Embryonic Stem Cells as a Module for Systems Biology - Systems Biology: Networks, Models, and Applications - page 307

SEQUENCE - TRANSCRIPTION BINDING SITE DISCOVERY BY MICROARRAY AND

BASED ANALYSIS OF C h IP

The miniature circuitry described above marks the activities of master
regulatory genes at the top of the hierarchy of the ES gene regulatory
network. An obvious and important question is how then does this cir-
cuit reach out to other genes to generate the larger networks that must
be brought into play to orchestrate self-renewal, proliferation, and
decisions to differentiate of ES cells into the three primary embryonic
lineages of mesoderm, ectoderm, and mesoderm? Here we illustrate
how once again the availability of ES as a stem cell line together with
new powerful technologies driven by a systems biology approach can
accelerate the clarification of these fundamental questions.
Based on the similar principle of microarray probes for a multiplex
parallel screen of transcriptome, one high-throughput approach is to use
chromatin-immunoprecipitated DNAs as probes to hybridize against
arrayed nucleotide fragments scanning entire chromosomes. Such “ChIP
on CHIP” platforms have been used successfully for whole genome
localization analysis in yeast [35]. This technology is still not readily
available or accessible for robust application in mammalian cells due to
the large size and complexity of the mammalian genomes. Nevertheless,
small-scale ChIP on CHIP analysis in mammalian cells has been applied
to specific areas, such as identifying CpG islands or flanking sequences
associated with transcription start sites, and small chromosome arrays
[36,37]. Significantly, these partial arrays have shown that a transcription
factor can bind to a large number of sites not just in the promoter but to
introns, in distal locations from genes and in genome “desert” regions.
However, as the amount of ChIP DNA needed for such studies is sub-
stantial, requiring a huge number of homogenous stem cell populations,
only ES cells can provide such an inexhaustible supply of material.
A beautiful example of this is the recent work by Boyer et al. show-
ing the mapping of the binding sites of Oct4 , Nanog , and Sox-2 using
ChIP DNA hybridization to CHIPs containing probes for the promoters
of known genes [38]. The data revealed not only the network of genes
regulated by each transcriptional factor but also a core set of genes
coregulated by all three factors, thereby demonstrating the power and
relevance of such studies to systems biology. Another direct experi-
mental approach to identifying DNA cis -elements bound by specific
DNA trans -acting binding proteins is ChIP followed by sequencing of
the protein-bound DNA. The basic idea is to make a complete library
from the ChIP DNA fragments followed by high-throughput sequenc-
ing. With sufficient depth of sequencing the approach has the potential
to identify all DNA segments enriched by ChIP. The scale of coverage
can be enhanced by SAGE (series analysis of gene expression) tag-
based sequencing strategies.