Samples (n=2657) including 1155 of fresh feces, 1236 from rectal swab, 110 of feed, 108 of water and 48 of carcass swab were collected between October 2012 and March 2017 from 18 cattle farms and 1 cattle slaughterhouse in China (Tacheng, Bole, Yili, Wujiaqu, Changji, Wulumuqi and Aksu in Xinjiang). Details about these samples are shown in Table 3.
Samples (feces/feed/water) each either 1 g or 1 ml were aseptically added to 9 ml of a trypticase soya broth (TSB) containing 20 mg/l of novobiocin and were incubated for 6-8 h at 37 °C. One rectal swab was transferred into a separate tube containing 2 ml of a nutrient broth and cultured at 37 °C for 24 h . One carcass swab was put into a stomacher bag and added with 500 ml of a modified trypticase soya broth containing 8 mg/l novobiocin. Each sponge was mixed in the stomacher bag for 2 min and incubated for 20 h at 37 °C . This was streaked out onto sorbitol MacConkey agar supplemented with 1 mg/l potassium tellurite and incubated for one day at 37 °C. One or more pale colonies were each picked as presumptive E. coli O157 per sample. The prevalence of E. coli O157:H7 was assessed via PCR (rfbE and fliC genes ) (Table 4). The positive isolates were each inoculated into separate TSB and incubated for one day at 37 °C, from which glycerol stock was made and then stored at -80 °C for further analyses.
Virulence analysis of isolates
Each E. coli O157:H7 isolate was characterized for the presence of virulence genes stx1 , stx2 , eae, hly  and tccP by PCR as previously described (Table 3).
Antimicrobial susceptibility tests
The isolates were tested for the susceptibility to antibiotics using the Kirby-Bauer disc diffusion technique. The antibiotic discs obtained from OXOID, UK, including ampicillin (AMP), piperacillin (PIP), cefotaxime (CTX), ceftazidime (CAZ), cefepime (FEP), aztreonam (ATM), ampicillin-sulbactam (SAM), piperacillin- tazobactam (TZP), amoxicillin-clavulanic acid (AMC), gentamicin (GEN), amikacin (AMI), streptomycin (STR), trimethoprim-sulfamethylisoxazole (SXT), chloramphenicol (CHL), levofloxacin (LEV), ciprofloxacin (CIP), tetracycline (TET), and polymyxin B (PB) . E. coli ATCC25922 was used as a quality control strain in the susceptibility tests. The ESBLs-producing isolates were determined by double-disk synergy tests according to CLSI .
Detection of antibiotic resistance genes
The following resistance determinants were investigated by PCR: blaCTX-M (the CTX-M-type genes were detected using universal primers CTX-M-U , and the entire CTX-M-type genes were amplified by using the primers CTX-M-1G , CTX-M-2G  or CTX-M-9G ), blaTEM , and blaSHV  which encode β-lactamases; chloramphenicol (cmlA1 ) efflux pumps; sulfonamide resistance gene (sul1 ); and the tetA , tetE , and tetG  tetracycline efflux pumps. Primer for the different genes are listed in (Table 4). Purified PCR products were sequenced. The DNA sequences and deduced amino acid sequences were compared with sequences reported in GenBank to confirm the subtypes of the β-lactamase genes.
Conjugation experiments and plasmid analysis
Conjugation was conducted with blaCTX-M-producing isolates by the filter method, using a sodium azide-resistant E. coli J53 as the recipient. Transconjugants were selected on Mac Conkey agar containing cefotaxime or ceftazidime (4 μg/ml) and sodium azide (200 μg/ml). ESBLs and antibiotic susceptibility were also tested in selected transconjugants, and the presence of bla genes was determined using PCR as described above. The resistance plasmids carried by transconjugants were typed by using PCR-based replicon typing .
All available isolates were referred for PFGE and analyzed according to some existing criteria . XbaI-digested chromosomal DNA of isolates was characterized by PFGE using the CHEF-MAPPER System (Bio-Rad Laboratories, Hercules, CA, USA). A Salmonella serotype Braenderup H9812 (ATCC BAA-664) was chosen as the molecular weight marker. Electrophoresis was run at 6.0 V/cm for 18.5 h with an angle of 120° at 14°C.