poly A site Initiation,

Pafl complex association, cscape to elongation poly A site

Association with mRNA processing and liistonc modification factors

poly A site

Release of Pafl complex at poly Asite poly A site processing s poly A site

Release of Pafl complex at poly Asite poly A site processing s

Fig.3.1 The dynamic association of initiation, elongation and RNA processing factors with RNAPII during the transcription can then synthesize a complete RNA transcript (Pal and Luse, 2003). It appears that the RNAPII associated factors TFIIE and TFIIH play an important role in this transitional stage, although their specific role in the process is not yet clear (Dvir et al., 1997; Goodrich and Tjian, 1994b; Kumar et al., 1998; Yamamoto et al., 2001; Watanabe et al., 2003).

Concurrent with the promoter clearance step is the beginning of the RNAPII CTD phosphorylation cycle. RNAPII exists in a form in which the CTD of its largest subunit is either unphosphorylated, hypo- (IIA) or hyperphosphorylated (IIO) (Dahmus, 1981). The RNAPII CTD is unphosphorylated in the preinitiation complex (Laybourn and Dahmus, 1989; Lu et al., 1991) and is heavily phosphorylated at specific serine residues (Ser 2 and Ser 5) during the transcription process (Christmann and Dahmus, 1981). Phosphorylation at Ser 5 in the YSPTSPS repeat of the CTD correlates with transcription initiation and early elongation, whereas phosphorylation at Ser 2 occurs further along in the elongation process (O'Brien et al., 1994; Komarnitsky et al., 2000) and plays a central role in recruiting elongation factors to the transcribing polymerase.

The process of transcription elongation is now viewed as a dynamic and highly regulated step in the synthesis of mRNA, and is an important factor in the coordination of downstream events as well. Recent studies have identified a multitude of factors that modulate the activity of RNAPII (active elongation factors), as well as many factors that associate with RNAPII but do not directly regulate its catalytic activity (passive elongation factors). These proteins, in association with RNAPII are often referred to as the transcription elongation complex (TEC) (reviewed in Sims, et al., 2004). Factors within the TEC often associate and disassociate in a very specific manner throughout elongation, making the TEC a highly regulatable and ever-changing complex. The following sections will discuss some of the important factors, associated with RNAPII, that facilitate transcription elongation.

A: "Active" Elongation Factors That Modify RNAPII Catalytic Activity

Several impediments, such as transcriptional pause, arrest, and termination, may represent a general, intrinsic feature of RNAPII catalytic activity (Bentley, 1995). There are many elongation factors whose job it is to prevent or alleviate such blocks.


TFIIF influences the overall rate of transcription by modulating transcriptional pause, the temporary and reversible halting of the addition of NTPs to the nascent RNA transcript. TFIIF was originally identified as a factor that directly binds immobilized RNAPII (Sopta et al., 1985). During elongation, it controls the rate of transcription by decreasing the amount of time that RNAPII is paused (Price et al., 1989; Bengal et al., 1991; Tan et al., 1994). TFIIF also has the ability to associate with both the hypo- and hyperphosphorylated forms of RNAPII (Zawel et al., 1995), consistent with a role in both initiation and elongation. Mutational analyses of both subunits of TFIIF have indicated an important role for TFIIF in promoter clearance and the earliest stages of elongation, preventing arrest (Yan et al., 1999). This is in agreement with chromatin immunoprecipitation experiments showing that TFIIF is primarily localized near promoter regions and not evenly distributed throughout coding regions (Krogan et al., 2002a; Pokholok et al. 2002), and it appears that its association with RNAPII is required predominantly at times when the elongation machinery is stalled.

A2: The ELL Family

Both the ELL family and the Elongins appear to stimulate the rate of transcription in a similar manner to TFIIF, by interacting directly with RNAPII and suppressing transient pausing (Shilatifard et al., 1996; Bradsher et al., 1993b). The ELL gene was initially identified in humans (Thirman, et al., 1994) and belongs to a family that includes ELL2 and ELL3, both of which affect the rate of elongation in vitro (Shilatifard et al., 1997; Miller et al., 2000). Consistent with its role in elongation, the single Drosophila melanogaster ELL homolog (dELL) associates with active sites of transcription following heat shock and colocalizes with the elongation-competent form of RNAPII on polytene chromosomes (Gerber et al, 2001). In addition, mutations in dELL preferentially affect expression of long genes (Eissenberg et al., 2002).

