Toxin-antitoxin systems are widespread in bacterial genomes. They are usually composed of two elements: a toxin that inhibits an essential cellular process and an antitoxin that counteracts its cognate toxin. In the past decade, a number of new toxin-antitoxin systems have been described, bringing new growth inhibition mechanisms to light as well as novel modes of antitoxicity. However, recent advances in the field profoundly questioned the role of these systems in bacterial physiology, stress response and antimicrobial persistence. This shifted the paradigm of the functions of toxin-antitoxin systems to roles related to interactions between hosts and their mobile genetic elements, such as viral defence or plasmid stability. In this Review, we summarize the recent progress in understanding the biology and evolution of these small genetic elements, and discuss how genomic conflicts could shape the diversification of toxin-antitoxin systems.

Type II toxin-antitoxin (TA) systems are small genetic elements composed of a toxic protein and its cognate antitoxin protein, the latter counteracting the toxicity of the former. While TA systems were initially discovered on plasmids, functioning as addiction modules through a phenomenon called postsegregational killing, they were later shown to be massively present in bacterial chromosomes, often in association with mobile genetic elements. Extensive research has been conducted in recent decades to better understand the physiological roles of these chromosomally encoded modules and to characterize the conditions leading to their activation. The diversity of their proposed roles, ranging from genomic stabilization and abortive phage infection to stress modulation and antibiotic persistence, in conjunction with the poor understanding of TA system regulation, resulted in the generation of simplistic models, often refuted by contradictory results. This review provides an epistemological and critical retrospective on TA modules and highlights fundamental questions concerning their roles and regulations that still remain unanswered. FIG 1 Type II TA systems, postsegregational killing and distribution. (A) Nonviable segregant or postsegregational killing model. TA genes, as well as proteins, are represented in red (toxins) and green (antitoxins).Rectangles denote TA genes encoded on a plasmid, and round shapes denote TA proteins produced from these genes. A TA-encoding plasmid can be lost during division in a way that one of the daughter cells does not inherit a plasmid copy. In these cells, TA proteins cannot be replenished due to the absence of TA genes. Since the antitoxin is degraded while its cognate toxin is stable, the free toxin concentration will increase, exert its activity, and, in time, induce cell death, therefore killing plasmid-free segregants. (B) Distribution of type II TA systems in various E. coli reference strains generated by TAfinder (23). Asterisks indicate systems that were not validated experimentally. Parentheses include name of the prophage a TA is encoded on when applicable. The strains are MG1655 (NCBI U00096.3), a common lab strain from phylogroup A; W (CP002967.1), a soil isolate from phylogroup B1; EDL933 (AE005174.2), an enterohemorrhagic pathogen from phylogroup E; and UTI89 (CP000243.1), a uropathogen from phylogroup B2. No TA systems are conserved within these four distantly related E. coli strains.

Persistence is a reversible and low-frequency phenomenon allowing a subpopulation of a clonal bacterial population to survive antibiotic treatments. Upon removal of the antibiotic, persister cells resume growth and give rise to viable progeny. Type II toxin-antitoxin (TA) systems were assumed to play a key role in the formation of persister cells in Escherichia coli based on the observation that successive deletions of TA systems decreased persistence frequency. In addition, the model proposed that stochastic fluctuations of (p)ppGpp levels are the basis for triggering activation of TA systems. Cells in which TA systems are activated are thought to enter a dormancy state and therefore survive the antibiotic treatment. Using independently constructed strains and newly designed fluorescent reporters, we reassessed the roles of TA modules in persistence both at the population and single-cell levels. Our data confirm that the deletion of 10 TA systems does not affect persistence to ofloxacin or ampicillin. Moreover, microfluidic experiments performed with a strain reporting the induction of the yefM-yoeB TA system allowed the observation of a small number of type II persister cells that resume growth after removal of ampicillin. However, we were unable to establish a correlation between high fluorescence and persistence, since the fluorescence of persister cells was comparable to that of the bulk of the population and none of the cells showing high fluorescence were able to resume growth upon removal of the antibiotic. Altogether, these data show that there is no direct link between induction of TA systems and persistence to antibiotics.

Bacterial persistence refers to the capacity of small subpopulations within clonal populations to tolerate antibiotics. Persisters are thought to originate from dormant cells in which antibiotic targets are less active and cannot be corrupted. Here, we report that in exponentially growing cultures, ofloxacin persisters originate from metabolically active cells: These cells are dividing before the addition of ofloxacin and do endure DNA damages during the treatment, similar to their nonpersister siblings. We observed that growth rate, DNA content, and SOS induction vary among persisters, as in the bulk of the population and therefore do not constitute predictive markers for persistence. Persister cells typically form long polynucleoid filaments and reach maximum SOS induction after removal of ofloxacin. Eventually, cell division resumes, giving rise to a new population. Our findings highlight the heterogeneity of persister cells and therefore the need to analyze these low-frequency phenotypic variants on a case-by-case basis at the single-cell level.

BackgroundEscherichia coli W3110 and a group of six isogenic derivatives, each displaying distinct specific rates of glucose consumption were characterized to determine levels of GFP production and population heterogeneity. These strains have single or combinatory deletions in genes encoding phosphoenolpyruvate:sugar phosphotransferase system (PTS) permeases as PtsG and ManX, as well as common components EI, Hpr protein and EIIA, also the non-PTS Mgl galactose/glucose ABC transporter. They have been transformed for expressing GFP based on a lac-based expression vector, which is subject to bistability.ResultsThese strains displayed specific glucose consumption and growth rates ranging from 1.75 to 0.45 g/g h and 0.54 to 0.16 h−1, respectively. The rate of acetate production was strongly reduced in all mutant strains when compared with W3110/pV21. In bioreactor cultures, wild type W3110/pV21 produced 50.51 mg/L GFP, whereas strains WG/pV21 with inactive PTS IICBGlc and WGM/pV21 with the additional inactivation of PTS IIABMan showed the highest titers of GFP, corresponding to 342 and 438 mg/L, respectively. Moreover, we showed experimentally that bistable expression systems, as lac-based ones, induce strong phenotypic segregation among microbial populations.ConclusionsWe have demonstrated that reduction on glucose consumption rate in E. coli leads to an improvement of GFP production. Furthermore, from the perspective of phenotypic heterogeneity, we observed in this case that heterogeneous systems are also the ones leading to the highest performance. This observation suggests reconsidering the generally accepted proposition stating that phenotypic heterogeneity is generally unwanted in bioprocess applications.Electronic supplementary materialThe online version of this article (10.1186/s12934-019-1073-5) contains supplementary material, which is available to authorized users.