The effects of trace narasin on the biogeochemical N-cycle in a cultivated sandy loam

The effects of trace narasin on the biogeochemical N-cycle in a cultivated sandy loam

Science of the Total Environment 716 (2020) 137031 Contents lists available at ScienceDirect Science of the Total Environment journal homepage: www...

2MB Sizes 0 Downloads 4 Views

Science of the Total Environment 716 (2020) 137031

Contents lists available at ScienceDirect

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

The effects of trace narasin on the biogeochemical N-cycle in a cultivated sandy loam Stephanie L. DeVries a,⁎, Karin A. Block b,c, Madeline Loving b, Laura Logozzo b, Pengfei Zhang b,c a b c

Department of Biology, Geology & Environmental Science, University of Tennessee – Chattanooga, 615 McCallie Avenue, Chattanooga, TN 37403, USA Department of Earth and Atmospheric Sciences, City College of New York, 160 Convent Avenue, New York, NY 10031, USA Department of Earth and Environmental Sciences, Graduate School and University Center, City University of New York, 365 5th Avenue, New York, NY 10016, USA

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• The effects of low doses of antibiotic on soil nitrogen cycling are not wellknown. • Field moist soils experienced an increase in NH+ 4 from exposure to trace narasin. • Nitrate decreased in response to trace narasin except in near-saturated conditions. • N2O increased as a result of narasin exposure via all production pathways. • Trace antibiotic exposure may lead to soil N depletion from increased N2O flux.

a r t i c l e

i n f o

Article history: Received 23 October 2019 Received in revised form 27 January 2020 Accepted 29 January 2020 Available online 01 February 2020 Editor: Jay Gan Keywords: Nitrogen cycle Antibiotic Environmentally relevant concentration Nitrous oxide Nitrate Microbial processes

a b s t r a c t Narasin is an antibiotic administered to broiler chickens to prevent coccidiosis. After storage, excreta containing parent narasin compounds is commonly spread as nitrogen fertilizer, yielding initial soil concentrations in the low μg·kg−1 range. In soil, antibiotics have been found to modify one or more pathways in the biogeochemical nitrogen cycle. The concentrations tested are often too high to be considered environmentally relevant, despite evidence that sub-therapeutic doses may also be disruptive. We conducted soil mesocosm experiments to determine the overall impact of trace narasin on major nitrogen pools and fluxes in soils treated with 0, 1, 10, 100, or 1000 ng·kg−1 narasin. Water content in the mesocosms varied from 40% to 80% water-filled pore space (WFPS), simulating a range of different redox conditions. Under aerobic conditions (40% WFPS), exposure to narasin inhibited nitrification, yielding increases in soil ammonium by up to 76%, perhaps by targeting nitrifying fungi. Under the same conditions, narasin caused soil nitrate concentrations to decline 17–39%. When the soil was near saturation (80% WFPS), nitrate increased by an average of 30%. Mass balances and isotopic enrichment of N2O indicate that NAR may also affect anammox and the rate of nitrifier nitrification/denitrification. In aerobic soils, N2O flux increased with antibiotic dose and the rise in flux strongly correlates to the N2O:N2 product ratio from dentification. This relationship suggests that N2O flux may increase in soils exposed to narasin even when total denitrification is modestly inhibited. We conclude that trace concentrations of narasin can significantly modify biogeochemical activities in soil on short timescales. Our results indicate the potential for extremely low concentrations of antibiotics to impact agricultural productivity, terrestrial N2O flux, and nonpoint source nitrogen pollution. © 2020 Elsevier B.V. All rights reserved.

⁎ Corresponding author. E-mail addresses: [email protected] (S.L. DeVries), [email protected] (P. Zhang).

https://doi.org/10.1016/j.scitotenv.2020.137031 0048-9697/© 2020 Elsevier B.V. All rights reserved.

