Aquaporin-4 (AQP4), the main astrocytic water channel, assembles into supramolecular structures known as Orthogonal Arrays of Particles (OAPs). While AQP4-mediated water fluxes are known to influence cytoskeletal dynamics at the single-cell level, their role in collective migration remain unclear. Using primary WT and OAP-deficient (OAP-null) astrocytes, we investigated cell migration under control and proinflammatory conditions. In wound healing assays, control OAP-null astrocytes exhibited greater migratory capacity, while inflamed cells were largely immotile. Particle Image Velocimetry revealed genotype and condition-specific migration patterns, with OAP-null astrocytes displaying more linear trajectories. Cytokine treatment also reduced AQP4 and Connexin-43 expression and impaired gap-junctional communication in both genotypes. Calcein-AM quenching assays showed reduced water permeability in OAP-null cells, independently of treatment. These findings highlight the supramolecular organization of AQP4 as a key determinant of astrocyte collective migration and suggest that chronic inflammation induces a poorly coordinated phenotype through dysregulation of AQP4 and gap-junctional networks.

Directional persistence is essential for efficient immune cell migration in tissues, yet how cytoskeletal systems stabilize migration in complex three-dimensional environments remains unclear. Using intravital subcellular microscopy and quantitative analysis of membrane dynamics, we identify two spatially distinct architectures of non-muscle myosin II (NMII) that coordinate protrusion dynamics during neutrophil migration. In vivo and in collagen matrices, NMII assembles at the leading edge into lattice-like structures that are structurally and functionally distinct from rear contractile actomyosin bundles. Protrusion-resolved analyses reveal that directional persistence correlates strongly with protrusion lifetime and sustained NMII engagement, with rear NMII load showing the strongest association with protrusion persistence. Strikingly, directional migration is not determined by the abundance of favorable protrusions but by their temporal organization during migration. Pharmacological perturbations that redistribute NMII activity disrupt this temporal organization and alter migration trajectories. Together, these findings reveal that spatially distinct NMII architectures coordinate protrusion dynamics across time to stabilize directional migration in complex environments.

The brain rapidly adapts to new contexts and learns from limited data, a coveted characteristic that artificial intelligence (AI) algorithms struggle to mimic. Inspired by the mechanical oscillatory rhythms of neural cells, we developed a learning paradigm utilizing link strength oscillations, where learning is associated with the coordination of these oscillations. Link oscillations can rapidly change coordination, allowing the network to sense and adapt to subtle contextual changes without supervision. The network becomes a generalist AI architecture, capable of predicting dynamics of multiple contexts, including unseen ones. These results make our paradigm a powerful starting point for models of cognition. Because our paradigm is agnostic to specifics of the neural network, our study opens doors for introducing rapid adaptive learning into leading AI models.

Estradiol (E2), a sex steroid hormone molecule, plays a key role in regulating the actin and shape dynamics of cells in a multitude of normal and pathophysiological conditions. While cytoskeletal rearrangements, membrane dynamics, and cellular protrusions are intimately involved in cell motility and invasiveness, little is known about the impact of E2 on these processes in estrogen-dependent epithelial cells. In this study, we quantified the impact of E2 on epithelial cell shape and actin dynamics. 12Z human endometriotic epithelial cells were transfected with LifeAct-GFP and observed with lattice lightsheet microscopy, a new imaging technique fast enough to capture 3D dynamics on second timescales. E2, when applied for 24 h, significantly decreased cell circularity, solidity, and rate of change of circularity, indicating a transition to a more elongated and less variable morphology. 24-h E2 treatment also induced the formation of large membrane protrusions reminiscent of invadopodia and led to a more disordered flow of actin within those protrusions. However, these effects were not seen after 15 min of E2 treatment, suggesting that longer-term signaling is required to drive these structural changes. Together, these results suggest that E2 modulates actin polymerization and membrane protrusion dynamics in endometriotic epithelial cells and may prime them for cell invasion. This work highlights a role for hormonal signaling in mediating cytoskeletal plasticity and migratory cell phenotypes.

