Effects of dicopper oxide and copper sulfate on growth performance and gut microbiota in broilers A. Forouzandeh,* L. Blavi,*,1 N. Abdelli,* D. Melo-Duran,* A. Vidal,y M. Rodríguez,z A. N. T. R. Monteiro,x J. F. Perez,* L. Darwich,y and D. Sola-Oriol* * Animal Nutrition and Welfare Service (SNiBA), Departament de Ciencia Animal i dels Aliments, Universitat onoma de Barcelona, 08193 Bellaterra, Spain; yDepartment of Animal Health and Anatomy, Universitat Aut onoma de Barcelona, 08193 Bellaterra, Spain; zPigCHAMP Pro Europa, Segovia, Spain; and xAnimine, Annecy, Aut France ABSTRACT An experiment was conducted to deter- Cu dose inclusion increased from 15 mg/kg to mine the effects of two sources of copper (Cu) from cop- 150 mg/kg. Supplementation of 150 mg/kg of Cu from per sulfate (CuSO4) and dicopper oxide (Cu2O, Cu2O decreased the abundance (P < 0.01) of some fami- CoRouge) at three levels of inclusion (15, 75, and 150 lies such as Streptococcaceae and Corynebacteriaceae mg/kg) on growth performance and gut microbiota of and increased the abundance (P < 0.05) of some com- broilers. A total of 840 one-d-old male chickens (Ross mensal bacteria like Clostridiaceae and Peptostrepto- 308) were weighed and randomly allocated to seven die- coccaceae. Phenotypic AMR was not different among tary treatments: negative control (NC, a basal diet treatments on d 14 and 28. Isolated Enterococcus spp. without Cu addition), and the NC supplemented with from broilers fed the NC diet on d 42 showed higher (P 15, 75, or 150 mg Cu/kg from CuSO4 or Cu2O (12 repli- < 0.05) resistance to enrofloxacin, gentamicin, and cate pens/treatment, 10 chicks per pen). Broilers were chloramphenicol compared with Cu treatments. By con- challenged by reusing an old litter with high concentra- trast, the isolated Escherichia coli from broilers fed tions in Clostridium perfringens to promote necrotic 150 mg/kg of Cu, either from CuSO4 or Cu2O, showed enteritis. Broiler performance was registered at d 21, 35, higher (P < 0.05) resistance to streptomycin and chlor- and 42. Excreta samples were collected at d 14, 28, and amphenicol compared to the NC. This study suggests 42 for antimicrobial resistance (AMR) analyses. At d that supplementing 150 mg/kg of Cu from Cu2O estab- 43, one broiler per pen was euthanized to obtain ileal lishes changes in the gut microbiota by regulating the content for microbial characterization. Body weight d bacterial population in the ileum, which may explain the 35 and daily gain d 42 improved (P < 0.05) in Cu2O as positive impact on broilers' growth performance. KEY WORDS: copper, broiler, growth performance, ileal microbiota, antimicrobial resistance 2021 Poultry Science 100:101224 https://doi.org/10.1016/j.psj.2021.101224 INTRODUCTION by EFSA (2012) in the European Union is 25 mg/kg. However, in many non-EU countries, therapeutic doses Copper (Cu) is an essential trace mineral in the poul- (125250 mg/kg of Cu) of Cu from copper sulfate pen- try diet (Davis and Mertz, 1987). It is involved in tahydrate are being widely used as a growth promoter immune function and oxidation, plays a significant role and antibacterial feed additive (Pang and Apple- in iron metabolism (Kim et al., 2008; Ognik et al., 2016), gate, 2006). and allows optimal growth performance by maintaining Therapeutic doses of Cu may improve growth perfor- body functions (Banks et al., 2004b). The copper mance in animals by modulating the microbial popula- requirement for broilers is 5 to 8 mg/kg diet according tion within the gastrointestinal tract (Arias and to NRC (1994) and 4 to 10 mg/kg according to Koutsos, 2006) and, therefore, improving nutrient FEDNA (2018), but the maximum dosage authorized absorption (Hawbaker et al., 1961; Bunch et al., 1965). On the other hand, high Cu dosages influence antibiotic Ó 2021 The Authors. Published by Elsevier Inc. on behalf of Poultry resistance development (Poole, 2017), and pollute the Science Association Inc. This is an open access article under the CC BY environment through higher Cu excretion (Malan et al., license (http://creativecommons.org/licenses/by/4.0/). Received June 29, 2020. 2015). However, the antibacterial properties of Cu may Accepted April 14, 2021. depend on its redox state: Cu(I), the reduced cuprous 1 Corresponding author: laia.blavi@uab.cat 1 2 FOROUZANDEH ET AL. form, has a stronger antibacterial effect in anaerobic and 18 h/d from d 11 to the last day of the experiment. conditions than Cu(II), the oxidized cupric form Broilers were challenged by a recycled NE litter. (Dunning et al., 1998). Besides, the different solubility and bioavailability of Cu sources may affect intestinal microbiota in a different way (Pang et al., 2009). Necrotic Enteritis Challenge Procedure Copper sulfate (CuSO4) is soluble in water (99%) and acidic solvents (Pang and Applegate, 2006) and has The selection of the recycled litter material was made a Cu concentration of 25.4% (Baker, 1999). On the other between four commercial poultry flocks based on signs side, dicopper oxide (Cu2O, CoRouge, Animine) is a of NE. The farm with the highest concentration of clini- water nonsoluble compound that has the highest Cu cal NE and previously characterized for its content of concentration in the market (75% of Cu). In a previous mesophilic aerobic bacteria (> 105 CFU/g), Enterobac- study, Hamdi et al., (2018) observed that therapeutic teriaceae (5.2 £ 103 CFU/g), filamentous fungi and doses (150 mg/kg of Cu) of Cu2O in broilers diet yeasts (2.2 £ 103 CFU/g), and Clostridium perfringens increased their body weight (BW), however, when (5.6 £ 104 CFU/g) was selected. The floor area was cov- 150 mg/kg of Cu from CuSO4 was supplied growth per- ered with 10% clean wood shavings and 90% recycled lit- formance was not modified, and feed efficiency reduced ter material on the first day of the experiment. The with 300 mg/kg addition. It was also suggested that challenging process comprised of exposing broilers to a excessive Cu accumulates in different organs, and free contaminated litter characterized by high Clostridium unbound copper in the blood may act as a strong oxidiz- perfringens counts was formerly used by ing agent and cause a toxic response (Banks et al., Abdelli et al. (2020). 2004b; Reece et al., 2015). Nevertheless, there is no information about the effect of Cu2O on gut microbiota. Taking into account all the effects, we have hypothe- Experimental Diets sized that using the most effective source (Cu2O) could A 3-phase feeding program was used, a starter phase enhance performance at a therapeutic dose of from d 0 to d 21, a grower phase from d 22 to d 35, and a 150 mg/kg, or even lower dosage, through changes in finisher phase from d 36 to d 42 (Table 1). Seven diets the gut microbiota. It was also hypothesized that differ- ences between the sources may lead to a reduction in antimicrobial resistance (AMR) development caused Table 1. Composition and nutrient content of the basal diet. by high Cu concentration. Item Starter Grower Finisher Therefore, the objective of our study was to explore Ingredients, % the effect of 75 or 150 mg/kg dose of Cu from Cu2O on Ground corn 59.99 60.68 61.01 growth performance, intestinal microbiota profile, and Soybean meal 47% 33.40 31.22 29.62 AMR when it is compared to CuSO4 in broilers chal- Soybean oil 2.55 4.27 5.66 Monocalcium phosphate 1.37 1.29 1.20 lenged with recycled necrotic enteritis (NE) litter. Calcium carbonate 1.28 1.21 1.21 Mineral-vitamin premix1 0.30 0.30 0.30 Sodium chloride 0.31 0.30 0.24 DL-Methionine 0.25 0.25 0.24 L-Lysine.HCl 0.15 0.10 0.12 MATERIALS AND METHODS L-Threonine 0.05 0.05 0.06 Sodium bicarbonate 0.20 0.20 0.2 All experimental animal procedures were approved by Choline chloride 0.15 0.13 0.14 Calculated composition, % the Animal Ethics Committee of the Universitat ME Kcal/kg 2950 3050 3150 Autonoma de Barcelona and complied with the Euro- Crude Protein 21.20 20.00 18.50 pean Union guidelines for the care and use of animals in Calcium 0.98 0.90 0.78 Phosphorus 0.68 0.65 0.62 research (European Commision, 2010). Lysine 1.22 1.11 1.07 Methionine 0.56 0.54 0.52 Methionine + cysteine 0.91 0.87 0.84 Bird Management and Husbandry Threonine 0.81 0.77 0.75 Tryptophan 0.23 0.23 0.21 The study was carried out at a commercial growing Analyzed composition, % Dry matter 89.10 88.26 88.44 poultry unit (Tarragona, Spain). The room was pro- Crude Ash 5.69 5.48 5.08 vided with 84 solid-sided pens (0.8 £ 1 m) in 4 lines of 21 Crude Protein 20.49 19.02 18.52 pens divided by a central feeding aisle. A total of 840 Crude Fat 5.43 6.73 7.76 Crude Fiber 3.04 2.96 3.43 one-d-old male chickens (Ross 308) were randomly allo- cated to one of 7 dietary treatments (12 replicate pens/ Abbreviation: ME, metabolizable energy. 