Environ. Sci. Technol. 2004, 38, 3092-3097
Adaptation of Soil Biological Nitrification to Heavy Metals J A M E S A . R U S K , † R E B E C C A E . H A M O N , * ,‡ DARYL P. STEVENS,‡ AND M I K E J . M C L A U G H L I N †,‡ School of Earth and Environmental Sciences, Faculty of Sciences, University of Adelaide, Waite Campus, South Australia, Australia 5005, and CSIRO Land and Water, PMB 2, Glen Osmond, South Australia, Australia 5064
The adaptive response of soil biological nitrification to Zn and Pb was assessed using an in situ method we have developed. The method is based on reinoculating a sterilized metal contaminated soil with the same soil that is either uncontaminated or has been incubated with metal. This approach excludes the potentially confounding effects of metal aging reactions in soils. We found added Zn concentrations which gave rise to a decrease in nitrification to 50% that of the uncontaminated soil (i.e. EC50) of 210 mg/ kg for communities not previously exposed to Zn and 850 mg/kg for communities exposed to Zn for 17 months, indicating that significant adaptation of the community to Zn had occurred. Similarly, this protocol was able to demonstrate adaptation of soil biological nitrification to Pb, with EC50 values of 1960 and 3150 mg/kg for the unexposed and exposed treatments, respectively. Exposure of unadapted and adapted microbial communities to a combination of Zn and Cd showed that the presence of Cd did not lead to greater toxicity in either community. Adapted communities were not more sensitive to decreases in soil pH than unadapted communities. Prior exposure to Zn was found to confer significantly greater tolerance of the community to Pb. Prior exposure to Pb similarly conferred significantly greater tolerance of the community to Zn. Implications of the adaptive capacity of soil microbes to the development of critical threshold values for heavy metals in soil based on ecotoxicity assessments are discussed.
Introduction Soil microorganisms can acclimate to elevated concentrations of heavy metals by inducing phenotypic changes at the individual level (1). Microbial communities may also adapt to elevated metal concentrations through shifts in community structure or through genotypic alteration e.g. through evolution of plasmids that can encode resistance systems for metal ions (2). In this study we do not differentiate between adaptation and acclimation but for simplicity refer to evidence of a increase in tolerance to metals with increasing exposure time as “adaptation”. Evidence from in situ ecotoxicity tests suggests that the soil microbial community is the most sensitive biological receptor for some metals, including Zn. This is one of the criteria being used to drive international regulations for * Corresponding author phone: +61 8 8303 8489; fax: +61 8 8303 8565; e-mail:
[email protected]. † University of Adelaide. ‡ CSIRO Land and Water. 3092
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 11, 2004
allowable inputs of these metals to soils. At present these ecotoxicity tests do not take into consideration the adaptive capacity of soil microorganisms. This means that such tests could be either underestimating or overestimating the risk of soil metal contamination. There have been many in vitro studies investigating changes in tolerance to metals of microorganisms extracted from contaminated soils. For example, Bååth (3) extracted soil bacteria from a Cucontaminated soil and investigated tolerance of the bacteria to fresh additions of soluble Cu using a thymidine incorporation technique. The results showed that the extracted population was tolerant to higher levels of Cu in solution than a population extracted from an uncontaminated control soil. Using the same methodology, Diaz-Ravin ´ a and Bååth (4) conducted an experiment to assess the effect of time of exposure to Zn on microbial tolerance to Zn and found that a tolerant population was established within 2 days following initial application of Zn to the soil. These and similar studies have concluded that metal resistance and metal adaptation capabilities are widespread