Impact of Treated Wastewater Irrigation on Antibiotic Resistance in

Apr 11, 2012 - wastewater (TWW) irrigation is becoming increasingly prevalent in arid ... The aim of this study was to assess the impact of TWW-irriga...
0 downloads 0 Views 3MB Size
Article pubs.acs.org/est

Impact of Treated Wastewater Irrigation on Antibiotic Resistance in Agricultural Soils Yael Negreanu,†,‡ Zohar Pasternak,‡ Edouard Jurkevitch,‡ and Eddie Cytryn†,* †

Institute of Soil, Water and Environmental Sciences, The Volcani Center, Agricultural Research Organization, POB 6, Bet Dagan, 50250, Israel ‡ Department of Plant Pathology and Microbiology, The Robert H. Smith Faculty of Agriculture, Food and Environment, The Hebrew University of Jerusalem, P.O. Box 12, Rehovot 76100, Israel ABSTRACT: Antibiotic resistance (AR) is a global phenomenon with severe epidemiological ramifications. Anthropogenically impacted natural aquatic and terrestrial environments can serve as reservoirs of antibiotic resistance genes (ARG), which can be horizontally transferred to human-associated bacteria through water and food webs, and thus contribute to AR proliferation. Treatedwastewater (TWW) irrigation is becoming increasingly prevalent in arid regions of the world, due to growing demand and decline in freshwater supplies. The release of residual antibiotic compounds, AR bacteria, and ARGs from wastewater effluent may result in proliferation of AR in irrigated soil microcosms. The aim of this study was to assess the impact of TWW-irrigation on soil AR bacterial and ARG reservoirs. Tetracycline, erythromycin, sulfonamide, and ciprofloxacin resistance in soil was assessed using standard culture-based isolation methods and culture-independent molecular analysis using quantitative real-time PCR (qPCR). High levels of bacterial antibiotic resistance were detected in both freshwater- and TWW-irrigated soils. Nonetheless, in most of the soils analyzed, AR bacteria and ARG levels in TWW-irrigated soils were on the whole identical (or sometimes even lower) than in the freshwater-irrigated soils, indicating that the high number of resistant bacteria that enter the soils from the TWW are not able to compete or survive in the soil environment and that they do not significantly contribute ARG to soil bacteria. This strongly suggests that the impact of the TWW-associated bacteria on the soil microbiome is on the whole negligible, and that the high levels of AR bacteria and ARGs in both the freshwater- and the TWWirrigated soils are indicative of native AR associated with the natural soil microbiome.



INTRODUCTION Although antibiotics are one of the most successful forms of chemotherapy developed by mankind, their extensive administration over the past seven decades has resulted in widespread antibiotic resistance (AR) among an array of potentially pathogenic bacteria in both hospitals and in the community.1 It is well established that clinical misuse of antibiotics significantly contributes to the spread of AR;1 however, there is increasing evidence that anthropogenic activities such as aquaculture, animal husbandry,2−4 agronomic practices5 and municipal wastewater treatment6−10 may contribute to environmental AR propagation due to the release of residual antibiotic compounds, AR bacteria, and antibiotic resistance genes (ARGs) into natural environments. Anthropogenically impacted natural aquatic and terrestrial environments may serve as reservoirs of ARGs, which can be transferred to communityassociated pathogens via water and food webs, potentially contributing to proliferation of AR.2,7−10 In both animals and humans, up to 95% of antibiotics can be excreted in an unaltered state.11 Many of these antibiotic compounds are not completely degraded in wastewater treatment plants (WWTPs) and as a result wastewater and sewage are natural reservoirs of residual concentrations of © 2012 American Chemical Society

antibiotics, of AR bacteria, and of ARGs. Indeed, wastewater treatment was found to cause a selective increase in the relative abundance of AR Acinetobacter spp. isolates,12 and ARGs are often identified in the WWTP’s final effluents indicating that these resistance determinants can be further disseminated in habitats downstream of the sewage plant.13 Several recent studies have targeted ARGs as source tracking indicators to monitor the dispersion and fate of AR from municipal and industrial wastewater effluents and biosolids to the natural aquatic and terrestrial environments.14−17 Storteboom et al. demonstrated that the abundance of selected ARGs in river water from Colorado was significantly higher in anthropogenically impacted areas than in pristine upstream regions of the river.18 The spread of AR among microorganisms is generally believed to be associated with natural selection due to bacterial exposure to antibiotic compounds. While high antibiotic concentrations that exceed minimal inhibitory concentrations Received: Revised: Accepted: Published: 4800

