Transmission of Staphylococcus aureus from Humans to Green Monkeys in The Gambia as Revealed by Whole-Genome Sequencing

Abstract

Staphylococcus aureus is an important pathogen of humans and animals. We genome sequenced 90 S. aureus isolates from The Gambia: 46 isolates from invasive disease in humans, 13 human carriage isolates, and 31 monkey carriage isolates. We inferred multiple anthroponotic transmissions of S. aureus from humans to green monkeys (Chlorocebus sabaeus) in The Gambia over different time scales. We report a novel monkey-associated clade of S. aureus that emerged from a human-to-monkey switch estimated to have occurred 2,700 years ago. Adaptation of this lineage to the monkey host is accompanied by the loss of phage-carrying genes that are known to play an important role in human colonization. We also report recent anthroponotic transmission of the well-characterized human lineages sequence type 6 (ST6) and ST15 to monkeys, probably because of steadily increasing encroachment of humans into the monkeys' habitat. Although we have found no evidence of transmission of S. aureus from monkeys to humans, as the two species come into ever-closer contact, there might be an increased risk of additional interspecies exchanges of potential pathogens.

Importance: The population structures of Staphylococcus aureus in humans and monkeys in sub-Saharan Africa have been previously described using multilocus sequence typing (MLST). However, these data lack the power to accurately infer details regarding the origin and maintenance of new adaptive lineages. Here, we describe the use of whole-genome sequencing to detect transmission of S. aureus between humans and nonhuman primates and to document the genetic changes accompanying host adaptation. We note that human-to-monkey switches tend to be more common than the reverse and that a novel monkey-associated clade is likely to have emerged from such a switch approximately 2,700 years ago. Moreover, analysis of the accessory genome provides important clues as to the genetic changes underpinning host adaptation and, in particular, shows that human-to-monkey switches tend to be associated with the loss of genes known to confer adaptation to the human host.

FIG 1

A maximum likelihood phylogenetic tree showing 5 major clades; branches are colored based on clade assignment. Tips are annotated by ST and colored by host. Reference genomes are annotated by name and CC and colored black. *, single locus variant of given ST.

FIG 2

A heat map nested into the phylogeny showing the proportion of accessory genes that are shared between isolates, determined by a pairwise comparison. Branches on the phylogeny are colored based on the clonal complex. Darker squares in the heat map indicate higher shared accessory genome content. To the right of the heat map is a bar chart showing the number of accessory genes for each genome.

FIG 3

BRIG. Whole-genome sequence analysis and comparison of USA300_FPR3757 to isolate a representative from each of the six subclusters in the monkey-associated clade 2. Circular diagram of the USA300 chromosome showing (from inner to outer), percent G+C, GC skew, and the homology based on BLASTN analysis of USA300_FPR3757 to H7, F2, G2, G11, F7, and H10. The sequence similarity ranges from 85% to 100%, as regions of homology of <85% were excluded. The outermost circle shows the location of large horizontally acquired pathogenicity islands and well-characterized phages present in USA300. *, single locus variant of given ST.