among different bacterial genera (5). However the requirement to extract or isolate microorganisms from the soil prior to assessment of adaptation limits the range of soil functions for which adaptation can be tested, as the extraction efficiency varies between soils (6), and culturable microbes only account for a small, unquantifiable fraction of the total microbial biomass (7). Similarly, it is possible that removal of microorganisms from their natural ecological niche may engender a stress response or other disruption to the community structure, both of which could impact on any assessment of changes in metal tolerance. There are several well documented in situ methods for assessing changes in soil function in response to elevated concentrations of toxic agents (8). One of these, the ISO international standard protocol for assessment of the influence of agents on nitrogen mineralization processes (9), is used regularly in ecotoxicological risk assessment. Nitrification is an important soil function that is highly sensitive to heavy metals (10). However contaminant-induced changes in the tolerance of soil biological nitrification cannot be assessed using a direct application of this method. This is because both microbial adaptation and metal aging processes lead to similar time dependent changes in the measured microbial response as follows: an increase over time in community tolerance is indicated by an increase in the EC50 of the nitrification test over time. However the bioavailability of metals added to soils can decrease over time as a result of transformation into phases that are not in rapid equilibrium with the soil solution (11, 12). This effect would therefore also be seen as an apparent increase in the EC50 of the nitrification test over time. Hence it is impossible to differentiate between the two effects to quantify microbial adaptation using the current test methodology. The potential for microbial adaptation to heavy metals in ecotoxicity assessments has been discussed in numerous reviews (13-15) but lacks quantitative data as to the magnitude, time scale, and possible burdens incurred as a result of adaptation. The first objective of this study was to develop an in situ nitrification test that would not be subject to interferences due to chemical aging in soil, to enable assessment of whether soil biological nitrification can adapt over time to elevated concentrations of soil Zn or Pb. Following the demonstration of adaptation to Zn and Pb, the second objective was an initial characterization of the selectivity of metal adaptation and whether adaptation increases sensitivity of communities to decreases in pH. 10.1021/es035278g CCC: $27.50
2004 American Chemical Society Published on Web 04/28/2004
Experimental Section Soil Characteristics. Topsoil was sampled from a site in South Australia used for intensive horticultural production of potatoes and onions. The soil properties were pHw 9.0, electrical conductivity 0.12 dS/m (1:5 soil:water) (16), cation exchange capacity 6.4 cmol+/kg (NH4Cl) (16), total carbon 0.55% (Leco furnace), water holding capacity (WHC) 30.7% (17), field capacity 7.4% (100 cm tension), and clay content 4% (16). Total concentrations of Zn and Pb in the different soil treatments were assessed by inductively coupled plasma atomic emission spectrometry following aqua-regia digestion (18). The soil was dried (40 °C), mixed thoroughly, and passed through a 2 mm sieve prior to any experimental uses. Inoculant Preparation. Subsamples of the soil were spiked with a solution containing different concentrations of ZnSO4 or Pb(NO3)2 giving 3 different soil Zn treatments and 2 different soil Pb treatments. To remove excess SO42or NO3- which has the potential to interfere with metal cation toxicity (19) the treatments were leached with two pore volumes of deionized H2O until the electrical conductivity of the leachate decreased to < 1 dS/m. The Zn concentrations in the soil treatments following leaching were as follows (mg/kg): 30, 670, 890. The Pb concentrations in the soil treatments were as follows (mg/kg): 20 and 3300. The treatments