Nuclear intermediate filaments formed by A- and B-type lamins are central components of the nucleoskeleton and are required for the architecture and integrity of the nucleus. There is growing evidence that lamins are also involved in regulatory pathways controlling cell proliferation and differentiation. Lamins affect the activity of several transcription factors, such as retinoblastoma protein and c-Fos, and signalling pathways, such as the ERK1/2 (extracellular-signal-regulated kinase 1/2) and Notch pathways, which are key regulators of cell-cycle progression and differentiation. During mitosis, lamins are dynamically reorganized and play active roles in spindle matrix formation and in post-mitotic nuclear reassembly. Several of the cell-cycle-regulating functions of lamins may be impaired in the diseases linked to mutations in lamins and lamin-associated proteins, including striated muscle diseases, lipodystrophies and premature aging syndromes, and contribute to the tissue-specific disease pathologies.
Lamins are nuclear IF (intermediate filament) proteins that, like all IF proteins, contain a central α-helical rod domain flanked by N- and C-terminal globular domains [1,2]. In vitro, lamin homodimers associate head-to-tail to form protofilaments, which assemble into 10 nm filaments and paracrystals . In vivo, lamins form a network-type scaffold at the NE (nuclear envelope) of metazoans, the nuclear lamina, in close association with the inner nuclear membrane and chromatin, and mechanically enforce nuclear morphology. The mammalian genome contains three lamin genes that encode seven lamin variants. B-type lamins, encoded by LMNB1 and LMNB2, are essential, and at least one B-type lamin is expressed in all cells throughout development. In contrast, the major A-type lamins, A and C, encoded by the LMNA gene, are expressed at later stages of development and in most differentiated cells. Unlike lamin C, lamin A and all B-type lamins are translated as prelamins and undergo a sequence of processing steps at the C-terminal CaaX (where a is an aliphatic residue) motif . These include farnesylation of the cysteine residue, endopeptidase cleavage of the C-terminal aaX tripeptide, and carboxymethylation. Whereas B-type lamins remain farnesylated, lamin A is cleaved further by the zinc metalloprotease Zmpste24, removing another 15 C-terminal residues with the farnesyl moiety. As a consequence, B-type lamins are tightly attached to the nuclear membrane even in mitosis, when the NE is disassembled, whereas mature lamin A and lamin C can also localize in lamina-independent complexes throughout the nucleoplasm [5,6]. The peripheral lamins in the lamina interact with numerous integral proteins of the inner nuclear membrane , chromatin components and proteins of nuclear pore complexes, thereby creating a tightly linked network that serves a multitude of structural functions, including maintenance of nuclear architecture, nuclear anchorage and migration, positioning of nuclear pore complexes and higher-order chromatin organization. There is growing evidence that lamins also regulate several transcription factors and signalling molecules involved in the control of cell-cycle progression and differentiation [1,2]. The physiological relevance and diversity of lamin functions is also reflected by the growing number of human diseases linked to genes encoding lamins or lamin-binding proteins . These diseases, known collectively as ‘laminopathies’ include more than ten overlapping or distinct syndromes, including striated muscle disorders, lipodystrophies, metabolic diseases and premature aging syndromes. The molecular disease mechanisms are poorly understood, but may be caused by a combination of effects of disease-causing mutations on structural and regulatory functions of lamins . In the present paper, we summarize and discuss the potential mechanisms of the active roles and dynamics of lamins through the cell cycle, including pathways affecting cell-cycle progression and cell differentiation (Figure 1).
Functions and modifications of lamins during the cell cycle
Lamins control cell-cycle entry and progression
Regulation of Rb (retinoblastoma protein)
Transition of proliferating cells from G1- to S-phase marks an irreversible event committing cells to another round of cell division. The tumour suppressor Rb is a major transcriptional regulator of this process. In its active (hypophosphorylated) state, Rb binds to the transactivation domain of E2F transcription factors and represses E2F target genes required for G1–S-phase transition, thereby arresting cells in the cell cycle and/or promoting transition to quiescence, senescence or differentiation . In a proliferative environment, Rb is sequentially phosphorylated by several cyclin–Cdk (cyclin-dependent kinase) complexes starting in mid-G1-phase and becoming hyperphosphorylated at the G1–S phase transition. Rb hyperphosphorylation releases and activates E2F transcription factors, thereby promoting progression into S-phase. Dephosphorylation of Rb occurs at the end of M-phase and is probably mediated by PP (protein phosphatase) 1 or PP2A . There is growing evidence that A-type lamins are involved in the regulation of Rb, but the molecular mechanisms are not fully understood. Lamin A/C and one of their binding partners in the nucleoplasm, LAP (lamina-associated polypeptide) 2α bind directly to Rb [12,13]. Lamin A and LAP2α have been shown to bind to E2F-dependent promoter sequences and repress E2F-dependent transcription in an Rb-dependent manner . In line with a role of lamin A–LAP2α complexes in promoting Rb repressor activity, overexpression of LAP2α in cultured cells favoured cell-cycle exit . Cells lacking either lamin A or LAP2α show abnormal regulation of G1–S transition, although differences have been observed between mice and humans. Whereas loss of lamin A or LAP2α in primary human fibroblasts induced cell-cycle arrest , loss of LAP2α  or lamin A/C [16–18] in mice favoured cell-cycle progression and impaired cell-cycle exit at the G1–S transition. The reasons for these discrepancies are not clear, but may be caused by different checkpoint pathways active in the two systems.
