Research Highlights

Host ANP32A mediates the assembly of the influenza virus replicase

Avian influenza (colloquially known as bird flu), is a viral infection caused by viruses that primarily affect birds, such as waterfowl and poultry. However, these viruses can also infect humans, leading to potentially lethal infections. Upon avian to mammalian transmission adaptive mutations occur in the viral genome, including in the genes encoding the RNA polymerase. Understanding how avian influenza viruses adapt to mammalian host species can provide the knowledge necessary for potential therapeutics against the disease, or the genetic engineering of domesticated animals resistant to infection.

Recently, Dr. Haitian Fan and graduate student Alex Walker in Professor Ervin Fodor’s group in the Dunn School, collaborating with Professor Jonathan Grimes’ group in the Division of Structural Biology at Oxford, have used cryo-electron microscopy to determine the structure of the RNA polymerase in influenza C viruses (FluPolC). They report the structure of FluPolC alone and in complex with both chicken and human acidic nuclear phosphoprotein 32A (ANP32A), an essential host protein for polymerase activity.

Importantly, these structures revealed possible molecular mechanisms by which influenza viruses replicate their RNA genome. An asymmetric dimer of FluPolC complexes with ANP32A, and the 627-domain of the PB2 subunit, where adaptive mutations accumulate, interacts with an acidic region of ANP32A. The acidic nature of this region could explain how a certain mutation eliminating an acidic residue in the equivalent PB2 subunit in FluPolA enables avian influenza A viruses to infect humans.

Additionally, this study proposes a potential mechanism by which one of the two RNA polymerases in the asymmetric dimer functions as a replicase to replicate the viral genome, while the other RNA polymerase functions as an “encapsidating” polymerase, which assembles nascent RNA product into ribonucleoprotein complexes, and eventually, viruses.

Written by Derek Xu @derekcxu

Carrique L, Fan H, Walker AP, Keown JR, Sharps J , Staller E, Barclay WS, Fodor E and Grimes JM (2020). Host ANP32A mediates the assembly of the influenza virus replicase

Nature (587) 638–643

Building complex circuits out of fluids

In biomedical research, ‘microfluidics’ refers to techniques used to manipulate samples, usually cells, through specially-designed devices and tiny amounts of fluid. Despite its great potential, uptake of microfluidics techniques by the biomedical community has been poor. The solid, often opaque walls of conventional microfluidic devices prevent essential physical and optical access to samples. Additionally, compatibility issues arise between what researchers need and what engineers produce.

To provide a solution to this problem, the Cook (Dunn School) and Walsh (Engineering Science) labs have developed a novel technology, ‘freestyle fluidics’, where the core idea is to replace solid walls confining liquids, with fluid ones. In a recent paper, they introduce a contactless method to build fluid walls, mostly using materials commonly found in bio-labs. Cell culture media is compartmentalised using a submerged micro-jet of a fluorocarbon (FC40) projected through the static layer of media onto the bottom of a conventional polystyrene cell culture dish. FC40 and water-based liquids, such as cell culture media, are immiscible, and thus the projected FC40 sweeps the media away to leave liquid fluorocarbon ‘walls’ pinned to the plate by interfacial forces. Such fluid walls can quickly be built into almost any imaginable 2D circuit. For instance, the image depicts the human circulatory system ‘drawn’ with an FC40 jet.

This work describes simple methods for miniaturizing common workflows in biology and provides solutions to various biomedical research problems, such as the use of irregular polygons to beat the troublesome Poisson limit in single-cell cloning while reducing reagent waste and costs, design of cell culture devices easily integratable into existing workflows, and even cell colony subculture without trypsin/EDTA, via the physical breakup of a clone using the momentum of the FC40 jet. The diversity and flexibility of this approach is expected to lead to widespread adoption of the technique for a variety of applications in biomedicine.

Written by Cristian Soitu. Edited by Alexandra Bisia

Soitu C, Stovall‐Kurtz N, Deroy C, Castrejón‐Pita A, Cook P and Walsh E. (2020). Jet‐Printing Microfluidic Devices on Demand

Advanced Science, p.2001854.

Jack of All Trades: Broad cellular functions of RSAD2/Viperin could provide novel strategies for antiviral therapeutics

Upon infecting human host cells, viruses hijack the cellular replication machineries to assemble additional virus particles and propagate infection, leading to illness. One important component for ensuring replication is the availability of nucleotides (ATP, CTP, GTP, and UTP), which are the building blocks of DNA and RNA. However, the activity of DNA or RNA polymerase can be inhibited by a variation of these nucleotides (also known as nucleotide analogues) that are missing the 3’-hydroxy group. Currently, multiple nucleotide analogues either have been utilised or are in development stages as antiviral or anticancer therapeutics.

