High-throughput profiling of antibiotic resistance genes in drinking water treatment plants and distribution systems

Environmental Pollution 213 (2016) 119e126

Contents lists available at ScienceDirect

Environmental Pollution journal homepage: www.elsevier.com/locate/envpol

High-throughput profiling of antibiotic resistance genes in drinking water treatment plants and distribution systems* Like Xu a, Weiying Ouyang b, Yanyun Qian a, Chao Su a, Jianqiang Su b, Hong Chen a, * a b

Department of Environmental Engineering, College of Environmental and Resource Sciences, Zhejiang University, Hangzhou 310058, China Key Lab of Urban Environment and Health, Institute of Urban Environment, Chinese Academy of Sciences, Xiamen 361021, China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 18 August 2015 Received in revised form 3 January 2016 Accepted 4 February 2016 Available online xxx

Antibiotic resistance genes (ARGs) are present in surface water and often cannot be completely eliminated by drinking water treatment plants (DWTPs). Improper elimination of the ARG-harboring microorganisms contaminates the water supply and would lead to animal and human disease. Therefore, it is of utmost importance to determine the most effective ways by which DWTPs can eliminate ARGs. Here, we tested water samples from two DWTPs and distribution systems and detected the presence of 285 ARGs, 8 transposases, and intI-1 by utilizing high-throughput qPCR. The prevalence of ARGs differed in the two DWTPs, one of which employed conventional water treatments while the other had advanced treatment processes. The relative abundance of ARGs increased significantly after the treatment with biological activated carbon (BAC), raising the number of detected ARGs from 76 to 150. Furthermore, the final chlorination step enhanced the relative abundance of ARGs in the finished water generated from both DWTPs. The total enrichment of ARGs varied from 6.4-to 109.2-fold in tap water compared to finished water, among which beta-lactam resistance genes displayed the highest enrichment. Six transposase genes were detected in tap water samples, with the transposase gene TnpA-04 showing the greatest enrichment (up to 124.9-fold). We observed significant positive correlations between ARGs and mobile genetic elements (MGEs) during the distribution systems, indicating that transposases and intI-1 may contribute to antibiotic resistance in drinking water. To our knowledge, this is the first study to investigate the diversity and abundance of ARGs in drinking water treatment systems utilizing highthroughput qPCR techniques in China. © 2016 Elsevier Ltd. All rights reserved.

Keywords: Antibiotic resistance genes Drinking water treatment plants Tap water High-throughput qPCR

1. Introduction Antibiotic resistance genes (ARGs) are emerging environmental pollutants (Pruden et al., 2006). ARGs are present in wastewater treatment plants (WWTPs), livestock and soil. They can enter into surface water and underground water through rain or surface runoff, resulting in high levels of ARGs that are found in the water environment (Wellington et al., 2013). Additionally, the association of ARGs and mobile gene elements (MGEs) can accelerate the proliferation of ARGs through horizontal gene transfer (HGT) in the water environment. HGT is widely recognized as the mechanism responsible for the widespread distribution of bacterial antibiotic resistance (de la Cruz and Davies, 2000; Gyles and Boerlin, 2014).

*

This paper has been recommended for acceptance by Maria Cristina Fossi. * Corresponding author. E-mail address: [email protected] (H. Chen).

http://dx.doi.org/10.1016/j.envpol.2016.02.013 0269-7491/© 2016 Elsevier Ltd. All rights reserved.

WWTPs are an important source of introducing ARGs into surface water. ARGs remaining in the finished water from WWTPs have a good chance of entering into rivers or lakes, contributing to antibiotic resistance pollution in surface water (Amos et al., 2014; Xu et al., 2015; Zhang et al., 2009). Also, aquaculture is the most direct way of introducing ARGs into the water environment. Antibiotic resistance levels in aquaculture systems are well documented and it has been suggested that this system may serve as a reservoir for antibiotic-resistant bacteria (ARB) and ARGs (Gao et al., 2012; Phuong Hoa et al., 2008; Su et al., 2011). In modern cities, the drinking water that is supplied to the population is often obtained from nearby surface water after its rigorous treatment in WWTPs. However, the high concentration of antibiotics and ARGs remaining in surface water might enter water supply pipelines through drinking water treatment systems(Jones et al., 2005), and this increases the potential for antibiotic resistance pollution of drinking water. The finished water from drinking water treatment plants (DWTPs) is provided to the local population through water supply

