Supporting materials Methods

Abstract

Zhu et al. 10.1073/pnas.1222743110 SI Materials and Methods Antibiotic Analyses. Extraction methods and analysis of tetracy-clines are given elsewhere (1). Sulfonamides and quinolones were extracted according to the method of Zhang et al. (2). Briefly, Na 2 EDTA (0.1 g) and 10 mL of a mixed solution containing phosphate buffer (pH = 3.0) and acetonitrile (1:1, vol/vol) were used as extraction solution. After sonication for 30 min, samples were then centrifuged at 7,000 × g for 10 min. The extraction process was performed three times in each sample. The obtained supernatants were combined and then diluted to 500 mL with deionized water, filtered through 0.45-μm filters, and acidified to pH 3.0 before SPE extraction. The samples were extracted using Oasis HLB (500 mg, 6 mL) extraction cartridges. Final extracts were transferred to 2-mL amber vials for liquid chromatography-MS/MS (ABI 3200 Q TRAP) analysis. Metal Analyses. Air-dried and milled samples (0.100 g) were weighed into 50-mL polypropylene digest tubes, and concentrated HNO 3 (2 mL), HCl (3 mL), and HClO 4 (1 mL) were added sequentially. Tubes were then capped and allowed to stand overnight. Tubes were randomized and heated in a microwave-assisted digestion system (CEM Microwave Technology Ltd.). The temperature was raised to 120 °C within 15 min with a holding time of 20 min. The temperature was then raised to 170 °C within 15 min with a holding time of 30 min. After cooling to room temperature, the solution was transferred and made to 50-mL volumetric flasks. The solutions were stored and before analysis were passed through a 0.45-μm × 13-mm nylon filter (Membrana). For quality control, reference materials (GBW-07401 and GBW-07405; IGGE IRMA) were included in all analysis, and average recovery rate was 92%. Concentrations of metals were measured by inductively coupled plasma mass spectrometry (7500cx; Agilent). Validation of the Primer Sets. In a previous study (3), a randomly selected subset of the assays was selected and validated. In this study, three validation methods were used to further confirm the specificity and accuracy of the designed primer sets. First, all primer sets were subjected to homology search by BLASTn, version 2.2.24 (www.ncbi.nlm.nih.gov), as an in silico PCR. Pa-rameters allowed for two mismatches to the forward and reverse primers each, but both were required to target a single organism within a distance of 1,000 base pairs. Second, an assumption of the ΔΔC T method is that the amplification efficiency of the target and reference gene are the same (4). To test this assumption, we performed the same high throughput qPCR of a dilution series of JM2, the farm sample with the most ARGs detected (Fig. 1). The ΔC T values were plotted and the slope determined. Primer sets that resulted in a slope >1.5 were considered too inefficient to allow in the ΔΔC T analysis. Third, type strains with sequenced genomes were used as templates on the qPCR arrays. Type strains used in three validation samples included either a single strain or a mixture of strains as follows: (i) Acinetobacter bau-mannii strain AYE, (ii) Pseudomonas aeruginosa strain PAO1, and (iii) equal genomic copies of Vibrio cholerae, Bacillus cereus, and Haemophilus influenzae strain RdKW20. First, their ge-nomes were analyzed by BLAST to determine which primer sets were expected to target the strain DNA. Then, qPCR was per-formed and results were compared with the BLAST results. Assays that were previously considered unspecific or not efficient were not considered in this analysis.