Transcription Elongation Factors in Health and Disease

Abstract

Gene expression is a complex process that establishes and maintains a specific cell state. Transcription, an early event during the gene expression, is fine-tuned by a concerted action of a plethora of transcription factors temporally and spatially in response to various stimuli. Most of the earlier research has focused on the initiation of transcription as a key regulatory step. However, work done over the last two decades has highlighted the importance of regulation of transcription elongation by RNA Pol II in the implementation of gene expression programs during development. Moreover, accumulating evidence has suggested that dysregulation of transcription elongation due to dysfunction of transcription factors can result in developmental abnormalities and a broad range of diseases, including cancers. In this chapter, we review recent advances in our understanding of the dynamics of transcription regulation during the elongation stage, the significance of transcriptional regulatory complexes, and their relevance in the development of potential accurate therapeutic targets for different human diseases.

1. Introduction

There are ~20,000 protein-coding genes in the human genome [1]. Cells modulate the expression of these genes in spatial and temporal manner in response to various stimuli. Gene expression is a highly regulated and complex process, which begins with the opening of chromatin, the transcription of the primary RNA transcript (hnRNA) from DNA, followed by processing of the hnRNA into mRNA, which is then translated into a protein that dictates cell functions. There has been an extensive study on precise control of gene expression at different stages by a plethora of factors leading to the concept of “gene-class specific” or selective gene control [2, 3, 4, 5, 6]. Tight control of gene expression is indispensable for normal cellular functions, and any dysregulation may lead to a wide range of diseases. The recent surge in knowledge and understanding of diseases that are caused by a mutation in regulatory sequences, transcription factors, cofactors, chromatin regulators, and non-coding RNA, such as diabetes, autoimmune disorders, neurological disorders, obesity, cardiovascular disease, and cancer, has altered our view of the underlying cause and primary focus of therapeutic targets.

It is imperative to understand the process of regulated gene expression to get insights into the mechanisms involved in the dysregulation of gene expression in various human diseases and to develop potential therapeutic targets for these diseases. Transcription has been considered the most important rate-limiting step during the expression of a gene. For a long time, initiation of transcription was considered as the key regulatory event during transcription and thus was the focus of major research in this field. However, for over a decade, the focus of research has shifted toward other steps during transcription, such as elongation and termination. Moreover, the regulation of transcription elongation has emerged to be the center of most therapeutic studies targeting fine-tuning of gene expression in diseases such as cancers.

In this chapter, I begin with a brief review of steps involved in gene regulation and the fundamentals of transcriptional control of gene expression. I further focus on a detailed understanding of regulation of transcription elongation by different transcription elongation factors and complexes and how a mutation or dysfunction of any of these factors contributes to altered transcription leading to progression of various diseases and cancer. I also highlight the recent advances in the development of precision tailored therapeutics by mediating transcriptional control of a gene.

2. Stages of regulation of gene expression

In eukaryotes, modulation of gene expression occurs at seven different steps (Figure 1).

Figure 1.

Schematic representation of steps involved in gene regulation. Regulation of gene expression occurs at seven different steps.

2.1 Chromatin structure

DNA wraps around proteins called histones to form nucleosomes. Each nucleosome is further condensed into chromatin. The condensation of eukaryotic DNA in chromatin acts to suppress the expression of genes by acting as a physical barrier to the transcription machinery [2]. The opening of chromatin, which allows access to genomic DNA, is indispensable for gene expression and formation of RNA from DNA template. This unwrapping of DNA from histone proteins is called chromatin remodeling and is carried out by enzymes that interact with histones and covalently attach functional groups to the amino terminal tail of histones. These histone-modifying proteins form complexes known as chromatin remodeling complexes. The most common modifications include methylation or acetylation of lysine residues on histone tail [3]. The outcome of these two modifications is entirely different: acetylation results in an open conformation of chromatin, thereby causing activation of gene expression; methylation results in a more compact chromatin conformation, hindering DNA accessibility to transcription factors and thus repressing transcription. Apart from histone modifications, methylation of DNA may also lead to a transcriptionally inactive state. A balance between the active and inactive state of chromatin is of profound importance for the maintenance of a healthy cellular environment. Dysfunctional chromatin remodeling complexes have been implicated in several disease conditions, including Williams syndrome [4, 5], Rett syndrome [6] breast cancer [7], and several other primary tumors [8].

2.2 Transcription

Transcription is the key regulatory step for gene expression in eukaryotes. It involves a concerted action of different proteins, such as transcription factors, mediator complex components, and RNA Pol II to produce RNA using DNA as a template. In eukaryotes, RNA Pol II is responsible for the synthesis of protein-coding genes as well as some non-coding RNAs such as small nuclear RNA (snRNA), microRNA (miRNA), cryptic unstable transcripts (CUTs), small nucleolar RNA (SnoRNA), and stable unannotated transcripts (SUTs) [9]. RNA Pol II mediated transcription is composed of three main steps: initiation, elongation, and termination, all of which are subjected to regulatory controls. During the stage of transcription initiation, the RNA Pol II in association with different transcription factors is recruited to the promoter region of the gene and forms a complex called pre-initiation complex (PIC). This opens the DNA, and the template strand positions itself in the active site of RNA Pol II, which then initiates the synthesis of the first few nucleotides of RNA. When the RNA length reaches ~10 nucleotides, RNA Pol II escapes the promoter and enters the gene body leading to productive elongation. Once Pol II reaches the end of the gene, RNA Pol II ceases RNA synthesis, which signals the release of the nascent RNA transcript as well as Pol II from the DNA template, thus terminating the process of transcription. All three stages of transcription are subject to tight control. Perhaps, due to its foremost position in the transcription cycle, the initiation step is extensively studied for mechanism and regulation. More recently, the transition of initiation to elongation has emerged to be the hub of major research in the field of transcription regulation. However, the mechanism and regulation of transcription termination have been less investigated.