1 Provided per kg of diet: vitamin A (retinyl acetate), 17,000 IU; vita- treatment, 10 chicks per pen, and 0.64 m2 per chick) min D3 (Cholecalciferol), 3,500 IU; vitamin E (dl-a-tocopheryl acetate), according to initial BW and continuously controlled 15 IU; vitamin K3 (menadione sodium bisulfate), 2 mg; vitamin B1, 1.6 over 42 d. The average temperature was maintained at mg; vitamin B2, 4.16 mg; vitamin B6, 2 mg; vitamin B12, 0.012 mg; nico- tinic acid, 21.2 mg; pantothenic acid (D-Ca pantothenate), 10.58 mg; bio- 35 § 1°C and was decreased gradually (at the rate of 3° tin, 0.048 mg; folic acid, 0.8 mg; Zn (ZnO) 60.19 mg; Fe (FeSO4¢7H₂O), 24 C per wk) to 20°C until d 42. The light cycle was pro- mg; Mn (MnSO₄¢H₂O), 54.06 mg; I (KI), 0.6 mg; and Se (NaSeO3), 0.18 vided 24 h/d from d 1 to d 2, 23 h/d from d 3 to d 10, mg; antioxidant, 0.8 mg. EFFECTS OF COPPER SOURCES ON BROILERS 3 Table 2. Calculated and analyzed Cu concentration in the exper- (ISO, 2002), and crude protein (ISO, 1997). Crude fat imental diets. was analyzed with the Soxhlet method using Foss Sox- Cu level, mg/kg tec/Hydrotec 8000 System for total fat analysis, consist- ing of Soxtec 8000 extraction unit and Hydrotec Calculated Analyzed1 Item Starter Grower Finisher hydrolysis unit, (FOSS Analytical, Denmark). The crude fiber content was also measured using the Weende Negative Control 7 6 9 6 CuSO4 15 22 23 20 method (NF V03−040). The copper content in all the 75 78 103 88 diets was determined using Inductively Coupled Plasma 150 131 213 138 Optical Emission Spectroscopy (ICP-OES, model Cu2O 15 30 23 20 75 95 84 81 Optima 4300DV, PerkinElmer Inc.; Waltham, MA). 150 139 152 169 1 The values expressed as mean based on duplicate determinations. Microbiota 16S rRNA Gene Analysis Library Preparation and Sequencing Bacterial DNA for each phase (21 diets in total) were prepared in a pel- was taken out from 250 mg of ileal content following the leted form (with a size of 1.8 mm for the starter phase manufacturer's instructions with the commercial Mag- and 3 mm for the grower and finisher). Dietary treat- MAX CORE Nucleic Acid Purification Kit 500RXN ments were negative control (NC) diet without Cu sup- (Thermo Fisher, Barcelona, Spain). Mock community plementation and six additional diets in which 15, 75, or DNA was involved as a control (Zymobiomics Microbial 150 mg/kg of Cu from CuSO4 (Copper sulfate, 24.1% Community DNA). Samples were amplified using spe- Cu, Manica Cobre S.L, Spain) or Cu2O (CoRouge, cific primers to the V3-V4 regions of the 16S rRNA 75.4% Cu, Animine, Sillingy, France) were added to the DNA (V3-V4-Forward 50 -TCGTCGGCAGCGTCA- NC diet. The analyzed Cu concentration of each diet is GATGTGTATAAGAGACAGCCTACGGGNGGCW presented in Table 2. The mineral-vitamin premix GCAG-30 , V3-V4-Reverse 50 GTCTCGTGGGCTCGGA included in the diet was formulated and mixed without GATGTGTATAAGAGACAGGACTACHVGGGTAT Cu. Diets were formulated to be isonutritive and to meet CTAATCC-30 ) (Klindworth et al., 2013). The library current estimates for nutrient requirements for growing preparation was performed in Microomics Systems SL broilers (FEDNA, 2018) and without antibiotics and (Barcelona, Spain). growth promoters. Feed and water were offered ad libi- tum. Each diet was sampled in duplicate, grounded, and stored at 4°C for their subsequent analysis. Amplicon Sequences Processing and Analysis Performance Measurements and Sample Forward and reverse reads of raw demultiplexed were Collection processed by following the methods and pipelines as implemented in QIIME2 version 2019.4 with defaulting All the birds were weighed individually on d 0, 21, 35, parameters unless indicated (Bolyen et al., 2019). and 42, and feed intake was recorded at d 21, 35, and 42. DADA2 was used for quality filtering, denoising, pair- The average daily gain (ADG), average daily feed end merging, and amplicon sequence variant calling intake (ADFI), and feed conversion ratio (FCR) were (ASV, i.e., phylotypes) using qiime dada2 denoise- calculated at the end of each phase and for the global paired method (Callahan et al., 2016). Q20 was used as period. Mortality and cause of death were also recorded. a quality threshold to define read sizes for trimming Excreta samples were collected at d 14, 28, and 42, before merging (parameters: −p-trunc-len-f and −p- from three animals per pen of all treatments (10 repli- trunc- len-r). Reads were truncated at the place when cate pens/treatment), and a pool of excreta samples was the 75th percentile Phred score felt below Q20 for both made for each pen and day to analyze AMR. At d 43, 1 forward and reverse reads. After quality filtering steps, broiler per pen with a similar BW to the average of the the average sample size of reads was resolved and phylo- pen was selected. The broilers were stunned using an types were detected. ASVs were aligned using the qiime electrical stunner (Reference: 105523, FAF, France) and alignment mafft method (Katoh and Standley, 2013). immediately exsanguinated to obtain ileal content. The alignment was used to generate a tree and to calcu- Subsequently, based on the performance results, ileal late phylogenetic relations between ASVs using qiime content and excreta samples of NC and 150 mg/kg of phylogeny FastTree method (Price et al., 2010). To Cu from CuSO4 and Cu2O treatments were used for the even sample sizes for the diversity analysis using qiime analysis of microbiota 16S rRNA gene, and AMR, diversity core-metrics-phylogenetic pipeline, ASV tables respectively. were subsampled without replacement. The sample with the smallest size was discarded to take advantage of the Chemical Analysis sequencing depth of the dataset. Afterward, subsam- pling to the next lowest sample size was used for each All the diets were analyzed according to standard comparison. Unweighted and weighted Unifrac distan- methods for dry matter (ISO, 1999), crude ash ces were calculated to compare community structure 4 FOROUZANDEH ET AL. (Lozupone et al., 2011). Taxonomic assignment of ASVs their interaction as a main effect and period as a random was performed using a Bayesian Classifier trained with effect. The LSMeans statement was used to calculate Silva V4 database (i.e., 99% OTUs database) using the mean values for each parameter. The AMR data were qiime feature-classifier classify-sklearn method analyzed using the chi-squared test (Fisher Exact Test). (Pedregosa et al., 2011). Unifrac distance matrices and For microbiota, Alpha and Beta diversity were analyzed ASV tables were used to calculate principal coordinates using Vegan package and taxa differences with the and construct ordination plots using the R software MetagenomeSeq package in open source software RStu- package version 3.6.0 (http://www.R-project.org). dio v.3.5.1. Alpha diversity was calculated with raw counts based on Simpson, Shannon, and Inverse-Simp- son estimators. Beta diversity was evaluated by multi- Antimicrobial Resistance Analysis variate ANOVA based on dissimilarities through envfit and adonis function. Finally, differential abundance Excreta samples (10 replicate pens/treatment) were analysis was performed with taxa relative abundances analyzed for microbiological isolation of Enterococcus under a zero-inflated log-normal mixture model, P-val- spp. and Escherichia coli (E. coli), using Slanetz-Bart- ues were corrected by the false-discovery rate with meta- ley (Oxoid, UK) for 48 h at 37 °C and McConkey agar genomeseq package (Paulson et al., 2017). plates (Oxoid, UK) for 24 hours at 37°C, respectively. The experimental unit was the replicate, and statisti- Compatible colonies with Enterococcus spp. and E. coli cal significance and tendencies were considered at P ≤ were confirmed and identified by PCR (Dutka- 0.05 and 0.05 < P ≤ 0.10, respectively. Malen et al., 1995). Genotypic AMR analysis was done in all the bacterial isolates to detect the resistance genes for vancomycin (vanC1 and vanC2) (Dutka- RESULTS Malen et al., 1995; Kariyama et al., 2000), tetracycline tet(M), and erythromycin erm(B) (Jacob et al., 2008). Growth Performance The detection of extended-spectrum beta-lactamases Growth performance was lower than Ross 308 stand- (ESBL) [blaSHV, blaCTX-M, blaCMY1, blaCMY2, ards, which confirmed that the experimental challenge and blaTEM] and carbapenemase-resistance (OXA-48) impaired the