Fig. 1.1 Cis-control elements in RNA polymerase II promoters can be located near the transcription start site or at great distances away. Some of the cis-control elements identified by early studies are shown. Abbreviations: TATA, TATA box; GC, GC box; SRE, sterol response element; NRE, nuclear hormone receptor response element; HRE, heat shock response element.

A Shaft of Light: Sequence Specific Transcription Factors

After a flurry of intense promoter bashing experiments with all manner of DNA templates, cell-types and gene systems, we were confronted with the daunting task of determining what was actually recognizing and keying off these composite arrays of cis-control DNA sequences to govern gene specific transcription. One important step along this pathway of discovery was the rapid deployment of various elegant in vitro mutagenesis techniques such as linker scanning clustered point mutations and deletions (McKnight et al., 1981; McKnight and Kingsbury, 1982; Myers and Tjian, 1980). At the same time, powerful new biochemical assays such as DNase I footprint protection were being developed (Galas and Schmitz, 1978). Perhaps the single most influential strategy for those of us attempting to dissect the molecular identity of transcriptional regulatory factors was the promoter selective in vitro transcription assay (Manley et al., 1980; Rio et al., 1980; Weil et al., 1979; Wu, 1978). This "bucket biochemistry" approach allowed us to use cloned DNA fragments containing well mapped and carefully defined promoters to drive accurate and factor dependent transcription by partially purified RNA polymerases. The tacit assumption in establishing such in vitro promoter dependent assays was that purified eukaryotic RNA polymerase II was necessary but not sufficient to direct accurate initiation of transcription. We therefore assumed that one or more additional transcription factors (whose identity and mode of action had remained unknown) was needed in order to instruct or otherwise impart upon RNA pol II the ability to discriminate one promoter from another. Indeed, since no such cellular factors in eukaryotes had yet been identified or isolated in 1980, we had little clue as to the biochemical properties of such factors (i.e. were these factors proteins, nucleic acids, carbohydrate, etc.?). The closest candidate at that time was the SV40 T-ag, a viral encoded protein that displayed many of the hallmarks of a bona fide promoter recognition factor (Rio et al., 1980; Tjian, 1978). Also, whether they would directly bind RNA polymerase a la cr-factors or they would behave more like CAP in the lac operon system and bind DNA in a sequence specific manner was a big question.

Indeed, one of the unappreciated and hidden advantages of using fairly crude nuclear extracts (i.e. from Hela cells or Drosophila embryos) to carry out systematic biochemical "complementation" tests in vitro allowed us the freedom to be unbiased and simply search for whatever molecules stimulated transcription of one promoter but not another. Using this approach, factors such as Spl were first identified as functional transcriptional activators for RNA pol II that could discriminate, for example, between the SV40 and AdML promoters (Fig. 1.2) (Carthew et al., 1985; Dynan and Tjian, 1983a; Dynan and Tjian, 1983b; Sawadogo and Roeder, 1985). Similar biochemical fractionation and in vitro assays led to the isolation of TFIIIA for Pol III and UBF for Pol I (Engelke el al, 1980; Learned et al., 1985; Learned et al., 1986; Pelham and Brown, 1980; Wu, 1978). However, it did not take long given the availability of various discriminating DNA binding assays available at that time to determine that these transcription factors were indeed sequence specific DNA binding "activators". And thus, there was a nice alignment of cis-regulatory elements and DNA binding transcription factors. We anticipate that a similar biochemical dissection and reconstitution of in vitro transcription reactions that are responsive to distal enhancers, tethering elements, silencers and boundary elements are still needed to fill-in our gaps of knowledge vis-à-vis the molecular players and mechanisms that govern "long distance" regulation so prevalent in metazoan organisms.

Identification and purification of activators

Identification and purification of activators

Proximal Promoter

Fig.1.2 Trans-acting factors bind to RNA polymerase II promoters. Abbreviations: Spl, specificity protein 1; SREBP, sterol response element binding protein; NHR, nuclear hormone receptor; HSF, heat shock factor.

Proximal Promoter

Fig.1.2 Trans-acting factors bind to RNA polymerase II promoters. Abbreviations: Spl, specificity protein 1; SREBP, sterol response element binding protein; NHR, nuclear hormone receptor; HSF, heat shock factor.

