Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Elucidation of quantitative structural diversity of remarkable rearrangement regions, shufflons, in IncI2 plasmids

Abstract

A multiple DNA inversion system, the shufflon, exists in incompatibility (Inc) I1 and I2 plasmids. The shufflon generates variants of the PilV protein, a minor component of the thin pilus. The shufflon is one of the most difficult regions for de novo genome assembly because of its structural diversity even in an isolated bacterial clone. We determined complete genome sequences, including those of IncI2 plasmids carrying mcr-1, of three Escherichia coli strains using single-molecule, real-time (SMRT) sequencing and Illumina sequencing. The sequences assembled using only SMRT sequencing contained misassembled regions in the shufflon. A hybrid analysis using SMRT and Illumina sequencing resolved the misassembled region and revealed that the three IncI2 plasmids, excluding the shufflon region, were highly conserved. Moreover, the abundance ratio of whole-shufflon structures could be determined by quantitative structural variation analysis of the SMRT data, suggesting that a remarkable heterogeneity of whole-shufflon structural variations exists in IncI2 plasmids. These findings indicate that remarkable rearrangement regions should be validated using both long-read and short-read sequencing data and that the structural variation of PilV in the shufflon might be closely related to phenotypic heterogeneity of plasmid-mediated transconjugation involved in horizontal gene transfer even in bacterial clonal populations.

Introduction

The shufflon, one of the members of the site-specific recombination system, was discovered in 1986 in incompatibility (Inc) I1 plasmid R641, 2. Shufflons have also been identified in IncI1γ3, IncI24, 5, IncK6, 7 and IncZ plasmids8, 9. Shufflons generate variants of the PilV protein, a minor component of the thin pilus. The shufflon regions of R64 (IncI1), CoIIb-P9 (IncI1) and R721 (IncI2) plasmids consist of four (A, B, C and D), three (A, B and C) and three (A, BD and C) segments, respectively. Each segment, A, B, C and BD, includes two different partial open reading frames (ORFs) on the C-terminal region of PilV (ORF A, ORF A′; ORF B, ORF B′; ORF C, ORF C′; and ORF B′, ORF D′, respectively), whereas segment D only includes ORF D. These segments are rearranged in the conserved repeat region at the end of each segment by Rci, which has the activity of a site-specific tyrosine recombinase and is encoded by the rci gene1, 10. Komano et al. suggested that seven possible variants of pilV encode seven different PilV tip adhesins in IncI1 plasmid R643, 11. PilV recognises specific lipopolysaccharide (LPS) structures on the surface of recipient cells during liquid mating4, 12. In particular, it has been reported that the ligands of the PilVA, PilVB′, PilVC and PilVC′ adhesins are the GlcNAc(β1–3)Glc, GlcNAc(α1–2)Glc, GlcNAc(β1–7)Hep and Glc(α1–2)Glc or Glc(α1–2)Gal structures, respectively, of LPS of Escherichia coli type R1 (E. coli O8), E. coli K-12 or Salmonella enterica Typhimurium LT26, 8, 13,14,15. Thus, the shufflon rearrangement is closely related to plasmid transmission to a broad range of the Enterobacteriaceae.

Colistin has emerged as a treatment option for infectious diseases caused by carbapenemase-producing multidrug-resistant Enterobacteriaceae 10, 11, 16,17,18. In 2015, Liu et al. reported the emergence of a plasmid-mediated colistin resistance mechanism, MCR-1, in the Enterobacteriaceae 11, 19, 20. The mcr-1 gene, which encodes phosphoethanolamine transferase MCR-1, was first identified in the pHNSHP45 plasmid, which is classified as an IncI2 plasmid, from the E. coli strain SHP45. In Japan, Suzuki et al. also identified five IncI2 plasmids carrying mcr-1 in animal isolates12, 18, 20, 21. This gene has been reported in various plasmid types (namely IncF13, 14, 22,23,24,25, P15, 26,27,28, I211, 16,17,18, 29, HI219, 20, 30, repB (pO111)31, HI120 and X418, 20, 21, 32) of the Enterobacteriaceae with diverse origins worldwide, suggesting horizontal gene transfer and plasmid-mediated conjugal transfer of mcr-1 among the Enterobacteriaceae. Therefore, analysis of plasmids carrying mcr-1 has become important for the surveillance and control of drug-resistant bacteria.

Next-generation sequencers (NGSs) are powerful tools to reveal genomic features of organisms and are typically represented by SOLiD/Ion Torrent PGM from Life Sciences, Genome Analyzer/HiSeq 2000/MiSeq from Illumina and GS FLX Titanium/GS Junior from Roche22,23,24,25, 33, 34. As of September 2016, 76,793 records of genome sequences were submitted in the Assembly Database of the National Center for Biotechnology Information. Moreover, NGSs not only provide draft and complete genome sequences but are also used for quantitative analyses, including transcriptomic and metagenomic approaches5, 26,27,28. However, NGS short reads are too difficult to assemble because of frequent recombination regions, which include shufflon regions of bacterial plasmids3, 29. The DNA recombination-based pilus phase and antigenic variation systems have been found in several virulent bacteria (such as Neisseria spp., Borrelia spp., Treponema pallidum and Mycoplasma spp.)30, 35. These rearrangement regions are also difficult to assemble into a sequence structure. The introduction of the single-molecule, real-time (SMRT) sequencing, i.e. PacBio RS II (Pacific Biosciences), has dramatically changed the method for determination of bacterial complete genome sequences, and we have been able to easily obtain a gapless bacterial chromosomal sequence32. SMRT sequencing also enables uncovering long structural variations in human genomes29, 33, 34. However, little has been reported on quantitative structural variation analysis of isolated bacterial clones, even though it is known that several bacteria have DNA recombination-based variation systems.

