Healthcare and Medicine Reference
In-Depth Information
( cont'd )
BOX 3
called Tiam1 (Kunda et al., 2001; Fukata et al., 2002).
When either of these proteins is overexpressed in a devel-
oping hippocampal cell, all the processes become axons;
when their function is reduced, all the processes become
dendrites. It is thought that these proteins help micro-
tubules invade actin networks. External influences
also affect the axonal versus dendritic decision perhaps
by affecting these cytoskeletal-associated proteins. For
instance, if a hippocampal cell is plated at the interface
between a laminin and polylysine, the axon almost invari-
ably grows on the laminin substrate.
Dendritic growth-promoting factors can also be found
in the environment. Dendritic outgrowth from mouse
cortical neurons was specifically enhanced by astrocytes
derived from the forebrain (Le Roux and Reh, 1994).
Similar results were obtained with glial conditioned
medium. Superior cervical ganglion neurons grown in
serum-free medium in the absence of nonneuronal cells
were unipolar and only grew axons. When the same
neurons were exposed to serum, they became multipolar
and developed processes that could be categorized as
dendrites by morphological and antigenic criteria (Bruck-
enstein and Higgins, 1988a,b). Thus, serum contains
factors that stimulate dendritic extension. The bone mor-
phogenetic proteins, BMP2 and BMP6, were subsequently
found to selectively induce the formation of dendrites
and the expression of microtubule-associated protein-2
(MAP2) in sympathetic neurons in a concentration-
dependent manner (Lein et al., 1995; Guo et al., 1998).
Dermatan sulfate also specifically enhances dendritic
growth. Dendritic retraction occurs in many regions of the
developing brain. Leukemia inhibitory factor (LIF) and
ciliary neurotrophic factor (CNTF) specifically cause den-
dritic retraction in SCG cells (Guo et al., 1997; Guo et al.,
1999). Axon growth is unaffected by these factors. Taken
together, these results suggest the existence of separate
but extensive molecular mechanisms for promoting and
inhibiting dendrite outgrowth that parallel the growth-
promoting and collapsing mechanisms known to be
involved in axonogenesis.
It has been thought that mature dendrite formation is
somehow dependent on the axon making proper connec-
tions to its target. The situation, however, may not be so
one-sided as it seems that dendrites can also “search” for
their inputs (Jan and Jan, 2003). In some cases, growing
dendrites are tipped with dendritic growth cones that
appear as miniature equivalents to axonal growth cones.
Dendritic growth cones express receptors for several
classes of guidance factors and indeed, factors such as
Slits, Netrins, and Semaphorins can influence the growth
of dendrites (Whitford et al., 2002). Interestingly the same
factors that do one thing in axons can do a different thing
in dendrites. For example, Sema3A can attract the apical
dendrites of cortical neurons toward the pial surface,
while the same factor can repel the axons of the same
pyramidal cells.
Even though the final size, shape, and complexity of
the dendritic tree are sensitive to innervation, the dendrite
is able to develop in a largely autonomous fashion. One
of the interesting examples of the independence of den-
dritic growth from innervation is the case of the Purkinje
cells in weaver mutant mice. In these mice, the granule
cells do not migrate properly into the cortex of the cere-
bellum and thus fail to make synapses on the Purkinje
cells. The Purkinje cells nevertheless make a dendritic tree
that, although smaller and less well formed than the trees
of properly innervated Purkinje cells, are still characteris-
tically complex (Bradley and Berry, 1978). The most
dramatic demonstrations of the ability of neurons in the
absence of synaptic input to produce dendritic trees come
from culture experiments. Conditions have been worked
out in which pyramidal neurons from the hippocampus,
principal neurons of the SCG, and even cerebellar Purk-
inje cells are able to develop a characteristic dendritic tree
in dissociated cell culture.
Each type of neuron has a characteristic dendritic tree.
In some neurons, like the Purkinje cell, the dendritic tree
is enormously complex and supports synaptic input from
thousands of presynaptic fibers. In other neurons, like
some sensory neurons, the dendritic tree is simple, con-
sisting of a single postsynaptic process. In the central
nervous system of a cockroach or a leech, the dendritic
tree of each identified cell has a unique signature branch-
ing pattern recognizable from individual to individual.
What drives these particular morphologies? The Rho
family of GTPases is critical for proper dendritic mor-
phology. These GTPases are key regulators of actin poly-
merization. Cdc42 is required for multiple aspects of
dendritic morphogenesis (Luo, 2000, 2002). For example,
in neurons that are mutant for Cdc42, dendrites are longer
than normal, branch abnormally, and have a reduced
number of spines. Extra activation of RhoA, via experi-
mental expression of a constitutively active form of the
molecule, leads to a dramatic simplification of dendritic
branch patterns. Such experiments suggest that different
Rho family members have distinct roles in regulating
dendritic morphogenesis. Several transcription factors,
identified in Drosophila , appear to dramatically regulate
Search Pocayo ::




Custom Search