growth of the animals (Table 3). The mor- genes was performed as previously described by tality rate was 2.5% for the overall experiment, with no Vidal et al. (2020). Also, copper (tcrB) and zinc (czcA) differences among the dietary treatments (results not resistance genes were analyzed, as previously described presented). Broilers fed 150 mg/kg of Cu from Cu2O by Hasman et al. (2006). had higher (P = 0.033) BW at d 35, and tended to have In parallel, all Enterococcus spp. and E. coli isolates higher BW (P = 0.053) at d 42 than broilers fed were tested for phenotypic antimicrobial sensitivity 15 mg/kg of Cu from Cu2O, a result which was not using the disk diffusion method, described by observed with CuSO4. Broilers fed 150 mg/kg of Cu Bauer et al., (1966). Thirteen antimicrobial agents were from Cu2O had higher ADG (P = 0.019) than birds fed used: penicillin G (10mg, Oxoid, Basingstoke, UK) ampi- 15 mg/kg of Cu2O, or NC at d 42. cillin (25 mg, BD), imipenem (10 mg, BD), vancomycin Supplementation of Cu from Cu2O, irrespective of (30 mg, BD), erythromycin (15 mg, BD), tetracycline dose, tended to have lower ADFI (P = 0.076) and FCR (30 mg, BD), ciprofloxacin (5 mg, BD), enrofloxacin (5 (P = 0.063) than CuSO4 supplementation at d 42. mg, BD), clindamycin (2 mg, BD), gentamicin (10 mg, BD), kanamycin (30 mg, BD), streptomycin (10 mg, BD) and chloramphenicol (30 mg, BD). Cut-off values were Microbiota 16S rRNA Gene Analysis those defined by the Clinical Laboratory and Standards Both alpha (Shannon, Simpson, and Inverse Simpson Institute. index) and beta diversity metrics were used to estimate Also, minimum inhibitory concentration tests were microbial communities' diversity. Alpha diversity indices performed to assess the susceptibility of Enterococcus showed higher diversity and evenness (P < 0.05) in ileal spp. and E. coli strains to copper (II) sulfate pentahy- microbiota of chickens fed 150 mg/kg of Cu from Cu2O drate (CuSO4¢5H2O) using the broth microdilution compared with NC and 150 mg/kg of Cu from CuSO4 in method as previously reported (Hasman et al., 2006). Shannon index and the NC diet with all the indexes (Table 4). However, there were no differences in beta diver- Statistical Analysis sity among treatments (PENVFIT = 0.4, data not shown). The relative abundance of phyla, families, and genera Growth performance data were analyzed as a com- detected among the experimental groups are illustrated plete randomized design with ANOVA using the GLM in Figure 1. Among 25 recognized phyla, Firmicutes was procedure of SAS software (SAS 9.4 Institute Inc., Cary, the major phyla (average 85.47%), followed by Cyano- NC). Homoscedasticity and variances normal distribu- bacteria and Actinobacteria (average 5.81% and 5.49%, tion were checked before the analysis using the Shapiro- respectively). At the family level, out of 222 different Wilk test and Levene's test from UNIVARIATE and families, 92% of the operational taxonomic unit (OTU) GLM procedures, respectively. For growth performance was allocated to 9 families of Lactobacillaceae parameters, the model included Cu source, Cu dose, and (3053%), Streptococcaceae (826%), Enterococcaceae EFFECTS OF COPPER SOURCES ON BROILERS 5 Table 3. Growth performance (BW, ADFI, ADG, and FCR) of broilers fed dietary treatments.1 CuSO4, mg/kg Cu2O, mg/kg P-value NC SEM Item 15 75 150 15 75 150 Source Dose Source*Dose BW, g d 21 628.7 639.1 629.6 630.9 601.5 631.0 649.7 11.92 0.608 0.421 0.121 d 35 1684.5ab 1718.3ab 1715.4ab 1710.7ab 1616.0b 1718.2ab 1783.6a 29.21 0.748 0.038 0.033 d 42 2420.8xy 2505.6xy 2489.7xy 2473.8xy 2404.4y 2514.9xy 2585.5x 37.51 0.741 0.025 0.053 ADG, g/d d 0-21 28.1 28.4 28.1 28.2 26.8 28.1 29.1 0.56 0.647 0.319 0.166 d 21-35 75.0 77.0 77.6 76.7 72.5 77.7 80.6 1.70 0.912 0.060 0.113 d 35-42 105.2 112.5 108.5 112 108.2 113.8 114.6 2.57 0.615 0.019 0.307 d 0-42 56.6b 58.6ab 58.3ab 57.9ab 55.6b 58.9ab 60.5a 0.88 0.932 0.013 0.019 ADFI, g/d d 0-21 43.3AB 44.0A 43.0AB 41.6B 41.2B 41.8AB 43.3AB 0.53 0.127 0.371 0.001 d 21-35 133.1ab 141.3a 133.2ab 137.2ab 131.1b 131.5b 136.7ab 2.03 0.032 0.062 0.043 d 35-42 180.6 194.2 187.4 187.6 183.3 190.8 189 3.49 0.545 0.052 0.179 d 0-42 96.1B 101.5A 97.4AB 97.1AB 94.8B 96.5B 98.7AB 1.16 0.076 0.267 0.004 FCR, g/g d 0-21 1.54 1.55 1.53 1.48 1.54 1.49 1.49 0.02 0.425 0.035 0.612 d 21-35 1.78 1.84 1.73 1.80 1.81 1.70 1.70 0.04 0.142 0.027 0.594 d 35-42 1.72 1.73 1.73 1.68 1.70 1.69 1.65 0.04 0.374 0.584 0.960 d 0-42 1.70 1.73 1.67 1.68 1.71 1.64 1.63 0.02 0.063 0.004 0.663 Abbreviations: ADG, average daily gain; ADFI, average daily feed intake; BW, body weight; FCR, feed conversion ratio; NC, negative control. a-b Means with different superscripts within a row indicate a significant difference of source*dose (P ≤ 0.05). x-y Means with different superscripts within a row indicate a tendency toward the significance of source*dose (P ≤ 0.1). 1 Data are means of 12 replicates per treatment. (415%), Clostridiaceae (210%), Peptostreptococca- Streptococcaceae (2.02 fold decrease; P < 0.0001), Cory- ceae (310%), Rivulariaceae (47%), Turicibacteraceae nebacteriaceae (1.67 fold decrease; P < 0.0001), and (16%), Corynebacteriaceae (25%), and Enterobac- Enterobacteriaceae (2.94 fold increase; P = 0.0006), teriaceae (0.21.7%), respectively. At the genus level, compared with broilers fed the NC diet (Figure 3). 85.88% of the OTU was assigned to 7 genera of Lactoba- The comparison between Cu sources revealed that the cillus, Streptococcus, Enterococcus, Clostridium, Calo- addition of Cu at 150 mg/kg from Cu2O increased the thrix, Turicibacter, and Alkaliphilus. abundance of Peptostreptococcaceae (1.37 fold; A more in-depth examination of the individual meta- P = 0.0004), Clostridiaceae (1.11 fold; P = 0.032), and genomics profile changes was detected on the dietary tended to increased the abundance of Enterobacteria- treatments using log2 changes. ceae (0.94 fold; P = 0.076), but reduced the amount of Broilers fed 150 mg/kg of Cu from CuSO4 levels com- Firmicutes (0.15 fold; P < 0.0001) phyla, and Strepto- pared with those fed the NC (Figure 2) had significant coccaceae (1.19 fold; P = 0.001) family compared with differences in the relative abundance of Firmicutes (0.28 150 mg/kg of Cu from CuSO4 (Figure 4). fold decrease; P < 0.0001) phyla, and some main families like Lactobacillaceae (0.29 fold decrease; P < 0.0001), Antimicrobial Resistance Analysis Streptococcaceae (0.82 fold decrease; P < 0.0001), Cory- nebacteriaceae (0.52 fold decrease; P < 0.0001), Entero- E. coli was isolated from more than 80% of the excreta coccaceae (0.33 fold increase; P < 0.0001), samples and Enterococcus spp. from all samples. As Peptostreptococcaceae (0.37 fold increase; P = 0.001), regards to Enterococcus spp., E. faecalis was the most Clostridiaceae (0.80 fold increase; P = 0.004), and frequently isolated, representing 70% (in the CuSO4 Enterobacteriaceae (1.99 fold increase; P = 0.013). group) and 90% (in both NC and Cu2O groups) of the Broilers supplementation with 150 mg/kg of Cu from total isolates at d 42. Interestingly, E. faecalis detection Cu2O significantly changed the abundance of Firmicutes was increasing according to days of the study (almost (0.43 fold decrease; P < 0.0001) phyla, and families of exclusively isolated from samples at d 42, and only Lactobacillaceae (1.35 fold decrease; P < 0.0001), detected in 2 samples from the CuSO4 at d 14 and 28). In the genotypical analysis, isolates were negative for ESBL and OXA-48 genes. The percentage of resistant Table 4. Differences in a-diversity indices in ileal microbiota of strains of Enterococcus spp. was higher for the rest of the broilers fed the NC diet or 150 mg/kg of Cu from CuSO4 and studied genes, in all the treatments, and days compared to Cu2O at d 42.1 E. coli strains (Table 5). Moreover, the frequency of tcrB NC CuSO4, mg/kg Cu2O, mg/kg SEM P-value resistant strains had an increasing trend over time in all Item 150 150 treatments (P > 0.1). For the E. coli isolates, the rate of Shannon 1.41B 1.62B 2.13A 0.13 0.003 vancomycin-resistant strains was lower (< 20%) for Simpson 0.57B 0.65AB 0.81A 0.05 0.007 vanC1 and vanC2 genes in all treatments and days. VanA Invsimpson 3.01b 3.30ab 4.99a 0.54 0.030 and VanB genes were not detected in any isolate. Abbreviation: NC, negative control. There were no differences among dietary treatments a-b Means with different superscripts within a row indicate significant differences (P ≤ 0.05). on the phenotypic AMR at d 14 and 28. However, for 1 Data are means of 12 replicates per treatment. the