The artificial cell: biology-inspired compartmentalisation of chemical function

The endocytic pathway is the gateway between cells and their environment, and promotes diverse events such as nutrient uptake and cell migration. Membrane proteins (“cargoes”) enter endocytic vesicles and reach the endosome. From here they are recycled to the surface, or sorted to intralumenal vesicles (ILVs) to generate a multivesicular body (MVB) en route to the lysosome. The generation of transport intermediates at all points of the endocytic pathway involves the ordered recruitment of an array of membrane-sculpting adaptor proteins, which recognise cargo and coordinate the local induction of membrane curvature. The impairment of these adaptor proteins, or their defective regulation, gives rise to errors in compartmentalisation. One class of endocytic cargo that is transported to the MVB is exemplified by the epidermal growth factor receptor (EGFR). For these cargoes, the sorting signal is generated by receptor ubiquitination, a reversible post-translational modification. A broad range of ubiquitin-binding adaptor proteins recognise ubiquitinated cargoes and promote their endocytic sorting. These adaptors themselves are regulated by reversible ubiquitination, generating a network of ubiquitin homeostasis that governs endocytic trafficking. The key regulators of this network, ubiquitin ligases and deubiquitinating enzymes, must also coordinate their functions with membrane curvature.

Autophagy is a pivotal recycling pathway that operates in all eukaryotic cells. More than 40 AuTophaGy related (Atg) proteins are known in yeast and many of them are conserved in humans. In non-stressed, vegetative cells, autophagy mainly recycles unwanted or damaged cytoplasmic components by exclusively engulfing them into autophagic membrane sacks, termed phagophores. In stressed cells, however, autophagy loses its selectivity and degrades bulk cytoplasm. The regulation of this dramatic loss in selectivity is intimately intertwined with the onset of human diseases such as neurodegeneration or cancer. Its molecular bases remained, however, not well understood. Through a combination of classical biochemistry and cell biology with cutting edge in vitro reconstitutions, Dr Wollert and colleagues have been able to solve the long standing question how selectivity is regulated at a molecular level in yeast. They found that the decision is made at the earliest step in autophagy, i.e. nucleation of the phagophore.

Diffusion of biomolecules in both the cytosol and in the plasma membrane are not simple random walks, instead the measured diffusivity depends on the timescale of observation. Our understanding of why and how this anomalous diffusion exists remains incomplete, and the presence of slower-moving obstacles, pinning sites, and compartmentalisation have all been suggested as potential contributors to anomalous diffusion. Ultimately, this confinement and compartmentalisation must affect the reaction rates of biomolecules, and understanding and exploiting these effects would be a useful tool in designing artificial cellular systems. Here Mark Wallace will present his group’s recent work to build artificial model membranes with defined nanoscale topographic and topological properties to control anomalous behaviour. Mark exploits a combination of single-molecule Total Internal Reflection Fluorescence and Interferometric Scattering microscopies to characterise diffusion using single-particle tracking that spans over four orders of magnitude in time.

Eukariotic cells are highly compartmentalized to achieve local special functions, but nevertheless have to exchange lipids and proteins between compartments and with the exterior of the cell. Thus, traffic intermediates (vesicles, tubules) have to be formed and detached from donor membranes to be transported and deliver their cargos and membrane to an acceptor compartment. Cell compartmentalization thus works hand-in-hand with membrane scission to allow for these exchanges. Reconstituted membrane systems have been instrumental to reveal mechanisms that drive membrane scission of bud neck in vivo, such as constriction by the ATPase dynamin involved in endocytosis or by line tension at the edge of membrane domains. Reconstitution approaches have similarly allowed recently to reveal a new scission mechanism, not based on constriction but on the friction between a protein scaffold and the membrane that produces a viscoelastic-like response and a scission when membrane tension exceeds lysis tension: Friction-Driven Scission (FDS). Professor Bassereau will show how this mechanism was shown using a combination of reconstituted systems and theoretical modelling.

Compartmentalization is generally regarded as one of the key prerequisites for life. In living cells, not only the cell itself is a compartment, with its properties controlled by the semipermeable cell membrane, but also the organelles play a crucial role in protecting and controlling biological processes. To better understand the role of compartmentalization, there is clear need for model systems that can be adapted in a highly controlled fashion, and in which life-like properties can be installed. Polymer-based compartments are robust and chemically versatile, and as such are a useful platform for the development of life-like compartments. In this contribution both an artificial organelle and cell system will be discussed. The artificial organelles are composed of biodegradable amphiphilic block copolymers that self-assemble into vesicular structures. These so-called polymersomes are loaded with enzymes and are semi-permeable for small molecule substrates. Upon introduction in living cells, they affect metabolic pathways as artificial organelles. A different type of polymersome is created via a shape change process in which a bowl-shaped structure is obtained. Within the cavity of the bowl enzymes are loaded which provide the nanostructure with motility upon conversion of chemical energy into kinetic energy. The synthetic cell platform is composed of a complex polymer coacervate, stabilized by a biodegradable block copolymer. The specific feature of the polymer membrane is its semipermeable character. Enzymes inside the protocell can therefore still be reached by their substrates, and small molecule products can be excreted. This allows protocell communication with this robust synthetic platform.

