Limits...
Impact of target site distribution for Type I restriction enzymes on the evolution of methicillin-resistant Staphylococcus aureus (MRSA) populations.

Roberts GA, Houston PJ, White JH, Chen K, Stephanou AS, Cooper LP, Dryden DT, Lindsay JA - Nucleic Acids Res. (2013)

Bottom Line: A limited number of Methicillin-resistant Staphylococcus aureus (MRSA) clones are responsible for MRSA infections worldwide, and those of different lineages carry unique Type I restriction-modification (RM) variants.We experimentally demonstrate that this RM system is sufficient to block horizontal gene transfer between clinically important MRSA, confirming the bioinformatic evidence that each lineage is evolving independently.This analysis of the identification and distribution of target sites explains evolutionary patterns in a pathogenic bacterium.

View Article: PubMed Central - PubMed

Affiliation: EaStCHEM School of Chemistry, University of Edinburgh, The King's Buildings, Edinburgh EH9 3JJ, UK and Division of Clinical Sciences, St. George's, University of London, Cranmer Terrace, London, SW17 0RE, UK.

ABSTRACT
A limited number of Methicillin-resistant Staphylococcus aureus (MRSA) clones are responsible for MRSA infections worldwide, and those of different lineages carry unique Type I restriction-modification (RM) variants. We have identified the specific DNA sequence targets for the dominant MRSA lineages CC1, CC5, CC8 and ST239. We experimentally demonstrate that this RM system is sufficient to block horizontal gene transfer between clinically important MRSA, confirming the bioinformatic evidence that each lineage is evolving independently. Target sites are distributed randomly in S. aureus genomes, except in a set of large conjugative plasmids encoding resistance genes that show evidence of spreading between two successful MRSA lineages. This analysis of the identification and distribution of target sites explains evolutionary patterns in a pathogenic bacterium. We show that a lack of specific target sites enables plasmids to evade the Type I RM system thereby contributing to the evolution of increasingly resistant community and hospital MRSA.

Show MeSH

Related in: MedlinePlus

Plasmids have fewer CC5-2 target sites than expected. (A) Average target recognition sites (TRS) per kb for CC5-1, CC5-2 and CC1-2 enzymes in S. aureus sequences of whole-genomes (n = 18), plasmid (n = 233), bacteriophage (n = 50) and Staphylococcal cassette chromosomes with mecA (n = 35). By Mann–Witney two-tailed test, enzyme 5-1 had significantly more TRS in plasmids, and SCCmec than genomes (P < 0.01); enzyme 5-2 had significantly less TRS in plasmids than genomes (P < 0.0001); enzyme 1-2 had significantly less TRS in phage (P < 0.01) but significantly more TRS in SCCmec (P < 0.0001). Asterisk indicates significant, P < 0.01, Double asterisk indicates significant, P < 0.0001. (B) Percentage of MGEs lacking target sites for CC5-1, CC5-2 and CC1-2 in sequences of plasmid (n = 233), bacteriophage (n = 50) and Staphylococcal cassette chromosomes with mecA (n = 35). There are significantly more plasmids missing TRS for 5-2 than missing TRS for 5-1 or 1-2 (Chi square, P < 0.0001, indicated by asterisk) (C) TRS distribution profile of plasmid sequences (n = 233) ordered by size shows small plasmids (<10 kb) are more likely to be missing a CC5-2 TRS than missing a CC5-1 or CC1-2 TRS (Chi square, P < 0.0001), and that large conjugative plasmids (tra+; indicated by red dash, n = 14) are more likely to have zero CC5-2 TRS than zero CC5-1 or CC1-2 TRS (Chi square, P < 0.001). Each horizontal line represents a plasmid and is shaded according to the number of TRS.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3753647&req=5

