe before use. In NO-3–amended experiments, KNO3 was added to the medium (before adding to the column) from an anaerobic, sterile stock solution to a concentration of 0.5 g L-1. Where indicated, glucose was added to a final concentration of 0.2 g L-1 from an anaerobic filter-sterilized stock solution.
Sand Columns
Pyrex cylinders were constructed with multiple sampling ports (Fig. 1)
and were fitted with rubber septa, tygon tubing, and a stopper in preparation for use. Cylinders were filled with 435 g of silica sand (300–600 µm diam.) that was washed with detergents and H2O2 as previously described (Crawford et al., 1998b). The sand was considered free of organics when CO2 evolution was not observed upon H2O2 addition. Packing of the added sand was not necessary and there was no visual evidence for air pockets remaining inside the columns. The sand-filled columns and tubing apparatus were sterilized by autoclaving on 3 consecutive days (90 min, 121°C) before use. Anaerobic liquid media were added by syringe through Port 2 while the column was purged with N2 (through the gas inlet port). When applicable, air (for aeration) or N2 (for anoxia) was introduced with a needle valve regulator through a 0.2-µm in-line filter into a gas dispersion tube, humidified with sterile distilled water, and then released into the column through one of the side ports. Before use, the water was autoclaved and cooled under N2 to remove dissolved O2.
|
Column Inoculation with the Atrazine-Degrading Bacterium M91-3
An anaerobically grown culture of M91-3 was inoculated (10% v/v) into 60 mL of liquid media containing 21.6 mg atrazine and 200 mg glucose L-1 in shake flasks. The culture flasks were incubated aerobically with shaking for 5 d. Sand-packed columns received 60 mL of the culture and 60 mL of the atrazine–glucose medium. The columns were aerated through the gas inlet, which also provided mixing. After 2 d, fresh glucose-free medium was added, and the columns were made anaerobic by N2–sparging. Atrazine concentration was monitored by high-pressure liquid chromatography (HPLC) analysis on withdrawn samples, and experiments were initiated when atrazine disappearance commenced without a lag period upon addition of fresh medium.
Experimental Conditions
Atrazine biodegradation experiments conducted with inoculated sand columns included aerobic (oxygen as electron acceptor) and anaerobic conditions with NO-3 as electron acceptor (Table 1) . Where applicable, Na2S (0.5 mL of 2.5% Na2S·9H2O) was used to establish reducing conditions, evident by negative redox potential values. Some experiments involved glucose amendment as a C source for the atrazine-degrading bacteria. All experiments were also performed under abiotic conditions in sterile, uninoculated sand columns. Samples from inoculated and sterile column experiments were taken and processed in exactly the same manner.
|
Before the experiments were started, all columns were anaerobic and contained atrazine medium. In two cases, inoculated columns had to be reused for subsequent experiments. One column was used first for Experiment V, then for Exp. IV. The other was used for Exp. III and then VI. In this case, the spent medium was drained but the column was not flushed before the addition of fresh medium. All inoculated experiments were conducted first, then the columns were washed and re-sterilized before initiating the abiotic control experiments.
Sampling
The columns were sampled (1 mL) from Ports 1, 3, and 5 (Exp. I and II, completely aerobic or anaerobic conditions) or Ports 1, 3, 4, and 5 (Exp. III–VI, redox potential gradient conditions). The sample was removed with a syringe, and the redox potential and then pH were measured immediately. The samples were then vortexed, and the optical density at 660 nm was determined before filtration through a 0.2-m filter (PVDF, Whatman, Clifton, NJ) for atrazine, NO-3, and NO-2 analysis. The possibility of retention on the syringe or filter was tested by HPLC analysis on the sample before and after contact with a syringe and filter. This procedure did not influence the analytical recovery of atrazine.
