The microtubule cytoskeleton is an intracellular polymer network involved in cell shape, cell motility, and cell division. Microtubules are self-assembling, dynamic polymers composed of tubulin proteins that alternate between phases of growth and shrinkage, a behaviour known as “dynamic instability”. A key feature of microtubules is their ability to generate pushing and pulling forces through controlled assembly and disassembly, guiding for example cell division. The stability of microtubules is largely governed by the progressive hydrolysis of incorporated tubulin dimers. In this thesis we sought to resolve microtubule assembly under force in the presence of regulatory proteins. Through the in vitro reconstitution of microtubule growth against novel micro-fabricated barriers, we studied the biochemical changes leading up to microtubule destabilization. We found that the resulting microtubule lifetimes can be understood with a simple phenomenological 1D model based on noisy microtubule growth and a single hydrolysis rate. Additionally, we explored the implementation of light-inducible protein-protein interactions to gain spatiotemporal control over microtubule function in particular and over in vitro reconstitutions in general. As a proof-of-principle, we developed a light-inducible microtubule gliding assay based on the reversible recruitment of motor proteins to a surface.@en

All living organisms share the need to replicate and proliferate to ensure the survival of their species. In prokaryotes, this is generally guaranteed by a process of cell division where a mother cell is split into two equally sized daughter cells, and it is a complex and heterogeneous process across all the different species. When looking into the Crenarchaea phylum of the archaea, we find a very particular set of proteins that are responsible for orchestrating this process of cell division: the Cdv system. This system is closely related to the ESCRT machinery, which is also responsible for cell division and many other membrane deforming processes in eukaryotes. This close similarity is one of the many common traits that point towards a common origin between archaea and eukaryotes. Although the eukaryotic machinery has been thoroughly and extensively studied, very little is known about the archaeal division system. For this reason, in this work we aimed at better understanding these archaeal proteins, making use of in vitro techniques, with the long-term view of using them to build a synthetic cell form the bottom up...@en

How a system of genetically identical biological cells organizes into spatially heterogeneous tissues is a central question in biology. Even when the molecular and genetic underpinnings of cell-cell interactions are known, how these lead to multicellular patterns is often poorly understood. Of particular interest are dynamic patterns such as traveling waves, which confer spatiotemporal control over key developmental processes such as differentiation, segmentation and cell division. Theoretical approaches based on mathematical descriptions of underlying physical and chemical processes provide a promising avenue to explore biological pattern formation. In particular, theoretical models connect processes on the molecular scale to biological function on the tissue level and may provide mechanistic descriptions of how patterns are generated and maintained.@en

The past decades have seen the rapid development of many aspects of synthetic biology. For example, attempts to build synthetic cells under controlled conditions in the laboratory have led to significant achievements. Following a bottom-up approach, scientists aim at building a self-reproducing synthetic cell with a minimum number of biological modules. A functional synthetic cell should accomplish at least four processes during one cycle: growth,DNAreplication,DNA segregation, and division. DNA segregation, as a vital step in the life cycle of a synthetic cell, is in focus in this thesis. Segregation of replicated DNA in eukaryotes depends highly on microtubules. Microtubules are protein polymers which form cylindrical tubes, a dynamic structure with which they can exert both pulling and pushing forces. During cell division, microtubules form a spindle-like structure, named the mitotic spindle. Microtubules in the mitotic spindle apparatus attach to the replicated chromosomes through kinetochores and exert pulling forces to separate the sister chromosomes and place them in the newly born daughter cells. These hollow tubes can grow in the presence of GTPbound tubulin dimers. When all the GTP-tubulin subunits at the end of a filament turned into GDP-tubulin, themicrotubule exhibits a transition to shrinkage. While the addition of GTP-tubulin to the end of microtubulesmay cause pushing forces, shrinking microtubules are able to exert pulling forces with the help of microtubule adapter proteins. The random transitions between growth and shrinkage of a microtubule is called dynamic instability and has been studied throughout decades. DNA segregation components of a synthetic cell should preferably be fully expressible in a cell-free manner which employs reconstituted transcription-translation factors. Although mitosis in eukaryotes is the best-studied DNA segregation system to date, microtubules are unsuitable for cell-free expression due to their complex chaperondependent folding as well as essential post-translational modifications. Interesting alternatives for DNA segregation in synthetic cells are provided by bacterial segregation systems. An active bacterial segregation system is ideally composed of three elements: a cytomotive NTPase filament, a connecting or adapter protein, and a centromere-like DNA sequence. This thesis will examine various bacterial cytoskeletal filaments with a central focus on their suitability for building a DNA segregation systemfor synthetic cells...@en

