Biology Reference
In-Depth Information
inhibition VCA first binds actin and then the Arp2/3 complex. Although the
Arp2/3 complex has only micromolar a
nity for the sides of actin filaments,
the presence of filaments increases the a
nity of the Arp2/3 complex for the
VCA domain. Thus VCA binding will increase the a
nity of the Arp2/3
complex for actin filaments. This thermodynamic coupling suggests that
VCA and filaments favour the same active conformation of the Arp2/3
complex.
Although we had evidence that the Arp2/3 complex binds to and forms
branches on the sides of existing filaments (Mullins et al., 1998a; Blanchoin et
al., 2000a), Pantaloni et al. (2000) suggested that the branches actually form at
the barbed end of growing filaments, with one of the Arps incorporated into
the mother filament and one into the daughter filament. Direct observation of
branching by total internal reflection microscopy (Amann and Pollard, 2001b;
Fujiwara et al., 2002) and confocal microscopy (Ichetovkin et al., 2002)
confirms that branches form on the sides of mother filaments. These real-time
assays also confirmed the observation of static samples (Amann and Pollard,
2001a) that branching is favoured on newly polymerized filaments. The
mechanism of this bias toward new filaments is not established, but is likely to
be related to nucleotide hydrolysis and/or a structural change in the mother
filament as it ages.
We do not yet know how the nucleation-promoting factors are organized in
cells, but most of their activators of WASp and N-WASp are associated with
membranes: PIP
2
is part of the lipid bilayer; Cdc42 is tethered to the bilayer by
a prenyl group; and Grb2 associates with active receptor tyrosine kinases.
Thus the active fraction of WASp (and perhaps Scar) proteins is expected to
be associated with the plasma membrane, making the inner surface of the
plasma membrane a favoured site for activating the Arp2/3 complex. In fact,
when PIP
2
and active Cdc42 are incorporated into small lipid vesicles, a huge
cloud of branched actin filaments grows from their surface (Higgs and
Pollard, 2000). A bias toward branching from newly formed filaments would
also favour nucleation near the plasma membrane where growing filaments
interact with the membrane.
A key unresolved point is why branches are so much more stable than the
low a
nity binding of the Arp2/3 complex to the sides of actin filaments
(Mullins et al., 1998b). The branch half-life in vitro is about 6min (Blanchoin
et al., 2000b) and can be prolonged by inhibiting the release of phosphate
from ADP-P
i
subunits in the branch with phalloidin or using the stably bound
phosphate analogue, BeF
3
. On the other hand, ADF/cofilin promotes both the
dissociation of the g-phosphate and dissociation of branches. Thus, phosphate
release from subunits in the branch seems to trigger dissociation of the pointed
end of the daughter filament from Arp2/3 complex. Insight into this matter is
likely to require a high-resolution model of branches from cryo-electron
microscopy.
