L - Technische Universität Braunschweig

2013 

Dr. Dina Grohmann 

Junior Research Group Leader / Akademischer Rat a.Z. 

Institute of Physical and Theoretical Chemistry 

- NanoBioSciences - 

Technische Universität Braunschweig 

Hans-Sommer-Straße 10 

38106 Braunschweig 

Tel: +49 (0)531 391 5395 

Fax: +49 (0)531 391 5334 

d.grohmann@tu-bs.de 

APPLICATION 

for the W1 Junior Professorship in 

Physical Chemistry 

Faculty of Chemistry and Earth Sciences 

Jena University 

Cover letter 

Curriculum vitae 

Funding ID 

Research plan 

Research accomplishments 

List of publications 

Statement of teaching 

Teaching concept 

References 

Certificates 

Reprints

2 APPLICATION DR. DINA GROHMANN 

Physikalische und Theoretische Chemie -NanoBioSciences 


An Herrn Prof. Dr. J. Popp 

Institut für Physikalische Chemie 

Helmholtzweg 4 

D-07743 Jena 

Technische Universität Braunschweig 

Institut für Physikalische 

und Theoretische Chemie 

Abt. NanoBioSciences 

Hans-Sommer-Str. 10 



Tel. +49 (0) 531 391-5395 

Fax +49 (0) 531 391-5334 

d.grohmann@tu-braunschweig.de 

Braunschweig, 22.08.2013 

BEWERBUNG AUF DIE JUNIORPROFESSUR (W1) IN PHYSIKALISCHER CHEMIE 

Sehr geehrter Herr Professor Dr. Popp, 

Vielen Dank für Ihre persönliche Einladung, mich auf die Juniorprofessur in Physikalischer 

Chemie an der Universität Jena zu bewerben. Momentan leite ich eine interdisziplinäre 

Nachwuchsgruppe (1 Postdoktorandin, 4 Doktoranden, 1 Master- und 1 Bachelorstudent) am 

Institut für Physikalische und Theoretische Chemie / Abteilung NanoBioSciences an der 

Technischen Universität Braunschweig. 

Meine Arbeit gliedert sich in zwei Themenschwerpunkte: ein Teil meiner Forschung ist auf 

die Erforschung molekularer Mechanismen, die der Transkription und dem RNA-induzierten 

gene-silencing (RISC) zugrunde liegen, ausgerichtet. Auf der anderen Seite beschäftige ich 

mich in zunehmendem Maße mit der Entwicklung neuer Nanostrukturen, die als Biosensoren 

genutzt werden können. Diese Arbeiten schließen DNA Origami und Protein-basierte 

Nanostrukturen ein. 

In diesem Zusammenhang nutze ich v.a. fluoreszenzbasierte Einzelmolekülspektroskopie, 

um die Architektur von Transkriptionskomplexen zu studieren und 

Konformationsänderungen, die die katalytische Aktivität und intermolekulare Interaktionen 

begleiten, zeitaufgelöst zu verfolgen. Einen ähnlichen Ansatz verfolge ich für das Argonaute 

Protein und den RISC-Komplex, der jedoch nicht in vitro rekonstituiert werden kann. Daher 

setze ich modernste Techniken ein („single-molecule pulldown“), um den Komplex direkt aus 

der Zelle zu isolieren und für die Einzelmolekülmessung zugänglich zu machen oder direkt in 

lebenden Zellen zu verfolgen (TIRF-Mikroskopie). Für die fluoreszenzbasierte 

Einzelmolekülanalyse müssen die Biomoleküle ortsspezifisch und effizient mit


Fluoreszenzmarkierungen versehen werden. Daher habe ich ein umfassendes Repertoire an 

Markierungstechniken für die Markierung von Proteinen und Peptiden mit organischen 

Fluoreszenzfarbstoffen, Biotinen, Spinmarkierungen, über Aptamere und fluoreszierende 

Fusionsproteine, u.a. basierend auf modernen bioorthogonalen Strategien, erfolgreich 

realisiert. Diese Kopplungstechniken sind auch in der Entwicklung von Nanomaterialien 

vorteilhaft. 

Neben der Erforschung biologischer Maschinen beschäftigt sich ein Teil meiner Gruppe mit 

der Entwicklung neuer Nanomaterialien. Eine erste erfolgreiche Kombination von Biochemie 

und Nanotechnologie gelang mir in der Entwicklung von DNA Origami als neue 

biokompatible Oberfläche für Einzelmolekülmessungen. Ich strebe eine Weiterentwicklung 

der DNA- aber auch neuer Protein-basierter Nanostrukturen an, sodass diese als 

Biosensoren dienen können. Ziel ist es, Biomoleküle auf einer Oberfläche positionsgenau zu 

arrangieren und die Aktivität der Enzyme in Abhängigkeit von der Probenzusammensetzung 

fluoreszenzspektroskopisch auszulesen („Lab on a chip“). 

In meiner Laufbahn habe ich stets in einer interdisziplinären Umgebung gearbeitet (Max- 

Planck-Institut Dortmund, ISMB am University College London) und diese sehr geschätzt. 

Diese interdisziplinären Kontakte pflege ich auch als Mitglied der Forschungsverbünde in 

Braunschweig (z.B. BRICS, Graduiertenschule an der Universität Braunschweig). Ich freue 

mich darauf, meine Expertise im Bereich Aktivität und Struktur komplexer molekularer 

Maschinerien aber auch meine Kompetenz im Bereich NanoBioScience in Lehre und 

Forschung an der Universität Jena einzubringen. Beide Themenbereiche würden sich 

ausgezeichnet in die Forschungsinitiativen der Universität eingliedern (z.B. „Dynamik 

komplexer biologischer Systeme“, BMBF-Projekt „Jenaer Biochip-Initiative, Mikrobiologie 

und Biodervisität). 

Ich hoffe, dass meine wissenschaftliche Qualifikation überzeugt und freue mich auf die 

Gelegenheit, meine Forschung und Zukunftspläne persönlich vorzustellen. 

Mit freundlichen Grüßen, 

Dina Grohmann 

Anlagen: 

- Lebenslauf (und eingeworbene Drittmittel) 

- Forschungsplan 

- Wissenschaftlicher Werdegang 

- Publikationen 

- Lehrveranstaltungen und Lehrkonzept 

- Referenzen 

- Zeugnisse und Urkunden 

- Reprints


PERSONAL DETAILS 

Date of birth 25. Mai 1978 

Place of birth Stollberg / Germany 

Telefon +49 - 0531 – 391 5395 

Email 

d.grohmann@tu-bs.de 

Nationality 

German 

RESEARCH EXPERIENCE 

Apr/2011 - dato 

Jan/2007 – Mar/2011 

Jul-Aug 2009 

Aug/2006 – Oct/2006 

Sep/2002 – Jul/2006 

Oct/2001 – Jun/2002 

Junior Research Group Leader and Junior Lecturer (Akademischer Rat a.Z.), 

Institute of Physical and Theoretical Chemistry, Dpt. NanoBioSciences (Prof. 

Dr. P. Tinnefeld), Technische Universität Braunschweig 

Postdoctoral Research Fellow, Department of Structural and Molecular 

Biology (Prof. Dr. F. Werner), University College London, UK, “Fluorescence 

mapping of archaeal transcription complexes” 

Visiting Researcher, Department of Chemistry and Biological Chemistry 

(Prof. Dr. R.H. Ebright), Rutgers University, USA, “Incorporation of unnatural 

amino acids into proteins and azide-specific fluorescent labelling of 

biomolecules by Staudinger-Bertozzi ligation” 

Postdoctoral Fellow, Institute of Molecular Medicine (Prof. Dr. T. Restle), 

University Lübeck, Germany 

PhD thesis, Max-Planck-Institute Dortmund and University Lübeck, 

Germany (Prof. Dr. T. Restle) 

“Development of antiviral strategies using HIV-1 reverse transcriptase as a 

target“ 

Master thesis (Biology), University of Düsseldorf (Prof. Dr. P. Westhoff), 

Germany 

“Biochemical analysis of protein-protein-interactions of HCF-proteins in 

Arabidopsis thaliana”


UNIVERSITY EDUCATION 

Jul 2006 

Ph.D. degree (magna cum laude) 

2002 - 2006 Ph.D. Student in Physical Biochemistry 

Max-Planck Institute of Molecular Physiology, Dortmund (Dr. T. Restle / 

Prof. Dr. R. Goody) continued at the Institute of Molecular Medicine, 

Universität Lübeck (Prof. Dr. T. Restle), Germany 

1999 - 2002 M.S. (“Diplom”, degree: 1,3) in Biological Sciences from the University of 

Düsseldorf, Germany. 

Specialisation in Biophysics, Developmental Biology of Plants and Organic 

Chemistry 

1996 - 1999 B.S. (“Vordiplom”) in Biological Sciences from the University of Düsseldorf, 

Germany 

AWARDS 

2010 Award for the best presentation given by a young scientist at the 9th 

Annual UK Meeting on Genetics & Molecular Mechanisms in Archaea 

(Birmingham, UK) 

2009 Bogue Research Fellowship, The Bogue-Fellowship supports outstanding 

young Scientists enabling them to carry out research in laboratories in the 

USA and Canada in order to enrich their research experience and help 

develop the scientific career of the Fellow. This Fellowship allowed me to 

visit the laboratory of Prof. Dr. R.H. Ebright, Department of Chemistry and 

Biological Chemistry, Waksman Institute, Rutgers University, New Jersey, 

USA. 

2009 Wellcome Trust VIP (Value in People) Award, The aim of the Wellcome 

Trust Value in People award is to help retain excellent academic staff at the 

University by e.g. providing bridging support for postdoctoral 

researchers. This award covered my salary from Jan/2010 to Dec/2010. 

PATENTS 

2006 A.Mescalchin, W.Wünsche, S.Laufer, D.Grohmann, T.Restle, G.Sczakiel. 

Hexanucleotide-Wirkstoffe. Patentverwertungsagentur, Kiel, Schleswig- 

Holstein (PVA).


FUNDING ID 

Year Funding Body Project Title Time period Amount 

2009 Bogue Foundation (UK) Research Fellowship Jul – Aug 2012 4 500 £ 

2010 Wellcome Trust (UK) Postdoctoral stipend 2011 - 2012 33 035 £ 

2012 Deutsche 

Forschungsgemeinschaft 

(D) 

2013 German Israel 

Foundation (D) 

2013 Fonds der Chemischen 

Industrie (D) 

2013 Boehringer-Ingelheim- 

Stiftung 

Dynamics and Mechanisms of 

Argonaute proteins 

Single molecule analysis of 

transcription complex dynamics 

Sachkostenzuschuss für den 

Hochschullehrernachwuchs 

Protein-based scaffolding of 

nanostructures 

Apr 2012 - 2015 422 858 € 

Jan – Dec 2013 34 000 € 

2013 – 2016 10 000 € 

pending


RESEARCH OBJECTIVES 

Biological machines are multisubunit macromolecular complexes that drive cellular functions in a 

highly efficient and specific manner. Molecular machines are functional units that spontaneously 

form in a cellular environment with correct stoichiometry and a preserved architecture to ensure full 

activity. Over the last two decades – aiming to imitate efficient biological machineries – artificial 

nanostructures have been developed that form defined 2D- and 3D-shapes including carbonnanotubes, 

DNA Origami and cyclodextrins. Among these, DNA Origami found a remarkable rise of 

interest spurred by its exceptional properties to allow a precise and accurate arrangement of 

functional molecules like proteins, nanoparticles, DNA “docking stations” for biomolecular assays and 

fluorescent probes. These Origami structures are based on a ‘scaffold’ DNA strand (the singlestranded 

DNA genome of bacteriophage M13), which can be folded into pre-defined 2D and 3D 

assemblies at the nanometre scale with the help of hundreds of short oligonucleotides called ‘staple 

strands’. Research in my group exploits the knowledge we gain from studying biological 

machineries to build new functional nanostructures. 

The detailed characterization of complex biological systems is still a challenge and therefore I use an 

interdisciplinary approach to reach a comprehensive picture of cellular machineries. My research is 

focused on two cellular machineries involved in RNA production and posttranscriptional regulation, 

the transcriptional apparatus and the RNA induced silencing complex (RISC). My lab combines 

classical biochemical methods with sophisticated biophysical techniques (e.g. single molecule 

fluorescence spectroscopy and electron paramagnetic resonance spectroscopy) and chemical 

biology. I have a long-standing expertise in the chemical modification of biomolecules which allows 

me to site-specific incorporate labels like fluorescent dyes or spin labels into the biomolecules in 

order to allow biophysical studies and to build new functional materials based on these 

biomolecules. To this end, I work on the extension of the chemical toolkit to provide new labelling 

strategies and functional molecules. Biology, Biophysics, Chemical Biology and Nanotechnology are 

essential ingredients for my long-term project that aims to develop biosensors and nanomaterials 

(“lab on a chip”). Ideally, these sensors allow a sensitive measurement of cellular parameters that 

can be quantified via single molecule measurements.


I. Biology meets Nanotechnology 

Nanotechnology is aiming to create complex functional structures on the nanometre scale. 

Combining nanotechnology, biology and biophysics I envisage the development of biosensors and 

biomaterials. Biological systems are offering numerous advantages that recommend them for 

nanotechnological applications. Suitable model system can provide the following key features (i) the 

highly specific interaction between chosen molecules, (ii) reversible conformational changes in 

reaction to external stimuli and (iii) a precise molecular behaviour that is reflected in clear ON-OFF 

states. We already developed DNA origami as self-assembled platform on which for examples 

enzymes can be positioned with high accuracy retaining their catalytic activity. Using single-molecule 

fluorescence spectroscopy we showed that DNA origami can be used to match ensemble and singlemolecule 

measurements serving as a transport platform for a biomolecular assay and to prevent 

surface-induced artefacts often encountered in TIRF-microscopy. These platforms could potentially 

be used to screen for inhibitors or to quantify metabolic compounds as single molecule techniques 

allows the detection of molecules in picomolar range opening up the opportunity for highly sensitive 

detection in a high-throughput format. 

Additionally, modifying proteins rather than DNA to produce new nanometric structures endowed 

with novel properties is an innovative and exciting combination of bionanotechnology and synthetic 

biology. To this end, I am currently establishing nanostructures based on an extremely stable 

hexameric archaeal protein that has been purified in my group as building unit for protein-scaffolded 

nanostructures. Ultimately, the integration of the protein-based scaffold with DNA origami 

nanotechnology is envisaged to form higher-order assemblies. Combining nanotechnology, biology 

and biophysics I plan to establish the protein-based building block and to characterise the properties 

of protein-scaffolded nanostructures created in order to pave the way towards fully functional 

protein-DNA hybrid structures (Grant application for this project with the Boehringer Ingelheim 

Stiftung is currently pending). In conjugation with single DNA oligonucleotides or DNA origami the 

hexamer-based scaffold can form the foundation of a variety of different higher-order nanostructures 

and nanomachines. A site-specific introduction of modifications into the protein as well as the DNA 

opens up numerous possibilities to functionalise the nanostructure forming a material with new 

properties, among them i) biomimetic DNA-protein carpets and ii) nanospheres for drug delivery. 

Figure 2: Reactions on a heaxmeric protein-scaffold. A: Site-specific engineering of natural or unnatural amino acids with a 

unique reactive side group (e.g. a) thiol or f) azide) in MjHfq facilitates reactions that decorate the protein-hexamer with 

functional units. Shown is a selection of a wide range of modifications possible. Additionally, genetically encoded protein 

fusions (i.e. shown in g) GFP, not to scale) at the free N-terminus of the MjHfq subunit is conceivable. Conjugation of a DNA


oligonucleotide supports the formation of a biomimetic monolayer of DNA and protein (carpet), (B) or a nanosphere (C). D: 

More extended structures can be built when a DNA Origami like the six-helix bundle is coupled to the hexamer via a DNAlinkage. 

Linkage to the protein-scaffold can be followed by FRET between a donor (green) and acceptor dye (red) while 

super-resolution microscopy reports on the creation of the symmetric structure with sixfold architecture (visualised using 

the two red dyes). 

II. Development of functional reagents for site-specific labelling of 

biomolecules 

Many biophysical techniques require a modification of the biomolecule with a reporter group (e.g. 

fluorescent dye, spin label, biotin, etc). Currently we are aiming at a bi-funtional reagent that permits 

biotinylation and fluorescence-labelling in one step. In addition, in a collaborative effort with Prof. Dr. 

H.J. Steinhoff (University Osnabrück) I am working on the development of spin labels that react with 

the side chains of unnatural amino acids. So far, the introduction of spin labels for EPR measurements 

has been carried out via cysteines. Therefore, labelling cysteine-rich proteins has been complicated 

and in analogy to fluorescence-labelling via the Staudinger-Bertozzi Ligation the EPR field would 

benefit from new types of labels. We are able to synthesize unnatural amino acids which are not 

commercially available and test the labelling efficiency of newly developed labels on a wide range of 

model proteins. 

Currently, we also establish the labelling schemes for eukaryotic proteins based on the copper-free 

click-reaction. This approach enables the direct labelling of eukaryotic proteins without the need to 

overexpress them in a heterologous expression system which frequently fails and provides only 

inactive protein. Instead, we incorporate an unnatural amino acid containing a strained cyclic alkine 

in vivo. The strained alkine can readily react with the azide moiety of a fluorophore directly in the cell 

lysate. Selection of the labelled proteins species from cell lysates is planned to be achieved using the 

zero-mode waveguide technology. This completely new and unique combination of technologies 

opens up the possibility to study eukaryotic proteins hitherto not amenable to a detailed 

investigation as the labelling of native proteins can be achieved and no further purification and 

isolation step is required. 

III. Dynamics und mechanisms of molecular machines involved in 

transcriptional and posttranscriptional regulation 

The controlled and differential gene expression is of uppermost importance for every living organism 

and is executed – among others – at the transcriptional and posttranscriptional level. My work mainly 

focuses on two molecular machineries involved in transcriptional and posttranscriptional regulation, 

the transcriptional apparatus and the RNA induced silencing complex (RISC). The combination of 

chemical biology, sophisticated fluorescence-based single molecule analysis with biochemical 

methods in my laboratory is highly synergistic and enabled me to address open questions regarding 

the dynamic aspects of protein-nucleic acid complexes. 

Structural changes in the transcriptional machinery 

The organisation of transcription complexes and the structural arrangement of the RNAP itself are 

subject to constant change during the transcription cycle. For example, the transition into the 

elongation phase necessitates that a set of interactions is established (e.g. the interaction between


elongation factors and RNAP) while others have to be disrupted (contacts between RNAP and the 

promoter DNA and initiation factors). During my postdoctoral work I established a fluorescently 

labelled version of the previously developed wholly recombinant transcription system derived from 

the hyperthermophilic archaeal model organism Methanocaldococcus jannaschii. With the possibility 

to site-specifically introduce fluorescent probes into the twelve-subunit archaeal RNAP as well as into 

the basal transcription factors TBP and TFE fluorescence-based technologies are employed in my lab 

to gain a deeper understanding of the organisation of large transcription complexes. Single molecule 

fluorescence measurements are an excellent tool to analyse dynamic protein-nucleic acid complexes 

and we were just able to describe the dynamic behaviour of the RNAP clamp domain throughout the 

transcription cycle (manuscript in preparation). Furthermore, my group is focusing on the 

transcription initiation factor TBP which induces a severe bend in the promoter DNA that can be 

quantitatively described using a FRET-based bending assay. Even though the structures of TBP from 

different organisms is highly conserved we found that the kinetics and of the bending process is 

dramatically different in the archaeal and eukaryotic domain of life (manuscript submited). Our data 

suggest that the TBP-promoter DNA interaction is highly adapted to environmental conditions 

resulting in fundamentally different factor requirements. 

RNA-induced silencing complex (RISC) 

Targeted gene silencing by RNA interference (RNAi) represents a very critical mechanism to control 

cellular transcript and protein levels and is therefore involved in a multitude of important cellular 

functions. All RNA-silencing processes are based on large ribonucleoprotein assemblies, termed RNAinduced 

silencing complexes (RISCs). At the functional core of RNA-silencing pathways, every RISC 

contains a member of the Argonaute family (Ago). This key protein is bound to a small non-coding 

RNA and is responsible to direct RISC to the target mRNA which then leads to Ago-catalysed 

degradation of the mRNA or translational inhibition. Misregulated RNAi-based processes cause 

cancer and it is of special interest to gain a detailed understanding of the molecular mechanisms that 

drive RNAi in order to develop specific and effective therapeutics. A description of the dynamics of 

the mobile PAZ-domain of Ago is still missing and the mechanisms that underlie the transfer of the 

RNA between the RISC proteins are poorly understood. I started to use single-molecule methods in 

conjunction with biochemical approaches to unravel the conformational changes of site-specifically 

fluorescently labelled Ago proteins (manuscript summarising our first results is currently under 

review). Here, we succeeded in a stochastical labelling of a protein with a fluorescent donor and 

acceptor via two unnatural amino acids. Our in vitro data are currently complemented by ex situ 

experiments that allow the direct immobilization of the endogenous RISC complex from cellular 

extracts on a cover slip making it amenable to single molecule studies (single-molecule pulldown 

assay). As the RISC complex has not been successfully reconstituted in larger amounts so far 

structural information of the complete complex could just be obtained using cryo-electron 

microscopy. Our approach potentially allows us to gain more insights into the structural organization 

and the dynamics of the complex.


RESEARCH ACCOMPLISHMENTS 

Whether I worked on proteins that are involved in the assembly of the photosynthetic complex II, the 

HIV-1 Reverse Transcriptase or currently on the more complex system of multisubunit RNA 

polymerases and transcriptional regulation I have always been highly interested in the molecular 

mechanisms that govern the function of a protein/enzyme interacting with other proteins or nucleic 

acids. This included very often a detailed characterization of not just a single protein but a whole 

protein-nucleic acid interaction network and a dissection of the factors that influence the activity of 

an enzyme. Instrumental to the success of my work has been the development of cutting edge 

labelling strategies, e.g. the site-specific incorporation of fluorescent probes via unnatural amino 

acids into proteins. 

During my PhD thesis I developed alternative strategies for the inhibition of HIV focusing on the 

polymerase of the virus, the Reverse Transcriptase. With the emergence of drug-resistant HIV strains 

new inhibitory approaches are required and the HIV-1 Reverse Transcriptase is an ideal target protein 

as it is vital for viral replication copying the viral RNA into cDNA for subsequent integration into the 

host genome. I worked on the development of peptide, nucleic acid (aptamers, hexamers) and small 

molecule inhibitors that are targeted at the reverse transcriptase. Most notably, I characterized the 

first small molecule inhibitor that prevents the dimerization of the two Reverse Transcriptase 

subunits thereby inactivating the enzymatic function that is correlated with the dimeric form of the 

enzyme. Based on my mutational studies a structure-based ligand design combined with docking 

studies was carried out that led to the identification of several small molecules with the potential to 

disrupt RT dimerization. Using an in vitro dimerization assay I demonstrated that one of the selected 

molecules strongly reduced the association of the two RT subunits. Additionally, I showed that the 

compound simultaneously inhibited both the polymerase as well as the RNaseH activity of the 

enzyme. This study represented the first successful rational screen for a small molecule HIV RT 

dimerization inhibitor. 

Furthermore, I described the effect of an additional viral protein, the Nucleocapsid (NC)-protein on 

reverse transcription. NC is a so-called nucleic acid chaperon and is associated with the viral RNA and 

promotes various steps during reverse transcription. In order to describe in a quantitative fashion the 

effect of NC on reverse transcription I carried out single-turnover, single-nucleotide incorporation 

studies. I complemented these functional studies with time-resolved FRET experiments to determine 

the influence of NC on the dissociation of the RT-substrate complex. My studies revealed that NC 

considerably enhances the stability of RT-substrate complexes by reducing the observed dissociation 

rate constants, which more than compensates for the observed drop in the polymerization rate 1 . 

These data shed a new light on the activity spectrum of NC as it not only indirectly assists the reverse 

transcription process by its nucleic acid chaperoning activity but also positively affects the RTcatalysed 

nucleotide incorporation reaction by increasing polymerase processivity presumably via a 

physical interaction of the two viral proteins. 

During my work as Postdoctoral Fellow at the University College London I focused on the application 

of fluorescence techniques in combination with other biophysical methods to carry out structural and 

functional studies on one of the most complex cellular machines – the multi-subunit RNA


polymerase. This included work with proteins from the third domain of life, the Archaea, and the 

application of a broad range of biochemical techniques to investigate the molecular mechanisms of 

transcription initiation and elongation. I established assays that monitor the interaction of up to 17 

individual molecules. Additionally, I developed labelling and purification protocols for six out of the 

12 subunits of the archaeal RNA polymerase (RNAP) and for three additional transcription factors. In 

addition, in a collaborative effort with Chemists at UCL I worked on the expansion of the chemical 

toolkit for biological molecules, e.g. the reversible biotinylation of proteins. During my time as 

Postdoctoral Fellow I was awarded a Bogue Fellowship that allowed me to visit the laboratory of Prof. 

Dr. Richard Ebright (Ruttgers University, New Jersey) to learn how to incorporate unnatural amino 

acids into proteins for site-specific labelling with fluorophores. I was able to successfully establish this 

technique at UCL and to apply this to the transcription system. My work resulted in a unique fully 

recombinant transcription system that can be labelled with dyes, biotins or spin labels at any chosen 

position. This ultimately allowed the study of i) the interaction of isolated RNAP subunits with the 

RNAP core, (ii) the dynamics of isolated subunits upon binding to their nucleic acid interaction 

partner, (iii) the architecture and activity of initiation complexes and (iv) the dynamics of RNAP 

domains. The work included fluorescence measurements on the ensemble and the single molecule 

level and electron paramagnetic resonance spectroscopy (EPR) measurements. Bringing together the 

biochemical and the biophysical approach ultimately allowed a comprehensive study of the 

molecular mechanisms that govern transcription initiation and elongation which has been published 

in Molecular Cell. I precisely mapped the position of the transcription factor TFE in the transcriptional 

pre-iniation complex using the FRET-based nano-positioning system. Alternative structural methods 

have not been successful up to this date to capture this complex as the complex is too big for NMR 

and seems to be too flexible for X-ray studies. I furthermore showed that the binding sites for 

initiation and elongation factors are overlapping and that the binding of the factors to the RNAP is 

mutually exclusive proposing a factor swapping mechanism for archaeal-eukaryotic RNA polymerases 

that supports phase transition during the transcription cycle. 

Over the last 2 years I have been working as Junior Research Group Leader at the Technische 

Universität Braunschweig heading a group of 4 PhD students and a Postdoctoral Fellow. During that 

time I expanded my methodological repertoire to Nanotechnology, more specifically the use of DNA 

origami as flexible three-dimensional structure and molecular breadboard. Again, using singlemolecule 

fluorescence spectroscopy I showed that DNA origami can be used to match ensemble and 

single-molecule measurements serving as a transport platform for a biomolecular assay and to 

prevent surface-induced artifacts often encountered in TIRF-microscopy. This work is currently 

extended by a completely new approach to build nanostructures. Counterbalancing my efforts in the 

nanotechnology sector are my interests in biological machineries, e.g. (i) the dynamics of the RISC 

complex and individual proteins that are part of RISC and (ii) the dynamics of transcription factors 

and the RNAP during the transcription cycle. These highly ambitious projects are financially 

supported by the Deutsche Forschungsgemeinschaft (DFG) and the German Israel Foundation and 

three manuscripts covering our recent results are in preparation already.


BOOK CHAPTER 

Finn Werner and Dina Grohmann (2011). Chapter 6: Structure, function and evolution of 

archaeo-eukaryotic RNA polymerases –gatekeeper of the genome. in Molecular Machines in 

Biology – Workshop of the Cell. Editor: Joachim Frank, Cambridge University Press. 

PUBLICATIONS 

1. Gietl A., Holzmeister P., Blombach F., Schulz S., Lamb D., Hahn S., Werner F., Tinnefeld 

P., Grohmann D. *corresponding author 2013). Single-molecule analysis reveals diverse 

pathways during early steps of transcription initiation. (submitted to Nat. Struct. Mol. 

Biol.) 

2. Zander A., Holzmeister P., Klose D. and Grohmann D. *corresponding author (2013). 

Fluorescent probes report on the structural dynamics of archaeal Argonaute. RNA 

Biology (under review) 

3. Holzmeister P., Acuna G.P., Grohmann D. and Tinnefeld P. (2013) Breaking the 

Concentration Limit of Optical Single-Molecule Detection. Chemical Society Reviews 

(accepted) 

4. Grohmann D. *corresponding author , Werner F. and Tinnefeld P. (2013). Making Connections 

– Strategies for Single Molecule Fluorescence Biophysics. Current Opinion in Chemical 

Biology. 17(4): 691-8. 

5. Gietl A. and Grohmann D. *corresponding author (2012). Modern biophysical approaches 

probe transcription factor induced DNA bending and looping. Biochem Soc Trans. 

41(1):368-73. 

6. Blombach F., Daviter T., Fielden D., Grohmann D., Smollett K. and Werner F. (2012). 

Archaeology of RNA polymerase – factor swapping during the transcription cycle. 

Biochem Soc Trans. 41(1):362-7. 

7. Klose D., Klare J., Grohmann D., Kay C.W.M., Werner F. and Steinhoff H-J. (2012). 

Simulation vs. reality: a comparison of in silico predictions with DEER and FRET 

measurements. PLOS one, 7(6):e39492.


8. Gietl A., Holzmeister P., Grohmann D. *shared corresponding author and Tinnefeld P. (2012). 

DNA origami as biocompatible surface to match single-molecule and ensemble 

experiments. Nucleic Acids Research, epub. 

9. Grohmann D., Nagy J., Chakraborty A., Klose D., Fielden D., Ebright R.H., Michaelis J., 

Werner F. (2011). The Initiation Factor TFE and the Elongation Factor Spt4/5 Compete 

for the RNAP Clamp during Transcription Initiation and Elongation. Molecular Cell, 

43:263-74. 

10. Grohmann D. *corresponding author and Werner F. (2011) Recent advances in the 

understanding of archaeal transcription. Curr Opin Microbiol. 2011 May 17. 

11. Ryan C.P., Smith M.E., Schumacher F.F., Grohmann D., Papaioannou D., Waksman G., 

Werner F., Baker J.R., Caddick S. (2011) Tunable reagents for multi-functional 

bioconjugation: reversible or permanent chemical modification of proteins and 

peptides by control of maleimide hydrolysis. Chem Commun. 47:5452-4. 

12. Grohmann D., Klose D., Fielden D., Werner F. (2011) FRET (fluorescence resonance 

energy transfer) sheds light on transcription. Biochem Soc Trans. 39:122-7. 

13. Werner F. and Grohmann D. (2011) Evolution of multisubunit RNA polymerases in the 

three domains of life. Nature Rev Microbiol.9:85-98. 

14. Grohmann D. and Werner F. (2011) Cycling through Transcription with the RNA 

polymerase F/E (RPB4/7) Complex - Structure, Function and Evolution of Archaeal RNA 

polymerase., Research in Microbiology 162:10-8. 

15. Grohmann D., Klose D., Klare J.P., Kay C.W., Steinhoff H.J., Werner F. (2010) RNAbinding 

to archaeal RNA polymerase subunits F/E: a DEER and FRET study. J Am Chem 

Soc., 132:5954-5. 

16. Hirtreiter A., Damsma G.E., Cheung A.C., Klose D., Grohmann D., Vojnic E., Martin A.C., 

Cramer P., Werner F. (2010), Spt4/5 stimulates transcription elongation through the 

RNA polymerase clamp coiled-coil motif. Nucleic acids research 38:4040-51. 

17. Hirtreiter A., Grohmann D., Werner F. (2010) Molecular mechanisms of RNA 

polymerase – the F/E (Rpb4/7) complex is required for high processivity in vitro. 

