Bacterial strains, plasmids, and growth conditions
The bacterial strains and plasmids used in this study are listed in Table S1. B. subtilis WB600 (BS), B. amyloliquefaciens (BA) and B. licheniformis WX-02 (BL) were stored in our lab. E. coli TOP10 (TIANGEN Biotech, Beijing, China) was used as the cloning host. B. vietamensis 151-6 and B. marisflavi 151-25 were isolated from Cd-contaminated soil in Hunan Province (27°46′N, 112°52′E), China. These strains were all grown in Luria–Bertani (LB) medium. CdCl2·H2O was used as the source of Cd. Ampicillin (100 µg mL-1), kanamycin (10 µg mL-1), or tetracycline (5 µg mL-1) was added as necessary.
Isolation of Cd-resistant bacteria
Cd-contaminated soil was collected from a former industrial site in Hunan Province (27°46′N, 112°52′E) and analyzed for Cd content through acid digestion followed by the use of a 7,700 × inductively coupled plasma mass spectrometer (Agilent Technologies, Tokyo, Japan). To isolate Cd-resistant and bio-safe bacteria, aerobic Bacillus sp. were isolated from the soil by plating on LB agar plates containing progressively higher concentrations of cadmium chloride (0, 0.5, 1 and 2 mM). Then, in order to isolate aerobic Bacillus sp., the bacterial enrichment cultures were heat-shocked at 80℃ for 20 min. Bacterial genomic DNA was isolated using the TIANamp Bacteria DNA kit (TIANGEN Biotech). The 16S rRNA gene was amplified from the extracted DNA using the universal primers 16S rRNA-F/16S rRNA-R (Table S2) and the amplification products were cloned in the pGM-T (TIANGEN Biotech) vector using competent E. coli TOP10 cells (TIANGEN Biotech). Sequencing was carried out using T7 and SP6 primers and compared to the GenBank database using the NCBI BLAST program.
Evaluation of cadmium resistance and growth curve
To evaluate growth in a liquid medium of isolated bacteria, the minimum inhibitory concentration (MIC) of Cd2+ (MIC-Cd) was determined. LB medium (800 μL) with different concentrations of Cd2+ was dispensed into 96-well (12 × 8) microtiter plates (96 × 2-mL wells) with a multi-channel micropipette (rows A to H: 0 mM, 0.02 mM, 0.04 mM, 0.06 mM, 0.08 mM, 0.1 mM, 0.2 mM,0.3 mM, 0.4 mM, 0.5 mM, 0.6 mM, 0.7 mM, 0.8 mM, 0.9 mM, 1.0 mM and 1.1 mM). Single colonies of the test strains were inoculated into 3 mL of LB medium and cultured overnight. The test culture (15 μL) was then inoculated into each well of the prepared 96-well plate. After 24 h at 37℃ and 750 rpm in an incubator (Heidolph, Viertrieb, Germany), 200 μL of the cell suspension was transferred to a 96-well plate and the turbidity at OD600 was measured.
To determine the Cd2+ tolerance of isolated bacteria, growth curves at different concentrations (0 mM, 0.1 mM, 0.3 mM and 0.5 mM) of Cd2+ were analyzed. For the growth assay, single colonies of the test strains were cultured overnight and then diluted 1:100 into 100-well plates containing 200 μL of LB and various concentrations of Cd2+ in quintuplicate. The growth curve was measured at 1-h intervals using a Bioscreen C automatic growth curve analyzer (Bioscreen, Helsinki, Finland).
Genome sequencing and analysis
Bacterial genomic DNA was extracted using the sodium dodecyl sulfate (SDS) method . A total of 5 μg DNA was used to generate each library, and this DNA was sheared using Covaris g-Tubes to generate sheared fragments >10 kb in length. The sheared DNA fragments were then prepared using the SMRT bell template preparation kit (Pacific Biosciences, Menlo Park, CA, USA) according to the manufacturer’s instructions. Whole-genome sequencing was performed on the Pacbio RSII platform. All high-quality paired reads were assembled using SOAPdenovo (http://soap.genomics.org.cn/soapdenovo.html) onto a number of scaffolds . Then, the filtered reads were transferred for the next step of gap closing. Transfer RNA (tRNA) genes were predicted with tRNAscan-SE . Ribosomal RNA (rRNA) genes were analyzed using rRNAmmer . Coding genes were identified with the GeneMarkS program . The predicted coding genes were annotated based on the non-redundant protein database (NR) of the National Center for Biotechnology Information  and Gene Ontology (GO) .
RNA sequencing and transcriptome analysis
An overnight culture was diluted 1:100 in LB medium in the presence (0.1 mM) and absence of Cd2+, and these cultures were grown at 37℃ and 200 rpm to the exponential phase. Total RNA from both groups (each group has three replicates) were extracted using a TRIzol kit (TIANGEN Biotech) according to the manufacturer’s instructions. A total of 1 μg RNA per sample was used as input material for RNA sample preparation. Sequence libraries were generated using the Illumina® TruSeq® Stranded Total RNA Sample Preparation kit (NEB, Ipswich, MA, USA). TruSeq Stranded Total RNA was prepared with the Ribo-Zero™ Bacteria Kit (Epicenter, Madison, WI, USA). The clustering of index-coded samples was performed on a cBot Cluster Generation System using the TruSeq HiSeq 4000 PE 150 Cluster Kit (Illumina, San Diego, CA, USA) according to the manufacturer’s instructions. After cluster generation, the prepared libraries were sequenced on the Illumina HiSeq 4000 platform by Allwegene Technology (Beijing, China) and 150-bp paired-end reads were generated. Bowtie2 was used to map mRNA reads to the genome, and HTSeq v 0.5.4 p3 was used to count the number of reads mapped to each gene. Then, the fragments per kilobase of exon model per million mapped reads (FPKM) of each gene was calculated based on the length of, and read count mapped to, the gene .