were incubated in loose plastic bags at approximately 50% WHC in a glasshouse for up to 21 months with moisture content maintained through periodic addition of deionized H2O to weight. Subsamples from the treatments were used as inoculants in the experiments described below and were sampled and dried at 40 °C for 24 h prior to their use. We refer to the 30 mg/kg Zn treatment and 20 mg/kg Pb treatment as “unexposed” inocula. For the “exposed” inocula, the 3300 mg/kg Pb treatment was used, while the 670 and 890 mg/kg Zn treatments were combined just prior to use to provide a single inoculum with a Zn concentration of 780 mg/kg. The metal concentrations used for the exposed inocula were equivalent to the N-mineralization EC50 value determined from a preliminary test with this soil (using lucerne meal as the substrate) following incubation of the soil with Zn or Pb for 4 months. General Experimental Protocol. Soil (180 g), which was to later receive the inoculant material, was divided into separate treatments, and then metals were added at the rates described in the following section. After application of metal treatments, soil samples were dried overnight at 40 °C, leached (19), and dried overnight again before being placed in sealable plastic bags and moistened to 50% of WHC to increase the radiosensitivity of the microbial cells (20). Soil samples were gamma irradiated at an effective dose of 2.5Mrad (25 kGy) (20, 21) a dose that is sufficiently low to avoid the possibility of radiation-induced breakdown of compounds under heavy radiation treatments (e.g. 4-12 Mrad) (20). Sterility of soil treatments was confirmed by plating sterile suspensions of gamma-irradiated soil and phosphate-buffered saline onto nutrient agar plates (Oxoid) and checking for microbial growth on the plates after incubating for 4 days at 25 °C. Sterilized soil from each metal treatment was divided in half to provide 90 g duplicates of each metal treatment. The treatments were placed into 250 mL plastic incubation containers. One duplicate from each metal rate was inoculated (3% w/w) with air-dry “unexposed” (see above) soil, and the other duplicate was inoculated (3% w/w) with airdry “exposed” (see above) soil. The treatments were then thoroughly mixed. All of the unexposed inoculation treatments were incubated under the same conditions and for the same period (see below) as the exposed treatments. The soil moisture contents of all treatments were adjusted to 50% WHC (determined from a preliminary test to be the
optimum moisture content for nitrate production in this soil), and the soil was remixed and then incubated at 25 °C for 4 days with lids kept loose throughout the experiment to prevent anoxic conditions. Nitrification tests (9) were then carried out as follows: a solution of (NH4)2SO4 was mixed thoroughly with each soil giving a final concentration of 100 mg N/kg soil. The treatments were then returned to incubate at 25 °C. During the 28 day incubation soil moisture contents were maintained at 50% WHC by twice weekly additions of deionized H2O. Three replicate soil samples from each of the treatments were extracted with KCl (0.1 M, 1:5 soil:solution) at 0 and 28 days after addition of nitrogen (22). Extracts were filtered through a 0.45 µm filter and analyzed for (NH4+)-N (23) and (NO3- + NO2-)-N (referred to hereafter as NOx) by a colorimetric assay (24). The production of NOx over the duration of the test was calculated by subtracting the concentration of NOx present in the soil at day 0 from that present in the soil at day 28. It should be noted that the amount of nitrification that had occurred in the control soils by day 28 was approximately equivalent to the total NH4 dose that had been supplied. The soil pHw (1:5) was measured before and after nitrogen addition at day 0 and at the end of the 28-day test. Soil Treatments. Experiment 1: seven rates of Zn were applied to 180 g samples of soil by spraying on a solution containing ZnSO4. The treatments were thoroughly mixed, allowed to equilibrate for 1 day, and then leached and sterilized as described above. The final Zn concentrations