72
C. Rochette-Egly
labeled with fluorophores and hybridized to microarrays composed of single stranded
DNA molecules corresponding to promoter sequences or covering complete genome
sequences of a selected species (Fig. 3.2). ChIP-chip with an Agilent extended promoter array allowed the identification of 462 RARγ bound loci in mESCs in which
the genes encoding RARα and RARγ were modified by homologous recombination
to introduce a TAP tag at the C terminus (Delacroix et al. 2010). In these cells, the
RAR bound loci were located in the regulatory regions of well-known RA regulated
genes such as Cdx1, Lefty1 and Rarb. Unexpectedly, such an approach revealed that
the majority of the binding sites did not contain any consensus RAREs but either
degenerate RAREs with multiple mismatches or anomalously spaced consensus half
sites. ChIP-chip was widely used until the arrival of ChIP-seq technology.
In Silico RA Response Element Analysis
The recent availability of an increasing number of genome sequences, allowed the
development of the in silico analysis of RAREs. It is a computational technique,
which analyzes all positions in a gene and which overcomes the cellular context
(tissue and cell specificity) and the chromatin structure. Thus it provides a direct
glance on the whole repertoire of possible RAREs.
Most RAREs are represented by two direct repeats of the hexametric motif (A/G)
G (G/T) TCA, separated by five (DR5), two (DR2) or one (DR1) nucleotides. Lalevee
et al. conducted a genome-wide in silico study of DR5 (Lalevee et al. 2011) and
DR2 (unpublished results) RAREs. They screened the masked human and mouse
genomes and identified more than 15,000 DR5 RAREs that were analyzed for their
localization and conservation in vertebrates. A set of 138 elements located −10 kb
from transcription start sites and gene ends and conserved across more than 6 species
was selected. The computational analysis was combined with experimental analysis
to determine whether the selected RAREs bind RARs and respond to RA.
Such a strategy expanded the repertoire of experimentally-validated RARE and
associated new genes involved in cell signaling and development. The limit of this
approach is that it focuses on classical RARE motifs and excludes RARs that bind
RAREs with other spacings or motifs. Another limitation is the high rate of identification of false positives, i.e. non-regulated genes containing RAREs. However,
lack of regulation of a gene in a given cell type is not sufficient to eliminate it as a
potential target in another context.
Current State of the Field
Recently, massively parallel sequencing technology known as NGS has revolutionized the field. With its ultra high throughput, scalability and speed, NGS has enabled
researchers to perform a wide variety of applications and to study biological systems
C. Rochette-Egly
labeled with fluorophores and hybridized to microarrays composed of single stranded
DNA molecules corresponding to promoter sequences or covering complete genome
sequences of a selected species (Fig. 3.2). ChIP-chip with an Agilent extended promoter array allowed the identification of 462 RARγ bound loci in mESCs in which
the genes encoding RARα and RARγ were modified by homologous recombination
to introduce a TAP tag at the C terminus (Delacroix et al. 2010). In these cells, the
RAR bound loci were located in the regulatory regions of well-known RA regulated
genes such as Cdx1, Lefty1 and Rarb. Unexpectedly, such an approach revealed that
the majority of the binding sites did not contain any consensus RAREs but either
degenerate RAREs with multiple mismatches or anomalously spaced consensus half
sites. ChIP-chip was widely used until the arrival of ChIP-seq technology.
In Silico RA Response Element Analysis
The recent availability of an increasing number of genome sequences, allowed the
development of the in silico analysis of RAREs. It is a computational technique,
which analyzes all positions in a gene and which overcomes the cellular context
(tissue and cell specificity) and the chromatin structure. Thus it provides a direct
glance on the whole repertoire of possible RAREs.
Most RAREs are represented by two direct repeats of the hexametric motif (A/G)
G (G/T) TCA, separated by five (DR5), two (DR2) or one (DR1) nucleotides. Lalevee
et al. conducted a genome-wide in silico study of DR5 (Lalevee et al. 2011) and
DR2 (unpublished results) RAREs. They screened the masked human and mouse
genomes and identified more than 15,000 DR5 RAREs that were analyzed for their
localization and conservation in vertebrates. A set of 138 elements located −10 kb
from transcription start sites and gene ends and conserved across more than 6 species
was selected. The computational analysis was combined with experimental analysis
to determine whether the selected RAREs bind RARs and respond to RA.
Such a strategy expanded the repertoire of experimentally-validated RARE and
associated new genes involved in cell signaling and development. The limit of this
approach is that it focuses on classical RARE motifs and excludes RARs that bind
RAREs with other spacings or motifs. Another limitation is the high rate of identification of false positives, i.e. non-regulated genes containing RAREs. However,
lack of regulation of a gene in a given cell type is not sufficient to eliminate it as a
potential target in another context.
Current State of the Field
Recently, massively parallel sequencing technology known as NGS has revolutionized the field. With its ultra high throughput, scalability and speed, NGS has enabled
researchers to perform a wide variety of applications and to study biological systems