Configurational fingerprints of multicellular living systems

Signatures of structural disorder in developing epithelial tissues

Epithelial cells generate functional tissues in developing embryos through collective movements and shape changes. In some morphogenetic events, a tissue dramatically reorganizes its internal structure - often generating high degrees of structural disorder - to accomplish changes in tissue shape. However, the origins of structural disorder in epithelia and what roles it might play in morphogenesis are poorly understood. We study this question in the Drosophila germband epithelium, which undergoes dramatic changes in internal structure as cell rearrangements drive elongation of the embryo body axis. Using two order parameters that quantify volumetric and shear disorder, we show that structural disorder increases during body axis elongation and is strongly linked with specific developmental processes. Both disorder metrics begin to increase around the onset of axis elongation, but then plateau at values that are maintained throughout the process. Notably, the disorder plateau values for volumetric disorder are similar to those for random cell packings, suggesting this may reflect a limit on tissue behavior. In mutant embryos with disrupted external stresses from the ventral furrow, both disorder metrics reach wild-type maximum disorder values with a delay, correlating with delays in cell rearrangements. In contrast, in mutants with disrupted internal stresses and cell rearrangements, volumetric disorder is reduced compared to wild type, whereas shear disorder depends on specific external stress patterns. Together, these findings demonstrate that internal and external stresses both contribute to epithelial tissue disorder and suggest that the maximum values of disorder in a developing tissue reflect physical or biological limits on morphogenesis.

Pulling the strings on solid-to-liquid phase transitions in cell collectives

Cell collectives must dynamically adapt to different biological contexts. For instance, in homeostatic conditions, epithelia must establish a barrier between body compartments and resist external stresses, while during development, wound healing or cancer invasion, these tissues undergo extensive remodeling. Using analogies from inert, passive materials, changes in cellular density, shape, rearrangements and/or migration were shown to result in collective transitions between solid and fluid states. However, what biological mechanisms govern these transitions remains an open question. In particular, the upstream signaling pathways and molecular effectors controlling the key physical axes determining tissue rheology and dynamics remain poorly understood. In this perspective, we focus on emerging evidence identifying the first biological signals determining the collective state of living tissues, with an emphasis on how these mechanisms are exploited for functionality across biological contexts.

Learning Dynamics from Multicellular Graphs with Deep Neural Networks

The inference of multicellular self-assembly is the central quest of understanding morphogenesis, including embryos, organoids, tumors, and many others. However, it has been tremendously difficult to identify structural features that can indicate multicellular dynamics. Here we propose to harness the predictive power of graph-based deep neural networks (GNN) to discover important graph features that can predict dynamics. To demonstrate, we apply a physically informed GNN (piGNN) to predict the motility of multi-cellular collectives from a snapshot of their positions both in experiments and simulations. We demonstrate that piGNN is capable of navigating through complex graph features of multicellular living systems, which otherwise can not be achieved by classical mechanistic models. With increasing amounts of multicellular data, we propose that collaborative efforts can be made to create a multicellular data bank (MDB) from which it is possible to construct a large multicellular graph model (LMGM) for general-purposed predictions of multicellular organization.

Topological packing statistics of living and nonliving matter

Complex disordered matter is of central importance to a wide range of disciplines, from bacterial colonies and embryonic tissues in biology to foams and granular media in materials science to stellar configurations in astrophysics. Because of the vast differences in composition and scale, comparing structural features across such disparate systems remains challenging. Here, by using the statistical properties of Delaunay tessellations, we introduce a mathematical framework for measuring topological distances between general three-dimensional point clouds. The resulting system-agnostic metric reveals subtle structural differences between bacterial biofilms as well as between zebrafish brain regions, and it recovers temporal ordering of embryonic development. We apply the metric to construct a universal topological atlas encompassing bacterial biofilms, snowflake yeast, plant shoots, zebrafish brain matter, organoids, and embryonic tissues as well as foams, colloidal packings, glassy materials, and stellar configurations. Living systems localize within a bounded island-like region of the atlas, reflecting that biological growth mechanisms result in characteristic topological properties.

Biophysical informatics reveals distinctive phenotypic signatures and functional diversity of single-cell lineages

MOTIVATION

In this work, we present an analytical method for quantifying both single-cell morphologies and cell network topologies of tumor cell populations and use it to predict 3D cell behavior.

