The genetic diversity of cereulide biosynthesis gene cluster indicates a composite transposon Tnces in emetic Bacillus weihenstephanensis

Abstract

Background: Cereulide is a cyclic dodecadepsipeptide ionophore, produced via non-ribosomal peptide synthetases (NRPS), which in rare cases can lead to human death. Early studies had shown that emetic toxin formation belongs to a homogeneous group of Bacillus cereus sensu stricto and the genetic determinants of cereulide (a 24-kb gene cluster of cesHPTABCD) are located on a 270-kb plasmid related to the Bacillus anthracis virulence plasmid pXO1.

Results: The whole genome sequences from seven emetic isolates, including two B. cereus sensu stricto and five Bacillus weihenstephanensis strains, were compared, and their inside and adjacent DNA sequences of the cereulide biosynthesis gene clusters were analyzed. The sequence diversity was observed, which classified the seven emetic isolates into three clades. Different genomic locations of the cereulide biosynthesis gene clusters, plasmid-borne and chromosome-borne, were also found. Potential mobile genetic elements (MGEs) were identified in the flanking sequences of the ces gene cluster in all three types. The most striking observation was the identification of a putative composite transposon, Tnces, consisting of two copies of ISces element (belonging to IS6 family) in opposite orientations flanking the ces gene cluster in emetic B. weihenstephanensis. The mobility of this element was tested by replacing the ces gene cluster by a KmR gene marker and performing mating-out transposition assays in Escherichia coli. The results showed that Tnces::km transposes efficiently (1.04 × 10(-3) T/R) and produces 8-bp direct repeat (DR) at the insertion sites.

Conclusions: Cereulide biosynthesis gene clusters display sequence diversity, different genomic locations and association with MGEs, in which the transposition capacity of a resistant derivative of the composite transposon Tnces in E. coli was demonstrated. Further study is needed to look for appropriate genetic tools to analysis the transposition of Tnces in Bacillus spp. and the dynamics of other MGEs flanking the ces gene clusters.

Figure 1

Phylogenetic analysis based on the sequences of genomes and ces genes of B. cereus group strains. (A) Phylogenetic overview in Gegenees of the genomes. The scale bar represents a 7% difference in average BLASTN score similarity. The heat-map is asymmetric because the variable contents of genomes differ in sizes and a similarity is calculated as a fraction of similar sequences in each genome. (B) Dendrogram based on the seven concatenated ces gene sequences by an NJ phylogenetic tree with a bootstrap of 1,000.

Figure 2

Genomic location of the cereulide gene cluster. (A) Genomic location of the cereulide gene cluster of emetic B. cereus group isolates determined by plasmid profiling (L) and hybridization (R). Lane 1: IS075, lane 2: MC118, lane 3: MC67, lane 4: CER057, lane 5: BtB2-4, lane 6: non cereulide-producing B. cereus isolate CER071, lane 7: AH187. The probe used was cesB internal fragment amplified with EmF and EmR primers from the reference strain AH187. pMC118 and pMC67, displaying a larger size than pCER270, are indicated by a dark triangle. (B) Linear arrangement of the contig containing the ces gene cluster of (L) CER057 with the chromosome of KBAB4 and (R) IS075 with the plasmid pCER270. Aligned segments are represented as dots (20 ~ 65 bp) and lines (>65 bp), with red and blue colors refer to forward and reverse matching substrings, respectively.

Figure 3

Physical map of the sequences flanking the emetic gene clusters. About 5 kb DNA sequences upstream of cesH and downstream of cesD were analyzed for CER057, CER074, BtB2-4, IS075 and F4810/75, respectively, and due to the available sequences are shorter, about 5 kb DNA sequences upstream of cesH and 2.2 kb downstream of cesD were analyzed for MC67 and MC118. The composite transposon Tnces in emetic B. weihenstephanensis MC67 and MC118 is indicated by black triangles. The Tnces consists of ces gene cluster flanked by two copies of IS element at each end in the opposite direction, containing a transposase gene and 16 bp invert repeats (IRL and IRR) at both ends. Sign and color codes are indicated on the right hand side. Physical map is not at scale.

Figure 4

Sketch drawing of the replicative transposition of Tnces::km into recipient chromosome and the strategy of hybridization. The transposase-mediated fusion of pTnkm and the target molecules generate a third copy of ISces. There are two theoretically possible results of transposition, depending on which ISces is duplicated. Three probes 1, 2, and 3, indicated by dotted lines, represent an internal fragment of bla in cloning vector pUC18, ISces, and Km, respectively, were used for the survey of the transposition. The NdeI sites in km^Rsm^R transconjugants were indicated. No matter which ISces was duplicated, hybridization with probe 1 and 3, a 3.5 kb band and a 1.6 kb band is expected, respectively; with probe 2, besides the 1 kb and 3.5 kb expected bands, extra bands with variable sizes in each independent transconjugant are probably detected due to multi-transpositions. Although there is also a (remote) possibility for the duplication of the whole Tnces::km element, the result will be similar except that more bands with probe 2 are expected.

Figure 5

Southern blot hybridization analysis of the transconjugants of Tn ces ::km transposition in E. coli HB101. Two independent hybridizations were performed. A: lane 1–4, independent Km^RSm^R transconjugants, lane 5, HB101, lane 6, JM109 (pTnkm); and B: lane 1–5, independent Km^RSm^R transconjugants, lane 6, HB101, lane 7, JM109 (pTnkm). Three probes of Km (a), ISces (b) and blapuc18 (c), respectively were used for hybridization as illustrated in Figure 4.