Supervisors: Dr Jack Mellor (University of Bristol), Dr Krasimira Tsaneva-Atanasova (University of Exeter), Dr Tony Pickering (University of Bristol)

The hippocampus is critical for the encoding and recall of episodic memories in the brain. It is believed that the process of pattern separation, whereby components of episodic memories are dissected, is performed at inputs to dentate granule cells whereas pattern completion, where the details of episodic memories are brought together, is performed at synapses between CA3 pyramidal cells. The link between these two regions is the mossy fibre pathway that transmits information from granule cells in the dentate gyrus to pyramidal cells and interneurons in the CA3 region. The hippocampus is also strongly implicated in the encoding of novelty via inputs from the noradrenergic system, which densely innervates the hippocampus and in particular the dentate gyrus and CA3 regions. Therefore, it is hypothesized that dentate gyrus and CA3 network function will be transformed by the release of noradrenaline. We have recently discovered/studied the impact of neuromodulators on hippocampal circuitry and have a long track record of studying the properties of mossy fibre synaptic transmission, which has led to a series of predictions for the impact of noradrenaline at this synapse.

The project will involve the design and implementation of a network model for the effect of mossy fibre transmission on CA3 network activity building on similar models already implemented in our labs. The aim will be to combine biophysically plausible descriptions of the electrical activity of principal excitatory neurons and inhibitory interneurons in the hippocampus with the hypothesised actions of noradrenaline on distinct neuronal subtypes. This model will then be tested experimentally by investigating the CA3 network response to mossy fibre activity in hippocampal slices and the alterations caused by activation of noradrenergic receptors.

Supervisors: Prof Daniel Robert (University of Bristol), Dr Heather Whitney (University of Bristol), Prof Dudley Shallcross (University of Bristol), Dr Gregory Sutton (University of Bath), Dr Mike Birkett (Rothamsted Research), Jason Chapman (Rothamsted Research), Andy Reynolds (Rothamsted Research), Allison Haughton (Rothamsted Research), Prof Juliet Osborne (University of Exeter)

Our research has shown that bumblebees can detect the weak electrostatic fields around flowers. Such electric fields constitute new floral cues, and enhance a bee’s capacity to learn the difference between two floral colours. Electric fields thus appear to play a role in plant-pollinator interactions. Several fundamental questions arise from these findings. What is the structure and ultimate relevance of these flower- and beegenerated electric fields? Central to this project is the question of how bees measure floral and environmental electric fields. What is the sensory mechanism for electroreception? Can bees measure their own field? This studentship aims at investigating the sensory basis of bee electroreception, with regard to characterising for the first time the type and range of electrical information the bee’s sensory system evolved to detect.

The project aims at establishing the bee’s sensitivity to weak electrical fields in both laboratory and field experiments. In the lab, hypotheses will be tested as to how fine hairs covering the bee’s body detect efields. Preliminary mechanical and electrophysiological evidence indicates that cranial hairs can sense electric fields. Alternative hypotheses will address the putative role of antennae, wings and other hairs in electroreception.

Electroreception will be behaviourally tested with training and psychophysical tests using the proboscis extension reflex. Measuring electrics in natural flowers and then simulate them in artificial e-flowers will allow determining the structure, shape and magnitude of the electric fields bees can detect. Mathematical modelling will be used to investigate the structure and diversity of weak electric fields around bees, flowers and from the atmospheric environment to inform sensory experiments above. The large scale foraging trips of bees in the open environment will be monitored using radar techniques, in conjunction with measurements of atmospheric potential gradients and weather conditions. Electric fields may affect, perhaps even facilitate chemical signalling between herbivores and plants. Pollinator and/or parasitoid behavior may in turn initiate floral electric fields, which may alter plant semiochemistry and therefore relationship to insects.

