The novel isolated strain Bt X022 (CCTCC No. M2014158) was previously described by Liu et al. (2015). Bt X022 cells was cultured overnight at 30°C with agitation at 200 rpm in liquid LB medium, comprised of polypeptone 10 g; yeast extract 5 g and NaCl 10 g/L (pH 7.6). Subsequently, 400 μL of the overnight culture was transfered into a 300 mL shaker-flask containing 20 mL fresh LB medium and grown for approximately another 3 h at 30°C until the optical density at 600 nm (OD₆₀₀) reached 1.0.

Total DNA of Bt X022 was extracted with Bacterial DNA Kit (Omega Bio-Tek Inc., Lilburn, GA, USA) according to the manufacturer’s instructions. Briefly, 3 mL bacterial culture at an OD₆₀₀ of 1.0 was harvested and the cell pellets were resuspended in 180 μL TE buffer containing 5 mg/mL lysozyme and incubated at 30°C for 10 min. Thereafter, the pellet digested cells were collected and subjected to effect complete lysis through incubation at 55°C in a shaking water bath with 200 μL Buffer BTL containing 25 μL Proteinase K solution. Five microliter RNase A was then added and incubated at room temperature for 30 min, followed by treatment with 220 μL Buffer BDL and incubation at 65°C for 10 min. Two hundred and twenty microliter absolute ethanol was then added and mixed thoroughly, and the entire sample was transfered to a DNA binding column and centrifuged at 10,000 × g for 1 min to bind DNA. After washed twice with washing buffer, the total DNA was finally eluted from the column and collected into a new 2 mL tube. Subsequently, the extracted total DNA sample was quantitated using a Nanodrop 2000 (Thermo scientific, Waltham, MA, USA) and subjected to electrophoresis examination in 0.7% agarose gel for 30 min at 120 V and another 10 h at 20 V. Finally, The samples packed on dry ice were sent to HuaDa Genomic Co. Ltd (Shenzhen, Guangdong, China) for genome sequencing.

The whole genomic sequencing of Bt X022 was completed with Illumina Hiseq 2000 by HuaDa Genomic Co. Ltd (Shenzhen, Guangdong, China). Briefly, total DNA sample of X022 was firstly randomly broken into smaller pieces of no longer than 800 bp by sonication with Bioruptor instruments. Next, the generated fragmented DNA are blunted using a mixture of T4 DNA Polymerase, Klenow DNA Polymerase and T4 PNK, followed by further ligation of specific adapter sequences to the ends. The resulted fragments were purified, amplified by PCR, and were finally used for library construction and sequencing. For genome assembly, the closest species Bt serovar kurstaki str. HD73 was used as the reference genome, and the resulting 244 scaffolds were firstly compared with the genome sequence of HD73 on line using the blast tool¹ respectively to determine the relative position of each scaffold in the chromosome. Thereafter, these scaffolds were assembled in order using Gene Construction Kit 3.0, and the gaps between contigs were filled by PCR amplification.

Automatic genome annotation was performed using the RAST server². Genome structural analysis contains tandem repeat analysis, non-coding RNA analysis (including rRNA, tRNA, and sRNA), and gene prediction. Genome functional analysis comprises conventional gene annotation, gene family analysis and phylogeny analysis. And the predicted ORFs were further functionly annotated by employing GO, KEGG, Swiss-Prot, NR and COG.

The extraction of total bacterial proteins from X022 cells, trypsin digestion and 2D-LC-MS/MS analysis, database search and functional classification processes were previously described by Liu et al. (2015).

The presence of putative genes for proteins Cry1Ca and Cry1Da were confirmed by PCR. The primers used for cry1Ca and cry1Da amplification were designed based on the N-terminal conservative region of the published sequences in Bt strains³. Meanwhile, three genes encoding for insecticidal proteins (Cry1Ia, Cry2Ab and Vip3A) were also detected. All primers used for detection of these five genes were listed in Table 1. The PCR amplification was performed in 20 μL volume containing 2 μL 10 × buffer, 200 ng genomic DNA, 1.6 μL dNTP mixture (2.5 mM), 50 pmol primers and 1.25 U Ex Taq DNA polymerase. The obtained PCR products were subjected to an electrophoresis examination in 1.0% agarose gel for 30 min at 120 V, and finally sent to Thermo Fisher Scientific (China) Co. Ltd (Shanghai, China) for sequence confirmation.