120

L. Xu et al. / Environmental Pollution 213 (2016) 119e126

systems. Thus, the quality of the drinking water can potentially cause disease to the citizens who consume the water. Jiang et al. investigated the ARGs present in the Huangpu River and found that the abundance of ARGs in water samples near the drinking water sources was higher than other samples that were distant from those sources (Jiang et al., 2013). DWTPs have an implication on the behavior of ARGs as it might increase the antibiotic resistance of surviving bacteria, and the finished water from DWTPs have been demonstrated to contain ARGs (Guo et al., 2014; Pruden et al., 2006; Xi et al., 2009). The conventional treatment process to generate drinking water includes coagulation, sedimentation, sand filtration and chlorination. The second generation of drinking water treatment adopts the combination of ozone and biological activated carbon (BAC). Thus far, the research of ARGs in municipal works is mainly focused on WWTPs, and there is a lack of knowledge about the abundance of ARGs in urban DWTPs. Moreover, the existing research on ARGs in DWTPs is incomplete and does not reflect a comprehensive profile of ARGs in drinking water. The aforementioned influence of ARGs on human health necessitates a comprehensive investigation of ARGs in DWTPs and distribution systems. The Qiantang River is an important commercial artery passing through the provincial capital Hangzhou before flowing into the East China Sea. To insure the safety of drinking water in this metropolitan area, various types of drinking water treatment process are employed, including conventional and advanced treatment processes. In this study, we investigated ARGs in two representative drinking water treatment plants and the distribution systems in Hangzhou City. Both drinking water suppliers use Qiantang River as the drinking water source for over 15 years and different technologies are employed during the treatment processes. In total, 285 ARGs, 8 transposes, and intI-1 were detected by using highthroughput quantitative PCR. This technique can be used to comprehensively profile ARGs in different environmental samples (Wang et al., 2014). The main goals of the current study are to detect the abundance of ARGs in drinking water treatment plants and distribution systems, as well as to analyze the potential for ARGs to propagate in DWTPs and water distribution systems.

2. Materials and methods 2.1. DWTPs and sample collection Water samples from two representative drinking water treatment plants (DWTPs) were collected in June and November of 2014 in Hangzhou city, eastern China. DWTP-1 employs the advanced O3/BAC treatment while DWTP-2 adopts conventional treatment. The water supply capacities are 100,000 m3/d and 600,000 m3/d in DWTP-1 and DWTP-2, respectively. The two DWTPs both use Qiantang River as the drinking water source. Qiantang River is one of the main rivers among coastal areas of southeast China. A summarization of the treatment scheme for each plant is shown in Table 1. For the analysis of drinking water distribution system, two residential areas located in the supplying area of DWTP-2 were selected in this study. The locations of DWTPs and residential areas (RA) were presented in Fig. 1.

Water samples were collected from source and finished water, as well as at every treatment procedure of the two DWTPs. The tap water from two different pipes of each residential area were also collected. Water samples were all collected using small-scale vacuum pump and suction filtration in each site due to the special properties of drinking water. Water samples were filtered through 0.22 mm membrane filters to capture bacteria and three samples were simultaneously filtered from each site. The membrane filters were carefully stored in prepared sterile silver paper bag and transported to the laboratory in an ice box. 2.2. DNA extraction Genomic DNA from the water samples were extracted using FastDNA SPIN Kit (MP Bio, USA) according to the manufacturers' instructions. For the DNA analysis of each sample, we mixed the DNA extracted from three water samples of each site as the final DNA sample. The concentration of the purified DNA was quantified spectrophotometrically (NanoDrop ND-2000c, Thermo, USA) and stored at
20  C until subsequent analysis. 2.3. High-throughput quantitative PCR (HT-qPCR) All high-throughput qPCR reactions were performed using the Wafergen SmartChip Real-time PCR system. A majority of primer sets have been validated and used in previous study (Ouyang et al., 2015; Zhu et al., 2013). There were altogether 295 primer sets targeting 285 ARGs, 8 transposases, 1 class1 intergron and 16S rRNA gene. The 285 ARG assays in this research conferred resistance to almost all major antibiotics and covered three resistance mechanisms. Amplification was conducted in 100 nL reaction containing (final concentration) 1  LightCycler 480 SYBR Green I Master Mix (Roche Inc., USA), Nuclease-free PCR-Grade water, 1 ng mL
1 BSA, 3 ng mL
1 DNA template, 1 mM of each forward and reverse primer. The thermal cycle was: initial denaturation at 95  C for 10 min, followed by a 40 cycles of denaturation at 95  C for 30s, annealing at 60  C for 30s, finally with amelting curve analysis auto-generated by the program. For each primer set, amplification was conducted in triplicate and a non-template control was included. The results of the high-throughput qPCR were analyzed using SmartChip qPCR software (V2.7.0.1), wells with multiple melting peak as well as wells with amplification efficiency beyond the range (1.8e2.2) were discarded. A threshold cycle (Ct) 31 was used as the detection limit. Only samples with three replicates that had amplification were regarded as positive. Data processing was done with Microsoft Excel 2010, while diagramming and correlation analysis were done with Origin 9.0. Heatmap graphs were produced using Heatmap Illustrator 1.0.1. 3. Result and discussion 3.1. Antibiotic resistance genes in DWTPs 3.1.1. Diversity of ARGs in DWTPs Among all of the targeted 285 ARGs in this study, a total of 184 ARGs were detected in DWTP-1 while 192 were detected in DWTP-