December 28, 2011 April 11, 2012 April 11, 2012 April 11, 2012 dx.doi.org/10.1021/es204665b | Environ. Sci. Technol. 2012, 46, 4800−4808

Environmental Science & Technology

Article

Table 1. Site Description and Soil Characterization of the Four Experimental Sites Investigated in This Studya experimental site/ geographic location Akko (Western Galilee) Rishon LeZion (Central Coast) Gilat (Northern Negev Desert) Kedma (Central Plain)

soil type vertisol (60% clay) dune quartz sand loam (20% clay) vertisol (52% clay)

TWW quality secondary effluent Agamit WWTP secondary effluent WWTP secondary effluent Timan WWTP secondary effluent Sorek WWTP

ShomratShafdan SdeJerusalem

FW quality well water coastal aquifer aquifer injected TWWb fresh water well water coastal aquifer

irrigation period April− October Yearround April− October April− October

crop

plot type

duration of TWW irrigation

avocado

Orchard

12 years

oroblanco citrus trees cotton and wheat olive trees

200 L steel barrel lysimeters field

12 years

orchard

6 years

15 years

a

Each site contained parallel plots irrigated in-tandem with either freshwater or TWW. btertiary infiltrated water- secondary effluent from the Shafdan plant that is recharged through the soil and the vadose zone into groundwater reservoirs where it undergoes natural physical, biological and chemical refinement.

used annually for agriculture.34,35 Many antibiotic compounds are not fully degraded in conventional wastewater treatment processes, indicating that antibiotic residues potentially accumulate in soils during TWW-irrigation. Indeed, sulfamethoxazole contamination has been detected in groundwater samples under agricultural land irrigated for several decades with TWW from a wastewater treatment plant adjacent to Tel Aviv.36 Residual antibiotic compounds can potentially exert selective pressure that may result in proliferation of AR bacteria and ARGs in the irrigated soils. Nonetheless, Israeli37 and international38 microbiological standards for WWTP effluents used for irrigation are limited to quantification of fecal coliforms, and do not address potential propagators of AR such as antibiotic resistant bacteria, residual concentrations of antibiotic compounds or ARGs. The objective of this study was to assess the impact of TWW-irrigation on the soil resistome. Four geographically and physiochemically distinct agricultural soils containing separate experimental plots concomitantly irrigated with either freshwater or treated wastewater were used. Relative levels of antibiotic resistance in the parallel soil plots were assessed using both standard serial-dilution culture-based methods and culture-independent real-time PCR analyses targeting selected clinically associated antibiotic resistance genes.