Redox Potential Measurements
Redox potential was measured using a combination platinum Ag–AgCl (4 M KCl) electrode (Model 13-62-082, Fisher Scientific, Pittsburgh, PA) interfaced with an Orion 920A pH/ISE meter (Orion Research, Beverly, MA). The electrode was calibrated with standard solutions of (i) 0.1 M potassium ferrocyanide and 0.05 M potassium ferricyanide (192 mV), and (ii) 0.01 M potassium ferrocyanide, 0.05 M potassium ferricyanide, and 0.36 M potassium fluoride (258 mV). The redox potential values presented in this paper have been normalized to the standard hydrogen scale. Care was taken to minimize sample aeration during sample transfer and redox measurement, which was completed before removing the next sample from the column. Each sample (1 mL) was removed from the column with a 25-gauge needle and syringe and was carefully dispensed into a conical pyrex tube (upper i.d. = 14 mm) by directing the liquid to the wall of the tube, then immersing the needle while dispensing to minimize aeration. The Pt of the redox potential electrode (which had a 10-mm diam. at the widest portion) was polished with fine sandpaper and immediately submerged in the tube. Less than 15 s elapsed during sample removal from the column and electrode immersion in the sample, and redox potential measurement was complete within 5 to 15 min per sample. The accuracy and reproducibility of these redox potential values were measured in triplicate cultures and were comparable to those obtained by immersing the electrode directly into liquid cultures.
Chemical Analysis
Nitrate and nitrite were analyzed with a colorimetric assay using a Lachat QuickChem autoanalyzer (Lachat Instruments, Milwaukee, WI) (Crawford et al., 1998b). Atrazine concentrations were monitored with HPLC using an RP-C18 column and UV detection at 220 nm (Radosevich et al., 1995). An isocratic mobile phase was used, which consisted of 65 parts methanol:35 parts water. All samples and standards were diluted 1:1 with mobile phase before analysis.
Statistics
To estimate rate constants of biodegradation, atrazine decline curves were fitted to a first-order decay equation using SigmaPlot 4.0 for Windows (SPSS, Chicago) and the nonlinear regression equation for exponential decay,
, where y = atrazine concentration, a = initial atrazine concentration, k = rate constant, and x = time.
|
Results and discussion |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
Redox Potential Measurement and Interpretation
In several experiments, the bottom and top zone redox potentials differed by several hundred millivolts, which constituted the basis of the redox potential gradient. This redox potential gradient was unstable and diminished with time as the proportion of the oxidized and reduced chemical species changed. The redox potential gradients in abiotic experiments were relatively more stable, presumably because of the lack of biological transformations of glucose, NO-3, and NO-2.
The column design did not permit a reasonable way to install redox electrodes for in situ measurement in such a way that the electrodes could also be easily removed and polished before measurement. Biofouling of installed electrodes would have also presented a problem with the accuracy of Eh measurement, since colonization of an electrode results in substantially lower redox potential values (Jacob, 1970). Thus, removal of samples from the column was necessary for redox potential measurement. Exposure to air was a concern for anaerobic samples because it may increase the redox potential. It was rationalized that reproducible redox potential measurements could be obtained by applying consistent sampling technique and by minimizing exposure of samples to air during measurement. Conical pyrex tubes were chosen with an inner diameter slightly greater than that of the electrode (a difference of 
4 mm) but large enough that the electrode junction was immersed but not in contact with the wall of the tube. Thus, only a small fraction of the sample surface was exposed to air in the short time required for measurement.
To test the accuracy and reproducibility of redox potential measurements, denitrifying liquid cultures were prepared in sealed vials in which the electrode could be inserted for in situ measurement. Following each in situ measurement, samples were transferred to a tube for redox potential measurement as described for the columns. The triplicate average values were identical for both in situ and transferred samples (standard deviation 20 and 10 mV, respectively). An average of 2 min was required for electrode stabilization in these samples. Triplicate redox potential values for the Na2S-amended cultures differed only by 9 mV from the mean (standard deviations 4 and 1 mV). An average of 10.5 min elapsed before electrode stabilization in these samples, presumably because of the effect of sulfide on redox potential.
Thus, the technique used to sample and measure redox potentials was concluded to be both reproducible and representative of conditions within the columns. The interpretations of redox potential in the current study are limited to the systems used and are useful in the context of trends observed over the time courses at the various depths.
Homogenous Column Studies
The following studies were conducted before the redox potential gradient studies to characterize aerobic and anaerobic atrazine metabolism in sand columns colonized by M91-3.
