Biology Reference
In-Depth Information
complex by an adaptor bound via its GPR loop offers the attractive possibility
that a comparable mechanism is active in higher eukaryotes where the PRD of
class I myosins and other linker proteins might do the job. Finally, beside
these three ways of linking to the Arp2/3 complex, class I myosins make use of
two different ways to concentrate G-actin, the fuel for F-actin elongation.
Class I myosins either bind profilin-actin through a Pro-rich domain (MyoK
GPR loop) or bind G-actin indirectly via the WH2 domains of the adaptor
proteins WASp, CARMIL and Vrp1p (Paunola et al., 2002). Concentrating
monomeric actin is not essential for actin-based motility (Loisel et al., 1999)
but greatly enhances the eciency and speed of the process (Zalevsky et al.,
2001; Machner et al., 2001; Cossart, 2000; Geese et al., 2000; Loisel et al.,
1999).
It is attractive to speculate that actin filament assembly might involve
clusters of membrane-anchored class I myosins and bound adaptors that link
to the Arp2/3 complex. In such a position, class I myosins could stay
associated both with the plasma membrane and with the barbed end of actin
filaments, which usually point out to the cell periphery, and could facilitate
addition of actin monomers. They might also be targeted to spots of actin
dynamics/endocytosis through the complex interaction networks mentioned
above and thereby increase tremendously the local concentration of both
nucleation, elongation and capping activities, favouring a highly dynamic and
branched meshwork.
Alternatively, myosins could transport Arp2/3 complexes towards the
barbed ends of actin filaments, accounting for the apparent requirement for
myosin I motor activity in actin assembly in the reconstituted system of
Lechler et al. (2000). In the light of such a model, it is intriguing to note that
in amoeba the complex might be kept at the pointed end of a very short and
capped actin filament (Jung et al., 2001). They might also capture Arp2/3
complexes at the rear of the ageing actin meshwork, where it is
depolymerized, and recycle the nucleation activity to the front of the
advancing polymerization zone. Such models potentially suffer from two
conceptual problems: the lack of processivity of myosin I motors and the
concurrent necessity for an uncapping activity (Falet et al., 2002). It is well
known that class I myosins are not processive because, unlike kinesin
motors, they spend most of their time dissociated from the actin filament,
and Ostap and Pollard (1996) predicted that clusters of 20 myosin I
molecules would be needed for processive motility. On the other hand, a
complex consisting for example of Las17p, Vrp1p and the Arp2/3 complex
may be able to cluster su
cient myosin I molecules to support processive
movement along an actin filament (Machesky, 2000). If class I myosins are
used to recruit the Arp2/3 complex at its site of action, then the role of the
WASp family proteins in this model is unclear. One possibility is that WASp
family proteins are also transported there by class I myosins, or otherwise