(MIC) of exposed bacteria are thought to strongly select for AR, recent experimental evidence has shown that subtherapeutic drug concentrations up to several hundred-fold below MICs (25 and 0.46 ng mL−1 of tetracycline and ciprofloxacin, respectively) can also enrich for resistant bacteria.19 These studies have substantial implications for environmental AR reservoirs because they suggest that discharge of residual concentrations of antibiotics into natural environments may generate propagation of AR. Antibiotic concentrations detected in WWTP effluents are often above these values (levels of 1.20 and 0.85 ng mL−1 and 0.40 and 0.12 ng mL−1 have been detected in WWTP influent and effluent for tetracycline and ciprofloxacin, respectively20). In addition, several studies have shown that exposure to compounds other than antibiotics can also result in AR proliferation in a process known as crossresistance.21,22 These include quaternary ammonium compounds (QACs),23 triclosan and triclocarban (used in soaps and other household compounds),24 and heavy metals25 that are often detected in WWTPs at significant concentrations. The concept of the “antibiotic resistome” was first proposed by D′Costa et al. as a framework for understanding the ecology of AR on a global scale.26,27 A myriad of recent culturedependent and molecular-based studies have exposed the immense scope of the soil resistome, illustrating its potential as a reservoir of AR bacteria and ARGs.28−32 Soil AR can be divided into native resistance mechanisms that are naturally found in soil, and acquired resistance that stems from anthropogenic impact. The presence of native mechanisms that long preceded clinical use of antibiotics has been established through recent studies by the authors of ref 31, who identified a highly diverse collection of genes encoding resistance to beta-lactam, tetracycline and glycopeptide antibiotics in the metagenomes of 30,000-year-old permafrost sediments; and by Allen et al.,28 who used functional metagenomics to identify diverse beta-lactamases in a pristine Alaskan soil. Anthropogenic factors may exert selective pressure and thereby expand the level of native resistance in soil. For example, a recent study33 showed increased abundance of ARGs in archived Dutch soils from the 1940s (when antibiotics were first administered) to the present time, indicating that the soil resistome is also anthropogenically impacted. Treated-wastewater (TWW) irrigation is becoming increasingly prevalent in arid regions of the world, due to continuously growing demand and decline in freshwater supplies. Israel has been using treated wastewater (TWW) for irrigation for over forty years, and currently, approximately 480 million cubic meters (MCM) of the 510 MCM of sewage produced each year is recycled for irrigation. This supplies over 50% of the water



MATERIALS AND METHODS Study Sites and Sample Collection. Four agricultural soils with distinct physicochemical properties from diverse geographic locations in Israel were used in this study. Each of the soils were affiliated with long-term (6−18 years) experiments that contained replicate plots irrigated in-tandem with either freshwater or TWW (freshwater plots were not irrigated with TWW and vice versa for the entire duration of the longterm experiments). Two of the soils (Akko and Rishon Lezion) were used for both cultivation and direct molecular targeting of ARGs, while the other two (Gilat and Kedma) were only used for molecular analyses (see below). An overview of the sampling locations, soil type, sources of TWW and freshwater, irrigation period (certain plots were not irrigated during the rainy season between November and March) and the experimental durations are shown in Table 1. Soil Samples. Approximately 50 g of soil were collected from the top 5 cm of soil from the four (field, orchard and lysimeter) experimental plots following removal of surface debris. All of the experimental plots and lysimeters were drip irrigated on a yearly or seasonal basis (Table 1). In the field and orchard plots triplicate biological replicates were sampled directly under irrigation drippers from three different locations along the 4801

dx.doi.org/10.1021/es204665b | Environ. Sci. Technol. 2012, 46, 4800−4808

Environmental Science & Technology

Article

Table 2. List of Quantitative Real-Time PCR Primers Used for in This Study primer

target

sequences

annealing temp (°C)

amplicon size (bp)

references

sul1-FW sul1-RV sul2-FW sul2-RV TetO-FW TetO-RV erm(F)-189f erm(F)-497r erm(B)-91 fc erm(B)-454rc qnrA32F qnrA155R

sul(1)

CGCACCGGAAACATCGCTGCAC TGAAGTTCCGCCGCAAGGCTCG TCCGGTGGAGGCCGGTATCTGG CGGGAATGCCATCTGCCTTGAG ACGGARAGTTTATTGTATACC TGGCGTATCTATAATGTTGAC CGA CAC AGCTTTGGTTGAAC GGACCTACCTCATAGACAAG GATACCGTTTACGAAATTGG GAATCGAGACTTGAGTGTGC AGGATTTCTCACGCCAGGATT CGCTTTCAATGAAACTGCA

55.9

163

53

60.8

191

60

171

54

56

309

55

58

364

57

124

sul(2) tet(O) erm(F) erm(B) qnr(A)