gkt535-F4: Plasmids have fewer CC5-2 target sites than expected. (A) Average target recognition sites (TRS) per kb for CC5-1, CC5-2 and CC1-2 enzymes in S. aureus sequences of whole-genomes (n = 18), plasmid (n = 233), bacteriophage (n = 50) and Staphylococcal cassette chromosomes with mecA (n = 35). By Mann–Witney two-tailed test, enzyme 5-1 had significantly more TRS in plasmids, and SCCmec than genomes (P < 0.01); enzyme 5-2 had significantly less TRS in plasmids than genomes (P < 0.0001); enzyme 1-2 had significantly less TRS in phage (P < 0.01) but significantly more TRS in SCCmec (P < 0.0001). Asterisk indicates significant, P < 0.01, Double asterisk indicates significant, P < 0.0001. (B) Percentage of MGEs lacking target sites for CC5-1, CC5-2 and CC1-2 in sequences of plasmid (n = 233), bacteriophage (n = 50) and Staphylococcal cassette chromosomes with mecA (n = 35). There are significantly more plasmids missing TRS for 5-2 than missing TRS for 5-1 or 1-2 (Chi square, P < 0.0001, indicated by asterisk) (C) TRS distribution profile of plasmid sequences (n = 233) ordered by size shows small plasmids (<10 kb) are more likely to be missing a CC5-2 TRS than missing a CC5-1 or CC1-2 TRS (Chi square, P < 0.0001), and that large conjugative plasmids (tra+; indicated by red dash, n = 14) are more likely to have zero CC5-2 TRS than zero CC5-1 or CC1-2 TRS (Chi square, P < 0.001). Each horizontal line represents a plasmid and is shaded according to the number of TRS.

Mentions: The distribution of each identified TRS in sequenced S. aureus genomes (Figure 4A) revealed a random distribution of each amongst whole genomes of S. aureus from various lineages, as well as several MGEs. The exceptions were plasmids, which specifically harboured fewer sites for the CC5-2 RM enzyme and often lacked these sites altogether (Figure 4B). A direct comparison of 233 plasmids showed that this was not confined to small plasmids, which might be expected to carry fewer TRS by chance owing to their small size (Figure 4C). Notably, multiple large conjugative plasmids, identified by the carriage of the tra gene locus for transfer (red), were particularly deficient in the TRS for the CC5-2 RM enzyme (Figure 4C). We hypothesized that this represents evolution of the larger plasmids to escape this enzyme and tested this experimentally.Figure 4.


Impact of target site distribution for Type I restriction enzymes on the evolution of methicillin-resistant Staphylococcus aureus (MRSA) populations.

Roberts GA, Houston PJ, White JH, Chen K, Stephanou AS, Cooper LP, Dryden DT, Lindsay JA - Nucleic Acids Res. (2013)

Plasmids have fewer CC5-2 target sites than expected. (A) Average target recognition sites (TRS) per kb for CC5-1, CC5-2 and CC1-2 enzymes in S. aureus sequences of whole-genomes (n = 18), plasmid (n = 233), bacteriophage (n = 50) and Staphylococcal cassette chromosomes with mecA (n = 35). By Mann–Witney two-tailed test, enzyme 5-1 had significantly more TRS in plasmids, and SCCmec than genomes (P < 0.01); enzyme 5-2 had significantly less TRS in plasmids than genomes (P < 0.0001); enzyme 1-2 had significantly less TRS in phage (P < 0.01) but significantly more TRS in SCCmec (P < 0.0001). Asterisk indicates significant, P < 0.01, Double asterisk indicates significant, P < 0.0001. (B) Percentage of MGEs lacking target sites for CC5-1, CC5-2 and CC1-2 in sequences of plasmid (n = 233), bacteriophage (n = 50) and Staphylococcal cassette chromosomes with mecA (n = 35). There are significantly more plasmids missing TRS for 5-2 than missing TRS for 5-1 or 1-2 (Chi square, P < 0.0001, indicated by asterisk) (C) TRS distribution profile of plasmid sequences (n = 233) ordered by size shows small plasmids (<10 kb) are more likely to be missing a CC5-2 TRS than missing a CC5-1 or CC1-2 TRS (Chi square, P < 0.0001), and that large conjugative plasmids (tra+; indicated by red dash, n = 14) are more likely to have zero CC5-2 TRS than zero CC5-1 or CC1-2 TRS (Chi square, P < 0.001). Each horizontal line represents a plasmid and is shaded according to the number of TRS.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3753647&req=5

