Push or pull: how cytoskeletal crosstalk facilitates nuclear movement through 3D environments

Abstract

As cells move from two-dimensional (2D) surfaces into complex 3D environments, the nucleus becomes a barrier to movement due to its size and rigidity. Therefore, moving the nucleus is a key step in 3D cell migration. In this review, we discuss how coordination between cytoskeletal and nucleoskeletal networks is required to pull the nucleus forward through complex 3D spaces. We summarize recent migration models which utilize unique molecular crosstalk to drive nuclear migration through different 3D environments. In addition, we speculate about the role of proteins that indirectly crosslink cytoskeletal networks and the role of 3D focal adhesions and how these protein complexes may drive 3D nuclear migration.

1. Introduction

Cellular forces are essential for cell migration (Bodor et al., 2020; Yamada and Sixt, 2019). When crawling across two-dimensional (2D) surfaces cells use actomyosin contractility generated by the non-muscle myosin II (NMII), which is regulated by the small GTPase RhoA, to probe the stiffness of the underlying substrate (Schiller et al., 2013), control focal adhesion formation (Chrzanowska-Wodnicka and Burridge, 1996), and retract the trailing edge (Worthylake et al., 2001). Microtubules help to polarize and steer the cell in part by guiding the constitutive secretion of membrane and proteins to the cell’s leading edge (Bergmann et al., 1983), while intermediate filaments like vimentin or lamin provide structural strength to the cytoplasm and nucleus, respectively (Patteson et al., 2020). Importantly, there are no confining surfaces or 3D matrix fibrils slowing the passage of the bulky nucleus in cells migrating on 2D. As a result, no additional significant forces are necessary to keep this essential nuclear cargo moving with the bulk of the cytoplasm.

In contrast, 3D environments such as type I collagen or narrow channels pose a barrier to movement to cells due to physical factors such as pore size, crosslinking, and varying levels of confinement. Rather than relying on one type of migration mechanism or mode to move through these structurally complex environments, a single type of cell can use up to three distinct modes of migration (Liu et al., 2015; Petrie et al., 2012) that are selected partially based on the cell’s physical environment and the extent of cell-matrix adhesion and actomyosin contractility (Petrie and Yamada, 2016). Despite this array of distinct 3D migration mechanisms, a universal principle has emerged that is remarkably consistent across multiple diverse cell types, from fibroblasts to epithelial cells, and from leukocytes to malignant cancer cells. Specifically, these cells all must overcome the problem of how to move their unwieldy nucleus through the narrow openings commonly found in 3D environments (Harada et al., 2014; Wolf et al., 2013). Interestingly, cells have the ability to modify the relative stiffness and deformability of their nuclei independently of cellular stiffness by controlling the expression of lamin A/C (Harada et al., 2014) and, potentially, nuclear pressure (Mistriotis et al., 2019).

Fascinatingly, not only can cells successfully move the nucleus through the narrow openings in 3D using a variety of strategies, the nucleus, with its myriad of cytoskeleton connections (Lombardi et al., 2011), is in fact required for 3D cell migration (Graham et al., 2018; Petrie et al., 2014). Further, recent work has placed the nucleus at the center of the cellular decision-making process dictating how cells move in 3D environments (Lomakin et al., 2020; Renkawitz et al., 2019; Venturini et al., 2020). Even though this field is relatively new, it is becoming clear that molecular crosstalk between the cytoskeletal networks and at the nucleoskeletal-cytoskeletal interface are critical to moving the nucleus through complex 3D environments. Additionally, there remain many questions about the role of focal adhesions in nuclear movement during 3D migration. While it is clear cell-matrix adhesions are absolutely required for the migration of non-amoeboid modes of 3D migration, such as the lobopodia- and lamellipodia-based movement of mesenchymal cells (Doyle et al., 2021; Petrie et al., 2012), it remains uncertain how these structures contribute to the cytoskeletal crosstalk that is essential for the movement of the nucleus in 3D.