In this study, we report the complete genome sequences, including those of IncI2 plasmids carrying mcr-1, of three E. coli strains, as well as quantitative structural variation and comparative analysis of whole-shufflon structures in the IncI2 plasmids.

Results

Genome sequencing and error corrections in shufflon regions of IncI2 plasmids

To determine the whole-genome sequences of colistin-resistant E. coli strains (MRY16-002 = 20Ec-P-124, MRY15-117 and MRY15-131), de novo assembly and circularisation were performed using SMRT sequencing data. The detailed de novo assembly results are summarised in Table S1. Sequences of IncI2 plasmids carrying mcr-1 were detected in the PacBio assemblies of all strains. It has already been reported that the shufflon is a multiple DNA inversion system present in IncI2 plasmids5, 36 as described in the Introduction.

Figure 1A shows the schematic illustration for the rearrangement of shufflons and relations between PilV proteins and specific ligand structures. Because the misassembling in a shufflon region might be caused by a heterogeneous population of shufflon structures in isolates, the error-corrected sequences of the shufflons were analysed by the pacBioToCA program with PacBio subreads and Illumina short reads as shown in Fig. 1B. The corrected sequences were composed of several combination patterns as follows: one segment (BD) in pMRY15–131_2; two segments (BD and C) in pMRY15-117_2; and three segments (A, B, D and C) in pMRY16-002_4. Moreover, the orientation and direction of the segments showed diversity of the detected shufflon structures. The corrected sequences carrying the most predominant shufflon structure were used in the subsequent analysis. The mapping validation using Illumina short reads was performed against the PacBio assemblies and corrected sequences, indicating broken orientations of paired-end reads intensively detected in the shufflon regions (Fig. 1C). In particular, unmapped and redundant coverage regions were detected in the PacBio assemblies of pMRY15-117_2 and pMRY16-002_4 (Fig. 1C), suggesting that these PacBio assemblies might have been misassembled in the shufflon regions. On the other hand, the mapping data of pMRY15-131_2 indicated a normal coverage (data not shown). The mapping validation against corrected sequences revealed that segment BD was deleted from the PacBio assemblies of pMRY15-117_2 and pMRY16-002_4 (Fig. 1C). In particular, segment BD in the PacBio assembly of pMRY15-117_2 was replaced by sequences that were nonhomologous to shufflons. Although broken paired-end reads were still detected in the shufflon regions of corrected sequences, the misassembled regions were resolved by error corrections, suggesting that the shufflon regions could not be simply completed by automatic de novo assembling because of the heterogeneous population of shufflon structures in the isolates.

Figure 1
figure 1

Error correction and validation of shufflons in IncI2 plasmids. (A) Schematic illustration of the model of shufflon rearrangement and relationship between PilV proteins and specific ligands. Site-specific recombination between any inverted repeat sequences, mediated by the site-specific recombinase Rci, yields the inversion of segments of a shufflon independently or among other segments. The various pilus PilV proteins recognise specific ligand structures in LPS. (B) Schematic representation of error correction in shufflon sequences. (C) Validations of shufflons using short-read mapping data. The X-axis and Y-axis represent the nucleotide positions of PacBio assemblies or corrected sequences and the coverage depth of the mapped short reads, respectively. Grey and red areas represent the coverage of the correct orientation and broken paired-end reads, respectively. Yellow boxes show identical regions between PacBio assemblies and corrected sequences.

Quantitative structural variation analysis of shufflon regions

The IncI2 plasmids with the corrected shufflon regions revealed that the number of shufflon segments was different among the three plasmids, as described above. It has been previously reported that rearrangements occur in shufflon segments and that possible variants of pilV can encode several different PilV tip adhesins3, 37. Our mapping validation data also showed the remarkable rearrangement of shufflon regions as shown in Fig. 1C. To reveal the abundance ratio of this rearrangement, quantitative structural variation analysis was performed as follows. Structural patterns for all shufflons were calculated according to the formula: number of shufflon structure combinations = 2n × n! (where n = number of segments). This formula is suitable for only segments containing two 3′ ends of pilV, but not for segments containing one 3′ end of pilV (e.g. segment D as described for IncI1 shufflons). These numbers were 2, 8 and 48 for pMRY13-131_2, pMRY15-117_2 and pMRY16-002, respectively (Fig. 2A). All potential combinations of shufflon sequence structures were constructed, followed by actual shufflon structure detection using a BLASTN homology search with SMRT sequencing subreads. The quantitative structural variation analysis revealed that all potential combinations of structural variations were detected in pMRY13-131_2 and pMRY15-117_2. On the other hand, 26 combination structures were detected in the 48 potential combination patterns of the shufflon structures of pMRY16-002_4 (Fig. 2A). The quantitative analysis yielded the following results: (i) in pMRY15-131_2, two structures [i.e. gene loci pilV–ORF B′ (B′)–D′–rci and pilV–D′–B′–rci] were detected, and the rci–B′–D′–pilV pattern accounted for 60.4% of all patterns; (ii) in pMRY15-117_2, the pilV–B′–D′–C–C′–rci, pilV–D′–B′–C–C′–rci, pilV–D′–B′–C′–C–rci, pilV–C–C′–B′–D′–rci, pilV–C′–C–B′–D′–rci, pilV–B′–D′–C′–C–rci, pilV–C′–C–D′–B′–rci and pilV–C–C′–D′–B′–rci patterns accounted for 29.2%, 20.8%, 15.6%, 12.5%, 6.3%, 6.3%, 5.2% and 4.2%, respectively; (iii) in pMRY16-002_4, pilV–A′–A–C′–C–B′–D′–rci, pilV–A–A′–C′–C–B′–D′–rci, pilV–A–A′–C–C′–B′–D′–rci and other patterns accounted for 16.8%, 12.1%, 7.5% and 63.6%, respectively. Interestingly, the abundance ratio of shufflon structure patterns was uneven in two of the IncI2 plasmids, pMRY15-117_2 and pMRY16-002_4 (Fig. 2A). Variants of the PilV protein are generated by the rearrangement between the pilV 3′-end and a shufflon segment. The abundance ratio of pilV gene variants was also calculated on the basis of SMRT sequencing subreads and Illumina short reads using the junction between the pilV 3′-end and a shufflon segment (Fig. 2B). The quantitative analysis using short reads was valid because these reads included both the pilV 3′-end and shufflon segment regions. To validate the data of the quantitative structural variation analysis as described above, the abundance ratio of the pilV gene was compared between sequencing reads obtained using the two different platforms, i.e. SMRT sequencing subreads and Illumina short reads (Fig. 2C). These detected abundance ratios showed a high correlation between the two distinct platforms (R2 = 0.94), suggesting that the SMRT sequencing data is valid for the quantitative analysis of combination patterns not only of the pilV gene but also of whole-shufflon structures. The complete IncI2 plasmid sequences were determined using the predominant shufflon structures as follows: pilV–B′–D′–rci in pMRY15-131_2, pilV–B′–D′–C–C′–rci in pMRY15-117_2 and pilV–A′–A–C′–C–B′–D′–rci in pMRY16-002_4 (¶ in Fig. 2A).

