Mechanics and functional consequences of nuclear deformations

Abstract

As the home of the cell’s genetic information, the nucleus plays a critical role in determining the cell’s fate and function in response to various signals and stimuli. In addition to biochemical inputs, the nucleus is constantly exposed to intrinsic and extrinsic mechanical forces that trigger dynamic changes in nuclear structure and morphology. Emerging data suggest that the physical deformation of the nucleus modulates many cellular and nuclear functions, which have long been considered downstream of cytoplasmic signaling pathways and dictated by DNA genomic sequences. In this Review, we discuss an emerging perspective on the mechanoregulation of the genetic machinery that considers the physical connections from chromatin to nuclear lamina and cytoskeletal filaments as a single mechanical unit. We describe key mechanisms of the spatial and temporal coordination of nuclear deformations and provide a critical review of the structural and functional adaptive responses of the nucleus to deformations. We then consider the contribution of nuclear deformations to the regulation of important cellular functions, including muscle contraction, cell migration, and human disease pathogenesis. Collectively, these emerging insights shed new light on the dynamics of nuclear deformations and their roles in cellular mechanobiology.

Introduction

As the largest and stiffest organelle of eukaryotic cells¹, the nucleus is constantly subjected to intrinsic and extrinsic forces that can lead to small and large scale nuclear deformations. Accumulating evidence suggests that the nucleus contributes to the cell’s perception of mechanical stimuli and the corresponding cellular response through dynamic changes of its structure and morphology^2,3. The nucleus must therefore be considered not only as the primary site of gene replication and transcription but also as a fundamental mechanical component of the cell. Emerging views of the nucleus indicate a more dynamic organelle than anticipated, capable of mechanosensing and rapid remodeling, thereby orchestrating key cellular functions in response to mechanical stimulation, making the nucleus a key mechanoresponsive and mechanosensitive organelle.

Here, we define “mechanoresponsive” as responding to mechanical stimulation, either directly or downstream of other processes triggered by the mechanical stimulus. In contrast, “mechanosensitive” refers to the cellular elements or processes directly involved in the sensing of mechanical forces or deformations by converting the mechanical inputs into biochemical signals, a process known as “mechanotransduction”. Since ‘mechanotransduction’ is often found more broadly defined as also encompassing cellular responses downstream of the initial mechanosensing (or ‘transduction’) event, we will refer to ‘mechanosensing’ throughout the text to avoid any confusion.

The mechanoresponsive properties of the nucleus are now well recognized, including its ability to adapt to the cell’s physical microenvironment with changes in nuclear morphology or the expression of specific genes^4,5. In contrast, the role of the nucleus as a mechanosensitive organelle has only recently begun to emerge. Nuclear deformation is a key component of the correct regulation of cell function in vivo and its importance has been highlighted by the observation of abnormally shaped nuclei⁶ and impaired mechanosensing mechanisms⁷ in many human diseases. Nuclear deformability plays crucial roles in activating and modulating cellular mechanoresponsive signaling. For example, several lines of evidence indicate that forces acting on the nucleus can induce sufficient nuclear deformations to modulate chromatin structure and trigger important protein conformational changes, thereby activating or repressing mechanoresponsive genes^8,9.

In this Review, we provide an overview of the current understanding of the physical properties of the nucleus, describe the contribution of specific components, discuss the physical connections between the NE and the cytoskeleton, and shed light on the role of nuclear deformation in cellular mechanosensing and mechanoresponses.

Section 1 – Nuclear structure, organization, and connections to the cytoskeleton

Nuclear deformations are determined by the balance between the mechanical properties of the nucleus and the mechanical forces acting on it. Here we describe components constituting the nuclear structure from the inside out, as well as the physical connections between the nucleus and the cytoskeleton (Fig. 1). Importantly, extensive physical interactions between these components ensure that forces from the cytoskeleton are transmitted to and across the nuclear interior (see textbox “The interconnected nucleus”).

Figure 1 – The nuclear envelope and nucleo-skeletal interactions.

(a) The nuclear envelope (NE) is composed of the outer (ONM) and inner (INM) nuclear membranes, which form a double lipid bilayer. The nuclear lamina is attached to the INM and in close contact with condensed chromatin, while nuclear pore complexes (NPCs) are surrounded by less condensed chromatin. The genomic regions connected to the lamina are lamina-associated chromatin domains (LADs), which have low transcriptional activity. The nuclear interior is connected to cytoskeletal filaments by nesprins and SUN domain proteins. Nesprin-1 and nesprin-2 bind to actin filaments, whereas nesprin-3 interacts with intermediate filaments. Nesprins-1, −2, and −4 can interact with microtubules via kinesin and dynein molecular motors. (b) NPCs allow controlled nuclear import and export of large molecules. The nuclear lamina meshwork composed of A-type and B-type lamins binds to the INM. Lamins, along with other INM proteins, such as LBR and emerin, anchor chromatin to the NE. Nesprins, ONM, SUN domain proteins and INM form together the LINC complex.

Textbox 1 - The interconnected nucleus.

Chromatin binds to scaffolding at the nuclear periphery, particularly interacting with the nuclear lamina and NPCs. Heterochromatin interacts with the nuclear lamina via lamina-associated domains (LADs), while euchromatin associates with NPCs (Fig. 1a)^223,224. Chromatin tethered to the nuclear periphery in LADs contains mostly silent or weakly expressed genes²²⁵. In contrast to LADs, genes associated with NPCs are often transcriptionally active^226,227. LADS arise from many interactions, since lamins can bind non-specifically to histones and chromatin through bridging proteins such as LAP2α, lamin B receptors (LBR), Barrier-to-Autointegration Factor (BAF) and emerin (Fig. 1b)²²⁸. Specific NE proteins often interact with different DNA regions²²⁹, and the exact mechanisms that determine chromatin-nuclear lamina association remain to be elucidated²³⁰. The physical connections between chromatin and the NE provide not only control over gene expression, but also provide increase nuclear stiffness and stability, akin to the mechanical reinforcement used in composite materials or cross-linked polymer networks^231–233.

Within the nuclear interior, chromatin from different chromosomes occupies distinct territories (Fig. 2). Inter-chromosome interactions are supported by mechanical crosslinking of neighboring chromatin at intervals of around 25 kb^14,16–18. Heterochromatin protein 1 (HP1a) is a putative crosslinker that acts through either DNA, H3K9me3 heterochromatin, or phase separation^236–238. Other candidates for linking the chromatin interior include proteins like CTCF and cohesin involved in chromatin looping and topologically associating domains (TADs)²³⁹ (Fig. 2), RNA scaffolding²⁴⁰, and nucleoli, whose periphery is covered with heterochromatin²⁴¹. Obtaining a better understanding of the precise spatial organization of chromatin within the nuclear interior, and how this organization affects transcriptional regulation, remains a topic of extensive research ^233,242.

The nuclear interior

The nuclear interior primarily consists of chromatin and nuclear bodies such as nucleoli, Cajal bodies, and promyelocytic leukemia (PML) bodies, which are membrane-less structures with specific signaling and processing functions¹⁰. Chromatin is composed of DNA and DNA-binding proteins, particularly histones (Fig. 2). Chromatin can be classified into two categories, depending on its level of compaction, transcriptional activity, and histone modifications. The loosely packed euchromatin is transcriptionally accessible and mostly localized in the nuclear interior and near nuclear pores. Densely packed heterochromatin is considered transcriptionally repressed and tends to be located at the nuclear periphery and around the nucleoli, with likely connections in between¹¹.

Figure 2 – Chromatin organization and consequences of nuclear deformations in DNA organization.

Chromosomal DNA is packaged inside the cell nucleus with the help of histones. At the simplest level, chromatin is a double-stranded helical structure of DNA. The negatively charged DNA double helix is complexed with histones, which are positively charged proteins, to form nucleosomes. Inside the interphase nucleus, chromosomes occupy distinct territories (highlighted by different colors). Within each chromosome territory, the chromatin is folded into multiple loops and segregated into two distinct compartments: compartment A clustered around nucleolus and nuclear bodies (permissive region, in grey), and compartment B (repressive region, in red) located at the nuclear periphery. Within compartments, chromatin is further partitioned into topologically associating domains (TADs), which have preferential intradomain interactions compared to interdomain interactions with the neighboring cis chromatin domains. Histone methylation, particularly at residues H3K9 and H3K27, are often associated with heterochromatin, whereas histone acetylation, particularly at residue H3K9 or histone methylation at residue H3K4 are typically associated with euchromatin.