in the treatments were 20, 120, 330, 510, 670, 790, and 1000 mg/kg. These soils were then inoculated with 17-month incubated Zn-exposed and unexposed inoculant materials as described above. Experiment 2: seven rates of Pb were applied to the soil as a solution of Pb(NO3)2. Final Pb concentrations in the treatments were 20, 1750, 2680, 2750, 3210, 4220, and 5140 mg/kg. These soils were then inoculated with 21-month incubated Pb-exposed and unexposed inoculant materials as described above. Experiment 3: the same rates of Zn as experiment 1 were applied to the soil, along with one hundredth the amount of Cd (as CdCl2). Cadmium application rates were set at one hundredth of Zn application rates to approximate the ratio of these two elements in contaminated soils. The final Zn and Cd concentrations in the treatments were 20, 110, 280, 540, 620, 780, and 1000 mg/kg and 0.3, 1, 3, 6, 7, 9, and 12 mg/kg, respectively. These soils were then inoculated with 21-month incubated Zn-exposed and unexposed inoculant materials. Experiment 4: soil was spiked with either Zn or Pb at the same rates as used in experiments 1 and 2. The Zn-treated soil was inoculated with Pb-exposed inoculant material, whereas the Pb-treated soil was inoculated with Zn-exposed inoculant material. Experiment 5: the pH tolerance of zinc-adapted nitrifiers was tested by applying dilute HCl to soil subsamples to give a range of pH treatments. Soil moisture content was adjusted to 30% WHC and equilibrated overnight. The treatments were then leached to ensure there was no effect of applied Clcontributing to the measured toxicity (26, 27) and sterilized as above. The sterile pH-treatments were inoculated with 17-month incubated Zn-exposed and unexposed inoculant materials. Statistical Approach. Nitrification EC50 values were estimated from log transformed total extractable soil metal concentrations by fitting sigmoidal dose response curves using the method of Scholze et al. (28). Bootstrapping was used to estimate confidence intervals around the EC50 value. As recommended by Scholze et al. (28), 1000 bootstrap iterations were performed to estimate bootstrap-based confidence intervals. To ascertain whether differences in the VOL. 38, NO. 11, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3093
TABLE 1. Experiment EC50 Values (mg/kg) for Exposed and Nonexposed Populationsa experiment
metal treatment
1
Zn
2
Pb
3
Zn + Cd (100:1)
4A
Zn
4B
Pb
5
pH
inoculant treatment
EC50
unexposed (30 mg Zn/kg) exposed (780 mg Zn/kg) unexposed (20 mg Pb/kg) exposed (3300 mg Pb/kg) unexposed (30 mg Zn/kg) exposed (780 mg Zn/kg) unexposed (20 mg Pb/kg) exposed (3300 mg Pb/kg) unexposed (30 mg Zn/kg) exposed (780 mg Zn/kg) unexposed (30 mg Zn/kg) exposed (780 mg Zn/kg)
210 mg Zn/kg (180-240) 850 mg Zn/kg (850-870) 1960 mg Pb/kg (1760-2380) 3150 mg Pb/kg (3140-3190) 350 mg Zn/kgb (270-420) >1000 mg Zn/kgc (not applicable) 230 mg Zn/kg (160-300) 650 mg Zn/kg (600-690) 2590 mg Pb/kg (2580-2610) 3880 mg Pb/kg (3670-3900) 7.96 pH (7.94-7.98) 7.79 pH (7.77-7.81)
a Values in brackets represent the 5th and 95th percentile. Inoculant exposure time (months): Zn 17m experiments 1, 3, and 4B; Zn 21m experiments 3 and 5; Pb 21m experiments 2 and 4A. b Soil Cd concentration at EC50 ) 4 mg/kg. c Soil Cd concentration at EC50 > 11 mg/kg.
EC50 values were significant, the bootstrap results were analyzed using a t-test using SPSS Version 8.0.
Results and Discussion Adaptation of Biological Nitrification to Zn and Pb (Experiments 1 and 2). Previous studies conducted in vitro have demonstrated increased tolerance of extractable and culturable soil microbes to Zn following preexposure of the organisms to Zn (4, 7, 29-35) and to Pb following preexposure to Pb (33, 36-38). In general, in the few in situ studies performed to investigate changes in tolerance of microbial communities to metals, the aging effects of metals in soils were not taken into account (e.g. ref 39). Hence it is impossible to determine whether the apparent increase in community tolerance with time is actually due to adaptation or due to decreased toxicity of the metals as a result