In this review, we discuss what is known about how cytoskeleton and nucleoskeleton crosstalk within the cell dictates 3D nuclear movement. We then consider how the cytoskeleton is linked to the nucleus through the linker of nucleoskeleton and cytoskeleton (LINC) complex and allows the nucleus to be pulled through the tight spaces within 3D extracellular matrices. Lastly, we speculate about the role of focal adhesion complexes in driving 3D nuclear movement based on evidence that focal adhesion turnover and 3D migration are linked and highlight some key questions in the field of nuclear mechanotransduction.

1.1. Moving the Nucleus through 3D Environments

During 3D migration, the nucleus is the rate-limiting step to cell migration due to its bulky nature (Davidson et al., 2014; Harada et al., 2014; Mistriotis et al., 2019; Wolf et al., 2013). In fact, the forces that have to be applied to the nucleus to move it through the narrow spaces in 3D environments generate significant mechanical stresses. These stresses manifest as significant deformation of the nucleus and can pressurize the nucleus relative to the cytoplasm, causing the nuclear envelope to rupture periodically in migrating cells (Denais et al., 2016; Mistriotis et al., 2019; Xia et al., 2018). For example, it was shown in human mesenchymal stem cells that increasing confinement causes decreased cell motility due to nuclear deformation (Doolin and Stroka, 2018). However, the nucleus is not just a burden during this process, it can actively participate in the process of migration. This is highlighted by the fact that cells migrating on 2D continue to do so after the nucleus is removed (Graham et al., 2018). In contrast, the presence of the nucleus is required for cells migrating through 3D environments. This contradiction can begin to be understood by the fact that the nucleus sits at the center of the extensively connected cytoskeleton network in cells (Lombardi et al., 2011). Further, these connections are used to help the cell make decisions on how to migrate. For example, leukocytes use the nucleus to determine the size of potential 3D migration paths, using it is a mechanical sensor to determine the width of the passage before committing the cell to moving into it (Renkawitz et al., 2019). Additionally, mesenchymal cells use the cytoskeleton to move the nucleus like a piston in a cylinder to pressurize the cytoplasm and generate lamellipodia-independent protrusions when migrating in 3D matrices (Lee et al., 2021; Petrie et al., 2014). It is now clear that cellular machinery exists to either push or pull the nucleus forward when moving through 3D environments and cytoskeletal crosstalk plays a critical role in this decision.

1.2. The Plasticity of 3D Migration

3D migration employs migratory plasticity wherein a cell can utilize specific mechanisms to move through different types of environments (Charras and Sahai, 2014; Even-Ram and Yamada, 2005; Hui and Pang, 2019; Yamada and Sixt, 2019). 3D cell migration can be categorized based on the level of integrin engagement and actomyosin contractility utilized by the cell to move through a certain environment (Petrie and Yamada, 2016). Amoeboid migration is driven by the robust retrograde flow of cortical actin coupled with low levels of integrin engagement and high levels of actomyosin contractility (Gabbireddy et al., 2021; Liu et al., 2015). It has been well-illustrated in both leukocytes and Dictyostelium that F-actin crosslinking and actomyosin contractility-generated cortical tension can help drive fast amoeboid migration in 3D environments (Alvarez-Gonzalez et al., 2015; Friedl et al., 2001). Cells migrating in confinement with low adhesion and contractility can also take advantage of the directional flow of water being taken in at the leading edge and expelled at the trailing edge to power the osmotic engine of 3D cell movement (Stroka et al., 2014). In contrast, mesenchymal migration is driven by both integrin engagement and actomyosin contractility. For example, cells migrating through collagenous matrices utilize similar mechanisms to cells on 2D surfaces. Specifically, they engage integrins, utilize actomyosin contractility, and form branched actin-based protrusions called lamellipodia at their leading edge (da Rocha-Azevedo et al., 2013; Doyle et al., 2021; Petrie et al., 2012). When mesenchymal cells use strong cell-matrix adhesions to migrate in 3D environments increase their actomyosin contractility to help move the nucleus, they switch from low-pressure lamellipodial migration to high-pressure lobopodial migration. During this high-pressure mode of migration mesenchymal cells use their nucleus like a piston to compress the cytoplasm and increase its hydraulic pressure (Lee et al., 2021; Petrie et al., 2014). Tumor cells can also oscillate between amoeboid and lamellipodia-based migration based on changes in confinement and adhesion (Friedl and Wolf, 2010; Sanz-Moreno et al., 2011).