isolated Enterococcus spp., broilers fed the NC diet 6 FOROUZANDEH ET AL. Figure 1. Relative abundance (%) of Top 10: phyla (A); families (B); and genera (C), in the ileum of different experimental groups. The rest of the taxonomic groups are pooled together (those representing less than a mean of 0.01, 1.3, and 1% of phyla, families, and genera, respectively). Abbreviations: NC, negative control; CuSO4 = 150 mg/kg of Cu from CuSO4; Cu2O, 150 mg/kg of Cu from Cu2O. had higher (P < 0.05) resistance to enrofloxacin, genta- this result is in line with Abdelli et al. (2020) who micin, and chloramphenicol compared to animals fed observed a reduction of 21% by reusing commercial litter 150 mg/kg of Cu from CuSO4 and Cu2O on d 42 with NE. In this frame, our result revealed that Cu sup- (Figure 5A). Conversely, the addition of 150 mg/kg of plementation with Cu2O at 150 mg/kg dose increased Cu from CuSO4 and Cu2O in the diet increased the E. the BW up to 10% at d 35 and numerically improved coli resistance to streptomycin (78 and 56%, respec- BW up to 7.5% at d 42 compared with 15 mg/kg dose. tively) and chloramphenicol (56% on average) compared Whereas, supplied Cu as CuSO4 has not modified to the NC diet (11 and 0%, respectively; Figure 5B). growth performance at therapeutic doses (150 mg/kg). Remarkably high levels of AMR were observed in all E. A numerical improvement of ADFI (up to 2%) and FCR coli strains, even in the NC group. (up to 1.8%) was also observed with Cu2O supplementa- tion in comparison to CuSO4. The hypotheses by which Cu stimulates growth DISCUSSION include regulation of intestinal microflora (Pang et al., Copper Effects on Growth Performance 2009), enhancement of neuropeptide Y and its mRNA expression level (Li et al., 2008), and improvement of In this experiment, Clostridium perfringens challenge dietary fat digestibility as a result of stimulated lipase established by reusing 90% recycled commercial litter and phospholipase activities (Luo and Dove, 1996). resulted in reduced growth performance in comparison Previous researches have described that dietary Cu with the standard Ross 308 values (17.8% decrease); can be beneficial for growth performance when fed over EFFECTS OF COPPER SOURCES ON BROILERS 7 Figure 2. Differentially abundant taxa at the phylum and family level from the ileum on d 42 between 150 mg/kg of Cu from CuSO4 and NC. Positive values (green color) and negative values (red color) indicate higher and lower abundance, respectively. Taxa are sorted by level of signifi- cance (from higher to lower). Data are means of 12 observations per treatment. Abbreviation: NC, negative control. Figure 3. Differentially abundant taxa at the phylum and family level from the ileum on d 42 between 150 mg/kg of Cu from Cu2O and NC. Positive values (green color) and negative values (red color) indicate higher and lower abundance, respectively. Taxa are sorted by level of signifi- cance (from higher to lower). Data are means of 12 observations per treatment. Abbreviation: NC, negative control. 8 FOROUZANDEH ET AL. Figure 4. Differentially abundant taxa at the phylum and family level from the ileum on d 42 between 150 mg/kg of Cu from CuSO4 and Cu2O. Positive values (green color) and negative values (red color) indicate higher and lower abundance, respectively. Taxa are sorted by level of signifi- cance (from higher to lower). Data are means of 12 observations per treatment. the minimum requirements in poultry and swine. In a growth improvement and decreased FCR was observed study by Arias and Koutsos (2006) supplementing by Samanta et al. (2011) when broilers fed 150 mg/kg of broilers' diet with 188 mg/kg Cu from CuSO4 or tribasic CuSO4 for 42 days. Similar positive effects of Cu on pigs copper chloride was also improved growth compared were reported by Villagomez-Estrada et al. (2020), with those fed a non-supplemented diet, and growth where 160 mg/kg Cu from CuSO4 or Cu hydroxychlor- improvement was the same with supplementation of ide was able to increase performance at d 42. sub-therapeutic antibiotics at d 45 under immune-chal- The Cu response may, however, depend on the source. lenging conditions (recycled vs. fresh litter). An 8.9% In an experiment conducted by Hamdi et al. (2018), Table 5. Genotypical antimicrobial resistance in isolates of Enterococcus spp. and isolates of E. coli of broilers fed the NC diet and 150 mg/kg of Cu from CuSO4 and Cu2O at d 14, 28, and 42.1 Day 14 Day 28 Day 42 AMR genes2 NC CuSO4, mg/kg Cu2O, mg/kg NC CuSO4, mg/kg Cu2O, mg/kg NC CuSO4, mg/kg Cu2O, mg/kg Enterococcus spp. vanC1 70% 70% 100% 30% 20% 10% 100% 100% 70% vanC2 60% 70% 100% 90% 90% 80% 100% 100% 100% tetM 100% 90% 100% 100% 100% 100% 100% 100% 100% ermB 100% 100% 100% 100% 100% 100% 100% 100% 100% ESBL 0% 0% 0% 0% 0% 0% 0% 0% 0% OXA-48 0% 0% 0% 0% 0% 0% 0% 0% 0% tcrB 90% 80% 80% 80% 60% 90% 100% 100% 100% czcA 10% 10% 0% 0% 10% 0% 50% 40% 0% E. coli vanC1 20% 0% 0% 0% 20% 0% 0% 0% 0% vanC2 20% 0% 30% 0% 10% 0% 0% 0% 0% tetM 89% 10% 0% 89% 90% 86% 0% 0% 0% ermB 78% 100% 10% 78% 80% 100% 50% 70% 20% ESBL 0% 0% 0% 0% 0% 0% 0% 0% 0% OXA-48 0% 0% 0% 0% 0% 0% 0% 0% 0% tcrB 0% 0% 0% 0% 10% 0% 70% 20% 40% czcA 100% 50% 50% 78% 100% 43% 0% 0% 0% 1 Data are means of 10 replicates per treatment. 2 Antimicrobial resistance genes: vancomycin (vanC1, and vanC2); tetracycline (tetM); erythromycin (ermB); penicillin, aminopenicillin and last gener- ation cephalosporine (ESBL); imipenem (OXA-48); Cu (tcrB); Zinc (czcA). EFFECTS OF COPPER SOURCES ON BROILERS 9 Figure 5. Phenotypic antimicrobial resistance in Enterococcus spp. isolates (A), and E. coli isolates (B) of broilers fed the NC diet and 150 mg/kg of Cu from CuSO4 and Cu2O at d 42. Abbreviations: NC, negative control; AMR, antimicrobial resistance. dietary level of 150 mg/kg of Cu from Cu2O increased bacteria or mold by denaturation or an oxidation mech- BW at d 35, whereas supplied Cu as CuSO4 at the same anism (Kim et al., 2007; Lok et al., 2007). The electro- dosage did not improve growth performance compared static attraction combines ionic Cu with the plasma with 15 mg/kg. Our findings were in line with membrane and results in the cell membrane penetration Hamdi et al. (2018). In another study on broilers, through opening or closing of the membrane channel. Lu et al. (2010) indicated that adding 200 mg/kg of Cu This process leads to the leakage of intracellular ions from tribasic copper chloride improved ADG without and low molecular-weight metabolites by altering the increasing ADFI compared with 200 mg/kg of Cu from permeability of cellular membranes (Tong et al., 2005). CuSO4 or other doses of both sources. Meanwhile, Cu2+ enters into the cell, induces plasmid The negligible impact of CuSO4 on growth compared DNA degradation (Giannousi et al., 2014), and leads to with other sources of Cu can be attributed to 1) damage bacterial death (Tong et al., 2005). to the mucosa and muscular layer in the intestinal tract Some researchers have reported that high dietary Cu (Chiou et al., 1999); 2) higher solubility (Pang and has affected gut microbiota profile and reduced the Applegate, 2006); 3) higher oxidation (Miles et al., growth of pathogenic bacteria in animals (Højberg et al., 1998); 4) reduced phytase efficacy and decreased appar- 2005; Zhang et al., 2017). Xia et al. (2004) reported that ent phosphorus retention; and 5) toxicity (Banks et al., the positive effect of Cu on weight gain in broiler chick- 2004a; Lu et al., 2010; Hamdi et al., 2018). ens might be an outcome of the significant reduction of the total pathogenic organism in the gut that intervenes with weight gain. Copper Effects on the Gut Microbiota Profile The analysis of ileal microbiota in the present study showed significant changes in some families of the gas- One of the growth-promoting actions of Cu has been trointestinal tract in broilers fed Cu. Supplementation credited to its antimicrobial effect in the gastrointestinal of Cu (CuSO4 or Cu2O) in broilers' diet compared with tract. Copper ions are toxic and can effectively kill the non-supplemented diet suppressed the abundance of 10 FOROUZANDEH ET AL. Streptococcaceae. The genera Streptococcus is active in supplementation than those supplemented with Cu the process of simple sugar fermentation into lactate (CuSO4 or Cu2O). Lactobacillaceae (Lactobacillus (Garvie, 1980; Zoetendal et al., 2012). Whereas, some spp.) is one of the main lactic acid-producing bacteria, species of the genus (e.g., Streptococcus bovis) are con- and the primary end product of these bacteria is lactic