Accumulation of Tetracycline Resistance Genes in ... - ACS Publications

Sep 18, 2009 - Antibiotic resistance genes (ARGs) are emerging contaminants found in the water and sediments surrounding animal feedlots. In this stud...
13 downloads 0 Views 951KB Size
Download PDF
Environ. Sci. Technol. 2009, 43, 7643–7650

Accumulation of Tetracycline Resistance Genes in Aquatic Biofilms Due to Periodic Waste Loadings from Swine Lagoons W E N Z H A N G , † B E L I N D A S . M . S T U R M , * ,† C H A R L E S W . K N A P P , ‡,§ A N D D A V I D W . G R A H A M †,‡ Department of Civil, Environmental, and Architectural Engineering, University of Kansas, Lawrence, KS, USA 66045, and School of Civil Engineering and Geosciences, Newcastle University, Newcastle Upon Tyne, UK NE1 7RU

Received May 15, 2009. Revised manuscript received August 25, 2009. Accepted September 06, 2009.

Antibiotic resistance genes (ARGs) are emerging contaminants found in the water and sediments surrounding animal feedlots. In this study, the fate of five tetracycline-resistance and 16S-rRNA genes released in swine waste were monitored for 21 days in the water column and biofilms in 12 mesocosms mimicking different natural receiving water bodies. Four treatments were employed in triplicate: two light exposures (light/ dark) and two loading scenarios (single/periodic). As seen previously,lightexposurehadasignificanteffectondisappearance rates of tet genes in both the water column and biofilms, although absolute rates were significantly lower in the biofilms. Further, periodic versus single loading events resulted in >2 orders of magnitude higher tet gene levels in associated tanks. Regardless of treatment, ARGs migrated quickly to biofilms, with 3% and >85% of detected tet determinants found in biofilms on days 1 and 4, respectively. Overall, these are the first quantitative data on specific ARG disappearance rates in biofilms, and also the first evidence of progressively accumulating ARG levels in biofilms under loading conditions typical of natural receiving waters. In summary, ARGs migrate rapidly to biofilms where they persist longer than adjacent waters, which suggests biofilms likely act as reservoirs for ARGs in nature.

Introduction Antibiotic resistance genes (ARGs) have been identified as a class of emerging contaminants (1). ARGs primarily enter receiving waters via fecal matter within bacterial hosts that have been previously exposed to elevated antibiotic levels, usually in higher organisms (2-4), which is a common practice in animal husbandry and agriculture. Whereas antibiotics themselves degrade fairly quickly in the environment (5), elevated ARG levels have been often observed in sediments downstream of intensive agricultural areas (6). Despite these observations, disappearance kinetics and fate * Corresponding author phone: 785-864-1739; fax: 785-864-5379; e-mail: [email protected]. † University of Kansas. ‡ Newcastle University. § Currently with David Livingstone Centre for Sustainability, Department of Civil Engineering, University of Strathclyde, Glasgow, UK G1 1XN. 10.1021/es9014508 CCC: $40.75

Published on Web 09/18/2009

 2009 American Chemical Society

mechanisms of ARGs are relatively unknown in aquatic systems. For example, the rate of ARG disappearance only has been quantified for genes present in the water column after a single pulse addition of waste, which revealed that selected tetracycline-resistance genes disappear rapidly under elevated light exposures compared with dark conditions (7, 8). These studies illustrate the importance of environmental conditions on the ultimate fate and persistence of ARGs in the environment. However, neither study addressed the persistence and/or migration of ARGs associated with periodic wastes releases that better mimic discharge patterns from agricultural lagoons to the environment. The purpose of this study was to quantitatively determine disappearance and migration rates of ARGs in aquatic and sediment environments using field mesocosms. Different environmental factors were addressed, including light penetration (shaded versus dark) and loading rate (single addition of lagoon waste versus periodic addition), each in triplicate. Swine lagoon wastewater was used as the waste source, collected from an oxytetracycline-fed piglet facility, and provided to each mesocosm. Five tetracycline resistance determinants were examined, including tet(L), tet(M), tet(O), tet(Q), and tet(W), which are often found in swine (9) and cattle lagoon wastes (10). These five determinants were biomarkers for tetracycline resistance and were compared to 16S-rRNA gene abundances in the mesocosms. They were monitored using quantitative PCR over a 21 day exposure period in the water column and biofilms of parallel mesocosms.