14

was subsequently used to calculate mean and standard deviations of resistance levels at a site. Plates were incubated for 1−4 days at 30 °C. Water samples were also serially diluted in saline and applied to plates as described above. DNA Extraction. DNA was extracted from soil and water samples using a modified bead-beating method as described previously.39 Initial samples consisted of 0.5 g of soil or 30 mL of TWW filtered onto 0.2 μm nitrocellulose filters (described above). Cells were initially lysed by bead-beating (Fast Prep FP 120, Bio101, Savant Instruments Inc., Holbrook, NY) at maximal speed for 50 s in an extraction buffer (100 mM Tris HCl, pH 8.0; 100 mM potassium phosphate buffer pH 8.0; 1% cetyltrimethylammonium bromide (CTAB); and 2% sodium dodecyl sulfate (SDS)). The crude extracts were mixed with KCl to a final concentration of 0.5 M, incubated for 5 min, followed by ethanol precipitation. Pellets were resuspended in Tris-EDTA (TE). DNA present in the supernatant was bound to glass milk (0.5−10 μm silica particles, Sigma Chemical Co., St. Louis, MO) with NaI as described elsewhere.40 The silica was then resuspended in an ethanol-based wash buffer solution40 and this solution, containing the glassmilk, was transferred to a centrifuge tube filter (0.22 μm cutoff nylon filter, Costar, Corning Inc., Corning, NY). After the glass milk was dried via centrifugation, DNA was eluted in 50 μL of TE buffer and quantified with a NanoDrop 1000 Spectrophotometer (ThermoScientific). Extracted DNA samples had an A260/280 ratio of ≥1.6 and an A260/230 ratio of ≥2.0. All DNA samples were diluted to 2 ng/μL in TE buffer prior to qPCR amplification. Real Time Quantitative PCR (qPCR). qPCR was applied to quantify the relative abundance of six selected AR genes (sul(1), sul(2), erm(B), erm(F), tet(O), and qnr(A), associated with sulfonamide, macrolide, tetracycline, and fluoroquinolone resistance, respectively), in the soil and water samples from the four agricultural sites. All qPCR conditions and primers are listed in Table 2. Plasmids containing the target AR genes were used as standard DNA templates for each of the qPCR reactions. The plasmids were decimally diluted to range from 109 to 100 copy numbers. All of the qPCR amplification and quantification was performed using an MX 3000 Real Time PCR system (Stratagene, La Jolla, CA). Each 25 μL reaction contained 12.5 μL Absolute Blue SYBR Green ROX Mix (Thermo Fisher Scientific, Surrey, UK), 1 μL of each primer (0.2 mM), 1 μL of DNA template and 9.5 μL ultra pure PCR grade water. The PCR program began with an initial hot start step of 15 min at 95 °C required for the activation of the DNA polymerase, followed by 40 cycles each consisting of the

irrigation lines of the freshwater (FW)- and TWW-irrigated crop rows. In the Rishon Lezion lysimeter experiment, separate barrels were sampled for each biological replicate (3 FWirrigated, and 3 TWW-irrigated samples) in the same manner as the field samples. Soil samples were transferred on ice from field sites to the laboratory within 4 h of sampling. Samples were taken for serial dilutions and plating (described below), and subsets were frozen at −80 °C for DNA extraction and subsequent molecular analyses. Water Samples. Water samples from the four experimental plots were taken from central FW and TWW taps that feed the irrigation pipes. Triplicate samples were taken for each water type in 1.5 L polyethylene terephthalate bottles. Bottles were flushed three times before filling and taps were opened such that water flowed for 30 s between each replicate. Water samples were transferred to the laboratory on ice and divided into two parts: for culture analyses, water was serially diluted and plated as described below. Concomitantly, approximately 30 mL of the remaining water samples were filtered onto 0.2 μm pore size nitrocellulose filters (Whatman, Brentford, UK) and stored at −80 °C for subsequent molecular analyses. WWTP Samples. Raw sludge and activated sludge samples from the Shafdan WWTP, whose effluent feeds the TWW used to irrigate the Rishon Lezion lysimeters were taken in triplicates from different areas of the sewage inlet and activated sludge reactors, respectively, in 50 mL sterile tubes and stored on ice until arrival at the laboratory. Activated sludge samples (20 mL) were filtered and processed as described from the water samples above. Tubes containing raw sludge were centrifuged at 4000g for 10 min, the supernatant was removed and the pellets were stored at −80 °C for DNA extraction. Enumeration of Resistant Soil Bacteria. Bacteria were dislodged from soil matrices by suspending 1 g (wet weight) of soil in 9 mL of saline (0.85% NaCl) and vortexing at maximal speed for 30 s followed by shaking on a reciprocating shaker (Edmund Bühler GmbH, Hechingen, Germany) for 30 min at 130 rpm. Samples were then serially diluted 10-fold in saline and applied to solid dilute (25%) Luria−Bertani (LB) growth medium amended with: (a) no antibiotic (control), (b) 20 mg/ L of tetracycline, (c) 4 mg/L of ciprofloxacin, or (d) 10 mg/L of erythromycin using standard plating techniques. Antibiotic resistance levels were calculated as the ratio of CFUs growing on plates supplemented with antibiotics compared to the number of CFUs growing on plates without antibiotics. All initial plate counts were done in triplicate and averaged before calculating resistance levels for each soil sample. The average resistance levels for each of three soil samples collected at a site 4802

