bs_bs_banner Chicken liver is a potential reservoir of bacteriophages and phage-derived particles containing antibiotic resistance genes Pedro Blanco-Picazo,1 Clara Go  mez-Go mez,1 the presence of phages and phage particles contain- Sergi Aguilo  -Castillo, Dietmar Ferna 2 ndez-Orth,3 ing antibiotic resistance genes (ARGs). Both bacteria Marta Cerda -Cue llar, 4 Maite Muniesa1 and and phages were detected in liver tissues, including Lorena Rodrıguez-Rubio * 1 those dissected under sterile conditions. The phages 1 Department de Gene  tica, Microbiologia i Estadıstica, were able to infect Escherichia coli and showed a Universitat de Barcelona, Diagonal 643, Planta 0, Siphovirus morphology. The chicken livers con- Barcelona, 08028, Spain. tained from 103 to 106 phage particles per g, which 2 Spanish National Bioinformatics Institute (INB)/ELIXIR- carried a range of ARGs (blaTEM, blaCTx-M-1, sul1, ES, Barcelona Supercomputing Center, Barcelona, qnrA, armA and tetW) detected by qPCR. The pres- Spain. ence of phages in chicken liver, mostly infecting E. 3 Department of Bioinformatics and Molecular Biology, coli, was confirmed by metagenomic analysis, Cerba Internacional Pl. Ramon Llull, 7-10, Sabadell, although this technique was not sufficiently sensitive Barcelona, 08203, Spain. to identify ARGs. In addition, ARG-carrying phages 4 IRTA, Centre de Recerca en Sanitat Animal (CReSA, were detected in chicken faeces by qPCR in a previ- IRTA-UAB), Campus de la Universitat Auto  noma de ous study of the group. Comparison of the viromes Barcelona, Bellaterra, Barcelona, 08193, Spain. of faeces and liver showed a strong coincidence of species, which suggests that the phages found in the liver originate in faeces. These findings suggests Summary that phages, like bacteria, can translocate from the gut to the liver, which may therefore constitute a Poultry meat production is one of the most important potential reservoir of antibiotic resistance genes. agri-food industries in the world. The selective pres- sure exerted by widespread prophylactic or thera- peutic use of antibiotics in intensive chicken farming Introduction favours the development of drug resistance in bacte- rial populations. Chicken liver, closely connected The European Union (EU) is one of the world’s largest with the intestinal tract, has been directly involved in producers of poultry meat and a net exporter of poultry food-borne infections and found to be contaminated products, with an annual production of around 13.4 mil- with pathogenic bacteria, including Campylobacter lion tonnes (Poultry | European Commission). Poultry is and Salmonella. In this study, 74 chicken livers, the only meat category whose production has expanded divided into sterile and non-sterile groups, were ana- during the Covid-19 pandemic and the only one lysed, not only for microbial indicators but also for expected to grow between 2020 and 2030 in the EU (European Commission). The consumption of poultry meat is also projected to increase on a global scale Received 9 December, 2021; accepted 15 March, 2022. (OECD-FAO AGRICULTURAL OUTLOOK 2018–2027), For correspondence. *E-mail lorenarodriguez@ub.edu; Tel. as it is regarded as a cheap, healthy and sustainable +34934034625; Fax +34934039047. Microbial Biotechnology (2022) 15(9), 2464–2475 product by consumers. doi:10.1111/1751-7915.14056 Intensive poultry production, one of the fastest growing Funding information industries in the world, strongly depends on antimicro- This work was supported by the Spanish Ministerio de Innovacion y Ciencia (AGL2016-75536-P and PID2020-113355GB-I00), the bials to prevent or treat diseases. In several countries, n (AEI) and the European regional Agencia Estatal de Investigacio including Spain, the poultry production sectors are com- fund (ERF). The study was partially supported by the Union Iberoa- mitted to a responsible use of antibiotics and are gradu- mericana de Universidades (UIU-OHOW) and the Generalitat de Catalunya (2017SGR170). CERCA Programme from the Generalitat ally reducing their application (Programa REDUCE de Catalunya is also acknowledged. P. B-P has a grant from the Pollos Broiler | PRAN). Nevertheless, in other parts of Spanish Ministerio de Economıa, Industria y Competitividad (BES- the world, the usage of antibiotics as prophylactics, or 2017-081296). L. R.-R is a lecturer of the Serra-Hu nter program, Generalitat de Catalunya. even as growth promoters, a forbidden practice in the ª 2022 The Authors. Microbial Biotechnology published by Society for Applied Microbiology and John Wiley & Sons Ltd. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and repro- duction in any medium, provided the original work is properly cited. Phages with ARGs in chicken liver 2465 EU and USA, is difficult to eradicate due to the fear of detected in different human systems (Brown-Jaque infections spreading through the poultry farms. The con- et al., 2016; Blanco-Picazo et al., 2020), ARG-carrying sequence of such an intensive use of antibiotics is a phages are also expected to occur in animal organs. In very high selective pressure on bacterial populations chickens, the largest source of meat intended for human (Xiong et al., 2018), which promotes mutations, the consumption, the liver is of particular concern as it is emergence of antibiotic resistances and the dissemina- closely connected with the gastrointestinal tract and has tion of antibiotic resistance genes (ARGs) by mobile been identified as a vehicle for food-borne infection genetic elements (World Health Organization, 2018). (Lanier et al., 2018). Pathogens such as Campylobacter Many pathogenic or commensal bacteria isolated from or Salmonella are thought to spread from the gastroin- poultry encode antimicrobial resistances (Nhung et al., testinal tract to the liver through the biliary, lymphatic, or 2017), which can be transmitted to humans by direct or vascular systems (Lanier et al., 2018). The ready detec- indirect contact (e.g., through food, water and the envi- tion of bacterial pathogens in internal tissues of chicken ronment). Multi-drug resistant bacteria arising from ani- liver (Firlieyanti et al., 2016; Lanier et al., 2018), and the mal production practices have been found in everyday numerous studies of the circulation of phages in different consumer products (Antimicrobial Resistance in the organs (Ka zmierczak et al., 2021), including liver Environment Summary Report of an FAO Meeting of (Rusckowski et al., 2008; Daz browska, 2019; Podlacha et Experts FAO Antimicrobial Resistance Working Group, al., 2021) suggests that phages containing ARGs can 2018). also be present in the liver. To date, ARG-carrying Among the diverse elements that