Abstract
In 1995 a severe haemolytic-uremic syndrome (HUS) outbreak in Adelaide occurred. A recent genomic analysis of Shiga toxigenic Escherichia coli (STEC) O111:H- strains 95JB1 and 95NR1 from this outbreak found that the more virulent isolate, 95NR1, harboured two additional copies of the Shiga toxin 2 (Stx2) genes encoded within prophage regions. The structure of the Stx2-converting prophages could not be fully resolved using short-read sequence data alone and it was not clear if there were other genomic differences between 95JB1 and 95NR1. In this study we have used Pacific Biosciences (PacBio) single molecule real-time (SMRT) sequencing to characterise the genome and methylome of 95JB1 and 95NR1. We completely resolved the structure of all prophages including two, tandemly inserted, Stx2-converting prophages in 95NR1 that were absent from 95JB1. Furthermore we defined all insertion sequences and found an additional IS1203 element in the chromosome of 95JB1. Our analysis of the methylome of 95NR1 and 95JB1 identified hemi-methylation of a novel motif (5′-CTGCm6AG-3′) in more than 4000 sites in the 95NR1 genome. These sites were entirely unmethylated in the 95JB1 genome, and included at least 177 potential promoter regions that could contribute to regulatory differences between the strains. IS1203 mediated deactivation of a novel type IIG methyltransferase in 95JB1 is the likely cause of the observed differential patterns of methylation between 95NR1 and 95JB1. This study demonstrates the capability of PacBio SMRT sequencing to resolve complex prophage regions and reveal the genetic and epigenetic heterogeneity within a clonal population of bacteria.
Highlights
In 1995 a large outbreak of hemolytic uremic syndrome (HUS) occurred in Adelaide, South Australia
Using Pacific Biosciences (PacBio) Single Molecule Real Time (SMRT) sequencing to determine their complete genome assemblies and methylomes, we completely resolve the genetic structure of all prophage-encoding regions, including two distinct, but closely related, Shiga toxin 2 (Stx2)-converting prophage in 95NR1 that are tandemly inserted in the same genomic location and appear to have been deleted from 95JB1
In contrast to the draft Illumina assemblies of 95JB1 and 95NR14, the small pO111_4 and pO111_5 colicin plasmids were not detected in the PacBio assemblies or in the raw PacBio read data consistent with their exclusion during the library preparation process
Summary
In 1995 a large outbreak of hemolytic uremic syndrome (HUS) occurred in Adelaide, South Australia. Based on Southern blot analysis, whereas early patient isolates (pre January 25th) had both stx[1] and stx[2] Shiga toxin genes, later patient isolates (post January 25th) were predicted to encode a second copy of stx2AB1,3 This genetic difference was hypothesised to account for the afore-mentioned difference in virulence between isolates collected before and after January 25th (represented by 95JB1 and 95NR1, respectively). Consistent with a heterogeneous population of O111 STEC within the primary source for the contaminated sausage (such as a livestock herd or individual animal), all core genome single nucleotide polymorphisms (SNPs) differentiating 95JB1 from 95NR1 (and the O111 STEC reference strain 11128) had occurred in 95JB14 These results indicated that 95JB1 was a derivative of 95NR1, even though 95JB1 was isolated earlier in the outbreak, and suggests that the two additional Stx[2] prophages were likely deleted from 95JB1 rather than acquired by 95NR14. We unambiguously determine the copy number and context of all IS elements in the genomes of 95NR1 and 95JB1 and reveal that a single difference in the IS complement could be directly responsible for differences in their methylome profiles
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
