Impact of crowding on the diversity of expanding populations

Carl F. Schreck, Diana Fusco, Yuya Karita, Stephen Martis, Jona Kayser, Marie-Cécilia Duvernoy, Oskar Hallatschek

Crowding effects critically impact the self-organization of densely packed cellular assemblies, such as biofilms, solid tumors, and developing tissues. When cells grow and divide, they push each other apart, remodeling the structure and extent of the population’s range. Recent work has shown that crowding has a strong impact on the strength of natural selection. However, the impact of crowding on neutral processes, which controls the fate of new variants as long as they are rare, remains unclear. Here, we quantify the genetic diversity of expanding microbial colonies and uncover signatures of crowding in the site frequency spectrum. By combining Luria–Delbrück fluctuation tests, lineage tracing in a novel microfluidic incubator, cell-based simulations, and theoretical modeling, we find that the majority of mutations arise behind the expanding frontier, giving rise to clones that are mechanically “pushed out” of the growing region by the proliferating cells in front. These excluded-volume interactions result in a clone-size distribution that solely depends on where the mutation first arose relative to the front and is characterized by a simple power law for low-frequency clones. Our model predicts that the distribution depends on a single parameter—the characteristic growth layer thickness—and hence allows estimation of the mutation rate in a variety of crowded cellular populations. Combined with previous studies on high-frequency mutations, our finding provides a unified picture of the genetic diversity in expanding populations over the whole frequency range and suggests a practical method to assess growth dynamics by sequencing populations across spatial scales.

Evolutionary rescue of resistant mutants is governed by a balance between radial expansion and selection in compact populations

Serhii Aif, Nico Appold, Lucas Kampman, Oskar Hallatschek, Jona Kayser

Mutation-mediated treatment resistance is one of the primary challenges for modern antibiotic and anti-cancer therapy. Yet, many resistance mutations have a substantial fitness cost and are subject to purifying selection. How emerging resistant lineages may escape purifying selection via subsequent compensatory mutations is still unclear due to the difficulty of tracking such evolutionary rescue dynamics in space and time. Here, we introduce a system of fluorescence-coupled synthetic mutations to show that the probability of evolutionary rescue, and the resulting long-term persistence of drug resistant mutant lineages, is dramatically increased in dense microbial populations. By tracking the entire evolutionary trajectory of thousands of resistant lineages in expanding yeast colonies we uncover an underlying quasi-stable equilibrium between the opposing forces of radial expansion and natural selection, a phenomenon we term inflation-selection balance. Tailored computational models and agent-based simulations corroborate the fundamental nature of the observed effects and demonstrate the potential impact on drug resistance evolution in cancer. The described phenomena should be considered when predicting multi-step evolutionary dynamics in any mechanically compact cellular population, including pathogenic microbial biofilms and solid tumors. The insights gained will be especially valuable for the quantitative understanding of response to treatment, including emerging evolution-based therapy strategies.

Wie mikrobielle Modellsysteme helfen, Tumorevolution zu entschlüsseln

Maximilian Eiche, Jona Kayser

While cellular evolution is one of the most fundamental concepts of<br>life, its consequences are among the most pressing issues of modern<br>health care, including cancer and the emergence of therapy resistance.<br>We currently still lack the ability to accurately predict evolutionary trajectories,<br>especially in spatially dense, pathogenic cellular populations<br>such as microbial biofi lms or solid tumors. Here, we discuss the conceptual<br>framework of evolution in dense populations and the potential<br>of tailored microbial model systems to systematically study the underlying<br>mechanisms.

