Fig.1.6 Multisubunit and Modular Co-regulatory Complexes.

EM analysis revealed that ARC and CRSP, the mammalian counterpart of the yeast mediator are structurally related (yellow subunits are common) but distinct (orange and green subunits are unique) co-factors that display dramatically different functional properties. The larger ARC complex is inactive while the smaller CRSP complex is highly potent as a co-activator in vitro. Remarkably, the 3D structure of CRSP can undergo dramatic conformational changes dependent on the activator bound to target subunits within the CRSP assembly. Thus, the 3D structure of the unliganded, VP16-bound and SREBP-bound CRSP complexes display distinct structures as determined by negative stain EM and single particle reconstruction.

Although it may seem that the TAFs and the Mediator, which are ubiquitous transcriptional co-activators, would be sufficient for activating all genes, eukaryotic transcription once again proved to me more elaborate than imagined. Many other coactivators have now been identified. CBP, which was first identified as

CRSP/ARC-L superposition head

CRSP/ARC-L superposition head a co-activator for phosphorylated CREB, and p300 are two highly related proteins that are now known to function in transcriptional activation at many genes (Chrivia et al., 1993; Eckner et al., 1994; Kwok et al.,

1994). The mechanism by which these two factors, as well as others (e.g. GCN5), co-activated transcription was partly illuminated by the finding that these proteins harbored histone acetyltransferase activity (Bannister and Kouzarides, 1996; Ogryzko et al., 1996). Activator-specific, cell type-specific, and developmentally regulated co-activators soon followed: for example, OCA-B co-activates Oct transcription (Luo and Roeder,

1995), TAFn105 (TAF4b) is found in a cell type specific version of TFIID in B cells (Dikstein et al., 1996b; Freiman et al., 2002), and multiple testis-specific TAF isoforms have been found to function in spermatid development (Hiller et al., 2004; Hiller et al., 2001). Clearly, when it comes to transcriptional regulation in eukaryotes, complexity is the dominant theme. Only time and considerably more research will reveal how vast the co-activator universe is and the diverse spectrum of mechanisms they use to potentiate transcriptional activation.

Paving the Way: Remodeling Nucleosomes at Promoters

While many labs were focusing considerable effort on identifying and characterizing the transcriptional machinery, a few bold researchers had the foresight to ask how activators, co-activators, and the general machinery could possibly overcome the repressive effects of nucleosomes and higher order chromatin structures present in eukaryotic nuclei. Inroads in this area came from the integration of complementary findings from experiments in yeast, Drosophila, and human, which showed that nucleosomes could be remodeled (Cote et al., 1994; Kwon et al., 1994; Pazin et al., 1994; Tsukiyama et al., 1994). The yeast SWI/SNF complex, subunits of which had been discovered in genetic screens, turned out to be an ATP-dependent chromatin remodeling complex and effector of transcription (Cote et al., 1994). Other complexes that could assemble chromatin were found to have similar activity. These observations led to the idea that activators capable of binding native chromatin might recruit remodeling complexes to promoters, thereby opening the chromatin and allowing access to other transcriptional activators, co-activators, and the general transcription machinery. This proved to be the case (Neely et al., 1999; Yudkovsky et al., 1999), and the role of chromatin structure and its modulation was brought to the forefront of transcription research (Fig 1.7).

A Missing Link: Histone Modifications

For years, it had been known that in cells histones were differentially modified with acetyl, methyl, ubiquitin, and other post-translationally added groups. Dogma had it that histones in euchromatin, which was transcriptionally active, were hyper-acetylated, while histones in transcriptionally silenced heterochromatic regions were hypo-acetylated. Theories abounded to explain the correlation between histone acetylation and transcriptional competence, but for the most part, the transcription community paid little attention to these theories. This all changed with the identification of a nuclear histone acetyltransferase purified from Tetrahymena (Brownell et al., 1996). Surprisingly, the Tetrahymena HAT had high sequence similarity to a known yeast co-activator, Gcn5p. Instantaneously, the

Gene-specific nucleosome remodeling and histone modification

Gene-specific nucleosome remodeling and histone modification

Fig.1.7 Nucleosome remodeling and histone modifying complexes are recruited to promoters via interactions with activators.

collective eyes of the transcription community opened to the possibility that many transcription factors might bear HAT activity. When the dust settled, multiple previously identified co-activators were found to be HATs, and ultimately it was realized that histones were not the only substrates of these acetyltransferases; indeed, activators themselves could be acetylated. With the subsequent discovery of deacetylases (Taunton et al, 1996), acetylation was added to phosphorylation as a reversible post-translational modification used by intracellular signaling pathways to regulate gene expression. Ultimately, enzymes placing other modifications on histones (e.g. methylation, phosphorylation, and ubiquitination) were identified and characterized, and in some cases also found to be co-activators or co-repressors of transcription. Moreover, these enzymes can be recruited to promoters by gene specific activators and repressors to control levels of transcription.