A detailed description of regulation during transcription is described in the subsequent sections of this chapter.

2.3 RNA processing

The primary or nascent transcript is further processed to form functional mRNA. During processing, the primary transcript undergoes three types of modifications: 5′ end modification (capping), removal of introns (splicing), and 3′ end modification (polyadenylation) [10].

The newly synthesized RNA is stabilized by the addition of a 7-methylguanosine cap, which protects nascent from attack by nuclear exonucleases and helps in promoting transcription, splicing, polyadenylation, and nuclear export [11, 12]. Factors responsible for the capping of 5′ end of RNA, for example, eIF4E-antisense oligonucleotides, are being extensively used therapeutically in clinical trials that aim to curb dysregulated gene expression in cancer [13]. Small ribonucleoprotein particles (snRNPs) along with auxiliary proteins form a spliceosome complex, which mainly carries out the process of splicing by recognizing the splice sites and catalyzing the splice reaction [14]. Any dysregulation of the splicing mechanism results in diseased conditions [15]. Polyadenylation of 3’end of nascent RNA includes cleavage at polyadenylation site (PAS) of RNA and addition of poly (A) tail [16, 17, 18]. Alternate polyadenylation (APA) is yet another mechanism adopted by the cell to produce diversity in the mRNA pool. APA results in different isoforms of the same gene with varying 3’UTR [19]. Poly(A) tails are responsible for stability, translation efficiency, and degradation of RNA. Alteration in polyadenylation is associated with a plethora of diseased conditions, such as neonatal diabetes, fragile X-associated premature ovarian insufficiency, IPEX (immune dysfunction, polyendocrinopathy, enteropathy, X-linked), ectopic Cushing syndrome, and several cancer conditions such as endocrine tumor [20].

2.4 RNA transport

Once the mature RNA is made post RNA processing events, it is rearranged in an export competent mRNP complex with RNA-binding factors and shuttling proteins and transported to the cytoplasm. This process is tightly regulated. After the transport, in the cytoplasmic side, the DEAD-box helicase remodels the mRNP to dissociate RNA binding and shuttling proteins, preventing the mRNA from returning to the nucleus [21]. Furthermore, cap-binding protein (CBP), which binds to the 7-methylguanylate cap on 5′ end of processed RNA, is recognized by nuclear pore complex and exported to the cytoplasm, where it is replaced by translation factors eIF4E and eIF4G [21]. Few lines of research have shown that the transport machinery co-transcriptionally associates with mRNA [21, 22, 23], while others have shown that 3′ end processing of transcript marks the event responsible for the loading of export complex [24, 25]. For example, shuttling proteins, Hrp1p and Nab2p, associate with poly (A) tail [26, 27].

Any defect in mRNA transport or nuclear retention results in human disease such as lethal congenital contracture syndrome 1.

2.5 RNA stability and degradation

The life span of mRNA in cytosol determines the protein turnover. This is an important step to control gene expression since modulation of mRNA abundance allows cell to adapt and respond to various situations adequately. Generally, proteins with housekeeping functions are encoded by mRNA with a long half-life, while the proteins required only at specific developmental stages are encoded by mRNA with a short half-life [28, 29]. mRNA decay is a highly regulated process, resulting from interactions between mRNA, RNA-binding proteins, non-coding RNA, and various decay factors. The stringency of mRNA degradation depends on the presence of regulatory RNA elements, consisting of specific sequences found anywhere in mRNA, including 5′ and 3’ UTR. These sequences are recognized by RNA-binding proteins, forming mRNP complexes [30]. These RNA-binding proteins determine the fate of bound mRNA—to be translated, decayed, or stored in untranslated form as cytoplasmic granules. mRNA degradation begins with deadenylation or shortening of poly(A) tail carried out by Pan2/Pan3 complex and Ccr4/Pop2/Not complexes. After deadenylation, the mRNA is degraded either by 3′ → 5’ mRNA decay pathway mediated by exosome or 5′ → 3’ mRNA decay pathway mediated by exonuclease Xrn1 after decapping by Dcp1/Dcp2 decapping complex [31]. More recent studies have focused on the role of non-coding RNA molecules called miRNA in regulation of mRNA degradation and subsequently gene expression. miRNA works by either repressing translation or promoting degradation of mRNA having sequence complementary to miRNA [32]. Dysregulation of mRNA stability has been implicated in several diseases, including tumors. For instance, in myeloma and human T-cell leukemia, the stability of c-Myc RNA (an oncogene) due to loss of 3’UTR, which is responsible for its decay, results in its stability up to 4–8-fold higher compared with the wild type [33, 34].