In a series of recent publications, Dr. Kourosh Honarmand Ebrahimi, a postdoctoral fellow at the Department of Chemistry in collaboration with Professor William James’s lab at the Dunn School, reported possible antiviral functions for a 3’-hydroxy group-lacking CTP analogue, ddhCTP. ddhCTP is generated by an interferon-inducible protein RSAD2/Viperin, a key enzyme of the innate immune response. Besides catalysing CTP-to-ddhCTP conversion, RSAD2/Viperin has previously been shown to play a role in various cellular processes. Thus, how RSAD/Viperin and the generation of ddhCTP exert antiviral functions have remained elusive.

In their current work, Dr. Ebrahimi and colleagues showed that the conversion of CTP to ddhCTP depletes available cellular concentration of CTP and interferes with the functions of mitochondria. To investigate how RSAD/Viperin could alter broad cellular processes during viral infection, Dr. Ebrahimi and colleagues utilised human induced pluripotent stem cells (hiPSC)-derived macrophages that lack the expression of RSAD/Viperin protein. They showed that ddhCTP inhibits NAD+-dependent activity of several enzymes, possibly by direct competition of NAD+ binding to these important metabolic enzymes.  

Overall, Dr. Ebrahimi and colleagues propose that disruption of NAD+-dependent cellular activity by RSAD2/Viperin-catalysed ddhCTP during pro-inflammatory cues, such as virus infection, could have wide range of downstream effects, protective or suicidal, depending on cell types, in order to restrict the pathogen threats. These studies open up new avenues for exploring novel antiviral drug targets.

The three relevant papers can be found here:



Written by Derek Xu @derekcxu

PP1 – turbocharging mitotic exit

The cell cycle is fundamental in biology, as its (mis)regulation is in involved development, ageing, and cancer. In this Molecular Biology of the Cell paper, researchers from Ulrike Gruneberg’s lab in the Dunn school, in collaboration with the Barr lab in the Department of Biochemistry, explore the requirement for the phosphatase PP1 at the metaphase-to-anaphase transition.

Anaphase, the process by which sister chromatids are equally divided between daughter cells during cell division, begins with the degradation of cyclin B which inactivates the mitotic Cdk1-cyclin B kinase. This is facilitated by the anaphase promoting complex (APC/C) together with its essential co-activator CDC20. Cdk1-cyclin B inhibits anaphase onset by phosphorylating CDC20, which prevents its efficient binding to the APC/C. When cyclin B degradation starts, this inhibition is relieved, further driving APC/C activation. The de-phosphorylation of CDC20 which enables this positive feedback loop is therefore a key regulatory step initiating anaphase onset.

In this paper, James Bancroft, James Holder, Zoë Geraghty and their colleagues use a combination of biochemical and live cell imaging approaches in human cells to demonstrate that phosphatase PP1 counters the phosphorylating activity of CDK1 by dephosphorylating CDC20, which results in an active APC/C complex, cyclin B degradation, and appropriate metaphase-to-anaphase transition.

These results add important new insights to our understanding of how PP1 regulates the metaphase-to-anaphase transition in human cells and set the scene for future investigations into further PP1 targets at this stage of the cell cycle.


Written by Alexandra Bisia @AlexandraBisia

Bancroft J, Holder J, Geraghty Z, Alfonso-Pérez T, Murphy D, Barr FA, Gruneberg U. PP1 promotes cyclin B destruction and the metaphase-anaphase transition by dephosphorylating CDC20

Mol Biol Cell. mbcE20040252. doi:10.1091/mbc.E20-04-0252


Bacteria use liquid crystalline armour to shield against antibiotics

Pathogenic rod-shaped bacteria are responsible for causing many human infectious diseases such as meningitis and cholera. A wide variety of these bacteria are becoming tolerant to current antibiotics, rendering treatments ineffective. It is therefore of critical importance to understand the mechanisms used by these bacteria to evade antibiotics.

Abul Tarafder and colleagues from Tanmay Bharat’s group in the Dunn School have identified a mechanism by which the rod-shaped bacterium, Pseudomonas aeruginosa, can evade antibiotics by surrounding its cells with a self-made protective casing. The bacteria produce a symbiotic filament-shaped phage, Pf4, that phase-separates into spindle-shaped liquid crystals. These encapsulate bacterial cells, preventing effective concentrations of antibiotic reaching the cell, thus ensuring bacterial survival. Interestingly, the authors found that this phage-mediated antibiotic tolerance mechanism is profoundly influenced by biophysical size and shape complementarity rather than the biochemical properties of the phage and bacteria, as the phage liquid crystals could encapsulate inanimate rods of comparable size to bacteria. This suggests that the mechanism of encapsulation by protective casings could be a general strategy adopted by many bacteria to evade antibiotics. This new knowledge is applicable to a wide variety of pathogenic bacteria so could have widespread implications on the development of novel methods to combat antibiotic tolerance.

