Cell condensation initiates organogenesis: the role of actin dynamics in supracellular self-organizing process

Abstract

The emergence of complex tissue architectures from homogeneous stem cell condensates persists as a central enigma in developmental biology. While biochemical signaling gradients have long dominated explanations of organ patterning, the mechanistic interplay between tissue-scale forces and thermodynamic constraints in driving symmetry breaking remains unresolved. This review unveils supracellular actin networks as mechanochemical integrators that establish developmental tensegrity structures, wherein Brownian ratchet-driven polymerization generates patterned stress fields to guide condensate stratification. Central to this paradigm is the dynamic remodeling of actin branches, which transduce mechanical loads into adaptive network architectures through force-modulated capping kinetics and angular reorientation. Such plasticity enables fluid-to-solid phase transitions, stabilizing organ primordia through viscoelastic microdomain formation. Crucially, these biophysical processes are functionally coupled with metabolic reprogramming events, where cytoskeletal strain modulates glycolytic flux and nuclear mechanotransduction pathways to inform differentiation decisions, forging a feedback loop between tissue mechanics and cellular fate specification. Building on these insights, we argue that limitations in current organoid self-organization may originate from incomplete reconstitution of actin-mediated mechanical coherence, and modeling of heterogeneous mesenchymal condensation dynamics offers a strategic framework to decode self-organization trajectories, bridging developmental principles with regenerative design. By synthesizing advances from molecular biophysics to tissue mechanics, this work reframes organogenesis not as a hierarchy of molecular commands, but as an emergent continuum where biochemical, mechanical, and thermodynamic constraints coevolve to sculpt living architectures.

Background

Organogenesis is a critical process in biological morphogenesis, establishing mature structures and functions. It begins with stem/progenitor cell condensation, a dynamic process involving the regulation of signaling pathways, extensive cell-cell interactions, and the spatial organization of cells and extracellular matrix into distinct tissues and functional organs, such as mesenchymal condensation for odontogenesis and skeletogenesis. Indeed, mesenchymal progenitors exhibit an intrinsic ability to aggregate, forming high-density condensates that give rise to the organ bud in the early stages of development [1–4]. Accurate condensation is essential for organogenesis to follow the correct developmental trajectory, leading to the formation of precise functional tissues [5, 6]. However, a key question remains poorly understood: how do homogeneous stem cell condensates undergo self-organized remodeling to form organs with complex structures? Recent advances in genomic sequencing have provided insights into the spatiotemporal regulation of signaling pathways during early organogenesis [7, 8], but they have not yet elucidated how symmetry breaking is initiated within stem cell condensates to achieve robust pattern formation. Unraveling this mechanism requires a comprehensive understanding of the coupled mechanochemical regulatory networks involved.

Increasing evidence suggests that the mechanical state of the organ bud is crucial for organogenesis, and large-scale coordination requires feedback between mechanical and biochemical processes [9, 10]. This is evident in the development of skin appendages, limb skeleton formation, and tooth germ development [11–13]. A pressing challenge is to elucidate the structures that integrate biochemical and mechanical signals, respond to spatiotemporal developmental cues, and facilitate mechanochemical signal conversion to direct tissue architecture dynamics. In early embryonic development, actin filaments (F-actin) are central to force generation in various cellular processes, including cell shape regulation, migration, morphogenesis, endocytosis, and organelle dynamics, also a critical component of the cytoskeleton regulated by both biochemical signals and mechanical forces [14–17]. Understanding how F-actin generates force in response to signaling cues and how the actin cytoskeleton architecture responds to mechanical loads is fundamental to many cellular, developmental, and pathological processes [17, 18]. In organogenesis, forces operate across scales from cells to tissues, highlighting the importance of supracellular F-actin structures in mesenchymal condensation and the formation of macroscopic mechanical architectures [19]. Mesenchymal condensates, characterized by clear boundaries and continuous internal organization, form an integrated entity in which F-actin networks create a supracellular “organoskeleton” through intercellular adhesions, guiding robust mechanical pattern formation and initiating organogenesis. The formation of this structure requires energy, and its unity is reinforced by energy flow, driving tissue remodeling toward a state of minimal free energy [17, 20].