Nucleic acids research, 38:585-96. 

18. Grohmann, D. & Werner, F. (2010) Hold on!: RNA polymerase interactions with the 

nascent RNA modulate transcription elongation and termination. RNA Biol, 7: 310- 

315.


19. Grohmann D., Hirtreiter A., Werner F. (2009), RNAP subunits F/E (Rpb4/7) are stably 

associated with archaeal RNA polymerase: using fluorescence anisotropy to monitor 

RNAP assembly in vitro. Biochem J, 421:339-43. (selected by the 'Faculty of 1000 

Biology' for its originality and important contribution to the field) 

20. Grohmann D., Hirtreiter A., Werner F. (2009), Molecular mechanisms of archaeal RNA 

polymerase. Biochem Soc Trans. 37:12-7. 

21. Grohmann D., Godet J., Mely Y., Darlix JL. and Restle, T. (2008). HIV-1 NC traps the 

reverse transcriptase on primer/template substrates. Biochemistry, 47:12230-40. 

22. Di Pasquale F, Fischer D, Grohmann D, Restle T, Geyer A, Marx A. (2008) Opposed 

steric constraints in human DNA polymerase beta and E.coli DNA polymerase I. J Am 

Chem Soc 130:10748-57. 

23. Grohmann D., Corradi V., Horenkamp F., Laufer, S.D., Manetti, F. Botta, M. and Restle, 

T. (2008). Small molecule inhibitors targeting HIV-1 dimerization. Chembiochem 9:916- 

22. 

24. Yamazaki S., Tan L., Mayer G., Hartig J.S., Song J.N., Reuter S., Restle T., Laufer S.D., 

Grohmann D., Kräusslich H.G., Bajorath J., Famulok M. (2007). Aptamer displacement 

identifies alternative small-molecule target sites that escape viral resistance. Chem 

Biol.14:804-12. 

25. Mescalchin A., Wünsche W., Laufer S.D., Grohmann, D., Restle, T. and Sczakiel, G. 

(2006). Specific binding of a bioactive hexanucleotide to HIV-1 reverse transcriptase: a 

novel class of small oligomeric nucleic acid drugs. Nucleic acids research, 34, 5631– 

5637. 

26. Plücken H, Müller B., Grohmann D., Westhoff P., Eichacker LA. (2002). The HCF136 

protein is essential for assembly of the photosystem II reaction center in Arabidopsis 

thaliana. FEBS Letters 532:85-90.


STATEMENT OF TEACHING 

2012/2013 Lecture “Biophysical Chemistry” (english), 1 st and 2 nd year graduate 

students (Master Chemistry/Master Biotechnology), Technische 

Universität Braunschweig 

2011 - 2012 Seminar “Basics in Physical Chemistry”, 3rd year undergraduate 

students (Bachelor Chemistry), Technische Universität Braunschweig 

2011/2012 Lecture and Seminar “NanoBioSciences” (englisch), 1 st and 2 nd year 

graduate students (Master Chemistry), Technische Universität 

Braunschweig 

2008 - 2010 Seminar “Advanced Biomolecular Mechanisms” (englisch), 

Department of Structural and Molecular Biology, 3rd year 

undergraduate students (Biochemistry), University College London, 

UK 

2003 – 2005 Seminar “Application of nucleic acid-based inhibitors”, 3 rd year 

undergraduate students (Bachelor Biotechnology), Universität Lübeck 

2003 – dato Supervision of 

 

various students for 1-2 month internships on a one to one 

basis 

final year projects of Diploma (1), Bachelor (7) and Master (2) 

students 

 

supervision of currently four PhD students working in my 

group 

Other relevant experiences 

2012 Supervision of the Fellows of the Fonds of the Chemical Industry 

(Fonds der Chemischen Industrie, Germany) – Supporting the 

education of women in science


TEACHING CONCEPT 

I am grateful to the University College London and the Technische Universität Braunschweig for the opportunity 

to teach undergraduate and graduate students. I am giving lectures and seminars on Biophysical Chemistry, 

NanoBioSciences and Biochemistry and enjoy teaching at the interface between Physics, Biology and Chemistry. 

I would like to pursue teaching these subjects in the future as they range from the fundamental nature of the 

biomolecules to current applications like single-molecule sequencing. It is a joy to see students realizing und 

understanding for the first time that the laws of the macroscopic world are not the same in the nanoscopic 

world. As I have taught Chemists, Biochemists, Biologist and Biotechnologists alike I had to adapt to the 

different levels of knowledge taking them from for example the discovery of the DNA double helix to the 

chemical fundament of DNA from which students can start to understand the nature of DNA and the 

information processing in cells. Personally, I think that good teaching is equipping students with a scientific 

fundament that enables them to follow their curiosity to discover the inner workings of life. 

I successfully used a variety of teaching methods ranging from direct discussion of difficult subjects with the 

students to exercises that accompany my lectures. I always seek to involve the students in the exploration of 

a subject starting sometimes from daily 

experiences and focusing more and more 

on the biological and chemical basis of 

these observations. I also incorporate 

example from the scientific frontiers to 

present current research questions 

pointing out the many open questions. 

Modern technology can help to include 

the students in a lecture. For example, I 

have been using the so-called “Eduvote” 

system. Here, students can chose an 

answer on questions that have been 

incorporated into a power point 

presentation using nothing more than a 

smart phone. This gives a fast feedback in 

real-time on how well the students 

understood the topic leaving the 

possibility to re-iterate difficult passages. 

In my seminars I often support students 

to work in small teams in order to include 

every single student encouraging them to 

discuss and present a topic guided by the 

figures of appropriate publications. 

Overall, I believe that teaching and research is about creativity, inspiration and productivity. Students need to 

be challenged intellectually but they also have to learn that creativity, ethics and perseverance is important in 

research. That is why I would like to realize a project I have been planning without the time to realize it yet. I 

called it “Scientists with a vision” (see flyer above) and it is designed to empower students to discover their 

personal vision in Science and to equip them with the skills needed to pursue a successful (scientific) career.


REFERENCES 

1. Prof. Dr. Philip Tinnefeld 

Institut für Physikalische und Theoretische Chemie 

Abteilung NanoBioSciences 

Fakultät für Lebenswissenschaften 

Hans-Sommer-Straße 10 


Tel: +49 (0)531 391 5330 

Fax: +49 (0)531 391 5334 

Email: p.tinnefeld@tu-bs.de 

2. Prof. Dr. Finn Werner 

Principal Investigator and Senior Lecturer 

Department of Structural and Molecular Biology 

University College London 

Room 308, Darwin Building 

Gower Street 

London, WC1E 6BT 

Tel: +44 (0)20 7679 0147 

Email: f.werner@ucl.ac.uk 

3. Prof. Dr. Tobias Restle 

Institut für Molekulare Medizin 

UK-SH Lübeck, 

Ratzeburger Allee 160 

23538 Lübeck 

Tel: +49 (0)451 500 2745 

Fax: +49 (0)451 500 2729 

Email: restle@imm.uni-luebeck.de 

4. Prof. Dr. Heinz-Jürgen Steinhoff 

Fachbereich Physik 

Universität Osnabrück 

Barbarastrasse 7 

D-49076 Osnabrück 

Tel: +49 541 969 2675 o. 2820 

Email: hsteinho@uni-osnabrueck.de 

5. Prof. Dr. Richard H. Ebright 

Howard Hughes Medical Institute 

Waksman Institute 

Rutgers University 

190 Frelinghuysen Road 

Piscataway, NJ 08854 

Tel: +1-848-445-5179 

Fax: +1-732-445-5735 

Email: ebright@waksman.rutgers.edu

Nucleic Acids Research Advance Access published April 20, 2012 

Nucleic Acids Research, 2012, 1–10 

doi:10.1093/nar/gks326 

DNA origami as biocompatible surface to match 

single-molecule and ensemble experiments 

Andreas Gietl, Phil Holzmeister, Dina Grohmann* and Philip Tinnefeld* 

Physikalische und Theoretische Chemie - NanoBioSciences, Technische Universität Braunschweig, 

Hans-Sommer-Strasse 10, 38106 Braunschweig, Germany 

Received February 2, 2012; Revised March 30, 2012; Accepted April 3, 2012 

ABSTRACT 

Single-molecule experiments on immobilized 

molecules allow unique insights into the dynamics of 

molecular machines and enzymes as well as their 

interactions. The immobilization, however, can 

invoke perturbation to the activity of biomolecules 

causing incongruities between single molecule and 

ensemble measurements. Here we introduce the 

recently developed DNA origami as a platform to 

transfer ensemble assays to the immobilized single 

molecule level without changing the nanoenvironment 

of the biomolecules. The idea is a 

stepwise transfer of common functional assays first 

to the surface of a DNA origami, which can be checked 

at the ensemble level, and then to the microscope 

glass slide for single-molecule inquiry using the DNA 

origami as a transfer platform. We studied the 

structural flexibility of a DNA Holliday junction and 

the TATA-binding protein (TBP)-induced bending of 

DNA both on freely diffusing molecules and attached 

to the origami structure by fluorescence resonance 

energy transfer. This resulted in highly congruent 

data sets demonstrating that the DNA origami does 

not influence the functionality of the biomolecule. 

Single-molecule data collected from surfaceimmobilized 

biomolecule-loaded DNA origami are in 

very good agreement with data from solution measurements 

supporting the fact that the DNA origami 

can be used as biocompatible surface in many 

fluorescence-based measurements. 

INTRODUCTION 

In recent years single-molecule experiments became a 

valuable tool to study dynamics of biomolecules on a 

molecular level (1,2). In particular Fluorescence (Fo¨ rster)- 

Resonance-Energy-Transfer (FRET)-based approaches 

can resolve conformational changes in the range of few 

nanometres and with a time-resolution of microseconds 

to minutes (3–6). To resolve such conformational 

changes, biomolecules are commonly immobilized to the 

surface of a cover slip. The surface, however, represents a 

potential perturbation that has to be carefully taken into 

account in each single-molecule experiment (7,8). 

Laborious control experiments have to be carried out to 

show that single-molecule experiments adequately reflect 

the ensemble solution experiment and often doubts 

remain whether obtained distributions of properties 

describe the heterogeneity of the system or that of the 

surface immobilization. Immobilization strategies have 

been an issue since single-molecule FRET has been established 

to study biomolecular dynamics more than a decade 

ago. Typical immobilization schemes include BSA 

passivated cover slips which have successfully been used 

for nucleic acid dynamics, polyethyleneglycol passivated 

glass slides and vesicle encapsulation (8). The encapsulation 

in immobilized unilamellar vesicles provides an environment 

for systems that do not interact with the membrane 

but the exchange of reagents remains challenging (9). 

A trustworthy immobilization strategy is so important 

because biomolecular reactions are usually characterized 

first in ensemble measurements on freely diffusing molecules 

and a reproduction of similar reaction conditions 

on the single molecule level is required for direct comparison. 

Groll et al.(6) could clearly demonstrate that RNAseH 

refolding after denaturation is prevented when immobilized 

on a PEO (polyethylene oxide) brush. Another example of 

surface-induced artefacts has been published by Talaga 

et al. Here, the direct immobilization of a peptide led to 

reduced conformational fluctuations (10). The continuous 

effort to find an universal and reliable method of 

immobilization is furthermore reflected in the incessant 

appearance of publications that primarily deal with the 

immobilization strategy of single molecules (5–9,11–17). 

Here, we present an immobilization scheme that allows 

matching of single-molecule and ensemble experiments 

(Figure 1). In contrast to previous approaches aiming at 

Downloaded from http://nar.oxfordjournals.org/ at UniversitÃ¤tsbibliothek Braunschweig on April 23, 2012 

*To whom correspondence should be addressed. Tel: +49 531 391 5330; Fax: +49 531 391 5334; Email: p.tinnefeld@tu-bs.de 

Correspondence may also be addressed to Dina Grohmann. Tel: +49 531 391 5395; Fax: +49 531 391 5334; Email: d.grohmann@tu-bs.de 

ß The Author(s) 2012. Published by Oxford University Press. 

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/ 

by-nc/3.0), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

2 Nucleic Acids Research, 2012 

improving the biocompatibility of surfaces, we focus on the 

convergence of conditions in ensemble and single-molecule 

experiments. By immobilizing the biomolecules of interests 

on a DNA origami, ensemble and single-molecule experiments 

can be carried out without changing the nanoenvironment. 

The DNA origami fulfils two functions: it 

serves as bio-compatible surface and represents a transportable 

entity for the biomolecular assay to passage it between 

fluorescence methods. For the single-molecule measurements, 

the DNA origami serves as an adapter between a 

glass slide and the biomolecular assay (Figure 1B). Since 

DNA origami is a promising scaffold for manifold applications 

including molecular computing, molecular assembly 

lines or nanorobots the biocompatibility of the DNA 

nanostructure is of particular importance (18–23). 

DNA origami structures are based on a ‘scaffold’ DNA 

strand (the single-stranded DNA genome of bacteriophage 

M13), which can be folded into 2D and 3D assemblies at the 

nanometre scale with the help of hundreds of short oligonucleotides 

called ‘staple strands’ (24). DNA origami represent 

a self-assembled system; the formation of the DNA 

nanostructure is achieved by a simple heat denaturation 

step followed by a slow cooling down of the scaffold 

DNA/staple strands mixture. Modifications (e.g. biotin) 

can be introduced into the DNA origami by replacing individual 

staple strands with a biotinylated version of the oligonucleotide. 

If biotins are positioned at one ‘face’ of a 

rectangular DNA origami an oriented immobilization of 

the rectangle on a streptavidin-covered glass surface 

becomes possible. Consequently, the opposite side of the 

DNA rectangle that faces away from the surface is available 

for the attachment of DNA sequences that mediate the 

specific interaction for biomolecule immobilization, e.g. 

via protein–DNA interactions. For this purpose, a single 

staple (‘anchor strand’) strand is extended by a sequence 

of interest that allows hybridization with complementary 

sequences to form a specific and functional DNA structure. 

The assembly of a DNA origami including modified staple 

strands, oligonucleotides complementary to the anchor 

staple strand and additional oligonucleotides required for 

the functional DNA entity follows the same convenient 

protocol of temperature-controlled self-assembly described. 

We studied the interconversion of the two conformational 

states of the well-described Holliday junction 


Figure 1. Schematic drawing of the experimental strategy that allows a direct comparison of ensemble and single-molecule experiments as the DNA 

origami provides an identical nano-environment in all experiments. (A) DNA origami structures are based on a ‘scaffold’ DNA strand (the 

single-stranded DNA genome of bacteriophage M13), which can be folded into 2D and 3D assemblies at the nanometre scale with the help of 

hundreds of short oligonucleotides called ‘staple strands’ (24). DNA origami represent a self-assembled system; the formation of the DNA 

nanostructure is achieved by a simple heat denaturation step followed by slowly cooling down the scaffold DNA/staple strands mixture. 

Modifications (e.g. biotin, fluorophores) can be introduced into the DNA origami by replacing individual staple strands with a modified version 

of the oligonucleotide. (B) The decorated DNA origami represents a transportable pseudo-surface for a biomolecular reaction that can be used in a 

range of fluorescence-based methods. Importantly, the DNA origami ensures an identical nano-environment for the biomolecular assay (shown here 

is the fluorescently labelled double-stranded TATA–box containing oligonucleotide) leading to comparable reaction conditions. If biotins are positioned 

at one ‘face’ of a rectangular DNA origami (blue spheres) an oriented immobilization of the rectangle on a streptavidin-covered glass surface 

becomes possible. Consequently, the opposite side of the DNA rectangle that faces away from the surface is available for the DNA sequences that 

mediate the specific interaction for biomolecule immobilization, e.g. via protein–DNA interactions.

Nucleic Acids Research, 2012 3 

(HJ) and the interaction of the transcription factor TBP 

(TATA-binding protein) with a TATA–box containing 

DNA either as isolated biomolecules or attached to a 

rectangular DNA origami platform. In both cases, a 

FRET signal served as readout to distinguish between 

the alternative HJ conformations and to follow the 

TBP-induced bending of the TATA–DNA (25). Both molecular 

processes are not affected by the presence of the 

DNA origami and are in very good agreement to datasets 

collected on freely diffusing molecules either in ensemble 

or in a single-molecule setup. Therefore, we demonstrate 

that the DNA origami provides an ideal biocompatible 

surface which (i) closes the gap between ensemble measurements 

and single-molecule studies on surfaces as the 

ensemble solution can be directly used to immobilize 

biomolecule-decorated origami on a cover slide and (ii) 

does not influence the biomolecule under investigation 

but creates a biocompatible environment ideally suited 

to monitor conformational changes without the risk of 

significant loss of activity or flexibility due to surface 

interactions. 

MATERIALS AND METHODS 

Buffers 

All experiments were carried out at room temperature, 

unless indicated otherwise. DNA annealing buffer was 

1 TAE (40 mM Tris/Acetate pH 8.3, 2.5 mM EDTA) 

with 12.5 mM MgCl 2 . HJ experiments were carried out in 

1 PBS (8.06 mM Na 2 HPO 4 , 1.94 mM KH 2 PO 4 , 2.7 mM 

KCl and 137 mM NaCl, pH 7.4) with varying MgCl 2 concentrations. 

TBP titrations as well as the TBP singlemolecule 

experiments on surfaces were carried out in 

50 mM Tris–HCl pH 7.5, 1 M NaCl and 12.5 mM MgCl 2 . 

Immobilization of the TATA–DNA oligonucleotides 

(10 pM) and TATA-origami (100 pM) was realized in 

1 PBS, 12.5 mM MgCl 2 . In order to reduce blinking ON 

and OFF states of the fluorescent dyes 2 mM Trolox 

(dissolved in DMSO, 100 mM) was added in all experiments; 

oxygen was removed by the GOC (glucose oxidase 

catalase) oxygen scavenger system and 0.5% w/w glucose 

(26,27). 

Preparation of TATA–double-stranded DNA, 

HJ and DNA origami 

All DNA constructs were hybridized and folded in 1 TAE 

in the presence of 12.5 mM MgCl 2 in an Eppendorf 

Thermocycler. All modified DNA sequences were 

purchased from IBA (Go¨ ttingen). The single-stranded 

DNA origami scaffold (genomic DNA of bacteriophage 

M13mp18) was prepared as described in Douglas et al. 

(28) but can also be purchased from New England 

Biolabs. The complete set of staple strands need for the 

rectangular DNA origami was ordered from MWG [a 

complete list of staple strand sequences required for the 

rectangular origami can be found in Rothemund et al.(24)]. 

The following DNA sequences were used for 

‘origami-free’ measurements: TATA–double-stranded 

DNA (dsDNA) 5 0 -Biotin-CGG ACC GAA AGC GCG 

ACC ATC GCC GGA GA (Cy3B) TGG AGT AAA 

GTT TAA ATA CTG-3 0 and 5 0 -ATTO647N-CAG TAT 

TTA AAC TTT ACT CCA ATC TCC GGC GAT GGT 

CGC GCT TTC GGT CCG-3 0 . Strands were hybridized at 

a final concentration of 25 mM for each strand. The HJ is 

composed of four strands called R, H, X and B (Figure 2A) 

with the following sequence: R: 5 0 -Biotin-CCC ACC GCT 

CGGC TCA ACT GGG-3 0 ,H:5 0 -Cy3-CCG TAG CAG 

CGCG AGC GGT GGG-3 0 ,X:5 0 -CCC AGT TGA GCG 

CTT GCT AGG G-3 0 ,B:5 0 -Cy5-CCC TAG CAA GCC 

GCT GCT AGG G-3 0 . The strands were annealed using a 

concentration ratio of 8.3 mM: 11.1 mM: 13.8 mM: 16.6 mM 

Figure 2. (A) The HJ is composed of four single-stranded DNAs 

(B, H, X, R) that can adopt two different conformations, iso I or 

iso II (B). The donor fluorophore Cy3 is attached to the 5 0 -end of 

strand H and the acceptor Cy5 is attached to the 5 0 end of strand B. 

Here we determined the kinetic properties of the interconversion 

between the two conformational states using a FRET signal between 

Cy3 and Cy5 as readout. The iso I conformation leads to a low FRET 

whereas the iso II state causes a high FRET signal. The kinetics have 

been determined for freely diffusing molecules or for HJs anchored to a 

rectangular origami that additionally can be immobilized to a quartz 

slide (Neutravidin–Biotin linkage) via strand R (B). (C) Single-molecule 

ALEX measurements on freely diffusing HJ molecules that are labelled 

with a Cy3–Cy5 FRET pair. FRET efficiency distributions together 

with Gaussian fits to the data are shown for the HJ (top) or a HJ 

attached to the origami (bottom). Measurements were carried out in 

the presence (right) or absence (left) of 100 mM MgCl 2 . At low magnesium 

concentrations the fast interconversion between the conformational 

states iso I and iso II cannot be resolved resulting in an 

averaged FRET efficiency of E = 0.53. The interconversion rate is 

reduced in the presence of 100 mM MgCl 2 and the low and high 

FRET states iso I and iso II can be detected (E iso I = 0.36 and 

E iso II = 0.74). An attachment of the HJ to the origami does not 

exert an influence on the conformational states of the HJ as judged 

by the distribution between iso I and iso II and the corresponding 

FRET efficiencies (E iso I = 0.35 and E iso II = 0.72). 

Downloaded from http://nar.oxfordjournals.org/ at UniversitÃ¤tsbibliothek Braunschweig on April 23, 2012


for R:H:X:B. Hybridization was carried out by a heating 

step (95 C for 30 s) and the sample was immediately cooled 

down to 20 C in 0.1 C steps every 6 s. A biotin modification 

at one of the HJ or TATA–oligonucleotide strands allowed 

a direct immobilization of the TATA–oligonucleotide or 

the HJ to a PEG–Biotin–Neutravidin glass surface as 

required for TIRF measurements. 

The rectangular DNA origami design from Rothemund’s 

original publication (24) was used for all experiments. 

Alternative DNA origami structures can be designed with 

the help of the freely available software ‘cadnano’ (http:// 

cadnano.org) that also allows a visualization of the 

nanostructure. For folding of the DNA origami 3 nM 

single-stranded DNA (M13mp18), 30 nM unmodified 

staples and 300 nM labelled staples were used. In order to 

attach the TATA sequence or the HJ to the DNA origami 

the standard staple strand r3t12f [compare notation in (24)] 

has been replaced by an alternative version of the strand that 

has been extended either by one of the TATA–oligonucleotides 

(the 5 0 -to3 0 -strand) or strand X of the HJ (Table 1). In 

this case, none of the TATA–oligonucleotides or the HJ 

strands were modified with biotin. In addition, two of 

the standard staple strands (e.g. r-5t4f and r-5t6f) were 

replaced by a single staple strand that carries a 

biotin-modification at the 5 0 -end. This has been done for 

three sets of standard staple strand pairs to introduce 

three biotins at one ‘face’ of the rectangular DNA origami 

in order to allow an oriented immobilization of the 

decorated DNA origami to a streptavidin-coated glass 

surface. The assembly of the modified DNA origami 

including biotinylated staple strands, the modified strand 

r3t12f and additional oligonucleotides for the formation 

of the TATA–oligonucleotide (5 0 -ATTO647N-CAG TAT 

TTA AAC TTT ACT CCA ATC TCC GGC GAT GGT 

CGC GCT TTC GGT CCG-3 0 ) or the HJ (R: 5 0 –CCC ACC 

GCT CGGC TCA ACT GGG-3 0 ,H:5 0 -Cy3-CCG TAG 

CAG CGCG AGC GGT GGG-3 0 ,B:5 0 -Cy5-CCC TAG 

CAA GCC GCT GCT AGG G-3 0 ) follows the same 

protocol of temperature-controlled self-assembly (95 C 

for 30 s and slow cooling down to 20 C in 0.1 C steps 

every 6 s). Additional oligonucleotides that form the 

double-stranded TATA–oligonucleotide or the HJ have 

been added directly to the DNA origami mix (at a concentration 

of 300 nM each) and were therefore part of the 

self-assembly process. 

After folding the excess of staple strands was removed 

by filtration using Amicon Ultra-0.5 ml Centrifugal 

Filters (100 000 MWCO) according to manufacturer’s instructions. 

Choosing a filter with a cut-off of 100 000 

MWCO ensures that the DNA origami (4.6 MDa) is 

retained whereas the staple strands (10 kDa) and 

non-attached but properly folded individual HJs 

(28 kDa) and double-stranded TATA–oligonucleotides 

(32 kDa) are passing through the filters pores and can 

be washed away. The DNA origami solution has been 

washed three times with 1 TAE buffer and 

concentrated to a final volume of 20 ml. Filtering has 

turned out to be an efficient way to remove the excess 

of biotinylated strands (29), HJs and TATA–oligonucleotides 

that are not attached to the surface of the 

DNA origami. The wash fractions have been analysed 

by single molecule TIRF microscopy checking the 

amount of fluorescently labelled molecules left in the 

flow-through that can be immobilized on a 

PEG-surface via the Biotin-anchor. After four rounds 

of washing no fluorescence signal could be detected 

ensuring that biotinylated strands were completely 

removed and there were no competing biotinylated 

staple strands in addition to the DNA origami left. 

The DNA origami can be stored at 20 nM in 1 TAE 

and 12.5 mM MgCl 2 at 4 C for several days. 

Preparation of TBP 

TBP was expressed and purified as described earlier (30). 

Ensemble fluorescence measurements 

All fluorometer measurements were carried out on a 

temperature-controlled Cary Eclipse spectrometer in a 

fluorescence cuvette (50 ml volume, Hellma Germany). 

HJs oligo (10 nM) and HJ–origami (3 nM) experiments 

were executed at 25 Cin1 PBS with varying MgCl 2 

concentrations (Supplementary Figure S3). The TBP– 

TATA–DNA titration was performed at 55 C with a 

TATA–dsDNA oligo concentration of 10 nM as well as 

for the TATA–dsDNA on origami. Excitation was set to 

515 nm and the emission was recorded at 660 nm (slit 

width: 20 nm, detector voltage: medium, integration time: 

0.5 s). Data were evaluated using the program OriginPro. 

The relative fluorescence was plotted against the 


Table 1. Modifications to the design are shown in the table below (Supplementary Figure S6) 

Unmodified strand 

r-5t4f, TTTCATGAAAATTGTGTCGAAATCTGTACAGA 

r-5t6f, CCAGGCGCTTAATCATTGTGAATTACAGGTAG 

r518f, GCGCAGAGATATCAAAATTATTTGACATTATC 

r5t20f, ATTTTGCGTCTTTAGGAGCACTAAGCAACAGT 

r-1t14f, AGGTAAAGAAATCACCATCAATATAATATTTT 

r-1t16f, GTTAAAATTTTAACCAATAGGAACCCGGCACC 

r3t12f, GGTATTAAGAACAAGAAAAATAATTAAAGCCA 

Modified strand 

Biotin-TTTTTCGAAATCTGTACAGACCAGGCGTTAATCAT 

Biotin-TTTTATTATTTGACATTATCATTTTGCGTCTTTAGG 

Biotin-TTTTCATCAATATAATATTTTGTAAAATTTTAACC 

HJ: GGTATTAAGAACAAGAAAAATAATTAAAGCCATTTCCCACCGCTC 

GGCTCAACTGGG 

TATA: GGTATTAAGAACAAGAAAAATAATTAAAGCCACGGACCGAAA 

GCGCGACCATCGCCGGAGA-Cy3b-TGGAGTAAAGTTTAAATACTG


corresponding TBP concentration and the dissociation 

constant was calculated using a quadratic equation with 

the error given as standard error. 

Surface preparation 

Studies on immobilized molecules using a widefield setup 

were carried out on a PEG surface attached to a flow 

chamber for custom built PRISM-based TIRF microscope. 

Quartz slides were thoroughly cleaned and dried 

with nitrogen. The quartz slides were first silanized and 

afterwards PEGylized according to Roy et al. (31) 

followed by washing with 1 PBS and Neutravidin 

(1 mg/ml) incubation for 10 min. 

Cover slides for solution measurements were rinsed with 

Acetone p.A., EtOH p.A. and deionized water. After a 

small chamber (Press-to-Seal 2.5 mm, Sigma Aldrich) 

had been glued to the cover slide, a solution of 5 mg/ml 

BSA in 1 PBS was incubated for 10 min. Excessive BSA 

was removed by washing with 1 PBS. 

Widefield single-molecule detection and analysis 

We used a homebuilt PRISM-TIRF setup based on an 

Olympus IX71 to perform widefield measurements. 

Fluorophores were excited with 532 nm (Coherent 

Sapphire, Clean-up filter 532/2 MaxLine Semrock, AHF 

Go¨ ttingen, circa 3 kW/cm 2 ) diode laser. The fluorescence 

was collected by an 60 Olympus 1.20 N.A. waterimmersion 

objective and split by wavelength with a 

dichroic mirror (640 DCXR, AHF) into two channels 

that were further narrowed by a bandpass filter Semrock 

BrightLine 582/75 in the green and a 633 nm RazorEdge 

longpass filter (Semrock, AHF) in the red detection range. 

Both detection channels were recorded by one EMCCD 

camera (Andor IXon 789DU, preGain 5.1, gain 1000, 

integration time 20 ms) in a dualview configuration and 

the videos were analysed by custom made software 

based on LabVIEW 2011 64 bit (National Instruments). 

The molecule spots were selected by an automated 

spotfinder and the resulting transients were filtered with 

the built in cubic filter of LabVIEW 2011. The fluorescent 

intensities were background corrected by subtracting the 

neighbour-pixels’ intensity. The transients were also corrected 

in leakage from the donor into the red detection 

channel and direct excitation of the acceptor by the 

532-nm laser excitation. The HJ transition states were 

analysed with the HaMMy software freely available at 

the TJ Ha group’s homepage. Statistical errors are given 

as standard deviation. 

Confocal measurements 

The concentrations of fluorescently labelled molecules 

were adjusted to an average of less than one molecule 

per confocal volume in order to identify bursts from 

single molecules, i.e. in the picomolar range. 

To study fluorescence and FRET on the level of single 

molecules, a custom built confocal microscope was used. 

The setup allowed alternating laser excitation of donor 

and acceptor fluorophores on diffusing molecules with 

separate donor and acceptor detection. Fluorophores were 

excited with continuous wave at 532 nm (TECGL-30, 

World Star Tech; 80 mW forCy3,60mW forCy3B)and 

with 80 MHz pulsed at 640 nm (LDH-D-C-640, 

Picoquant, 60 mW for Cy5, 30mW for ATTO647N). 

Alternation of both wavelengths with 100 -ms period was 

achieved by use of an acousto-optical tunable filter 

(AOTFnc-VIS, AA optoelectronic). The laser beam 

entered an inverse microscope and was coupled into an 

oil-immersion objective (TBP: 60X, NA 1.35, UPLSAPO 

60XO; HJ: 60X, NA 1.49, APON 60XO TIRFM, both 

Olympus) by a dual-band dichroic beam splitter 

(Dualband z532/633, AHF). The resulting fluorescence 

was collected by the same objective, focused onto a 50 mm 

pinhole, and split spectrally at 640 nm by a dichroic beam 

splitter (640DCXR, AHF). Two avalanche photodiodes 

(t-SPAD-100, Picoquant) detected the donor and acceptor 

fluorescence with appropriate spectral filtering (Brightline 

HC582/75, Bandpass ET 700/75 m, AHF). The detector 

signal was registered using a PC card for single-photon 

counting (SPC-830, Becker&Hickl) and evaluated using 

custom made LabVIEW (National Instruments) software. 

Data evaluation for ALEX measurements 

In solution measurements, fluorescence bursts from single 

molecules diffusing through the laser focus are identified 

by a burst search algorithm applied to the sum of donor 

and acceptor photons (parameters used for free HJ: 

T = 500 ms, M = 30, L = 60, for TBP oligo: T = 500 ms, 

M = 30, L = 50, for origami: T = 500 ms, M = 30, 

L = 100) (32). Molecules are alternately excited and the 

fluorescence of donor and acceptor is separately detected. 

