Structure and function of Transcription Initiation Factor IID
created by Charlotte Campbell
Transcription Initiation Factor IID (TFIID) in Encephalitozoon cuniculi is one of six transcription factors (in conjunction with TFII-A-B-E-F-and-H) that drive transcription of protein-encoding genes (Romier, 2007). TFIID forms a pre-initiation complex with RNA polymerase II (Pol II) and is responsible for positioning Pol II at the transcription “start site.” TFIID is the largest of the transcription factors (McClean & Johnson). TFIID (pdb ID = 3OCI) has a molecular weight of 24462.41 Da, and its isolectric point (pI) is 9.78 (Artimo, 2012).
Both yeast and human forms of TFIID, as observed with low-resolution electron microscopes, have a symmetrical tertiary structure consisting of three lobes, each measuring about 60 Angstroms in diameter. The lobes are connected by thin links about 20 Angstroms wide, arranged around a cavity with an open channel measuring 40 Angstroms (Yarris, 1999). The protein, in essence, resembles a clamp (Romier, 2007). This clamp-like structure facilitates binding to DNA promoters. TFIID has approximately 15 subunits, the most significant of which is the TATA box-binding protein (TBP) (McClean & Johnson). In addition to TBP, TFIID has about 14 TBP-associated factors, called TAFs (Mencía, 2002). TBP is responsible for binding the TATA box on a specific DNA sequence. The TATA box is located about 25-35 base pairs upstream of a gene’s transcription unit, and TBP is the first protein to bind to this promoter region (McClean & Johnson).
TBP is a saddle-shaped protein located in the minor groove of the lobed TFIID. TBP has a twisting beta sheet at its center, and it has alpha helices at the sides with interspersed random coil (Nakatani, 1996).
It is likely the minor groove of TFIID that makes contact with the TATA box, as adenine- and thymine-rich DNA sequences favor minor groove compression, and the guanine- and cytosine-rich sequences that often flank the TATA region would favor the stabilization of TFIID’s major grooves (Horikoshi, 1992). Minor groove binding proteins tend to have lower sequence specificity relative to major groove binding proteins, but TFIID shows specificity of 2 x 10-9 M affinity for its sequence. Possible structural arrangements for the minor groove interaction include an alpha-helix in the H1 globular domain, an antiparallel beta ribbon, and a proline-rich beta-ribbon found in H1 termini (Lee, 1991). TFIID binding with DNA shows phosphate contacts at residues also essential for methylation interactions of TFIID, suggesting that the direct repeats of TFIID form two arms of beta-ribbons that contact DNA through the minor groove (Horikoshi, 1992). Bulky hydrophobic residues have been shown to improve transcriptional activation (Gill, 1994). Residues 118-129 in TFIID mainly fold along the backbone of the coding DNA strand, creating a binding pocket and positioning for transcription. Residue Phe-57 is an important residue involved in the hydrophobic interactions of TBP’s DNA-binding groove, and Gln-116 stacks with a hydrophobic DNA moiety at the TBP groove (Wollmann, 2012). These interactions with DNA cause the DNA strands to bend, further initiating a cascade toward transcription (McClean & Johnson). Several other transcription factors, including TFIIA and TFIIB, aid in securing TFIID to the DNA strand and in recruiting and initiating RNA polymerase II. Residue Val-162 plays a possible role in the formation of the TBP-TFIIB-DNA complex (Romier, 2007). Once mRNA elongation begins, all transcription factors except for TFIID separate from the RNA polymerase II complex. When transcription is complete, Pol II and then TFIID dissociate from the strand. After TBP unclasps the bent strand, the DNA returns to its normal form and shape (McClean & Johnson).