Written by Heather Jeffery @HeatherMJeffery

Tarafader, AK, von Kügelgen A, Mellul AJ, Schulze U, Aarts DGAL, Bharat TAM (2020). Phage liquid crystalline droplets form occlusive sheaths that encapsulate and protect infectious rod-shaped bacteria.

PNAS pii: 201917726. doi: 10.1073/pnas.1917726117. 

It takes two to tango: PARP1 active site is completed by HPF1

PARP1 is a poly [ADP-ribose] polymerase that can sense DNA damage and facilitate the choice of repair pathway. Currently, PARP1 inhibitors are the preferred treatment for carcinomas which are already deficient in DNA damage repair through acquired BRCA1/2 mutations.

In their previous work, Ivan Ahel group showed that the PARP1 inhibitor efficacy is greatly increased in cells lacking HPF1, a PARP1 interacting protein. In cells, PARP1 preferentially adds poly ADP-ribose to serine residues of proteins, while in vitro PARP1 modifies aspartate and glutamate residues. Strikingly, adding HPF1 to an in vitro reaction corrected PARP1 specificity to serine. Therefore, the group concluded that HPF1 likely plays a crucial part in PARP1 function.

In a recent publication from Ivan Ahel lab, Marcin Suskiewicz, Florian Zobel, and colleagues show that HPF1 directly contributes to the PARP1 active site with substrate binding and catalytic residues. The strong HPF1-PARP1 interaction is opposed by an autoinhibitory region of PARP1. This region is known to locally unfold on binding DNA lesions. Therefore, the authors propose that the complex formation is a regulatory mechanism restricting PARP1 activity until suitable cues, such as DNA damage induced PARP1 DNA binding, present themselves. This work hugely contributed to our mechanistic understanding of ADP-ribosylation synthesis and reversal. Most importantly, the authors provided the clinical community with an important puzzle piece that could help explain and predict reactions to PARP1 inhibitors.

Written by Lucie Kafkova

Suskiewicz MJ, Zobel F, Ogden TEH, Fontana P, Ariza A, Yang JC, Zhu K, Bracken L, Hawthorne WJ, Ahel D, Neuhaus D, Ahel I (2020). HPF1 completes the PARP active site for DNA damage-induced ADP-ribosylation.

Nature (2020).

The immune system gets TICK-ed off by a new class of complement pathway inhibitors

The human body defends itself from infection by utilising a complex and diverse array of weapons held by the immune system. One important element of the innate immune response is the complement system (also known as the complement cascade), which enhances the ability of immune soldiers such as antibodies to attack and destroy invading pathogens. Overactivation of the complement system can lead to adverse effects such as hyper-inflammation, autoimmunity, etc. A number of disease-causing organisms, including biting ticks, have evolved ways to suppress the complement system and can therefore overcome this attack. The prospect of identifying inhibitors of this system has long been sought, with the promise of therapeutic potential.

Martin Reichhardt and colleagues from Susan Lea’s lab have now uncovered a novel class of complement inhibitors, named the CirpT family, from tick saliva. They showed binding of CirpT1 to complement C5, one of the pointy ends of the complement pathway spear, via a site not targeted by already known inhibitors. Using cryo-electron microscopy and X-ray crystallography, they uncovered that CirpT1 obstructs the interaction and cleavage of complement C5 by C5-convertase, a critical final step in the complement pathway activation. The mechanistic insight into this unique mode of action by CirpT1 certainly provides a platform to further investigate the effect of other complement system inhibitors and also to design potential therapeutic agents.



Written by Iqbal Dulloo

Reichhardt MP, Johnson S, Tang T, Morgan T, Tebeka N, Popitsch N, Deme JC, Jore MM, Lea SM (2020).  An inhibitor of complement C5 provides structural insights into activation

PNAS 117 (1): 362-370

A role for R-Loops in promoting antisense transcription

R-Loops are nucleic acid structures that have been implicated in both DNA damage and DNA repair processes. They tend to form during transcription, when a growing RNA strand invades the DNA to form an RNA:DNA hybrid. R-Loops contain single-stranded DNA, a type of DNA which has been shown to possess the potential to initiate transcription.

