Search

For many cellular activities including escaping from toxins or acquiring nutrition, mobility is of significant importance. To mimic and study cellular motion, many strategies have been proposed. However, most reported motile systems cannot be applied to biological environments as either the moving objects or the environment media is not biocompatible, which confines their further applications. In this thesis, droplet-based cell-mimics were produced, of which the ability to acquire mobility and act as micro-reactors were separately demonstrated, shedding light in applying these cell-mimics to target delivery. Firstly, a polyethylene glycol/dextran aqueous two-phase system and liposomes were used to produce liposome-stabilized droplets, i.e., droplet-based cell-mimics. The morphology and stability of these droplets were then studied systematically. It shows that liposomes locate at the droplet surface to form a coating preventing droplet coalescence. The stability of emulsion droplets are both influenced by liposome concentration and size, and higher concentration/larger size leads to better stability. Secondly, a polymer gradient was introduced to the emulsion, which led to a concomitant interfacial tension gradient, inducing the Marangoni effect. Propelled by the Marangoni flow, droplets can achieve directional movement with the desorption of liposomes from droplet surface. These phenomena were explained by theoretical analysis. Finally, photo-responsive liposomes were incorporated into the emulsion system to achieve the controllable release of substrate molecules from liposome lumen to droplet lumen, which enables the enzymatic reactions to occur in a controllable way within droplets. This result proves that the emulsion droplets hold the potential to work as functional carriers for biomolecules.

One of the main drivers within the field of bottom-up synthetic biology is to develop artificial chemical machines, perhaps even living systems, that have programmable functionality. Bottom-up methods take an engineering approach to biology, with multiple downstream applications. Such applications that require very specific quantities of product e.g., antibodies, drug, or vaccines, necessitate the design of vesicles with total and precise control of a vesicle's molecular machinery. There exists a wide range of toolkits to produce and engineer artificial cells with an ever-increasing array of functionality and capability. However, little attention has been given to the quality control of vesicle production; so, to date, there is a paucity of techniques that are able to measure their molecular constituents precisely upon formation and with absolute quantification. The work outlined here presents the development of a microfluidic-based methodology that is able to characterise artificial cells at the single vesicle level with single-molecule resolution. High content fluorescence microscopy was used to assess the properties of Giant unilamellar vesicles (GUVs) and their statistics at the population level. Since this technique is semi-quantitative, the relative concentration of protein across the GUV population was determined however the absolute concentration of protein within each GUV was not. To overcome this, a lab-on-a-chip approach was taken to determine the precise number of biomolecules within each GUV and across the population of GUVs. The pulldown-5 array single-molecule high-throughput (PASH) chip was developed and capable of determining the encapsulation efficiency of protein within GUVs produced by phase transfer of an inverted emulsion. Using this approach, it was possible to determine the variability of the encapsulation efficiency between GUVs across a population as well as between different populations while testing batch-to-batch variation. The encapsulation efficiency measured to be 11.4 ± 6.2% across all tested parameters. This has consequences for the use of GUVs as precise biological models as well as for their development in a variety of biotherapeutic applications. To determine whether GUVs with specific concentrations of biomolecules could be produced by phase transfer, changes in the concentration of the seeding materials and reagents were investigated; while inefficiencies in encapsulation could be overcome, the variation in concentration could not. The results presented in this thesis help to understand the limitation of the phase transfer technique applied to systems biology research. In biological systems, measuring changes in gene expression at the transcriptomic and proteomic level is important. When used to develop simplified models of protein expression, these results indicate that vesicles produced using phase transfer may only be confidently used to produce 10-fold changes in encapsulant protein. It is possible to distinguish different GUV populations with a precision down to a two-fold change in encapsulant protein but with significant population overlap. The understanding of the limitations of encapsulation efficiency provides insight into the potential relevance of such systems for a variety of therapeutic applications such as smart drug delivery.

Antibiotic resistance (ABR) is a serious global health problem necessitating new bactericidal approaches as well as biosensors which can detect bacterial infection. Vesicular nanostructures have shown widespread promise for numerous biomedical applications in drug delivery and biosensing. A recently reported vesicle class composed of amphiphilic Janus dendrimers (AJDs), termed dendrimersomes (DSs), represent a promising system marrying the diverse chemistry of polymers and exact molecular definition of lipids. However, DSs have not yet been explored in depth for potential biomedical applications and, as yet, have not been used to construct nanoreactors - nanocompartments which encapsulate enzymes and permit reactions within a confined and localised space. Nanoreactors are of interest for myriad biomedical applications, some of which have been inspired by natural organelles. Herein, synthesis of a specific AJD was conducted and DS self-assembly investigated using various characterisation techniques revealing key properties of the DSs, particularly, their inherently semipermeable membranes, an essential property for use as a nanoreactor. Consequently, and inspired by the microbicidal function of the neutrophil phagosome, a DS-based nanoreactor was developed for an antibiotic-free bactericidal application. This was achieved by encapsulation of glucose oxidase and myeloperoxidase within DSs, enabling localised conversion of glucose to the highly potent microbicide, hypochlorite. This nanoreactor exerted a potent bactericidal effect against two bacterial pathogens on the WHO list for which new antibiotics are needed. Furthermore, a proof-of-concept is presented for harnessing the activity of membrane-lytic bacterial toxins to activate the nanoreactors through glucose release from separate lipid vesicles. Overall, the repertoire of potential DS biomedical applications has been expanded by this thesis and sets the basis for future nanoreactor applications. Lastly, again inspired by the activity of bacteria toxins, a liposome-based biosensor is presented towards the in vivo detection of Staphylococcus aureus lung infection using a colorimetric urinary readout. This system is composed of a liposome encapsulating renally clearable gold nanoclusters (AuNCs) with peroxidase-like activity. AuNC loading and in vitro release upon enzyme and toxin incubation was achieved and initial in vivo experiments performed to assess the utility of the system for detecting lung infection. Furthermore, initial work towards imparting an extra logic gate into the AuNC-liposome sensor for more specific disease detection is discussed. Overall, this thesis presents two vesicle-based systems towards treatment and detection of bacterial infection, which are in urgent need of alternative strategies due to the spread of ABR.

Surfactants are one of the most versatile chemical compounds as it is utilised in a wide range of products and processes such as pharmaceuticals, oil extractions, agrochemicals, food industry, personal care industry and household cleaning products. The use of high levels of surfactants in some of these applications, e.g. detergents, bring about several challenges including, product stability (shelf-life) and physiological safety. Countering these challenges leads to extensive research and development time, enduring high cost, which does not always result in a commercially viable product. The broad aim of this thesis is to address these challenges through understanding the molecular interactions within surfactant formulations involving low temperature stability and also understanding the interactions between these concentrated surfactant solutions with lipid membranes, in the context of ocular irritancy. The effect of changing organic solvent content on the Krafft point of the model surfactant sodium dodecyl sulfate (SDS) was determined through the surface tension pendant drop technique, which proved to be a robust method, and together with the sensitivity and accuracy of X-ray scattering and DSC techniques, the transitions near the solubility boundary of hydrated crystals to lyotropic liquid crystals (and vice-versa) were explored. This enabled SDS phase diagrams to be established for a series of water-glycerol mixtures, identifying and distinguishing between the numerous phases and the conditions under which they form. This knowledge base now provides a platform for understanding the performance of structured, surfactant-based systems, as well as controlling and fine-tuning the route of negative transformations in future studies. Concentrated surfactant formulations also have to comply with health hazard and safety regulations, particularly regarding ocular irritancy. There is a lack of ethical and reproducible eye irritancy test and a need for a better understanding of how surfactant systems interact with lipid membranes. Here, this understanding was gained through examining the permeation and solubility of model membrane, liposomes, by actual detergent formulations and later by surfactant(s) only systems. Through developing a liposome-calcein release assay (LCRA), a meaningful, negative correlation was established between the calcein release (liposome permeation) and corneal swelling, a parameter from the ICE test, suggesting that slow and incomplete solubilisation of lipid membranes leads to higher corneal swelling in the epithelial membrane. To gain a better insight, the three surfactants used in the detergent were tested in various ratios to analyse exclusively their interactions with the lipid bilayer, with the aid of the LCRA and real-time DLS analysis. It was gathered that NI follows the conventional three-stage solubilisation mechanism due to its structure consisting of a large headgroup to tail ratio, while the two anionic surfactants involved interact via a time-limiting flip-flop mechanism. It is proposed that the negative correlation between permeation and ICE test corneal swelling measurements, is due to this prolonged flip-flop mechanism. The LCRA developed provides a high-throughput and a less expensive framework, which shows great promise in developing novel, complimentary assays for predicting ocular irritancy.

