Science of the Total Environment 479–480 (2014) 284–291

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Dissipation of sulfamethoxazole in pasture soils as affected by soil and environmental factors Prakash Srinivasan a,b, Ajit K. Sarmah a,⁎ a b

Department of Civil & Environmental Engineering, Faculty of Engineering, The University of Auckland, Private Bag 92019, Auckland, New Zealand Landcare Research, Private Bag 3127, Hamilton, New Zealand

H I G H L I G H T S • • • •

Sulfamethoxazole dissipation was a combined effect of biotic and abiotic factors, with microbes being the major contributors. SMO dissipation rate in soils was independent of initial spiked concentration. Phospholipid fatty acid analysis was indicative of higher bacterial presence as compared to fungal community. Sulfamethoxazole is unlikely to persist more than 5–6 months in pasture soils at either depth.

a r t i c l e

i n f o

Article history: Received 9 January 2014 Received in revised form 5 February 2014 Accepted 5 February 2014 Available online 22 February 2014 Keywords: Sulfamethoxazole Dissipation Kinetics Dehydrogenase PLFA

a b s t r a c t The dissipation of sulfamethoxazole (SMO) antibiotic in three different soils was investigated through laboratory incubation studies. The experiments were conducted under different incubation conditions such as initial chemical concentration, soil depth, temperature, and with sterilisation. The results indicate that SMO dissipated rapidly in New Zealand pasture soils, and the 50% dissipation times (DT50) in Hamilton, Te Kowhai and Horotiu soils under non-sterile conditions were 9.24, 4.3 and 13.33 days respectively. During the incubation period for each sampling event the soil dehydrogenase activity (DHA) and the variation in microbial community were monitored thorough phospholipid fatty acid extraction analysis (PLFA). The DHA data correlated well with the dissipation rate constants of SMO antibiotic, an increase in the DHA activity resulted in faster antibiotic dissipation. The PLFA analysis was indicative of higher bacterial presence as compared to fungal community, highlighting the type of microbial community responsible for dissipation. The results indicate that with increasing soil depth, SMO dissipation in soil was slower (except for Horotiu) while with increase in temperature the antibiotic loss was faster, and was noticeable in all the soils. Both the degree of biological activity and the temperature of the soil influenced overall SMO dissipation. SMO is not likely to persist more than 5–6 months in all three soils suggesting that natural biodegradation may be sufficient for the removal of these contaminants from the soil. Its dissipation in sterile soils indicated abiotic factors such as strong sorption onto soil components to play a role in the dissipation of SMO. © 2014 Elsevier B.V. All rights reserved.

1. Introduction An estimated 9000 tonnes of antibiotics is annually used in the livestock industry by the US, 5000 tonnes by the European Union, and 6000 tonnes by China. After administration, a high proportion (30–90%) of the antibiotics is excreted by livestock animals in unchanged form and/or sometimes as metabolite/s (Sarmah et al., 2006). Occurrences of antibiotic residues are common in many parts of the world and have been detected in environmental media such as soils, ⁎ Corresponding author at: Department of Civil & Environmental Engineering, Faculty of Engineering, Private Bag 92019, Auckland 1142, New Zealand. Tel.: + 64 9 9239385; fax: + 64 9 3737462. E-mail address: [email protected] (A.K. Sarmah).

http://dx.doi.org/10.1016/j.scitotenv.2014.02.014 0048-9697/© 2014 Elsevier B.V. All rights reserved.

surface water, and ground water (Hamscher et al., 2003; Luo et al., 2011; Perret et al., 2006; Zuccato et al., 2005). Although the use of antibiotics in livestock industry in New Zealand (NZ) is not as widespread as in many other parts of the world, intra-mammary injectable antibiotics dominate the dairy industry (Srinivasan et al., 2013). According to the New Zealand Food Safety Authority, the sulfonamide group contributes ~17% of total antibiotic usage, and is a common class of antibiotics widely used in livestock industries in NZ. Free pasture grazing by millions of cattle is common in many parts of NZ, and dairy industry has been also expanding at a rapid rate especially in South Island of NZ. Because of direct excretal inputs by grazing animals and permitted activity such as land-application of animal waste by farmers, there is a concern that antibiotic residues may be entering the environment and could potentially impact the aquatic and terrestrial ecosystems.