Aerobic Column Studies (Experiment I)
The time course of atrazine biodegradation was investigated in a sand column containing atrazine and NO-3 under aerobic conditions. In this experiment, NO-3 was tested aerobically as a N source by M91-3, in anticipation of adding NO-3 to anaerobic columns for redox potential gradient studies. Biodegradation decreased atrazine concentration to undetectable levels within 50 h at each column depth (Fig. 2A)
. Atrazine degradation appeared to be first-order, with similar degradation rates at all column depths (Table 2)
. Homogeneity was maintained within the column by turbulence created by aeration through the gas inlet port of the column. Redox potential values remained positive throughout the time course (Fig. 2B). Nitrate was not depleted (Fig. 2C), indicating that NO-3 utilization by M91-3 was negligible under these conditions. A parallel column experiment under aerobic conditions without NO-3 revealed a comparable time course of atrazine biodegradation (data not shown). Thus, NO-3 did not appear to inhibit the biodegradation of atrazine. The pH values in this column remained at about 7.6 and there was no change in the optical density (data not shown). Abiotic column experiments showed that atrazine concentration did not decrease with time in sterile media.
|
|
|
The redox potential values at all depths remained negative except for a transient increase in Port 1 depth (Fig. 3B). Nitrate concentrations varied, making it difficult to discern any decrease (Fig. 3C). The Na2S amendment increased the pH to 11 at the onset of the experiment followed by a gradual decrease (Fig. 3D). Atrazine concentration remained constant under comparable abiotic conditions.
Redox Potential Gradient Column Studies
The following studies were designed to permit aerobic metabolism in the topmost zone of the column (Port 1 depth) and anaerobic metabolism in the remaining zones (Port 3, 4, and 5 depths). The Port 2 depth was not sampled because this port was used for aeration or adding media. Ports 3 and 4 could not be distinguished from the aerobic Port 1 on the basis of redox potential, since the mere removal of oxygen does not sufficiently change the measured redox potential of the liquid (Jacob, 1970; Kaspar and Tiedje, 1994). Redox potential values in NO-3–amended experiments were comparable in all depths (aerobic and anaerobic) that did not receive Na2S. The NO-3–N2 couple is not reversible and thus the denitrification process cannot be discerned by redox potential measurement (Tiedje, 1988). Thus, these measurements substantiate the conclusion that aerobic and anaerobic redox processes can occur at the same redox potential plateau.
To create an environment that could be characterized with a measured low redox potential, Na2S was added as a reducing agent to the Port 5 depth. Studies with this column type, that is, with a continuum of aerobic, anaerobic, and reduced conditions, were considered as redox potential gradient systems. The ability of the solid matrix to support distinct redox zones without a cross-mixing between the sampling depths has been tested and described previously (Crawford et al., 1998b).
Experiment III
Inoculated sand columns were filled with anaerobic atrazine and NO-3 medium and subjected to a redox potential gradient by the addition of Na2S (0.5 mL of 2.5% Na2S·9H2O) only to Port 5. The column was initially aerated through Port 1 for 15 min to create aerobic conditions in the surface zone. Atrazine was completely degraded in all zones, but the relative rates differed (Fig. 4A)
. Most rapid biodegradation of atrazine occurred under anaerobic conditions at Port 4 depth (Table 2). Port 5 data were not used for these calculations because of the previously described uncertainty about the initial concentration. A redox potential gradient spanning about 700 mV between the surface and bottom zones was established at the onset of the experiment (Fig. 4B, Table 2), but the gradient gradually diminished. It is noteworthy that the redox potential values were comparable for the aerobic (Port 1) and anoxic (Port 3) zones in this column experiment.
|
Decrease in NO-3 concentration in the initially aerated top layer in the otherwise anaerobic column (Exp. III) is attributed to bacterial denitrification in anaerobic micropockets remaining after 15 min of purging with air through Port 2. Nitrate reduction required anaerobic conditions, as evident from the lack of NO-3 utilization in aerobic column experiments (Fig. 2).
