Session W04: Collective Behavior in Biology I

Sensitive thermometry through criticality in voltage-coupled TRP channels

Sensory systems must integrate information from many individually noisy receptors. Here we study the TRP family of ion channels, which are sensitive to a range of physiological stimuli in animals, including hot and cold temperatures. While individual channels open over a 3-5 Kelvin temperature range, thermal imaging organs found in bats and snakes respond to infrared radiation that heats them by just a few milli-Kelvin. Thus, these organs have responses thousands of times more sensitive than individual molecular sensors, but the mechanism through which they achieve this dramatic amplification is unknown. The simple quantifiable function of these organs makes them ideal for studying how information is integrated into coherent and sensitive responses. Here we propose a theoretical model wherein sensitivity arises due to the diverging susceptibility close to a critical point. Individual channels are embedded into a larger dynamical system with positive feedback from the collective response. These dynamics, combined with slower adaptation naturally tunes the dynamical process to the vicinity of a bifurcation critical point. In our model, positive feedback is mediated through voltage, which indeed activates all TRP channels, irrespective of whether they are hot or cold sensitive channels. Our work could shed light into a design principle common for many sensory systems and TRP channels in particular.   

Tissue confinement governs cell size regulation in epithelium

While populations of single-celled organisms increase exponentially, animal cell growth must be coupled to organism growth for tissues to maintain their structure. These spatial constraints lead to a different regime of growth and division regulation known as contact inhibition of proliferation. We still lack a general framework to describe contact inhibition across different biological systems. Here we use model epithelial monolayers with varying spatial constraints to explore how contact inhibition affects cell growth and division. We introduce a concept of tissue confinement which describes the extent to which spatial constraints suppress cell growth in different tissues. Interestingly, confinement has no effect on cell division leading to a decoupling between rates of cell growth and division. In confined tissues cell division outpaces growth causing cell size to decrease. However, when cell size decreases below a specific value cell division becomes arrested. This final cell size is near a physical limit set by the amount of space occupied by DNA in the cell. By perturbing cell division regulation, it is possible to push cells closer to this limit, however, this leads to DNA damage suggesting loss of size regulation could play a role in the development of cancer.

Identifying key network features coordinating robust collective signaling oscillations

A major challenge in understanding emergent cellular behaviors such as collective oscillations is identifying which features in single-cell signaling networks are essential for robustly coordinating population phenomena. To address this, we focus on one of the most notable examples, the starvation response of the social amoeba Dictyostelium discoideum (Dicty), where cells use collective intracellular cyclic AMP (cAMP) oscillations to mediate a transition from a unicellular to a multicellular state. Taking advantage of recent experimental measurements that a single Dicty cell displays an internal cAMP spike with characteristic height and length in response to an external cAMP input, we were able to normalize the response and time scales of five major models describing group cAMP oscillations in Dicty. This allows us to directly compare models based on how well they reproduce recent experimental data on both single-cell and population behaviors. Our systematic analysis suggests that excitability and fold-change detection are two key features in single-cell signaling networks that coordinate robust population-wide behaviors. This work sheds light on the basic principles in single-cell signaling networks that drive robust population oscillations in biochemically-coupled systems.   

Modeling collective cell migration on substrates with topological defects

Collective movement, arrangement and proliferation of cell monolayers are important for wound healing and tissue development, and an example of active matter physics in biology. Recent experiments have highlighted the importance of liquid crystal order and topological defects within these layers, in particular suggesting that +1 defects have a role in organizing tissue morphogenesis. In this work we perform 2D active Monte Carlo simulations to investigate cell organization, motion and proliferation on a substrate guided with ridges that induce a +1 defect. This models experiments on fibroblasts from the Serra group, who found that cells align with the ridges, and that cells are denser and more isotropic toward the center of the defect. We model different cell types as self-propelled deformable ellipses that interact via a Gay-Berne potential. We propose two potential mechanisms that can lead to increased density at the defect core: first, collective spiraling motion driven by coherent migration, and second, the regulation of proliferation by cell area and aspect ratio. By observing individual cell movements and analyzing timescales of processes in experiments we argue that the proliferation mechanism is more compatible with experiments.   

