bioRxiv Science⌕ Search

Biology subjects

Rahm, A. D.

Publications and source records attributed to Rahm, A. D..

3 recordsLinked to original sources
bioRxiv2022

Informing plasmid compatibility with bacterial hosts using protein-protein interaction data

The compatibility of plasmids with new host cells is significant given their role in spreading antimicrobial resistance (AMR)1 and virulence factor genes. Evaluating this using in vitro screening is laborious and can be informed by computational analyses of plasmid-host compatibility through rates of protein-protein interactions (PPIs) between plasmid and host cell proteins. We identified large excesses of such PPIs in eight important plasmids, including pOXA-48, using most known bacteria (n=4,363). 23 species had high rates of interactions with four blaOXA-48-positive plasmids. We also identified 48 species with high interaction rates with plasmids common in Escherichia coli. We found a strong association between one plasmid and the fimbrial adhesin operon pil, which could enhance host cell adhesion in aqueous environments. An excess rate of PPIs could be a sign of host-plasmid compatibility, which is important for AMR control given that plasmids like pOXA-48 move between species with ease. HighlightsO_LIWe tested for protein interactions between key plasmids and 4,363 bacterial hosts C_LIO_LI23 host species had high rates of protein interactions with four blaOXA-48 plasmids C_LIO_LIPseudomonas species showed high rates of interactions with these plasmids C_LIO_LIHost-plasmid protein-protein interaction rates may be associated with compatibility C_LI

genomics↗
bioRxiv2022

Bacterial plasmid-associated and chromosomal proteins have fundamentally different properties in protein interaction networks

Plasmids facilitate horizontal gene transfer, which enables the diversification of pathogens into new anatomical and environmental niches, implying that plasmid-encoded genes can cooperate well with chromosomal genes. We hypothesise that such mobile genes are functionally different to chromosomal ones due to this ability to encode non-essential functions like antimicrobial resistance and traverse distinct host cells. The effect of plasmid-driven gene gain on protein-protein interaction network topology is an important question in this area. Moreover, the extent to which these chromosomally- and plasmid-encoded proteins interact with proteins from their own groups compared to the levels with the other group remains unclear. Here, we examined the incidence and protein-protein interactions of all known plasmid-encoded genes across representative specimens from most bacteria using all available plasmids. We found that such plasmid-encoded genes constitute [~]0.7% of the total number of genes per bacterial sample, and that plasmid genes are preferentially associated with different species but had limited taxonomical power beyond this. Surprisingly, plasmid-encoded proteins had both more protein-protein interactions compared to chromosomal proteins, countering the hypothesis that genes with higher mobility rates should have fewer protein-level interactions. Nonetheless, topological analysis and investigation of the protein-protein interaction networks connectivity and change in the number of independent components demonstrated that the plasmid-encoded proteins had limited overall impact in >96% of samples. This paper assembled extensive data on plasmid-encoded proteins, their interactions and associations with diverse bacterial specimens that is available for the community to investigate in more detail. Significance statementIt is well-established that plasmids drive new traits in their bacterial hosts, but the extent to which host-plasmid co-evolution is evident at the level of protein-protein interactions remains unclear. To address this, we compiled and analysed all available valid bacterial plasmids and associated proteins to explore the compositional differences between chromosomal and plasmid-encoded proteins and their interaction levels. We found that plasmid-encoded genes were highly correlated across the bacterial samples such that they had a high association with taxonomic context. Contrasting with the complexity hypothesis, plasmid-encoded proteins had far more interactions on average than chromosomal ones, though they had minimal effects on protein-protein interaction network structure. This demonstrated that host-plasmid co-evolution is evident and detectable at the level of protein interactions.

genomics↗