Figure 2
figure 2

Quantitative structural variation analysis of shufflon regions in IncI2 plasmids. (A) Quantitative structural variation analysis of IncI2 plasmids. In pMRY13-131_2 and pMRY15-117_2, all shufflon structures were detected, whereas 26 shufflon structures were detected in pMRY16-002_4. (B) Comparison of the pilV gene ratios between different sequencer platforms. (C) Scatter plot of the pilV gene 3′ region ratios between the PacBio and MiSeq reads. The data indicate that there was a high correlation between the PacBio and MiSeq sequencer reads (R2 = 0.94).

Genomic information of strains MRY16-002, MRY15-117 and MRY15-131

The complete chromosomal and plasmid sequences were determined, and it was found that E. coli MRY16-002, MRY15-117 and MRY15-131 harboured six, two and two plasmids, respectively (Table 1). Two short plasmids, pMRY16-002_5 (6,078 bp) and pMRY16-002_6 (4,073 bp), were only detected using the Illumina data because the average length of PacBio long-read DNA library was approximately 30 kb. Multilocus sequence typing (MLST) analysis revealed that MRY16-002 and the other two strains (MRY15-117 and MRY15-131) belonged to two different sequence types (STs), ST117 and ST457, respectively. In silico serotype analysis showed different patterns in the three strains, and the chromosomal virulence gene profiles were also different between MRY15-117 and MRY15-131 (Table S2). However, the plasmid profiles indicated the same pattern for MRY15-117 and MRY15-131. Plasmid copy number analysis showed that the IncI2 plasmids carrying mcr-1 were present in a single-copy per genome in each strain (Table 1). Comparative analysis indicated that the sequence structures were highly conserved among the three IncI2 plasmids. A single-nucleotide variation (SNV) was only detected in the stbE gene (nucleotide position 53,712 in pMRY16-002_2; Fig. 3A). Insertion/deletion sequences were only detected in the shufflon regions. All plasmids contained segment BD in the shufflon region, but segment A only existed in pMRY16-002_4 (Fig. 3B). The multiple alignments showed that 5′-GTGCCAATCCGGTNNGTGGA-3′ (20 bp) was a highly conserved sequence among the repeat regions in the shufflon segments (Fig. 3C and Table S3). The repeat sequences of each segment A and BD showed complete identity. However, those of segment C included three nucleotide variations, consistent with a previous report29, 35. A more detailed analysis showed that 23 bp at the ends of the shufflon segments were conserved, but variations from one to six nucleotides existed across these segment ends. The minimum spanning network analysis revealed low stability of repeat regions of ORF C′ (Fig. 3D and Table S3). The methylome analysis revealed the following results: (i) N6-Methyladenosine (m6A) methylation sites were found spread all over the genome; (ii) Dam methylation sequence pattern (i.e. 5′-GATC-3′) was detected most frequently; and (iii) unique methylation motifs were detected in each strain (Figure S1 and Table S4). However, the m6A methylation site was not detected in all shufflon inverted repeat regions (Figure S1).

Table 1 General features of three isolates and their complete genome sequences.
Figure 3
figure 3

Comparative analysis of IncI2 plasmids pMRY13-131_2, pMRY15-117_2 and pMRY16-002_4. (A) Schematic comparative analysis of whole IncI2 plasmid sequences. The red and blue bars between chromosomal DNA sequences represent individual nucleotide matches in the forward and reverse directions, respectively. (B) Schematic representation and comparison of shufflon regions and flanking regions. Six coloured arrows (orange, sky blue, purple, light green, red and blue) represent the 3′ ORF region of the pilV gene in IncI2 plasmid shufflon segments. A grey arrow and black box represent the rci gene and main pilV gene, respectively. (C) Multiple alignments of six repeat regions of shufflon segments. These repeat regions were analysed by the sequence logo program. (D) Identical ratio plot of the 3′-end of shufflon regions and minimum spanning network analysis of SNVs in 23-bp conserved repeat regions. The numbers on the edges of the network show the number of SNVs between the repeat sequences.