Every living organism consists of cells. Even for the simplest single-cell organism, this cell is extremely complex. Thousands of components (such asDNA, cytoskeletal filaments, proteins, lipids, nutrients and energy) are organized both spatially and temporally to ensure proper functioning of vital cellular processes. One of those processes is pattern formation, or cell polarity. Cell polarity is defined as the morphological and functional differentiation of cellular compartments in a directional manner. This directionality is crucial for processes like cell division, cell migration and cell growth, which require an asymmetric action of the cell. When polarity is inhibited by silencing of cell polarity proteins, cells become deformed and have trouble to function, if viable at all. It is well-known that cell polarity is the result of reaction-diffusion and cytoskeleton-based mechanisms, but the exact mechanisms remain unknown. In this thesis, we focus on the latter, and more specifically on one type of cytoskeletal filaments: microtubules (MTs). Our goal is to study the role of MTs in cell polarity establishment. @en

The research described in this doctoral thesis was part of the BaSyC consortium, which aims to build synthetic cells from bottom up. My small role in this large collaboration was to study the theoretical basis for cell division. This is a very complex process in living organisms, involves many components and can be accomplished in a variety of ways. To successfully replicate cell division in synthetic organisms it will be important to identify the simplest mechanisms and understand the basic components of the cell division machinery, that come together to reshape membranes and divide one compartment into two. The theoretical framework used to study these self-assembly processes, that lead to membrane deformations and division, is the framework of membrane mediated interactions. Let me use an example to explain this framework. Imagine you sit down on a couch, while holding a saucer with a coffee cup on it. The coffee is too hot, so you set it down next to you ont the couch and while you wait you get lost in your thoughts. . .You feel like getting up and walking around in the room. While you are still lifting your body up from the sitting position, you hear the clinking of the cup against the saucer, and you see from the corner of your eye that the coffee cup is shaking. Without having touched the cup you managed to spill the coffee. The surface of the couchwas deformed and pulled tight because youwere sitting on it. When you got up the surface relaxed back to its original shape. This movement of the surface caused the cup to shake and spill its contents (see fig. 1 for an artist’s impression). Membrane mediated interactions are just like that! When something deforms a membrane in one place, that deformation is felt at a distance by other objects that inhabit the membrane. Exactly how far and how strongly such signals are felt depends on how floppy or tense a membrane is. When we study membrane mediated interactions, our main goal is to quantify the effects membrane deforming objects can have on each other.@en

Biological membranes are selective soft barriers that compartmentalize internal structure of a cell into organelles and separate them as a whole from the external environment. Due to their innate feature of being able to undergo constant reshaping, cellular membranes spatially attain diverse shapes ranging from simple spherical vesicles to more peculiar structures like the interconnected network of tubes found in the endoplasmic reticulum. Membranes are not only composed of lipids, but also host an enormous number of inclusions like proteins. Recent studies of biological membranes have revealed that such inclusions play a key role in diverse biological processes through either sensing or inducing perturbations to the membrane shape. In this dissertation, we studied the interplay between the shape of membrane and the spatial organization of attached curvature inducing objects using mathematical tools and numerical simulations in highly curved spherical and cylindrical geometries. First, we investigated the interaction between inclusions of different shapes embedded in/adhered to tubular membranes. Our combined theoretical analysis and numerical simulation results evinced that tubular membranes, in contrast to their planar counterpart, transmit an attractive force between inclusions, stemming from their closed and curved geometry. We then elucidated that collective interaction between proteins results in the formation of line-like and ring-like clusters, depending on the their intrinsic shape (Chapters 2–4). We further showed how curvature sensing crescent-like proteins in high densities can constrict tubular membranes and facilitate their splitting, demonstrating that both the curvature-sensing and curvature-inducing property of proteins are two sides of the same coin. Moreover, we used our simulation results to explain how mitochondorial machinery triggers, facilitates and drives membrane fission in its tubular network to avoid entanglements (Chapter 3). Next, we examined the interaction of spherical proteins adhered to closed vesicles. Our simulation results – supported by recent experimental evidence – revealed membrane curvature as a common physical origin for interactions between any membrane deforming objects, from nanometre-sized proteins to micrometre-sized particles (Chapter 5). Our further simulations unraveled how introducing curvature variation on the surface of a closed vesicle can be exploited by inanimate particles to regulate their pattern formation (Chapter 6). Finally, through theoretical calculations,we analyzed the interplay between the shape of a cell and the rearrangement of attached microtubules (Chapter 7). Our results particularly suggested that the commonly reported parallel structure and bundling of microtubules can be induced by membrane mediated interactions.@en

Cells of the most common organisms like plants and animals are filled with polymeric networks that fulfil important functions of the cell. There is however no analytically solvable model that describes diffusion in such a cell. This thesis presents a model for diffusion in polymeric environments, and some predictions about the behaviour of the model are made and confirmed by simulations. Furthermore, the Fokker-Planck equation of this problem is studied, in order to solve the problem. Certain approximations are presented and solved, and it is investigated when the approximations are sound. Moreover, a method is described that can derive a solution to an equation with certain boundary conditions, from a solution to the same equation with different boundary conditions. This thesis also shows how this novel method can be applied to this model to find a non-approximated solution to the equation, where this was not possible without this method. Finally, it is described how a multitude of partial differential equations that are linked to each other via the boundary conditions can be solved, can be solved using the method described.