Collective motion conceals fitness differences in crowded cellular populations

Jona Kayser, Carl F. Schreck, Matti Gralka, Diana Fusco, Oskar Hallatschek

Many cellular populations are tightly packed, such as microbial colonies<br> and biofilms, or tissues and tumours in multicellular organisms. The<br> movement of one cell in these crowded assemblages requires motion of<br> others, so that cell displacements are correlated over many cell<br> diameters. Whenever movement is important for survival or growth, these<br> correlated rearrangements could couple the evolutionary fate of<br> different lineages. However, little is known about the interplay between<br> mechanical forces and evolution in dense cellular populations. Here, by<br> tracking slower-growing clones at the expanding edge of yeast colonies,<br> we show that the collective motion of cells prevents costly mutations<br> from being weeded out rapidly. Joint pushing by neighbouring cells<br> generates correlated movements that suppress the differential<br> displacements required for selection to act. This mechanical screening<br> of fitness differences allows slower-growing mutants to leave more<br> descendants than expected under non-mechanical models, thereby<br> increasing their chance for evolutionary rescue. Our work suggests that,<br> in crowded populations, cells cooperate with surrounding neighbours<br> through inevitable mechanical interactions. This effect has to be<br> considered when predicting evolutionary outcomes, such as the emergence<br> of drug resistance or cancer evolution.

Emergence of evolutionary driving forces in pattern-forming microbial populations

Jona Kayser, Carl F. Schreck, QinQin Yu, Matti Gralka, Oskar Hallatschek

Evolutionary dynamics are controlled by a number of driving forces, such<br> as natural selection, random genetic drift and dispersal. In this<br> perspective article, we aim to emphasize that these forces act at the<br> population level, and that it is a challenge to understand how they<br> emerge from the stochastic and deterministic behaviour of individual<br> cells. Even the most basic steric interactions between neighbouring<br> cells can couple evolutionary outcomes of otherwise unrelated<br> individuals, thereby weakening natural selection and enhancing random<br> genetic drift. Using microbial examples of varying degrees of<br> complexity, we demonstrate how strongly cell-cell interactions influence<br> evolutionary dynamics, especially in pattern-forming systems. As pattern<br> formation itself is subject to evolution, we propose to study the<br> feedback between pattern formation and evolutionary dynamics, which<br> could be key to predicting and potentially steering evolutionary<br> processes. Such an effort requires extending the systems biology<br> approach from the cellular to the population scale.

SCWISh network is essential for survival under mechanical pressure

Morgan Delarue, Gregory Poterewicz, Ori Hoxha, Jessica Choi, Wonjung Yoo, Jona Kayser, Liam Holt, Oskar Hallatschek

Cells that proliferate within a confined environment build up mechanical<br> compressive stress. For example, mechanical pressure emerges in the<br> naturally space-limited tumor environment. However, little is known<br> about how cells sense and respond to mechanical compression. We<br> developed microfluidic bioreactors to enable the investigation of the<br> effects of compressive stress on the growth of the genetically tractable<br> model organism Saccharomyces cerevisiae. We used this system to<br> determine that compressive stress is partly sensed through a module<br> consisting of the mucin Msb2 and the cell wall protein Sho1, which act<br> together as a sensor module in one of the two major osmosensing pathways<br> in budding yeast. This signal is transmitted via the MAPKKK kinase<br> Ste11. Thus, we term this mechanosensitive pathway the "SMuSh" pathway,<br> for Ste11 through Mucin/Sho1 pathway. The SMuSh pathway delays cells in<br> the G1 phase of the cell cycle and improves cell survival in response to<br> growth-induced pressure. We also found that the cell wall integrity<br> (CWI) pathway contributes to the response to mechanical compressive<br> stress. These latter results are confirmed in complimentary experiments<br> in Mishra et al. [Mishra R, et al. (2017) Proc Natl Acad Sci USA,<br> 10.1073/pnas. 1709079114]. When both the SMuSh and the CWI pathways are<br> deleted, cells fail to adapt to compressive stress, and all cells lyse<br> at relatively low pressure when grown in confinement. Thus, we define a<br> network that is essential for cell survival during growth under<br> pressure. We term this mechanosensory system the SCWISh (survival<br> through the CWI and SMuSh) network.

Excess of mutational jackpot events in expanding populations revealed by spatial Luria-Delbruck experiments

Diana Fusco, Matti Gralka, Jona Kayser, Alex Anderson, Oskar Hallatschek

The genetic diversity of growing cellular populations, such as biofilms,<br> solid tumours or developing embryos, is thought to be dominated by rare,<br> exceptionally large mutant clones. Yet, the emergence of these<br> mutational jackpot events is only understood in well-mixed populations,<br> where they stem from mutations that arise during the first few cell<br> divisions. To study jackpot events in spatially structured populations,<br> we track mutant clones in microbial populations using fluorescence<br> microscopy and population sequencing. High-frequency mutations are found<br> to be massively enriched in microbial colonies compared with well-shaken<br> liquid cultures, as a result of late-occurring mutations surfing at the<br> edge of range expansions. Thus, jackpot events can be generated not only<br> when mutations arise early but also when they occur at favourable<br> locations, which exacerbates their role in adaptation and disease. In<br> particular, because spatial competition with the wild type keeps most<br> mutant clones in a quiescent state, strong selection pressures that kill<br> the wild type promote drug resistance.

