Introduction

Eukaryotic cells share a similar strategy to divide their genetic material at the end of somatic cell division, which is essential to ensure the faithful transmission of genetic information in daughter cells: mitosis. However, one major difference lies in the remodeling of the nuclear envelope (NE)1. Indeed, fungal species, including budding and fission yeasts, have evolved toward closed mitosis, without the complete dismantlement of the NE, although local disassembly of the NE during mitosis has been reported in fission yeast2,3, while some species, such as Drosophila, undergo semi-closed mitosis4. In contrast, vertebrate cells, including mammalian cells undergo fully open mitosis. This implies a NE breakdown (NEBD), to allow microtubule access to the chromosomes in pro-metaphase and metaphase, and the reassembly of the NE after chromosome separation, between anaphase and telophase. Alterations in the tight regulation of NE dynamic during mitosis often lead to mitotic catastrophes, aneuploidy, cell death or senescence5.

The mammalian NE is a complex structure, consisting of two phospholipid bilayers, the outer and inner nuclear membranes (ONM and INM), along with associated transmembrane proteins. All the components of the NE must disassemble upon the entry into mitosis and reassemble upon exit from mitosis. Among these components, the nuclear pore complex (NPC), ensures the nuclear import and export of soluble molecules6,7. In addition, the NE is connected to the cytoskeleton through the transmembrane LINC complex (Linkers of the Nucleoskeleton to the Cytoskeleton Complex), which interacts with the lamina in the inner part of the nucleus8, and with the actin in the cytoskeleton9,10. Finally, LEM-Domain proteins (LAP2-Emerin-MAN1 domain, LEM-D) bridge the NE to chromatin, and interact with A-type lamins11. The LEM-D group gathers proteins that share the ability to bind both DNA and BAF (Barrier-to-Autointegration Factor), a DNA and histone-binding protein, that also interacts with lamin A12,13.

Beneath the INM, the lamina forms a fibrous meshwork of type V intermediate filaments, the lamins, and provides a strong mechanical support for the nuclear integrity14. Besides their central role in maintaining nuclear integrity, lamins have several other functions including chromatin organization, DNA repair, DNA replication, transcription, telomere maintenance and mitosis14,15,16,17. Lamins are subdivided into A and B- types, which form two separate but interconnected fibrous networks18,19. A-type lamins include lamin isoforms A, C, AΔ10 and C2. These are splicing variants encoded by the LMNA gene20. B-type lamins, on the other hands, are composed of lamin B1 (encoded by the LMNB1 gene21) and lamins B2 and B3 (encoded by the LMNB2 gene)22. All lamins contain a N-terminal head domain, followed by three coil domains that form the central rod domain, next to an Ig-like domain, and a C-terminal tail domain, which is the most variable region of lamins23. The head, rod and tail domains are important for lamin assembly, while the Ig-like domain is responsible for multiple protein-protein interactions23. A key difference between the lamin types is their maturation process. During this process, lamins A, B1 and B2 undergo farnesylation, a post-translational modification involving the addition of a farnesyl moiety to the C-terminal cysteine residue in the CAAX motif, which facilitates the anchoring of proteins to the membrane24. Of note, lamin C, which lacks the CAAX box in its C-terminal domain, is not farnesylated. While mature lamin B1 and B2 remain farnesylated, allowing their direct interaction with the INM23, pre-lamin A is cleaved by the endopeptidase Zmpste24, which removes the farnesyl group and contributes to making lamin A more soluble than lamin B125. In addition, lamins interact with different transmembrane proteins of the INM, including the LEM proteins for A-type lamins and lamin B receptor for B-type lamins. Some of these interactions are thought to contribute to the anchoring of the nuclear lamina to the NE26,27. However, a pool of A and B- type lamins is found to be nucleosoluble, and dispersed in the nucleoplasm28,29. Besides being involved in common cellular processes, A- and B-type lamins display distinct expression patterns during embryogenesis and cell differentiation, and ensure non-redundant functions15,16,23.

In this review, we focus on lamin functions during mitosis, especially their localization, assembly state, and regulation throughout mitosis, from NEBD to NE reformation, and outline their contributions to mitotic processing. We also discuss the links between lamina dysregulation and mitotic failure in connection with human pathologies, from rare genetic disorders to cancer, and highlighting the importance of lamin proteins in mitosis and further chromosome stability.