gkt535-F4: Plasmids have fewer CC5-2 target sites than expected. (A) Average target recognition sites (TRS) per kb for CC5-1, CC5-2 and CC1-2 enzymes in S. aureus sequences of whole-genomes (n = 18), plasmid (n = 233), bacteriophage (n = 50) and Staphylococcal cassette chromosomes with mecA (n = 35). By Mann–Witney two-tailed test, enzyme 5-1 had significantly more TRS in plasmids, and SCCmec than genomes (P < 0.01); enzyme 5-2 had significantly less TRS in plasmids than genomes (P < 0.0001); enzyme 1-2 had significantly less TRS in phage (P < 0.01) but significantly more TRS in SCCmec (P < 0.0001). Asterisk indicates significant, P < 0.01, Double asterisk indicates significant, P < 0.0001. (B) Percentage of MGEs lacking target sites for CC5-1, CC5-2 and CC1-2 in sequences of plasmid (n = 233), bacteriophage (n = 50) and Staphylococcal cassette chromosomes with mecA (n = 35). There are significantly more plasmids missing TRS for 5-2 than missing TRS for 5-1 or 1-2 (Chi square, P < 0.0001, indicated by asterisk) (C) TRS distribution profile of plasmid sequences (n = 233) ordered by size shows small plasmids (<10 kb) are more likely to be missing a CC5-2 TRS than missing a CC5-1 or CC1-2 TRS (Chi square, P < 0.0001), and that large conjugative plasmids (tra+; indicated by red dash, n = 14) are more likely to have zero CC5-2 TRS than zero CC5-1 or CC1-2 TRS (Chi square, P < 0.001). Each horizontal line represents a plasmid and is shaded according to the number of TRS.
Mentions: The distribution of each identified TRS in sequenced S. aureus genomes (Figure 4A) revealed a random distribution of each amongst whole genomes of S. aureus from various lineages, as well as several MGEs. The exceptions were plasmids, which specifically harboured fewer sites for the CC5-2 RM enzyme and often lacked these sites altogether (Figure 4B). A direct comparison of 233 plasmids showed that this was not confined to small plasmids, which might be expected to carry fewer TRS by chance owing to their small size (Figure 4C). Notably, multiple large conjugative plasmids, identified by the carriage of the tra gene locus for transfer (red), were particularly deficient in the TRS for the CC5-2 RM enzyme (Figure 4C). We hypothesized that this represents evolution of the larger plasmids to escape this enzyme and tested this experimentally.Figure 4.

Bottom Line: A limited number of Methicillin-resistant Staphylococcus aureus (MRSA) clones are responsible for MRSA infections worldwide, and those of different lineages carry unique Type I restriction-modification (RM) variants.We experimentally demonstrate that this RM system is sufficient to block horizontal gene transfer between clinically important MRSA, confirming the bioinformatic evidence that each lineage is evolving independently.This analysis of the identification and distribution of target sites explains evolutionary patterns in a pathogenic bacterium.

View Article: PubMed Central - PubMed

Affiliation: EaStCHEM School of Chemistry, University of Edinburgh, The King's Buildings, Edinburgh EH9 3JJ, UK and Division of Clinical Sciences, St. George's, University of London, Cranmer Terrace, London, SW17 0RE, UK.

ABSTRACT
A limited number of Methicillin-resistant Staphylococcus aureus (MRSA) clones are responsible for MRSA infections worldwide, and those of different lineages carry unique Type I restriction-modification (RM) variants. We have identified the specific DNA sequence targets for the dominant MRSA lineages CC1, CC5, CC8 and ST239. We experimentally demonstrate that this RM system is sufficient to block horizontal gene transfer between clinically important MRSA, confirming the bioinformatic evidence that each lineage is evolving independently. Target sites are distributed randomly in S. aureus genomes, except in a set of large conjugative plasmids encoding resistance genes that show evidence of spreading between two successful MRSA lineages. This analysis of the identification and distribution of target sites explains evolutionary patterns in a pathogenic bacterium. We show that a lack of specific target sites enables plasmids to evade the Type I RM system thereby contributing to the evolution of increasingly resistant community and hospital MRSA.

Show MeSH
Related in: MedlinePlus