2

S.L. DeVries et al. / Science of the Total Environment 716 (2020) 137031

1. Introduction Nitrogen (N) compounds are a vital component of all living systems on Earth. In terrestrial and aquatic environments N is predominantly + found as nitrate (NO− 3 ) and ammonium (NH4 ). These reactive compounds provide essential nutrition to phytoplankton and plants that are the foundation of the global food chain. In excess, these compounds can also be harmful to the environment. Accelerated use of organic and inorganic N fertilizers in agriculture has led to a surplus of nitrogen in soil and water, contributing to eutrophication, toxic algal blooms, biodiversity loss, fishery collapse and increased production of paralytic shellfish toxins (Camargo et al., 2005; Hautier et al., 2009; Rabalais et al., 2009; Selander et al., 2008). N fertilizer use is also a leading cause of elevated NO− 3 in freshwater and contributes to terrestrial nitrous oxide (N2O) flux (Davidson, 2009). Elevated NO− 3 in drinking water has been linked to a number of human health risks including methemoglobinemia (blue-baby syndrome), colon cancer, and reproductive disorders (Ward et al., 2005). N2O, on the other hand, is an atmospheric pollutant that acts as a powerful greenhouse gas and is the leading modern contributor to stratospheric ozone depletion (Ravishankara et al., 2009). The nitrogen cycle's ecological and environmental importance has led to its intensive study for more than a century. However, the impacts of anthropogenic compounds, including C-rich pharmaceuticals and antibiotics on the microbial nitrogen cycle have only begun to be investigated thoroughly in the last ten years (DeVries and Zhang, 2016; Grenni et al., 2018; Roose-Amsaleg and Laverman, 2016). At therapeutic doses (Nmg·kg−1), all steps of the N-cycle are shown to be impacted by one or more antibiotics and most antibiotics inhibit soil microbial processes. Broad-spectrum antibiotics appear most likely to impact nitrification. In instances where nitrification was not affected, positive shifts in the ratio of ammonia oxidizing archaea (AOA) to ammonia oxidizing bacteria (AOB) were used to attribute this outcome to functional redundancy in the soil microbial community (Campos et al., 2001; Kotzerke et al., 2008). In some cases, no observable effect is more easily explained by the specific activity of the antibiotics tested. Nitrobacteraceae, a family of micro-organisms that includes the majority of bacterial nitrifiers, e.g., Nitrobacter and Nitrosomonas (Belser, 1979) are gram-negative bacteria. When introduced to soil, neither bacitracin (Banerjee and D'Angelo, 2013) nor monensin (Konopka et al., 2015; Toth et al., 2011), inhibited nitrification, probably because neither is active against gram-negative bacteria. Recent literature reviews show that sub-therapeutic antibiotic exposure in soil also affects the biogeochemical N cycle (DeVries et al., 2015; Roose-Amsaleg and Laverman, 2016). Several of these studies have shown that antimicrobial compounds at low (bmg·kg−1) and trace (bμg·kg−1) soil concentrations can either inhibit or stimulate denitrification (Chen et al., 2019; D'Alessio et al., 2019; DeVries et al., 2015). At least three studies reported an increase in N2O flux from low concentrations of tetracycline, sulfamethazine, and narasin, an outcome that was proposed to result from either a shift in the N2O:N2 ratio from denitrification or towards other N2O-producing pathways (DeVries et al., 2015; Semedo et al., 2018; Shan et al., 2018). In contrast, Chen et al. (2019) observed reduced N2O flux from paddy soils treated with triclosan and triclocarban. The antibiotics demonstrating activity at trace concentrations in soil include narasin (NAR), an anti-coccidiodal drug approved for therapeutic and prophylactic use and for growth-promotion in large-scale poultry production. When NAR is administered as feed to broiler chickens (50–80 mg·kg−1), the parent compound is excreted at rates that yield 6.5 mg·kg−1 NAR in fresh poultry litter (Elanco Animal Health, 2017). Applying fresh litter as a nitrogen fertilizer at a typical rate of 3 tons/ acre (Camberato, 2016) can yield up to 30 μg·kg−1 NAR in the upper 10 cm of soil. This concentration can be significantly reduced if litter is stored prior to land application. In storage, NAR in poultry litter has a half-life of approximately 7 days (Elanco Animal Health, 2017),

therefore b1 μg·kg−1 NAR is consistent with application of litter that was stored for ≥4 weeks prior to soil application. Soil concentrations in that range (2.2 μg·kg−1) have been reported (Bak et al., 2013). Detection of lesser concentrations is precluded by limits of quantification of NAR in soils, which range from 0.68 μg·kg−1 (Bak et al., 2013) in soil to 1 μg·kg−1 in sewage sludge (Herrero et al., 2013). NAR is a relatively sorptive compound (Kd = 38.8–98.4 L·kg−1) (Hussain and Prasher, 2011), so it is unlikely to undergo significant leaching. Consequently, NAR has the potential to persist in the upper soil profile at trace soil concentrations for months after initial exposure. In 2018, over 9 billion chickens were produced in the U.S. and 30 states reported production in excess of 500,000 broilers (USDA, 2019). Ritz and Merka (2009) report that each broiler chicken produces 1.3 kg of litter whereas laying hens and breeders can produce up to 10× that amount. Therefore, the potential for NAR contamination resulting from litter applications to soil is significant and warrants environmental impact studies at relevant soil concentrations. The objective of this study is to examine the effects of trace (1, 10, 100, and 1000 ng·kg−1) concentrations of NAR on environmentally important N pools and related transformation pathways in an agricultural soil. − This was accomplished by quantifying the changes in NH+ 4 , NO3 , and N2O flux, as well as mineralization, nitrification, and denitrification rates in soils with low, medium, and high soil moisture. Our working hypotheses were (1) that the NH+ 4 pool would not be significantly affected by NAR exposure across redox conditions, but that inhibited denitrification would lead to an increased accumulation of NO− 3 , especially under high soil moisture; and (2) that NAR exposure would increase N2O emissions regardless of redox conditions. 2. Materials & methods 2.1. Soil sampling and preparation The soil used in this study is an Ultisol (89.0% sand, 5.5% silt, 5.5% clay) sampled from a coastal farm along the Upper Indian River Bay near Milford, Delaware (Fig. 1). Total organic carbon (TOC), measured by loss on ignition (LOI), is 1.86%. The climate at the study site is temperate with hot, humid summers and cool winters. The agricultural fields are bordered by a narrow, steep riparian zone adjacent to the Indian River Bay, which receives runoff and groundwater discharge from this site. The history of the site is known beyond 20 years by personal communication with the farmer who leases the land and the authors are assured that the soils have not previously been amended with organic fertilizers (manure, wastewater sludge, etc.) that would have introduced antibiotics or other pharmaceutical products to the soil. Groundwater sampling conducted at this site in 2012 (unpublished data) reveals no trace of pharmaceutical contamination and corroborates this claim. Sandy loam topsoil samples were composited from 10 cm cores, collected along 5 transects, air-dried, sieved to 2 mm, and stored at 4 °C. 2.2. Experimental setup Three sets of soil incubation experiments were performed. Each set consisted of 120 soil mesocosms (5 antibiotic concentrations × 2 different 15N-enrichments × 4 days, each prepared in triplicate). Seventy-five grams of soil were placed into 50 cm3 lidded polycarbonate containers treated with the same NAR dose. One milliliter of NAR solution (0, 0.075, 0.75, 7.5, or 75 μg·L−1) prepared from a 1 mg·L−1 NAR standard (5 mg, ≥98% purity in 5 mL methanol, Sigma Aldrich, St. Louis, MO) was transferred to the soil chambers, resulting in soil concentrations of 0, 1, 10, 100, or 1000 ng kg−1 NAR. One milliliter of 15N-enriched nutrient solution was administered to each mesocosm, delivering 93 μg·g−1 NH+ 4 −1 N, and 43 μg g−1 NO− surface application 3 -N (equivalent to 72 kg N ha rate). Half of the samples were enriched to 10% atom excess 15N as 15N15 NH+ N-(NH4)2SO4, Cambridge Isotope Laboratories, Tewksbury, 4 (99%

S.L. DeVries et al. / Science of the Total Environment 716 (2020) 137031

3

Fig. 1. Map showing the location of the field site, Bull's Eye Farm (inset), which is approximately 5 miles east of the Millsboro, DE. 15 MA), the other half as 15N-NO− N-KNO3, Cambridge Isotope Lab3 (99% oratories, Tewksbury, MA). Soil moisture was adjusted to 40% (Set 1), 60% (Set 2), or 80% (Set 3) water-filled pore space (WFPS) to replicate conditions ranging from mostly aerobic to mostly anaerobic. Following these additions, the Day 0 samples (Fig. 2) were immediately sampled for initial nutrient concentrations and 15N enrichment of the NH+ 4 and NO− 3 -N-pools. The remaining containers were placed in the bottom of 500 mL glass jars, and subsequently capped with stainless steel lids outfitted with two gas-tight sampling ports and three-way stopcocks (Fig. 3). Once sealed, the jars were incubated in the dark at room temperature (~23 °C).

2.3. Extraction and analysis of soil headspace Headspace samples were collected 24, 48, and 72 h after the start of each incubation experiment to quantify 15N-N2, 15N-N2O, and N2O flux. Two 25-milliliter sample lock syringes were attached to the sampling ports on the jar lids. The syringes were primed by withdrawing 5 mL headspace into each syringe and discarding the gas to the atmosphere through the three-way stopcock. The remaining headspace gas was mixed by extracting and purging 25 mL into one syringe, followed by the other, for 3 cycles. After mixing, 25 mL headspace gas was withdrawn and flushed through a pre-evacuated Exetainer vial (Labco Ltd, High Wycombe, UK); the final 12 mL was retained for analysis. 30N2O, % atom excess of 15N in N2O, and total N2O flux were measured for all samples; 30N2 was measured for soils incubated at 40% and 60% WFPS

only. N2O flux data from 40% WFPS was reported by DeVries et al. (2015). 2.4. Extraction and analysis of soil nitrogen Two ~5 g soil “cores” were collected from each mesocosm using a hollowed-out 3 mL plastic syringe. The gravimetric water content of each soil was measured by weighing one of the cores before and after − oven drying at 105 °C. NH+ 4 and NO3 were extracted from the second “core” using 40 mL of 2 M KCl and the concentrations in the soil extracts were determined by automated colorimetric analysis (SEAL AQ2 Discrete Nutrient Analyzer, Seal Analytical, Mequon, Wisconsin, USA). 15 − To separate NH+ N analysis, the extracts were se4 and NO3 for quentially diffused onto acidified glass fiber filters (Holmes et al., 1998; Sigman et al., 1997; K. Kroeger, personal communication). The filters were placed in a desiccator with fuming H2SO4 and then wrapped in a tin capsule (Costech Analytical Technologies, Inc., Santa Clara, CA). 15 N enrichment was determined by Isotope Ratio Mass Spectrometry (IRMS) at the UC-Davis Stable Isotope Facility using a PDZ Europa ANCA-GSL elemental analyzer interfaced to a PDZ Europa 20–20 isotope ratio mass spectrometer (Sercon Ltd., Cheshire, UK). Samples were combusted at 1000 °C in a reactor packed with chromium oxide and silvered copper oxide. Following combustion, oxides were removed in a reduction reactor (reduced copper at 650 °C). The helium carrier then flows through a water trap (magnesium perchlorate) and a CO2 trap (for N-only analyses). N2 and CO2 were separated on a Carbosieve GC

4

S.L. DeVries et al. / Science of the Total Environment 716 (2020) 137031

Fig. 2. Schematic diagram the experimental set-up for the 1 ng·kg−1 dose. 1 mL of 75 ng·L−1 NAR and 1 mL of nutrient solution was added to each soil aliquot (75 g dry soil). Each nutrient −1 + solution contained 1.24 μg·mL−1 NH+ NO− 4 -N as (NH4)2SO4, and 0.57 ng·mL 3 -N as NaNO3. For the “A” samples (shaded) the NH4 -N amendment was enriched with 10% atom excess 15 15 N-NH+ N-NO− 4 whereas the “B” samples (unshaded) were enriched with 10% atom excess 3 .

column (65 °C, 65 mL/min) before entering the IRMS. Samples are interspersed with several replicates of at least two different laboratory standards. These laboratory standards, which are selected to be compositionally similar to the samples being analyzed, have been previously calibrated against NIST Standard Reference Materials (IAEA-N1, IAEA-N2, IAEA-N3, USGS-40, and USGS-41). Each sample's preliminary isotope ratio was measured relative to reference gases analyzed with each sample. These preliminary values were finalized by correcting the values for the entire batch based on the known values of the included laboratory standards. 2.5. Calculating rates of mineralization, nitrification, and denitrification Gross mineralization (m) and nitrification (n) rates (μg N·g−1 soil·d−1) were determined from 15N enrichment data using the theoretical model described by Barraclough (1995). This approach, modified to simplify Kirkham and Bartholomew's (1954) classic theoretical model, is based on dilution of an enriched nitrogen pool such as NH+ 4 or NO− 3 . To calculate gross mineralization rate, the total concentration 14+15 −1 of NH+ NH+ dry soil) and 4 is assigned the variable M (μg 4 -N·g the abundance of 15N-NH+ (% atom excess) is assigned the variable A 4 −1 (μg 15NH+ dry soil). The rate equation (Eq. (1)) is based upon 4 -N·g the initial and final values of M and A, where the final value occurs at

time t.   A0 M0 −Mt At m¼  log ð1 þ ðM0 −Mt Þ=M0 Þ t log

ð1Þ

The same rate equation is applied to nitrification rate, however the −1 measured N pools (M and A) are taken as μg 14+15NO− dry soil 3 -N·g 15 − and % atomic excess NO3 -N, respectively, and the symbol m is replaced by n. Overall mineralization (m) and nitrification (n) rates were also calculated from the non-linear curve-fitting function A(t): Aðt Þ ¼

A0 m  ðM 0 −Mt Þ ðM0 −Mt Þ t 1þ M0

ð2Þ

Denitrification (d) rate (μg N·g−1 soil·d−1) in the soil samples was calculated using the classic equations of Hauck et al. (1958) as modified by Siegel et al. (1982). The modified equations simplify the original by assuming the change in 28N2 in closed systems is negligible. The ion current ratio, r′, is measured from the ion currents of N2 at masses 28, 29,

Fig. 3. Soil incubation chambers. Individual soil samples were placed inside 500 mL Kilner Jars with three-way stopcocks used for headspace sampling.

S.L. DeVries et al. / Science of the Total Environment 716 (2020) 137031

5

%15Nflux.

and 30, where ion current at mass 30 r ¼ ion current of mass 28 þ ion current of mass 29 0

ð3Þ

The fraction of N2 evolved from denitrification, d, is subsequently determined by: r0 d ¼  2 X 15 N

ð4Þ

15 N-NO− where X15 N is the mole fraction of 3 .

2.6. Calculating mass balance − The measured soil concentrations of NH+ 4 and NO3 were compared to a mass balance in which

 þ   NH 4 t ¼ NHþ 4 0 þ t ðm−nÞ

ð5Þ

15

% N of mix ¼

     ðNair Þ %15 N air þ Nflux %15 Nflux

ð7Þ

N air þ Nflux

where %15N of mix = % atom excess measured Nair = μmol of N2O-N in air, assumed to be 326 ppb %15Nair = % atom excess 15N, assumed to be 0.3663 Nflux = μmol of N2O-N measured - Nair 2.8. Calculating relative contributions of nitrification and denitrification to N2O flux The relative contributions of mineral nitrogen from denitrification and nitrification to total N2O flux DN 2O and NN 2O, respectively, were calculated from Eqs. (8) and (9). Eq. (8) assumes that denitrification is the only source of 15N-N2O in 15NO− 3 labeled soils, thus the contribution of denitrification to total N2O is directly determined from those measurements. Eq. (9) assumes that both nitrification and nitrification-coupleddenitrification (NCD) contribute to 15N-N2O in 15NH+ 4 labeled soils and includes a correction to account for 15N-N2O resulting from NCD to determine the contribution of nitrification. This correction becomes irrelevant when denitrification is the dominant N2O source.

and DN2 O ¼ 15 NNO−3

 −   NO3 t ¼ NO− 3 0 þ t ðn−dÞ

15

NN2 O ¼

where m = net mineralization rate (μg N·g−1 soil·d−1) n = net nitrification rate (μg N·g−1 soil·d−1) d = denitrification rate (μg N·g−1 soil·d−1) t = time (d) 2.7. Calculating 15N-N2O enrichment

15

N NHþ − 4

15

N

NO− 3

 15

N NHþ

! ð9Þ

4

NNO−3

15 NNH+4 is N2O where 15NNO−3 is total N2O flux from 15NO− 3 amended soil, 15 15 − is nitrate from flux from 15NH+ amended soil and N NO− 4 NO 3 3 amended soil.

3. Results and discussion

15 N isotopic enrichment of N2O collected from headspace was used to estimate the relative contributions of nitrification and denitrification to total N2O flux. Measured %15N-N2O enrichment values were “corrected” using the mixing ratio of the general form and solving for

40% WFPS NH+4 -N (mg kg-1)

ð8Þ

ð6Þ

3.1. Influence of NAR on NH+ 4 + −1 Figs. 4 and 5 show the concentration of NH+ ) and 4 (μg NH4 -N·g −1 − NO− (μg NO -N·g ) under different moisture conditions (40%, 60%, 3 3

60% WFPS

200

80% WFPS

Day 1

Day 1

Day 1

Day 2

Day 2

Day 2

Day 3

Day 3

Day 3

150 100 50

NH+4 -N (mg kg-1)

0 200 150 100 50

NH+4 -N (mg kg-1)

0 200 150 100 50 0

0

1

10

100

Narasin Dose (ng kg -1)

1000

0

1

10

100

Narasin Dose (ng kg -1)

1000

0

1

10

100

Narasin Dose (ng kg -1)

Fig. 4. Box-whisker plots illustrating the NH+ 4 -N pool over a three-day incubation period in moist soils (40%, 60%, and 80% WFPS) treated with NAR.

1000

6

S.L. DeVries et al. / Science of the Total Environment 716 (2020) 137031

40% WFPS

60% WFPS

80

80% WFPS

Day 1

Day 1

Day 1

Day 2

Day 2

Day 2

Day 3

Day 3

Day 3

60 40

NO-3 -N (mg kg-1)

20

80

NO-3 -N (mg kg-1)

NO-3 -N (mg kg-1)

100

80

60 40 20

60 40 20 0

0

1

10

100

1000

0

Narasin Dose (ng kg -1)

1

10

100

1000

0

1

Narasin Dose (ng kg -1)

10

100

1000

Narasin Dose (ng kg -1)

Fig. 5. Box-whisker plots illustrating the NO− 3 -N pool over a three-day incubation period in moist soils (40%, 60%, and 80% WFPS) treated with NAR.

3.2. Influence of NAR on NO− 3

and 80% WFPS) as a function of time and NAR dose. Results of one-way ANOVA to assess the statistical significance of the resulting doseresponse curves are available in the Supplemental information (Tables S1 and S2). Out of the three moisture regimes, NH+ 4 was most distinctly affected by exposure to NAR in semi-dry soils (40% WFPS) where aerobic conditions are favorable for ammonia oxidation reactions. Under these conditions, NAR exposure increases the size of the ex−1 tractable NH+ dose, 4 pool at most of the doses tested. At the 10 ng·kg + however, the NH4 pool decreased relative to the control. At this dose, the corresponding mineralization and nitrification rates (Table 1) suggest that strongly inhibited mineralization is the principal cause for reduced NH+ 4 . Although mineralization rates were inhibited in all the NAR-treated soils, the degree of inhibition is nearly 4× greater at the 10 ng·kg−1 dose than at any other dose tested. Biphasic dose responses like this are not uncommon at extremely low concentrations (Calabrese and Baldwin, 2001) and have previously been observed for the growth of soil fungi exposed to trace concentrations of fungicide (Garzon and Flores, 2013). At the remaining doses, increased NH+ 4 indicates inhibited nitrification. Inhibition is also observed at larger doses (100 and 1000 ng·kg−1 NAR) in soils with higher moisture content (60% and 80% WFPS). Since NAR has no activity against gram-negative nitrifying bacteria but is known to have antifungal properties (Berg and Hamill, 1978), this result indicates that nitrifying fungi are inhibited by NAR at trace concentrations. This explanation is supported by previous work showing that low doses of fungicides decrease ammonification (Wainwright and Pugh, 1973).

Results of one-way ANOVA show that statistically significant changes to the NO− 3 pool occurred in soils incubated with 1, 10, 100, or 1000 ng·kg−1 NAR under all experimental conditions (Table S2). At lower soil moisture (40% and 60% WFPS), the addition of NAR results in overall losses of NO− 3 from the soil system. These losses roughly increase with dose and up to 50% less NO− 3 was extracted from soils treated with NAR than in the soil controls. Denitrification rates, nitrification rates (Table 1) and related trends in the NH+ 4 pool indicate that a loss of NO− 3 in soils treated with NAR results from inhibited nitrification, which reduces additions of NO− 3 , rather than from accelerated denitrification, which depletes the NO− 3 pool. This is especially evident at higher doses, e.g. 100 ng·kg−1 or 1000 ng·kg−1 NAR. At 80% WFPS, extractable NO− 3 roughly increased with NAR dose. This is consistent with results from other antibiotic studies, in which at least 12 different antibiotics were found to inhibit denitrification after exposure to low or therapeutic concentrations (DeVries et al., 2015; Grenni et al., 2018; RooseAmsaleg and Laverman, 2016). More importantly, these results demonstrate that even trace concentrations may have environmentally relevant impacts on soil nitrogen. Mass balances (Figs. S1 and S2) were used to examine whether sec− ondary NH+ 4 and NO3 production or consumption processes, i.e., remineralization, immobilization, anaerobic ammonia oxidation (anammox), nitrifier nitrification/denitrification, or dissimilatory nitrate reduction (DNRA) may also have contributed to changes in the

Table 1 Mineralization (M), nitrification (N), and denitrification (D) rates (μg N·g−1·d−1) over 3-day incubation period for 40% and 60% WFPS. NAR Dose (ng·kg−1)

40% WFPS M

0 1 10 100 1000

2.2 1.4 0.3 1.1 1.2

± ± ± ± ±

0.1 0.1 0.06 0.05 0.08

60% WFPS N

D

8.2 ± 0.4 2.0 ± 0.2 5.0 ± 0.3 0.7 ± 0.07 −0.3 ± 0.1

17.8 14.7 14.9 18.8 15.8

M ± ± ± ± ±

9.9 5.2 4.8 8.1 5.5

1.5 1.7 1.1 1.1 0.3

± ± ± ± ±

0.02 0.04 0.05 0.07 0.04

N

D

1.5 ± 0.04 1.2 ± 0.05 1.1 ± 0.04 0.01 ± 0.1 −0.7 ± 0.05

22.2 22.7 15.7 17.3 10.5

± ± ± ± ±

2.0 4.2 4.2 11.3 1.8

S.L. DeVries et al. / Science of the Total Environment 716 (2020) 137031

mineral N content during incubation. In the 40% WFPS experiment, the mass balance overestimates NH+ 4 on Days 2 and 3, particularly at the lowest doses (1 and 10 ng·kg−1 NAR). The isotopic composition of N2O flux (see Section 3.3) indicates that nitrifier nitrification and/or nitrifier denitrification may account for some portion of this imbalance. − These pathways reduce NH+ 4 to N2O without adding to the NO3 pool and the losses are therefore not captured by nitrification rate measure− ments. Anammox, which also consumes NH+ 4 without producing NO3 (NH+ ➔ NO ➔ N ) could result in lower than expected concentrations 4 2 2 of NH+ 4 . Any of these three pathways could also explain the mass balance of NO− 3 in the 40% WFPS experiment, which underestimates soil −1 NO− NAR on Day 1 and at 3 concentration at 1 and 10 ng·kg −1 N10 ng·kg NAR on Days 2 and 3. At 60% WFPS, the mass balances gen− erally agree with measured NH+ 4 and NO3 , indicating that these processes were less important under the higher moisture conditions. This inference is supported by the relative contributions of nitrifier nitrification and/or nitrifier denitrification to N2O which were also found to be much less important in wetter soils, i.e., ≥60% WFPS. 3.3. Influence of NAR on N2O N2O-N flux rates (ng·g−1·day−1) for each set of soil incubation experiments are shown in Fig. 6. The N2O flux rate at 40% WFPS was previously reported to be very low, b6 ng g−1 d−1 by DeVries et al. (2015). However, one-way ANOVA shows that the effects of NAR exposure were statistically significant on all 3 days (Table S3). Initially, N2O production is suppressed relative to the control in all NAR-treated soils, but by Day 3 N2O flux increases relative to the controls by 70% (1 ng·kg−1 NAR) to 114% (1000 ng·kg−1 NAR). A similar pattern of increased N2O flux with NAR dose was observed in soils incubated at 60% WFPS. In these experiments, N2O flux was nearly two orders of magnitude greater than in drier soils, ranging from 53 to 161 ng N2O-N g−1 d−1. Although median daily flux rates declined over the course of the 3-day incubation period, N2O flux from NAR-treated soils remained elevated throughout. N2O results from at least three known pathways: denitrification (NO− → NO− → NO → N2O), nitrifier nitrification (NH+3 2 4 − → NH2OH → N2O), and nitrifier denitrification (NH+ 4 → NO2 → N2O). The relative contributions of denitrification and nitrifier nitrification

6 4 2

N2 O-N (ng g -1 d-1)

4 2 10

200 100

8 6 4 2 0

1

10

100

Narasin Dose (ng/kg)

1000

Day 1

80 60 40 20 100

Day 2 300 200 100

Day 2

80 60 40 20 100

400

Day 3

N2 O-N (ng g -1 d-1)

N2 O-N (ng g -1 d-1)

300

N2 O-N (ng g -1 d-1)

6

80% WFPS

100

Day 1

400

Day 2

8

N2 O-N (ng g -1 d-1)

N2 O-N (ng g -1 d-1)

10

The risk of soil contamination by NAR is substantial due to its widespread use in the broiler chicken industry and associated spreading of poultry litter fertilizer. This study shows that even trace concentrations of NAR in soil have the potential to modify soil microbial activity and Nturnover rates. Understanding these processes is important to

Day 3 300 200 100

0

1

10

100

Narasin Dose (ng/kg)

1000

N2 O-N (ng g -1 d-1)

N2 O-N (ng g -1 d-1)

8

3.4. Environmental and agricultural implications

60% WFPS

400

Day 1

(NN) or nitrifier denitrification (ND) to soil N2O flux is illustrated in Fig. 7. The importance of each pathway to net N2O flux depends upon the physiochemical properties of soil and the composition of the microbial community. Denitrification is frequently deemed the principal source of soil N2O but additional flux from NN and ND can be significant under some conditions (Ma et al., 2017; Zhu et al., 2013). At low water content (40% WFPS), the sum of NN and ND accounted for the majority of N2O flux in nearly all samples measured. However, total flux was extremely low for all NAR doses (0.38–2.40 ng·kg−1·d−1) and the ratio of N2O:N2 flux (Fig. 8) indicates that the dose-related increase in N2O is more directly related to denitrification. At both 40% and 60% WFPS, NAR treatments resulted in a positive shift in the N2O:N2 product ratio that strongly correlates to measured N2O flux (Table 2). This relationship shows that when net N2O flux is very low and primarily results from NN and ND, the composition of the denitrifying community still exerts influence that is measurably altered by trace additions of NAR. At 60% WFPS, where denitrification was shown to be the dominant N2O source, a similar shift in the N2O:N2 ratio demonstrates that exposure to NAR also has the potential to increase N2O flux when total denitrification is modestly inhibited. Because N2O flux in these semi-moist soils is also the highest observed (51–159 ng·kg−1·d−1), this effect of NAR is particularly significant and results in more than a 2-fold increase in N2O flux for two days when 1000 ng·kg−1 NAR was administered. At 80% WFPS, N2O flux was moderate, ranging from 4 to 39 ng N2ON g−1 d−1. Initially, NAR stimulated N2O flux at the lowest doses (1 and 10 ng·kg−1) and had no significant effect at highest doses (100 and 1000 ng·kg−1). After 72 h, N2O flux was inhibited at all but the 1 ng·kg−1 dose. This trend agrees with the observations reported by Chen et al. (2019), whose study found that N2O flux was inhibited by triclosan and triclocarban at all but the lowest dose (10 μg·kg−1).

N2 O-N (ng g -1 d-1)

40% WFPS

10

7

Day 3

80 60 40 20 0

1

10

100

1000

Narasin Dose (ng/kg)

Fig. 6. Box-whisker plots illustrating the N2O-N flux over a three-day incubation period in moist soils (40%, 60%, and 80% WFPS) treated with NAR. 40% WFPS data from DeVries et al. (2015).

8

S.L. DeVries et al. / Science of the Total Environment 716 (2020) 137031

40% WFPS

2.0

60% WFPS

80% WFPS

Day 1

Day 1

Day 1

Day 2

Day 2

Day 2

Day 3

Day 3

Day 3

1.5 1.0

Relative Contribution to N2 O flux

0.5

1.5 1.0 0.5

2.0 1.5 1.0 0.5 0

0

1

10

100

Narasin Dose (ng kg -1)

1000

0

1

10

100

1000

0

Narasin Dose (ng kg -1)

1

10

100

1000

Narasin Dose (ng kg -1)

Fig. 7. Relative contribution of denitrification (black) and nitrifier denitrification (white) to total N2O flux.

agricultural productivity and indicates that the effects of trace exposure to NAR and other antibiotics on crop nutrition and non-point source N pollution should be further investigated. N2O is a powerful greenhouse gas and ozone-reducing molecule produced by soil microorganisms and

our results showed that exposure to 1–1000 ng kg−1 soil increased N2O flux in unsaturated soils. Long-term incubations and field-studies would help to determine whether the short-term fluxes observed in the present study are persistent enough to impact projections of terrestrial N2O flux.