RESULTS

We utilized a supervised deep learning approach to perform instance segmentation on label-free live cell images across a wide range of cell densities. We measured cell shape properties and characterized network topologies for 136 single-cell clones derived from the YUMM1.7 and YUMMER1.7 mouse melanoma cell lines. Using an unsupervised clustering algorithm, we identified six distinct morphological subclasses. We further observed differences in tumor growth and invasion dynamics across subclasses in an in vitro 3D spheroid model. Compared to existing methods for quantifying 2D or 3D phenotype, our analytical method requires less time, needs no specialized equipment and is capable of much higher throughput, making it ideal for applications such as high-throughput drug screening and clinical diagnosis.

AVAILABILITY AND IMPLEMENTATION

https://github.com/trevor-chan/Melanoma_NetworkMorphology.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Disorder to order transition in cell-ECM systems mediated by cell-cell collective interactions

Cells in functional tissues execute various collective activities to achieve diverse ordered processes including wound healing, organogenesis, and tumor formation. How a group of individually operating cells initiate such complex collective processes is still not clear. Here, we report that cells in 3D extracellular matrix (ECM) initiate collective behavior by forming cell-ECM network when the cells are within a critical distance from each other. We employed compaction of free-floating (FF) 3D collagen gels with embedded fibroblasts as a model system to study collective behavior and found a sharp transition in the amount of compaction as a function of cell-cell distance, reminiscent of phase transition in materials. Within the critical distance, cells remodel the ECM irreversibly, and form dense collagen bridges between each other resulting in the formation of a network. Beyond the critical distance, cells exhibit Brownian dynamics and only deform the matrix reversibly in a transient manner with no memory of history, thus maintaining the disorder. Network formation seems to be a necessary and sufficient condition to trigger collective behavior and a disorder-to order transition. STATEMENT OF SIGNIFICANCE: Macroscopic compaction of in vitro collagen gels is mediated by collective mechanical interaction of cells. Previous studies on cell-induced ECM compaction suggest the existence of a critical cell density and phase transition associated with this phenomenon. Cell-mediated mechanical remodeling and global compaction of ECM has mostly been studied at steady state. Our study reveals a link between a transition in cell dynamics and material microstructure as cells collectively compact collagen gels. It underscores the significance of temporal evolution of these cell-ECM systems in understanding the mechanism of such collective action and provides insights on the process from a mechanistic viewpoint. These insights can be valuable in understanding dynamic pathological processes such as, cancer progression and wound healing, as well as engineering biomaterials and regenerative tissue mimics.

Reciprocity of Cell Mechanics with Extracellular Stimuli: Emerging Opportunities for Translational Medicine

Human cells encounter dynamic mechanical cues in healthy and diseased tissues, which regulate their molecular and biophysical phenotype, including intracellular mechanics as well as force generation. Recent developments in bio/nanomaterials and microfluidics permit exquisitely sensitive measurements of cell mechanics, as well as spatiotemporal control over external mechanical stimuli to regulate cell behavior. In this review, the mechanobiology of cells interacting bidirectionally with their surrounding microenvironment, and the potential relevance for translational medicine are considered. Key fundamental concepts underlying the mechanics of living cells as well as the extracelluar matrix are first introduced. Then the authors consider case studies based on 1) microfluidic measurements of nonadherent cell deformability, 2) cell migration on micro/nano-topographies, 3) traction measurements of cells in three-dimensional (3D) matrix, 4) mechanical programming of organoid morphogenesis, as well as 5) active mechanical stimuli for potential therapeutics. These examples highlight the promise of disease diagnosis using mechanical measurements, a systems-level understanding linking molecular with biophysical phenotype, as well as therapies based on mechanical perturbations. This review concludes with a critical discussion of these emerging technologies and future directions at the interface of engineering, biology, and medicine.

Shear-Driven Solidification and Nonlinear Elasticity in Epithelial Tissues

Biological processes, from morphogenesis to tumor invasion, spontaneously generate shear stresses inside living tissue. The mechanisms that govern the transmission of mechanical forces in epithelia and the collective response of the tissue to bulk shear deformations remain, however, poorly understood. Using a minimal cell-based computational model, we investigate the constitutive relation of confluent tissues under simple shear deformation. We show that an initially undeformed fluidlike tissue acquires finite rigidity above a critical applied strain. This is akin to the shear-driven rigidity observed in other soft matter systems. Interestingly, shear-driven rigidity can be understood by a critical scaling analysis in the vicinity of the second order critical point that governs the liquid-solid transition of the undeformed system. We further show that a solidlike tissue responds linearly only to small strains and but then switches to a nonlinear response at larger stains, with substantial stiffening. Finally, we propose a mean-field formulation for cells under shear that offers a simple physical explanation of shear-driven rigidity and nonlinear response in a tissue.