This review discusses the self-organizing mechanisms governing mesenchymal condensation and condensation-induced differentiation during organogenesis. At the supracellular level, the three-dimensional (3D) actin organoskeleton of mesenchymal tissues orchestrates the formation and evolution of condensate structures, driven by the minimization of free energy. This process progressively forms complete organ primordia and dictates the fate trajectories of cells differentiating into mature tissues. Coupling macro-scale tissue analysis with micro-scale organization within cells offers deeper insights into the mechanisms driving diverse molecular and cellular behaviors during development and may elucidate how feedback loops between cell-microenvironment interactions facilitate tissue development.

Redefined mesenchymal condensation: a new framework for organogenesis

Research on multicellular condensation has long focused on mesenchymal condensation during development, a key early event in the formation of organs such as the skeletal system, muscles, kidneys, and teeth [3, 6, 21]. Beyond its role as the main character of initial event in morphogenesis, mesenchymal stem cells (MSCs) offer a clear model for studying this process [6, 7, 22]. A paradigm shift from a planar local perspective to a three-dimensional (3D) global framework has revealed the fundamental forces driving mesenchymal condensation and shaping organogenesis [1, 23–26].

Early models conceptualized mesenchymal condensation as a discrete phase transition rather than a progressive process, primarily focusing on two-dimensional (2D) cell connectivity alterations [27]. This framework emphasized cell migration mediated by cell-matrix adhesion and chemotactic signaling, with limited consideration of 3D tissue folding dynamics. Crucially, prior to these direct aggregation mechanisms, molecular prepatterns or protocondensations were proposed to establish spatial templates through reaction-diffusion principles [28]. In vertebrate limbs, Turing-type activator-inhibitor systems — governed by bone morphogenetic protein (BMP)-Wnt-Sox9 (BSW) networks — preconfigure skeletal domains, with homeobox (HOX) genes modulating the wavelength of these patterning waves to refine digit morphology [29, 30]. Simultaneously, galectin-1a/galectin-8 (2GL)-mediated clustering initiates protocondensation by bridging ECM and cell-surface glycans, while TGF-β-fibronectin-FGF (TFF) networks further stabilize positional cues [31–33]. These prepatterns operate upstream of cell-matrix adhesion remodeling, defining conserved developmental blueprints through biochemical self-organization rather than direct mechanical condensation. Building upon these molecular templates, studies revealed that the assembly of fibronectin matrix and the precise patterning of morphogen gradients guided by prepattern-derived positional cues directly instruct mesenchymal condensation and subsequent tissue morphogenesis [34–37]. Notably, MSC heterogeneity and subtype-specific fibronectin organization across developmental niches contribute to tissue-specific mesenchymal condensation, leading to the development of distinct tissues or organs [38, 39]. However, mechanistic crosstalk between these systems remained unresolved in early paradigms. Moreover, how tissues achieve feedback regulation to robustly generate mesenchymal condensates of appropriate size and shape remained unclear [5, 34]. This knowledge gap was ultimately addressed through later advancements integrating mechanical force transduction and supracellular-scale regulatory network analyses.