This defines three different photon counts: donor emission 

due to donor excitationF D D , acceptor emission due to 

acceptor excitation F A A and acceptor emission due to 

donor excitation F D A . Upon correction of these values 

from background signals, the stoichiometry parameter S 

and the proximity ratio E are defined, where S describes 

the ratio between donor and acceptor dyes of the sample 

and E stands for the proximity ratio between the dyes in 

terms of energy-transfer efficiency (33). Statistical errors 

are given as standard deviation. 

FD A 

E ¼ 

F D A +F D D 

FD A +FD D 

S ¼ 

F D A +F D D +FA A 

RESULTS AND DISCUSSION 

DNA origami has emerged as a new way to build 

nanoarchitectures that combine the advantage of 

self-assembly with free addressability of the structure to 

add functionality (19,34). DNA origami are 3D folded 

DNA structures, highly ordered on the nanoscale that 

allow the site-directed tethering of DNA strands on its 

surface (24). DNA-nanostructures have been suggested 

to be a valuable tool for a multitude of future applications 

and initial work included (i) origami as scaffold for 

dye-based photonic wires (35), (ii) DNA nanostructures 



as host for macromolecular molecules e.g. proteins (36) 

and G-quadruplexes (37) and (iii) DNA origami as a 

molecular breadboard that allows the precise positioning 

of e.g. fluorescently labelled oligonucleotides (38) or 

nanotubes (39). Here, we add an alternative application 

to the DNA origami toolbox presenting origami as a 

bio-compatible surface that can match ensemble and 

single-molecule measurements. 

In order to evaluate whether DNA origami can serve as 

a transfer platform we chose two biological relevant 

reporter systems, the DNA HJ and the TBP-induced 

bending of DNA. For both model systems the conformational 

flexibility of the DNA component(s) engineered to 

the origami surface is mandatory for functionality and can 

be monitored via the change of a FRET signal. Thereby, 

employing ensemble and single-molecule fluorescence 

spectroscopy, we were able to assess in detail whether a 

biological assay can be established on an origami surface 

that can be directly used in both types of fluorescence 

measurements (Figure 1). 

Holliday junction 

The HJ is a mobile DNA junction composed of four individual 

DNA strands (Figure 2A). In presence of bivalent 

metal ions (here Mg 2+ ) the HJ switches between two 

conformational states iso I and iso II (Figure 2B). These 

states can be detected and distinguished when the 

FRET-efficiency of a Cy3–Cy5 donor-acceptor pair 

attached to strands H and B (Figure 2A) is monitored. 

The iso I state causes a low FRET signal whereas in the 

iso II state the fluorophores are in close proximity resulting 

in a high FRET signal. Bulk fluorescence measurements 

average the dynamic behaviour and the resulting FRET efficiency 

is independent of the MgCl 2 concentration 

(Supplementary Figures S2 and S3). In contrast, at the 

single-molecule level the fast switching between conformational 

states can be resolved. The fluorescence of freely 

diffusing HJ molecules can be monitored in a confocal 

fluorescence microscope equipped with alternating laser 

excitation (ALEX). Because of the fast transition rates at 

low magnesium concentrations the FRET efficiency of the 

two states cannot be resolved and leads to an averaged 

E-value of 0.5 (Figure 2C). In contrast, at higher magnesium 

concentrations of 100 mM the transition rate slows 

down so that the single molecule under investigation 

stays in either of the states while it is diffusing through the 

focus, resulting in two FRET populations with defined 

mean FRET efficiency values of E iso I = 0.36 (±0.10) and 

E iso II = 0.74 (±0.08) (Figure 2C). The kinetics can be 

monitored in real-time in a widefield setup with total 

internal reflection (TIR, Figure 3). Transition rates are 

calculated with the help of two-state hidden Markov 

modelling (40). Increasing the magnesium concentration 

from 40 to 400 mM leads to a 2.5-fold reduced transition 


Figure 3. (A) Pseudo coloured wide-field image of HJ–origami immobilized on a PEG surface with 200 mM MgCl 2 added to the buffer, excited with 

532-nm laser light (scale bar = 10 mm). Green spots indicate donor (Cy3) only HJs, orange spots reveal FRET to the acceptor Cy5. (B) Fluorescence 

time transient of the circled HJ in (A) (upper trace). The Cy3 and Cy5 intensities show anti-correlated behaviour until the acceptor bleaches (at 9 s). 

The FRET efficiency changes rapidly between 0.2 (iso I) and 0.7 (iso II) (lower trace). Based on a two-state Hidden Markov Model (HMM) these 

changes are recognized and counted. Only FRET efficiencies between 0.15 and 0.8 are taken into account to rule out dye bleaching or dark states. 

(C) Transition rates in dependence of the magnesium concentration. All curves show decreasing rate values at higher Magnesium concentrations.


rate from 4 transitions/s to 1.5 transitions/s (Figure 3C). 

These data are in very good agreement with previously 

reported results (41,42). As judged from the FRET efficiency 

distributions and calculated transition rates, the 

kinetics of the transition and the spatial arrangement of 

HJ arms are not affected if the mobile HJ is attached to 

the origami. These results suggest that the HJ is able to 

freely undergo conformational switching without any 

effect of the DNA-nanostructure. An additional benefit of 

the origami pseudo-surface is the slower diffusion of the 

high-molecular weight origami (4.6 MDa) accompanied 

by an unusual form factor in solution that leads to a 

higher photon count per burst. It also increases the potential 

to study biomolecular dynamics in the lower millisecond 

timescale on diffusing molecules. Based on the Stokes– 

Einstein equation the difference in the diffusion constant 

can be calculated in relation to the molecular weight of molecules. 

The molecular weight of the DNA origami is 4600 

times higher as compared to a fluorescence dye (1 kDa). 

Since the cube of the diffusion constant scales approximately 

with the inverse of the molar mass the diffusion 

co-efficient should be of the order of 17-fold smaller than 

that of the single dye. For the TATA oligo (51 bp) we expect 

decrease of the diffusion constant by 5-fold. 

Experimentally, we found a decrease of the diffusion 

constant by a factor of 4 by burst length analysis. 

TBP induced bending of TATA–box DNA 

The TBP is one of the general transcription initiation 

factors found in Archaea and Eukaryotes (43). 

Recognition of the TATA–box motif upstream of the 

transcription start site by TBP allows the assembly of the 

transcription initiation complex at the promoter. Binding 

of TBP to the minor groove of the DNA is driven by hydrophobic 

interactions and leads to a significant bending of the 

DNA helix (Figure 4B) (44). We monitored the TBPinduced 

bending of DNA via a FRET signal between a 

donor (Cy3B) and acceptor (Atto647N) fluorophore 

attached to either end of a TATA–box containing 

dsDNA in free solution or attached to the origami surface 

(Figure 4A). The addition of TBP led to an increase in 

FRET, and fitting of the data yielded a dissociation 

constant of 21.5 (±1.1) nM (Figure 4C). The affinity 

between TBP and the TATA–box oligonucleotide 

attached to the origami is identical within experimental 

errors (K d = 21.7 ± 2.0 nM). The congruent affinities 

indicate that the four identical (and after assembly of the 

DNA origami double stranded) TATA–box sequence 

motifs (5 0 -TTTAAA-3 0 ) found in the M13mp18 scaffold 

are not accessible for TBP. Otherwise these motifs would 

act as competing TBP-binding sites and the K d would 

clearly deviate from the measurements with the isolated 

TATA–oligonucleotide. To further investigate this we 

carried out a competition experiment between a pre-formed 

labelled TATA–oligonucleotide/TBP complex at halfsaturation 

(50 nM TBP) and the DNA origami (Supplementary 

Figure S7). Addition of the origami did not 

result in any significant change in FRET efficiency 

providing clear evidence that the TATA–box motifs in 

the DNA origami are not available for TBP. Mei et al. 

Figure 4. (A) TATA–oligonucleotide used in this study showing the 

position of the donor fluorophore Cy3B and the acceptor Atto647N. 

The TATA–box sequence is highlighted (cyan box). Coupling of a 

biotin at the 5 0 -end of the upper strand allows direct immobilization 

of the TATA–oligo on a cover slide. For the attachment of the TATA– 

DNA to the origami surface the upper strand has been extended by the 

staple strand sequence (see ‘Materials and Methods’ section). (B) 

Addition of the TBP to the TATA–oligonucleotide induces a bending 

of the DNA. Bending of the DNA brings the fluorophores in close 

proximity which results in an increased FRET efficiency. (C) 

Ensemble fluorescence titration of 25 nM free TATA–oligonucleotide 

(filled circles) or 10 nM of origami-attached TATA–oligonucleotide 

(filled rectangle) was titrated with TBP while the FRET signal was 

observed (excitation at 515 nm, emission at 660 nm). The curve shows 

the best fit of the data to a quadratic equation yielding a dissociation 

constant (K d ) of 21.5 ± 1.1 (free oligonucleotide) and 21.7 ± 2.0 (oligo 

on DNA origami). The titration was carried out at 55 C. (D) Single 

molecule ALEX measurements on freely diffusing molecules using the 

Cy3B–Atto647N labelled TATA–DNA (top) or the labelled TATA– 

DNA attached to the origami (bottom). Measurements were carried 

out in the presence (right) or absence (left) of TBP. Upon addition 

of TBP the dsDNA bends and the donor and acceptor fluorophores 

get in close proximity. Consequently, in addition to the low FRET 

population of the straight dsDNA (E oligo = 0.29, E origami = 0.29) a 

high FRET population (E oligo = 0.47 and E origami = 0.51) can be 

detected in the presence of TBP (right). The emergence of the high 

FRET population is independent on whether or not the TATA–oligo 

is attached to the origami. Measurements were carried out at 25 C. 

showed that DNA origami are even stable in cell lysate 

making it suitable or for in vivo applications (45). It is noteworthy 

that these measurements were carried out at 

elevated temperatures (55 C) because the TBP protein 

used in this study originates from a hyperthermophilic 

organism (Methanocaldococcus jannaschii). Stability tests 

showed that the origami structure is stable and does not 

unfold when exposed to temperatures up to 55 C for at 

least up to 20 min (Supplementary Figure S1). Hence, the 

origami can additionally serve as firm platform to study 



Figure 5. (A) A transient of an immobilized origami-bound TATA–oligo visualizes the rapid binding and unbinding of TBP to the TATA–box (at 

35 C, 1 mM TBP). At 40 s the acceptor ATTO647N bleaches and the donor fluorophore Cy3B shows increased fluorescence. (B) The TBP-induced 

bending is shown in high-time resolution of 20 ms/frame (Seconds 2–5 of panel A). (C and D). FRET efficiency over time is changing between 0.19 

(low FRET of straight DNA) and 0.47 (TBP induced high FRET due to DNA bending). 

biomolecular reactions that require more extreme 

conditions. These data demonstrate that (i) the attached 

DNA is freely available for the TBP interaction and (ii) 

the TBP–TATA–box interaction occurs specifically 

between the exposed dsDNA that encodes the TATA– 

box but not between similar sequence motifs found in the 

M13 DNA sequence that builds up the origami structure. 

At the single-molecule level the TBP-induced increase in 

FRET was monitored employing either TIRF spectroscopy 

(immobilized molecules) or confocal fluorescence 

microscopy (freely diffusing molecules). In the absence of 

TBP a well-defined population at low FRET values 

(E = 0.28 ± 0.07) can be measured for the TATA–oligo 

(Figure 4C). The attachment of the TATA–DNA to the 

origami did not affect the FRET value (E = 0.29 ± 0.07). 

The addition of TBP leads to an increase in FRET 

efficiency as seen in ensemble measurements. Again, the 

presence of the origami does not interfere with the 

TBP–TATA–DNA interaction as the mean FRET values 

of the bent DNA are almost identical (TATA–DNA: E = 

0.47 ± 0.10, TATA–DNA/origami: E = 0.51 ± 0.08). 

Immobilization of the TATA–oligo decorated origami on 

a quartz slide makes it possible to study the TBP–TATA 

interaction with a time-resolution in the millisecond range 

(Figure 5). Hence, the association and dissociation of a 

nucleic acid-binding protein like TBP can be monitored in 

real-time adding another level of information, namely 

kinetic parameters like complex lifetime and transition rates 

that cannot be obtained from ensemble measurements. 

Taken together, we introduced the recently developed 

DNA origami as a biocompatible platform that paves the 

way for a reliable and direct transfer of a fluorescence-based 

experiment from the ensemble to the single-molecule level. 

Making use of the DNA origami platform, the conditions 

for a biomolecular reaction can be optimized first on the 

ensemble level and subsequently the very same experiment 

can be transferred to the single-molecule level. Here, the 

DNA origami provides an identical nano-environment 

that ensures highly congruent reaction conditions, which 

consequently leads to comparable results and adds all the 

benefits of the single-molecule technique. 

For optimal comparability of single-molecule and 

ensemble data we suggest the following procedure that 

can be applied to a wide range of studies on biomolecular 

mechanisms: (i) the functionality of the biomolecules is 

proven in an ensemble fluorescence assay using a cuvette 

and a spectrometer; (ii) the same assay is carried out with 

the biomolecules attached to the DNA origami and any 

effect of the origami can directly be detected. In case the 

reaction is affected by the DNA origami pseudo-surface 

the biocompatibility can be further improved e.g. by using 

pegylated oligonucleotides to convert the DNA origami 

into a PEG surface that can still be used to match 

single-molecule and ensemble measurements; and (iii) a 

transfer of the assay to the single-molecule setup is now 

possible as the DNA origami platform can be immobilized 

on the cover slide via its biotin-moieties. Following this 

scheme the researcher has a high confidence that surface 

induced artefacts can be excluded. The DNA origami 

approach is of special interest for studies that require 

immobilized nucleic acids as these can be easily introduced 

into the self-assembled DNA origami and error-prone 

immobilization reactions are avoided. Therefore, 

biomolecular reactions placed on the DNA origami 

surface can be easily studied on the single-molecule level 

and a deeper understanding of the system can be achieved 

as structurally different complexes can be identified and 

time-resolved data are obtained without the need to 

synchronize molecules. The DNA origami immobilization 

scheme is of interest for studies that are compatible with 

immobilized nucleic acids or immobilized proteins via 

linkers or tags. It is especially simple when the 

immobilization is straightforward such as for (i) nucleic 

acids that can adopt multiple structures dependent on 

co-factor, ligand or metal binding (e.g. Q-quadruplexes, 

ribozymes, complex RNA structures, DNA walkers) and 



(ii) protein–nucleic acids interactions (e.g. aptamer–ligand 

interaction, molecular motors acting on nucleic acids, 

transcription factors). We therefore suggest the origamiplatform 

to bridge the often encountered gap between 

ensemble and single-molecule methods. 

Finally, the demonstration of biocompatibility and 

unchanged kinetics in solution and on the DNA origami 

is fundamental for the design of new DNA origami applications 

that imply biomolecular dynamics including 

molecular computing, aritifical molecular machines, molecular 

assembly lines and nanorobots (23,46,47). 

SUPPLEMENTARY DATA 

Supplementary Data are available at NAR Online: 

Supplementary Figures S1–S7. 

ACKNOWLEDGEMENTS 

The authors like to thank Emmanuel Margeat for his 

advice and help in constructing a PRISM TIRF microscope 

as well as Ingo H. Stein for his support with 

respect to confocal solution measurements. 

FUNDING 

ERC starting grant (SiMBA), and a grant from the DFG 

[TI329/5-1]. Funding for open access charge: ERC starting 

grant [ERC-2010-StG-20091118]. 

Conflict of interest statement. None declared. 

REFERENCES 

1. Myong,S., Stevens,B.C. and Ha,T. (2006) Bridging conformational 

dynamics and function using single-molecule spectroscopy. 

Structure, 14, 633–643. 

2. Tinnefeld,P. and Sauer,M. (2005) Branching out of 

single-molecule fluorescence spectroscopy: challenges for chemistry 

and influence on biology. Angew. Chem. Int. Ed. Engl., 44, 

2642–2671. 

3. Kim,H.D., Nienhaus,G.U., Ha,T., Orr,J.W., Williamson,J.R. and 

Chu,S. (2002) Mg2+-dependent conformational change of RNA 

studied by fluorescence correlation and FRET on immobilized 

single molecules. Proc. Natl Acad. Sci. USA, 99, 4284–4289. 

4. Schwille,P., Haupts,U., Maiti,S. and Webb,W.W. (1999) 

Molecular dynamics in living cells observed by fluorescence 

correlation spectroscopy with one- and two-photon excitation. 

Biophys. J., 77, 2251–2265. 

5. Amirgoulova,E.V., Groll,J., Heyes,C.D., Ameringer,T., Rocker,C., 

Moller,M. and Nienhaus,G.U. (2004) Biofunctionalized polymer 

surfaces exhibiting minimal interaction towards immobilized 

proteins. Chemphyschem., 5, 552–555. 

6. Groll,J., Amirgoulova,E.V., Ameringer,T., Heyes,C.D., Rocker,C., 

Nienhaus,G.U. and Moller,M. (2004) Biofunctionalized, ultrathin 

coatings of cross-linked star-shaped poly(ethylene oxide) allow 

reversible folding of immobilized proteins. J. Am. Chem. Soc., 

126, 4234–4239. 

7. Okumus,B., Wilson,T.J., Lilley,D.M. and Ha,T. (2004) Vesicle 

encapsulation studies reveal that single molecule ribozyme 

heterogeneities are intrinsic. Biophys. J., 87, 2798–2806. 

8. Rasnik,I., McKinney,S.A. and Ha,T. (2005) Surfaces and 

orientations: much to FRET about? Acc. Chem. Res., 38, 

542–548. 

9. Okumus,B., Arslan,S., Fengler,S.M., Myong,S. and Ha,T. (2009) 

Single molecule nanocontainers made porous using a bacterial 

toxin. J. Am. Chem. Soc., 131, 14844–14849. 

10. Talaga,D.S., Lau,W.L., Roder,H., Tang,J., Jia,Y., 

DeGrado,W.F. and Hochstrasser,R.M. (2000) Dynamics and 

folding of single two-stranded coiled-coil peptides studied by 

fluorescent energy transfer confocal microscopy. Proc. Natl 

Acad. Sci. USA, 97, 13021–13026. 

11. Cisse,I., Okumus,B., Joo,C. and Ha,T. (2007) Fueling protein 

DNA interactions inside porous nanocontainers. Proc. Natl Acad. 

Sci. USA, 104, 12646–12650. 

12. Dickson,R.M., Norris,D.J., Tzeng,Y.L. and Moerner,W.E. (1996) 

Three-dimensional imaging of single molecules solvated in pores 

of poly(acrylamide) gels. Science, 274, 966–969. 

13. Ishitsuka,Y., Okumus,B., Arslan,S., Chen,K.H. and Ha,T. (2010) 

Temperature-independent porous nanocontainers for 

single-molecule fluorescence studies. Anal. Chem., 82, 9694–9701. 

14. Koopmans,W.J., Schmidt,T. and van Noort,J. (2008) Nucleosome 

immobilization strategies for single-pair FRET microscopy. 

Chemphyschem., 9, 2002–2009. 

15. Kufer,S.K., Dietz,H., Albrecht,C., Blank,K., Kardinal,A., Rief,M. 

and Gaub,H.E. (2005) Covalent immobilization of recombinant 

fusion proteins with hAGT for single molecule force 

spectroscopy. Eur Biophys. J., 35, 72–78. 

16. Lee,J.Y., Okumus,B., Kim,D.S. and Ha,T. (2005) Extreme 

conformational diversity in human telomeric DNA. Proc. Natl 

Acad. Sci. USA, 102, 18938–18943. 

17. Wennmalm,S., Edman,L. and Rigler,R. (1997) Conformational 

fluctuations in single DNA molecules. Proc. Natl Acad. Sci. USA, 

94, 10641–10646. 

18. Seeman,N.C. (2010) Nanomaterials based on DNA. Annu. Rev. 

Biochem., 79, 65–87. 

19. Torring,T., Voigt,N.V., Nangreave,J., Yan,H. and Gothelf,K.V. 

(2011) DNA origami: a quantum leap for self-assembly of 

complex structures. Chem. Soc. Rev., 40, 5636–5646. 

20. Pinheiro,A.V., Han,D., Shih,W.M. and Yan,H. (2011) Challenges 

and opportunities for structural DNA nanotechnology. Nat. 

Nanotechnol., 6, 763–772. 

21. Rajendran,A., Endo,M. and Sugiyama,H. (2012) Single-molecule 

analysis using DNA origami. Angew. Chem. Int. Ed. Engl., 51, 

874–890. 

22. Jungmann,R., Scheible,M., Kuzyk,A., Pardatscher,G., Castro,C.E. 

and Simmel,F.C. (2011) DNA origami-based nanoribbons: 

assembly, length distribution, and twist. Nanotechnology, 22, 

275301. 

23. Simmel,F.C. (2012) DNA-based assembly lines and nanofactories. 

Curr. Opin. Biotechnol., In press. 

24. Rothemund,P.W. (2006) Folding DNA to create nanoscale shapes 

and patterns. Nature, 440, 297–302. 

25. Grohmann,D., Klose,D., Fielden,D. and Werner,F. (2011) FRET 

(fluorescence resonance energy transfer) sheds light on 

transcription. Biochem. Soc. Trans., 39, 122–127. 

26. Cordes,T., Vogelsang,J. and Tinnefeld,P. (2009) On the 

mechanism of Trolox as antiblinking and antibleaching reagent. 

J. Am. Chem. Soc., 131, 5018–5019. 

27. Rasnik,I., McKinney,S.A. and Ha,T. (2006) Nonblinking and 

long-lasting single-molecule fluorescence imaging. Nat. Methods, 3, 

891–893. 

28. Douglas,S.M., Dietz,H., Liedl,T., Hogberg,B., Graf,F. and 

Shih,W.M. (2009) Self-assembly of DNA into nanoscale 

three-dimensional shapes. Nature, 459, 414–418. 

29. Steinhauer,C., Jungmann,R., Sobey,T.L., Simmel,F.C. and 

Tinnefeld,P. (2009) DNA origami as a nanoscopic ruler for 

super-resolution microscopy. Angew. Chem. Int. Ed. Engl., 48, 

8870–8873. 

30. Werner,F. and Weinzierl,R.O. (2002) A recombinant RNA 

polymerase II-like enzyme capable of promoter-specific 

transcription. Mol. Cell, 10, 635–646. 

31. Roy,R., Hohng,S. and Ha,T. (2008) A practical guide to 

single-molecule FRET. Nat. Methods, 5, 507–516. 

32. Nir,E., Michalet,X., Hamadani,K.M., Laurence,T.A., 

Neuhauser,D., Kovchegov,Y. and Weiss,S. (2006) Shot-noise 

limited single-molecule FRET histograms: comparison between 

theory and experiments. J. Phys. Chem. B., 110, 22103–22124. 



33. Kapanidis,A.N., Lee,N.K., Laurence,T.A., Doose,S., Margeat,E. 

and Weiss,S. (2004) Fluorescence-aided molecule sorting: analysis 

of structure and interactions by alternating-laser excitation of 

single molecules. Proc. Natl Acad. Sci. USA, 101, 8936–8941. 

34. Pinheiro,A.V., Han,D., Shih,W.M. and Yan,H. (2011) Challenges 

and opportunities for structural DNA nanotechnology. Nat. 

Nanotechnol., 6, 763–772. 

35. Stein,I.H., Steinhauer,C. and Tinnefeld,P. (2011) Single-molecule 

four-color FRET visualizes energy-transfer paths on DNA 

origami. J. Am. Chem. Soc., 133, 4193–4195. 

36. Berardi,M.J., Shih,W.M., Harrison,S.C. and Chou,J.J. (2011) 

Mitochondrial uncoupling protein 2 structure determined by 

NMR molecular fragment searching. Nature, 476, 109–113. 

37. Sannohe,Y., Endo,M., Katsuda,Y., Hidaka,K. and Sugiyama,H. 

(2010) Visualization of dynamic conformational switching of the 

G-quadruplex in a DNA nanostructure. J. Am. Chem. Soc., 132, 

16311–16313. 

38. Vogelsang,J., Steinhauer,C., Forthmann,C., Stein,I.H., Person- 

Skegro,B., Cordes,T. and Tinnefeld,P. (2010) Make them blink: 

probes for super-resolution microscopy. Chemphyschem, 11, 

2475–2490. 

39. Maune,H.T., Han,S.P., Barish,R.D., Bockrath,M., Iii,W.A., 

Rothemund,P.W. and Winfree,E. (2010) Self-assembly of carbon 

nanotubes into two-dimensional geometries using DNA origami 

templates. Nat Nanotechnol., 5, 61–66. 

40. McKinney,S.A., Joo,C. and Ha,T. (2006) Analysis of 

single-molecule FRET trajectories using hidden Markov modeling. 

Biophys. J., 91, 1941–1951. 

41. Joo,C., McKinney,S.A., Lilley,D.M. and Ha,T. (2004) Exploring 

rare conformational species and ionic effects in DNA Holliday 

junctions using single-molecule spectroscopy. J. Mol. Biol., 341, 

739–751. 

42. McKinney,S.A., Declais,A.C., Lilley,D.M. and Ha,T. (2003) 

Structural dynamics of individual Holliday junctions. Nat. Struct. 

Biol., 10, 93–97. 

43. Werner,F. and Grohmann,D. (2011) Evolution of multisubunit 

RNA polymerases in the three domains of life. Nat. Rev. 

Microbiol, 9, 85–98. 

44. Masters,K.M., Parkhurst,K.M., Daugherty,M.A. and 

Parkhurst,L.J. (2003) Native human TATA-binding protein 

simultaneously binds and bends promoter DNA without a slow 

isomerization step or TFIIB requirement. J. Biol. Chem., 278, 

31685–31690. 

45. Mei,Q., Wei,X., Su,F., Liu,Y., Youngbull,C., Johnson,R., 

Lindsay,S., Yan,H. and Meldrum,D. (2011) Stability of 

DNA origami nanoarrays in cell lysate. Nano Lett., 11, 

1477–1482. 

46. Gu,H., Chao,J., Xiao,S.-J. and Seeman,N.C. (2010) A 

proximity-based programmable DNA nanoscale assembly line. 

Nature, 465, 202–205. 

47. Jungmann,R., Steinhauer,C., Scheible,M., Kuzyk,A., Tinnefeld,P. 

and Simmel,F.C. (2010) Single-molecule kinetics and 

super-resolution microscopy by fluorescence imaging of transient 

binding on DNA origami. Nano Lett., 10, 4756–4761. 


Published on Web 04/12/2010 

RNA-Binding to Archaeal RNA Polymerase Subunits F/E: A DEER and FRET 

Study 

Dina Grohmann, † Daniel Klose, ‡ Johann P. Klare, ‡ Christopher W. M. Kay, † Heinz-Jürgen Steinhoff, ‡ 

and Finn Werner* ,† 

Institute of Structural and Molecular Biology, UniVersity College London, Darwin Building, Gower Street, London 

WC1E 6BT, United Kingdom, and Department of Physics, UniVersity of Osnabrück, Barbarastrasse 7, 

49076 Osnabrück, Germany 

Received February 26, 2010; E-mail: werner@biochem.ucl.ac.uk 

DNA-dependent RNA polymerases (RNAPs) are complex multisubunit 

enzymes that facilitate transcription of all cellular genomes. 

The archaeal RNAP and eukaryotic RNAP are closely related in 

terms of subunit composition, ternary architecture, and requirements 

for basal transcription factors. Recently hyperthermophilic archaeal 

RNAPs have emerged as versatile model systems for the less 

tractable mesophilic eukaryotic RNAPII. 1,2 During transcription 

RNAP interacts with double-stranded DNA, a 9 base pair DNA/ 

RNA hybrid, and the nascent RNA chain. Biochemical studies have 

shown that the eukaryotic RNAPII subunits RPB4/7 and their 

archaeal homologues F/E bind the emerging RNA transcript in a 

nonsequence-specific manner over a range of approximately 20 

nucleotides. 3,4 This interaction stimulates the processivity of the 

archaeal RNAP. 5 During transcription initiation, prior to RNA 

synthesis, F/E facilitates interaction with the basal transcription 

factor TFE and stimulates DNA-strand separation. 6,7 The molecular 

mechanisms that underlie the interplay between TFE, F/E and RNA 

during transcription are poorly understood but are likely to involve 

dynamic conformational changes of the RNAP-nucleic acid 

complex. 8 Subunits F/E form a stable heterodimer 9 that associates 

stably with the RNAP core 10 (Figure 1a). Structural and biophysical 

studies have provided insights into the topology of downstream 

and upstream DNA and the DNA/RNA hybrid in the context of 

the elongating RNAP. 11,12 However, the nascent transcript and its 

interaction with F/E-like subunits have not yet been captured. 

To characterize the interaction between subunits F/E and RNA 

in greater detail, we carried out an equilibrium fluorescence titration 

monitoring the fluorescence signal of a Cy3-labeled 27nt RNA upon 

addition of increasing amounts of subunits F/E (Figure 1b). The 

addition of F/E led to an increase of fluorescence, and fitting of 

the data yielded a dissociation constant of 0.34 ((0.06) µM, which 

is in good agreement with semiquantitative studies using electrophoretic 

mobility shift assays (0.54 µM). 3 The accurate determination 

of the affinity between RNA and F/E was essential to choose 

the appropriate conditions for the subsequent distance measurements 

in the presence and absence of RNA. 

In this study we have employed pulsed double electron-electron 

resonance (DEER) 13 and fluorescence resonance energy transfer 

(FRET) to probe for subtle alterations in the F/E structure upon binding 

of RNA. Both techniques report reliably on distances and changes 

thereof. 13-15 We introduced single cysteine residues at positions 36 

or 63 in subunit F and 49, 65, or 123 in subunit E, taking advantage 

of the fact that F and E can be expressed in a recombinant form, 

purified, and labeled individually prior to heterodimerization. These 

labeling positions were chosen because (i) they provide a good spatial 

† University College London. 

‡ University of Osnabrück. 

Figure 1. (a) Sketch of the transcription elongation complex. The archaeal 

RNAP is represented by the S. shibatae X-ray structure (PDB: 2WAQ, 8 

gray). RNAP subunits F and E are highlighted in magenta and blue. The 

path of the RNA transcript is indicated with a red dotted line, and the 

downstream duplex DNA binding channel is depicted as a cylinder. (b) 

Equilibrium titration of Cy3-labeled RNA (50 nM) with subunits F/E at 65 

°C. Analysis of the data using a single-site binding model yielded a K d of 

0.34 ((0.06) µM. (c) The probe derivatization sites are highlighted as red 

spheres in the crystal structure of the F/E complex 9 (PDB: 1GO3). The 

probe interactions are indicated with double headed arrows (Cβ-Cβ 

distances in angstrom). The location of the RNA binding interface is 

indicated with a yellow dashed circle. 

coverage along both vertical and horizontal axes of the F/E complex (see 

Figure 1c) so that possible conformational changes in either direction could 

be monitored and (ii) they do not obstruct RNA binding (Figure 1c). 3,5 

This approach allowed labeling with either nitroxide spin-label 

(MTSSL, the resulting spin label side chain is denoted R1) or the 

desired donor-acceptor fluorophor combination suitable for FRET at 

chosen sites in F or E. The structural and functional integrity of the labeled 

F/E complexes was ascertained in transcription elongation assays. 5 

Figure 2 illustrates the resulting DEER data for two double spin 

labeled F/E variants in the presence and absence of RNA. The 

chosen protein and RNA concentration ensured complete saturation 

5954 9 J. AM. CHEM. SOC. 2010, 132, 5954–5955 

10.1021/ja101663d 

© 2010 American Chemical Society

COMMUNICATIONS 

Figure 2. (a) Four-pulse DEER (16, 32 ns for π/2, π pulses and a frequency 

offset of 65 MHz pumping on the center of the nitroxide spectrum) form factors 

F/E ( RNA (blue and black, with the respective fits in green and red) after 

correction with a homogeneous 3D background, temperature T ) 50 K. (b) 

DEER interspin distance distributions obtained by Tikhonov Regularization 

of the traces in (a) for F/E ( RNA (blue and black, respectively). 

of F/E with RNA according to the affinity data derived from the 

equilibrium binding studies. The DEER distance distributions 

obtained by Tikhonov Regularization 13 result in a single interspin 

distance of 27 ( 3 Å for F C36R1 /E C123R1 and 36 ( 4 Å for F C36R1 / 

E C65R1 . These distances correspond well to the Cβ-Cβ distances 

in the F/E structure 9 [PDB: 1GO3] (27 Å for F C36 /E K123C and 34 Å 

for F C36 /E S65C ). The presence of RNA did not alter the interspin 

distance distributions significantly as judged by the form factors 

(Figure 2b). The result indicates that no gross structural rearrangement 

of F/E occurs upon RNA binding. 