Classically, crosstalk between the Rac1 and RhoA pathways govern the switch between lamellipodia-dependent and -independent modes of migration (Sanz-Moreno et al., 2008; Wolf et al., 2003). More recently, the pathways downstream of the small GTPases have garnered attention in how they dictate cell movement. As a result, we now understand that the ability of cells to switch between different modes of 3D migration could be mirrored in the plasticity of the actomyosin network and how it is used to move the nucleus through 3D environments. Actomyosin filaments can generate contractility either in front of (Cao et al., 2016; Davidson et al., 2020; Petrie et al., 2017) or behind (Hetmanski et al., 2019; Lee et al., 2021; Lomakin et al., 2020; Poincloux et al., 2011) the nucleus to help force the nucleus through 3D environments depending on the migration mode being utilized. For example, in lobopodial cells migrating through 3D cell-derived matrix (CDM), tropomyosin 1.6 positive actomyosin filaments are concentrated in front of the nucleus to pull the nucleus forward (Sao et al., 2019). Recently, it was discovered that mesenchymal stem cells (MSCs) placed in protease-resistant alginate-based hydrogels utilize a form of the nuclear piston mechanism wherein actomyosin contractility at the rear of the nucleus pushes the nucleus forward (Lee et al., 2021). Similarly, fibrosarcoma cells migrating in 3D CDM and confined channels show increased NMII activity behind the nucleus, suggesting the nucleus is pushed forward from the back (Mistriotis et al., 2019). Finally, it was illustrated in neuronal cells migrating through both brain tissue and Matrigel that a “cup-like” organization of actomyosin filaments pushed the nucleus through the dense tissue (Martini and Valdeolmillos, 2010; Schaar and McConnell, 2005). Thus, actomyosin contractility plays a role in both pushing and pulling the nucleus, as needed, through varying substrates and is a key player in mediating migratory plasticity. Interestingly, while the requirement for actomyosin contractility is clear regardless of whether the cell is pushing or pulling the nucleus, the functional connections between the cytoskeleton and the nucleus can be distinct (Lee et al., 2021; Petrie et al., 2014).

2. Nucleo-cytoskeleton crosstalk and 3D nuclear movement

In addition to cross-talking with each other, cytoskeletal networks can also interact indirectly with the nucleoskeleton on the inner surface of the nuclear envelope. This interaction is mediated by the LINC (linker of nucleoskeleton and cytoskeleton) complex (Kim et al., 2014; Lombardi et al., 2011; Luxton et al., 2010), which can help form a continuous mechanical connection from the nucleus to the extracellular matrix (ECM) to move the nucleus forward (Fig. 1).

Figure 1. The LINC complex connects the nucleoskeleton with the cytoskeleton to facilitate 3D nuclear movement.

Nuclear lamins are linked to the cytoskeleton through inner nuclear membrane-associated proteins SUN1/2 which bind to outer-nuclear membrane KASH proteins called nesprins. This LINC complex-mediated association between the nucleus and cytoskeleton drives 3D nuclear movement. For example, nesprin-2G mediated actin connections to the nucleus are necessary to pull the nucleus through narrow constrictions (Davidson et al., 2020).