Table 1 Water sources and treatment schematics of the two DWTPs targeted in this study. DWTP

Source

Treatment processes

1 2

Qiantang River Qiantang River

Raw rivera/Pre-ozonationa/Coagulation/flocculation/Sedimenta/Sand filtera/Ozonationa/BACa/Chlorinationa Raw rivera/Coagulation/flocculationa/Sedimenta/Sand filter Chlorinationa

a

Samples were collected after each corresponding treatment process.

L. Xu et al. / Environmental Pollution 213 (2016) 119e126

121

Fig. 1. Geographic map of sampling sites. DWTP: drinking water treatment plant; RA: residential area.

2. The number of detected ARGs in all of the samples in two DWTPs is shown in Fig. 2. The presence of ARGs varied during the drinking water treatment procedure. In the first half of the treatment procedure in DWTP-1, the number of detected individual ARG declined gradually from 74 in raw water to 65 after sand filtration. Notably, there was a rise in the detected number of ARGs after BAC treatment, ranging from 76 to 150 (p < 0.05). Among the ARGs present, MLSB resistance genes increased the most and the detected number rose from 5 to 23 (p < 0.01). erm genes which belong to the class of macrolide resistance genes accounted for 33% of MLSB resistance genes in this study. Previous research has shown that erm genes can be easily

captured by mobile gene elements such as plasmids and transposes, easily being transferred among different host bacteria (Roberts, 2003). After chlorination, the number of detected ARGs decreased from 150 to 120 in the finished water. As we can see from Fig. 2, the detected number of ARGs in DWTP-2 gradually declined from 168 in raw water to 115 in finished water. 3.1.2. Abundance of ARGs in DWTPs samples An overview of the absolute abundance of ARGs throughout the treatment processes at the DWTPs is shown in Fig. 3. The total concentration of ARGs among all of the samples was between 105~1010 copies/L. A trend of decrease of ARGs concentrations can

Fig. 2. Average number of unique resistance genes detected in A: DWTP-1; B: DWTP-2. The resistance genes detected in all samples were classified based on the antibiotic to which they confer resistance. FCA, fluroquinolone, quinolone, florfenicol, chloramphenicol, and amphenicol resistance genes; MLSB, Macrolide-Lincosamide-Streptogramin B resistance.

122

L. Xu et al. / Environmental Pollution 213 (2016) 119e126

Fig. 4. Relative abundance of ARGs in A:DWTP-1;B:DWTP-2. Fig. 3. Absolute abundance of ARGs in A:DWTP-1;B:DWTP-2.

be observed in both of DWTPs and the total removal of ARGs in DWTP-2 (2.46 log) was higher than in DWTP-1 (1.89 log). The relative abundance of ARGs (normalized to the corresponding 16SrRNA gene copy numbers) in DWTPs is shown in Fig. 4. In DWTP-1, which employed advanced treatment processes, the raw water was first treated by pre-ozonation instead of prechlorination in order to reduce the production of disinfection byproducts. The relative abundance of ARGs decreased slightly after both pre-ozonation and ozonation treatments (Fig. 4A). However, the concentrations of almost all types of ARGs (copies/L) increased after pre-ozonation but decreased significantly (p < 0.05) after ozonation (Fig. 3A). This discrepancy could be explained by the different ozone dosage used for the two ozonation procedures (0.30e0.60 mg/L for pre-ozonation and 1.50~1.70 mg/L for ozonation). Previous research under lab conditions has demonstrated that the efficiency of ARG removal was enhanced by increasing the ozone dosage simultaneously between the range of 0e7 mg/L (Oh et al., 2014). In addition, ozone can react with organic matter and metals in water and slightly affect bacteria and ARGs (Guo et al., 2014). Sand filtration is often regarded as an efficient and stable technology in water treatment plants. Suspended solids and waterborne pathogens are removed by sand filtration using both physical and biological processes (Li and Zhang, 2013), and this process may influence ARB. The relative abundance of a small