of aging processes. One exception to this is the study by Witter et al. (7). In this study, as well as investigating changes in the tolerance of microorganisms extracted from soil treatments which had historical inputs of different rates of metal contaminated sewage sludge, the effect of metal salt addition on substrate induced respiration in the treatments was also assessed in situ. The authors found that communities inhabiting the treatments with high sludge metal concentrations had significantly higher specific growth rates than the control (no sludge) treatment, following the addition of Cd or Zn salts which were added shortly after the substrate (glucose) was provided. Since metal salt additions to each of the treatments were made at the same time, this study is not confounded by metal aging effects and demonstrates increased tolerance of the metal exposed communities to further additions of metals. In our study, there was no difference between the aging status, and hence bioavailability, of the soil metals that the unexposed and exposed communities were further exposed to. Therefore differences in the dose-response curves between the unexposed and exposed communities are the result of an inherent quality of the particular community and not the consequence of soil chemistry. There was a significant difference (p e 0.05) between the EC50 values (Table 1) obtained for nitrification by the Znexposed compared to the unexposed treatments when reinoculated into sterilized Zn-spiked soil (Figure 1A) and between the Pb-exposed compared to the unexposed treatments when reinoculated into sterilized Pb-spiked soil (Figure 1B). Our study was therefore able to demonstrate significant adaptation of biological nitrification to both Zn and Pb. Toxicity in the Presence of Cd (Experiment 3). The nitrification EC50 value for the Zn-exposed community was significantly different (p e 0.05) compared to the unexposed community when both were reinoculated into sterilized Zn 3094
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 11, 2004
+ Cd-spiked soil (Figure 1C, Table 1) demonstrating that the presence of Cd at concentrations relevant to many contaminated site scenarios did not constrain the tolerance developed by the Zn-exposed community. A comparison of response curves found no significant differences (p > 0.05) between the unexposed community reinoculated into Znspiked soils (Figure 1A) and the unexposed community reinoculated into Zn + Cd spiked soils (Figure 1C). Chaudri et al. (40) and Renella et al. (41) argued that Zn and not Cd causes the toxic microbial response in multimetal contaminated sludge-treated soils where Zn:Cd ratios are typically close to 100:1. In contradiction to these findings, Smith and Giller (42) suggested that Cd is the metal causing the most adverse effects on the soil microbial biomass and activity in sludge-treated soils. That we observed no increase in toxicity in unexposed communities in soils spiked with Zn+Cd compared to soils spiked with Zn alone supports the theory that Zn rather than Cd is more likely to be responsible for observed toxic responses in nitrification, at least in soils where Zn:Cd ratios are close to 100:1. Specificity of Adaptation (Experiment 4A,B). There was a significant difference (p e 0.05) between the nitrification EC50 values in the Zn-exposed community compared to the unexposed community when reinoculated into sterilized, Pbspiked soils (Figure 1D, Table 1). Similarly, there was a significant difference (p e 0.05) between the nitrification EC50 values in the Pb-exposed community compared to the unexposed community when reinoculated into sterilized, Znspiked soils (Figure 1E, Table 1). Note that it was necessary to remove the 500 mg/kg Zn data points from consideration in order to fit a sigmoidal curve to data for both the unexposed and Pb-exposed treatments. These results indicate that adaptation to Zn confers tolerance to Pb and vice versa. These results support the findings of Diaz-Ravin ´ a et al. (32) who used the thymidine incorporation technique to measure tolerance of soil microorganisms to a range of