2.6 Translation

Regulation of translation is a crucial mechanism for spatial control of gene expression [35]. There has been an explosion of studies highlighting the importance of regulation of translation in various physiological processes such as in normal development [36], in apoptosis [37], in stress response [38], etc. Dysregulation of translation has been implicated in different cancers [39]. In cells lacking nucleus, such as neurons, regulation of gene expression by translation has been shown to be of utmost importance. Several studies have demonstrated the initiation of translation as one of the crucial regulatory steps of protein synthesis [38, 40, 41, 42]. The role of P-bodies and small RNAs in translation regulation has also been elucidated. Initially, P-bodies were discovered as small foci rich in mRNA decay enzymes [43, 44, 45, 46, 47, 48]. When encountered with stress, yeast cells block translation initiation, as is evident by reduced polysome number and increased size of P-bodies [49]. Brengues et al. [50] have demonstrated that after the removal of stress and absence of new transcription, there is a decrease in the size of P-bodies and the reformation of polysomes [50]. This shows that P-bodies also act as storage sites for mRNA without undergoing degradation. On the other hand, small RNAs also the regulate translation and stability of mRNA [51, 52].

2.7 Posttranslational modifications and protein degradation

After its synthesis, the polypeptide is folded and modified by the addition of various chemical groups or by removal of certain amino acids from polypeptide (proteolytic cleavage). These protein modification steps act as targets for regulation. For example, phosphorylation of eIF2 results in inactivity and thus blocking of translation [53]. However, in some instances, phosphorylation may enhance the activity of a protein. Moreover, when not required by the cell, these proteins undergo degradation via the process of polyubiquitination, which is again regulated.

3. Regulation of transcription

In eukaryotes, there are three types of RNA polymerases responsible for transcription: RNA polymerase I is responsible for the synthesis of rRNA; RNA Pol II is responsible for the production of protein-coding mRNA, long non-coding RNA, snRNA, and miRNA; and RNA Pol III (RNA Pol III) is responsible for the synthesis of tRNA, some small non-coding RNA, 5S, and 5.8S RNA. Although transcription by all these enzymes is amenable to regulation, we will focus on the transcription of protein-coding genes in this chapter. Transcription is one of the most critical steps during gene expression and is regulated by a plethora of transcription factors working in a concerted manner with RNA Pol II to ensure proper initiation, elongation, and termination of transcription. RNA Pol II transcription involved initiation, elongation, and termination of transcription. For a long time, regulation of transcription was majorly focused on the initiation step. However, for over a decade, there has been an increase in mechanistic insights of regulation of transcription elongation, marking it as another regulatory event during transcription.

3.1 Transcription initiation

Initiation of transcription begins with recognizing and binding of RNA Pol II to the promoter. However, since RNA Pol II cannot recognize the promoter, it needs assistance from other proteins called as “general transcription factors” (GTFs) [54]. The GTFs are evolutionarily conserved proteins, and their ordered recruitment to RNA Pol II is necessary to initiate RNA synthesis. There are six types of GTFs: TFIIA, TFIIB, TFIID, TFIIE, TFIIF, and TFIIH. Table 1 summarizes the function of these GTFs.

General Transcription Factor	Subunits	Function
TFIIA	3 (α, β, γ)	Interact with TBP subunit of TFIID and stabilize PIC
TFIIB	1	Help in transcription start site selection, recruitment of TFIIH/RNA polymerase IIcomplex and assist in promoter escape
TFIID	15 [TBP (TATA box binding protein) and 14 TAFs (TBP associating factors)]	Recognize promoter by binding to TATA box, mediate interaction between activators and basal transcription machinery
TFIIE	2 (TFA1, TFA2 in Saccharomyces cerevisiae or α, β in human)	Facilitate TFIIH recruitment
TFIIF	3 in S. Cerevisiae (Tfg1, Tfg2, Tfg3); 2 in human (RAP74, RAP20)	Promote binding of RNA polymerase IIto TFIID-TFIIB DNA ternary complex, help in transcription start site selection
TFIIH	10 [7 core subunits, 3 kinase modules (CAK)]	DNA dependent ATPase, ATP dependent DNA helicase and CTD kinase, involved in promoter escape, promoter proximal pausing, elongation, RNA processing and termination

Table 1.

General transcription factors in eukaryotes and their functions.

Recruitment of RNA Pol II and general transcription factors (known as PIC or “Pre-Initiation Complex”) to the promoter is highly regulated during the initiation of transcription. Figure 2 shows the process of transcription initiation in eukaryotes. Specific transcription factors bind to the regulatory regions of the promoter. They work by modulating the assembly and activity of transcription machinery, either through direct interaction with components of basal transcription machinery or through action on chromatin [55, 56]. The mediator complex facilitates the connection between the activators associated with UAS and the PIC bound to the promoter region. The role played by the mediator complex, activators, and repressors marks yet another event of regulation during the initiation of transcription. The mediator complex is a multiprotein complex primarily comprised of head, middle, tail, and kinase modules. The head and middle domain form the core of the mediator complex, while the tail and kinase domains serve as regulatory modules [57, 58, 59, 60]. Although the mediator complex is conserved across evolution, the number of subunits vary in different species, comprising 19, 25, or up to 30 subunits in S. pombe, Saccharomyces cerevisiae, or humans, respectively. Five subunits have been reported to be metazoan-specific: MED23, MED25, MED26, MED28, and MED30 [61]. Figure 3 represents the subunit composition of the mediator complex in yeast and human. The mediator complex functions as a bridge between basal transcription machinery and specific transcription factors, resulting in the assembly of the pre-initiation complex (PIC) at the core promoter [62]. Besides recruitment of basal transcription machinery during initiation, the role of the mediator complex has been established in transition from initiation to elongation [63], during elongation [64, 65], as well as during termination [66], mRNA export [67], DNA repair [68], and S. pombe cell separation [69].