Our model organism M. jannaschii is viable between 48 and 94 

°C. Therefore, we also performed FRET measurements of fluorescently 

labeled F/E complexes at the biologically relevant elevated temperature 

of 65 °C. F/E complexes F C63 /E C49 ,F C63 /E C65 , and F C63 /E C123 were 

labeled with the donor fluorophor Alexa350 in subunit F at position 

63 and with the acceptor fluorophor Alexa488 in subunit E at three 

different positions (49, 65, and 123). The emission spectra of all single 

and double labeled F/E complexes are shown in Figure 3. In the 

presence of the acceptor the donor fluorescence decreased and the 

acceptor emission increased accordingly indicating a high FRET 

efficiency. A distinct FRET signal is a prerequisite to monitor subtle 

changes in FRET efficiencies and thereby the distances between two 

probes. We calculated the interprobe distances from the decrease in 

donor emission using the Förster equation (F G63C -E VS65C , -E K123 , 

-E V49C are 52 ( 2, 53 ( 5, and 46 ( 1 Å, respectively; X-ray [pdb 

1GO3 9 ]: 48.2, 52.8, and 27.4 Å). The discrepancy between the 

measured F G63C -E V49C distance and the X-ray structure is likely to 

be due to the relatively long Förster radius of the Alexa 350-488 

pair (R 0 ) 50 Å) compared to the distance between the probes (27.4 

Å). The addition of RNA at saturating concentrations (20 µM) to the 

donor-acceptor labeled F/E complexes had no influence on the relative 

donor and acceptor fluorescence intensities. Both FRET and DEER 

data are in good agreement and support the hypothesis that RNA 

binding has no influence on the overall structure of F/E. 

In summary, we have shown that interspin distance distributions 

within the archaeal F/E complex can be accurately determined by 

DEER. Likewise, FRET can provide information about the structural 

flexibility of a subcomplex of the RNAP machinery, or the lack of 

flexibility. Our results demonstrate that RNAP subunits F/E are 

conformationally stable upon RNA binding, which suggests that the 

F/E complex acts as a rigid guiding rail that directs the emerging RNA 

Figure 3. FRET within the F/E complex. (a-c) The fluorescence emission 

spectra are shown for donor only (F C63 /E C49*A350 , blue line), acceptor only 

(F C63*A488 /E C49 , red line), and donor-acceptor (DA) labeled F/E (black line) 

varying the position of the donor located in E (a: C49, b: C65, c: C123). 

(d-f) Fluorescence emission spectra of DA labeled F/E in the presence 

(red dashed line) or absence (gray line) of a 27nt RNA (20 µM). The protein 

concentration was 50 nM, and the excitation was carried out at 320 nm. 

away from the RNAP. This could possibly prevent the entanglement 

of the transcript with the DNA template during transcription. However, 

the results presented here do not rule out a movement of F/E as a 

rigid body relative to the RNAP core (Figure 1), which in turn could 

trigger a conformational change within the RNAP. 

Acknowledgment. Research at the ISMB was funded by project 

grants from the Wellcome Trust (079351/Z/06/Z) and BBSRC (BB/ 

E008232/1) to Finn Werner. Work at the University of Osnabrück 

was funded by DFG SFB 431 P18. 

References 

(1) Werner, F. Mol. Microbiol. 2007, 65, 1395–404. 

(2) Werner, F.; Weinzierl, R. O. Mol. Cell 2002, 10, 635–46. 

(3) Meka, H.; Werner, F.; Cordell, S. C.; Onesti, S.; Brick, P. Nucleic Acids 

Res. 2005, 33, 6435–44. 

(4) Ujvari, A.; Luse, D. S. Nat. Struct. Mol. Biol. 2006, 13, 49–54. 

(5) Hirtreiter, A.; Grohmann, D.; Werner, F. Nucleic Acids Res. 2010, 38, 585– 

96. 

(6) Andrecka, J.; Lewis, R.; Bruckner, F.; Lehmann, E.; Cramer, P.; Michaelis, 

J. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 135–40. 

(7) Werner, F.; Weinzierl, R. O. Mol. Cell. Biol. 2005, 25, 8344–55. 

(8) Korkhin, Y.; Unligil, U. M.; Littlefield, O.; Nelson, P. J.; Stuart, D. I.; 

Sigler, P. B.; Bell, S. D.; Abrescia, N. G. PLoS Biol. 2009, 7, e102. 

(9) Todone, F.; Brick, P.; Werner, F.; Weinzierl, R. O.; Onesti, S. Mol. Cell 

2001, 8, 1137–43. 

(10) Grohmann, D.; Hirtreiter, A.; Werner, F. Biochem. J. 2009, 421, 339–43. 

(11) Kettenberger, H.; Armache, K. J.; Cramer, P. Mol. Cell 2004, 16, 955–65. 

(12) Andrecka, J.; Treutlein, B.; Arcusa, M. A.; Muschielok, A.; Lewis, R.; Cheung, 

A. C.; Cramer, P.; Michaelis, J. Nucleic Acids Res. 2009, 37, 5803–9. 

(13) Pannier, M.; Veit, S.; Godt, A.; Jeschke, G.; Spiess, H. W. J. Magn. Reson. 

2000, 142, 331–340. 

(14) Jeschke, G.; Chechik, V.; Ionita, P.; Godt, A.; Zimmermann, H.; Banham, 

J.; Timmel, C. R.; Hilger, D.; Jung, H. Appl. Magn. Reson. 2006, 30. 

(15) Heyduk, T. Curr. Opin. Biotechnol. 2002, 13, 292–6. 

JA101663D 

J. AM. CHEM. SOC. 9 VOL. 132, NO. 17, 2010 5955

Molecular Cell 

Article 

The Initiation Factor TFE and the Elongation 

Factor Spt4/5 Compete for the RNAP Clamp 

during Transcription Initiation and Elongation 

Dina Grohmann, 1,6 Julia Nagy, 2 Anirban Chakraborty, 3,4,5 Daniel Klose, 1 Daniel Fielden, 1 Richard H. Ebright, 3,4,5 

Jens Michaelis, 2 and Finn Werner 1, * 

1 

University College London, Institute for Structural and Molecular Biology, Division of Biosciences, Darwin Building, Gower Street, 

London WC1E 6BT, UK 

2 

Department of Chemistry and Center for Integrated Protein Science München, Ludwig-Maximilians-Universität München, 

Butenandtstrasse11, 81377 München, Germany 

3 

Howard Hughes Medical Institute 

4 

Waksman Institute 

5 

Department of Chemistry and Chemical Biology 

Rutgers University, Piscataway, NJ 08902, USA 

6 

Present address: Institute for Physical and Theoretical Chemistry, Braunschweig University of Technology, 38106 Braunschweig, Germany 

*Correspondence: werner@biochem.ucl.ac.uk 

DOI 10.1016/j.molcel.2011.05.030 

SUMMARY 

TFIIE and the archaeal homolog TFE enhance DNA 

strand separation of eukaryotic RNAPII and the 

archaeal RNAP during transcription initiation by an 

unknown mechanism. We have developed a fluorescently 

labeled recombinant M. jannaschii RNAP 

system to probe the archaeal transcription initiation 

complex, consisting of promoter DNA, TBP, TFB, 

TFE, and RNAP. We have localized the position of 

the TFE winged helix (WH) and Zinc ribbon (ZR) 

domains on the RNAP using single-molecule FRET. 

The interaction sites of the TFE WH domain and the 

transcription elongation factor Spt4/5 overlap, and 

both factors compete for RNAP binding. Binding of 

Spt4/5 to RNAP represses promoter-directed transcription 

in the absence of TFE, which alleviates 

this effect by displacing Spt4/5 from RNAP. During 

elongation, Spt4/5 can displace TFE from the RNAP 

elongation complex and stimulate processivity. Our 

results identify the RNAP ‘‘clamp’’ region as a regulatory 

hot spot for both transcription initiation and transcription 

elongation. 

INTRODUCTION 

RNA polymerases (RNAPs) are responsible for DNA-dependent 

transcription in all living organisms (Jun et al., 2011; Werner 

and Grohmann, 2011). In contrast to eukaryotes, who employ 

between three (animal) and five (plant) distinct nuclear RNAPs 

to transcribe distinct and nonoverlapping subsets of genes, 

archaea only have one RNAP. However, the subunit composition 

of the archaeal RNAP, its structure, and its requirements for 

general transcription factors bear close resemblance to those 

of eukaryotic RNAPII (Werner and Grohmann, 2011). The 

archaeal RNAP system offers substantial experimental advantages 

over the eukaryotic counterparts. Thus, it is possible to 

reconstitute an archaeal RNAP from its 12 individual recombinant 

subunits in vitro under defined conditions, a feat that has 

not been achieved in any eukaryotic system to date (Naji et al., 

2007; Werner and Weinzierl, 2002). The ability to reconstitute 

archaeal RNAP in vitro has enabled us to site-specifically introduce 

molecular probes into separate RNAP subunits with the 

aim of characterizing dynamic properties of transcription 

complexes (Grohmann et al., 2010). 

In eukaryotes and archaea, TBP and TFIIB (TFB in archaea) are 

necessary and sufficient to direct transcription initiation from 

strong promoters in vitro (Parvin and Sharp, 1993; Qureshi 

et al., 1997; Werner and Weinzierl, 2002). A third evolutionary 

conserved factor, TFIIE (TFE in archaea), is not strictly required, 

but stimulates initiation by enhancing DNA strand separation 

(Forget et al., 2004; Naji et al., 2007) and in eukaryotes by aiding 

the recruitment of the RNAPII-specific transcription factor TFIIH 

(Holstege et al., 1995; Holstege et al., 1996). TFIIE (TFE) homologs 

can be found in several different RNAP systems. For 

example, eukaryotic RNAPIII includes two subunits, C82 and 

C34, that are homologous to TFIIEa and b, respectively (Geiger 

et al., 2010; Carter and Drouin, 2010). Archaeal TFE consists of 

two principal domains, a winged helix (WH) and a Zinc ribbon 

(ZR) domain, which together are homologous to the N-terminal 

part of the eukaryotic TFIIEa subunit (Bell et al., 2001). In yeast 

the corresponding region of the TFIIEa subunit is sufficient for 

TFIIE activity (Kuldell and Buratowski, 1997). While it has not 

been possible to determine the structure of the full-length 

factors, the structure of the archaeal WH domain from Sulfolobus 

shibatae has been determined by X-ray crystallography (Meinhart 

et al., 2003) and the structure of the ZR domain from human 

TFIIEa by NMR spectroscopy (Okuda et al., 2004). Recently, an 

archaeal homolog of the TFIIEb subunit was identified in a subset 

of archaeal genomes, but nothing is known about its function 

(Blombach et al., 2009). In the absence of complete structural 

Molecular Cell 43, 263–274, July 22, 2011 ª2011 Elsevier Inc. 263


Mechanisms of TFE and Spt4/5 

information about TFE, mechanistic insights into its role in 

transcription initiation come from a variety of biochemical 

experiments. TFE enhances promoter DNA melting during the 

formation of the RNAP-promoter open complex, possibly by interacting 

directly with the DNA nontemplate strand (NTS), and it 

preferentially binds to transcription initiation complexes formed 

on artificially melted ‘‘heteroduplex’’ promoter variants (Naji 

et al., 2007; Werner and Weinzierl, 2005). This is corroborated 

by biochemical evidence from the RNAPII system, where TFIIE 

can be crosslinked to the promoter DNA in the transcription 

bubble (Kim et al., 2000). Using a recombinant in vitro reconstituted 

RNAP system, we have shown that the activity of TFE 

crucially depends on the RNAP ‘‘stalk’’ consisting of subunits 

Rpo4/7 (Todone et al., 2001), which suggested a functional 

and possibly physical interaction between the RNAP stalk and 

TFE (Ouhammouch et al., 2004; Werner and Weinzierl, 2005). 

In order to explore proximities between transcription factors 

and RNAPII in the eukaryotic PIC, Hahn and coworkers derivatized 

yeast RNAP subunits with a photoactivatable crosslinker 

inserted in RPB1 and 2 (corresponding to Rpo1 and 2 in the 

archaeal annotation) and showed that TFIIE could be crosslinked 

to the RNAP clamp motif (Chen et al., 2007). However, this work 

could not provide information on a possible proximity between 

the RNAP stalk and TFIIE. The Rpo4/7 stalk promotes DNA 

melting at suboptimal temperatures (Naji et al., 2007) and plays 

a pivotal role during transcription elongation by enhancing 

processivity in vitro and in vivo (Hirtreiter et al., 2010a; Runner 

et al., 2008). In addition to RNAP subunits Rpo4/7, which 

suppress pausing (Hirtreiter et al., 2010b), several transcription 

elongation factors can release paused transcription elongation 

complexes, among them Spt4/5 (eukaryotes and archaea) and 

NusG (the bacterial homolog of Spt5). Not all NusG homologs 

have the same effect on RNAP, e.g., T. thermophilus NusG has 

been shown to reduce transcription elongation rather than 

increasing it (Sevostyanova and Artsimovitch, 2010). Spt4/5 

and NusG associate with their cognate RNAPs by highly 

conserved interactions between the RNAP clamp coiled-coil 

motif and a hydrophobic depression in the Spt5 and NusG 

(NGN) domains (Hirtreiter et al., 2010a; Mooney et al., 2009b; 

Klein et al., 2011). 

While the last couple of years have seen some new structural 

information on the architecture of transcription initiation 

complexes (Kostrewa et al., 2009; Liu et al., 2010), the position 

and conformation of TFIIF and TFIIE in the complexes has 

remained covert. Protein crosslinking combined with mass 

spectrometry has been used to obtain information about the 

interactions between RNAPII and TFIIF (Chen et al., 2010). For 

complexes where structural information is difficult to obtain 

from standard methodologies, measurement of fluorescence 

resonance energy transfer (FRET) followed by triangulation has 

proven to be successful (Mekler et al., 2002). An extension of 

this technique to the level of single molecules (Joo et al., 2008) 

allows us to obtain information about dynamic aspects (Margittai 

et al., 2003; Rasnik et al., 2004). Triangulation of single-molecule 

FRET (smFRET) distance information, combined with structural 

information and rigorous statistical analysis referred to as nanopositioning 

system (NPS), has been used to study the position of 

the exiting RNA (Andrecka et al., 2008), the influence of transcription 

factor TFIIB on the position of the nascent RNA 

(Muschielok et al., 2008), and the position of nontemplate and 

upstream DNA (Andrecka et al., 2009) in yeast RNAPII transcription 

elongation complexes. 

Here we have used a recombinant in vitro transcription system 

based on the hyperthermophilic archaeon Methanocaldococcus 

jannaschii to investigate the structure and molecular mechanisms 

of the initiation and elongation factors TFE and Spt4/5, 

respectively. Using fluorescently labeled RNAP and TFE variants, 

we have applied the NPS to determine in solution the position 

of TFE in an archaeal preinitiation complex (PIC) consisting 

of RNAP, TBP, TFB, TFE, and promoter DNA. We find that the 

TFE WH domain binds to the RNAP clamp close to the clamp 

coiled-coil motif, and the TFE ZR domain binds at a position 

between the RNAP clamp and the RNAP stalk. Furthermore, 

using in-gel fluorescence quenching experiments, we have 

analyzed the spatial relationship between TFE domains and the 

DNA NTS. Since the binding site on RNAP for TFE identified in 

this work overlaps with the binding site on RNAP for Spt4/5 

identified in previous work (Hirtreiter et al., 2010a), we carried 

out binding competition experiments and compared effects of 

TFE and Spt4/5 on RNAP activity during the initiation and elongation 

phases of transcription. We find that TFE and Spt4/5 

compete for binding to RNAP and RNAP-containing complexes 

and that the relative binding affinities of TFE and Spt4/5 differ 

during initiation and elongation. During initiation, Spt4/5 can 

inhibit transcription, and TFE can efficiently displace Spt4/5 

and overcome this inhibition. In contrast, during elongation, 

Spt4/5 efficiently displaces TFE. Our results identify the RNAP 

clamp as an important interaction site and regulatory hotspot 

for both initiation and elongation factors. They suggest 

that structural differences between RNAP in the PIC and 

TEC—e.g., in the clamp and/or in the position of the NTS—alter 

the affinity for TFE and Spt4/5 in a way that is important for the 

molecular mechanisms of transcription initiation, promoter 

escape, and transcription elongation. 

RESULTS 

TFE Can Interact with Free RNAP, with RNAP 

in the PIC, and with RNAP in the TEC 

In order to characterize the binding of TFE to RNAP, we 

produced fluorescently labeled TFE variants and carried out 

native gel electrophoresis experiments. The structure of 

M. jannaschii TFE has not been solved yet. In order to illustrate 

the size of the two principal TFE domains and to highlight the 

probe incorporation sites, we built a homology model (Experimental 

Procedures) using structural information on the WH 

(Sulfolobus solfataricus TFE, PDB: 1Q1H) and the ZR domains 

(Homo sapiens TFIIEa, PDB: 1VD4) (Figure 1A) and approximating 

the conformations of the interdomain linker and 

the C-terminal tail using minimum-energy considerations. The 

models of the WH and ZR domains show a good overall 

structural alignment with their parental structures (Figure S1). 

Recombinant TFE variants containing p-azido phenylalanine 

at positions 44 (WH domain), 108 (interdomain linker), and 

133 (ZR domain) were produced, purified, and derivatized 

with the fluorescent probe DyLight 549 using Staudinger 

264 Molecular Cell 43, 263–274, July 22, 2011 ª2011 Elsevier Inc.



A 

B 

- 

G133 

C 

RNAP 

F108 

RNAP*TFE 

complex 

C 

G44 

N 

ZR 

WH 

Figure 1. TFE Can Interact Directly with RNAP as 

Component of the Transcription Preinitiation 

Complex and the Ternary Elongation Complex 

(A) A homology model of TFE from M. jannaschii. The 

winged helix domain is highlighted in purple-blue, the ZR 

domain in lemon green. Fluorophore attachment sites are 

shown in red. 

(B) RNAP-TFE complexes; EMSA using TFE 133 *DL549 

(0.74 mM) and RNAP (0.2, 0.4, 1, and 2 mM). 

(C) Fluorescence anisotropy using labeled TFE 44 *Cy3B 

(50 nM) and wild-type RNAP (red curve) or RNAPDRpo4/7 

(black curve). Direct fitting of the titration curves yields a 

K d of 0.2 ± 0.01 mM (wild-type RNAP) and 1.7 ± 0.15 mM 

(RNAPDRpo4/7). 

(D) Complete PICs. EMSA using TFE 133 *DL549 (0.74 mM), 

SSV T6 DNA (666 nM), TBP (8.7 mM), TFB (1 mM), and 

RNAP (1.2 mM). 

(E) TEC-TFE complexes. EMSA using TFE 133 *DL549 

(0.74 mM), TS DNA (15 mM), NTS DNA (20 mM), RNA 

(68 mM), and RNAP (1.2 mM). 

D 

TFE* 

E 

- + + + + RNAP 

- + + + + + RNAP 

- - + - + DNA 

- - - + + + 

- 

- 

- 

- 

- 

+ 

+ 

+ 

+ 

+ 

TFB 

TBP 

- 

- 

- 

- 

- 

+ 

- 

+ 

+ 

- 

+ 

+ 

PIC 

RNAP*TFE 

TFE* 

ligation (Chin et al., 2002) (Experimental Procedures). When 

labeled TFE was incubated with increasing amounts of RNAP, 

a species with lower electrophoretic mobility, corresponding 

to the RNAP-TFE complex, was formed in a concentrationdependent 

manner, indicating that TFE and RNAP can form 

a complex (Figure 1B). To confirm and quantify the interaction, 

we performed fluorescence-anisotropy experiments (Figure 1C). 

Upon addition of RNAP to fluorescently labeled TFE, fluorescence 

anisotropy increased in a concentration-dependent 

manner, with an apparent dissociation constant in the submM 

range (K d = 0.2 ± 0.01 mM). We next investigated the incorporation 

of TFE into the archaeal PIC. The PIC was assembled 

using SSV T6 promoter DNA oligonucleotides (Bell et al., 1999; 

Werner and Weinzierl, 2002), TBP, TFB, RNAP, and fluorescently 

labeled TFE. We utilized a promoter variant containing 

a 4 nucleotide (nt) heteroduplex region ( 3/+1), which previ- 

TS 

NTS 

RNA 

TEC 

RNAP*TFE 

TFE* 

ously has been shown to form very stable 

PICs in the open complex conformation (Figure 

S4) (Werner and Weinzierl, 2005). In the 

presence of all components, a species with 

lower electrophoretic mobility than the RNAP- 

TFE complex was observed, corresponding to 

the complete archaeal PIC (Figure 1D). The 

assembly of the PIC was absolutely dependent 

on TBP and TFB. In order to test whether TFE 

also could associate with RNAP during the 

elongation phase of transcription, we assayed 

the binding of fluorescently labeled TFE to an 

archaeal TEC. RNAP can be recruited in a 

promoter-independent manner to synthetic 

elongation scaffolds consisting of a DNA 

template strand (TS), a nontemplate strand 

(NTS), and a 14 nt RNA oligomer to form a catalytically 

competent TEC (Hirtreiter et al., 

2010a). We find that fluorescently labeled TFE 

can be recruited to the TEC, resulting in the 

formation of a species with slightly but unambiguously 

decreased electrophoretic mobility in a manner 

dependent on the TS, the NTS, and RNA (Figure 1E). 

The Location of TFE within the Archaeal PIC Complex 

After we had established that TFE stably associates with RNAP, 

we sought to identify its precise binding site(s) on RNAP using 

NPS (Muschielok et al., 2008). In NPS, the location of a first entity 

(in this case TFE) relative to a second entity (in this case RNAP) is 

determined through the use of smFRET to obtain distance information 

for a fluorescent probe incorporated within the first entity 

and a set of complementary fluorescent probes incorporated at 

reference sites within the second entity. The use of Bayesian 

parameter estimation allows the computation of the most likely 

position and the three-dimensional uncertainty of the position 

of the fluorescent probe in the first entity (Figure S2). We incorporated 

a fluorescent probe at one site in each TFE domain (i.e., 




Figure 2. The Two TFE Domains Interact 

with Distinct Sites of the RNAP Clamp 

(A) Inferred locations of a fluorescent probe 

attached to residue 44 in the TFE WH domain 

(purple volume) and a fluorescent probe attached 

to residue 133 in the ZR domain (green volume). 

The size of each surface corresponds to 68% 

credible volumes. The X-ray structure of the 

archaeal polymerase of S. solfataricus (Hirata 

et al., 2008) (PDB: 2PMZ) is represented as ribbon, 

and each subunit is color-coded according to the 

convention. 

(B) Histogram of 898 sp-FRET trajectories for 

the FRET pair TFE-Rpo2 00 (TFE 44APA *Cy3B and 

Rpo2 00373APA *DL649 ). The single peak can be fitted 

with a Gaussian distribution that is centered at 

E = 0.74. 

(C) Histogram of 197 sp-FRET trajectories for 

the FRET pair TFE-Rpo7 (TFE 44APA *Cy3B and 

Rpo7 S65C *A647 ), a main peak and a smaller side 

peak, which are fitted with Gaussian distributions 

centered at E = 0.29 and E = 0.56, respectively. 

residue 44 in the TFE WH domain and residue 133 in the TFE ZR 

domain), and we incorporated a complementary fluorescent 

probe at each of five reference sites in RNAP (i.e., residue 257 

of Rpo1 0 , residue 373 of Rpo2 00 , residue 11 of Rpo5, residue 49 

of Rpo7, and residue 65 of Rpo7). Archaeal PICs were formed 

by incubating the SSV T6 promoter DNA oligonucleotides with 

TBP, TFB, TFE, and RNAP. For each single-molecule measurement, 

complexes having a fluorescence donor molecule 

attached to a TFE domain and a fluorescent acceptor attached 

to one of the five reference sites on RNAP were prepared. The 

complexes were immobilized and measured in a homebuilt 

TIRF microscope (Experimental Procedures). At least three 

smFRET measurements were performed for each pair of labeling 

sites. The FRET efficiency from all molecules was plotted as 

histograms and fitted with one or two Gaussian functions to 

extract the mean FRET efficiency. Corresponding histograms 

are shown in Figures 2B and 2C. All other histograms are shown 

in Figure S3, and the extracted data are summarized in Tables S1 

and S2. For the NPS localization analysis of the position of 

the WH and the ZR domains of TFE in the PIC, first, the uncertainties 

due to the presence of flexible linkers between the probe 

and RNAP were computed (Figure S2), and the fluorescence 

anisotropies and the isotropic Förster radii were determined 

experimentally (Table S4). Three-dimensional probability densities 

were then calculated as in Andrecka et al., 2009 (Figure 2A 

and Table S3). The results indicate that the TFE WH domain 

interacts with RNAP in the PIC at or near the tip of the RNAP 

clamp coiled-coil motif (see purple volume in Figure 2A, denoting 

position of probe at TFE residue 44) and that the TFE ZR domain 

interacts with the RNAP within the PIC at 

or near the base of the RNAP clamp and 

the RNAP Rpo4/7 stalk (see green 

volume in Figure 2A, denoting position 

of probe at TFE residue 44). For each 

TFE domain, at least one smFRET histogram 

showed an additional minor subpopulation 

(%20% of molecules) (Figures 2C and S3). No 

dynamic switching between the major and minor subpopulations 

was observed. We infer that each TFE domain may have an alternative, 

less favorable, but long-lived binding position. The NPS 

results indicate that, for each TFE domain, the inferred alternative 

binding position is immediately adjacent to the inferred 

primary binding position (Figure S6). 

The WH Domain of TFE Is Located Proximal 

to the Upstream Edge of the Transcription Bubble 

In order to map the relative proximities of the two TFE domains 

and the interdomain linker to the NTS in the context of the PIC, 

we developed a fluorescence quenching assay by assembling 

PICs containing a fluorescence quencher (black hole quencher, 

BHQ-2) incorporated into the NTS at positions 21, 12, 1, +8, 

or +20 (Figure 3A). As in the above experiments, in order to 

ensure that the PIC was in the open complex conformation, we 

used a premelted heteroduplex promoter variant (Figure 3A). 

PICs were assembled with TFE fluorescently labeled at residue 

44 (WH), 108 (linker), or 133 (ZR) and BHQ-2 derivatized or 

wild-type promoter DNA. The complexes were separated on 

native gels, and the PIC TFE fluorescence signal was quantitated 

in situ (Figures 3B–3D). For a positive control, we used fluorescently 

labeled TBP, which exhibited maximal quenching (86% 

quenching efficiency) when BHQ-2 was incorporated at position 

21 just downstream of the TATA element (Figures 3 and S4). 

The TFE WH domain exhibited maximal quenching efficiency 

when BHQ-2 was incorporated at position 12 (76%), which is 

close to the upstream edge of the transcription bubble in the 




A 

B 

-21 -12 -1 +8 +20 

BHQ BHQ BHQ BHQ BHQ 

GATTGATAGA GTAAAGTTTAAATA CTTATATAGATAGAGTATAGATAGAGGGTTCAAAAAATGGTT 

CTAACTATCTCATTTCAAATTTAT GAATATATCTATCTCATAT TCGCCTCCCAAGTTTTTTACCAA 

BRE/TATA 

TSS 

- 

- 

- 

+ 

- 

+ 

+ 

+ 

+ 

+ 

+ 

Q 

RNAP 

TBP/TFB 

DNA 

PIC 

C 

WH [G44] 

L [F108] 

ZR [G133] 

wt 

-21 -12 -1 +8 +20 

NTS 

TS 

Figure 3. Fluorescence Quenching between TFE 

and NTS 

(A) Sequence of the SSV T6 promoter (transcription start 

site, TSS) and the location of BH quenchers. 

(B) PIC EMSA using TFE DL549 (246 nM), RNAP (1.2 mM), 

TBP (8.7 mM), TFB (1 mM), and DNA (667 nM). The 

quencher (Q) incorporated into the DNA nontemplate 

strand reduces fluorescence emission of fluorophores 

incorporated into TFE (shown for TFE 44 *DL549 ). 

(C) PIC EMSAs (concentrations as in B) using individually 

labeled TFE domains (WH, winged helix; L, linker; ZR, 

Zinc ribbon) or labeled TBP (control). The promoter nontemplate 

strand DNA carried the BHQ-2 quencher molecule 

at positions 21, 12, 1, +8, or +20. 

(D) The fluorescence intensity of the PIC band was 

quantified and normalized to nonquenched wild-type (WT) 

PIC (based on at least three independent experiments). 

TFE* 

TBP 

D 

relative Quenching % 

BHQ-position Winged helix [G44] Linker [F108] Zn ribbon [G133] 

wt - - - - 

-21 54 ± 6 63 ± 9 25 ± 3 86 ± 0.3 

-12 76 ± 7 66 ± 12 26 ± 4 48 ± 21 

1 22 ± 6 26 ± 1 13 ± 5 17 ± 3 

8 18 ± 16 23 ± 5 4 ± 2 -7 ± 10 

20 25 ± 9 0 ± 18 -9 ± 12 -10 ± 18 

TEC (Andrecka et al., 2009). The TFE linker exhibited substantial 

quenching when BHQ-2 was incorporated at position 12 (66%) 

or position 21 (63%). The TFE ZR domain did not display 

substantial position-dependent differences in the fluorescence 

signal, suggesting that it is located approximately equidistant 

from the tested BHQ-2 incorporation positions in the NTS. 

The RNAP Clamp Coiled Coil and RNAP Stalk 

Are Required for TFE Binding and Activity 

In order to confirm the identified TFE domain binding sites, we 

made use of two previously described mutant variants of 

RNAP: a mutant in which ten residues of the tip of the RNAP 

clamp coiled-coil motif have been replaced by a tetra-glycine 

linker (the CC-Gly4 mutant) (Hirtreiter et al., 2010a) and a tensubunit 

RNAP subassembly lacking Rpo4/7 (RNAPDRpo4/7) 

(Hirtreiter et al., 2010a, 2010b; Ouhammouch et al., 2004; 

Werner and Weinzierl, 2005) . In electrophoretic mobility shift 

assays (EMSAs), the addition of wild-type RNAP to fluorescently 

labeled TFE yielded a fluorescently labeled species with lower 

electrophoretic mobility, corresponding to the RNAP-TFE 

complex (Figure 4A). In contrast, the addition of the mutant variants 

RNAP CC-Gly4 and RNAPDRpo4/7 failed to yield this 

species. We infer that the tip of the RNAP clamp coiled-coil motif 

and the Rpo4/7 stalk both are important for RNAP-TFE complex 

formation. Control experiments confirmed that both RNAP 

CC-Gly4 and DRpo4/7 are able to form stable PICs in a TBP/ 

TFB-dependent fashion (Figure 4B). In order to quantify the 

contribution of the Rpo4/7 stalk to TFE binding, we repeated 

the fluorescence anisotropy experiments using RNAPDRpo4/7 

and found that the affinity for TFE was lower 

by approximately an order of magnitude (Figure 

1C) (K d = 1.7 ± 0.15 mM). We conclude that 

TBP 

the Rpo4/7 complex is important for the binding 

of TFE to RNAP. We infer that the Rpo4/7 

complex physically interacts with TFE, in agreement 

with the NPS localization of the ZR domain 

described above, and/or allosterically affects 

the conformation of the binding site for TFE. 

