TY - JOUR
AU - Edwards, Elizabeth, A
AB - ABSTRACT Solid organic waste is a significant source of antibiotic resistance genes (ARGs) and effective treatment strategies are urgently required to limit the spread of antimicrobial resistance. Here, we studied ARG diversity and abundance as well as the relationship between antibiotic resistome and microbial community structure within a lab-scale solid-state anaerobic digester treating a mixture of food waste, paper and cardboard. A total of 10 samples from digester feed and digestion products were collected for microbial community analysis including small subunit rRNA gene sequencing, total community metagenome sequencing and high-throughput quantitative PCR. We observed a significant shift in microbial community composition and a reduction in ARG diversity and abundance after 6 weeks of digestion. ARGs were identified in all samples with multidrug resistance being the most abundant ARG type. Thirty-two per cent of ARGs detected in digester feed were located on plasmids indicating potential for horizontal gene transfer. Using metagenomic assembly and binning, we detected potential bacterial hosts of ARGs in digester feed, which included Erwinia, Bifidobacteriaceae, Lactococcus lactis and Lactobacillus. Our results indicate that the process of sequential solid-state anaerobic digestion of food waste, paper and cardboard tested herein provides a significant reduction in the relative abundance of ARGs per 16S rRNA gene. antibiotic resistance genes, anaerobic digestion, food waste INTRODUCTION Antimicrobial resistance (AMR) is a widely recognized public health risk. The extensive use of antimicrobial compounds since World War II has triggered the rapid spread and evolution of AMR mechanisms to such an extent that AMR is considered one of today's top medical concerns globally (WHO 2015). Currently, 700 000 deaths are attributed to resistant microbial infections annually, which is projected to increase to 10 million deaths by 2050 resulting in the loss of 100 trillion USD of economic output (O'Neill 2016). In light of this gloomy prospect, a One Health approach has been proposed to tackle AMR by recognizing the connections between human and animal health and the environment (Robinson et al. 2016). Although the vast majority of antimicrobial research is focusing on the clinical setting, the natural environment has gained attention as a possible reservoir and dispersal route of AMR determinants, including antibiotic-resistant bacteria and antibiotic resistance genes (ARGs; Huijbers et al. 2015; Bengtsson-Palme, Kristiansson and Larsson 2018). Various natural ecosystems, such as ancient permafrost sediments (D'Costa et al. 2011), pristine soil (Perron et al. 2015; Durso et al. 2016), and oceanic (Hatosy and Martiny 2015) and freshwater bodies (Czekalski et al. 2015), have been shown to possess a diversity of ARGs collectively defined as the resistome. However, higher concentrations of resistance determinants have been observed in environments with anthropogenic impact where selective pressure for ARGs is present, such as wastewater treatment plants, animal husbandry facilities, aquaculture farms and pharmaceutical manufacturing (Berendonk et al. 2015). Various types of solid organic waste may also serve as potential sources of antibiotic-resistant bacteria and ARGs (Pepper, Brooks and Gerba 2018). In addition to the extensively studied resistomes of sewage sludge and animal manure, the organic fraction of municipal solid waste (MSW) may also contribute to the dissemination of AMR. Global production of MSW is estimated at around 2 billion tons per year, of which 34–53% is organic biodegradable waste, including food waste (FW) as the primary component (Braguglia et al. 2018). Paper waste forms 17% of MSW globally, including different lignocellulosic fibers such as cardboard, newspaper, magazines, wrapping paper, shredded paper, boxes, bags and beverage cups (Hoornweg and Bhada-Tata 2012). Several studies have highlighted the role of the food chain in AMR as a direct link to human health (Friedman 2015; Bengtsson-Palme 2017) and identified antibiotic-resistant pathogens in various food products such as meat (Chajęcka-Wierzchowska, Zadernowska and Łaniewska-Trokenheim 2016; Noyes et al. 2016; Yang et al. 2016a), fruit and vegetables (Oliveira et al. 2012; Holvoet et al. 2013; Marti et al. 2013; Rahube et al. 2014), poultry (Diarra et al. 2014; Yang et al. 2014), fish (Elhadi 2014; Moore et al. 2014) and dairy (Silveira-Filho et al. 2014). Therefore, it is reasonable to assume that ARGs are also present in food waste: Lee et al. (2017) detected a variety of ARGs in the wastewater from food waste recycling in Korea, although the abundance of ARGs remained below that of manure and sewage sludge. To mitigate the harmful environmental impacts of organic solid waste, anaerobic digestion (AD) is widely implemented as a treatment strategy, providing waste stabilization as well as the production of renewable energy (Vasco-Correa et al. 2018). AD changes the structure of the microbial community of the substrates, which, in turn, affects the resistome present in digestion products. Although it is generally accepted that ARG abundance is reduced overall during AD, enrichment of some ARGs and rebound effects have been reported in several studies (summarized by Youngquist, Mitchell and Cogger 2016). For example, Pu et al. (2018) studied the impact of applying pig manure to fields and found that AD reduced the relative abundance of macrolide–lincosamide–streptogramin (MLS) and tetracycline resistance genes, while resistance genes for sulfa, aminoglycoside, florfenicol, amphenicol and chloramphenicol were enriched. Similarly, the effect of aerobic composting of organic waste remains contradictory: while some evidence suggests composting can reduce the abundance of antibiotic-resistant bacteria and ARGs, other studies have shown increased abundance and diversity of ARGs (Youngquist, Mitchell and Cogger 2016). Recently, there has been a growing interest in AD of food waste due to its high energetic value (Braguglia et al. 2018). However, limited information is available on the effect of AD on the microbial community and resistome in food waste. Zhang et al. (2018) identified 11 ARGs and class 1 integron-integrase gene intI1 in digested food waste: while tetA, tetB, tetX, sul1, cmlA, floR and intI1 were significantly reduced by AD, enrichment of tetM, tetW, tetQ and tetO was recorded. Similarly, another study detected both increase and reduction of specific ARGs in the co-digestion of sewage sludge and food waste following microwave pre-treatment (Zhang et al. 2016). Thus, the effect of AD on the food waste resistome and the associated microbial community is still unclear. In addition to quantification of individual ARGs, detecting potential host organisms of these genes is important to evaluate the risk to human health. Several studies have used correlation and network analysis to detect relationships between individual ARGs and bacterial genera (Li et al. 2015; Zhang et al. 2016; Jang et al. 2017; Luo et al. 2017). High mobility of ARGs due to horizontal gene transfer (HGT) is responsible for the spread of AMR between different bacteria, including human pathogens as well as non-pathogenic environmental bacteria that often serve as reservoirs of ARGs (Stokes and Gillings 2011). The objective of this study was to measure ARG abundance and resistome diversity before and after anaerobic co-digestion of food waste, paper and cardboard. Samples were collected from a lab-scale solid-state AD system that exhibited stable methane production and substrate destruction rates (Guilford 2017; Guilford et al. 2019). Using small subunit (SSU) rRNA gene sequencing, total community metagenome sequencing and targeted quantitative polymerase chain reaction (PCR), we aimed at quantifying changes in the diversity and size of the resistome before and after digestion together with changes in the taxonomic and genomic profiles of the microbial community. MATERIALS AND METHODS Anaerobic digestion system and feedstock A lab-scale anaerobic digestion system designed for the treatment of solid organic waste was operated for a total of 88 weeks. System design, its operating parameters and feedstock have been described in detail by Guilford et al. (2019). In short, the anaerobic digestion system consisted of six leach beds (8.5 L each), an upflow anaerobic sludge blanket reactor (UASB; 27.5 L) treating leach bed leachate, a UASB feed tank (17.5 L), a leach bed feed tank (17.5 L), three peristaltic pumps to recirculate leachate, two wet-tip gas meters for biogas measurement and an automated control system. The system was maintained at 37–39°C with continuous recirculation of the leachate. It was operated in sequential batch feeding mode: each week one of the leach beds was filled with a mixture of lignocellulosic fibers and food waste making up the leach bed feed (LBF), recovered from local residential waste recycling programs. Lignocellulosic fibers included shredded cardboard, boxboard, fine paper and newsprint; food waste was collected from a residential green bin program and sorted manually to remove bones and inorganic materials. Solids retention time in the system was 6 weeks. Sampling and DNA extraction Samples for microbial community analysis were collected during weeks 75–84 of the experiment, during which FW contributed 21.7% of the total chemical oxygen demand (COD) in the feedstock and the system exhibited stable performance with a methane yield of 246 L CH4 kg VSadded−1 and a substrate destruction efficiency, as volatile solids (VS), of 63.5% (Guilford et al. 2019). Ten 50 mL samples were collected: three food waste samples FW1, FW2 and FW3 (FW source and composition has been described by Guilford et al. 2019); three leach bed feed samples LBF1, LBF2 and LBF3 consisting of the respective food waste mixed with lignocellulosic fibers; three digestate samples DG74, DG76 and DG78 collected after 6 weeks of digestion; and one sample from the microbial granular sludge of the UASB reactor (UASB). It should be noted that DG74, DG76 and DG78 were not direct digestion products of LBF1, LBF2 and LBF3, respectively, although derived from the same source of FW and lignocellulosic fibers used for LBF1, LBF2 and LBF3. All samples were preserved at –20°C until DNA extraction. Total community DNA was extracted using the PowerMax Soil DNA Isolation Kit (MoBio Laboratories, Carlsbad, CA) from 5 g of sample according to the manufacturer's protocol. FW and LBF samples were further purified with 5 M NaCl and 100% ethanol as recommended by the manufacturer. The quantity and quality of DNA extracts were confirmed using NanoDrop spectrophotometer ND-1000 (Thermo Fisher Scientific, Wilmington, DE). All DNA extracts were stored at –80°C. Quantitative PCR analyses Quantitative PCR (qPCR) was used to quantify total bacterial (method 1.1 in Fig. 1) and archaeal (method 1.2 in Fig. 1) 16S rRNA gene copies using a CFX96 Touch Real-Time PCR Detection System (Bio-Rad Laboratories, Hercules, CA) and previously published primers targeting most bacterial 16S rRNA genes: Bac1055f (5′-ATGGCTGTCGTCAGCT-3′) and Bac1392r (5′-ACGGGCGGTGTGTAC-3′; Amann, Ludwig and Schleifer 1995; Ferris, Muyzer and Ward 1996); and most archaeal 16S rRNA genes: Arch787f (5′-ATTAGATACCCGBGTAGTCC-3′) and Arch1059r (5′-GCCATGCACCWCCTCT-3′; Yu et al. 2005). qPCR reactions were performed in 20 μL comprising 10 μL SsoFast EvaGreen Supermix (Bio-Rad Laboratories, Hercules, CA), 0.5 μM each forward and reverse primer, 2 μL template DNA and sterile UltraPure distilled water. The thermocycling program was as follows: initial denaturation at 98°C for 2 min, followed by 40 