drive the mobiliza- phages in chicken have only been confirmed in meat tion of ARGs, the most extensively studied are plasmids samples and faeces (Go mez-Gomez et al., 2019). and transposons (Brown-Jaque et al., 2015). Recently, In this study, internal tissues of chicken liver samples phages and phage-derived particles have been recog- were analysed for the presence of phages using a meta- nized as underestimated mobile genetic elements genomics approach and the liver virome was compared responsible for ARG transfer and as ARG reservoirs in with that of faecal samples. Infectious phages were eval- the environment (Colomer-Lluch et al., 2011a; Ross and uated in the same liver samples. Finally, ARG-carrying Topp, 2015; Subirats et al., 2016; Blanco-Picazo et al., phages in total liver and internal tissues were analysed 2020). First discovered in the second decade of the 20th by quantitative real-time PCR (qPCR). century (D’Herelle, 1917), phages have been receiving renewed attention for their role as regulators of bacterial Results and discussion populations in natural ecosystems (Fuhrman, 1999). Many metagenomic analyses of human and animal A high percentage of non-sterile liver samples contained microbiomes have described the extremely high abun- moderate amounts of culturable microbial indicators dance of phages, generally greater than that of eukary- (Table 1). To confirm that these indicators originated otic viruses (Breitbart et al., 2003; Colomer-Lluch et al., from the liver tissues and were not the result of contami- 2011b; Minot et al., 2011; Virgin, 2014). Phages have nation from the peritoneal cavity, a second set of livers been reported in the lung, vaginal, skin, oral or intestinal were analysed in which the inner tissues were dissected microbiota (Oh et al., 2014) and more recently, infectious under more strict sterile conditions, thus avoiding the phages have been found in clinical samples, such as potentially contaminated external parts. Although fewer ascitic fluids and urine (Brown-Jaque et al., 2016). It has sterile livers contained culturable microorganisms, when been shown that phages can translocate from the intesti- present, the average values were not remarkably lower nal tract to the peritoneal cavity, bloodstream and other (Table 1), suggesting that the detected bacteria and organs (Go rski et al., 2006; Daz browska, 2019; Podlacha phages, at least partially, did not originate from external et al., 2021). In animals, phages infecting Bacteroides contamination. Somatic coliphages confirmed the pres- have been found in serum (Keller and Traub, 1974) and ence of phages infectious for the E. coli WG5 strain. phages infecting E. coli O157 show good persistence in the circulatory system of mouse (Capparelli et al., 2006). Observation of phage particles in sterile liver samples Also, phages translocate from blood to mouse foetal tis- sues in pregnant mice (Srivastava et al., 2004) and from To confirm these results, total phage particles in sterile intestinal tract to blood in Goldfish (Kawato and Nakai, livers were monitored by TEM. As titres of at least 107–8 2012). PFU are necessary for phage detection by electron Considering that phages are found in different systems microscopy (Brown-Jaque et al., 2016), a pool of phage and organs of the human and animal bodies suspensions from 10 livers were 1009 concentrated as (Daz browska, 2019; Kazmierczak et al., 2021; Podlacha described in the methods section and used for TEM et al., 2021), and that those carrying ARGs have been observation. Phage particles with icosahedral heads of ª 2022 The Authors. Microbial Biotechnology published by Society for Applied Microbiology and John Wiley & Sons Ltd., Microbial Biotechnology, 15, 2464–2475 2466 P. Blanco-Picazo et al. Table 1. Bacterial and viral indicators in the chicken liver samples. irrespective of whether the livers were sterile or not. Except for blaTEM, ARG prevalence was lower in the viral Ampicillin- Total resistant total fraction of sterile (Fig. 2) versus non-sterile liver. qnrA Somatic aerobic aerobic was only detected in sterile liver, and only in one sample coliphages bacteria bacteria (qnrA is not included in Fig. 3). This single positive does Sample Value (PFU g 1) (CFU g 1) (CFU g 1) not allow conclusions to be drawn about the prevalence Non- n 47 47 47 of qnrA in the two groups. sterile % Positive 66.0 100.0 95.0 Antibiotic resistance gene abundance (Fig. 3) did not liver Average 2.6 9 102 1.5 9 104 2.1 9 103 Max–min 103–5 2.2 9 105– 2.0 9 103– differ significantly (P > 0.05) between sterile and non- 4.2 9 103 5.0 9 102 sterile livers. In non-sterile liver, there were no significant Sterile liver N 27 27 27 differences (P > 0.05) between the abundances of the % Positive 60.6 96.4 67.9 Average 2.2 9 102 1.2 9 104 1.5 9 103 different ARGs. In the sterile liver, the abundance of Max–min 103–5 1.6 9 105– 2.7 9 103– blaCTX-M-1 showed significant (P < 0.05) differences 5.0 9 102 5.0 9 101 when compared with armA, sul1, or tetW, respectively. As qnrA was only detected in one sterile liver sample, it was not included in the chart. 3.07 log10 of qnrA gene copies were detected per g of liver. 57–62 nm and non-contractile tails were observed (Fig. 1). In the same pool, the visualized phages had Liver and faecal viromes long straight tails (Fig. 1A) as well as curly tails, sug- gesting that the concentrated samples contained at least The viromes of a pool of sterile liver samples were ana- two different types of phages (Fig. 1B). It should be lysed to gain more insight into the phage fraction. The noted that the concentration step and the requirement purification of encapsidated DNA to achieve the quality for high titres of phages for TEM observation (Blanco- required for metagenomics resulted in the generation of Picazo et al., 2020) may have reduced the diversity of a relatively low amount of contigs; 62% of the observable phages to those reaching densities above sequences were unclassified when compared against 107–8 PFU. general databases, and did not correspond to eukaryotic, archaeal or bacterial DNA, while the rest of contigs were identified as bacterial (28%) and viral (11%) (Table 2). Presence of ARGs in the phage DNA fraction of chicken The great majority of viral sequences (89%) corre- liver sponded to bacteriophages of the Caudovirales order, Antibiotic resistance genes were detected in the phage particularly of the Drexlerviridae family (Fig. 4), which DNA fraction of a large percentage of liver samples, with comprises phages with an icosahedral head and long variable results for the five targeted ARGs (Fig. 2), non-contractile tail, compatible with the TEM (A) (B) Fig. 1. Transmission electron