A3: Elongins

Elongin, also called SIII, is a heterotrimeric protein composed of Elongin A, B, and C (Aso, et al., 1995, 1996; Garrett et al., 1995; Takagi et al., 1996). It was first purified to homogeneity from rat liver nuclei by its ability to stimulate the rate of RNAPII elongation in vitro (Bradsher et al., 1993a). Although the precise mechanism by which it controls the rate of elongation is not completely understood, recent evidence suggests that it functions to properly align the 3'-OH end of a nascent transcript into the active site of RNAPII (Takagi et al., 1995). There is also evidence suggesting that

Elongin is an E3 ubiquitin ligase and functions to activate ubiquitylation of RNAPII, targeting it for proteasomal degradation. However, where and how it functions in this process in unclear (reviewed in Shilatifard et al, 2003).

A4: Spt4/Spt5 (DSIF)

DSIF is a heterodimeric protein complex composed of the human homologs of the yeast Spt4 and Spt5 proteins (Wada et al., 1998). Early studies in yeast showed that Spt4 and Spt5 act as transcription factors that modify chromatin structure (Winston and Carlson, 1992). However, more recent genetic and biochemical evidence demonstrates that Spt4 and Spt5 are elongation factors that form a complex in which Spt5 binds RNAPII and functions in transcription elongation (Hartzog et al., 1998; Wada et al., 1998). Upon the discovery that DSIF functioned as an elongation factor, it became classified as a negative factor in elongation because adding it to a partially purified transcription reaction resulted in transcription inhibition. However, later experiments showed that the transcription reaction used in those studies also contained NELF, a factor which acts in concert with DSIF to inhibit elongation (Yamaguchi et al., 1999). More recent studies, focusing on the yeast Spt4 protein suggest that DSIF actually promotes transcription (Rondon et al., 2003). As will be discussed later in this chapter, transcription elongation and chromatin remodeling are intricately connected. Therefore, it is not surprising that DSIF has been shown to genetically and physically interact with many chromatin-related factors, including Spt6, FACT, Chdl, as well as members of the Pafl complex (Hartzog et al., 1998; Orphanides et al., 1999; Costa and Arndt, 2000; Krogan et al., 2002b; Mueller and Jaehning, 2002; Simic et al., 2002; Squazzo et al., 2002; Lindstrom et al, 2002).


While the previously discussed elongation factors alleviate pausing and are considered positive elongation factors, the NELF (Negative Elongation Factor) complex actually works to promote pausing (Yamaguchi et al, 1999). NELF binds to an assembled RNAPII/DSIF complex, but cannot bind DSIF or RNAPII alone, and this association is required for NELF to exhibit its negative effects on transcription (Yamaguchi et al, 2002). When NELF dissociates from the TEC upon phosphorylation of the RNAPII CTD, as well as the Spt5 component of DSIF, pausing is overcome and elongation resumes (Ivanov et al, 2000; Kim and Sharp, 2001). Once elongation resumes, it appears that the RNAPII/DSIF complex remains with the TEC, but NELF does not. In vivo studies in Drosophila melanogaster show that NELF/RNAPII/ DSIF are localized to the hsp70 heat-shock promoter before induction; upon induction by heat-shock, RNAPII and DSIF localize to active sites of transcription, but NELF does not (Andrulis et al, 2000; Wu et al,

2003). It has been proposed that this mechanism of inhibiting elongation occurs to allow time for the assembly of other factors involved in mRNA maturation (reviewed in Sims et al, 2004).


TFIIS was originally identified by its ability to stimulate transcription in vitro (Natori et al, 1973) and subsequently the first transcription elongation factor to be purified (Sekimizu et al, 1976). TFIIS promotes RNAPII read-through at transcriptional arrest sites (reviewed in Fish and Kane, 2002; Conaway et al, 2003; Wind and Reines, 2000) thereby increasing the efficiency of transcription elongation rather than its rate (Wind-Rotolo et al, 2001; reviewed in Sims et al,

2004). Evidence for this comes from a series of experiments using artificial arrest sites in vivo in which TFIIS counteracted RNAPII arrest at a position far from the promoter (Kulish and Struhl, 2001). It is generally thought that the read-through of these transcriptional blocks is overcome by TFIIS-stimulated RNA cleavage, which creates new 3' ends, repositions the RNA in the active site of RNAPII, and allows RNAPII to continue elongating. However, a new X-ray crystallographic model of the complete 12-subunit RNAPII in complex with TFIIS suggests that the TFIIS-induced repositioning of RNA may result in a different elongation complex conformation that is less prone to stalling. It also suggests that the acidic residues of a TFIIS hairpin structure could assist in NTP binding (Kettenberger et al, 2004). Although the precise mechanism by which TFIIS alleviates transcriptional arrest is not completely understood, it does appear to be a major contributor to the efficiency of elongation.