The nuclear envelope

The NE serves multiple pivotal functions: it controls access of cytoplasmic proteins to the genome, provides structural stability to the nucleus, and physically connects the nuclear interior and cytoskeleton (Fig. 1a). The NE is comprised of the nuclear membranes, the nuclear lamina, and nuclear pore complexes (NPCs). The inner and outer nuclear membranes (INM and ONM, respectively) are two concentric lipid bilayers, each ≈4 nm thick, separated by the ≈20–50 nm wide perinuclear space¹² (Fig. 1b). The ONM is contiguous with the endoplasmic reticulum (ER) and provides an external reservoir of lipids to adapt nuclear surface area upon deformation, although membrane recruitment to the NE may be limited by resistance from the ER. Nuclear membrane wrinkling and folds present at low membrane tension provide an additional membrane reservoir for adjusting nuclear shape¹³. NPCs regulate the active nuclear transport of macromolecules larger than ≈50 kDa into and out of the nucleus^14,15. NPCs are homogeneously distributed over the nuclear membrane surface by the underlying structure provided by lamins and nucleoporin ELYS¹⁵. NPCs expanding in response to mechanical stress can account for up to 10% of nuclear surface expansion during nuclear deformations^16–18.The NE and ER additionally contain mechanosensitive ion channels such as Piezo1¹⁹ and inositol triphosphate receptors (InsP3Rs)²⁰ that can respond to nuclear membrane tension (see also textbox “Nuclear mechanoresponses and mechanosensing”). The nuclear lamina, a dense protein network underlying INM, is primarily comprised of lamins. Lamins assemble into 300–400 nm long and ≈3.5 nm thick non-polar filaments, and form a ≈14–30 nm thick meshwork^21,22. Recent electron cryotomography imaging has revealed that the lamin meshwork organization in mammalian somatic cells is heterogenous²¹, substantially deviating from the regular meshwork structure reported for Xenopus oocytes.

Textbox 2 – Nuclear mechanoresponses and mechanosensing.

Although it is now well recognized that nuclear deformation has both rapid and long-lasting consequences on nuclear and cellular function, the precise mechanisms by which nuclear deformation is translated into biochemical signals, and to what degree the nucleus itself serves as a cellular mechanosensor, remain incompletely understood. As a note of caution, many nuclear changes cited as indicators of nuclear mechanosensing in response to external mechanical stimuli (e.g., altered nuclear shape, chromatin organization, gene expression), may reflect at least in part downstream effects of signaling pathways initiated in the cytoplasm or cell surface, rather than direct nuclear mechanosensing. In the following, we highlight recent findings and novel insights into established and proposed nuclear mechanosensing mechanisms. For a more detailed discussion, we refer the reader to some excellent recent reviews^243–246.

Stretch-activated opening of channels in the nuclear membranes

NPCs allow passage of small molecules while excluding larger molecules that do not contain nuclear localization sequences or are transported by other proteins. Recent live cell imaging, electron microscopy, and cryoelectron-tomography studies found that NPCs are highly sensitive to nuclear membrane tension^18,215,247, increasing their diameter in response to elevated nuclear membrane tension and thus facilitating nuclear import, including of the mechanoresponsive transcription factor YAP (Fig. 5c)²⁴⁸. The NE and ER contain various other stretch-sensitive ion channels, such as Piezo1 and InsP3R. Increased nuclear membrane tension, in response to cell compression, osmotic swelling, or stretching application, may trigger opening of these channels and release of calcium from the ER and perinuclear space, which can lead to changes in chromatin organization¹⁹ and increased cell contractility^125,169. However, it remains unclear whether opening of these ion channels in response to cellular deformation occurs at the NE, ER, or the plasma membrane. One interesting hypothesis is that all three locations contribute to cellular mechanotransduction, and spatial coordination between ion channels in the these different membranes allow cells to distinguish between different sources of nuclear membrane strain, such as osmotic swelling and compression^125,249.

Mechanosensing of the nuclear membranes and nuclear envelope proteins

Changes in the tension or curvature of the nuclear membranes alone can be sufficient to induce nuclear mechanoresponses by altering the packing and/or composition of nuclear membrane phospholipids, thereby facilitating binding of cytosolic phospholipase A2 (cPLA2) to the INM, which is further amplified by increased intranuclear calcium concentrations^246,250,251. Recruitment of cPLA2 to the INM leads to the production of arachidonic acid and lysophosphatidic acid (LPA) production that can trigger increased cell contractility and other downstream responses (Fig. 5d)^{75,88,165,240}.

Besides altering protein-interaction with the nuclear membranes, forces acting on the nucleus can also lead to local unfolding, conformational changes, and increased phosphorylation of lamins^{100,103,253–255} (Fig. 5e), although the functional relevance of these changes remain to be fully characterized. Furthermore, force application to the nucleus via nesprins leads to phosphorylation of emerin via Src kinases, resulting in the recruitment of lamins to the NE and nuclear stiffening²⁵⁶. Although it remains unclear whether the increased phosphorylation is due to mechanically induced activation of nuclear Src kinase or emerin becoming more accessible to the kinase, this study, which was conducted on isolated nuclei, provided some of the most direct evidence for nuclear mechanosensing.

Force-induced changes in chromatin organization

Several studies have demonstrated mechanically induced changes in chromatin organization that could affect gene expression, including in neutrophils that had migrated through tight constrictions²⁵⁷, macrophages under spatial confinement²⁰⁶, and a 3D chemomechanical model of the nuclear interior and its connections to the cytoskeleton²⁵⁸, but they did not completely address whether the effects were nucleus-intrinsic or mediated by cytoplasmic signals. Support for direct nuclear mechanosensing comes from two recent studies, which found that force application to the cell surface leads to near instantaneous chromatin deformation, visualized by tracking multiple GFP-LacI labeled genomic loci, and rapid (<15 sec) increase in transcription of the corresponding transgene and other genes^203,204. The magnitude of the response was directly related to the extent of chromatin deformation and histone methylation status. Of note, the chromatin ‘stretching’ reported in these studies likely does not reflect stretching of the DNA itself, but rather partial unpacking of the chromatin, which may promote access to transcriptional regulators or polymerases²⁰⁴ (Fig. 5f). Depletion of lamins, emerin or LINC complex components abolished the force induced gene expression²⁰³, pointing to the importance of nucleo-cytoskeletal coupling. The effect of LINC complex disruption on the activation of mechanoresponsive genes contrasts with a previous study, in which LINC complex disruption did not alter expression of a several mechanoresponsive genes despite reducing nuclear deformation³², possibly reflecting differences in cell type, force application, or extent of nuclear deformation.

Another intriguing thought is that liquid-liquid phase separation (LLPS) of intrinsically disordered proteins within the nucleus could contribute to nuclear mechanosensing. LLPS inside the nucleus can exert significant mechanical forces that alter chromatin-organization and rearrangements^84,259. Although the previous studies did not address whether these changes were associated with altered gene expression, one could speculate that externally applied forces and nuclear deformation could similarly affect LLPS events in the nucleus and thereby modulate nuclear function. In addition to LLPS, changes in molecular crowding resulting from mechanically induced alterations in nuclear volume^13,162 could modulate various nuclear functions.

In mammalian somatic cells, the nuclear lamina is predominantly composed of four lamin isoforms: two A-type lamins (A and C), and two B-type lamins (B1 and B2)²³. The LMNA gene encodes for lamin A and C and some rare isoforms, which arise from alternative splicing, and the LMNB1 and LMNB2 genes encode lamin B1 and lamin B2, respectively²³. Each lamin isoform forms separate but interacting meshworks^24,25. B-type lamins are modified by farnesylation and are thus primarily located at the nuclear membranes (Fig. 1b), whereas A-type lamins either lack (lamin C) or have their farnesylated C-terminus removed (lamin A) and can be localized both at the nuclear lamina and the nuclear interior²⁶, with the intranuclear distribution of lamins mediated by LAP2α and other proteins²⁷. Lamins interact with various binding partners, including NPC proteins, INM proteins, chromatin, and various transcriptional regulators²³. Accordingly, the lamina has many structural and other functions, including contributing to nuclear shape, mechanical stability, nucleo-cytoskeletal coupling, nuclear positioning, genome organization, and mechanosensing^28–30.

Physically connecting the nucleus and the cytoskeleton

Force transmission between the cytoskeleton and the nucleus is required for nuclear movement and positioning, for example, during cell migration, nucleokinesis, and muscle fiber regeneration³¹ (Fig. 3). Cytoskeletal connections to the large and rigid nucleus are also important for cytoskeletal organization, affecting stress fiber organization, focal adhesions, and cell-cell adhesion^32,33. The physical coupling between the cytoskeleton and the nuclear interior is achieved by the linker of nucleoskeleton and cytoskeleton (LINC) complexes that span the NE^31,32 (Fig. 1a), although additional mechanisms, such as molecular motors binding to NPCs³⁴ or microtubules connecting to emerin and other nuclear envelope proteins³⁵ may further contribute to nucleo-cytoskeletal coupling. LINC complexes are composed of nesprins (nuclear envelope spectrin repeat proteins) localized within the ONM that bind across the perinuclear space to SUN (Sad1p, UNC-84) domain containing proteins located on the INM via their C-terminal KASH (Klarsicht/ANC-1/Syne Homology) domain^31,36,37. This interaction appears to be at least in part responsible for controlling the spacing between the INM and ONM³¹. On the cytoplasmic side, nesprin-1 and nesprin-2 bind to actin filaments³⁸ and—via kinesins³⁹ and dynein⁴⁰ to microtubules. Nesprin-3 binds to intermediate filaments via plectin⁴¹. Nesprin-4, which is found in polarized epithelial cells, plays an important role in nuclear positioning via kinesin-1⁴². KASH5 is a germ-cell specific KASH-domain protein required for proper meiosis⁴³. On the nucleoplasmic side, SUN domain proteins bind to the nuclear lamina, nuclear pores, and chromatin. The current model considers that LINC complexes balance part of the cytoskeletal tensile force exerted on the ONM, with maximal stress values at nuclear poles⁴⁴.