number of ARGs families was increased after sand filter while the absolute ARGs concentrations decreased, which is consistent with Guo's research on ARGs in DWTPs (Guo et al., 2014). The reaction mechanisms of different sand media with ARB or ARGs are still unclear and warrants further study. We observed a noteworthy phenomenon in DWTP-1. The relative abundance of ARGs increased consistently after biological activated carbon (BAC) and chlorination while the ARGs concentrations decreased significantly after the two processes (p < 0.05). Micropollutants such as antibiotics and ARB can adhere to the biofilm generated on the surface of activated carbon during BAC processing, resulting in high concentrations of biomass and antibiotic contaminants in biofilms. Michael et al. summarized the absorption of antibiotics by BAC and found that a wide range of antibiotics (e.g. b-lactam, macrolides, quinolones, and tetracycline) can be effectively removed by BAC (Michael et al., 2013). Since it is widely accepted that antibiotic contaminants in the environment are the main reason for selection and transmission of resistance (Baquero et al., 2008; Martinez, 2009), we can deduce that selective pressure driven by high levels of antibiotics may give rise to ARB containing various ARGs in biofilm. In addition, dynamic changes between adsorptive microbes and planktonic microbes has been shown to occur in biofilms, suggesting that gene exchanges exist between these bacterial communities (Farkas et al., 2013). In this study, we detected both high concentrations of MGEs (1.13  108 copies/L) and increased number of detected ARGs in the effluent of

L. Xu et al. / Environmental Pollution 213 (2016) 119e126

BAC treatment. The significant positive correlation between ARGs and MGEs (Tables S1 and S2) indicated that horizontal gene transfer might occur not only in bacteria adsorbed by biofilm but also in bacteria contained in water phase flowing near the biofilm. Although BAC can adsorb different kinds of microorganisms through the biofilm generated on the surface of activated carbon, it is still unknown whether these specific adsorbed microorganisms influence the induction or transmission of ARGs. Nevertheless, in another study, the relative abundance of some tetracycline and sulfonamide resistance genes also increased after BAC during DWTPs (Guo et al., 2014). Chlorination can result in the enrichment of ARB and the proliferation of ARGs (Xi et al., 2009), and this study showed an increase of the relative abundance of ARGs in the finished water after chlorination. In previous research conducted by Shi et al., chlorination was shown to have a strong effect on the structure of the microorganism community in drinking water. Also the abundance and diversity of ARGs increased significantly after chlorination (Shi et al., 2013). In the clear water tank, the surviving bacteria were more likely to obtain co-resistance to both the disinfectant and antibiotics, beginning a “secondary growth” phase, and would eventually increase the levels of ARGs. A co-resistance phenomenon was observed after chlorination in the finished water from both DWTPs. The increased relative abundance of ARGs despite the ARGs concentrations decrease indicated that the final treatment might have enhanced the resistance of a subpopulation of bacteria in the finished water resulting in the enrichment of ARGs. In this study, the number of detected ARGs in raw water in DWTP-2 (n ¼ 168) was much higher than DWTP-1 (n ¼ 74) (Fig. 2), and there was a significant difference between them (p < 0.05). Besides, the absolute abundance of ARGs in raw water of DWTP-2 (2.52  109 copies/L in average) was also higher than in DWTP-1 (4.10  108 copies/L in average). Previous research in Huangpu River indicated that the levels of ARGs in areas with anthropogenic activities was much higher than in areas that were not affected by human activities (Jiang et al., 2013). Comparing the geographic locations of the two DWTPs, DWTP-2 is at the downstream region of DWTP-1. There are many residential and commercial areas located along the river, thus the interference of human activities may have contributed to the significant increase in the levels of ARGs in the raw water in DWTP-2. Despite of the significant difference (p < 0.05) of both the abundance and detected numbers of ARGs in raw water found between the two DWTPs, the finished water of the two DWTPs had similar levels of detected ARGs number ranging between 115 and 120 and absolute abundance of ARGs between 105-108 copies/L. Overall, the removal of ARGs in DWTP-2 (2.46 log) was higher than in DWTP-1 (1.89 log), indicating that the advanced drinking water treatment used by DWTP-1 did not add an expected benefit in eliminating ARGs compared to conventional methods. 3.2. Dynamics of antibiotic resistance genes in the distribution systems 3.2.1. Diversity of ARGs in tap water samples In this study, we selected two residential areas that are located in the supplying area of DWTP-2. To explore the variation of ARGs in drinking water distribution systems, the raw water and finished water from DWTP-2 and tap water in residential areas were compared in the following analyses. The raw water and finished water mentioned below specifically refer to water samples in DWTP-2. The total number of ARGs detected in tap water samples ranged from 112 to 122 (Fig. S1). Among the detected ARGs in raw water, finished water, and tap water, antibiotic deactivation and efflux pump were the two most