cycles of denaturation at 98°C for 5 s, annealing at Tm (55°C for bacterial 16S rRNA, 60°C for archaeal 16S rRNA) for 10 s and final melting curve analysis in the range of 65–95°C (steps of 0.5°C for 5 s). All samples were measured in three technical replicates and every qPCR run included negative controls containing all reaction components except for template DNA. Serial dilutions of plasmid stocks containing corresponding 16S rRNA gene fragments were used to generate standard curves. Amplification data were analyzed using CFX Manager Software v.3.1. Copy numbers of bacterial and archaeal 16S rRNA genes per gram of sample dry weight were calculated as follows: Absolute abundance = starting quantity × DNA elution volume/(sample weight × sample DW), where starting quantity represents the gene copies in 1 μL DNA extract as estimated by the CFX Manager Software, DNA elution volume was 5000 μL for DG74, DG76, DG78 and UASB and 100 μL for FW1, FW2, FW3, LBF1, LBF2 and LBF3, sample weight was the mass of the sample used for DNA extraction and sample DW was the dry weight of the sample material used for DNA extraction. Figure 1. Open in new tabDownload slide Experimental workflow of anaerobic digestion experiment. Total community DNA extracts from 10 samples including food waste (FW1, FW2, FW3), leach bed feed (LBF1, LBF2, LBF3), digestate (DG74, DG76, DG78) and microbial granules from UASB reactor (UASB) were subjected to SSU rRNA gene amplicon sequencing (2), total community metagenome sequencing (3) and quantitative PCR analysis (1, 4). Sequencing data were analyzed using a variety of bioinformatic tools shown and numbered in italics. SSU rRNA gene sequencing data were analyzed using QIIME1 (tool 2.1). The workflow for total community metagenome sequencing data included quality trimming of the short sequencing reads (tools 3.1 and 3.2), taxonomic annotation of the quality trimmed reads (tools 3.3 and 3.4), detection of ARGs from quality trimmed reads (tool 3.5), assembly of the quality trimmed reads into contigs (tools 3.6 and 3.7), identification of plasmid sequences (tool 3.8) and ARGs (tool 3.10) from assembled contigs, taxonomic annotation of assembled contigs (tool 3.9), binning of the assembled contigs into metagenome-assembled genomes (MAGs; tools 3.11, 3.12, 3.13) and detection of ARGs in MAGs by a local Python script (tool 3.14). Quantitative PCR was performed for bacterial (1.1) and archaeal (1.2) 16S rRNA gene as well as for ARGs (4.1) and MGEs (4.2) using a high-throughput qPCR (HT-qPCR) array. Results covered microbial community composition, ARG diversity, abundance, genomic location and potential bacterial host organisms of ARGs. Figure 1. Open in new tabDownload slide Experimental workflow of anaerobic digestion experiment. Total community DNA extracts from 10 samples including food waste (FW1, FW2, FW3), leach bed feed (LBF1, LBF2, LBF3), digestate (DG74, DG76, DG78) and microbial granules from UASB reactor (UASB) were subjected to SSU rRNA gene amplicon sequencing (2), total community metagenome sequencing (3) and quantitative PCR analysis (1, 4). Sequencing data were analyzed using a variety of bioinformatic tools shown and numbered in italics. SSU rRNA gene sequencing data were analyzed using QIIME1 (tool 2.1). The workflow for total community metagenome sequencing data included quality trimming of the short sequencing reads (tools 3.1 and 3.2), taxonomic annotation of the quality trimmed reads (tools 3.3 and 3.4), detection of ARGs from quality trimmed reads (tool 3.5), assembly of the quality trimmed reads into contigs (tools 3.6 and 3.7), identification of plasmid sequences (tool 3.8) and ARGs (tool 3.10) from assembled contigs, taxonomic annotation of assembled contigs (tool 3.9), binning of the assembled contigs into metagenome-assembled genomes (MAGs; tools 3.11, 3.12, 3.13) and detection of ARGs in MAGs by a local Python script (tool 3.14). Quantitative PCR was performed for bacterial (1.1) and archaeal (1.2) 16S rRNA gene as well as for ARGs (4.1) and MGEs (4.2) using a high-throughput qPCR (HT-qPCR) array. Results covered microbial community composition, ARG diversity, abundance, genomic location and potential bacterial host organisms of ARGs. An HT-qPCR array was used for the quantification of ARGs (method 4.1 in Fig. 1) and mobile genetic elements (MGEs; method 4.2 in Fig. 1) in FW1, FW2, DG74 and DG78. DNA extracts were sent to Dr Robert Stedtfeld at Michigan State University (USA) who performed the analysis. A total of 384 primer sets were used targeting 315 ARGs and 57 MGEs; additionally, taxonomic marker genes were included in the array (Stedtfeld et al. 2018). All HT-qPCR reactions were performed using Takara (previously WaferGen) SmartChip real-time PCR system as described previously (Wang et al. 2014). In brief, 5184 parallel qPCR reactions (100 nL) were dispensed into the SmartChip using the SmartChip Multisample Nanodispenser, followed by thermal cycling in the SmartChip Cycler, automatic melting process and initial data processing with SmartChip qPCR Software (v.2.7.0.1) as described before (Wang et al. 2014). Wells with multiple melt peaks and wells with amplification efficiency outside the range of 1.75 to 2.25 were removed from analysis. Each sample was analyzed by three technical replicates. Genes detected in only one of the replicates were considered false positives and removed from analysis. Genomic copy numbers were estimated using the formula in Looft et al. (2012) with the exception of setting detection limit to threshold cycle (Ct) 28 as suggested by Stedtfeld et al. (2018). Relative abundances of the studied genes were calculated as ratios to 16S rRNA: Relative abundancegene = genomic copy numbergene/genomic copy number16S. Absolute abundances were determined by multiplying the relative abundance of a gene by 16S rRNA absolute abundance (determined by regular qPCR analysis described above): Absolute abundancegene = relative abundancegene × absolute abundance16S. ARG absolute abundances were visualized in R software using the package ComplexHeatmap (Gu, Eils and Schlesner 2016). Spearman rank order correlation was used to assess the relationship between the relative abundance values obtained by HT-qPCR array and by metagenomic analysis for a subset of 61 selected ARGs. SSU rRNA gene amplicon sequencing and taxonomic analyses SSU rRNA gene sequencing with the universal primers 926f-modified (5′-AAACTYAAAKGAATWGRCGG-3′) and 1392r-modified (5′-ACGGGCGGTGWGTRC-3′; modified from Engelbrektson et al. 2010) targeting the V6–V8 variable region of the 16S rRNA gene from bacteria and archaea as well as the 18S rRNA gene in eukarya was performed at McGill University and Génome Québec Innovation Centre (Canada) on Illumina MiSeq System (PE300). The analysis of sequencing reads was performed with the QIIME1 package (tool 2.1 in Fig. 1; Caporaso et al. 2010) by joining forward and reverse reads (multiple_join_paired_ends.py, min. overlap 50 bp, max. mismatch allowance 8), removing low-quality sequences (multiple_split_libraries_fastq.py, quality threshold 19), removing chimeric sequences and identifying operational taxonomic units (OTUs) at 97% similarity [identify_chimeric_seqs.py, pick_open_reference_otus.py, method usearch61 (Edgar 2010), reference database SILVA release 128]. Total community metagenome sequencing and bioinformatic workflow Total community metagenome sequencing of the samples FW1, FW2, LBF1, LBF2, DG74, DG78 and UASB was performed at University of Toronto Centre for the Analysis of Genome Evolution and Function (Canada) on Illumina NextSeq500 Desktop Sequencer v2 using high output flowcell (PE150). Bioinformatic workflow included quality control and trimming of the sequencing reads, taxonomic classification, metagenomic assembly and binning, and detection of ARGs. Sequence reads were trimmed with Trimmomatic v.0.32 (Bolger, Lohse and Usadel 2014; tool 3.1 in Fig. 1) using default settings for paired-end mode; additional trimming to remove polyG sequences (≥30 bp) was performed with AfterQC v.0.9.1 (Chen et al. 2017; tool 3.2 in Fig. 1). Quality-trimmed reads were subjected to taxonomic classification using Kaiju v.1.4.5 (Menzel, Ng and Krogh 2016; tool 3.3 in Fig. 1) in greedy mode (allowed substitutions 5, minimum required match length 11, minimum required match score 70) with bacterial, archaeal, eukaryotic and viral protein sequences from the NCBI nr database as reference. To further investigate microbial community composition, metaxa2 v.2.1 (Bengtsson-Palme et al. 2015; tool 3.4 in Fig. 1) was used to extract small subunit rRNA gene reads in metagenome mode with default search criteria and reliability score cutoff 80. Quality-trimmed reads were assembled into longer DNA contigs using MEGAHIT v.1.1.1 (Li et al. 2016; tool 3.6 in Fig. 1) and metaSPAdes v.3.10.1 (Nurk et al. 2017; tool 3.7 in Fig. 1) with default settings. Assemblies were compared using metaQUAST v.4.5 (Mikheenko, Saveliev and Gurevich 2016), which also provided assembly statistics. Metagenome binning was performed using MaxBin v.2.2.3 (Wu, Simmons and Singer 2016; tool 3.11 in Fig. 1) and MetaBAT v.2.12 (Kang et al. 2015; tool 3.12 in Fig. 1) with default settings on the MEGAHIT assemblies. All genome bins produced by both binning methods were dereplicated using dRep v.1.4.3 (Olm et al. 2017; tool 3.13 in Fig. 1) with default settings (ANI cutoff 99%, minimum genome completeness 75%, maximum contamination 25%) and evaluated with CheckM v.1.0.7 (Parks et al. 2015). Quality of the resulting MAGs was defined as (completeness − 5× contamination) as suggested by Parks et al. (2017). MAGs with quality > 50 and taxonomic classification to at least phylum level were submitted to the JGI Integrated Microbial Genomes and Microbiomes (IMG) database for annotation. Anvi'o v.5 (Eren et al. 2015) was used for visualizing MAGs clustered based on the relative abundance of MAGs in the samples: Euclidian distances were calculated from relative abundance estimates of each MAG and clustering was performed using the ward linkage algorithm. Relative abundance of each MAG was calculated as number of reads recruited to the MAG divided by total reads recruited to that MAG across all samples. ARGs from subsets of quality-trimmed sequencing reads (20 million reads/sample) were determined with ARGs-OAP v.1.2 (Yang et al. 2016b; tool 3.5 in Fig. 1) using the default reference database SARG and default settings. The same reference database was also used to detect ARGs on assembled contigs by local BLASTp (e-value 1e-7, identity % 80, minimal alignment length 25) using protein coding genes annotated by JGI IMG/M 4 version (Markowitz et al. 2014) as query sequences (tool 3.10 in Fig. 1). Contigs with multiple ARGs were identified by a local python script that counted how many times a contig ID appeared in the list of contigs that carried ARGs (Method S1 available at https://github.com/kkanger/PythonScripts). PlasFlow v.1.1 (Krawczyk, Lipinski and Dziembowski 2018) with default settings was used to predict plasmid sequences from the contigs with ARGs (tool 3.8 in Fig. 1). Taxonomic classification of the non-plasmid contigs carrying ARGs was determined with Kaiju v.1.4.5 (Menzel, Ng and Krogh 2016) in greedy mode (maximum mismatches allowed 5, minimal required match length 11, minimal required match score 70, filtering of query sequences containing low-complexity regions on; tool 3.9 in Fig. 1) using NCBI RefSeq reference database and visualized with Cytoscape v.3.6.1 (Shannon et al. 2003) by selecting 10 most abundant ARGs from all genera above relative abundance of 1%. The presence of ARGs in MAGs was determined by a local python script comparing a list of ARG-carrying contigs to the list of contigs included