Biology Reference
In-Depth Information
et al., 1994) and spectrin (Holleran et al., 2001). Since mDia1 is an actin-
binding protein, it might, in principle, bind Arp1. This could provide a
mechanism for a possible link between mDia1 and dynactin, and, via
dynactin, to other microtubule-end-tracking proteins (CLIP-170, LIS1, EB1,
etc.), that may affect microtubule dynamics.
Of course, less direct mechanisms of mDia1 effects on microtubules could
also be envisioned. For example, it was suggested in a recent study that
activation of mDia1 leads to activation of the Rac1 protein (Tsuji et al., 2002).
Rac1 activation could stabilize microtubules via PAK1-mediated phosphor-
ylation (and inactivation) of a tubulin-sequestering protein, Op18/stathmin
(Daub et al., 2001) or via direct effect on IQGAP protein that binds CLIP-170
(Fukata et al., 2002). All these possibilities deserve to be tested experimentally.
Finally, we must consider how the effects of mDia on microtubule dynamics
described above could be integrated into the general scheme of the interplay
between contractile actin cytoskeleton, mechanosensory focal adhesions and
contraction-suppressing microtubules outlined in the previous section. The
first question that should be addressed in this connection is, where and when
does the activation of mDia1 occur? Since the only known mDia1 activator is
Rho, the spatial and temporal distribution of active mDia1 should be similar
to that of active Rho. While no studies have been reported relating to the
distribution of active Rho, there are some data describing gradients of Rac
and Cdc42 activity, obtained by various modifications of the fluorescence
resonance energy transfer (FRET) technique (Gardiner et al., 2002; Itoh et al.,
2002; Kraynov et al., 2000). Rac activation can occur as a result of integrin
signalling (Price et al., 1998), and there is good evidence that formation of the
new cell-ECM adhesions at the cell leading edge is responsible for establishing
the spatial gradient of Rac activity (Del Pozo et al., 2002; Tzima et al., 2002).
Integrin signalling can also activate Rho, although the mechanism is less clear;
this activation depends in a complex way on the integrin type (Danen et al.,
2002; Miao et al., 2002), and is biphasic over time (activation is preceded by a
decay) (Ren et al., 1999). Nevertheless, it is reasonable to suggest that the
integrin-mediated signal at the focal adhesion can at some point activate Rho,
and the gradient of Rho activity (and therefore mDia1 activity) will depend on
focal adhesion distribution.
Based on our model, gradients of mDia1 activity should locally remodel
microtubule behaviour (Figure 5.7). Minus end stabilization will lead to
selective protection of microtubules released from the centrosome. Alterations
in plus end dynamics may increase the average time these ends will spend in
the cell region with high mDia1 activity. Thus, mDia1 should promote an
elevated concentration of microtubules (especially their plus ends) in
proximity to the source of integrin signalling. This, in turn, may trigger
some of the following effects: (1) suppression of actomyosin contractility in
that region of the cell; (2) increase of microtubule-dependent transport to and