123

dominant resistance mechanisms (Fig. 5A). Surprisingly, the diverse set of resistance genes detected in tap water could potentially confer resistance to all major classes of antibiotics including those that are extremely important for human medicine(e.g. aminoglycosides, beta-lactam, marcolides, and vancomycin) (Zhu et al., 2013). Resistance genes for beta-lactam (19.1%e22.8%), aminoglycoside (14.3%e16.5%) and MLSB (12.2%e13.7%) were the three most dominate types in the raw, finished, and tap water samples (Fig. 5B) followed by resistance genes for FCA (11.3%e12.9%), tetracycline (10.1%e12.5%), vancomycin (5.4%e7.0%), and sulfonamide (1.1%e 1.8%). The number of vancomycin resistance genes in tap water (6.0%) was even higher than that in raw water (5.4%). Since vancomycin is recognized as a “last-resort” life-saving antibiotic (Ziglam and Finch, 2001), our detection of vancomycin resistance genes in tap water indicates that the water being consumed in this area may potentially affect human health and increase the resistance of the human body to certain antibiotics. The broad-spectrum distribution of different types of ARGs in tap water samples detected in this study reveals that the pollution of ARGs has already sprawled into the drinking water. 3.2.2. Abundance of ARGs in distribution systems The absolute abundance of ARGs significantly increased (p < 0.01) after the pipeline transportation (Fig. S2-a), especially for beta-lactam resistance genes which had grown from 1.08  107copies/L in finished water to 5.12  108 copies/L (average value) in tap water. Similarly, the relative abundance of most ARGs families except for other/efflux increased in tap water compared to finished water (Fig. S2-b), indicating the regrowth of resistance bacteria in drinking water distribution systems. The elevated abundance suggested that the distribution systems might be an important ARGs reservoir and should be paid more attention to. Resistance gene profiles indicated the patterns and degrees of enrichment of ARGs for each water sample (Fig. 6). The fox5 gene which belongs to the beta-lactam resistance gene family was the most enriched ARG after pipeline transportation with an enrichment up to 109.2-fold. Previous study has confirmed that Enterobacter and Aeromonas spp. harbor plasmid-mediated fox5 gene and this gene was also found to be located on plasmids in K. pneumoniae strains isolated from clinical samples (Coudron et al., 2003; Queenan et al., 2001). On the resistance profile, several ARGs (e.g. aacC, blaCTX-M, blaSHV, qacH), which are commonly embedded in integron resistance cassettes (Monstein et al., 2007; Singh et al., 2005) were detected with the higher absolute abundance in tap water samples, indicating horizontal gene transfer exacerbated the dissemination of these ARGs. The sul II gene which had a 9.7-fold enrichment in tap water is previously found to be located on IncN plasmids in WWTPs effluent and these IncN plasmid have also been detected in clinical samples, which were regarded as having human health relevance (Eikmeyer et al., 2012). The finished water of DWTPs experienced a complicated path before reaching the residential areas. During transportation, it is common to have biofilm formation due to microbial regrowth on tube walls. Biofilm exerts a secondary pollution of drinking water during transportation. Moreover, the water supply networks are like a huge reactor, providing an ideal environment for the continuous reaction between different micropollutants. Consequently, ARB in drinking water may experience complex reactions during transportation, resulting in variations of ARGs abundance. Lv et al. summarized some possible reasons for the obtainment of resistance in ARB during the drinking water distribution systems: 1) horizontal gene transfer; 2) cross- or co-resistance to heavy metals or other antibacterial agents; and 3) chromosomal mutations (Lv et al., 2014). They further confirmed that some disinfection

124

L. Xu et al. / Environmental Pollution 213 (2016) 119e126

Fig. 5. Resistance genes detected in raw water, finished water of DWTP-2 and tap water were classified based on A: the mechanism of resistance, and B: the antibiotic to which they confer resistance. FCA, fluroquinolone, quinolone, florfenicol, chloramphenicol, and amphenicol resistance genes; MLSB, Macrolide-Lincosamide-Streptogramin B resistance.