Total RNA was extracted from the bacterial cells harvested at the time point of 24 h cultivation using TRIzol Reagent (Invitrogen Biotechnology Co., Ltd., Shanghai, China). Following digestion of the residual genomic DNA by DNase I (Thermo Fisher Scientific Co. Ltd., Shanghai, China), 1 μg of total RNA was reverse transcribed to cDNA utilizing RevertAid^TM First Strand cDNA Synthesis Kit (Thermo Fisher Scientific Co. Ltd., Shanghai, China) according to the manufacturer’s instruction. The two-step real-time RT-PCR analysis were conducted using Power SYBR^® Green PCR Master Mix (Thermo Fisher Scientific Co. Ltd., Shanghai, China) in a CFX Connect Real-Time PCR Detection System (Bio-Rad Laboratories, Inc., Hercules, CA, USA), as previously described (Liu et al., 2015). qRT-PCR primers used in this study were summarized in Table 2. 16S rRNA was used as a normalization control for gene expression.

Whole genome sequencing of Bt X022 was completed with the Illumina Hiseq 2000 sequencing platform, which resulted in a total of 706 Mb clean data containing 8,151,360 reads. And all the paired reads (~96.23% of the total) were further assembled into 244 scaffolds (470 contigs in total) using Short Oligonucleotide Alignment Program (version 2.04). The above scaffolds were assembled in order based on the reference genome of closest species Bt serovar kurstaki str. HD73 in NCBI. The gaps between scaffolds were filled by PCR amplification. Assembly results revealed that Bt X022 contained a circular chromosome DNA, which consisted of 170 contigs with a total size of 5,460,989 bp and an average G + C content of 32.67%. The longest contig contained 580,476 bp and the length of N50 contig was 32,069 bp. The draft genome for strain X022 also revealed that eight plasmids with high homology to pDNA-involved, pCT72, pHT8-2, pBTHD789-4, pK1S1, pHT73, pCT14 and pCT281 (failed assembly) were possibly harbored (Table 3).

Annotation results of the chromosome DNA sequence based on RAST server showed that the total 170 contigs of the X022 strain contained 5646 coding sequences (CDS) and 148 ncRNAs. The length of total CDS was 4,538,678 bp, which corresponded to 83% of the whole chromosome DNA sequence. The subsystem statistics of the genome based on annotation by RAST server were presented in Figure 1: Genes associated with amino acids and derivatives (558 ORFs), carbohydrates (416 ORFs), and protein metabolism (276 ORFs) were ranked the first three abundant categories among the whole subsystem (Figure 1). Genome annotation results showed that Bt X022 elaborated copper homeostatic systems which comprises Copper-translocating P-type ATPase, Multicopper oxidase, Copper resistance protein D and Cytoplasmic copper homeostasis protein CutC. These functional proteins attempted to maintain the normal physiological activities of bacterial cells under certain levels of extracellular copper. To analyze the gene expression differences between the two conditions (fermentation medium and Cu²⁺-containing condition), five genes coding for copper homeostasis(P-type ATPase1, P-type ATPase2, Multicopper oxidase, Copper resistance protein D and CutC) were subjected to real-time RT-PCR analysis. We found that both mRNA levels of P-type ATPase1 and Multicopper oxidase were upregulated, and the other three genes were downregulated in the Cu²⁺-containing condition (Figure 2).

Encoding genes responsible for PHB biosynthesis and metabolism, which were associated with the increased ICP production caused by Cu²⁺, were also found located in the chromosome. The biosynthetic gene cluster for PHB in Bt X022 consisted of six ORFs, namely, phaJ, phaP, phaQ (a repressor of phaP), phaR (a subunit of PHB synthase), pha B (acetoacetyl-CoA reductase), and pha C (PHB synthase). The genetic organization of this biosynthetic gene cluster for PHB in Bt X022 was slightly different from that in B. megaterium and Bt strain BMB171 (Chen et al., 2012). However, in Bt strain BMB171, this biosynthetic gene cluster was composed of phaP, phaQ, phaR, pha B, pha C and an oxidoreductase (Figure 3).

Preliminary analysis also presented that the well assembled seven plasmids carried 242 predicted CDSs, with a total length of 152,490 bp for CDS regions (Table 1). Hence, the genomic features of Bt X022 and its comparative analysis with that of Bt 4.0718 were listed in Table 4. The chromosomal size of Bt 4.0718 was about 5.6 Mb, which contained 6797 predicted CDSs. These results indicated that Bt 4.0718 exhibited a more complicated genetic background compared to Bt X022.