Figure 2.

Transcription initiation. Diagram representing stepwise recruitment of factors leading to initiation of transcription. Transcription activators bind to the UAS, which recruits co-activators including chromatin remodeling complexes. This opens the chromatin and facilitates association of RNA Pol II along with GTFs at the gene promoter, forming PIC. Association between activators and PIC is mediated by mediator complex, which phosphorylates RNA Pol II at CTD and initiates the process of transcription.

Figure 3.

Subunit composition of mediator complex in yeast and human. Diagram representing different modules comprising various subunits of mediator complex in yeast and human.

3.2 Transcription elongation

Promoter clearance is the transit phase between transcription initiation and elongation. Several factors determine the ability of RNA Pol II to move out of the promoter and enter the elongation phase. Co-crystallization of RNA Pol II with TFIIB has demonstrated that TFIIB obstructs the exit channel for newly synthesized RNA, and therefore, removal of TFIIB is imperative to promoter escape [70]. Transcription elongation is divided into two phases: early elongation and productive elongation [71]. The phosphorylation status of CTD of RNA Pol II is also an important determinant of the stage of transcription: during PIC assembly and initiation of transcription, CTD remains unphosphorylated [72, 73], whereas phosphorylation at serine 5 marks promoter clearance [74]. Serine 5 phosphorylated RNA Pol II associates with promoter-proximal regions during transcription initiation and early elongation, while serine 2 phosphorylation is associated with distal promoter regions during transcription elongation [75]. The CDK8 subunit of kinase module of mediator complex and CDK7 subunit of TFIIH are responsible for phosphorylation of CTD during initiation and early elongation phase, signaling RNA Pol II to clear promoter [76]. Phosphorylation of serine 2 on RNA Pol II CTD by P-TEFb triggers the passage into productive elongation from the early elongation phase [77, 78].

During the early elongation phase, RNA Pol II encounters various hurdles including transcriptional pause, arrest, or termination. The phenomenon of “promoter-proximal pausing” is characterized by transient pausing of RNA Pol II after synthesis of 20–60 nucleotides long RNA before resuming transcription elongation [71, 79]. Promoter-proximal pausing is well established in metazoans but less frequently observed in yeast [80, 81].

Evidence from biochemical studies has shown that pausing/arrest occurs as a result of backtracking of RNA Pol II on DNA template, thereby displacing the 3′ end of nascent RNA from the active site in RNA Pol II. This process can be spontaneously reversed (“pausing”) or not (“arrest”) [82]. The release of RNA Pol II from pause/arrest has emerged as an important mechanism to ensure continued and effective transcription elongation.

3.2.1 Transcription elongation factors

Biochemical experiments have shown that purified RNA Pol II proceeds optimally at rates of only 100–300 nucleotides/min in vitro as compared with the rate of 1200–2000 nucleotides/min in vivo [83, 84, 85]. The in vitro slow rate of mRNA synthesis was reported to be due to frequent pausing or arrest of RNA Pol II along the DNA [86, 87], suggesting elongation to be an inherently discontinuous process. An array of proteins known as “transcription elongation factors” function to regulate the rate of elongation.

These transcription elongation factors have been classified into different classes:

• Factors that assist RNA Pol II to traverse through transient pausing sites, e.g., P-TEFb [73], DRB sensitivity inducing factor [88, 89];

• Factors that can assist RNA Pol II to transcribe through chromatin, e.g., FACT, Swi/Snf [90, 91].

• Factors that can increase the overall rate of transcribing RNA Pol II, e.g., Elongin [92, 93], ELL [94, 95], ELL2 [96].

• Factors that suppress the activity of RNA Pol II, e.g., NELF [97].

  Some transcription elongation factors increase transcription of all protein coding genes, while others expedite transcription of only a set/class of genes [86]. Moreover, there are several GTFs as well as elongation factors that regulate transcription during either early or productive elongation phase of transcription.

3.2.2 Transcription factors regulating early elongation phase and pause release

3.2.2.1 TFIIF

TFIIE, TFIIF, and TFIIH have been implicated in post-initiation functions regulating early elongation and promoter escape during transcription [98].

In mammals, it is composed of two subunits, RAP30 and RAP74 [99]. Documented evidence suggests that TFIIF functions in suppressing RNA Pol II-associated transient pausing during active elongation via direct interaction with the ternary complex [100, 101]. Several lines of research have established that distinct functions of TFIIF during initiation and elongation are carried out by its different functional domains [100, 101, 102]. For instance, the initiation activity of TFIIF is mediated by a DNA binding domain in the C-terminal region of RAP30 while the elongation activity is carried out by the RNA Pol II binding regions in the upstream sequence of the RAP30 C-terminus [98].

3.2.2.2 TFIIH

TFIIH is a conserved protein composed of 10 subunits, seven of which form the core, while three subunits comprise a catalytic module called CDK activating kinase (CAK) comprising CDK7, ATP-dependent helicase XBP, and XPD [103].