Comparative analysis of the whole IncI2 plasmid sequences

To reveal the conservation of the whole IncI2 plasmid sequences, comparative analysis was performed for pMRY16-002_4, pMRY15-117_2, pMRY15-131_2 and the following five plasmids: (i) pUHKPC45-77, pDMC1097-77.775 kb and pKPC_CAV1596-78 using raw sequencing data deposited in the Short-Read Archive (SRA) database; (ii) pHNSHP45, a mcr-1-positive plasmid, using unregistered data in SRA; and (iii) R721, the first reported IncI2 plasmid, used as a reference. The length of the conserved region was 47,362 bp or 76.63% of the 61,805-bp sequence of pMRY16-002_4 (Fig. 4). The backbones of the IncI2 plasmids were divided into three types as follows: (i) pMRY15-131_2, pMRY15-117_2, pMRY16-002_4 and pHNSHP45; (ii) R721; and (iii) pUHKPC45-77, pDMC1097-77.775 kb and pKPC_CAV1596-78. The sequence structure of pHNSHP45 showed high similarity to those of three plasmids (pMRY15-131_2, pMRY15-117_2 and pMRY16-002_4); there were differences in the shufflon region and an insertion of a mobile element containing the IS683 sequence (Fig. 4). The rci gene and the pil operon were conserved in all plasmids; conversely, the composition of the shufflon segments showed variability. Shufflon segments were detected using assembled sequences and NGS sequencing raw reads because of the reevaluation of the shufflons in the deposited complete IncI2 plasmid sequences. Segment BD was detected in all plasmids; however, segment C was only detected in pMRY15-117_2 and pMRY16-002_4 (Fig. 4). Although 18 complete IncI2 plasmid sequences are available in the GenBank database, NGS raw sequencing reads of only three of them were retrieved from the SRA database (Table S5). Among the other 15 IncI2 plasmids, whose sequences have only been deposited in GenBank, 12 plasmids harbour segments A and BD but not segment C. These results showed that segment BD shows a noticeable stability in the shufflon of IncI2 plasmids.

Figure 4
figure 4

Comparative plasmid analysis of eight complete IncI2 plasmids. The area included shows that the percentage of identity between similar regions in the reference plasmid, pMRY16-002_4, and the other plasmids was at least 80%. The existence patterns of shufflon segments are shown on the right side.

Discussion

The SMRT sequencer and the hierarchical genome assembly process (HGAP) assembler are able to generate long reads and easily produce long assembled contigs without gap regions5, 29, 32. Chromosomal and plasmid sequences were assembled into one contig in this study. However, misassembled sequences were detected in the shufflon regions of IncI2 plasmids pMRY15-117_2 and pMRY16-002_4 because of high heterogeneity of these regions in the isolated clones. Brouwer et al. also discussed the utility of IncI1 plasmids and shufflon region sequences using Roche-454 and Illumina platforms29, 35, 38, 39. Although error correction programs for PacBio unitigs, iCORN26, 8, 36 and Pilon29, 37 have been released, the heterogeneous remarkable rearrangement region could interfere with the correction software, preventing obtaining a correct complete sequence. Therefore, it is suggested that careful hybrid analysis using SMRT and short-read sequencing is necessary to resolve sequencing errors in the high rearrangement region. In contrast, the quantitative analysis of the pilV gene structure was performed using SMRT and Illumina sequencing data; however, the Illumina reads were too short to analyse the structural variation of the whole-shufflon sequence. The discordant quantitative analysis between SMRT and Illumina sequencing data might be caused by long single-nucleotide and/or tandem repeat regions that were difficult to sequence in Illumina data. Thus, it is strongly suggested that SMRT sequencing is used for the quantitative structural variation analysis.

It has previously been reported that the combination pattern of shufflon segments in IncI1 plasmids is conserved within the same STs29, 40. Our complete genome sequence analysis revealed that the IncI2 plasmids, excluding the shufflon region, were highly conserved among E. coli MRY16-002 (ST117), MRY15-117 (ST457) and MRY15-131 (ST457). In particular, the only difference between the IncI2 plasmids from the two ST457 strains was the insertion or deletion of segment C (Fig. 3B). However, the compositions of virulence factor genes were different between MRY15-117 and MRY15-131 (Table S2). Thus, it is suggested that the combination pattern of shufflon segments is variable within the same ST and among different STs, similar to transposable virulence factor genes, which constitute part of accessory genes.

The quantitative analysis of shufflon regions revealed that there are various shufflon structures in each strain (Fig. 2A). Previous studies have discussed the mechanism of shufflon rearrangement and the variability of shufflon sequences among IncI-type plasmids5, 29, 41. Our study is the first report on quantitative analysis of the whole-shufflon structure rearrangement in each isolated bacterial clone. Initially, we supposed that the abundance ratio in the shufflon structure was stochastically equal for each isolated clone. Indeed, the IncI2 plasmid pMRY15-131_2, which only has shufflon segment BD, showed nearly perfect evenness in the structural variation. In contrast, the ratios of structural variation showed deviations in pMRY16-002_4 and pMRY15-117_2, which possess two and three shufflon segments, respectively. In particular, the abundance ratios of pilVC and pilVC′ were lower than those of other pilV genes in pMRY16-002_4 and pMRY15-117_2 (Fig. 2B). These deflections in recombination events might correlate with the types of inverted repeat sequences in the shufflon. It has previously been reported that the site-specific recombinase Rci promotes site-specific recombination between any two of 19-bp inverted repeat sequences of the shufflon35, 38, 39, 42. Our results showed that the nucleotide identity ranged from 85% to 100% among 20-bp repeat regions of the six segments; as shown above, the length of the conserved repeat region was up to 23 bp (Fig. 3D). In the 23-bp conserved region, the repeat sequences of ORF C′ showed low identify (approximately 74 to 78%) to other repeats; there were two-nucleotide, one-nucleotide and six-nucleotide differences in the 23-bp conserved region of segments A, BD and C, respectively. Hence, the deflection in the shufflon structural variation might be related to the nucleotide polymorphism in the 23-bp repeat regions, and the low stability of the repeat region in segment C might also lead to low abundance ratios of pilVC and pilVC′. Moreover, the plasmid copy number analysis showed that the IncI2 plasmids were single-copy plasmids (Table 1), suggesting that a single bacterial cell carries an IncI2 plasmid with a single pilV type. Taken together, these results imply that pili consisting of various PilV proteins are unevenly present in bacterial clonal populations even though individual bacterial cells may express a single PilV protein.