micrographs of phages purified from a pool of 10 chicken liver samples from group 2 (sterile inner tissue). Two morphologies were observed: phages with icosahedric capsids and a long non-contractile tail (A) or a short and curly non-contractile tail (B). Bar 50 nm. ª 2022 The Authors. Microbial Biotechnology published by Society for Applied Microbiology and John Wiley & Sons Ltd., Microbial Biotechnology, 15, 2464–2475 Phages with ARGs in chicken liver 2467 Fig. 2. Antibiotic resistance genes (ARGs) in the phage DNA fraction of chicken liver samples. Percentage of positive samples for each ARG (blaTEM, blaCTX-M-1, sul1, armA and tetW) in each matrix. Two types of samples were analysed; group 1: the whole liver (non-sterile) and group 2: inner tissues of the liver (sterile), recovered by dissection with a sterile scalpel. Fig. 3. Abundance of each antibiotic resistance gene (ARG) in phage DNA isolated from chicken liver samples of group 1: non-sterile (47 sam- ples) and group 2: sterile (27 samples). Box plot of the average values (log10 GC g 1 sample) of all ARGs in the positive samples. The cross- pieces of each box represent (from top to bottom) the maximum, upper-quartile, median, lower quartile, and minimum values. The diamond (black) shows the mean value. The upper boxes (white) in the box plot include samples showing values within the 75th percentile and lower box samples (grey) show values within the 25th percentile. observations (Fig. 1). Most phages in the liver virome corresponded to gyroviruses (5%) and avian adeno- showed homology with E. coli or Shigella phages, associated viruses (6%) (Fig. 4). although the high number of sequences of Enterobac- Although phage presence in the analysed samples teria phages, particularly those of E. coli, in the data- was confirmed and ARGs were detected in the pack- bases may have influenced this result. Considering the aged DNA by qPCR, metagenomic sequencing of the widespread genome mosaicism reported in phages (Hat- same DNA was of insufficient depth to identify any full, 2008), those infecting Enterobacteria should share ARGs. Less sensitive than qPCR, metagenomics cannot several DNA fragments. The non-phage viral sequences determine genes present in moderate concentrations, ª 2022 The Authors. Microbial Biotechnology published by Society for Applied Microbiology and John Wiley & Sons Ltd., Microbial Biotechnology, 15, 2464–2475 2468 P. Blanco-Picazo et al. Table 2. Metagenomic parameters in chicken liver and chicken faecal samples. Size of the longer #contigs #contigs #contigs #contigs Origin #contigs/Total length (bp) contigs (bp) Unclassified (%) Archaea (%) Bacteria (%) Viruses (%) Chicken liver (PL) 11 539 / 164 020 1646 7121 (61.71) 5.8 (0.05) 3196 (27.70) 1216 (10.54) Chicken faeces (HP1) 643 753 / 295 827 925 110 470 535 005 (83.11) 329 (0.05) 105 241 (16.35) 2638 (0.41) Chicken faeces (HP2) 1 163 538 / 558 526 824 237 483 954 610 (82.04) 611 (0.05) 204 160 (17.55) 3282 (0.28) Fig. 4. Pie graphs showing the diversity of the viral fraction of chicken liver simples according to Kraken classification of metagenomics data. The graphs compare the distribution and relative abundance of viruses identified in the viromes with regard to the total viral sequences identi- fied. Interactive pie graphs are available upon request. such as ARGs, particularly in complex samples (e.g., in the liver virome suggests a translocation from the liver, food and faeces) (Ferna ndez-Orth et al., 2019; rski et al., 2006; Brown-Jaque et al., intestinal virome (Go Blanco-Picazo et al., 2020). On the other hand, not 2016; Chen et al., 2018). To gain further information, two being limited to specific genes, metagenomics can iden- chicken faecal viromes were analysed using the same tify a more extensive repertoire of ARGs. It should also protocol for phage purification and DNA extraction. The be mentioned that the successful elimination of all traces first finding was that the faecal virome contained a signif- of non-packaged DNA, which required additional steps in icantly higher amount of DNA, and consequently more the protocol, was achieved at the cost of reducing the sequences, compared with the liver virome (Table 2). recovery and quality of the extracted DNA. However, a higher percentage of sequences remained The presence of phages (including those carrying unclassified, and a lower percentage were identified ARGs) infectious for Enterobacteria, particularly E. coli, as viral (Table 2). Within the viral fraction, most ª 2022 The Authors. Microbial Biotechnology published by Society for Applied Microbiology and John Wiley & Sons Ltd., Microbial Biotechnology, 15, 2464–2475 Phages with ARGs in chicken liver 2469 bacteriophages were of the Caudovirales order (93% in transporters. Since ABC transporters confer unspecific HP1 and 88% in HP2) (Table S2). resistance to drugs, and cannot be classified as ARGs, A great diversity of ARGs was detected in the viromes they have not been included in these results despite of chicken faeces in accordance with the qPCR results they can be horizontally mobilized, either by phage parti- obtained in the previous studies (Go mez-Gomez et al., cles or by plasmids (Kadlec and Schwarz, 2009). 2019) (Table 3) and in accordance with the high amount The viromes show a large set of bacterial sequences. of ARGs in the faecal virome of humans (Ferna ndez- Since the controls used have discarded that bacterial Orth et al., 2019) and viromes of faecally polluted envi- DNA contamination might have contributed significantly ronments (Colombo et al., 2016; Lekunberri et al., 2017; to these results, our assumption supported by previous Chen et al., 2021; Wang et al., 2021). Only two ARGs studies (Fernandez-Orth et al., 2019; Fillol-Salom et al., were detected by both the metagenomic analysis (using 2021; Humphrey et al., 2021) is that phages package either Prokka, CARD or Resfinder) and qPCR, blaTEM bacterial DNA from their hosts more frequently than pre- and tetW, which were the most abundant (more than 6 viously thought. This information, together with the and 8 log10 GC/g, respectively) (Table 3) according to sequences of viral species, can then be used to deter- the qPCR results. The other ARGs targeted by qPCR mine the route of transmission of these phages. Since went undetected by the metagenomic approach, regard- we speculate that phages can come from faecal micro- less of the bioinformatic platform used. biomes, we searched for coincidences between the Several ARGs other than those targeted by qPCR chicken liver virome (PL) and two different faecal vir- were determined in the faecal viromes, conferring resis- omes (HP1 and HP2). When comparing bacterial and tances to