sidered as major opportunistic pathogens, which can acid (Garvie, 1980; Rajilic-Stojanovic and de Vos, 2014) result in many diseases (Abdulamir et al., 2011; which has positive effects on growth. Interestingly, in a Munita et al., 2012; Qiao et al., 2014). A decreased study by Gharib-Naseri et al. (2019), it has been Streptococcus abundance in colonic microbiota and asserted that increased Lactobacillus in the intestine increased growth performance have been observed in may not indicate a healthier gut. Modified microbiota pigs fed 160 mg/kg of Cu (Villagomez-Estrada et al., composition in challenging conditions could affect avail- 2020). Copper supplementation also declined the abun- able nutrients for bacteria, and therefore, bacterial dance of Corynebacteriaceae. Members of this family dynamics in the intestine (Stanley et al., 2012). Similar have been positively correlated with a wide range of results were published by Park and Kim (2018). The severe infections, including opportunistic infections in authors reported an increase in the Lactobacillus popula- both humans and animals (Prada et al., 1994; Zhi et al., tion in the ileal digesta of broilers that received essential 2017). Therefore, a lower proportion of Corynebacteria- oils, but their BW were not significantly different from ceae may indicate a healthier intestinal environment. broilers in the non-supplemented group. Another family that responded to the treatments was Comparing ileal microbiota of broilers fed Cu sources Enterobacteriaceae, whose abundance was increased by indicated that supplementation of Cu from Cu2O was Cu supplementation. The family Enterobacteriaceae more effective than CuSO4 towards the reduction of includes 51 genera, which consist of commensal and Streptococcaceae and development of Peptostreptococ- pathogenic microorganisms (Janda, 2006). Opportunis- caceae and Clostridiaceae which have beneficial proper- tic pathogens, for example, Citrobacter, Enterobacter, ties. Bactericidal action of Cu contributed to the Escherichia, Klebsiella, Serratia, and Proteus have been concentration of free ionic Cu in solution associated with diarrhea, urinary tract infections, masti- (Menkissoglu and Lindow, 1991). Therefore, reduced tis, arthritis, and meningitis (Fairbrother et al., 2005; copper states, such as Cu2O, can provide Cu ion release Nagy and Fekete, 2005). However, low levels (0.86 to more sustainably (Ren et al., 2011), and may exhibit a 1.52%) of the mentioned pathogens were present in higher antibacterial activity (Dunning et al., 1998). broilers fed Cu supplementation. Moreover, microbiota from the Cu2O treatment group The reduction of families containing pathogenic bacteria was more diverse, and OTU's were more evenly distrib- as a result of high Cu supplementation may lay the ground uted, compared to the NC and CuSO4 treatment groups. for the growth of other families. Adding 150 mg/kg of Cu A correlation between FCR and richness and evenness from CuSO4 into the diet promoted the colonization of indices has been previously observed by Enterococcaceae, Peptostreptococcaceae, and Clostridia- Stanley et al. (2016). In their experiments, broilers with ceae compared with the non-supplemented diet. low FCR showed higher diversity than those with high Enterococcaceae (Enterococcus spp.) belongs to the FCR. Furthermore, in a recent study, Villag omez- group of lactic acid bacteria. The genera consist of more Estrada et al. (2020) discuss that reduction of opportu- than 20 species (Gomes et al., 2008). Some strains of this nistic pathogens from one hand and development of sap- genus are capable of inhibiting the development of specific rophytic bacteria from the other hand, could lead to a pathogens (Foulquie Moreno et al., 2006), and exhibit pro- significant improvement in intestinal nutrient absorp- biotic properties (O  Cuív et al., 2013; Yadav and tion and, eventually, feed efficiency in pig fed 160 mg/kg Jha, 2019). Peptostreptococcaceae and Clostridiaceae are of Cu from CuSO4 or Cu hydroxychloride. members of Firmicutes phylum. Peptostreptococcaceae In agreement with Villagomez-Estrada et al., (2020), reported as normal commensal bacteria with a higher pro- our results suggest that adding 150 mg/kg of Cu, partic- portion in the gut microbiota of healthy animals than ularly Cu2O, appears to improve intestinal microbiota those experiencing dysbiosis of the intestinal microbiota. profile and enhance chickens' performance by increasing It indicates that this family helps preserve gut homeostasis the abundance of reportedly beneficial bacteria, such as (Fan et al., 2017). Clostridiaceae is one of the potential Peptostreptococcaceae and Clostridiaceae, reducing the phylotypes involved in butyrate production from glucose, colonization of harmful bacteria, and increasing the succinate, and lactate (Esquivel-Elizondo et al., 2017). diversity and evenness of ileal microbiota. However, fur- Also, it has been highly correlated to protein and fat ther work is s required to understand how these changes digestibility in dogs (Bermingham et al., 2017). The abun- in bacterial composition relate to metabolic changes in dance of Peptostreptococcaceae and Clostridiaceae fami- the host that ultimately lead to improved performance. lies have been previously shown to be associated with improved performance in broilers fed a mixture of organic acids with aromatic compounds or organic acids with Copper Effects on the Antimicrobial medium-chain fatty acid plus aromatic compound Resistance (Abdelli et al., 2020). On the other hand, the abundance of Lactobacillaceae As an alternative to antibiotics, metal poisoning is was remarkably greater in broilers without Cu used to destroy bacteria (Hao et al., 2016). This has EFFECTS OF COPPER SOURCES ON BROILERS 11 resulted in the emergence and prevalence of AMR, repre- The use of avoparcin, whose structure is similar to senting a severe threat to public health worldwide that of vancomycin, has decreased over the years in ani- (Hammerum and Heuer, 2009). High dietary Cu may mal feed, and it has resulted in a reduction of the num- have undesired effects, such as the growth of Cu-resis- ber of vancomycin-resistant isolates (Yazdankhah et al., tant bacteria (Pang et al., 2009). Copper resistance 2014). Regarding the vancomycin resistance genes, high genes are usually located on plasmids and, in most cases, prevalences for the vanC1 and vanC2 genes were are transferable (Hasman and Aarestrup, 2002). These detected. The vanC genes are associated with low-level plasmids conferring resistance to copper (tcrB) have vancomycin resistance (Watanabe et al., 2009; been identified in several Enterococcus species, including de Moura et al., 2013), and are considered intrinsic in E. faecium and E. faecalis, in pigs, poultry, calves, and some enterococcal species, such as E. gallinarum and E. also humans (Torres et al., 2018). In Enterococcus spp. casseliflavus (Monticelli et al., 2018). The presence of the most common genes conferring resistance to antibi- these genes could be explained by the presence of this otics are for erythromycin, tetracycline, and vancomycin latter species on the excreta samples, and the possible (Oravcova et al., 2019; Tian et al., 2019). transmission of plasmids containing the resistance genes On the other hand, there is high diversity and variants between enterococcal species. Also, the presence of the of E. coli strains integrating the normal gut microbiota vanC gene can explain the intermediate phenotypical and can cause severe diseases in both animals and resistance found in some of the analyzed isolates. humans, such as urinary tract infections, diarrhea, enteri- The phenotypical resistance in Enterococcal and E. tis, and septicemia (Li et al., 2019). The routine use of coli isolates was high in almost all the antimicrobial antimicrobials in livestock for either “prophylaxis” or agents. This high prevalence may be due to the fact that “metaphylaxis” has represented a serious hazard for the this study was carried in the north-east part of Spain, selection of multidrug-resistant Enterobacteriaceae strains which has a very high density of pig production, where (Angulo et al., 2004). The effectiveness of treatments antimicrobial agents are widely used and the prevalence against E. coli is threatened by the dramatic increase of of AMR genes, principally in E. coli strains, has been extended-spectrum beta-lactamases producing isolates highly reported (Vidal et al., 2020). Subsequently, worldwide (Livermore et al., 2007; Carattoli et