Artificial cells (ACs) are biomimetic constructs that aim to reconstitute the functions and behaviours of living systems. Many recent developments have led to a variety of ACs of differing composition, however such structures often lack the ability to respond to their local environments or user-designed external cues. This is especially the case for lipid vesicles composed of phospholipid amphiphiles. To overcome this limitation, this research has focused on the development of lipid vesicles that can respond to elements of their environment. To achieve this, a molecular functionalisation strategy has been adopted using chemical functional groups, membrane biophysics and membrane protein (mechanosensitive channel) reconstitution to generate vesicles capable of triggered release in response to optical, thermal and enzymatic/ionic stimuli respectively. To illustrate the utility of mechanosensitive channel functionalisation, lipid vesicles have been designed that can respond to enzymatic (secretory phospholipase and protease) elements of prostate cancer microenvironments through protein-membrane-membrane protein (P1-M-P2) and protein- protein (P1-P2) interactions. Control of vesicle composition and mechanosensitive channel number enables modulation of triggered release across different biological microenvironments. Secondly, functionalised nanoscale vesicles have been used as synthetic 'organelles' in multi-compartment, nested giant vesicles. Such cell mimics can be generated in a modular approach via the emulsion phase transfer method. This has been used to create a light-responsive microreactor with the capability for user-controlled enzymatic catalysis, where irradiation time can be used to control the rate of catalysis. A second project has used calcium flux to control P1-M-P2 communication within nested vesicles, creating a signalling pathway that mimics various elements of biological signal transduction. These achievements highlight the flexibility of this modular construction approach, which can also incorporate elements of chemistry and nanotechnology challenging to introduce into living systems. These advances further the development of bottom-up synthetic biology, providing a framework for the design of increasingly biomimetic artificial cells for use in applications across biotechnology.

A critical driver in the development of single-cell analysis platforms has been the recognition that cellular heterogeneity is crucial to understanding disease. Single cell proteomics offer significant insights of cellular function; however, currently suffers from low-throughput. The work outlined here presents the development and application of several single-cell protein analysis systems. Each aim to address technological gaps regarding throughput, cell selectivity and amenability to processing samples directly from patients. To achieve higher-throughput, we have developed the CellWell platform, a high-density microwell array which can capture thousands of cells within minutes; however, posed challenges relating to the simultaneous lysis of these cells. We developed a facile method to produce surface microelectrodes to achieve single-cell lysis on-chip, but the demanding surface chemistry requirements imposed by the necessity to support simultaneously both the microelectrodes and single-molecule antibody microarrays proved difficult to overcome. Instead, we investigated how implementing semi-permeable hydrogel-based microwells could overcome these issues. To assay cells in patient blood samples with the CellWell, pre-processing is necessary. With a clinical setting in mind, it would be advantageous to process raw samples directly from patients with little or no off-chip pre-processing. To address this, we develop our methodology into the Hydrodynamic Trapping Centrifugal Release (HTCR) chip which is specifically designed to isolate cancer cells from patient liquid biopsies. The HTCR implements a method by which cells can be easily released from hydrodynamic traps and subsequently moved to isolated compartments. We conclude in validating the single-molecule single-cell method using fluorescence and immunofluorescence microscopy. While necessary to validate our single-molecule approach, we also show that the equivalence of these measurements of the steady-state distribution of protein abundance can be exploited to pave the way for absolute quantitation by fluorescence and immunofluorescence microscopy.

Over 80% of therapeutic monoclonal antibody (mAb) products are expressed in Chinese Hamster Ovary (CHO) cells. While cell-based expression platforms are traditionally used for the exploration of novel mAb therapeutics in the discovery phase, cell-based technologies remain particularly time- and resource intensive for mAb screening. To circumvent these shortcomings, cell-free protein synthesis (CFPS) platforms have emerged as versatile alternative allowing for rapid and flexible production of therapeutic proteins. In contrast to in vivo systems, CFPS platforms do not require intact host cells, which make them completely independent of host cell metabolism in determining product yield and quality. Although the open-production system bears several advantages over the traditional cell-based platforms, CFPS platforms still face limitations such as low product yields, challenges with post-translational modifications (PTMs), and poor cost efficiency of CFPS. It is therefore of great interest, to develop a simple, more cost-efficient and active CHO CFPS platform for accelerated screening of therapeutic mAbs. This work addresses the current limitations by developing a cell-free protein synthesis platform using CHO extracts for the rapid production and evaluation of industrially relevant mAbs. This study is divided into three main parts: 1) Development of a CHO CFPS system, 2) Yield optimization of CHO CFPS, 3) Production and screening of therapeutic mAbs. Our platform was able to express and characterise four functional mAbs in the supernatant fraction of 25 μl coupled batch reactions including two low in vivo expressers and the blockbuster drug Trastuzumab. Total synthesis yields were increased up to 50-fold by supplementing the system with two accessory proteins, GADD34 and K3L. Using our optimized platform, mAb yields of up to 31.06 μg/ml could be achieved. The trend in cell-free expressed functional mAbs replicated previously reported results from cell-based expression. Based on the success of our platform, we suggest its use for rapid, low-scale antibody expression screening to accelerate molecule selection and development for the biopharmaceutical industry.

The phase behaviour of aqueous dioctadecyl dimethylammonium chloride (DODAC) solutions were studied to model the stability of liquid fabric enhancers, when exposed to the environmental stresses of cooler climate regions during transportation or storage. These effects were studied using; SAXS and WAXS, 1H NMR, DOSY and polarised light microscopy, before and after freeze-thawing (FT) of the DODAC bilayers. The effects upon FT resulted in a FT induced swelling, where the magnitude of this swelling decreased with the addition of CaCl2 to the aqueous DODAC solution. Additionally, in the absence of CaCl2, the cooling properties of the bilayers resulted in a temperature shift of the Lα→Lβ' transition from ~31 to ~35 oC after FT, instead of the two transitions given at ~31 and ~35 oC in the presence of CaCl2 after FT. This suggests that CaCl2 decreases the changes to the DODAC bilayers and properties, upon FT. The DODAC aqueous solution is studied for the addition of glycerol and an ethoxylated alcohol surfactant. Glycerol caused swelling of the DODAC bilayers and the ethoxylated alcohol surfactant caused a lateral phase separation of a spherical to planar bilayers transition. These alterations to the DODAC bilayers, minimised the differences between the bilayers before and after FT. Also, it was shown that small unilamellar vesicles promoted the Lα→Lβ' transition temperature shift from ~31 to ~35 oC and a spherical→planar bilayers transition, after FT. The freezing behaviour of the DODAC bilayers were observed as a cooling induced dehydration which took place after the Lα→Lβ' phase transition instead of ice formation in the bulk water phase, at high pressure and low temperature. This resulted in a metastable freezing state with tilted bilayers, at ~ -2.0 oC which did not lead to a transition into the Lc phase until incubation at -18 oC. The effects upon thawing showed a lack of osmotic responsiveness between -5.0 and 15 oC, and an influx of water at 20 oC resulted in greater swelling of the DODAC bilayers than before FT.