P. Srinivasan, A.K. Sarmah / Science of the Total Environment 479–480 (2014) 284–291

In the last decade several studies have been conducted on the biodegradation of sulfonamide antibiotics such as sulfadiazine (SDZ), sulfamethazine (SMZ), and sulfachloropyridazine (SCP) in soils under diverse laboratory conditions (Accinelli et al., 2007; Fan et al., 2011; Halling-Sørensen et al., 2003; Kreuzig and Holtge, 2005; Thiele-Bruhn and Peters, 2007; Wang et al., 2006b; Yang et al., 2009). Most of the published studies reported in the literature are difficult to compare, as no two studies were similar in terms of the antibiotics investigated, and the experimental conditions and environmental matrices used. Limited studies on sulfonamide degradation in soils have shown that 50% dissipation half-life (DT50) values for sulfonamides ranged from as low as 1 day to 2 weeks under varied initial concentration and temperature (SI. Table 1). These studies are not appropriate to compare each other because of the differences in their experimental approaches and objectives. Some of the major findings from previous studies which focused on degradation of SMO or other compounds within the same group in soils suggest that sulfonamides may be more persistent than would be predicted from laboratory controlled studies (Bialk et al., 2005). Studies conducted by Accinelli et al. (2007) found that high initial concentration (100 mg kg−1) did not affect the dissipation rate of SMZ and SCP suggesting that at environmental concentrations (ppb or ppt level) there would be little or no effects. Elsewhere, SMO and trimethoprim showed higher dissipation than tylosin in soils, owing to greater sorption potential for the latter in soils (Liu et al., 2010). At a spiked concentration of 10 mg kg− 1, SDZ half-lives in aerobic nonsterile soils ranged from 12 to 18 days while it was more persistent in anoxic non-sterile soils with half-lives ranging between 57 and 237 days (Yang et al., 2009). While there have been a number of studies on sulfonamide degradation in manure, sludge amended soils, studies on pasture soils have hitherto been neglected. NZ pasture soils are high in organic carbon content and minerals such as allophane, and these properties have been found to have pronounced effects on the sorption of these compounds which could indirectly influence their degradation behaviour in soils (Srinivasan et al., 2012). Dissipation half-life is an important input parameter required in antibiotic fate modelling exercises and for risk assessment purposes. Sulfamethoxazole (SMO), which belongs to the sulfonamide group of antibiotics, was chosen for this study, as limited information exists about its fate in soil (Holtge and Kreuzig, 2007). Given the varied soil and climatic conditions of NZ, extrapolating degradation data obtained from overseas studies to NZ conditions may not reflect the true nature of SMO degradation behaviour. Although the terms ‘dissipation’ and ‘degradation’ have been used interchangeably in the literature, it would be appropriate to use the term ‘dissipation’ instead of ‘degradation’ in the work presented here as we did not attempt to identify metabolites formed during the experimental period. The main objective of this study was to conduct laboratory incubation experiments to investigate the dissipation kinetics of SMO antibiotic in topsoils and subsoils collected from three pasture soils (Te Kowhai, Hamilton and Horotiu). The use of subsoils in this study was necessitated by the marked differences in the values for pH, organic carbon, and microbial biomass of the top and subsoils, which could affect the overall dissipation behaviour of SMO. The incubation conditions were maintained at 60% maximum water holding capacity (MWHC) and with varying initial antibiotic spiked concentrations, different depth profiles, temperature regimes (7.5 °C and 25 °C) and with sterilisation at 60% MWHC. The principal focus was to derive the dissipation times (50%, 90%, and 99%) of SMO under each condition, and compare them to values that were reported in the literature. In order gain a better understanding about the dynamics of the SMO degradation under varied treatment conditions, we also performed phospholipid fatty (PLFA) analysis of samples and discuss our results in relation to the microbial community composition and their effects on the fate of SMO in the selected soils.

285

2. Materials and methods 2.1. Chemicals SMO (N98% purity), triphenyl tetrazolium chloride, Tris (hydroxymethyl) aminomethane (TRIS) buffer and triphenyl formazan were obtained from Sigma Aldrich, Australia. Acetonitrile (Mallinckrodt ChromAR, ≥ 99.8% purity), chloroform, acetone, methanol and dichloromethane (Mallinckrodt UltimAR, ≥ 99.9% purity) were obtained from Thermo Fischer Scientific Ltd. NZ. High Performance Liquid Chromatography (HPLC) grade deionised water was obtained from an onsite Arium® 61316 high performance reverse osmosis system (Sartorius Stedim Biotech GmbH, Germany). 2.2. Soils Topsoil and subsoil of three soils (Te Kowhai silt loam, Hamilton clay loam, and Horotiu silt loam) representative of dairy farming areas of Waikato region in the North Island of NZ were collected fresh from two depths (0–10, 30–40), sieved (2 mm), and stored at 4 °C until use. The soil pH was measured using a PHM62 standard pH meter, and organic carbon (OC) content was determined using an IL550 TOCTN analyser. The microbial biomass carbon (MBC) of the soils was measured by the fumigation extraction method (Wu et al., 1990). The moisture content (MC) of soils was determined gravimetrically at 105 °C and the water content was adjusted to 60% of MWHC. The soil was pre-incubated at 25 °C and 7.5 °C for 2 days before spiking with the antibiotic. These two temperatures were selected based on the typical summer and winter temperatures observed in the regions where the soils were collected from. The soils varied in their pH, OC, clay content and MBC as shown in Table 1. A full description of the soils and the methods used to determine their physico-chemical properties can be found elsewhere (Blakemore et al., 1987). 2.3. Dissipation experiments For all analyses, destructive soil samples (5 g) were performed to investigate the effect of each factor (temperature, soil depth, sterile vs non-sterile, concentration, and DHA) on SMO dissipation. Overall the experimental protocol in the dissipation experiment involved a total of 36 samples (12 each for temperature, depth and concentration effect) each with 2 replicates. Furthermore, another individual 18 samples (8 for sterile control and 10 for DHA measurement) each with 2 replicates were set up separately. Soil samples (5 g) were placed in 35 mL Kimax centrifuge tubes and appropriate amounts of SMO stock solution (1000 mg L− 1) prepared in methanolic solution earlier were spiked onto the soil to obtain an initial concentration of 5 or 0.5 mg kg−1. The amount of methanol present in the antibiotic solution spiked onto soil was unlikely to have any effect on soil microorganisms as we allowed the methanol to evaporate immediately after spiking inside a fume cupboard. The contents were then thoroughly mixed by vortexing before incubating in the dark at 25 °C and 7.5 °C respectively. The moisture content in each vial was maintained gravimetrically to 60% of its field capacity (− 33 kPa) by adding de-ionised water once every 3 days during the experiment, and the tubes were also aerated everyday to ensure a constant oxygen atmosphere. The entire experiment was conducted in closed incubators with temperature control, and wrapping individual tubes with aluminum foil in order to avoid photodegradation. To establish the role of microorganisms in the degradation of the antibiotic, the experiments were also conducted on sterile soils. Sterilisation was achieved by means of autoclaving twice (121 °C, 103 kPa for 30 min). Spiking procedure in sterile control treatment was similar to what was used in non-sterile treatments, except that sterile deionised water was used to maintain the moisture content at 60% of its field capacity during the fortification of soil samples in sterile experiment. All equipment used during sterile treatment was swabbed with

286

P. Srinivasan, A.K. Sarmah / Science of the Total Environment 479–480 (2014) 284–291

Table 1 Selected properties of soils used in the study. Soils

pH 1:2 (water)

OC (%)

Sand (%)

Silt (%)

Clay (%)

MC (%)

MWHC 60 (%)

MBC (μg C g−1)

DHA (μg g−1 h−1)

