Embryonic Stem Cells as a Module for Systems Biology - Systems Biology: Networks, Models, and Applications - page 311

cells (unpublished data). This switch from a differentiated-cell tran-
scriptome to that of an ES cell involves a reactivation of ES-specific
genes and requires silencing of non-ES cell genes. To date, no transcrip-
tion factors have been shown to be able to convert a differentiated cell
to an ES-like cell or the reactivation of specific ES genes. As discussed
above, Oct4 , Sox2 , and Nanog form a key trinity of ES transcriptional
factors essential for ES state and during reprogramming these must be
reactivated. It is plausible that several transcription activators may be
required. Interestingly, the overexpression of Oct4 in non-ES cells does
not appear to elicit reprogramming in cell lines, but does lead to tumor
induction in animal models [59]. This does suggest that certain popu-
lations of uncommitted cells in different niches within an adult animal
are responsive to the reintroduction of Oct4 protein.
Reprogramming by ncRNAs
The recent demonstration that double-stranded RNA targeted against
complementary DNA segments can induce methylation of CpG residues
leading to a reduction in transcription or complete gene silencing further
increases the scope of the role of noncoding RNAs in regulating the
transcriptome [60-62] Indeed, naturally occurring microRNAs in plants
have been shown to direct this mechanism toward silencing gene
regions that act as transposon elements, shutting down their ability to
excise and recombine. Whether naturally occurring microRNAs or
other ncRNAs are involved in epigenetic regulation and modulating
silencing of transcription of coding genes remains to be demonstrated.
If verified, the incorporation of this network of regulatory RNAs into
the conventional transcriptional network will present yet another level
of complexity.
CONCLUSIONS
Major advances in genomics technology continue to drive the recovery
of large biological data sets. This windfall of information presents, for
the first time, possibilities of breaking new ground in biological under-
standing based on a comprehensive global consideration of the relation
of networks of genes. The endpoint is to arrive at unifying principles
underlying the dynamics of networks of molecules that will allow com-
putational prediction of biochemical and cellular responses to specific
perturbations and signals. ES cells offer a unique and versatile stem cell
model system that allows derivation of transcriptome and proteome
data from cells to tissues and whole organism. By generating predictive
models using ES cells we should be able to test, improve and, hope-
fully, apply such systems strategies in research on tissue engineering,
pathogenesis of developmental disorders and other diseases, and tar-
gets for therapeutics and gene therapy.