*2.2. Sludge Sample and Medium*

The sludge was a semisolid slurry (total solids 0.87 g/L, pH 6.7) that was produced as sewage sludge from the Dihus wastewater treatment plant in Taipei, Taiwan. Fresh sludge was used for antibiotic adaptation. The temperature of the sampling site was 30 ◦C. The medium used in aerobic experiments contained the following chemicals (mg/L): K2HPO4, 65.3; KH2PO4, 25.5; Na2HPO4·12 H2O, 133.8; NH4Cl, 5.1; CaCl2, 82.5; MgSO4·7H2O, 67.5; and FeCl3·6H2O, 0.75. The medium used in anaerobic experiments contained the following chemicals (mg/L): NH4Cl, 2.7; MgCl2•6H2O, 0.1; CaCl2•2H2O, 0.1; FeCl2•4H2O, 0.02; K2HPO4, 0.27; KH2PO4, 0.35 and resazurin, 0.001. The pH of the medium was adjusted to 7.0 using potassium hydroxide or nitric acid. Resazurin is an indicator, which exhibits red color under aerobic conditions and is colorless under anaerobic conditions. Titanium citrate of 0.9 mM was used as a reducing reagent. All anaerobic operations were performed in an anaerobic glove box.

## *2.3. Sludge Adaptation*

Aerobic adaptation was performed by adding 1 mg/L CTC, SMX and AMO, simultaneously, to 1000 mL serum bottles containing 450 mL of aerobic medium, 50 mL sludge, and incubated on a rotary shaker (120 rpm) at 30 ◦C without light for 6 months. Anaerobic adaptation was performed by adding 1 mg/L CTC, SMX and AMO, simultaneously, to 1000 mL serum bottles containing 450 mL of anaerobic medium, 50 mL sludge and capped with butyl rubber stoppers and crimp seals, wrapped in aluminum foil, and then incubated without shaking at 30 ◦C without light for 6 months. All anaerobic operations were performed in an anaerobic glove box. In this paper, the sludge was referred to as antibiotic-adapted sludge.

#### *2.4. Enrichment, Isolation, and Identification of Antibiotic-Degrading Bacteria*

The enrichment procedure was performed using 5 mL of antibiotic-adapted sludge in a 125-mL serum bottle containing 45 mL of aerobic or anaerobic medium, with CTC, SMX and AMO of final concentration of 0.2 mg/L, simultaneously, and incubated at 30 ◦C. The second to the fourth enrichment transfers were amended with gradually increasing concentrations of CTC, SMX and AMO from 0.2 to 1 mg/L. An elevated concentration is commonly used for enrichment to ensure that the CTC, SMX and AMO degraders are selected. After the fourth enrichment, aerobic or anaerobic medium agar plates containing CTC, SMX and AMO (1 mg/L) were inoculated with 100 μL of the liquid part of the sludge by streaking to isolate pure strains of bacteria. The AnaeroPack system was used for anaerobic cultivation, including AnaeroPack®-Anaero and AnaeroPouch®-Anaero. All anaerobic operations were performed in the anaerobic glove box. To confirm that the bacterial strains were antibiotic degraders, degradation experiments were performed using 5 mL (106 CFU/mL) of bacterial culture and 45 mL of medium with 1 mg/L antibiotics on a shaker (120 rpm) at 30 ◦C in the dark. Samples were taken periodically to analyze residual CTC, SMX and AMO.

The 16S rRNA gene of the isolated bacterial strains were amplified by PCR with the 5 -primer F8 (5 -AGAGTTTGATCCTGGCTCAG-3 ) and the 3 -primer R1510 (5 -GGTTACCTTGTTACGACTT-3 ). The PCR parameters included initial denaturation at 94 ◦C for 10 min, followed by 35 cycles of 45 s at 94 ◦C, 1 min at 60 ◦C, and 1 min at 72 ◦C, with a final extension at 72 ◦C for 10 min. The PCR products were sequenced on an ABI Prism automatic sequencer. The 3 -end sequence was converted into the reverse complementary sequence. The overlapping parts of sequences from the 5 and 3 -ends were identified with the Align Sequences Nucleotide BLAST tool at the National Center for Biotechnology Information (NCBI) website. Finally, the two sequences were assembled into a single contig sequence based on the overlapping sequences. The 16S rRNA sequences of the four isolates have been submitted to Genbank; accession number: A12: MT678104, B: MT678105, SANA: MT678106, SF1: MT678107. The 16S rRNA gene sequences of the four isolated bacterial strains were used to search the NCBI 16S rRNA database with the Basic Local Alignment Search Tool (BLASTn). The top five sequences (with the highest scores) in the Blast results for each 16S rRNA sequence were retrieved and used to construct the phylogenetic tree. Phylogenetic analysis was performed using Clustal X 2.0 with 1000 bootstrapping repetitions [29]. The neighbor-joining algorithm was used to construct the phylogenetic tree.