The regulated recruitment of TFIIH to the promoter is orchestrated by TFIIE [104, 105, 106]. TFIIE and TFIIH work together in suppressing premature arrest of the early RNA Pol II, thereby facilitating promoter escape [98, 107]. The CDK7 phosphorylates CTD and initiates elongation [108]. Furthermore, it has been shown that the CDK7 subunit of TFIIH recruits ELL at sites of DNA damage and helps in transcription restart after repair of damaged DNA [109].

3.2.2.3 DSIF

A nucleoside analog, 5,6-dichloro-1-β-D-ribofuranosylbenzimidazole (DRB), works by obstructing the transition from initiation to elongation by inhibiting the phosphorylation of CTD of RNA Pol II [110, 111, 112]. DRB sensitivity inducing factor (DSIF) was initially identified from HeLa cell nuclear extracts as a transcription factor that promotes pausing of RNA Pol II in response to DRB [88, 97]. The two subunits of DSIF, Spt4 and Spt5, regulate the activity of RNA Pol II by genetically and physically interacting with it [88, 89, 97, 113, 114]. Association of DSIF with RNA Pol II after promoter escape is controlled by CDK7 dependent phosphorylation of the Spt5 subunit [115, 116]. Reduction in levels of the nascent and processed snRNA transcripts upon knockdown of DSIF in HeLa cell lines has pointed toward the function of DSIF as a transcriptional elongation regulator [117].

3.2.2.4 NELF

Negative elongation factor (NELF) is composed of five subunits, namely NELF-A, NELF-B, NELF-C, NELF-D, and NELF-E [118]. Interestingly, NELF is essential in Drosophila melanogaster and mammalian cells but is absent in Saccharomyces cerevisiae, C. elegans, and Arabidopsis thaliana [119, 120]. NELF promotes pausing of RNA Pol II by binding to RNA Pol II-DSIF complex through its NELF-E subunit [97, 118]. Yamaguchi et al. [97] demonstrated that DSIF/NELF interacts with the hypophosphorylated CTD of RNA Pol II to suppress its elongation function [97]. DSIF/NELF-mediated pausing of RNA Pol II gives sufficient time for recruitment of capping enzyme and addition of 5’cap to the nascent RNA [119, 121].

3.2.2.5 P-TEFb

Positive transcription elongation factor (P-TEFb) was first identified from a partially purified transcription system as a kinase inhibited by DRB [97]. It is composed of two subunits: cyclin T and CDK9. P-TEFb alleviates the NELF/DSIF-induced RNA Pol II pausing by phosphorylation of Spt5 subunit of DSIF and Ser2 residue of CTD heptad resulting in dissociation of NELF and transition from paused state to productive elongation state [71, 97, 119, 122, 123, 124]. Moreover, the FACT complex cooperates with P-TEFb to mitigate NELF/DSIF-mediated inhibition of transcription elongation [125]. In the cell, P-TEFb is subjected to stringent regulation and exists in an active as well as the inactive state. In mammals, most of the inactive P-TEFb exists as a part of complex, which includes an inactive ribonucleoprotein complex called 7SK snRNP, which consists of 7SK snRNA, P-TEFb, HEXIM1/2 (hexamethylene bisacetamide inducible protein), LARP7 (La-related protein 7), and MePCE (methyl phosphate capping enzyme) [126, 127]. However, when rapid transcription induction is required, P-TEFb is released from the inactive complex and is recruited to the transcription site by a specific activator or chromatin remodeling protein, bromodomain-containing protein 4 (BRD4) [79, 128, 129]. P-TEFb is a component of yet another multiprotein complex called super elongation complex (SEC), which is involved in productive elongation by RNA Pol II [130]. P-TEFb has been implicated as a potential therapeutic target in multiple myeloma, autoimmune diseases, cardiac hypertrophy, and infectious diseases [131, 132, 133, 134, 135, 136, 137]. An emerging concept of indirect therapeutic targeting using synthetic transcription elongation factor has shown enhanced gene expression in diseased conditions where a particular gene is downregulated. A synthetic transcription elongation factor is composed of a programmable DNA-binding ligand attached to a small molecule that can bind to and recruit the transcription elongation machinery, thereby regulating the gene expression. Syn-TEF1 has been shown to engage P-TEFb via recruitment of BRD4 at GAA repeats and restored the expression of the FXN gene in Fredrich Ataxia cell line to the levels observed in healthy cells [138].

Figure 4 depicts the role of P-TEFb in regulating the transition from the early elongation phase to productive elongation.

Figure 4.

Role of P-TEFb in regulating transcription elongation. (a) P-TEFb is recruited to the paused RNA Pol II as a result of the negative effects of DSIF and NELF. (b) At pause sites, P-TEFb phosphorylates RNA Pol II CTD as well as NELF and DSIF. (c) Due to phosphorylation, NELF dissociates, DSIF functions as a positive factor in the absence of NELF, recruits other factors for mRNA elongation and processing, resulting in productive elongation.

3.2.3 Transcription factors regulating productive elongation phase

A separate class of transcription factors regulate the process of transcription during productive elongation, which is described below:

3.2.3.1 Elongin

Elongin increases the overall rate of transcription by suppressing the transient pausing of RNA Pol II through its interaction with RNA Pol II and stabilizing it in an active conformation for an extended period [139, 140]. Elongin is composed of three subunits: Elongin A, Elongin B, and Elongin C. Elongin A was found to be enriched at the transcriptionally active sites in association with an active form of RNA Pol II on polytene chromosomes of Drosophila melanogaster [141]. Elongin B and C were also found to be components of various Cullin2/Cullin5-based ubiquitin ligase complexes and interact with different proteins in the complex via the BC box motif. Elongin BC serves as an adaptor linking Elongin A to Cul2/Cul5 and RING finger protein Rbx1/2 containing modules [142]. Several studies have shown that the association of Elongin ABC with Cul5/Rbx2 contributes significantly to the degradation of stalled RNA Pol II at sites of DNA damage [143, 144].