The NE breakdown triggers lamina disassembly

Open mitosis implies the dismantlement of the NE at mitotic entry to allow microtubules to access chromosomes. NEBD is initiated by two events, described as simultaneous: NPC disassembly, and spindle microtubule-driven forces that trigger a mechanical tearing of the NE. The disassembly of the NPC is driven by the phosphorylation of several of its main components, the nucleoporins (NUPs), which is mediated by several mitotic kinases, including Polo-like kinase 1 (PLK1), Cyclin-dependent kinase 1 (CDK1), NIMA-related kinases (NEKs) and Aurora-A kinase. In parallel, spindle microtubules polymerize from the MTOC (microtubule-organizing center) in the cytoplasm towards the NE, and exert forces on the nucleus30,31. These mechanical forces result in invaginations and stretching of the NE, ultimately leading to rupture of the NE at the opposite site from the invaginations30,31,32 (Fig. 1). This initial rupture event accelerates the chromosome condensation and induces synchronous phosphorylation of A- and B-type lamins. This results in lamina disassembly owing to the depolymerization of lamin filaments30,31. Meanwhile, both the INM and ONM, along with some NE proteins - including LEM proteins and some nucleoporins -, are redistributed into the ER membrane33,34,35. It is worth noting that although lamins are in a monomeric state after NEBD, they are not degraded, and their contribution to the mitotic process is described in more detailed below.

Fig. 1: Disassembly and localization of lamins during the nuclear envelope breakdown.
figure 1

Before mitotic entry (from G1 to late G2), lamins are mostly organized into separated A-type (purple) and B-type (red) filament networks that lie beneath the inner nuclear membrane, and form the so-called lamina (although a fraction of lamins is also present as a soluble nucleoplasmic pool). During the nuclear envelope breakdown (NEBD), lamina dismantlement relies on: 1) microtubules (orange lines) driven forces (black arrows) acting on the nuclear envelope (NE), creating stretching forces (black arrows) at the distal site of microtubule polymerization. This ultimately results in the rupture of the NE under these forces; 2) the phosphorylations of A- and B-type lamins by the kinases CDK1 and PKC (and Src for lamin A), on both sides of their rod domain, abolish their assembly into nucleofilaments. The main mitotic phosphorylation events are indicated in the boxes, and summarized in Table 1. Once the NE is disassembled, lamins localize homogeneously in the nucleoplasm, with a higher concentration of B-type lamin close to the centrosomes, at the mitotic spindle pole. Figure created with BioRender.com.

Full size image

The phosphorylation of lamins as a cause for their dismantlement during mitosis was proposed in the early 80 s’ by L. Gerace36,37. These reports showed that during mitosis, lamin A exists in a soluble and diffuse state. This is in contrast to lamin B1, which remains more insoluble and is associated with membrane fragments36,38. Later on, hyperphosphorylation of both A and B-type lamins during mitosis by CDK1/cdc2 was demonstrated39,40,41. CDK1 phosphorylates residues Ser23 and Ser393 in lamin B142 (residue positions in UniProtKB P20700), and residues Thr34, Ser37 and Ser405 in lamin B243,44 (residue positions in UniProtKB Q03252). CDK1 also phosphorylates lamin A at Thr19, Ser22, Ser390 and Ser392 residues45,46,47. Furthermore, Protein kinase C (PKC) has been reported to phosphorylate lamin B248, as well as lamin B1 at residues Ser395 and Ser405, respectively, during mitosis42,49. Consequently, a delay of lamin disassembly can be observed after PKC inhibition50,51. Lamin A/C also interacts with PKC-α52, and murine lamin A/C is a substrate for PKC at Thr416, Thr480, Thr199 and Ser625 during both mitosis and interphase53. Interestingly, PKC also phosphorylates murine lamin A at Ser5 and Ser525 only during interphase53, and it has recently been reported to contribute to lamin A mobility outside of mitosis54 (Table 1).

Table 1 Phosphorylation of Lamins During Mitosis Drives Their Depolymerization
Full size table

Interestingly, the conservation of phosphorylated regions across different species and lamin types, on both sides of the central rod domain, highlights the mechanism behind their dissociation. Indeed, the phosphorylation of the C- and N-terminal regions of lamins by CDK1 and PKC destabilizes the head-to-tail interaction between the coil 1a domain and the C-terminal part of the coil 2 domain of lamins. This abolishes the interactions between lamin dimers, ultimately leading to a transition from assembled type V intermediate filaments to monomeric lamins40,55.