The pH values remained constant at about pH 7.9 in Ports 1, 3, and 4. The initial pH 11 in Port 5 caused by the Na2S addition decreased to about pH 8.5 in 3 d (Fig. 4D); this coincided with increasing redox potential (Fig. 4B) and the beginning of atrazine degradation. Atrazine and NO-3 concentrations remained unchanged in the corresponding abiotic experiment.
Experiment IV
Anaerobic biodegradation of atrazine was also tested without NO-3 in a redox potential gradient. An inoculated column was filled with anaerobic atrazine medium; Na2S (0.5 mL of 2.5% Na2S·9H2O) was added to Port 5; and the column was aerated for 15 min through Port 1. Atrazine biodegradation occurred at Port depths 1, 3, and 4 (Fig. 5A)
. Atrazine biodegradation in Port 1, 3, and 4 depths could be described by first-order regression according to the r2 values (Table 2). Atrazine loss and biodegradation activity at Port 5 depth were inconclusive because of the low concentration resulting from initial dilution with residual from a previous experiment.
|
Experiment V
The degradation of atrazine by M91-3 was studied in a redox potential gradient in the presence of both glucose and NO-3. Glucose was used as an additional substrate because it enhanced the growth of the bacterium with concurrent aerobic use of atrazine (Radosevich et al., 1995). The inoculated column was filled with anaerobic atrazine-, glucose-, and NO-3–containing medium; Na2S (0.5 mL of 2.5% Na2S·9H2O) was added to Port 5; and the column was aerated for 15 min through Port 1. Rate constants indicated faster biodegradation with the addition of glucose than in the glucose-free experiments (Fig. 6A
vs. Fig. 4A, Table 2). Although cell density measurements were not within the scope of these column experiments, this difference was attributed to the biomass effect because glucose at 200 mg L-1 is known to support higher cell density than atrazine at 21 mg L-1 (Radosevich et al., 1995).
|
Experiment VI
The biodegradation of atrazine was also monitored in a column that was amended with atrazine and glucose but no NO-3 (i.e., no external electron acceptor). Na2S (0.5 mL of 2.5% Na2S·9H2O) was added to Port 5 and the column was aerated for 15 min through Port 1. Atrazine biodegradation occurred at all column depths (Fig. 7A)
; however, degradation at the Port 1 depth was limited. Although this port was initially aerobic, it was postulated that available oxygen was consumed due to glucose-dependent respiration.
|
Residual NO-3 was present at the Port 5 depth in the carryover from the previous experiment, and decreased from 45 to 30 mg NO-3 L-1 between hours 14 and 60 (data not shown). Transient NO-2 levels (<0.07 mM) were also detected. The initial redox potential gradient of 700 mV gradually diminished to 300 mV at the termination of the experiment (Fig. 7B). The pH at the depth of Port 5 was initially 11.5 and decreased to <8, whereas at the overlying depths, the pH remained essentially unchanged (Fig. 7C). Atrazine depletion also commenced at the onset of the experiment at the Port 5 depth, which had a redox potential of -520 mV.
|
Conclusions |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
In the present work, the anaerobic biodegradation was accelerated by glucose in the presence of NO-3, but not in the absence of NO-3. In general, atrazine degradation by M91-3 within the redox potential gradient could be described with a first-order rate expression, evident by the observed r2 values of the nonlinear regressions. Atrazine degradation rate constants ranged from 30 to 0.43 h-1, and good fits of biodegradation data with first-order rate equations suggest strong dependence on substrate concentration. Biodegradation of atrazine occurred in all redox potential gradients, which may be construed to encompass redox potentials characteristic to aerobic, microaerophilic, and anaerobic conditions (-400 to 400 mV). All these findings indicate that atrazine biodegradation can occur within oxic–anoxic redox regimes. Active biodegradation of atrazine should occur in vadose and saturated zones, as well as in the capillary fringe, unless limited by the presence of atrazine-degrading microorganisms.
|
ACKNOWLEDGMENTS |
|---|
Received for publication September 5, 1998.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. van Bochove, S. Beauchemin, and G. Theriault Continuous Multiple Measurement of Soil Redox Potential Using Platinum Microelectrodes Soil Sci. Soc. Am. J., November 1, 2002; 66(6): 1813 - 1820. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