B: "Passive" Elongation Factors and Factors That Remodel and Modify Chromatin

In addition to transcriptional pausing and arrest, another major impediment to elongation which adds a new level of complexity to its regulation is the requirement that RNAPII must transcribe through chromatin. Recent explorations into covalent histone modifications (reviewed in Strahl and Allis, 2000; Jenuwein and Allis, 2001; Gerber and Shilatifard, 2003), as well as the mechanisms by which RNAPII elongates through chromatin (including nucleosome mobilization and histone depletion models) (reviewed in Studitsky et al, 2004) have provided valuable insights into regulation of elongation in a chromatin environment. Some of the key players in this process are discussed below.

Bl: Swi-Snf

The Swi-Snf complex is an ATP-dependent chromatin remodeler, and, while it is generally thought of as a factor involved in initiation by acting at the promoters of active genes, it appears to also function as an elongation factor required to overcome the enhanced transcriptional pausing that occurs on chromatin templates. Evidence for this comes from the fact that while human heat shock factor 1 (HSF1) can stimulate both initiation and elongation and recruit Swi-Snf to a chromatin template, recruitment of Swi-Snf is greatly impaired when residues in HSF1 responsible for elongation are mutated (Brown et al, 1996; Sullivan et al, 2001). In addition, mutations in yeast Swi-Snf components (SWI1, SNF5, SWI2, or SNF2) are synthetically lethal in combination with a disruption in PPR2 (DST1), the gene encoding the transcription elongation factor TFIIS (Davie and Kane, 2000). The specific mechanism by which Swi-Snf facilitates elongation is not known.

B2: Chdl

Chdl is an ATP-dependent chromatin remodeler that appears to function in both elongation and transcription (Tran et al, 2000). Much evidence suggests a role for Chdl in elongation. Analyses of Drosophila melanogaster polytene chromosomes showed that Chdl associates with highly active sites of transcription (Stokes et al, 1996). Chdl genetically interacts with Set2 and members of the IWI family, and Swi-Snf, all of which have been implicated in elongation (Tsukiyama et al, 1999; Krogan et al, 2003b; Tran et al, 2000). In addition, Chdl physically interacts with the elongation factors DSIF and FACT (discussed later in this chapter) (Kelley et al, 1999; Simic et al, 2003). Again, like Swi-Snf, little is known about how Chdl facilitates elongation through chromation.


Cockayne Syndrome group B (CSB) is DNA-dependent ATPase and a member of the Swi-Snf family of chromatin remodelers. CSB has been shown to enhance the rate of transcription elongation on naked DNA in vitro (Selby and Sancar, 1997), and can remodel chromatin in vitro (Citterio et al, 2000). In addition, CSB can bind directly to RNAPII and affect the elongation activity of TFIIS (Tantin et al, 1997). In addition to its link to elongation, CSB also plays a role in transcription-coupled nucleotide repair and base excision repair (reviewed in Licht et al., 2003). Therefore, while it is possible that CSB may function as a chromatin elongation factor, its ability to remodel chromatin may simply reflect the ability of CSB to remodel protein-DNA interactions such as those between a stalled RNAPII and a DNA lesion (reviewed in Svejstrup, 2002).