Finally, inhibiting both CDK1 and PKC is insufficient to completely block lamin B1 disassembly. While this partial inhibition may be due to insufficient inhibitor efficiency, it also suggests that other kinases or other post-translational modifications, such as dephosphorylation or deacetylation56, may contribute to lamina disassembly during NEBD50. However, these remain to be discovered. Some factors recently involved in lamina stability could be involved. For example, Src (non-receptor protein tyrosine kinase) has recently been shown to phosphorylate lamin A at Tyr45. This phosphorylation triggers lamin A disassembly and appears to occur preferentially during mitosis, although it is not known whether Src also phosphorylates lamin B157. Altogether, these findings suggest that lamina disassembly relies on the synchronization of multiple kinases to ensure proper NEBD. Recently, the PP2A phosphatase regulatory subunit B55SUR-6 has been involved in the NEBD in C. elegans, possibly by targeting the B-type lamin LMN-158. Whether phosphatases, such as PP2A, are also involved in lamina disassembly in mammalian cells remains to be addressed.

After complete NEBD, lamin B1 partially localizes at the centrosome until early metaphase30 and remains associated with the endoplasmic reticulum membrane32,59. In contrast, lamin A is more diffusely and homogeneously distributed in the cytoplasm36,60. These different localization patterns between A and B-type lamins suggest that they could ensure different functions after NEBD, during the next stages of mitosis.

Consequences of altered dismantlement of the lamin networks

Preventing lamina depolymerization can compromise mitosis progression. Indeed, preventing lamin A disassembly by expressing a non-phosphorylatable mutant (S22A, S392I and deletions of adjacent prolines) in CHO (Chinese hamster ovary) cells results in lamin A aggregation in prophase and blockage in metaphase45. These findings raise questions about the fate of the cells that are unable to process normal NEBD. It remains to be determined whether these cells would strictly undergo apoptosis after the mitotic checkpoint, or whether they could exhibit an intermediate phenotype that could lead to an altered cytokinesis and potentially result in aneuploidy or polyploidy, as it has been previously described following prolonged mitotic arrest61,62. Interestingly, Heald and colleagues have shown that a single mutation at residues Ser22 or Ser392 does not induce mitotic blockage as frequently as observed for the double mutant form of lamin A45, suggesting a non-redundant role of residues Ser22 and Ser392 for lamin A disassembly. On the other hand, the inhibition of both CDK1 and PKC has been reported to delay, but not arrest mitotic progression, despite lamin B1 aggregation50. These studies used either overexpression of human mutated lamin A in mouse model45, or partial inhibition of lamin phosphorylations50. In addition, LMNB2 knockout cardiomyocytes harbor an important NE disassembly defect in pro-metaphase. This suggests that, although lamins must be disassembled prior to metaphase, their presence may be necessary to process a complete and correct NEBD63.

Thus, the precise common and differential contributions of each lamin type to the mitotic process, as well as the potential cooperation between A- and B-type lamins during NEBD, remain to be fully elucidated. To this matter, it would be of great interest to express non phosphorylatable forms of lamin B1, both individually and in combination with a non-phosphorylatable form of lamin A in human cells.

