Bacterial Community Shift Drives Antibiotic ... - ACS Publications

Article pubs.acs.org/est

Bacterial Community Shift Drives Antibiotic Resistance Promotion during Drinking Water Chlorination Shuyu Jia,† Peng Shi,*,† Qing Hu,† Bing Li,‡ Tong Zhang,‡ and Xu-Xiang Zhang*,† †

State Key Laboratory of Pollution Control and Resource Reuse, Environmental Health Research Center, School of the Environment, Nanjing University, Nanjing 210023, China ‡ Environmental Biotechnology Lab, Department of Civil Engineering, The University of Hong Kong, Hong Kong SAR, China S Supporting Information *

ABSTRACT: For comprehensive insights into the effects of chlorination, a widely used disinfection technology, on bacterial community and antibiotic resistome in drinking water, this study applied high-throughput sequencing and metagenomic approaches to investigate the changing patterns of antibiotic resistance genes (ARGs) and bacterial community in a drinking water treatment and distribution system. At genus level, chlorination could effectively remove Methylophilus, Methylotenera, Limnobacter, and Polynucleobacter, while increase the relative abundance of Pseudomonas, Acidovorax, Sphingomonas, Pleomonas, and Undibacterium in the drinking water. A total of 151 ARGs within 15 types were detectable in the drinking water, and chlorination evidently increased their total relative abundance while reduced their diversity in the opportunistic bacteria (p < 0.05). Residual chlorine was identified as the key contributing factor driving the bacterial community shift and resistome alteration. As the dominant persistent ARGs in the treatment and distribution system, multidrug resistance genes (mainly encoding resistance-nodulation-cell division transportation system) and bacitracin resistance gene bacA were mainly carried by chlorine-resistant bacteria Pseudomonas and Acidovorax, which mainly contributed to the ARGs abundance increase. The strong correlation between bacterial community shift and antibiotic resistome alteration observed in this study may shed new light on the mechanism behind the chlorination effects on antibiotic resistance.

■

INTRODUCTION Misuse and overuse of antibiotics during the past decades have contributed to continuous emissions of antibiotics into the environment1−3 and emergence of antibiotic resistant bacteria (ARB).4 Recent studies have shown that a variety of antibiotic resistance genes (ARGs) occur in drinking water around the world.5,6 Escalation of antibiotic resistance bestowed by the ARGs can reduce the disease susceptibility,7 which has been an intractable environmental health issue.8 Recently, it has been reported that multidrug resistant NDM-1 and enterohemorrhagic Escherichia coli emerge in drinking water in India,9,10 posing great threat to the health of local residents. As a disinfection method, chlorination is widely used in China and other countries for removal of microorganisms in drinking water.11 Recently, microbial safety of disinfection methods has received growing research concerns.12−14 Our previous study has revealed that chlorination can elevate the total relative abundance of ARGs in opportunistic bacteria, and individual ARGs show highly variable responses to the chlorine stress.15 However, universal changing patterns of ARGs during chlorination have not been fully explored to date. Usually, excessive replication of mobile genetic elements (MGEs) may take place in bacterial cells under external stress,16 which is © XXXX American Chemical Society

considered as the major factor that drives the alteration of resistome in drinking water,17 ocean sediments18 and swine manure.19 Recent studies have also demonstrated that bacterial community shift plays a more important role than MGEs in shaping soil and sludge resistome.20,21 It is well-known that chlorination can alter bacterial community structure in drinking water,22,23 but the contribution of the bacterial community shift to the resistome alteration still remains unclear. Technically, polymerase chain reaction (PCR) and quantitative real time PCR (q-PCR) have been widely applied to detect environmental ARGs,24 but limited primers are available for ARGs, and bias may exist during the amplification for environmental samples.25 Application of hybridization-based techniques in environmental ARGs detection often encounters difficult technical problems, such as low detection limit and complex sample pretreatment.24 Combined use of highthroughput sequencing (HTS) and bioinformatics analysis exactly compensates for the technical deficiencies due to the Received: July 21, 2015 Revised: September 21, 2015 Accepted: September 23, 2015

A

DOI: 10.1021/acs.est.5b03521 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology

DNA extracted from each sample was subjected to Illumina HTS by Novogene Bioinformatics Technology Co., Ltd. (Beijing, China) using Illumina Hiseq 2000 (Illuminna Inc., San Diego, CA). The sequencing strategy of Index 101 PE (Paired End sequencing, 101-bp reads and 8-bp index sequence) was applied to generate about 6 G clean reads (FASTQ format) for each sample. An online tool kit (Galaxy, https://usegalaxy.org/) was utilized to filter low quality reads to ensure (1) less than three ambiguous nucleotides in one read; (2) more than 90% bases with quality score greater than 30; and (3) no sequencing artifacts.32 BLAST tool was applied to annotate sequencing reads against an offline database of ARGs previously established,15 to determine ARGs abundance and their potential hosts following Kristiansson et al.26 MGEs (integrons, plasmids, and insertion sequences (ISs)) were also quantified using metagenomic methods according to our previous study.15 The metagenomic data have been deposited in NCBI Sequence Read Archive under accession number SRA178672. Pyrosequencing of 16S rRNA Gene and Bioinformatics Analysis. To investigate the bacterial community structure of the drinking water, 16S rRNA gene was amplified from each metagenomic DNA sample with a set of primers (Table S2) targeting the hypervariable V1−V3 region (526 bp), and a 10-nucleotide barcode was added to the 5′ end of each primer set to differentiate samples during amplification. PCR reactions were performed in a 20 μL mixture containing 4 μL of 5 × FastPfu Buffer, 2 μL of 2.5 mM dNTPs, 0.8 μL of each primer (5 μM), 0.4 μL of FastPfu Polymerase and 10 ng of template DNA. The PCR amplification condition was: degeneration at 95 °C for 2 min, followed by 25 cycles at 95 °C for 30 s, 55 °C for 30 s, and 72 °C for 30 s, and a final extension at 72 °C for 5 min. After purification by the AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, Union City, U.S.) and quantification by QuantiFluor-ST (Promega, Fitchburg, U.S.), appropriate amounts of purified amplicon products of each sample were mixed and sent to the Shanghai Majorbio Biopharm Biotechnology Co., Ltd. (Shanghai, China) for pyrosequencing on Roche 454 FLX Titanium platform (Roche 454 Life Sciences, Branford, U.S.). The generated sequences were processed and analyzed using both Ribosomal Database Project (RDP, http://rdp.cme.msu.edu) and Mothur (http://www.mothur.org) as described in the Supporting Information (Text S1) following Wang et al.33 For each sample, the number of high-quality pyrosequencing reads was normalized to 8542 for operational taxonomic units (OTU) picking, diversity indices calculation and microbial community annotation. Statistical Analysis. The relative abundance of ARGs was determined by using the unit of “ppm”, namely one hit of the ARG in one million sequencing reads. The percentage of ARGs abundance (%) was calculated as the proportion of the annotated reads of one ARG in the total reads of all ARGs. Persistent ARGs indicate the common ones shared by all the samples. Canonical correspondence analysis (CCA) and Pearson correlation analysis were conducted to reveal the correlations between the percentages of the genera and the tested chemical parameters including residual chlorine. Correlation between bacterial community and persistent ARGs was examined by Mantel test and redundancy analysis (RDA), and bacterial genera were considered as the environmental factors affecting the ARGs diversity and abundance in RDA.34,35 Variation partitioning analysis (VPA) was further

