Journal: Nature Communications

Article Title: Diversification of bacterial genome content through distinct mechanisms over different timescales

doi: 10.1038/ncomms6471

Figure Lengend Snippet: ( a ) Maximum likelihood phylogeny based on the core genome annotated according to the distribution of sequence clusters. The branches of the phylogeny are coloured according to a maximally parsimonious reconstruction of CSP pherotype. The ‘CSP-3-like’ sequence was identical to the previously described CSP-3 (ref. ) but lacking an FNIFNF peptide. ( b ) Variation in accessory RMSs. The columns to the left indicate which of the three RMSs is present at the dpn locus by black bars in the appropriate rows. The eight columns to the right indicate the presence of other putative accessory RMSs, as inferred from the distribution of the relevant methylase COGs. Columns are labelled with the accession code of the sequence in , with black bars again indicating the presence of an RMS in the corresponding isolate. ( c ) Variation in the ivr locus. The left columns show reads corresponding to the 5′ part of the full-length spnIVRhsdS gene assigned to the two alternative 5′ TRD-encoding sequences A or B. The heatmap indicates the proportion of reads corresponding to the spnIVRhsdS gene that matched each allele, with red indicating a higher proportion and blue a lower proportion. The right columns show reads likely corresponding to the 3′ part of spnIVRhsdS assigned to the three alternative TRD-encoding sequences a, b or c. ( d ) Variation in the tvr locus. Eleven different spnTVRhsdS TRD-encoding sequences were identified across the population. When the TRD-encoding sequence was present as part of a full-length CDS, the cell is coloured red, if the TRD was found in the 5′ half (these are labelled with uppercase Roman numerals), and orange, if found in the 3′ half (these are labelled with lowercase Roman numerals). Where the TRD-encoding sequence was present as a lone fragment, the corresponding cell in the grid is coloured blue. Empty cells indicate the TRD-encoding sequence was absent from the corresponding isolate.

Article Snippet: The one other change at the dpn locus within a SC involved replacement of a Type II RMS (designated Dpn III and represented by SPN23F18640-18650 in the genome of S. pneumoniae ATCC 700669 (ref. )) present in all but one isolate of SC13, in which it had been replaced by Dpn I. Dpn III likely targets a different motif to Dpn I and Dpn II, both sensitive to adenine methylation , as functional domain information suggested the Dpn III RMS modified cytosine bases.

Techniques: Sequencing

Journal: Nature Communications

Article Title: Diversification of bacterial genome content through distinct mechanisms over different timescales

doi: 10.1038/ncomms6471

Figure Lengend Snippet: ( a ) Rate of change in different components of the accessory genome. The horizontal axis represents a threshold maximum cophenetic distance separating isolates, based on the core genome maximum likelihood phylogeny; the vertical axis represents the similarity observed in different aspects of the genome when considering all pairwise comparisons below this cophenetic distance threshold. The black line uses the median Jaccard similarity metric to trace the change in overall COG content between isolates. The blue and purple lines represent the median value of a similar metric calculated only using the subset of COGs characteristic of different MGE classes. Prophage-associated COG content (excluding the prophage remnant GI) diverged considerably within sequence clusters, indicating these MGEs are relatively transiently associated with pneumococcal hosts. By contrast, PRCI and ICE content (excluding the ICE ‘scars’) were stable within sequence clusters, but varied substantially between them. When considering the distribution of RMSs, each pairwise comparison was coded one where both isolates shared the same profile (as calculated from the data in ), and therefore the system could not be effective in preventing an MGE transmission, and zero otherwise. In the case of the ivr locus, the TRDs most commonly predicted to form the spnIVRhsdS gene were used to calculate this metric; in the case of the tvr locus, the profile of all TRDs at this locus, including whether or not they were present in a full-length spnTVRhsdS gene, was used. The plotted lines show the proportion of pairwise comparisons in which isolates had identical profiles based on the same core genome cophenetic distance thresholds as for the similarities in terms of COGs. This found the dpn locus and other accessory RMS to be conserved over relatively long evolutionary timescales, whereas the ivr and tvr loci were divergent between even very closely related isolates. ( b ) The distribution of pairwise cophenetic distances, calculated from a maximum likelihood core genome phylogeny , represented as a histogram.

Article Snippet: The one other change at the dpn locus within a SC involved replacement of a Type II RMS (designated Dpn III and represented by SPN23F18640-18650 in the genome of S. pneumoniae ATCC 700669 (ref. )) present in all but one isolate of SC13, in which it had been replaced by Dpn I. Dpn III likely targets a different motif to Dpn I and Dpn II, both sensitive to adenine methylation , as functional domain information suggested the Dpn III RMS modified cytosine bases.

Techniques: Sequencing, Comparison, Transmission Assay