In non-equilibrium (active) systems, increased driving is commonly assumed to produce proportionally greater energy dissipation. Using picowatt-sensitive calorimetry, entropy-production analysis, rheology, and microscopy in reconstituted actomyosin networks, we show that this relationship breaks down as the material reorganizes under motor activity. Dissipation initially increases with myosin abundance but subsequently decreases despite continued network stiffening, indicating that energetic cost becomes regulated by the evolving mechanical state rather than actuator abundance alone. Comparing catch-bond (α-actinin) and slip-bond (fascin) crosslinked networks reveals that bond mechanics shift the critical motor concentration at which this transition occurs, marking the onset of state-dependent energy conversion. Although these networks generate distinct active stresses and mechanical states, they can exhibit comparable dissipation rates, revealing that state-dependent energy conversion can produce degenerate dissipation, whereby mechanically distinct states exhibit similar energetic costs. These findings demonstrate that dissipation in active materials is governed not only by the magnitude of driving but also by the mechanical state that emerges in response to that driving, providing an experimental example of state-dependent energy conversion far from equilibrium.

The cytoskeleton is a quintessential active material, in which interpenetrating networks (IPNs) of filaments are restructured by molecular motors to drive essential cellular functions. However, the interplay between the mechanics, structure, and dynamics of these composite networks remains poorly understood. Here, we use confocal fluorescence microscopy and particle image velocimetry to elucidate the dynamics of IPNs composed of rigid microtubules and semiflexible actin filaments driven by tetrameric kinesin motors. We discover that both microtubule speed and shear are maximized at intermediate motor concentrations and suppressed at high motor concentrations, corresponding to a switch from extensile to contractile dynamics. We demonstrate that the emergent fast dynamics, arising from extensile shearing of filaments, directly correlates with peak network stiffness. Conversely, the transition to slow, contractile dynamics at higher motor loads is driven by increased network drag. Our results suggest that the cytoskeleton can exquisitely tune its active motion by sculpting its mechanical properties, enabling the facile switching between different dynamical states required for cellular processes like division and migration.

Living matter consumes energy to build and actuate dynamic protein machines, but the way energy enters these systems is often assumed to be externally imposed and independent of state. This review advances a unifying perspective, that across diverse cytoskeletal active matter, energy injection itself is state dependent, regulated by the mechanical and organizational configuration of the system, rather than acting as a fixed external bath. We synthesize recent experimental evidence, from excitable cortical waves to reconstituted actomyosin networks , showing that feedback between system state and energy input leads to nonmonotonic dissipation, selective energy routing, and relaxation-driven contractility. By organizing these findings within a common framework, we highlight how state-dependent energy injection challenges classical nonequilibrium assumptions and opens new questions for the design and control of active matter.

A modular programming framework for controlling microtubule-based active matter using light is introduced, enabling the precise design and manipulation of dynamic micrometre-scale fluid flows for tasks such as mixing, transport and separation in microfluidic applications.

Self-organized criticality can occur in earthquakes, avalanches and biological processes, and is characterized by intermittent, scale-free energy dissipation. In living cells, the actin cytoskeleton undergoes dynamic structural reorganization, particularly during migration and division, where molecular motors generate mechanical stresses that drive large dissipative events. However, the mechanisms governing these critical transitions remain unclear. Here we show that cytoskeletal criticality emerges from the interplay between F-actin organization and active stress generation. Our study focuses on a minimal actomyosin system in vitro, which is composed of F-actin filaments, myosin II motors and nucleation-promoting factors. By systematically varying the actin connectivity and nematic order, we demonstrate that ordered and sparsely connected networks exhibit exponential stress dissipation, whereas disordered and highly connected networks show heavy-tailed distributions of energy release and the 1/f noise characteristic of self-organized criticality. Increased disorder leads to stress localization, shifting force propagation into stiffer mechanical modes, reminiscent of Anderson localization in condensed-matter systems. Furthermore, we show that network architecture directly regulates the myosin II filament size, establishing a chemical-mechanical feedback loop that modulates criticality. Our findings provide insights into the collective cytoskeletal dynamics, energy localization and cellular self-organization.