Mapping Internal Stress of in Vitro Cytoskeletal Networks with UV-Laser Ablation

Martina Lindauer, Jona Kayser, Andreas R. Bausch

The Small Heat Shock Protein Hsp27 Affects Assembly Dynamics and Structure of Keratin Intermediate Filament Networks

Jona Kayser, Martin Haslbeck, Lisa Dempfle, Maike Krause, Carsten Grashoff, Johannes Buchner, Harald Herrmann, Andreas R. Bausch

The mechanical properties of living cells are essential for many<br> processes. They are defined by the cytoskeleton, a composite network of<br> protein fibers. Thus, the precise control of its architecture is of<br> paramount importance. Our knowledge about the molecular and physical<br> mechanisms defining the network structure remains scarce, especially for<br> the intermediate filament cytoskeleton. Here, we investigate the effect<br> of small heat shock proteins on the keratin 8/18 intermediate filament<br> cytoskeleton using a well-controlled model system of reconstituted<br> keratin networks. We demonstrate that Hsp27 severely alters the<br> structure of such networks by changing their assembly dynamics.<br> Furthermore, the C-terminal tail domain of keratin 8 is shown to be<br> essential for this effect. Combining results from fluorescence and<br> electron microscopy with data from analytical ultracentrifugation<br> reveals the crucial role of kinetic trapping in keratin network<br> formation.

Neurofilament sidearms modulate parallel and crossed-filament orientations inducing nematic to isotropic and re-entrant birefringent hydrogels

Joanna Deek, Peter J. Chung, Jona Kayser, Andreas R. Bausch, Cyrus R. Safinya

Neurofilaments are intermediate filaments assembled from the subunits<br> neurofilament-low, neurofilament-medium and neurofilament-high. In<br> axons, parallel neurofilaments form a nematic liquid-crystal hydrogel<br> with network structure arising from interactions between the<br> neurofilaments' C-terminal sidearms. Here we report, using small-angle<br> X-ray-scattering, polarized-microscopy and rheometry, that with<br> decreasing ionic strength, neurofilament-low-high,<br> neurofilament-low-medium and neurofilament-low-medium-high hydrogels<br> transition from the nematic hydrogel to an isotropic hydrogel (with<br> random, crossed-filament orientation) and to an unexpected new<br> re-entrant liquid-crystal hydrogel with parallel filaments-the<br> bluish-opaque hydrogel-with notable mechanical and water retention<br> properties reminiscent of crosslinked hydrogels. Significantly, the<br> isotropic gel phase stability is sidearm-dependent:<br> neurofilament-low-high hydrogels exhibit a wide ionic strength range,<br> neurofilament-low-medium hydrogels a narrow ionic strength range,<br> whereas neurofilament-low hydrogels lack the isotropic gel phase. This<br> suggests a dominant regulatory role for neurofilament-high sidearms in<br> filament reorientation plasticity, facilitating organelle transport in<br> axons. Neurofilament-inspired biomimetic hydrogels should therefore<br> exhibit remarkable structure-dependent moduli and slow and fast<br> water-release properties.