Materials and Methods Experimental Design. The experimental design was a 2 × 2 factorial design with three mesocosms per treatment; factors represented light conditions (light/dark) and pattern of waste addition (single/periodic). Twelve cylindrical fiberglass tanks (1.5 m depth, 3.2 m diameter) were prepared similarly to previous experiments (8, 11) at the Nelson Environmental Studies Area (NESA) near Lawrence, Kansas. “Light” treatments were covered by shade cloth lids (International Greenhouse Company, Danville, IL) to reduce sunlight by 90% to establish an environment representative of a natural receiving pond (unshaded fiberglass tanks reflected too much light). “Dark” mesocosms were covered with black vinyl plastic lids that blocked 100% light to represent deeper waters nearer sediments. Light intensity measurements were taken at depths 0, 0.5, and 1.0 m on five days at midday (LI-COR quantum sensor; Lincoln, NE). Mesocosm Preparation. Twelve mesocosms were filled with 11.3 m3 water from an aquifer-fed reservoir located at NESA to initiate the experiment. Before filling, three 15.5 × 30.5 cm plastic trays containing sediments collected from the reservoir were placed on the bottom of each tank to provide microbial inoculum to the units. The surrounding ponds were filled with reservoir water at the same time, and the mesocosms were allowed to acclimate for four weeks before the experiment commenced. Three weeks before the experiment, small plexiglass plates containing 24 circular disks of 600-grit, waterproof sandpaper (2.4 cm diameter; GatorGrit, Finland) were suspended by fishing line near the bottom of all the tanks to provide retrievable surfaces on which biofilms could develop. Shade lids were then placed on the tanks one week before the experiment began. Lagoon wastewater was collected from an oxytetracyclinefed piglet containment area. The experiment was initiated by adding 20 L of waste slurry to each tank, whereas for the pulse-addition tanks, 10 L of slurry was added (each) on VOL. 43, NO. 20, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7643

days 3, 6, 9, 12, 15, and 18. Fresh lagoon waste for these days was collected on days 0, 2, 5, 10, and 14 and stored at 4 °C prior to use; waste was never stored >4 days. Waste was added at dusk and allowed 12 h to equilibrate before sampling at 8 am the next morning. Waste addition rates were determined based on quantification of gene abundances in the waste and previous experience of the required concentration to detect differences among treatments (8). The minimum volume of waste was added to minimize excessive nutrient inputs, but to allow qPCR detection. Sample Collection and Processing. Water Chemistry. Samples for both water chemistry and molecular biological analyses were collected on days 0, 1, 2, 4, 7, 10, 16, and 21. In situ dissolved oxygen (DO), pH, turbidity, temperature, and conductivity were measured at three depths (0.2, 0.7 and 1.2 m) using a field meter (model U-10, Horiba Instruments). Duplicate depth-integrated water column samples were collected using 1.2 m length, 25 mm diameter PVC samplers. Samples were stored in two sterile 1000 mL acid-washed, amber glass bottles equipped with Teflon-lined caps, and temporarily placed on ice. A different PVC sampler was used for each treatment, and the PVC samplers and sample bottles were rinsed three times with tank water before sampling. At the laboratory, 100 mL of each sample was removed immediately for total organic carbon (TOC), total nitrogen (TN), and total phosphorus (TP) analysis; 250 mL of each sample was separated and filtered through prerinsed Whatman GF/F glass-fiber filters. The filtrate was divided for duplicate dissolved organic carbon (DOC) analysis, and the filter was retained for chlorophyll a analysis. Molecular Biology. Triplicate 500 mL samples were filtered through 47 mm diameter, presterilized 0.20 µm Nalgene filters (NNI, Rochester, NY). The filters were placed in sterile centrifuge tubes using sterile forceps and stored at -20 °C. Biofilm samples were collected in triplicate from each tank by slowly drawing the plexiglass plates to the surface and transferring the disks from the plates into sterile centrifuge tubes using ethanol-flamed sterile forceps. The samples were temporarily stored on ice, and then at -20 °C. Analytical Methods. Chemical Analysis. TP and TN were analyzed spectrophotometrically (Shimadzu, Columbia, MD) following standard methods 4500-nitrogen and 4500phosphorus (12). Spectrophotometric analysis of chlorophyll a was performed after extraction with hot ethanol (13). DOC and TOC were determined using a Dohrmann organic carbon analyzer with potassium hydrogen phthalate (KHP) standards ranging from 0 to 20 ppm. Soluble oxytetracycline was measured by LC-MS-MS as reported previously (14). Sample Processing and Quantitative PCR. DNA were extracted from filters and biofilm disks using MoBio UltraClean Soil DNA kits (Carlsbad, CA). The filters, beads, and extraction buffer were combined and agitated using a FastPrep bead beater (QBiogene, Irvine, CA) for 30 s (5 speed), incubated at 70 °C for 10 min to enhance the lysis of Grampositive bacteria, and reagitated in the bead beater for 30 s (5 speed). The remaining purification steps followed the manufacturer’s protocols, with final DNA elution in 200 µL of DNase-free water. Extracts were stored at -80 °C prior to gene quantification. qPCR was used to quantify five tetracycline resistance (tet) genes, including tet(L), tet(M), tet(O), tet(Q), and tet(W) because they are the five most common tetracycline resistance genes in current databases, being found in over 90% of identified resistant species (15, 16). 16S-rRNA were also quantified. The TaqMan probe and primer sequences, reaction conditions, and PCR programs have been reported previously for tet(O), tet(Q), and tet(W) (10); tet(L), tet(M) (14); and16S-rRNA assays (17). The PCR reaction mixture included 5 µL of template DNA, primers/TaqMan probe, and 1× BioRad iQ Supermix (BioRad, Hercules, CA) in a 20 µL 7644