The methylation analysis revealed that m6A methylation sites were found spread all over the genome in three strains; however, these methylation sites were not detected in all shufflon inverted repeat regions (Figure S1 and Table S4). It was reported that the methylated nucleotides affect DNA-protein interaction, and m6A is involved in many biological phenomenon in bacteria (i.e. bacterial defence against bacteriophage, regulation of chromosome replication, mismatch repair, conjugal transfer of plasmids, packaging of phage DNA, and transcriptional regulation of virulence genes)43. Thus, it is suggested that m6A methylation does not directly influence the interaction between Rci recombinase and shufflon inverted repeat regions and rearrangement of the shufflon structure.

It has previously been reported that rearrangements in the shufflon determine the specificity of bacterial recipients6, 8, 44, i.e. that PilV structures specifically recognise the LPS pattern. Comparative analysis revealed that the four mcr-1-positive plasmids, i.e. pMRY15-131_2, pMRY15-117_2, pMRY16-002_4 and pHNSHP45, were highly conserved in all sequence regions, except the shufflon regions, which showed the presence of six, four and two PilV structures in pMRY16-002_4, pMRY15-117_2 and pMRY15-131_2, respectively (Fig. 3A). These results suggest that the IncI2 plasmid encoding six PilV structures, pMRY16-002_4, might have diverse transconjugation ability among the Enterobacteriaceae. Moreover, comparative analysis of the shufflon regions among 21 IncI2 plasmids revealed the deletion of segment C in 16 plasmids (Table S5). Brouwer et al. discussed that careful review of contigs produced by automated assembly may be needed29, 45, which is consistent with the results of this study. Among the 18 IncI2 plasmid sequences deposited in GenBank, only those of three plasmids (pUHKPC45-77, pDMC1097-77.775 kb and pKPC_CAV1596-78) were available in the SRA database (Table S5). The patterns of shufflon segments were determined using Illumina and/or PacBio raw reads in six plasmids (pMRY15-131_2, pMRY15-117_2, pMRY16-002_4, pUHKPC45-77, pDMC1097-77.775 kb and pKPC_CAV1596-78) and were the same as those in the assembled sequence data. Although there is a possibility that some misassembled sequence information has been deposited in public databases, segments C and BD of IncI2 plasmids might tend to carry insertions or deletions and have noticeable stability, respectively.

In conclusion, we demonstrated that SMRT sequencing is effective in quantitative structural variation analysis of regions with high rearrangement rates, including the shufflon. This analysis could reveal not only the diversity of shufflon structures but also the actual rearrangement status for the shufflon in isolated bacterial clones. The shufflon is important for IncI2 conjugative plasmid transfer in various Enterobacteriaceae. Moreover, detection of mcr-1-carrying IncI2 plasmids of diverse origin has been reported worldwide, and multidrug-resistant bacteria, including those resistant to colistin, have become a global issue. Thus, the detection of whole-shufflon structures and pilV gene variants might be effective for control and surveillance of antimicrobial-resistant bacteria harbouring IncI2 plasmids.

Methods

Bacterial strains and DNA extraction

Three strains of E. coli were used in the present study (Table 1). These single-clone strains were cultured in Luria–Bertani broth at 37 °C for 24 h under aerobic conditions. To prepare long-chain genomic DNA, cultured isolates were treated with 0.1% sodium dodecyl sulphate in TE buffer for 30 min at 65 °C, followed by proteinase K treatment for 4 h at 55 °C. Phenol/chloroform was used for removing proteins from crude DNA samples, followed by dialysis against TE buffer. The quality of the genomic DNA was analysed by 1% agarose gel electrophoresis with GelRed (Biotium, Inc., Hayward, CA, USA), and the purity of DNA was assessed using a NanoDrop spectrophotometer (NanoDrop Technologies). Plasmid isolation was performed by pulsed-field gel electrophoresis with S1 nuclease-treated agarose plugs. The gel-extracted plasmids were purified using the Wizard gel purification kit (Promega, Madison, WI, USA).

Sequencing

Short-insert DNA libraries (approximately 400 bp in length) were constructed using a NexteraXT sample prep kit according to the manufacturer’s instructions (Illumina, Inc., San Diego, CA, USA), and the libraries were sequenced using MiSeq with the MiSeq version 3 600 cycle reagent kit (Illumina, Inc.). Genomic DNA was sheared by g-TUBE (Covaris, Inc., Woburn, MA, USA) followed by size selection using BluePippin (Sage Science, Inc., Beverly, MA, USA). Long-insert DNA libraries (approximately ≥20 kb in length) were constructed using the SMRTBell template prep kit, version 1.0, according to the manufacturer’s instructions (Pacific Biosciences, Menlo Park, CA, USA). The samples were sequenced using the PacBio RSII (Pacific Biosciences) with a movie length of 240 min; one SMRT cell was used for each sample.

De novo assembly and gap closing

To remove the adapter and low-quality regions from Illumina short reads, short paired-end 300-mer reads were analysed by the Skewer software version 0.240, 46 and Sickle version 1.33 (https://github.com/najoshi/sickle). The trimmed short reads were assembled using the A5 MiSeq software version 2014060441, 47. For adapter removal and de novo assembly with long reads, SMRT raw reads were analysed using RS HGAP Assembly in SMRT Analysis version 2.3. The error correction of long-subread sequences was also performed using pacBioToCA in Celera Assembler version 8.242, 48 for SMRT subreads and quality-trimmed Illumina short reads. The assembled contigs of long subreads were manually removed at 5′-end and 3′-end low-coverage regions (≤×100 coverage). A BLASTN homology search44, 49 was performed against each prime-end region; the gap regions were closed at completely identical regions. In IncI2 plasmids, the shufflon region was replaced using the corrected long subreads. To detect incorrect gap closing and misassembled sequences, SMRT long subreads and Illumina short reads were aligned to tentative complete genomes using BLASR version 150 and BWA-MEM45, 51, respectively.