Horotiu TS Horotiu SS Te Kowhai TS Te Kowhai SS Hamilton TS Hamilton SS

5.7 6.6 6.7 5.7 5.8 5.1

8.2 1.7 5.0 0.5 4.0 0.8

34.0 34.0 9.0 12.3 13.7 13.4

48.0 48.0 54.0 62.8 51.0 40.3

17.0 17.0 37.0 24.9 30.4 46.2

49.0 60.9 23.9 41.4 23.7 23.8

121.2 134.8 79.6 84.5 77.6 75.5

816 584 1126 536 1724 620

28.66 3.18 13.78 1.42 16.69 1.28

TS = topsoil; SS = subsoil.

methanol, and autoclaved. All handling and operations in relation to the sterile treatment were performed inside a laminar flow cabinet. 2.4. Extraction and HPLC analysis Extraction of the antibiotic contained in the whole 5 g soils in each tube was performed at selected time intervals using sonication. Briefly, duplicate samples (5 g) at each treatment were extracted with 10 mL dichloromethane (DCM), vortexed (1 min), followed by 15 min of sonication, and shaken for 12 h in a rotary drum shaker. The tubes were centrifuged at 1750 g for 5 min, and an aliquot of 1.5 mL of DCM extract was evaporated to dryness under a gentle stream of nitrogen, reconstituted in methanol (0.5 mL), and immediately analysed using HPLC-UV/Fluorescence detection. The isocratic elution scheme used for the analysis of SMO antibiotic has been described in detail in an earlier study (Srinivasan et al., 2012). The use of DCM as a solvent to extract SMO from the soils resulted in acceptable recoveries ranging from 72 to 88% in the three topsoils and subsoils, with SMO recovery being the lowest (72%) for the Te Kowhai topsoil (SI. Fig. 4). Similar recovery of 76–85% (Blackwell et al., 2007) and 65% (Kay et al., 2004) was reported for sulfachloropyridazine in soils.

2.5. Measurement of soil bioactivity Dehydrogenase is a cellular enzyme only active in living organisms and thus is an indicator for soil microbial activity (Friedel et al., 1994). Experimental protocol earlier developed by Ghaly and Mahamoud (2006) was used to determine DHA of soils at selected sampling times during the incubation study. Full description of DHA determination is given in Appendix A and the Supplementary data.

2.6. Phospholipid fatty acid extraction (PLFA) For Hamilton top and subsoil (spiked with 5 mg kg−1) at selected incubation times, microbial community structure was characterised by determining the phospholipid fatty acid (PLFA) composition. Soil lipids were extracted from soil samples (5 g) at various selected sampling times using the method of Bardgett et al. (1996), and detailed experimental protocol has been described in the Supplementary data. PLFAs were separated by Gas Chromatography on a SGE 25QC3 BP-5 25 m × 0.32 μm film thickness with flame ionisation detector (at 150 °C, 400 mL min− 1, hydrogen at 30 mL min− 1 for 25 min). The separated fatty acids were identified by and quantified from chromatographic retention time comparison to bacterial methyl esters (Supelco Bacterial Acid Methyl Esters CP Mix 47080-U) as external standard. For each soil, the abundance of individual fatty acids was expressed as relative nmol g−1 of dry soil and standard nomenclature. The branched phospholipids i15:0, a15:0, i16:0, i17:0, and a17:0 were used as indicators for gram-positive bacteria markers, while the PLFAs 18:1 ω7, cy17:0, 18:1 ω9c, and cy19:0 were considered as gram-negative bacteria markers. The unsaturated PLFAs 18:1 ω6 and 18:2 ω6 were used as a fungal-biomass indicator (Frostegård et al., 1993; Hammesfahr et al., 2011).

2.7. Data and statistical analysis The dissipation of SMO in soils under varied treatments was modelled using simple first-order kinetics (SI. Fig. 3). Eq. (1) below represents the most simple form of the concentration–time relationship, where t is time (days), k1 is the dissipation rate constant (day−1) and M0 and Mt are the initial and final concentrations respectively, at time t ð−k1 t Þ

Mt ¼ M 0 exp

:

ð1Þ

The first-order dissipation rate and the corresponding DT50 (time required for 50% of the initial dose of sulfamethoxazole to be degraded) values were determined using Eq. (1). In order to compare the 90% and 99% dissipation times, we also calculated DT90 and DT99 to provide an overall perspective on the fate of compounds in the environment. A two-sample t-test was used to statistically examine two different kinetic data sets and to distinguish whether there was any significant difference in measured SMO concentration on sampling days between different soil types, initial concentrations used, soil depth, incubation temperature, and sterile treatment. In addition, Pearson's correlation coefficients (R) were also calculated to evaluate the influence of soil parameters on the rate of antibiotic dissipation. Other statistical analysis performed was a single one-way analysis of variance (ANOVA) to evaluate the influence of the factors soil depth and temperature on the dependent DT50 values. All significant differences were accepted at level p b 0.05. 3. Results and discussion 3.1. Dissipation in soils No lag phase was observed in the SMO dissipation in soils under any of the treatments investigated, implying the absence of acclimation period for the microbial population involved in the dissipation process. However, initial rapid dissipation observed could also account for some abiotic loss of SMO. Dissipation of SMO at 60% MWHC (−10 kPa) under varying initial concentrations (0.5 and 5 mg kg−1), soil depth (0–10 cm and 30–40 cm), temperature (25 °C and 7.5 °C) and under sterile and non-sterile conditions followed simple first order kinetics. In general, the SMO dissipation rate was found to be rapid irrespective of the soil depth at both initial concentrations, during the initial incubation time. The coefficients of determination (R2) for the first-order model fits ranged from 0.80 to 1.00 (Table 2) with the exception of Hamilton topsoil under sterile treatment, where dissipation deviated from the first-order kinetic as evident with low R2 value (0.61). No statistically significant difference (p b 0.05) was observed for the dissipation kinetic data sets for all the three soils between the experimental conditions within the incubation period. A simple Pearson's correlation matrix (SI. Table 2) was obtained using with the soil variables taken from Table 1, and using k values from Table 2. The significance test was carried out at 0.05 levels for the Pearson's correlation coefficient (R). The correlation matrix shows that pH, % clay, MBC and DHA to be positively correlated to the rate constant (k). An increase in MBC and DHA values implied greater biological activity, hence faster dissipation of SMO in the soils. Both pH and % clay content reduce SMO bioavailability as they