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 20, 2009

reaction. The 16S rRNA assay, however, used 1× LD Taq Mastermix (Applied Biosystems, Foster City, CA). To determine and control for the presence of inhibitory substances, a subset of samples were screened by performing PCR for each gene at six different dilutions. All subsequent samples were run at the lowest dilution without inhibition (typically a 1:10 dilution). Reactions were performed using a BioRad iCycler with an iCycler iQ fluorescence detector. Gene concentrations were quantified in duplicate for each assay using calibration curves created by 10-fold serial dilution of plasmid standards (3 × 101 to 3 × 106 genes/reaction); a calibration curve was measured in each plate to quantify samples included on that plate. The quantitative detection limit was calculated separately as the lowest standard concentration detectable within a linear standard curve. Construction of Plasmid Standards for qPCR. Plasmid standards for tet genes were obtained by groups as previously reported (9, 18-20). A 16S-rRNA PCR product from E. coli was cloned into the 2.1-TOPO vector using TOPO TA cloning kit (Invitrogen, Carlsbad, CA). Plasmid extraction from clones was performed using QIAprep Spin Miniprep Kit (Qiagen, Valencia, CA). Plasmid DNA concentrations were measured using a NanoDrop spectrophotometer ND-1000 (Thermal Scientific, San Diego, CA). Statistical Analysis. Mean physical and chemical water conditions for each treatment were based on triplicate (physical) and duplicate (chemical) samples taken for each tank, with three tanks per treatment. Confidence intervals (CI) were estimated from standard deviations around each mean. Mean tet and 16S-rRNA gene abundances were calculated similarly. The disappearance rate coefficient (kd) for individual genes was estimated using regression analysis with a first-order model. Means and standard errors of kd were calculated using a ln-transformed linear regression of all tanks within one treatment. Given that most gene disappearances were biphasic, two kd estimates were made for each treatment. The Kruskal-Wallis (KW) test was used to compare physical and chemical measurements among data points taken from three depths within each tank, and among treatments over time. ANCOVA, with time as a covariant, was used to assess whether gene abundances differed significantly among treatments over time. All analyses were performed using Minitab and SPSS.

Results Determination of Background Tet Genes and Waste Addition Rates. Prior to the addition of lagoon waste on Day 0, tet and 16S-rRNA gene abundances were determined for the reservoir water used to fill all mesocosms, in the acclimated mesocosms after inoculation with sediment trays, and in the lagoon waste on each day of addition. Table 1 shows that water from the reservoir averaged 3.09 × 106 16SrRNA genes/ml, whereas only tet(L), tet(O), and tet(W) genes were detectable, averaging 4.77 × 101, 2.66 × 102, and 2.46 × 102 genes/mL, respectively. After a four-week acclimation, all tet genes were detected (except tet(L) in the water column) with tet(M) being highest in the water column, averaging 1.01 × 102 genes/mL across the tanks and tet(Q) being highest in the biofilms, averaging 1.17 × 106 genes/cm2. On day 0, before the addition of wastes, only water column levels of tet(Q) and tet(W) differed significantly (p < 0.05) between the dark and light tanks, but when these values were normalized to ambient 16S-rRNA-gene levels, no significant difference was seen among treatments. Therefore, each tank had functionally the same distribution of ARG per bacteria, and the initial conditions for all tanks were similar. The actual concentration of genes varied in the lagoon wastes from day-to-day, but total tet averaged 1.43 × 107 ( 1.23 × 107 genes/mL lagoon waste for all addition events. On day 0, for the only addition to the single-addition units, the

TABLE 1. the Mean Concentration of Measured Tetracycline-Resistance and 16s-Rrna Genes in the Reservoir Water (Used to Fill the Mesocosms) and in the Mesocosm (Water Column and Biofilms) on Day 0 of the Experiment, After Acclimation to Natural Conditions and before the Addition of Lagoon Waste. the Standard Deviation of the Mean Is Presented in Parentheses sample reservoir water (genes/ml) day 0 water columnc (genes/ml) day 0 biofilm (genes/cm2)

tet(L)

tet(M)

tet(O)

tet(Q)

tet(W)

not-detected 2.46 × 102

16S-rRNA gene

5.59 × 102

normalized total tetb 1.81 × 10-4

4.77 × 101

not-detected 2.66 × 102

not-detected

3.42 × 101 (1.10 × 102)

5.11 × 10-1 1.91 × 101 (5.68 × 10-1) (2.91 × 101)

4.35 (6.15)

7.77 × 102 (2.64 × 103)

3.21 × 105 (1.04 × 106)

3.97 × 104 (9.54 × 104)

7.17 × 103 1.65 × 108 1.54 × 106 8.49 × 10-3 (6.31 × 103) (1.25 × 108) (5.13 × 106) (2.73 × 10-2)