Subsequent studies delineated how mechanochemical feedback loops enforce morphological commitment: these efforts identified mitogen-activated protein kinase (MAPK), Yes-associated protein/transcriptional co-activator with PDZ-binding motif (YAP/TAZ), and Rho family of GTPases/Rho-associated protein kinase (Rho/ROCK) pathways as core mechanotransduction components [13, 24, 25], thereby establishing mechanical induction as a fundamental morphogenetic driver. This paradigm shift revealed that mesenchymal condensation initiation is autonomously coordinated through cortical tension gradients and cytoskeletal remodeling driven by Rho GTPase-mediated and Myosin II-dependent actomyosin contractility [4, 19, 24, 40, 41]. Crucially, these mechanical perturbations act as biochemical switches: actomyosin-driven deformation simultaneously activates BMP/fibroblast growth factor (FGF) signaling while suppressing transforming growth factor-beta (TGF-β) pathways [41]. At the actin-extracellular matrix (ECM) interface, focal adhesion complexes mediate reciprocal mechanical feedback. ECM remodeling alters substrate rigidity and tension, which is sensed by integrins to trigger MAPK activation [15, 23, 42]. This regulatory circuit controls F-actin dynamics and adhesion molecule expression, thereby modulating cellular morphology and motility [23, 42, 43]. Through such mechanochemical integration, self-reinforcing feedback loops emerge that progressively stabilize condensed tissues into cohesive mechanical units. Force transmission via cell-cell junctions and remodeled ECM drives subsequent morphogenetic events [42, 44]. However, fundamental questions persist regarding the spatiotemporal coordination of these pathways across tissue-level dynamic patterns, particularly given current technical limitations in tracking dynamic multiscale interactions.

Organogenesis occurs in 3D space, necessitating a shift in mesenchymal condensation research to consider the 3D global aspect of self-organization [23, 26]. In vitro experiments have demonstrated that cells exhibit different physiological characteristics in 2D and 3D aggregates, highlighting the necessity of analyzing 3D condensation [1, 23, 26]. Exploring 3D mechanisms has deepened our understanding of spatial structural heterogeneity during organogenesis. In spatial configuration, the chemical chemoattraction of morphogens and the mechanical action of the extracellular matrix are intricately coupled, leading to symmetry breaking in early mesenchymal tissue development [7, 45–47]. The diffusion of morphogens induces directed cell migration, while migrating cells exert mechanical forces on the surrounding matrix through adhesion molecules, resulting in matrix deformation and the formation of adhesion gradients [48]. These gradients further influence cell movement and intercellular communication [49]. Recent advances in understanding supracellular morphogenesis have shifted the focus from individual cell signaling responses to the global features of tissue architecture [7, 42]. For instance, Rho GTPases regulate supracellular polarization and cell adhesion, while morphogens such as FGF and BMP significantly impact tissue viscoelasticity. This influence prompts the condensation zone to undergo phase transitions, forming a structure with a “solid-like core” and a “fluid-like periphery” [7]. These supracellular dynamic phase transitions correspond to new tissue states and the formation of new 3D tissue scaffolds. Actin dynamics play a crucial role in this process, contributing to molecular responses and supracellular heterogeneity [14, 41]. Downstream signaling pathways activate various gene expressions, and the interface pressure generated by this new structure influences tissue morphogenesis and cell differentiation fate [7, 41].

Transitioning from a 2D focus on local signaling to a 3D perspective emphasizing global phase transitions, research on mesenchymal condensation has gained a holistic view of early organogenesis. This shift underscores the necessity of viewing the actin cytoskeleton within collective cells as an integral component of the tissue’s dynamic framework. From a thermodynamic perspective, the tissue’s dissipative structure, driven by energy flow, reorganizes the scaffold toward a state of minimal free energy [50–52], offering deeper insights into the dynamic evolution of actin cytoskeletal organization and the heterogeneity that emerges during condensation.

From mesenchymal condensation to differentiation: actin-driven self-organization

Ratchet model for actin cytoskeleton dynamics

F-actin, a semi-flexible polymer, forms a dynamic network through actin-related protein 2/3 (ARP2/3)-mediated branching and crosslinking, ensuring structural integrity [53]. Myosin II generates contractile forces contributing to cortical tension, while actin filament polymerization at the leading edge drives cellular protrusion [54]. F-actin polymerization is driven by the preferential addition of monomeric globular actin (G-actin) to the barbed end, a process regulated by actin’s intrinsic polarity [55, 56]. In structures like the lamellipodium of migrating cells, filament growth oriented toward the plasma membrane generates protrusive force by pushing the cell edge forward, with each polymerizing filament contributing a modest force of approximately 1 pN in vitro [19].