P. Srinivasan, A.K. Sarmah / Science of the Total Environment 479–480 (2014) 284–291

287

Table 2 Average first-order rate constants (day−1) and associated dissipation times (days) for SMO in three different soils under varying treatment conditions. Soils

Hamilton

Te Kowhai

Horotiu

Parameters

DT50 DT90 DT99 k1 M0 R2 DT50 DT90 DT99 k1 M0 R2 DT50 DT90 DT99 k1 M0 R2

25 °C non-sterile 0.5 mg kg−1

25 °C non-sterile 5 mg kg−1

7.5 °C non-sterile 0.5 mg kg−1

25 °C sterile 0.5 mg kg−1

TS

SS

TS

SS

TS

SS

TS

SS

11.36 37.75 75.49 0.06 3.10 0.99 NA*

12.38 41.12 82.24 0.06 4.62 0.94

9.24 30.70 61.40 0.08 0.28 0.84 4.31 14.30 28.60 0.16 0.31 0.97 13.33 44.28 88.56 0.05 0.22 0.98

11.75 39.03 78.05 0.06 0.39 0.97 14.15 46.99 93.98 0.05 0.40 0.96 12.38 41.12 82.24 0.06 0.36 0.97

25.39 84.34 168.69 0.03 0.36 0.97 20.69 68.73 137.47 0.04 0.31 0.94 23.18 77.01 154.02 0.03 0.26 0.91

29.88 99.25 198.50 0.02 0.42 0.91 20.95 69.56 139.13 0.03 0.40 1.00 19.69 65.41 130.83 0.04 0.35 0.96

11.00 36.55 73.10 0.06 0.40 0.93 13.00 43.20 86.40 0.05 0.32 0.93 18.10 60.12 120.24 0.04 0.37 0.89

34.66 115.13 230.26 0.02 0.35 0.61 22.43 74.52 149.03 0.03 0.40 0.90 22.65 75.25 150.50 0.03 0.38 0.88

NA*

k1 is the first order rate constant; M0 is the initial observed concentration; DT50, DT90, and DT99 are the degradation endpoints for 50, 90 and 99% antibiotic dissipation; R2 is the measure of the goodness of fit of the model. TS = topsoil; SS = subsoil. NA* = not applicable (For Te kowhai and Horotiu soil, data are not available, as the experiment on the effect of initial concentration was performed only in Hamilton soil).

enhance the process of sorption leading to permanently bound nonextractable residues in soils, and thus affecting the overall recoveries and leading to underestimation of k values.

magnitude greater than the typical environmental values for these compounds. The authors attributed this to more specific effects of the compounds on single microbial species that were possibly compensated by

3.2. Effect of initial concentration

120

7

Hamilton TS 100

5 mg kg

6

-1

0.5 mg kg

80

-1

DHA: 5 mg kg

5 -1

DHA: 0.5 mg kg

-1

4 3

40 2 20

1

0

0 -1 7

120

Hamilton SS 6

100

5

80

4 60 3

Dehydrogenase activity TPF in µg g-1h-1

60

% remaining

The concentration dependency of SMO dissipation was studied only in Hamilton soil (top and subsoils) under non-sterile and sterile treatments, at 25 °C using initial concentrations of 0.5 mg kg− 1 and 5 mg kg− 1 and the results are summarised in Table 2. There was no marked difference on DT50 values for Hamilton top and subsoils at 5 mg kg− 1 which were 11.4 and 12.4 days respectively. Ten-foldreduction in the initial spiked concentration down to 0.5 mg kg−1 also did not have a marked effect on the DT50 values for SMO (9.2 and 11.8 days respectively for top and subsoils). Though there was a 15–20% increase in the rate constant (k1) of SMO at low concentration, it was inconclusive whether the persistence of SMO was affected by initial chemical concentration given conflicting results reported in the literature which is discussed later. Thiele-Bruhn and Beck (2005) showed that the extractability of antibiotic sulfapyridine depended on its initial concentration and that it was most effective at smaller spiking levels. An examination of literature data suggests that SMO dissipation kinetics are dependent on the initial concentration and dissipation being slower at a higher concentration. A plausible explanation is that, since the higher initial concentration of antibiotic often involves spiking larger volumes of methanolic stock solution, the stock itself could indirectly inhibit microbial activity or even the high dosage of the antibiotic itself could be lethal to soil microbes (Ma et al., 2001), thus lowering the microbial activity of the soil in both cases. As shown in Fig. 1, the DHA activity of the topsoil (5 μg g−1 h−1 TPF), and subsoils (at about 0.5 μg g−1 h−1 TPF) at both initial concentrations (low and high) was similar indicating that microbial activity was not affected by the increase in the chemical concentration. Similar findings were also reported earlier involving other sulfonamides such as sulfachloropyridazine and sulfamethazine (Accinelli et al., 2007). The authors concluded that the concentrations greater than 100 mg kg−1 would be necessary to affect the microbial process involved in sulfonamide degradation. Thiele-Bruhn and Beck (2005) also reported that the antibiotic sulfapyridine and oxytetracycline had no effect on DHA, even at concentrations 1000 mg kg− 1, which are several orders of

40 2 20

1

0

0 -1 0

10

20

30

40

Days Fig. 1. Dissipation of SMO as a function of time in Hamilton soil (TS = topsoil and SS = subsoil) at initial spiked concentrations of 0.5 mg kg−1 (circle, red) and 5 mg kg−1 (triangle, blue). Error bars show deviation of the duplicate samples. The plots also show the DHA activity for 0.5 mg kg−1 (circle, red) and 5 mg kg−1 (triangle, blue) in the secondary axis.