al., 2017). observing any significant effect on interventional groups Regarding Cu resistance genes in the present study, a is challenging, given the high background levels of AMR high prevalence of tcrB was detected in Enterococcus in the non-supplemented group. spp. isolates mainly on d 42, without any difference Likewise, most of the E. coli strains were resistant to between groups. In the case of E. coli, tcrB was less fre- vancomycin. This resistance was expected as vancomy- quent, but a higher prevalence was found at d 42. The cin has been designed to kill a different type of microbe: presence of tcrB is not associated with Cu addition in gram-positive cocci. However, there were differences the diet, as the presence of this gene was higher in the between Cu non-supplemented and supplemented in E. NC group. Another study has reported lower frequencies coli isolates for streptomycin and chloramphenicol, of tcrB gene in enterococcal isolates (34%) in broiler where birds supplemented with Cu had higher resistance chickens, but higher levels of prevalence (76%) in pigs than non-supplemented. Although in Agga et al. (2014) (Hasman and Aarestrup, 2002). study with pigs, they did not observe any difference in The presence of the zinc resistance gene czcA was the resistance prevalence of Streptomycin between Cu barely detected in enterococcal isolates in the first two supplementations or without, but higher chlorampheni- samplings (d 14 and 28). However, in the third sampling col prevalence was found in non-Cu-supplemented pigs. (d 42), the prevalence reached 50% in NC. Contrarily, in However, the mean prevalence was similar for both stud- E. coli, the prevalence diminished significantly between ies. For other antimicrobial agents (ampicillin, gentami- samplings, going from 100% at d 14 to 0% at d 42. These cin, and kanamycin), the prevalences were similar differences were observed in all the treatments; therefore, between both studies, which suggests that Cu supple- they cannot be correlated with Cu addition in the diet. mentation could have a similar effect in antimicrobial Overall, a high prevalence of AMR genes was resistance for pigs and poultry. Controversially, in the observed in Enterococcus spp. isolates. Erythromycin Enterococcus isolates, no resistance was observed in use is permitted in chickens, and laying hens and tetra- vancomycin, and higher resistance prevalence was cyclines have been widely used as a growth promoter in observed for non-Cu-supplemented birds in enrofloxacin, animal husbandry (Granados-Chinchilla and Rodrí- gentamicin, and chloramphenicol. Therefore, further guez, 2017). Therefore, the high prevalence of ermB and studies with more controlled environmental conditions tetM could be explained by the frequent use of these should implement in this field to assess the origin of antimicrobials in broiler farms over the years. A high these AMR genes. prevalence of ermB and tetM in broiler excreta has been reported before in Enterococcus spp. isolates (Cauwerts et al., 2007; Tremblay et al., 2011; CONCLUSION Hasan et al., 2018). Also, Cauwerts et al. (2007) found a correlation between the presence of ermB and phenotyp- To conclude, supplementation of the high dose of ical resistance to tetracyclines, mediated by several tet dicopper oxide (150 mg Cu/kg) was able to enhance the genes, tetM being among them. growth performance of broiler chickens raised under 12 FOROUZANDEH ET AL. challenging conditions by modulating bacterial commu- A. M. Caraballo-Rodríguez, J. Chase, E. K. Cope, R. Da Silva, nities in the ileum. Finally, Cu addition did not alter the C. Diener, P. C. Dorrestein, G. M. Douglas, D. M. Durall, C. Duvallet, C. F. Edwardson, M. Ernst, M. Estaki, J. Fouquier, AMR genes in this study, which suggests that using J. M. Gauglitz, S. M. Gibbons, D. L. Gibson, A. Gonzalez, broilers in a reused litter does not seem an appropriate K. Gorlick, J. Guo, B. Hillmann, S. Holmes, H. Holste, method to check for AMR genes. C. Huttenhower, G. A. Huttley, S. Janssen, A. K. Jarmusch, L. Jiang, B. D. Kaehler, K. Bin Kang, C. R. Keefe, P. Keim, S. T. Kelley, D. Knights, I. Koester, T. Kosciolek, J. Kreps, ACKNOWLEDGMENTS M. G. I. Langille, J. Lee, R. Ley, Y. X. Liu, E. Loftfield, C. Lozupone, M. Maher, C. Marotz, B. D. Martin, D. McDonald, L. J. McIver, A. V. Melnik, J. L. Metcalf, S. C. Morgan, This research (Project E! 11780) was sponsored by the J. T. Morton, A. T. Naimey, J. A. Navas-Molina, L. F. Nothias, Eurostars-2 joint program with co-financing from the S. B. Orchanian, T. Pearson, S. L. Peoples, D. Petras, BPI in France and CDTI in Spain, from the Horizon M. L. Preuss, E. Pruesse, L. B. Rasmussen, A. Rivers, 2020 Research and Innovation Framework Program of M. S. Robeson, P. Rosenthal, N. Segata, M. Shaffer, A. Shiffer, R. Sinha, S. J. Song, J. R. Spear, A. D. Swafford, L. R. Thompson, the European Union. This work was also supported by P. J. Torres, P. Trinh, A. Tripathi, P. J. Turnbaugh, S. Ul-Hasan, Agencia de Gesti o d’Ajusts Universitaris i de Recerca de J. J. J. van der Hooft, F. Vargas, Y. V azquez-Baeza, la Generalitat de Catalunya (Ph.D. grant 2019FI_B E. Vogtmann, M. von Hippel, W. Walters, Y. Wan, M. Wang, 00282). J. Warren, K. C. Weber, C. H. D. Williamson, A. D. Willis, Z. Z. Xu, J. R. Zaneveld, Y. Zhang, Q. Zhu, R. Knight, and J. G. Caporaso. 2019. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat. Biotech- DISCLOSURES nol. 37:852–857. Bunch, R. J., J. T. Mccall, V. C. Speer, and V. W. Hays. 1965. Copper The authors declare that they have no known compet- supplementation for weanling pigs. J. Anim. Sci. 24:995–1000. ing financial interests or personal relationships that Callahan, B. J., P. J. McMurdie, M. J. Rosen, A. W. Han, could have appeared to influence the work reported in A. J. A. Johnson, and S. P. Holmes. 2016. DADA2: high-resolution sample inference from Illumina amplicon data. Nat. Methods this paper. 13:581–583. Carattoli, A., L. Villa, C. Feudi, L. Curcio, S. Orsini, A. Luppi, G. Pezzotti, and C. F. Magistrali. 2017. Novel plasmid-mediated REFERENCES colistin resistance mcr-4 g ene in Salmonella and Escherichia coli, Italy 2013, Spain and Belgium, 2015 to 2016. Euro surveill Abdelli, N., J. F. Perez, E. Vilarrasa, I. Cabeza Luna, D. Melo-Duran, 22:30589. M. D’Angelo, and D. Sol a-Oriol. 2020. Targeted-release organic Cauwerts, K., A. Decostere, E. M. De Graef, F. Haesebrouck, and acids and essential oils improve performance and digestive function F. Pasmans. 2007. High prevalence of tetracycline resistance in in broilers under a necrotic enteritis challenge. Animals 10:259. Enterococcus isolates from broilers carrying the erm(B) gene. Abdulamir, A. S., R. R. Hafidh, and F. Abu Bakar. 2011. The associa- Avian Pathol 36:395–399. tion of Streptococcus bovis/gallolyticus with colorectal tumors: Chiou, P. W. S., C. L. Chen, K. L. Chen, and C. P. Wu. 1999. Effect the nature and the underlying mechanisms of its etiological role. J. of high dietary copper on the morphology of gastro-intestinal tract Exp. Clin. Cancer Res. 30:1–13. in broiler chickens. Asian-Australasian J. Anim. Sci. 12:548–553. Agga, G. E., H. M. Scott, R. G. Amachawadi, T. G. Nagaraja, Davis, K. G., and W. Mertz. 1987. Trace elements in human and ani- J. Vinasco, J. Bai, B. Norby, D. G. Renter, S. S. Dritz, mal nutrition. Pages 301364 in Copper. (5th ed). academic press, J. L. Nelssen, and M. D. Tokach. 2014. Effects of chlortetracycline New York, NY. and copper supplementation on antimicrobial resistance of fecal Dunning, J. C., Y. Ma, and R. E. Marquis. 1998. Anaerobic killing of Escherichia coli from weaned pigs. Prev. Vet. Med. 114:231–246. oral Streptococci by reduced, transition metal cations. Appl. Envi- Angulo, F. J., V. N. Nargund, and T. C. Chiller. 2004. Evidence of an ron. Microbiol. 64:27–33. association between use of anti-microbial agents in food animals Dutka-Malen, S., S. Evers, and P. Courvalin. 1995. Detection of gly- and anti-microbial resistance among bacteria isolated from copeptide resistance genotypes and identification to the species humans and the human health consequences of such resistance. J. level of clinically relevant enterococci by PCR. J. Clin. Microbiol. Vet. Med. B 51:374–379. 33:24–27. Arias, V. J., and E. A. Koutsos. 2006. Effects of copper source and EFSA, (European Food Safety Authority). 