3.2.3.2 CSB

A mutation in Cockayne syndrome A (CSA) or Cockayne syndrome B (CSB) genes results in an accelerated neurodegenerative disorder called cocaine syndrome [145, 146]. Clinical studies using cells from Cockayne syndrome patients exhibited a defect in transcription coupled repair (TCR) but not in global genome repair, indicating the role of CSA/CSB in TCR [147, 148]. CSB was shown to be transiently associated with DNA, RNA Pol II, and nascent RNA, and in vitro studies have shown that purified CSB stimulated the rate of elongation by RNA Pol II [139, 149]. Transcribing RNA Pol II stalls upon detecting lesions in DNA [150]. Blocked RNA Pol II is either retracted or dissociated by CSA and CSB proteins, thereby making the damage site on DNA accessible to repair proteins, accomplishing DNA damage repair and augmenting the resumption of transcription [151, 152, 153].

3.2.3.3 TFIIS

TFIIS, a zinc finger transcription factor, is known to stimulate the rate of RNA transcript synthesis. TFIIS is required for stimulating transcription elongation by reducing pause time, and it also increases the processivity of RNA Pol II on nucleosomes as well as stimulates translational elongation [124, 154, 155]. Interaction of TFIIS with ELL-Associated Factor 2 (EAF2) promotes transactivation by FESTA/EAF2 in murine embryonic stem cells [156].

3.2.3.4 ELL family

ELL (Eleven nineteen Lysine rich Leukemia) was first identified as a translocating partner to trithorax-like mixed-lineage leukemia (MLL) gene, located on 11q23 chromosomal locus observed in acute myeloid leukemia [157]. Functional characterization and mechanistic studies have shown that ELL plays a role during recruitment of PIC, promoter clearance, and release of RNA Pol II from pause sites, thereby stimulating the overall rate of transcription [94, 158]. The functions of ELL in connection with various disease conditions have also been reported. The expression of HIF-1α, as well as its downstream target genes, is elevated in the absence of ELL as observed in PC3 prostate cancer cell lines [159]. The Tax protein of Human T-cell Leukemia Virus Type 1 (HTLV-1) interacts with ELL and incorporates it into the p300 and P-TEFb containing complexes, which enhance the transcription of immediate early genes [160].

In humans, two other ELL homologs, namely ELL2 and ELL3, were identified based on sequence similarity with ELL [124]. ELL2 regulates pre-mRNA processing by assisting RNA Pol II to select weak promoter-proximal poly(A) sites in immunoglobulin heavy gene (IgH) [161, 162, 163]. ELL2 has been found to be upregulated in neuroendocrine prostate tumor, while its mRNA level was reduced in prostate adenocarcinoma and multiple myeloma plasma cells [164, 165]. The role of ELL3 has been implicated in breast cancer and B-cell lymphoma [166, 167].

3.2.3.5 EAF family

Yeast two-hybrid screens are carried out to identify proteins that associate with human ELL and identified two ELL interacting partners, namely EAF1 and EAF2 (ELL Associated Factor) [168, 169]. EAF2, also known as androgen-upregulated 19 (U19), was recognized as a novel testosterone-regulated protein that induces apoptosis in the prostrate [170]. EAF family of proteins act as cofactors to ELL, stimulating its transcription elongation activity in vitro [171]. The importance of EAF in diseases was first highlighted by a study that showed that the association of EAF with ELL is essential for the immortalization of hematopoietic progenitor cells and the development of acute myeloid leukemia [169, 172]. Since then, there has been an explosion in the studies describing the role of EAF1 and EAF2 in the development and progression of various tumors. A reduced expression of EAF2 in human prostate cancer specimen, lower survival rates upon complete loss of EAF2, and increased cell migration and proliferation in prostate tumor cell lines upon EAF2 knockdown underscore the role of EAF2/U19 as a tumor suppressor in the prostate [173, 174, 175, 176, 177]. A murine model for EAF2 knockout has further implicated the role of EAF2 in other diseased conditions such as B-cell lymphoma, hepatocellular carcinoma, prostate intraepithelial neoplasia (PIN), lung adenocarcinoma, and enlarged cardiac cells with an abnormal vascular system as well as abnormalities in spermatogenesis [178, 179]. EAF2 knockdown is also associated with heightened humoral immune response and excessive generation of autoantibody. The immune balance function of EAF2 is mediated by apoptosis of germinal center B cells [180]. A study has identified a frameshift mutation that resulted in the loss of EAF2 function in colorectal cancer as well as gastric cancer [181]. Recently, a study established that absence of EAF1 in mouse prostate triggers pre-neoplastic prostatic intraepithelial neoplasia lesions. However, a combined loss of both EAF1 and EAF2 significantly enhanced the proliferation and inflammation in the murine prostate and resulted in a more aggressive tumor when compared with the individual loss of either EAF1 or EAF2, indicating that coordination between the two homologs is required for maintaining homeostasis in the prostate [182].