This dynamic process is governed by the principle of free energy minimization, as elaborated in the following discussion. Before polymerization, the thermal motion of G-actin follows a random diffusion process, accompanied by thermal fluctuations on the order of a few k_BT [18, 57]. However, once bound, the energy consumed results in a significant decrease in free energy, indicating that new monomer must surmount a large energy barrier to be dislodged, and so depolymerization is very unlikely. The free energy landscape of this process can be depicted as a ‘staircase’ function resembling a ratchet, hence the name “Brownian ratchet” [56, 58] (Fig. 1A). The ratchet operates in a unidirectional manner, driven by the minimization of free energy, facilitating the growth of F-actin and its sustained interaction with the plasma membrane [54, 57]. Thus, essentially, the Brownian ratchet underlies the ability of growing actin filaments to generate force by converting thermal fluctuations into mechanical energy stored in their elastic structure and transferring this energy to the external environment through contact with the cell membrane [17, 18]. This energy transfer is essential for driving cell movement and shaping during mesenchymal condensation.

Fig. 1.

Force-coupled dynamics of actin filament growth and remodeling. (A) Actin cytoskeletal networks exert reciprocal forces with the plasma membrane through barbed-end contacts, maintaining cellular morphology. The schematic depicts actin polymerization dynamics at filament tips, where thermal fluctuations create transient gaps between barbed ends and the membrane, enabling insertion and binding of globular actin (G-actin) monomers. Progressive filament elongation generates outward protrusive forces, while concurrent membrane tension imposes retrograde resistive forces. This energy-coupled interaction operates via a Brownian ratchet mechanism, driven by minimal free energy principles to directionally bias monomer addition. The system evolves toward a mechanical equilibrium state where polymerization forces balance membrane resistance. (B) Adapted from Li et al.. (2022). This figure is used with permission, and no copyright applies [18]. Stress-dependent steady-state remodeling of branched actin networks. Applied mechanical loads linearly reduce nucleation, elongation, and capping rates of filaments. The ratio of nucleation-to-capping rates reflects the density of free barbed ends, which escalates proportionally with increasing stress magnitude. This load-adaptive response suggests a feedback mechanism linking network architecture to external mechanical constraints

Initiation of mesenchymal condensation

Mesenchymal condensation marks the site of future organs through its localized increase in cell density, clear boundaries and specific patterns [3, 4]. This process is governed by two interdependent mechanisms: chemotactic cell migration and tissue mechanical compression, both dynamically regulated by actomyosin-driven cytoskeletal rearrangements, signaling networks, and energy dissipation [4, 7, 11, 14, 17].

Chemotactic gradients of morphogens — including TGF-β, WNT, FGF, BMP, and Hedgehog proteins — originate from ectodermal sources and direct mesenchymal cell migration [7, 44, 45, 59]. For example, Fgf20 from hair follicle epithelium and Fgf8 and semaphorin 3 F (Sema3f) from dental epithelium establish signaling centers that polarize cells via microtubule reorganization and Rac-dependent F-actin polymerization, driving pseudopod formation and directed migration [44, 54, 60, 61]. Migrating cells exhibit collective behavior, with cytoskeletal linkages between cells and the ECM enabling synchronized tissue condensation [62]. System sensitivity to chemotactic cues is constrained by information entropy, which balances feedback mechanisms and energy distribution to optimize migration efficiency [63].

In loosely packed tissues, mechanical compaction serves as a fundamental driver of condensation. Interestingly, the force driving compaction is not primarily mediated by the cell adhesion molecule E-cadherin, but by actomyosin tension at the cell-matrix interface [14, 52]. Actomyosin tension propagates through the cytoskeleton to surface adhesion molecules (e.g., integrins), establishing supracellular mechanical coupling that coordinates tissue-wide compression [15, 52]. During contraction, ECM remodeling occurs prominently, with fibronectin and type I collagen forming structural scaffolds to stabilize condensed tissues [37, 42, 64, 65]. Fibronectin mediates cell-ECM adhesion via its heparin-binding domain, binding integrins (e.g., α5β1) and syndecans to establish focal adhesion complexes essential for cell surface interactions and mesenchymal condensation [34, 66–68]. Exogenous TGF-β enhances this process by upregulating fibronectin gene expression, thereby promoting cell-matrix adhesion and mechanical coupling [59, 64, 69]. Concurrently, type I collagen governs actomyosin contractility through strain-stiffening viscoelasticity, amplifying integrin-ROCK signaling to synchronize cell-matrix force reciprocity and collective cytoskeletal remodeling [23]. In vivo validation of collagen I’s mechanical role remains an active area of investigation, particularly in scaffold-free 3D models that better recapitulate physiological ECM dynamics [70]. Importantly, the whole energy-consuming process of mechanical compression is guided by interfacial tension minimization, which determines tissue contact morphology (e.g., compaction degree and boundary angles) and is ultimately governed by energy dissipation principles [50, 52].