FACT (facilitates chromatin transcription) was discovered using an assay designed to identify factors that support RNAPII transcription on chromatin templates (Orphanides et al, 1998). This highly conserved heterodimer is comprised of hSptl6 and SSRP1 (Orphanides et al, 1999) which are homologous to the yeast Sptl6/Cdc63 and Pob3 proteins. FACT is one of the few, if only, known factors that can stimulate transcription through chromatin in a highly purified system (Orphanides et al, 1998). It now appears that FACT facilitates RNAPII-induced displacement of the H2A-H2B histone dimer from the nucleosome by a direct interaction with H2A and/or H2B (Belotserkovskaya et al, 2003). This is supported by experiments demonstrating that FACT activity is impaired when nucleosomes are covalently cross-linked and therefore cannot be displaced (Orphanides et al, 1999; Belotserkovskaya et al, 2003). Studies in yeast also suggest that not only does FACT disrupt nucleosomes to allow RNAPII access to DNA, but it also reassembles the nuclesomes afterward (Formosa et al, 2002). Additional studies in yeast implicate FACT in the regulation of elongation through chromatin. Sptl6 has been found to genetically and physically interact with the elongation factor DSIF, as well as the chromatin remodeler Chdl (Orphanides et al, 1999; Krogan et al, 2002b; Lindstrom et al, 2003; Simic et al, 2003). In addition, FACT is associated with actively transcribed genes on Drosophila melanogaster polytene chromosomes (Saunders et al, 2003). These data all support the idea that FACT functions after transcription initiation, allowing RNAPII to elongate through chromatin templates (reviewed in Belotserkovskaya et al, 2004).

85: Elongator

The Elongator complex was initially identified in yeast as a multi-subunit complex associated with the hyperphosphorylated, elongating form of RNAPII. High speed eentrifugation and high salt concentrations allowed for separation of a mediator-containing, hypophosphorylated complex from a hyperphosphoiylated RNAPII complex associated with chromatin. Elongator, consisting of Elpl, Elp2, and Elp3 purified with both fractions, while little, if any, mediator was found in the chromatin-associated fraction (Otero et al., 1999). Further isolation studies identified three more components of Elongator- Elp4, Elp5, and Elp6—which exist as a discrete subcomplex of the six-subunit "holo-Elongator" (Winkler et al., 2001; Krogan and Greenblatt, 2001). Characterization of Elongator components identified genetic interactions with the transcription elongation factor TF1IS and a delay in growth recovery when elpA cells were subjected to changes in growth medium. A similar delay in gene activation was also observed, leading to the conclusion that Elongator is a novel form of the RNAPII holoenzyme functioning in transcription elongation (Otero et al., 1999). Because Elongator was isolated from yeast chromatin, it has been speculated that it plays a role during transcription through chromatin. The discovery that Elp3 contains motifs with homology to the GNAT family of histone acetyltransferases (HATs) (Wittschieben et al., 1999; 2000) and acetylates histone H3 and H4 in vivo (Winkler et al., 2002), supports this idea. In addition, deletion of the histone H4 tail is lethal in the absence of Elp3 (Wittschieben et al., 1999). However, a role for Elongator in transcription elongation through chromatin is somewhat controversial. Subcellular localization studies demonstrated that it is primarily cytoplasmic, and chromatin immunoprecipitation experiments failed to detect Elongator recruitment to DNA (Pokholok et al., 2002). In addition, a recent proteomics approach failed to detect an interaction between Elongator and RNAPII in yeast (Krogan et al., 2002b). These data are refuted by more recent results from the Svejstrup laboratory in which an RNA immunoprecipitation (RIP) procedure was utilized to find Elongator associated with RNA along the entire coding region of genes in vivo (Gilbert et al., 2004).

B6: Histone Methyltransferases Setl and Set2

Another histone modification that has been linked to transcription elongation is methylation (reviewed in Gerber and Shilatifard, 2003). The pattern of methylation on histone lysine residues is specific to each transcriptional state of a gene and this methylation appears to be an important transcriptional regulator (reviewed in Lachner et al., 2003; Sims et al., 2003). In yeast, Setl is a specific histone H3 lysine 4 (H3-K4) methyltransferase that associates with the RNAPII CTD.

It was originally purified in a large complex termed COMPASS (Complex of Proteins Associated with Setl) (Briggs et al., 2001; Miller et al., 2001; Roguev et al., 2001; Krogan et al., 2002a; Nagy et al., 2002; Noma and Grewal, 2002; Ng et al., 2003). Setl may play a role in the early stages of transcription elongation, as it primarily interacts with the Ser 5 phosphorylated form of RNAPII associated with promoters and early elongation complexes (Krogan et al., 2003a; Ng et al.,

2003). In addition, Setl has recently been shown to catalyze tri-methylation of H3-K4, while di-methylation occurs on a genome-wide scale, tri-methylation is present exclusively at active genes (Santos-Rosa et al.,

2004). Set2 is an H3-K36-specific methyltransferase (Strahl et al., 2002) that preferentially associates with the Ser 2 phosphorylated form of RNAPII, indicating a role in transcription elongation. In addition, deletion of Ctkl, the CTD Ser 2 kinase, results in defective H3-K4 methylation (Li et al., 2003) and disrupts the interaction between Set2 and RNAPII, thus linking histone methylation to transcription elongation (Krogan et al., 2003b). The activities of Setl and Set2 have also been found to require the Pafl complex, although a precise role for this interaction is not completely understood. Continuing identification of RNAPII-associated factors will allow for better understanding of how transcription initiation and elongation are coordinately regulated.