Roles and localization of lamins from prometaphase to anaphase

Although the NE disassembly is required for mitosis to proceed, the NE components, especially lamins, are not degraded and remain detectable in close vicinity to the mitotic chromosomes (Fig. 2). Indeed, lamin B1 was first reported to be homogeneously localized in the nucleoplasm during metaphase in rodent models30,64,65. The localization and function of B-type lamins after prometaphase have been further characterized in human cells. In HeLa cells, lamin B1 and B2 co-localize with the mitotic spindle during metaphase, along the microtubules attached to the kinetochores66. Interestingly, depletion of either lamin B1 or B2 in human cells induces significant spindle asymmetry and defects in spindle formation, coupled with a prolonged prometaphase-metaphase duration. This phenotype is more severe upon lamin B2 depletion63,66. This suggests that B-type lamins are involved in spindle assembly, although the precise role of lamin B2 during pro-metaphase remains to be fully characterized. Furthermore, in Xenopus eggs arrested in M-phase, RanGTP can stimulate the assembly of lamin B3 (major lamin B isoform in Xenopus eggs) into a matrix-like network, by abolishing its interaction with importin βα (responsible for the nuclear import of Nuclear Localization Signal (NLS)-carrying proteins). This model is consistent with previous study from Adam and colleagues, which showed that importin α maintains Xenopus lamin B3 in a soluble and depolymerized state by interacting with this latter67. In addition to this mechanism, the formation of the Xenopus lamin B3 matrix also depends on cytoplasmic dynein, a microtubule-based cytoskeletal motor, and on its major regulator Nudel, both of which contribute to spindle organization68,69,70. Therefore, it has been proposed that B-type lamins form the spindle-associated matrix, which is involved in the gathering of all the components necessary for spindle assembly during mitosis. In support of this model, the depletion of lamin B2 in colorectal carcinoma cell lines (HCT116 and RKO) induces defects in spindle formation and in chromosome condensation71. In addition, lamin B2 deficiency also leads to aneuploidy71, which is a common consequence of spindle formation defects and subsequent chromosome missegregation72. In agreement with these previous observations, LMNB1 knockout in mice and LMNB2 inactivation in cardiomyocytes from perinatal mice, have been shown to induce polyploidization, alterations in metaphase progression and spindle attachment defects63,73. However, in these contexts, it remains unclear whether polyploidy results from spindle misregulation due to B-type lamin deficiency. Altogether, these studies highlight a clear contribution of B-type lamin homeostasis in ensuring proper spindle formation, chromosome segregation and maintenance of ploidy.

Fig. 2: Localizations of lamins during metaphase.
figure 2

During metaphase, B-type lamins are able to form a matrix-like network in a RanGTP- and Dynein-dependent manner. In addition, LAP2α influences the localization of B-type lamins towards the spindle. A fraction of B-type lamins is also detected diffusely in the nucleoplasm. The localization of A-type lamin during metaphase is less clear. While one study shows that lamin A also localizes at the spindle in a dynein-dependent manner and is associated with BAF and LAP2α, other studies have shown that A-type lamins are soluble and homogeneously distributed in the nucleoplasm during metaphase36,60,76. Figure created with BioRender.com.

Full size image

Regarding lamin A, most studies investigating its localization between the NEBD and mitotic exit in both human and mouse cells have reported a diffuse pattern of lamin A, described as soluble in the cytoplasm64,74,75,76. This soluble state of lamin A has recently been reported to depend on its interaction with LAP2α, which maintains a pool of lamin A in an unassembled state during both interphase and mitosis76. Importantly, LAP2α is also found in a ternary complex with BAF1 and lamin A, localized to the spindle during mitosis in HeLa cells77. Qi and colleagues demonstrated that this complex associates with the spindle through dynein, and influences spindle assembly and positioning77. In addition, the lamin A - LAP2α - BAF1 complex could influence the localization of lamin B1 to the spindle77, suggesting a potential cooperation between lamin B1 and lamin A for proper spindle assembly in metaphase77.

Although the localization of lamin A before anaphase remains uncertain, the organization of both lamin B1 and lamin A during metaphase appears to be regulated by dynein. Taken together, these findings suggest a model in which dynein facilitates the assembly of both lamin B and lamin A into the spindle matrix which may then help for the proper assembly of mitotic spindle, and subsequent accurate chromosome segregation.

It is noteworthy that the NE already contributes to centrosome regulation during prophase. Indeed, centrosomes, which are responsible for mitotic spindle assembly at the onset of mitosis, are separated along the NE in prophase due to the action of motor proteins, such as kinesin-5 and dynein78. Centrosome positioning is essential for the efficient mitotic process. In mammals, the LINC complex contributes to establish a connection between centrosome and the NE79. The LINC complex is also required for the positioning of the centrosome during prophase80. Additionally, it has been reported that NPCs interact with dynein in late G2/prophase, contributing to its recruitment at the NE and further centrosome positioning81. Finally, lamins are also involved in regulating the dynein-based pulling forces and even NPC distribution required for centrosome separation during prophase82. Indeed, compromising the nuclear lamina leads to asymmetrical NPC distribution, caused by unbalanced dynein forces. This results in defects in centrosome separation and NPC positioning in different species, including mouse82 and C. elegans83.

NE reformation during mitosis exit

Open mitosis ends with cytokinesis and the formation of two daughter cells. This implies an urgent need for NE reformation in order to restore the compartmentalization between the nucleus and the cytoplasm of the two daughter cells. While one of the main drivers of NEBD is phosphorylation events, NE reassembly occurs through the dephosphorylation of many targeted proteins during NEBD.