observed elevated resistance to some antibiotics during water treatment and in tap water, indicating that the water distribution systems may serve as a reservoir for the spread of antibiotic resistance (Xi et al., 2009). 3.3. Correlation between ARGs and MGEs In this study, the absolute abundance of MGEs after transportation increased from 1.20  107 copies/L in finished water to 1.41  109 copies/L (average value) in tap water (Fig. S3-a), which was close to raw water (2.27  109 copies/L). The relative abundance of MGEs followed the order: tap water > finished water > raw water (Fig. S3-b). Among the eight transposase genes, five were detected (TnpA-02, TnpA-03, TnpA-04, TnpA-05, IS613) in both the finished water and tap water. TnpA-04 was the most enriched transposase with a striking enrichment up to 124.9-fold. The correlations between the abundance of ARGs and transposase, intI-1, and MGEs are shown in Table 2. Most of ARGs have significant correlations with total abundance of MGEs (P < 0.05), among which FCA, MLSB, other/efflux and tetracycline resistance genes have the highest correlations (P < 0.01) (Fig. 7). Table 2 The correlations between absolute abundance of ARGs and MGEs in drinking water distribution systems.

Fig. 6. A heatmap of the ARGs showing distinct pattern between raw water (RW), finished water (FW) of DWTP-2 and tap water (TW).

byproducts could induce antibiotic resistance, even generating multidrug resistance in drinking water under lab conditions. In short, the drinking water treatment and distribution systems can affect the propagation of ARGs. A previous study in America

ARGs

Transposase

intI-1

MGEs

Aminoglycoside Beta-lactam FCA MLSB Other/efflux Sulfonamide Tetracycline Vancomycin

0.741* 0.618* 0.828** 0.818** 0.821** 0.421 0.863** 0.732*

0.532 0.724* 0.675* 0.760* 0.257 0.697* 0.605* 0.497

0.739* 0.628* 0.851** 0.837** 0.835** 0.435 0.881** 0.729*

*P < 0.05, **P < 0.01. MGEs refers to the sum of the transposase and intI-1.

L. Xu et al. / Environmental Pollution 213 (2016) 119e126

125

Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.envpol.2016.02.013. References

Fig. 7. Correlations of total FCA, MLSB, other/efflux, tetracycline resistance with MGEs abundance. The data used was logarithmic transformation.

Previous studies have demonstrated that several tetracycline resistance genes are linked on plasmids and other novel MGEs (Ciric et al., 2014, 2011). Okitsu found that the ermB gene was present on transposons, including Tn917 and Tn1545 (Okitsu et al., 2005). ARGs that are associated with MGEs can propagate among species through horizontal gene transfer mechanisms (Chen and Zhang, 2013), contributing to the persistence and spread of ARGs in the environment as well as the variations of ARGs in drinking water distribution systems. The significant correlations observed between ARGs and MGEs suggested that gene transfer occurred during the transportation process. The residual amount of ARGs and MGEs remaining in tap water may possibly interact with human intestinal flora and spread the fragment of resistance genes to indigenous and intestinal bacteria in the human body (Mathur and Singh, 2005; Salyers et al., 2004), potentially compromising human health. 4. Conclusion In summary, this study has revealed the impact of DWTPs and distribution systems on the profile of ARGs. Comparison between DWTPs suggested that the advanced treatment of drinking water did not add a significant benefit for the removal of ARGs compared to conventional treatments. The elevated abundance of ARGs in tap water indicates that distribution systems could be an important reservoir of ARGs. Overall, our research helps evaluate the pollution of ARGs in drinking water and we demonstrated that the drinking water treatment and distribution systems could affect the propagation of ARGs. Further research is needed to explore the potential transfer mechanisms of ARGs in specific drinking water treatment procedures including sand filtration and BAC, to decrease the harm of ARGs on human health. Acknowledgments We would like to thank the managers of drinking water treatment plants for providing the samples and the information needed for this study. Ms. Lam Sze Ying from University College London helped with the proof reading of this article. This work was supported from the Natural Science Foundation of China (No. 21277117, No. 20210008).