Within the body, the above two mechanisms do not operate independently but are closely intertwined at supracellular scales [4, 14]. Chemotactic cell influx elevates local density, enhancing adhesion molecule expression and promoting mechanical compaction [7, 44, 46]. Conversely, tissue compression alters ECM connectivity, modulating morphogen diffusion and pathway activation [23, 42]. These coupled mechanisms establish spatial boundaries for condensation [3]. Actin cytoskeleton remodeling serves as a shared node, linking force transmission and signal transduction to coordinate global tissue condensation [7]. Energy flow and interfacial tension dynamics further unify these mechanisms, ensuring robust spatiotemporal control of organogenesis.

Internal structural formation and remodeling of mesenchymal condensation

Mesenchymal cells condense around the signaling center to form a quasi-spherical structure (which may represent only a portion of a sphere depending on whether the condensation signal is induced by a point or a surface; for instance, the condensation induced by dental epithelium results in a hemispherical shape) [6]. Inside the sphere, densely packed cells and the surrounding stroma adhere to each other [23]. The cytoskeleton of each cell is connected to that of neighboring cells or to the surrounding ECM through surface adhesion structures and mechanical contact, creating a mechanically conductive architecture at the supracellular scale [19, 23]. Notably, ECM remodeling during condensation not only transmits mechanical forces but also acts as a pressure sensor: matrix compression alters integrin-mediated signaling, modulating cell proliferation and motility to fine-tune condensation dynamics [70]. As condensation increases and the tissue grows, the internal structure of the condensed mesenchyme undergoes further remodeling. This remodeling is essentially driven by mechanical adaptation, with mechanical forces arising from increased cell density and the non-equilibrium stresses generated by densely packed ECM structures (18, 46, 52). Adjustments in the tissue architecture during this adaptation process are accompanied by energy flow, where the minimization of free energy serves as the underlying impetus for the reorganization of energy distribution within the tissue, ultimately leading to a new equilibrium [17, 52]. This specific process will be discussed in the following sections.

Within the cell, F-actin self-assembles into an actin network through branching and cross-linking in the cortical region near the plasma membrane, constituting a semiflexible polymer matrix ADDIN EN.CITE (14, 56). The elasticity and plasticity of the actin network enable it to convert elastic potential energy into mechanical work output, while also adapting its architecture to changing loads ADDIN EN.CITE (18). It therefore plays a pivotal role in cell remodeling. The remodeling comprises the reorientation of filaments within the network and an increase in the steady-state number and density of growing filaments ADDIN EN.CITE (18, 53). This process involves the molecular mechanism of how mechanical strain affects the geometric and kinetic properties of branched actin ADDIN EN.CITE (53). The key-ingredients of the mechanism are the force-velocity relation determining filament nucleation rates, the capping rates, and the geometry of protruding networks ADDIN EN.CITE (18) (Fig. 1B). The steady-state characteristics of the network constrain the total rates of nucleation and capping of branched actin (Eqs. 1&2), indicating that the capping rate is related to the filament density within the network, which increases under certain loading conditions. Concurrently, the branching angle regulated by Arp2/3 increases, leading to the formation of a densely branched network structure that is better able to withstand and distribute stress ADDIN EN.CITE (71) (Fig. 2A & B). If the stress exceeds the elastic resistance of the filaments, some filaments may undergo elastic buckling, resulting in the formation of new network cross-links ADDIN EN.CITE (72, 73). In concert with these changes is the distribution of cell surface adhesion sites, which directly link the actin cytoskeleton and the ECM and thereby coupling the mechanical transmission between the two. This allows both intracellular and extracellular components to collectively adapt to varying loads.