Elimination of plasmids
Plasmid elimination was performed using SDS, as previously described , with some modifications. Specifically, B. vietamensis 151-6 and B. marisflavi 151-25 were treated with 0.0005% and 0.002% SDS, respectively. Single colonies were selected on LB agar plates to screen for strains that had eliminated the p6 plasmid or the p25 plasmid. Primers 4108CDS-F/4108CDS-R and 4109CDS-F/4109CDS-R (Table S2) were used to identify gene located on the chromosome of B. vietamensis 151-6. The primer pairs 4963CDS-F/4963-CDS-R, 4967CDS-F/4967CDS-R, 4982CDS-F/4982CDS-R, 4983CDS-F/4983CDS-R, 5014CDS-F/5014CDS-R and 5018CDS-F/5018CDS-R (Table S2) were employed to identify the genes in the plasmid p6 (Fig. S4) For B. marisflavi 151-25, the primers 4163CDS-F and 4163CDS-R (Table S2) were used to identify genes located on the chromosome, while the primer pairs 4779CDS-F/4779-CDS-R, 4780CDS-F/4780CDS-R and 4803CDS-F/4803CDS-R (Table S2) were used to detect the p25 plasmid (Fig. S5). Meanwhile, the primers for quantitative reverse transcription-polymerase chain reaction (qRT-PCR) (Table S2) located on plasmid p25 were also used to detect the plasmid (Fig. S5).
RT-PCR and qRT-PCR
Total RNA of B. vietamensis 151-6 and B. marisflavi 151-25 were isolated as described above under RNA sequencing. cDNA samples were prepared using TransScript One-Step gDNA Removal and cDNA Synthesis SuperMix (TransGen Biotech, Beijing, China) with random primers according to the manufacturer’s instructions. qRT-PCR was performed using the ChamQ Universal SYBR qPCR Master Mix (Vazyme Biotech, Beijing, China) according to the manufacturer’s instructions. 16S rRNA, amplified with the primers 16S-F/16S-R, was used as an internal control. All reactions were performed in biological triplicate, and the normalized fold changes of the relative expression ratios were quantified using the 2-△△CT method .
Construction and screening of an B. vietamensis 151-6 genomic fosmid library
High-molecular-weight genomic DNA from B. vietamensis 151-6 was extracted using the bacterial genomic DNA extraction kit (BioTeke Biotech, Beijing, China) according to the manufacture’s instructions. And a fosmid library was constructed with the CopyControl Fosmid Library Production Kit (vector pCC1FOS; Epicentre, Madison, WI, USA) according to the manufacture’s instructions. A total of 1204 clones were used to construct the library in 13 96-well plates. To screen for clones with Cd2+ resistance, overnight cultures of the 1204 clones were inoculated into LB liquid medium with 1.0 mM and 1.2 mM Cd2+, respectively. Meanwhile, these clones were spotted onto LB agar plates with 1.2 mM and 1.5 mM Cd2+. Fosmid DNA was isolated from the positive fosmid clones by Axyprep Plasmid Miniprep Kit (Axygen, USA) and sequenced by pCC1 sequencing primers pCC1-F/ pCC1-R (Table S2). The results were aligned with the genome sequence of B. vietamensis 151-6 to confirm the sequence of the inserted fragments.
Construction of plasmids for gene fragments heterologous expression
To verify that Cd2+ resistance is conferred by genes, recombinant pUC19 and pUBC19 plasmids were constructed. The 3444-bp Sac I-Xba I DNA fragment amplified from B. vietamensis 151-6 genome with the primers 4111-4112-4113-F/4111-4112-4113-R (Table S2), containing copper-translocating P-type ATPase gene (orf4111), cadmium efflux system accessory gene cadC (orf4112), cadmium resistance gene cadD (orf4113) and 217-bp upstream of orf4113, was cloned into E. coli pUC19 vector and E. coli-B. subtilis shuttle vector pUBC19, respectively. The gene fragments 4087-4088, 4093-4094-4095, 4102-4103 and 4108-4109 amplified from B. vietamensis 151-6 genome with corresponding primers (Table S2) were also cloned into pUC19 and pUBC19. For B. marisflavi 151-25, the gene fragments 4774-4775, 4776-4777, 4779-4780, 4781-4782 and 4802-4803 amplified from plasmid p25 of B. marisflavi 151-25 with the corresponding primers (Table S2) were cloned into pUC19 and pUBC19. The ligation mixture was transformed into E. coli TOP10, and the correct plasmids were identified through colony PCR with corresponding sequencing primers (Table S2). After confirmation via sequencing, recombinant pUBC19 plasmids were extracted from E. coli TOP10 and transformed into B. subtilis, as described previously . Transformants were harvested by screening the clones on LB agar containing 10 µg/mL kanamycin, and the correct plasmids were verified through colony PCR with the pUBC19 sequencing primers CX-F/CX-R (Table S2).
MIC-Cd and growth curve determination
MIC-Cd and the growth curve of plasmid-eliminated strains and wild strains of B. vietamensis 151-6 and B. marisflavi 151-25 were determined as described above in 2.3. MIC-Cd values for the recombinant strains of E. coli TOP10 and B. subtilis were also evaluated using 96-well microtiter plates, as described in 2.3.