in the MAG (tool 3.14 in Fig. 1; Method S2 available at https://github.com/kkanger/PythonScripts). Sequencing data availability Sequencing data from SSU rRNA gene sequencing and total community metagenome sequencing (accession numbers provided in Table S1, Supporting Information) are available on NCBI Sequence Read Archive under BioProject PRJNA501900, SRA study SRP167436 and accession numbers SRX4965138–47 (SSU rRNA gene sequencing) and SRX4986160–6 (total community metagenome sequencing). Additionally, metagenome assemblies (accession numbers provided in Table S1, Supporting Information) and MAGs (accession numbers provided in Table S7, Supporting Information) have been deposited to the JGI Integrated Microbial Genomes and Microbiomes (https://img.jgi.doe.gov/cgi-bin/m/main.cgi). RESULTS A total of 10 samples were collected from a lab-scale solid-state anaerobic digestion system described in detail by a recent Ph.D. thesis and corresponding publication (Guilford 2017; Guilford et al. 2019). The microbial community of the system has been previously described by Lee (2018), which was used as a reference for selecting representative samples for this follow-up study. DNA was extracted from each sample in order to analyze the microbial community composition and ARG diversity, abundance, genomic location and host organisms using a multipronged approach combining metagenome sequence analyses, taxonomic profiling and quantitative PCR (Fig. 1). Microbial community Microbial community structure from a variety of samples collected from the studied anaerobic digestion system over 88 weeks of operation has been described previously by Lee (2018) showing a stable microbial community structure that varied as a function of food waste content in digester feed. In this study, microbial community structure of 10 selected samples was analyzed by SSU rRNA gene-fragment sequencing (method 2 in Fig. 1) capturing bacterial, archaeal and eukaryotic diversity (Table S1, Supporting Information). A clear distinction in the communities of digester feed and digestion products was detected: food waste and leach bed feed microbial communities were dominated by bacterial and eukaryotic taxa while anaerobic digestate and microbial community of granules from the UASB reactor possessed a large proportion of archaeal phylotypes mainly belonging to the methanogenic genus Methanosaeta (Fig. 2). The most abundant bacterial OTUs in digester feed belonged to the family Enterobacteriaceae, including the genera Citrobacter, Enterobacter, Erwinia, Kluyvera, Pantoea, Serratia and Lelliottia. Additionally, representatives of the phylum Firmicutes were detected in FW2, including the genera Lactobacillus and Leuconostoc, probably indicating fermentation processes occurring in the respective food waste. In addition to bacterial sequences, fungal and plant material was detected in digester feed (FW, LBF), but not in digestion products (DG) or the UASB. Across all samples, 36–62% of OTUs were detected at low relative abundances comprising <1% of the total community. Figure 2. Open in new tabDownload slide Taxonomic composition of bacterial, archaeal and eukaryotic OTUs in food waste (FW1, FW2, FW3), leach bed feed (LBF1, LBF2, LBF3), digestate (DG74, DG76, DG78) and UASB reactor (UASB) samples detected by SSU rRNA gene sequencing. Each rectangle represents an individual OTU with relative abundance ≥1%. The remaining OTUs with relative abundance <1% are summed and represented in gray as separate rectangles for bacteria, archaea and eukaryota. Bacterial OTUs are presented in solid colors, archaeal OTUs with vertical lines and eukaryotic OTUs in diamond pattern. OTUs belonging to the bacterial family Enterobacteriaceae are presented in various shades of green. Additional information on all detected OTUs is available in Table S2 (Supporting Information). Figure 2. Open in new tabDownload slide Taxonomic composition of bacterial, archaeal and eukaryotic OTUs in food waste (FW1, FW2, FW3), leach bed feed (LBF1, LBF2, LBF3), digestate (DG74, DG76, DG78) and UASB reactor (UASB) samples detected by SSU rRNA gene sequencing. Each rectangle represents an individual OTU with relative abundance ≥1%. The remaining OTUs with relative abundance <1% are summed and represented in gray as separate rectangles for bacteria, archaea and eukaryota. Bacterial OTUs are presented in solid colors, archaeal OTUs with vertical lines and eukaryotic OTUs in diamond pattern. OTUs belonging to the bacterial family Enterobacteriaceae are presented in various shades of green. Additional information on all detected OTUs is available in Table S2 (Supporting Information). A comparison of genus level taxonomic composition detected from SSU rRNA gene amplicon sequence analysis using QIIME (tool 2.1 in Fig. 1) to that derived from total community metagenome sequencing analysis using Kaiju (tool 3.3 in Fig. 1) or metaxa2 (tool 3.4 in Fig. 1) showed relatively good agreement between the three different annotation tools used in this study (Figure S1, Supporting Information). Comparison of the 10 samples analyzed in this study to previous samples collected from the same AD system showed that the bacterial community of digester feed was similar to the bacterial community of fresh leachate samples collected over 50 weeks of operation straight after adding new feedstock to the system (Figure S2A, Supporting Information). Highest similarity was detected at periods of high food waste content in the system. Similarly, digestion products closely resembled matured leachate and microbial granules collected from various locations of the AD system (Figure S2B, Supporting Information). Thus, samples analyzed in this study are representative of the whole AD system over a longer operating period and particularly at high food waste concentrations. ARG diversity and abundance Total community metagenome sequencing data from seven samples (Table S1, Supporting Information) were analyzed using the ARGs Online Analysis Pipeline (ARGs-OAP; Yang et al. 2016b; tool 3.5 in Fig. 1) to characterize the distribution and diversity of ARGs before and after anaerobic digestion. The richness of ARGs, measured as the number of distinct ARGs identified in one sample type, was highest in digester feed with 330 and 336 different ARGs detected in FW and LBF samples, respectively. The richness of ARGs in digestion products remained two times lower with 115 different ARGs detected in DG and UASB samples indicating reduced diversity of ARGs after anaerobic digestion. Twenty-three ARGs were found to be unique to digestion products belonging to aminoglycoside [aac(3)-I, ant(9)-I], beta-lactam (OXA-10, OXA-205, OXA-251, OXA-34, OXA-46, OXA-75), chloramphenicol (catQ), MLS (ereB, lnuB, mphA, carA), sulfonamide (sul3), tetracycline (tet44, tetT), trimethoprim (dfrA5) and vancomycin (vanA, vanG, vanH, vanN, vanU, vanX) type. In addition to richness, the relative abundance of ARGs per 16S rRNA gene was determined from total community metagenome sequencing data (Table S3, Supporting Information). Relative abundances of ARGs per 16S rRNA gene were significantly higher in digester feed than in digestion products, indicating the ability of anaerobic digestion to reduce ARGs (Fig. 3). Fold change of most ARG types, defined as a ratio of average relative abundance in digestion products to average relative abundance in digester feed, showed a decrease ranging between 0.001 (kasugamycin) and 0.70 (sulfonamide) of the input abundances (Table S3, Supporting Information). ARGs conferring resistance to fosfomycin, fusaric acid and tetracenomycin C were completely removed and not detected in digestion products, while trimethoprim and vancomycin ARG types showed a fold increase of 1.40 and 1.54 times, respectively. Total relative abundances of ARGs per 16S rRNA gene in food waste samples FW1 (0.78 ARG/16S rRNA) and FW2 (0.40 ARG/16S rRNA) were similar to the respective values in the mixtures of food waste with lignocellulosic fibers (LBF1 0.84 ARG/16S rRNA, LBF2 0.44 ARG/16S rRNA), indicating the role of food waste as the primary source of ARGs in leach bed feed. Most of the ARGs identified in digester feed conferred multidrug resistance (Fig. 3A). Additionally, many potential ARGs with unclassified resistance type were detected. The highest relative abundance value for an individual gene (Fig. 3B) was recorded for AcrAB-TolC multidrug efflux complex subunit acrB reaching 0.05 copies/16S rRNA gene in samples FW1 and LBF1. Other highly abundant genes in digester feed included subunits of multidrug resistance complexes (acrA, mdtB, mdtC, tolC), several regulatory protein genes associated with AMR (cpxR, arlR) and bacitracin resistance gene bacA. Relative abundances of individual ARGs in digestion products remained below 0.01 copies/16S rRNA in all cases. Similarly to digester feed, the highest value for an individual gene in digestion products was recorded for acrB with 0.007 copies/16S rRNA in UASB. Figure 3. Open in new tabDownload slide Relative abundance of ARGs per 16S rRNA gene categorized by type (A) and by gene (B) detected in total community metagenome sequencing reads using ARGs-OAP. Samples analyzed were from food waste (FW1, FW2), leach bed feed (LBF1, LBF2), digestate (DG74, DG78) and UASB reactor (UASB). The 10 most abundant ARGs from each sample are depicted in panel B. Genes not detected are marked in gray. Relative abundances of all detected ARGs are provided in Table S3 (Supporting Information). Figure 3. Open in new tabDownload slide Relative abundance of ARGs per 16S rRNA gene categorized by type (A) and by gene (B) detected in total community metagenome sequencing reads using ARGs-OAP. Samples analyzed were from food waste (FW1, FW2), leach bed feed (LBF1, LBF2), digestate (DG74, DG78) and UASB reactor (UASB). The 10 most abundant ARGs from each sample are depicted in panel B. Genes not detected are marked in gray. Relative abundances of all detected ARGs are provided in Table S3 (Supporting Information). An HT-qPCR array was also used as an alternative approach to quantify 315 ARGs (tool 4.1 in Fig. 1) and 57 MGEs (tool 4.2 in Fig. 1) in food waste and digestate DNA samples (Table S4, Supporting Information). In agreement with the results from the metagenomic data, the diversity of ARGs and MGEs was higher in food waste with 161 different genes detected, while only 32 different genes were detected in digestate (Fig. 4). Among these, 10 ARGs were found only in digestate samples and not in food waste. Notably, MLS resistance genes mphA and lnuB, phenicol resistance gene catQ and tetracycline resistance gene tet44 were found to be unique to digestate by both methods. Figure 4. Open in new tabDownload slide Absolute abundance of ARGs and MGEs in food waste (FW1, FW2) and digestate (DG74, DG78) samples detected using an HT-qPCR array. Absolute abundance of all detected genes is presented on log-scale as gene copies per gram of dry weight of food waste or digestate samples. Genes not detected are marked in gray. Panel (A) shows all ARGs detected in at least one of the four samples. Panel (B) shows all ARGs detected in at least one of the digestate samples. Panel (C) shows all MGEs detected in at least one of the four samples. Raw data for all analyzed genes together with abundance calculations are provided in Table S4 (Supporting Information). Figure 4. Open in new tabDownload slide Absolute abundance of ARGs and MGEs in food waste (FW1, FW2) and digestate (DG74, DG78) samples detected using an HT-qPCR array. Absolute abundance of all detected genes is presented on log-scale as gene copies per gram of dry weight of food waste or digestate samples. Genes not detected are marked in gray. Panel (A) shows all ARGs detected in at least one