Fig. 2.

Mechanoadaptive remodeling of supracellular actin networks during mesenchymal condensation. (A) Schematic of condensation as a dynamic self-organizing process driving tissue densification and viscoelastic enhancement. Supracellular actin networks establish load-bearing architectures that enable cohesive tissue-scale mechanical behavior while generating structural heterogeneities through continuous network reorganization. (B) Magnified view of force-mediated actin restructuring at condensation boundaries. Locally amplified mechanical stress triggers adaptive network densification through enhanced actin branching. This compression-induced cytoskeletal reinforcement propagates supracellular-scale stiffening, creating microenvironmental asymmetries that guide subsequent tissue patterning events

In tissues undergoing mesenchymal condensation, intracellular actin networks dynamically reorganize under mechanical stress, forming mechanically integrated ensembles through cell-cell and cell-ECM interactions. This mechanical coupling is mediated by force-dependent reinforcement of E-cadherin/β-catenin/αE-catenin complexes and F-actin at cell-cell junctions, which collectively stabilize the organoskeletal framework [74]. The spheroidal geometry of condensation zones establishes radial symmetry in stress distribution, enabling supracellular-level analysis. Radially symmetric tissue states emerge, where regions equidistant from the condensation center exhibit similar mechanical and structural properties, forming distinct inner and outer layers [18].

Structural stratification is driven by energy-regulated actin rearrangements. In self-organized condensates (e.g., within embryos), surface free energy minimization induces inward compression, concentrating cortical actin at the tissue periphery [18, 75] (Fig. 2A). High-energy consumption by branched actin networks at the edge triggers outward energy flux [17], increasing surface tension [47]. This promotes epithelialization and polarity of surface cells. Conversely, in condensates formed via mechanical compaction and chemotactic signaling, sustained internal compression elevates tissue viscoelasticity [7, 52]. Non-equilibrium stress and energy distributions drive gradual separation between the contractile outer layer and viscoelastic core, resulting in adaptive remodeling from a homogeneous to a stratified architecture [7].

1

2

R_cap: the network-level rate of capping, in units of sec^–1 μm^–2.

R_nucleate: the network-level rate of nucleation, in units of sec^–1 μm^–2.

k_cap: appropriate capping rate constant, in units of µM^–1sec^–1.

CP: the soluble capping protein concentration, in units of µM.

E: the surface density of free barbed ends, in units of µm^–2.

Mesenchymal condensates achieve structural symmetry breaking through energy-guided actomyosin reorganization, resolving radial homogeneity into architecturally stratified states. This hierarchy, a contractile cortex coupled with a viscoelastic core, amplifies force transmission via coordinated tension and energy dissipation. By converting free energy minimization into biomechanical feedback, condensates emerge as self-organizing primordia that spatially encode morphogenetic trajectories [17, 46, 51].

Cell fate specification after mesenchymal condensation

Following mesenchymal condensation, tissues undergo further differentiation and development, ultimately forming functional organs [5, 6]. The establishment of symmetry-breaking patterns during mesenchymal condensation introduces heterogeneity in the tissue structure and microenvironment, activating distinct signaling pathways and establishing a supracellular signaling regulation network [76]. Within condensation zones, the 3D architecture and biomechanical properties of ECM — including stiffness and viscoelasticity — couple with the cytoskeleton through integrin, collectively guiding stem cell fate determination [23, 77]. In regions of high contractility, dynamic actin cytoskeletal remodeling, through WNT, focal adhesion kinase (FAK), and other pathways, promotes stem cell proliferation and osteogenic, adipogenic, or dentogenic differentiation [78]. In high viscoelastic condensation regions, F-actin fibers transmit compressive forces to the nucleus, inducing nuclear flattening and nuclear pore stretching, which enhances nuclear import of YAP and influences downstream stem cell paracrine signaling [25].