P. Srinivasan, A.K. Sarmah / Science of the Total Environment 479–480 (2014) 284–291

the remainder of the microbial community. In contrast to our findings, Wang et al. (2006a) observed that the degradation rate constant decreased with increasing sulfadimethoxine concentration in manure, suggesting that the activity of degrading microorganisms was inhibited at high concentrations. In a separate study Wang et al. (2006b) observed decreasing bioactivity of the microorganisms with increases in the initial sulfamethoxine concentrations in manure thereby leading to lessened antibiotic degradation. Yang et al. (2009) varied the initial concentration for antibiotic sulfadiazine (1, 10 and 25 mg kg−1) and found half-life values of 2, 18 and 34 days respectively. The DT50 values obtained in our study for SMO were in agreement to values obtained for sulfadiazine by Yang et al. (2009), however, the initial concentrations used by the authors were 2 to 5 fold greater than those in our study (Table 2).

3.3. Effect of soil type and depth Figs. 1 and 2 show the general trend in the dissipation kinetics of SMO in three soils at two depths, and the associated DHA at each sampling event. SMO dissipated at a faster rate in Te Kowhai topsoil with N95% of the applied amount being lost within the half of the incubation period. The dissipation rate for SMO was initially slower in subsoils than the topsoils, especially in Te Kowhai soil; however, only 10% of the SMO remained on the last sampling day. Subsoil properties (clay content, OC, and MBC) correlated well with the lower dissipation rate for SMO in all soils except Horotiu soil (Table 1). However, the ANOVA results showed

120

7

Horotiu 100

TS: 0-10 cm SS: 30-40 cm DHA: 0-10 cm DHA: 30-40 cm

80

6 5 4 3

40 2

% remaining

20

1

0

0 -1

120

7

Te Kowhai

6

100

5

80

4 60 3

Dehydrogenase activity TPF in µµg g-1h-1

60

40 2 20

that DT50 values were not influenced by soil depth (Fcrit = 7.7; p = 0.23). It is important to note that the testing could have been biased given the small sample size in this case (n = 3). The DHA activity for subsoils was low at about 0.5 μg g−1 h−1TPF as compared to the topsoils which had an order of magnitude greater (N5 μg g−1 h−1TPF) values across the three soils (Figs. 1 and 2). However, despite low DHA in subsoils, SMO continued to dissipate at a similar rate as observed in the topsoil, implying that factors other than biotic processes influenced SMO dissipation rate in subsoils. There was a good correlation between the dissipation rate of SMO and the bioactivity of topsoils in our study. For instance, the Hamilton soil having the highest MBC gave the lowest DT50 values as compared to other soils. The opposite was true for Horotiu soil, which had the lowest MBC amongst other soils that gave high DT50 values. This was in contrast to some earlier studies (Monteiro and Boxall, 2009; Thiele-Bruhn and Beck, 2005), where no correlation was observed between the rate of dissipation and the soil bioactivity. Fig. 3 shows the fungal to bacterial PLFA ratios for Hamilton soil (5 mg kg−1). Analysis of the relative abundance of the major microbial groups during the incubation time revealed higher proportion of bacterial biomass over fungal biomass (fungal to bacterial ratio b1) for each sampling event. PLFA markers i-15:0, a-15:0, 15:0, i-16:0, i-17:0, cy-17:0, 17:0, and cy-19:0 showed bacterial community, while PLFA marker 18:2 ω6 was indicative of fungal presence (AntizarLadislao et al., 2008). For the topsoil, the ratio of fungal to bacterial PLFAs was higher and ranged between 0.12 and 0.21 when compared to subsoil which ranged between 0.05 and 0.15 respectively. Most PLFA markers for bacteria and fungi increased around the 12th and 7th days of incubation period for TS and SS (Fig. 3). The dissipation kinetic data also correlated well with the PLFA data, while an increased presence in microbes resulted in faster rate of dissipation for SMO, which was in agreement with studies conducted by Frostegård et al. (1993) indicating the relative importance of the bacterial and fungal energy channels in antibiotic dissipation. An interesting aspect of this work was the slower dissipation of SMO in Horotiu topsoil compared to its subsoil which could be attributed to a multitude of factors. Horotiu soil is volcanically derived and contains allophane (an alumino-silicate clay mineral), and has the highest % OC amongst the other soils studied. A recent work by Srinivasan et al. (2014) found that SMO sorption by Horotiu surface soil was much higher than the Te Kowhai and Hamilton soils. Thus high sorption of Horotiu soil could play a role in contaminant bioavailability, and influence the overall dissipation behaviour of SMO. The soil microbial activity in the Horotiu subsoil was 10-fold lower than the topsoil and in general, reduced levels of OC and low MBC are common characteristics 0.25 Hamilton-TS Hamilton-SS

0.20

F:B ratio

288

0.15

0.10

1

0

0

0.05

-1 0

10

20

30

40

Days

0.00

0

33

0.3

1

3

7

12

24

31

36

Days Fig. 2. Dissipation of SMO as a function of time in Horotiu and Te Kowhai topsoils (triangle, blue) and subsoils (circle, red), at initial spiked concentration of 0.5 mg kg−1. Error bars show deviation of the duplicate samples. The plots also show the DHA activity for topsoils (triangle, blue) and subsoils (circle, red) in the secondary axis.

Fig. 3. Distribution of the fungal:bacterial community ratio for Hamilton topsoil and subsoil as a function of the incubation time. Spiked concentration of SMO was 5 mg kg−1 while the incubated temperature was 7.5 °C.