capability of broad-spectrum scan and accurate quantification of ARGs in complicated environmental samples, which has been successfully applied to screen ARGs in river sediments,26 activated sludge27 and gull feces.28 In this study, HTS-based metagenomic approaches were applied to investigate the occurrence and fate of ARGs, bacterial community shift and their correlations in a large-scale waterworks. Environmental variables were also measured to determine their relative contributions to the ARGs variation. The results may help to understand chlorination effects on microbial antibiotic resistance in drinking water and to reveal the underlying molecular ecological mechanisms.

■

MATERIALS AND METHODS Sample Collection and DNA Extraction. Source water (SW), effluent of sedimentation tank (ES), filtered water (FW) and chlorine-disinfected water (DW) were collected from the water pumping station I (2 L), horizontal-flow sedimentation tank (4 L), ordinary rapid filtration tank (25 L), and clear water tank (1000 L) of the BHK Waterworks of Nanjing (China), respectively (Figure S1). Tap water (800 L each) was collected along the distribution pipeline at three locations, 10 (TWA), 30 (TWB) and 40 (TWC) kilometers away from the waterworks (A secondary chlorine station is located between TWA and TWB) (Figure S1). Detailed operational parameters of the waterworks and distribution system are summarized in Table S1. To minimize the temporal variation, three replicate water samples were simultaneously collected from each location (21 samples in total) in June 2013. The water sampling and pretreatment methods were previously described.15 DNA extraction was performed using the FastDNA SPIN Kit for Soil (MP Biomedicals, CA), and the DNA concentration and purity were measured by microspectrophotometry (NanoDropND-2000, NanoDrop Technologies, Willmington, DE). Chemical Analysis. Concentrations of 21 antibiotics, including ampicillin, oxacillin, penicillin G, ceftazidime, cefazolin, cefotaxime, cefalexin, sulfamethoxazole, sulfadiazine, sulfamethazine, norfloxacin, ciprofloxacin, ofloxacin, tetracycline, oxytetracycline, chlortetracycline, roxithromycin, erythromycin, vancomycin, trimethoprim, and chloramphenicol in the samples, were determined using ultra performance liquid chromatography with tandem mass spectrometry. In brief, 250 mL of each sample was concentrated to 1 mL of 5% aqueous acetonitrile solution by optimized parameters of solid phase extraction. Isotopically labeled 13C3-caffein solution was used as surrogate standard to control the recovery efficiency and machine performance. Chromatographic separation of the antibiotics was performed in the Acquity Ultra Performance LC system (Waters, Milford, MA) equipped with a BEH C18 column (1.7 μm, 50 × 2.1 mm). Acquity TQ Detector, a tandem quadrupole mass spectrometer (Waters, Milford, U.S.) equipped with the Z-spray electrospray ionization (ESI) source, was applied for antibiotics detection. The detailed operation parameters and pretreatment methods have been previously described.29 Heavy metals in the samples were measured by using inductively coupled plasma-mass spectrometry (ICP-MS, Agilent 7500, Agilent).30 Total organic carbon (TOC), total nitrogen (TN), total phosphorus (TP) and total residual chlorine were analyzed following the standard methods.31 All the chemical parameters were measured in triplicate for each sample. HTS and Bioinformatics Analysis. For comprehensive investigation of ARGs in drinking water, the metagenomic B


Article

Environmental Science & Technology conducted to determine the contributions of bacterial communities and MGEs to the variations of persistent ARGs. Principal coordinate analysis (PCoA) were performed to evaluate the difference of ARGs profiles among the samples based on the Bray−Curtis distance of ARGs relative abundance. Adonis test was conducted to determine the significance of the difference in ARGs profiles or bacterial community structure in the different samples. All the statistical analyses were conducted using R software (version 3.0.3) with “vegan” and “psych” packages. One-way analysis of variance (ANOVA) was carried out to assess the variations by using SPSS V20.0 (IBM) and p < 0.05 was considered statistically significant.