a range of antibiotics (e.g., chloramphenicol, viral species detected in the liver virome with those macrolides, lincosamide and vancomycin) (Table 3). detected in the faecal viromes, 61 of the 68 (89.7%) spe- Resfinder, a more restrictive database, gave fewer cies found in the PL virome were also present in HP1 results than Prokka or CARD, which also identified (Fig. 5, Table S3). The seven absent organisms were Table 3. ARGs in chicken faeces detected by metagenomics compared with the ARGs detected by qPCR. METAGENOMICS qPCR Gene Resistance to Average (log GC g 1) Resistance to HP1 tet M (1-2)a, tet 32, tet40, tet W Tetracycline tet W(8.55) Tetracycline bla TEM Betalactamics bla TEM (6.33) Betalactamics blaCTX-M-1 (5.20) ant(6)-Ia, aph(2’’)-Ig, aph(3’)-III, aac(6’)-Im Aminoglycoside armA (3.54) Amynoglycoside drrA (2-15) Daunorubicin/doxorubicin qnrA (3.63) Quinolone linA Lincosamide sul1 (5.59) Sulphonamide mecI (1-3) Methicillin mupB Mupirocin vanA (1-2), van B (1-3), van C (1-2) Vancomycin A, B, C-type lnuC Lyncomycin cfrC Linezolid ermB Macrolide cat Chloramphenicol HP2 tet 32(1-2), tet M(1-3), tet O, tet W Tetracycline tet W (8.00) Tetracycline bla TEM Betalactamics bla TEM (6.64) Betalactamics blaCTX-M-1 (5.12) aph(2’’)-Ig, ant(6)-Ia, aph(3’)-III, aac(6’)-Im Aminoglycoside armA (4.76) Aminoglycoside ble Bleomycin sul1 (5.85) Sulfonamide drrA(1-34) Daunorubicin/doxorubicin qnrA (4.10) Quinolone cat Chloramphenicol cfrC Linezolid ermB, mefA, erm G Erythromycin lnuC Lyncomycin fsr Fosmidomycin linA (1-2) Lincosamide mecI(1-16) Methicillin mupA, mupB (1-2) Isoleucine--tRNA ligase. Mupirocin tmrB (1-2) Tunicamycin vanA (1-13), vanB (1-6),vanC (1-8), vanW (1-8) Vancomycin A, B and C-type In bold, ARGs detected in the virome by metagenomics (either by Prokka or Resfinder) as well as by qPCR a. In brackets are indicated the range of gene subtypes. ª 2022 The Authors. Microbial Biotechnology published by Society for Applied Microbiology and John Wiley & Sons Ltd., Microbial Biotechnology, 15, 2464–2475 2470 P. Blanco-Picazo et al. resident phages. Mutant phages able to circumvent the mononuclear phagocyte system have been reported (Merril et al., 1996); therefore, if phages resistant to mac- rophages can be selected under laboratory conditions, it is plausible that natural subpopulations of phages with natural resistance to phagosomal degradation can be selected as well. In addition, the efficiency of the circu- lating phages removal and the tolerance of the immune system depend on many variables, such as the type of phages, the size of the virions, the amount of phages and even the administration method, if any (Daz browska, 2019; Podlacha et al., 2021). All this together could explain the presence of infectious phages in the livers analysed in this study. Assuming that a fraction of the phage population could be retained within the inner tissues of the liver for a long enough period of time to inject their DNA if a suitable Fig. 5. Venn diagram showing the number of unique and sharing host is found, what happens next will depend on whether sequences between the chicken liver virome (PL) and the chicken the phages are complete (with the entire phage genome faecal viromes HP1 or HP2 respectively. in the capsid) or transducing particles (phage capsids containing only bacterial DNA). Phages can propagate in the host strain and transduce some of the genes they Cloacibacterium normanense, Rogunavirus, Rtpvirus, encode (including ARGs), whereas transducing particles Escherichia phage vB_EcoS_AHS24, Enterobacteria can inject their DNA in a recipient cell, which can phage vB_EcoS_Rogue1, avian gyrovirus 2 and chicken acquire the genes through recombination. ARG transfer anaemia virus. Similar results were observed when com- to a suitable host is possible in both scenarios. Adminis- paring the PL virome with HP2 (Fig. 5, Table S3), with tering a strong antibiotic treatment to an animal colo- 97% of the species (66 of the 68) in chicken liver being nized by bacteria transduced by an ARG-carrying phage found in the faecal virome, only avian gyrovirus 2 and would impose strong selective pressure in favour of the chicken anaemia virus being absent. transduced strains. The high proportion of species in chicken liver virome Chicken liver is a reservoir of Campylobacter and Sal- matching those found in the faecal viromes (Fig. 5) monella (Firlieyanti et al., 2016; Lanier et al., 2018; points towards an intestinal origin of the viruses found in Laconi et al., 2021) and also of avian pathogenic E. coli the chicken liver. It should be noted that although livers (Abdelhamid et al., 2020). Moreover, Campylobacter and faecal samples come from the same farm and type gastrointestinal colonization in chickens seems to con- of chicken, samples were taken at different times, mak- tribute to the translocation of E. coli from the gut to ing the coincidence even more relevant. Translocation extra-intestinal organs, leading to a high prevalence of from the intestinal tract seems more likely than phages E. coli in the spleen and liver (Awad et al., 2016). Like reaching the liver through systemic colonization. High their bacterial hosts, phages might easily translocate out- accumulation of phages in liver tissues have been previ- side the intestinal environment to other organs, which ously demonstrated (Daz browska, 2019; Van Belleghem can serve as reservoirs of phage particles and their et al., 2019; Kazmierczak et al., 2021; Podlacha et al., genetic content and constitute environments for the 2021). However, pharmacokinetics studies suggest that emergence of new resistant strains. although phages accumulate very efficiently in the liver, they are also quickly inactivated because liver is the first Experimental procedures responsible of bacteriophage clearance (Daz browska, 2019; Van Belleghem et al., 2019), particularly by the This study did not include animal experimentation. Animal macrophages found in the Kupffer cells (Inchley, 1969). liver samples were collected at the slaughterhouse at the Therefore, doubts arise whether the liver could be a res- time of sacrifice following the guidelines of the Ethical ervoir of phages able to transduce ARGs to active recipi- Committee for Animal Experimentation (CEEA), the autho- ent cells, despite the high accumulation. Noteworthy is rized body by the Catalan Government. The guidelines the fact that the removal of phages in liver has been are defined in European Parlament 2010/63/EU; RD 53/ evaluated using phage suspensions, usually of a single 2013; Directive 2010/63/UE; Decret 214/1997/GC, Ordre phage type administered in animal models and not using ECC/566/2015 and regulate the use of animals for ª 2022 The