From NEBD to anaphase, the NE membranes are contained within the ER1. At the end of mitosis, the chromatin-binding protein BAF, LEM proteins, LBR84 and the endosomal sorting complex required for transport (ESCRT)-III mediate the recruitment and fusion of the NE membranes to the chromosome surface85,86,87. Furthermore, BAF also promotes the targeting of LEM proteins and lamin A/C to the decondensing chromatin88. Simultaneously, the reassembly of NPCs is initiated by the recruitment of the Nup107-160 complex to chromatin89. The reforming NE membranes jointly enclose these assembling NPCs (for reviews, see7,90,91). Lamins have been reported to associate with the newly formed NE after the recruitment of NPCs and transmembrane proteins, and to shuttle through the NPCs into the nucleus92,93. Notably, some components of the NPCs (i.e. Nup153 and the nucleoporin ELYS) have been reported to be involved in targeting B-type lamins to the INM during NE reformation94,95.

Lamin dephosphorylation and incorporation in the newly formed NE

Like other NE components, lamins reassemble after dephosphorylation, a mechanism that is mainly ensured by the protein phosphatase 1 (PP1)96. However, the activity of PP1 on lamin A and lamin B1 depends on different partners. The interaction of PP1 with its regulatory subunit, A-kinase anchoring protein (AKAP149), an integral protein of the ER, allows PP1 to dephosphorylate lamin B197,98, while its interaction with Repo-Man, another PP1 regulatory subunit, triggers lamin A dephosphorylation at Ser2299,100. These dephosphorylations induce re-polymerization of A- and B-type lamins during anaphase97,99. PP2A appears to be involved in lamin reassembly, since inhibiting its activity decreases lamin dephosphorylation. However, whether PP2A contributes directly to this process by dephosphorylating lamins or by targeting proteins important for this process remains unknown101. Moreover, PKC-mediated phosphorylation of both AKAP149102 and Repo-Man103,104 impairs their interaction with PP1, thereby preventing lamin polymerization and premature lamina reassembly prior to mitotic exit.

Notably, compared with lamin phosphorylation during NEBD, which has been extensively studied and described, lamin dephosphorylation and the associated phosphatases are not well understood. Thus, whether lamin dephosphorylation is fully ensured by PP1, or whether phosphatases like PP2A, which has been proposed to participate, are directly responsible for lamin dephosphorylation, remains elusive.

Since dephosphorylation of the lamins drives their assembly state, these events also lead to changes in their localizations. Indeed, lamin B1 and B2 are the first, before lamin A, to reassemble and localize to the chromosome surface, after PP1 dephosphorylation, as early as anaphase, both in mouse cell lines expressing human lamins and in rat models64,65,74. In particular, lamins B2 and B1, together with LBR and LAP2β, have been shown to concentrate at the chromosome periphery, in late anaphase, before being detected all around the chromosome surface in telophase84,105,106 (Fig. 3). In contrast, Haraguchi and colleagues showed that lamin A accumulates at the core region of the chromosomes (the central region of the chromosomes near to the spindle attachment sites) in association with BAF, emerin and LAP2α, between late anaphase and early telophase, in a BAF-dependent manner. This core structure is important for the reconstitution of a normal NE in interphase106,107,108. However, this localization is transient, and lamin A is then incorporated in the NE only in late telophase/early G1, while a significant fraction of the protein remains unassociated with the nuclear periphery64,65,74. These observations have been further characterized in several other studies performed in mouse60 and in human cell lines75,76,99, in which lamin A has been shown to be incorporated into the NE during telophase. These different localizations could contribute to chromatin reorganization after mitosis. During NE reassembly, lamins contribute to the organization of chromosome territories. Indeed, an altered state of lamin assembly can lead to abnormal chromosome positioning. Defects in lamin B1109, or loss of lamin A/C and emerin110 disrupt normal chromosome positioning. Unlike lamin A, lamin C is not detected in the NE before the G1 phase and remains nucleoplasmic during telophase and early G1. The late reassociation of lamin C with the NE has also been proposed to be important for chromatin reorganization after mitosis111.