Absolute quantification of cell-to-cell variability arising from the stochastic nature of gene expression is concealed within bulk measurement approaches. While microfabricated technologies for single-cell biology have revealed and advanced our understanding of this heterogeneity, limitations remain in transforming this progress for plant biology. Therefore, this thesis provides single plant cell proteomic profiling technologies to challenge conventional bulk cell analysis to further our knowledge in biological variability. Section one details the establishment, of the first known, miniaturised technology undertaking absolute protein quantification in single plant cells and organelles. Fabricated as an array of sub-nanoliter assay chambers containing micro-printed antibody spots, permitting multiplexed proteomic pull-down assays subsequent to the lysis of isolated protoplasts or organelles. Preliminary evaluation, employing a bi-molecular assay to assess fluorescent gene expression in protoplasts, presents novel experimental insight towards synergistic detection advantages through volume reduction coupled with single-molecule fluorescence microscopy. To demonstrate the quantification of unlabeled proteins a tri-molecular assay is developed for Rubisco, providing a new single-cell perspective towards Rubisco expression during mesophyll protoplast development. Section two challenges the restrictions of antibody affinity and experimental throughput, by merging techniques to produce novel technologies. Initially, a coverslip unifying pull-down assay within a microwell array was developed to overcome throughput limitations and is evaluated by quantifying p53 expression in mammalian cell lines. While the pull-down assay faithfully reproduces the precision in quantitative measurements concerning epi-fluorescence microscopy, a more relevant tool, a secondary standard for quantitative fluorescence microscopy is created. Through ratiometrically calibrating chloroplast autofluorescence against the absolute fluorescent protein content of three transgenic cell lines. These technologies permitted measurements, over six orders of magnitude for organelle, plant, and mammalian protein copy numbers, examined in Section three. Suggesting that unperturbed protein abundance does not scale with absolute expression noise, consequential to a linear relationship between the measured variance and mean.

Human hair is comprised of a highly complex biomaterial, α-keratin. Some of the key features of keratin are its robust nature and strong mechanical stability. These qualities originate from the molecular assembly of the proteins and are described by a two-phase model. Highly ordered crystalline intermediate filaments are interspersed within a less-ordered amorphous matrix. Developing a further understanding of the internal sub-structures of hair fibres, particularly with relation to the effect of chemical agents on these structures is imperative to the personal care and cosmetic industries. This information could direct the growth and expansion of new cosmetic treatments. Nuclear magnetic resonance experiments have been utilised as a method to measure and quantify "damage" to hair fibres caused by chemical modifications. X-ray scattering experiments were employed to investigate the effect of chemical agents on the sub-structures of hair. Equilibrium and time-resolved hydration experiments were used to ascertain both the lateral and axial swelling of the intermediate filaments and how "damage" affects this. Micro-focus x-ray scattering techniques provided information on the scattering patterns produced by the three principal hair structures; the cuticle, the cortex and the medulla. Some of the key findings of this research include the use of spin-lattice relaxation times as a robust and reproducible method for quantifying hair damage. Time-resolved hydration experiments demonstrated that both the magnitude and rate of swelling a fibre undergoes can be altered through chemical agents. The distribution of intermediate filaments within the cortex was measured for the first time by x-ray scattering, showing a clear increase in the density towards the centre of the fibres. These findings and use of the various 3 techniques begin to produce a tool-kit for measuring and quantifying the effect of chemical treatments on human hair fibres.

Cell membranes are a complex mixture of lipids and proteins, all of which vary in size, shape and composition. The ability for a cell to control its membrane, and regulate its contents, is in large part due to the regulation of local composition, curvature and tension. These changes to membrane properties can in turn alter the behaviour of membrane bound proteins. Understanding how these membranes, and their fundamental properties, can influence proteins in vitro is essential for building true synthetic cells, and building artificial cellular networks. In this thesis, work is presented which investigates the effect of membrane composition on the activity of the mechanosensitive channel of large conductance (MscL) from E.Coli. Using different model membrane systems, and many activation mechanisms of the protein, a greater understanding of MscL function has been gained. The first reconstitution, and activity measurement, of MscL into droplet interface bilayers (DIBs) is presented. Being able to measure MscL activity in the DIB model membrane has then lead to whole new avenues of research. The DIB system allowed for electrophysiological recording of single MscL channels, exploring the ability to see in-depth changes to MscL activity as DIB composition changes. Then, by linking multiple DIBs together, a network of bilayers connected by MscL channels has been presented. The first time a synthetic bilayer network linked by a gated ion channel has been demonstrated. Finally, a new device has been built allowing the generation of a measurable amount of shear force inside a DIB, and MscL activity was measured. It is hoped that the results of this thesis, will lead to further studies of more complex biological systems, deeper mechanistic studies of membrane proteins, and further synthetic networks of membrane protein interactions.

Stability determination of formulated products represents an ongoing challenge. These solutions have complex energy landscapes and experience a wide variety of conditions during manufacturing, transportation, storage and use. The mesostructure of such multicomponent mixtures is sensitive to external influences and time, which affects their performance in real-life applications. Understanding product metastability, in particular phase transitions such as crystallisation, is important both academically and practically. Current methods of stability determination are often time consuming and do not provide the causes of such phase changes. Here this challenge is addressed by focusing on a ubiquitous surfactant system: sodium dodecyl sulfate in water micellar solutions. A fundamental understanding of the system's crystallisation kinetics was obtained by developing and implementing a range of experimental and analytical techniques. Three different approaches were employed - the effects of both isothermal and linear temperature profiles were initially examined, before assessing the impact of various additives under both isothermal and linear cooling conditions. Lastly the consequence of different flow rates and flow types using microfluidics, including straight channel flow, oscillatory and push-pull, was examined and quantified. For the temperature profiles both an increase in isothermal hold temperature and lower cooling rates were found to decrease the rates of crystallisation, with the linear cooling data mapped onto the isothermal results. A wide variety of morphologies formed across the investigative window, which could be classified as the mono- and hemihydrate polymorphs. Additives with structures comparable to SDS were found to have potent effects on the crystallisation kinetics and morphologies of SDS even at the lowest concentrations, with their mode of action deemed to be kinetic rather than thermodynamic. Lastly, crystallisation in microfluidic devices was found to be sensitive to device type, temperature control and flow profiles. Overall this project has resulted in three first-author and two other publications.

Over recent years, there has been a growing interest in the field of artificial cell development, as they offer tremendous potentials across many fields of science and technology. A significant amount of progress has been made, particularly in the field of bottom-up artificial cells, due to the advancement in synthetic biology which has allowed us to assemble artificial cells from biological components systematically. The ability to sense and respond to their environment, in particular, has been the centre of much attention as it plays a key role in many important biological processes such as controlled release, signalling and communication between cells. However, most of the developments to date are largely limited to simple structures with a single function due to technological limitations. Areas such as artificial tissues which required high structural complexity and collective functionality have been largely unexplored. This thesis details the development of environment responsive artificial cells as well as tissue mimics in the form of water in oil droplets and aims to address these lack of platform technologies. We developed an optofluidic platform technology DROPLAY, capable of generating functional droplet networks with user-defined 2D structures. Instead of assembling the droplet network mechanically, a stem-cell-like approach was used. By incorporating a light-responsive mechanism into droplets, we have induced droplet differentiation to generate functional droplet networks using a laser-based setup. The results from this technology provide the foundation for future development of artificial tissues or other complex membrane structures. In addition, numerous light-responsive mechanisms, suitable for the construction of artificial cells, both chemical and biological have been investigated. In particular, we have showed and enhanced Bacteriorhodopsin (BR) activity in the acidification of lipid vesicles. Furthermore, a light-induced calcium controlled release system has been established and shown great potential. These light-responsive mechanisms can be incorporated in DROPLAY technology in the future or be introduced to other membrane structures for constructing more complex environment sensing artificial cells.