Supervisors: Prof Mike Mendl (University of Bristol), Prof Iain Gilchrist (University of Bristol), Prof Peter Dayan (UCL), Dr Elizabeth Paul (University of Bristol)

We have developed a conceptual framework and translational task (Harding et al. 2004. Nature 427,312; Mendl et al. 2010. Proc Roy Soc B 277, 2895-2904) to examine the influence of emotional (affective) states on decision-making under ambiguity in non-human animals. Humans in negative emotional states are more likely to make negative judgements about ambiguous stimuli (e.g. words, phrases, facial expressions) than happier people. Over 30 published studies using our approach have shown that animals in induced negative states are also more likely to judge ambiguous stimuli negatively (e.g. auditory, visual cues). This suggests that decision-making under ambiguity is a valuable new indicator of animal affect and welfare.

However, the psychological processes underlying these findings remain unknown. Affective states might alter expected utility by influencing subjective probabilities or valuations of decision outcomes; they might also influence sensory perceptual processes such as stimulus generalization. Disentangling these processes will offer important new insights into the links between affect and decision-making. The aim of this PhD will be identify the processes by which affective states alter decisions, and the extent to which these show parallels across humans and rodents. Carefully controlled experiments, e.g. sensory discrimination paradigms that are designed and interpreted via quantitative computational and mathematical models, will be employed.

Supervisors: Dr Andrew Davidson (University of Bristol), Dr Séverine Tasker (University of Bristol), Prof Linda Wooldridge (University of Bristol)

Feline infectious peritonitis (FIP) is a fatal disease in young cats arising due to (type 1) feline coronavirus (FCoV) infection. There is an urgent need to develop an effective FCoV vaccine that protects against FIP. However the type 1 FCoV cell entry receptor (CER) is currently unknown, which has impeded the propagation of field-derived FCoVs in vitro. This has prevented the production of recombinant attenuated FCoVs, which are essential for FCoV vaccine development.

With recent technical advances, we now have the tools to circumvent these barriers. Firstly, using a novel high-throughput proteomic approach to identify the FCoV CER, we can generate feline cell lines that allow field-derived FCoV propagation. Following a recently established protocol, we will exploit the binding of the FCoV spike (S) protein to its CER to identify the latter, making use of the University of Bristol’s state-of-the-art proteomic and bioinformatics expertise. Secondly, we have recently sequenced the genomes of multiple field-derived FCoVs. This will enable us to use a rapid synthetic biology approach to establish a reverse genetic system for a representative FCoV, which will be a powerful technological platform for vaccine development as it allows us to manipulate the FCoV genome and produce novel viruses. Mutations that are predicted to attenuate the ability of FCoV to evade the host innate immune response can be introduced into the viral genome, allowing us to produce recombinant attenuated FCoVs that can serve as candidate vaccine strains, which would be a major step forward in our aim of preventing the devastating disease of FIP.

Supervisors: Prof Jeremy Henley (University of Bristol), Prof Dek Woolfson (University of Bristol)

Realistic, but simplified models for complex systems, such as synaptic networks, would facilitate both basic and applied research geared at developing our understanding of the brain, and testing the efficacy and side effects of new reagents and drugs. The aim of this project is optimize and exploit the hSAF (hydrogelating self-assembling fibers) peptide hydrogel system developed in the Woolfson lab to grow cultured rat neurons and form directionally orientated synaptic networks in vitro in the Henley lab.

Once established, these aligned cultures will provide a new and extremely useful tool for investigating specifically pre- or postsynaptic events. This will be achieved using viral vectors already available in Henley’s lab to shRNA knock-down one or more protein of interest and, where appropriate, use molecular replacement strategies to express shRNA insensitive WT or mutated fluorophore-tagged protein.

A major goal for the project will be to develop and use custom designed hydrogels incorporating specific axonal or dendritic growth cues to align and coordinate the growth and synapses of hippocampal or cortical neurons seeded into the matrix. Initially, these will be linear but the aim is to develop a torus-shaped scaffold using techniques in development in the Woolfson lab. This will provide the scaffold for a closed and orientated neuronal network analogous to single autaptic cultured neurons, but much more useful because the network will comprise identifiable and aligned neurons, each of which can be individually manipulated.