Amos, G.C.A., Zhang, L., Hawkey, P.M., Gaze, W.H., Wellington, E.M., 2014. Functional metagenomic analysis reveals rivers are a reservoir for diverse antibiotic resistance genes. Veterinary Microbiol. 171, 441e447. Baquero, F., Martinez, J.L., Canton, R., 2008. Antibiotics and antibiotic resistance in water environments. Curr. Opin. Biotechnol. 19, 260e265. Chen, H., Zhang, M., 2013. Occurrence and removal of antibiotic resistance genes in municipal wastewater and rural domestic sewage treatment systems in eastern China. Environ. Int. 55, 9e14. Ciric, L., Mullany, P., Roberts, A.P., 2011. Antibiotic and antiseptic resistance genes are linked on a novel mobile genetic element: Tn6087. J. Antimicrob. Chemother. 66, 2235e2239. Ciric, L., Brouwer, M.S.M., Mullany, P., Roberts, A.P., 2014. Minocycline resistance in an oral Streptococcus infantis isolate is encoded by tet( S) on a novel small, low copy number plasmid. Fems Microbiol. Lett. 353, 106e115. Coudron, P.E., Hanson, N.D., Climo, M.W., 2003. Occurrence of extended-spectrum and AmpC beta-lactamases in bloodstream isolates of Klebsiella pneumoniae: isolates harbor plasmid-mediated FOX-5 and ACT-1 AmpC beta-lactamases. J. Clin. Microbiol. 41, 772e777. de la Cruz, F., Davies, J., 2000. Horizontal gene transfer and the origin of species: lessons from bacteria. Trends Microbiol. 8, 128e133. Eikmeyer, F., Hadiati, A., Szczepanowski, R., Wibberg, D., Schneiker-Bekel, S., Rogers, L.M., Brown, C.J., Top, E.M., Puhler, A., Schluter, A., 2012. The complete genome sequences of four new IncN plasmids from wastewater treatment plant effluent provide new insights into IncN plasmid diversity and evolution. Plasmid 68, 13e24. Farkas, A., Butiuc-Keul, A., Ciataras, D., Neamtu, C., Craciunas, C., Podar, D., DraganBularda, M., 2013. Microbiological contamination and resistance genes in biofilms occurring during the drinking water treatment process. Sci. Total Environ. 443, 932e938. Gao, P., Mao, D., Luo, Y., Wang, L., Xu, B., Xu, L., 2012. Occurrence of sulfonamide and tetracycline-resistant bacteria and resistance genes in aquaculture environment. Water Res. 46, 2355e2364. Guo, X., Li, J., Yang, F., Yang, J., Yin, D., 2014. Prevalence of sulfonamide and tetracycline resistance genes in drinking water treatment plants in the Yangtze River Delta, China. Sci. Total Environ. 493, 626e631. Gyles, C., Boerlin, P., 2014. Horizontally transferred genetic elements and their role in Pathogenesis of bacterial disease. Veterinary Pathol. 51, 328e340. Hoa, Phuong, Nonaka, P.T., Hung, L., Viet, P., Suzuki, S., 2008. Detection of the sul1, sul2, and sul3 genes in sulfonamide-resistant bacteria from wastewater and shrimp ponds of north Vietnam. Sci. Total Environ. 405, 377e384. Jiang, L., Hu, X., Xu, T., Zhang, H., Sheng, D., Yin, D., 2013. Prevalence of antibiotic resistance genes and their relationship with antibiotics in the Huangpu River and the drinking water sources, Shanghai, China. Sci. Total Environ. 458e460, 267e272. Jones, O.A., Lester, J.N., Voulvoulis, N., 2005. Pharmaceuticals: a threat to drinking water? Trends Biotechnol. 23, 163e167. Li, B., Zhang, T., 2013. Different removal behaviours of multiple trace antibiotics in municipal wastewater chlorination. Water Res. 47, 2970e2982. Lv, L., Jiang, T., Zhang, S.H., Yu, X., 2014. Exposure to Mutagenic disinfection byproducts leads to increase of antibiotic resistance in Pseudomonas aeruginosa. Environ. Sci. Technol. 48, 8188e8195. Martinez, J.L., 2009. Environmental pollution by antibiotics and by antibiotic resistance determinants. Environ. Pollut. 157, 2893e2902. Mathur, S., Singh, R., 2005. Antibiotic resistance in food lactic acid bacteria - a review. Int. J. Food Microbiol. 105, 281e295. Michael, I., Rizzo, L., McArdell, C.S., Manaia, C.M., Merlin, C., Schwartz, T., Dagot, C., Fatta-Kassinos, D., 2013. Urban wastewater treatment plants as hotspots for the release of antibiotics in the environment: a review. Water Res. 47, 957e995. Monstein, H.J., Ostholm-Balkhed, A., Nilsson, M.V., Nilsson, M., Dornbusch, K., Nilsson, L.E., 2007. Multiplex PCR amplification assay for the detection of (SHV)S-bla, (TEM)-T-bla and (CTX)-C-bla-M genes in enterobacteriaceae. Apmis 115, 1400e1408. Oh, J., Salcedo, D.E., Medriano, C.A., Kim, S., 2014. Comparison of different disinfection processes in the effective removal of antibiotic-resistant bacteria and genes. J. Environ. Sci. China 26, 1238e1242. Okitsu, N., Kaieda, S., Yano, H., Nakano, R., Hosaka, Y., Okamoto, R., Kobayashi, T., Inoue, M., 2005. Characterization of ermB gene transposition by Tn1545 and Tn917 in macrolide-resistant Streptococcus pneumoniae isolates. J. Clin. Microbiol. 43, 168e173. Ouyang, W.Y., Huang, F.Y., Zhao, Y., Li, H., Su, J.Q., 2015. Increased levels of antibiotic resistance in urban stream of Jiulongjiang River, China. Appl. Microbiol. Biotechnol. 99, 5697e5707. Pruden, A., Pei, R.T., Storteboom, H., Carlson, K.H., 2006. Antibiotic resistance genes as emerging contaminants: studies in northern Colorado. Environ. Sci. Technol. 40, 7445e7450.