Annotation

Gene prediction was performed for complete genome sequences with the PROKKA version 1.1146, followed by InterProScan47 and BLASTP searches using the nr database for validation. Searches for drug resistance genes, virulence factor genes, plasmid Inc types and serotypes were performed using the ABRicate program version 0.2 (https://github.com/tseemann/abricate) with ResFinder48 and the Lahey β-lactamase database (https://www.ncbi.nlm.nih.gov/projects/pathogens/beta-lactamase-data-resources/), VirulenceFinder49 and the VFDB51 database, PlasmidFinder52 database and SerotypeFinder53 database, respectively. MLST of E. coli was performed by the MLST program version 1.2 (https://github.com/tseemann/mlst/issues) against the PubMLST typing database.

Quantitative structural variation analysis

The putative structural variation of the shufflon was calculated with the number of detected shufflon segments in the corrected long subreads; all possible shufflon structural sequence patterns were constructed. To detect and calculate the real structural variation, these possible sequences were searched against PacBio subreads using BLASTN; the detected subreads were counted only if they contained adjacent shufflon regions of the pilV and rci genes. The recombined pilV sequence type was also analysed by BLASTN using the MiSeq-trimmed reads and counted with only short reads containing a part of the main pilV region, an inverted repeat sequence and a part of the shufflon segment.

Methylome analysis

Detection of the N6-Methyladenosine (m6A) methylation site and methylated motif was performed by RS_Modification_and_Motif_Analysis protocol in the SMRT Analysis version 2.3 using the standard mapping protocol. The interpulse duration (IPD) ratio of detected methylation sites was visualised using Circos version 0.6954.

Comparative plasmid analysis

A BLASTN homology search was performed for comparative sequence analysis of pMRY15-131_2, pMRY15-117_2 and pMRY16-002_4, followed by alignment visualisation with the ACT55 and Easyfig56 programs. SNVs in 23-bp repeat regions were compared by minimum spanning network analysis of PopART (http://popart.otago.ac.nz). Whole-plasmid comparative analysis was performed using the GView server with the default parameters (https://server.gview.ca/)57 for eight complete IncI2 plasmid sequences. The shufflon segments were searched by BLASTN against complete IncI2 plasmid sequences and NGS raw reads.

Data deposition

Raw sequence reads were deposited in the DDBJ Sequence Read Archive under accession number DRA004888 (BioProject: PRJDB5007, BioSample: SAMD00056130–SAMD00056132 and Experiment: DRX060089–DRX060094). The complete genome sequences were deposited in DDBJ under the following accession numbers: chromosome of strain MRY16-002, AP017610; pMRY16-002_1, AP017611; pMRY16-002_2, AP017612; pMRY16-002_3, AP017613; pMRY16-002_4, AP017614; pMRY16-002_5, AP017615; pMRY16-002_6, AP017616; chromosome of strain MRY15-117, AP017617; pMRY15-117_1, AP017618; pMRY15-117_2, AP017619; chromosome of strain MRY15-131, AP017620; pMRY15-131_1, AP017621l; and pMRY15-131_2, AP017622.

Reference

  1. Gyohda, A., Furuya, N., Kogure, N. & Komano, T. Sequence-specific and non-specific binding of the Rci protein to the asymmetric recombination sites of the R64 shufflon. J Mol Biol 318, 975–983 (2002).

    CAS  Article  PubMed  Google Scholar 

  2. Komano, T., Kubo, A., Kayanuma, T., Furuichi, T. & Nisioka, T. Highly mobile DNA segment of IncI alpha plasmid R64: a clustered inversion region. J Bacteriol 165, 94–100 (1986).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  3. Komano, T., Kubo, A. & Nisioka, T. Shufflon: multi-inversion of four contiguous DNA segments of plasmid R64 creates seven different open reading frames. Nucleic Acids Res 15, 1165–1172 (1987).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  4. Ishiwa, A. & Komano, T. The lipopolysaccharide of recipient cells is a specific receptor for PilV proteins, selected by shufflon DNA rearrangement, in liquid matings with donors bearing the R64 plasmid. Mol Gen Genet 263, 159–164 (2000).

    CAS  Article  PubMed  Google Scholar 

  5. Komano, T., Fujitani, S., Funayama, N., Kanno, A. & Sakuma, K. Physical and genetic analyses of IncI2 plasmid R721: evidence for the presence of shufflon. Plasmid 23, 248–251 (1990).

    CAS  Article  PubMed  Google Scholar 

  6. Ishiwa, A. & Komano, T. Thin pilus PilV adhesins of plasmid R64 recognize specific structures of the lipopolysaccharide molecules of recipient cells. J Bacteriol 185, 5192–5199 (2003).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  7. Cottell, J. L., Saw, H. T. H., Webber, M. A. & Piddock, L. J. V. Functional genomics to identify the factors contributing to successful persistence and global spread of an antibiotic resistance plasmid. BMC Microbiol 14, 168 (2014).

    Article  PubMed  PubMed Central  Google Scholar 

  8. Ishiwa, A. & Komano, T. PilV adhesins of plasmid R64 thin pili specifically bind to the lipopolysaccharides of recipient cells. J Mol Biol 343, 615–625 (2004).

    CAS  Article  PubMed  Google Scholar 

  9. Venturini, C. et al. Sequences of two related multiple antibiotic resistance virulence plasmids sharing a unique IS26-related molecular signature isolated from different Escherichia coli pathotypes from different hosts. PLoS ONE 8, e78862 (2013).

    ADS  CAS  Article  PubMed  PubMed Central  Google Scholar 

  10. Falagas, M. E., Karageorgopoulos, D. E. & Nordmann, P. Therapeutic options for infections with Enterobacteriaceae producing carbapenem-hydrolyzing enzymes. Future Microbiol 6, 653–666 (2011).