1.17 × 106 (3.99 × 106)

3.09 × 106

total teta

8.70 × 106 5.84 × 101 3.64 × 10-2 (5.87 × 106) (1.08 × 102) (1.26 × 10-1)

a

“Total tet” refers to the sum of all tet-determinant concentrations that were measured. b “Normalized total tet” values refer to the “total tet” concentration normalized to 16S-rRNA gene abundance. c In order to average a data set containing detected and nondetected samples by qPCR, the concentration of a “non-detect” was set equal to half the quantitative detection limit. The detection limits ranged from 0.4 to 1.6 genes/mL in water-column samples and 3.54 to 14.16 genes / cm2 in biofilm samples.

distribution of gene in the lagoon waste was: not detectable tet(L); 1.24 × 107 tet(M); 3.41 × 106 tet(O); 6.83 × 105 tet(Q); 3.26 × 106 tet(W); and 6.00 × 108 16S-rRNA genes/mL waste. The concentration and distribution of tet genes on subsequent days of lagoon waste addition is provided in Supporting Information (SI) Figure SI1. It should be noted that measured total tet concentrations were greater than expected on day 1, which was due to variations in the lagoon waste, which was not homogenized before addition, or in growth of organisms overnight between waste addition and sampling. Physical, Chemical, and Biological Conditions as a Function of Treatment. Chlortetracycline, a derivative of oxytetracycline, was detected in the lagoon waste at a concentration of 10 µg/L. Oxytetracycline was not detected in any mesocosm or waste sample from day 1 to 21 (detection limit of 1 µg/L). Other measurements were taken at three different depths. Dark units functionally had zero light penetration, whereas the Light units received 4, 3, and 2% of atmospheric light intensity (3417.75 µE/m2/s) at 0, 0.5, and 1.0 m depths, respectively. Overall, water temperature, pH, conductivity, turbidity, and DO did not vary with depth; therefore, the water column was treated as a completely mixed environment and measures were averaged over depth (SI Table SI1). Significant differences did exist between treatments during the experiment. For example, turbidity differed significantly between the single and periodic treatments (KW; p e 0.002) and conductivity differed between light-single and dark-single treatments (p ) 0.019). Significant differences (KW; p < 0.05) also were observed for TKN, TP, TOC, and DOC measurements between single and periodicaddition units, presumably due to different amounts of waste added. Finally, chlorophyll a levels were significantly lower in dark vs light treatments (p < 0.01) due to reduced photosynthesis. Although DO did not vary with depth in each tank, DO did vary with time for each treatment. The DO level in all mesocosms dropped from day 0 to day 4, but remained relatively constant after day 10. Tanks with repeated waste addition became anoxic after day 10, whereas single-addition tanks remained aerobic (>2 mg/L DO) throughout the experiment. DO trends over time were not significantly different between any of the treatments (KW; p > 0.05). Distribution of tet Genes in the Water Column and Biofilm. On Day 1, the average total tet gene abundance among the twelve tanks was 1.15 × 107 ( 7.19 × 106 genes/ mL water. In the single addition tanks, all ARGs in the water column decreased significantly over the 21 day experiment, whereas ARGs in the water columns with pulsed additions were relatively constant over time throughout the same period (around 6.0 × 106 genes/mL). In the biofilms, total tet levels

were 1.67 × 107 ( 6.87 × 106 genes/cm2 on day 1. In singleaddition units, tet abundance typically peaked at 9.97 × 107 ( 1.86 × 106 genes/cm2 on day 4 and then diminished. When wastes were added every three days, levels of tet genes in water columns and biofilms were 2-3 orders of magnitude higher than in single-addition tanks at the end of the experiment. Figures 1 and 2 present mean abundances of individual determinants in the Light treatments throughout the experiment; SI Figures SI2 and SI3 show that similar patterns existed in the dark treatments, although absolute concentrations differed. tet(Q) was the dominant detected ARG in the water column among treatments, and tet(M) was the dominant ARG in biofilms in all treatments except the dark-periodic units. Migration of Tet Genes from the Water Column to Peripheral Biofilms. Upon addition of waste, the total concentration of tet genes increased from 1.54 × 106 genes/ cm2 (day 0) to over 1 × 107 genes/cm2 in the biofilms in all tanks, indicating an immediate influx of tet genes to the biofilms (Figure 1). In mesocosms receiving a single waste input, tet levels continued to increase in the biofilms during the first 4 days. Using a mass-balance approach to tet distribution in the tanks, total percentages of ARGs in the water column versus biofilm were calculated by extrapolating the gene concentrations in the samples to the entire tank volume and surface area (the product of genes per cm2 disk and the surface area of the tank walls and base). On day 1, only 3-4% of total tet genes were in the biofilms, but by day 4, >85% of all tet genes in the single-addition tanks were biofilm-associated. For the mesososms receiving multiple waste loadings, the concentration of tet genes increased in the biofilms over time, as shown in Figure 2. In contrast, tet gene levels remained relatively constant in the water column over time, suggesting relatively balanced gene inputs (from waste addition) and losses (due to gene decay and migration out of the water column). At the end of the experiment in the biofilms, 1.97 × 108 ( 1.38 × 108 gene/cm2 of total tet (the sum of detected tet determinants) were detected in lightpulse addition tanks, and 2.78 × 108 ( 8.39 × 107 genes/cm2 of total tet were detected in dark-pulse addition tanks. Disappearance Rate Coefficients as a Function of Treatment. Disappearance rate coefficients (kd) were calculated for individual tet determinants in each tank (Table 2), total tet genes combined, and 16S-rRNA gene levels (Figure 3). Since tet disappearance was not observed in pulse-addition tanks, rates were not calculated for those units. During the first seven days, all genes disappeared at a faster rate from the water column than between days 7 and 21; the initial VOL. 43, NO. 20, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7645