For millennia, humans have used microbes to produce industrial products of social and economical value through fermentation processes. In recent years, the application of engineering principles to microbiology have dramatically expanded our ability to modify and optimize microbes for the production of a wide variety of commercial products from renewable feedstocks: food and commodity chemicals, to biofuels and fine chemicals such as pharmaceuticals, fragrances, cosmetics or dyes. The use of microbial bioprocesses for the production of natural products represents an attractive and sustainable alternative to current industrial production methods, which mainly rely on chemical synthesis and/ or extraction from native producers. Advanced biomanufacturing technologies would not only provide sustainable economic benefits (by reducing the monetary cost of production of useful chemicals), but also offer social and environmental benefits. Synthetic biology has allowed engineering the production of many industrial compounds within microbes that do not naturally produce them – this is called “heterologous microbial biosynthesis”. In addition to replacing current manufacturing processes, heterologous microbial biosynthesis likely offers the only viable platform to produce certain natural products at industrial scales. Indeed, many relevant compounds cannot be viably manufactured through chemical synthesis, and/or are produced at undetectable/insufficient levels in native organisms. However, many heterologous bioprocesses remain in their infancy to fully enable an economically viable delivery of relevant natural products to the market. In order to build and sustain the promise of a bioeconomy for the 21st century, metabolic engineering is under pressure to continue to provide largescale, sustainable and cost-competitive bioprocesses that meet global needs. In this thesis, we focus on the development of microbial strains to accelerate the microbial production of 2 different families of high-value compounds of prominent biotechnological relevance within the established microbial chassis Escherichia coli: antibiotics and isoprenoids. The fight against antimicrobial resistance is considered one of the greatest public health challenges of the 21st century. Recent technologies have uncovered new antibiotics that, if harnessed, might help alleviate this crisis. However, most of these new antibiotic compounds are far too complex for economical chemical synthesis, and are naturally produced by unculturable and/or genetically intractable microbes. Developing new heterologous microbial platforms for antibiotic production may be an efficient solution for harnessing the clinical potential of these molecules and their commercialization. Isoprenoids represent one of the largest families of natural compounds (over 50,000 molecules) with an incredible number of practical uses, and of great commercial value: from high-value compounds such as many pharmaceuticals, fragrances and flavors, to commodity chemicals such as solvents, rubber or advanced biofuels. We focus in particular on relevant obstacles associated with the development of proof-of-principle strains for the laboratory-scale production of these high-value chemicals. @en

One of the biggest scientific challenges to be tackled this century is how traits of living organisms originate from genes, the so-called genotype-phenotype map, and conversely how traits influence genes through a process called evolution. The solution will yield a large societal impact, with applications in food (e.g., engineering drought-resistant crops), industry (e.g., material production through microorganisms) and health care (e.g., personalized medicine). The complexity of the genotype-phenotype map lies in how it typically spans multiple, interwoven scales (e.g., in size). This dissertation builds on the ambition that ultimately, a solution is found by generalizations of simpler systems. Therefore, we unravel here the map for a tractable example, polarization in budding yeast, and make insightful how evolution can couple to the map. During polarization, the unicellular organism budding yeast chooses a direction in which it will divide. This involves self-organizing many proteins, in particular Cdc42p, to a single region on its cell membrane. While starting on the molecular scale, the process ultimately affects population traits such as doubling time. To understand the transition in scales in detail, we start bottom-up by experimentally verifying the molecular theory behind polarity success for different genetic backgrounds. The theoretical model treats, amongst others, proteins that activate Cdc42p, which are mechanistically included for the first time. Concretely, we test resulting predictions on sharp lower Cdc42p concentration bounds for viability using, inter alia, growth assays on strains variably producing fluorescent Cdc42p. The experiments confirmed the theory that allows reconstitution of molecular mechanisms underlying polarity establishment. To advance to population traits, I constructed a tractable growth model, fed by simple rules emerging from the aforementioned theory (only implicitly encompassing the molecular information). Essentially, Cdc42p is stochastically produced, diluted by basic volume expansion, and must exceed a concentration threshold to divide. Despite disregarding many details, quantitative agreement between unintuitive, experimentally validated traits documented in literature and those from model simulations is reached. The simplicity of the model assumptions also allows new insights in evolution. I elaborate theoretically how lucky cells that by chance produce above average amounts of protein, proliferate better to bias the observed population. Therefore, protein levels promptly adapt non-genetically, also in response to e.g., environmental changes, in a reversible and almost automatic manner. Based on existing experimental data, I predict this noise-based mechanism to notably expand the ease of evolution for essential genes (in yeast for 25%-60% of these). Due to its simple nature, I conjecture that it should be found in many organisms. In conclusion, we find a successful strategy to tractably analyze the genotype-phenotype map in yeast polarity. The map can be expanded to other functions than polarity, provided that sufficient bio-functional information is available. The analysis also elucidates a new evolutionary coupling to this map. At a step above genes, noisy protein production can freely be utilized for short-term adaptation. Experimentally confirming the presence of this evolutionary mechanism in other model systems, and applying to these the same strategy to predict traits, will generate a completer picture of how traits of living systems are formed and shaped by evolution. @en