Biology Reference
In-Depth Information
are differentially expressed by peripherally-injured (regenerating) versus
centrally injured (nonregenerating) DRGs, of which 30 were transcription
factors (
Stam et al., 2007
). Similarly,
Zou et al. (2009)
identified several
hundred genes regulated in injured DRGs in the first day following injury,
including 19 transcription factors (
Zou et al., 2009
). Other groups have
used bioinformatics approaches to detect transcription factors whose
activity
,
but not necessarily abundance, might regulate PNS regeneration. This
approach is motivated by the high degree of regulation of transcription factor
activity by posttranslational modification and binding partners (
Moore &
Goldberg, 2011
).
Smith et al. (2011)
assembled a list of genes that are
enriched in DRG neurons compared to cerebellar granular neurons (CGNs)
using both subtractive hybridization and microarray data mining. Genomic
sequences upstream of these DRG-enriched genes were then scanned for
known transcription factor binding sites, with overrepresentation of sites
that bind a particular transcription factor being indicative of high activity.
By combining this binding site analysis with a network analysis based on
known protein-protein interactions, the authors assembled a list of 32 tran-
scription factors predicted to be more active in DRG neurons. Finally, a
study by
Michaelevski et al. (2010)
examined the early signaling events that
initiate the regenerative cell body response in injured DRG neurons. Focus-
ing on time points between 1 and 28 h postinjury, this study combined mass
spectrometry analysis of axonal proteins with microarray analysis of DRG
gene expression, identifying 900 axonal proteins, 2500 phosphorylation
sites, and 2700 transcripts that changed in abundance. An analysis of
potential transcription factor binding sites, similar in principle to the one
described earlier, was combined with analysis of potential protein interac-
tions to generate a list of 39 transcription factors that were common to both
analyses, and thus candidates to regulate the regenerative response.
How do these four analyses compare? The
Stam et al. (2007)
and
Zou et al. (2009)
datasets, both of which derive directly from microarray
expression data, share three transcription factors (c-Jun, Atf3, and Isl2)
and identify related members of two more families (Sox and KLFs)
(
Fig. 3.1
). The datasets from
Smith et al. (2011)
and
Michaelevski et al.
(2010)
, which used separate bioinformatics-based approaches, share six
identical or related transcription factors, or about 20% of each list. Looking
at all four sets in aggregate, c-Jun is common to all four; Atf3 and members of
the KLF and Smad families are found in three; and Yy1, Fos, Egr1, Isl2, and
members of the Stat, Sox, and Gata families are shared by two datasets. It is
encouraging that
these
separate
studies have
identified common