The neurodegenerative disorder Alzheimer’s disease (AD) is widely known for biomarkers such as amyloid beta plaques and tauopathy, as well as functional differences in memory and cognitive ability. Despite this devastating functional impact, a large body of work only focuses on molecular biomarkers of AD. In this study, we investigate collective neural dynamics in vitro and assess how network-level properties differ between a well-established model of familial AD (FAD) and a newly developed in vitro accelerated model (acAD). The new model system reliably develops the key structural characteristics of AD in three weeks, but its calcium dynamics had not been characterized previously. Spontaneous network dynamics influences information processing as part of the internal network state. Here we measure this spontaneous activity of a network of hundreds of cells in each field of view. We find that the FAD model has a larger fraction of hyperactive cells, while the acAD model displays similar characteristics to healthy cells. Additionally, the FAD model has altered cooperation between cells, losing a proportion of highly correlated cellular activities, both for fast and slow coupling among cells. The acAD model is again consistent with healthy networks. Since the acAD model does not show the same spontaneous network dysfunction seen in FAD, it can enable measurements of changes in learning and memory associated with the plasticity, rather than the structure of the internal network state.

Astrocyte ion channels, water channels, and calcium signaling play vital roles in maintaining brain homeostasis, coordinating neural network activity, supporting cognitive functions, and mediating neurovascular coupling. Disruptions in astrocyte dynamics are causally linked to gliosis occurring in pathologies, such as brain injuries and neurological diseases. Despite their importance, the potential to probe and modulate astrocyte molecular, functional, and morphological properties using intracellular light adsorbing nanoscale interfaces has been limited. In this study, we investigate the effects of fluorescent gold nanoclusters (fAuNCs) bound to bovine serum albumin (BSA) on the morphological, molecular, and functional properties of primary rat cortical astrocytes over an extended period. Treatment with fAuNCs‐BSA is not toxic over long term and induces notable morphological changes alongside alterations in whole‐cell chloride currents, cell volume regulation, and calcium signaling magnitude. To leverage the light‐absorbing properties of fAuNCs‐BSA, we expose acutely fAuNCs‐BSA‐treated astrocytes to LED blue light (λ_ex = 450 nm), which modulates potassium currents. Nanodiamond thermometry suggests that this effect results from local heating caused by photoexcitation of the nanoclusters. The potential of nanoclusters as a transformative approach for studying astrocyte function and their role in the biophysical mechanisms underlying neural communication is discussed.

Mechanical properties of biological tissues, driven by passive and active forces, play a vital role in several processes ranging from development to cancer metastasis. However, the dynamical responses of cells in tissues, subject to mechanical deformations such as shear and the associated rheological properties, are not well characterized. Here, we use three-dimensional agent-based models for normal and cancer tissues to investigate their responses to simple shear as a function of cell stiffness and stochastic active forces. In the normal epithelium, with uniform strength of active force, the yield stress as a function of shear rate follows the Herschel-Bulkley form over a range of cell volume fraction. Strikingly, the shear rate dependence and the elasticity-dependent changes in the yield stress fall on master curves upon suitable scaling. To model cancer-like behavior, a certain fraction (N_p) of cells was chosen to have enhanced activity and decreased stiffness. As N_p increases, the extent of collective cell movement decreases, transitioning from affine (collective) to non-affine (individualistic) movement, a finding that is in accord with imaging experiments. Simulations of a model of a stiff solid tumor, with radius R_s embedded in normal tissue, show that as R_s increases, the yield stress increases. Interestingly, the cells migrate collectively as R_s increases. A Gaussian Mixture Model (GMM) and a mean field theory quantitatively account for the simulation as well as experimental results on cancerous, non-cancerous, and a mixture of these two types. The combined theoretical and experimental study establishes that heterogeneity in stiffness and activity determines non-affine movements in normal and cancer tissues.