Successful proof-of-concept experiments have already shown that PC12 cells grow well in and hSAFs decorated with the fibronectin RGDS motif that mediates cell attachment to the extracellular matrix. As the project develops, we intend to explore of other motifs to functionalise the hSAFs; for example, incorporation BDNF or netrin within specific regions of the hydrogel promote and attract neurite outgrowth and synapse formation.

Supervisors: Prof Eamonn Kelly (University of Bristol), Dr Richard Sessions (University of Bristol)

G protein-coupled receptors (GPCRs) are highly dynamic proteins that display complex patterns of behaviour in signalling, for example the same GPCR can couple to multiple signalling pathways in a cell, whilst different ligands acting at the same GPCR can lead to distinct signalling outputs (ligand bias). This is of particular interest to the pharmaceutical industry as it can lead to the development of novel drugs with enhanced therapeutic efficacy and fewer adverse effects. Perhaps surprisingly, the computational modelling of fundamental concepts such as ligand bias at GPCRs is not well advanced. The aim of this project therefore is to use Molecular dynamics simulations (MDs) as a tool to understand ligand bias at a biologically important GPCR.

Prof Eamonn Kelly has made detailed studies of ligand interaction with, and molecular signalling of, the μ opioid receptor (MOPr), an extremely important GPCR which is crucial for pain and reward pathways in mammals. Prof Eamonn Kelly also has wide experience of ligand bias at this receptor. Dr Richard Sessions is a highly experienced protein modeller, including the use of MDs to model membrane proteins. Together they wish to understand the molecular basis of ligand bias at the mu opioid receptor, using a combination of computer modelling and measures of receptor cell signalling. Based upon the published crystal structure of the mu opioid receptor and related GPCRs, the student would build a model of the mu opioid receptor for MDs, using these simulations to:

• Determine the nature of the interaction of biased/unbiased ligands with the MOPr binding pocket, as well as receptor conformational changes induced/stabilised by biased and unbiased ligands to produce active, presumably distinct receptor conformations
• Use the MD models to screen other ligands to predict biased/unbiased ligand phenotype and predict the effect of mutations

Furthermore the student will test and confirm these modelling outcomes by expressing mu opioid receptor and relevant mutants in mammalian cell lines and determining the binding and signalling of biased ligands at this receptor.

Supervisors: Dr James Spencer (University of Bristol), Prof Adrian Mulholland (University of Bristol)

Beta lactams are the most widely prescribed antibacterial drugs, and are of particular value against infections by Gram-negative bacteria, where few alternatives exist. In these organisms β-lactamase production is the main resistance mechanism. While these enzymes are many and varied, several have been structurally and biochemically characterised, making β-lactamases both relevant to human (and animal) health and suitable model systems for developing computational methods to study enzyme-catalysed reactions.

Carbapenems are the newest, most potent β-lactams and key antibiotics for severe bacterial infections. While most β-lactamases are inhibited by carbapenems, enzymes that efficiently hydrolyze these substrates (carbapenemases) are disseminating. The molecular basis for this expansion of activity remains obscure. We have previously (J Am Chem Soc 134 18275 (2012); Chem Comm, in press (2014)) used structural and computational (molecular dynamics (MD) and quantum mechanical (QM)) approaches to study interaction of β-lactamases with carbapenems with the aim of both understanding the basis for this variation in activity and developing methodologies for predicting the activity of uncharacterised systems. We propose to extend this work to i) study how activity against carbapenems may be acquired by point mutation, ii) predict the reactivity towards carbapenems of uncharacterised enzymes from environmental sources and iii) develop and apply in silico tools to identify strategies for carbapenemase inhibition.

β-lactamase complexes will be studied computationally using MD, with QM/molecular mechanics (QM/MM) approaches used to simulate the acylation and deacylation stages of the reaction. Simulations will be based upon available crystal structures, and upon structures generated by homology modelling and/or in silico ligand docking. Experimental work will encompass crystallographic structure determination for both enzyme:ligand complexes and mutated enzyme variants; as well as uncharacterised β-lactamases; complemented with pre-steadystate kinetic measurements of free energy barriers for the different reactions. As the project progresses experimental focus will shift to investigating the properties of systems modelled in silico, to assess the predictive power of our simulations, and of the activity of potential inhibitors.