Authors. Microbial Biotechnology published by Society for Applied Microbiology and John Wiley & Sons Ltd., Microbial Biotechnology, 15, 2464–2475 Phages with ARGs in chicken liver 2471 experimentation and other scientific purposes. All experi- supernatant was filtered through 0.22 lm low protein- mental procedures were conducted in accordance with the binding polyethersulfone membranes (PES) (Millex -GP, biosafety regulations for microorganisms and approved Millipore, Bedford, MA). The filtrates were used for the by the Office of Safety, Health and the Environment analysis of somatic coliphages and the extraction of viral (OSSMA) of the University of Barcelona. DNA. Samples Culturable bacterial and viral indicators Sampling was performed in two slaughterhouses from Two groups of culturable indicators were analysed in the northeastern Spain between February and September liver samples from groups 1 and 2. Total aerobic micro- 2020. Healthy animals with no prophylactic antibiotic organisms and ampicillin (amp)-resistant total aerobic treatment were randomly selected. During the eviscera- microorganisms were evaluated in the homogenate of tion process, a first group of livers was collected from the livers for the presence of culturable bacteria. To do carcasses. Whole livers were removed aseptically using this, the homogenates were diluted 1/10 and 1/100. clean gloves and ethanol-disinfected scissors and placed 0.1 ml of the homogenate or the dilutions were plated on in new, clean plastic bags. The livers of three carcasses Tryptone Soy agar (TSA) and TSA with amp per flock were pooled and analysed as a single sample. (100 lg ml 1) and incubated at 37°C. Each homogenate These correspond to group 1 comprising 47 whole liver was analysed in duplicate. samples. Somatic coliphages, proposed as viral indicators of However, after the analysis of culturable microorgan- faecal pollution (Jofre, 2007), were evaluated for the isms, the high levels of bacteria and phages detected presence of infectious faecal viruses in the samples. To raised doubts about possible contamination from the do this, 10 ml of homogenates were centrifuged for peritoneal cavity. Therefore, a second group of samples 15 min at 4000 g and the supernatants were filtered was collected for this study (group 2). Group 2 com- through 0.22 lm low protein-binding polyethersulfone prised 27 internal liver tissue samples, each sample is a membranes (PES) (Millex -GP, Millipore, Bedford, MA). pool of three internal liver tissues from the same flock. Decimal dilutions of the filtrates were performed and To sample internal tissues, livers were placed on a ster- 1 ml of each was analysed by double agar layer in dupli- ile Petri dish and external parts of the liver were sani- cate for the presence of somatic coliphages following the tized by searing the surface with a permanent cautery ISO standard method (Anonymous, 2000) that uses E. spatula. Then, sterile scissors and tweezers were used coli strain WG5 (ATCC 700078) as the bacterial host. to remove the upper external segment of the liver and to Plates were incubated at 37°C for 18 h. Each homoge- collect fragments of the internal tissue, avoiding contact nate was analysed in duplicate. with the external surface of the liver even if it was previ- ously sanitized. Extraction of phage particles and removal of non- In addition, archive samples of chicken faeces (caecal packaged DNA contents) of broiler chickens from different flocks and farms in northeastern Spain collected from a slaughter- To extract the DNA of viral particles in the sample, house in 2015 and stored at 20C were used to ana- regardless they were infective or not, the homogenates lyse the chicken faecal viromes. obtained as described above were treated with chloro- form (1:10 (v/v)) to remove any extracellular vesicles. The samples were vortexed for 5 min and centrifuged at Homogenization of the liver samples 16 000 g for 5 min. The supernatant was recovered and Ten grams of liver samples (requiring three livers from digested with DNase (100 units ml 1 of the supernatant) chickens of the same batch and farm) of Group 1 and 2 to eliminate any DNA that might be present in the sam- were homogenized in 40 ml of phage buffer. Tissues ples outside the phage particles. DNase was inactivated were disrupted using 1 and 3 mm glass beads (Hecht by heating for 5 min at 75°C and an aliquot was taken Assistant, Sondheim vor der Rho €n. Germany) following as a control to verify the removal of non-packaged DNA the manufacturer’s instructions and the homogenates by qPCR amplification of bacterial 16S rRNA and ARGs were recovered. as described below. Additional controls were performed At this point, the homogenates were used for the anal- to confirm the correct inactivation of DNase by heat ysis of bacterial indicators, as described below. Subse- treatment, as described previously (Colomer-Lluch et al., quently, the homogenates were centrifuged at 4000 g for 2014). 25 min. The supernatant was collected and centrifuga- Once the control was confirmed as negative, lysates tion was repeated under the same conditions. The final were used to observe phage particles by electron ª 2022 The Authors. Microbial Biotechnology published by Society for Applied Microbiology and John Wiley & Sons Ltd., Microbial Biotechnology, 15, 2464–2475 2472 P. Blanco-Picazo et al. microscopy and to extract the DNA in the phage parti- lactam antibiotics (blaTEM, blaCTX-M-1 group), a quinolone cles (packaged DNA). resistance gene (qnrA), sul1, which confers resistance to sulphonamides and is frequently found in environmental and clinical bacterial populations (Pruden et al., 2012), Transmission electron microscopy armA, which encodes aminoglycoside resistance and is Viruses from the homogenates of liver samples of group widely distributed in Enterobacteriaceae (Galimand et 2 were purified by filtration, treated with chloroform as al., 2005), and tetW, conferring resistance to tetracycline, described above and 1009 concentrated by protein con- an antibiotic commonly used in intensive poultry farming centrators (100-kDa Amicon Ultra centrifugal filter units; (Mehdi et al., 2018). Millipore, Bedford, MA). Seven microlitres of the concen- For quantification, serial dilutions of gBlocksTM Gene trated phage suspension was dropped onto copper grids Fragments (Integrated DNA Technologies, Coralville, with carbon-coated Formvar films and negatively stained USA) of known concentration were used to generate the with 2% ammonium molybdate (pH 6.8) for 1.5 min. standard curves. All samples were run in triplicate Phages were visualized using a Jeol 1010 transmission (including the standards and negative controls). The