RNA Polymerase II and mRNA Processing mRNA processing, which includes 5' end capping, splicing, and 3' end formation by cleavage and polyadenylation is tightly coupled to transcription. Like the many protein complexes that associate with RNAPII to regulate transcription initiation and elongation, a large number of mRNA processing factors are also associated with the elongating RNAPII in a complex often referred to as the "mRNA factory" (Bentley, 2002; Zorio and Bentley, 2004). It is now becoming increasingly clear that the events comprising mRNA processing occur co-transcriptionally while the dynamic mRNA factoiy transcribes, processes, and packages transcripts for proper export from nucleus to cytoplasm. The coordination of processing events is directed primarily by the RNAPII CTD, which serves as a platform for many processing factors. Indeed, deletion of the CTD in many vertebrate cells inhibits capping, splicing, and poly (A) site cleavage (McCracken et al., 1997a; McCracken et al., 1997b). The phosphorylation state of the CTD is critically important in specifying where and when the array of processing factors will associate with RNAPII throughout the transcription cycle.

A: 5' End Capping

RNAPII transcripts are capped at their 5' end by a methyl guanosine cap which serves to stabilize the mRNA against 5'-to- 3' exonucleolytic degradation and promote splicing, 3' end formation, transport and translation (Lewis et al., 1995; reviewed in Shuman, 2001). Nascent pre-mRNA is capped early in transcription, when only about 25-30 bases of RNA have been synthesized (Coppola et al., 1983). Capping occurs via the activities of three enzymes acting in order: RNA triphosphatase (RT), guanylyltransferase (GT), and 7-methyltransferase (MT). In yeast, RT and GT are represented by two polypeptides, Cetl and Cegl, respectively, which form a heterodimer referred to the capping enzyme (Ho et al., 1998). Capping enzymes are recruited to the transcription complex shortly after transcription initiation by binding to the phosphorylated CTD (McCracken et al., 1997b; Yue et al., 1997; Cho et al., 2001). This recruitment requires Kin28, the TFIIH-associated kinase that phosphoiylates the CTD at Ser 5 (Ho and Shuman, 1999; Moteki and Price, 2002). Once a cap has been added, release of the capping enzymes correlates with the removal of Ser 5 phosphates from the elongating polymerase; however, whereas GT is rapidly released, MT may remain associated with RNAPII even at the 3' end of the transcript (Schroeder et al., 2000; Komarnitsky et al., 2000). The capping enzymes appear to manipulate RNAPII function, perhaps by repressing transcription reinitiation (Myers et al., 2002), which serves as a checkpoint to ensure proper cap addition (reviewed in Orphanides and Reinberg, 2002).

B: 3' End Formation: Cleavage and Polyadenylation

Most protein-encoding mRNAs possess a uniform 3' end with a poly(A) tail. Prior to the addition of this tail, pre-mRNA must be cleaved. In mammalian cells, cleavage is directed by a highly conserved AAUAAA sequence and a GU- rich downstream sequence element (DPE) within the RNA. In yeast, the sequence elements that constitute the poly(A) signal are more complex and not as well defined (Graber et al., 2002; also reviewed in Guo and Sherman, 1996; Keller et al., 1997; Zhao et al, 1999). Cleavage and polyadenylation in mammalian cells is mediated by a complex of proteins including poly(A) polymerase (PAP), cleavage and polyadenylation specificity factor (CPSF), cleavage stimulation factor (CstF), and cleavage factors I and II (CFIm and CFIIm). Yeast homologs of these factors have also been identified, including Rnal4, Rnal5, Pcfll, Clpl, and

Papl. Like capping, 3' end formation requires factors that associate with the RNAPII CTD. More specifically, several components of the 3' end processing machinery have been shown to interact with the Ser 2 phosphorylated CTD (Barilla et al., 2001; Licatalosi et al., 2002). Despite the discovery of many of these interactions with RNAPII and transcription factors, the precise mechanism by which cleavage and polyadenylation occur remains somewhat of a mystery.

0 0

Post a comment