of the four samples. Panel (B) shows all ARGs detected in at least one of the digestate samples. Panel (C) shows all MGEs detected in at least one of the four samples. Raw data for all analyzed genes together with abundance calculations are provided in Table S4 (Supporting Information). Although the richness of ARGs and MGEs was higher in food waste samples, absolute abundances of individual genes normalized per sample dry weight reached higher levels in digestates (Fig. 4) according to HT-qPCR array results calculated based on abundance of 16S rRNA gene per gram of dry sample (tool 1.1 in Fig. 1; Figure S3, Supporting Information). Most prevalent ARG types in final digestates (Fig. 4B) included MLS [genes ere(A), erm(F), erm(O), lnu(F), lnuB, mef(B), mphA], aminoglycoside (genes aac3-Via, aadE, aph4ib) and tetracycline (genes tet44, tetD, tetM, tetW) with absolute abundances of individual genes ranging between 107 and 108 copies/g DW. Absolute abundances of ARGs in food waste samples ranged between 104 and 107 copies/g DW. MGEs followed a similar pattern to ARGs with more genes detected in food waste but higher absolute abundances of individual genes per gram of sample dry weight detected in digestates (Fig. 4C). In order to compare ARG levels obtained by metagenomic analysis and by HT-qPCR array, a subset of 61 ARGs in four samples was selected. In 84.8% of cases, the two methods agreed on whether an ARG was detected or not (44.7% both methods detected the gene, 40.1% both methods did not detect the gene); disagreement of the two methods was recorded for 15.2% of cases. Spearman rank order correlation was used to analyze the correspondence between 16S-normalized relative abundance values obtained by the two methods (Figure S4, Supporting Information). Using gene-level data, statistically significant correlation between the two methods was detected (R = 0.66, P < 0.001); however, the strength of the correlation was found to be dependent on ARG type. For sulfonamides and MLS type, the correlation between HT-qPCR array and metagenomic analysis was very strong (R = 0.93–0.99, P < 0.001); multidrug and beta-lactam types showed strong correlation (R = 0.73–0.77, P < 0.001); a statistically significant correlation was also recorded for phenicol and tetracycline types (R= 0.61–0.62, P < 0.01); and no correlation was found for trimethoprim, vancomycin and aminoglycoside resistance types. Genomic context of ARGs In order to study the genomic context of ARGs, total community metagenome sequencing reads were assembled into longer contigs using MEGAHIT (tool 3.6 in Fig. 1) and metaSPAdes (tool 3.7 in Fig. 1) assemblers (Table S1, Supporting Information), followed by detection of ARGs on contigs (tool 3.10 in Fig. 1) and identification of plasmid sequences carrying ARGs (tool 3.8 in Fig. 1; Table S5, Supporting Information). On average, 0.10% of all assembled contigs carried ARGs in digester feed, while only 0.003% of contigs identified in digestion products included ARGs (Table S6, Supporting Information). More than 90% of ARG-carrying contigs in all samples included only one ARG; however, contigs with multiple ARGs were also observed with up to five ARGs per contig (Table S6, Supporting Information) detected in digester feed. Contigs with multiple ARGs typically carried subunits of multidrug efflux systems such as MdtABC-TolC coupled with two-component regulatory systems for efflux proteins (such as BaeSR). Identification of plasmids from ARG-carrying contigs revealed that on average 32% of ARGs detected in digestion substrates were located on plasmids (Table 1) indicating their potential for HGT. The distribution of plasmid ARGs resembled the pattern of ARGs identified from short sequencing reads with multidrug resistance genes being the most abundant type in digestion substrates (Figure S5, Supporting Information). Proportion of plasmid ARGs in digestion products varied between 33.3 and 60.0%, although the number of detected ARGs remained low (11 ARGs on plasmids in DG74, 24 ARGs in DG78 and 20 ARGs in UASB, respectively; Table 1). ARGs identified as chromosomal accounted for 8.3–28.9% of ARGs across all samples, remaining below the respective values for plasmid ARGs in all cases (Table 1). Table 1. Genomic location of ARGs detected in total community metagenome sequencing data assembled with MEGAHIT. ARGs were detected by BLAST analysis of protein coding genes using a local SARG reference database. Plasmid or chromosomal location of the ARGs was predicted by PlasFlow. All ARG BLAST hits detected in MEGAHIT-assembled metagenome sequencing data are provided in Table S5 (Supporting Information). Sample . Total assembly length (bp) . Protein coding genes . Number of ARG BLAST hits (% from protein coding genes) . Plasmid ARGs (%) . Chromosomal ARGs (%) . FW1 1285 684 727 2088 107 2407 (0.12%) 30.2 28.6 FW2 941 870 975 1796 492 2484 (0.14%) 34.7 23.0 LBF1 1201 537 563 1821 640 1908 (0.10%) 30.0 28.9 LBF2 1044 915 087 1906 512 1938 (0.10%) 33.2 21.9 DG74 726 434 840 1137 099 11 (0.001%) 36.4 9.1 DG78 729 045 788 1138 132 24 (0.002%) 33.3 8.3 UASB 941 623 397 1636 065 20 (0.001%) 60.0 10.0 Sample . Total assembly length (bp) . Protein coding genes . Number of ARG BLAST hits (% from protein coding genes) . Plasmid ARGs (%) . Chromosomal ARGs (%) . FW1 1285 684 727 2088 107 2407 (0.12%) 30.2 28.6 FW2 941 870 975 1796 492 2484 (0.14%) 34.7 23.0 LBF1 1201 537 563 1821 640 1908 (0.10%) 30.0 28.9 LBF2 1044 915 087 1906 512 1938 (0.10%) 33.2 21.9 DG74 726 434 840 1137 099 11 (0.001%) 36.4 9.1 DG78 729 045 788 1138 132 24 (0.002%) 33.3 8.3 UASB 941 623 397 1636 065 20 (0.001%) 60.0 10.0 Open in new tab Table 1. Genomic location of ARGs detected in total community metagenome sequencing data assembled with MEGAHIT. ARGs were detected by BLAST analysis of protein coding genes using a local SARG reference database. Plasmid or chromosomal location of the ARGs was predicted by PlasFlow. All ARG BLAST hits detected in MEGAHIT-assembled metagenome sequencing data are provided in Table S5 (Supporting Information). Sample . Total assembly length (bp) . Protein coding genes . Number of ARG BLAST hits (% from protein coding genes) . Plasmid ARGs (%) . Chromosomal ARGs (%) . FW1 1285 684 727 2088 107 2407 (0.12%) 30.2 28.6 FW2 941 870 975 1796 492 2484 (0.14%) 34.7 23.0 LBF1 1201 537 563 1821 640 1908 (0.10%) 30.0 28.9 LBF2 1044 915 087 1906 512 1938 (0.10%) 33.2 21.9 DG74 726 434 840 1137 099 11 (0.001%) 36.4 9.1 DG78 729 045 788 1138 132 24 (0.002%) 33.3 8.3 UASB 941 623 397 1636 065 20 (0.001%) 60.0 10.0 Sample . Total assembly length (bp) . Protein coding genes . Number of ARG BLAST hits (% from protein coding genes) . Plasmid ARGs (%) . Chromosomal ARGs (%) . FW1 1285 684 727 2088 107 2407 (0.12%) 30.2 28.6 FW2 941 870 975 1796 492 2484 (0.14%) 34.7 23.0 LBF1 1201 537 563 1821 640 1908 (0.10%) 30.0 28.9 LBF2 1044 915 087 1906 512 1938 (0.10%) 33.2 21.9 DG74 726 434 840 1137 099 11 (0.001%) 36.4 9.1 DG78 729 045 788 1138 132 24 (0.002%) 33.3 8.3 UASB 941 623 397 1636 065 20 (0.001%) 60.0 10.0 Open in new tab Bacterial hosts of ARGs MEGAHIT-assembled contigs that were not identified as plasmids were further annotated for their taxonomic affiliation (tool 3.9 in Fig. 1). Figure 5 depicts the frequency of specific host–ARG pairs identified for most abundant bacterial genera in digester feed. Gammaproteobacterial genera Stenotrophomonas and Acinetobacter clustered separately and were characterized by species-specific multidrug resistance efflux complexes and the corresponding regulatory systems. Stenotrophomonas carried subunits for SmeABC and SmeDEF efflux pumps as well as the SmeRS regulatory system formerly identified in Stenotrophomonas maltophilia (Alonso and Martínez 2000; Li, Zhang and Poole 2002). Acinetobacter carried subunits for AdeIJK and AdeFGH multidrug efflux pumps together with multidrug resistance genes adeB and abeM, transcriptional activator mexT and aminoglycoside resistance gene APH(3′)-Ia. In addition to Stenotrophomonas and Acinetobacter, other identified ARG host genera in the digester feed included Pseudomonas, Enterobacter, Klebsiella, Raoultella, Rahnella, Rouxiella, Pantoea, Serratia, Erwinia, Citrobacter, Leclercia and Lelliottia that formed a network by sharing connections with ARGs belonging mainly to multidrug resistance type. For example, Pseudomonas was characterized by the highest number of ARG connections carrying subunits of the MexEF-OprN efflux system; similarly, Enterobacter and Klebsiella shared connections to subunits of MdtABC-TolC and AcrAB-TolC efflux systems. In addition to multidrug resistance genes, genes conferring resistance to polymyxin were related to multiple host genera in digester feed: both rosA and rosB were found in Serratia, Rahnella and Rouxiella, arnA was found in Citrobacter and Pseudomonas, pmrE in Enterobacter and Klebsiella, and pmrA in Raoultella. Figure 5. Open in new tabDownload slide Most abundant bacterial host genera of non-plasmid ARGs detected in digester feed (FW1, FW2, LBF1, LBF2). ARGs were identified on MEGAHIT-assembled contigs by BLAST analysis of protein coding genes using a local SARG reference database (Table S5, Supporting Information). Non-plasmid contigs carrying ARGs were further annotated for their taxonomic affiliation using Kaiju. ARGs are depicted in ellipses colored by resistance type and bacterial genera are depicted in gray rectangles. The width of the connecting edge corresponds to detection frequency of individual host–ARG pairs. Figure 5. Open in new tabDownload slide Most abundant bacterial host genera of non-plasmid ARGs detected in digester feed (FW1, FW2, LBF1, LBF2). ARGs were identified on MEGAHIT-assembled contigs by BLAST analysis of protein coding genes using a local SARG reference database (Table S5, Supporting Information). Non-plasmid contigs carrying ARGs were further annotated for their taxonomic affiliation using Kaiju. ARGs are depicted in ellipses colored by resistance type and bacterial genera are depicted in gray rectangles. The width of the connecting edge corresponds to detection frequency of individual host–ARG pairs. Only a limited number of ARGs were detected and their host organisms taxonomically determined in digestion products (Table 2). Bacterial hosts of ARGs in digestion products included genera formerly identified in anaerobic digesters associated with microbial degradation processes, such as Fermentimonas, and several sulfate-reducing genera (Desulfobacca, Desulfococcus). More importantly, Burkholderia and Arcobacter (both carrying isoniazid resistance gene katG), Streptococcus (carrying MLS resistance gene mefA) and Escherichia (carrying MLS resistance gene mefB) were identified as ARG host organisms of clinical relevance. Table 2. Bacterial host organisms of non-plasmid ARGs detected in digestion products (DG74, DG78, UASB). ARGs were identified on MEGAHIT-assembled contigs by BLAST analysis of protein coding genes using a local SARG reference database (Table S5, Supporting Information). Non-plasmid contigs carrying ARGs were further annotated for their taxonomic affiliation using Kaiju. Kingdom/phylum/class . Family/genus . ARG . ARG type . Firmicutes not determined ANT(6)-Ia aminoglycoside Terrabacteria group not determined ANT(9)-Ia aminoglycoside Bacteroidetes Proteiniphilum bacA bacitracin Betaproteobacteria Thauera bacA bacitracin Deltaproteobacteria Desulfomicrobium bacA bacitracin Firmicutes Alkaliphilus bacA bacitracin Firmicutes Thermoanaerobacterium bcrA bacitracin Kiritimatiellaeota Kiritimatiella OXA-119 beta-lactam Betaproteobacteria Burkholderia katG isoniazid Epsilonproteobacteria Arcobacter katG isoniazid Bacteroidetes Fermentimonas ermF MLS Firmicutes not determined linB MLS Deltaproteobacteria