This work outlines efforts towards the development of immunosorbent assays for quantifying post-translationally modified p53, a key tumour suppressor protein, in single cancer cells. These assays utilised the microfluidic antibody capture (MAC) chip to selectively detect post-translationally modified p53 at single molecule sensitivity. Three assays for post-translationally modified p53 were developed; for acetylated-p53, ubiquitinated-p53 and phosphorylated-p53. The assay for acetylated p53, utilising click chemistry, was unsuccessful, as was an approach using antibodies. The assay for ubiquitinated p53 was successful in cell lysate, though could not be made successful in single cells. However, an assay for phosphorylated-p53 was successful and was multiplexed alongside a previously developed total p53 assay to elucidate the effects of a range of drugs on the phosphorylation and total p53 expression in single cells from immortalised cancer cell lines. This assay was then successfully applied, alongside the total p53 assay, to clinically relevant patient-derived xenograft material, the results of which were published in the peer-reviewed literature due to its novelty. This constituted one of the first examples of the measurement of a potentially clinically relevant biomarker in tumour derived material at single cell sensitivity. Further development of the MAC chip aimed at improving its throughput and range of accessible analytes could allow its use as a diagnostic or prognostic device when linked to clinically relevant outcomes such as remission, progression free survival, overall survival or measures of quality of life. This would be a significant contribution towards the personalisation of cancer treatment.

Investigations into characterising and measuring bio-membrane permeability have been ongoing for over a century. Driven by both industry and academia, a variety of in vivo, in vitro, and in silico techniques have been developed and employed to understand the mechanisms and thermodynamics of drug and agrochemical transport in living systems. Unfortunately, many of the techniques suffer from a lack of high-throughput implementation or suffer physical restrictions such as bulk diffusion limitations. Moreover, it is apparent that membrane permeability data is lacking, particularly with respect to the agricultural industry and non-mammalian biological sciences. With the invention of the droplet interface bilayer (DIB) a decade ago, there have been several breakthroughs in membrane technologies for electro-physiology, membrane protein reconstitution, and membrane permeability assay methods. In this thesis, I have studied DIBs as a possible candidate for a rapid and high-throughput membrane permeability assay that can be used to measure rates for speci c agrochemicals in varying lipid systems in situ. The rst output of this research includes device engineering, design and fabrication techniques of novel micro uidic chips for improved DIB formation, droplet rendering, and manipulation. More speci cally, on demand and high through-put DIB manufacture has been achieved for the rst time, the results of which has been published in the peer reviewed journal, Lab on a Chip. Furthermore, to prove that the application of DIBs are not categorically limited to a small subset of lipid types, it has been proven that DIBs can be formed with a variety of lipids, including some plant lipid extracts. For the rst time, DIB model membranes have been formed to mimic soy, Arabidopsis, tobacco, and oat plasma membranes. A successful permeability assay was performed with these DIBs, and the results were published in the peer reviewed journal, Biomicro uidics. A serious challenge of measuring membrane permeability in DIBs is the limitation of bulk diffusion, which often leads to underestimates in intrinsic membrane permeability rates. To further the understanding of this limitation, the uid dynamics of coupled advection-diffusion in stirred droplets has been investigated experimentally and computationally. As a result, a novel micro uidic device has been developed to induce shear stress along the membrane to disrupt the effects of the bulk uid stagnation in the permeability assay, which allows for more accurate measurements of the intrinsic membrane permeability. To the best of my knowledge, this is the most accurate technique available, and is a breakthrough tool for future applications, such as supplying permeability data to systems transport models. The results of the intrinsic membrane permeability of various lipid types have been published in the journal, Nature Scienti c Reports. Furthermore, the physical properties of DIBs have been investigated including surface energy driven morphology, formation dynamics, and bilayer surface tension measurements. For the rst time, the effect of membrane curvature in DIBs has been thoroughly scrutinized, and new insights into DIB behaviour have been established.

There is a growing body of evidence which suggests post-translational modifications occurring under oxidative stress (oxPTMs) play an important role in both human health and disease. The focus of the work described in this thesis has been on the use of coherent multidimensional spectroscopy (CMDS) to perform detection and quantification of oxPTMs in a label-free and non-destructive manner. Electron-Vibration-Vibration (EVV) two-dimensional infrared (2DIR) spectroscopy is a CMDS technique which is able to directly observe intra- and intermolecular interactions. As a result, EVV 2DIR spectroscopy is particularly useful for characterising (oxPTMs). EVV 2DIR spectroscopy employs one near-IR and two mid-IR picosecond excitation beams to probe vibrational couplings in a sample via a four-wave mixing process. This results in the spread of vibrational coupling information across two dimensions, which leads to spectral decongestion and the ability to directly analyse vibrational modes within complex molecules, such as proteins. Here, tyrosine (Tyr) nitration is used as a study model due to its importance in inflammatory diseases, amongst other pathologies. Results are presented for various nitration models and will demonstrate EVV 2DIR spectroscopy's ability to identify, relatively quantify and characterise the effect of nitration of tyrosine side-chains.

Microfluidic Antibody Capture (MAC) chips are small devices capable of quantifying biomarkers in single cells. These devices offer an all-optical approach for cell manipulation, lysis and single molecule quantification of a specific protein. This thesis details various developments to this device, both in terms of throughput and improvements to the single molecule counting process. The tumour suppressor protein p53 is a central hub for cellular stresses such as DNA damage, overproliferation and ribosomal biogenesis stress. Under stressed conditions p53 brings about the expression of a host of downstream effectors ultimately leading to DNA repair, temporary cell cycle arrest, senescence or apoptosis. The specifics of how p53 can lead to a number of different cell fate decisions are still unknown and require the development of quantitative biochemical techniques. In this thesis MAC chips are used to quantify p53 in single cells under a number of conditions. The chip data is used to create a quantitative model of p53 expression. This involved the use of stochastic simulation techniques such as the Gillespie algorithm and Approximate Bayesian Computation (ABC). These simulations determined that differences in p53 expression are best described as changes in the p53 degradation rate. This agrees with previous reports describing the p53-MDM2 relationship and its associated negative feedback loop. Lastly, attempts were made to obtain absolutely quantifiable data from the MAC chip platform. This involved calibrating the platform with known amounts of recombinant p53. By providing absolutely quantifiable data to the model of p53 expression the simulations could potentially provide real, biologically relevant parameters.

Keratins are one of the most abundant proteins found in mammals. They possess intrinsic properties that are as a result of their extensive and complex molecular architecture. They are physicochemically and biomechanically robust. Their complex molecular assembly is what confers them their stability. For this reason understanding their internal structure could yield essential information regarding their function and properties. In addition, quantifiable changes detected in the structure could give rise to new insights that could be exploited in the dermatologic and cosmetic industries. Their ubiquitous nature means that the fundamental biophysical understanding, could aid in the development of analogous protein-based fibrous biomaterials for a wide range of purposes. Equilibrium and time-resolved x-ray diffraction along with multinuclear natural abundance NMR (1H and 13C) techniques have been utilised to study the effect various modifications have on the structure of keratin fibres, and their associated proteins and lipids. The core structural intermediate filament units have been analysed using x-ray diffraction. The molecular mobilities of the keratin protein networks have been investigated using solid-state NMR. The intermediate filaments, which make up the internal scaffolding of keratin fibres, were found to be extremely robust and this was ascertained through lateral and axial swelling experiments. With a range of modification, the sensitive detection of protein reduction and denaturing was quantified. The hydration of lithium bromide modified keratin was found to achieve large degrees of lateral swelling without protein degradation at equilibrium. Some key findings of the research include the cooperative behaviour of the intermediate filaments and their constituent α-helical proteins, these behaved analogously upon hydration. This was observed through time-resolved hydration experiments which were carried out for the first time on human keratin fibres. In addition, the amorphous intracellular matrix within keratin fibres displayed interesting and varied polymorphic behaviour. Furthermore, the presence of two distinct lipid bilayer domains has been detected: an intracellular domain and a cuticle domain. These core findings may be applicable to other fibrous proteins, detection of distal pathological pathways and the development of analogous bio-materials.