Biological rhythms coordinate adaptive sensing and computation, spontaneously demonstrating concept drift detection abilities. We demonstrate that our oscillatory learning scheme, called rhythmic sharing, can autonomously detect concept drift. Inspired by astrocytic oscillations, the algorithm has recurrent links that vary sinusoidally, producing emergent sensitivity to distributional drift. We introduce a new measure, called per-input synchrony, which harnesses this sensitivity to enable early and precise detection of hidden or complex drifts. Across three datasets, NASA C-MAPSS, SWaT, and WADI, the output of our per-input synchrony features improves detector performance, culminating in new state-of-the-art F1-scores on the complex SWaT and WADI datasets. These industrial datasets highlight the ability of our model to detect drift in highly-complex, real-world systems. Additionally, these results suggest that oscillatory link dynamics may serve as a general computational principle for adaptive sensing, with implications for neuromorphic hardware and astrocytic network biology.

Traditional artificial neural networks take inspiration from biological networks, using layers of neuron-like nodes to pass information for processing. More realistic models include spiking in the neural network, capturing the electrical characteristics more closely. However, a large proportion of brain cells are of the glial cell type, in particular astrocytes which have been suggested to play a role in performing computations. Here, we introduce a modified spiking neural network model incorporating artificial astrocytes and assess their impact on learning. We implement the network as a liquid state machine and task the network with performing a chaotic time-series prediction task. We varied the number and ratio of artificial neurons and astrocytes in the network to examine the latter units' effect on learning. We show that networks combining both neurons and astrocytes together, as opposed to neural- and astrocyte-only networks, are critical for driving learning. Interestingly, we found that the highest learning rate was achieved when the ratio between artificial astrocytes and neurons was roughly 2:1, mirroring some estimates of the ratio of biological astrocytes to neurons. Our results demonstrate that incorporating artificial astrocytes which represent information across longer timescales can alter the learning rates of neural networks, and the proportion of astrocytes to neurons should be tuned appropriately to a given task.

With laboratory experiments we investigate the ejecta of low-velocity (~m/s) impacts into multi-scale granular media and compare them against ejecta from impacts into mono-scale media. Impacts are into a 50 cm diameter galvanized washtub filled with fine sand that has larger diameter gravel buried below the surface is filmed with two high-speed cameras. The resulting ejecta curtain consists mainly of fine sand, and has a complex asymmetric structure that depends on the location and interaction of the ejecta with the larger gravel grains mixed into the sand. To characterize the highly heterogeneous ejecta curtain we combine three analysis techniques: Particle tracking measures the ejecta velocities and ejecta angles best in low density regions, while particle image velocimetry (PIV) elucidates average motion in dense regions, and histogram of oriented gradients (HOG) which captures directions of motion against a patterned background. We find significant asymmetries in the multi-scale ejecta's velocity distributions and ejection angles compared to the symmetry seen in the ejecta from impacts into mono-scale media. Our experiments show that larger grains under the surface impede and direct ejecta along preferential paths during the impact process.

Waves and oscillations are key to information flow and processing in the brain. Recent work shows that, in addition to electrical activity, biomechanical signaling can also be excitable and support self-sustaining oscillations and waves. Here, we measured the biomechanical dynamics of actin polymerization in neural precursor cells (NPC) during their differentiation into populations of neurons and astrocytes. Using fluorescence-based live-cell imaging, we analyzed the dynamics of actin and calcium signals. The size and localization of actin dynamics adjusts to match functional needs throughout differentiation, enabling the initiation and elongation of processes and, ultimately, the formation of synaptic and perisynaptic structures. Throughout differentiation, actin remains dynamic in the soma, with many cells showing notable rhythmic character. Arrest of actin dynamics increases the slower time scale (likely astrocytic) calcium dynamics by 1) decreasing the duration and increasing the frequency of calcium spikes and 2) decreasing the time-delay cross-correlations in the networks. These results are consistent with the transition from an overdamped system to a spontaneously oscillating system and suggest that dynamic actin may dampen calcium signals. We conclude that mechanochemical interventions can impact calcium signaling and, thus, information flow in the brain.