LINC complexes, consisting of KASH and SUN proteins embedded within the outer and inner nuclear membranes, respectively, bridge the nuclear envelope and help to connect the nucleoskeleton to the cytoskeleton (Sosa et al., 2012). The nuclear envelope consists of the inner and outer nuclear membrane, which are linked through SUN and KASH proteins, respectively. At the inner nuclear membrane, the proteins SUN1 and SUN2 link the inner membrane to the outer membrane by interacting with and controlling the positioning of outer nuclear membrane associated nesprins (Crisp et al., 2006; Padmakumar et al., 2005). This interaction between SUN proteins and nesprins is necessary for the transmission of force to the nucleus, which drives nuclear translocation and cell movement (Cain et al., 2018; Hao and Starr, 2019; Lombardi et al., 2011). In fact, interactions between SUN proteins and Nesprin-2G mediate actin accumulation at the front of the nucleus which aids in pulling the nucleus through in vitro constrictions (Davidson et al., 2020). Also, it has been demonstrated that vimentin intermediate filaments are linked to the nucleus via nesprin-3 and play a role in nuclear movement (Ketema et al., 2013; Postel et al., 2011; Wilhelmsen et al., 2005). Further, nesprin-3 mediated vimentin organization has been shown to mediate 3D lobopodial migration of cells in crosslinked matrices by helping to pull the nucleus forward in combination with NMII activity (Petrie et al., 2014). In contrast, nesprin 3 is not required for the nuclear piston mechanism used by MSCs in protease-resistant hydrogels. These cells rely on cytoskeleton-nucleoskeleton connection mediated by nesprin 2 to help push the nucleus forward to trigger the formation of pressure-based protrusions (Lee et al., 2021).

Nuclear lamins are type V intermediate filaments that lie just beneath the inner nuclear membrane and provide structural support to the nucleus. Lamins play a key role in regulating nuclear deformation and cell migration (Denais et al., 2016; Gerace et al., 1978; Harada et al., 2014; Krause et al., 2019; Lautscham et al., 2015). In 3D ECM, lamins play a role in nuclear protection and the stiffening of the nucleus through lamin A poses a hindrance to 3D migration (Harada et al., 2014). For example, leukocyte migration through collagen I barriers is slowed down by lamin A overexpression, however migration is sped up when these leukocytes migrate through endothelial barriers (Yadav et al., 2018). Further, when bone marrow derived mesenchymal stem cells (BMSCs) are placed in stiff gel matrices, lamin A expression is decreased allowing for nuclear deformability and increased migratory rates compared to BMSCs in soft gels (Lin et al., 2019). Critically, it has been demonstrated that nuclear lamins are necessary for 3D but not 2D migration (Khatau et al., 2012). Specifically, lamin A/C and nesprins are both necessary for actin association with the nucleus through the formation of an actin cap that plays a role in 3D migration. When either lamins or nesprins are depleted, these connections are severed and 3D cell migration slows, indicating that these nucleo-cytoskeletal connections are required for efficient 3D migration.

2.1. Nesprin 2-microtubule linkages drive protease-based 3D migration through collagen

For nearly two decades, we have known that interactions between SUN1/2-nesprin1/2 and dynein motors plays a role in centrosome positioning relative to the nucleus and driving nuclear movement (Malone et al., 2003; Zhang et al., 2009). Recently, it was demonstrated in MDA-MB-231 breast cancer cells that nesprin 2-mediated microtubule connections to the nucleus facilitates nuclear movement through confined collagen matrices (Infante et al., 2018). In these cells, the dynein motor Lis1 plays a role in anchoring microtubules and driving force to the nucleus to pull the nucleus forward. These cells escape confinement through protease activity via MT1-MMP, which drives invasion through collagen by invadopodia formation. Microtubule connections to the nucleus via nesprin 2 and Lis1 are necessary for proper positioning of MT1-MMP endosomes and invadopodia formation. In summary, nesprin 2-mediated microtubule connections and Lis1 are necessary for breast cancer cell invasion through collagen as it aids in invadopodia formation and creates the mechanical anchorage needed to move the nucleus through 3D collagen.

