Biology Reference
In-Depth Information
Arp2/3 complex-independent actin nucleation
Whilst the WASp-Arp2/3 pathway of actin filament nucleation is now
becoming fairly well documented, the possibility remains for actin filament
nucleation by alternative mechanisms. It is suggested that although the Arp2/3
complex is responsible for nucleation of actin in the formation of branched
microfilaments (Mullins et al., 1998), a separate mechanism may exist for the
production of unbranched filaments such as those found in stress fibres.
Current research has identified additional proteins capable of nucleating
unbranched actin polymerization from barbed ends.
The L. monocytogenes protein ActA has been known to be involved in the
intracellular motility of the bacterium for a long time, recruiting actin-binding
proteins to enable actin tail formation (Loisel et al., 1999). Recently Fradelizi et
al. (2001) showed that ActA, and the related human protein zyxin, which
localizes to stress fibres and focal contacts, are capable of nucleating actin
polymerization in cell-free and permeabilized cell systems. Actin assembles on
ActA-coated beads as well as on ActA and zyxin targeted to mitochondria. This
actin polymerization is dependent upon VASP in the case of zyxin, but
independent of the Arp2/3 complex in both cases. Currently no mechanisms for
control of this potential actin nucleation system have been described.
Two groups have recently implicated the budding yeast formin Bni1 as an
actin nucleator (Pruyne et al., 2002; Sagot et al., 2002b). This formin,
previously identified as being required for polarized arrays of actin cables in
budding yeast (Evangelista et al., 2002; Sagot et al., 2002a), is thought to be
involved in the formation of long unbranched actin filaments which are
organized into bundles. A purified constitutively active C-terminal construct
of Bni1 containing the formin homology 1 (FH1) and 2 (FH2) domains
(Figure 8.3) causes a concentration-dependent increase in the rate of in vitro
actin filament assembly. Removal or mutation of the FH2 domain but not the
FH1 domain inhibits this apparent nucleating activity (Sagot et al., 2002b),
and addition of profilin shortens the lag time of polymerization. The effect of
profilin is dependent upon its capacity to bind to both actin and the
polyproline-rich FH1 domain of Bni1 (Sagot et al., 2002b).
Several formins are Rho GTPase effectors, believed to be activated by the
relief of an intramolecular interaction between a GTPase binding domain
(GBD) and an auto-regulatory C-terminal domain named DAD (Dia
autoregulatory domain) upon binding of an activated GTPase (Figure 8.3)
(Alberts, 2001). Specifically, the GTPases Rho1p and Cdc42p, interact with
Bni1 at the N-terminal GBD. This causes the formin to adopt an open or
active conformation in a similar manner to WASp (Evangelista et al., 1997;
Kohno et al., 1996). Activation of formins allows interaction with a variety of
binding partners at either the FH1 or FH2 domains to facilitate actin filament
nucleation.