3.2.4 Elongation complexes

Several studies have demonstrated the existence of different multiprotein complexes called “Super Elongation Complex” (SEC), which are recruited to RNA Pol II to enhance its catalytic activity and thereby the rate of transcription elongation. These complexes are composed of transcription elongation factors such as p-TEFB, ELL1/2/3, EAF1/2 along with other proteins such as AFF4, ENL, AFF1, AFF9 [79]. In different organisms, super elongation complexes vary in composition and display functional diversity. For instance, in mammals, different combinations of four members of AFF family proteins (AFF1-4) and three members of ELL family protein (ELL, ELL2, and ELL3) confer functional variation to the SEC and form SEC-like complexes, SEC-L2 and SEC-L3 as shown in Figure 4. In place of AFF1/4, AFF2 and AFF3 represent the AFF family component of SEC-L2 and SEC-L3, respectively. Interestingly, biochemical studies have shown the absence of the one or more ELL family members in SEC-L2 and SEC-L3. The AFF4 and ELL2 containing SEC have been implicated in transcription elongation checkpoint control (TECC) in mammals. A similar super elongation complex was detected in D. melanogaster, comprising ELL, EAF1, P-TEFb, Lilli (AFF family member), and EAR (ENL) proteins [183]. Additionally, there is another complex present in D. melanogaster, known as Little Elongation Complex (LEC), comprising ICE1, ICE2, ELL, and EAF [184]. The role of LEC is found both in the initiation and elongation of snRNA transcription. Yet another elongation complex identified in hematopoietic cells called as “AEP complex” contains P-TEFb, AFF4, Afq31, and ENL protein [185]. The role of SEC in releasing paused RNA Pol II is very well described by several studies. However, other studies have also identified the role of SEC in rapid transcription by non-paused RNA Pol II in Drosophila embryos and in mouse embryonic stem cells [186, 187]. Recent work by Gopalan et al. [188] has recognized a rudimentary SEC in S. pombe consisting of only three members, ELL, EAF, and a newly identified AFF4 homolog EBP1 (ELL binding protein) [188]. The SECs were first reported two decades ago by their association with the viral transactivator of transcription (Tat) protein of HIV-1 and MLL-fusion partners involved in leukemogenesis [183, 189, 190, 191, 192]. During HIV infection, the Tat protein interacts with the P-TEFb component of SEC and recruits it to the HIV-1 long terminal repeat (LTR) to stimulate the expression of HIV-1 in host cells. Different AFF family members dictate gene-class specific recruitment of SEC [193]. For instance, AFF1 containing SEC is important for interaction with Tat protein in HIV pathogenesis. On the other hand, AFF4 containing SEC is involved in Hsp70 gene expression upon heat shock. Since several MLL fusion partners are components of SEC, the fusion protein-mediated recruitment of SEC to hox genes points toward dysregulation of developmental genes in leukemia [183, 191, 192]. Furthermore, mutations in SEC components also result in dysregulation of the transcription elongation process leading to several diseases. A missense mutation in AFF4 results in CHOPS syndrome, a developmental disorder [194]. Given the importance of regulation of transcription elongation in gene expression in healthy and diseased states, several research groups have worked toward delineating the roles of transcription elongation complexes as therapeutic targets in diseases including cancers. Small-molecule inhibitors targeting SEC, such as KL-1 and KL-2, have been implicated in targeted therapy of myc-driven tumors [195].

3.3 Transcription termination

In mammals, the transcription termination of protein-coding genes is mainly dependent on termination complex or “CPSF-CF complex” comprising cleavage and polyadenylation specificity factor (CPSF or CPF in yeast), cleavage stimulating factor (CstF or CF1A in yeast), cleavage factor I (CFI), and cleavage factor II (CFII) [196, 197]. CPSF directly binds to the body of Pol II and recognizes polyadenylation signal (PAS) (AAUAAA). Association of CPSF with PAS and Pol II triggers transcription pausing. CstF, which associates with Ser2 phosphorylated residues on Pol II CTD, recognizes GU-rich processing signal downstream of PAS and facilitates cleavage of transcripts. This dislodges the CPSF, and RNA Pol II is released from pause. In eukaryotes, two models have been proposed to facilitate the termination of Pol II-mediated transcription after cleavage of nascent RNA [196, 197, 198]. The allosteric model postulates that the transcription terminates following destabilization of the elongation complex triggered by the loss of elongation factors/conformational change in Pol II after transcription of PAS. The second “Torpedo model” posits that the transcription termination occurs due to the dismantling of Pol II elongation complex following the degradation of nascent RNA by exonuclease. SETX (Sen1 in yeast) resolves the R-loop formed by short RNA left after cleavage and DNA. This allows the recruitment of 5′-3′ exoribonuclease XRN2 (Rat1 in yeast), which chews up the nascent RNA downstream of the cleavage site and releases Pol II from DNA [198, 199]. However, an emerging view in this field is that the combination of these two models likely explains the process of termination [200, 201]. The mechanism of transcription termination is depicted in Figure 5. A detailed mechanistic understanding of transcription termination is still missing, despite a surge in studies highlighting the role of transcription termination in controlling gene expression.

Figure 5.