Figure 3 – Physiological sources of nuclear deformations.

(a) Actomyosin contraction (in red) produces tension in actin fibers spanning the nucleus (in blue), which are connected to the NE via LINC complexes (in orange). Tension in apical actomyosin fibers generates vertical compressive forces that result in nuclear flattening. (b) Contraction and stretching of myofibers induce nuclear deformations, including NE wrinkling and expansion. Apical microtubules (in green) form cage-like structure around nuclei and exert compressive forces during myofibers elongation. (c) Formation and regeneration of skeletal myofibers require migration of nuclei along the myofiber axis through the interplay between LINC complex and microtubule associated motors such as kinesin-1. Myofibril contraction drives nuclear movement from the center to the periphery of the myofiber during muscle fiber maturation. This process requires myofibrils to exert contractile forces on the nucleus, resulting in large nuclear deformations. (d) Epithelial cell intercalation within dense tissues requires cellular elongation and nuclear deformation. Lateral compressive forces are exerted on both nuclear sides by ventral fibers, which are thick actomyosin bundles connected from their both ends to focal adhesions at the bottom of the cell. (e) Nucleokinesis events are observed during the development of the neuroepithelium of the central nervous system and is accompanied by considerable nuclear deformations. This mechanism occurs in densely packed tissues and involves pulling forces on the nucleus exerted by a microtubule cage towards the centrosome and pushing forces at the cell rear generated by actomyosin contraction, depending on the system. In mammals, microtubules exert pulling forces on the nuclear lamina through LINC complexes that move the nucleus towards the centrosome. (f) Immune cells and tumor cells can breach the endothelial barrier of blood vessels by inserting protrusion between or inside endothelial cells. Transendothelial migration through the small gaps (can be associated with a nuclear softening. Migration through the small openings (a few micrometers in diameter) is associated with large nuclear deformations. (g) Migrating cells translocate and deform their nucleus through narrow ECM pores or in between cells by using a combination of “push” and “pull” mechanisms. Nuclear deformations result from the balance between the amount and direction of the applied cytoskeletal force and the mechanical properties of the nucleus. Nuclear translocation requires both rear and front actomyosin contraction leading to pushing/pulling forces, respectively. At the front, microtubule motors are recruited to generate pulling forces. Together, the balance of forces results in the forward movement of the nucleus through the narrow constriction. High level of lamin A/C results in stiffer nuclei and highly invasive cells, whereas actin sleeve can be recruited at the site of the constriction to locally deform stiffer nuclei with higher level of lamin AC/ during the translocation of less invasive cells

LINC complex localization at the NE is associated with specific cellular functions. For example, LINC complex proteins are organized along apical stress fibers interacting with the cell nucleus^45,46 and at the front of the nucleus as cells squeeze their nuclei through small pores⁴⁷. Although our current understanding of how LINC complex localization and force transmission is regulated is still incomplete, recent findings indicate that disulfide bonds between the SUN and KASH domains can serve as a crucial modulator of nucleo-cytoskeletal coupling^37,31. Several additional components have been identified that mediate LINC complex function and force transmission, including FHOD1 (Formin Homology 2 Domain Containing 1)⁴⁸, torsinA⁴⁹, Samp1⁵⁰ and lamins A/C⁵¹. Nesprins can also contribute to nucleo-cytoskeletal coupling independent of their actin and KASH-domains via their spectrin repeats⁵². Nonetheless, many questions remain regarding the precise regulation of LINC complex assembly and function.

Section 2 – Nuclear mechanics

Determining time and length scales of nuclear deformations is one of the key pieces of information for understanding how nuclear deformations mediate cellular functions. Here we analyze the nucleus from a mechanobiology perspective and discuss changes that influence nuclear mechanical properties and responses.

The physical properties of the nucleus

Insights from various experimental assays⁵³ indicate that the nucleus behaves as a viscoelastic material, i.e., it exhibits both elastic and viscous behavior when subjected to external forces⁵⁴. In this context, elastic deformations are defined as instantaneous, reversible deformations, like a spring that extends under an applied force and snaps back to its original length when the force is removed. In contrast, viscous, i.e., liquid-like materials, exhibit flow and thus a time-dependent, irreversible deformation when subjected to force.

Numerous assays have been developed to quantitatively capture the rheological properties of the nucleus, ranging from micropipette aspiration and microindentation to stretching intact cells or isolated nuclei⁵⁵. A major challenge lies in the fact that the viscoelastic response of the nucleus reflects a complex coupling between chromatin, lamins, and other nuclear components, and thus the exact behavior can vary depending on the nature of the applied force/deformation and the molecular composition and organization of the cells being examined. Illustrating this challenge, some studies using micropipette aspiration found that the nucleus gradually deformed under an applied pressure before reaching a plateau, while in other cases the nucleus continued to deform under applied pressure, exhibiting a fluid-like behavior^{56,54,57–59}. Stretching isolated nuclei at physiologically relevant strain rates revealed that for small deformations (<30% of the original length), the nuclear resistance is dominated by chromatin, whereas resistance to larger deformations is dominated by lamins A/C⁶⁰. Furthermore, the nucleus undergoes strain stiffening, i.e., becomes more difficult to deform upon direct force application^60,61.

After the removal of a mechanical strain, the elongated nucleus can relax with a nearly elastic response^62–65 or with a delayed response and even residual plastic deformation, characteristic of viscoelastic material properties^66,67. The elastic response requires the presence of lamin A/C, SUN-domain protein linkages and vimentin⁶³. These differences of nuclear deformation and restoration dynamics may be explained by variations in nuclear lamina composition, chromatin organization, and cytoskeletal structure, composition, and remodeling.

Contribution of specific nuclear components to the mechanical properties of the nucleus

Although A- and B-type lamins share similar biochemical properties and filament structure, it is primarily the levels and assembly status of A-type lamins that determine nuclear stiffness and viscoelastic properties. Nonetheless, B-type lamins also contribute to nuclear stiffness and stability^68,69, and loss of either lamin type results in abnormal nuclear shape and increased NE rupture^62,70–73. Besides lamins, chromatin histone modification state and composition are major determinants of the mechanical properties of the nucleus, particularly for low nuclear deformation regimes^74,60. Increasing the euchromatin content with histone deacetylase inhibitors, decreasing heterochromatin with histone methyltransferase inhibitors, or disrupting dynamics of the linker histone, all lead to softer nuclei and more nuclear blebbing events, without perturbing lamin levels^74,60.

Although determining the physical state of condensed chromatin is critical for understanding mechanisms that modulate genome function, the mechanisms by which 10-nm chromatin fibers are packaged into higher-order heterochromatin and euchromatin domains and form phase-separated condensates at high cation concentrations^75,76 have not been fully established.

The rheological behavior of the genome is highly complex and it is possible that the viscoelastic properties of chromatin are heterogeneous, locally tuned across the nucleus to accommodate different DNA-related biochemical processes such as transcription, replication, or DNA repair. Future experiments exploring the ATP dependence of DNA-related biochemical transactions as well as different time- and length-scales of chromatin organization both in vitro⁷⁷, in vivo and in silico⁷⁸ might provide further insights into the physical state of the genome. New evidence suggests that chromatin proteins such as HP1a, WDR5, BAF, and Numa also provide mechanical support to chromatin and regulate nuclear shape^79–82. Interactions between chromatin and the NE further contribute to nuclear stiffness by forming an interconnected network. In addition, it is increasingly recognized that liquid-liquid phase separation (LLPS) of nucleoplasmic components, which leads to the formation of biomolecular condensates such as the nucleolus or heterochromatin⁸³, may serve as a key principle governing nuclear organization^84–87. The propensity to form liquid droplets is significantly enhanced in the vicinity of regions of low chromatin density because the higher mechanical energy required to deform the dense chromatin to create space for a growing protein droplet would generate an energetic penalty⁸⁸. The growth of liquid droplets within the low chromatin density areas can lead to two distinct mechanical effects⁸⁹. First, chromatin can be repelled as the drops growth by creating an effective repulsive interaction. A second effect can be driven by the tendency of the droplets to merge to minimize their surface energy. Indeed, regions of chromatin initially far apart and in separate droplets can be brought into close proximity when the droplets merge, creating an effective attractive interaction. The interplay between LLPS and chromatin is thus able to generate significant mechanical forces that can result in chromatin rearrangement⁹⁰. Nonetheless, the relative contributions of LLPS versus other molecular mechanisms in determining the static and dynamic organization of chromatin within the nucleus remains to be fully elucidated.

Furthermore, the contribution of condensed chromatin to the mechanical integrity of the nucleus and its ability to respond to extranuclear forces are difficult to reconcile with a liquid state. Indeed, nuclear chromatin is mechanically responsive and can resist significant applied force⁹¹ which is a more consistent with a solid or gel state. Further studies that will consider chromatin fibers as viscoelastic filaments that can behave as both a viscoelastic solid and as a viscous liquid at different time- and length-scales may reconcile some of the apparently contradictory observations and ultimately provide a physical framework for the genome organization in space and time.