By forming these structures in vitro, researchers from the Proudfoot Lab demonstrated that R-Loops are indeed able to promote transcription. The team then moved into cells to find out what sort of transcripts these R-Loops might be initiating. When they removed the R-Loops using a specific enzyme called RNase H1, they found that levels of antisense long non-coding RNAs (lncRNA) were reduced when compared with cells which had not been expressing the RNase. Furthermore, the researchers showed that these RNase-sensitive lncRNA transcripts were often formed adjacent to R-Loops, suggesting that their formation is R-Loop dependent.

lncRNA may form from protein-coding genes but do not code for proteins themselves. Their function is enigmatic, and little was previously known about how lncRNA are produced. This study offers insight into their origin, as well as positing a novel function for R-Loops in our genome.

Written by Laura Hankins

Tan-Wong SM, Dhir S, Proudfoot NJ (2019). R-Loops Promote Antisense Transcription across the Mammalian Genome.

Mol. Cell 76(4):600-616.e6

RNA modification protects genome stability

RNA, a nucleic acid important for protein production and regulation, consists of only 4 bases. However, a large amount of variation can be achieved by adding modifications. One such modification is N6-methyladenosine (m6A), where the adenosine in RNA has a methyl group (-CH3) added. Therefore, the genomic language can be more complicated than at first sight.

Natalia Gromak’s group from the Dunn School, in collaboration with Alexey Ruzov’s group at the University of Nottingham and others, has identified a role of m6A in the regulation of genome stability in human pluripotent stem cells. They found it performs this role through control of the number of R-loops. These structures consist of a RNA:DNA hybrid and unpaired single-stranded DNA and are involved in regulating gene expression and telomere length. It is important to regulate R-loop numbers as an excess of these structures can lead to cell growth retardation and an increase in DNA double-strand breaks, associated with neurodegeneration and cancer.

The researchers found that the number of R-loops containing m6A on the RNA portion varied throughout the cell cycle, rising in levels during the lead up to mitosis. By investigating a knock-out cell line, they identified that an increase in m6A in R-loops leads to mRNA degradation through the m6A reader YTHDF2. Therefore, m6A in R-loops acts as a signal to promote R-loop removal. This is important for maintaining genome stability and healthy cells.

Written by Heather Jeffery @HeatherMJeffery

Abakir A, Giles TC, Cristini A, Foster JM, Dai N, Starczak M, Rubio-Roldan A, Li M, Eleftheriou M, Crutchley J, Flatt L, Young L, Gaffney DJ, Denning C, Dalhus B, Emes RD, Gackowski D, Corrêa Jr IR, Garcia-Perez JL, Klungland A, Gromak N, Ruzov A. (2020). N6-methyladenosine regulates the stability of RNA:DNA hybrids in human cells

Nature Genetics

Exploring DNA damage repair through new angles: how the flexibility of DNA affects RAD51 and homologous repair

An organism’s genome, encoding all the genes and proteins used by an organism for life, is composed of DNA sequences. As it is passed down from generation to generation, the genome needs to be copied and maintained with as little error as possible.

Damage to DNA, such as through double-strand breaks (DSBs), can lead to genome instability and tumorigenesis if left unrepaired. The best error-free repair mechanism, homologous recombination (HR), requires the creation of single-stranded DNA (ssDNA) at the damaged site. ssDNA is rapidly recognized by RAD51, a recombinase protein that catalyses HR-mediated repair. However, exactly how RAD51 preferentially binds to ssDNA over dsDNA is unknown.

The Esashi and Dushek labs worked together to tackle this question. Collaborating with colleagues from the Department of Chemistry and the University of California, Irvine, Federico Paoletti from the Esashi lab explored the problem from a new angle: the difference between the flexibility of ss- and dsDNA. 

Federico and colleagues used surface plasmon resonance and small-angle X-ray scattering to measure the kinetics of RAD51 binding and the flexibility of different DNA substrates, respectively. Combined with mathematical modeling, the interdisciplinary collaborators were able to show that RAD51 is a mechano-sensor, able to polymerize faster on more flexible DNA. While thermodynamically counterintuitive, as stably bound ssDNA has less entropy than unfettered ssDNA, the investigators show that the entropic penalty is offset by the strong RAD51 self-interaction. They also demonstrate that RAD51 binds to DNA through a two-step “Bend-to-Capture” interaction, which is facilitated by more flexible DNA.

In this way, RAD51 is able to rapidly accumulate at sites of DNA damage to start HR repair, and could potentially explain some previously unexplained phenomenon, such as the accumulation of DNA damage at “stiff DNA,” such as poly-A rich regions of the genome.


Written by Derek Xu @derekcxu

Federico P, El-Sagheer AH, Allard J, Brown T, Dushek O, Esashi F (2020). Molecular Flexibility of DNA as a Key Determinant of RAD51 Recruitment

EMBO 39(7):e103002. doi: 10.15252/embj.2019103002.