metals including Zn and Pb. They found that metal-exposed populations from polluted soils were tolerant to metals other than the metal originally added to the soil, thus providing evidence for a multiple heavy metal tolerance mechanism at the community level. Tolerance to Soil pH of Zn Adapted Nitrifiers (Experiment 5). Adaptation to toxicants requires a net increase, either at the organism or at the community level, in the expenditure of energy on detoxification strategies. This may divert energy from other cellular processes and could lower the resilience of the adapted microorganism and/or community to other stressors. Soil biological nitrification is sensitive to changes in soil pH (43). This may be due in part to shifts in equilibrium from NH3 f NH4+ with decreases in pH (pKa 9.25) where NH3 is the preferred substrate for
FIGURE 1. Comparisons of NO2- + NO3- (i.e. NOx, mg/L) produced over a 28-day incubation in gamma irradiated soil. All data points (three replicates) are shown. In some cases replicates were so close that they directly overlay each other: (A) Zn-spiked soil reinoculated with Zn-exposed or unexposed communities, (B) Pb-spiked soil reinoculated with Pb-exposed or unexposed communities, (C) Zn and Cd (added at 1/100 the Zn concentration) spiked soil reinoculated with Zn-exposed or unexposed communities, (D) Pb-spiked soil reinoculated with Zn-exposed or unexposed communities, and (E) Zn-spiked soil reinoculated with Pb-exposed or unexposed communities. nitrification (44). However studies have also suggested that NH4+ may be utilized as a substrate for nitrification and that acid sensitive nitrifiers can adapt to low pH conditions (44). To investigate whether adaptation to Zn resulted in an increased sensitivity of the nitrifying community to soil acidity, we compared the pH tolerance of Zn exposed and unexposed communities. As expected, we observed a significant effect of pH on nitrification in both Zn-exposed and unexposed communities (Figure 2) with respective EC50 values of pH 7.79 and 7.96 and little nitrification in either community occurring at pH < 6.5. The finding that the EC50 for the exposed community was not higher than the unexposed community indicates that adaptation to Zn did not increase sensitivity to changes in pH in this soil. Further studies are required to confirm this result in other soils and to investigate the interaction between metal adaptation and microbial resilience to variations in other environmental factors such as temperature, nutrition, and water regime. In soil spiking studies designed to assess the impact of metal contamination, particular care is needed to ensure
that changes in the measured response are due to the metal contaminant and not a result of metal-induced pH changes (19, 45). Spiking our soil with Zn or Pb and then leaching caused the soil pH to decrease from 9.0 (in unspiked soils) to 7.0 to 7.5 at the highest rates of metal addition. Spiking the soil with Zn + Cd and then leaching resulted in a greater decrease in pH than spiking the soil with Zn alone. The pH of the Zn + Cd treatments ranged from 9.0 (unspiked soil) down to 6.9 at the highest rate of added Zn + Cd. To discriminate between the effect of pH change and the effect of metal toxicity, we plotted the nitrification results obtained from experiment 1 as a function of the pH of the different Zn treatments from experiment 1 and compared this to the effect on nitrification of pH alone (Figure 2). Despite the strongly negative effect of soil acidification on nitrification, the presence of the metal led to a greater inhibition of nitrification for the unexposed community than could be attributed to pH alone (Figure 2), confirming that toxicity in the metal treated soil was Zn induced. In contrast for the Zn-exposed community, nitrification rates were greater in VOL. 38, NO. 11, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3095
Further quantitative studies assessing microbial adaptation and the burden associated with different levels of adaptation are urgently required to enhance our ability to make meaningful interpretations and comparisons between experimental results from short-term laboratory ecotoxicity studies with results from monitoring of long-term exposure of microbial communities to heavy metals in the field.