The total protein expressions profile in Bt X022 by proteomic analysis based on 2D-LC-MS/MS approach were described previously by Liu et al. (2015). A total of 813 proteins in X022 were identified in the spore-release period (44 h): 651 and 566 proteins were identified in fermentation medium (CK) and Cu²⁺-containing condition, respectively. Among these proteins, 404 proteins were commonly detected in both conditions (Figure 4A). According to the functional classification system reported by Huang et al. (2012), these identified proteins of Bt X022 were classified into seven functional groups by using the Gene Ontology (GO), KEGG and BioCyc metabolic pathway database of Uniprot KB: Small molecular metabolism (39.32%), Macromolecular metabolism (24.42%), Unknown function (18.13%), Cell processes (7.83%), Regulation (2.76%), Environmental information processing (6.76%) and Extra-chromosomal proteins (0.77%). We preliminarily estimated that the total 813 proteins detected in both conditions, which constitute the expressed genome in spore-release period for Bt X022, accounted for 13.8% of the total predicted proteome encoded by the whole genome (Figure 4B).

Since Bt X022 was found to exert a strong insecticidal activity against several target insects, it was tempted to mine the functional gene resources that were related with its entomopathogenic characteristics based on the above genomic and proteomic data. The assembled genome for strain X022 contained one complete ICP gene which showed a 100% pairwise amino acid identity with pesticidal crystal protein cry 1Ac from B. cereus (Table 5). Moreover, we found that this strain also included two other complete crystal protein gene (cry1Ia and cry2Ab) and a vegetative insecticidal protein gene (vip3A) in two scaffolds of the failed assembled plasmid pCT281 by screening the comprising scaffolds with BLASTx analysis. The putative cry1Ia gene encoded a shorter insecticidal protein of 719 amino acids (aa) with 100% pairwise identity to the 746 aa protein of other Bt. Insecticidal genes cry2Ab and vip3A exhibited 100 and 99% pairwise amino acids identity to those of B. cereus and Bt, respectively. The annotated genome for strain X022 also included up to 20 CDSs which were all located in the chromosome and related with insecticidal activity, including non-hemolytic enterotoxin A, immune inhibitor A, chitinase, hemolysin BL, hemolysin III, cytotoxin K and SpoIISA-like protein. However, the 2D-LC-MS/MS results showed that three ICPs: Cry1Ca, Cry1Ac and Cry1Da were detected and revealed to be mainly expressed during the spore-release period of strain X022. To confirm the presence of putative genes for crystal proteins (Cry1Ca and Cry1Da), an PCR detection of these five genes (cry1Ca, cry1Da, cry1Ia, cry2Ab and vip3A)was conducted. All these PCR amplifications generated products with the expected size (Figure 5). And sequencing of cry1Ca and cry1Da PCR products further confirmed their presence in Bt X022. We also try to confirm the expression of cry1Ia and cry2Ab by qRT-PCR. However, transcription of cry1Ia and cry2Ab in these two culture conditions were not detected (data not shown).

Liu et al. (2015) have reported that Cu²⁺ addition caused significant improvement of crystal protein production were probably associated with carbon flow changes, which was indicated by the downregulation of glycan biosynthesis and metabolism-related proteins. Genome annotation combined with KEGG database search results revealed that up to 20 CDSs that coded for proteins participated in PHB synthesis and metabolism located in the chromosome (Table 6). However, only PhaJ (3-hydroxybutyryl-CoA dehydratase), Hbd (3-hydroxybutyryl-CoA dehydrogenase), EchA (Enoyl-CoA hydratase), AtoB (Acetyl-CoA acetyltransferase), HADH (3-hydroxyacyl-CoA dehydrogenase), PhaP and PhaR were detected by LC-MS/MS in the spore-release period of this strain.

Accordingly, a network for PHB biosynthesis and metabolism regulation of X022 was finally drawn based on the related enzymes annotated from genome and KEGG database (Figure 6). Figure 6 indicated that in the logarithmic phase, the excess carbon supply fluxed to the PHB biosynthetic pathway through an important intermediate metabolite acetyl-CoA. However, when X022 cells grew into the stationary phase and the carbon and energy source around the extracellular environment were gradually exhausted, the reserved carbon resources could be reutilized through degradation of PHB and convertion into acetyl-CoA, finally entering into the tricarboxylic acid (TCA) cycle to release and provide energy for the important physiological processes, such as sporulation and ICP synthesis. Importantly, Liu et al. (2015) found that two PHB metabolism-related proteins, namely, PhaR and 3-hydroxybutyrate dehydrogenase (BDH), were upregulated in response to Cu²⁺ treatement, accompanied by an increased ICP production. It was therefore speculated that Cu²⁺ might have changed the direction of carbon flow in strain X022, finally resulting in more energy reserved and further reuse for ICP synthesis.