Determinants of nuclear volume and intranuclear pressure

Although the initial observation that the ratio between cellular and nuclear volumes is largely constant was made over 100 years ago⁹², and it is now well recognized that nuclear volume changes with chromatin organization and DNA content, the precise mechanisms underlying nuclear volume regulation remain incompletely understood. The nuclear volume is determined by the balance between outward pressures that originate from the nucleoplasm and tend to expand the nucleus, and inward pressures that originate from the cytoplasm and compress the nucleus. The outward pressure includes contribution from both the chromatin and the fluid inside the nucleus. Notable, despite the presence of NPCs that facilitate flow of fluid either into or out of the nucleus, cells are able to establish hydrostatic pressure differences between the nucleoplasm and cytoplasmic compartments^93–95. To further understand, this mechanisms, biological factors implicated in nuclear size determination⁹² must be translated into quantifiable physical quantities to establish the force balance between the nucleus and cytoplasm that in mechanical—but not necessarily thermodynamic—equilibrium determines nuclear volume. Based on the concept that the interior of living cells is “crowded”, colloid osmotic pressure was introduced as a simple crowding metric to explain how mechanical works, such as inflating the nucleus, can be explained from protein aggregation and phase separation⁹⁶. Very recent theoretical works suggest that the dominant pressure within the nucleus and cytoplasm originates from the osmotic pressure of the preferentially localized soluble molecules rather than the mechanical properties of large complexes such as the chromatin and cytoskeleton^97,98. To go a step further, more sensitive subcellular osmometers⁹⁷, such as genetically encoded biosensors, are needed to establish definitive physiological values of colloid osmotic pressure and to determine how crowding inside cells is regulated as a function of subcellular location and physiological inputs.

Adaptive changes in nuclear mechanics

Deformation of cells and the nucleus can lead to changes in chromatin organization and compaction, thus changing the mechanical properties of the nucleus and providing a mechanism to protect the nucleus from mechanical stress^19,99 (see also textbox “Nuclear mechanoresponses and mechanosensing”). Furthermore, mechanical force application can lead to phosphorylation of emerin and subsequent recruitment of lamins to the NE, causing rapid stiffening of the nucleus, whereas reducing cytoskeletal tension can soften the nucleus by increasing lamin phosphorylation and turnover¹⁰⁰, highlighting the importance of the interplay between the nucleus and the cytoskeleton. Of note, emerin is a recognized actin-binding protein that promotes actin polymerization¹⁰¹. Mechanically induced translocation of emerin from the INM to the ONM can thus lead to increased perinuclear actin polymerization¹⁰², which could alter nuclear deformability.

Differences in lamin expression between various tissues that affect the deformability and mechanical stability of nuclei may indicate tissue-specific adaptations to particular mechanical demands of the local microenvironment^29,103–108 but may also reflect the role of lamins in tissue-specific gene expression^104,105. For example, neutrophil nuclei have a particular lobulated morphology with characteristic low lamin A level and elevated condensed chromatin level¹⁰⁹ that promote their perfusion and migration through tight spaces¹¹⁰, such as lung capillaries that are only few microns in diameter, or even smaller gaps between endothelial cells. However, whether individual cells can dynamically adapt their nuclear stiffness on short timescales to promote migration through tight spaces is still under debate. Confocal Brillouin microscopy revealed nuclear softening during transendothelial migration of breast cancer cells¹¹¹. However, the origin and temporality of such nuclear softening remain poorly understood. Interestingly, metalloproteinase (MMP) inhibitor treatment leads to nuclear softening via lamin A/C phosphorylation, which is essential for migration through pores with subnuclear diameter (see also next section)^112,113. This response requires an intact connection between the nucleus and the centrosome via the LINC complex protein nesprin-2 and the dynein adaptor Lis1¹¹². Chromatin remodeling can further modulate nuclear stiffness and cell migration in 3D environments⁸⁰. These findings suggest that dynamic chromatin modification and changes in lamin levels and organization can mediate nuclear mechanics and promote cell migration in confined 3D environments^114,115, although reducing lamin A/C levels below a critical threshold may reduce cell survival under mechanical stress^71,116–118.

Section 3 – Physiological sources of nuclear deformations

The nucleus is constantly exposed to forces from the surrounding cytoskeleton, including from active positioning of the nucleus during cell polarization¹¹⁹, migration¹¹⁹ or differentiation¹²⁰. Recent advances in intravital imaging and modeling physiological microenvironments in vitro have documented large scale nuclear deformations in striated muscle^121,122 and during ‘confined migration’, i.e., cells squeezing through three-dimensional spaces with pore sizes smaller than the size of the nucleus^71,72,123, although similar nuclear deformations and functional consequences are expected to also occur during numerous other situations, such as embryonic development^124,125 or nucleokinesis events¹²⁶ (Fig. 3).

Nuclear deformation in cells adhering to flat and rigid substrates

Actin stress fibers and actomyosin contractility can impose vertical and lateral inward compressive forces on the nucleus. Lateral actin fibers can lead to nuclear deformation when cells migrate or are stretched^127,128. Vertical compressive forces are exerted by apical actin stress fibers that form a dome-like structure across the nucleus and that are physically attached to the nuclear lamina through LINC complexes¹²⁹. On flat rigid substrates, these forces flatten the nucleus during cell spreading (Fig. 3a) and can cause nuclear envelope rupture events^130–132. In contrast, the nucleus remains more rounded in cells on soft substrates¹³³, which are characterized by a lower amount of cytoskeletal tension and fewer actin stress fibers¹³⁴, or when the actin cytoskeleton or LINC complex are disrupted¹³⁰. Indeed, ventral actin fibers, which are thick actomyosin bundles connected from their both ends to focal adhesions at the bottom of the cell, can exert lateral compressive forces on both nuclear sides¹³⁵. The high level of tension in ventral actin stress fibers can lead to nuclear indentations of the order of a few microns that are characterized by local enrichment of LINC complexes and segregated domains of condensed chromatin^46,136. Collectively, these findings suggest that the amount of tension within the perinuclear actin fibers is an important parameter of nuclear deformation and nuclear mechanoresponses.

Nuclear deformation in skeletal and cardiac muscle

Actomyosin contractility also plays an important role in nuclear deformation in striated muscle cells. Large nuclear deformations were recently visualized in cardiac and skeletal muscle contraction in living fly larvae¹²¹. Increased expression of lamins A/C in muscle cells is essential to protect their nuclei from mechanical damage caused by muscle contraction¹³⁷ (Fig. 3b). Another, more surprising mechanism responsible for mechanical stress on the nucleus are the cytoskeletal forces required to position muscle nuclei along the length of the muscle fiber and the nuclear periphery during muscle cell maturation and repair^138,139. LINC complex proteins such as nesprin-1, together with the microtubule associated motors kinesin-1 and dynein, and other NE proteins such as emerin play a crucial role in moving myonuclei along the microtubule network to distribute them along the length of the muscle fiber^138,140,141 (Fig. 3c). The physical stress during nuclear positioning results in nuclear rotation and nuclear deformation^154,161,170. In lamin A/C-deficient or mutant cells, which have mechanically weaker nuclei, the kinesin-mediated forces can result in large-scale nuclear deformations and damage¹⁴³. In later stages of muscle fiber maturation, myofibril contraction is needed to move skeletal muscle nuclei to the periphery of muscle fibers, incurring nuclear deformation in the process (Fig. 3e), particularly in lamin A/C-deficient cells¹⁴⁴. Intriguingly, in lamin A/C-deficient and mutant mouse models that develop severe muscular dystrophy and dilated cardiomyopathy (see textbox “Human pathologies associated with nuclear deformations”), reducing the cytoskeletal forces acting on the fragile muscle cell nuclei by disrupting the LINC complex, prevents nuclear damage and results in improved muscle function and viability in vitro and in vivo^118,145, pointing to promising new therapeutic approaches for these devastating diseases. However, given that mutations in nesprins and SUN proteins can lead to muscular dystrophy and heart disease¹⁴⁶,further studies will need to evaluate the long-term risks and consequences of LINC complex disruption, using for example inducible LINC complex disruption models¹⁴⁷.

Textbox 3 - Human pathologies associated with nuclear deformations.

Abnormalities in nuclear and chromatin organization, including defects in the lamina, are hallmarks of many diseases ranging from heart disease to premature aging and cancer²⁶⁰. For many years, aberrations in nuclear morphologies have been used by pathologist as a signature for disease such as cancer, where they can indicate metastatic potential^261–263. Hundreds of mutations and variants have been found in genes encoding for NE components ranging from ONM proteins (e.g., nesprins) to INM proteins (e.g., emerin and SUN proteins) and lamins A/C. These mutations result in a set of dramatic diseases called nuclear envelopathies²⁶⁴. Mutations in the LMNA gene, which encodes Lamin A/C, are the second most frequent cause of congenital dilated cardiomyopathy (DCM), comprising 5–8% of cases²⁶⁵. So far, over 450 mutations associated with the LMNA gene have been identified²⁶⁶ and LMNA mutations are believed to cause over 13 human diseases. These diseases include various types of muscular dystrophy²⁶⁷, familial partial lipodystrophy²⁶⁸, and progeria²⁶⁹.

