P1: IwX
052182091Xc10.xml CB786/Lax 0 521 82091 X November 4, 2005 5:8
228
alistair j lax
of signals, both from soluble growth factors that interact with the cell via
G-protein coupled receptors and receptor tyrosine kinases, and from cell–cell
contact mechanisms. Thus, toxins that attack Rho can produce many differ-
ent cellular effects. While conventional toxins that target Rho either activate
or inactivate its function, some type III delivered effectors act to mimic the
function of normal cellular regulators such as Rho GEFs. These latter toxins
transiently activate signalling and thus can induce more subtle subversion of
signalling.
Some toxins have recently been shown to affect regulation of the cell
cycle and the linked process of apoptosis. In this case, however, much less is
known about the molecular mechanisms involved.
Despite over a century of work on bacterial toxins, in particular those
encoded by the well-known major pathogens, new toxins are still being iden-
tified. Indeed, new toxins are still being discovered from well-characterised
bacteria, such as the recently described Cif effector from E. coli (March
`
es et al.,
2003). There remains a wealth of possibilities for the discovery of novel toxic
activities, particularly from bacteria not viewed as mainstream pathogens or
from bacteria not yet identified. In addition, homology searching of sequence
data will identify toxins related to known toxins. As more completed bacterial
genome sequences become available, homology searching is likely to lead to
further identification of toxins.
While many of these “new” toxins are likely to fall into existing categories
of toxin action, it is also likely that new mechanisms will be discovered, both in
terms of targets and the induced chemical modification. In a similar manner
to the effector toxins found to act non-covalently on Rho, other bacterial
effectors may act on other signalling components to mimic the action of
natural eukaryotic effector molecules. Likewise, it is to be expected that other
toxins will be identified that act as mitogens, using either a similar set of
molecular targets as the Pasteurella multocida toxin (PMT), or entirely different
ones. There are also likely to be more toxins that target the machinery of the
cell cycle. Such toxins are likely to be valuable tools for the analysis of cell
function.
Although the thrust of much recent research has been on the molecular
mode of action of toxins, there is clearly a need to understand toxin action
in vivo. Some toxins have been suggested to play no part in pathogenesis.
One example is the dermonecrotic toxin of Bordetella species. Mutants in
this toxin were shown to display an identical LD
50
to wild-type bacteria in a
murine challenge model (Weiss and Goodwin, 1989). However, this toxin is
highly conserved across all the Bordetella species and is regulated by the bvg
His-Asp phosphorelay system, and it thus seems unlikely that the bacteria