We directly observed TFE recruitment to PIC 

using fluorescently labeled TFE in EMSAs. 

Neither RNAP mutant variant was able to recruit TFE into the 

PIC (Figure 4C). In order to monitor the impact of TFE on transcription 

initiation, we developed a promoter-directed transcription 

runoff assay using the SSV T6 promoter. In the presence of 

TBP and TFB, RNAP initiates start-site-specific transcription 

from this strong viral promoter. The linearized plasmid template 

directs the synthesis of a 70 nt runoff transcript (Figure 4D). The 

addition of increasing amounts of TFE stimulates transcription 

without qualitatively altering the transcript pattern (Figure 4D). 

The TFE binding-deficient RNAP variants RNAP CC-Gly4 and 

RNAPDRpo4/7 were able to synthesize the runoff transcript, 

albeit at reduced levels (Figure 4D). However, while transcription 

by the wild-type RNAP was stimulated by TFE about 5-fold, 

neither of the mutant variants was able to respond to TFE to an 

extent comparable to the wild-type RNAP (Figure 4D). 

The Elongation Factor Spt4/5 Can Inhibit PIC Formation 

and Transcription Initiation 

Spt4/5 stimulates the processivity of RNAP (Hirtreiter et al., 

2010a), and while the molecular mechanisms are still not 

completely understood, it is believed that Spt4/5 modulates 

the DNA binding properties of RNAP (Grohmann and Werner, 

2010). We tested the influence of Spt4/5 on the recruitment of 

RNAP to the PIC in EMSAs using fluorescently labeled DNA, 

TBP, and TFB. Interestingly, the addition of Spt4/5 prevented 

the formation of the minimal PIC in a dose-dependent manner 

(Figure 5A). The effect was specific. Thus, a mutant variant of 

Spt4/5 carrying a single substitution (A4R) in the Spt5 NGN 

domain that abrogates RNAP binding failed to exhibit this 




A 

B 

- 

wt Δ Rpo4/7 CC-Gly 4 

wt Δ Rpo4/7 CC-Gly 

- + + + + + + + + + 

- 

- 

- 

- 

- 

+ 

+ 

+ 

- 

- 

- 

+ 

+ 

+ 

- 

- 

- 

+ 

+ 

+ 

4 

RNAP 

TFB 

TBP 

PIC 

DNA* 

RNAP*TFE 

TFE* 

TBP*DNA 

activity (Figure 5A) (Hirtreiter et al., 2010a). In order to determine 

whether this activity was also reflected in transcription initiation, 

we carried out promoter-directed runoff transcription assays. 

Consistent with the results of the recruitment experiments, the 

results of the transcription assays show that the addition of 

Spt4/5 to minimal transcription complexes consisting of DNA, 

TBP, TFB, and RNAP inhibited transcription (Figure 5B) (IC 50 = 

9.6 ± 5 mM) and that Spt4/5-A4R had no effect. 

TFE Efficiently Prevents Inhibition of Transcription 

Initiation by Spt4/5 

Our NPS results (Figure 2A) and our molecular genetics results 

with the RNAP CC-Gly4 mutant (Figure 4A) indicate that the 

binding site of the TFE maps to the same part of RNAP that 

previously has been shown to serve as the binding site for the 

Spt5 NGN domain, i.e., the tip of the RNAP clamp coiled-coil 

motif (Hirtreiter et al., 2010a). To determine whether the binding 

sites for TFE and Spt4/5 overlap, we performed binding competition 

experiments using fluorescently labeled RNAP-TFE 

complexes. The addition of Spt4/5 prevented the formation of 

RNAP-TFE complexes in a concentration-dependent fashion, 

indicating that Spt4/5 and TFE compete for binding to RNAP 

(Figure 5C). The RNAP binding-deficient mutant variant Spt4/5 

A4R had no effect on the RNAP-TFE complexes (Figure 5C). 

The IC 50 of Spt4/5 for the negative effect on the RNAP-TFE 

complex was 0.55 ± 0.14 mM (Figure S5). 

D 

C 

100 

90 

80 

70 

60 

50 

40 

wt 

Δ Rpo 4/7 

CC-Gly 

+ 

4 

+ + 

+ + + 

+ + + 

wt 

RNAP 

DNA 

TFB 

TBP 

PIC 

RNAP*TFE 

TFE* 

Δ 

RNAP 

Rpo4/7 CC-Gly 4 

- - - TFE 

1 3.1 4.9 1 1.2 1.4 1 0.9 1.3 

run-off 

transcript 

Figure 4. Mutations in the RNAP Clamp 

Coiled Coil and the Rpo4/7 Stalk Complex 

Interfere with TFE Recruitment and Activity 

(A) The RNAP-TFE complex; EMSA of TFE-RNAP 

complexes using TFE 133 *DL549 (0.74 mM) and wildtype 

RNAP, RNAPDRpo4/7, or CC-Gly4 (0.5, 1, 

and 2 mM). 

(B) PIC EMSA using fluorescently labeled DNA 

(Alexa 555) as tracer (67 nM), RNAP (1.2 mM), TBP 

(8.7 mM), and TFB (1 mM). 

(C) PIC EMSA using fluorescently labeled TFE 

(TFE DL549 , 0.74 mM), RNAP (1.2 mM), TBP (8.7 mM), 

and TFB (1 mM). 

(D) Promoter-directed transcription assay using 

RNAP (1.2 mM), TBP (17.4 mM), TFB (2 mM), and 

TFE (0, 0.32, and 8 mM). The TFE stimulation is 

tabulated under the lanes. 

We subsequently investigated the 

combined effects of TFE and Spt4/5 on 

formation of the PIC. We assessed effects 

of Spt4/5 on the formation of the 

complete PIC (RNAP, TBP, TFB, TFE, 

and promoter DNA) in EMSAs using fluorescently 

labeled DNA as tracer and 

found that the presence of TFE prevented 

the inhibition of PIC formation by Spt4/5, 

reducing inhibition to the level observed 

with the mutant variant Spt4/5 A4R (Figure 

5A). We repeated the PIC EMSAs 

using fluorescently labeled TFE as tracer 

in order to test whether Spt4/5 could 

displace TFE from the PIC (Figure 5D). We found that Spt4/5 

could displace TFE from the PIC (Figure 5D), but that it could 

do so only very inefficiently, requiring a 50-fold higher concentration 

to displace TFE from the PIC than to displace TFE from 

RNAP-TFE (IC 50 =29±17mM versus IC 50 = 0.55 ± 0.14 mM) 

(Figures 5C and S5). We analyzed whether TFE could prevent 

the inhibition of transcription initiation by Spt4/5. The addition 

of TFE to minimal transcription reactions (DNA, TBP, TFB, and 

RNAP) increased the transcript synthesis by approximately 

5-fold, in agreement with previous observations (Bell et al., 

2001; Naji et al., 2007; Werner and Weinzierl, 2005). In contrast, 

the addition of Spt4/5 inhibited transcript synthesis by more than 

10-fold (Figure 5E). The addition of TFE prevented the Spt4/5- 

dependent inhibition of transcription initiation by Spt4/5, leading 

to transcript levels identical to those in reactions in which Spt4/5 

was omitted (Figure 5E). For a negative control, we used the 

RNAPDRpo4/7 variant, which is defective in TFE binding (Figure 

3A). Under these conditions TFE stimulated transcription 

less than 1.2-fold, Spt4/5 repressed transcription similarly to 

the wild-type RNAP, and TFE only marginally compensated for 

this repression (Figure 5E). 

Spt4/5 Displaces TFE from the TEC 

Our data showed that relative affinities of TFE and Spt4/5 

to RNAP are context dependent: Spt4/5 efficiently displaces 

TFE from the RNAP-TFE complex (IC 50 = 0.55 ± 0.14 mM), but 




A 

- 

B 

100 

- TFE + TFE 

Spt 4/5 Spt 4/5 A4R Spt 4/5 

90 

80 

70 

60 

50 

- 

PIC 

DNA*TBP 

DNA* 

Spt4/5 Spt4/5 A4R run-off 

transcript 

D 

E 

100 

90 

80 

70 

60 

50 

+ 

- 

+ 

- 

- 

+ 

- 

- 

+ 

+ 

+ 

wt 

+ - + 

+ 

- 

- 

TBP/TFB 

Spt 4/5 

PIC 

TFE* 

ΔRpo4/7 

- + - + 

- 

- 

+ 

+ 

RNAP 

TFE 

Spt 4/5 

run-off 

transcript 

Figure 5. Spt4/5 and TFE Compete for 

RNAP Binding during Transcription Initiation, 

and TFE Alleviates the Repression of 

Spt4/5 

(A) The PIC complex is destabilized by Spt4/5 

and rescued by TFE. Fluorescently labeled SSV 

T6 promoter DNA (67 nM) was incubated with 

1.2 mM RNAP, 8.7 mM TBP, and 1 mM TFB in the 

presence or absence of TFE (8 mM) and 

increasing amounts of WT Spt4/5 or the RNAP 

binding-deficient mutant Spt4/5 A4R (5, 18, 60, and 

147 mM). 

(B) Spt4/5 represses promoter-directed transcription 

in the absence of TFE. Reactions 

included RNAP (1.2 mM), TBP (17.4 mM), TFB 

(2 mM), TFE (0, 0.32, and 8 mM), and Spt4/5 or 

Spt4/5 A4R (5, 18, and 55 mM). 

(C) Spt4/5 displaces RNAP-bound TFE. Increasing 

amounts of Spt4/5 or Spt4/5 A4R (0.33, 1, 

7.5, and 25 mM) were added to a preformed 

RNAP * TFE 133 *DL549 (0.74 mM) complex. 

(D) Addition of increasing amounts of WT Spt4/5 

(0, 16, 32, and 137 mM) to preformed PICs using 

fluorescently labeled TFE (0.75 mM). 

(E) Promoter-directed transcription using either 

WT RNAP or RNAPDRpo4/7 (1.2 mM), TFE 

(8 mM), and Spt4/5 (55.2 mM). The effect of Spt4/ 

5 and TFE on transcription is tabulated under 

the lanes. 

40 

40 

C 

- 

Spt 4/5 Spt 4/5 A4R TFE* 

RNAP*TFE 

only inefficiently displaces TFE from the PIC (IC 50 =29±17mM). 

In order to test the binding characteristics of TFE and Spt4/5 

during transcription elongation, we carried out binding and 

transcription assays using synthetic elongation scaffolds consisting 

of DNA TS, NTS, and a short RNA primer (RNA). As 

observed previously, TFE forms a complex with RNAP (Figure 

6A). In the presence of TS, NTS, and RNA, a species with 

lower electrophoretic mobility than that of RNAP-TFE appears, 

corresponding to the TEC-TFE complex (Figure 6A). The addition 

of increasing amounts of Spt4/5 efficiently prevented the 

formation of the TEC-TFE complex, with a half-maximal inhibitory 

concentration comparable to the RNAP-TFE complex: 

IC 50 = 0.79 ± 0.07 mM. For negative controls, we made use of 

the RNAP binding-deficient Spt4/5 A4R mutant, which had no 

effect on the TEC-TFE complex (Figure 6A). We complemented 

1 3.2 0.3 2.8 1 1.3 0.2 0.3 

the binding studies with transcription 

elongation assays using synthetic elongation 

scaffolds. RNAP can be recruited 

factor-independently to the scaffolds 

and upon NTP addition extends the 

14-mer RNA primer to form a 72 nt runoff 

transcript (Figure 6B). Whereas the addition 

of TFE has no substantial effect 

on elongation, the addition of Spt4/5 

stimulates the synthesis of the runoff 

transcript, as observed previously (Figure 

6B) (Hirtreiter et al., 2010a). Importantly, 

the Spt4/5 stimulation was not 

significantly reduced by the addition of 

TFE, indicating that Spt4/5 remains associated with the TEC in 

the presence of TFE (Figure 6B). 

DISCUSSION 

The FRET and mutational analyses presented here identify 

two discrete positions on the RNAP clamp as the binding site 

for TFE: the TFE WH domain interacts with the tip of the RNAP 

clamp coiled-coil motif (subunit Rpo1 0 ), and the TFE ZR domain 

interacts with base of the RNAP clamp (Rpo1 0 and Rpo2 00 ) and 

is in close proximity to the RNAP stalk (Rpo4/7) (Figure 2A). 

These binding sites provide a framework for understanding 

published results on TFE and, by inference, TFIIE. First, both 

TFE and TFB interact with the RNAP clamp coiled coil and 

possibly with each other, which may account for why TFE can 




A 

wt 

CC-Gly 

- + + + + + + + + + + + 

wt A4R wt 

- - 

- 

4 

RNAP 

DNA/RNA 

Spt4/5 

TEC 

RNAP*TFE 

TFE* 

B 

100 

90 

80 

70 

60 

50 

40 

- 

- 

+ 

- 

- + 

+ + 

TFE 

Spt4/5 

time 

run-off 

transcript 

Figure 6. Spt4/5 Displaces TFE from 

Transcription Elongation Complexes 

(A) Spt4/5 efficiently competes for TFE binding in 

the TEC. EMSAs were conducted using WT RNAP 

or RNAP CC-Gly4 (0.21 mM), TEC (NTS, 20 mM; TS, 

15 mM; RNA, 68 mM), fluorescently labeled 

TFE G133 *DL549 (0.74 mM), and increasing amounts 

of Spt4/5 or Spt4/5 A4R (1, 2.5, and 15 mM). 

(B) Transcription elongation assay using WT RNAP 

(420 nM), TFE (2.5 mM), and Spt4/5 (10 mM). Spt4/5 

stimulates elongation in the presence of TFE. 

Reactions were stopped at 1.5, 3, and 10 min. The 

runoff transcript levels were quantified and are 

indicated under the lanes. 

complement mutations in the TFB linker region in RNAP recruitment 

and transcription assays (Werner and Weinzierl, 2005). 

Second, the proximity of the WH domain and the NTS at the 

upstream edge of the transcription bubble (Figure 3) accounts 

for the reported crosslinking between TFE and the NTS (Grünberg 

et al., 2007) and between eukaryotic TFIIE and promoter 

DNA in the transcription bubble (Kim et al., 2000). Third, the 

proximity between the TFE ZR domain and the RNAP stalk 

provide a rationale for the Rpo4/7 dependency of TFE activity 

(Naji et al., 2007; Werner and Weinzierl, 2005). The two binding 

sites on the archaeal RNAP for the TFE WH and ZR domains 

are in excellent agreement with the results obtained in the eukaryotic 

RNAPII system by Hahn and coworkers (Chen et al., 

2007) and recent studies in the RNAPIII system. Yeast RNAPII 

subunits Rpb1 and 2 were derivatized with crosslinkers, and 

eukaryotic TFIIE could be crosslinked to residues RPB1 

His213 and 286 (corresponding to Lys186 and Gln259 in 

S. solfataricus) after formation of the PIC. Both residues reside 

in the RNAP clamp domain and are proximal to the location of 

the TFE WH domain identified with NPS (Figure S6). In the 

RNAPIII system, a subcomplex of subunits C82/C34/C31 

(C82/C34 are homologs of TFIIEa and TFIIEb) is stably associated 

with the RNAPIII core and essential for transcription initiation. 

A comparison between the crystal structure of yeast 

RNAPII and the cryo-EM surface envelope of RNAPIII has 

allowed the identification of additional densities that have 

been assigned to RNAPIII-specific subunits (Fernández-Tornero 

et al., 2007; Lefèvre et al., 2011). In congruence to our NPS data, 

both the C82/C34/C31 subcomplex as well as hRPC62 (a 

human ortholog of C82) have been assigned to densities next 

to the clamp and the stalk. 

Our NPS data have enabled us to position the two individual 

WH and ZR domains of archaeal TFE on discrete parts of the 

clamp motif. These two binding sites provide the basis to 

suggest specific structural hypotheses for the mechanism of 

action of TFE and, by inference, TFIIE. First, our results suggest 

that the WH and ZR domains both interact with the RNAP clamp. 

In principle, contacts of the two TFE domains with two different 

sites on the RNAP clamp might help ‘‘prise’’ the RNAP clamp into 

a specific open or closed conformation (Figure 7A). Second, our 

results suggest that the TFE ZR domain interacts with the base 

of the RNAP clamp close to the RNAP stalk. In principle, the 

TFE ZR domain could ‘‘wedge’’ between the RNAP clamp and 

the remainder of RNAP and/or between the RNAP clamp and 

the RNAP stalk, helping to ‘‘lock’’ the RNAP clamp in a specific 

open or closed conformation. By either of the above two hypotheses, 

TFE would induce or stabilize a conformational change that 

would affect the width of the DNA binding channel and thereby 

would affect loading of the template DNA. Our results also 

suggest that the TFE WH domain interacts with the NTS at the 

upstream edge of the transcription bubble. In principle, interactions 

of TFE with the NTS could help favor promoter melting and 

open complex formation. 

The eukaryotic and archaeal transcription elongation factor 

Spt4/5 and its bacterial counterpart NusG previously have 

been shown to interact with the tip of the RNAP clamp coiledcoil 

motif and to stimulate transcription elongation (Hirtreiter 

et al., 2010a). Here, we show that Spt4/5 additionally has an 

opposite effect on transcription initiation: Spt4/5 inhibits PIC 

formation and transcription initiation. Since the Spt4/5 (NusG) 

binding site is located on the tip of the RNAP clamp and is close 

to the RNAP DNA binding channel, it is likely that Spt4/5 (NusG) 

modulates the interaction of RNAP with the DNA and/or with the 

DNA-RNA hybrid (Grohmann and Werner, 2010). In principle, 

Spt4/5 (NusG) might allosterically favor closed conformational 

states of the RNAP clamp, thereby indirectly interfering with 

entry of DNA into the RNAP DNA binding channel and/or 

departure of DNA from the RNAP DNA binding channel. Our 

results provide two additional lines of support for this hypothesis. 

First, Spt4/5 inhibits formation of the PIC and inhibits promoterdependent 

transcription initiation. Second, Spt4/5 stimulates 

transcription elongation in assays that do not involve 

promoter-dependent transcription initiation but instead utilize 

linear DNA-RNA scaffolds. A low-resolution cryo-EM structure 

of the RNAP-Spt4/5 complex and a model based on an X-ray 

structure of a recombinant clamp-Spt4/5NGN complex (both 

from Pyrococcus furiosus) confirm our results (Klein et al., 

2011; Martinez-Rucobo et al., 2011). A density corresponding 

to the Spt4/5NGN core closes the gap across the DNA binding 

channel; it could prevent entry to and release of template DNA 

from the RNAP. 

We discovered that TFE is able to overcome the inhibitory 

effects exerted by Spt4/5 during transcription initiation by virtue 

of competitive displacement of Spt4/5. Figure 7 illustrates 

a working model of TFE and Spt4/5 action during transcription 

initiation and elongation. In our experiments, TFE prevails over 




Figure 7. Molecular Mechanisms of TFE during 

Open Complex Formation 

(A) The TFE WH (highlighted in green) and ZR domains 

(purple blue) interact with the RNAP clamp on the tip of the 

Rpo1 coiled coil (gray spheres) and with Rpo2 (orange 

spheres) at the base of the Rpo4/7 stalk, respectively. The 

Spt5 NGN domain (red) interacts with the clamp coiled coil 

(gray spheres). RNAP structure representation is based on 

S. shibatae (PDB: 2WAQ), and TFE is a homology model of 

M. jannaschii TFE (Experimental Procedures). Rpo2 is 

highlighted in orange, Rpo4 in magenta, Rpo7 in sky blue, 

and Rpo1 and all other RNAP subunits in gray. We 

envisage that the bidentate RNAP-TFE interaction mode 

(indicated with gray block arrows) provides the necessary 

purchase for TFE to close/open the RNAP clamp (dashed 

black circle) and thereby alter the width of the DNA binding 

channel (red block arrow). The movement of the clamp 

(indicated with a spring) is likely to play an important role 

during DNA melting and the loading of the template strand 

into the active site. 

(B) Recruitment pathways during transcription elongation. 

The TATA/TBP/TFB platform can recruit the RNAP-TFE 

complex (1) or first RNAP and subsequently TFE (2) to 

form the preinitiation complex (PIC). Free RNAPs can 

associate with TFE or Spt4/5. The RNAP-Spt4/5 complex 

is barred from efficient recruitment (red cross) to the 

TATA-TBP-TFB platform, but TFE overcomes this 

impediment by displacing RNAP-bound Spt4/5 (3) to form 

RNAP-TFE complexes that are readily recruited to the 

promoter. 

(C) Recruitment of Spt4/5 during transcription elongation. 

Following promoter escape, TFE can remain associated 

with RNAP, forming a TEC-TFE complex. Spt4/5 can 

efficiently displace TFE from the TEC-TFE complex and 

stimulate processivity (4). Alternatively, Spt4/5 engages 

with the PIC at the transition of transcription initiation and 

elongation during promoter escape (5). 

Spt4/5 in the competition for binding to RNAP in the context of 

PIC (possibly due to contacts between TFE and TFB and/or 

possibly due to PIC-specific contacts between TFE and 

RNAP and/or between TFE and the NTS). Once RNAP has 

escaped the promoter, the relative affinities of Spt4/5 and 

TFE are reversed: in the context of the TEC, Spt4/5 prevails 

over TFE in the competition of binding to RNAP (Figure 7C). It 

is also possible that Spt4/5 displaces TFE from the PIC earlier, 

during promoter escape, and thus stimulates transcription at 

the transition between initiation and elongation (Figure 7C). 

Our data thus provide in vitro evidence for a mechanism of 

transcription initiation and elongation factors that compete for 

RNAP binding. The RNAP clamp coiled-coil motif is a conserved 

binding site for the initiation factors TFB and TFE (eukaryotes 

and archaea) and sigma70 (bacteria) and for 

the elongation factors Spt4/5 (eukaryotes and 

archaea) and NusG (bacteria) (Belogurov 

et al., 2007; Hirtreiter et al., 2010a; Kostrewa 

et al., 2009). Moreover, NusG and its paralog 

RfaH compete with sigma70 for binding to 

RNAP (Sevostyanova et al., 2008), highly reminiscent 

of Spt4/5 and TFE in the archaea and, 

by inference, in eukaryotes. NusG has pleiotropic effects on 

elongation; it is a positive elongation factor that increases 

processivity but enhances transcription termination in the 

context of rho (Mooney et al., 2009a). Similarly, Spt4/5 may 

modulate transcription in more than one way. Spt4/5 stimulates 

processivity (Hirtreiter et al., 2010a), and our results presented 

here demonstrate that it inhibits transcription initiation. The eukaryotic 

Spt4/5 complex has multiple KOW domains and 

C-terminal repeat regions and interacts with a plethora of 

factors involved in chromatin remodeling, RNA processing, 

and polyA site selection (Cui and Denis, 2003; Lindstrom 

et al., 2003; Schneider et al., 2006). In summary, Spt5-like transcription 

factors are not only universally conserved in evolution, 

but also highly versatile. 




EXPERIMENTAL PROCEDURES 

Recombinant Protein Production and Labeling 

Unlabeled transcription factors TBP, TFB, TFE, and Spt4/5 were produced as 

described previously (Hirtreiter et al., 2010a; Werner and Weinzierl, 2005). 

Recombinant RNAP was reconstituted as described previously (Werner and 

Weinzierl, 2002). Rpo5 and 7 were labeled with fluorescent probes as 

described previously (Grohmann et al., 2009). Rpo1 0 , Rpo2 00 , and TFE were 

labeled using a nonsense suppressor strategy (Chin et al., 2002) (Supplemental 

Information). 

Comparative Modeling 

The TFE WH domain was modeled based on the S. solfataricus TFE N-terminal 

domain crystal structure (PDB: 1Q1H, resolution 2.9 Å) (Meinhart et al., 2003), 

and the TFE ZR domain was modeled based on the human TFIIEa NMR 

structure ensemble (PDB: 1VD4) (Okuda et al., 2004), both using Modeler 

9.7 (build 6923) (Sali and Blundell, 1993). Stereochemistry was checked using 

Procheck V3.4 (Laskowski et al., 1993). The TM score for the WH domain 

model is 0.93 and the average rmsd 1.24 Å, and for the ZR domain model 

0.62 and 2.09 Å, respectively (Figure S1). The two domains were connected 

by an initially coiled linker of 12 aa missing in the templates and energy minimized 

in a 3 ns unconstrained molecular dynamics simulation (in explicit water 

with ions) using simulated annealing energy minimization with the force field 

Amber99 (Wang et al., 2000) as implemented in Yasara (Krieger et al., 2002). 

Fluorescence Anisotropy 

Fluorescence anisotropy of labeled TFE and TFE-RNAP complexes was 

recorded as previously described (Grohmann et al., 2009). 

PIC Preparation and NPS Experiments 

Nucleic acid scaffolds were used to assemble preinitiation complexes consisting 

of a 65 nt long double-stranded DNA with template and nontemplate DNA 

strands containing a 4 bp mismatch around the active site (m3 template 

[Werner and Weinzierl, 2005]). For surface immobilization of the complexes, 

the nontemplate DNA strand had Biotin attached at the 5 0 end via a C6-amino 

linker. The DNA strands were purchased from IBA (Göttingen, Germany). The 

DNA strands were annealed as described before (Andrecka et al., 2008). The 

PIC complexes were assembled by adding 1 ml each of nucleic acid scaffold 

(2 mM), TBP (10 mM), TFB (10 mM), RNAPDRpo4/7 (2 mM), and Rpo4/7 

(10 mM) to 10 ml HMNE buffer (40 mM HEPES [pH 7.3], 250 mM sodium 

chloride, 2.5 mM magnesium chloride, 0.1 mM EDTA, 5% glycerol and 

10 mM dithiothreitol). The mixture was then incubated at 55 C for 10 min, 

and complete PIC complexes were purified using Microcon-YM100 centrifugal 

filters (Millipore) against HMNE buffer. Then 1 ml TFE (12.4 mM) was added to 

the purified complexes and incubated for 10 min at 55 C. 

NPS was carried out as described previously (Andrecka et al., 2008, 2009; 

Muschielok et al., 2008). For a detailed description of NPS setup and calculations, 

refer to the Supplemental Information. 

Electrophoretic Mobility Shift Assays 

The reaction components indicated in the figure legends were combined on 

ice in 13 HNME buffer, incubated for 20 min at 65 C, and separated on 

10%–12% native Tris-glycine gels or 4%–12% Tris-glycine gradient gels 

(Bio-Rad and Invitrogen) at 200 V for 45 min at room temperature (Werner 

and Weinzierl, 2005). For PIC promoter templates and synthetic elongation 

scaffolds, complementary DNA strands and RNA were annealed by incubation 

for 5 min at 95 C and slowly cooled down to room temperature. The final 

concentrations were as follows: TFE, 740 nM; RNAP, 1.2 mM; TBP, 8.7 mM; 

TFB, 1 mM; TS, 667 nM; NTS, 667 nM; Heparin, 6.7 mg/ml. Fluorescently 

labeled TFE and TFE-containing complexes were visualized on a Fuji 

FLA2000 scanner, and signals were quantified using Image Gauge software 

(Fuji Science Lab 2003). 

Transcription Assays 

Promoter-directed transcription runoff assays were carried out by combining 

666 nM RNAP, 17.5 mM TBP, and 2 mM TFB with 1.5 mg pGEM-SSV T6 linearized 

with NcoI in a total volume of 15 ml (Werner and Weinzierl, 2002). All 

components were combined on ice, and transcription was initiated by the 

addition of 0.75 mM ATP, UTP, and GTP substrates containing 2 mM CTP 

and 75 pM [a- 32 P]CTP (0.3 ml of 3000 Ci/mmol, Perkin Elmer). Ten microliters 

of the reactions were stopped by the addition of 15 ml formamide loading 

buffer. The 32 P-labeled fragments were separated on 10% urea PAGE for 

80 min at 80 W and visualized using a Fuji FLA2000 scanner, and the signals 

were quantified using Image Gauge software (Fuji Science Lab). Transcription 

elongation assays using synthetic elongation scaffolds were carried out as 

previously described (Hirtreiter et al., 2010b). 

SUPPLEMENTAL INFORMATION 

Supplemental Information includes seven figures, four tables, Supplemental 

Experimental Procedures, and Supplemental References and can be found 

with this article online at doi:10.1016/j.molcel.2011.05.030. 

ACKNOWLEDGMENTS 

This work was supported by Wellcome Trust grant 079351/Z/06/Z and BBSRC 

grants BB/E008232/1 and BB/H019332/1 to F.W., DFG SFB 646 and additional 

financial support by Nanosystems Initiative Munich (NIM) to J.M., and 

by National Institutes of Health grant GM41376 and a Howard Hughes Medical 

Institute Investigatorship to R.H.E. D.K. received support from grant DFG SFB 

431 P18 to Heinz-Jürgen Steinhoff. We thank Peter Schultz for plasmids. 

Received: December 20, 2010 

Revised: March 9, 2011 

Accepted: May 24, 2011 

Published: July 21, 2011 

REFERENCES 

Andrecka, J., Lewis, R., Brückner, F., Lehmann, E., Cramer, P., and Michaelis, 

J. (2008). Single-molecule tracking of mRNA exiting from RNA polymerase II. 

Proc. Natl. Acad. Sci. USA 105, 135–140. 

Andrecka, J., Treutlein, B., Arcusa, M.A., Muschielok, A., Lewis, R., Cheung, 

A.C., Cramer, P., and Michaelis, J. (2009). Nano positioning system reveals 

the course of upstream and nontemplate DNA within the RNA polymerase II 

elongation complex. Nucleic Acids Res. 37, 5803–5809. 

Bell, S.D., Kosa, P.L., Sigler, P.B., and Jackson, S.P. (1999). Orientation of the 

transcription preinitiation complex in archaea. Proc. Natl. Acad. Sci. USA 96, 

13662–13667. 

Bell, S.D., Brinkman, A.B., van der Oost, J., and Jackson, S.P. (2001). The 

archaeal TFIIEalpha homologue facilitates transcription initiation by enhancing 

TATA-box recognition. EMBO Rep. 2, 133–138. 

Belogurov, G.A., Vassylyeva, M.N., Svetlov, V., Klyuyev, S., Grishin, N.V., 

Vassylyev, D.G., and Artsimovitch, I. (2007). Structural basis for converting 

a general transcription factor into an operon-specific virulence regulator. 

Mol. Cell 26, 117–129. 

Blombach, F., Makarova, K.S., Marrero, J., Siebers, B., Koonin, E.V., and van 

der Oost, J. (2009). Identification of an ortholog of the eukaryotic RNA polymerase 

III subunit RPC34 in Crenarchaeota and Thaumarchaeota suggests 

specialization of RNA polymerases for coding and non-coding RNAs in 

Archaea. Biol. Direct 4, 39. 

Carter, R., and Drouin, G. (2010). The increase in the number of subunits in 

eukaryotic RNA polymerase III relative to RNA polymerase II is due to the 

permanent recruitment of general transcription factors. Mol. Biol. Evol. 27, 

1035–1043. 

Chen, H.T., Warfield, L., and Hahn, S. (2007). The positions of TFIIF and TFIIE in 

the RNA polymerase II transcription preinitiation complex. Nat. Struct. Mol. 

Biol. 14, 696–703. 

