Actin Isoforms in Neuronal Development and Function - International Review of Cell and Molecular Biology - page 170

epitopes, Bassell et al. (1998) reported that β-actin was specifically enriched
in growth cones and present at much lower levels in the developing neurite
shafts. In contrast, γ-actin appeared to be uniformly distributed and was
essentially indistinguishable from phalloidin-stained F-actin ( Bassell et al.,
1998 ). It is important to note, however, that quantitative data was not pre-
sented here. Because growth cones are among the most actin-rich structures
within neurons, and neurites themselves contain relatively little actin, poor
binding of one antibody compared to another could give a similar staining
pattern. Therefore, caution must be taken in interpreting non-normalized
data comparing two different antibodies and concluding that the staining
represents distinct localization patterns of, in this case, β-and γ-actin.
These same antibodies have also been used to stain fixed brain sections
and examine the distribution of β- and γ-actin in vivo. Here again, distinct
localization patterns for β- and γ-actin were reported. In one study, γ-actin
was reported to be expressed throughout cell bodies and neurites while
β-actin expression became progressively restricted to dendritic spines with
age based on immunogold studies ( Micheva et al., 1998 ). A similar restric-
tion in β-actin distribution was reported by another group where β-actin
expression in medulla axons progressively decreased from embryonic stages
to adults, while γ-actin staining remained strong ( Weinberger et al., 1996 ).
Collectively, these studies suggest that β-actin may be restricted to more
dynamic structures such as dendritic spines as development progresses, and
not present in significant amounts in relatively less dynamic structures such
as axon shafts, where γ-actin staining is prominent. Thus, although β-actin
levels may decrease with development, the relative composition of β- to
γ-actin within tightly confined yet dynamic compartments may actually
increase, potentially altering the properties of the underlying actin cytoskel-
eton and influencing neuronal function.
The reports discussed above place β-actin expression and localization
within dynamic neuronal structures, supporting models that β-actin is the
major actin isoform fueling actin dynamics. Nevertheless, some caution is
warranted. First and foremost, the specificity of the antibodies used under
the staining conditions described has not always been demonstrated with
KO tissue. Secondly, there is now evidence that actin isoforms in some
tightly packed filament actin structures may be undetectable with labeled
secondary antibodies ( Perrin et al., 2010 ). Finally, the binding of actin-iso-
form-specific antibodies may also be masked by formaldehyde fixation. We
have found that postfixation with methanol dramatically increases staining
with actin-isoform-specific antibodies, which was also recently confirmed