A New Era of Transcription Biochemistry Arrives: Clone, Sequence, Express & Reconstitute

The next big hurdle was to actually purify, clone, and characterize these seemingly powerful transcriptional activators. As often happens in emerging fields, advances in concepts and techniques must go hand in hand. For the transcription field, the development of sequence specific DNA affinity chromatography and a host of affiliated techniques revolutionized our capacity to detect, purify and clone the genes encoding sequence-

specific transcription factors (Briggs et al., 1986; Jones et al., 1985; Kadonaga el al., 1987; Kadonaga and Tjian, 1986). Once the genes encoding the first few bona fide transcriptional activators (and repressors) such as Spl, TFIIIA, CTF, API, GCN4, Gal4, GR, and HSF were characterized — a flood of paradigm shifting concepts emerged (Berg, 1988; Bohmann et al., 1987; Courey et al., 1989; Kadonaga et al., 1987; Kadonaga et al., 1988; Mermod et al, 1989; Miller et al., 1985; Mitchell et al., 1987; Triezenberg et al., 1988; Turner and Tjian, 1989). For instance, the remarkably modular nature of transcriptional activators was revealed (Ma and Ptashne, 1987a; Ma and Ptashne, 1987b). The subsequent cloning and sequencing of transcription factors rapidly advanced our ability to recognize DNA binding motifs (i.e. Zn finger, B-HLH, homeodomains, etc.) dimerization domains (LZ, histone folds) activation domains (gin-rich, acidic, etc.) and regulatory/ligand binding domains (AF2).

Initially, as a result of the pioneering work on transcription factor structures derived from studies of the ^.-repressor and other phage and bacterial transcription factors (Anderson et al., 1985; Wharton et al., 1984), there was a tendency to assume that all transcription factors would utilize a helix-turn-helix DNA binding domain and an "acidic" activation domain. However, the structure/ function analysis of eukaryotic transcription factors such as Spl, TFIIIA, steroid receptors, Jun/Fos API, C/EBP, CTF etc. quickly dispelled the over-simplified notion that there were only one or two motifs for DNA binding and transcription activation (Gill and Ptashne, 1988). Indeed, it became clear that in eukaryotes and especially metazoan organisms, the repertoire of structural domains that had evolved to accommodate transcriptional specificity was astoundingly diverse and elaborate.

One of the most impressive accomplishments during this rich middle period (1985 — 1995) of transcription research was not only the rapid identification, cloning and characterization of hundreds of sequence specific transcription factors, but also a quantum leap in our understanding of the relationship between function and structure — particularly with regards to DNA binding motifs (Pabo and Sauer, 1992). The high resolution X ray structures of countless DNA binding domains were solved and this rich body of information continues to provide a basis for rapid genome wide functional analysis of novel gene products. The discovery of thousands of different transcriptional activators (repressors) and their pivotal role in complex biological processes such as anterior-posterior and dorsal-ventral patterning in metazoans firmly cemented the importance of this vast family of proteins. Indeed, after the first few different metazoan genomes were determined, it became apparent that between 5%~10% of the coding capacity of eukaryotes is devoted to encoding such transcriptional regulators. These findings provided another inexorable clue to the essential, universal, and yet diverse nature of transcriptional control mechanisms. However, despite this exponential growth in knowledge about transcription factors, not everything was rosy or well understood about transcriptional regulation. Indeed, although DNA binding motifs and their structures had proven to be highly informative with respect to structure/function relationships, a similar understanding of activation domains was sorely lacking and largely remains so even today.