Chen, Z.A., Jawhari, A., Fischer, L., Buchen, C., Tahir, S., Kamenski, T., 

Rasmussen, M., Lariviere, L., Bukowski-Wills, J.C., Nilges, M., et al. (2010). 

Architecture of the RNA polymerase II-TFIIF complex revealed by cross-linking 

and mass spectrometry. EMBO J. 29, 717–726. 




Chin, J.W., Santoro, S.W., Martin, A.B., King, D.S., Wang, L., and Schultz, P.G. 

(2002). Addition of p-azido-L-phenylalanine to the genetic code of Escherichia 

coli. J. Am. Chem. Soc. 124, 9026–9027. 

Cui, Y., and Denis, C.L. (2003). In vivo evidence that defects in the transcriptional 

elongation factors RPB2, TFIIS, and SPT5 enhance upstream poly(A) 

site utilization. Mol. Cell. Biol. 23, 7887–7901. 

Fernández-Tornero, C., Böttcher, B., Riva, M., Carles, C., Steuerwald, U., 

Ruigrok, R.W., Sentenac, A., Müller, C.W., and Schoehn, G. (2007). Insights 

into transcription initiation and termination from the electron microscopy structure 

of yeast RNA polymerase III. Mol. Cell 25, 813–823. 

Forget, D., Langelier, M.F., Thérien, C., Trinh, V., and Coulombe, B. (2004). 

Photo-cross-linking of a purified preinitiation complex reveals central roles 

for the RNA polymerase II mobile clamp and TFIIE in initiation mechanisms. 

Mol. Cell. Biol. 24, 1122–1131. 

Geiger, S.R., Lorenzen, K., Schreieck, A., Hanecker, P., Kostrewa, D., Heck, 

A.J., and Cramer, P. (2010). RNA polymerase I contains a TFIIF-related 

DNA-binding subcomplex. Mol. Cell 39, 583–594. 

Grohmann, D., and Werner, F. (2010). Hold on!: RNA polymerase interactions 

with the nascent RNA modulate transcription elongation and termination. RNA 

Biol. 7, 310–315. 

Grohmann, D., Hirtreiter, A., and Werner, F. (2009). RNAP subunits F/E (RPB4/ 

7) are stably associated with archaeal RNA polymerase: using fluorescence 

anisotropy to monitor RNAP assembly in vitro. Biochem. J. 421, 339–343. 

Grohmann, D., Klose, D., Klare, J.P., Kay, C.W., Steinhoff, H.J., and Werner, F. 

(2010). RNA-binding to archaeal RNA polymerase subunits F/E: a DEER and 

FRET study. J. Am. Chem. Soc. 132, 5954–5955. 

Grünberg, S., Bartlett, M.S., Naji, S., and Thomm, M. (2007). Transcription 

factor E is a part of transcription elongation complexes. J. Biol. Chem. 282, 

35482–35490. 

Hirata, A., Klein, B.J., and Murakami, K.S. (2008). The X-ray crystal structure of 

RNA polymerase from Archaea. Nature 451, 851–854. 

Hirtreiter, A., Damsma, G.E., Cheung, A.C., Klose, D., Grohmann, D., Vojnic, 

E., Martin, A.C., Cramer, P., and Werner, F. (2010a). Spt4/5 stimulates transcription 

elongation through the RNA polymerase clamp coiled-coil motif. 

Nucleic Acids Res. 38, 4040–4051. 

Hirtreiter, A., Grohmann, D., and Werner, F. (2010b). Molecular mechanisms of 

RNA polymerase—the F/E (RPB4/7) complex is required for high processivity 

in vitro. Nucleic Acids Res. 38, 585–596. 

Holstege, F.C., Tantin, D., Carey, M., van der Vliet, P.C., and Timmers, H.T. 

(1995). The requirement for the basal transcription factor IIE is determined 

by the helical stability of promoter DNA. EMBO J. 14, 810–819. 

Holstege, F.C., van der Vliet, P.C., and Timmers, H.T. (1996). Opening of an 

RNA polymerase II promoter occurs in two distinct steps and requires the 

basal transcription factors IIE and IIH. EMBO J. 15, 1666–1677. 

Joo, C., Balci, H., Ishitsuka, Y., Buranachai, C., and Ha, T. (2008). Advances in 

single-molecule fluorescence methods for molecular biology. Annu. Rev. 

Biochem. 77, 51–76. 

Jun, S.H., Reichlen, M.J., Tajiri, M., and Murakami, K.S. (2011). Archaeal RNA 

polymerase and transcription regulation. Crit. Rev. Biochem. Mol. Biol. 46, 

27–40. 

Kim, T.K., Ebright, R.H., and Reinberg, D. (2000). Mechanism of ATP-dependent 

promoter melting by transcription factor IIH. Science 288, 1418–1422. 

Klein, B.J., Bose, D., Baker, K.J., Yusoff, Z.M., Zhang, X., and Murakami, K.S. 

(2011). RNA polymerase and transcription elongation factor Spt4/5 complex 

structure. Proc. Natl. Acad. Sci. USA 108, 546–550. 

Kostrewa, D., Zeller, M.E., Armache, K.J., Seizl, M., Leike, K., Thomm, M., and 

Cramer, P. (2009). RNA polymerase II-TFIIB structure and mechanism of 

transcription initiation. Nature 462, 323–330. 

Krieger, E., Koraimann, G., and Vriend, G. (2002). Increasing the precision of 

comparative models with YASARA NOVA—a self-parameterizing force field. 

Proteins 47, 393–402. 

Kuldell, N.H., and Buratowski, S. (1997). Genetic analysis of the large subunit 

of yeast transcription factor IIE reveals two regions with distinct functions. 

Mol. Cell. Biol. 17, 5288–5298. 

Laskowski, R.A., MacArthur, M.W., Moss, D.S., and Thornton, M. (1993). 

PROCHECK: a program to check the stereochemical quality of protein structures. 

J. Appl. Cryst. 26, 283–291. 

Lefèvre, S., Dumay-Odelot, H., El-Ayoubi, L., Budd, A., Legrand, P., Pinaud, 

N., Teichmann, M., and Fribourg, S. (2011). Structure-function analysis of 

hRPC62 provides insights into RNA polymerase III transcription initiation. 

Nat. Struct. Mol. Biol. 18, 352–358. 

Lindstrom, D.L., Squazzo, S.L., Muster, N., Burckin, T.A., Wachter, K.C., 

Emigh, C.A., McCleery, J.A., Yates, J.R., 3rd, and Hartzog, G.A. (2003). Dual 

roles for Spt5 in pre-mRNA processing and transcription elongation revealed 

by identification of Spt5-associated proteins. Mol. Cell. Biol. 23, 1368–1378. 

Liu, X., Bushnell, D.A., Wang, D., Calero, G., and Kornberg, R.D. (2010). 

Structure of an RNA polymerase II-TFIIB complex and the transcription initiation 

mechanism. Science 327, 206–209. 

Margittai, M., Widengren, J., Schweinberger, E., Schröder, G.F., Felekyan, S., 

Haustein, E., König, M., Fasshauer, D., Grubmüller, H., Jahn, R., and Seidel, 

C.A. (2003). Single-molecule fluorescence resonance energy transfer reveals 

a dynamic equilibrium between closed and open conformations of syntaxin 

1. Proc. Natl. Acad. Sci. USA 100, 15516–15521. 

Martinez-Rucobo, F.W., Sainsbury, S., Cheung, A.C., and Cramer, P. (2011). 

Architecture of the RNA polymerase-Spt4/5 complex and basis of universal 

transcription processivity. EMBO J. 30, 1302–1310. 

Meinhart, A., Blobel, J., and Cramer, P. (2003). An extended winged helix 

domain in general transcription factor E/IIE alpha. J. Biol. Chem. 278, 

48267–48274. 

Mekler, V., Kortkhonjia, E., Mukhopadhyay, J., Knight, J., Revyakin, A., 

Kapanidis, A.N., Niu, W., Ebright, Y.W., Levy, R., and Ebright, R.H. (2002). 

Structural organization of bacterial RNA polymerase holoenzyme and the 

RNA polymerase-promoter open complex. Cell 108, 599–614. 

Mooney, R.A., Davis, S.E., Peters, J.M., Rowland, J.L., Ansari, A.Z., and 

Landick, R. (2009a). Regulator trafficking on bacterial transcription units 

in vivo. Mol. Cell 33, 97–108. 

Mooney, R.A., Schweimer, K., Rösch, P., Gottesman, M., and Landick, R. 

(2009b). Two structurally independent domains of E. coli NusG create regulatory 

plasticity via distinct interactions with RNA polymerase and regulators. 

J. Mol. Biol. 391, 341–358. 

Muschielok, A., Andrecka, J., Jawhari, A., Brückner, F., Cramer, P., and 

Michaelis, J. (2008). A nano-positioning system for macromolecular structural 

analysis. Nat. Methods 5, 965–971. 

Naji, S., Grünberg, S., and Thomm, M. (2007). The RPB7 orthologue E 0 is 

required for transcriptional activity of a reconstituted archaeal core enzyme 

at low temperatures and stimulates open complex formation. J. Biol. Chem. 

282, 11047–11057. 

Okuda, M., Tanaka, A., Arai, Y., Satoh, M., Okamura, H., Nagadoi, A., 

Hanaoka, F., Ohkuma, Y., and Nishimura, Y. (2004). A novel zinc finger 

structure in the large subunit of human general transcription factor TFIIE. 

J. Biol. Chem. 279, 51395–51403. 

Ouhammouch, M., Werner, F., Weinzierl, R.O., and Geiduschek, E.P. (2004). A 

fully recombinant system for activator-dependent archaeal transcription. 

J. Biol. Chem. 279, 51719–51721. 

Parvin, J.D., and Sharp, P.A. (1993). DNA topology and a minimal set of basal 

factors for transcription by RNA polymerase II. Cell 73, 533–540. 

Qureshi, S.A., Bell, S.D., and Jackson, S.P. (1997). Factor requirements for 

transcription in the Archaeon Sulfolobus shibatae. EMBO J. 16, 2927–2936. 

Rasnik, I., Myong, S., Cheng, W., Lohman, T.M., and Ha, T. (2004). 

DNA-binding orientation and domain conformation of the E. coli rep helicase 

monomer bound to a partial duplex junction: single-molecule studies of 

fluorescently labeled enzymes. J. Mol. Biol. 336, 395–408. 




Runner, V.M., Podolny, V., and Buratowski, S. (2008). The Rpb4 subunit of 

RNA polymerase II contributes to cotranscriptional recruitment of 3 0 processing 

factors. Mol. Cell. Biol. 28, 1883–1891. 

Sali, A., and Blundell, T.L. (1993). Comparative protein modelling by satisfaction 

of spatial restraints. J. Mol. Biol. 234, 779–815. 

Schneider, D.A., French, S.L., Osheim, Y.N., Bailey, A.O., Vu, L., Dodd, J., 

Yates, J.R., Beyer, A.L., and Nomura, M. (2006). RNA polymerase II elongation 

factors Spt4p and Spt5p play roles in transcription elongation by RNA polymerase 

I and rRNA processing. Proc. Natl. Acad. Sci. USA 103, 12707–12712. 

Sevostyanova, A., and Artsimovitch, I. (2010). Functional analysis of Thermus 

thermophilus transcription factor NusG. Nucleic Acids Res. 38, 7432–7445. 

Sevostyanova, A., Svetlov, V., Vassylyev, D.G., and Artsimovitch, I. (2008). 

The elongation factor RfaH and the initiation factor sigma bind to the same 

site on the transcription elongation complex. Proc. Natl. Acad. Sci. USA 105, 

865–870. 

Todone, F., Brick, P., Werner, F., Weinzierl, R.O., and Onesti, S. (2001). 

Structure of an archaeal homolog of the eukaryotic RNA polymerase II 

RPB4/RPB7 complex. Mol. Cell 8, 1137–1143. 

Wang, J., Cieplak, P., and Kollman, P.A. (2000). How well does a restrained 

electrostatic potential (RESP) model perform in calculating conformational 

energies of organic and biological molecules? J. Comput. Chem. 21, 1049– 

1074. 

Werner, F., and Grohmann, D. (2011). Evolution of multisubunit RNA polymerases 

in the three domains of life. Nat. Rev. Microbiol. 9, 85–98. 

Werner, F., and Weinzierl, R.O. (2002). A recombinant RNA polymerase II-like 

enzyme capable of promoter-specific transcription. Mol. Cell 10, 635–646. 

Werner, F., and Weinzierl, R.O. (2005). Direct modulation of RNA polymerase 

core functions by basal transcription factors. Mol. Cell. Biol. 25, 8344–8355. 


REVIEWS 

Evolution of multisubunit RNA 

polymerases in the three domains of life 

Finn Werner and Dina Grohmann 

Abstract | RNA polymerases (RNAPs) carry out transcription in all living organisms. All 

multisubunit RNAPs are derived from a common ancestor, a fact that becomes apparent 

from their amino acid sequence, subunit composition, structure, function and molecular 

mechanisms. Despite the similarity of these complexes, the organisms that depend on them 

are extremely diverse, ranging from microorganisms to humans. Recent findings about the 

molecular and functional architecture of RNAPs has given us intriguing insights into their 

evolution and how their activities are harnessed by homologous and analogous basal factors 

during the transcription cycle. We provide an overview of the evolutionary conservation of 

and differences between the multisubunit polymerases in the three domains of life, and 

introduce the ‘elongation first’ hypothesis for the evolution of transcriptional regulation. 

Last universal common 

ancestor 

The last common ancestor of 

all three contemporary 

domains of life. 

Transcription factor 

A protein that transiently 

interacts with RNA polymerase, 

modulates its protein‐, DNAand 

RNA-binding or catalytic 

properties, and thereby 

regulates transcription. 

Convergent evolution 

The acquisition of the same 

trait in unrelated lineages. 

Homologous 

Pertaining to genes: derived 

from the same ancestor. 

RNA Polymerase Laboratory, 

Institute for Structural and 

Molecular Biology, 

Division of Biosciences, 

University College London, 

Darwin Building, Gower Street, 

London WC1E 6BT, UK. 

Correspondence to F.W. 

e-mail: 

werner@biochem.ucl.ac.uk 

doi:10.1038/nrmicro2507 

RNA plays key roles in the expression of genes in all 

living organisms. As RNA is omnipresent, it is not 

surprising that the enzymes that synthesize it — RNA 

polymerases (RNAPs) — are highly conserved in evolution 

1 . Even though the polymerization of RNA has probably 

been ‘invented’ several times during evolution, as 

judged by the five structurally discrete and evolutionarily 

unrelated folds of RNAP active sites, all multisubunit 

RNAPs responsible for the transcription of cellular 

genomes have a common structural framework and 

operate by closely related molecular mechanisms 2 , suggesting 

that the last universal common ancestor (LUCA) of 

the Archaea, Bacteria and Eukarya had an RNAP very 

similar to the simplest form of contemporary RNAPs 

found in the Bacteria. Whereas the Archaea and the 

Bacteria use a single type of RNAP to transcribe their 

entire gene repertoire, the Eukarya use several classes 

of RNAPs that are specialized to transcribe distinct 

and non-overlapping subsets of genes 3 . A plethora of 

transcription factors interact with their cognate RNAP to 

modulate its activities during the three distinct phases 

of the transcription cycle: initiation, elongation and termination 

(FIG. 1). The structure and function of some 

of these factors are conserved across the three domains, 

whereas some non-homologous factors show an intriguing 

level of structural and functional similarity, suggesting 

that convergent evolution has led to alternative means 

of facilitating the same process. Some homologous proteins 

that are permanently incorporated into RNAPs in 

one system are reversibly incorporated in another RNAP. 

Thus, the boundary between core RNAP subunits and 

associated transcription factors is diffuse. Therefore, an 

understanding of the evolution of transcription machineries 

requires an appreciation of both RNAP subunits and 

transcription factors, and for hypotheses to be rooted on 

our structural, functional and mechanistic knowledge of 

transcription. Our deep understanding of multisub unit 

RNAPs in all three domains of life allows us to explain 

the molecular mechanisms of RNAP and transcription 

in unprecedented detail. In this Review, we focus on the 

current understanding of the structure and function of 

RNAP in an evolutionary context, and propose the ‘elongation-first’ 

hypothesis, according to which the RNAP of 

the LUCA was regulated during the elongation phase 

of transcription rather than the initiation phase. 

Molecular anatomy of RNAP 

Multisubunit RNAPs differ fundamentally from the 

single-subunit ‘right-handed’ RNAPs encoded by bacteriophages, 

such as T7 or SP6, which directly recognize 

promoter sequences without any requirements for accessory 

and regulatory factors 4,5 . All multisubunit RNAPs 

resemble a crab claw, the ‘jaws’ of which interact with 

the downstream duplex DNA template 6–10 (FIG. 2). In the 

transcribing RNAP, the downstream DNA projects 

along the floor of the major DNA-binding channel until 

it encounters the active centre at the RNAP ‘wall’. The 

DNA–RNA hybrid rises up from the active site perpendicular 

to the downstream duplex DNA, and the strands 

are separated by the RNAP ‘lid’ (REF. 7).The DNA–RNA 

hybrid and the DNA immediately downstream of the 

RNAP active centre are both secured by the ‘RNAP clamp’. 

nature reviews | Microbiology Volume 9 | february 2011 | 85 

© 2011 Macmillan Publishers Limited. All rights reserved

REVIEWS 

Rpo4–Rpo7 

RNAP 

TBP 

TFB 

TFE 

TFE 

TATA box 

Abortive 

RNAs 

Initiation 

(Recruitment) 

Initiation 

(Closed complex) 

Initiation 

(Open complex) 

Initiation 

(Abortive) 

In the archaeal and eukaryotic enzymes, the transcript is 

guided away from the elongating RNAP by interactions 

with the RNAP ‘stalk’ (REF. 11), whereas in the bacterial 

system, the flap tip helix might fulfil a similar function. 

The pore, or secondary channel, located under the active 

site, allows substrates and cleavage factors to access the 

active site, and allows extrusion of the transcript during 

backtracking 12 . 

All RNAP subunits can be divided into three overlapping 

functional classes 1,2 : assembly platform sub units, 

which nucleate RNAP assembly; catalytic subunits, which 

form the catalytic core that harbours the active site, 

including the Mg 2+ -chelating residues, the bridge and 

trigger helices, the downstream DNA-binding and 

DNA–RNA hybrid-binding sites, the secondary NTP 

entry channel, and the loop and switch regions that are 

instrumental for the handling of the nucleic acid scaffold 

(including DNA strand separation); and auxiliary 

subunits, which include the RNAP stalk. The combination 

of assembly platform and catalytic subunits is the 

minimal configuration of active RNAPs. The auxiliary 

RNAP subunits are not strictly required for basic RNAP 

operations, including promoter-directed transcription, 

but add interaction sites with basal transcription factors 

and/or nucleic acids. 

TATA box 

DNA 

U 5–8 

Nascent RNA 

Spt4–Spt5 

TFS 

Full-length RNA 

Rpo4–Rpo7 

Elongation 

Termination 

Structural and functional complexity of RNAP 

All multisubunit RNAPs in the three domains of life 

are evolutionary related and contain homologues 

of the bacterial RNAP β-, β’-, α- and ω-subunits 

(FIG. 3a; TABLE 1). The eukaryotic RNAPII subunits are 

called RPB1–RPB12 (numbered from largest to smallest 

polypeptide, based on the subunits in Saccharomyces 

cerevisiae), whereas the archaeal RNAP subunits are 

named Rpo1–Rpo13, although an alternative letter 

designation is often used in the literature (TABLE 1). 

The two largest RNAP subunits — the β-subunit and 

the β’-subunit in the Bacteria, RPB1 and RPB2 in the 

Eukarya, and Rpo1 (also known as RpoA) and Rpo2 (also 

known as RpoB) in the Archaea — contain two doublepsi 

β-barrel motifs that form the active site and are derived 

from a common ancestor, although it is difficult to recognize 

the structural similarity beyond the double-psi 

β-barrels owing to large insertions and deletions. 

RNAP clamp 

A flexible RNA polymerase 

(RNAP) domain that is 

predicted to move over the 

DNA-binding channel during 

open-complex formation. 

Bridge and trigger helices 

Flexible RNA polymerase 

motifs that unfold and refold 

repeatedly during nucleotide 

addition and translocation. 

RNAP in the RNA world. According to the ‘RNA world’ 

Figure 1 | The transcription cycle Nature of the Reviews archaeal | Microbiology RNA 

polymerase. Multisubunit RNA polymerases (RNAPs) carry 

hypothesis, the primordial RNAP was a ribozyme. It has 

out transcription by repeatedly cycling through initiation, been suggested that the ancestor of the contemporary 

elongation and termination phases. RNAP activity is β-subunit and β’-subunit was a homodimeric RNAbinding 

dependent on, and modulated by, exogenous transcription 

protein without any catalytic activity 13 . After 

factors. In the Archaea, TATA box-binding protein (TBP), the emergence of protein synthesis, this homodimer 

transcription factor B (TFB) and TFE facilitate transcription could have acted as a chaperone by binding to and 

initiation, whereas TFS and Spt4–Spt5 regulate 

improving the fitness of a ribozyme RNAP. According to 

transcription elongation and Rpo4 and Rpo7 (also known this hypothesis, the homodimer subsequently evolved 

as RpoF and RpoE, respectively) increase the processivity 

into a heterodimer with useful chemical properties at 

of the RNAP. Some initiation factors (for example, TBP) 

its interface, and the active site was transferred from 

remain associated with the promoter ready for the 

recruitment of the next RNAP in the subsequent cycle, 

the RNA to the protein cofactor. Then, the functionally 

obsolete RNA component was lost and the pro 

whereas other factors (for example, TFE) remain associated 

with elongating RNAPs 120 . Transcription termination in the tein subunit complexity increased, resulting in the 

Archaea is facilitated by a poly(T) signal at the 3′ end of contemporary multisubunit RNAPs. However, this 

the template gene. 

hypothesis remains speculative without the existence 

86 | february 2011 | Volume 9 www.nature.com/reviews/micro 


REVIEWS 

Assembly 

platform 

Wall 

Stalk 

Catalytic 

centre 

Clamp 

Figure 2 | Overall architecture of RNA polymerase. This Nature simplified Reviews diagram | Microbiology 

of RNA 

polymerase shows important structural and functional features discussed in the main 

text, including the assembly platform, the active site, the DNA-binding channel, the jaws 

and the wall, clamp and stalk domains. The two active-site Mg 2+ ions are indicated as 

magenta spheres. The structural information was obtained from eukaryotic RNAPII 

Protein Data Bank entry 1Y1W. 

Double-psi β-barrel motif 

A structural module in the two 

largest RNA polymerase 

subunits that reveals the 

common origin of these 

subunits and constitutes the 

active site. 

Ribozyme 

An enzyme made exclusively of 

RNA. 

RNAi RNAP 

An RNA polymerase (RNAP) 

involved in RNA interference 

(RNAi); it produces 

double-stranded RNA from 

aberrant single-stranded RNA 

templates to trigger RNAi. 

Jaw 

Jaw 

Duplex DNA-binding 

channel 

of experimental data or a naturally occurring ribozyme 

RNAP 14 . During evolution, the template specificity 

of RNAP gradually changed from RNA to DNA; bacterial 

RNAP and eukaryotic RNAPII still can use RNA templates 

in special circumstances, such as the regulation 

of transcription by 6S RNA 15 and the replication of the 

hepatitis delta virus genome 16 , respectively. 

The minimal core of RNAP. The two largest RNAP subunits 

are each encoded by one gene in most bacteria and 

eukaryotes, whereas many of the homologous archaeal 

subunits are split into two genes that are transcribed as 

one large polycistronic operon 2 . The split sites do not 

affect the structure or function of the subunits, as has 

been demonstrated by introducing the archaeal split 

sites into the bacterial RNAP 17 and by fusing the archaeal 

subunits (F.W., unpublished observations). In some 

bacterial lineages, including the genera Helicobacter 

and Campylobacter, the two large RNAP subunits are 

fused, presumably as a result of a frameshift in the intercistronic 

region 18 ; furthermore, a synthetic fusion protein 

of the two large Escherichia coli RNAP subunits can 

be assembled to yield a catalytically active enzyme 19 . The 

evolutionary conservation of the large subunits and the 

presence of highly conserved blocks of sequence can be 

easily recognized in all multisubunit RNAPs 20 , a fact that 

was further refined and elaborated in an analysis of multiple 

sequence alignments using a large sequence data 

set across all domains of life 21,22 . The availability of highresolution 

structural information about RNAPs allows 

the conserved sequence blocks to be mapped onto the 

three-dimensional structure of RNAP; these conserved 

regions correspond to mechanistically important motifs 

of RNAP, thereby linking evolution of sequence with 

RNAP function 21,22 . Such analyses make it possible to 

predict a hypothetical minimal RNAP core, which in 

essence is composed of the two double-psi β-barrels. 

Interestingly, two hypothetical proteins encoded by 

Corynebacterium glutamicum (ORF Cgl1702) and the 

yeast ‘killer’ plasmid pGKL2 correspond to this design 23 , 

although no experimental data on the expression, structure 

or catalytic activity of these putative ultra-minimal 

RNAPs are available. Other examples of minimal RNAPs 

are RNAi RNAPs such as QDE1, which consist of two identical 

subunits with one active site in each subunit 24 . The 

active sites of QDE1 in Neurospora crassa have the same 

architecture as multisubunit DNA-dependent RNAPs, 

encompassing two double-psi β-barrels 25 . 

Despite a low sequence identity, the two main classes 

of multisubunit RNAP — from the Bacteria and from 

the Archaea and the Eukarya, respectively — share an 

extensive structural homology 8 (FIG. 3a–e). The majority 

of the strictly conserved residues cluster around the 

RNAP active site, form the pore and are involved in 

the handling of the template and non-template DNA 

strands and RNA strands, or form flexible motifs, including 

the RNAP clamp and the switch regions 23 . In addition to 

the well-characterized RNAP subunits that are conserved 

in the three domains of life, the archaeal and eukaryotic 

RNAPs share a complement of additional subunits 1 

(TABLE 1). The archaea–eukaryote-specific subunits 

interact with many of the universally conserved RNAP 

subunits (FIG. 3c–e) and are not clustered at one particular 

site of the enzyme. The following section discusses how 

these subunits contribute to RNAP function. 

Domain-specific RNAP subunits 

A subset of RNAP subunits emerged following the split 

of the bacterial and archaeal–eukaryotic branches of 

the universal tree of life. These subunits are present in 

archaeal and/or eukaryotic RNAPs but have no homologues 

in bacteria. Although most of them are essential 

for cell viability in yeast, they are not strictly required for 

RNA polymerization. They make important interactions 

with basal transcription factors and the DNA–RNA scaffold 

of transcription complexes. Some of these subunits 

facilitate interactions between distinct RNAP subunits 

and between RNAP and basal transcription factors. 

Others make important contacts with the DNA template 

or the transcript RNA, resulting in a modulation 

of RNAP function during the transcription cycle. 

The stalk. The most pronounced difference between 

the RNAPs of archaeal and eukaryotic cells and that 

of bacteria is the stalk (FIG. 2), which is located at the 

periphery of the archaeal and eukaryotic enzymes and 

comprises Rpo4–Rpo7 (also known as RpoF–RpoE) in 

archaea and RPB4–RBP7 in eukaryotes (FIG. 3b,d,e). The 

stalk is the most thoroughly characterised sub unit of the 

archaea–eukaryote-specific RNAP subunits. The two 

subunits form a stable complex that reversibly associates 

with RNAPII in S. cerevisiae 26 , but is stably (nonreversibly) 

incorporated into the archaeal RNAP 27 . The 



REVIEWS 

a Bacteria 

RNAP clamp 

β′ 

β 

α I 

α II 

ω 

b Archaea 

Rpo1 

Rpo2 

Rpo3 

Rpo11 

Rpo6 

Downstream 

DNA-binding site 

RNAP 

clamp 

Downstream 

DNA-binding site 

Rpo4 

Rpo5 

Rpo7 

Rpo8 

Rpo10 

Rpo12 

Rpo13 

Catalytic centre 

Catalytic centre 

c Bacteria d Archaea 

Rpo4–Rpo7 e Eukaryotes 

RPB4–RPB7 

Rpo13 

Rpo5 

RPB5 

Rpo10 and Rpo12 

RPB10 and RPB12 

RPB9 

Figure 3 | Important structural and functional features of multisubunit RNA polymerases. a,b | The structures of 

the bacterial RNA polymerase (RNAP) (part a; based on the Thermus aquaticus Protein Data Bank Nature entry Reviews 1I6V) | and Microbiology the 

archaeal RNAP (part b; based on the Sulfolobus shibatae Protein Data Bank entry 2WAQ) reveal an evolutionarily 

conserved architecture. Homologous RNAP subunits are colour coded according to the key. The Mg 2+ ion is highlighted in 

magenta, and the bridge helix is in orange. c–e | The locations of RNAP subunits that are absent in bacteria (part c) but 

present in archaea (part d) and eukaryotes (part e) are shown. Universally conserved RNAP subunits are blue, and those 

that are unique to the archaeal–eukaryotic lineage are magenta. 

Open-complex formation 

The structural transition of 

RNA polymerase concomitant 

with the melting of DNA 

strands during initiation. 

stalk has multiple roles during the transcription cycle 14,28 : 

it promotes open-complex formation during transcription 

initiation 29 and facilitates the action of the basal transcription 

factor TFE 30,31 . During elongation, Rpo4–Rpo7 

and RPB4–RPB7 increase the processivity of RNAP, and 

Rpo4–Rpo7 also augment transcription termination 11 . 

Assembly platform. The assembly of RNAPs requires an 

assembly platform 32 . In eukaryotes, RPB10 and RPB12 

(homologous to archaeal Rpo10 (also known as RpoN) 

and Rpo12 (also known as RpoP)) form a stable complex 

with RPB3–RPB11 (homologous to archaeal Rpo3– 

Rpo11 (also known as RpoD–RpoL)), and this complex 

is homologous to the bacterial α-subunit homodimer 33 

(FIG. 3a,b; TABLE 1). In contrast to the bacterial RNAP, 

which only requires the α-subunit homodimer (FIG. 3a), 

all four archaeal subunits are necessary for the efficient 

assembly and stability of archaeal RNAP 32 (FIG. 3b). 

Rpo10 and Rpo12, and RPB10 and RPB12, fill concave 

depressions in the second-largest RNAP sub unit (Rpo2 

and RPB2, respectively) and thereby act as structural 

adaptors between Rpo2 and Rpo3 or RPB2 and RPB3, 

respectively (FIG. 3d,e); this explains, at least in part, their 

role during RNAP assembly and stability. RPB10 and 

RPB12 are found in the three main classes of eukaryotic 

RNAPs, but their interaction partners differ in each class; 

the assembly platform subunits of RNAPI and RNAPIII 

(AC19–AC40 in both) are distinct from RPB3–RPB11, 

and the second-largest catalytic subunits of RNAPI and 

RNAPIII (A135 and C128, respectively) are distinct 

from RPB2, suggesting that RPB10 and RPB12 have 

additional functions beyond RNAP assembly. It is 



REVIEWS 

Table 1 | Conserved RNA polymerase (RNAP) subunits and transcription factors* 

Bacteria Archaea Eukaryotes 

RNAP subunits 

RNAPII RNAPIII RNAPI Plant RNAPIV ‡ Plant RNAPV ‡ 

β-subunit Rpo1 (RpoA) RPB1 C160 A190 NRPD1 NRPE1 

β-subunit Rpo2 (RpoB) RPB2 C128 A135 NRPD/E2 NRPD/E2 

a-subunit Rpo3 (RpoD) RPB3 AC40 AC40 RPB3 [1] RPB3 [1] 

a-subunit Rpo11 (RpoL) RPB11 AC19 AC19 RPB11 RPB11 

w-subunit Rpo6 (RpoK) RPB6 RPB6 RPB6 RPB6 [1] RPB6 [1] 

Rpo5 (RpoH) RPB5 RPB5 RPB5 RPB5 [3] NRPE5 

Rpo8 § (RpoG) RPB8 RPB8 RPB8 RPB8 [1] RPB8 [1] 

Rpo10 (RpoN) RPB10 RPB10 RPB10 RPB10 RPB10 

Rpo12 (RpoP) RPB12 RPB12 RPB12 RPB12 RPB12 

Rpo4 (RpoF) RPB4 C17 A14 NRPD/E4 NRPD/E4 

Rpo7 (RpoE) RPB7 C25 A43 NRPD7 [1] NRPE7 

Rpo13 § 

Transcription factors 

RPB9 C11 A12 NRPD9b RPB9 

TFIIFα (RAP74) C53 (C4) A49 

TFIIFβ (RAP30) C37(C5) A34.5 

TFEα TFIIEα C82 

TFEβ/C34 § TFIIEβ C34 

C31 

TBP TBP TBP TBP 

TFB TFIIB BRF1 

TFIIA 

TFIIH 

TFS TFIIS TFIIS 

Spt4 SPT4 SPT4 

NusG Spt5 SPT5 SPT5 

NusA 

Rho 

σ-factors 

NusA 

TBP, TATA box-binding protein. *Alternative names that are common in the literature are shown in brackets. ‡ The numbers in square 

brackets indicate the number of orthologues of RNAPIV and RNAPV subunits. § Found in some but not all archaeal species. 