Desulfococcus lnuF MLS Deltaproteobacteria Geobacter lnuF MLS Firmicutes Streptococcus mefA MLS Gammaproteobacteria Escherichia mefBa MLS Firmicutes Clostridium mel MLS Firmicutes Desulfitobacterium vatB MLS Betaproteobacteria Thauera acrB multidrug Deltaproteobacteria Desulfobacca acrB multidrug Deltaproteobacteria Syntrophobacter acrBa multidrug Planctomycetes Thermogutta acrB multidrug Firmicutes not determined lsaE multidrug Deltaproteobacteria Desulfobacca smeE multidrug Gammaproteobacteria Sedimenticola smeR multidrug Betaproteobacteria Roseateles arnA polymyxin Gammaproteobacteria Enterobacteriaceae arnA polymyxin Bacteria not determined sul1 sulfonamide Bacteria not determined sul1 sulfonamide Kingdom/phylum/class . Family/genus . ARG . ARG type . Firmicutes not determined ANT(6)-Ia aminoglycoside Terrabacteria group not determined ANT(9)-Ia aminoglycoside Bacteroidetes Proteiniphilum bacA bacitracin Betaproteobacteria Thauera bacA bacitracin Deltaproteobacteria Desulfomicrobium bacA bacitracin Firmicutes Alkaliphilus bacA bacitracin Firmicutes Thermoanaerobacterium bcrA bacitracin Kiritimatiellaeota Kiritimatiella OXA-119 beta-lactam Betaproteobacteria Burkholderia katG isoniazid Epsilonproteobacteria Arcobacter katG isoniazid Bacteroidetes Fermentimonas ermF MLS Firmicutes not determined linB MLS Deltaproteobacteria Desulfococcus lnuF MLS Deltaproteobacteria Geobacter lnuF MLS Firmicutes Streptococcus mefA MLS Gammaproteobacteria Escherichia mefBa MLS Firmicutes Clostridium mel MLS Firmicutes Desulfitobacterium vatB MLS Betaproteobacteria Thauera acrB multidrug Deltaproteobacteria Desulfobacca acrB multidrug Deltaproteobacteria Syntrophobacter acrBa multidrug Planctomycetes Thermogutta acrB multidrug Firmicutes not determined lsaE multidrug Deltaproteobacteria Desulfobacca smeE multidrug Gammaproteobacteria Sedimenticola smeR multidrug Betaproteobacteria Roseateles arnA polymyxin Gammaproteobacteria Enterobacteriaceae arnA polymyxin Bacteria not determined sul1 sulfonamide Bacteria not determined sul1 sulfonamide a ARG was affiliated to the respective genus on several contigs. Open in new tab Table 2. Bacterial host organisms of non-plasmid ARGs detected in digestion products (DG74, DG78, UASB). ARGs were identified on MEGAHIT-assembled contigs by BLAST analysis of protein coding genes using a local SARG reference database (Table S5, Supporting Information). Non-plasmid contigs carrying ARGs were further annotated for their taxonomic affiliation using Kaiju. Kingdom/phylum/class . Family/genus . ARG . ARG type . Firmicutes not determined ANT(6)-Ia aminoglycoside Terrabacteria group not determined ANT(9)-Ia aminoglycoside Bacteroidetes Proteiniphilum bacA bacitracin Betaproteobacteria Thauera bacA bacitracin Deltaproteobacteria Desulfomicrobium bacA bacitracin Firmicutes Alkaliphilus bacA bacitracin Firmicutes Thermoanaerobacterium bcrA bacitracin Kiritimatiellaeota Kiritimatiella OXA-119 beta-lactam Betaproteobacteria Burkholderia katG isoniazid Epsilonproteobacteria Arcobacter katG isoniazid Bacteroidetes Fermentimonas ermF MLS Firmicutes not determined linB MLS Deltaproteobacteria Desulfococcus lnuF MLS Deltaproteobacteria Geobacter lnuF MLS Firmicutes Streptococcus mefA MLS Gammaproteobacteria Escherichia mefBa MLS Firmicutes Clostridium mel MLS Firmicutes Desulfitobacterium vatB MLS Betaproteobacteria Thauera acrB multidrug Deltaproteobacteria Desulfobacca acrB multidrug Deltaproteobacteria Syntrophobacter acrBa multidrug Planctomycetes Thermogutta acrB multidrug Firmicutes not determined lsaE multidrug Deltaproteobacteria Desulfobacca smeE multidrug Gammaproteobacteria Sedimenticola smeR multidrug Betaproteobacteria Roseateles arnA polymyxin Gammaproteobacteria Enterobacteriaceae arnA polymyxin Bacteria not determined sul1 sulfonamide Bacteria not determined sul1 sulfonamide Kingdom/phylum/class . Family/genus . ARG . ARG type . Firmicutes not determined ANT(6)-Ia aminoglycoside Terrabacteria group not determined ANT(9)-Ia aminoglycoside Bacteroidetes Proteiniphilum bacA bacitracin Betaproteobacteria Thauera bacA bacitracin Deltaproteobacteria Desulfomicrobium bacA bacitracin Firmicutes Alkaliphilus bacA bacitracin Firmicutes Thermoanaerobacterium bcrA bacitracin Kiritimatiellaeota Kiritimatiella OXA-119 beta-lactam Betaproteobacteria Burkholderia katG isoniazid Epsilonproteobacteria Arcobacter katG isoniazid Bacteroidetes Fermentimonas ermF MLS Firmicutes not determined linB MLS Deltaproteobacteria Desulfococcus lnuF MLS Deltaproteobacteria Geobacter lnuF MLS Firmicutes Streptococcus mefA MLS Gammaproteobacteria Escherichia mefBa MLS Firmicutes Clostridium mel MLS Firmicutes Desulfitobacterium vatB MLS Betaproteobacteria Thauera acrB multidrug Deltaproteobacteria Desulfobacca acrB multidrug Deltaproteobacteria Syntrophobacter acrBa multidrug Planctomycetes Thermogutta acrB multidrug Firmicutes not determined lsaE multidrug Deltaproteobacteria Desulfobacca smeE multidrug Gammaproteobacteria Sedimenticola smeR multidrug Betaproteobacteria Roseateles arnA polymyxin Gammaproteobacteria Enterobacteriaceae arnA polymyxin Bacteria not determined sul1 sulfonamide Bacteria not determined sul1 sulfonamide a ARG was affiliated to the respective genus on several contigs. Open in new tab To further investigate the potential hosts of ARGs, the assembled contigs were binned into 201 MAGs (tools 3.11, 3.12 and 3.13 in Fig. 1) with >75% completeness and <25% redundancy (Table S7 and Figure S6, Supporting Information). Notably, MAGs found in digestion products (DG74, DG78, UASB) accounted for 75% of all assembled MAGs and did not contain ARGs. On the contrary, four MAGs containing ARGs were identified in digester feed (FW1, FW2, LBF1, LBF2) (Table 3). Interestingly, a genome belonging to the plant pathogenic genus Erwinia carried 19 ARGs conferring resistance to aminoglycoside, bacitracin, polymyxin, quinolone and sulfonamide antibiotics while also harboring several genes for multidrug resistance. Lactic acid bacteria Lactococcus lactis and Lactobacillus that are generally associated with probiotic features displayed resistance for MLS (L. lactis genes lmrC, lmrD, lmrP), tetracycline (L. lactis genes tetM, tetS) and trimethoprim (Lactobacillus gene dfrE). Additionally, a MAG belonging to the family Bifidobacteriaceae carried the inner membrane transporter gene mdsB of multidrug and metal efflux complex MdsABC. The four ARG-containing MAGs that were highly abundant in digester feed were not identified in digestion products (Figure S6, Supporting Information). Table 3. MAGs containing ARGs originating from digester feed (FW1, FW2, LBF1, LBF2). ARGs were not detected in MAGs from digestion products. MAGs were produced by MaxBin and metaBAT, dereplicated with dRep and evaluated with CheckM for genome size, GC%, completeness, contamination and estimated quality. ARGs that were detected in two separate copies in a given MAG are indicated by parenthesis (2×). All ARG-containing MAGs have been deposited to the JGI Integrated Microbial Genomes and Microbiomes (IMG/M). Additional information about all MAGs is available in Table S7 and Figure S5 (Supporting Information). MAG . IMG/M ID . Genome size (bp) . GC% . Completeness (%) . Contamination (%) . Estimated quality . ARG . ARG type . Erwinia 2772190821 5042 753 56.0 97.9 1.9 88.2 acrD aminoglycoside bacA bacitracin acrB multidrug cpxA multidrug cpxR (2×) multidrug CRP multidrug emrR multidrug H-NS multidrug mdtB (2×) multidrug mdtC multidrug arnA polymyxin phoP polymyxin pmrF polymyxin rosA polymyxin rosB polymyxin emrB quinolone folP sulfonamide Lactococcus lactis 2773857626 3416 378 33.7 92.0 5.7 63.4 lmrC MLS lmrD MLS lmrP multidrug tetM tetracycline tetS tetracycline Lactobacillus 2772190822 2569 052 30.9 97.0 3.6 79.1 dfrE trimethoprim Bifidobacteriaceae 2806310579 2067 166 64.6 90.7 2.4 78.6 mdsB multidrug MAG . IMG/M ID . Genome size (bp) . GC% . Completeness (%) . Contamination (%) . Estimated quality . ARG . ARG type . Erwinia 2772190821 5042 753 56.0 97.9 1.9 88.2 acrD aminoglycoside bacA bacitracin acrB multidrug cpxA multidrug cpxR (2×) multidrug CRP multidrug emrR multidrug H-NS multidrug mdtB (2×) multidrug mdtC multidrug arnA polymyxin phoP polymyxin pmrF polymyxin rosA polymyxin rosB polymyxin emrB quinolone folP sulfonamide Lactococcus lactis 2773857626 3416 378 33.7 92.0 5.7 63.4 lmrC MLS lmrD MLS lmrP multidrug tetM tetracycline tetS tetracycline Lactobacillus 2772190822 2569 052 30.9 97.0 3.6 79.1 dfrE trimethoprim Bifidobacteriaceae 2806310579 2067 166 64.6 90.7 2.4 78.6 mdsB multidrug Open in new tab Table 3. MAGs containing ARGs originating from digester feed (FW1, FW2, LBF1, LBF2). ARGs were not detected in MAGs from digestion products. MAGs were produced by MaxBin and metaBAT, dereplicated with dRep and evaluated with CheckM for genome size, GC%, completeness, contamination and estimated quality. ARGs that were detected in two separate copies in a given MAG are indicated by parenthesis (2×). All ARG-containing MAGs have been deposited to the JGI Integrated Microbial Genomes and Microbiomes (IMG/M). Additional information about all MAGs is available in Table S7 and Figure S5 (Supporting Information). MAG . IMG/M ID . Genome size (bp) . GC% . Completeness (%) . Contamination (%) . Estimated quality . ARG . ARG type . Erwinia 2772190821 5042 753 56.0 97.9 1.9 88.2 acrD aminoglycoside bacA bacitracin acrB multidrug cpxA multidrug cpxR (2×) multidrug CRP multidrug emrR multidrug H-NS multidrug mdtB (2×) multidrug mdtC multidrug arnA polymyxin phoP polymyxin pmrF polymyxin rosA polymyxin rosB polymyxin emrB quinolone folP sulfonamide Lactococcus lactis 2773857626 3416 378 33.7 92.0 5.7 63.4 lmrC MLS lmrD MLS lmrP multidrug tetM tetracycline tetS tetracycline Lactobacillus 2772190822 2569 052 30.9 97.0 3.6 79.1 dfrE trimethoprim Bifidobacteriaceae 2806310579 2067 166 64.6 90.7 2.4 78.6 mdsB multidrug MAG . IMG/M ID . Genome size (bp) . GC% . Completeness (%) . Contamination (%) . Estimated quality . ARG . ARG type . Erwinia 2772190821 5042 753 56.0 97.9 1.9 88.2 acrD aminoglycoside bacA bacitracin acrB multidrug cpxA multidrug cpxR (2×) multidrug CRP multidrug emrR multidrug H-NS multidrug mdtB (2×) multidrug mdtC multidrug arnA polymyxin phoP polymyxin pmrF polymyxin rosA polymyxin rosB polymyxin emrB quinolone folP sulfonamide Lactococcus lactis 2773857626 3416 378 33.7 92.0 5.7 63.4 lmrC MLS lmrD MLS lmrP multidrug tetM tetracycline tetS tetracycline Lactobacillus 2772190822 2569 052 30.9 97.0 3.6 79.1 dfrE trimethoprim Bifidobacteriaceae 2806310579 2067 166 64.6 90.7 2.4 78.6 mdsB multidrug Open in new tab DISCUSSION AD is widely used for treatment of solid organic waste, providing waste stabilization as well as energy production. More recently, the potential of AD to reduce antibiotic-resistant bacteria and ARGs has been investigated with mixed results (Youngquist, Mitchell and Cogger 2016). In this study, we examined the effect of co-digestion of food waste, paper and cardboard on microbial community composition and resistome using SSU rRNA gene sequencing, total community metagenome sequencing and quantitative PCR. The studied AD system was a lab-scale solid-state leach bed reactor with leachate recycle via a UASB reactor and has been thoroughly described elsewhere (Guilford 2017; Guilford et al. 2019). The system performed very well over 1.5 years of operation, also maintaining a stable microbial community (Lee 2018). Microbial community composition of digester feed and digestion products Microbial community composition has been suggested as one of the main drivers of ARGs in anaerobic digesters (Miller et al. 2013; Luo et al. 2017). The microbial communities of the digester feed—a mix of food waste and lignocellulosic