■

RESULTS Changes of Chemical Pollutants. Chemical analysis showed that the drinking water treatment processes could effectively remove the detectable antibiotics, TOC, TN, TP, and heavy metals, but the pollutants concentrations were relatively stable along the distribution system (Table S3). Residual chlorine level increased after chlorination, but then decreased along the transportation pipeline (Table S3). Among the 21 antibiotics tested in this study, sulfamethoxazole, sulfadiazine, sulfamethazine, and erythromycin were detectable in SW, and coagulation and sedimentation significantly increased the erythromycin concentration, and showed no effects on the removal of other antibiotics (Figure S2, p < 0.05). Sand filtration was found to remove 49.46% of sulfamethoxazole and 42.94% of sulfadiazine in the drinking water (Figure S2, p < 0.05). Moreover, the removal efficiencies of the detectable antibiotics ranged from 77.38% to 100% during chlorination. Linkages between Bacterial Community and Chemical Parameters. Based on the pyrosequencing of 16S rRNA gene, phylogenetic classification demonstrated that bacterial community shifted along the treatment and distribution system at both phylum (Figure S3) and genus (Figure 1) levels. Proteobacteria phylum dominated in each sample and its abundance increased after chlorination, while Actinobacteria and Bacteroidetes seemed sensitive to chlorine (Figure S3). Bacteroidetes could gain regrowth after chlorine consumption in the transportation pipeline, but Actinobacteria could not (Figure S3). At genus level, Methylophilus, Methylotenera, Limnobacter, and Polynucleobacter dominated in the samples before chlorination, but almost disappeared after chlorination (Figure 1). However, the relative abundance of bacterial genera, including Pseudomonas, Acidovorax, Pleomonas, Sphingomonas, and Undibacterium within Proteobacteria phylum, seemed to greatly increase after chlorination (Figure 1). Furthermore, bacterial diversity decreased after chlorination and increased along the transportation pipeline (Figure S4), and the overall patterns of bacterial community were significantly altered by chlorination (Adonis test, p < 0.01). CCA revealed bacterial community shift after chlorination, which was significantly correlated with TOC, sulfamethoxazole, sulfadiazine, sulfamethazine, erythromycin and residual chlorine in the drinking water (p < 0.05 each, Figure 2). TOC, sulfamethoxazole, sulfadiazine, sulfamethazine, and erythromycin were the major factors driving bacterial community shift in the drinking water before chlorination (SW, ES, and FW), but residual chlorine played the most important role in affecting bacterial communities in the transportation system (Figure 2). Pearson correlation analysis also indicated that the relative abundance of Pseudomonas, Acidovorax, and Pleomonas was significantly

Figure 1. Relative abundance of each taxonomic genus (>1% in any sample) in source water (SW), effluent of sedimentation tank (ES), filtered water (FW), chlorine-disinfected water (DW), tap water A (TWA), tap water B (TWB) and tap water C (TWC). The relative abundance of each bacterial genus means was calculated as the percentages of the number of the reads assigned to one genus vs. the total number of pyrosequencing reads. The 35 genera include all the 23 persistent genera in each sample, which are marked as *. The color intensity in each panel shows the percentage of each genus in one sample.

Figure 2. Quantitative correlation between the genera and chemical parameters in all the samples revealed by canonical correspondence analysis (CCA). The quantity of each bacterial genus (relative abundance) indicates the proportion of the number of the reads assigned to one genus vs. the total number of pyrosequencing reads. Arrows represent the positive correlations between the bacterial communities and the chemical parameters (p < 0.05).

C


Article


Figure 3. Occurrence and fate of the antibiotic resistance genes (ARGs) in source water (SW), effluent of sedimentation tank (ES), filtered water (FW), chlorine-disinfected water (DW), tap water A (TWA), tap water B (TWB) and tap water C (TWC) determined by high-throughput sequencing and metagenomic analysis.

Table 1. Relative Abundance of Each ARGs Type in Source Water (SW), Effluent of Sedimentation Tank (ES), Filtered Water (FW), Chlorine-Disinfected (DW), Tap Water A (TWA), Tap Water B (TWB) and Tap Water C (TWC) Revealed by HighThroughput Sequencing and Metagenomic Analysis relative abundance (ppm) ARGs type multidrug bacitracin sulfonamide aminoglycoside M-L-S β-lactam tetracycline chloramphenicol fosmidomycin trimethoprim polymyxin quinolone fosfomycin vancomycin others a

SW 13.18 7.97 4.82 2.94 0.42 1.40 0.71 0.41 0.33 0.11 0.06 0.05 0.00 0.00 0.02

± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

ES 3.42 3.70 2.88 0.59 0.26 0.78 0.43 0.44 0.18 0.07 0.03 0.05 0.00 0.00 0.03

14.77 11.14 7.33 3.42 1.11 1.16 0.41 0.38 0.51 0.23 0.00 0.00 0.00 0.00 0.00

± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

FW 4.11 1.70 3.78 1.50 0.94 0.49 0.30 0.32 0.11 0.16 0.00 0.00 0.00 0.00 0.00

16.97 7.57 4.11 3.08 0.81 2.10 2.10 0.24 0.74 0.15 0.02 0.02 0.02 0.02 0.05

± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

DW 3.18 4.60 1.39 2.54 0.67 2.90 0.23 0.34 0.22 0.22 0.03 0.03 0.03 0.03 0.05

71.10 15.45 2.63 1.04 0.11 1.43 0.47 0.59 0.35 0.00 0.35 0.05 0.68 0.00 0.06

± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

4.81a 2.62a 0.05a 1.57 0.07 0.26 0.30a 0.65 0.03 0.00a 0.09a 0.05 0.18a 0.00 0.07

TWA 15.77 6.67 2.90 0.65 0.03 0.27 0.48 0.29 0.14 0.00 0.08 0.11 1.92 0.00 0.00

± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

3.33a 0.59a 0.54 0.18 0.05 0.05a 0.26 0.21 0.05 0.00 0.03a 0.05 0.46a 0.00 0.00

TWB 28.99 4.92 6.44 0.74 0.00 0.68 0.09 0.32 0.27 0.00 0.15 0.02 2.12 0.00 0.00

± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

10.90a 1.92 5.61 0.17 0.00 0.90 0.08a 0.40 0.12 0.00 0.09 0.03 1.08 0.00 0.00

TWC 16.65 22.46 4.28 0.53 0.00 0.63 0.15 0.63 1.34 0.00 0.12 0.02 0.62 0.00 0.02

± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

5.00a 8.12a 2.69 0.56 0.00 0.28 0.07 0.14 0.21a 0.00 0.06 0.03 0.25 0.00 0.03

Significantly different from its front sample (p < 0.05).