P. Srinivasan, A.K. Sarmah / Science of the Total Environment 479–480 (2014) 284–291

of subsoils (Sarmah et al., 2009). Thus, it is conceivable that relative increase in the rate of dissipation for Horotiu subsoil compared with the other two subsoils could be due to decreased sorption, and greater bioavailability. Another plausible explanation for the faster dissipation in the subsoil despite a lower biological activity as shown by reduced DHA is that dissipation might be due to the existence of microbial species which are more specific in degrading the target compound in the subsoil (Di et al., 1998). In the present study, DT50, DT90 and even DT99 for all the three topsoils were obtained within the length of sampling time (40 days), except in subsoils, where DT90 and DT99 values occurred outside the sampling period. In general, DT50 values for SMO in subsoil increased when compared with the topsoils, which is consistent with the earlier assumption that reduced organic carbon and lower microbial activity at increased depths hinder degradation process.

289

3.4. Effect of incubation temperature SMO dissipation in soils as affected by the incubation temperature allowed a relative assessment of the effect of two temperatures in both top and subsoils (Fig. 4). Summarised datasets in Table 2 show that the dissipation rate constants (k1 day−1) for SMO were higher in all soils incubated at 25 °C than at 7.5 °C. For example, in Te Kowhai topsoil at 7.5 °C, DT99 for SMO was nearly 5-fold greater than at 25 °C. Overall, lower temperature resulted in reduced dissipation rate irrespective of the soil depth, and such a result is consistent either with microbial or chemical degradation. The ANOVA results also confirmed that DT50 values for SMO were most influenced by temperature (Fcrit = 7.7; p b 0.001). More than 80–90% of the applied antibiotic dissipated in the topsoils and subsoils incubated at 25 °C by days 20 and 40 respectively. However, at 7.5 °C it required nearly the whole duration of the

120

Hamilton TS

Hamilton SS

25oC

100

7.5oC Sterile

80 60 40 20 0

% remaining (initial concentration)

0

10

20

30

40

0

10

20

30

40

30

40

30

40

120

Horotiu TS

Horotiu SS

100 80 60 40 20 0 0

10

20

30

40

0

10

20

120

Te Kowhai TS

Te Kowhai SS

100 80 60 40 20 0 0

10

20

Days

30

40

0

10

20

Days

Fig. 4. Dissipation kinetics of SMO in Hamilton, Te Kowhai and Horotiu topsoils (TS) and subsoils (SS) at 25 °C and 7.5 °C, at initial spiked concentration of 0.5 mg kg−1 together with sterile control datasets at 7.5 °C. Error bars show deviation of the duplicate samples.

290

P. Srinivasan, A.K. Sarmah / Science of the Total Environment 479–480 (2014) 284–291

experiment (40 days) to reach N70% dissipation in topsoils and subsoils. Wang et al. (2006a) reported an increase in the dissipation rate constant of the antibiotic sulfadimethoxine from 0.332 to 0.777 day−1in manure when the temperature was increased from 25 °C to 40 °C. Elsewhere, at 6 °C, dissipation rates of four sulfonamides in activated sludge were found to be three to four-fold slower than at 20 °C (Ingerslev and Halling-Sørensen, 2000). In general, our study showed that when incubation temperature was decreased, DT50 values increased nearly two-fold for all soils irrespective of the soil depth. The soil biological activity as measured by the DHA indicated little or no bioactivity for soils incubated at 7.5 °C as opposed to the soils incubated at 25 °C, emphasising the role of microorganisms in the degradation of SMO and the role of temperature in moderating this. A likely explanation for this could be that during the dissipation experiment the contaminant bioavailability remains constant, and the overall dissipation follows first order kinetics. However, when the bioactivity is significantly altered by reducing the incubation temperature, the rate constant, which is temperature-dependent, also decreases resulting in greater persistence of the compound in soils. 3.5. Effect of sterilisation In order to investigate the relative role of microorganisms in the SMO dissipation, soils were sterilised, and Fig. 4 displays the dissipation pattern of SMO in non-sterile and sterile soils at both depths. An examination of data in Table 2 reveals that overall, rates of dissipation were slower and the associated DT50, DT90 and DT99 values were higher in sterile soil compared with non-sterile soils, and this was evident in all three soil types, except in Horotiu soil where dissipation rate for SMO was similar to non-sterile treatment (Fig. 4). Relatively smaller variation in dissipation parameters for SMO in sterile and non-sterile Horotiu soils was indeed surprising. The apparent consistency in the results between the two treatments in Horotiu soil led us to postulate that factors other than biotic processes may have played a role in the dissipation of SMO. Fig. 4 shows that 99% of the applied antibiotic disappeared by 73– 120 days in sterile topsoils, and within 149–230 days in sterile subsoils. Several explanations can be offered to support these incongruent findings; the possibility of chemical hydrolysis, chemical reduction, photolysis, or sorption to glass bottles. Given sulfamethoxazole contains nitroaromatic moiety like many other sulfa drugs, it is conceivable that the chemical species such as reduced sulphur, or iron compounds could potentially play a role on the abiotic dissipation of SMO (Mohatt et al., 2011; Zeng et al., 2012; Zhang and Weber, 2013). For example, Mohatt et al. (2011) demonstrated microbially-mediated abiotic transformation of SMO under ion-reducing soil conditions in a soil microcosm study. Remarkable dissipation of SMO under Fe(III)-reducing conditions with as much as N95% loss within 24 h was attributed to the abiotic reactions between SMO and Fe(II) generated by microbial reduction of Fe(III) soil minerals. Furthermore, a recent study by Zhang and Weber (2013) involving p-cyanonitrobenzene (pCNB) demonstrated that surface-associated Fe(II) and reduced dissolved organic carbon (DOC) acted as the key reductants in natural sediments. Although a justifiable explanation in our study could not be offered because of the absence of data for soil mineral such as ferrous, sulphur or DOC, the studies conducted in recent years shed lights on the fate of some nitro-aromatic compounds in the environment, and these findings can serve as a basis for understanding the abiotic loss mechanisms for other related compounds. Even though all precautions were taken to avoid photolysis, during manipulation of the samples (e.g. shaking, extraction, and analysis), some photolysis is bound to occur. Furthermore, dissipation of SMO could have also occurred as a result of other means. For example, given the presence of high amounts of allophane (alumina-silicate mineral) in Horotiu soil, it is conceivable that surface-induced abiotic transformation of SMO could have occurred due to the catalytic effect