dx.doi.org/10.1021/es204665b | Environ. Sci. Technol. 2012, 46, 4800−4808

Environmental Science & Technology

Article

following steps: 95 °C for 30s, the respective annealing temp (Table 2) for 30s, 72 °C for 30s, and a final melt curve stage with temperature ramping from 55 to 95 °C. Melting curve analysis of the PCR products was conducted following each assay to confirm that the fluorescence signal originated from specific PCR products. Quantification of total bacteria was also performed using the general bacterial primer set Eub519F and Univ90741,42 as previously described. Baseline and threshold calculations were performed with the MxProTM qPCR software analysis tools (Stratagene, La Jolla, CA). Following real-time PCR, the products were confirmed by agarose gel electrophoresis. PCR products of each targeted gene were cloned (CloneJET PCR Cloning Kit, Fermentas, Lithuania) and sequenced to confirm specificity of the reaction. All qPCR reactions were done in triplicate for both the standards and the microbial community DNA samples. Statistical Analysis. The isolation-based and qPCR experimental data were analyzed with the use of one-way analysis of variance (ANOVA) using the SPSS statistical package (SPSS, Chicago, IL). Nonparametric Kruskal−Wallis tests (performed in SPSS) with P values less than 0.05, were considered to be significant.



RESULTS AND DISCUSSION Heterotrophic Bacterial Plate Counts on AntibioticSelective Media. The relative abundance of cultivatable tetracycline-, ciprofloxacin- and erythromycin-resistant bacteria was evaluated in two profiles from FW- and TWW-irrigated Avocado orchard plots in Akko and one profile from the FWand TWW-irrigated citrus lysimeters in Rishon Lezion. Total bacterial cell levels in the FW- and TWW-irrigated soils in the Akko and Rishon Lezion profiles ranged between 1 and 3 × 106 CFU/g soil (ww). Despite differences in conditions, statistically significant differences (p < 0.05) in bacterial abundance were not observed between sites or between the FW- and TWWirrigated soils. The abundance of tetracycline-, ciprofloxacin-, and erythromycin-resistant strains relative to the total bacterial levels in the three profiles are shown in Figure 1. Although significant differences were sometimes seen in the relative abundance of AR bacteria between profiles, the relative abundance of AR bacteria was never significantly higher in TWW-irrigated soils than in corresponding FW-irrigated plots, although the opposite phenomenon was observed on two occasions: the September 2010 Akko profile, where the relative abundance of tetracycline resistant bacteria in the FW-irrigated soil was significantly higher (p < 0.05) than in the TWW-soils (Figure 1a); and the Rishon Lezion profile, where higher ciprofloxacin resistance (p < 0.05) was found in the FWirrigated soil than in the TWW-irrigated soil (Figure 1b). Although, serial dilution plating on heterotrophic medium targets a very small fraction of the aerobic, heterotrophic soil microbiome, and resistance to only three antibiotics was tested in this study, these results fail to show any correlation between TWW-irrigation and elevated soil AR. This phenomenon was supported by additional platting experiments that assessed levels of ampicillin- and tetracycline-resistant bacteria in two additional sites (a clay rich soil in northern Israel and a sandy soil in southern Israel) where parallel plots (duration of four and ten years parallel irrigation, respectively) were irrigated with either freshwater or TWW (Negreanu, unpublished). In contrast to water quality (TWW vs FW), soil moisture did have a significant inducing effect on both tetracycline and ciprofloxacin resistance levels and may be a principal factor