Salt-Responsive Liquid Crystal Hydrogels: Neurofilament Network Structure and Mechanical Modulation

Joanna Deek, Peter J. Chung, Jona Kayser, Andreas Bausch, Cyrus R. Safinya

Towards constructing extracellular matrix-mimetic hydrogels: An elastic hydrogel constructed from tandem modular proteins containing tenascin FnIII domains

Shanshan Lv, Tianjia Bu, Jona Kayser, Andreas Bausch, Hongbin Li

Protein-based hydrogels have been developed for various biomedical<br> applications where they provide artificial extracellular<br> microenvironments that mimic the physical and biochemical<br> characteristics of natural extracellular matrices (ECMs). In natural<br> ECMs, a large number of proteins are tandem modular proteins consisting<br> of many individually folded functional domains that confer structural<br> and biological functionalities. Such tandem modular proteins are<br> promising building blocks for constructing ECM-mimetic biomaterials.<br> However, their use for such purposes has not been explored extensively.<br> Tenascin-C (TNC) is an ECM tandem modular protein and plays an important<br> role in mechanotransduction by regulating important cell matrix<br> interactions. The third FnIII domain of TNC (TNfn3) contains an RGD<br> sequence and is known to bind integrins. Here we use the TNfn3 domain<br> and resilin sequence-based tandem modular protein FRF4RF4R (F represents<br> the TNfn3 domain and R represents the resilin sequence, respectively) as<br> a building block to construct protein-based ECM-mimetic hydrogels. The<br> tandem modular protein-based building block FRF4RF4R closely mimics the<br> architecture of the naturally occurring tandem modular ECM protein TNC<br> and incorporates intact RGD-containing FnIII domains. Our results<br> demonstrate that tandem modular proteins containing TNfn3 can be readily<br> photochemically crosslinked into elastic hydrogels, whose Young's<br> modulus can be tuned by the concentration of the tandem modular protein<br> solution. In vitro studies demonstrate that none of the photochemical<br> crosslinking reaction components are cytotoxic at the level tested, and<br> the hydrogel supports the spread of human lung fibroblast cells. Our<br> results demonstrate that FRF4RF4R-based hydrogel is a novel ECM-mimetic<br> hydrogel. (C) 2013 Acta Materialia Inc. Published by Elsevier Ltd. All<br> rights reserved.

Assembly kinetics determine the structure of keratin networks

Jona Kayser, Heinrich Grabmayr, Markus Harasim, Harald Herrmann, Andreas R. Bausch

Living cells exhibit an enormous bandwidth of mechanical and morphological properties that are mainly determined by the cytoskeleton.<br> In metazoan cells this composite network is constituted of three<br> different types of filamentous systems: actin filaments, microtubules<br> and intermediate filaments. Keratin-type intermediate filaments are an<br> essential component of epithelial tissues, where they comprise networks<br> of filaments and filament bundles. However, the underlying mechanisms<br> leading to this inherently polymorphic structure remain elusive. Here,<br> we show that keratin filaments form kinetically trapped networks of<br> bundles under near-physiological conditions in vitro. The network<br> structure is determined by the intricate interplay between filament<br> elongation and their lateral association to bundles and clusters.

Heat Shock Proteins Regulate Structure of Intermediate Filament Networks

Jona Kayser, Martin Haslbeck, Harald Herrmann, Johannes Buchner, Andreas R. Bausch

Slow dynamics and internal stress relaxation in bundled cytoskeletal networks

O. Lieleg, Jona Kayser, G. Brambilla, L. Cipelletti, A. R. Bausch

Crosslinked and bundled actin filaments form networks that are essential<br> for the mechanical properties of living cells. Reconstituted actin<br> networks have been extensively studied not only as a model system for<br> the cytoskeleton, but also to understand the interplay between<br> microscopic structure and macroscopic viscoelastic properties of<br> network-forming soft materials. These constitute a broad class of<br> materials with countless applications in science and industry. So far,<br> it has been widely assumed that reconstituted actin networks represent<br> equilibrium structures. Here, we show that fully polymerized<br> actin/fascin bundle networks exhibit surprising age-dependent changes in<br> their viscoelastic properties and spontaneous dynamics, a feature<br> strongly reminiscent of out-of-equilibrium, or glassy, soft materials.<br> Using a combination of rheology, confocal microscopy and space-resolved<br> dynamic light scattering, we demonstrate that actin networks build up<br> stress during their formation and then slowly relax towards equilibrium<br> owing to the unbinding dynamics of the crosslinking molecules.