Fig. 3: Regulation of lamin network reformation during mitosis exit.
figure 3

After chromosome separation occurring in anaphase, lamins reassociate with the NE following reassembly of the membranes, NPC and other transmembrane proteins. The phosphatase PP1 dephosphorylates both B- and A-type lamins. For B-type lamins, this event depends on the interaction between PP1 and its regulatory subunit Repo-Man (red exclamation mark, see magnification of the left). For A-type lamins, it depends on the binding of PP1 and the kinase anchor protein AKAP149 (purple exclamation mark, see magnification on the right). During late anaphase (upper box), lamin B1 and B2 are first detected at the periphery of chromosomes. At this stage of mitotic exit, lamin A transiently localizes to the core region of chromosome, in association with BAF, LAP2α and emerin. Later on, during telophase (middle box), all detectable lamin B1 and B2 are associated with the NE. However, lamin A is not incorporated into the NE at this time, despite being dephosphorylated. The interactions between LAP2α and BAF with lamin A regulate its restricted assembly state. Finally, in early G1 (lower box), lamin A is found associated with the NE, as well as in a nucleosoluble form. Figure created with BioRender.com.

Full size image

Importantly, LEM4/ANKL2 has been shown to be involved in post-mitotic reassembly of NE proteins, including lamins, by promoting PP2A-dependent BAF dephosphorylation, a function conserved in C.elegans, Drosophila and Human112,113,114. In human cells, LEM4/ANKLE-2 is required for lamin A reassociation with chromosomes113. LEM4/ANKEL-2 also promotes the association of BAF and LAP2α with chromatin113, allowing the proper regulation of lamin A reassembly at the NE. Indeed, Naetar and colleagues have shown that LAP2α decreases the ability of lamin A/C to assemble into filament, and favors its localization to nuclear interior76. Furthermore, emerin and BAF are also involved in controlling lamin A incorporation into the NE by preventing its aggregation at the nuclear periphery27. These mechanisms could explain the differential localization of B- and A-type lamins toward the NE at the end of mitosis, despite their similar timing of dephosphorylation. Furthermore, these regulations shed light on mechanisms controlling lamin assembly other than phosphorylation.

Consequences of lamina reassembly defects

Impairment of lamina reformation after mitosis has been reported to contribute to mitotic defects and could have dramatic effects on genomic stability. In particular, inhibition of lamin B1 reassembly by preventing its dephosphorylation in HeLa cells induces rapid proteolysis of lamin B1, and apoptosis within the next 6 h98 despite that nuclear assembly of lamin A/C is not affected. These data suggest that lamin B1 network play important role in cell survival in cancer cells98. However, it is noteworthy, that in other cellular models, lamin B1 deficiency did not induce apoptosis, such as in keratinocytes, hepatocytes115 or mouse fibroblasts, although nuclear shape abnormalities, altered proliferation capacity and chromosomal instability (CIN) have been observed73.

In contrast to lamin B1 deficiency, impairment of lamin A reassembly does not induce cell death. However, LEM4/AKEL-2 deficiency impairs the global NE reformation by affecting LAP2α, BAF and lamin A, and induces major defects in NE integrity as well as hyperploidy in HeLa cells. Notably, delaying lamin A dephosphorylation by depleting Repo-Man does not result in striking defects in cell viability after mitosis99, although the consequences for genome organization and integrity toward the next cell cycle could be deleterious, given the important roles of lamin A in genome organization, replication, transcription and repair116. In addition, HeLa cells expressing the phosphomimetic Y45D mutant of lamin A, which alters lamin A assembly capacity, exhibit nuclear dysmorphia, micronuclei and altered DNA repair response57.

In the light of these studies, the reassociation of lamin networks appears to be critical for proper mitotic exit and subsequent cell viability. This is further illustrated by the loss of the PP1-AKAP149 interaction in G1 nuclei, which triggers depolymerization of lamin A and B filaments, leading to G1 arrest and apoptosis117.

Lamin-related mitotic defects in disease

Pathologies resulting from lamin dysregulation and/or mutations in genes encoding lamins are called laminopathies. Lamin A mutations most commonly cause lipodystrophies, muscular dystrophies, cardiomyopathies, and progeroid disorders118, whereas genetic alterations affecting lamins B1 and B2 are less common and are responsible for microcephaly (LMNB1 mutations), autosomal dominant leukodystrophy (ADLD, LMNB1 duplication) and lipodystrophy (LMNB2 mutation)119. In addition, dysregulations of both A and B-type lamins have been reported in many cancers and have been correlated with poor prognosis in several tumor types119, although the underlying molecular mechanisms are not well understood.