The number of possible combinations of covalent modifications on the eight histones in any single nucleosome was dumbfounding. What was the function of all of these histone modifications? A seductive idea was posited: perhaps, specific patterns of post-translational modifications on the core histones in nucleosomes in individual promoters or regions of the genome help set the levels of transcription from those genes (Jenuwein and Allis, 2001; Strahl and Allis, 2000). For example, activation correlates with acetylation of specific lysines, while repression is observed upon acetylation or methylation of other lysines. Thus was born the Histone Code Hypothesis (Jenuwein and Allis, 2001; Strahl and Allis, 2000). While the putative histone "code" is far from understood, or even the notion of a true code accepted, it is clear that modification of histones adds another level of dynamic encoded information to the static DNA sequence present in a genome.

Escaped and On the Run: Regulation of Postinitiation Steps of Transcription

During the time that activators, co-activators, and general factors were being discovered and their roles in forming preinitiation complexes were initially characterized, some of the same labs and others embarked on understanding the mechanism and regulation of the RNA synthesis steps of the RNA polymerase II reaction. RNA synthesis is not simply the monotonous creation of phosphodiester bonds, but instead is a phase of the reaction rich in regulation (Fig. 1.8). TFIIF and the TFIIH helicase function during promoter escape (Chang et al, 1993; Goodrich and Tjian, 1994). The

TFIIH kinase phosphorylates the CTD of RNA polymerase II as the enzyme leaves the promoter (Lu et al, 1992). At the HSP70 promoter polymerase pauses after synthesis of a short (-20 nt) RNA, and is poised to fire the moment heat shock is sensed (via the Heat Shock Factor) (Gilmour and Lis, 1986; Rougvie and Lis, 1988). P-TEFb and DSIF/NELF have opposing effects on elongation (Marshall and Price, 1995; Wada et al, 1998; Yamaguchi et al, 1999). Elongation factors were discovered, including TFIIF, TFIIS, Elongin, etc, and indeed, the overall rate of elongation can be controlled globally and in a gene specific fashion (Aso et al, 1995; Reinberg and Roeder, 1987). HIV TAT, regulates the transcription reaction by binding a TAR element in the nascent RNA, which is reminiscent of bacterial phage factors that control transcriptional termination by binding the RNA transcript (Kao et al, 1987). RNA itself has recently appeared in the transcriptional regulatory picture, as a number of small noncoding RNAs have been found to control the RNA polymerase II transcription reaction via association with transcription factors and RNA polymerase II (Allen et al, 2004; Espinoza et al, 2004; Kwek et al, 2002; Nguyen et al, 2001; Yang et al, 2001). It seems that evolution has taken advantage of many different regulatory mechanisms beyond simply controlling the formation of preinitiation complexes, and we have only begun to appreciate and understand the multiple layers of regulation that can come into play.

Keeping the End in Sight: Coupling RNA Processing to Transcription

As transcription factors were identified and characterized using biochemical and genetic approaches, individual pieces of data began to support the notion that the transcriptional apparatus in eukaryotic cells is tightly coupled to the RNA processing machinery, and moreover that the transcription reaction itself can be influenced by factors that add the 5' Cap, splice the RNA, process the 3' end of the transcript, and transport the mature transcript out of the nucleus (Fig. 1.8) (Cho et al, 1997; Dantonel et al, 1997; Fong and Zhou, 2001; Hirose et al, 1999; McCracken et al, 1997a; McCracken et al, 1997b; Strasser et al, 2002). In hindsight, the coupling between transcription and RNA processing is logical, however, observations of splicing factors influencing transcription, and indications that RNA processing factors are recruited via interaction with the Pol II CTD were surprising, and the implications profound. We now envision that the nucleus contains mRNA synthesis/processing machines,

Promoter escape and transcript elongation: coupling transcription to mRNA processing

Promoter escape and transcript elongation: coupling transcription to mRNA processing


Fig.1.8 The post-initiation steps of transcription (e.g. promoter escape and elongation) are coupled to RNA processing (e.g. Capping, splicing, termination, polyadenylation, and RNA transport). Abbreviations: CPSF, cleavage and polyadenylaation specificity factor; GT, guanyl transferase; MT, methyl transferase; CstF, cleavage-stimulation factor; TREX, transcription/export complex.

which in response to gene specific activators coordinate the entire process from chromatin remodeling/ modification through transcription, processing and export of mature mRNAs to the cytosol—perhaps akin to an assembly line where all aspects of building and refining a final product are carried out in the same location by workers in intimate contact and orchestrated in a coordinated fashion. All parts of this massive molecular machine must work collaboratively to produce a functional mRNA. Current studies of the integration and cooperativity between the transcription and mRNA processing machineries have provided glimpses of the network that exists, and future studies on the integration of transcription with other nuclear processes (repair, replication, and recombination) will undoubtedly provide many new surprises and a true appreciation for the level to which nuclear events are functionally (and physically) connected.


Allen, T. A., Von Kaenel, S„ Goodrich, J. A., and Kugel, J. F. (2004). The SINE-encoded mouse B2 RNA represses mRNA transcription in response to heat shock. Nat Struct Mol Biol 11, 816-821.

Andel, F., 3rd, Ladurner, A. G., Inouye, C., Tjian, R., and

Nogales, E. (1999). Three-dimensional structure of the human

TFIID-IIA-IIB complex. Science 286, 2153-2156.

Anderson, J. E., Ptashne, M., and Harrison, S. C. (1985). A phage repressor-operator complex at 7 A resolution. Nature 316,

0 0

Post a comment