Biology Reference
In-Depth Information
(LetD) toxin inactivates the host gyrase (topoisomerase II) by trapping it in an inactive DNA
complex and preventing it from supercoiling the chromosomal DNA. A CcdB dimer binds to the
central cavity of the
-terminal portion of the gyrase subunit A. If no antitoxin is present, cell
growth is inhibited, causing Ýlamentation, and the affected cell will eventually die. The antitoxin
CcdA (LetA) displaces CcdB from the gyrase by forming a CcdAÏCcdB complex, which detaches
from the gyrase and restores its activity. The ParE toxin of the parDE system of the broad host-
range plasmid RK2/RK2 is also a gyrase inhibitor. The F plasmid carries two additional addiction
systems,
N
srn
(stable RNA degradation) and
Þm
(F leading maintenance), which function as inde-
pendent killing systems.
The Kid (killing determinant) and PemK (plasmid emergency maintenance killer) toxins of
plasmids R1 and R100, which turned out to be identical, are part of another well-studied addiction
module (Santos-Sierra et al., 2002). These toxins are inhibitors of DnaB, a crucial enzyme in DNA-
chain elongation, and as such prevent the initiation of chromosomal-DNA replication. This will
not kill the cell instantly but will preclude any further cell divisions, and the lineage will eventually
be lost accidentally. The antitoxins Kis (killing suppression) and PemI (pem inhibitor) form
complexes with their cognate toxins and thus neutralize their toxicity.
The
pas
(plasmid addiction system) system on the broad host-range plasmid pTF-FC2 originally
found in
is special in that it consists of three, rather than two, components
(Rawlings, 1999). The third protein functions as an enhancer of toxinÏantitoxin neutralization and
in this way decreases the Ýtness costs of the system for its carrier. It is an interesting example of
how a selÝsh element adapts to its host.
The mobility aspect of addiction systems becomes evident in the
Thiobacillus ferrooxidans
bacteriophage P1,
which carries the phd/doc (prevents host death/death on curing) module. The precise cellular target
of Doc is not yet known, but death occurs by cell-wide inhibition of protein synthesis. However,
there is now evidence that this phage-borne addiction system acts through a chromosomal addiction
system of
E. coli
. The free Doc toxin inhibits the translation of the chromosomal MazE
antitoxin, and this in turn leads to free MazF toxin, which either kills the cell directly or may
initiate another Ñdeath cascadeÒ (Hazan et al., 2001).
In yet another type of plasmid- and chromosome-encoded modiÝcationÏrescue systems, the
rescue factors are short, unstable, untranslated antisense RNAs of only some 60 nucleotides that
inhibit the translation of very stable modiÝcation-factor-encoding mRNAs. In the best-studied
system,
E. coli
,
mazEF
(host killing) gene codes for a
small hydrophobic toxin of 52 amino acids that presumably integrates into the cytoplasmic mem-
brane with the C-terminus protruding into the periplasm (Gerdes et al., 1997). This and many related
toxins of the Hok family functionally resemble a group of bacteriophage-encoded proteins known
as holins that create pores in the cytoplasmic membrane. The
hoh/sok
on plasmid AD1 of
Enterococcus faecalis
, the
hok
gene (suppression of killing)
constitutes the antitoxin. It codes for an antisense RNA that is complementary to the
sok
hok
mRNA
leader region. Like the
pas
system, a third gene,
mok
(modulation of killing), regulates
hok
translation.
mRNA is stable and constitutively expressed from a weak promoter, whereas sok
mRNA is unstable and expressed from a strong promoter. This system is quite different from the
proteic addiction systems discussed earlier. The proteic systems are characterized by a bicistronic
organization whereby one promoter controls the transcription of both toxin and antitoxin, in that
precise order. The only exception to this rule so far is the
hok
hig
(host inhibition of growth) addiction
system on the Rts 1 plasmid of
Escherichia coli
. The toxin gene is upstream of the antitoxin gene,
and the
locus contains two promoters (Tian et al., 2001). The disequilibrium between toxin and
antitoxin in the
hig
system requires an additional dimension compared to the proteic systems
in order to work. The toxin gene
hok/sok
is transcribed as a full-length mRNA. Because of its secondary
structure, the full-length form hides the binding site for the short, corresponding antisense RNA,
the
hok
antitoxin gene. The secondary structure also prevents translation of the mRNA. The full-
length mRNA is only slowly truncated at the 3
sok
end by exonucleases. This leads to an accumulation
of the full-length mRNA in the cell. The truncation triggers a refolding of the mRNA into a