2012. Scientific opinion on level on intestinal physiology and growth of broiler chickens. the safety and efficacy of copper compounds (E4) as feed additives for Poult. Sci. 85:999–1007. all animal species: cupric sulphate pentahydrate based on a dossier Baker, D. H. 1999. Cupric oxide should not be used as a copper sup- submitted by Manica S.p.A. EFSA J. 10:1–38. plement for either animals or humans. J. Nutr. 129:2278–2279. Esquivel-Elizondo, S., Z. E. Ilhan, E. I. Garcia-Pe~ na, and Banks, K. M., K. L. Thompson, P. Jaynes, and R. Krajmalnik-Brown. 2017. Insights into butyrate production in a T. J. Applegate. 2004a. The effects of copper on the efficacy of phy- controlled fermentation system via gene predictions. mSystems 2:1–13. tase, growth, and phosphorus retention in broiler chicks. Poult. European Commision. 2010. Directive 2010/63/EU of the European Sci. 83:1335–1341. Parliament and of The Council of 22 September 2010 on the pro- Banks, K. M., K. L. Thompson, J. K. Rush, and tection of animals used for scientific purposes. Off. J. Eur. Union T. J. Applegate. 2004b. Effects of copper source on phosphorus L276:33–79. retention in broiler chicks and laying hens. Poult. Sci. 83:990–996. Fairbrother, J. M., E. Nadeau, and C. L. Gyles. 2005. Escherichia coli Bauer, A. W., W. M. M. Kirby, J. C. Sherris, and M Turck. 1966. in postweaning diarrhea in pigs: an update on bacterial types, Antibiotic susceptibility testing by a standardized single disk pathogenesis, and prevention strategies. Anim. Heal. Res. Rev. method. Am. J. Clin. Pathol. 36:493–496. 6:17–39. Bermingham, E. N., P. Maclean, D. G. Thomas, N. J. Cave, and Fan, P., P. Liu, P. Song, X. Chen, and X. Ma. 2017. Moderate W. Young. 2017. Key bacterial families (Clostridiaceae, Erysipelo- dietary protein restriction alters the composition of gut micro- trichaceae and Bacteroidaceae) are related to the digestion of pro- biota and improves ileal barrier function in adult pig model. tein and energy in dogs. PeerJ. 2017 5:e3019. Sci. Rep. 7:1–12. Bolyen, E., J. R. Rideout, M. R. Dillon, N. A. Bokulich, C. C. Abnet, FEDNA (Fundaci on Espa~ nola para el Desarrollo de la Nutrici on Ani- G. A. Al-Ghalith, H. Alexander, E. J. Alm, M. Arumugam, mal). 2018. Necesidades Nutricionales para Avicultura: Pollos de F. Asnicar, Y. Bai, J. E. Bisanz, K. Bittinger, A. Brejnrod, Carne y Aves de Puesta. In FEDNA. R. Lazaro and G. G. Mateos, C. J. Brislawn, C. T. Brown, B. J. Callahan, eds. FEDNA, Madrid, Spain. EFFECTS OF COPPER SOURCES ON BROILERS 13 Foulquie Moreno, M. R., P. Sarantinopoulos, E. Tsakalidou, and surveillance isolates of vancomycin-resistant enterococci. J. Clin. L. De Vuyst. 2006. The role and application of enterococci in food Microbiol. 38:3092–3095. and health. Int. J. Food Microbiol. 106:1–24. Katoh, K., and D. M. Standley. 2013. MAFFT multiple sequence Garvie, E. I. 1980. Bacterial lactate dehydrogenases. Microbiol. Rev. alignment software version 7: improvements in performance and 44:106–139. usability. Mol. Biol. Evol. 30:772–780. Gharib-Naseri, K., S. K. Kheravii, C. Keerqin, N. Morgan, Kim, J. S., E. Kuk, K. N. Yu, J. H. Kim, S. J. Park, H. J. Lee, S. H. Kim, R. A. Swick, M. Choct, and S. B. Wu. 2019. Two different Clos- Y. K. Park, Y. H. Park, C. Y. Hwang, Y. K. Kim, Y. S. Lee, tridium perfringens strains produce different levels of necrotic D. H. Jeong, and M. H. Cho. 2007. Antimicrobial effects of silver enteritis in broiler chickens. Poult. Sci. 98:6422–6432. nanoparticles. Nanomed. Nanotechnol. Biol. Med. 3:95–101. Giannousi, K., K. Lafazanis, J. Arvanitidis, A. Pantazaki, and Kim, B. E., T. Nevitt, and D. J. Thiele. 2008. Mechanisms for copper C. Dendrinou-Samara. 2014. Hydrothermal synthesis of copper acquisition, distribution and regulation. Nat. Chem. Biol. 4:176– based nanoparticles: antimicrobial screening and interaction with 185. DNA. J. Inorg. Biochem. 133:24–32. Klindworth, A., E. Pruesse, T. Schweer, J. Peplies, C. Quast, Gomes, B. C., C. T. Esteves, I. C. V. Palazzo, A. L. C. Darini, M. Horn, and F. O. Gl€ ockner. 2013. Evaluation of general 16S ribo- G. E. Felis, L. A. Sechi, B. D. G. M. Franco, and somal RNA gene PCR primers for classical and next-generation E. C. P. De Martinis. 2008. Prevalence and characterization of sequencing-based diversity studies. Nucleic Acids Res 41:1–11. Enterococcus spp. isolated from Brazilian foods. Food Microbiol Li, Q., W. Chang, H. Zhang, D. Hu, and X. Wang. 2019. The role of 25:668–675. plasmids in the multiple antibiotic resistance transfer in ESBLs- Granados-Chinchilla, F., and C. Rodríguez. 2017. Tetracyclines in producing Escherichia coli isolated from wastewater treatment food and feedingstuffs: from regulation to analytical methods, bac- plants. Front. Microbiol. 10:1–8. terial resistance, and environmental and health implications. J. Li, J., L. Yan, X. Zheng, G. Liu, N. Zhang, and Z. Wang. 2008. Effect Anal. Methods Chem. 2017:1–24. of high dietary copper on weight gain and neuropeptide Y level in Hamdi, M., D. Sola, R. Franco, S. Durosoy, A. Romeo, and the hypothalamus of pigs. J. Trace Elem. Med. Biol. 22:33–38. J. F. Perez. 2018. Including copper sulphate or dicopper oxide in Livermore, D. M., R. Canton, M. Gniadkowski, P. Nordmann, the diet of broiler chickens affects performance and copper content G. M. Rossolini, G. Arlet, J. Ayala, T. M. Coque, in the liver. Anim. Feed Sci. Technol. 237:89–97. I. Kern-Zdanowicz, F. Luzzaro, L. Poirel, and N. Woodford. 2007. Hammerum, A. M., and O. E. Heuer. 2009. Human health hazards CTX-M: changing the face of ESBLs in Europe. J. Antimicrob. from antimicrobial-resistant Escherichia coli of animal origin. Chemother. 59:165–174. Clin. Infect. Dis. 48:916–921. Lok, C. N., C. M. Ho, R. Chen, Q. Y. He, W. Y. Yu, H. Sun, Hao, X., F. L€ uthje, R. Rønn, N. A. German, X. Li, F. Huang, P. K. H. Tam, J. F. Chiu, and C. M. Che. 2007. Silver nanopar- J. Kisaka, D. Huffman, H. A. Alwathnani, Y. G. Zhu, and ticles: partial oxidation and antibacterial activities. J. Biol. Inorg. C. Rensing. 2016. A role for copper in protozoan grazing − two bil- Chem. 12:527–534. lion years selecting for bacterial copper resistance. Mol. Microbiol. Lozupone, C., M. E. Lladser, D. Knights, J. Stombaugh, and 102:628–641. R. Knight. 2011. UniFrac: an effective distance metric for micro- Hasan, K. A., S. A. Ali, M. Rehman, H. Bin-Asif, and S. Zahid. 2018. bial community comparison. ISME J 5:169–172. The unravelled Enterococcus faecalis zoonotic superbugs: emerg- Lu, L., R. L. Wang, Z. J. Zhang, F. A. Steward, X. Luo, and ing multiple resistant and virulent lineages isolated from poultry B. Liu. 2010. Effect of dietary supplementation with copper sulfate environment. Zoonoses Public Health 65:921–935. or tribasic copper chloride on the growth performance, liver copper Hasman, H., and F. M. Aarestrup. 2002. tcrb, a gene conferring concentrations of broilers fed in floor pens, and stabilities of vita- transferable copper resistance in Enterococcus faecium: occur- min e and phytase in feeds. Biol. Trace Elem. Res. 138:181–189. rence, transferability, and linkage to macrolide and glycopep- Luo, X. G., and C. R. Dove. 1996. Effect of dietary copper and fat on tide resistance. Antimicrob. Agents Chemother. 46:1410– nutrient utilization, digestive enzyme activities, and tissue mineral 1416. levels in weanling pigs. J. Anim. Sci. 74:1888–1896. Hasman, H., I. Kempf, B. Chidaine, R. Cariolet, A. K. Ersbøll, Malan, M., F. M€ uller, L. Raitt, J. Aalbers, L. Cyster, and H. Houe, H. C. B. Hansen, and F. M. Aarestrup. 2006. Copper L. Brendonck. 2015. Farmyard manures: the major agronomic resistance in Enterococcus faecium, mediated by the tcrB gene, is sources of heavy metals in the Philippi Horticultural Area in the selected by supplementation of pig feed with copper sulfate. Appl. Western Cape Province of South Africa. Environ. Monit. Assess. Environ. Microbiol. 72:5784–5789. 187:708. Hawbaker, J. A., V. C. Speer, V. W. Hays, and D. V. Catron. 1961. Menkissoglu, O., and S. E. Lindow. 1991. Relationship of free ionic Effect of copper sulfate and other chemotherapeutics in growing copper and toxicity of bacteria in solutions of organic compounds. swine rations. J. Anim. Sci. 20:163–167. Phytopathology 81:1258–1263. Højberg, O., N. Canibe, H. D. Poulsen, M. S. Hedemann, and Miles, R. D., S. F. O. Keefe, P. R. Henry, C. B. Ammerman, and B. B. Jensen. 2005. Influence of dietary zinc oxide and copper sul- X. G. Luo. 1998. The effect of dietary supplementation with cop- fate on the gastrointestinal ecosystem in newly weaned piglets. per sulfate or tribasic copper chloride on broiler performance, rela- Appl. Environ. Microbiol. 71:2267–2277. tive copper bioavailability, and dietary prooxidant activity 1. International Organization for Standardization (ISO). 