correlated with residual chlorine levels (R2 > 0.5, p < 0.05, Table S4). Diversity and Abundance of ARGs. HTS-based metagenomic analysis revealed that a total of 151 ARGs were identified in all the water samples, mainly grouped into 15 types (Figure 3 and Table S5). Multidrug, bacitracin and sulfonamide resistance genes were the dominant types of ARGs in the drinking water, which accounted for 75.44−94.60% of the total ARGs abundance (Figure S5). The abundance of the genes

(mainly mex efflux genes) within resistance-nodulation-cell division (RND) transportation system accounted for 99.33− 100% of the total abundance of multidrug resistance genes (Figure S6). Inverse Simpson and Shannon indices calculations consistently indicated that coagulation, sedimentation and filtration posed no obvious effects on the ARGs diversity, but chlorination could significantly reduce the diversity (Figure S7) since the ARGs types decreased along the transportation pipeline (Figure S8A). D


Article


Figure 4. Relative abundance of potential hosts for ARGs within RND transportation system, undecaprenyl pyrophosphate phosphatase, sulfonamide-resistant dihydropteroate synthase, potassium antiporter system, aminoglycoside N-acetyltransferase, aminoglycoside O-nucleotidylyltransferase, and aminoglycoside O-phosphotransferase in all the samples revealed by high-throughput based metagenomic analysis. Relative abundance (%) of the hosts indicates the proportion of the reads assigned to each type of ARGs carried by the corresponding genus in all the annotated reads of the type of ARGs. Genera with percentage of over 10% in any sample are shown. Color intensity in each panel shows the percentages of the genera in each sample.

Figure 5. Redundancy analysis (RDA) of the quantitative correlation between genera (>1% in any sample) and persistent ARGs (A) and variation partitioning analysis (VPA) differentiating effects of bacterial community (BC) and mobile genetic elements (MGEs) on the resistome alteration (B). The VPA was conducted based on percentages of major phyla and relative abundance of integrons, plasmids, and insertion sequences. Monte Carlo permutation test and variance inflation factor were performed to remove the redundant variables in RDA modeling. Arrows represent the six genera (Pseudomonas (G1), Sphingomonas (G2), Gemmata (G3), Polynucleobacter (G4), Limnohabitans (G5), and Methylotenera (G6)) positively correlated with persistent ARGs distribution (p < 0.02). Green and yellow nodes represent the indicated variable partitioned into the relative effects of BC and MGEs, respectively. The blue node represents the joint effect of BC and MGEs. The square represents the portion unexplained by the two selected variable groups.

The total abundance of the detectable ARGs showed no significant differences among the samples of SW, ES and FW (Figure S8B, p > 0.05), but was greatly increased in drinking water after chlorination, and transportation effectively reduced the ARGs abundance (Figure S8B, p < 0.05). The abundance of multidrug resistance genes in SW was 13.18 ± 3.42 ppm, and was significantly increased to 71.10 ± 4.81 ppm after chlorination but was reduced to 15.77 ± 3.33 ppm in TWA (Table 1, p < 0.05). The stimulation effect was confirmed by the results of TWB and TWC after secondary chlorination (Table 1). Additionally, the relative abundance of bacitracin, polymyxin and fosfomycin resistance genes was also significantly increased after chlorination (Table 1, p < 0.05). On the

contrary, sulfonamide, tetracycline, and trimethoprim resistance genes had lower abundance after chlorination (Table 1, p < 0.05). PCoA of the resistome patterns revealed a significant shift (Adonis test, p < 0.01) after chlorination, and the drinking water samples were grouped into three clusters: cluster 1 including SW, FW and EW, cluster 2 including DW, and cluster 3 including TWA, TWB, and TWC, revealing that chlorination greatly altered the ARGs abundance and diversity (Figure S9). In addition, the relative abundance of MGEs including integrons, plasmids and ISs was significantly increased after chlorination with the highest abundance of 84.23 ± 10.20 ppm for integrons, 3869.65 ± 1246.01 ppm for plasmids and 182.46 ± 87.38 ppm for ISs in DW (Figure S10). E


Article


can affect functional gene distribution by altering bacterial community.21,32 This study indicated that residual chlorine was the key contributing factor driving the bacterial community shift in the drinking water. The difference in chlorine sensitivity among bacterial populations in drinking water may be the main reason for the succession dynamics and diversification of microbial community.39 Based on a comprehensive investigation of ARGs with metagenomic analyses, this study revealed that chlorination significantly altered the antibiotic resistome in drinking water by increasing the total relative abundance and decreasing diversity of the ARGs in the opportunistic bacteria, which has been confirmed by using molecular methods in previous studies.6,15 Similarly, it has been reported that environmental stresses of antibiotics, heavy metals, heat shock, oligotrophic environments, or oxidative stresses can obviously accelerate ARGs replication in microbial population.40,41 We found that multidrug resistance genes dominated in the drinking water and their relative abundance was greatly elevated after chlorination, and the ARGs encoding RND transportation system were mainly responsible for the enhancement of total ARGs abundance after chlorination. RND transportation system is capable of binding multiple structurally unrelated compounds to confer broad resistance phenotypes, and crossresistance between disinfectants and antibiotics may occur if the two antibacterial agents share the same approaching pathway or mechanism of action.42 Although no cross-resistance between chlorine and antibiotics has been documented to date, RND transportation system is known to have broad substrate specificity and play key roles in accommodation of various biocides including silver, chlorhexidine, triclosan, and benzalkonium chloride.43,44 Especially, this study revealed that multiple mex efflux genes, the main functional genes of RND transportation system, were significantly enriched by chlorination. The mex genes are members of MexAB−OprM, MexCD− OprJ, MexEF−OprN, and MexXY−OprM efflux systems, which can confer resistance to nearly all the antibiotics,45 and can also pump out acriflavine, ethidium bromide and sodium dodecyl sulfate.43 BacA gene that encodes bacitracin resistance was another dominant persistent ARGs in the drinking water, and the increase of its abundance by chlorination also greatly contributed to the increase of the total ARGs abundance. A previous survey has reported that 52% of swine operations use bacitracin as growth promoter and prophylaxis,46 so it is not surprising that the resistance gene is present in drinking water at high levels. Chlorination can kill bacteria by destroying cell wall, but bacA gene product is essential for the biosynthesis of peptidoglycan and other cell wall components,47 which may result in the survival of the bacteria harboring bacA under chlorine stress. Residual chlorine was found to be the most important factor affecting the antibiotic resistome in drinking water treatment and distribution system. However, cross-resistance to chlorine and antibiotics have not been fully understood due to the complexity of microbial community in natural environments, incapability of cultivating most microorganisms, and variation of bacteria lineages in drinking water treatment systems.34 In order to reveal the potential molecular ecological mechanisms underlying its stimulation effects on antibiotic resistance, this study applied different statistical methods to determine the correlations between the core bacteria and persistent ARGs and found that Pseudomonas, Acidovorax, Sphingomonas, and Gemmata played crucial roles in the dissemination and variation