Muscular dystrophy

Several mutations in myopathic lamins associated with muscular dystrophy and dilated cardiomyopathy result in more deformable nuclei. Nuclear softening was exclusively associated with myopathic lamin mutations¹⁸⁰. In skeletal muscle cells, lamin mutations were recently observed to cause extensive NE damages in skeletal muscle cells in vitro an in vivo, resulting from mechanical stress on the more fragile nuclei¹¹⁸. Lamin mutations associated with muscular dystrophy can also impair LINC complex function^180,270,271 and other cellular processes. Abnormal YAP activity, known to be responsive to nuclear deformation and lamin A levels^30,248, has been reported in muscular dystrophy and rhabdomyosarcoma²⁷². In LMNA-related congenital muscular dystrophy, lamin mutations increase YAP nuclear localization via an increased nuclear import, implicating YAP as a potential pathogenic contributor in muscular dystrophies caused by NE defects²⁷³.

Hutchinson-Gilford Progeria Syndrome (HGPS)

HGPS is an exceptionally rare and dramatic premature aging disease caused by mutations in the LMNA gene. Most cases of HGPS result from a mutation that leads to alternative splicing, causing a truncated form of prelamin A (LAΔ50) that remains farnesylated. Cells from HGPS patients have irregular nuclear shapes²⁷⁴, increased nuclear stiffness, and increased sensitivity to mechanical stress^275–278, which may be responsible for the progressive loss of vascular smooth muscle cells in HGPS. The formation of orientationally ordered microdomains of lamins in HGPS cells reduces the ability of the NE to dissipate mechanical stress²⁷⁸. Restoring the loss of heterochromatin alone in Hela cells expressing LAΔ50 and in patient HGPS cells is sufficient to restore normal nuclear shape, suggesting that heterochromatin loss may be responsible for many of the phenomena associated with HGPS^279–281.

Neurodevelopmental defects

Deficiency of lamin B1 and lamin B2, but also increased expression of lamin B1, are associated with neurodevelopmental defects and distinct nuclear shape abnormalities in neurons. Altered nuclear lamin interferes with proper nucleokinesis, a nuclear translation process required during neuronal migration⁶⁹. Defective migration of cortical neurons was observed in lamin B1- and lamin B2- deficient embryos, leading to neuronal layering abnormality in the cerebral cortex along with neonatal mortality^282–284. The neuronal migration abnormality can be explained by a weakened nuclear lamina, as B-type lamin depletion was previously shown to affect nuclear mechanical properties²⁸⁵. Duplication of the gene encoding lamin B1 results in autosomal dominant leukodystrophy, which is characterized by widespread loss of myelin loss in the central nervous system²⁸⁶.

Tauopathies

Tauopathies refer to a class of neurodegenerative diseases involving the aggregation of Tau protein, a neuronal microtubule-associated protein into neurofibrillary or gliofibrillary tangles (NFTs) in the brain. Pathological accumulation of Tau is known to form Tau nuclear rods (TNRs) or Tau-positive nuclear indentations (TNIs)²⁸⁷. These structures have been identified in several neurodegenerative disorders ranging from Alzheimer’s disease, frontotemporal dementia to Huntington’s disease^288,289. Several studies identify the Tau protein (MAPT) as central during pathogenesis associated to dementia^290,291. However, the mechanism underlying Tau mediated pathogenesis is still unclear. Mutations in the Tau encoding gene MAPT result in Tau mislocalization to the cell bodies rather than neuronal axon. This leads to abnormal microtubule organization, which deform the NE via LINC complex based coupling²⁹², causing large nuclear lamin invaginations and defects in nucleocytoplasmic trafficking^291,293.

Why are laminopathies involved in a wide spectrum of diseases?

Although the pathological mechanisms underlying the diverse laminopathies are still not fully understood, various hypotheses have been put forward to explain the tissue-specific defects found in many laminopathies and the diverse phenotypes associated with specific LMNA mutations. The key role of lamins in establishing the mechanical properties of the nucleus suggests that defects in NE/lamina proteins can result in impaired nuclear stability and increased nuclear fragility, which can lead to mechanically induced nuclear damage and perturbations of mechanosensing pathways. This hypothesis is supported by numerous in vitro and in vivo observations of abnormalities in nuclear morphology (e.g., wrinkling, irregularities, blebs, and invaginations)^{180,65,294,295}. Mechanically induced damage to the fragile laminopathic nuclei was associated with increased DNA damage, resulting from direct force transmission to the genome or NE rupture^29,196,296. Besides their mechanical function, lamins have a key role in tethering and organizing chromatin, as well as in signaling involved in transcriptional regulation. In support of this, laminopathic nuclei often display alterations in the organization of chromatin, signaling, and broad alterations in gene expression^{297,298,280,299,300}, which could contribute to the tissue specific phenotypes.

Nuclear deformation during development

In early Drosophila embryo, severe nuclear deformations occur when somatic nuclei at the periphery of the syncytial embryo move as the plasma membrane invaginates to form membranes around each nucleus, a process called ‘cellularization’. The nuclear deformations are caused by the polymerization of microtubules in bundles organized by dynein¹⁴⁸. The nuclear deformations may be particularly pronounced because A-type lamin is not expressed in Drosophila during cellularization, leading to more deformable nuclei¹⁴⁹. Nuclear movement during development also results in substantial nuclear deformations in the nematode Caernorhabditis elegans, which require cytoskeletal force transmission to the nucleus via the LINC complex¹⁵⁰.

In epithelial systems, cellular intercalation is a common process occurring throughout development, where neighboring cells exchange their place to maintain epithelium integrity. Depending on the cell density, cellular intercalation can lead to transient cellular squeezing events and nuclear deformation (Fig. 3d), likely due to compression by neighboring cells and cytoskeletal remodeling^151,152. Interkinetic nuclear migration is observed during the development of the neuroepithelium of the central nervous system and is accompanied by considerable nuclear deformations in the zebrafish embryos¹⁵³ and within the mouse retinal tissue from P0 to P15 stages¹⁵⁴, when A/C type lamins and Lamin B receptor (LBR) are not expressed. Interestingly, suppression of nuclear deformation in the mouse retina results in impairment of chromocenter clustering, suggesting that dynamic nuclear deformation could be an underlying driving force of nuclear architecture and spatiotemporal genomic reorganization¹⁵⁵.

During brain development, nucleokinesis in neurons (Fig. 3e) is a beautiful example of how pulling the nucleus in densely packed tissues through the cytoskeletal leads to nuclear deformation¹⁵⁶. Interkinetic nuclear migration is a nucleokinesis event occurring in progenitor cells that involves up and down movements of the nucleus in elongated cells, which are attached on their both ends. Both actin and microtubules have been involved in the process, depending on the system¹⁵⁷. In mammals, microtubules exert pulling forces on the nuclear lamina through LINC complexes that move the nucleus towards the centrosome. Loss of either lamin B1 or lamin B2 cause both defective migration of cortical neurons in the developing brain and lead to severe nuclear defects (e.g. chromatin protrusions) during that phase, likely explaining the severe brain development defects and reduced neuronal survival⁶⁹. It remains to be determined whether these defects are caused by disrupted transmission of force during saltatory nuclear movement or a more fragile nucleus unable to bear the stress generated during nucleokinesis. Besides nucleokinesis, live imaging studies have found remarkable nuclear deformation and rotation during the migration of cerebellar granule cells through narrow intercellular spaces in neural tissues¹⁵⁸. During this process, microtubules steer the nucleus and drive its rotation and deformation through a dynamic interaction of nesprins with kinesin-1 and dynein. Given the apparent diversity of cytoskeletal organization in neuron species, further studies will be needed to obtain a better understanding of nuclear dynamics and nuclear shape regulatory mechanisms in neuronal cells.

Nuclear deformation during confined migration

Nuclear deformation is also a hallmark of important physiological and pathological situations involving cell migration. For instance, immune cells or invasive cancer cells must navigate through small interstitial spaces ranging from 1 to 20 μm in diameter^159,160, which requires cells to deform their nucleus to squeeze through the available spaces (Fig. 3f). In the absence of matrix metalloproteinase (MMPs) proteolysis, the nucleus is often the main physical hindrance to cell migration through confined spaces^162,173,174. Leucocytes can insert basolateral protrusion within (paracellular) or between (transcellular) endothelial cells to breach the endothelial barrier (Fig. 3f) and use actomyosin forces to push the nucleus through the pore, resulting in substantial nuclear deformation.

Tumor cells face similar challenges when invading tissues and intra- and extravasating blood vessels to metastasize to distant tissues¹⁶¹ (Fig. 3f). The primary sources of cytoskeletal forces to translocate and deform the nucleus are (i) actomyosin contractility that can cause both tension and compression of the nucleus by actin stress fibers pulling or pushing on the nucleus^174,177,178 or by generating hydrostatic pressure within cellular compartments that acts on the nucleus¹⁶², and (ii) microtubule-associated motors, i.e., kinesins and dyneins¹⁶³, which directly attach to the nucleus via nesprins and other NE proteins. Whether the nucleus is pulled and/or pushed is still debated¹⁶⁴, although it is likely that cells can use multiple independent mechanism, depending on the particular context (Fig. 3f). For example, LINC complexes recruit dynein and kinesin-1 to pull nuclei towards the minus ends of polarized microtubule networks during C. elegans development ¹⁸¹, whereas evidence for nuclear pushing has been reported in breast cancer cells and in glioma invasion, where non-muscle myosin IIB (NMIIB) at the cell rear pushes the nucleus forward^165–167. An additional mechanism has been observed in dendritic cells that uses Arp2/3 to generate lateral pushing forces that deform the nucleus to facilitate the migration process through narrow ECM pores¹⁶⁸. In the context of cancer cells, fibroblasts and macrophages from the stromal microenvironment may further aid in the invasion process by physically pulling on the tumor cells or degrading the extracellular matrix.