This thesis presents the design, development and application of several platforms through which to generate giant vesicles for biophysical and mechanical membrane studies. There has been a growing focus on manufacturing model membrane systems with improved throughput and structural properties in recent years. GUVs are a popular model membrane system for studying lipid membrane-associated phenomena due to their inherent similarity to biological cells. Traditional methods to construct vesicles offer little control over nuanced membrane properties such as asymmetry and patterning, which has paved the way for more refined techniques to be developed. This thesis details the development of a microfluidic platform technology that addresses this chasm in sophisticated GUV fabrication strategies. The technique presented offers control over key structural features such as vesicle size dispersity, internal content, membrane composition and asymmetry. Vesicles were investigated using contour detection and fluctuation analysis in order to quantify the bending rigidity in membranes constructed by microfluidics for the very first time. Furthermore, the emulsion phase transfer (EPT) method was refined for the construction of GUVs with phase separated membranes across three of compositions. This is the first investigation concerning domain formation in membranes constructed from emulsion precursors at a range of a compositions. The progress made in advancing platform technologies opens up various avenues through which to further explore biophysical phenomena such as a lipid flip-flop dynamics, as well as for the high-throughput generation of artificial cell systems, with potential relevance for therapeutic applications such as smart drug delivery.

Allostery is the regulation of a protein's activity through a perturbation at a location distant from its active site. Such regulation is central to many biochemical processes. Targeting allosteric sites with drugs promises to allow fine-tuning of protein activity. However, proteins are complex systems composed of thousands of atoms interacting over multiple temporal and spatial scales. Direct observation of the non-equilibrium response of proteins to allosteric perturbations is still a major challenge. This limits our understanding of how the signal induced by the perturbation propagates across the protein and hampers our ability to predict the location of novel allosteric sites. Graph theory provides a way of representing proteins in a reduced form that still captures the full complexity of their underlying physico-chemical interactions. In this thesis, we develop a number of novel graph-theoretic methods for analysing allosteric behaviour. We start by constructing an atomistic, energy-weighted graph representation of a protein. We then use the behaviour of dynamic processes on this graph to explore how signals propagate within the protein. We use three distinct, but related methods. Markov stability identifies hierarchical community structure in the graph; Markov transients identifies anisotropic pathways of flow; and our bond-bond propensity measure quantifies the effect of instantaneous bond fluctuations propagating through the protein. These methods are applied to a number of biologically important allosteric proteins. Markov stability identifies dynamic coupling between the active and allosteric sites in caspase-1. The pathways involved in this coupling are revealed by combining a Markov transients analysis with computational mutagenesis. In caspase-1, CheY and h-Ras, the bond-bond propensity correctly predicts the location of the allosteric site and identifies key allosteric interactions. Evaluating the Markov transients and bond-bond propensity methods against a larger set of allosteric proteins, we demonstrate that these measures are good predictors of a site's allosteric propensity.

In recent years there has been an increasing interest in using lipid vesicles and related membrane structures as (i) artificial cells that mimic biological processes and (ii) bio-inspired micro-machines that serve functional purposes. To date, vesicles have largely been single-compartment structures with homogenous interiors, which has impeded the fulfilment of these goals. This thesis details the development of technologies to address this. We develop droplet-based methods to controllably generate multi-compartment vesicles (MCVs) for the first time. The potential of these novel structures as artificial cells capable of hosting a range of biological and bio-mimetic processes is explored. Most notably, we introduce spatial segregation of function, thus mimicking eukaryotic organelles, and incorporate an artificial enzymatic signalling cascade to transmit chemical signals between distinct vesicle regions. We also construct microfluidic devices to generate related structures known as multisomes. Microfluidic technologies enable the size of these constructs to be scaled-down (approaching characteristic cellular sizes), and the production throughput to be scaled-up (hundreds of multisomes produced a minute). We demonstrate their use as programmable modular microdroplet 'factories' for in situ chemical synthesis in physiological environments, with potential relevance for therapeutic applications. The above technologies provide a platform for further developments in bottom-up synthetic biology and in microreactor technologies, and will pave the way for the fulfilment of some of the ambitious goals of these fields.

This thesis reports on work performed in the development and application of an automated plate reading microscope implementing wide field time gated fluorescence lifetime imaging technology. High content analysis (HCA) imaging assays enabled by automated microscopy platforms allow hundreds of conditions to be tested in a single experiment. Though fluorescence lifetime imaging (FLIM) is established in life sciences applications as a method whereby quantitative information may be extracted from time-resolved fluorescence signals, FLIM has not been widely adopted in an HCA context. The FLIM plate reader developed throughout this PhD has been designed to allow HCA-FLIM experiments to be performed and has been demonstrated to be capable of recording multispectral, FLIM and bright field data from 600 fields of view in less than four hours. FLIM is commonly used as a means of reading out Förster resonance energy transfer (FRET) between fluorescent fusion proteins in cells. Using the FLIM plate reader to investigate large populations of cells per experimental condition without significant user input has allowed statistically significant results to be obtained in FRET experiments that present relatively small changes in mean fluorescent lifetime. This capability has been applied to investigations of FOXM1 SUMOylation in response to anthracycline treatment, and to studies of the spatiotemporal activation profiles of small GTPases. Furthermore, the FLIM plate reader allows FLIM-FRET to be applied to protein-protein interaction screening. The application of the instrument to screening RASSF proteins for interaction with MST1 is discussed. The FLIM plate reader was also configured to utilise ultraviolet excitation radiation and optimised for the measurement of autofluorescence lifetime for label-free assays of biological samples. Experiments investigating the autofluorescence lifetime of live cells under the influence of metabolic modulators are presented alongside the design considerations necessary when using UV excitation for HCA-FLIM.

Stress response systems are prevalent throughout all organisms with several functioning to maintain the cell envelope. One widely distributed system in bacteria is the Phage Shock Protein (Psp) response which is involved in pathogenicity, biofilm and persister cell formation. Induced under conditions proposed to cause membrane (often Inner-Membrane, IM) stress, the Psp response appears to stabilize the IM and so prevent dissipation of the proton motive force. The central component, PspA, is a peripheral IM protein that acts as both the effector and negative regulator of the Psp response. PspA has a counterpart Vipp1, which functions for chloroplast envelope maintenance and thylakoid biogenesis in plants, algae and photosynthetic bacteria. Mechanistic insight into how PspA and Vipp1 undertake their respective effector functions is limited but thought to be through their direct interactions with cellular membranes. Rigorously controlled, in vitro methodologies with lipid vesicles, purified proteins and peptides were established and used in this study, providing the first biochemical and biophysical characterisation of membrane binding by PspA and Vipp1. Direct membrane association of PspA and Vipp1 was shown to occur through their conserved N-terminal amphipathic helices. Both proteins are found to sense stored curvature elastic (SCE) stress and anionic lipids within the membrane. PspA has enhanced sensitivity for SCE stress while Vipp1 partitioning is most sensitive to membranes with a high net-negative charge. Experimental data points to alleviation of SCE stress by Amphiphatic Helix (AH) insertions as an attractive mechanism for membrane maintenance by PspA and Vipp1. Furthermore, by probing PspA's regulatory role we show that its transcription inhibition, though binding to the transcription activator PspF, can be relieved upon bilayer exposure in a SCE stress specific manner. The identification of a physical, stress related membrane signal suggests a unilateral mechanism that promotes both membrane binding of PspA and a stress triggered induction of the Psp response.