Among all the detectable ARGs, 21 persistent ones within seven types were commonly shared by all the water samples (Figure S11). Although the diversity of persistent ARGs was low (most were multidrug resistance genes) (Figure S12A), their abundance accounted for 81.80−95.00% of the total ARGs abundance in the drinking water (Figure S12B). Among the persistent ARGs, bacitracin resistance gene bacA (encoding undecaprenyl pyrophosphate phosphatase) had the highest abundance in SW, followed by sulfonamide resistance gene sulI and multidrug resistance gene mexF. Among the seven types, bacA gene and RND-related genes had increased abundance in drinking water after chlorination (p < 0.05), while other types of ARGs were effectively removed (Table S6). Correlation between Bacterial Community and Persistent ARGs. Potential host analysis showed that Pseudomonas was the main potential host of the persistent ARGs encoding RND transportation system, and had increased abundance after chlorination (Figure 4). Most copies of bacA gene were located in the cells of Polaromonas and Polynucleobacter genera in prechlorination samples (SW, ES and FW), but the gene was mainly carried by Pseudomonas and Acidovorax after chlorination (Figure 4). Salmonella, the major host of sulfonamide resistance genes in the drinking water, also had increased abundance after chlorination (Figure 4). However, Pseudomonas and Escherichia carrying most of aminoglycoside resistance genes seemed sensitive to chlorine in the drinking water (Figure 4). RDA of the bacterial communities and persistent ARGs showed that the drinking water samples could be grouped into three clusters: cluster A including SW, ES, and FW, cluster B including DW and TWB, and cluster C including TWA and TWC (Figure 5A). Among the identified genera, six genera were found significantly correlated with the ARGs distribution in drinking water (p < 0.02) (Figure 5A). Within cluster A, Polynucleobacter, Limnohabitans, and Methylotenera, dominating in the prechlorination samples, showed quantitatively positive correlation with sulfonamide and aminoglycoside resistance genes (sulI, sulII, aac(6′)-Ib, ant(2″)-Ia, ant(3″)-Ia, aph(3′)-Ia, and rosB). Within cluster B, the abundance of Pseudomonas, Sphingomonas, and Gemmata were positively correlated with the copies of bacA gene and the ARGs encoding RND transportation system (mexF, mexW, mexB, mexI, mexD, mexY, macB, acrB, ceoB, smeB, smeE, adeB, amrB, and mdtF) (Figure 5A). VPA showed that bacterial community shift contributed to 57.22% of the resistome variation in the drinking water samples, which was much higher than MGEs alterations (16.63%) and their joint effects (5.31%) (Figure 5B). In addition, Mantel test showed that bacterial community were significantly correlated with ARGs profiles based on Bray− Curtis distance (r = 0.54, p < 0.001, permutations = 9999).

■

DISCUSSION This study showed that drinking water contained low concentrations of several antibiotics, and coagulation and filtration posed no obvious effects on removal of the antibiotics because of their high water solubility.36 However, chlorination removed most of the antibiotics, which may result from the strong oxidizing property of chlorine.37 Among the detectable ARGs in the samples, sulfonamide, and tetracycline resistance genes have gained much attention due to their widespread in environments.38 Chlorination was found to effectively remove the two types of ARGs, which agrees with previous studies.6,15 Environmental variables, such as temperature and heavy metals, F


Article

Environmental Science & Technology of bacA gene and the ARGs encoding RND transportation system in drinking water. This result was confirmed by pyrosequencing analysis of 16S rRNA genes demonstrating that chlorination greatly increased the relative abundance of Pseudomonas and Acidovorax (most important hosts of bacA and RND-related ARGs) in the drinking water. Previous studies have demonstrated that Pseudomonas, Acidovorax, and Sphingomonas are the core microbiota in chlorinated water.15,22,36 Each species within Pseudomonas is known to encode RND transportation system,48 and also serves as hosts of bacA gene.49 Genome sequencing has shown that Acidovorax sp. JS42 (GenBank accession number: CP000539.1, unpublished results) and MR-S750 contain multiple ARGs of RND family (e.g., mexB), and the former also carried bacA gene (UniPort accession number: A1W560). Sphingomonas isolates from drinking water exhibited various antibiotic resistance patterns including β-lactams, ciprofloxacin, and cotrimoxazole, but the genotypes have rarely been analyzed.51 This study indicated that the bacterial community shift, rather than MGEs reproduction or transfer, is the major driver shaping the antibiotic resistome in drinking water. Horizontal gene transfer via MGEs is often regarded as the underlying mechanism responsible for the resistome formation under different ecological niches.52,53 It has been reported that lowdose chlorine may promote the frequency of conjugative transfer via alteration of cell permeability, but the transfer can hardly occur when the bacterial concentration is below 104 CFU/mL.14 Filtered water usually contains less than 103 CFU/ mL of bacteria in waterworks,15,54 and bacA and RND-related ARGs are usually located on chromosome DNA,47,55 so conjugative transfer may contribute little to the ARGs abundance increase observed in this study. Therefore, this study revealed that exposure to low-dose chlorine resulted in the increase of Pseudomonas and Acidovorax abundance, and subsequently increased the abundance of bacA and RNDrelated ARGs in drinking water. Similarly, previous studies indicated that benzalkonium chloride disinfectants56 and mutagenic disinfection byproducts57 can promote the antibiotic resistance of human pathogen Pseudomonas aeruginosa, and RND transportation system plays a key role in conferring resistance to the two chemicals. In conclusion, this study reveals that chlorination can alter antibiotic resistome in drinking water via bacterial community shift. Enrichment of chlorine-resistant bacteria (e.g., Pseudomonas and Acidovorax) increases the abundance of bacA gene and RND-related ARGs, which mainly contributes to the increase of total relative abundance of ARGs in opportunistic bacteria after chlorination. This is the first study providing the solid evidence for the correlation between antibiotic resistome and bacterial community in a real waterworks, which may help to fully understand the promotion mechanisms of antibiotic resistance by chlorination. To reduce the potential risk, more efforts should be devoted to controlling the antibiotic resistance promotion induced by chlorination.