Synchrony and causality in stimulated communicating cell populations

Cells routinely respond to both input stimuli and communication from neighbors, and it is still poorly understood how these two types of signals are integrated at the population level. In the case of temporal stimuli, the presence of cell-cell communication raises several possibilities: non-communicating cells may exhibit asynchronous responses to the stimulus due to cell-to-cell heterogeneity, whereas strongly communicating cells may exhibit highly synchronized responses to the stimulus. However, at intermediate communication strength it is unclear what temporal patterns may emerge, e.g., due to particularly responsive cells triggering subsequent responses in their neighbors. Here we investigate the relationships between cell-cell communication, synchrony, and causality using a simple excitable-response model to describe experiments performed in monolayers of kisspeptin neurons stimulated by time-varying ATP concentrations. Surprisingly, the model predicts that as cell-cell communication strength is increased, the cell responses first desynchronize before ultimately becoming fully synchronous. This prediction is confirmed by perturbing the degree of communication in the experiments. Our work elucidates the complex interplay between cell-cell communication and synchrony in interacting multicellular systems responding to time-varying signals.   

Developmentally driven formation and dissolution of living chiral crystals

The emergent dynamics exhibited by collections of living organisms often shows signatures of symmetries that are broken at the single-organism level. At the same time, early organism development itself is accompanied by a sequence of symmetry breaking events that eventually establish the body plan. Combining these key aspects of collective phenomena and embryonic development, we describe here the spontaneous formation of hydrodynamically stabilized active crystals made of hundreds of starfish embryos during early development. As development progresses and embryos change morphology, crystals become increasingly disordered and eventually stop forming. We introduce a minimal hydrodynamic model that is fully parameterized by experimental measurements of single embryos and can quantitatively describe the stability, formation and rotation of emerging living crystals. A detailed analysis of developmental changes in the swimming properties of single embryos and associated sources of effective noise in embryo-embryo interactions provides insights into the mechanisms that eventually lead to the dissolution of clusters. Our work thereby quantitatively connects developmental symmetry breaking events on the single-embryo level with the remarkable macroscopic properties of a novel living chiral crystal system.   

Optical flow-based characterization of serotonergic modulation of crayfish hindgut motility

Peristalsis is a series of gastrointestinal (GI) contractions carrying food through the digestive tract. Due to its major role in maintaining homeostasis, several drivers have evolved for its smooth functioning: from motor input from the central nervous system (CNS) or enteric nervous system (ENS) lining the gut. Recent studies have shed light on the shared neurochemistry and structure between these, which have often led to comorbid GI and neurological disorders while dysfunctional. To fully understand this interaction between ENS and CNS, it is crucial to study a shared signaling pathway. Receptors for Serotonin, a common neurotransmitter, were previously observed to be embedded along the lining of the gut, suggesting its role in muscular contraction. However, it is not clear to what degree peristalsis is driven by local Serotonin concentration as compared to the CNS input. To explore this relationship further, we study a simple crayfish model and capture its hindgut motility under denervation and serotonergic excitation. Using optical flow as a motion detection tool, we identify and characterize patterns of gut motility. We study whether adding Serotonin could compensate for the lack of CNS input and how these motility patterns differ from the typical gut motion.   

The role of positional information in collective gradient sensing

Eukaryotic cells sense chemical gradients to decide where and when to move. While individual cells struggle to sense shallow gradients, clusters of cells increase their sensing accuracy by integrating measurements of concentration made across the cluster. However, the integration of these measurements may be constrained in the case where cells have limited information on their position within the cluster. We apply maximum likelihood estimation to study gradient sensing accuracy when cells have limited positional information. We compare our results with a tug-of-war model in which cells respond to the gradient by a collective guidance mechanism without relying on their positional information, but only using cells at the edge of the cluster. As cell positional uncertainties increase, a trade-off occurs in which the tug-of-war model responds more accurately to the chemical gradient despite using fewer cells. However, for sufficiently large cell groups or shallow chemical gradients, the tug-of-war model will always be suboptimal to one that integrates information from all cells, even if positional uncertainty is high.   