This thesis discusses the development of time resolved fluorescence imaging techniques and their use in the study of cellular signalling pathways, in particular the ability to perform multiplexed imaging of a number of pathways in live cells. These techniques are applied to investigate chemotaxis, the ability of cells to migrate directionally in response to a chemoattractant gradient, which requires precise spatiotemporal coordination of signalling events. Fluorescence lifetime imaging (FLIM) is widely applied to obtain quantitative information from fluorescence signals, particularly using Förster Resonant Energy Transfer (FRET) biosensors to map protein-protein interactions in live cells. The development of a software tool for the global analysis of large FLIM datasets is presented which allows simultaneous analysis of hundreds of FLIM images in minutes and the use of complex models, for example a four-exponential model of an ECFP FRET system, with relatively low photon-count data. Live cell imaging with optimised FRET biosensors is used to investigate the role of Phospholipase C epsilon (PLCε) in fibroblast chemotaxis. It is demonstrated that PLCε-null fibroblasts show a compromised chemotactic response to platelet derived growth factor and spatial defects in Rac1 activation and phosphoinositide signalling. The ability to image multiple functional reporters simultaneously in a single cell is desirable when investigating complex signalling networks with significant cross-talk such as chemotaxis. A number of approaches for multiplexed measurements are investigated, in particular using homo-FRET between two spectrally identical fluorophores, which presents a promising approach to reduce the spectral bandwidth compared to conventional hetero-FRET biosensors. The optimisation and automation of a to perform multiplexed time-resolved fluorescence anisotropy imaging of homo-FRET biosensors is discussed. The development and multiplexed imaging of homo-FRET reporters for phosphoinositide signalling using a polarisation resolved confocal time correlated single photon counting (TCSPC) microscope is presented. Potential approaches for multiplexed imaging three functional reporters are discussed.

Positron emission tomography (PET) is increasingly being used by the pharmaceutical industry in drug development. Drugs are designed to bind to a specific target, which is usually a membrane embedded receptor or an enzyme. PET is used to establish the utility of radiolabelled drugs and other radioligands in vivo before embarking on expensive clinical trials. In developing new PET radioligands, a common reason for candidate rejection is that the non-specific binding signal obscures the specific binding signal and, thus reduces the quality of the PET scan data. A high non-specific/specific binding ratio is a major reason for radiotracer failure and there is no completely satisfactory predictor of its magnitude. Non-specific binding is a poorly understood phenomenon but is believed to be related to the binding of labelled molecules to tissue membranes. A series of Spiperone analogues have been synthesised and tested in a variety of biophysical assays to increase the understanding of the molecular basis of drug-lipid interactions. Rapid cyclic voltammetry experiments were also performed to examine the interaction of drugs with a DOPC monolayer on a Pt/Hg electrode. The Spiperone analogues were tested in this system and the larger molecules were shown to have a larger interaction with the DOPC monolayer. A set of well characterised central nervous system drugs were also tested in this system and it was found that the strength of their interaction has a strong correlation with the measured in vivo non-specific binding. This technique could therefore be used to screen candidate drugs and radioligands to predict their non-specific binding. An assay that can select compounds that display the greatest likelihood of success would be extremely valuable due to the high cost and low-throughput nature of PET imaging.

This thesis presents the design, development and application of a number of platform technologies for the study of lipid-phase behaviour and bio-membrane mechanics. This includes an automated laboratory based X-ray beamline with a multi-capillary sample chamber capable of undertaking small angle X-ray scattering measurements on a maximum of 104 samples at a time as a function of temperature between 5 and 85 C. The modular format of the system enables the user to simultaneously equilibrate samples at eight different temperatures with an accuracy of +/-0.005 C. This system couples a rotating anode generator and 2-D optoelectronic detector with Franks X-ray optics, leading to typical exposure times of less than ve minutes for lyotropic liquid crystalline samples. Beamline control including sample exchange and data acquisition has been fully automated via a custom designed LabVIEW framework. In addition this thesis presents an overview of the development of a suite of tools for undertaking fluctuation analysis measurements of lipid vesicles under a variety of conditions including as a function of hydrostatic pressure. These and other biophysical techniques have been used to study a variety of binary lipidic systems determining key parameters ranging from spontaneous curvature and bending rigidity through to p-T dependent phase behaviour.

Biological membranes are a dynamic and complex mixture of lipids and proteins, all of which vary in size, shape and composition. The ability of cells to compartmentalise and regulate their contents is in large part due to the regulation of membrane stress and composition. This in turn can alter the behaviour of proteins associated with the membrane. For this reason, understanding membranes and how their fundamental properties can influence proteins in vitro is essential for improved in vivo knowledge, and is also of great interest to the pharmaceutical industry. In this thesis work is presented which investigates the effects of membrane stress on the mechanosensitive channel of large conductance (MscL) from E. Coli using a fluorescence based assay. Membrane compositional sweeps were used to control the stress and to prove that changes in membrane stress directly affect MscL activity. Corresponding X Ray scattering and solid state NMR measurements probed the fundamental properties of the membranes used during the MscL activity assay sweeps. Not only were buffers found to have a significant effect on the membranes but a unique result from the solid state NMR data showed that for a specific lipid composition the membranes were more susceptible to deformation within the magnetic field. This result correlates directly with a change in activity observed in the MscL activity assay. This is to the best of our knowledge the first direct correlation between MscL activity and a fundamental membrane parameter. One key result of this thesis is the applicability of the assay and the technology to other systems, lipids and proteins. We present preliminary data for the translation of this assay into droplet interface bilayer (DIB) format, thereby creating a platform technology enabling the study of the effects of membrane asymmetry. This assay is already being adapted for use by other researchers.

Lipids in a bilayer determine the stresses and the topology of the membrane they form. An understanding of the link between lipid composition and biomechanical parameters such as the spontaneous curvature and bending rigidity is key to elucidate the mechanisms behind protein‐membrane interactions. Despite their relatively low abundance in‐vivo, one of the most important class of lipids involved in signalling cascades in cells are the phosphatidylinositols (PIs) lipids. Experimental studies of the effect of phosphatidylinositol lipids upon model and cellular membrane systems remain in their infancy and have not matched the pace of discovery with respect to their role in regulating key cellular processes. Synthesised at the endoplasmic reticulum, one route for the distribution of the PIs to other organelle membranes is via translocation by the phosphatidylinositol transfer proteins (PITP). In order to study the link between membrane composition and PITP function, these proteins have been assayed for their interaction and binding with model membranes containing differing amounts of PI. Corresponding studies of the phase behaviour of these systems have been conducted using Small Angle X‐ray scattering specifically investigating the influence PI has in a bilayer membrane formed by dioleoylphosphatidylcholine, as well as in an inverted hexagonal phase formed by dioleoyl‐phosphatidylethanolamine. Additionally, a novel platform based upon a BODIPY fluorescent probe is presented, which is able to sense the stored stresses within lipid bilayers, and whose measurements are correlated these with the make‐up of membranes.

We have designed and developed a novel form of biosensor with implications to both the understanding of the mechanistics of the cell function and as a biomedical diagnostic tool. As part of a platform of technologies orientated to single cell proteomics, this project focuses on the development of single molecule microarrays (SMM), which is intended to produce proteome snap shots of individual cells. The single cell proteomics platform includes technologies for microfluidics based live cell analysis, manipulation, lysis and analysis of the proteome of a single cell. Analysis of single cells to high and detailed resolution will provide both quantitative and qualitative information on discrete events and protein dynamics, usually overlooked through ensemble measurements of large populations of inhomogeneous cells. With the aim of developing a quantitative and qualitative tool oriented to single cell proteomics; we present an antibody microarray capable of simultaneous quantification of multiple proteins to the single molecule level. The principle enhancements of design enabling the capabilities described are based around single molecule detection of fluorescently labelled analytes bound to affinity patches of antibodies; the affinity patches are limited to the microscope field of view and incorporated into microfluidic devices, detection is performed with a single resolution through the use of total internal reflection fluorescence (TIRF) microscopy. The cell is handled and lysed in a microfluidic device to minimize dilution of the sample, so that the greater majority of analyte is bound to the sensing surface at equilibrium and so detectable at the surface through TIRF evanescent wave. This technology also bears significance for the detection of many forms of analyte, such as glycan profiling, direct mRNA quantification and the analysis of post translational modification (PTM). The SMM produces rapid snap shots of a proteome; this makes it suitable for rapidly changing and dynamic cell signalling cascades, of which PTM makes a significant contribution to cell fate. We present here the basic principles of design and the initial testing of the design and current setup. Also discussed are intensions and capabilities for the technology, as well inevitable potential pitfalls and intrinsic limitations.