The architecture of the actin cortex determines the generation and transmission of stresses, during key events from cell division to migration. However, its impact on myosin-induced cell shape changes remains unclear. Here, we reconstitute a minimal model of the actomyosin cortex with branched or linear F-actin architecture within giant unilamellar vesicles (GUVs, liposomes). Upon light activation of myosin, neither the branched nor linear F-actin architecture alone induces significant liposome shape changes. The branched F-actin network forms an integrated, membrane-bound "no-slip boundary" -like cortex that attenuates actomyosin contractility. By contrast, the linear F-actin network forms an unintegrated "slip boundary" -like cortex, where actin asters form without inducing membrane deformations. Notably, liposomes undergo significant deformations at an optimized balance of branched and linear F-actin networks. Our findings highlight the pivotal roles of branched F-actin in force transmission and linear F-actin in force generation to yield membrane shape changes. Investigating shape changes of GUVs with actomyosin cortex suggests that neither branched nor linear F-actin alone can induce the shape change; the combination of branched and linear F-actin is essential to yielding membrane shape changes.

Substrate geometry and topography alter cellular F-actin organization through the activation of adhesionassociated intracellular signaling, playing pivotal roles in cell adhesion, migration, and division. However, it remains unclear how the deformation of the cell membrane under microenvironments impacts F-actin organization. Here, we reconstitute a minimal model of the actin cortex within adherent giant unilamellar vesand local topography. For global geometry, liposome shapes were altered to triangular upon adhesion to the triangular pattern. For local topography, 1-mm-wide ridges induce local wavy membrane deformations at the bottom of the adherent liposomes. We find that neither substrate feature alters the organization of the branched F-actin network, whereas the linear F-actin network significantly changes its orientation and chirality. Our findings reveal how microenvironmental geometry and topography impact F-actin organization, providing insights into cellular geometry sensing mediated by cytoskeletal organization.

Mechanical work serves as the foundation for dynamic cellular processes, ranging from cell division to migration. A fundamental driver of cellular mechanical work is the actin cytoskeleton, composed of filamentous actin (F-actin) and myosin motors, where force generation relies on adenosine triphosphate (ATP) hydrolysis. F-actin architectures, whether bundled by crosslinkers or branched via nucleators, have emerged as pivotal regulators of myosin II force generation. However, it remains unclear how distinct F-actin architectures impact the conversion of chemical energy to mechanical work. Here, we employ in vitro reconstitution of distinct F-actin architectures with purified components to investigate their influence on myosin ATP hydrolysis (consumption). We find that F-actin bundles composed of mixed polarity F-actin hinder network contraction compared to non-crosslinked network and dramatically decelerate ATP consumption rates. Conversely, linear-nucleated networks allow network contraction despite reducing ATP consumption rates. Surprisingly, branched-nucleated networks facilitate high ATP consumption without significant network contraction, suggesting that the branched network dissipates energy without performing work. This study establishes a link between F-actin architecture and myosin energy consumption, elucidating the energetic principles underlying F-actin structure formation and the performance of mechanical work. F-actin architecture modulates transmission and generation of stresses in cells, yet its impact on myosin ATP hydrolysis remains unknown. The authors perform experiments measuring myosin ATP hydrolysis rates, showing that F-actin architecture can control myosin energy consumption.

Living systems are driven far from thermodynamic equilibrium through the continuous consumption of ambient energy. In the cell cortex, this energy is invested in the formation of diverse patterns in chemical and mechanical activities, whose spatial and temporal dynamics determine the cell phenotypes and behaviours. How cells partition internal energy between these activities is unknown. Here we measured the entropy production rate of both chemical and mechanical subsystems of the cell cortex across a variety of patterns as the system is driven further from equilibrium. We do this by manipulating the Rho GTPase pathway, which controls the cortical actin filaments and myosin-II. At lower levels of GTPase-activating protein expression, which produce pulses or choppy Rho and actin filament waves, energy is proportionally partitioned between the two subsystems and is subject to the constraint of Onsager reciprocity. Within the range of reciprocity, the entropy production rate is maximized in choppy waves. As the cortex is driven into labyrinthine or spiral travelling waves, reciprocity is broken, marking an increasingly differential partitioning of energy and an uncoupling of chemical and mechanical activities. We further demonstrate that energy partitioning and reciprocity are determined by the competing timescales between chemical reaction and mechanical relaxation.