fibers—revealed aerobes and facultative aerobes that were distinct from the strictly anaerobic bacteria and archaea inside the digester. The FW alone and the blended LBF samples had very similar taxonomic profiles indicating that food waste was the main contributor to the microbial community in the digester feed, consisting mainly of microorganisms native to food products or of the microbes that colonized food waste during collection and storage prior to AD. Previous studies on food microbiomes have highlighted the role of fermentative organisms such as lactic acid bacteria in fermentation processes (De Filippis, Parente and Ercolini 2018). Several OTUs belonging to the genera Lactobacillus and Leuconostoc were identified in the food waste used in this study, indicating the presence of fermented food products (e.g. cheese, sourdough) or the start of degradation processes during the collection, storage and pre-processing of food waste prior to AD. Lactobacillus and Leuconostoc may also indicate spoilage of meat as previously shown for vacuum-packed pork (Nieminen, Dalgaard and Björkroth 2016), beef (Ferrocino et al. 2016; Jääskeläinen et al. 2016; Säde et al. 2017), minced meat (Stoops et al. 2015) and sausages (Benson et al. 2014; Fougy et al. 2016). Besides lactic acid bacteria, several genera from the family Enterobacteriaceae were detected in digester feed, including Citrobacter, Enterobacter, Erwinia, Kluyvera, Pantoea, Serratia and Lelliottia. Jackson et al. (2013) identified Serratia, Erwinia, Enterobacter and Pantoea from leafy salad vegetables, which are also common in food waste. While many of the identified Enterobacteriaceae are widely recognized as plant pathogens or symbionts (e.g. Erwinia), potential human pathogens were also detected (e.g. Enterobacter, Serratia). Special attention should be given to the occurrence of ESKAPE pathogens—Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa and Enterobacter species—that are common causes of nosocomial infections and characterized by various AMR mechanisms (Rice 2008). A total of 15 OTUs annotated to S. aureus, K. pneumoniae, A. baumannii or P. aeruginosa were found in very low abundance in digester feed (each with relative abundance below 0.12% of total OTUs), while none of these were detected in digestion products. One hundred and two (102) OTUs were annotated to Enterobacter comprising up to 0.13% of OTUs in digester feed, while none was detected in digestion products. Thus, ESKAPE pathogens were detectable in relatively low abundance in the digester feed used in this study but did not survive anaerobic treatment. The microbial community after AD was completely different from that in the feed. Hydrolytic, fermenting, acidogenic, acetogenic and methanogenic microorganisms comprise a typical microbial community in a stable AD process. Representatives from the phyla Actinobacteria, Bacteroidetes, Chloroflexi, Firmicutes and Proteobacteria that have been connected to hydrolysis, fermentation and acidogenesis (Wang et al. 2018) were identified in the bacterial communities of the digestate samples in this study. As expected, digestates exhibited abundant archaeal communities responsible for methane production. The acetoclastic methanogen Methanosaeta was identified as the dominant archaeal genus (reaching up to 72.6% of total archaeal OTUs in DG74), which has been noted before for digesters with high acetate concentration (Supaphol et al. 2011; Lin et al. 2012). Members of the orders Methanobacteriales and Methanomicrobiales commonly identified in anaerobic digesters treating food waste (Yi et al. 2014; Kim et al. 2015; Peng et al. 2018) were also present in the digestates of the current study. Thus, a diverse methanogenic community achieved stable methane production, as evidenced by substantial feedstock degradation and biogas production rates (Guilford et al.2019). ARGs in digester feed and digestion products ARGs were identified in all samples, in digester feed as well as in AD products. Higher richness as well as higher relative abundance of individual ARGs per 16S rRNA gene was found in digester feed compared to digestion products, indicating the ability of AD to reduce ARG diversity and abundance. Total relative abundance of ARGs per 16S rRNA gene reached 0.84 ARG/16S rRNA gene in digester feed (LBF1), which is comparable to ARG levels in feces and wastewater from livestock farms (Li et al. 2015), while the respective value in digestion products was below 0.05 ARG/16S rRNA (0.04 ARG/16S rRNA in UASB), comparable to natural environments such as soils, sediments and river water (Li et al. 2015). Thus, AD clearly reduced ARG values from levels associated with high contamination to environmental background levels. Yet, although AD significantly reduced the abundance of ARGs, a limited number of ARGs were also detected in digestion products. Both total community metagenome sequencing and HT-qPCR array detected ARGs in digestion products that were low in abundance relative to the total microbial population (as shown by total community metagenome sequencing), but achieved high absolute abundance per gram dry mass of digestate (reaching 108 copies/g DW for individual genes and 109 copies/g DW in total) due to high bacterial loads in the digestates (as shown by HT-qPCR array). Same order of magnitude absolute abundance values of individual ARGs detected by qPCR have also been reported previously for anaerobic digestates of FW (Zhang et al. 2017). Thus, the effectiveness of AD on treating waste containing ARGs is a function of which microbes can grow in the AD system. In this case study, ARGs from food waste microbes were eliminated while ARGs present in the microbes involved in the digestion process were more abundant in the final digestate, and were at high absolute abundance owing to the high concentration of microbes generally after digestion. Future experiments should focus on post-treatment of digested solids, such as additional aerobic curing to reduce odor and stabilize waste, to determine the ultimate environmental load of ARGs associated with land application of anaerobic digestates. Special attention should be given to the mobility of ARGs that can be transferred from environmental bacteria to human pathogens or vice versa as shown previously (Forsberg et al. 2012). In this study, on average, 32% of ARGs in assembled contigs from digester feed were located on plasmids, indicating their potential mobility. Similar plasmid proportions from total assembled contigs have also been found in photobioreactor microbial communities (Nõlvak et al. 2018) and microbial mats inhabiting mine waters (Krawczyk, Lipinski and Dziembowski 2018). The HT-qPCR array detected a wide array of MGEs, including insertional sequences, integrases, transposases and plasmids mostly not only in food waste but also in digestates. Class 1 integron-integrase gene intI1 has been suggested as a proxy for anthropogenic pollution (Gillings et al. 2015) and was also found in digester feed (in the range of 105–106 copies/g DW) as well as in digestion products (108 copies/g DW) of this study. The occurrence of intI1 in digestion products correlates with the limited number of ARGs detected in the digestates of this study, showing that anthropogenic impact remains detectable after AD. Potential bacterial host organisms of ARGs Correlation and network analysis have been used in previous studies to link ARGs to their potential host organisms (Li et al. 2015; Zhang et al. 2016; Jang et al. 2017; Luo et al. 2017). Here, we used a combination of metagenomic assembly and binning methods to detect taxonomically annotated contigs and MAGs with ARGs. Taxonomic annotation of ARG-containing contigs showed several connections between bacterial genera and ARGs. Among others, the genera Enterobacter, Klebsiella, Acinetobacter and Pseudomonas that include medically critical ESKAPE-pathogens were connected to ARGs in digester feed conferring resistance to multidrug, MLS, polymyxin, aminoglycoside, triclosan and bacitracin antibiotics. The high number of multidrug resistance genes is especially worrisome as these genes have the potential to confer resistance to several types of antibiotics. Besides that, 12 out of 14 of the most abundant ARG-containing genera in digester feed also included resistance genes to polymyxin, a last-resort antibiotic used against Gram-negative bacteria. This highlights the spread of ARGs against last-resort antibiotics among various bacterial genera. ARG-containing genera that include clinically important species were also detected in digestion products, including Burkholderia, Arcobacter, Streptococcus and Escherichia indicating the risks associated with the possible use of anaerobic digestates. Using metagenomic binning, we were able to detect four ARG-containing MAGs originating from digester feed. A MAG annotated as Erwinia was found to carry 19 ARGs conferring resistance to several classes of antibiotics. Although Erwinia is commonly known as a plant pathogen, few cases of human infections have also been reported (O'Hara et al. 1998; Shin, Song and Ko 2008). As Erwinia is closely related to other Enterobacteriaceae that include several known human pathogens, horizontal transfer of ARGs from plant-pathogenic Erwinia to human pathogenic genera may occur. Additionally, MAGs annotated as lactic acid bacteria L. lactis and Lactobacillus were found to carry ARGs. These bacteria are extensively used for food fermentation processes and are often regarded as probiotics; however, increasing evidence suggests they are reservoirs of potentially transmissible ARGs and may play a crucial role in the acquisition of AMR via food (Devirgiliis, Zinno and Perozzi 2013). It is noteworthy that although 75% of all MAGs originated from digestion products, these MAGs did not contain ARGs. This correlates with the low abundance of ARGs in digestion products detected by other methods used in this study. Methodological aspects of ARG detection In this study, total community metagenome data and HT-qPCR data were used to characterize the diversity and abundance of ARGs. While the metagenomic approach has the potential to detect higher diversity of ARGs due to large reference databases, qPCR of ARGs, although being limited in a number of ARGs even in high-throughput applications, can provide higher sensitivity (Ju and Zhang 2015). Our results showed that both metagenomic and HT-qPCR approach detected a broad spectrum of ARGs with higher diversity of ARGs in digester feed compared to digestion products. Among the limited number of ARGs detected in digestion products, mphA, lnuB, catQ and tet44 were found to be unique to digestate by both metagenomic and HT-qPCR approach. Additionally, HT-qPCR provided absolute abundance values for individual ARGs that could be related back to the number of gene copies present per gram of digester feed and in digested solids. This type of quantitative information on ARGs is required for assessing the risks associated with food waste and the use of anaerobic digestates as agricultural fertilizers. Thus, a combination of shotgun metagenomic and qPCR approach is recommended for a comprehensive view of a sample's resistome and for the assessment of AMR risks. In addition to the annotation of short sequencing reads, we used assembled contigs to provide genomic context of ARGs, which cannot be determined from short sequencing reads. Although assembled contigs provide better resolution for taxonomic and functional annotation than short sequencing reads, assembly may also introduce biases (Ju and Zhang 2015). Similar problem has been noted for metagenomic binning, which often has limited power for analysis of complex microbial communities. This was also evident in the results of the current study with 75% of MAGs (151 MAGs) originating from digestion products and only 25% of MAGs (50 MAGs) originating from the heterogeneous microbial communities of digester feed. Thus, our findings should be interpreted in light of the limitations of current methods used for taxonomic and functional analysis of complex microbial communities. In conclusion, using a combination of metagenome sequencing, assembly and binning as well as quantitative PCR analysis is recommended to estimate the diversity and abundance of ARGs and to situate the ARGs within their genomic context and potential host organisms. The detection of potential host organisms and genomic context allows for a more accurate risk evaluation associated with solid organic wastes. ACKNOWLEDGEMENTS The authors wish to thank Dr Robert Stedtfeld from Michigan State University for his help in advising and running the HT-qPCR analysis. FUNDING This work was supported by the Natural Sciences and Engineering Research Council of Canada (Collaborative Research and Development Grant) and Miller Waste Systems Inc. Conflicts of interest. None declared. REFERENCES Alonso A , Martínez JL . Cloning and characterization of SmeDEF, a novel multidrug efflux pump from Stenotrophomonas maltophilia . Antimicrob Agents Ch . 