Mix and Match: Combinatorial Control, Modularity, and Enhanceosomes

As activators and genes were being characterized in greater detail, it became apparent that the simple paradigm of a single activator or single repressor controlling transcription of a gene, as was the case in some bacterial and even yeast systems, did not apply in higher eukaryotes. The regulatory regions of mammalian and Drosophila genes, enhancers and silencers, contain binding sites for many transcriptional regulators. An enhancer might bind 10 or more DNA binding factors, including many different activators as well as multiple copies of a single activator. This complexity was further amplified by the presence of large activator families (Homeo-box, FOXO, API etc.) in which individual members had similar DNA binding specificities, but distinct activation domains and presumably different functions (Mitchell and Tjian, 1989). Combinatorial control and the notion of cis-regulatory networks help explain observations indicating that it is the precise complement of activators and repressors present at a promoter that gives rise to gene specific activation in a spatial and temporally regulated pattern (DeFranco and Yamamoto, 1986; Diamond et al, 1990). Cooperativity in DNA binding and synergy in transcriptional activation further contribute to an uncanny level of control over gene transcription. Our understanding of enhancers and activators was substantially advanced with the detailed characterization of the interferon-P and T-cell receptor a enhancers, where the correct function of the enhancers requires not only the presence of the appropriate array of transcriptional activator proteins but also the association of architectural proteins, and the proper spatial orientation of all of these factors dictates the ultimate outcome (Giese et al, 1992; Giese et al., 1995; Thanos and Maniatis, 1995).

Unimagined complexity: The General Transcription Factors, PIC formation, and Promoter-Specific Transcription

From early in vitro studies it was realized that while core RNA polymerase II was capable of synthesizing an RNA product, it required additional factors to initiate transcription at specific promoters (Weil et al., 1979). The general concept of dissociable and essential transcription factors had been firmly established in bacteria, where core RNA polymerase required a sigma subunit for promoter-specific transcription. While this paradigm provided a useful framework for studying eukaryotic transcription, the requirement for a single sigma-like subunit was quickly dispelled in eukaryotic transcription systems. Employing biochemical fractionation and promoter specific DNA templates to drive in vitro transcription reactions, an unexpectedly large number of critical accessory factors were painstakingly teased out and characterized, initially as crude fractions eluted from columns (Matsui et al., 1980). Of course, like any good biochemist, once you have an assay, next you want to purify the critical activity, characterize its biochemical properties, and identify the gene encoding the factor. After many hundreds of researcher years, all of the general factors and their genes from human, Drosophila and yeast eventually were isolated (Aso et al., 1992; DeJong and Roeder, 1993; Eisenmann et al., 1989; Finkelstein et al., 1992; Fischer et al, 1992; Ha et al, 1991; Hahn et al, 1989a; Hahn et al, 1989b; Hoey et al, 1990; Horikoshi et al., 1989; Kao et al, 1990; Ma et al, 1993; Peterson etal., 1991; Peterson et al, 1990; Schaeffer et al, 1993; Shiekhattar et al, 1995; Sopta et al, 1989; Yokomori et al, 1993). Thus, the general or basal transcription factors TFII-A, -B, -D (TBP), -E, -F, and -H were identified (Fig. 1.3).

The general transcription factors were unlike the sequence specific activators in that most of them showed little or no propensity to bind DNA in ft sequence dependent manner, but instead associated with RNA polymerase II and participated in complex ways towards the assembly of the pre-initiation complex (PIC). Among this large clan of general transcription factors—one that stood out early on was the fraction originally designated TFIID which revealed a weak tendency to bind TATA elements (Reinberg et al, 1987; Sawadogo and Roeder, 1985). Attempts to purify and characterize this activity proved to be particularly intransigent. After many attempts and failures on the part of several labs, through a combination of persistent biochemistry and fortuitous genetics—the all important TATA binding protein (TBP) was isolated and cloned in the late 1980s (Hahn et al., 1989b; Hoey et al., 1990; Horikoshi et al., 1989; Kao et al., 1990; Peterson et al., 1990). TBP, the central subunit of the TFIID complex, itself has the ability to bind specifically to TATA box elements found in many, but by no means most RNA polymerase II promoters. The surprising observation that the single subunit TBP could replace the crude TFIID function in directing preinitiation complex assembly and basal transcription in vitro enabled biochemical experiments to establish an order of assembly for the preinitiation complex—TBP, TF1IA, TFIIB, TFIIF/RNA polymerase II, TFIIE, and TFIIH (Buratowski et al., 1989; Buratowski et al., 1988; Flores et al., 1992). In later in vitro experiments, RNA polymerase II was also found in larger complexes containing some of the general transcription factors and provided an alternative mode of preinitiation complex assembly (Koleske and Young, 1994), in which a RNA polymerase II complex is recruited to TFIID and TFIIA pre-assembled on promoter DNA.

Identification and purification of Lhegeneral factors

0 0

Post a comment