Supervisors: Dr Dafydd Jones (Cardiff University), Prof Saverio Russo (University of Exeter), Dr Emyr Macdonald (Cardiff University), Dr Martin Elliott (Cardiff University)

Bacterial resistance to antibiotics is one of the most significant crises in modern healthcare. The most widely utilised class of antibiotics (and therapeutics in general) are the β-lactams, which include ampicillin, amoxicillin and methicillin. By far the main mechanism bacteria use to confer β-lactam resistance is the production of β-lactamases that hydrolyses the pseudo-peptide bond in the β-lactam ring critical for it is antibacterial activity. The aim of this project to generate a high information content miniaturised sensor with the capacity to rapidly detect the presence∧ type of β-lactamase to allow more effective clinical use of current β-lactams. To achieve this engineered β-lactamase inhibitory protein (BLIP) will be selectively interfaced with a graphene field-effect transistor so that protein-protein interactions are transduced into a change in resistance of graphene-bridged microelectrodes through both bulk binding effects and electrostatic surface interactions. This system has the potential to sense events connected to single protein molecules. BLIP is the ideal sensor as it binds a wide variety of clinically relevant β-lactamases with very distinct surface-facing electrostatics and interaction properties, which will potentially offer discrete electronic signatures. A key requirement is defined and directed interfacing of BLIP with graphene to maximise communication between the two components. This will! be achieved using a synthetic biology approach to design in silico residue positions for the recombinant incorporation of non-native chemical handles (e.g. phenyl azide chemistry) that will allow high precision and optimal interfacing of protein with graphene. Detailed single!molecule analysis (AFM/STM)!of! the protein-graphene interface will lead to the microfabrication of the protein-graphene-electrode biosensing platform and begin monitoring and assessing binding events through changes in conductivity.

Supervisors: Prof Rob Honey (Cardiff University), Prof Kevin Fox (Cardiff University)

This project combines expertise and training opportunities from two laboratories at Cardiff University to tackle an important question in an interdisciplinary context: How does experience with similar stimuli increase the distinctiveness of their central representations? This question has been investigated by behavioural scientists in the context of the phenomenon perceptual learning, where animals given simple exposure to similar stimuli (e.g., two textures) will later discriminate between them more readily than between two novel stimuli (e.g., Montuori & Honey, 2014, Neuroscience and Biobehavioral Reviews; see also, Mundy, Downing, Dwyer, Honey & Graham, 2013, J. Neuroscience). It has also been examined by neuroscientists studying the rodent whisker system and experience-based changes in the barrel cortex (e.g., Fox, 2008, Barrel Cortex). There are generic computational models of perceptual learning, but whether these apply to tactile stimuli in rodents is unknown, and the neural substrates of perceptual learning resulting from simple exposure are also unknown (cf. Kuhlman, O’Connor, Fox & Svoboda, 2013, J. Neuroscience). The overarching aim of the project is to use new evidence from behavioural, electrophysiological recording and optical imaging studies to develop a biologically constrained computational model of perceptual learning using the rodent whisker system as a model.

Supervisors: Dr Andrew Young (University of Exeter), Dr Alastair Wilson (University of Exeter)

Collaborators: Dr Xavier Harrison (Institute of Zoology, London)

Negative effects of inbreeding on animal health and performance (inbreeding depression) are well documented, but it is unknown whether an organism’s social environment can mitigate such effects. This is surprising as social species exhibit both a high potential for inbreeding (living in kin-structured populations) and the potential for cooperative behaviour to mitigate the negative effects of inbreeding. Quantifying inbreeding depression requires the calculation of proxies for genome-wide heterozygosity. While research to date has often been hampered by a reliance on crude proxies for genome-wide heterozygosity (e.g. pedigree-based inbreeding coefficients or estimates using few genetic loci), the advent of Restriction site Associated DNA (RAD) sequencing now offers the opportunity to obtain accurate measures of genome-wide heterozygosity by typing thousands of loci across the genome of nonmodel organisms. RAD-Seq therefore has the potential to revolutionise our ability to investigate, and then localise on the genome, effects of inbreeding depression.

