Biology Reference
In-Depth Information
localization for Dyn2 (Okamoto et al., 2001) where it may play a role during
glutamate receptor endocytosis (Carroll et al., 1999). As very little work has
focused on the testes- and brain-specific dynamin, Dyn3, it is important to
understand whether this third isoform performs distinct functions in the
neuron. Recently, we have provided evidence implicating Dyn3 in the actin-
dependent process of filopodial outgrowth and dendritic spine maturation
(Gray et al., 2003).
In this study we found that Dyn3 spliced-forms induced differential effects
on the dendritic spine morphology of transfected cultured neurons. Dendritic
spines, small actin-rich protrusions that decorate the dendrites of excitatory
pyramidal neurons in the hippocampus and cortex, usually emerge after 2-3
weeks in culture, replacing the filopodia that populate the dendrites prior to
complete synaptogenesis. When overexpressed in immature neurons, the
Dyn3baa spliced-form caused a proliferation of dendritic filopodia at the
expense of mushroom-shaped dendritic spines, even in temporally mature
cultures (Figure 12.1D). Contrasting with this, the expression of a second
Dyn3 spliced-form, Dyn3aaa, did not have any effect on the normal
maturation process of the dendrites and spines (Figure 12.1D, inset).
Interestingly, these spliced-forms differ by only 10 amino acids, a cassette
that is spliced into Dyn3baa, but is lacking in Dyn3aaa. These Dyn3baa-
induced filopodia did not contain any postsynaptic markers, nor did they
associate with the presynaptic terminals of non-transfected neurons,
indicating that these dendritic protrusions are morphologically and function-
ally immature. The GTPase activity of Dyn3baa was necessary to induce
filopodia, as a dominant-negative mutant (Dyn3baaK44A) did not have the
same effect on transfected neurons but actually had the opposite effect of
causing an early development of mushroom-shaped spines (N.W.G. and
M.A.M., unpublished observations). These results are consistent with studies
in non-neuronal cells that demonstrate that GTPase-deficient dynamin affects
actin-dependent membrane processes, including podosome dynamics (Ochoa
et al., 2000) and actin comet formation and movement (Orth et al., 2002; Lee
and De Camilli, 2002).
The concept of a postsynaptic dynamin regulating the growth and
development of the dendrites is an intriguing proposition. The study that
identified Dyn3 as a postsynaptic dynamin also found this isoform to bind to a
biochemical complex containing other postsynaptic proteins, such as the
scaffolding molecule Homer and the glutamate receptor mGluR5 (Gray et al.,
2003). Overexpression of Homer leads to an accelerated maturation of
dendritic spines, with these protrusions becoming more functionally and
synaptically active weeks before they normally do (Sala et al., 2001). This was
attributed to the possibility that as a scaffolding molecule, an abundance of
Homer could provide more opportunities to 'capture' postsynaptic protein
and allow for the early development of the postsynaptic density (PSD). Thus,