TATA box-binding protein 

One of the two minimal 

transcription factors required 

for initiation by archaeal RNA 

polymerase (RNAP) and 

eukaryotic RNAPII. It binds to a 

sequence recognition motif in 

promoters that is called the 

TATA element or TATA box. 

TFIIB 

The second of the two minimal 

transcription factors required 

for initiation by archaeal RNA 

polymerase (RNAP) and 

eukaryotic RNAPII. 

possible that RPB10 and RPB12 are incorporated into 

multiple classes of RNAPs owing to functional and/or 

physical interactions with basal transcription factors 

that facilitate transcription of all eukaryotic and archaeal 

RNAPs, such as TATA box-binding protein (TBP). Indeed, 

RPB10 and RPB12 are localized in proximity to TBP in a 

structural model of the DNA–TBP–TFIIB–RNAPII transcription 

initiation complex 34 . In addition, the archaeal 

homologue Rpo12 plays a part during transcription initiation 

by promoting DNA melting and stabilizing the 

open complex 35 . 

DNA melting. Like RPB10 and RPB12, RPB5 is present 

in all eukaryotic RNAPs, and the archaeal homologue 

Rpo5 (also known as RpoH) plays a part in DNA melting 

and early transcription 36 . RPB5 consists of two discrete 

domains; the carboxy‐terminal domain, which corresponds 

to the full-length Rpo5, makes intricate contacts 

with the C terminus of the largest RNAP subunit 

(RPB1 and Rpo1; FIG. 3b). Similarly to RPB5 and RPB1 in 

eukaryotes, Rpo5 and a fragment of Rpo1 form the lower 

jaw domain of RNAP in archaea, which is more extended 

than its bacterial counterpart 8,9 (FIG. 3). The jaw interacts 

with the downstream duplex DNA and undergoes substantial 

conformational changes between the initiation 

and elongation phases of transcription 36,37 . 

RPB8 and Rpo8 are located at the underside of the 

RNAP between the assembly platform and the pore. 

These homologues contain a nucleic acid-binding 

OB‐fold 38 and could potentially interact with the 3′ end 



REVIEWS 

s-factor 

A bacterial transcription 

initiation factor that binds 

specific sequences in the 

promoter. Each σ-factor 

regulates the transcription of 

a specific set of genes. 

–10 and –35 elements 

Key sequence motifs of 

bacterial promoters that are 

recognized by the transcription 

factor σ 70 . Promoter strength 

is in part regulated by the 

strength of the interaction 

between σ 70 and these 

elements. 

of the nascent transcript in backtracked elongation complexes, 

as this end is extruded through the pore 39 . Yeast 

Rpb8 is essential, but its precise function during transcription 

is not clear 40 . In archaea, Rpo8 is present only 

in the Crenarchaeota, one of two main archaeal phyla, 

reflecting the closer kinship of the Crenarchaeota to 

the Eukarya compared with the relationship between 

the Eukarya and the other main archaeal phylum, the 

Euryarchaeota 41 (TABLE 1). 

Domain-specific subunits. Some subunits are specific 

for a single domain. RPB9 is probably the only subunit 

found exclusively in eukaryotic RNAPs. RPB9 influences 

the interaction of RNAP with the basal factor TFIIF, and 

therefore transcription start site selection and transcription 

fidelity 42,43 . Rpo13 is the only archaea-specific subunit; 

its function is unknown, and it is only present in a 

subset of archaeal genomes 9 . 

Mechanisms of transcription initiation 

Promoter-directed transcription requires sequencespecific 

recruitment of RNAP to the promoter, initiation 

of RNA polymerization in a primer-independent manner 

and efficient escape from the promoter (FIG. 1), all of 

which are stimulated by evolutionarily unrelated basal 

initiation factors in the Bacteria and in the Eukarya and 

the Archaea. However, as the molecular mechanisms of 

initiation are the same in all three domains, the mechanisms 

of action of the non-homologous factors are 

closely related. They have similar structures, utilize the 

same RNAP-binding sites and carry out nearly identical 

functions. This is an important insight into the evolution 

of transcription machineries and provides an excellent 

example of convergent evolution. 

Bacteria. The bacterial core RNAP associates with a 

σ-factor to form ‘holo’-RNAP, which is directly recruited 

to the promoter in a sequence-specific manner (FIG. 4a,b). 

The canonical σ-factors consist of four conserved regions 

that contribute in distinct ways to transcription initiation 

44 . The σ-factors determine promoter specificity by 

binding directly to the –10 element in the promoter 

through region 2 and to the –35 element through region 4; 

they also interact with RNAP in a complex way to recruit 

the polymerase (predominantly through region 2 and 

region 4, but also through other regions), facilitate DNA 

melting and template strand loading during the closedto-open 

complex transition (through region 2), stabilize 

the binding of the initiating nucleotide substrate 

and affect abortive initiation (through region 3.2), and 

modulate promoter escape (through multiple regions) 45 . 

In addition to the ‘housekeeping’ factor σ 70 (also 

known as RpoD), numerous alternative σ-factors (up to 

60 in Streptomyces coelicolor), including σ S (also known 

as RpoS), σ 54 and σ 32 , direct RNAP to subsets of genes 

that are induced, for example, in stationary phase, under 

phage shock conditions and under heat shock conditions, 

respectively 46 . In Bacillus subtilis, one alternative 

σ-factor consists of two subunits, YvrHa (also known as 

RsoA) and YvrI (also known as SigO) 47,48 . YvrI binds to 

the RNAP flap motif and mediates promoter recognition 

of the –35 element, similarly to region 4 in σ 70 , whereas 

YvfHa interacts with the –10 element of the promoter 

and the coiled coil of the RNAP clamp, thereby stabilizing 

the open complex 47 , similarly to region 2 in σ 70 

(FIG. 4a,b). 

Whereas transcription initiation facilitated by the 

σ 70 holo-RNAP is a spontaneous, energy-independent 

process, transcription initiation by the σ 54 holo-RNAP 

requires an activator from the enhancer-binding protein 

family (which is part of the AAA+ family) and hydrolysis 

of ATP 49 . Transcription initiation complexes formed by 

σ 54 holo-RNAP are trapped in a conformation that is not 

conducive to DNA strand separation, and ATP hydrolysis 

mediated by the RNAP-bound enhancer-binding 

proteins induces a conformational change in the transcription 

initiation complex that overcomes this restriction 

and results in efficient induction of genes under the 

control of σ 54 promoters 49 . Conceptually, this mechanism 

is reminiscent of the RNAPII system, in which transcription 

initiation from many promoters is strongly enhanced 

by, if not dependent on, the hydrolysis of ATP by the 

basal transcription factor TFIIH, which facilitates DNA 

strand melting 50 . However, the underlying mechanisms 

are distinct and not conserved in evolution 49,50 . 

Archaea and eukaryotes. The archaeal RNAP and 

eukaryotic RNAPII have identical minimal requirements 

for homologous basal transcription initiation factors, 

and these factors operate via the same mechanisms 

(TABLE 1). Neither RNAP can be recruited to the promoter 

without the aid of additional factors; instead, the RNAP 

recognizes a complex of basal factors, including TBP and 

TFIIB, pre-assembled on the promoter 2 . Unlike bacterial 

σ-factors 51 , TBP and TFIIB can be recruited to the 

promoter independently of RNAP. TBP consists of a 

highly conserved DNA-binding core domain and a less 

conserved eukaryote-specific amino‐terminal domain 52 . 

The TBP core domain has a bipartite symmetrical structure 

encoded by a sequence repeat, which suggests that it 

evolved by a gene duplication event. However, no eukaryotic 

or archaeal protein with a single TBP repeat has been 

identified. Only one other protein, a bacterial RNAse H 

subtype (RNase HIII), contains a domain that is homologous 

to a single TBP repeat, and this domain probably 

facilitates interactions with its DNA–RNA substrate 52 . 

TFIIB consists of two discrete domains connected 

by a flexible linker. The C‐terminal core domain of 

TFIIB binds to the TATA box–TBP complex and makes 

contacts with the DNA both upstream and downstream 

of the TATA box. From a structural perspective, 

the TBP–TFIIB–TATA box complex is reminiscent of the 

complex formed by region 4 of σ 70 and the –35 element 

of bacterial promoters, which forms independently of 

RNAP 53 . Although the structure of the DNA–TBP– 

TFIIB–RNAPII complex has not been solved, partial 

crystal structures combined with cross-linking and 

cleavage experiments have allowed a model of the initiation 

complex to be generated 34,54,55 . In this model, 

eukaryotic and archaeal RNAPs make very few direct 

contacts with the DNA in the closed complex, similar to 

bacterial RNAP. Instead, archaeal and eukaryotic RNAPs 



REVIEWS 

Bacteria 

a 

σ4 

σ3.2 

σ3 

σ2 

b 

σ3 

σ2 

β-flap 

σ3.2 

RNAP clamp 

coiled-coil 

motif 

RNAP clamp 

coiled-coil motif 

Eukaryotes 

c 

TFIIB 

Zn-ribbon 

RNAP dock 

TFIIB 

B-reader 

RNAP clamp 

coiled-coil 

motif 

d 

TFIIB 

B-reader 

TFIIB core N-terminal 

cyclin fold 

TFIIB 

B-linker 

RNAP clamp 

coiled-coil 

motif 

TFIIB 

B-linker 

TFIIB core N-terminal 

cyclin fold 

Figure 4 | Transcription initiation — holo-RNA polymerase structures from bacteria and eukaryotes. a,b | The 

bacterial core RNA polymerase (RNAP) forms a stable holoenzyme complex with the initiation Nature factor Reviews σ 70 (also | known Microbiology as 

RpoD) (based on the Thermus thermophilus Protein Data Bank entry 2A6E). The holoenzyme structure is shown in top 

(part a) and front (part b) views. The density of the bacterial β’-subunit downstream of region 2 is lineage specific and not 

representative of all bacterial RNAPs. c,d | Eukaryotic RNAPII forms a stable complex with basal transcription factor IIB 

(TFIIB) (based on the Saccharomyces cerevisiae Protein Data Bank entry 3K1F). The RNAPII–TFIIB structure is shown in top 

(part c) and front (part d) views. Important features in RNAP (the β-flap, RNAP dock and clamp coiled-coil motifs), TFIIB 

(the Zn-ribbon, the B‐reader, the B-linker and the core amino‐terminal cyclin repeats) and σ 70 (regions 2, 3, 3.2 and 4) are 

indicated. All RNAP subunits are colour coded as in FIG. 1. 

Zn-ribbon domain 

A domain found in many 

transcription factors, including 

TFIIB, TFIIS and TFIIE in 

eukaryotes (and TFB, TFS and 

TFE, respectively, in archaea). 

The amino‐terminal Zn-ribbon 

domain of TFIIB and TFB 

interacts with the RNA 

polymerase (RNAP) dock 

domain and is important for 

efficient RNAP recruitment. 

are anchored to the promoter via multiple interactions 

with TFIIB. The N‐terminal Zn-ribbon domain of TFIIB 

interacts with the dock domain of RNAP (FIG. 4c), and the 

C‐terminal core domain is positioned across the DNAbinding 

channel (FIG. 4c,d). The RNAP clamp coiled-coil 

motif, which is conserved across the three domains of 

life, is an important binding site for region 2 of σ 70 and 

for TFIIB 34,56 (FIG. 4). The highly flexible linker region 

that connects the TFIIB domains (consisting of the 

B-reader helix and the B‐linker) penetrates deep into 

the active centre of RNAP (FIG. 4c,d). The linker can be 

cross-linked to the template DNA strand 57 . The B‐reader 

is displaced by the growing RNA transcript (longer than 

5 nucleotides), whereas the B‐linker is displaced by the 

rewinding of upstream DNA during TFIIB release and 

promoter escape 58 . The position of the TFIIB Zn-ribbon 

and its interaction with RNAP, as well as the TFIIB core 

domain and the B‐reader, are reminiscent of region 4, 

region 3 and region 3.2 of σ 70 , respectively 34,59 (FIG. 4a,c). 

Bacterial, archaeal and eukaryotic RNAPs form 

a ‘composite’ active site that is complemented by the 

cognate initiation factor, resulting in a higher affinity 



REVIEWS 

Paralogous 

Pertaining to genes: separated 

by a gene duplication event. 

for the initiating nucleotide substrate 60 and stimulating 

catalysis 31 . Thus, despite a lack of similarity between 

these transcription factors at the sequence level, the 

interaction networks between RNAP and σ-factors in 

bacteria and between RNAP and TFIIB in archaea and 

eukaryotes are strikingly similar. 

All RNAPs enter a non-productive phase of transcription 

following recruitment to the promoter called 

abortive cycling, during which the downstream DNA 

template is repeatedly reeled in 61 , and small transcripts 

of 3 to 9 nucleotides are synthesized and released without 

the RNAP disengaging from the promoter 62 . The exact 

mechanical nature of abortive initiation is still unclear, 

but it is likely to be caused by several factors. First, the 

linker regions of σ 70 (region 3.2) and TFIIB clash sterically 

with the growing RNA chain 34,63 . Second, the interactions 

within the initiation complex between RNAP and 

σ-factors or TFIIB need to be disrupted during promoter 

escape, and this presents a substantial energy barrier. In 

all initiation complexes, multiple low-affinity interactions 

between initiation factors and RNAP combine to form a 

stable complex 34,45,59 . These interactions guarantee efficient 

recruitment of RNAP to the promoter and simultaneously 

enable dissociation by small conformational 

changes of the complex, during which the individual 

contacts are broken in a stepwise manner 45 . 

In archaea and eukaryotes, additional basal factors 

contribute to transcription initiation, including TFE, 

TFIIE, TFIIF and TFIIH (TABLE 1). The interaction 

of TFIIE with the RNAP clamp is not strictly required for 

initiation but stimulates DNA melting and stabilizes the 

open complex 29,31,64 . The α-subunit of TFIIE shows very 

weak sequence homology to region 2 and region 4 of σ 70 , 

and the β-subunit shows a weak homology to region 3 

(REFS 65,66). The eukaryote-specific factor TFIIF also 

displays very weak sequence similarities to σ 70 region 4 

(REF. 67). However, it is unlikely that these proteins are 

bona fide homologues, as there is no apparent structural 

homology or shared RNAP-binding sites between TFIIE, 

TFIIF and the σ-factors. TFIIF interacts with RNAPII 

during transcription initiation and elongation 68–70 . 

Paralogous TFIIF-like factors are involved in transcription 

in all three major eukaryotic RNAP systems: the 

general transcription factor TFIIF (the α-subunit and the 

β-subunit; also known as RAP74 and RAP30, respectively) 

in RNAPII, A49–A34.5 in RNAPI and C53–C37 

(also known as C4–C5) in RNAPIII 71 (TABLE 1). 

Modulation of elongation and termination 

Following promoter escape, RNAP enters the elongation 

phase of transcription, which is modulated by DNA 

and RNA sequences (including secondary structures) and 

RNAP subunits, and regulated by general and genespecific 

transcription factors 72,73 . Transcription elongation 

is intrinsically discontinuous and is interrupted by 

frequent pausing, stalling and arrest. The two important 

parameters that vary during elongation are the translocation 

rate (the average number of nucleotides polymerized 

per second) and the processivity (the number of nucleotides 

polymerized per initiation event). Transcription 

pausing can be instrumental for the regulation and 

timing of gene expression but can also decrease RNA 

synthesis, as pausing reduces processivity 74 . The catalytic 

cycle and the resulting mechanism of RNAP translocation 

along the DNA template involves alternative structures 

of the bridge and trigger helices in the active site 75 . 

In the ternary elongation complex (TEC), composed 

of RNAP–DNA–RNA (FIG. 5), RNAP interacts with the 

downstream duplex DNA template, the DNA–RNA 

hybrid and the nascent RNA transcript 76 . The interaction 

of the archaea–eukaryote-specific RNAP stalk with 

the RNA is dynamic, and the path of the transcript could 

not be resolved by X‐ray crystallography. Whereas the 

regions that interact with the duplex DNA and DNA– 

RNA hybrid are highly conserved in all RNAPs, the stalk 

(consisting of RNAP subunits RPB4–RPB7) (FIG. 5c,d) is 

not present in bacterial RNAP (FIG. 5a,b). In the bacterial 

TEC, the nascent transcript is secured by the RNAP flap 

domain (FIG. 5a), which also serves as the primary interaction 

site with the elongation termination factor NusA 77 

(TABLE 1). Both bacterial and archaeal NusA interact with 

RNA, but it is unclear whether archaeal NusA modulates 

transcription or indeed interacts with the RNAP, as it 

lacks the N‐terminal domain that is present in bacterial 

NusA and that facilitates recruitment to the RNAP 2,78,79 . 

Two lines of evidence demonstrate that the stalk is 

involved in elongation for eukaryotic and archaeal 

RNAP: recombinant archaeal RNAP lacking Rpo4–Rpo7 

has a substantially lowered processivity in vitro 11 , and 

S. cerevisiae RNAPII lacking RPB4 is depleted in the 

3′ regions of long ORFs in vivo 80 . Rpo4–Rpo7 modulates 

both transcription elongation and termination via 

two mechanisms: by interacting with the nascent transcript, 

and by an allosteric mechanism that probably 

involves a repositioning of the RNAP clamp 81 (FIG. 5c,d). 

As Rpo4–Rpo7‐like complexes increase the processivity 

of transcription, they may enable RNAPs to transcribe 

longer genes 14,28 . This probably applies only to eukaryotes, 

as eukaryotic genes are up to two orders of magnitude 

longer than bacterial and archaeal genes. However, 

it does not argue against the hypothesis that stalk-like 

complexes in ancestral eukaryotic RNAPs enabled the 

expansion of ORF size. 

Rescue of arrested elongation complexes 

Paused RNAPs have a tendency to move in a retrograde 

direction along the DNA template in vivo and 

in vitro. During this ‘backtracking’, the RNA 3′ end is 

extruded from the RNAP through the pore and RNA 

polymerization cannot occur. RNAPs can overcome 

this impediment by cleaving the transcript internally, 

releasing short (3 to 18 nucleotides) RNA 3′-cleavage 

products and creating a new 3′-OH on the RNA that 

is aligned in the active site and conducive to catalysis 

82,83 . This endonucleolytic cleavage activity of RNAP 

is prominent at elevated pH and stimulated by transcript 

cleavage factors under physiological conditions. 

In eukaryotes and archaea, TFIIS and TFS, respectively, 

stimulate transcript cleavage 84,85 , and the structure of the 

yeast RNAPII–TFIIS complex provides insights into its 

molecular mechanism 86 . TFIIS is recruited to the TEC 

via interactions between TFIIS domain II and the RNAP 



REVIEWS 

Bacteria 

a 

RNAP flap 

RNAP clamp 

RNAP clamp 


b 

RNAP clamp 

coiled-coil 

motif 

RNAP clamp 

Downstream 

DNA 

Eukaryotes 

c 

RNA 

RNAP clamp 

RNAP clamp 


d 

RNAP clamp 


RNAP clamp 

Downstream 

DNA 

Figure 5 | The architecture of the transcription elongation complex. The ternary elongation complexes (TECs) of 

Thermus thermophilus RNA polymerase (RNAP) (parts a,b; Protein Data Bank entry 2O5I) and Nature Saccharomyces Reviews | cerevisiae Microbiology 

RNAPII (parts c,d; Protein Data Bank entry 1Y1W) are shown in top (parts a,c) and front (parts b,d) views. RPB4 is the 

magenta ribbon, and RPB7 is the blue ribbon. The RNAP clamp domain coiled-coil motif is the red ribbon, and the bridge 

helix is the orange ribbon. Template and non-template DNA strands are green, and the RNA transcript is red. The RNAP flap 

secures the nascent transcript in the bacterial TEC. The arrows indicates an inwards closing movement of the RNAP clamp 

over the DNA-binding channel in response to an association of RNAP with RPB4–RPB7 in eukaryotes, and possibly with the 

elongation factor NusG in bacteria, and with Spt4–Spt5 and SPT4–SPT5 in archaea and eukaryotes, respectively. 

jaw, while domain III is inserted into the active site 

through the pore (FIG. 6). Domain II forms a Zn-ribbon 

domain with a thin protruding β-hairpin, which complements 

the active site without either denying access of 

NTP substrates to the active site or blocking the extrusion 

of the transcript through the pore. Two invariant 

acidic residues on the tip of the hairpin are essential for 

the stimulatory effect of TFIIS on transcript cleavage 87 

(FIG. 6). They are brought into close proximity to the 

Mg 2+ ions in the active site and probably alter the binding 

characteristics of the metal ions and/or modulate 

the structure and location of the RNA or DNA–RNA 

hybrid in the active site in a manner that stimulates 

endonucleolysis 86 (FIG. 6). The archaeal TFS resembles 

both TFIIS and RPB9 at the sequence level; it has the 

domain structure of RPB9, but carries out the function 

of TFIIS 84 . Like TFIIS and TFS, the bacterial GreB 

stimulates transcript cleavage of its cognate RNAP and 

thereby augments transcription elongation 85,88 . Although 

GreB is not evolutionarily related to TFIIS in sequence 

or structure, their interactions with RNAP and their 

molecular mechanisms of action are very similar. GreB 

is recruited to the bacterial TEC by interacting with the 

RNAP jaw domain, while its N‐terminal coiled-coil 

domain is inserted into the active site through the pore. 

As in TFIIS, two acidic residues positioned at the tip of 

the coiled-coil domain are brought into close proximity 

to the active site Mg 2+ ions and are crucial for GreB 

activity 89 . As most if not all genes contain frequent 

pause sites, TFIIS and GreB regulate RNAP activity by 

increasing its processivity and thereby increasing the 

overall transcription elongation rate. 



REVIEWS 

a 

TFIIS domain III 

TFIIS domain linker 

TFIIS domain II 

b 

Figure 6 | The transcript cleavage complex. The transcript cleavage factor TFIIS (red) forms a stable complex with RNA 

polymerase II (RNAPII) (based on the Saccharomyces cerevisiae Protein Data Bank entry 1Y1Y), Nature shown Reviews in bottom | Microbiology 

(part a) 

and front (part b) views. The TFIIS domain II interacts with the RNAPII jaw (resolved at lower resolution and shown as dots), 

while domain III penetrates deeply into the RNAPII active site through the pore. Two acidic residues (green) are located in 

close proximity to the Mg 2+ ion (magenta sphere). The RNAPII bridge helix and RNAP subunits RPB4 and RPB7 are orange, 

magenta and blue ribbons, respectively. 

The mechanisms of TFIIB, σ-factors, TFIIS and GreB 

bear some similarity, as all of these factors invade the 

catalytic centre of RNAP (either via the major DNAbinding 

channel or by the pore), complement the active 

site, alter the interactions with nucleic acids, Mg 2+ ions 

and NTP substrates, and thereby modulate the catalytic 

properties of RNAP. Notably, neither initiation factors 

(TFIIB and σ-factors) nor cleavage factors (TFIIS and 

GreB) are evolutionarily conserved between the Archaea 

and Eukarya and the Bacteria; instead, they have adapted 

a similar structure to interact with RNAP and execute 

the same function, by convergent evolution. 

Transcription elongation, NusG and Spt5 

Decreased processivity and transcription termination 

are widely used to regulate gene expression in bacteriophages, 

and in ribosomal (rrn) operons in bacteria 90 . The 

association of bacteriophage-encoded (such as phage λ 

anti-termination protein Q) and/or bacterial (such as 

NusA, NusB, NusE and NusG) elongation factors with 

RNAP converts the TEC into a termination-resistant 

anti-termination complex, which can transcribe genes 

beyond termination signals in the phage λ genome or 

downstream of the 5′ leader region of rrn operons 91–93 . 

Only NusG is universally conserved (FIG. 7; TABLE 1). 

NusG and its archaeal and eukaryotic homologues, 

Spt5 and SPT5, respectively, associate with their cognate 

RNAPs and enhance transcription elongation by 

stimulating processivity and possibly the elongation rate 

of RNAP. Archaeal Spt5 and bacterial NusG consist of 

an N‐terminal NusG (NGN) domain and a C‐terminal 

Kypridis–Ouzounis–Woese (KOW) domain (FIG. 7a). In 

contrast to the non-homologous basal factors that facilitate 

transcription initiation, the NGN domain structure, 

the NusG–RNAP interaction sites and the stimulatory 

activity on transcription elongation are conserved in 

archaeal Spt5 (REFS 94,95). Eukaryotic SPT5 is much 

larger owing to the presence of four to six copies of the 

KOW domain, and two heptad repeat motifs that regulate 

SPT5 through phosphorylation 96 . 

Archaeal and eukaryotic NGN domains form stable 

complexes with Spt4 and SPT4, respectively, a protein 

that is not conserved in bacteria (FIG. 7). The bacterial 

and archaeal NGN domains are sufficient for RNAP 

binding and stimulation of transcription elongation 94,95 . 

Archaeal Spt5 and bacterial NusG associate with RNAP 

via interactions between a hydrophobic cavity in the 

NGN domain (FIG. 7b,c) and the tip of the RNAP clamp 

coiled-coil domain, which protrudes from the RNAP 

surface 94,95 (FIG. 4). Interestingly, the clamp coiled-coil 

domain is also an important RNAP recruitment site 

during initiation in all RNAPs, through the interaction 

with region 4 of σ 70 and with TFIIB 34,56 . This overlapping 

binding site might have an important role during 

promoter escape or a stalled RNAP proximal to the promoter, 

where elongation factors (Spt5, SPT5, NusG or 



REVIEWS 

a b c d 

NGN 

RNAP-binding site 

KOW 

Spt4 and SPT4 

Bacteria 

E. coli NusG NGN 

Archaea 

M. jannaschii Spt4–Spt5 NGN 

Eukaryotes 

S. cerevisiae SPT4–SPT5 NGN 

Figure 7 | Universal evolutionary conservation of the elongation factor Spt4–Spt5 and NusG. a | The X‐ray 

structure of full-length Pyrococcus furiosus Spt4–Spt5, containing the carboxy‐terminal Kypridis–Ouzounis–Woese 

Nature Reviews | Microbiology 

(KOW) domain (K. S. Murakami, personal communication). b | The bacterial amino-terminal NusG (NGN) domain (based 

on the Escherichia coli Protein Data Bank entry 2K06). c | The archaeal Spt5 NGN domain bound to Spt4 (based on the 

Methanocaldococcus jannaschii Protein Data Bank entry 3LPE). Note that the Spt5 NGN domain is homologous to NusG, 

whereas Spt4 does not have a bacterial homologue. d | The eukaryotic SPT5 NGN domain bound to SPT4 (based on the 

Saccharomyces cerevisiae Protein Data Bank entry 2EXU). The C‐terminal residue of the NGN domain that connects to 

the KOW domain is indicated with a dashed circle. 

the paralogue RfaH) could displace initiation factors 

(TFIIB and region 4 of σ 70 ) and thereby release RNAP 

into the elongation phase of transcription 56,97,98 . Efficient 

binding of SPT5 to RNAPII requires contacts with the 

RNA transcript in the eukaryotic system 99,100 . 

Thus, non-conserved RNAP subunits (Rpo4–Rpo7) 

and universally conserved transcription factors (Spt5, 

SPT5 and NusG) modulate RNAP elongation, with both 

the conserved factors and non-conserved factors interacting 

with the RNAP clamp. These interactions could 

lead to subtle alterations in the RNAP clamp that have 

the potential to alter the catalytic properties of RNAP 

directly via an allosteric signal to the bridge and trigger 

helices 94,101–103 . In addition, a closure of the clamp, which 

forms one side of the DNA-binding channel, is likely to 

affect the interactions of RNAP with the downstream 

template DNA or the DNA–RNA hybrid and thereby 

alter the ‘traction’ of the RNAP in the TEC 14,28 ; the 

interaction of Rpo4–Rpo7 and eukaryotic Spt5 with 

the nascent transcript could stabilize the TEC and result 

in increased processivity 94 . 

The ‘elongation-first hypothesis’ 

Despite the high degree of homology between all 

RNAPs, the basal transcription factors that are required 

for transcription initiation in bacteria and in archaea and 

eukaryotes are not evolutionarily related. By contrast, the 

only RNAP-associated transcription factor that is universally 

conserved in evolution, the Spt5–SPT5–NusG 

family, controls the elongation phase of transcription. 