electron microscope (JEOL Inc. Peabody, MA, USA) number of gene copies (GC) was defined as the mean operating at 80 kV. of the triplicate data obtained. To quantify the ARGs, we considered the results obtained within the threshold cycle (Ct) and within the limit of quantification (LOQ). Extraction of packaged DNA for ARG quantification This was determined by the last valid Ct for each ARG The protocol for DNA extraction was as previously assay (GC ll 1 are shown in Table S1) in the standard described (Sambrook and Russell, 2001) with some modi- curve that was consistent in the diverse replicates. fications. After performing the chloroform and DNase treat- gBlocks were also used as positive controls. ments and obtaining negative results for ARGs and 16S rRNA amplification, which ruled out the presence of non- Metagenomic analysis encapsidated DNA, capsids of the viral suspensions were degraded with proteinase K (20 mg ml 1) in 250 ll of pro- Ten liver samples from group 2 (10 g) were pooled and teinase K buffer and incubated for 1 h at 55°C. Packaged used for sequencing the viral fraction (Virome PL). DNA was then extracted by the phenol–chloroform (1:1) Chicken faecal samples were divided into two pools (v:v) method and the aqueous phase was again treated (chicken faeces HP1 and HP2). For each pool of liver or with chloroform (1:1) (v:v), centrifuging at the same speed faeces, 10 g was homogenized together in 40 ml of and time as in the previous step. The remaining phenol/ phage buffer using a horizontal agitator at 800 rpm for chloroform was removed by placing the mixture in Phase 10 min. Viruses were recovered from the samples after Lock Gel Tubes (5-Prime, Hucoa Erlo €ss, Madrid, Spain) filtration, chloroform and DNAse treatment as described for centrifugation following the manufacturer’s instructions. above. Before breaking the capsids, the removal of non- DNA was precipitated using 100% ethanol and 3 M packaged DNA was confirmed by qPCR amplification of sodium acetate and resuspended in a final volume of bacterial 16S rRNA as a control. 100 µl of ultrapure water. The use of phenol in the DNA extraction protocol described above meant that the DNA quality was insuffi- cient for metagenomic analysis, as previously reported qPCR analysis (Ferna ndez-Orth et al., 2019). Therefore, the phage The ARGs were evaluated by qPCR TaqMan assays lysates were used to extract the DNA for metagenomic (Table S1). The absence of bacterial 16S rRNA was ver- analysis using the QIAmp Viral RNA Mini Kit (Hilden, ified by qPCR using Power SYBR Green PCR Master Germany), which broke down the capsids and allowed Mix (Thermo Fisher Scientific, Waltham, MA, USA) the viral DNA to be recovered in a final volume of 80 lL (Table S1). of bidistilled water. The DNA from the samples was qPCR using TaqMan hydrolysis probes was conducted pooled and further purified using the DNA Clean & with the StepOne Real-Time PCR System (Applied Bio- ConcentratorTM-5 Kit (Zymo Research, Irving, CA, USA) systems, Waltham, MA, USA) in a 20 ll reaction mixture to a final volume of 50 ll, which contained the required with the TaqMan Environmental Master Mix 2.0 (Applied DNA concentration for subsequent sequencing analysis. Biosystems). The results were analysed with the Applied Biosystems StepOneTM Instrument program. Sequencing Six qPCR assays targeting ARGs were performed in 9 ll of the sample DNA or quantified DNA. The targets The DNA concentration for metagenomics was evaluated were clinically relevant genes that confer resistance to b- using a Qubit Fluorometer (Life Technologies, CA, ª 2022 The Authors. Microbial Biotechnology published by Society for Applied Microbiology and John Wiley & Sons Ltd., Microbial Biotechnology, 15, 2464–2475 Phages with ARGs in chicken liver 2473 USA) and the DNA quality was further confirmed using Data availability the 2100 Bioanalyzer system (Agilent Technologies, CA, USA). 0.2 ng ll 1 of DNA was used to prepare the The metagenomic data set generated was deposited in libraries. The DNA was fragmented and used to prepare BioProject (PRJNA782068). the libraries with the Nextera XT Kit (Illumina. Inc. San Diego, CA, USA) protocol for paired-end libraries Statistics (2 9 150 bp). For extension, 14 PCR cycles of 2.5 min were performed to increase the tagmentation process. Statistical tests were performed using the Statistical Libraries were purified using AmPure beads (Beckman Package for Social Science Software (SPSS). A paired Coulter Inc., California, USA), checked for the distribu- Student’s-t test (P < 0.05) was used to evaluate the dif- tion and size of the fragments in a 2100 Bioanalyzer and ferences between ARGs in sterile and non-sterile liver the Agilent High Sensitivity DNA Chip (DNA 1000) (Agi- samples. lent Technologies, CA, USA) and quantified in a Quan- tusTM Fluorometer (Promega, WI, USA). An equimolar Acknowledgements pool of the seven samples was sequenced in a NextSeq Illumina in a High Output run of 300 cycles. This work was supported by the Spanish Ministerio de Innovacion y Ciencia (AGL2016-75536-P and PID2020- 113355GB-I00), the Agencia Estatal de Investigacio n Bioinformatic analysis (AEI) and the European regional fund (ERF). The study Reads were analysed with FastQC (version 0.11.9) and was partially supported by the Union Iberoamericana de summarized with MultiQC (version 1.11). As the files Universidades (UIU-OHOW) and the Generalitat de showed low sequence quality, high duplication levels, Catalunya (2017SGR170). CERCA Programme from the overrepresented sequences and the presence of Generalitat de Catalunya is also acknowledged. P. B-P adapters, a preprocessing step was performed. First, the has a grant from the Spanish Ministerio de Economıa, duplications for each file were removed with Seqkit Industria y Competitividad (BES-2017-081296). L. R.-R (v0.16.1). Then, the adapters for the paired-end files nter program, Generalitat de is a lecturer of the Serra-Hu were cut with Trimmomatic (v0.39). Finally, the remain- Catalunya. ing duplicates and the sequences with quality below 20 for the paired-end files were removed with PRINSEQ Conflict of interest (v0.20.4). Reads were de novo assembled with SPAdes (version None declared. 