The actin cytoskeleton’s heterogeneous structure not only regulates downstream signaling but also influences cellular metabolic states to control differentiation [79]. During mesenchymal condensation, energy demands exceed the capacity of oxidative phosphorylation (OXPHOS) due to immature mitochondrial function in MSCs [80], leading to an upregulation of glycolysis supported by cellular metabolic plasticity [81]. This metabolic shift is facilitated by the actin cytoskeleton’s response to mechanical cues. Stress fiber assembly mechanically regulates phosphofructokinase (PFK) expression, a glycolytic rate-limiting enzyme, via the E3 ligase tripartite motif-containing protein 21 (TRIM21), thereby enhancing glycolysis [82]. Additionally, the hypoxic microenvironment at the core of condensed tissues promotes reactive oxygen species (ROS) generation and hypoxia-inducible factor (HIF) upregulation, further inducing glycolytic metabolism [83, 84]. Glycolysis, which converts glucose to lactate through enzymatic redox reactions without oxygen, rapidly generates adenosine triphosphate (ATP) to meet tissue metabolic demands [79, 82]. Lactate export increases extracellular acidity and raises intracellular pH [85], influencing protein structure, function, and stability. For instance, elevated intracellular alkalinity promotes β-catenin acetylation, activating the WNT signaling pathway [85]. In diverse microenvironments, dynamic metabolic regulation is sufficient to drive homogeneous cells towards different differentiation fates [76, 80, 86].

During organogenesis, mesenchymal condensation commonly leads to the formation of highly differentiated, heterogeneous structures, such as the layered tissues in tooth development (outer dental follicle and inner papilla), hair follicle development (hair matrix and papilla), and cartilage formation (membrane and internal tissue structures) [3, 6, 7]. The supracellular actin cytoskeleton plays a pivotal role in initiating this differentiation through intricate molecular regulatory mechanisms. These structures are stabilized by continuous energy and signal flux via feedback loops, with energy flow influencing structural integrity rather than specific molecular identities, contributing to the conserved developmental patterns observed across diverse species [50]. Acquiring a comprehensive understanding of underlying mechanisms could enable the prediction of diversions in cell differentiation fates and facilitate the exploration of key regulatory mechanisms in organogenesis.

Investigating the actin organoskeleton framework in morphogenesis and its applications in regenerative medicine

Recreating complex physiological architectures in vitro remains a central challenge in regenerative medicine. Current organoid technologies predominantly rely on guiding stem cell differentiation through precise spatiotemporal delivery of biochemical cues (e.g., morphogen gradients), mechanical stimuli (e.g., matrix stiffness modulation), and metabolic conditioning (e.g., oxygen/glucose tuning) to activate lineage-specific signaling pathways, or alternatively, reassemble multicellular components into predefined structures (45, 87, 88). However, a critical bottleneck persists: engineering single-stem-cell populations to undergo multidirectional differentiation and self-organize into sustainable, developmentally competent tissues. This necessitates decoding the mechanisms underlying mesenchymal condensation, a developmental process governed by supracellular actin organoskeleton remodeling and phase transitions during early organogenesis [7, 46, 52].

Recent advances highlight scaffold-free organoids, 3D MSC aggregates mimicking embryonic primordia, as potent models for studying self-organized morphogenesis in systems ranging from teeth to hair follicles [89–92]. Their efficacy hinges on two pillars: [1] recapitulating physiological 3D architectures and [2] sustaining long-term proliferative fidelity. Central to this is the actin organoskeleton, which establishes mechanical coherence across the organoid primordium. Supracellular F-actin networks dynamically interface with adhesion complexes to regulate collective migration and tissue fluidity, thereby orchestrating structural maturation [19, 93]. Quantitative analysis of these frameworks, particularly their viscoelastic coupling with microscale cellular processes (e.g., actomyosin contractility), could refine predictive models of organoid development. For instance, matching hydrogel moduli to native tissue viscoelasticity enhances ECM mimicry and sustains growth [91, 94].