2.2. Perinuclear actin nucleation drives 3D nuclear movement

Another mechanism by which nuclear movement through 3D spaces is regulated is through perinuclear actin polymerization controlled by actin nucleating proteins (Lammerding and Wolf, 2016). Perinuclear Arp2/3-driven actin polymerization facilitates deformability of the nucleus to move dendritic cells through 3D spaces (Thiam et al., 2016). In leukocytes migrating through constricting microchannels, Arp2/3 mediated perinuclear actin polymerization caused weakening of the nuclear lamina, which allowed the nucleus to be deformed and migrate through the microchannel. Moreover, in C. elegans, Arp2/3 activity promotes UNC-84/Sun associated-microtubule growth which pushes the male pronucleus towards the female pronucleus and promotes cell movement (Xiong et al., 2011). However, LINC-independent actin polymerization can also drive nuclear migration. It was discovered in C. elegans that TOCA-1, which interacts with both Cdc42 and actin-nucleating protein WASP can also move nuclei through actin-dependent mechanisms (Chang et al., 2013). Nuclear movement can also be mediated by the F-actin bundling protein fascin, which is known to interact with nesprin-2, thus mediating F-actin-nuclear interactions (Groen et al., 2015; Jayo et al., 2016). It was found that this coupling of fascin to the nucleus was necessary for transmission of actomyosin contractility-based force to the nucleus and that this controlled nuclear deformability in invasive breast cancer cells through collagen matrices (Jayo et al., 2016).

3. Cytoskeletal crosstalk and 3D nuclear movement

Crosstalk between F-actin, intermediate filaments, and microtubules within the cytoplasm also plays an important role in propelling the nucleus past constrictions in 3D environments (Fig. 2). This crosstalk can occur either through direct interactions between the distinct cytoskeletal filaments or indirectly through crosslinking proteins. For example, vimentin intermediate filaments directly bind F-actin through their tail domain (Esue et al., 2006) and indirectly crosslinked with F-actin via the cytolinker protein plectin (for a comprehensive review, see (Wiche et al., 2015). In this section, we review how unique configurations of direct and indirect cytoskeletal crosstalk can drive nuclear movement through 3D brain tissue, collagen, and CDMs.

Figure 2. Cytoskeletal crosstalk underlies migratory plasticity and 3D nuclear movement.

Mesenchymal cells such as fibroblasts overcome physical barriers to movement when migrating through crosslinked matrices by coordinated crosstalk between vimentin and actomyosin filaments, which pull the nucleus through the matrix fibers (left panel). Neuronal cells in non-crosslinked matrices push their nuclei forward due to actomyosin contractility located behind the nucleus along with microtubule polymerization ahead of the nucleus, which moves the whole cell forward (right panel). Thus, different cell types can utilize unique combinations of cytoskeletal crosstalk to drive 3D nuclear movement through varying 3D environments.

3.1. Vimentin-actin crosstalk drives nuclear movement through 3D crosslinked matrices

While neuronal cells require cytoskeletal crosstalk to move through brain tissue, dermal fibroblasts and breast cancer cells also rely on cytoskeletal crosstalk to drive nuclear movement through crosslinked matrices. Dermal fibroblasts and MDA-MB-231 breast cancer cells utilize a unique method of migration that relies on crosstalk between the vimentin intermediate filament network and the actomyosin network to pull the nucleus through crosslinked matrices, such as CDM (Petrie et al., 2017; Petrie et al., 2014). In this 3D environment, stiffness and crosslinking of the matrix adds an additional physical barrier to cell movement, particularly with regards to nuclear translocation. This is overcome by vimentin filaments wrapping around the nucleus and pulling the nucleus forward through force generated by actomyosin contractility in the anterior compartment of the cell. The resultant high-pressure protrusions are called lobopodia and are distinct from 3D lamellipodia formed by cells in collagenous matrices. It is of note that both the vimentin filaments interacting with the nucleus and the actomyosin force in the anterior compartment of the cell are critical to nuclear movement in these cells. A similar crosstalk between actomyosin filaments and vimentin filaments drives migration of mouse embryonic fibroblasts (mEFs) through confined microchannels (Patteson et al., 2019). In this study, the authors found that vimentin filaments play a nucleo-protective role in 3D migration and vim −/− mEFs migrated faster than vim +/+ mEFs through microchannels. When vim−/− mEFs were treated with cytochalasin D or nocodazole, which inhibit actin polymerization and disrupts microtubule assembly respectively, cell migration slowed down. Although the authors did not specifically look at nuclear movement, these findings suggest that robust crosstalk between all three networks are needed for the cells to migrate through the microchannels. Interestingly, in 2D cells actomyosin filaments and perinuclear vimentin interact such that myosin contractility controls vimentin organization and vimentin controls protrusion formation at the leading edge of the cell (Jiu et al., 2015). It is remarkable that the interaction between actomyosin filaments and the perinuclear vimentin network plays distinct roles in 2D and 3D migration and the potential significance of this difference remains to be explored (Costigliola et al., 2017; De Pascalis et al., 2018).

