Introduction

Essential and fundamental to all organisms, transcription is the process for RNA synthesis from template DNA. At the heart of this activity is the large multisubunit enzyme called RNA polymerase. RNA polymerase, abbreviated RNAP and officially known as DNA-directed RNA polymerase, is found in all living organisms as well as many viruses. Present in bacteria, archaea, and even eukaryotes, these RNAPs all share similar protein core structures as well as mechanisms.

Fundamentals

RNAP is the multisubunit enzyme that transcribes template DNA into RNA. While bacteria (prokaryotes) and archaea contain only one RNAP, eukaryotes contain three RNAPs: RNAP I, RNAP II, and RNAP III. Though there are drastic differences between these multisubunit RNAPs, they are also many significant similarities. Clearly related and forming a family, these RNAPs have three highly conserved subunits. Though archaea RNAP is more complex than bacterial RNAP and is more closely related to eukaryotic RNAP II, about 50% of the enzyme is still conserved.[1] Thus, scientists have been able to glean many of RNAP’s important structure and function by studying the simpler bacterial RNAP. Because bacterial RNAP is most heavily researched and has the most significant clinical relevance, most of this activity will focus on this RNAP.

All RNAPs contain a core. This RNA-synthesizing core generates 5’ to 3’ RNA chains by hydrolyzing pyrophosphate from nucleoside triphosphates. Bacterial core RNAP is the simplest, comprised of five subunits: beta, beta prime, two alphas, and omega. Together, the large beta and beta prime subunits form a claw with the reactive magnesium ion.[2] In the center of the claw is a catalytically active site. The initiation of the core assembly occurs by the dimerization of the N-terminal domain of the alpha subunits, followed by beta and with omega subsequently serving as the chaperone for beta prime. A flexible linker tethers the C-terminal domains of alphas and serves important regulatory roles. Archaea and eukaryotic core subunits are also highly homologous to bacterial RNAP. Bacterial, archaeal, and eukaryotic RNAPs all resemble a crab claw with an enzyme active site located at the bottom cleft of the claw.[3][4] Here, a catalytic metal magnesium ion as well as an absolutely-conserved motif of NADFDGD and three invariant residues are found.[3][4] The architecture surrounding the cleft is also highly conserved among all three domains of life, suggesting that this mechanism of RNA synthesis gets conserved from bacteria to humans.

Issues of Concern

Issues of concern arise when RNAP mistranscribes, also known as transcription infidelity. However, this area of research has been difficult to study since the error of translation is significantly higher, obscuring phenotypic results.[5][6] Along with misincorporation of nucleotides, another mechanism of transcription infidelity involves bacterial RNAP elongation slippage on homopolymeric A/T tracks, disrupting open reading frames.[7]

Cellular Level

RNAP is a highly abundant enzyme which catalyzes the formation of phosphodiester bonds linking nucleotides together, subsequently producing linear chain. The growing RNA chain is extended by one nucleotide at a time in the 5’ to 3’ direction using nucleoside triphosphates (ATP, CTP, UTP, and GTP) as substrates. Estimates are that about 20 nucleotides undergo synthesis for each gene with over a thousand transcripts formed in an hour from one gene. Unlike DNA polymerase, RNAP has increased infidelity. One mistake is made every 10,000 nucleotides while DNA polymerase makes one mistake every 10,000,000 nucleotides. Despite lower accuracy, RNAP does have a proofreading mechanism. RNAP can back up, excise the misincorporated ribonucleotide, and insert the proper ribonucleotide.

