Compensatory evolution facilitates the acquisition of multiple plasmids in bacteria

Abstract

17The coexistence of multicopy plasmids is a common phenomenon. However, the 18 evolutionary forces promoting these genotypes are poorly understood. In this study, we 19 have analyzed multiple ColE1 plasmids (pB1000, pB1005 and pB1006) coexisting 20 within Haemophilus influenzae RdKW20 in all possible combinations. When 21 transformed into the naïve host, each plasmid type presented a particular copy number 22 and produced a specific resistance profile and biological cost, whether alone or 23 coexisting with the other plasmids. Therefore, there was no fitness advantage associated 24 with plasmid coexistence that could explain these common plasmid associations in 25 nature. Using experimental evolution, we showed how H. influenzae Rd was able to 26 completely compensate the fitness cost produced by any of these plasmids. Crucially, 27 once the bacterium has compensated for a first plasmid, the acquisition of new 28 multicopy plasmid(s) did not produced any extra biological cost. We argue therefore 29 that compensatory adaptation pave the way for the acquisition of multiple coexisting 30 ColE1 plasmids. 31 Importance 32Antibiotic resistance is a major concern for human and animal health. Plasmids play a 33 major role in the acquisition and dissemination of antimicrobial resistance genes. In this 34 report we investigate, for the first time, how plasmids are capable to cohabit stably in 35 populations. This coexistence of plasmids is driven by compensatory evolution 36 alleviating the cost of a first plasmid, which potentiates the acquisition of further 37 plasmids at no extra cost. This phenomenon explains the high prevalence of plasmids 38 coexistance in wild type bacteria, which generates multiresistant clones and contributes 39 to the maintenance and spread of antibiotic resistance genes within bacterial 40 populations. 41 42Antibiotic resistance is a serious problem in animal and human health and bacterial 43 plasmids play an essential role in the dissemination of resistance (1). In last years, 44 numerous works have described the importance of small ColE1-like plasmids in the 45 dissemination of resistance genes (2-15). These plasmids replicate via two RNAs (16). 46Natural SNPs in these RNAs allow different ColE1-like plasmids to stably cohabit 47 within the same cell (17, 18). If the plasmids bear antibiotic resistance genes, this 48 cohabitation confers antibiotic multiresistance to the host bacteria (9). 49The acquisition of plasmids usually entails a biological cost to the host bacterium that 50 will generate a selection against plasmid-bearing clones (19, 20). Thus, it is reasonable 51 to assume that the accumulation of various plasmids will decrease the fitness of bacteria 52 and therefore clones bearing several plasmids will be outcompeted in bacterial 53 populations. Notwithstanding, ColE1 plasmids cohabitation is a common phenomenon 54 in nature (9, 11, 13, 14, 21). In this study we test two hypotheses that could explain the 55 high prevalence of plasmid coexistence in nature: i) Positive epi...

Positive selection inhibits plasmid coexistence in bacterial genomes

“…Shared targets of compensatory evolution among divergent plasmids have been reported in other plasmid-host systems [15,19], for example compensatory mutations in chromosomal helicases [19]. Consistent with previous work on other host-plasmid systems [15,19], our findings suggest that compensatory evolution could indeed promote plasmid coinfection by reducing the cost of multiple plasmids simultaneously. This has the effect of reducing the efficiency of negative selection against plasmid costs over time.…”

“…One explanation for abundant plasmid coinfection is that the fitness costs of acquiring multiple plasmids could be less than additive. Positive epistasis between plasmid costs could permit the accumulation of multiple plasmids by reducing the cost for plasmid-bearers of acquiring additional plasmids [4,15,16], however, positive epistatic interactions among plasmid costs are not universal [4]. Moreover, as we will show in this study, the methods by which positive epistasis has been previously estimated (i.e., competition of plasmid-carriers against plasmid-free cells [4]) may not measure the actual cost of plasmid coinfection.…”

Hotspot mutations and ColE1 plasmids contribute to the fitness of Salmonella Heidelberg in poultry litter

“…Recently, a study showed that the acquisition of three ColE1 plasmids in naïve H . influenzae hosts did not impose a fitness cost as long as the bacterium compensated for the first plasmid [ 94 ]. The presence of one plasmid favored the presence of a second plasmid and the presence of this second plasmid or third plasmid did not impose an additional significant biological cost, even though these plasmids produced a cost when they are alone in the cell.…”

Positive Selection Inhibits Plasmid Coexistence in Bacterial Genomes

“…One explanation for abundant plasmid coexistence in bacterial genomes is that the fitness costs of acquiring multiple plasmids could be less than additive. Positive epistasis between plasmid costs could permit the accumulation of multiple plasmids by reducing the cost for plasmid bearers of acquiring additional plasmids (4,16,17); however, positive epistatic interactions among plasmid costs are not universal (4). Moreover, as we show in this study, the methods by which positive epistasis has been previously estimated (i.e., competition of plasmid carriers against plasmid-free cells) (4) may not measure the actual cost of plasmid coinfection.…”