Actin Isoforms in Neuronal Development and Function - International Review of Cell and Molecular Biology - page 194

and assembly of snRNPs mediating mRNA splicing ( Monani, 2005 ). Why
decreased levels of SMN leads selectively to the loss of motor neurons is a
question that has plagued the field for decades. One intriguing hypothesis
is that SMN may have unique and essential functions in motor neurons,
potentially beyond its classical role in the nucleus. Evidence supporting this
hypothesis emerged when SMN protein was identified throughout neuro-
nal axons and growth cones, where it exhibited fast bidirectional motility
( Fan and Simard, 2002 ; Pagliardini et al., 2000 ; Tizzano et al., 1998 ; Zhang
et al., 2003 ). Additionally, knockdown of SMN in a number of neuronal
cell lines and in zebrafish resulted in decreased neurite outgrowth and axon
branching abnormalities, raising the possibility that SMN somehow regu-
lates the functions of the neuronal cytoskeleton underlying neurite out-
growth and guidance ( Bowerman et al., 2007 ; McWhorter et al., 2003 ;
Oprea et al., 2008 ; Tadesse et al., 2008 ; van Bergeijk et al., 2007 ).
Studies by two independent groups revealed a potential link between
SMN and the neuronal cytoskeleton when it was discovered that SMN
interacts with at least two proteins known to bind the zipcode sequence of
β-actin mRNA: KSRP, the human homolog of ZBP2 ( Tadesse et al., 2008 ),
and hnRNP-R ( Rossoll et al., 2002 , 2003 ). Because SMN undergoes bidi-
rectional axonal transport, a tantalizing hypothesis is that SMN functions
within a ribonucleoprotein complex similar to ZBP1 and facilitates the
shuttling of β-actin mRNA to motor neuron axons and growth cones.
Data supporting this hypothesis emerged when cultured primary motor
neurons from an SMA mouse model showed decreased β-actin protein in
the growth cone which correlated with deficits in axon outgrowth, growth
cone size, and β-actin mRNA localization ( Rossoll et al., 2003 ). Yet, evi-
dence supporting a critical role for SMN-mediated β-actin localization in
SMA mouse models in vivo has been rare.
While there is one report of motor axons “overshooting” the motor
endplate in the diaphragm muscle of an SMA mouse model ( Kariya et al.,
2009 ), most studies have not found any motor axon guidance defects in
multiple SMA mouse models ( Kariya et al., 2008 ; McGovern et al., 2008 ;
Murray et al., 2010 ). Recent reports suggest instead that motor axon growth
and guidance is normal in SMA mouse models, but that neuromuscular
junction (NMJ) development or function is impaired. Further characteriza-
tion of these mouse models has revealed NMJs with an abnormal, simplified
morphology, and impaired synaptic vesicle release from motor axon nerve
terminals ( Cifuentes-Diaz et al., 2002 ; Kariya et al., 2008 ; Kong et al., 2009 ;
McGovern et al., 2008 ; Murray et al., 2008 ). Because NMJ development