The transport of compounds around the body has been a topic of interest for many years, and the advent of non-invasive biological imaging in living tissue has made huge advances in the characterisation and localisation of cellular receptors for use in drug targeting. However, there remains a significant paucity of knowledge regarding how the majority of drug molecules are transported about the body, when they often exhibit negligible aqueous solubility and the body expresses no trans-membrane pumps or chaperone proteins that recognise them and facilitate their movement. This leads to large attrition rates in drug discovery programmes, as compounds with high binding constants or inhibitive activity in vitro fail to perform in vivo, due to poor bioavailability or non-specific sequestration away from the tissue of interest. In this study, the interactions between a number of drug and lipid molecules were investigated and the effects upon both the lipids’ chemical and bulk membrane structures were analysed. This revealed some of the mechanistic causes of the previously observed hydrolytic activity a number of common drug compounds exhibit toward lipid membranes and identified parameters affecting the observed rates of reaction. The findings also suggest approaches by which this behaviour might be predicted, or even tuned to deliver optimum pharmacological characteristics.

This thesis describes the development of integrated microfluidic technology for single cell proteomic analysis, focusing on circulating tumour cells (CTCs). While single cell proteomic analysis has wide applicability across biology and medicine, CTCs form an ideal first application. Circulating tumour cells are intimately involved in metastasis, the step in cancer overwhelmingly responsible for death, yet have proved hard to study. Single cell microfluidic technology is ideal first because the quantity of material available is inherently at the level of a few cells and second because cell to cell variation is of great interest. Chapter 1 is an introduction to the field. In chapter 2 a microfluidic sandwich assay for quantification of protein at the single cell level is described. In chapter 3 the isolation of CTCs in a microfluidic device is described. This relies on taking the output of the CellSearch® system and inputing it to a microfluidic device. While CTCs were identified, the result showed that a more systematic approach is required for counting and integration with the single cell assay previously described. Chapters 4 and 5 describe development of technology suitable for counting and isolation of CTCs integrated into a microfluidic device with single cell proteomic analysis, although the work done here makes use of fluorescently labelled beads and model cell lines rather than CTCs from patient samples. Chapter 4 describes microfluidic cytometry that can be used to count and identify a labelled population of cells, such as stained CTCs. Chapter 5 describes the prelimary development of a sorting system suitable for isolation of CTCs integrated with the cytometer.

The current set of three studies comprises preparative synthetic organic chemistry, membrane biophysics, and enzymology. Specifically, racemic and chiral synthetic organic preparations of distearoylphosphatidylinositol-4-phosphate (DSPIP) are described, along with tangential studies that inform use of protecting group strategies and the limits and behaviour of key myo-inositol-based intermediates. The biophysical behaviour of this lipid as a function of temperature, pressure, and concentration in dioleoyl phosphatidylcholine and hydration has been examined. Unsaturated analogues of DSPIP, comprising oleoyl (SOPIP) and γ-linolenoyl (SGPIP) have also been prepared, both racemically and chirally. The Enzymology study comprised kinetic assays of the three enantio-pure lipids DSPIP, SOPIP and SGPIP prepared in the organic study with Salmonella (bacterial) phosphatase SopB and has established that increasing numbers of unsaturated bonds in the glyceride residue of the inositide has a decreasing effect on activity of the enzyme and are in agreement with previous preliminary results for this enzyme. This body of work provides further evidence that the sensitivity of biological systems to fatty acids is primarily physical.

Rab proteins are the largest subfamily of the Ras superfamily of small GTPases, with more than 60 known members, that are involved in a multitude of different processes regulating membrane traffic. Rab proteins cycle between the cytosol and association with membranes, whereby each Rab exhibits a characteristic and specific subcellular localisation. It remains obscure how Rab proteins, in spite of high sequence and structure identity, distinguish between different membranes in the cell with such specificity. Membrane biophysical properties, such as stored curvature elastic stress and bending rigidity, are increasingly found to be determinants for protein recruitment and activity, and other Ras related proteins have recently been shown to exhibit sensitivity towards lipid species and elastic membrane properties. In this study Rab membrane binding is for the first time correlated to membrane bending rigidity, suggesting that biophysical properties of lipid membranes may play a role in the regulation of Rab targeting. Furthermore, all Rab proteins tested were observed to bind membranes in the absence of other protein factors, questioning the function of protein targeting factors for the Rab membrane recruitment process. Another aspect of Rab membrane interaction is Rab extraction from membranes by GDI. A large scale in vitro screening of 17 Rab proteins revealed a broad range of extractability from membranes with GDI. No correlation was found between extractability and the C-terminal prenylation motif, and no difference in extractability was observed in direct comparison of the extraction potential with GDIα and β. However, Rab proteins that exhibited low extractability from membranes are involved in secretory processes, suggesting a functional correlation to extractability. Furthermore, Rab40c as the first mammalian Rab protein to date was shown to be palmitoylated.

Understanding how drug molecules interact with our body and what effects they may induce as a result is of fundamental concern to the pharmaceutical industry. Crucially we want to use this knowledge to our advantage during the drug discovery process in order to manipulate a drug’s efficacy in vivo - therefore the development of new technologies, able to effectively screen numerous desirable drug-membrane interactions, is of key importance. The first half of this thesis details the development of a vesicle leakage assay, as a means to assess the effect of cationic amphiphilic drugs (CADs) on model lipid membranes. Having demonstrated the reproducibility of the assay, the assay was transferred into a microfluidic format where water-in-oil droplet systems act as individual experimental vessels. As such, it has been demonstrated that the use of fluorescence lifetime techniques can provide a way in which to translate this assay into a high-throughput format. The second part of the thesis is concerned with using droplet interface bilayers (DIBs) as a means to probe the effect of exogenous species upon a single lipid bilayer, as opposed to bulk vesicle populations. Several advantages exist for using such a system, including for example the ease with which one is able to control the composition of the aqueous compartments on either side of the bilayer and to form asymmetric bilayers. An assay, involving the use of pH gradients, is detailed, where proof-of-concept experiments illustrate that pH sensitive dyes could be used to report the extent to which lipid bilayers are perturbed by drug molecules for example. Furthermore, a novel automated approach has been developed, offering advantages over the manual manipulation of lipid-containing droplets for DIB formation, where a microfluidic approach is used for their generation in high-throughput. Consequently, this approach enables the formation of multiple DIBs, where the composition may be differed and the droplet dimensions controlled, enabling the formation of DIB networks that can be arranged in either two- or three-dimensions.