Acknowledgments The authors gratefully acknowledge financial support from the International Lead Zinc Research Organization. The authors thank Enzo Lombi for helpful suggestions on the manuscript, Tunde Heinrich for her technical support, and Ray Correll and Mary Barnes for envirometrics support.
Literature Cited FIGURE 2. Comparisons of NO2- + NO3- (i.e. NOx, mg/L) produced over a 28-day incubation between Zn exposed (hollow symbols) and unexposed (filled symbols) communities which were reinoculated into gamma irradiated soil which was treated prior to irradiation either with acid (triangles, dashed fitted curve) or with Zn (circles, solid fitted curve) to give a series of treatments which range in soil pH. the Zn-treated soil compared to acidified soil at equal pH (Figure 2). Implications of Adaptation for Ecotoxicity Testing. This study has provided clear evidence of adaptation of soil biological nitrification to elevated concentrations of Zn and Pb. Use of functional endpoints (respiration, nitrification, substrate incorporation, enzymes activities, etc.) are therefore inappropriate for assessing changes in metal bioavailability over time, unless special precautions are taken to isolate the effects of microbial adaptation from physicochemical reactions affecting metal availability. This has not been considered in previous studies (46, 47); however, the protocol developed in our work offers a technique that allows this distinction. If adaptation to metals is a rapid process, as was suggested by the results of Diaz-Ravin ´ a and Bååth (4), then the development of metal tolerance may even be occurring during the short incubation times used in ecotoxicity tests. This poses problems in using the data from such assays for risk assessment purposes. Is the development of critical thresholds based on adapted microorganisms appropriate? The development of metal resistance and enhanced metal tolerance mechanisms have been described as the very processes that express deterioration of the soil microbial community - pollution-induced community tolerance (PICT) (48). Hence apparent “normal” responses of metal-adapted microbial communities in the presence of elevated metal concentrations may conceal a system under stress. Ecotoxicity tests will therefore underestimate risk if the soil microorganisms or communities have adapted to metals during the test, and this adaptation is an adverse effect. However, the ability to adapt to at least small changes in environmental conditions is essential for, and inherent to, all living organisms. Indeed, there is evidence that adaptation to metals occurs at background concentrations in soil, with soils that have low background Zn concentrations being more sensitive to added Zn than soils having high background Zn concentrations (49). Adaptation is a response of the microbial community to natural background variation in metal concentrations in soil and is also likely to be a continuous response by the microbial community from ultralow background concentrations right through to high metal levels in the most polluted soils. Ecotoxicity tests will therefore overestimate the risk if the soil microorganisms or communities have not adapted to metals during the test, and adaptation is not an adverse effect. 3096
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 11, 2004