2000 ; 44 : 3079 – 86 . Google Scholar Crossref Search ADS WorldCat Amann RI , Ludwig W , Schleifer KH . Phylogenetic identification and in situ detection of individual microbial cells without cultivation . Microbiol Rev . 1995 ; 59 : 143 – 69 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Bengtsson-Palme J . Antibiotic resistance in the food supply chain: where can sequencing and metagenomics aid risk assessment? . Curr Opin Food Sci . 2017 ; 14 : 66 – 71 . Google Scholar Crossref Search ADS WorldCat Bengtsson-Palme J , Hartmann M , Eriksson KM et al. . Metaxa 2: improved identification and taxonomic classification of small and large subunit rRNA in metagenomic data . Mol Ecol Resour . 2015 ; 15 : 1403 – 14 . Google Scholar Crossref Search ADS PubMed WorldCat Bengtsson-Palme J , Kristiansson E , Larsson DGJ . Environmental factors influencing the development and spread of antibiotic resistance . FEMS Microbiol Rev . 2018 ; 42 : fux053 . Google Scholar Crossref Search ADS WorldCat Benson AK , David JRD , Gilbreth SE et al. . Microbial successions are associated with changes in chemical profiles of a model refrigerated fresh pork sausage during an 80-day shelf life study . Appl Environ Microb . 2014 ; 80 : 5178 – 94 . Google Scholar Crossref Search ADS WorldCat Berendonk TU , Manaia CM , Merlin C et al. . Tackling antibiotic resistance: the environmental framework . Nat Rev Microbiol . 2015 ; 13 : 310 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat Bolger AM , Lohse M , Usadel B . Trimmomatic: a flexible trimmer for Illumina sequence data . Bioinformatics . 2014 ; 30 : 2114 – 20 . Google Scholar Crossref Search ADS PubMed WorldCat Braguglia CM , Gallipoli A , Gianico A et al. . Anaerobic bioconversion of food waste into energy: a critical review . Bioresource Technol . 2018 ; 248 : 37 – 56 . Google Scholar Crossref Search ADS WorldCat Caporaso JG , Kuczynski J , Stombaugh J et al. . QIIME allows analysis of high-throughput community sequencing data . Nat Methods . 2010 ; 7 : 335 – 6 . Google Scholar Crossref Search ADS PubMed WorldCat Chajęcka-Wierzchowska W , Zadernowska A , Łaniewska-Trokenheim Ł . Diversity of antibiotic resistance genes in Enterococcus strains isolated from ready-to-eat meat products . J Food Sci . 2016 ; 81 : M2799 – 807 . Google Scholar Crossref Search ADS PubMed WorldCat Chen S , Huang T , Zhou Y et al. . AfterQC: automatic filtering, trimming, error removing and quality control for fastq data . BMC Bioinformatics . 2017 ; 18 : 80 . Google Scholar Crossref Search ADS PubMed WorldCat Czekalski N , Sigdel R , Birtel J et al. . Does human activity impact the natural antibiotic resistance background? Abundance of antibiotic resistance genes in 21 Swiss lakes . Environ Int . 2015 ; 81 : 45 – 55 . Google Scholar Crossref Search ADS PubMed WorldCat D'Costa VM , King CE , Kalan L et al. . Antibiotic resistance is ancient . Nature . 2011 ; 477 : 457 – 61 . Google Scholar Crossref Search ADS PubMed WorldCat De Filippis F , Parente E , Ercolini D . Recent past, present, and future of the food microbiome . Annu Rev Food Sci Technol . 2018 ; 9 : 589 – 608 . Google Scholar Crossref Search ADS PubMed WorldCat Devirgiliis C , Zinno P , Perozzi G . Update on antibiotic resistance in foodborne Lactobacillus and Lactococcus species . Front Microbiol . 2013 ; 4 : 301 . Google Scholar Crossref Search ADS PubMed WorldCat Diarra MS , Delaquis P , Rempel H et al. . Antibiotic resistance and diversity of Salmonella enterica serovars associated with broiler chickens . J Food Protect . 2014 ; 77 : 40 – 9 . Google Scholar Crossref Search ADS WorldCat Durso LM , Wedin DA , Gilley JE et al. . Assessment of selected antibiotic resistances in ungrazed native Nebraska prairie soils . J Environ Qual . 2016 ; 45 : 454 – 62 . Google Scholar Crossref Search ADS PubMed WorldCat Edgar RC . Search and clustering orders of magnitude faster than BLAST . Bioinformatics . 2010 ; 26 : 2460 – 1 . Google Scholar Crossref Search ADS PubMed WorldCat Elhadi N . Prevalence and antimicrobial resistance of Salmonella spp. in raw retail frozen imported freshwater fish to Eastern Province of Saudi Arabia . Asian Pac J Trop Biomed . 2014 ; 4 : 234 – 8 . Google Scholar Crossref Search ADS PubMed WorldCat Engelbrektson A , Kunin V , Wrighton KC et al. . Experimental factors affecting PCR-based estimates of microbial species richness and evenness . ISME J . 2010 ; 4 : 642 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat Eren AM , Esen ÖC , Quince C et al. . Anvi'o: an advanced analysis and visualization platform for ‘omics data . PeerJ . 2015 ; 3 : e1319 . Google Scholar Crossref Search ADS PubMed WorldCat Ferris MJ , Muyzer G , Ward DM . Denaturing gradient gel electrophoresis profiles of 16S rRNA-defined populations inhabiting a hot spring microbial mat community . Appl Environ Microb . 1996 ; 62 : 340 – 6 . Google Scholar Crossref Search ADS WorldCat Ferrocino I , Greppi A , La Storia A et al. . Impact of nisin-activated packaging on microbiota of beef burgers during storage . Appl Environ Microb . 2016 ; 82 : 549 – 59 . Google Scholar Crossref Search ADS WorldCat Forsberg KJ , Reyes A , Wang B et al. . The shared antibiotic resistome of soil bacteria and human pathogens . Science . 2012 ; 337 : 1107 – 11 . Google Scholar Crossref Search ADS PubMed WorldCat Fougy L , Desmonts M-H , Coeuret G et al. . Reducing salt in raw pork sausages increases spoilage and correlates with reduced bacterial diversity . Appl Environ Microb . 2016 ; 82 : 3928 – 39 . Google Scholar Crossref Search ADS WorldCat Friedman M . Antibiotic-resistant bacteria: prevalence in food and inactivation by food-compatible compounds and plant extracts . J Agric Food Chem . 2015 ; 63 : 3805 – 22 . Google Scholar Crossref Search ADS PubMed WorldCat Gillings MR , Gaze WH , Pruden A et al. . Using the class 1 integron-integrase gene as a proxy for anthropogenic pollution . ISME J . 2015 ; 9 : 1269 – 79 . Google Scholar Crossref Search ADS PubMed WorldCat Gu Z , Eils R , Schlesner M . Complex heatmaps reveal patterns and correlations in multidimensional genomic data . Bioinformatics . 2016 ; 32 : 2847 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Guilford NGH . The anaerobic digestion of organic solid wastes of variable composition. Ph.D. Thesis. University of Toronto 2017 . Guilford N , Lee H , Kanger K et al. . Solid state anaerobic digestion of mixed organic waste: the synergistic effect of food waste addition on the destruction of paper and cardboard . Environ Sci Technol . 2019 ; 53 : 12677 – 87 . Google Scholar Crossref Search ADS PubMed WorldCat Hatosy SM , Martiny AC . The ocean as a global reservoir of antibiotic resistance genes . Appl Environ Microb . 2015 ; 81 : 7593 – 9 . Google Scholar Crossref Search ADS WorldCat Holvoet K , Sampers I , Callens B et al. . Moderate prevalence of antimicrobial resistance in Escherichia coli isolates from lettuce, irrigation water, and soil . Appl Environ Microb . 2013 ; 79 : 6677 – 83 . Google Scholar Crossref Search ADS WorldCat Hoornweg D , Bhada-Tata P . What a Waste : A Global Review of Solid Waste Management . Washington, DC : World Bank , 2012 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Huijbers PMC , Blaak H , de Jong MCM et al. . Role of the environment in the transmission of antimicrobial resistance to humans: a review . Environ Sci Technol . 2015 ; 49 : 11993 – 2004 . Google Scholar Crossref Search ADS PubMed WorldCat Jääskeläinen E , Hultman J , Parshintsev J et al. . Development of spoilage bacterial community and volatile compounds in chilled beef under vacuum or high oxygen atmospheres . Int J Food Microbiol . 2016 ; 223 : 25 – 32 . Google Scholar Crossref Search ADS PubMed WorldCat Jackson CR , Randolph KC , Osborn SL et al. . Culture dependent and independent analysis of bacterial communities associated with commercial salad leaf vegetables . BMC Microbiol . 2013 ; 13 : 274 . Google Scholar Crossref Search ADS PubMed WorldCat Jang HM , Shin J , Choi S et al. . Fate of antibiotic resistance genes in mesophilic and thermophilic anaerobic digestion of chemically enhanced primary treatment (CEPT) sludge . Bioresource Technol . 2017 ; 244 : 433 – 44 . Google Scholar Crossref Search ADS WorldCat Ju F , Zhang T . Experimental design and bioinformatics analysis for the application of metagenomics in environmental sciences and biotechnology . Environ Sci Technol . 2015 ; 49 : 12628 – 40 . Google Scholar Crossref Search ADS PubMed WorldCat Kang DD , Froula J , Egan R et al. . MetaBAT, an efficient tool for accurately reconstructing single genomes from complex microbial communities . PeerJ . 2015 ; 3 : e1165 . Google Scholar Crossref Search ADS PubMed WorldCat Kim YM , Jang HM , Lee K et al. . Changes in bacterial and archaeal communities in anaerobic digesters treating different organic wastes . Chemosphere . 2015 ; 141 : 134 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat Krawczyk PS , Lipinski L , Dziembowski A . PlasFlow: predicting plasmid sequences in metagenomic data using genome signatures . Nucleic Acids Res . 2018 ; 46 : e35 . Google Scholar Crossref Search ADS PubMed WorldCat Lee H . Characterization of the microbial community in a sequentially fed anaerobic digester treating solid organic waste. MAS Thesis. University of Toronto 2018 . Lee J , Shin SG , Jang HM et al. . Characterization of antibiotic resistance genes in representative organic solid wastes: food waste-recycling wastewater, manure, and sewage sludge . Sci Total Environ . 2017 ; 579 : 1692 – 8 . Google Scholar Crossref Search ADS PubMed WorldCat Li X-Z , Zhang L , Poole K . SmeC, an outer membrane multidrug efflux protein of Stenotrophomonas maltophilia . Antimicrob Agents Ch . 2002 ; 46 : 333 – 43 . Google Scholar Crossref Search ADS WorldCat Li B , Yang Y , Ma L et al. . Metagenomic and network analysis reveal wide distribution and co-occurrence of environmental antibiotic resistance genes . ISME J . 2015 ; 9 : 2490 – 502 . Google Scholar Crossref Search ADS PubMed WorldCat Li D , Luo R , Liu C-M et al. . MEGAHIT v1.0: a fast and scalable metagenome assembler driven by advanced methodologies and community practices . Methods . 2016 ; 102 : 3 – 11 . Google Scholar Crossref Search ADS PubMed WorldCat Lin J , Zuo J , Ji R et al. . Methanogenic community dynamics in anaerobic co-digestion of fruit and vegetable waste and food waste . J Environ Sci . 