Toxins & Antitoxins: Presentation
Toxin/Antitoxin (TA) modules are generally
made of two adjacent co-oriented but antagonist genes, where one
encodes a stable toxin harmful to an essential cell process, and the
second a labile antitoxin, capable of blocking the toxin's aggressive
behavior by DNA- or protein-binding. The toxic component can be:
- an antisense-RNA; these have until now been
linked to plasmid stabilization by means of a post-segregational killing effect,
and have thus been named PSK loci (for review see Gerdes 1997).
- a fully translated protein. For consistency with previous studies, we
shall refer to them throughout this paper as TA systems.
Discovered in 1983 by Ogura and co-workers, they were first called "plasmid addiction systems"
as they fulfilled plasmid maintenance. Ten years later, chromosomal
homologs were discovered in Escherichia coli (Masuda), renewing interest in them.
This lead to the discovery of new systems in various bacteria, and of their
implication in programmed cell death (PCD). Aizenman even hypothesized in 1996
that under severe starvation conditions TA-mediated PCD of moribund subpopulations
could provide the remaining healthy cells with nutrients, thus benefiting to the specie.
In fact, proof was later established that some of them actually provoked a static state
in some adverse conditions, with cells being still viable but unable to proliferate,
and that this state was fully reversible on cognate antitoxin induction
(Pedersen et al. 2002), however possibly within a certain timeframe
(Amitai et al. 2004).
TA systems are presently distributed in 8 families, depending on their structural
features or modes of action (Gerdes 2005).
- One 3-component family:
- Little is known about the omega-epsilon-zeta (ω-ε-ζ)system from
plasmid pSM19035, except that the additional gene (ω) acts as a repressor
regulating the transcription of the operon (De la Hoz 2000). ω-ε-ζ systems
are only found in Gram-positive bacteria.
- Seven two-component families:
- the ParDE system, found in Gram-negative and Gram-positive bacteria as
well as in archea, targets DNA gyrase (Sobecky 1996).
- HigBA, unique in that its toxin is located upstream of its antitoxin
(Tian 2001), is found in Gram-negative and Gram-positive bacteria, and its
action has recently been shown to be mRNA cleavage (Christensen-Dalsgaard 2006).
- the phd/doc locus, found in all types of prokaryotes, is
believed to inhibit translation (Hazan 2001).
- The vapBC locus, found both on plasmids and chromosomes, seems
to be the highest copy-number TA system among prokaryotes which bare them, but has no
known cellular target for the moment, although VapC toxins contain a PIN
domain (ribonuclease involved in nonsense-mediated mRNA decay and RNA
interference in eukaryotes), making the system a possible quality
control of gene expression agent (Anantharaman 2003).
- The ccdAB locus, found only in some Gram-negative bacteria, fulfils plasmid stabilization at
the replication level by targeting DNA gyrase (Bernard 1993).
- the RelBE members, present in Gram-negative, Gram-positive and archea, inhibit
cell-growth by impairing translation on mRNA cleavage through the A-site
of the ribosome (Christensen 2003; Pedersen 2003).
- Finally, the MazEF/PemIK family's toxins, sometimes referred to as "RNA
interferases" (Zhang 2005), are ribonucleases that cleave cellular
mRNA, thus depriving the ribosomes with substrates to translate (Zhang
2003). They have been identified in Gram-negative and Gram-positive
Chromosome-borne TA systems are activated by various extreme conditions, such as:
- presence of antibiotics (Sat 2001)
- presence of infecting phages (Hazan 2004a)
- thymine starvation or other DNA damage (Sat 2003)
- high temperatures (Hazan 2004b)
- oxidative stress (Hazan 2004b)
- response to amino acid starvation (Hayes 2003)
Indeed, TA modules are believed to fulfill a backup system to the stringent
response during stasis. As a reduced translational
rate means less translational errors, TA loci most likely function in
quality control of gene expression, helping the cells cope with
nutritional stress (Gerdes 2005). Therefore, it remains a priority to
exhaustively identify TA loci in prokaryotic organisms, in order to
improve our comprehension of them and more broadly of the cellular
mechanisms behind bacterial adaptation.
TA systems, which are widespread among both bacteria and archeae, thus reveal
to be of high importance to the prokaryotic world. Their role in programmed cell
death moreover constitutes a promising target for the design of a new class of
antibiotics. Previous studies have searched for TA in completely sequenced
genomes (archaea and eubacteria) based on standard sequence alignment tools (BLASTP and
TBLASTN). To overcome the limitation of such methods, and because we are
concerned by the lack of annotation of small ORFs, we developed a simple method
for identifying all potential TA systems in any given bacterial genome:
Rapid Automated Scan for Toxins and Antitoxins
in Bacteria (RASTA-Bacteria).
Our tool not only relies on the existence of conserved functional domains in
toxins and antitoxins, but also takes into account the genomic features of these genes.
We hope our tool's predictions will help scientists increase the knowledge of prokaryotic behaviour.
For further information, please write to Emeric Sevin
or Frédérique Hubler.
Disclaimer: as of October 2010, and until at least end of 2011,
insufficient credits/mantime have been allocated to RASTA-Bacteria, which is not maintained anymore.
If you have any questions, you can try writing to Emeric Sevin
but with no guarantee that you'll get an answer soon. We apologize for this situation, and thank
you for your understanding.