Mechanism of transcription termination at protein-coding genes in metazoans. Termination of transcription is triggered by the recruitment cleavage and polyadenylation factor complex (CPSF-CF complex) on the transcript. When RNA Pol II transcribes polyadenylation signal (PAS), the CPSF subunit binds to the PAS sequence on RNA while remaining bound to Pol II body. This results in pause of RNA Pol II. CstF subunit recognizes the GU-rich sequence downstream of PAS and creates a conformational change in the CPSF-CF complex, dissociating CPSF from Pol II and cleaving the nascent transcript between PAS and GU-rich sequence. R-loops formed by leftover transcript are resolved by Senataxin, allowing subsequent recruitment of XRN2 exonuclease. The remaining transcript downstream of the cleavage site is chewed up releasing Pol II and elongation complex by torpedo mechanism, thereby terminating the process of transcription.

3.4 Recent advances in regulation of gene expression

Numerous studies have shown that transcription is carried out in membraneless phase-separated compartments when biomacromolecules such as proteins and transcription factors undergo self-assembly via liquid-liquid-like phase separation (LLPS). These phase-separated condensates act as a hub of transcription, where specific transcription factors are exchanged to ensure proper temporal and spatial gene expression patterns [202]. It has been reported that Pol II forms phase-separated condensate at active genes through its low-complexity CTD to concentrate transcription regulators and initiate the process of transcription [203, 204, 205]. SEC can also dynamically extract P-TEFb from inactive HEXIM1-P-TEFb complex through phase separation, resulting in activation of transcription elongation. This was confirmed by disruption of SEC phase-separated droplets, which reduces the transcription efficiency [206]. The roles of LLPS in disease states have also been explored recently. Oncogenic fusion of ENL, a component of SEC, and MLL has been shown to enhance the phase separation capabilities contributing to overactivation of leukemic genes [206]. Mutation in the YEATS domain of ENL has also been shown to promote self-assembly of ENL into LLPS, resulting in augmented recruitment of SEC to promote transcription of oncogenes in Wilms’ tumor [207].

Another emerging concept of transcription regulation is through non-coding RNA and enhancer RNA (eRNA). Gene expression is regulated at multiple levels by long non-coding RNA (lncRNA): modulating chromatin structure and function, regulating transcription of neighboring or distal genes, RNA stability and translation [208, 209, 210, 211, 212, 213, 214]. The role of lncRNA has also recently been described in the regulation of nuclear condensates. Owing to the functional significance of lncRNA in transcription regulation, their role in the progression of diseases such as cancers and neurological disorders has been extensively studied. The transcription repression of tumor suppressors such as INK4A/ARF/INK4B has been associated with lncRNA ANRIL. ANRIL works by recruiting PRC1 and PRC2 complexes to promoters of these genes. Any dysregulation in ANRIL function may lead to silencing of these tumor suppressor genes contributing to tumor progression [215, 216, 217, 218]. BACE1-AS, an antisense of gene encoding BACE1 protein, a precursor of amyloid plates in Alzheimer’s disease (AD), promotes stability of BACE1 mRNA leading to accumulation of amyloid plates in the brains of AD patients. BACE1-AS also serves as a biomarker for AD and could be a potential therapeutic target to treat AD [219, 220]. lncRNAs are also involved in the suppression of gene expression through altered recruitment of transcription factors or Pol II or through reduced chromatin accessibility. For instance, following nerve injury, the lncRNA Silc1 is necessary for activation of the SOX11 transcription program for nerve regeneration [221].

4. Conclusion

Regulation of gene expression is imperative to the normal physiological functioning of the cell. In recent years, we have observed remarkable progress in our understanding of the regulation of gene expression. Transcriptional regulation has emerged to be the most critical and well-studied stage during the expression of a gene. In line of its foremost position in the transcription process, the initiation step is most extensively researched and was considered a major rate-limiting step during transcription. However, now the focus has shifted from activation to elongation stage, which has been established as another key regulatory event during gene expression under normal physiological state. Regulation of transcription termination has only recently started to become the focus of several studies, and more mechanistic insights are required to fully understand regulatory events during this stage of transcription.

Accumulating evidence in the last few years has suggested the prevalence of “gene-class specific” transcription elongation factors, adding another layer to transcriptional regulation. These transcription elongation factors have been reported to be part of several multiprotein elongation complexes, which enhance the probability of cross talk between these factors and increase the regulatory potential of cells. Identification of phase-separated assemblies of transcription complexes has provided a biophysical basis of dynamic regulation of transcription in response to cellular cues. Another less-explored layer of transcription regulation is through non-coding RNA. Although there has been a tremendous increase in our understanding of the regulatory capabilities of lncRNA, we still lack a rigorous investigation to relate sequence and structural features of non-coding RNA to their regulatory functions.

A recent surge in targeted gene therapy has opened the doors for therapeutic targeting of “gene-class specific” transcription elongation factors in various diseases including cancers. An interesting concept in therapeutic targeting is of synthetic transcription elongation factors, which modulate the expression of a particular gene by selectively engaging the transcription elongation machinery at a specific gene locus. However, more efforts and research are required to dwell into the genome-wide perturbation of gene expression as a result of the binding of synthetic transcription factors to other less specific loci. In the context of personalized medicines, disease-related non-coding RNAs are gaining attention due to their specific expression patterns, which makes them a good candidate for disease biomarkers. Growing mechanistic insight into the regulation of transcription elongation and the interplay between different steps of regulation of gene expression would offer new aspects for intervention with aberrant modulation of gene expression and precisely tuned therapeutics.

Acknowledgments

The author would like to thank Mr. Anurag Saroha for his help with the figures.