Most of the mechanisms known to contribute to laminopathies to date mainly involve the role of lamins in chromatin organization, transcription, or DNA repair118. However, the complexity and diversity of laminopathy-associated mechanisms are still not completely understood. In the next chapter, we will summarize the mitotic defects that have been described in association with laminopathies, as well as in cancers with dysregulated lamin expression level, and discuss to what extent they could potentially contribute to the pathogenesis of these diseases, in addition to already known underlying mechanisms.

Mitotic defects in laminopathies

One of the major laminopathies is the Hutchinson-Gilford Progeria syndrome (HGPS), a rare genetic disorder that causes premature ageing. HGPS is caused by mutations in the LMNA gene, that result in the expression of a truncated form of lamin A called progerin. Unlike lamin A, progerin is not matured and remains farnesylated, resulting in a greater anchoring to the NE120.

Interestingly, progerin expression induces important mitotic defects. Indeed, the ectopic expression of progerin in HeLa cells has been shown to delay cytokinesis and NE reformation after mitosis121,122. In addition, Eisch and colleagues have shown that HGPS fibroblasts (expressing progerin) are prone to similar mitotic defects123. These cells also harbor a mislocalization of CENP-F, a centromeric protein crucial for kinetochore attachment in mitosis, which is no longer associated with the kinetochore during metaphase, resulting in lagging chromosomes. Furthermore, Eisch et al. proposed that progerin delays NE reformation after mitosis by altering the recruitment of SUN1 (a protein from the LINC complex) and emerin123. These reports suggest that, in addition to the accumulation of DNA damage - a major contributor of the HGPS symptoms124, progerin-induced mitotic defects could also participate to predispose HGPS cells to genetic instability and premature senescence.

Besides the LMNA mutations responsible for HGPS, a subset of mutations can lead to cardiomyopathy. Two of these, S22L and Y45C, have been associated with cardiomyopathy in patients125,126. These mutations affect key residues that are phosphorylated during the NEBD and dephosphorylated at the end of mitosis. Therefore, the Y45C and S22L mutations may convert lamin A into a non-phosphorylatable form, which could have deleterious effects during mitosis. Although no molecular data are available for patients carrying these mutations, the expression of a non-phosphorylatable form of lamin A at Ser22 in mouse cells results in incomplete disassembly of lamin A. On the other hand, delayed dephosphorylation of the Ser22 residue of lamin A in HeLa cells has not yet been reported to affect mitotic exit. However, the non-dephosphorylation of lamin B1 has been shown to do so99,117. Further studies are therefore required to characterize the precise consequences of these mutations targeting phosphorylation sites.

In contrast to lamin A, there is no direct evidence that dysregulation of B-type lamins can lead to mitotic defects in a pathological context. Mutations in mitotic proteins, mainly centrosomal and mitotic spindle factors, are often involved in primary microcephaly127, resulting in alterations of mitotic spindle orientation and mitotic progression, highlighting the potential involvement of mitotic defects in the pathogenesis of microcephaly. Among B-type lamin-related pathologies, mutations in LMNB1 and LMNB2 genes have been reported to result in primary microcephaly128,129. Given the role of B-type lamins in mitotic spindle assembly and orientation66, it was proposed that these mutations could lead to mitotic spindle defects, thereby suggesting a possible mechanism contributing to the phenotype of microcephaly. However, despite the abnormal nuclear shapes resulting from the expression of lamin B1 mutants, no abnormal spindle formation was observed, nor ploidy alterations in cells from patients with LMNB1 mutations causing microcephaly. Although the contribution of mitotic failure has not been studied for lamin B2-related, others mechanisms such as the impact of nuclear integrity on neural migration and/or expression of genes involved in neurogenesis could underly type-B lamin deficiency-associated microcephaly. However, lamin B1 knockout in mice, which causes perinatal death, results in the induction of aneuploidy73,130. Thus, we cannot exclude that a defect in spindle formation may also contribute to the aneuploidy observed in Lmnb1-/- mouse fibroblasts, given the role of lamin B1 in mitotic spindle formation73,130. Nonetheless, Coffinier et al. have shown that fibroblasts from Lmnb2-/- mice are able to maintain a normal ploidy, despite the role of lamin B2 in spindle assembly and chromosome segregation131. Although we could not exclude that lamin B1 may compensate for the lamin B2 defect in this case, the contribution of mitotic defects in the pathogenesis of B-type associated laminopathies still remains elusive.