What is the significance of this observation, and what 

does it tell us about the regulation of the ancestral form of 

RNAPs in the LUCA? Owing to the complete absence 

of any bona fide σ-factor homologues in archaea and 

eukaryotes, and of any TBP or TFIIB homologues in 

bacteria, it is unlikely that the RNAP of the LUCA initiated 

transcription aided by σ‐like, TBP-like or TFIIB-like 

transcription factors. There are four possible scenarios 

for the emergence of ancestral forms of the transcription 

initiation factors in the three domains of life (FIG. 8). In 

the first scenario, RNAP in the LUCA did not use any 

transcription initiation factors, and ancestral variants 

of σ-factors and of TBP and TFIIB emerged independently 

in the bacterial and archaeal–eukaryotic lineages, 

following their split. In the second scenario, RNAP in 

the LUCA used both σ-factors and TBP and TFIIB 

factors, followed by loss of TBP and TFIIB in the bacterial 

lineage and loss of σ-factors in the archaeal–eukaryotic 

lineage. In the third scenario, RNAP in the LUCA used 

σ-factors, which were lost and replaced by TBP and 

TFIIB in the archaeal–eukaryotic lineage. In the fourth 

scenario, RNAP in the LUCA used TBP and TFIIB proteins, 

which were lost and replaced by σ-factors in the 

bacterial lineage. Based on parsimony, we favour the first 

scenario, as it requires only three independent evolutionary 

events, versus four events in the third scenario, five 

events in the fourth scenario and six events in second 

scenario (FIG. 8). Rather than regulating transcription by 

recruiting RNAPs to proto-promoters such as the TATA 

box or the –35 and –10 elements, RNAPs could have initiated 

transcription non-specifically by directly associating 

with the template DNA, without basal transcription 

factors. Suitable candidates for these RNAP ‘entry sites’ 

are sequences that have a high A and T content, as they 

have a propensity to distort the DNA topology (that is, 

to cause DNA bending) 104 and they melt readily and thus 

allow loading of the template DNA strand into the RNAP 

active site. It may not be accidental, therefore, that both 

TATA boxes and –10 elements in contemporary promoters 

are rich in T and A residues. Auxiliary protein factors 

could have evolved independently in the bacterial and 

archaeal–eukaryotic lineages to enhance this process, 

and eventually these sequences could have co-evolved 

with their cognate factors, resulting in the TBP–TATA 



REVIEWS 

a 

Bacteria 

c 

σ 

LUCA 

3 events 

Archaea 

Eukaryotes 

TBP and 

TFIIB 

4 events d 

5 events 

Figure 8 | Emergence of transcription initiation factors in the three domains of 

life. The bacterial RNA polymerase (RNAP) strictly depends Nature on σ-factors Reviews for | transcription 

Microbiology 

initiation, whereas archaeal and eukaryotic RNAPs require the initiation factors TATA 

box-binding protein (TBP) and transcription factor IIB (TFIIB). There are four potential 

scenarios for the evolution of these factors. As the first scenario requires the fewest 

independent evolutionary events, it is the simplest explanation for the occurrence of 

transcription initiation factors in all extant life. a | As there are no σ-factor homologues in 

archaea and eukaryotes, and no TBP and TFIIB homologues in bacteria, it is likely that the 

ancestral RNAP of the last universal common ancestor (LUCA) used neither σ-factors nor 

TBP or TFIIB factors, and that these factors evolved independently in the bacterial and 

archaeal–eukaryotic lineages, respectively, after their split. b | An alternative scenario is 

that the RNAP of the LUCA used both σ-factors and TBP and TFIIB factors in parallel, and 

then lost the relevant factors in each lineage. c | A third scenario is that the LUCA used 

σ-factors, and then the archaeal–eukaryotic lineage lost these factors are gained TBP and 

TFIIB factors. d | The final scenario is that the LUCA used TBP and TFIIB factors, and that 

these were then lost in the bacterial lineage and σ-factors were gained. 

b 

Bacteria 

TBP and 

TFIIB 

TBP and 

TFIIB 

6 events 

Archaea 

Eukaryotes 

Bacteria Archaea Eukaryotes Bacteria Archaea Eukaryotes 

LUCA 

σ 

σ 

TBP and 

TFIIB 

TBP and 

TFIIB 

σ 

TBP and 

TFIIB 

LUCA 

LUCA 

box ensemble of extant archaea and eukaryotes and the 

σ-factor–DNA (–35 and –10 elements) ensemble of 

extant bacteria. Interestingly, single-subunit RNAPs, 

such as the T7 RNAP, can initiate transcription at a 

promoter in a sequence-dependent manner, without 

additional factors 5 . 

Owing to the extensive structural and functional 

homology between bacterial NusG, archaeal Spt5 and 

eukaryotic SPT5, and the highly conserved binding sites 

on RNAP, it is almost certain that a NusG‐ and Spt5-like 

transcription factor associated with the RNAP in the 

LUCA, modulated its properties and possibly regulated 

gene expression. This hypothesis implies that the regulation 

of evolutionarily ancient RNAPs could predominantly 

have targeted the elongation phase of transcription 

instead of initiation. It is unclear to what extent the 

ancestral forms of NusG‐like factors regulated transcription 

per se. However, regulation could have occurred by 

counteracting sequence-specific pausing, by modulating 

σ 

σ 

the elongation rates, by affecting the likelihood of entering 

the paused state or by decreasing the pause duration. 

All of these phenomena have been observed with NusG 

in E. coli 105 . Alternatively or in addition to regulating 

transcription by RNAP in the LUCA in a gene-specific 

manner, the NusG‐like factor could have acted as a general 

processivity factor, possibly even by improving the 

poor utilization of RNA templates by RNAP before 

the emergence of DNA as the main coding molecule 16 . 

It was recently demonstrated that NusG plays a crucial 

part in coupling transcription and translation in vivo by 

connecting elongating RNAPs and ribosomes, and this 

further underpins the crucial role of NusG–Spt5–SPT5 

family factors in the regulation of gene expression and 

their possible very early origin 106,107 . 

Conclusions and questions 

RNAPs have a fundamental role in the biology of all 

living organisms. In recent years, the study of RNAPs 

has made tremendous progress, which is partially due 

to a number of technological breakthroughs. X‐ray 

crystallography of RNAPs and of complexes with basal 

transcription factors and nucleic acid scaffolds has given 

us structural information at the atomic level, which has 

helped to formulate theories on the molecular mechanisms 

of transcription 9,34,108 . Single-molecule studies 

using either fluorescence-based methods 61,109,110 or optical 

tweezers 105 have enabled a functional characterization 

of RNAP at the single-molecule level, which is essential 

if we are to characterize and refine our understanding 

of the mechanisms. Systems biology approaches including 

‘deep’ sequencing transcriptomics and ChIP–seq 

(chromatin immunoprecipitation followed by sequencing) 

whole-genome occupancy studies have reported on 

the composition of all transcripts synthesized by RNAPs 

(including their accurate 5′ and 3′ termini) 111 and the 

genomic locations of RNAP and transcription factors 112 , 

respectively. The field of gene expression is currently in 

an exciting phase, during which it will be possible to collate 

all the available information — from the atomic scale 

to a global scale — in order to achieve a genuinely comprehensive 

understanding of transcription in all three 

domains of life. However, many important questions still 

remain unanswered. How is the subcellular organization 

of RNAP in ‘transcription factories or foci’ linked to the 

regulation of transcription 113 ? How do multiple RNAPs 

transcribing the same template affect each other 114,115 ? 

How does the coupling of transcription and translation 

regulate gene expression in the Archaea and the 

Bacteria 106,107 ? How are termination and initiation linked 

during re-initiation of the same RNAP on the same template 

116,117 ? Are the recently discovered plant complexes 

RNAPIV and RNAPV genuine RNA polymerases and, if 

so, what is the nature of their templates and which transcription 

factors regulate their activities 118 ? Not much is 

known about the unorthodox transcription machineries 

in ‘simple’ eukaryotes such as protists — how do they 

work, and what can they tell us about the evolution of 

transcription 119 ? Thus, even though parts of RNAPs are 

in some cases understood on an atomic level, there is still 

much to be learned about these important enzymes. 



REVIEWS 

1. Werner, F. Structural evolution of multisubunit RNA 

polymerases. Trends Microbiol. 16, 247–250, (2008). 

2. Werner, F. Structure and function of archaeal RNA 

polymerases. Mol. Microbiol. 65, 1395–1404 

(2007). 

3. Cramer, P. et al. Structure of eukaryotic RNA 

polymerases. Annu. Rev. Biophys. 37, 337–352 

(2008). 

4. Cheetham, G. M. & Steitz, T. A. Insights into 

transcription: structure and function of single-subunit 

DNA-dependent RNA polymerases. Curr. Opin. Struct. 

Biol. 10, 117–123 (2000). 

5. Steitz, T. A. The structural changes of T7 RNA 

polymerase from transcription initiation to elongation. 

Curr. Opin. Struct. Biol. 19, 683–690 (2009). 

6. Cramer, P., Bushnell, D. A. & Kornberg, R. D. 

Structural basis of transcription: RNA polymerase II at 

2.8 Ångstrom resolution. Science 292, 1863–1876 

(2001). 

7. Gnatt, A. L., Cramer, P., Fu, J., Bushnell, D. A. & 

Kornberg, R. D. Structural basis of transcription: an 

RNA polymerase II elongation complex at 3.3 Å 

resolution. Science 292, 1876–1882 (2001). 

8. Hirata, A., Klein, B. J. & Murakami, K. S. The X‐ray 

crystal structure of RNA polymerase from Archaea. 

Nature 451, 851–854 (2008). 

9. Korkhin, Y. et al. Evolution of complex RNA 

polymerases: the complete archaeal RNA polymerase 

structure. PLoS Biol. 7, e102 (2009). 

10. Zhang, G. et al. Crystal structure of Thermus aquaticus 

core RNA polymerase at 3.3 Å resolution. Cell 98, 

811–824 (1999). 

11. Hirtreiter, A., Grohmann, D. & Werner, F. Molecular 

mechanisms of RNA polymerase — the F/E (RPB4/7) 

complex is required for high processivity in vitro. 

Nucleic Acids Res. 38, 585–596 (2010). 

12. Batada, N. N., Westover, K. D., Bushnell, D. A., 

Levitt, M. & Kornberg, R. D. Diffusion of nucleoside 

triphosphates and role of the entry site to the RNA 

polymerase II active center. Proc. Natl Acad. Sci. USA 

101, 17361–17364 (2004). 

13. Iyer, L. M., Koonin, E. V. & Aravind, L. Evolutionary 

connection between the catalytic subunits of 

DNA‐dependent RNA polymerases and eukaryotic 

RNA-dependent RNA polymerases and the origin of 

RNA polymerases. BMC Struct. Biol. 3, 1 (2003). 

14. Grohmann, D., Werner, F. Cycling through 

transcription with the RNA polymerase F/E (RPB4/7) 

complex: structure, function and evolution of archaeal 

RNA polymerase. Res. Microbiol. 21 Sep 2010 

(doi:10.1016/j.resmic.2010.09.002). 

15. Wassarman, K. M. & Saecker, R. M. Synthesismediated 

release of a small RNA inhibitor of RNA 

polymerase. Science 314, 1601–1603 (2006). 

16. Lehmann, E., Brueckner, F. & Cramer, P. Molecular 

basis of RNA-dependent RNA polymerase II activity. 

Nature 450, 445–449 (2007). 

17. Severinov, K. et al. Structural modules of the large 

subunits of RNA polymerase. Introducing 

archaebacterial and chloroplast split sites in the β 

and β′ subunits of Escherichia coli RNA polymerase. 

J. Biol. Chem. 271, 27969–27974 (1996). 

18. Zakharova, N. et al. Fused and overlapping rpoB and 

rpoC genes in helicobacters, campylobacters, and 

related bacteria. J. Bacteriol. 181, 3857–3859 

(1999). 

19. Severinov, K., Mooney, R., Darst, S. A. & Landick, R. 

Tethering of the large subunits of Escherichia coli RNA 

polymerase. J. Biol. Chem. 272, 24137–24140 

(1997). 

20. Sweetser, D., Nonet, M. & Young, R. A. Prokaryotic 

and eukaryotic RNA polymerases have homologous 

core subunits. Proc. Natl Acad. Sci. USA 84, 

1192–1196 (1987). 

21. Lane, W. J. & Darst, S. A. Molecular evolution of 

multisubunit RNA polymerases: sequence analysis. 

J. Mol. Biol. 395, 671–685 (2010). 

22. Lane, W. J. & Darst, S. A. Molecular evolution of 

multisubunit RNA polymerases: structural analysis. 

J. Mol. Biol. 395, 686–704 (2010). 

23. Ruprich-Robert, G. & Thuriaux, P. Non-canonical DNA 

transcription enzymes and the conservation of twobarrel 

RNA polymerases. Nucleic Acids Res. 38, 

4559–4569 (2010). 

24. Makeyev, E. V. & Bamford, D. H. Cellular RNAdependent 

RNA polymerase involved in 

posttranscriptional gene silencing has two distinct 

activity modes. Mol. Cell 10, 1417–1427 (2002). 

25. Salgado, P. S. et al. The structure of an RNAi 

polymerase links RNA silencing and transcription. 

PLoS Biol. 4, e434 (2006). 

26. Orlicky, S. M., Tran, P. T., Sayre, M. H. & Edwards, 

A. M. Dissociable Rpb4-Rpb7 subassembly of RNA 

polymerase II binds to single-strand nucleic acid and 

mediates a post-recruitment step in transcription 

initiation. J. Biol. Chem. 276, 10097–10102 (2001). 

27. Grohmann, D., Hirtreiter, A. & Werner, F. RNAP 

subunits F/E (RPB4/7) are stably associated with 

archaeal RNA polymerase: using fluorescence 

anisotropy to monitor RNAP assembly in vitro. 

Biochem. J. 421, 339–343 (2009). 

28. Grohmann, D., Werner, F. Hold on! RNAP interactions 

with the nascent RNA modulate transcription 

elongation and termination. RNA Biol. 7, 310–315 

(2010). 

29. Naji, S., Grunberg, S. & Thomm, M. The RPB7 

orthologue E′ is required for transcriptional activity of 

a reconstituted archaeal core enzyme at low 

temperatures and stimulates open complex formation. 

J. Biol. Chem. 282, 11047–11057 (2007). 

30. Ouhammouch, M., Werner, F., Weinzierl, R. O. & 

Geiduschek, E. P. A fully recombinant system for 

activator-dependent archaeal transcription. J. Biol. 

Chem. 279, 51719–51721 (2004). 

31. Werner, F. & Weinzierl, R. O. Direct modulation of RNA 

polymerase core functions by basal transcription 

factors. Mol. Cell Biol. 25, 8344–8355 (2005). 

32. Werner, F. & Weinzierl, R. O. A recombinant RNA 

polymerase II‐like enzyme capable of promoterspecific 

transcription. Mol. Cell 10, 635–646 (2002). 

33. Werner, F., Eloranta, J. J. & Weinzierl, R. O. Archaeal 

RNA polymerase subunits F and P are bona fide 

homologs of eukaryotic RPB4 and RPB12. Nucleic 

Acids Res. 28, 4299–4305 (2000). 

34. Kostrewa, D. et al. RNA polymerase II–TFIIB structure 

and mechanism of transcription initiation. Nature 

462, 323–330 (2009). 

35. Reich, C. et al. The archaeal RNA polymerase 

subunit P and the eukaryotic polymerase subunit 

Rpb12 are interchangeable in vivo and in vitro. Mol. 

Microbiol. 71, 989–1002 (2009). 

36. Grunberg, S., Reich, C., Zeller, M. E., Bartlett, M. S. & 

Thomm, M. Rearrangement of the RNA polymerase 

subunit H and the lower jaw in archaeal elongation 

complexes. Nucleic Acids Res. 38, 1950–1963 

(2010). 

37. Bartlett, M. S., Thomm, M. & Geiduschek, E. P. 

Topography of the euryarchaeal transcription initiation 

complex. J. Biol. Chem. 279, 5894–5903 (2004). 

38. Kang, X. et al. Structural, biochemical, and dynamic 

characterizations of the hRPB8 subunit of human RNA 

polymerases. J. Biol. Chem. 281, 18216–18226 

(2006). 

39. Komissarova, N. & Kashlev, M. Functional topography 

of nascent RNA in elongation intermediates of RNA 

polymerase. Proc. Natl Acad. Sci. USA 95, 

14699–14704 (1998). 

40. Briand, J. F. et al. Partners of Rpb8p, a small subunit 

shared by yeast RNA polymerases I, II and III. Mol. 

Cell Biol. 21, 6056–6065 (2001). 

41. Cox, C. J., Foster, P. G., Hirt, R. P., Harris, S. R. & 

Embley, T. M. The archaebacterial origin of eukaryotes. 

Proc. Natl Acad. Sci. USA 105, 20356–20361 

(2008). 

42. Walmacq, C. et al. Rpb9 subunit controls transcription 

fidelity by delaying NTP sequestration in RNA 

polymerase II. J. Biol. Chem. 284, 19601–19612 

(2009). 

43. Ziegler, L. M., Khaperskyy, D. A., Ammerman, M. L. & 

Ponticelli, A. S. Yeast RNA polymerase II lacking the 

Rpb9 subunit is impaired for interaction with 

transcription factor IIF. J. Biol. Chem. 278, 

48950–48956 (2003). 

44. Campbell, E. A., Westblade, L. F. & Darst, S. A. 

Regulation of bacterial RNA polymerase σ factor 

activity: a structural perspective. Curr. Opin. 

Microbiol. 11, 121–127 (2008). 

45. Murakami, K. S. & Darst, S. A. Bacterial RNA 

polymerases: the wholo story. Curr. Opin. Struct. Biol. 

13, 31–39 (2003). 

46. Gruber, T. M. & Gross, C. A. Multiple sigma subunits 

and the partitioning of bacterial transcription space. 

Annu. Rev. Microbiol. 57, 441–466 (2003). 

47. MacLellan, S. R., Guariglia-Oropeza, V., Gaballa, A. & 

Helmann, J. D. A two-subunit bacterial σ-factor 

activates transcription in Bacillus subtilis. Proc. Natl 

Acad. Sci. USA 106, 21323–21328 (2009). 

48. MacLellan, S. R., Wecke, T. & Helmann, J. D. A 

previously unidentified σ factor and two accessory 

proteins regulate oxalate decarboxylase expression in 

Bacillus subtilis. Mol. Microbiol. 69, 954–967 

(2008). 

49. Wigneshweraraj, S. et al. Modus operandi of the 

bacterial RNA polymerase containing the σ 54 

promoter-specificity factor. Mol. Microbiol. 68, 

538–546 (2008). 

50. Kim, T. K., Ebright, R. H. & Reinberg, D. Mechanism 

of ATP-dependent promoter melting by transcription 

factor IIH. Science 288, 1418–1422 (2000). 

51. Schwartz, E. C. et al. A full-length group 1 bacterial 

sigma factor adopts a compact structure incompatible 

with DNA binding. Chem. Biol. 15, 1091–1103 (2008). 

52. Patikoglou, G. A. et al. TATA element recognition by 

the TATA box-binding protein has been conserved 

throughout evolution. Genes Dev. 13, 3217–3230 

(1999). 

53. Campbell, E. A. et al. Structure of the bacterial RNA 

polymerase promoter specificity σ subunit. Mol. Cell 9, 

527–539 (2002). 

54. Chen, H. T. & Hahn, S. Binding of TFIIB to RNA 

polymerase II: mapping the binding site for the TFIIB 

zinc ribbon domain within the preinitiation complex. 

Mol. Cell 12, 437–447 (2003). 

55. Chen, H. T. & Hahn, S. Mapping the location of TFIIB 

within the RNA polymerase II transcription preinitiation 

complex: a model for the structure of the PIC. Cell 

119, 169–180 (2004). 

56. Sevostyanova, A., Svetlov, V., Vassylyev, D. G. & 

Artsimovitch, I. The elongation factor RfaH and the 

initiation factor σ bind to the same site on the 

transcription elongation complex. Proc. Natl Acad. 

Sci. USA 105, 865–870 (2008). 

57. Renfrow, M. B. et al. Transcription factor B contacts 

promoter DNA near the transcription start site of the 

archaeal transcription initiation complex. J. Biol. 

Chem. 279, 2825–2831 (2004). 

58. Bushnell, D. A., Westover, K. D., Davis, R. E. & 

Kornberg, R. D. Structural basis of transcription: an 

RNA polymerase II‐TFIIB cocrystal at 4.5 Angstroms. 

Science 303, 983–988 (2004). 

59. Liu, X., Bushnell, D. A., Wang, D., Calero, G. & 

Kornberg, R. D. Structure of an RNA polymerase 

II‐TFIIB complex and the transcription initiation 

mechanism. Science 327, 206–209 (2010). 

60. Kulbachinskiy, A. & Mustaev, A. Region 3.2 of the σ 

subunit contributes to the binding of the 3′-initiating 

nucleotide in the RNA polymerase active center and 

facilitates promoter clearance during initiation. J. Biol. 

Chem. 281, 18273–18276 (2006). 

61. Kapanidis, A. N. et al. Initial transcription by RNA 

polymerase proceeds through a DNA-scrunching 

mechanism. Science 314, 1144–1147 (2006). 

62. Goldman, S. R., Ebright, R. H. & Nickels, B. E. Direct 

detection of abortive RNA transcripts in vivo. Science 

324, 927–928 (2009). 

63. Murakami, K. S., Masuda, S., Campbell, E. A., 

Muzzin, O. & Darst, S. A. Structural basis of 

transcription initiation: an RNA polymerase 

holoenzyme-DNA complex. Science 296, 1285–1290 

(2002). 

64. Holstege, F. C., van der Vliet, P. C. & Timmers, H. T. 

Opening of an RNA polymerase II promoter occurs in 

two distinct steps and requires the basal transcription 

factors IIE and IIH. EMBO J. 15, 1666–1677 (1996). 

65. Ohkuma, Y. et al. Structural motifs and potential σ 

homologies in the large subunit of human general 

transcription factor TFIIE. Nature 354, 398–401 

(1991). 

66. Sumimoto, H. et al. Conserved sequence motifs in the 

small subunit of human general transcription factor 

TFIIE. Nature 354, 401–404 (1991). 

67. Garrett, K. P. et al. The carboxyl terminus of RAP30 is 

similar in sequence to region 4 of bacterial sigma 

factors and is required for function. J. Biol. Chem. 

267, 23942–23949 (1992). 

68. Flores, O. et al. The small subunit of transcription 

factor IIF recruits RNA polymerase II into the 

preinitiation complex. Proc. Natl Acad. Sci. USA 88, 

9999–10003 (1991). 

69. Yan, Q., Moreland, R. J., Conaway, J. W. & Conaway, 

R. C. Dual roles for transcription factor IIF in promoter 

escape by RNA polymerase II. J. Biol. Chem. 274, 

35668–35675 (1999). 

70. Eichner, J., Chen, H. T., Warfield, L. & Hahn, S. 

Position of the general transcription factor TFIIF within 

the RNA polymerase II transcription preinitiation 

complex. EMBO J. 29, 706–716 (2010). 

71. Geiger, S. R. et al. RNA polymerase I contains a 

TFIIF-related DNA-binding subcomplex. Mol. Cell 39, 

583–594 (2010). 

72. Selth, L. A., Sigurdsson, S. & Svejstrup, J. Q. 

Transcript elongation by RNA polymerase II. Annu. 

Rev. Biochem. 79, 271–293 (2010). 



REVIEWS 

73. Vassylyev, D. G. Elongation by RNA polymerase: a race 

through roadblocks. Curr. Opin. Struct. Biol. 19, 

691–700 (2009). 

74. Landick, R. The regulatory roles and mechanism of 

transcriptional pausing. Biochem. Soc. Trans. 34, 

1062–1066 (2006). 

75. Brueckner, F., Ortiz, J. & Cramer, P. A movie of the 

RNA polymerase nucleotide addition cycle. Curr. Opin. 

Struct. Biol. 19, 294–299 (2009). 

76. Brueckner, F. et al. Structure–function studies of the 

RNA polymerase II elongation complex. Acta 

Crystallogr. D Biol. Crystallogr. 65, 112–120 (2009). 

77. Ha, K. S., Toulokhonov, I., Vassylyev, D. G. & 

Landick, R. The NusA N‐terminal domain is necessary 

and sufficient for enhancement of transcriptional 

pausing via interaction with the RNA exit channel of 

RNA polymerase. J. Mol. Biol. 401, 708–725 (2010). 

78. Shibata, R. et al. Crystal structure and RNA-binding 

analysis of the archaeal transcription factor NusA. 

Biochem. Biophys. Res. Commun. 355, 122–128 

(2007). 

79. Beuth, B., Pennell, S., Arnvig, K. B., Martin, S. R. & 

Taylor, I. A. Structure of a Mycobacterium tuberculosis 

NusA–RNA complex. EMBO J. 24, 3576–3587 

(2005). 

80. Runner, V. M., Podolny, V. & Buratowski, S. The Rpb4 

subunit of RNA polymerase II contributes to 

cotranscriptional recruitment of 3ʹ processing factors. 

Mol. Cell Biol. 28, 1883–1891 (2008). 

81. Armache, K. J., Mitterweger, S., Meinhart, A. & 

Cramer, P. Structures of complete RNA polymerase II 

and its subcomplex, Rpb4/7. J. Biol. Chem. 280, 

7131–7134 (2005). 

82. Deighan, P. & Hochschild, A. Conformational toggle 

triggers a modulator of RNA polymerase activity. 

Trends Biochem. Sci. 31, 424–426 (2006). 

83. Sigurdsson, S., Dirac-Svejstrup, A. B. & Svejstrup, 

J. Q. Evidence that transcript cleavage is essential for 

RNA polymerase II transcription and cell viability. Mol. 

Cell 38, 202–210 (2010). 

84. Hausner, W., Lange, U. & Musfeldt, M. Transcription 

factor S, a cleavage induction factor of the archaeal 

RNA polymerase. J. Biol. Chem. 275, 12393–12399 

(2000). 

85. Laptenko, O., Lee, J., Lomakin, I. & Borukhov, S. 

Transcript cleavage factors GreA and GreB act as 

transient catalytic components of RNA polymerase. 

EMBO J. 22, 6322–6334 (2003). 

86. Kettenberger, H., Armache, K. J. & Cramer, P. 

Complete RNA polymerase II elongation complex 

structure and its interactions with NTP and TFIIS. Mol. 

Cell 16, 955–965 (2004). 

87. Jeon, C., Yoon, H. & Agarwal, K. The transcription 

factor TFIIS zinc ribbon dipeptide Asp-Glu is critical for 

stimulation of elongation and RNA cleavage by RNA 

polymerase II. Proc. Natl Acad. Sci. USA 91, 

9106–9110 (1994). 

88. Borukhov, S., Lee, J. & Laptenko, O. Bacterial 

transcription elongation factors: new insights into 

molecular mechanism of action. Mol. Microbiol. 55, 

1315–1324 (2005). 

89. Opalka, N. et al. Structure and function of the 

transcription elongation factor GreB bound to 

bacterial RNA polymerase. Cell 114, 335–345 

(2003). 

90. Condon, C., Squires, C. & Squires, C. L. Control of 

rRNA transcription in Escherichia coli. Microbiol. Rev. 

59, 623–645 (1995). 

91. Arnvig, K. B. et al. Evolutionary comparison of 

ribosomal operon antitermination function. 

J. Bacteriol. 190, 7251–7257 (2008). 

92. Greenblatt, J., Nodwell, J. R. & Mason, S. W. 

Transcriptional antitermination. Nature 364, 

401–406 (1993). 

93. Dodd, I. B., Shearwin, K. E. & Egan, J. B. Revisited 

gene regulation in bacteriophage l. Curr. Opin. Genet. 

Dev. 15, 145–152 (2005). 

94. Hirtreiter, A. et al. Spt4/5 stimulates transcription 

elongation through the RNA polymerase clamp coiled 

coil motif. Nucleic Acids Res. 38, 4040–4051 (2010). 

95. Mooney, R. A., Schweimer, K., Roesch, P., Gottesman, 

M. & Landick, R. Two structurally independent 

domains of E. coli NusG create regulatory plasticity via 

distinct interactions with RNA polymerase and 

regulators. J. Mol. Biol. 2, 341–358 (2009). 

96. Zhou, K., Kuo, W. H., Fillingham, J. & Greenblatt, J. F. 

Control of transcriptional elongation and 

cotranscriptional histone modification by the yeast 

BUR kinase substrate Spt5. Proc. Natl Acad. Sci. USA 

106, 6956–6951 (2009). 

97. Nickels, B. E., Mukhopadhyay, J., Garrity, S. J., 

Ebright, R. H. & Hochschild, A. The s 70 subunit of RNA 

polymerase mediates a promoter-proximal pause at 

the lac promoter. Nature Struct. Mol. Biol. 11, 

544–550 (2004). 

98. Artsimovitch, I. Post-initiation control by the initiation 

factor sigma. Mol. Microbiol. 68, 1–3 (2008). 

99. Missra, A. & Gilmour, D. S. Interactions between DSIF 

(DRB sensitivity inducing factor), NELF (negative 

elongation factor), and the Drosophila RNA 

polymerase II transcription elongation complex. Proc. 

Natl Acad. Sci. USA 107, 11301–11306 (2010). 

100. Cheng, B. & Price, D. H. Analysis of factor interactions 

with RNA polymerase II elongation complexes using a 

new electrophoretic mobility shift assay. Nucleic Acids 

Res. 36, e135 (2008). 

101. Svetlov, V., Belogurov, G. A., Shabrova, E., Vassylyev, 

D. G. & Artsimovitch, I. Allosteric control of the RNA 

polymerase by the elongation factor RfaH. Nucleic 

Acids Res. 35, 5694–5705 (2007). 

102. Belogurov, G. A. et al. Structural basis for converting a 

general transcription factor into an operon-specific 

virulence regulator. Mol. Cell 26, 117–129 (2007). 

103. Tan, L., Wiesler, S., Trzaska, D., Carney, H. C. & 

Weinzierl, R. O. Bridge helix and trigger loop 

perturbations generate superactive RNA polymerases. 

J. Biol. 7, 40 (2008). 

104. Peters, J. P. 3rd & Maher, L. J. DNA curvature and 

flexibility in vitro and in vivo. Q. Rev. Biophys. 43, 

23–63 (2010). 

105. Herbert, K. M. et al. E. coli NusG inhibits backtracking 

and accelerates pause-free transcription by promoting 

forward translocation of RNA polymerase. J. Mol. Biol. 

399, 17–30 (2010). 

106. Proshkin, S., Rahmouni, A. R., Mironov, A. & 

Nudler, E. Cooperation between translating ribosomes 

and RNA polymerase in transcription elongation. 

Science 328, 504–508 (2010). 

107. Burmann, B. M. et al. A NusE:NusG complex links 

transcription and translation. Science 328, 501–504 

(2010). 

108. Vassylyev, D. G., Vassylyeva, M. N., Perederina, A., 

Tahirov, T. H. & Artsimovitch, I. Structural basis for 

transcription elongation by bacterial RNA polymerase. 

Nature 448, 157–162 (2007). 

109. Andrecka, J. et al. Nano positioning system reveals 

the course of upstream and nontemplate DNA within 

the RNA polymerase II elongation complex. Nucleic 

Acids Res. 37, 5803–5809 (2009). 

110. Greive, S. J. & von Hippel, P. H. Thinking quantitatively 

about transcriptional regulation. Nature Rev. Mol. Cell 

Biol. 6, 221–232 (2005). 

111. Wang, Z., Gerstein, M. & Snyder, M. RNA-Seq: a 

revolutionary tool for transcriptomics. Nature Rev. 

Genet. 10, 57–63 (2009). 

112. Mooney, R. A. et al. Regulator trafficking on bacterial 

transcription units in vivo. Mol. Cell 33, 97–108 (2009). 

113. Sutherland, H. & Bickmore, W. A. Transcription 

factories: gene expression in unions? Nature Rev. 

Genet. 10, 457–466 (2009). 

114. Epshtein, V., Toulme, F., Rahmouni, A. R., Borukhov, S. 

& Nudler, E. Transcription through the roadblocks: the 

role of RNA polymerase cooperation. EMBO J. 22, 

4719–4727 (2003). 

115. Saeki, H. & Svejstrup, J. Q. Stability, flexibility, and 

dynamic interactions of colliding RNA polymerase II 

elongation complexes. Mol. Cell 35, 191–205 (2009). 

116. Werner, M., Thuriaux, P. & Soutourina, J. Structure– 

function analysis of RNA polymerases I and III. Curr. 

Opin. Struct. Biol. 19, 740–745 (2009). 

117. Spitalny, P. & Thomm, M. A polymerase III-like 

reinitiation mechanism is operating in regulation of 

histone expression in archaea. Mol. Microbiol. 67, 

958–970 (2008). 

118. Landick, R. Functional divergence in the growing family 

of RNA polymerases. Structure 17, 323–325 (2009). 

119. Vanhamme, L. Trypanosome RNA polymerases and 

transcription factors: sensible trypanocidal drug 

targets? Curr. Drug Targets. 9, 979–996 (2008). 

120. Grunberg, S., Bartlett, M. S., Naji, S. & Thomm, M. 

Transcription factor E is a part of transcription 

elongation complexes. J. Biol. Chem. 282, 

35482–35490 (2007). 

Acknowledgements 

We thank S. Gribaldo from the Institute Pasteur, Paris, 

France, for stimulating discussions on evolution and the 

inspiration for figure 8. We also thank K. S. Murakami for 

sharing unpublished results. 

Competing interests statement 

The authors declare no competing financial interests. 

FURTHER INFORMATION 

Finn Werner’s homepage: http://www.smb.ucl.ac.uk/ 

molecular-microbiology/dr-finn-werner.html 

All links are active in the online pdf

L - Technische Universität Braunschweig

Create successful ePaper yourself

Delete template?

Save as template?