Filamentous-actin (F-actin) crosslinking within the cell cytoskeleton mediates the transmission of mechanical forces, enabling changes in cell shape, as occurs during cell division and cell migration. Crosslinking by actin binding proteins (ABPs) generally increases the connectivity of the F-actin network, but also increases network rigidity. As a result, there is a narrow range in the concentration of crosslinker protein at which F-actin networks are both connected and labile. Another ABP, cofilin, severs F-actin filaments at high pH through increasing their bending flexibility and concentrating mechanical stress, inducing fragmentation. By contrast, at lower pH, cofilin increases filament flexibility yet does not sever. Instead, it forms disulfide bonds, which crosslink F-actin into bundles, and bundles into networks. Here, we combine light microscopy and rheology to determine the impact of two potentially opposing effects on the mechanics of F-actin networks-increased flexibility at the filament level, and increased connectivity at the network level. Indeed, by linear rheology, we find that these mechanisms are counterbalanced, such that cofilactin network moduli are only weakly dependent on cofilin concentration over a broad range, in contrast to the dramatic stiffening that occurs with F-actin crosslinking protein. Further, by nonlinear rheology, the network stiffens at a higher stress than crosslinking protein, indicative of a broader range in which the material remains flexible. These results may enable F-actin networks to increase connectivity without heavy penalties to rigidity, and thus provide a new route to modulating active polymer mechanics unseen using traditional F-actin accessory proteins.

During cancer pathogenesis, cell-generated mechanical stresses lead to dramatic alterations in the mechanical and organizational properties of the extracellular matrix (ECM). To date, contraction of the ECM is largely attributed to local mechanical stresses generated during cell invasion, but the impact of “elastocapillary” effects from surface tension on the tumor periphery has not been examined. Here, we embed glioblastoma cell spheroids within collagen gels, as a model of tumors within the ECM. We then modulate the surface tension of the spheroids, such that the spheroid contracts or expands. Surprisingly, in both cases, at the far-field, the ECM is contracted toward the spheroids prior to cellular migration from the spheroid into the ECM. Through computational simulation, we demonstrate that contraction of the ECM arises from a balance of spheroid surface tension, cell–ECM interactions, and time-dependent, poroelastic effects of the gel. This leads to the accumulation of ECM near the periphery of the spheroid and the contraction of the ECM without regard to the expansion or contraction of the spheroid. These results highlight the role of tissue-level surface stresses and fluid flow within the ECM in the regulation of cell–ECM interactions.

The spatial and temporal dynamics of forces in cells coordinate essential behaviors like division, polarization, and migration. While intracellular signaling initiates contractile ring assembly during cell division, how mechanical forces coordinate division and their energetic costs remain unclear. Here, we develop an in vitro model where myosin-induced stress drives division-like shape changes in giant unilamellar vesicles (GUVs, liposomes). Myosin activity is controlled by light patterns globally or locally at the equator. Global activation causes slow, shallow cleavage furrows due to a tug-of-war between the equatorial and polar forces. By contrast, local activation leads to faster, deeper, and symmetric division as equatorial forces dominate. Dissociating the actin cortex at the poles is crucial for inducing significant furrowing. During furrowing, actomyosin flows align actin filaments parallel to the division plane, forming a contractile ring-like structure. Mechanical power is not greatest during contraction, but is maximized just before furrowing. This study reveals the quantitative relationship between force patterning and mechanical energy during division-like shape changes, providing insights into cell division mechanics. While biochemical regulation initiates cell division, how mechanical forces ensure successful division remains unclear. Here, the authors develop a biomimetic model of dividing cells to characterize the cytoskeleton’s mechanical power prior to division.