■

■

parameters of water samples; Table S4, correlations between the genera and residual chlorine; Table S5, relative abundance of the ARGs and their corresponding types; Table S6, relative abundance of persistent ARGs and their corresponding types and resistance mechanisms; Figure S1, the process flowchart and sampling locations; Figure S2, concentrations of four detectable antibiotics; Figure S3, bacterial community compositions at phylum level; Figure S4, OTU number and Chao1 index of bacterial community in drinking water; Figure S5, percentages of ARGs types; Figure S6, percentages of genes within RND transportation system; Figure S7, Shannon and Inverse Simpson index of ARGs; Figure S8, numbers of ARGs and their types; Figure S9, principal coordinate analysis of the ARGs; Figure S10, relative abundance of MGEs; Figure S11, occurrence and fate of persistent ARGs; Figure S12, percentages of numbers and abundance for persistent ARGs and nonpersistent ARGs (PDF)

AUTHOR INFORMATION

Corresponding Authors

*(S.P.) Phone: +86-25-89680363; fax: +86-25-89680363; email: [email protected]. *(Z.X.-X.) E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This study was financially supported by the National Natural Science Foundation of China (51290282 and 51278240) and the Fundamental Research Funds for the Central Universities. We also thank the High Performance Computing Center (HPCC) of Nanjing University for the help of computation.

■

REFERENCES

(1) López-Serna, R.; Pérez, S.; Ginebreda, A.; Petrović, M.; Barceló, D. Fully automated determination of 74 pharmaceuticals in environmental and waste waters by online solid phase extraction−liquid chromatography−electrospray−tandem mass spectrometry. Talanta 2010, 83 (2), 410−424. (2) Luo, Y.; Xu, L.; Rysz, M.; Wang, Y. Q.; Zhang, H.; Alvarez, P. J. J. Occurrence and transport of tetracycline, sulfonamide, quinolone, and macrolide antibiotics in the Haihe River Basin, China. Environ. Sci. Technol. 2011, 45 (5), 1827−1833. (3) Kleywegt, S.; Pileggi, V.; Yang, P.; Hao, C. Y.; Zhao, X. M.; Rocks, C.; Thach, S.; Cheung, P.; Whitehead, B. Pharmaceuticals, hormones and bisphenol A in untreated source and finished drinking water in Ontario, Canada  occurrence and treatment efficiency. Sci. Total Environ. 2011, 409 (8), 1481−1488. (4) Blaser, M. Antibiotic overuse: stop the killing of beneficial bacteria. Nature 2011, 476 (7361), 393−394. (5) Li, B.; Yang, Y.; Ma, L. P.; Ju, F.; Guo, F.; Tiedje, J. M.; Zhang, T. Metagenomic and network analysis reveal wide distribution and co− occurrence of environmental antibiotic resistance genes. ISME J. 2015, 1−13. (6) Xi, C. W.; Zhang, Y. L.; Marrs, C. F.; Ye, W.; Simon, C.; Foxman, B.; Nriagu, J. Prevalence of antibiotic resistance in drinking water treatment and distribution systems. Appl. Environ. Microbiol. 2009, 75 (17), 5714−5718. (7) Cosgrove, S. E.; Carroll, K. C.; Perl, T. M. Staphylococcus aureus with reduced susceptibility to vancomycin. Clin. Infect. Dis. 2004, 39 (4), 539−545. (8) Schwartz, T.; Kohnen, W.; Jansen, B.; Obst, U. Detection of antibiotic−resistant bacteria and their resistance genes in wastewater,

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.5b03521. Text S1, metagenomic analysis of pyrosequencing results; Table S1, operational parameters of the BHK Waterworks; Table S2, PCR primers; Table S3, chemical G