3.15.3), specifically MetaSPAdes (Nurk et al., 2017), which is recommended when assembling metagenomics References data sets. The obtained scaffolds from this assembly were used in the following analysis. Abdelhamid, M.K., Quijada, N.M., Dzieciol, M., Hatfaludi, T., Contigs were taxonomically classified with Kraken2 Bilic, I., Selberherr, E., et al. (2020) Co-infection of (version 2.0.7 beta) (Wood and Salzberg, 2014). chicken layers with Histomonas meleagridis and avian pathogenic Escherichia coli is associated with dysbiosis, The presence of ARGs in the viromes was deter- cecal colonization and translocation of the bacteria from mined by the specific software ResFinder (version 4.1) the gut lumen. Front Microbiol 11: 586437. (Zankari et al., 2012). Since this software is very Anonymous. (2000) ISO 10705-2: Water quality. Detection restrictive, ARGs search was completed with CARD and enumeration of bacteriophages -part 2: enumeration database (McArthur et al., 2013) and Prokka (version of somatic coliphages. 1.14.5) (Seemann, 2014) with the terms ‘resist’, ‘beta- Antimicrobial Resistance in the Environment Summary lactam’, ‘penicillin’ which includes antibiotic as well as Report of an FAO Meeting of Experts FAO Antimicrobial Resistance Working Group (2018). other resistances. Awad, W.A., Dublecz, F., Hess, C., Dublecz, K., Khayal, B., Comparison of sequences between the chicken liver Aschenbach, J.R., and Hess, M. (2016) Campylobacter and faecal viromes was also done by BLAST+ (version jejuni colonization promotes the translocation of Escheri- 2.10.1). For this, two Blastn queries (PL vs. HP1 and PL chia coli to extra-intestinal organs and disturbs the short- vs. HP2) were performed. The PL sequences were chain fatty acids profiles in the chicken gut. Poult Sci 95: inputted as a query, and the HP1 and HP2 files were 2259–2265. the database where the query was searched in each Blanco-Picazo, P., Ferna ndez-Orth, D., Brown-Jaque, M., Miro, E., Espinal, P., Rodrıguez-Rubio, L., et al. (2020) comparison. It was determined that the output matches Unravelling the consequences of the bacteriophages in must have an E-value < 1-10, which indicated an almost human samples. Sci Rep 10: 6737. identical match. ª 2022 The Authors. Microbial Biotechnology published by Society for Applied Microbiology and John Wiley & Sons Ltd., Microbial Biotechnology, 15, 2464–2475 2474 P. Blanco-Picazo et al. Blanco-Picazo, P., Roscales, G., Toribio-Avedillo, D., Fillol-Salom, A., Bacigalupe, R., Humphrey, S., Chiang, Go mez-Go mez, C., Avila, C., Balleste , E., et al. (2020) Y.N., Chen, J., and Penade s, J.R. (2021) Lateral trans- Antibiotic resistance genes in phage particles from Ant- duction is inherent to the life cycle of the archetypical Sal- arctic and Mediterranean seawater ecosystems. Microor- monella phage P22. Nat Commun 12: 6510. ganisms 8: 1293. Firlieyanti, A.S., Connerton, P.L., and Connerton, I.F. (2016) Breitbart, M., Hewson, I., Felts, B., Mahaffy, J.M., Nulton, J., Campylobacters and their bacteriophages from chicken Salamon, P., and Rohwer, F. (2003) Metagenomic ana- liver: the prospect for phage biocontrol. Int J Food Micro- lyses of an uncultured viral community from human feces. biol 237: 121–127. J Bacteriol 185: 6220–6223. Fuhrman, J.A. (1999) Marine viruses and their biogeochemi- Brown-Jaque, M., Calero-Ca ceres, W., and Muniesa, M. cal and ecological effects. Nature 399: 541–548. (2015) Transfer of antibiotic-resistance genes via phage- Galimand, M., Sabtcheva, S., Courvalin, P., and Lambert, T. related mobile elements. Plasmid 79: 1–7. (2005) Worldwide disseminated armA aminoglycoside Brown-Jaque, M., Muniesa, M., and Navarro, F. (2016) Bac- resistance methylase gene is borne by composite trans- teriophages in clinical samples can interfere with microbi- poson Tn1548. Antimicrob Agents Chemother 49: 2949– ological diagnostic tools. Sci Rep 6: 33000. 2953. Capparelli, R., Ventimiglia, I., Roperto, S., Fenizia, D., and Gomez-Go mez, C., Blanco-Picazo, P., Brown-Jaque, M., Iannelli, D. (2006) Selection of an Escherichia coli O157: Quiro s, P., Rodrıguez-Rubio, L., Cerda -Cuellar, M., and H7 bacteriophage for persistence in the circulatory system Muniesa, M. (2019) Infectious phage particles packaging of mice infected experimentally. Clin Microbiol Infect 12: antibiotic resistance genes found in meat products and 248–253. chicken feces. Sci Rep 9: 13281. Chen, M., An, X., Liao, H., Yang, K., Su, J., and Zhu, Y. Gorski, A., Wazna, E., Dabrowska, B.-W., Dabrowska, K., (2021) Viral community and virus-associated antibiotic Switała-Jelen , K., and Miedzybrodzki, R. (2006) Bacterio- resistance genes in soils amended with organic fertilizers. phage translocation. FEMS Immunol Med Microbiol 46: Environ Sci Technol 55: 13881–13890. 313–319. Chen, Y., Guo, J., Shi, D., Fang, D., Chen, C., and Li, L. Hatfull, G.F. (2008) Bacteriophage genomics. Curr Opin (2018) Ascitic bacterial composition is associated with clini- Microbiol 11: 447–453. cal outcomes in cirrhotic patients with culture-negative and Humphrey, S., Fillol-Salom, A., Quiles-Puchalt, N., Ibarra- non-neutrocytic ascites. Front Cell Infect Microbiol 8: 420. Cha vez, R., Haag, A.F., Chen, J., and Penade s, J.R. Colombo, S., Arioli, S., Guglielmetti, S., Lunelli, F., and (2021) Bacterial chromosomal mobility via lateral trans- Mora, D. (2016) Virome-associated antibiotic-resistance duction exceeds that of classical mobile genetic elements. genes in an experimental aquaculture facility. FEMS Nat Commun 12: 6509. Microbiol Ecol 92: fiw003. Inchley, C.J. (1969) The activity of mouse Kupffer cells fol- Colomer-Lluch, M., Calero-Ca ceres, W., Jebri, S., Hmaied, lowing intravenous injection of T4 bacteriophage. Clin Exp F., Muniesa, M., and Jofre, J. (2014) Antibiotic resistance Immunol 5: 173. genes in bacterial and bacteriophage fractions of Tunisian Jofre, J. (2007) Indicators of waterborne enteric viruses. In and Spanish wastewaters as markers to compare the Human Viruses in Water. Perspectives in Medical Virol- antibiotic resistance patterns in each population. Environ ogy. Bosch, A. (ed). Amsterdam, the Netherlands: Else- Int 73: 167–175. vier, pp. 227–249. Colomer-Lluch, M., Imamovic, L., Jofre, J., and Muniesa, M. Kadlec, K., and Schwarz, S. (2009) Novel ABC transporter (2011a) Bacteriophages carrying antibiotic resistance gene, vga(C), located on a multiresistance plasmid from a genes in fecal waste from cattle, pigs, and poultry. Antimi- porcine methicillin-resistant Staphylococcus aureus crob Agents Chemother 55: 4908–4911. ST398 strain. Antimicrob Agents Chemother 53: 3589– Colomer-Lluch, M., Jofre, J., and Muniesa, M. (2011b) Anti- 3591. biotic resistance genes in the bacteriophage DNA fraction Kawato, Y., and Nakai, T. (2012) Infiltration of bacterio- of environmental samples. PLoS One 6: e17549. phages from intestinal tract to circulatory system in gold- D’Herelle, F. (1917) Sur un microbe invisible antagonist des fish. Fish Pathol 47: 1–6. bacilles disenterique. Comptes Rendus L’academie Sci Ka zmierczak, Z., Majewska, J., Milczarek, M., Owczarek, Paris 165: 373–375. B., and Daz browska, K. (2021) Circulation of fluorescently Daz browska, K. (2019) Phage therapy: What factors shape labelled phage in a murine model. Viruses 13: 297. phage pharmacokinetics and bioavailability? Systematic Keller, R., and Traub, N. (1974) The characterization of and critical review. Med Res Rev 39: 2000. Bacteroides fragilis bacteriophage recovered from animal European Commission. 