FIGURE 1. Mean distribution of detected tetracycline-resistance genes over time in the water column and peripheral biofilms of the light-single treatment. Percentages refer to the proportion of tet genes in the biofilm relative to all tet genes in the mesocosm on each sample day. increased rate of disappearance coincided with an increase in ARG concentrations in the biofilms during the first four days, followed by disappearance during days 7-21 (Figure 3). kd values from biofilm samples were calculated using a first-order model on data from days 7 to 21 only, and only monophasic rates are reported. There were minimal changes in the 16S-rRNA gene levels over the 21-day experiment with kd ranging from -0.03 d-1 to +0.02 d-1 for light and dark single-addition treatments, rates that were not significantly different from 0. In contrast, tet-disappearance rates in the light treatment were significantly greater than those of the dark treatment (p < 0.05). Average tet disappearance rates for light-single treatments were -0.19 d-1 and -0.11 d-1 in the water columns and biofilms, respectively; compared to the dark-single averages of -0.17 d-1 and -0.08 d-1. This data confirm that light exposure is a significant factor affecting the disappearance of tet genes, but rates differ in the water column and biofilms.

Discussion Several studies have shown ARGs to accumulate in the vicinity of agricultural lagoon operations, including groundwater, soil, and river sediments (1, 10, 21, 22). Lagoon waste from animal feedlots is either released directly into a receiving water body, or applied on land as fertilizer (and subject to runoff). The frequency and loading rate of such wastes upon 7646

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 20, 2009

impacted water bodies varies greatly, but release is typically semiperiodic rather than a single event (23). A few studies have provided quantitative data on the fate of ARGs in aquatic systems and peripheral compartments (sediments and biofilms) (7, 8); however, previous work only focused on fatekinetics in the water column. The current study quantifies ARG fate within the aquatic system (water column and biofilm) under differing receiving water conditions, providing data which are essential for engineers and managers to develop best management practices for the storage and release of lagoon wastes with minimum impact on the environment. The current experiment confirms results from Engemann et al. (8), which showed that ambient light conditions significantly affected the disappearance rate of tet genes after a single release of feedlot waste to a standing body of water. That study showed the migration of ARGs from the water column to peripheral biofilms, although the disappearance rates in the biofilms were not quantified due to a shorter experimental duration and a lack of replicates. Here we show that tet disappearance rates in biofilms also are significantly greater in light-exposed units. Interestingly, although disappearance patterns observed here are functionally the same as previous work relative to light and dark effects, our absolute rate coefficients are 3-5 times lower than previous results (see Table 2). Figure 3 shows that tet gene disappearance

FIGURE 2. Mean distribution of detected tetracycline-resistance genes over time in the water column and peripheral biofilms of the light-periodic treatment. Percentages refer to the proportion of tet genes in the biofilm relative to all tet genes in the mesocosm on each sample day.

TABLE 2. Disappearance-Rate Coefficients (kd in d-1) for Tetracycline-Resistance and 16S-rRNA Genes under the Two Field Treatments Compared with Equivalent Rate Coefficients from a Previous Mesocosm Experiment (8) current research gene tet(M)

treatment

sample

D1 to D7

D7 to D21

D1 to D21

D1 to D14

water column biofilm water column biofilm water column biofilm water column biofilm water column biofilm water column biofilm water column biofilm water column biofilm water column biofilm water column

-0.704(0.100)

-0.356(0.057) -0.215(0.070) -0.246(0.037) -0.125(0.045) -0.166(0.079) -0.167(0.029) -0.109(0.016) -0.148(0.063) -0.079(0.096) -0.316(0.134) 0.057 (0.106) -0.217(0.106) -0.321(0.042) -0.307(0.097) -0.212(0.023) -0.251(0.056) -0.275(0.065) -0.168(0.103) -0.227(0.024)

-0.460(0.031)

-0.317(0.038)

-0.739(0.07) N/Ab -0.532(0.06) N/A -0.405(0.059)

-0.103(0.374)

-0.333(0.062)

-0.506 (0.016)

-0.792(0.116)

-0.483(0.087)

-0.575(0.101)