Self tuned criticality amplifies signals in bacterial chemotaxis

Many sensory systems must amplify weak signals that arrive at single receptors. One method to achieve such sensitivity is to embed receptors into a larger system tuned close to a critical point, such that small changes to input signal are amplified into a large change in output behavior. Here, we propose a model where chemoreceptors act as the edges for the percolation of activity between signaling proteins. This system achieves high amplification by exploiting the diverging susceptibility found near the percolation transition. Output from the signaling proteins is fed back into the system to drive it towards criticality and maintain amplification over a wide range of input signals. Our model provides a mechanistic explanation of the high signal amplification at the cost of large noise which occurs in E. coli chemotaxis signaling.   

Relating division, migration, and cell signaling during collective cell migration

Epithelial monolayers provide critical barrier functions for multicellular organisms and are among the most well-studied collective systems, maintaining strong cell-cell adhesion and exhibiting coordinated outwards growth. While much is known about the biomechanics of how force generation and substrate stiffness affect such collective motion, how cells collectively share information during migration is less clear. The ERK signaling pathway is key for many forms of cell-cell signaling, and here we study the interplay between ERK signaling, collective migration, and tissue proliferation in epithelia. Using the ERK-KTR reporter, we show that tissue migration induces both periodic, rapid ERK signals propagating back into the tissue from the leading edge, as well as slower domains of ERK activity propagating towards the leading edge. We hypothesize that this slower ERK signal is correlated with the spatiotemporal cell proliferation pattern. Complementing these studies are additional experiments we have performed on E-cadherin substrates that mimic cell-cell adhesion signaling. These substrates show extremely low motility due to a low active fluctuation state and further regulate cell cycle dynamics by mimicking cell-cell information.   

Tunning to Criticality in the Cochlea with Local Information

The human ear can detect sounds over a vast frequency range with incredible sensitivity. To explain these properties, it has been suggested that individual hair cells operate near an oscillatory instability, a so-called Hopf bifurcation. However, resonating modes are determined primarily by the passive hydrodynamics and mechanical stiffness of the cochlea, with active processes in individual hair cells only serving to counteract dissipative forces. In this picture, every frequency must have a mode tuned close to a Hopf bifurcation, but each hair cell impacts every mode to varying extents. It is an open question how these coupled hair cells could tune their behavior so as to bring the entire cochlea to this line of Hopf bifurcations. To approach this question, we develop a model where hair cells collectively tune their active processes while only using information measurable from their location on the cochlea. Our work suggests a new perspective on the hearing mechanism: Instead of individual hair cells being critical, their concerted individual action tunes the whole cochlea to a critical point revealing a collective organization principle in the inner ear.   

Defining "success" for pairwise maximum entropy models

Statistical physics approaches to understanding complex systems have been remarkably successful. Numerous scientific fields are undergoing a data revolution, and high accuracy simple models for complex phenomena are in high demand. In particular, the maximum entropy principle as formulated by Jaynes provides a powerful framework to capture the collective nature of an ensemble while keeping all but a few measurement constraints maximally random. For binary variables, these models are equivalent to Ising models with competing interactions. Despite their ubiquity, deciding whether these models "work" remains a challenge. Here, we study the activity of 1000+ simultaneously active cells in the brain of mice as they navigate a virtual environment. Leveraging the massive scale of these data, we compared 900 different models for different subgroups of 100 neurons each. All neurons are selected out of the same hippocampal population but randomly from gradually increasing spatial field. We find that the more spatially contiguous the subgroup is, the better the pairwise model captures its collective behavior. Systematic comparison of many different predictions across these many examples allows us to draw the boundaries for success of these models.