Stability of a Surface-Bound Oligonucleotide Duplex Inferred from Molecular Dynamics: A Study of Single Nucleotide Defects Using DNA Microarrays

Thomas Naiser, Jona Kayser, Timo Mai, Wolfgang Michel, Albrecht Ott

Microarray technology uses the sequence dependent hybridization<br> (binding) affinity of surface-bound oligonucleotide strands for the<br> quantification of complex nucleic acid mixtures. In spite of its huge<br> potential in life science and medicine, microarray oligonucleotide<br> hybridization remains far from being understood. Taking advantage of<br> microarray combinatorial possibilities we show that, although surface<br> bound, the hybridization affinities of single-base mismatched<br> oligonucleotides can be derived from first principles using parameters<br> from bulk.

Position dependent mismatch discrimination on DNA microarrays - experiments and model

Thomas Naiser, Jona Kayser, Timo Mai, Wolfgang Michel, Albrecht Ott

Background: The propensity of oligonucleotide strands to form stable<br> duplexes with complementary sequences is fundamental to a variety of<br> biological and biotechnological processes as various as microRNA<br> signalling, microarray hybridization and PCR. Yet our understanding of<br> oligonucleotide hybridization, in particular in presence of surfaces, is<br> rather limited. Here we use oligonucleotide microarrays made in-house by<br> optically controlled DNA synthesis to produce probe sets comprising all<br> possible single base mismatches and base bulges for each of 20 sequence<br> motifs under study.<br> Results: We observe that mismatch discrimination is mostly determined by<br> the defect position (relative to the duplex ends) as well as by the<br> sequence context. We investigate the thermodynamics of the<br> oligonucleotide duplexes on the basis of double-ended molecular zipper.<br> Theoretical predictions of defect positional influence as well as long<br> range sequence influence agree well with the experimental results.<br> Conclusion: Molecular zipping at thermodynamic equilibrium explains the<br> binding affinity of mismatched DNA duplexes on microarrays well. The<br> position dependent nearest neighbor model (PDNN) can be inferred from<br> it. Quantitative understanding of microarray experiments from first<br> principles is in reach.

Impact of point-mutations on the hybridization affinity of surface-bound DNA/DNA and RNA/DNA oligonucleotide-duplexes: Comparison of single base mismatches and base bulges

Thomas Naiser, Oliver Ehler, Jona Kayser, Timo Mai, Wolfgang Michel, Albrecht Ott

Background: The high binding specificity of short 10 to 30 mer<br> oligonucleotide probes enables single base mismatch ( MM) discrimination<br> and thus provides the basis for genotyping and resequencing microarray<br> applications. Recent experiments indicate that the underlying principles<br> governing DNA microarray hybridization - and in particular MM<br> discrimination - are not completely understood. Microarrays usually<br> address complex mixtures of DNA targets. In order to reduce the level of<br> complexity and to study the problem of surface-based hybridization with<br> point defects in more detail, we performed array based hybridization<br> experiments in well controlled and simple situations.<br> Results: We performed microarray hybridization experiments with short 16<br> to 40 mer target and probe lengths ( in situations without competitive<br> hybridization) in order to systematically investigate the impact of<br> point-mutations - varying defect type and position - on the<br> oligonucleotide duplex binding affinity. The influence of single base<br> bulges and single base MMs depends predominantly on position - it is<br> largest in the middle of the strand. The position-dependent influence of<br> base bulges is very similar to that of single base MMs, however certain<br> bulges give rise to an unexpectedly high binding affinity. Besides the<br> defect ( MM or bulge) type, which is the second contribution in<br> importance to hybridization affinity, there is also a sequence<br> dependence, which extends beyond the defect next-neighbor and which is<br> difficult to quantify. Direct comparison between binding affinities of<br> DNA/DNA and RNA/DNA duplexes shows, that RNA/DNA purine-purine MMs are<br> more discriminating than corresponding DNA/DNA MMs. In DNA/ DNA MM<br> discrimination the affected base pair ( C . G vs. A . T) is the<br> pertinent parameter. We attribute these differences to the different<br> structures of the duplexes ( A vs. B form).<br> Conclusion: We have shown that DNA microarrays can resolve even subtle<br> changes in hybridization affinity for simple target mixtures. We have<br> further shown that the impact of point defects on oligonucleotide<br> stability can be broken down to a hierarchy of effects. In order to<br> explain our observations we propose DNA molecular dynamics - in form of<br> zipping of the oligonucleotide duplex - to play an important role.