-0.483(0.044)

-0.761(0.091)

-0.386(0.044)

-0.354(0.098)

-0.468(0.039)

-0.71(0.07)

-0.416(0.015)

-0.47(0.02)

light-single dark-single

tet(Q)

light-single dark-single

tet(W)

light-single dark-single

Total tet

Engemann (2008)

light-single dark-single

tet(O)

a

light-single dark-single biofilm

-0.804(0.121) -0.712 (0.165) -0.882(0.098) -1.651 (0.129) -2.029(0.123) -0.849 (0.044) -0.978(0.054) -0.983 (0.036) -0.940(0.107) -0.162(0.057)

-0.402(0.016)

a Biphasic disappearance rates were calculated for days 1-7 and 7-21, respectively, and are compared to disappearance rates when all data are averaged (days 1-21) b N/A, Not available

VOL. 43, NO. 20, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7647

FIGURE 3. Water column tet (O) and 16S-rRNA (b) gene abundances over time for the mesocosms receiving a single loading of lagoon waste. “Total tet” is the sum of all detectable tetracycline-resistance determinants. Error bars indicate the standard deviation of triplicate mesocosms for each treatment. Trendlines indicate estimated disappearance curves assuming a first-order model. Both monophasic (shown in A and B) and biphasic (shown in C and D) trendlines are presented. was most rapid in the first 7 days and declined thereafter. When only the first 7 days are used in the calculation, the rate coefficients from this experiment are similar to previous data. A number of explanations are possible for the differences in observed rates, including the longer duration of this experiment than previous studies (21 days vs 14 days), undefined differences between cattle and swine wastes, and differences in tetracycline levels in the waste. The digestive systems, and thus composition of intestinal flora, of cattle and swine are quite different. Moreover, there is a wide range of agricultural practices within and between cattle and swine feedlots, which leads to a great variance in the reported prevalence of antibiotic resistant bacteria from a given waste source (23). Regarding the concentration of tetracycline, in the Engemann et al. (2008) study, only one light exposed treatment had measurable levels of oxytetraycline (OTC) at 240 µg/L; all other mesocosms had less than the detection limit of 0.5 µg/L. In that study, the presence of OTC had no statistically significant (p < 0.05) impact on kd. In the current study, all mesocosms had less than the detection limit of 1 µg/L of OTC throughout the experimental period, including mesocosms receiving periodic waste additions. OTC is known to disappear rapidly from the water phase due to cation exchange, cation bridging at clay surfaces, surface complexation, hydrogen bonding, and extreme photosensitivity (23, 24). Given the nondetectable levels of OTC in this and the previous study, it is unlikely that this caused differences in kd. DO levels were lower here than previous studies, which could have resulted in lower reaction rates. Specifically, tanks with periodic addition of wastes became anaerobic after Day 10 of the experimental period (see SI Table 1); however, since both the light-periodic and dark-periodic treatments had similar DO profiles, they can be directly compared. The prevalence of antibiotic resistance has steadily increased 7648

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 20, 2009

among anaerobic bacteria since the 1970s, with the most frequently isolated anaerobe being Bacteroides fragilis, a fecal bacterium (25). All known conjugative transposons in Bacteroides include resistance to tetracycline, with ∼70% of organisms in the B. fragilis group harboring these transposons. In the current experiment, anaerobic conditions may have influenced the growth of tetracycline-resistant anaerobes, or the transfer for resistance between organisms, although qPCR cannot discern specific resistant populations. Beyond the association of antibiotic resistance with anaerobic bacteria, there have not been any studies on the effect of DO directly on the disappearance or transfer rate of tetracyclineresistance genes. This may be an avenue of future study. Although ARG disappearance patterns were the same here versus previous work, actual rate coefficients between the water column and biofilms differed. For example, for both the light-single and dark-single treatments, the rate coefficient of each tet determinant was significantly greater in the water column versus the adjacent biofilm. There are two possible explanations for differences between observed biofilm and water column disappearance rates. First, lower ARG disappearance rates in biofilms might simply be a result of reduced light exposure. The typical light intensity at 1.2 m depth where the biofilms were placed was 23.8 umol photons/ m2/s, whereas light intensities at the surface averaged 147.67 umol photons/m2/s. As such, if ARG fate is light-driven, whether it is photolytic (unlikely) versus biological, greater primary production would occur in the water column versus biofilms. Alternately, biofilms are a network of cells and extracellular polymeric substances (EPS), including any substances of biological origin (26). EPS aid in keeping bacteria together in biofilms, cause the adherence of biofilms to surfaces, and protect bacteria against adverse environmental impacts, including predation (27). Therefore, the complex three-dimensional structure of cells, EPS, and