To migrate, divide, and change shape, cells must regulate the mechanics of their periphery. The cell surface is a complex structure that consists of a thin, contractile cortical actin network tethered to the plasma membrane by specialized membrane-to-cortex attachment (MCA) proteins. This active and constantly fluctuating system maintains a delicate mechanochemical state which permits spontaneous polarization and shape change when needed. Combining in silico, in vitro, and in vivo experiments we show how membrane viscosity and MCA protein length regulate cortical dynamics. We reveal a novel mechanism whereby caging of linker proteins in the actin cortex allows for the amplification of small changes in these key parameters, leading to major alterations of cortical contractility. In cells, this mechanism alone gives rise to symmetry breaking phenomena, suggesting that local changes in lipid composition, in combination with the choice of MCA proteins, contribute to the regulation of cellular morphogenesis and function. ### Competing Interest Statement The authors have declared no competing interest.

During mesenchymal migration, F-actin protrusion at the leading edge and actomyosin contraction determine the retrograde flow of F-actin within the lamella. The coupling of this flow to integrin-based adhesions determines the force transmitted to the extracellular matrix and the net motion of the cell. In tissues, motion may also arise from convection, driven by gradients in tissue-scale surface tensions and pressures. However, how migration coordinates with convection to determine the net motion of cellular ensembles is unclear. To explore this, we study the spreading of cell aggregates on adhesive micropatterns on compliant substrates. During spreading, a cell monolayer expands from the aggregate towards the adhesive boundary. However, cells are unable to stabilize the protrusion beyond the adhesive boundary, resulting in retraction of the protrusion and detachment of cells from the matrix. Subsequently, the cells move upwards and rearwards, yielding a bulk convective flow towards the centre of the aggregate. The process is cyclic, yielding a steady-state balance between outward (protrusive) migration along the surface, and ‘retrograde’ (contractile) flows above the surface. Modelling the cell aggregates as confined active droplets, we demonstrate that the interplay between surface tension-driven flows within the aggregate, radially outward monolayer flow and conservation of mass leads to an internal circulation.

Growth and turnover of actin filaments play a crucial role in the construction and maintenance of actin networks within cells. Actin filament growth occurs within limited space and finite subunit resources in the actin cortex. To understand how filament growth shapes the emergent architecture of actin networks, we developed a minimal agent-based model coupling filament mechanics and growth in a limiting subunit pool. We find that rapid filament growth induces kinetic trapping of highly bent actin filaments. Such collective bending patterns are long-lived, organized around nematic defects, and arise from competition between filament polymerization and bending elasticity. The stability of nematic defects and the extent of kinetic trapping are amplified by an increase in the abundance of the actin pool and by crosslinking the network. These findings suggest that kinetic trapping is a robust consequence of growth in crowded environments, providing a route to program shape memory in actin networks.

Membraneless organelles, composed of protein and nucleic acids, alter the biochemical and physical landscape of the cell. While specific membraneless organelles are found in stereotypical locations, little is known about the physical mechanisms that guide their positioning. Here, we investigate how stress granules, a type of cytoplasmic membraneless organelle, establish their stereotypical perinuclear positioning. We find that actin and microtubules play complementary roles. Lamellar actin confines stress granules, and its retrograde flow drives them toward the cell center. Microtubules, in turn, adhere to stress granules through capillary interactions, which tend to concentrate stress granules in micro-tubule rich regions near the nucleus. Similar physical mechanisms are likely to play a role in the positioning of other membraneless organelles.

The organization of actin filaments (F-actin) into crosslinked networks determines the transmission of mechanical stresses within the cytoskeleton and subsequent changes in cell and tissue shape. Principally mediated by proteins such as α-actinin, F-actin crosslinking increases both network connectivity and rigidity, thereby facilitating stress transmission at low crosslinking yet attenuating transmission at high crosslinker concentration. Here, we engineer a two-dimensional model of the actomyosin cytoskeleton, in which myosin-induced mechanical stresses are controlled by light. We alter the extent of F-actin crosslinking by the introduction of oligomerized cofilin. At pH 6.5, F-actin severing by cofilin is weak, but cofilin bundles and crosslinks filaments. Given its effect of lowering the F-actin bending stiffness, cofilin- crosslinked networks are significantly more flexible and softer in bending than networks crosslinked by α-actinin. Thus, upon local activation of myosin-induced contractile stress, the network bends out-of-plane in contrast to the in-plane compression as observed with networks crosslinked by α-actinin. Here, we demonstrate that local effects on filament mechanics by cofilin introduces novel large-scale network material properties that enable the sculpting of complex shapes in the cell cytoskeleton.