126

L. Xu et al. / Environmental Pollution 213 (2016) 119e126

Queenan, A.M., Jenkins, S., Bush, K., 2001. Cloning and biochemical characterization of FOX-5, an AmpC-type plasmid-encoded beta-lactamase from a New York City Klebsiella pneumoniae clinical isolate. Antimicrob. Agents Chemother. 45, 3189e3194. Roberts, M., 2003. Acquired tetracycline and/or macrolideelincosamidese streptogramin resistance in anaerobes. Anaerobe 9, 63e69. Salyers, A.A., Gupta, A., Wang, Y.P., 2004. Human intestinal bacteria as reservoirs for antibiotic resistance genes. Trends Microbiol. 12, 412e416. Shi, P., Jia, S., Zhang, X.X., Zhang, T., Cheng, S., Li, A., 2013. Metagenomic insights into chlorination effects on microbial antibiotic resistance in drinking water. Water Res. 47, 111e120. Singh, R., Schroeder, C.M., Meng, J.H., White, D.G., McDermott, P.F., Wagner, D.D., Yang, H.C., Simjee, S., DebRoy, C., Walker, R.D., Zhao, S.H., 2005. Identification of antimicrobial resistance and class 1 integrons in Shiga toxin-producing Escherichia coli recovered from humans and food animals. J. Antimicrob. Chemother. 56, 216e219. Su, H.C., Ying, G.G., Tao, R., Zhang, R.Q., Fogarty, L.R., Kolpin, D.W., 2011. Occurrence of antibiotic resistance and characterization of resistance genes and integrons in Enterobacteriaceae isolated from integrated fish farms in south China. J. Environ. Monit. 13, 3229e3236.

Wang, F.H., Qiao, M., Su, J.Q., Chen, Z., Zhou, X., Zhu, Y.G., 2014. High throughput profiling of antibiotic resistance genes in urban park soils with reclaimed water irrigation. Environ. Sci. Technol. 48, 9079e9085. Wellington, E.M.H., Boxall, A.B.A., Cross, P., Feil, E.J., Gaze, W.H., Hawkey, P.M., Johnson-Rollings, A.S., Jones, D.L., Lee, N.M., Otten, W., Thomas, C.M., Williams, A.P., 2013. The role of the natural environment in the emergence of antibiotic resistance in gram-negative bacteria. Lancet Infect. Dis. 13, 155e165. Xi, C., Zhang, Y., Marrs, C.F., Ye, W., Simon, C., Foxman, B., Nriagu, J., 2009. Prevalence of antibiotic resistance in drinking water treatment and distribution systems. Appl. Environ. Microbiol. 75, 5714e5718. Xu, J., Xu, Y., Wang, H., Guo, C., Qiu, H., He, Y., Zhang, Y., Li, X., Meng, W., 2015. Occurrence of antibiotics and antibiotic resistance genes in a sewage treatment plant and its effluent-receiving river. Chemosphere 119, 1379e1385. Zhang, X.X., Zhang, T., Fang, H.H., 2009. Antibiotic resistance genes in water environment. Appl. Microbiol. Biotechnol. 82, 397e414. Zhu, Y.G., Johnson, T.A., Su, J.Q., Qiao, M., Guo, G.X., Stedtfeld, R.D., Hashsham, S.A., Tiedje, J.M., 2013. Diverse and abundant antibiotic resistance genes in Chinese swine farms. Proc. Natl. Acad. Sci. U. S. A. 110, 3435e3440. Ziglam, H.M., Finch, R.G., 2001. Limitations of presently available glycopeptides in the treatment of Gram-positive infection. Clin. Microbiol. Infect. 7, 53e65.