4. Conclusions The primary objective of this study was to determine whether trace concentrations of NAR measurably impact major N-pools and fluxes − from soil. The results indicate that three major N pools (NH+ 4 , NO3 , and N2O) change in response to additions of 1 ng·kg−1 to 1000 ng·kg−1 NAR. Previously, only broad-spectrum antibiotics at high concentrations have been reported to inhibit nitrification, possibly because there is functional redundancy in the microbial nitrifying community. Inhibition by NAR, which is inactive against gram-negative bacteria, including most commonly referenced nitrifying bacteria, is an unanticipated outcome of this study. Since NAR has antifungal properties, our results suggest that nitrifying fungi may play a significant role in the resilience of nitrifying communities. Isotopic enrichment data showed that both aerobic and anaerobic N2O production pathways are affected by exposure to NAR, but a positive shift in the N2O:N2 ratio in relation to NAR dose appears to most strongly influence net N2O flux in unsaturated soils. This shift leads to a considerable increase in N2O flux even when total denitrification is modestly inhibited. To better understand the mechanism and importance of this shift, future studies of N2O flux from soils treated with NAR should include genetic analyses that can distinguish between fungal, archaeal, and bacterial activity. Additionally, isotopomer studies would be helpful to better constrain the pathways leading to N2O flux.

Table 2 Pearson correlation coefficient (R2) calculated between the proportion of total N2O flux attributed to denitrification and the N2O:N2 ratio. Fig. 8. Line plots of the N2O:N2 flux ratio from soils treated with 1, 10, 100, or 1000 ng·kg−1 NAR and incubated for 3 days. Results shown for 40% WFPS (top) and 60% WFPS (bottom).

40% WFPS 60% WFPS

N2O:N2 ratio N2O:N2 ratio

R2 R2

Day 1

Day 2

Day 3

0.94 0.99

0.75 0.93

0.98 0.98

S.L. DeVries et al. / Science of the Total Environment 716 (2020) 137031

Supplementary data to this article can be found online at https://doi. org/10.1016/j.scitotenv.2020.137031. CRediT authorship contribution statement Stephanie L. DeVries: Funding acquisition, Conceptualization, Methodology, Data curation, Visualization, Writing - original draft, Writing - review & editing.Karin A. Block: Writing - original draft, Visualization, Writing - review & editing.Madeline Loving: Data curation. Laura Logozzo: Data curation.Pengfei Zhang: Conceptualization, Methodology, Data curation. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements This work was partially supported by the Northeast Sustainable Agriculture Research and Education (SARE) program, administered by the University of Vermont and State Agricultural College on behalf of the National Institute of Food and Agriculture, U.S. Department of Agriculture. (Prime Award #2012-38640-19543, Subaward # GN#13-057). References Bak, S.A., Hansen, M., Pedersen, K.M., Halling-Sørensen, B., Björklund, E., 2013. Quantification of four ionophores in soil, sediment and manure using pressurised liquid extraction. J. Chromatogr. A 1307, 27–33. Banerjee, S., D’Angelo, E., 2013. Livestock antibiotic effects on nitrification, denitrification, and microbial community composition in soils. Open Journal of Soil Science 3, 203. Barraclough, D., 1995. 15N Isotope Dilution Techniques to Study Soil Nitrogen Transformations and Plant Uptake. Nitrogen Economy in Tropical Soils. Springer, Dordrecht, Netherlands, pp. 185–192. Belser, L.W., 1979. Population ecology of nitrifying bacteria. Annual Reviews in Microbiology 33, 309–333. Berg, D.H., Hamill, R.L., 1978. The isolation and characterization of narasin, a new polyether antibiotic. The Journal of Antibiotics 31, 1–6. Calabrese, E.J., Baldwin, L.A., 2001. Hormesis: U-shaped dose responses and their centrality in toxicology. Trends Pharmacol. Sci. 22, 285–291. Camargo, J.A., Alonso, A., Salamanca, A., 2005. Nitrate toxicity to aquatic animals: a review with new data for freshwater invertebrates. Chemosphere 58, 1255–1267. Camberato, J., 2016. Land Application of Poultry Manure. Confined Animal Manures Managers (CAMM): Poultry Training Manual. Clemson Cooperative Extension, Clemson, SC, pp. 1–12. Campos, L.J., Garrido, J., Méndez, R., Lema, J., 2001. Effect of two broad-spectrum antibiotics on activity and stability of continuous nitrifying system. Appl. Biochem. Biotechnol. 95, 1–10. Chen, S., Chee-Sanford, J.C., Yang, W.H., Sanford, R.A., Chen, J., Yan, X., et al., 2019. Effects of triclosan and triclocarban on denitrification and N2O emissions in paddy soil. Sci. Total Environ. 695, 133782. D’Alessio, M., Durso, L.M., Miller, D.N., Woodbury, B., Ray, C., Snow, D.D., 2019. Environmental fate and microbial effects of monensin, lincomycin, and sulfamethazine residues in soil. Environ. Pollut. 246, 60–68. Davidson, E.A., 2009. The contribution of manure and fertilizer nitrogen to atmospheric nitrous oxide since 1860. Nat. Geosci. 2, 659. DeVries, S.L., Zhang, P., 2016. Antibiotics and the terrestrial nitrogen cycle: a review. Current Pollution Reports 2, 51–67. DeVries, S.L., Loving, M., Li, X., Zhang, P., 2015. The effect of ultralow-dose antibiotics exposure on soil nitrate and N2O flux. Sci. Rep. 5, 16818.