with various clay minerals. The pH of soils used in the present study ranged from 5.1 to 6.7, and pH measurements of soils before and after autoclaving showed an increase by 0.2–0.4 log units, making sterilised soils slightly more alkaline. However, alkaline hydrolysis at this pH range is highly unlikely, and SMO has no structural features that can be hydrolysed (Loftin et al., 2008). Very little microbial activity was observed for sterile soils compared with non-sterile soils as evident by the DHA measurements (SI. Fig. 5). Even though the autoclaving was monitored by using autoclave tape and sterilisation indicators, the possibility of an artefact during the process of sterilising the soils cannot be ruled out. Autoclaving the soils can alter soil chemistry, and is known to change the physical, chemical, and microbiological properties of the soil due to the high treatment temperature and pressure involved (Fletcher and Kaufman, 1980; Wolf et al., 1989). For instance, autoclaving the soils has been found to increase the concentrations of dissolved organic carbon dramatically, providing a good environment for those bacterial spores that had survived sterile treatment (Tuominen et al., 1994). Since autoclaving kills the bacteria and not the spores (Nowak and Wronkowska, 1987), it is possible that the one autoclaving performed in this study may not have been sufficient to sterilise the soils. Other possible explanation for the abiotic loss of SMO in the soils could be attributed to the irreversible binding with the soil components through cross-coupling mechanisms forming non-extractable residues (Bialk et al., 2005), and binding onto the ketonic, carboxylic and phenolic carbon as well as aromatic C\H and methoxy/N-alkyl C sites (Kahle and Stamm, 2007). Regardless of the factors responsible for dissipation of SMO in the soils investigated, our findings suggest that SMO is unlikely to persist for a long duration on these NZ dairy farm soils.

4. Conclusion Concentration dependency of the SMO dissipation rate in the soils investigated remained unclear with disparity between the results obtained in this study with those reported so far in the literature. Overall dissipation rate of SMO varied with soil type, soil depth, and incubation temperature. The rate of dissipation was faster in topsoil as compared to subsoil in incubated samples collected from all the three sites, clearly demonstrating the prominence of microbes and their role in SMO dissipation. Both the degree of biological activity and the soil temperature influenced the overall SMO dissipation, and data revealed that this antibiotic is unlikely to persist more than 5–6 months in the three soils suggesting that natural attenuation may be sufficient for the removal of these contaminants from these pasture soils. There was a strong correlation between soil bioactivity (DHA) and soil MBC, consistent with the dissipation rate of SMO in all three soils. The PLFA analysis was indicative of higher bacterial presence as compared to fungal community, highlighting the type of microbial community responsible for dissipation. The dissipation rate constants were higher in soils incubated at higher temperature (25 °C), which was supported by the measured DHA showing little or no activity for soils incubated at 7.5 °C compared with 25 °C, emphasising the microorganism's role in the dissipation process that warmer temperatures can enhance SMO biodegradation. It was postulated that microbially mediated abiotic transformation of SMO under sulphur or iron reducing soil conditions could be one of the mechanisms of abiotic loss for SMO in the soils, however, much work is warranted to validate this argument under controlled laboratory conditions.

Acknowledgements This work was funded by the Foundation for Science, Research and Technology (New Zealand), through contract CO9X0705 (Landcare Research).

P. Srinivasan, A.K. Sarmah / Science of the Total Environment 479–480 (2014) 284–291

Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.scitotenv.2014.02.014. References Accinelli C, Koskinen WC, Becker JM, Sadowsky MJ. Environmental fate of two sulfonamide antimicrobial agents in soil. J Agric Food Chem 2007;55:2677–82. Antizar-Ladislao B, Spanova K, Beck AJ, Russell NJ. Microbial community structure changes during bioremediation of PAHs in an aged coal-tar contaminated soil by in-vessel composting. Int Biodeter Biodegr 2008;61:357–64. Bardgett R, Hobbs P, Frostegård Å. Changes in soil fungal:bacterial biomass ratios following reductions in the intensity of management of an upland grassland. Biol Fertil Soils 1996;22:261–4. Bialk HM, Simpson AJ, Pederson JA. Cross-coupling of sulfonamide antimicrobial agents with model humic constituents. Environ Sci Technol 2005;39:4463–73. Blackwell PA, Kay P, Boxall ABA. The dissipation and transport of veterinary antibiotics in a sandy loam soil. Chemosphere 2007;67:292–9. Blakemore LC, Searle PL, Daly BK. Methods for chemical analysis of soils. NZ Soil Bur Sci Rep 1987;80:103. Di HJ, Aylmore LAG, Kookana RS. Degradation rates of eight pesticides in surface and subsurface soils under laboratory and field conditions. Soil Sci 1998;163:404–11. Fan ZS, Casey FXM, Hakk H, Larsen GL, Khan E. Sorption, fate, and mobility of sulfonamides in soils. Water Air Soil Pollut 2011;218:49–61. Fletcher CL, Kaufman DD. Effect of sterilization methods on 3-chloroaniline behavior in soil. J Agric Food Chem 1980;28:667–71. Friedel JK, Molter K, Fischer WR. Comparison and improvement of methods for determining soil dehydrogenase-activity by using triphenyltetrazolium chloride and iodonitrotetrazolium chloride. Biol Fertil Soils 1994;18:291–6. Frostegård Å, Tunlid A, Bååth E. Phospholipid fatty acid composition, biomass, and activity of microbial communities from two soil types experimentally exposed to different heavy metals. Appl Environ Microbiol 1993;59:3605–17. Ghaly AE, Mahamoud NS. Optimum conditions for measuring dehydrogenase activity of Aspergillus niger using TTC. Am J Biochem Biotechnol 2006;2:186–94. Halling-Sørensen B, Sengelov G, Ingerslev F, Jensen LB. Reduced antimicrobial potencies of oxytetracycline, tylosin, sulfadiazin, streptomycin, ciprofloxacin, and olaquindox due to environmental processes. Arch Environ Contam Toxicol 2003;44:7–16. Hammesfahr U, Bierl R, Thiele-Bruhn S. Combined effects of the antibiotic sulfadiazine and liquid manure on the soil microbial-community structure and functions. J Plant Nutr Soil Sci 2011;174:614–23. Hamscher G, Pawelzick HT, Sczesny S, Nau H, Hartung J. Antibiotics in dust originating from a pig-fattening farm: a new source of health hazard for farmers? Environ Health Perspect 2003;111:1590–4. Holtge S, Kreuzig R. Laboratory testing of sulfamethoxazole and its metabolite acetyl-sulfamethoxazole in soil. Clean Soil Air Water 2007;35:104–10. Ingerslev F, Halling-Sørensen B. Biodegradability properties of sulfonamides in activated sludge. Environ Toxicol Chem 2000;19:2467–73. Kahle M, Stamm C. Sorption of the veterinary antimicrobial sulfathiazole to organic materials of different origin. Environ Sci Technol 2007;41:132–8. Kay P, Blackwell PA, Boxall ABA. Fate of veterinary antibiotics in a macroporous tile drained clay soil. Environ Toxicol Chem 2004;23:1136–44. Kreuzig R, Holtge S. Investigations on the fate of sulfadiazine in manured soil: laboratory experiments and test plot studies. Environ Toxicol Chem 2005;24:771–6. Liu F, Ying GG, Yang JF, Zhou LJ, Tao R, Wang L, et al. Dissipation of sulfamethoxazole, trimethoprim and tylosin in a soil under aerobic and anoxic conditions. Environ Chem 2010;7:370–6.