Notably, nuclear deformation during confined migration may also involves dynamic or persistent changes in nuclear mechanical properties. For example, transient nuclear softening has been reported during transendothelial migration of cancer cells¹¹¹, neutrophils develop highly lobulated and deformable nuclei during granulopoiesis that facilitates passages through tight spaces¹¹⁰, and highly invasive breast cancer cells are characterized by increased nuclear deformability and low lamin A/C levels¹¹³ (Fig. 3f).

Section 4 – Consequences of nuclear deformation

Given the central role of the cell nucleus in cellular function, it is easy to imagine how nuclear deformations can lead to various transient or persistent consequences, ranging from increased cell contractility, loss of NE integrity, DNA damage, and epigenetic modifications to altered cell differentiation. Notably, although these outcomes are now well established, the molecular mechanisms responsible, and whether the nuclear itself transduces mechanical signals into biological responses, often remains unresolved and a matter of active research (see textbox “Nuclear mechanoresponses and mechanosensing”).

Impact of NE tension on cellular proprioception and actomyosin contractility

Confinement of cells below a critical threshold, typically a fraction of the uncompressed nuclear height, results in nuclear flatting, an increase in nuclear membrane tension, and opening of nuclear membrane folds^125,169. These events trigger the release of calcium from the ER and the recruitment of phospholipase cPLA2 to the INM, where it catalyzes the production of arachidonic acid (AA), an omega-6 polyunsaturated fatty acid. AA has been implicated to regulate myosin II activity both directly¹⁷⁰ and indirectly via protein phosphorylation¹⁷¹. The resulting increase in cortical actomyosin contractility modulates cell morphology and promotes migration through narrow constrictions^125,169. Unfolding of the NE under increasing membrane tension allows the nucleus to deform without exceeding critical membrane tension in the nuclear membranes¹⁷², but may also trigger downstream signaling events¹⁷³. This nuclear mechanosensing of cellular confinement has been referred to as “cellular proprioception”. The physical properties of the large nucleus can also directly influence cellular processes. Recently, the microtubule mediated “frontward” positioning of the nucleus in amoeboid cell migration was shown to allow cells to use their nucleus as a mechanical gauge to determine the path of less resistance when encountering bifurcations of the path with pores of different size¹⁷⁴, providing an example of how deformation of the nucleus aids cells in their ‘decision making’ during migration through confined environments.

Deformation-associated NE rupture and repair

NE rupture describes the (transient) loss of nuclear membrane integrity at localized sites, rather than global breakdown of the NE. Spontaneous NE rupture events, persisting typically from a few minutes to several dozens of minutes, were first observed in vitro in cells expressing the HIV protein VPR¹⁷⁵, and subsequently in laminopathy patient fibroblasts¹⁷⁶ and cancer cells¹⁷⁷. Since then, it has become apparent that physical stress on the nucleus and the associated nuclear deformations can lead to transient NE ruptures, particularly during migration through confined environments, and that the probability of NE rupture increases with the degree of confinement^{71,72,178,179}. NE ruptures have been documented in vitro and in vivo. The NE ruptures are often associated with loss of A-type or B-type lamins^{49,109,154,197}, lamin mutations^182–184, peripheral heterochromatin disruption⁹¹ or high-level of mechanical stress, resulting from tensile or compressive forces on the nucleus^{70–72,116,185–188}. Based on super-resolution imaging and computational modeling, the NE rupture sites are estimated to be ≈100 nm in diameter^69,70,206. A current hypothesis proposes that NE ruptures occur at pre-exiting gaps or defects in the nuclear lamina, particularly where the lamin B meshwork is weaker and thus cannot sufficiently support the nuclear membranes, causing the membrane to form a bleb that expands under continued mechanical stress and ultimately ruptures (Fig. 4)^190,179. However, NE rupture and membrane blebs have also been observed in the absence of nuclear lamina gaps; they may thus generally arise when the nuclear membrane peels off the underlying nuclear lamina in response to increased nuclear pressure resulting from cytoskeletal forces^71,191,192. A better understanding of the nucleation mechanism of nuclear ruptures will require to study the dynamics of the heterogenous lamina meshwork and its interaction with the nuclear membranes during nuclear deformations.

Figure 4 – Nuclear envelope rupture and repair.

Migration through confined environments or external compression of cells can result in nuclear membrane bleb formation and NE rupture. The nuclear membrane rupture process is typically initiated by the formation of a nuclear membrane extrusion, or bleb. Blebs from at sites with high nuclear membrane curvature and where an initial defect in the nuclear lamina exists. Blebs are driven by increased hydrostatic pressure within the nucleus. Initially, only the nuclear membrane detaches from the lamina. Later, lamin A/C and chromatin can enter the bleb. The lifetime of blebs can be minutes to hours, but the rupture itself is usually quite short, on the orders of minutes. Blebs can have varying size and can contain chromatin or are just fluid filled. Nuclear blebs typically lack lamin B (green) and NPCs. Continued nuclear compression by confinement from the extracellular matrix, apical actin stress fibers, cell contractions, or external compression results in bleb expansion until the nuclear membranes in the bleb exceeds a critical threshold and ruptures, leading to the leakage of soluble proteins from the nucleoplasm into the cytoplasm and uncontrolled influx of cytoplasmic proteins into the nucleus. Following NE rupture, barrier-to-autointegration factor (BAF) is rapidly (< min) recruited to initiate NE repair. The recruitment of ESCRT III complexes further contributes to resealing the nuclear membranes. The process of repair/rescue is typically completed within 10–15 min and often associated with recruitment of lamin A/C to the site of rupture. Although the NE rupture is resealed, the bleb/protrusion often persists and is not fully resorbed.

As the transient nature of most NE ruptures documents, cells have robust mechanisms to repair their nuclear membrane during interphase, and even longer rupture events (few hours) can eventually be repaired¹⁹³. The mechanisms involved in interphase nuclear membrane repair are largely shared with those during resealing of the NE post mitosis. The nuclear membrane repair mechanism is based on the recruitment of specific proteins to the sites of NE rupture, particularly BAF, LEM-domain proteins, the endosomal sorting complexes required for transport (ESCRT)-III remodeling complex, and CHMP7 ^{72,181,186,193,194}. Considering that the extent of rupture is correlated with the amount of cytoplasmic BAF accumulating at the rupture site^72,193,194, a current model of nuclear membrane repair considers that the binding of cytosolic BAF to the exposed chromatin initializes recruitment of both new ER membranes to repair the membrane hole and the ESCRT-III complex to reseal the remaining gaps (Fig. 4). Interestingly, some nuclear processes such as transcription and DNA replication can be disturbed after NE rupture events, leading for instance to aneuploidy or extensive DNA damage such as double-stranded DNA breaks that cannot be repaired due to loss of DNA damage repair proteins¹³².

Mechanically induced DNA damage

Severe nuclear deformations, occurring for example during confined migration, external compression, or nuclear repositioning, can induce DNA damage upon NE rupture^{68,69,212,213} and even in the absence of NE rupture¹⁹⁶. NE rupture can cause DNA damage by allowing access of the ER-associated exonuclease TREX1 into the nucleus¹⁷², or by loss of DNA damage repair factors from the nucleus^195,197. Whereas NE rupture associated DNA damage occurs throughout all phases of the cell cycle, and more often in Ataxia Telangiectasia and Rad3-related protein (ATR)-defective cells¹⁹⁸, the deformation-induced-DNA damage (i.e., DNA damage in the absence of NE rupture) occurs primarily in S/G2 phases, i.e., during active DNA replication, and is linked to increased replication stress, possibly due to torsional stress on DNA resulting from the nuclear deformation during confined migration or mechanically compression of cells¹⁹⁶. Interestingly, different cell lines exhibit different propensities for these modes of DNA damage^172,196, but the exact molecular reasons for these cell type specific differences remain to be elucidated.

What are the long-term consequences of DNA damage and NE rupture for cells and tissues homeostasis? Repeated migration through tight constrictions can lead to accumulation of DNA damage and changes in chromosome copy number¹⁹⁵. Furthermore, TREX1-dependent DNA damage following NE rupture may favor tumor cell invasion by inducing a partial epithelial-to-mesenchymal transition (EMT) phenotype via ATM and SNAIL1 that leads to MMP dependant collagen degradation¹⁷². The precise mechanisms linking TREX1 and collagen degradation activity through MMP’s activity is still unknown, but is believed to be downstream of the ATM DNA damage response pathway^199,200. NE rupture can also lead to activation of the cyclic GMP-AMP synthase-stimulator of interferon genes (cGAS-STING) DNA-sensing pathway, as it allows access of cytosolic cGAS to the genomic DNA at sites of NE rupture^71,72,201. Intriguingly, a recent study found that increased cGAS-STING signaling can drive cancer metastasis in a mouse breast cancer model²⁰¹, although in this case the cGAS-STING activation was primarily due to NE rupture of micronuclei, not primary nuclei.