channels will dictate the transfer of materials to and from the liquid phase and the biofilm (28). Once organisms or genes migrate from the bulk liquid phase to the solid biofilm phase, the physical mechanisms affecting transport and gene disappearance will be unique to the biofilm phase, which may explain lower observed disappearance rates in biofilms. In this study, only mesoscale observations were made, but these are influences by the microscale structure of the biofilm (29). In tanks with periodic addition of wastes, ARG concentrations increased in the biofilms over the experimental period. Given that ARG waste inputs to any receiving water likely occur in pulses, this result might explain why one sees elevated ARG levels in peripheral compartments near waste outfalls, even where water column ARG levels are low. In theory, when ARGs in liquid wastes or runoff enter the water column, they either decay quickly or move to adjacent sediments and/or biofilms. However, once the ARG enters the adjacent compartment, it appears to disappear much more slowly, and if the rate of ARG supply to the compartment is adequately high, pseudosteady elevated basal levels of ARG will develop, which is what has been seen in previous field monitoring (1). Although this study focused on selected tetracycline resistance genes, future studies should focus on differences in persistence and risk associated with specific ARGs linked to pathogens. However, little information exists on this at this time because methods have not been developed that link resistance genes to pathogenicity in complex environments like receiving waters or sediments. Collectively, ARG accumulation in the environment can be minimized in the water column by maximizing light penetration to the system. However, when ARGs are released to quiescent water bodies on a semicontinuous basis, the sediments accumulate ARGs regardless of light conditions. In total, our results suggest that ARG accumulation in biofilms harbor such genes in the environment, although much more work is needed to link such accumulation to organisms of public health significance or mechanisms of ARG resilience in biofilms. These are both key questions for minimizing the transmission of ARGs away from sites of agricultural use, such as feedlots, which are significant for retaining effective antibiotics into the future.

Acknowledgments This research was funded in part by the University of Kansas, Transportation Research Institute from Grant No. DT0S5906-G-00047, provided by the U.S. Department of Transportation, Research and Innovative Technology Administration, and E.U. Marie Curie Excellence Programme (MEXT-CT2006-023469). We thank Mr. Scott Campbell (Kansas Biological Survey) and Mr. Jay Barnard (University of Kansas), who assisted in preparation of the mesocosms. Drs. Tim LaPara and Sudeshna Ghosh (Minnesota) and Dr. Trina McMahon provided the tetracycline-resistance genes for qPCR standards. Dr. Michael T. Meyer at the U.S. Geological Survey, Lawrence, KS, measured the tetracycline concentration in samples. Finally, Mr. Roger Flory supplied the lagoon waste for this experiment.

Supporting Information Available Additional information on the concentration and distribution of tet-resistant genes in the lagoon waste (Figure SI1), distribution of tet-resistant genes in dark-single mesocosms (Figure SI2) and Dark-Periodic mesocosms (Figure SI3), and the physical and chemical data for each treatment over the experimental period (Table SI1). This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Pruden, A.; Pei, R.; Storteboom, H.; Carlson, K. H. Antibiotic resistance genes as emerging contaminants: studies in northern