3.2. Microtubule-actin crosstalk drives nucleokinesis during 3D neuronal cell migration

In neuronal cells migrating through 3D collagen matrices, the nucleus is moved forward through the combined action of microtubule organization and actomyosin contractility (Schaar and McConnell, 2005). In these cells, the centrosome is positioned ahead of the nucleus via microtubules which is followed by generation of NMII contractility at the rear of the nucleus which pushes the nucleus forward. Through inhibition of both NMII contractility and disruption of microtubules, it was concluded that both networks were necessary for efficient nuclear movement. Specifically, NMII contractility was needed to push the nucleus forward and microtubule polymerization was needed to move the entire cell forward through protrusion formation. Importantly, movement of the microtubules away from the cell rear causes the increased NMII contractility that is needed for movement of the nucleus. Thus, a delicate cooperation is needed between the microtubule and actomyosin networks to achieve 3D neuronal migration through collagenous matrices. Similarly, it was shown in neuronal cells and glioma cells migrating through mouse brain tissue that actin and myosin contractility at the cell rear was needed to push or translocate the nucleus (Beadle et al., 2008; Martini and Valdeolmillos, 2010). In these cells, it was found that the stabilizing the microtubule network through Taxol treatment abrogated the dynamic changes to the actin network needed to translocate the nucleus. Hence, neuronal and glioma cells demonstrate that microtubule turnover can help to organize the actomyosin filaments driving 3D nuclear translocation.

4. Indirect crosstalk via crosslinking proteins

The CP, Arp2/3, myosin-I linker (CARMIL) family of proteins bind capping protein found at the barbed ends of actin filaments and help to control actin nucleation and the interaction of F-actin with other proteins. Recently, it was found that CARMIL2-mediated interactions between vimentin intermediate filaments and F-actin in HT1080 cells forming invadopodia (Lanier et al., 2015). This CARMIL2-mediated crosstalk was necessary for invadopodia formation and invasion into collagen and highlights the importance of the indirect crosstalk between these networks in driving 3D migration. However, this study measured nuclear and cellular displacement of CARMIL2-depeleted cells only on 2D surfaces and significance difference in velocity was reported. It will be important to determine whether CARMIL2 depleted cells are slowed down in 3D collagen and thus whether this indirect interaction between the actin and vimentin networks is necessary for 3D nuclear displacement in these cells.

The plakin family of crosslinking proteins are able to crosslink various cytoskeletal networks together. The large cytolinker protein plectin, which binds intermediate filaments to the nucleus, focal adhesions, hemidesmosomes, actin, and microtubules plays a role in both 2D cell migration (Osmanagic-Myers et al., 2006) and invadopodia formation (Sutoh Yoneyama et al., 2014) by facilitating cytoskeletal crosstalk. In 2D keratinocytes, plectin mediates nuclear mechanotransduction through organizing the keratin network perinuclearly (Almeida et al., 2015). The perinuclear arrangement of the vimentin network is controlled by contractile actin transverse arcs, that interact with vimentin in a plectin-specific manner. In response, the vimentin network regulates the width of lamellum formation at the leading edge of migrating cells (Jiu et al., 2015). The plectin-dependent effects on nuclear size were also dependent on actomyosin contractility, indicating that both plectin and actomyosin contractility are necessary for nuclear mechanotransduction. While our current understanding of the potential role of plectin in 3D migration is limited, based on its role in linking vimentin intermediate filaments to the nucleus via nesprin 3 (Ketema et al., 2013; Postel et al., 2011; Wilhelmsen et al., 2005), we speculate that it could help pull the nucleus forward during lobopodial 3D migration. Further, since plectin has been shown to play a role in actin-vimentin crosstalk that drives 2D lamellipodia formation (De Pascalis et al., 2018), more investigation should be done into its role in 3D lamellipodia migration.