291

Loftin KA, Adams CD, Meyer MT, Surampalli R. Effects of ionic strength, temperature, and pH on degradation of selected antibiotics. J Environ Qual 2008;37:378–86. Luo Y, Xu L, Rysz M, Wang YQ, Zhang H, Alvarez PJJ. Occurrence and transport of tetracycline, sulfonamide, quinolone, and macrolide antibiotics in the Haihe River Basin, China. Environ Sci Technol 2011;45:1827–33. Ma QL, Gan J, Papiernik K, Becker JO, Yates SR. Degradation of soil fumigants as affected by initial concentration and temperature. J Environ Qual 2001;30:1278–86. Mohatt JL, Hu L, Finneran KT, Strathmann TJ. Microbially mediated abiotic transformation of the antimicrobial agent sulfamethoxazole under iron-reducing soil conditions. Environ Sci Technol 2011;45:4793–801. Monteiro SC, Boxall ABA. Factors affecting the degradation of pharmaceuticals in agricultural soils. Environ Toxicol Chem 2009;28:2546–54. Nowak A, Wronkowska H. On the efficiency of soil sterilization in autoclave. Zentralbl Mikrobiol 1987;142:521–5. Perret D, Gentili A, Marchese S, Greco A, Curini R. Sulphonamide residues in Italian surface and drinking waters: a small scale reconnaissance. Chromatographia 2006;63: 225–32. Sarmah AK, Meyer MT, Boxall ABA. A global perspective on the use, sales, exposure pathways, occurrence, fate and effects of veterinary antibiotics (VAs) in the environment. Chemosphere 2006;65:725–59. Sarmah AK, Close ME, Mason NWH. Dissipation and sorption of six commonly used pesticides in two contrasting soils of New Zealand. J Environ Sci Health B 2009;44: 325–36. Srinivasan P, Sarmah AK, Manley-Harris M, Wilkins AL. Development of an HPLC method to analyze four veterinary antibiotics in soils and aqueous media and validation through fate studies. J Environ Sci Health A 2012;47:2120–32. Srinivasan P, Sarmah AK, Manley-Harris M. Co-contaminants and factors affecting the sorption behaviour of two sulfonamides in pasture soils. Environ Pollut 2013;180: 165–72. Srinivasan P, Sarmah AK, Manley-Harris M. Sorption of selected veterinary antibiotics onto dairy farming soils of contrasting nature. Sci Total Environ 2014;472:695–703. Thiele-Bruhn S, Beck IC. Effects of sulfonamide and tetracycline antibiotics on soil microbial activity and microbial biomass. Chemosphere 2005;59:457–65. Thiele-Bruhn S, Peters D. Photodegradation of pharmaceutical antibiotics on slurry and soil surfaces. Landbauforschung Volkenrode 2007;57:13–23. Tuominen L, Kairesalo T, Hartikainen H. Comparison of methods for inhibiting bacterial-activity in sediment. Appl Environ Microbiol 1994;60:3454–7. Wang QQ, Bradford SA, Zheng W, Yates SR. Sulfadimethoxine degradation kinetics in manure as affected by initial concentration, moisture, and temperature. J Environ Qual 2006a;35:2162–9. Wang QQ, Guo MX, Yates SR. Degradation kinetics of manure-derived sulfadimethoxine in amended soil. J Agric Food Chem 2006b;54:157–63. Wolf DC, Dao TH, Scott HD, Lavy TL. Influence of sterilization methods on selected soil microbiological, physical, and chemical-properties. J Environ Qual 1989;18:39–44. Wu J, Joergensen RG, Pommerening B, Chaussod R, Brookes PC. Measurement of soil microbial biomass C by fumigation extraction — an automated procedure. Soil Biol Biochem 1990;22:1167–9. Yang JF, Ying GG, Yang LH, Zhao JL, Liu F, Tao R, et al. Degradation behavior of sulfadiazine in soils under different conditions. J Environ Sci Health B 2009;44:241–8. Zeng T, Chin YP, Arnold WA. Potential for abiotic reduction of pesticides in prairie pothole porewaters. Environ Sci Technol 2012;46:3177–87. Zhang H, Weber EJ. Identifying indicators of reactivity for chemical reductants in sediments. Environ Sci Technol 2013;47:6959–68. Zuccato E, Castiglioni S, Fanelli R. Identification of the pharmaceuticals for human use contaminating the Italian aquatic environment. J Hazard Mater 2005;122:205–9.