Colorado. Environ. Sci. Technol. 2006, 40 (23), 7445–50. (2) Davies, J. Microbes have the last word. EMBO Reports 2007, 8, 616–621. (3) Witte, W. Medical consequences of antibiotic use in agriculture. Science 1998, 279, 996–997. (4) Roberts, M. C. Update on acquired tetracycline resistance genes. FEMS Microbiol. Lett. 2005, 263 (2), 176–182. (5) De Liguoro, M.; Cibin, V.; Capolongo, F.; Halling-Sorensen, B.; Montesissa, C. Use of oxytetracycline and tylosin in intensive calf farming: evaluation of transfer to manure and soil. Chemosphere 2003, 52 (1), 203–12. (6) Pei, R.; Cha, J.; Carlson, K. H.; Pruden, A. Response of antibiotic resistance genes (ARG) to biological treatment in dariy lagoon water. Environ. Sci. Technol. 2007, 41 (14), 5108–5113. (7) Engemann, C. A.; Adams, L.; Knapp, C. W.; Graham, D. W. Disappearance of oxytetracycline resistance genes in aquatic systems. FEMS Microbiol. Lett. 2006, 263, 176–182. (8) Engemann, C. A.; Keen, P. L.; Knapp, C. W.; Hall, K. J.; Graham, D. W. Fate of tetracycline resistance genes in aquatic systems: migration from the water column to peripheral biofilms. Environ. Sci. Technol. 2008, 42 (14), 5131–6. (9) Chee-Sanford, J. C.; Aminov, R. I.; Krapac, I. J.; GarriguesJeanjean, N.; Mackie, R. I. Occurrence and diversity of tetracycline resistance genes in lagoons and groundwater underlying two swine production facilities. Appl. Environ. Microbiol. 2001, 67 (4), 1494. (10) Smith, M. S.; Yang, R. K.; Knapp, C. W.; Niu, Y.; Peak, N.; Hanfelt, M. M.; Galland, J. C.; Graham, D. W. Quantification of tetracycline resistance genes in feedlot lagoons by real-time PCR. Appl. Environ. Microbiol. 2004, 70 (12), 7372–7377. (11) Knapp, C. W.; Engemann, C. A.; Hanson, M. L.; Keen, P. L.; Hall, K. J.; Graham, D. W. Indirect evidence of transposon-mediated selection of antibiotic resistance genes in aquatic systems at low-level oxytetracycline exposures. Environ. Sci. Technol. 2008, 42 (14), 5348–53. (12) Eaton, A. D.; Clesceri, L. S.; Rice, E. W.; Greenberg, A. E., Eds.; Standard Methods for the Examination of Water and Wastewater; American Public Health Association/ American Water Works Association/ Water Environment Federation: Washington DC, 2005. (13) Marker, A. F. H.; Nusch, E. A.; Rai, H.; Riemann, B. The measurement of photosynthetic pigments in freshwaters and standardization of methods: conclusions and recommendations. Arch. Hydrobiol. Beih. Ergebn. Limnol. 1980, 14, 91–106. (14) Peak, N.; Knapp, C. W.; Yang, R. K.; Hanfelt, M. M.; Smith, M. S.; Aga, D. S.; Graham, D. W. Abundance of six tetracycline resistance genes in wastewater lagoons at cattle feedlots with different antibiotic use strategies. Environ. Microbiol. 2007, 9 (1), 143–151. (15) Distribution of tet resistance genes among Gram-negative bacteria. Available at http://faculty.washington.edu/marilynr/ tetweb2.pdf. (16) Distribution of tetracycline resistance genes among Grampositive bacteria, Mycobacterium, Mycoplasma, Nocardia, Streptomcyes and Ureaplasma. Available at http://faculty.washington.edu/marilynr/tetweb3.pdf. (17) Harms, G.; Layton, A. C.; Dionisi, H. M.; Gregory, I. R.; Garrett, V. M.; Hawkins, S. A.; Robinson, K. G.; Sayler, G. S. Real-Time PCR quantification of nitrifying bacteria in a municipal wastewater treatment plant. Environ. Sci. Technol. 2003, 37 (2), 343– 351. (18) Auerbach, E. A.; Seyfried, E. E.; McMahon, K. D. Tetracycline resistance genes in activated sludge wastewater treatment plants. Water Res. 2007, 41 (5), 1143–1151. (19) Aminov, R. I.; Chee-Sanford, J. C.; Garrigues, N.; Teferedegne, B.; Krapac, I. J.; White, B. A.; Mackie, R. I. Development, validation, and application of PCR primers for detection of tetracycline efflux genes of gram-negative bacteria. Appl. Environ. Microbiol. 2002, 68 (4), 1786–93. (20) Ghosh, S.; LaPara, T. M. The effects of subtherapeutic antibiotic use in farm animals on the proliferation and persistence of antibiotic resistance among soil bacteria. ISME J. 2007, 1 (3), 191–203. (21) Koike, S.; Krapac, I. G.; Oliver, H. D.; Yannarell, A. C.; CheeSanford, J. C.; Aminov, R. I.; Mackie, R. I. Monitoring and source tracking of tetracycline resistance genes in lagoons and groundwater adjacent to swine production facilities over a 3-year period. Appl. Environ. Microbiol. 2007, 73 (15), 4813–4823. (22) Rysz, M.; Alvarez, P. J. J. Amplification and attenuation of tetracycline resistance in soil bacteria: aquifer column experiments. Water Res. 2004, 38 (17), 3705–3712. VOL. 43, NO. 20, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7649

(23) Chee-Sanford, J. C.; Mackie, R. I.; Koike, S.; Krapac, I. G.; Lin, Y. F.; Yannarell, A. C.; Maxwell, S.; Aminov, R. I. Fate and transport of antibiotic residues and antibiotic resistance genes following land application of manure waste. J. Environ. Qual. 2009, 38 (3), 1086–108. (24) Tolls, J. Sorption of veterinary pharmaceuticals in soils: a review. Environ. Sci. Technol. 2001, 35 (17), 3397–406. (25) Hecht, D. W. Prevalence of antibiotic resistance in anaerobic bacteria: worrisome developments. Clin. Infect. Dis. 2004, 39 (1), 92–7. (26) Geesey, G. G. Microbial exopolymers: ecological and economic considerations. ASM News 1982, 48, 9–14.

7650

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 20, 2009

(27) Wingender, J.; Neu, T. R.; Flemming, H.-C.; Wingender, J.; Neu, T. R.; Flemming, H.-C. In What Are Bacterial Extracellular Polymeric Substances?; Springer: Berlin, 1999; pp 1-20. (28) McSwain, B. S.; Irvine, R. L.; Hausner, M.; Wilderer, P. A. Composition and distribution of extracellular polymeric substances in aerobic floc and granular sludge. Appl. Environ. Microbiol. 2005, 71 (2), 1051–1057. (29) Picioreanu, C.; van Loosdrecht, M. C.; Heijnen, J. J. A theoretical study on the effect of surface roughness on mass transport and transformation in biofilms. Biotechnol. Bioeng. 2000, 68 (4), 355–69.

ES9014508