Horizontal gene transfer in human-associated microorganisms inferred by phylogenetic reconstruction and reconciliation

Abstract

Horizontal gene transfer (HGT) is widespread in the evolution of prokaryotes, especially those associated with the human body. Here, we implemented large-scale gene-species phylogenetic tree reconstructions and reconciliations to identify putative HGT-derived genes in the reference genomes of microbiota isolated from six major human body sites by the NIH Human Microbiome Project. Comparisons with a control group representing microbial genomes from diverse natural environments indicated that HGT activity increased significantly in the genomes of human microbiota, which is confirmatory of previous findings. Roughly, more than half of total genes in the genomes of human-associated microbiota were transferred (donated or received) by HGT. Up to 60% of the detected HGTs occurred either prior to the colonization of the human body or involved bacteria residing in different body sites. The latter could suggest ‘genetic crosstalk’ and movement of bacterial genes within the human body via hitherto poorly understood mechanisms. We also observed that HGT activity increased significantly among closely-related microorganisms and especially when they were united by physical proximity, suggesting that the ‘phylogenetic effect’ can significantly boost HGT activity. Finally, we identified several core and widespread genes least influenced by HGT that could become useful markers for building robust ‘trees of life’ and address several outstanding technical challenges to improve the phylogeny-based genome-wide HGT detection method for future applications.

Phenotypic reconstruction of the last universal common ancestor reveals a complex cell

“…Irrespective of any network structure due to HGT, we recover a strong phylogenetic signal in all traits from both trees, implying that HGT does not rapidly act to change whole phenotypes or that it occurs mostly between very closely related lineages. While a positive correlation between the frequency of HGTs in distant lineages and genome size has been suggested 52 , several studies indicate that HGT in prokaryotes also depends on both physical and genetic proximity [53][54][55] . Moreover, there is accumulating evidence to suggest that some of the traits that we present such as motility, cell envelope and sporulation, were subject to little or rare HGT events during their evolutionary history in a number of prokaryotic lineages 22,23,25,45,55 .…”

Regulatory issues for genetically modified animals

“…Moreover, bacteria and their plasmids contribute to human microbiomes [102][103][104] ; the human gut is considered an antimicrobial resistance reservoir [105] . Recent studies of plasmids in human microbiomes suggest that plasmids and the genes they carry are actively swapped [106][107][108] via natural processes that happen between bacterial cells but not between bacterial and human cells. Note that antibiotic resistance can be passed to humans through physicianprescribed fecal microbiota transplantation, an emerging therapy for certain bowel disorders [109] .…”

Distribution of Epsilon-Polylysine Synthetases in Coryneform Bacteria Isolated from Cheese and Human Skin

“…Genes of them were downloaded from NCBI and included in the analysis. For the human skin microorganism genome dataset from reference (22) , 124 genomes were downloaded from the NIH Human Microbiome Project (https://www.hmpdacc.org/hmp/catalog/grid.php?dataset=genomic). We downloaded the following Pls proteins to use as a reference: BAG68864.1, BAH85292.1, WP_014008652.1, AZL89021.1, CCE28893.1, and BBU42014.1.…”

“…Next we investigated the distribution of pls in a genome collection of 156 bacteria isolated from cheese from Europe and United States (20). As the microorganisms from cheese are known to originate from or be related to animal and human skin microbiota (21) and ε-PL is also used in cosmetic products (1), we further included a genome collection of 124 microorganisms isolates from human skin (22). We used experimentally confirmed Pls protein sequences as query to do BlastP against the two collections with a cutoff of 40% sequence identity and 80% sequence coverage.…”