9

Elanco Animal Health EL, 2017. Environmental Assessment for the Use of MaxibanTM (Narasin and Nicarbazin) in Feed for Prevention of Coccidiosis in Broiler Chickens Greenfield, IN. Garzon, C.D., Flores, F.J., 2013. Hormesis: biphasic dose-responses to fungicides in plant pathogens and their potential threat to agriculture. Agricultural and Biological Sciences. Fungicides–Showcases of Integrated Plant Disease Management from. Around the World 12, 311–328. Grenni, P., Ancona, V., Caracciolo, A.B., 2018. Ecological effects of antibiotics on natural ecosystems: a review. Microchem. J. 136, 25–39. Hauck, R.D., Melsted, S.W., Yankwich, P.E., 1958. Use of N-isotope distribution in nitrogen gas in the study of denitrification. Soil Sci. 86, 287. Hautier, Y., Niklaus, P.A., Hector, A., 2009. Competition for light causes plant biodiversity loss after eutrophication. Science 324, 636–638. Herrero, P., Borrull, F., Marcé, R.M., Pocurull, E., 2013. Determination of polyether ionophores in urban sewage sludge by pressurised liquid extraction and liquid chromatography–tandem mass spectrometry: study of different clean-up strategies. J. Chromatogr. A 1285, 31–39. Holmes, R., McClelland, J., Sigman, D., Fry, B., Peterson, B., 1998. Measuring 15N–NH 4+ in marine, estuarine and fresh waters: an adaptation of the ammonia diffusion method for samples with low ammonium concentrations. Mar. Chem. 60, 235–243. Hussain, S.A., Prasher, S.O., 2011. Understanding the sorption of ionophoric pharmaceuticals in a treatment wetland. Wetlands 31, 563–571. Kirkham, D., Bartholomew, W.V., 1954. Equations for following nutrient transformations in soil, utilizing tracer data. Soil Science Society Proceedings 18, 33–34. Konopka, M., Henry, H.A., Marti, R., Topp, E., 2015. Multi-year and short-term responses of soil ammonia-oxidizing prokaryotes to zinc bacitracin, monensin, and ivermectin, singly or in combination. Environ. Toxicol. Chem. 34, 618–625. Kotzerke, A., Sharma, S., Schauss, K., Heuer, H., Thiele-Bruhn, S., Smalla, K., et al., 2008. Alterations in soil microbial activity and N-transformation processes due to sulfadiazine loads in pig-manure. Environ. Pollut. 153, 315–322. Ma, C., Jensen, M.M., Smets, B.F., Thamdrup, B., 2017. Pathways and controls of N2O production in nitritation–anammox biomass. Environmental Science & Technology 51, 8981–8991. Rabalais, N.N., Turner, R.E., Díaz, R.J., Justić, D., 2009. Global change and eutrophication of coastal waters. ICES Journal of Marine Science: Journal du Conseil 66, 1528–1537. Ravishankara, A.R., Daniel, J.S., Portmann, R.W., 2009. Nitrous oxide (N2O): the dominant ozone-depleting substance emitted in the 21st century. Science 326, 123–125. Ritz, C.W., Merka, W.C., 2009. Maximizing poultry manure use through nutrient management planning. Bulletin 1245. Roose-Amsaleg, C., Laverman, A., 2016. Do antibiotics have environmental side-effects? Impact of synthetic antibiotics on biogeochemical processes. Environ. Sci. Pollut. Res. 23, 4000–4012. Selander, E., Cervin, G., Pavia, H., 2008. Effects of nitrate and phosphate on grazer-induced toxin production in Alexandrium minutum. Limnol. Oceanogr. 53, 523–530. Semedo, M., Song, B., Sparrer, T., Phillips, R.L., 2018. Antibiotic effects on microbial communities responsible for denitrification and N2O production in grassland soils. Front. Microbiol. 9, 2121. Shan, J., Yang, P., Rahman, M.M., Shang, X., Yan, X., 2018. Tetracycline and sulfamethazine alter dissimilatory nitrate reduction processes and increase N2O release in rice fields. Environ. Pollut. 242, 788–796. Siegel, R., Hauck, R., Kurtz, L., 1982. Determination of 30N2 and application to measurement of N2 evolution during denitrification. Soil Sci. Soc. Am. J. 46, 68–74. Sigman, D., Altabet, M., Michener, R., McCorkle, D., Fry, B., Holmes, R., 1997. Natural abundance-level measurement of the nitrogen isotopic composition of oceanic nitrate: an adaptation of the ammonia diffusion method. Mar. Chem. 57, 227–242. Toth, J.D., Feng, Y., Dou, Z., 2011. Veterinary antibiotics at environmentally relevant concentrations inhibit soil iron reduction and nitrification. Soil Biol. Biochem. 43, 2470–2472. USDA, 2019. Poultry-Production and Value 2018 Summary. United States Department of Agriculture. Wainwright, M., Pugh, G., 1973. The effect of three fungicides on nitrification and ammonification in soil. Soil Biol. Biochem. 5, 577–584. Ward, M.H., deKok, T.M., Levallois, P., Brender, J., Gulis, G., Nolan, B.T., et al., 2005. Workgroup report: drinking-water nitrate and health–recent findings and research needs. Environ. Health Perspect. 113, 1607–1614. Zhu, X., Burger, M., Doane, T.A., Horwath, W.R., 2013. Ammonia oxidation pathways and nitrifier denitrification are significant sources of N2O and NO under low oxygen availability. Proc. Natl. Acad. Sci. 110, 6328–6333.