Dry and wet approaches for genome-wide functional annotation of conventional and unconventional transcriptional activators

Abstract

Transcription factors (TFs) are master gene products that regulate gene expression in response to a variety of stimuli. They interact with DNA in a sequence-specific manner using a variety of DNA-binding domain (DBD) modules. This allows to properly position their second domain, called "effector domain", to directly or indirectly recruit positively or negatively acting co-regulators including chromatin modifiers, thus modulating preinitiation complex formation as well as transcription elongation. At variance with the DBDs, which are comprised of well-defined and easily recognizable DNA binding motifs, effector domains are usually much less conserved and thus considerably more difficult to predict. Also not so easy to identify are the DNA-binding sites of TFs, especially on a genome-wide basis and in the case of overlapping binding regions. Another emerging issue, with many potential regulatory implications, is that of so-called "moonlighting" transcription factors, i.e., proteins with an annotated function unrelated to transcription and lacking any recognizable DBD or effector domain, that play a role in gene regulation as their second job. Starting from bioinformatic and experimental high-throughput tools for an unbiased, genome-wide identification and functional characterization of TFs (especially transcriptional activators), we describe both established (and usually well affordable) as well as newly developed platforms for DNA-binding site identification. Selected combinations of these search tools, some of which rely on next-generation sequencing approaches, allow delineating the entire repertoire of TFs and unconventional regulators encoded by the any sequenced genome.

Keywords: HT-SELEX; Moonlighting transcriptional activators; Protein binding microarrays; Transcription factors; Transcriptional activator trap; Yeast/bacterial one-hybrid.

Fig. 1

Identification and functional validation of TFs and unconventional activators. a. Schematic representation of the transcriptional activator trap (TAT) approach, as applied to the identification and functional validation of AD-containing, conventional and unconventional transcriptional activators. Reporter gene expression (HIS3, URA3 and LacZ) is activated if the query TF (a selected subset or a whole cDNA library; green) fused to the Gal4-DBD (blue) behaves as a transcriptional activator — i.e., it is capable of recruiting RNA Pol II transcription machinery (red). UAS: upstream activating sequence (Gal4 DNA-binding site); TATA: TATA box. b. Nuclear transportation trap (NTT) assay used to test the autonomous nuclear localization capacity of putative unconventional activators. A chimeric protein (NLS-less TF, blue) comprising a modified bacterial DBD (LexA), a portion of the E. coli maltose binding protein and the yeast Gal4 AD, but lacking a nuclear localization signal (NLS), is fused to a candidate unconventional activator (UA, green). If the latter contains a NLS (either recognizable in silico or cryptic), it will direct the chimeric protein to the nucleus, thus leading to reporter gene (HIS3, LacZ) activation. The transcriptional machinery is in red. LBS: LexA binding site; TATA: TATA box.

Fig. 2

Heterologous in vivo approaches for TF DNA binding site identification. a. The yeast one-hybrid (Y1H) is a DNA-centered approach used to identify TFs capable of binding to a specific DNA element. The DNA sequence to be interrogated (“DNA bait”) is cloned into a selectable yeast plasmid, upstream of reporter genes such as HIS3 and LacZ, and subsequently integrated into a mutated marker locus within the yeast genome. A TF of interest (either a selected one or a whole cDNA library; green) is expressed as a fusion with the yeast Gal4 activation domain (Gal4 AD, shown in blue). Positive hits (i.e., TFs bearing a DBD capable of interacting with the bait sequence) activate reporter gene expression. The transcriptional machinery is in red. TATA: TATA box. b. The bacterial one-hybrid (B1H) is a TF-centered approach used to identify the DNA element bound by a (putative) TF or activator. A bi-cistronic vector bearing a randomized region (rainbowed) upstream of two reporter genes (HIS3 and URA3) is used as a “prey” to identify the DNA elements bound by the “bait” TF (or putative activator) (shown in green) fused to the ω subunit (blue) of bacterial RNA polymerase (orange). The yeast URA3 gene is used as negative selection marker (5-FOA counter-selection) to eliminate self-activating DNA elements; the yeast HIS3 gene is used as a positive selection marker to identify the DNA elements bound by the bait TF.