Mechanically induced epigenetic modifications

Local cytoskeletal forces such as actin-based indentation acting on the nucleus can severely deform the NE and trigger reversible formation of heterochromatin at the NE periphery^46,135,202. Local stresses applied to integrins can propagate to the LINC complex through the actin cytoskeleton and lead to chromatin un-packing²⁰³ and force-induced transcription that requires H3K9me3 demethylation²⁰⁴. Nuclear deformation during confined migration can also induce increased heterochromatin formation in a histone methylase and histone deacetylase dependent process, which promotes cell migration²⁰⁵. Furthermore, nuclear deformations associated with confined migration, cell compression, or cell stretching can lead to chromatin re-arrangements and the increased formation of heterochromatin that can last from hours to days²⁰⁵. In macrophages, spatial confinement can suppress late lipopolysaccharide (LPS)-activated inflammatory transcriptional programs (e.g., expression of IL-6, CXCL9, IL-1β, and iNOS) by modulating chromatin organization and epigenetic alterations^216,206. In the context of cell migration, the increase in H3K9me3 and H3K27me3 heterochromatin marks promotes confined cell migration through yet to be defined mechanisms^114,205. The molecular details by which mechanical deformation of the cell and nucleus result in increased heterochromatin formation remain incompletely understood, but two major contributors have emerged to date: an increase in intracellular cations (calcium and/or magnesium) by activation of stretch-activated ion channels, and remodeling of the nuclear and/or perinuclear actin network.

Repetitive stretching of mesenchymal stem cells activates mechanosensitive ion channels (Fig. 5a) such as Piezo1, leading to increased intracellular calcium levels and increased heterochromatin (H3K9me2,3) formation, and, ultimately promoting mesenchymal differentiation^207,208. In epithelial cells, cyclic mechanical stretch triggers immediate nuclear deformation that leads to Piezo1-mediated calcium release from the ER, reducing lamina-associated heterochromatin (H3K9me3 marks) within a ≈30 minutes window, and resulting in nuclear softening that decreases stress and DNA damage in the stretched cells¹⁹. Long-term (8–12 hours) cyclic uniaxial stretch application causes transcriptional repression, increased heterochromatin (H3K27me3), and silencing of differentiation gene expression¹⁹. Intriguingly, activation of mechanosensitive ion channels by increasing extracellular multivalent ion concentrations, even in the absence of cell stretching or compression, is sufficient to trigger similar increase in heterochromatin⁹¹. The increased heterochromatin content mechanically strengthened the nucleus, rescued abnormal nuclear morphology in LMNA mutant and breast cancer cells, reduced NE ruptures, and prevented DNA damage⁹¹. Collectively, these findings demonstrate that mechanosensitive ion channels respond to mechanical stimuli, causing an increase in intracellular calcium that leads to chromatin modifications, which mechanically protect the nucleus and influence cell fate decisions. These stretch-sensitive ion channels can be found on the plasma membrane, the ER, and potentially the NE, with the contribution of specific channels and their locations likely depending on the particular cellular context. At least in some cases, influx of extracellular calcium, rather than release from intracellular stores, appears to be sufficient to trigger chromatin remodeling^91,208.

Figure 5 – Schematic illustration of nuclear mechanoresponses and mechanosensing.

(a) High tension exerted on the NE during nuclear deformations induces unfolding of the wrinkled NE and the opening of stretch-activated ion channels. (b) Deformation of the nucleus induces enrichment of non-muscle myosin and emerin at the ONM. Relocalization of emerin to the ONM promotes perinuclear actin polymerization that leads to decreased levels of free nuclear monomeric actin, thereby reducing global transcriptional activity and increasing heterochromatin formation. The decrease of emerin at the INM leads to a loss of H3K9me2 and heterochromatin maintain their silenced state by recruiting H3K27me3. (c) Increased nuclear membrane tension stretch NPCs, leading to increased nuclear import of transcription factors (TFs) and mechanoresponsive transcriptional activators, such as YAP. (d) High NE tension resulting from nuclear deformation induces nuclear membrane unfolding, subsequent calcium release and the recruitment of cytosolic phospholipase A2 (cPLA2) activated by phosphorylation to the nuclear periphery, which promotes production of arachidonic acid (AA). The activation of the cPLA2-AA pathway leads to myosin II via AA-mediated Rhoa activation recruitment to the cell cortex, increasing actomyosin contractility. (e) Nuclear deformation induces phosphorylation of emerin and conformational changes in lamin A/C, which can alter the interaction with binding partners and induce further signaling events or recruit other proteins to the NE. (f) Forces acting on the nucleus may reposition or locally unfold chromatin domains, altering their transcriptional activity, and modulate the methylation level of histones by methyltransferases, regulating the transcriptional activity.

Changes in perinuclear actin polymerization, mediated by relocalization of emerin to the ONM (Fig. 5b), can result in increased facultative heterochromatin formation by depleting monomeric actin from the nucleus, reducing transcription, and activating the polycomb repressive complex (PRC2)¹⁰². Mechanically induced actin depolymerization can also lead to translocation of the histone deacetylase 3 (HDAC3) from the cytoplasm into the nucleus, resulting in increased heterochromatin formation²⁰⁹. Spatial confinement can similarly reduce actin polymerization, thereby reducing nuclear translocation of megakaryotic leukemia 1 (MKL1), a myocardin-related coactivator of the serum response factor that regulates many physiological processes, which can lead, for example, to altered inflammatory signaling in macrophages exposed to LPS²⁰⁶. Emerin-mediated actin polymerization can modulate nuclear translocation of the mechanoresponsive transcription factor MKL1, thereby altering expression of MKL1/SRF response genes such as Srf or vinculin²¹⁰.

Nuclear deformations affect transcription factor translocation and cell differentiation

Nuclear deformations can also modulate the balance of nuclear and cytoplasmic pools of the mechanoresponsive transcriptional regulators YAP (Yes-associated protein) and TAZ (Transcriptional coactivator with PDZ-binding motif)²¹¹ (Fig. 5c), which play crucial roles in regulating a wide range of key biological processes²¹². For instance, the differentiation of myoblasts into myotubes which requires nuclear deformations was associated with the nuclear export of epigenetic regulator and transcription factors such as SMYD3 lysine methyltransferase and YAP²¹³. Whereas activity and nuclear translocation of YAP/TAZ have been traditionally viewed as being controlled by cytoplasmic pathways²¹⁴, recent findings point to nuclear deformation as an additional modulator. For example, compressive forces lead to nuclear flattening and NPC opening¹⁸, which increases YAP nuclear import²¹⁵. More recently YAP nuclear export was associated with substrate curvature changes that impose nuclear deformation. Nuclei located on convex zones (i.e. crests) were flattened with YAP translocated to the nucleus and chromatin less condensed, whereas nuclei on concave zones (i.e. valleys) were highly elongated, contained more condensed chromatin, and YAP was predominantly cytoplasmic²¹⁶. These findings support the notion of a control of YAP/TAZ by nuclear deformation and highlight the importance of mechanical and cytoskeletal regulation of the nuclear shape in YAP/TAZ regulation. Accumulative evidence has shown that similar effects can be observed by imposing nuclear shape changes with higher cell density²¹⁷ or various external forces^{30,129,211,215}, without changing the matrix properties. However, precisely how the intracellular localization of YAP is modulated by nuclear volume changes²¹⁵ and how this observation relates to known regulators of YAP nuclear translocation remain to be elucidated.

Recent evidence suggests that the cytoskeleton can modify not only the physical state of the nucleus, but also the chromatin state. Sustained activity of MKL1, a key transcriptional co-activator of SRF that can be activated by mechanically induced actin polymerization and that regulates expression of many actin cytoskeletal genes, results in reduced nuclear volume and globally reduced chromatin accessibility²¹⁸, consistent with a recent study that found that actin polymerization reduced pluripotency in induced pluripotent stem cells by decreasing chromatin accessibility. An additional example of how forces on the nucleus can modulate cell fate decisions comes from recent work on myofibroblasts, which found that the persistence of the myofibroblast phenotype relies on chromatin remodeling mediated by nuclear mechanosensing of cytoskeletal forces via LINC complexes²¹⁹. Furthermore, human mesenchymal cells respond to matrix stiffening by increasing nuclear membrane tension and histone acetylation via deactivation of histone deacetylases (HDACs), leading to osteogenic fate determination²²⁰. On the other hand, LINC complex disruption, which presumably reduces nuclear membrane tension, leads to upregulation of HDACs and inhibits osteogenic differentiation²²⁰. In another example, migration of myoblasts through confined environments, which is associated with substantial nuclear deformations, led to delayed myoblast differentiation²²¹. These findings may be particularly relevant to fibrotic tissues (e.g., in aging, muscular dystrophies), in which the regeneration ability of muscle stem cells is compromised²²², and illustrate the impact of nuclear deformation on cell differentiation.

