TAF3

TAF3 is a TFIID subunit that couples core promoter recognition to the chromatin and signaling state of active genes, functioning as a histone-mark reader and an adaptor for alternative transcription complexes (PMID:11438666, PMID:18682226, PMID:23452851). Originally identified as a histone-fold-containing TFIID component that heterodimerizes with TAF10 via its HFD (PMID:11438666), TAF3 uses its PHD finger to selectively read the H3K4me3 mark through cation-pi interactions in a dedicated binding pocket, a recognition event blocked by asymmetric dimethylation of H3R2, establishing an H3R2/K4 methyl-methyl switch (PMID:18682226). This H3K4me3-PHD interaction directs global TFIID recruitment to active genes and stimulates preinitiation complex assembly, including cooperative regulation of gene-selective p53 target genes under genotoxic stress (PMID:23452851). The same PHD finger binds nuclear phosphoinositides at a site distinct from the H3K4me3 pocket, and this lipid binding modulates chromatin association, linking PIP4K2B-controlled nuclear PI5P levels to TAF3-dependent myogenic gene expression (PMID:25866244). Beyond canonical TFIID, TAF3 forms an alternative TRF3/TAF3 complex that replaces TFIID at the myogenin promoter to support MyoD-dependent myogenesis (PMID:18851836) and, in zebrafish, the selective Trf3-Taf3 interaction at the mespa promoter is required for early development and hematopoiesis (PMID:19777587). In embryonic stem cells, TAF3 is recruited by CTCF to promoter-distal sites and mediates long-range DNA looping to core promoters to safeguard lineage-balanced transcription (PMID:21884934). TAF3 also physically interacts with p53 and, when overexpressed, inhibits p53-dependent transcription (PMID:18549481).

1. 2001 Medium

   Establishing TAF3 as a bona fide TFIID subunit and defining its domain architecture was needed to place it in the general transcription machinery.

   Evidence Immunoprecipitation, sequence analysis, and HFD heterodimerization assay identifying the histone fold and PHD finger and TAF10 partnership

   PMID:11438666

   Open questions at the time

   • Did not assign a function to the PHD finger
   • Did not address whether TAF3 acts outside canonical TFIID
2. 2008 High

   Defining the PHD finger as an H3K4me3 reader explained how TFIID could be targeted to active chromatin via a histone mark.

   Evidence NMR structure of the TAF3-PHD finger in complex with H3K4me3 peptide plus mutagenesis and affinity measurements

   PMID:18682226

   Open questions at the time

   • Structural study alone did not establish genome-wide consequences of the interaction
   • The H3R2/K4 methyl switch was inferred biochemically without in vivo demonstration
3. 2008 High

   Discovery of the TRF3/TAF3 complex showed TAF3 can substitute for canonical TFIID at a tissue-specific promoter, revealing a promoter-selective transcription mechanism in myogenesis.

   Evidence Purified reconstituted in vitro transcription system, cell-based assays, and domain mapping at the myogenin promoter

   PMID:18851836

   Open questions at the time

   • Scope of TRF3/TAF3 target promoters beyond myogenin not defined
   • Mechanism of TFIID-to-TRF3/TAF3 switching not resolved
4. 2008 Medium

   The TAF3-p53 interaction indicated TAF3 can negatively modulate a sequence-specific activator, beyond its general TFIID role.

   Evidence Yeast two-hybrid, in vitro binding, and cell-based transcription assays with TAF3 overexpression in human and Drosophila

   PMID:18549481

   Open questions at the time

   • Effect shown by overexpression rather than endogenous loss-of-function
   • Mechanism reconciling repression here with later H3K4me3-enhanced p53 transcription unclear
5. 2009 Medium

   Demonstrating that a selective Trf3-Taf3 interaction is required in vivo extended the alternative-complex model to whole-organism developmental processes.

   Evidence Morpholino knockdown, ChIP, interaction assay, and mutant rescue at the mespa promoter in zebrafish

   PMID:19777587

   Open questions at the time

   • Direct biochemical reconstitution of the zebrafish complex not performed
   • Whether mammalian hematopoiesis uses the same mechanism not addressed
6. 2011 High

   Identifying CTCF-dependent recruitment of TAF3 to distal sites and TAF3-mediated DNA looping revealed a role in long-range chromatin architecture rather than only proximal PIC assembly.

   Evidence Genome-wide ChIP-seq, chromosome conformation capture, co-IP, and TAF3 knockdown with lineage differentiation readout in ES cells

   PMID:21884934

   Open questions at the time

   • Whether looping requires the PHD/H3K4me3 reading function not separated
   • Direct interface mediating the CTCF-TAF3 interaction not mapped
7. 2013 High

   Connecting the PHD-H3K4me3 interaction to genome-wide TFIID recruitment and stress-induced p53 target selection demonstrated the functional output of histone-mark reading.

   Evidence ChIP-seq, in vitro PIC assembly assays with TAF3 PHD mutants, and p53 target gene analysis under genotoxic stress

   PMID:23452851

   Open questions at the time

   • Interplay between H3K4me3-enhanced p53 transcription and earlier reported TAF3-mediated p53 repression unresolved
   • Relative contribution of TATA box versus H3K4me3 across the genome not quantified
8. 2015 High

   Showing that the PHD finger also binds nuclear phosphoinositides at a distinct site established a signal-dependent layer of control over TAF3 chromatin association.

   Evidence Targeted PI-binding screen, in vitro lipid-binding with separation-of-function PHD mutants, ChIP, and myoblast differentiation assay linking PIP4K2B/PI5P to TAF3

   PMID:25866244

   Open questions at the time

   • How PI binding is dynamically regulated at specific loci in vivo not resolved
   • Whether PI binding affects the CTCF-looping or p53 functions not tested

• How TAF3's distinct activities — canonical TFIID H3K4me3 reading, alternative TRF3/TAF3 complex formation, CTCF-mediated looping, phosphoinositide sensing, and p53 modulation — are coordinated within a cell remains unresolved.
• No unified model integrating PHD-mediated histone, lipid, and protein interactions
• Switching logic between canonical TFIID and TRF3/TAF3 complexes uncharacterized
• Endogenous regulation of TAF3 partner choice not defined