Future breakthroughs will require integrating developmental principles—specifically, how actin-driven mesenchymal condensates achieve phase separation and mechanical memory—with engineered microenvironments. Such synergy could unlock monolineage organoids capable of autonomous patterning, bridging the gap between reductionist models and authentic organotypic complexity.

Furthermore, the supracellular actin cytoskeleton offers a simple yet precise framework for computational modeling in tissue engineering, leveraging the rapidly advancing artificial intelligence (AI) and machine learning (ML) algorithms to enable the prediction and optimization of the mechanical properties of tissue structures (95, 96). This facilitates the creation of structures tailored to specific tissue environments and dynamic developmental processes [97]. Additionally, AI trained in image recognition can non-invasively detect the expression levels of particular markers and predict cell differentiation states [98]. For instance, Park and colleagues developed a convolutional neural network (CNN)-based method that utilizes bright-field microscopy images to predict the differentiation levels of kidney organoids, achieving superior accuracy compared to human classifiers [99]. This approach establishes dynamic links between tissue architecture and cellular states during organogenesis, facilitating the engineered regulation of structural growth and functional differentiation of organoids in culture. The emergence of these new technologies and their application in regenerative medicine represents a significant opportunity, holding the promise of addressing critical challenges and enhancing the translational potential of tissue engineering.

Conclusion

Organogenesis is driven by the spatiotemporal integration of mechanical forces and biochemical signals, with the actin organoskeleton serving as a master regulator of supracellular self-organization. This review underscores actin dynamics as a linchpin in initiating and coordinating tissue morphogenesis, where energy flow and free energy minimization principles direct mesenchymal condensates to evolve into complex organ architectures. The actin organoskeleton not only mediates force transmission and energy distribution but also bridges cellular-scale behaviors with emergent tissue patterns, reconciling global mechanical frameworks and local microenvironmental heterogeneity. For regenerative medicine, these insights offer a blueprint to engineer functional tissues by recapitulating developmental logic, yet challenges persist in decoding how homogeneous condensates diversify into specialized structures. Future research must unravel context-specific molecular regulators of supracellular actin dynamics, enabling precise manipulation of tissue architecture and function. By dissecting feedback between actin-mediated mechanochemical interplay and transcriptional programs, we pave the way for transformative therapies that exploit the self-organizing principles inherent to organogenesis.

Abbreviations

3D

Three-dimensional

2D

Two-dimensional

AI

Artificial intelligence

ARP2/3

Actin-related protein 2/3

ATP

Adenosine triphosphate

BMP

Bone morphogenetic protein

CNN

Convolutional neural network

CP

Capping protein

ECM

Extracellular matrix

FAK

Focal adhesion kinase

F-actin

Filamentous actin

FGF

Fibroblast growth factor

G-actin

Globular actin

HIF

Hypoxia-inducible factor

HOX

Homeobox gene

MAPK

Mitogen-activated protein kinase

ML

Machine learning

MSCs

Mesenchymal stem cells

OXPHOS

Oxidative phosphorylation

PFK

Phosphofructokinase

Rho/ROCK

Ras homolog/Rho-associated coiled-coil containing protein kinase

ROS

Reactive oxygen species

Sema3f

Semaphorin 3 F

TGF-β

Transforming growth factor-beta

TRIM21

Tripartite motif-containing protein 21

YAP/TAZ

Yes-associated protein/transcriptional co-activator with PDZ-binding motif

Author contributions

J.X.H. and B.D.S. conceived and designed the study, wrote the manuscript, and designed the figures. Y.J. assisted with manuscript revision. C.X.Z. and F.J. provided financial and administrative support, contributed to manuscript revision, and gave final approval of the manuscript. All authors have read and approved the current version of the manuscript.

Funding

This work was supported by grants from the National Natural Science Foundation of China (82170988, 82471011, 82371020 and 82301028) and the China Postdoctoral Science Foundation (BX20230485).

Data availability

Not applicable.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.