2020. EU agricultural outlook 2020- sera: observations on the nature of bacteroides phage 30: sustainability objectives to impact meat and dairy carrier cultures. J Gen Virol 24: 179–189. along the supply chain. https://ec.europa.eu/info/sites/ Laconi, A., Drigo, I., Palmieri, N., Carraro, L., Tonon, E., default/files/food-farming-fisheries/farming/documents/ Franch, R., et al. (2021) Genomic analysis of extra- agricultural-outlook-2020-report_en.pdf. intestinal Campylobacter jejuni and Campylobacter coli Ferna ndez-Orth, D., Miro , E., Brown-Jaque, M., Rodrıguez- isolated from commercial chickens. Vet Microbiol 259: Rubio, L., Espinal, P., Rodriguez-Navarro, J., et al. (2019) 109161. Faecal phageome of healthy individuals: presence of anti- Lanier, W.A., Hale, K.R., Geissler, A.L., and Dewey-Mattia, biotic resistance genes and variations caused by cipro- D. (2018) Chicken liver-associated outbreaks of campylo- floxacin treatment. J Antimicrob Chemother 74: 854–864. bacteriosis and salmonellosis, United States, 2000–2016: ª 2022 The Authors. Microbial Biotechnology published by Society for Applied Microbiology and John Wiley & Sons Ltd., Microbial Biotechnology, 15, 2464–2475 Phages with ARGs in chicken liver 2475 Identifying opportunities for prevention. Foodborne Pathog for potential transduction. Appl Environ Microbiol 81: Dis 15: 726–733. 7905–7913. Lekunberri, I., Subirats, J., Borrego, C.M., and Balca zar, Rusckowski, M., Gupta, S., Liu, G., Dou, S., and Hnatowich, J.L. (2017) Exploring the contribution of bacteriophages to D.J. (2008) Investigation of four 99mTc-labeled bacterio- antibiotic resistance. Environ Pollut 220: 981–984. phages for infection specific imaging. Nucl Med Biol 35: McArthur, A.G., Waglechner, N., Nizam, F., Yan, A., Azad, 433. M.A., Baylay, A.J., et al. (2013) The comprehensive anti- Sambrook, J., and Russell, D. (2001) Molecular Cloning: A biotic resistance database. Antimicrob Agents Chemother Laboratory Manual. Long Island, NY: Cold Spring Harbor 57: 3348–3357. Laboratory Press, p. 999. Mehdi, Y., Le tourneau-Montminy, M.-P., Gaucher, M.-L., Seemann, T. (2014) Prokka: rapid prokaryotic genome Chorfi, Y., Suresh, G., Rouissi, T., et al. (2018) Use of annotation. Bioinformatics 30: 2068–2069. antibiotics in broiler production: global impacts and alter- Srivastava, A.S., Chauhan, D.P., and Carrier, E. (2004) In natives. Anim Nutr 4: 170–178. utero detection of T7 phage after systemic administration Merril, C.R., Biswas, B., Carlton, R., Jensen, N.C., Creed, to pregnant mice. Biotechniques 37: 81–83. G.J., Zullo, S., and Adhya, S. (1996) Long-circulating bac- Subirats, J., Sanchez-Melsio, A., Borrego, C.M., Balca zar, teriophage as antibacterial agents. Proc Natl Acad Sci J.L., and Simonet, P. (2016) Metagenomic analysis USA 93: 3188–3192. reveals that bacteriophages are reservoirs of antibiotic Minot, S., Sinha, R., Chen, J., Li, H., Keilbaugh, S.A., Wu, resistance genes. Int J Antimicrob Agents 48: 163–167. G.D., et al. (2011) The human gut virome: inter-individual Van Belleghem, J.D., Daz browska, K., Vaneechoutte, M., variation and dynamic response to diet. Genome Res 21: Barr, J.J., and Bollyky, P.L. (2019) Interactions between 1616–1625. bacteriophage, bacteria, and the mammalian immune sys- Nhung, N.T., Chansiripornchai, N., and Carrique-Mas, J.J. tem. Viruses 11: 10. (2017) Antimicrobial resistance in bacterial poultry patho- Virgin, H.W. (2014) The virome in mammalian physiology gens: a review. Front Vet Sci 4: 126. and disease. Cell 157: 142–150. Nurk, S., Meleshko, D., Korobeynikov, A., and Pevzner, Wang, M., Sun, Y., Zeng, Z., and Wang, Z. (2021) Meta- P.A. (2017) metaSPAdes: a new versatile metagenomic genomics of wastewater phageome identifies an exten- assembler. Genome Res 27: 824–834. sively cored antibiotic resistome in a swine feedlot water OECD/FAO. (2018) OECD-FAO Agricultural Outlook 2018- treatment environment. Ecotoxicol Environ Saf 222: 2027. Rome, Italy: OECD Publishing, Paris/Food and 112552. Agriculture Organization of the United Nations. Wood, D.E., and Salzberg, S.L. (2014) Kraken: ultrafast Oh, J., Byrd, A.L., Deming, C., Conlan, S., Kong, H.H., and metagenomic sequence classification using exact align- Segre, J.A. (2014) Biogeography and individuality shape ments. Genome Biol 15: 46. function in the human skin metagenome. Nature 514: 59– World Health Organization. (2018) Antibiotic Resistance. 64. Geneva, Switzerland: World Health Organization. Podlacha, M., Grabowski, Ł., Kosznik-Kaw snicka, K., Zdro- Xiong, W., Wang, Y., Sun, Y., Ma, L., Zeng, Q., Jiang, X., jewska, K., Stasiłoj c, M., Wez grzyn, G., and Wez grzyn, A. et al. (2018) Antibiotic-mediated changes in the fecal (2021) Interactions of bacteriophages with animal and microbiome of broiler chickens define the incidence of human organisms—safety issues in the light of phage antibiotic resistance genes. Microbiome 6: 34. therapy. Int J Mol Sci 22: 8937. Zankari, E., Hasman, H., Cosentino, S., Vestergaard, M., Poultry | European Commission. Information on an overview Rasmussen, S., Lund, O., et al. (2012) Identification of of EU poultry, market measures and standards, trade acquired antimicrobial resistance genes. J Antimicrob measures, market monitoring, legal bases and commit- Chemother 67: 2640–2644. tees. https://ec.europa.eu/info/food-farming-fisheries/ animals-and-animal-products/animal-products/poultry_en Programa REDUCE Pollos Broiler | PRAN. https://www. resistenciaantibioticos.es/es/system/files/content_images/ Supporting information 2020.07.22_acuerdo_base_broiler_entidades.pdf Pruden, A., Arabi, M., and Storteboom, H.N. (2012) Correla- Additional supporting information may be found online in tion between upstream human activities and riverine anti- the Supporting Information section at the end of the biotic resistance genes. Environ Sci Technol 46: 11541– article. 11549. Table S1. Oligonucleotides used in this study. Ross, J., and Topp, E. (2015) Abundance of antibiotic resis- Table S2. Percentatges of viral species in fecal viromes. tance genes in bacteriophage following soil fertilization Table S3. Common species detected in the two viromes with dairy manure or municipal biosolids, and evidence compared (PL vs HP1 or PL vs HP2). ª 2022 The Authors. Microbial Biotechnology published by Society for Applied Microbiology and John Wiley & Sons Ltd., Microbial Biotechnology, 15, 2464–2475