This study explores collective learning in living neural networks, focusing on group-to-group Hebbian learning, i.e., strengthening and weakening of links dependent on the precise timing of their activities. While neuronal plasticity is now well understood for single pairs of neurons, recent research has demonstrated that groups of tens of neurons are required to encode information in mammalian brains. Thus, it is critical to understand how mechanisms of plasticity, in particular spike-timing-dependent plasticity, operate at the group scale. We find that neuronal groups can reach significant plasticity after only 45 stimuli when a proper tradeoff between pulse duration and photostimulation effectiveness is chosen. Random contextual stimulation, which enhances the reliability of response for the targeted neuronal groups, is necessary for rapid network-level Hebbian learning. By demonstrating enhanced learning in the presence of random contextual activity, this study underscores the highly cooperative character of neurons and the importance of investigating learning, information flow, and memory formation at the network scale.

Motility is a critical function of the gastrointestinal (GI) system governed by neurogenic and myogenic processes. Due to its major role in maintaining homeostasis, overlapping mechanisms have evolved for its adaptive operation including modulation by the central nervous system (CNS), enteric nervous system (ENS) and intrinsic pacemaker cells. Our understanding of the modulatory mechanisms that underlie intestinal motility remains incomplete. Crayfish provide a tractable ex vivo model to study the interplay between CNS and neurochemical regulation of GI motor patterns. Our study investigated the effects of CNS denervation and exogenously applied serotonin (5-HT) on crayfish hindgut motility. Multiscale spatial measurements showed stable motility parameters throughout 90 min of control conditions. Denervation, i.e. separating the gut from the CNS, resulted in a significant decrease in the magnitude and synchrony of hindgut contractions, while preserving the underlying frequency and directional bias of the waves. Subsequent application of 5-HT to the denervated preparation enhanced motility but disrupted spatiotemporal coordination. Treatment with TTX (a sodium channel blocker) had minor impacts on motility metrics, indicating a prominent role of myogenic mechanisms. Our model provides a multiscale analysis framework to dissect CNS and interrelated neurochemistry contributions to GI motor dynamics.

In natural environments, cells move in the presence of multiple physical and chemical guidance cues. Using a model system for such guided cell migration, Dictyostelium discoideum (Dicty), we investigate how chemical and physical signals compete in guiding the motion of cell groups. In Dicty cells, chemical signals can lead to collective streaming behavior, in which cells follow one another head-to-tail and aggregate into clusters of ∼10^{5} cells. We use experiments and numerical simulations to show that streaming and aggregation can be suppressed by the addition of a physical guidance cue of comparable strength to the chemical signals, parallel nanoridges. The bidirectional character of physical guidance by ridges is a determining factor in the suppression of streaming and aggregation. Thus, combining multiple types of guidance cues is a powerful approach to trigger or explain a broad range of collective cell behaviors.

Reactive Oxygen Species (ROS) and the associated condition of excess ROS, oxidative stress, has been implicated in a number of diseases including neurodegeneration. However, ROS are also crucial second messengers with beneficial impacts. Within neural cells, ROS signals are known to impact maturation of cells as well as memory and learning. Photobiomodulation (PBM), the use of light to impact cells/tissue, is a promising way to noninvasively modulate ROS. This study investigates the effects of PBM using 370 nm light to increase ROS levels in human neural progenitor cells (hNPC), and study potential impacts on calcium dynamics. We find 370 nm light to be effective at inducing ROS within hNPC. The photoinduction of ROS only impacts ROS levels in illuminated cells, with no measurable signal relay to non-illuminated cells within the acute time period we examined. The increase in ROS generated by our UV light exposure creates elevated basal levels of calcium, but does not impact spontaneous calcium signaling in networks of hNPC cells.