Phosphoinositide-specific phospholipase C (PLC) is an intensively studied family of enzymes constituting a junction between trans-membrane signal transduction processes and phosphoinositide lipid signalling. PLCs are activated in response to stimulation of cell surface receptors at the plasma membrane, and the signals are carried downstream by other transducers. PLCs catalyse the hydrolysis of phosphatidylinositol 4,5-bisphosphate [PtdIns(4,5)P2] to diacylglycerol and inositol 1,4,5-trisphosphate, which are both well known intracellular second messengers. This study mainly focused on the PLCε sub-family which is closely linked to the Ras oncogene and may play a role in tumorigenesis and development. The functions and regulatory mechanism of PLCε are not yet understood in detail. To address these issues activity and structural studies were performed. Activity studies were carried out in vivo using cell lines and in vitro using lipid vesicles in a model system. The latter was designed to study protein-protein and lipid-protein interactions using PLC purified to homogeneity and guanosine triphosphatases (GTPases) prenylated in vitro. Evidence was found for a direct interaction between the GTPases and the PLC that mediated activation of the phospholipase. Furthermore, the correlation between PLC activity and substrate presentation in lipid vesicles of various sizes and lipid compositions was analysed. For the first time, PLC activity was found to depend upon the electrostatic potential and the stored elastic curvature stress of the lipid bilayer of the vesicles. The binding and the activation process between GTPase (using H-Ras) and PLCε was also investigated at a molecular level in vitro. Functional studies were carried out using Förster resonance energy transfer (FRET) to determine if PLCε undergoes a conformational change upon H-Ras binding. This would distinguish whether conformational change or translocation of PLCε to the membrane interface (where GTPases are localised) is most likely to be the key event during PLCε activation; no conformational change was observed. Electron crystallographic structural studies, in which two-dimensional protein crystals are grown on a lipid monolayer followed by electron microscopy, were attempted. The aim was to retrieve structural information in a functional state that resembles the natural one. Protein and lipid monolayer compositions (lipid proportions, lipid concentration, protein concentration and incubation time) were screened to identify conditions where specific protein-lipid interaction would favour twodimensional crystal formation. The protein was expressed with a His6-tag that allowed specific binding to nickel chelating lipids included in the lipid monolayer. In addition, catalytically inactive PLCε mutants were generated and their ability to bind PtdIns(4,5)P2, and thereby to drive the crystallisation process, was investigated. Conditions that led to protein-lipid binding, but not to two-dimensional crystallisation, were identified.

In the era of proteomics, new technologies in separation and identification are required. Separation methods, such as capillary electrophoresis or liquid chromatography, are a crucial part of high throughput proteomic workflows. In this thesis, novel approaches to proteomics using capillary electrophoresis are presented. A platform of technologies based on capillary electrophoresis with continuous deposition of separated proteins onto metallic substrates enables subsequent analyses and identification. Since sample deposition and identification are decoupled, multiple readout techniques can be explored. Readout techniques used include matrix assisted laser desorption/ionisation mass spectrometry (MALDI-MS), electron-vibration-vibration two dimensional infrared spectroscopy (EVV 2DIR) and fluorescence microscopy. This technology was used without the deposition interface, to achieve advances in ribosomal separations or with the deposition interface, to develop new proteomic strategies of separation and readout. The eukaryotic ribosomal proteins were separated using capillary electrophoresis for the first time. Over 26 peaks were resolved in less than 10 minutes. An outstanding RSD migration time of < 0.5% was achieved, demonstrating that the readout could provide a ribosomal ' fingerprint'. Separations of proteins were successfully analysed using a standard MALDIMS instrument. This work was advanced by the offline coupling of CE to MALDI-imaging and applied to the ribosomal proteins to demonstrate a novel workflow from cell culture to protein identification. Quantitative analysis of protein levels is an important part of proteomics, but is difficult to achieve using mainstream technologies with high throughput and accuracy. EVV 2DIR is a non-linear spectroscopy which is able to achieve absolute quantification of proteins.[1] Coupling of EVV 2DIR to CE (CE- 2DIR) was demonstrated through the deposition and analysis of peptide and proteins. CE-2DIR offers great promise as a new proteomic tool.

Synaptojanin is unique in that it is the only mammalian inositol phosphatase to contain more than one catalytic domain; a CX5R Suppressor of Actin (Sac) domain and an Inositol Polyphosphate Phosphatase Catalytic (IPPc) domain. The enzyme has been shown to play a crucial role in synaptic vesicle recycling and its functioning has been implicated in the onset of Alzheimer’s disease. Many domains and sub-units in bi- and multi-functional enzymes are found to operate in a co-dependent manner. In this work, the possibility that the functioning of the Sac and IPPc domains in Synaptojanin are co-dependent was investigated. The kinetic parameters of the IPPc domain were assessed in a double Sac/IPPc phosphatase and compared to that in both a double phosphatase where the Sac activity had been rendered inactive and a single IPPc phosphatase. It was found that the VMax activity of the IPPc domain towards its lipid substrate PI(4,5)P2 is significantly lower when the Sac domain is not present or functional. Likewise, it was found that the VMax activity of the Sac domain towards PI(4)P is reduced when the IPPc domain is removed or its activity rendered inactive. Interfacial recognition and substrate channelling were investigated as mechanisms to explain the domain dependency. However, they were found not to contribute to the observed differences in VMax. Instead, it seems likely that protein-lipid interactions induce the dependency. In summary, this thesis presents the first evidence that the catalytic domains in Synaptojanin act in a co-operative manner and probes the mechanism by which the domains interact.

SNARE (Soluble NSF (N-ethylmaleimide Sensitive Fusion) Attachment Protein Receptors) proteins have been linked to the membrane fusion mechanism since 1993 as fusion proteins and have been suggested to be the minimal machinery. The complexity of the fusion process means that many questions remain unanswered as to how SNARE proteins perform their role. The most favoured model (the stalk model) does not involve proteins directly and so the influence of the SNARE proteins on lipid properties is of interest. In this thesis, work is presented which investigates how these proteins may manipulate membrane properties in order to promote fusion. Purified proteins solutions of His6-VAMP2 (Vesicle Associated Membrane Protein 2), His6-SNAP-25 (Synaptosomal-associated Protein 25) and a truncated form of Syntaxin 1A (His6-ΔN-Syx 1A, w.t aa 181-288) were obtained following bacterial over-expression. Fluorescent versions of His6-VAMP2 and His6-ΔN-Syx 1A were produced by the addition of cysteine residues to the C-terminus followed by labelling using Alexa Fluor® 488-C5-maleimide and Alexa Fluor® 555-C2-maleimide respectively. These fluorescent proteins were used to establish that the purified protein inserted into model lipid bilayers. The effect of SNARE protein incorporation on the relaxed curvature of bilayers was explored by examining giant unilamellar vesicles grown using electroformation. Bilayers containing either 1:300 His6-VAMP2: DOPC or 1:1:600 His6-SNAP-25: His6-ΔN-Syx 1A: DOPC were smaller than pure DOPC vesicles, indicating that SNARE proteins increase the relaxed curvature of the bilayer. Analysis of these vesicles by micropipette aspiration suggested that VAMP2 lowered the bending rigidity of the membrane and a reduction in the area expansion modulus relative to the pure lipid bilayer was found. The t-SNARE sample also indicated a reduction in bending rigidity but the area expansion modulus was found to increase. These latter results are thought to be due to the formation of protein aggregates. Lipid mixing assays were conducted to investigate how changes in the properties of liposome bilayers affected fusion rates. It was found that the addition of DOPE to DOPC bilayers increased the rate of hemifusion and this was also found for cholesterol addition, suggesting both components are fusogens. The rate of hemifusion rose continually upon DOPE addition but reached a plateau in the cholesterol study shortly after 10 mol%. Despite this, the fusion rates for the cholesterol study were generally higher than the same mol% DOPE added. The changes in fusion rates observed have been explained by considering the impact of the additives on the free energy and stored curvature elastic stress of membranes as well as the change in the energy of formation of intermediate structures. From the findings of this thesis it is proposed that the SNARE proteins are able to soften the membranes in which they reside. This allows the membrane to be deformed with less energy input. The strength of the SNARE complex and the force applied to the membranes during its formation increases membrane tension and reduces inter-membrane separation; promoting hemfusion. Following the action potential of the neuron it is proposed that a conformational change occurs in the synaptic SNARE complex, pulling on the hemifusion diaphragm and inducing the formation of a fusion pore.