    CAS  Article  PubMed  Google Scholar 

  11. Liu, Y.-Y. et al. Emergence of plasmid-mediated colistin resistance mechanism MCR-1 in animals and human beings in China: a microbiological and molecular biological study. Lancet Infect Dis 16, 161–168 (2016).

    Article  PubMed  Google Scholar 

  12. Suzuki, S., Ohnishi, M., Kawanishi, M., Akiba, M. & Kuroda, M. Investigation of a plasmid genome database for colistin-resistance gene mcr-1. Lancet Infect Dis 16, 284–285 (2016).

    CAS  Article  PubMed  Google Scholar 

  13. McGann, P. et al. Escherichia coli Harboring mcr-1 and blaCTX-M on a Novel IncF Plasmid: First Report of mcr-1 in the United States. Antimicrob Agents Chemother 60, 4420–4421 (2016).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  14. Xavier, B. B., Lammens, C., Butaye, P., Goossens, H. & Malhotra-Kumar, S. Complete sequence of an IncFII plasmid harbouring the colistin resistance gene mcr-1 isolated from Belgian pig farms. J Antimicrob Chemother 71, 2342–2344 (2016).

    CAS  Article  PubMed  Google Scholar 

  15. Malhotra-Kumar, S. et al. Colistin resistance gene mcr-1 harboured on a multidrug resistant plasmid. Lancet Infect Dis 16, 283–284 (2016).

    CAS  Article  PubMed  Google Scholar 

  16. Sun, J. et al. Complete Nucleotide Sequence of an IncI2 Plasmid Coharboring blaCTX-M-55 and mcr-1. Antimicrob Agents Chemother 60, 5014–5017 (2016).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  17. Yang, Y.-Q. et al. Co-occurrence of mcr-1 and ESBL on a single plasmid in Salmonella enterica. J Antimicrob Chemother 71, 2336–2338 (2016).

    CAS  Article  PubMed  Google Scholar 

  18. Du, H., Chen, L., Tang, Y.-W. & Kreiswirth, B. N. Emergence of the mcr-1 colistin resistance gene in carbapenem-resistant Enterobacteriaceae. Lancet Infect Dis 16, 287–288 (2016).

    CAS  Article  PubMed  Google Scholar 

  19. Zhi, C., Lv, L., Yu, L.-F., Doi, Y. & Liu, J.-H. Dissemination of the mcr-1 colistin resistance gene. Lancet Infect Dis 16, 292–293 (2016).

    Article  PubMed  Google Scholar 

  20. Zurfluh, K., Klumpp, J., Nüesch-Inderbinen, M. & Stephan, R. Full-Length Nucleotide Sequences of mcr-1-Harboring Plasmids Isolated from Extended-Spectrum-β-Lactamase-Producing Escherichia coli Isolates of Different Origins. Antimicrob Agents Chemother 60, 5589–5591 (2016).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  21. Di Pilato, V. et al. mcr-1.2, a New mcr Variant Carried on a Transferable Plasmid from a Colistin-Resistant KPC Carbapenemase-Producing Klebsiella pneumoniae Strain of Sequence Type 512. Antimicrob Agents Chemother 60, 5612–5615 (2016).

    Article  PubMed  PubMed Central  Google Scholar 

  22. Liu, L. et al. Comparison of next-generation sequencing systems. J Biomed Biotechnol 2012, 251364 (2012).

    PubMed  PubMed Central  Google Scholar 

  23. Hall, N. Advanced sequencing technologies and their wider impact in microbiology. J Exp Biol 210, 1518–1525 (2007).

    CAS  Article  PubMed  Google Scholar 

  24. Shendure, J. & Ji, H. Next-generation DNA sequencing. Nat Biotechnol 26, 1135–1145 (2008).

    CAS  Article  PubMed  Google Scholar 

  25. MacLean, D., Jones, J. D. G. & Studholme, D. J. Application of ‘next-generation’ sequencing technologies to microbial genetics. Nat Rev Microbiol 7, 287–296 (2009).

    PubMed  Google Scholar 

  26. van Vliet, A. H. M. Next generation sequencing of microbial transcriptomes: challenges and opportunities. FEMS Microbiol Lett 302, 1–7 (2010).

    Article  PubMed  Google Scholar 

  27. Suenaga, H. Targeted metagenomics: a high-resolution metagenomics approach for specific gene clusters in complex microbial communities. Environ Microbiol 14, 13–22 (2012).

    CAS  Article  PubMed  Google Scholar 

  28. Weinstock, G. M. Genomic approaches to studying the human microbiota. Nature 489, 250–256 (2012).

    ADS  CAS  Article  PubMed  PubMed Central  Google Scholar 

  29. Brouwer, M. S. M. et al. IncI shufflons: Assembly issues in the next-generation sequencing era. Plasmid 80, 111–117 (2015).

    CAS  Article  PubMed  Google Scholar 

  30. Vink, C., Rudenko, G. & Seifert, H. S. Microbial antigenic variation mediated by homologous DNA recombination. FEMS Microbiol Rev 36, 917–948 (2012).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  31. Anjum, M. F. et al. Colistin resistance in Salmonella and Escherichia coli isolates from a pig farm in Great Britain. J Antimicrob Chemother 71, 2306–2313 (2016).

    CAS  Article  PubMed  Google Scholar 

  32. Rhoads, A. & Au, K. F. PacBio Sequencing and Its Applications. Genomics Proteomics Bioinformatics 13, 278–289 (2015).

    Article  PubMed  PubMed Central  Google Scholar 

  33. Sudmant, P. H. et al. An integrated map of structural variation in 2,504 human genomes. Nature 526, 75–81 (2015).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  34. English, A. C. et al. Assessing structural variation in a personal genome-towards a human reference diploid genome. BMC Genomics 16, 286 (2015).