Article

Environmental Science & Technology surface water, and drinking water biofilms. FEMS Microbiol. Ecol. 2003, 43 (3), 325−335. (9) Walsh, T. R.; Weeks, J.; Livermore, D. M.; Toleman, M. A. Dissemination of NDM-1 positive bacteria in the New Delhi environment and its implications for human health: an environmental point prevalence study. Lancet Infect. Dis. 2011, 11 (5), 355−362. (10) Ram, S.; Vajpayee, P.; Shanker, R. Contamination of potable water distribution systems by multiantimicrobial−resistant enterohemorrhagic Escherichia coli. Environ. Health Perspect. 2008, 116 (4), 448−452. (11) McGuire, M. J. The Chlorine Revolution: Water Disinfection and the Fight to Save Lives; American Water Works Association: Denver, CO, 2013. (12) Zhang, S. H.; Ye, C. S.; Lin, H. R.; Lv, L.; Yu, X. UV disinfection induces a VBNC state in Escherichia coli and Pseudomonas aeruginosa. Environ. Sci. Technol. 2015, 49 (3), 1721−1728. (13) Turetgen, I. Induction of Viable but Nonculturable (VBNC) state and the effect of multiple subculturing on the survival of Legionella pneumophila strains in the presence of monochloramine. Ann. Microbiol. 2008, 58 (1), 153−156. (14) Guo, M. T.; Yuan, Q. B.; Yang, J. Distinguishing effects of ultraviolet exposure and chlorination on the horizontal transfer of antibiotic resistance genes in municipal wastewater. Environ. Sci. Technol. 2015, 49 (9), 5771−5778. (15) Shi, P.; Jia, S. Y.; Zhang, X.-X.; Zhang, T.; Cheng, S. P.; Li, A. M. Metagenomic insights into chlorination effects on microbial antibiotic resistance in drinking water. Water Res. 2013, 47 (1), 111−120. (16) Wegrzyn, G.; Wegrzyn, A. Stress responses and replication of plasmids in bacterial cells. Microb. Cell Fact. 2002, 1 (1), 2−11. (17) Chao, Y. Q.; Ma, L. P.; Yang, Y.; Ju, F.; Zhang, X.-X.; Wu, W. M.; Zhang, T. Metagenomic analysis reveals significant changes of microbial compositions and protective functions during drinking water treatment. Sci. Rep. 2013, 3, 3550. (18) Chen, B. W.; Yang, Y.; Liang, X. M.; Yu, K.; Zhang, T.; Li, X. D. Metagenomic profiles of antibiotic resistance genes (ARGs) between human impacted estuary and deep ocean sediments. Environ. Sci. Technol. 2013, 47 (22), 12753−12760. (19) Y, G.; Johnson, T. A.; Su, J. Q.; Qiao, M.; Guo, G. X.; Stedtfeld, R. D.; Hashsham, S. A.; Tiedje, J. M. Diverse and abundant antibiotic resistance genes in Chinese swine farms. Proc. Natl. Acad. Sci. U. S. A. 2013, 110 (9), 3435−3440. (20) Forsberg, K. J.; Patel, S.; Gibson, M. K.; Lauber, C. L.; Knight, R.; Fierer, N.; Dantas, G. Bacterial phylogeny structures soil resistomes across habitats. Nature 2014, 509 (7502), 612−616. (21) Su, J. Q.; Wei, B.; Ou-Yang, W. Y.; Huang, F. Y.; Zhao, Y.; Xu, H. J.; Zhu, Y. G. Antibiotic resistome and its association with bacterial communities during sewage sludge composting. Environ. Sci. Technol. 2015, 49 (12), 7356−7363. (22) Hwang, C.; Ling, F. Q.; Andersen, G. L.; LeChevallier, M. W.; Liu, W. T. Microbial community dynamics of an urban drinking water distribution system subjected to phases of chloramination and chlorination treatments. Appl. Environ. Microbiol. 2012, 78 (22), 7856−7865. (23) Lührig, K.; Canbäck, B.; Paul, C. J.; Johansson, T.; Persson, K. M.; Rådström, P. Bacterial community analysis of drinking water biofilms in southern Sweden. Microbes Environ. 2015, 30 (1), 99−107. (24) Zhang, X.-X.; Zhang, T.; Fang, H. H. P. Antibiotic resistance genes in water environment. Appl. Microbiol. Biotechnol. 2009, 82 (3), 397−414. (25) Volkmann, H.; Schwartz, T.; Kirchen, S.; Stofer, C.; Obst, U. Evaluation of inhibition and cross−reaction effects on real−time PCR applied to the total DNA of wastewater samples for the quantification of bacterial antibiotic resistance genes and taxon−specific targets. Mol. Cell. Probes 2007, 21 (2), 125−133. (26) Kristiansson, E.; Fick, J.; Janzon, A.; Grabic, R.; Rutgersson, C.; Weijdegård, B.; Söderström, H.; Larsson, D. G. J. Pyrosequencing of antibiotic−contaminated river sediments reveals high levels of resistance and gene transfer elements. PLoS One 2011, 6 (2), e17038.

(27) Zhang, T.; Zhang, X.-X.; Ye, L. Plasmid metagenome reveals high levels of antibiotic resistance genes and mobile genetic elements in activated sludge. PLoS One 2011, 6 (10), e26041. (28) Martiny, A. C.; Martiny, J. B. H.; Weihe, C.; Field, A.; Ellis, J. C. Functional metagenomics reveals previously unrecognized diversity of antibiotic resistance genes in gulls. Front. Microbiol. 2011, 2, 238. (29) Li, B.; Zhang, T.; Xu, Z. Y.; Fang, H. H. P. Rapid analysis of 21 antibiotics of multiple classes in municipal wastewater using ultra performance liquid chromatography−tandem mass spectrometry. Anal. Chim. Acta 2009, 645 (1), 64−72. (30) Yuan, C. G.; Shi, J. B.; He, B.; Liu, J. F.; Liang, L. N.; Jiang, G. B. Speciation of heavy metals in marine sediments from the East China Sea by ICP−MS with sequential extraction. Environ. Int. 2004, 30 (6), 769−783. (31) Water Environment Federation; American Public Health Association. Standard Methods for the Examination of Water and Wastewater; American Public Health Association (APHA): Washington, DC, 2005. (32) Miao, Y.; Liao, R. H.; Zhang, X.-X.; Wang, Y.; Wang, Z.; Shi, P.; Liu, B.; Li, A. M. Metagenomic insights into Cr (VI) effect on microbial communities and functional genes of an expanded granular sludge bed reactor treating high−nitrate wastewater. Water Res. 2015, 76, 43−52. (33) Wang, Z.; Zhang, X.-X.; Lu, X.; Liu, B.; Li, Y.; Long, C.; Li, A. M. Abundance and diversity of bacterial nitrifiers and denitrifiers and their functional genes in tannery wastewater treatment plants revealed by high−throughput sequencing. PLoS One 2014, 9 (11), e113603. (34) Yang, Y.; Li, B.; Zou, S. C.; Fang, H. H. P.; Zhang, T. Fate of antibiotic resistance genes in sewage treatment plant revealed by metagenomic approach. Water Res. 2014, 62, 97−106. (35) Huerta, B.; Marti, E.; Gros, M.; López, P.; Pompêo, M.; Armengol, J.; Barceló, D.; Balcázar, J. L.; Rodríguez-Mozaz, S.; Marcé, R. Exploring the links between antibiotic occurrence, antibiotic resistance, and bacterial communities in water supply reservoirs. Sci. Total Environ. 2013, 456, 161−170. (36) Hullman, R. Fate of Pharmaceuticals in Drinking Water Utilities; ProQuest LLC: Ann Arbor, MI, 2009. (37) Li, B.; Zhang, T. pH significantly affects removal of trace antibiotics in chlorination of municipal wastewater. Water Res. 2012, 46 (11), 3703−3713. (38) Gao, P. P.; Mao, D. Q.; Luo, Y.; Wang, L. M.; Xu, B. J.; Xu, L. Occurrence of sulfonamide and tetracycline−resistant bacteria and resistance genes in aquaculture environment. Water Res. 2012, 46 (7), 2355−2364. (39) Chiao, T. H.; Clancy, T. M.; Pinto, A.; Xi, C. W.; Raskin, L. Differential resistance of drinking water bacterial populations to monochloramine disinfection. Environ. Sci. Technol. 2014, 48 (7), 4038−4047. (40) Moro, M. H.; Beran, G. W.; Hoffman, L. J.; Griffith, R. W. Effects of cold stress on the antimicrobial drug resistance of Escherichia coli of the intestinal flora of swine. Lett. Appl. Microbiol. 1998, 27 (5), 251−254. (41) McMahon, M. A. S.; Xu, J. R.; Moore, J. E.; Blair, I. S.; McDowell, D. A. Environmental stress and antibiotic resistance in food−related pathogens. Appl. Environ. Microbiol. 2007, 73 (1), 211− 217. (42) Li, X. Z.; Nikaido, H. Efflux−mediated drug resistance in bacteria. Drugs 2004, 64 (2), 159−204. (43) Poole, K. Efflux−mediated antimicrobial resistance. J. Antimicrob. Chemother. 2005, 56 (1), 20−51. (44) Morita, Y.; Murata, T.; Mima, T.; Shiota, S.; Kuroda, T.; Mizushima, T.; Gotoh, N.; Nishino, T.; Tsuchiya, T. Induction of MexCD−OprJ operon for a multidrug efflux pump by disinfectants in wild−type Pseudomonas aeruginosa PAO1. J. Antimicrob. Chemother. 2003, 51 (4), 991−994. (45) Lambert, P. A. Mechanisms of antibiotic resistance in Pseudomonas aeruginosa. J. R. Soc. Med. 2002, 95 (Suppl 41), 22−26. (46) Mackie, R. I.; Koike, S.; Krapac, I.; Chee-Sanford, J.; Maxwell, S.; Aminov, R. I. Tetracycline residues and tetracycline resistance genes in H


Article

Environmental Science & Technology groundwater impacted by swine production facilities. Anim. Biotechnol. 2006, 17 (2), 157−176. (47) El Ghachi, M.; Derbise, A.; Bouhss, A.; Mengin-Lecreulx, D. Identification of multiple genes encoding membrane proteins with undecaprenyl pyrophosphate phosphatase (UppP) activity in Escherichia coli. J. Biol. Chem. 2005, 280 (19), 18689−18695. (48) Stoitsova, S. O.; Braun, Y.; Ullrich, M. S.; Weingart, H. Characterization of the RND−type multidrug efflux pump MexAB− OprM of the plant pathogen Pseudomonas syringae. Appl. Environ. Microbiol. 2008, 74 (11), 3387−3393. (49) Winsor, G. L.; Van Rossum, T.; Lo, R.; Khaira, B.; Whiteside, M. D.; Hancock, R. E. W.; Brinkman, F. S. L. Pseudomonas Genome Database: facilitating user−friendly, comprehensive comparisons of microbial genomes. Nucleic Acids Res. 2009, 37 (suppl 1), D483− D488. (50) Miura, T.; Kusada, H.; Kamagata, Y.; Hanada, S.; Kimura, N. Genome sequence of the multiple−β−lactam−antibiotic−resistant bacterium Acidovorax sp. strain MR−S7. Genome Announc. 2013, 1 (4), e00412−13. (51) Vaz-Moreira, I.; Nunes, O. C.; Manaia, C. M. Diversity and antibiotic resistance patterns of Sphingomonadaceae isolates from drinking water. Appl. Environ. Microbiol. 2011, 77 (16), 5697−5706. (52) Smillie, C. S.; Smith, M. B.; Friedman, J.; Cordero, O. X.; David, L. A.; Alm, E. J. Ecology drives a global network of gene exchange connecting the human microbiome. Nature 2011, 480 (7376), 241− 244. (53) Caro-Quintero, A.; Deng, J.; Auchtung, J.; Brettar, I.; Höfle, M. G.; Klappenbach, J.; Konstantinidis, K. T. Unprecedented levels of horizontal gene transfer among spatially co−occurring Shewanella bacteria from the Baltic Sea. ISME J. 2011, 5 (1), 131−140. (54) Hammes, F.; Berney, M.; Wang, Y. Y.; Vital, M.; Köster, O.; Egli, T. Flow−cytometric total bacterial cell counts as a descriptive microbiological parameter for drinking water treatment processes. Water Res. 2008, 42 (1−2), 269−277. (55) Stover, C. K.; Pham, X. Q.; Erwin, A. L.; Mizoguchi, S. D.; Warrener, P.; Hickey, M. J.; Brinkman, F. S. L.; Hufnagle, W. O.; Kowalik, D. J.; Lagrou, M. Complete genome sequence of Pseudomonas aeruginosa PAO1, an opportunistic pathogen. Nature 2000, 406 (6799), 959−964. (56) Tandukar, M.; Oh, S.; Tezel, U.; Konstantinidis, K. T.; Pavlostathis, S. G. Long−term exposure to benzalkonium chloride disinfectants results in change of microbial community structure and increased antimicrobial resistance. Environ. Sci. Technol. 2013, 47 (17), 9730−9738. (57) Lv, L.; Jiang, T.; Zhang, S. H.; Yu, X. Exposure to mutagenic disinfection byproducts leads to increase of antibiotic resistance in Pseudomonas aeruginosa. Environ. Sci. Technol. 2014, 48 (14), 8188− 8195.

I


Bacterial Community Shift Drives Antibiotic ... - ACS Publications

Recommend Documents