Figure 1. Cultivation based estimation of relative abundance of tetracycline resistant (A); ciprofloxacin resistant (B) and erythromycin resistant (C) bacteria in TWW- (black bars) and FW- (striped bars) irrigated soils from the Akko and Rishon Lezion sites. Asterisks indicate statistically significant (p < 0.05) differences between the TWW and freshwater-irrigated samples.

associated with propagation of AR in soil. Soil samples taken directly under the irrigation drippers showed significantly higher antibiotic resistance levels (p < 0.05) than samples taken from soils 50 cm from the drippers (data not shown). This may be explained by higher levels of microbial activity or horizontal gene transfer in the wet soils. Indeed previous studies have indicated that R-plasmid transfer significantly increases with increased soil−water-holding capacity and nutrient availability.43 Total bacterial counts and tetracycline- and ciprofloxacinresistant levels were also examined in TWW and FW irrigation water from the 2011 Akko and Rishon Lezion profiles (Figure 2). Average bacterial abundance in the FW samples was approximately 100 CFU/mL, whereas in the TWW samples levels were roughly 4 orders of magnitude higher (Figure 2). The abundance of resistant bacteria found in the TWW samples ranged between 50 and 450 CFU/mL for tetracycline, 4803

dx.doi.org/10.1021/es204665b | Environ. Sci. Technol. 2012, 46, 4800−4808

Environmental Science & Technology

Article

from soil particles. Therefore, qPCR was applied concomitant to the traditional isolation techniques described above in order to assess the impact of TWW-irrigation on the soil resistome. Six different ARGs, conferring resistance to four different clinically relevant antibiotic families, were targeted in treated wastewater (TWW); and in FW- and TWW-irrigated soil samples from the four experimental sites detailed in Table 1; and in raw sewage (RS) and activated sludge (AS) from the Rishon Lezion WWTP. The qnr(A) gene is associated with plasmid-borne fluoroquinolone resistance which has become increasingly prevalent in clinically relevant gram negative bacteria.45 Tet(O) is a ribosome protection protein that confers resistance to tetracycline that is generally found on plasmids in gram positive bacteria.46 Sul(1) and sul(2) are linked to sulfonamide resistance. The sul(1) gene is an excellent indicator of both horizontal gene transfer and multiple resistance because it is generally harbored in class 1 integrons on conjugative plasmids that often contain other resistance genes; while sul(2) is usually located on small nonconjugative plasmids47 or large transmissible multiresistance plasmids.48 Erm(B) and erm(F), that confer resistance to macrolide, lincosamide and streptogramin (MLS) antibiotics, are generally found on congugative genetic elements. Although originally believed to be carried by gram positives, there is substantial evidence showing their transfer to gram negative bacteria, especially within the Bacteroidetes phylum.49

Figure 2. Cultivation based estimation of bacterial abundance in water samples used for irrigation of the Akko and Rishon Lezion sites. Akko FW (diagonal striped bars); Rishon Lezion FW (v striped bars); Akko TWW (black bars); Rishon Lezion TWW (gray bars).

and between 700 and 1100 CFU/mL for ciprofloxacin, in the Akko and Shafdan profiles, respectively. In freshwater samples, resistant bacteria were not detected (Figure 2). Quantification of Resistance Genes at the Sites. Traditional cultivation-based approaches are limited in their capacity to estimate bacterial abundance because they detect only a small fraction (