2012 ; 24 : 1288 – 94 . Google Scholar Crossref Search ADS WorldCat Looft T , Johnson TA , Allen HK et al. . In-feed antibiotic effects on the swine intestinal microbiome . Proc Natl Acad Sci USA . 2012 ; 109 : 1691 – 6 . Google Scholar Crossref Search ADS PubMed WorldCat Luo G , Li B , Li L-G et al. . Antibiotic resistance genes and correlations with microbial community and metal resistance genes in full-scale biogas reactors as revealed by metagenomic analysis . Environ Sci Technol . 2017 ; 51 : 4069 – 80 . Google Scholar Crossref Search ADS PubMed WorldCat Markowitz VM , Chen I-MA , Chu K et al. . IMG/M 4 version of the integrated metagenome comparative analysis system . Nucleic Acids Res . 2014 ; 42 : D568 – 73 . Google Scholar Crossref Search ADS PubMed WorldCat Marti R , Scott A , Tien Y-C et al. . Impact of manure fertilization on the abundance of antibiotic-resistant bacteria and frequency of detection of antibiotic resistance genes in soil and on vegetables at harvest . Appl Environ Microb . 2013 ; 79 : 5701 – 9 . Google Scholar Crossref Search ADS WorldCat Menzel P , Ng KL , Krogh A . Fast and sensitive taxonomic classification for metagenomics with Kaiju . Nat Commun . 2016 ; 7 : 11257 . Google Scholar Crossref Search ADS PubMed WorldCat Mikheenko A , Saveliev V , Gurevich A . MetaQUAST: evaluation of metagenome assemblies . Bioinformatics . 2016 ; 32 : 1088 – 90 . Google Scholar Crossref Search ADS PubMed WorldCat Miller JH , Novak JT , Knocke WR et al. . Effect of silver nanoparticles and antibiotics on antibiotic resistance genes in anaerobic digestion . Water Environ Res . 2013 ; 85 : 411 – 21 . Google Scholar Crossref Search ADS PubMed WorldCat Moore JE , Huang J , Yu P et al. . High diversity of bacterial pathogens and antibiotic resistance in salmonid fish farm pond water as determined by molecular identification employing 16S rDNA PCR, gene sequencing and total antibiotic susceptibility techniques . Ecotox Environ Safe . 2014 ; 108 : 281 – 6 . Google Scholar Crossref Search ADS WorldCat Nieminen TT , Dalgaard P , Björkroth J . Volatile organic compounds and Photobacterium phosphoreum associated with spoilage of modified-atmosphere-packaged raw pork . Int J Food Microbiol . 2016 ; 218 : 86 – 95 . Google Scholar Crossref Search ADS PubMed WorldCat Nõlvak H , Truu M , Oopkaup K et al. . Reduction of antibiotic resistome and integron-integrase genes in laboratory-scale photobioreactors treating municipal wastewater . Water Res . 2018 ; 142 : 363 – 72 . Google Scholar Crossref Search ADS PubMed WorldCat Noyes NR , Yang X , Linke LM et al. . Resistome diversity in cattle and the environment decreases during beef production . Elife . 2016 ; 5 : e13195 . Google Scholar Crossref Search ADS PubMed WorldCat Nurk S , Meleshko D , Korobeynikov A et al. . metaSPAdes: a new versatile metagenomic assembler . Genome Res . 2017 ; 27 : 824 – 34 . Google Scholar Crossref Search ADS PubMed WorldCat O'Hara CM , Steigerwalt AG , Hill BC et al. . First report of a human isolate of Erwinia persicinus . J Clin Microbiol . 1998 ; 36 : 248 – 50 . Google Scholar Crossref Search ADS PubMed WorldCat O'Neill J . Tackling Drug-Resistant Infections Globally: Final Report and Recommendations. The Review on Antimicrobial Resistance , London : HM Government and Wellcome Trust , 2016 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Oliveira M , Viñas I , Usall J et al. . Presence and survival of Escherichia coli O157:H7 on lettuce leaves and in soil treated with contaminated compost and irrigation water . Int J Food Microbiol . 2012 ; 156 : 133 – 40 . Google Scholar Crossref Search ADS PubMed WorldCat Olm MR , Brown CT , Brooks B et al. . dRep: a tool for fast and accurate genomic comparisons that enables improved genome recovery from metagenomes through de-replication . ISME J . 2017 ; 11 : 2864 – 8 . Google Scholar Crossref Search ADS PubMed WorldCat Parks DH , Imelfort M , Skennerton CT et al. . CheckM: assessing the quality of microbial genomes recovered from isolates, single cells, and metagenomes . Genome Res . 2015 ; 25 : 1043 – 55 . Google Scholar Crossref Search ADS PubMed WorldCat Parks DH , Rinke C , Chuvochina M et al. . Recovery of nearly 8,000 metagenome-assembled genomes substantially expands the tree of life . Nat Microbiol . 2017 ; 2 : 1533 – 42 . Google Scholar Crossref Search ADS PubMed WorldCat Peng X , Zhang S , Li L et al. . Long-term high-solids anaerobic digestion of food waste: effects of ammonia on process performance and microbial community . Bioresource Technol . 2018 ; 262 : 148 – 58 . Google Scholar Crossref Search ADS WorldCat Pepper IL , Brooks JP , Gerba CP . Antibiotic resistant bacteria in municipal wastes: is there reason for concern? . Environ Sci Technol . 2018 ; 52 : 3949 – 59 . Google Scholar Crossref Search ADS PubMed WorldCat Perron GG , Whyte L , Turnbaugh PJ et al. . Functional characterization of bacteria isolated from ancient arctic soil exposes diverse resistance mechanisms to modern antibiotics . PLoS One . 2015 ; 10 : e0069533 . Google Scholar Crossref Search ADS PubMed WorldCat Pu C , Liu H , Ding G et al. . Impact of direct application of biogas slurry and residue in fields: in situ analysis of antibiotic resistance genes from pig manure to fields . J Hazard Mater . 2018 ; 344 : 441 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Rahube TO , Marti R , Scott A et al. . Impact of fertilizing with raw or anaerobically digested sewage sludge on the abundance of antibiotic-resistant coliforms, antibiotic resistance genes, and pathogenic bacteria in soil and on vegetables at harvest . Appl Environ Microb . 2014 ; 80 : 6898 – 907 . Google Scholar Crossref Search ADS WorldCat Rice LB . Federal funding for the study of antimicrobial resistance in nosocomial pathogens: no ESKAPE . J Infect Dis . 2008 ; 197 : 1079 – 81 . Google Scholar Crossref Search ADS PubMed WorldCat Robinson TP , Bu DP , Carrique-Mas J et al. . Antibiotic resistance is the quintessential One Health issue . Trans R Soc Trop Med Hyg . 2016 ; 110 : 377 – 80 . Google Scholar Crossref Search ADS PubMed WorldCat Säde E , Penttinen K , Björkroth J et al. . Exploring lot-to-lot variation in spoilage bacterial communities on commercial modified atmosphere packaged beef . Food Microbiol . 2017 ; 62 : 147 – 52 . Google Scholar Crossref Search ADS PubMed WorldCat Shannon P , Markiel A , Ozier O et al. . Cytoscape: a software environment for integrated models of biomolecular interaction networks . Genome Res . 2003 ; 13 : 2498 – 504 . Google Scholar Crossref Search ADS PubMed WorldCat Shin SY , Song JH , Ko KS . First report of human infection due to Erwinia tasmaniensis-like organism . Int J Infect Dis . 2008 ; 12 : e329 – 30 . Google Scholar Crossref Search ADS WorldCat Silveira-Filho VM , Luz IS , Campos APF et al. . Antibiotic resistance and molecular analysis of Staphylococcus aureus isolated from cow's milk and dairy products in Northeast Brazil . J Food Protect . 2014 ; 77 : 583 – 91 . Google Scholar Crossref Search ADS WorldCat Stedtfeld RD , Guo X , Stedtfeld TM et al. . Primer set 2.0 for highly parallel qPCR array targeting antibiotic resistance genes and mobile genetic elements . FEMS Microbiol Ecol . 2018 ; 94 : fiy130 . Google Scholar Crossref Search ADS WorldCat Stokes HW , Gillings MR . Gene flow, mobile genetic elements and the recruitment of antibiotic resistance genes into Gram-negative pathogens . FEMS Microbiol Rev . 2011 ; 35 : 790 – 819 . Google Scholar Crossref Search ADS PubMed WorldCat Stoops J , Ruyters S , Busschaert P et al. . Bacterial community dynamics during cold storage of minced meat packaged under modified atmosphere and supplemented with different preservatives . Food Microbiol . 2015 ; 48 : 192 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Supaphol S , Jenkins SN , Intomo P et al. . Microbial community dynamics in mesophilic anaerobic co-digestion of mixed waste . Bioresource Technol . 2011 ; 102 : 4021 – 7 . Google Scholar Crossref Search ADS WorldCat Vasco-Correa J , Khanal S , Manandhar A et al. . Anaerobic digestion for bioenergy production: global status, environmental and techno-economic implications, and government policies . Bioresource Technol . 2018 ; 247 : 1015 – 26 . Google Scholar Crossref Search ADS WorldCat Wang F-H , Qiao M , Su J-Q et al. . High-throughput profiling of antibiotic resistance genes in urban park soils with reclaimed water irrigation . Environ Sci Technol . 2014 ; 48 : 9079 – 85 . Google Scholar Crossref Search ADS PubMed WorldCat Wang P , Wang H , Qiu Y et al. . Microbial characteristics in anaerobic digestion process of food waste for methane production–a review . Bioresource Technol . 2018 ; 248 : 29 – 36 . Google Scholar Crossref Search ADS WorldCat WHO . Global Action Plan on Antimicrobial Resistance . Geneva, Switzerland : World Health Organization , 2015 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Wu Y-W , Simmons BA , Singer SW . MaxBin 2.0: an automated binning algorithm to recover genomes from multiple metagenomic datasets . Bioinformatics . 2016 ; 32 : 605 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat Yang B , Cui Y , Shi C et al. . Counts, serotypes, and antimicrobial resistance of Salmonella isolates on retail raw poultry in the People's Republic of China . J Food Protect . 2014 ; 77 : 894 – 902 . Google Scholar Crossref Search ADS WorldCat Yang X , Noyes NR , Doster E et al. . Use of metagenomic shotgun sequencing technology to detect foodborne pathogens within the microbiome of the beef production chain . Appl Environ Microb . 2016a ; 82 : 2433 – 43 . Google Scholar Crossref Search ADS WorldCat Yang Y , Jiang X , Cai B et al. . ARGs-OAP: online analysis pipeline for antibiotic resistance genes detection from metagenomic data using an integrated structured ARG-database . Bioinformatics . 2016b ; 32 : 2346 – 51 . Google Scholar Crossref Search ADS WorldCat Yi J , Dong B , Xue Y et al. . Microbial community dynamics in batch high-solid anaerobic digestion of food waste under mesophilic conditions . J Microbiol Biotechnol . 2014 ; 24 : 270 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Youngquist CP , Mitchell SM , Cogger CG . Fate of antibiotics and antibiotic resistance during digestion and composting: a review . J Environ Qual . 2016 ; 45 : 537 . Google Scholar Crossref Search ADS PubMed WorldCat Yu Y , Lee C , Kim J et al. . Group-specific primer and probe sets to detect methanogenic communities using quantitative real-time polymerase chain reaction . Biotechnol Bioeng . 2005 ; 89 : 670 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Zhang J , Chen M , Sui Q et al. . Fate of antibiotic resistance genes and its drivers during anaerobic co-digestion of food waste and sewage sludge based on microwave pretreatment . Bioresource Technol . 2016 ; 217 : 28 – 36 . Google Scholar Crossref Search ADS WorldCat Zhang J , Mao F , Loh K-C et al. . Evaluating the effects of activated carbon on methane generation and the fate of antibiotic resistant genes and class I integrons during anaerobic digestion of solid organic wastes . Bioresource Technol . 2018 ; 249 : 729 – 36 . Google Scholar Crossref Search ADS WorldCat Zhang J , Zhang L , Loh K-C et al. . Enhanced anaerobic digestion of food waste by adding activated carbon: fate of bacterial pathogens and antibiotic resistance genes . Biochem Eng J . 2017 ; 128 : 19 – 25 . Google Scholar Crossref Search ADS WorldCat © FEMS 2020. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Antibiotic resistome and microbial community structure during anaerobic co-digestion of food waste, paper and cardboard
JF - FEMS Microbiology Ecology
DO - 10.1093/femsec/fiaa006
DA - 2020-02-01
UR - https://www.deepdyve.com/lp/oxford-university-press/antibiotic-resistome-and-microbial-community-structure-during-H2YNWMFIS3
VL - 96
IS - 2
DP - DeepDyve
ER -