Molecular Level

RNAP is a multisubunit enzyme that contains both protein-protein as well as protein-DNA interactions. The specificity factors of bacterial RNAP, sigma, contain four distinct regions: region 1, region 2, region 3, and region 4. Except for sigma factors belonging to the sigma54 family, all sigma factors contain a highly conserved region 2 and region 4 while some also additionally contain a conserved region 3.[2] These regions are vital because they allow the core to recognize specific promoters, allowing for differential upregulation and or downregulation during particular times. These promoter elements include the following: UP elements, -35 element, and the extended -10 motif.  The tethered C-terminal domains of the alpha subunits interact with the UP elements which are AT-rich sequences located from -40 to -60 relative to the +1 transcriptional start site.[8][9] Region 4.2 of sigma interacts with the -35 element, which is a highly conserved -35 TTGACA-30 sequence.[10] Region 3.0 interacts with the extended -10 sequence which is a -15TGn-13 as well as the most upstream basepair, the -12 T of the -10 element.[11] Region 2.4 interacts with the -10 element, which is a highly conserved sequence comprised of -12 TATAAT -7.[11]

Along with sigma-DNA interactions, sigma also makes various extensive protein contacts with core RNAP. Important interactions include residues within Regions 2.1 and 2.2 with the beta prime subunit as well as residues with region 3.2 threading through a channel comprised of beta and beta prime.[12] Regions in 4.1 and 4.2 interact with the beta-flap, a domain within the beta subunit, while the beta-flap tip makes contact by the very C-terminus (known as helix 5) of the sigma factor.[13] Interestingly, primary sigma factors contain a highly charged region 1.1, which interacts with the beta and beta prime cleft and acts as a DNA mimic, helping hold DNA that is downstream of the transcription start site.[14]

The overall structure of RNAP resembles that of a crab claw. The beta and beta prime subunits form two opposing pincer-like structures, bordering the main channel.[15][16] The larger of the pincer, the beta prime subunit, contains a highly mobile domain that can hinge around a flexible region; this is known as the switch region and has five discrete elements SW1-SW5.[17] A cleft between the two pincers houses the RNAP catalytic site which contains a magnesium ion.[18] Leading to this catalytic active site are two channels. The primary channel contains the DNA-RNA hybrid as well as the downstream DNA while the secondary channel proves access for NTP substrates as well as the nascent RNA.[19]

Function

RNAP’s function is to transcribe DNA into RNA. The template stranded DNA is read from 3’ to 5’ fashion while RNA synthesis occurs in a 5’ to 3’ manner. Many different kinds of RNA get produced: mRNA, tRNA, rRNA, sRNA, snoRNA, and other noncoding RNAs.

Mechanism

RNAP’s mechanism of transcribing DNA into RNA is highly conserved across prokaryotes, archaea, and eukaryotes. To begin the process of bacterial transcription, along with core polymerase, an additional specificity factor is also necessary. These specificity proteins recognize certain elements in the promoter DNA and determine the start site of transcription. In bacteria, these specificity factors have the name of sigma subunits. Sigma, together with core, form the bacterial RNAP. Though research has now identified hundreds of sigma, the main sigma factor, sigma70 in Escherichia coli or sigma A in other bacteria, is responsible for exponential growth and regulation of housekeeping genes.[20] Other alternate sigma factors are used during times of stress and/or different growth conditions.[20]

To initiate the process of bacterial transcription, sigma and core together first binds to double-stranded DNA.[21][2] C-terminal domains of the alpha subunits both interact with the UP elements, Region 4.2 interactions with the -35, and Region 3.0 interacts with the extended -10.[2] These interactions between core and promoter DNA are very unstable and short-lived, forming a product called the unstable closed complex (RPc).[21] The thinking is that the double-stranded DNA lies across the face of the RNAP. RPc quickly transitions to the stable open complex (RPo) where the DNA unzips, bends about 90 degrees, and forms a bubble from -11/12 to about +5 relative to the +1 transcriptional start site.[21][22] Along with DNA changes, there are also major protein conformational changes within polymerase: Region 1.1 moves downstream to enter that beta and beta prime channel and beta prime clamp undergoes large conformational changes, allowing the downstream DNA to be fully secured.[22] Nucleotide triphosphates then enter through the secondary channel of the core, leading to the synthesis of RNA and forming a complex called the initiating complex (RPi). These RNA products are very short and are also known as abortives. Once the abortives reach a certain size, they can ultimately push Region 3.2 out of its current position within the RNA exit channel.[23] This change thus opens up the RNA exit channel, leading to the release of both the sigma factor as well as RNA. Core polymerase continues moving along the template stranded DNA until it reaches the termination site. In Rho-dependent termination, a factor called Rho is responsible for disrupting the complex of RNAP/RNA/template DNA while in Rho-independent termination, a loop forms at the end of the RNA molecule, allowing for RNAP to fall off.[24]