Finally, another member of the plakin family of cytolinker proteins, microtubule-actin crosslinking factor 1 (MACF1), is critical for efficient neuronal cell migration in the brain (Ka et al., 2014). It was found that MACF1 plays a role in both stabilizing microtubules and facilitating microtubule-actin crosslinking, both of which are crucial to efficient neuronal cell migration. However, more work is needed to elucidate upon the crosslinking between microtubule and actin networks and whether this MACF1-mediated association is necessary for 3D nuclear translocation specifically.

5. Concluding Remarks

Over the last three decades, our understanding of the mechanisms that drive 3D cell migration have advanced considerably (Klemke et al., 2010; Petrie and Yamada, 2016; Wu et al., 2014b). First, it has been well-established that the nucleus proves to be a hindrance to 3D migration due to its stiff and bulky nature (Davidson et al., 2014; Krause et al., 2019). Second, we have learned that the LINC complex, which comprises of inner nuclear membrane associated SUN proteins and outer nuclear membrane associated nesprins, plays a key role in nuclear translocation by linking the nucleus to the cytoskeleton (Luxton et al., 2010; Wu et al., 2014a; Zhang et al., 2009). For example, nesprin 3-associated vimentin filaments and nesprin 2-associated microtubules can pull the nucleus through crosslinked and non-crosslinked matrices, respectively (Infante et al., 2018; Petrie et al., 2017; Petrie et al., 2014). Third, it has been illustrated that cytoskeletal crosstalk, dynamics, and organization can itself drive nuclear movement. For example, microtubule dynamics and the re-organization of the microtubule network away from the cell rear can promote myosin II activity behind the nucleus to push the nucleus forward (Schaar and McConnell, 2005).

While we are aware of the importance of the LINC complex in driving 3D nuclear movement, we do not yet understand how all SUN and KASH proteins, actin, intermediate filaments, and microtubules interact cooperatively and at once to drive migration through varying 3D substrates. Additionally, it has been well-established that cooperation and crosstalk between cytoskeleton networks is important to nuclear movement, but we have not yet uncovered the importance of crosslinking proteins in driving 3D nuclear movement. While there is some data illustrating crosslinking proteins such as plectin and MACF1 play a role in invadopodia formation and 3D migration at large (Gad et al., 2008; Ka et al., 2014; Sutoh Yoneyama et al., 2014), more work is needed to uncover how their crosslinking of actin, intermediate filaments, and microtubules could facilitate nuclear movement. In addition, the interplay between molecular motors such as NMII, dynein, and kinesin need to be examined more closely. All these motors have individually been found to play a role in nuclear translocation (Infante et al., 2018; Malone et al., 2003; Petrie et al., 2014; Zhang et al., 2009), and one can speculate that there is robust crosstalk between these motors responsible for nuclear positioning and translocation. Lastly, there remains a gap in our knowledge about the role of focal adhesions in 3D nuclear movement.

Since we know that focal adhesions play an important role in 3D mesenchymal migration (Doyle et al., 2012; Doyle et al., 2021; Fraley et al., 2010; Kubow et al., 2013; Petrie et al., 2012), we speculate that there must be contacts between the cell and the surrounding matrix which anchors the cell and aids in dislocating the nucleus. Without such a continuous network between the nucleus and the matrix, the force generated to displace the nucleus could simply result in the collapse of the cell. Though this relatively new field has these and other open questions, it is clear that investigating nucleo- and cytoskeletal crosstalk will be critical to understanding the migratory plasticity that drives nuclear movement through complex 3D environments.