The social amoeba Dictyostelium discoideum is a standard model system for studying cell motility and formation of biological patterns. D. discoideum cells form protrusions and migrate via cytoskeletal reorganization driven by coordinated waves of actin polymerization and depolymerization. Assembly and disassembly of actin filaments are regulated by a complex network of biochemical reactions, exhibiting sensitivity to external physical cues such as stiffness, composition and surface topography of the extracellular matrix, as well as the presence of external electric fields. In this study, we investigate whether the cellular microenvironment, and in particular the presence of electric fields and the nano-topography type, can be directly inferred from images or videos of actin waves. We employ three machine learning techniques to analyze the resulting videos: dictionary learning, scattering transforms, and optical flow. We predict the type of the extracellular environment by observing actin waves frame-by-frame and identifying key visual features that help classify cell motion by the microenvironment type. Our analysis reveals that the decomposition of static images into an adaptive basis of visual primitives provides a robust approach to classifying cells by the nano-topography type. In contrast, predicting whether cells are moving under the influence of an external electric field requires tracking of stable cellular features such as corners and edges over a period of time. We expect our computational approach to be useful in many settings where non-trivial collective dynamics is observed with the help of fluorescent labeling and video microscopy.

The development of axons and dendrites (neurites) in a neural circuit relies on the dynamic interplay of cytoskeletal components, especially actin, and the integration of diverse environmental cues. Building on prior findings that actin dynamics can serve as a primary sensor of physical guidance cues, this work investigates the role of nanotopography in modulating and guiding actin waves and neurite-tip dynamics during early neural circuit development. Although actin dynamics is well known to contribute to pathfinding in wide axonal tips, typically referred to as growth cones, we also observe dynamic actin remodeling throughout neurites and at other, narrower, neurite tips. We find that actin-wave speeds do not change significantly in the first 2 weeks of neurite development on flat substrates, but decrease over the same period in neurites on nanoridges. The ability of nanoridges to guide actin waves and the neurite-tip direction also decreases as neurites mature, both for narrow tips and wide growth cones. This change in responsiveness to physical guidance cues with neuronal maturation may impact the regenerative capacity of developing neural cells that are inserted into mature brains.

Destruction of myelin internodes, oligodendrocyte (OL) apoptosis, and axonal degeneration characterize diseased or aged central nervous systems. While OLs can partially regenerate myelin sheaths, the remyelination process ultimately fails. Tissue mechanical and physical properties, such as stiffness and axonal curvature, play a role in this process. However, the complexity of existing models has hindered studies of OL mechanobiology. Here, a tissue-engineered model is presented to investigate the impact of stiffness and axonal diameter on OL myelination. The model consists of poly(dimethylsiloxane) micropillars with biologically relevant diameters (1-5 um), tunable rigidity, and amenable for surface functionalization. The optimized method enables the production of high-aspect-ratio, transparent micropillar arrays, in a reproducible and scalable system, serving as surrogate axons. Additionally, new protocols for quantifying myelin formation are introduced, which can be adapted to any myelination studies. Softer micropillars accelerate OL differentiation, while rigid ones promote the maintenance of mature OL states. Wrapping of OLs increased with micropillar diameter on rigid substrates, but not on softer ones, suggesting a complex interplay between curvature and rigidity. These processes involve calcium-sensitive channels, histone deacetylases, and microtubules dynamics. The proposed platform constitutes a versatile and user-friendly system, with applications from fundamental myelin research to drug discovery. ### Competing Interest Statement The authors have declared no competing interest.

IntroductionPhotomodifiable azopolymer nanotopographies represent a powerful means of assessing how cells respond to rapid changes in the local microenvironment. However, previous studies have suggested that azopolymers are readily photomodified under typical fluorescence imaging conditions over much of the visible spectrum. Here we assess the stability of azopolymer nanoridges under 1-photon and 2-photon imaging over a broad range of wavelengths.MethodsAzopolymer nanoridges were created via microtransfer molding of master structures that were created using interference lithography. The effects of exposure to a broad range of wavelengths of light polarized parallel to the ridges were assessed on both a spinning-disk confocal microscope and a 2-photon fluorescence microscope. Experiments with live Dictyostelium discoideum cells were also performed using alternating cycles of 514-nm light for photomodification and 561-nm light for fluorescence imaging.Results and DiscussionWe find that for both 1-photon and 2-photon imaging, only a limited range of wavelengths of light leads to photomodification of the azopolymer nanotopography. These results indicate that nondestructive 1-photon and 2-photon fluorescence imaging can be performed over a considerably broader range of wavelengths than would be suggested by previous research.