    Article  PubMed  PubMed Central  Google Scholar 

  35. Gyohda, A., Funayama, N. & Komano, T. Analysis of DNA inversions in the shufflon of plasmid R64. J Bacteriol 179, 1867–1871 (1997).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  36. Otto, T. D., Sanders, M., Berriman, M. & Newbold, C. Iterative Correction of Reference Nucleotides (iCORN) using second generation sequencing technology. Bioinformatics 26, 1704–1707 (2010).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  37. Walker, B. J. et al. Pilon: an integrated tool for comprehensive microbial variant detection and genome assembly improvement. PLoS ONE 9, e112963 (2014).

    ADS  Article  PubMed  PubMed Central  Google Scholar 

  38. Gyohda, A. & Komano, T. Purification and characterization of the R64 shufflon-specific recombinase. J Bacteriol 182, 2787–2792 (2000).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  39. Gyohda, A., Zhu, S., Furuya, N. & Komano, T. Asymmetry of shufflon-specific recombination sites in plasmid R64 inhibits recombination between direct sfx sequences. J Biol Chem 281, 20772–20779 (2006).

    CAS  Article  PubMed  Google Scholar 

  40. Jiang, H., Lei, R., Ding, S.-W. & Zhu, S. Skewer: a fast and accurate adapter trimmer for next-generation sequencing paired-end reads. BMC Bioinformatics 15, 182 (2014).

    Article  PubMed  PubMed Central  Google Scholar 

  41. Coil, D., Jospin, G. & Darling, A. E. A5-miseq: an updated pipeline to assemble microbial genomes from Illumina MiSeq data. Bioinformatics 31, 587–589 (2015).

    CAS  Article  PubMed  Google Scholar 

  42. Myers, E. W. et al. A whole-genome assembly of Drosophila. Science 287, 2196–2204 (2000).

    ADS  CAS  Article  PubMed  Google Scholar 

  43. Wion, D. & Casadesús, J. N6-methyl-adenine: an epigenetic signal for DNA-protein interactions. Nat Rev Microbiol 4, 183–192 (2006).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  44. Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J. Basic local alignment search tool. J Mol Biol 215, 403–410 (1990).

    CAS  Article  PubMed  Google Scholar 

  45. Li, H. & Durbin, R. Fast and accurate long-read alignment with Burrows-Wheeler transform. Bioinformatics 26, 589–595 (2010).

    Article  PubMed  PubMed Central  Google Scholar 

  46. Seemann, T. Prokka: rapid prokaryotic genome annotation. Bioinformatics 30, 2068–2069 (2014).

    CAS  Article  PubMed  Google Scholar 

  47. Jones, P. et al. InterProScan 5: genome-scale protein function classification. Bioinformatics 30, 1236–1240 (2014).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  48. Zankari, E. et al. Identification of acquired antimicrobial resistance genes. J Antimicrob Chemother 67, 2640–2644 (2012).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  49. Joensen, K. G. et al. Real-time whole-genome sequencing for routine typing, surveillance, and outbreak detection of verotoxigenic Escherichia coli. J Clin Microbiol 52, 1501–1510 (2014).

    Article  PubMed  PubMed Central  Google Scholar 

  50. Chaisson, M. J. & Tesler, G. Mapping single molecule sequencing reads using basic local alignment with successive refinement (BLASR): application and theory. BMC Bioinformatics 13, 238 (2012).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  51. Chen, L., Zheng, D., Liu, B., Yang, J. & Jin, Q. VFDB 2016: hierarchical and refined dataset for big data analysis–10 years on. Nucleic Acids Res 44, D694–7 (2016).

    Article  PubMed  Google Scholar 

  52. Carattoli, A. et al. In silico detection and typing of plasmids using PlasmidFinder and plasmid multilocus sequence typing. Antimicrob Agents Chemother 58, 3895–3903 (2014).

    Article  PubMed  PubMed Central  Google Scholar 

  53. Joensen, K. G., Tetzschner, A. M. M., Iguchi, A., Aarestrup, F. M. & Scheutz, F. Rapid and Easy In Silico Serotyping of Escherichia coli Isolates by Use of Whole-Genome Sequencing Data. J Clin Microbiol 53, 2410–2426 (2015).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  54. Krzywinski, M. et al. Circos: an information aesthetic for comparative genomics. Genome Res 19, 1639–1645 (2009).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  55. Carver, T. et al. Artemis and ACT: viewing, annotating and comparing sequences stored in a relational database. Bioinformatics 24, 2672–2676 (2008).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  56. Sullivan, M. J., Petty, N. K. & Beatson, S. A. Easyfig: a genome comparison visualizer. Bioinformatics 27, 1009–1010 (2011).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  57. Petkau, A., Stuart-Edwards, M., Stothard, P. & Van Domselaar, G. Interactive microbial genome visualization with GView. Bioinformatics 26, 3125–3126 (2010).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

This work was supported by Research Program on Emerging and Re-emerging Infectious Diseases from the Japan Agency for Medical Research and Development, AMED on the ‘16fk0108305j0003’.

Author information

Authors and Affiliations

Authors

Contributions

T.S., S.S. and M.K. designed the study. M.K. and M.O. contributed materials. A.S., K.K., M.M. and S.S. performed the experiments. T.S. and A.Y. performed the bioinformatic analysis. T.S. and M.K. wrote the paper.

Corresponding author

Correspondence to Tsuyoshi Sekizuka.

Ethics declarations

Competing Interests

The authors declare that they have no competing interests.

Additional information

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Sekizuka, T., Kawanishi, M., Ohnishi, M. et al. Elucidation of quantitative structural diversity of remarkable rearrangement regions, shufflons, in IncI2 plasmids. Sci Rep 7, 928 (2017). https://doi.org/10.1038/s41598-017-01082-y

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1038/s41598-017-01082-y

Further reading

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing