Mechanisms Underlying the Initiation and Dynamics of Neuronal Filopodia: From Neurite Formation to Synaptogenesis - International Review of Cell and Molecular Biology - page 105

of axons extending linearly (e.g. on top of previously guided axons) tend
to have relatively simple morphologies characterized by few filopodia and
lamellae. In contrast, when growth cones arrive at regions of the developing
nervous system where their trajectory must change significantly, they often
develop more complex morphologies ( Tosney and Landmesser, 1985 ). The
elaboration of growth cone morphology at these “decision points” may
represent an attempt by the growth cone to increase its ability to sample the
environment and arrive at a guidance decision.
Although a role for filopodia in growth cone guidance has been sup-
ported by in vitro and in vivo studies, a study investigating the effects of
inhibiting Ena/VASP protein function in developing Xenopus retinal gan-
glion cell axons in vivo has indicated that filopodia may not be obligatory
( Dwivedy et al., 2007 ). In Xenopus , retinal ganglion cells project their axons
into the developing optic tectum. Along the path, the growth cone has to
make guidance decisions at multiple points. Surprisingly, depletion of Ena/
VASP proteins, which decreases the mean number of growth cone filopodia
by 70-90% depending on the location of the growth cone along its path to
the tectum, did not impair the ability of axons to ultimately arrive at their
target without making overt guidance errors (i.e. projecting to inappropri-
ate regions along their path). However, the net rate of axon advance was
decreased at multiple points along the path to the tectum, and the decrease
was due to increased time spent stalling without net advance. While one
interpretation of these results is that growth cones with greatly diminished
numbers of filopodia, but not completely devoid of filopodia as noted above,
can ultimately pathfind correctly, an alternative interpretation could also
be considered. As noted previously, growth cones often stall at “decision”
points along their path and elaborate filopodia and lamellipodia. Presum-
ably, this increase in growth cone morphologic complexity reflects periods
during which the growth cone is sampling its environment using protrusive
structures in order to parse the guidance information in its environment.
Therefore, when the number of growth cone filopodia is decreased, but
not abolished, by Ena/VASP knock down, growth cones may be taking
more time to make filopodia-dependent guidance decisions (as noted in
Dwivedy et al., 2007 ). Indeed, Ena/Vasp depletion increased the stalling of
growth cones in vivo. Determination of the independence of growth cone
guidance from filopodia would therefore benefit from additional investiga-
tions in which all filopodia are abolished at growth cone, perhaps through
concerted depletion of multiple pathways that mediate filopodia forma-
tion. The depletion of Ena/VASP did not abolish lamellipodia formation at