Magnetic nanoparticles (MNPs) have many applications, particularly over a range of emerging biomedicine (diagnostic and therapeutic). For such applications the MNPs are required to be biocompatible and have a mono-disperse size and shape distribution to ensure their magnetic behaviour is consistent, which can require chemical synthesis under harsh and toxic conditions, which may compromise biocompatibility. Magnet bacteria synthesis nanoparticles of magnetite within lipid vesicles (known as magnetosomes) within their cells. Magnetosomes show great potential for such biomedical applications, however, magnetosomes themselves may not be ideal with respect to economy of production and adaptability for different specifications. Sarah Staniland’s group have therefore taken inspiration from magnetic bacteria to create artificial magnetosome-like theranostic magnetic nanovesicles in situ under mild reaction conditions. Here Sarah describes how her group mimics the magnetosomes with iron ions imported into base filled vesicles to form a range of magnetovesicles, from magnetic nanoparticles embedded in the membrane to magnetic nanoparticle core liposome similar to a native magnetosome. Sarah will show how different magnetovesicles vary in their hyperthermic and MRI response and demonstrate how the group can control and tune the particle size and number to tailor them for different applications.

Biological cells are highly organized with numerous subcellular compartments, many of which lack membranous boundaries such as those that surround organelles. Christine Keating’s group are developing simple experimental models for these membraneless organelles based on liquid-liquid phase separation, which is a common phenomenon in aqueous solutions of macromolecules. Solutes such as ions, small molecules, and biopolymers can become compartmentalized by partitioning into one of the phases. The group is studying mechanisms for, and consequences of, this type of compartmentalization using a variety of simple model systems composed of phase-separating aqueous polymer solutions. Through these types of studies, Christine hopes to uncover underlying physiochemical mechanisms in cellular organization and to identify new avenues for biomimetic systems for applications in biotechnology and materials science. For example, compartmentalization of catalysts and/or reactants into polymer-rich droplets can lead to control over the sites and rates of reactions. This approach has led to increased ribozyme reaction rates and development of liposome-stabilized water-in-water emulsions that act as artificial mineralizing vesicles.

Protein design, that is, the construction of entirely new protein sequences that fold into prescribed structures, has come of age. It is now possible to design proteins de novo using simple rules of thumb or computational design methods. The designs can be made rapidly via peptide synthesis or the expression of synthetic genes; and the resulting proteins can usually be characterised all the way through to high-resolution X-ray crystal structures. Contemporary questions in the protein-design field include: What do we do with these new-found skills? What protein structures and functions do we target? How far can we move past the confines of natural protein structures and functions? And, how can we take protein design from an exercise largely done in silico and in vitro into a truly synthetic biology and take it in vivo? This talk will address these questions with reference to simple through to complex and functional protein designs that we have explored over the past 5–10 years. These use a straightforward protein structure, called the alpha-helical coiled coil, which are bundles of 2 or more alpha helices found in many protein-protein interactions. As such it provides an excellent basis for building proteins from the bottom up. The vast majority of coiled-coil designs have been based on simple rules of thumb learnt from natural proteins or derived empirically through experiment. These rules relate sequence to structure to guide the specification of coiled-coil oligomerization state, strand orientation, partner selection, and, to some extent, stability. This has been extremely informative and productive, and design and engineering is probably more advanced for coiled coils than for any other protein structure. However, to move past the low-hanging fruit of coiled-coil design, and into the so-called dark matter of protein structures, we will all have to learn new tricks. To address this Professor Woolfson has begun to tackle coiled-coil design parametrically using computational methods. Professor Woolfson has developed easy-to-use computational modelling tools and a more-sophisticated suite of programs called ISAMBARD that allow the rapid generation and optimisation of protein designs in silico. The first part of the talk will describe how a serendipitous discovery of a 6-stranded alpha-helical barrel led his group to develop these computational methods, and how they have used these to deliver entirely new non-natural protein structures predictably. It will show the utility of this approach to make water-soluble protein-like barrels, which the group has engineered to form materials, bind small molecules, and catalyse simple reactions. Secondly, Professor Woolfson will demonstrate how we might take protein design in vivo with recent work with the Warren lab (Kent) and the Verkade lab (Bristol). He has engineered hybrids of a de novo heterodimer and a natural component of bacterial microcompartments. When expressed in E. coli, the hybrid assemblies to form a cytoscaffold that permeates the cells, and can act as a support for the co-localisation of functional enzymes.