AIM: To test for the first time whether inbreeding depression can be mitigated by the social environment, utilising existing large-scale genomic (RAD-Seq data from >300 individuals) and phenotypic data sets (growth, size, reproductive success, survival) from a long-term study of a social bird, the sparrow-weaver. Sparrow-weavers (i) live in kin-structured populations (high potential for inbreeding), and (ii) are cooperative breeders (parents are assisted with offspring care by 0-12 helpers), yielding a high potential for the parent’s social environment (helper number) to mitigate effects of parental heterozygosity on parental fitness. We will:

• Test for effects of RAD-Seq-derived genome-wide heterozygosity on parental fitness
• Establish whether such effects are mitigated by the presence and number of helpers
• Contrast conclusions from RAD-Seq approach with (traditional) pedigree-based approach
• Use linkage mapping to partition effects of inbreeding depression across the chromosomes, to establish whether mediated by many loci of small effect or few loci of large effect.

Supervisors: Dr Isabelle Jourdain (University of Exeter), Dr John Chilton (University of Exeter)

Cell polarity lies at the heart of a vast amount of cellular processes in virtually all kingdoms. Of all cells, neurons arguably have the most complex shapes, because of their elongated axons and dendrites. By contrast, with its rod-like shape, the fission yeast S. pombe displays one of the simplest geometries. Yet, dendrite tip growth and yeast tip growth are orchestrated by the same set of factors. We propose to exploit the simplicity and tractability of yeast cells to describe a fundamental system, before we challenge our findings against one of the most complex cellular morphologies.

Cellular morphology is the result of complex interplay between the cytoskeleton and the cell membrane. Actin tracks are believed to deliver secretory vesicles to specific areas of the cell surface, thereby controlling lipid and protein composition of the plasma membrane. The tethering of the incoming vesicle, before docking and fusion with the plasma membrane, is mediated by a complex of proteins called the exocyst. Recent evidence indicates that the exocyst in turn regulates the actin cytoskeleton but how this occurs is not understood.

A PhD student will tackle this question by exploiting the experimental power of the fission yeast S. pombe in conjunction with vertebrate neuronal cultures. Using a range of genetic, biochemical and fluorescent imaging techniques the PhD student will use yeast to identify the partners of the exocytic system in actin organisation and apply the conclusions to vertebrate neurons. Due to their inherent complexity, variations in neuron morphology are currently practically impossible to assess. In collaboration with the Jeremy Metz (Bioinformatics hub), an automated image processing framework will be developed to objectively extract morphological information from fluorescent images of the cells’ borders, from which quantitative conclusions may be drawn (eg. width, length, axiality, symmetry, polarity, spreading, etc). This tool will help probe features of cell shape across classes of organism. All in all, these data will provide novel insights into a universal mechanism that is utilised throughout evolution, in fungi, plants and animals.

Supervisors: Dr James Wakefield (University of Exeter), Prof Rob Beardmore (University of Exeter), Dr Jeremy Metz (Biomedical Informatics Hub), Dr Kate Heesom (University of Bristol)

Cell division is a fundamental biological process, driven by the formation of a microtubule (MT)-based mitotic spindle that aligns the chromosomes and ensures their segregation. Research over the last twenty years has led to a list of gene products with roles in spindle formation. The challenge of post-genomic biology is to understand how these gene products work together to define biological process. We have developed methodologies, based on high-resolution fluorescence microscopy, coupled to image analysis, and quantitative proteomics which allow us understand generation and organisation of MTs during mitotic spindle formation in the model organism, Drosophila melanogaster (Hayward et al., (2014) Dev Cell). The goal of this project is to determine the molecular mechanisms underpinning MT organisation during the early stages of mitotic spindle formation, referred to as “coalescence”.