Testing

Antibodies to RNAP can be commercially purchased and are routinely used in the laboratories to detect the presence of RNAP subunits. Furthermore, anti-RNAP III antibodies are considered to be highly sensitive and specific for systemic sclerosis or scleroderma and are also associated with diffuse cutaneous scleroderma and scleroderma renal crisis.[25] Within the diffuse cutaneous scleroderma population, the presence of anti-RNAP III antibodies correlates with a decrease in time of onset of initial symptoms and skin thickening peak.[26] Thus, patients with anti-RNAP III antibodies are at a higher risk of developing sclerodermal renal crisis.[27] Patients who are positive for anti-RNAP III antibodies had higher incidences of systemic scleroderma disease, ranging from 14% to 51%.[27]

Pathophysiology

Though mutations in eukaryotic RNAP have not been solely implicated in disease, dysregulation of ribosomal RNA synthesis via RNAP I has lead to a variety of different diseases. The most well-recognized diseases result from a loss of function mutations in ribosomes or factors associated with Pol I.[28] These diseases are known as ribosomopathies and include Diamond-Blackfan anemia, 5q minus syndrome, Treacher Collins syndrome, and Blooms and Werner syndrome.[28] There are also other ribosomopathies which are associated with mutations affecting RNA processing and modification; these diseases include Shwachman-Diamond syndrome, dyskeratosis congenita, cartilage hair hypoplasia, North American Indian childhood cirrhosis, Bowen-Conradi syndrome, and alopecia, neurological defect and endocrinopathy (ANE) syndrome. Unfortunately, ribosomopathies are rare, and treatment is typically palliative rather than curative.[28]

Clinical Significance

Prevention of gene expression is one way in which antibiotics can kill bacteria. Since transcription is an essential process for all organisms, the transcription machinery is an extremely attractive target for the development of new antibiotics. Rifampicin is a commonly used antibiotic against Mycobacterium tuberculosis. Belonging to the ansamycin class of antibiotics, rifampicin binds to the beta subunit of RNAP within the DNA/RNA channel and prevents the formation of the second or third phosphodiester bond, inducing the release of short abortives and subsequently blocking nascent RNA extension.[29][30][31] Interestingly, bacteria have quickly evolved to develop resistance to rifampicin. Mutations within the beta subunit of RNAP (S531L, H526Y, and D516V) account for about for 41, 36, and 9% of resistant tuberculosis strains, respectively.[31] Other antibiotics that block nascent RNA extension include sorangicin and GE23077.[32] Another commonly used antibiotic, fidaxomicin, also binds to Clostridium difficile RNAP. Instead of the beta subunit, fidaxomicin binds to the switch region of RNAP.[33] This binding prevents RPo formation. The proposed mechanism is that fidaxomicin prevents the correct spatial orientation of Region 2 and Region 4 for recognition of the -10 and -35 core promoter elements, respectively.[33][34] Other antibiotics that also target the RNAP switch region include squaramides, myxopyronin, corallopyronin, and ripostatin.[32] Though only very few antibiotics have reached the clinical market, there are also other many antibiotics which inhibit transcription by varying processes. The SB-2 series disrupt holoenzyme assembly, pseudoridmycin is a nucleoside analog, and salinamides targets the mobile elements of the primary channel.[32]

Along with prokaryotic transcription inhibition, researchers have discovered several compounds that target eukaryotic RNAP. Actinomycin D is a bacterial antibiotic used as an antitumor reagent. It intercalates DNA, thereby preventing the progression of both bacterial and eukaryotic RNAP I.[35][36] Along with actinomycin D, other chemicals also inhibit eukaryotic transcription. Alpha-amanitin binds to the bridge helix and trigger loop of RNAP II and III, preventing the incorporation of nascent RNA chains,[37][38] while 8-hydroxyquinoline chelates bivalent cations, manganese, and magnesium, within the active site of RNAP.[39]