3 Retinoic Acid-Regulated Target Genes During Development …
73
at a level never before possible. Taking advantage of the recent in silico integrative approaches, it has markedly expanded our knowledge of the repertoire of RAregulated genes, RAR binding sites, and the RA-driven transcriptome, primarily in
endodermal (F9 cells) and neuronal (P19 cells and ESCs) differentiation.
RNA-Seq: A Revolutionary Tool for Transcriptomics
The novel high-throughput DNA sequencing method, termed RNA-seq (RNA
sequencing) uses recently developed NGS technologies (Ozsolak and Milos 2011;
Zeng and Mortazavi 2012), including Illumina (Solexa), Roche, Ion torrent or SOLiD
sequencing. Briefly, poly(A) + RNA is purified, fragmented and converted to cDNA
fragments that are built into a library. In general, libraries are made by ligating specific adapters to the DNA fragments, allowing them to cluster on a platform and be
amplified. Then each molecule is sequenced in a high throughput manner using a
variety of programs and finally aligned to the reference genome (Fig. 3.4). RNAseq generates extremely large data sets, the analysis of which is computationally
challenging due to the complexity of the transcripts. However, such a method offers
several key advantages. First it is not limited to detecting transcripts corresponding to
spotted cDNAs or oligonucleotides and thus it allows the entire transcriptome to be
analyzed. Second it has very low, if any background signal. Third it is very sensitive,
requires low amounts of RNA sample, and is dynamic as it has no upper limit for
quantification.
RNA-seq analysis identified thousands of RA-regulated genes in mESCs during
neuronal differentiation (Al Tanoury et al. 2014; Moutier et al. 2012). As in previous
studies, the list of genes shown to be up-regulated by RA included the RARβ2 gene,
canonical RA target genes involved in RA metabolism (Cyp26a1, Cyp26b1, Cyp26c1
and Dhrs3), patterning genes exemplified by the Hox genes (Hoxa1, Hoxa3, Hoxa5,
Hoxb1 and Hoxb4), other genes encoding homeobox proteins (Meis2, Cdx1, Gbx2,
and Hnf1b) and genes with a wide variety of functions such as Lefty1, Arg1 and Stra8
(Al Tanoury et al. 2014; Table 3.1). Interestingly, Al Tanoury et al. corroborated the
importance of the RARγ subtype by comparing WT and RARγ null mESCs (Al
Tanoury et al. 2014).
RNA-seq has greatly increased our knowledge about RA-regulated transcriptome.
However, it must be stressed that it measures the steady-state level of a given RNA,
which is the equilibrium between transcription, processing, and degradation. Therefore, other NGS-based technologies such as GRO-seq (Global run-on sequencing)
have been developed to measure nascent RNA, i.e. the genes engaged in transcription (Gardini 2017; Fig. 3.4). Unfortunately this technique is limited by the amount
of starting material (10
7 cells) and has not been used yet to study RA-regulated
developmental genes.
73
at a level never before possible. Taking advantage of the recent in silico integrative approaches, it has markedly expanded our knowledge of the repertoire of RAregulated genes, RAR binding sites, and the RA-driven transcriptome, primarily in
endodermal (F9 cells) and neuronal (P19 cells and ESCs) differentiation.
RNA-Seq: A Revolutionary Tool for Transcriptomics
The novel high-throughput DNA sequencing method, termed RNA-seq (RNA
sequencing) uses recently developed NGS technologies (Ozsolak and Milos 2011;
Zeng and Mortazavi 2012), including Illumina (Solexa), Roche, Ion torrent or SOLiD
sequencing. Briefly, poly(A) + RNA is purified, fragmented and converted to cDNA
fragments that are built into a library. In general, libraries are made by ligating specific adapters to the DNA fragments, allowing them to cluster on a platform and be
amplified. Then each molecule is sequenced in a high throughput manner using a
variety of programs and finally aligned to the reference genome (Fig. 3.4). RNAseq generates extremely large data sets, the analysis of which is computationally
challenging due to the complexity of the transcripts. However, such a method offers
several key advantages. First it is not limited to detecting transcripts corresponding to
spotted cDNAs or oligonucleotides and thus it allows the entire transcriptome to be
analyzed. Second it has very low, if any background signal. Third it is very sensitive,
requires low amounts of RNA sample, and is dynamic as it has no upper limit for
quantification.
RNA-seq analysis identified thousands of RA-regulated genes in mESCs during
neuronal differentiation (Al Tanoury et al. 2014; Moutier et al. 2012). As in previous
studies, the list of genes shown to be up-regulated by RA included the RARβ2 gene,
canonical RA target genes involved in RA metabolism (Cyp26a1, Cyp26b1, Cyp26c1
and Dhrs3), patterning genes exemplified by the Hox genes (Hoxa1, Hoxa3, Hoxa5,
Hoxb1 and Hoxb4), other genes encoding homeobox proteins (Meis2, Cdx1, Gbx2,
and Hnf1b) and genes with a wide variety of functions such as Lefty1, Arg1 and Stra8
(Al Tanoury et al. 2014; Table 3.1). Interestingly, Al Tanoury et al. corroborated the
importance of the RARγ subtype by comparing WT and RARγ null mESCs (Al
Tanoury et al. 2014).
RNA-seq has greatly increased our knowledge about RA-regulated transcriptome.
However, it must be stressed that it measures the steady-state level of a given RNA,
which is the equilibrium between transcription, processing, and degradation. Therefore, other NGS-based technologies such as GRO-seq (Global run-on sequencing)
have been developed to measure nascent RNA, i.e. the genes engaged in transcription (Gardini 2017; Fig. 3.4). Unfortunately this technique is limited by the amount
of starting material (10
7 cells) and has not been used yet to study RA-regulated
developmental genes.