Conclusion and perspectives

Significant efforts in the recent years have started to shed light on the fascinating roles of nuclear deformations in cellular responses and human disease pathogenesis (see textbox “Human pathologies associated with nuclear deformations”). The disruption of nucleo-cytoskeletal coupling, alteration of nuclear mechanics, or defective mechanotransduction signaling can cause the emergence of various human diseases. Chromatin organization, compaction, stretching, and modification that arise from nuclear deformations control downstream expression of genes and cell fate decisions, although often the specific molecular mechanisms, and the precise site of where cellular mechanosensing occurs, remain to be explored in further detail. Altogether, these discoveries have revealed the remarkable mechanoresponsive nature of the nucleus and the key role of NE proteins in the cellular response to mechanical stimuli. However, many open questions remain, such as the mechanisms used by the nucleus to sense force and/or deformation, or how deformation of the nucleus may result in activation or epigenetic modification of specific genes. Although substantial progress has been made in the understanding of nucleo-cytoskeletal coupling, the precise mechanisms for the spatio-temporal regulation of force transmission across the LINC complex required for many cellular functions has yet to be fully elucidated. Connections between the nucleus, other organelles, and the plasma membrane have received far less attention and should be investigated in more detail. Inside the nucleus, a better understanding of the role nuclear F-actin and motor proteins, as well as LLPS processes, in the maintenance of the nuclear structure, genomic organization, and chromatin remodeling will require deeper investigation.

Deciphering the complex mechanical interplay between chromatin, the NE, cytoskeletal filaments, and the cell surface in mechanobiology will benefit from interdisciplinary and integrative approaches, combining live-cell imaging with high spatial and temporal resolution, genetic manipulation, and precise mechanical manipulation. Much of our knowledge in nuclear mechanotransduction has come from innovative technologies. Addressing current challenges in this field will require further technological innovations, for instance to visualize gene expression in live cells while exerting sub-cellular deformations, ideally on a genome-wide scale, and yet with single cell resolution. In addition to these experimental breakthroughs, mechano-chemical models of the nucleus developed by theoretical modelling will be essential to explore how the cooperation between mechanical and biochemical parameters regulates feedback loops in nuclear signaling pathways. A better understanding of the molecular mechanisms governing nuclear mechanobiology would not only clarify how the various cellular mechanotransduction pathways are combined to determine downstream cellular function but may also guide the development of novel therapeutic strategies to treat human diseases that arise from impaired nuclear mechanotransduction signaling and disturbed nucleo-cytoskeletal force transmission.

Glossary

Mechanosensing

Active cellular process through which cells or cellular components detect changes in external forces or mechanical properties of their microenvironment

Mechanotransduction

Molecular process in which mechanical stimuli are converted (or transduced) into biochemical signals. The mechanotransduction is the central process of cellular mechanosensing

Chromatin compaction

Process that occurs predominantly at the mesoscale during nucleosome packing leading to chromatin fibers. Epigenetic modifications associated with heterochromatin lead to increased chromatin compaction

Nuclear transport

Passive (cargo ≤ ≈50 kDa) or active (cargo ≥ ≈50 kDa) transport of molecules between the cytoplasm and nuclear interior through nuclear pore complexes

Stress

Expression of the mechanical loading in terms of force applied per cross-sectional area of an object. Units of stress are N/m² (or Pa)

Farnesylation

post-translational modification of proteins catalyzed by the enzyme farnesyltransferase which adds a 15-carbon unsaturated hydrocarbon chain to a cysteine residue via a thioether linkage, thus anchoring the protein to a lipid membrane

Focal adhesions

Integrin-mediated cell-substrate adhesion structure anchoring the ends of stress fibers. Focal adhesions mediate strong attachments to substrates and function as an integrin-signaling platform

Stress fiber

Actin filaments assembly resulting from the interaction and merging of pre-existing radial fibers and transverse arcs (10–30 filaments). Stress fibers can reach a diameter of several hundreds of nanometers and are under constant prestress due to actomyosin contractility

LINC complex

The Linker of Nucleoskeleton and Cytoskeleton complex consists of SUN-domain proteins at the INM and KASH domain proteins at the ONM and is crucial for force transmission across the NE. SUN-domain proteins are associated with both nuclear lamins and chromatin and cross the inner nuclear membrane. They interact with the KASH domain proteins in the perinuclear (lumen) space between the two membranes. The KASH domain proteins cross the outer nuclear membrane and interact with actin filaments, microtubule filaments (through dynein and kinesin motors), intermediate filaments (through plectin), centrosomes and cytoplasmic organelles

Intermediate filaments

Large family of cytoskeletal filaments that includes keratins (types I and II), desmin and vimentin (type III), neurofilaments (type IV), and lamins (type V). Intermediate filaments form dimers that then assemble into larger non-polar filament structures that are characterized by their ability to extend substantially under mechanical stress

Tensile force

Pulling force resulting in the extension of an object

Viscoelastic or Viscoelasticity

Rheological behaviour of natural or synthetic materials that exhibit a combination of elastic and viscous properties

Viscous or viscosity

Measure of the resistance of a fluid to deform under either shear or extensional stress, defined as the ratio of shear stress to shear flow. It is commonly perceived as the resistance of a liquid to flow, usually represented by a dashpot, and is related to the internal friction within the fluid. Units of viscosity are Pa/s

Strain

Geometric measure of the amount of deformation in the direction of the applied force divided by the initial length of the object (unitless number)

Strain stiffening

Mechanical material property corresponding to a sudden increase of the elastic modulus under strain, i.e., an increase in resistance to further deformation

Plastic deformation or plasticity

Ability of a solid material to undergo permanent deformation (i.e. non-reversible change of shape) without rupture in response to applied forces

BAF

Barrier-to-autointegration factor (BAF) is an essential 10 kDa chromatin-binding protein that is highly conserved in metazoan that helps DNA anchoring to the NE. BAF is involved in multiple pathways including NE reassembly (after mitosis and NE rupture), chromatin epigenetics, and DNA damage response. BAF’s function is controlled by phosphorylation/dephosphorylation waves that drive nuclear disassembly

Colloid osmotic pressure

Pressure generated by solutions of macromolecules in contact with pores that are permeable to water and ions but not macromolecules. Colloid osmotic pressure generates depletion forces that push macromolecules together in crowded solutions and thus promotes aggregation and phase separation

Confocal Brillouin microscopy

Optical technique combining Brillouin spectroscopy with confocal microscopy to provide a non-contact and direct readout of the mechanical properties of a material (i.e. stiffness, temperature or strain) at the micrometer scale. Spontaneous Brillouin light scattering arises from the interaction between photons and acoustic phonons (i.e., propagation of thermodynamic fluctuations) and permits to quantify intracellular longitudinal modulus without disturbing the cell

Interkinetic nuclear migration

periodic movement of the nucleus between apical and basal surfaces of neuroepithelial progenitor cells as they progress through the cell cycle

Epigenetic modification

Response and adaption of an organism to its environment by altering the (local) structure of chromatin, without changes in the genomic sequence, thereby altering the expression of genes. Epigenetic regulation is often achieved by acetylation or methylation of histones and/or DNA

Laminopathy

Over 180 mutations have been reported in the genes of the nuclear lamina, in particular LMNA, causing diseases termed ‘laminopathies.’ The number of identified laminopathies has steadily increased in recent years, currently including 13 known conditions. heterogeneous genetic disorders that have been associated with mutations in LMNA and, most recently, LMNB1 and LMNB2

LEM domain protein

The LAP2, Emerin and MAN1 (LEM)-domain is a ≈40-residue helix-loop-helix fold conserved both in eukaryotes and in prokaryotic DNA/RNA-binding proteins. Except Lap2 proteins, which have a second LEM-domain that binds DNA, the function of eukaryotic LEM-domains is to directly bind a conserved chromatin protein named barrier-to-autointegration factor (BAF)

Aneuploidy

Deviation from the normal diploid karyotype of 46 chromosomes with the presence of one or more extra chromosomes or the absence of one or more chromosomes

Epithelial-to-mesenchymal transition (EMT)

Transcriptionally governed process over which epithelial cells establish a front-rear polarity while acquiring a mesenchymal and motile phenotype

TREX1

Three prime repair exonuclease 1 is major 3′ → 5′ DNA exonuclease in mammalian cells, metabolizes preferentially single-stranded DNA (ssDNA). It cleans the cytosol from DNA fragments coming from endogenous elements. Unless degraded, the accumulation of these DNA fragments can activate innate immune signaling

cGAS-STING pathway

Cellular cytosolic double-stranded DNA (dsDNA) sensor, allowing innate immune response to infections, inflammation, and cancer

Micronuclei

Small DNA-containing nuclear structures that are spatially isolated from the main nucleus. Micronuclei form from lagging chromosomes or chromosome fragments following mitotic errors or DNA damage, respectively

Facultative heterochromatin

Condensed transcriptionally silent chromatin region that can decondense and adopt to allow transcription within temporal and spatial contexts. Facultative heterochromatin is not characterized by repetitive sequences, so at the DNA sequence level it is entirely different from constitutive heterochromatin

Polycomb repressive complex (PRC2)

Major repressive chromatin complex formed by the polycom Group (PcG) proteins

Serum response factor (SRF)

Transcription factor that plays a key role in the transduction of mechanical signals from cytoplasmic actin and extracellular matrix proteins to the nucleus. SRF is involved in various cellular processes, from cell proliferation to differentiation and devlopment

Topologically associating domains (TADs)

Self-interacting megabase-scale genomic blocks in which DNA sequences exhibit significantly higher interaction frequencies with other DNA sequences within the domain than with those outside of the block