Mitotic failures in cancers with lamin dysregulation

Chromosomal instability and aneuploidy are common features of cancer genomes132, and expression level of lamins are commonly found misregulated in various tumor types. Lamin B1 is either overexpressed or downregulated in different cancer types119,133. On the other hand, lamin A expression is mostly downregulated in tumors134, while lamin B2 is overexpressed135.

Given the various contributions of lamins to mitotic regulation discussed above, it is tempting to propose that both up- and down-regulation of A- and B-type lamins could induce aneuploidy in tumor contexts. However, reports demonstrating lamin-induced mitotic defects in a tumor context are limited. For instance, one study has shown that lamin B1 expression is downregulated in colon cancer136, and that the association of 5-fluorouracil treatment (5-FU, one of the most common anti-metabolic drug used in cancer therapy) with lamin B1 overexpression in colon cancer cells, induces polyploidy137. The mechanism leading to polyploidy in this context is elusive, and could involve different roles of lamin B1. Indeed, in addition to its role in spindle assembly66, lamin B1 is involved in the regulation of DNA damage repair pathways16, and both DNA damage accumulation and mitotic defects can result in polyploidy138. Conversely, the loss of lamin A in ovarian cancer is associated with aneuploidy139,140, which has been proposed to be a consequence of mitotic defects resulting from lamin A deficiency, although no data yet support this model. Regarding lamin B2, its overexpression promotes tumor progression, but no study has yet demonstrated that this effect is due to mitotic defects in this context135.

Concluding remarks

Because the lamina network, a key subcompartment of the NE, is disassembled at the onset of mitosis, the functions of lamina components outside of NE assembly and disassembly have long been underexplored.

In recent years, significant progress has been made in identifying important contributions of the lamina to the regulation of several aspects of mitosis that are important for proper chromosome segregation. Indeed, although the lamina is mostly disassembled during chromosome alignment and separation, its components – the lamins- contribute strongly to mitotic regulation and spindle assembly. These studies have shown that lamins are highly dynamic throughout mitosis, with two main types of lamins—A and B—exhibiting several differences in their assembly and disassembly pathways, especially at mitotic exit in terms of dynamics, timing and localization patterns. These dynamics are mediated by numerous partners, including kinases, phosphatases, interacting partners from the NE and DNA-binding proteins, which differ between lamin types. This highlights the non-redundant functions of lamins during mitosis. Moreover, the importance of lamins for normal mitosis progression is supported by the various studies reporting frequent mitotic defects in pathological contexts associated with lamin mutations or dysregulations. Taken together, these studies demonstre contributions of lamins during mitosis, particularly in spindle assembly and stability. This highlights their importance in theses processess, beyond their function as intermediate filaments providing mechanical support for the NE and their involvement in various nuclear processes. This sheds light on the poorly explored matrix-like aspect of lamin networks between the NE and chromosome separation.

However, despite these recent advances in understanding the involvement of lamins in mitotic regulation, many questions remain unanswered. First, for instance, what mechanism controls the amount of lamin A that must be associated with the NE or dispersed in the nucleoplasm, and why does lamin A only localize to the NE after mitotic exit, whereas B-type lamins are associated with the NE as early as anaphase? Secondly, are there other players involved in lamina assembly and disassembly during mitosis: other kinases and phosphatases? Or are there other types of modifications? Thirdly, could lamins contribute to other aspect of mitosis, besides spindle formation and NE dynamics? Finally, what are the underlying mechanisms that link lamin misregulation and mitotic failure in pathological conditions?

Finally, the implications of lamins during mitosis should be considered with those of other NE components, such as the LINC complex, which controls centrosome positioning80, the NPC, which participates in kinetochore functions141. Deciphering the cross talks between lamins and both centrosome and kinetochore components could provide some mechanistic insights into how lamina defects can affect these processes, thereby perturbing mitosis regulation and potential disease setting. As NE components collaborate during interphase to ensure nuclear integrity, mechanotransduction, transcription regulation and chromatin organization8,142,143,144, it is likely that they also cooperate to facilitate mitosis. The concept of a global system forming a mitotic scaffold has for long been debated, and future studies should improve our understanding of this complex mechanism.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.