1997. ISO Poult. Sci. 416–425. 5983:1997. Animal feeding stuffs - Determination of Nitrogen Con- Monticelli, J., A. Knezevich, R. Luzzati, and S. Di Bella. 2018. Clini- tent and Calculation of Crude Protein Content. Kjeldahl method. cal management of non-faecium non-faecalis vancomycin-resistant ISO, Geneva, Switzerland. enterococci infection. Focus on Enterococcus gallinarum and International organization for standardization (ISO). 1999. ISO Enterococcus casseliflavus/flavescens. J. Infect. Chemother. 6496:1999. Animal feeding stuffs - Determination of moisture and 24:237–246. other volatile matter content. ISO, Geneva, Switzerland. de Moura, T. M., A. P. V. Cassenego, F. S. Campos, A. M. L. Ribeiro, International organization for standardization (ISO). 2002. ISO A. C. Franco, P. A. d’Azevedo, J. Frazzon, and 5984:2005. Animal feeding stuffs - Determination of crude ash con- A. P. G. Frazzon. 2013. Detection of vanC1 gene transcription in tent. ISO, Geneva, Switzerland. vancomycin-susceptible enterococcus faecalis. Mem. Inst. Oswaldo Jacob, M. E., J. T. Fox, S. K. Narayanan, J. S. Drouillard, Cruz 108:453–456. D. G. Renter, and T. G. Nagaraja. 2008. Effects of feeding wet Munita, J. M., C. A. Arias, and B. E. Murray. 2012. Enterococcal corn distillers grains with solubles with or without monensin and endocarditis: can we win the war? Curr. Infect. Dis. Rep. 14:339– tylosin on the prevalence and antimicrobial susceptibilities of fecal 349. foodborne pathogenic and commensal bacteria in feedlot cattle. J. Nagy, B., and P. Z. Fekete. 2005. Enterotoxigenic Escherichia coli in Anim. Sci. 86:1182–1190. veterinary medicine. Int. J. Med. Microbiol. 295:443–454. Janda, J. M. 2006. New members of the family Enterobacteriaceae. National Research Council. 1994. Nutrient Requirements of Poultry. Pages 5−40 in The Prokaryotes. Springer, New York, NY. (9th ed). Natl. Acad. Press, Washington, DC. Kariyama, R., R. Mitsuhata, J. W. Chow, D. B. Clewell, and  Cuív, P., E. S. Klaassens, W. J. Smith, S. Mondot, A. S. Durkin, O H. Kumon. 2000. Simple and reliable multiplex PCR assay for D. M. Harkins, L. Foster, J. Mccorrison, M. Torralba, and 14 FOROUZANDEH ET AL. K. E. Nelson. 2013. Probiotic strain isolated from human feces. identification of performance enhancing probiotic bacteria. Front. Genome Announc 1:12–13. Microbiol. 7:1–13. Ognik, K., A. Stȩpniowska, E. Cholewi nska, and K. Koz»owski. 2016. Stanley, D., A. L. Keyburn, S. E. Denman, and R. J. Moore. 2012. The effect of administration of copper nanoparticles to chickens in Changes in the caecal microflora of chickens following Clostridium drinking water on estimated intestinal absorption of iron, zinc, perfringens challenge to induce necrotic enteritis. Vet. Microbiol. and calcium. Poult. Sci. 95:2045–2051. 159:155–162. Oravcova, V., M. Kolar, and I. Literak. 2019. Highly variable vanco- Tian, Y., H. Yu, and Z. Wang. 2019. Distribution of acquired antibi- mycin-resistant enterococci in the north-eastern part of the Czech otic resistance genes among Enterococcus spp. isolated from a hos- Republic. Lett. Appl. Microbiol. 69:16–22. pital in Baotou, China. BMC Res. Notes 12:12–16. Pang, Y., and T. J. Applegate. 2006. Effects of copper source and con- Tong, G., M. Yulong, G. Peng, and X. Zirong. 2005. Antibacterial centration on in vitro phytate phosphorus hydrolysis by phytase. effects of the Cu(II)-exchanged montmorillonite on Escherichia J. Agric. Food Chem. 54:1792–1796. coli K88 and Salmonella choleraesuis. Vet. Microbiol. 105:113–122. Pang, Y., J. A. Patterson, and T. J. Applegate. 2009. The influence of Torres, C., C. A. Alonso, L. Ruiz-Ripa, R. Le on-Sampedro, copper concentration and source on ileal microbiota. Poult. Sci. R. Del Campo, and T. M. Coque. 2018. Antimicrobial resistance in 88:586–592. Enterococcus spp. of animal origin. Microbiol. Spectr. 6:185–227. Park, J. H., and I. H. Kim. 2018. Effects of a protease and essential Tremblay, C. L., A. Letellier, S. Quessy, M. Boulianne, D. Daignault, oils on growth performance, blood cell profiles, nutrient retention, and M. Archambault. 2011. Multiple-antibiotic resistance of ileal microbiota, excreta gas emission, and breast meat quality in Enterococcus faecalis and Enterococcus faecium from cecal con- broiler chicks. Poult. Sci. 97:2854–2860. tents in broiler chicken and Turkey flocks slaughtered in Canada Paulson, J. N., H. Talukder, and H. C. Bravo. 2017. Longitudinal dif- and plasmid colocalization of tetO and ermB genes. J. Food Prot. ferential abundance analysis of microbial marker-gene surveys 74:1639–1648. using smoothing splines. bioRxiv 99457. Vidal, A., L. Aguirre, C. Seminati, M. Tello, N. Redondo, M. Martín, Pedregosa, F., G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, and L. Darwich. 2020. Antimicrobial resistance profiles and char- O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V. Dubourg, acterization of Escherichia coli strains from cases of neonatal diar- J. Vanderplas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, rhea in Spanish Pig Farms. Vet. Sci. 7:48. and E. Duchesnay. 2011. Scikit-learn: machine learning in python. Villagomez-Estrada, S., J. F. Perez, L. Darwich, A. Vidal, J. Mach. Learn. Res. 12:2825–2830. S. van Kuijk, D. Melo-Dur an, and D. Sol a-Oriol. 2020. Effects of Poole, K. 2017. At the nexus of antibiotics and metals: the impact of copper and zinc sources and inclusion levels of copper on weanling Cu and Zn on antibiotic activity and resistance. Trends Microbiol pig performance and intestinal microbiota. J. Anim. Sci. 98:1–15. 25:820–832. Watanabe, S., N. Kobayashi, D. Quin, and S. Hayakawa. 2009. Prada, J. L., J. L. Villanueva, J. Torre-Cisneros, F. Rodriguez, Genetic diversity of the low-level vancomycin resistance in Entero- M. Espinosa, and M. Anguita. 1994. Endocarditis due to coryne- coccus casseliflavus. Microb. Drug Resist. 15:1–9. bacteriaceae in a kidney transplant patient. Nephrol. Dial. Trans- Xia, M. S., C. H. Hu, and Z. R. Xu. 2004. Effects of copper-bearing plant. 9:1185–1187. montmorillonite on growth performance, digestive enzyme activi- Price, M. N., P. S. Dehal, and A. P. Arkin. 2010. FastTree 2 - approxi- ties, and intestinal microflora and morphology of male broilers. mately maximum-likelihood trees for large alignments. PLoS One 5: Poult. Sci. 83:1868–1875. e9490. Yadav, S., and R. Jha. 2019. Strategies to modulate the intestinal Qiao, Y., J. Sun, Z. Xie, and Y. and Shi, and. 2014G. Le. 2014. Pro- microbiota and their effects on nutrient utilization, performance, pensity to high-fat diet-induced obesity in mice is associated with and health of poultry. J. Anim. Sci. Biotechnol. 10:1–11. the indigenous opportunistic bacteria on the interior of Peyer’s Yazdankhah, S., K. Rudi, and A. Bernhoft. 2014. Zinc and copper in patches. J. Clin. Biochem. Nutr. 55:120–128. animal feed − development of resistance and co-resistance to anti- Rajilic-Stojanovic, M., and W. M. de Vos. 2014. The first 1000 cul- microbial agents in bacteria of animal origin. Microb. Ecol. Heal. tured species of the human gastrointestinal microbiota. FEMS Dis. 25:25862. Microbiol. Rev. 38:996–1047. Zhang, F., W. Zheng, R. Guo, and W. Yao. 2017. Effect of dietary Reece, William O., H. Howard, Jesse Erickson, P. Goff, and copper level on the gut microbiota and its correlation with serum E. E. Uemura. 2015. Dukes’ Physiology of Domestic Animals. inflammatory cytokines in Sprague-Dawley rats. J. Microbiol. (13th ed). Wiley Blackwell, Hoboken, NJ. 55:694–702. Ren, J., W. Wang, S. Sun, L. Zhang, L. Wang, and J. Chang. 2011. Zhi, X. Y., Z. Jiang, L. L. Yang, and Y. Huang. 2017. The underlying Crystallography facet-dependent antibacterial activity: the case of mechanisms of genetic innovation and speciation in the family Cor- Cu 2O. Ind. Eng. Chem. Res. 50:10366–10369. ynebacteriaceae: a phylogenomics approach. Mol. Phylogenet. Samanta, B., A. Biswas, and P. R. Ghosh. 2011. Effects of dietary cop- Evol. 107:246–255. per supplementation on production performance and plasma bio- Zoetendal, E. G., J. Raes, B. Van Den Bogert, M. Arumugam, chemical parameters in broiler chickens. Br. Poult. Sci. 52:573–577. C. C. Booijink, F. J. Troost, P. Bork, M. Wels, W. M. De Vos, and Stanley, D., R. J. Hughes, M. S. Geier, and R. J. Moore. 2016. Bacte- M. Kleerebezem. 2012. The human small intestinal microbiota is ria within the gastrointestinal tract microbiota correlated with driven by rapid uptake and conversion of simple carbohydrates. improved growth and feed conversion: challenges presented for the ISME J 6:1415–1426.