Cryo-electron microscopy, which preserves the structures of proteins in their natural conformations to a greater accuracy than negative stains, has confirmed the suspected three-lobed structure of TFIID. This technique, however, also allows speculation of the multiple conformations of isolated TFIID in cells. TFIID has been found to complex with unique arrangements of TAF subunits, which eventually alters transcriptional properties; for example, a form of TFIID thought to contribute to activation of estrogen receptors contains TAF30 but lacks TAF18 (Grob, 2006).
Nine of the 14 TAFs have been found to consist of histone-fold domains (HFDs) (Thuault, 2002). TAF6 and TAF9 contain domains similar to those of DNA core histones H4 and H3. TFIID may also be recruited to certain activated promoters by binding to a trimethylated lysine of H3 (Vermeulen, 2007), and TFIID-promoter complexes are so stable that TFIID is believed capable of packaging within densely compact chromosomes. This ability of TFIID to complex within nucleosomes may permit stable propagation and even promptly restart transcription after mitosis in cells is complete (Christova, 2002).
TFIID not only may associate with DNA, but the protein may also form a stable association with BTAF1 in humans and Mot1 in Saccharomyces cerevisiae (Pereira, 2004). Mot1 (pdb ID = 3OC3) and BTAF1 are forms of Swi2/Snf2 ATPases, which are enzymes that regulate transcription and other processes by catalyzing nucleosome remodeling or interfering with target protein-DNA interactions (Wollmann, 2012). BTAF1 and Mot1 have the capacity to use the energy from ATP hydrolysis to remove TBP from TATA DNA (Pereira, 2004). Mot1 has both significant sequence and folding pattern similarities to TFIID, as exhibited by the results of DALI (Z=26.0) (Holm & Rosenstrom, 2010) and protein Blast (E=6e-160) searches (National Library of Medicine, 2009). PSI-BLAST allows comparisons of proteins based on homology. E values compare primary structures of proteins to a query protein; a score below 0.5 indicates high similarity. DALI searches determine tertiary structure similarities. A Z score above 2 indicates similar folding mechanisms. TFIID also shares a similar fidelity with the transcription regulator NC2 alpha chain (pdb ID = 1JFI), DALI (Z=25.9) (Holm & Rosenstrom, 2010) and protein Blast (E=6e-134) (National Library of Medicine, 2009). These results may be indicators of the transcriptionally inactive, stable complexes TBP forms with NC2 and Mot1. NC2 blocks the incorporation of TFIIA and TFIIB in the pre-initiation complex when it binds with TBP. Mot1, on the other hand, affects TBP-DNA complexes by negatively regulating DNA promoters (Geisberg, 2002).
Mot1 contains multiple “HEAT (huntingtin, elongation factor 3, PP2A and lipid kinase TOR) repeats” that insert acidic loops on the undersurface of the TBP-DNA binding site (Wollmann, 2012). The TBP α-helices H1 and H2 are bound by the loop of Mot1 HR 4 (residues 209-221), α13 by HR 5, and α15 by HR 6. These interactions involve ion pairs between several TBP Arg and Lys residues and Mot1 Gln and Glu residues. These residues also act to stabilize the Mot1:TBP interactions as Mot1 packs against the Lys-103 side chain and forms a hydrophobic anchor at H1 and H2. This “latch” mechanism of Mot1 acts in the manner of a chaperone to prevent DNA re-association with TBP (Wollmann, 2012).
When considering these mechanisms and the immense involvement of TFIID in DNA chromosomal arrangements, transcription, and even repair, the benefits of studies concerning the function of TFIID become quite evident. Insight into the mechanistic causes and eventual prevention of diseases like Huntington’s disease – a neurogenerative polyglutamine disease – may be discovered when evaluating TAFs that bind glutamine-rich protein domains or when observing the affinity of HEAT repeats like those of Mot1 for transcription enhancers such as TBP (Neuwald, 2000). The profound ability of TBP to bind a TATA element, even at damaged genetic material, may provide the basis for future anti-cancer therapies as TBP becomes targeted to recognize damaged pieces of DNA (Vichi, 1997).