The main body of the PhD will see the student investigating the co-ordinated functions of the kinesin-like proteins known to function during spindle formation. For those for which antibodies are not available, the student will produce constructs to allow the bacterial expression and purification of pure protein, in order to generate polyclonal antibodies for usage. Under the guidance of Prof Beardmore and Dr Metz they will expand on the computational techniques they have been introduced to, gaining deeper understanding of the numerical approaches and connections with the appropriate theoretical modeling with potential for additional collaboration with Mathematical modeling groups within Exeter.

Supervisors: Dr Stefano Pagliara (University of Exeter), Prof Richard Titball (University of Exeter)

Populations of genetically identical cells are not homogeneous, but contain subpopulations with phenotypic heterogeneity in their response to external stimuli. This has severely constrained our scientific understanding of key problems such as the emergence of drug resistance, where antimicrobial screening at the population level often leads to ambiguous findings, for instance due to the presence of drug-tolerant persister cells. Consequently, several key issues are not understood including drug uptake at the single-cell level, the effect of the drug on the single cell and the cell response, and phenotypic/genotypic differences among sub-populations of cells responding to stress in different ways. Addressing these issues using cutting-edge single-cell techniques is critical for an understanding of the dynamics of the response to drugs in heterogeneous populations and will allow both the more effective use of existing drugs and better ways of screening novel drugs.

The aim of this project is to develop novel technologies that will then allow an investigation of the underlying mechanisms of cellular response to external stimuli, particularly exposure to antimicrobial drugs, at the single-cell level. The principal supervisor (SP) has recently introduced a novel assay to control and characterize single live cells in microfluidics (Nature Materials 13, 638 (2014)) and label-free technique for studying drug transport across lipid membranes (Lab on a Chip 14, 2303 (2014)). In this project the student will develop novel microfluidic and microscopy techniques to correlate external stresses and cellular responses at the single-cell level. Hundreds of individual cells will be confined in micro-chambers and exposed individually to similar stresses (e.g. antimicrobial drugs). The effect of the stress will be monitored in situ; for example the antimicrobial uptake in each single-cell will be quantified either by UV auto-fluorescence or by Raman scattering. Cellular response will be monitored by carrying out a multi-parametric analysis of cell health (viability, morphology, growth rate, deformability, motility, activation of signalling pathways, gene regulation and metabolic activity). Finally, cells will be independently harvested for ex situ analyses such as single-cell transcriptomics.

Supervisors: Dr Ivana Gudelj (University of Exeter), Prof Robert Beardmore, (University of Exeter), Prof Ed Feil (University of Bath), Prof Laurence Hurst (University of Bath)

Diverse bacterial infections present a serious problem because they are difficult to treat. While pathogens are expected to differ between continents, countries and even hospitals we are currently facing an unforeseen challenge. Increasing number of studies report that bacteria are able to rapidly diversify within a single infection site within a single patient, and our work has identified life-history trade-offs as a possible cause. Trade-offs are ubiquitous at all scales of life and are defined as the fitness costs experienced by an organism when a beneficial change in one trait is linked to a detrimental change in another. In particular bacteria are constrained by a trade-off between multiplication and survival where evolutionary adaptation to stressors (including the immune system and antibiotics) comes at the cost of slow multiplication. While we have a plethora of evidence that such trade-offs exist we lack the mechanistic understanding of what causes them.

We have recently developed a mathematical theory that uses a systems approach to predict the form of the multiplication-survival trade-off in bacteria from the underlying physiological and metabolic properties of the bacterial cell. The proposed project will test this new theory using a combination of synthetic biology and evolutionary experiments. In particular we have synthetically engineered strains of Escherichia coli with fixed resource allocations, which will enable us to accurately measure trade-offs between bacterial survival and multiplication in different environments.

This work is fundamental to our understanding of diseases. It will allow us to predict if and how trade-offs emerge from intracellular processes, if and how they evolve, and what their impact is on pathogen diversity. This understanding will subsequently be applied to bacterial clinical isolates. Using the MiSeq we will characterize within-patient genome-wide pathogen diversity, and apply our knowledge of trade-offs to provide a mechanistic underpinning of any such observed diversity.