(1) Tyler, G.; Pahlsson, M. B.; Bengtsson, G.; Bååth, E.; Tranvik, L. Water, Air, Soil Pollut. 1989, 47, 189. (2) Silver, S.; Walderhaug, M. Microbiol. Rev. 1992, 56, 195. (3) Bååth, E. Soil Biol. Biochem. 1992, 24, 1167. (4) Diaz-Ravin ´ a, M.; Bååth, E. Appl. Environ. Microbiol. 1996, 62, 2970. (5) van der Lelie, D.; Tibazarwa, C. In Metals in the Environment: Analysis by Biodiversity; Prasad, M. N. V., Ed.; Marcel Dekker: New York, 2001; pp 1-36. (6) Bakken, L. R. Appl. Environ. Microbiol. 1985, 49, 1482. (7) Witter, E.; Gong, P.; B Bååth, E.; Marstorp, H. Environ. Toxicol. Chem. 2000, 19, 1983. (8) McGrath, S. P. In Test methods to Determine Hazards of Sparingly Soluble Metal Compounds in Soils; Fairbrother, A. P., Glazebrook, N., van Straalen, N., Tarazona, J., Eds.; SETAC Press: Pensacola, FL, 2002; pp 17-36. (9) ISO. Soil quality Part 4 Biological methods Section 4.4 Effects of pollutants on microbes; British Standards Institute: 1997; Subsection 4.4.3. (10) Wuertz, S.; Margeay, M. In Modern Soil Microbiology; Van Elsas, J. D., Trevors, J. T., Wellington, E., Eds.; Marcel Dekker: U.S.A., 1997; pp 607-642. (11) Barrow, N. J. J. Soil Sci. 1986, 37, 277. (12) Brennan, R. F. Aust. J. Soil Res. 1990, 28, 303. (13) Bååth, E. Water, Air, Soil Pollut. 1989, 47, 335. (14) Giller, K.; Witter, E.; McGrath, S. P. Soil Biol. Biochem. 1998, 30, 1389. (15) van Beelen, P.; Doelman, P. Chemosphere 1997, 34, 455. (16) Rayment, G. R.; Higginson, F. R. Australian Laboratory Handbook of Soil and Water Chemical Methods; Inkata Press: Sydney, 1992. (17) Jenkinson, D. S.; Powlson, D. S. Soil Biol. Biochem. 1976, 8, 167. (18) Zarcinas, B. A.; McLaughlin, M. J.; Smart, M. K. Commun. Soil Sci. Plant Anal. 1996, 27, 1331. (19) Stevens, D. P.; McLaughlin, M. J.; Heinrich, T. Environ. Toxicol. Chem. 2003, 22, 3017. (20) Cawse, P. A. In Soil Biochemistry; Paul, E. A., McLaren, A. D., Eds.; Marcel Dekker: New York, 1975; pp 213-267. (21) Jenkinson, D. S.; Powlson, D. S. Soil Biol. Biochem. 1976, 8, 209. (22) Anon. In OECD guidelines for the testing of chemicals; Organisation for Economic Development (OECD): Paris, 2000; Guideline 216. (23) Perstorp Analytical. The Flow Solution Methodology for Ammonium Nitrogen; 1992; P/N 000857 and P/N 0000858. (24) Perstorp Analytical. Nitrate+Nitrite Nitrogen in Soil Extracts; 1993; P/N 001092 and P/N 001903. (25) McLaughlin, M. J.; Hamon, R. E.; McLaren, R. G.; Speir, T. W.; Rogers, S. L. Aust. J. Soil Res. 2000, 38, 1037. (26) Bongers, M.; Rusch, B.; Van Gestel, C. A. M. Effects of prepercolation and counterion on Pb toxicity to Folsomia candida; SETAC: Brighton, UK, 2000. (27) Smit, C. E.; Van Gestel, C. A. M. Environ. Toxicol. Chem. 1998, 17, 1132. (28) Scholze, M.; Boedeker, W.; Faust, W.; Backhaus, T.; Attenburger, R.; Grimme, L. H. Environ. Toxicol. Chem. 2001, 20, 448. (29) Angle, J. S.; Chaney, R. L.; Rhee, D. Soil Biol. Biochem. 1993, 25, 1443. (30) Bååth, E.; Diaz-Ravin ´ a, M.; Frostegård, Å.; Campbell, C. D. Appl. Environ. Microbiol. 1998, 64, 238. (31) Diaz-Ravin ´ a, M.; Bååth, E. Soil Biol. Biochem. 2001, 33, 241. (32) Diaz-Ravin ´ a, M.; Bååth, E.; Frostegård, Å. Appl. Environ. Microbiol. 1994, 60, 2238. (33) Olson, B.; Thornton, I. J. Soil Sci. 1982, 33, 272.
(34) Pennanen, T.; Frostegard, A.; Fritze, H.; Bååth, E. Appl. Environ. Microbiol. 1996, 62, 420. (35) Rutgers, M.; Verlaat, I.; Wind, B.; Posthuma, L.; Breure, A. Environ. Toxicol. Chem. 1998, 17, 2210. (36) Doelman, P.; Haanstra, L. Soil Biol. Biochem. 1979, 11, 487. (37) Richards, J.; Krumholz, G.; Chval, M.; Tisa, L. Appl. Environ. Microbiol. 2002, 68, 923. (38) Rother, J.; Millbank, J.; Thornton, I. Plant Soil 1982, 69, 239. (39) Kelly, J. J.; Haggblom, M.; Tate, R. L. Soil Biol. Biochem. 1999, 31, 1455. (40) Chaudri, A. M.; McGrath, S. P.; Giller, K. E.; Reitz, E.; Sauerbeck, D. R. Soil Biol. Biochem. 1993, 25, 301. (41) Renella, G.; Chaudri, A. M.; Brookes, P. C. Soil Biol. Biochem. 2002, 34, 121. (42) Smith, S. R.; Giller, K. E. Soil Biol. Biochem. 1992, 24, 781.
(43) Smolders, E. Brans, K. Coppens, E.; Merckx, R. Environ. Toxicol. Chem. 2001, 20, 2469. (44) De Boer, W.; Kowalchuk, G. A. Soil Biol. Biochem. 2001, 33, 853. (45) Speir, T. W.; Kettles, H. A.; Percival, H. J.; Parshotam, A. Soil Biol. Biochem. 1999, 31, 1953. (46) Doelman, P.; Haanstra, L. Plant Soil 1984, 79, 317. (47) Doelman, P.; Haanstra, L. Biol. Fertil. Soil 1989, 8, 235. (48) Blanck, H. Human Ecol. Risk Assess. 2002, 8, 1003. (49) McLaughlin, M. J.; Smolders, E. Environ. Toxicol. Chem. 2001, 20, 2639.
Received for review November 17, 2003. Revised manuscript received February 27, 2004. Accepted March 26, 2004. ES035278G
VOL. 38, NO. 11, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3097