Several non-mutually exclusive mechanisms of a lamin-directed control of Rb function can be envisaged: first, lamin complexes may regulate cell-cycle-dependent phosphorylation and dephosphorylation of Rb. In Lmna−/− fibroblasts, Rb was found to be hyperphosphorylated, impairing TGFβ (transforming growth factor β)-mediated cell-cycle arrest . The authors propose that A-type lamins serve as a platform to recruit Rb and PP2A, promoting efficient dephosphorylation of Rb. Secondly, lamins may be important for the stability of Rb, since Lmna−/− fibroblasts exhibited reduced Rb levels . The mechanism of lamin-mediated Rb stabilization is unclear, but is independent of the gankyrin and MDM2 (murine double minute 2) degradation pathways . Finally, lamin complexes may affect E2F/Rb-dependent promoter activity by recruiting additional transcriptional and/or epigenetic regulators.
Deregulation of Rb activity may also contribute to the pathologies of lamin-linked diseases. In support of this, gene-expression profiling of muscle biopsies from lamin-linked Emery–Dreifuss muscular dystrophy patients  revealed expression defects in genes regulated by Rb and by the transcription factor MyoD, which are required for muscle stem cell regulation and differentiation. In addition, skin fibroblasts from patients with Hutchinson–Gilford progeria, a premature aging syndrome resulting from an incomplete post-translational processing of lamin A, show reduced levels of Rb . However, not all disease-linked lamin mutations affect Rb stability, since introduction of many mutated laminopathy-causing lamin variants into Lmna−/− cells restored Rb levels .
Regulation of c-Fos and ERK1/2 (extracellular-signal-regulated kinase 1/2)
c-Fos and its activator kinases ERK1/2 are also targets of lamin regulation [22,23]. c-Fos heterodimerizes with c-Jun to form the transcriptionally active dimeric transcription factor AP-1 (activator protein 1), which participates in the regulation of cell proliferation, differentiation and apoptosis and is activated in response to various signals, including cytokines and growth factors. In serum-starved fibroblasts, c-Fos and ERK1/2 are found to be associated with the nuclear periphery in a strict lamin A/C-dependent manner. The binding of c-Fos to A-type lamins inhibits its heterodimerization with c-Jun, thereby suppressing AP-1-dependent transcription. .
Following serum stimulation, c-Fos is phosphorylated by ERK1/2 and dissociates from the NE, leading to a fast activation of AP-1 and binding to AP-1-regulated promoters. Thus lamina-bound ERK1/2 may function as a molecular switch for rapid mitogen-dependent AP-1 activation through phosphorylation-induced release of c-Fos from its inhibitory interaction with lamin A/C. In Lmna−/− fibroblasts, AP-1 transcriptional activity is up-regulated, suggesting that the increased proliferation of lamin A/C-deficient cells could be mediated by c-Fos hyperactivation.
Lamins affect tissue-specific differentiation pathways
Lamins have also been reported to regulate various transcription factors and signalling molecules involved in tissue-specific differentiation. SREBP1 (sterol-regulatory-element-binding protein 1), a basic helix–loop–helix-leucine zipper transcription factor that regulates genes involved in adipogenesis interacts with farnesylated prelamin A [24,25], which decreases the active pool of SREBP1. Overexpression of prelamin A, seen in several lamin-linked lipodystrophies may thus impair adipogenesis through SREBP1 inactivation. Lamin A also affects downstream events of the Notch signalling pathway, required for the differentiation of mesenchymal stem cells. Lamin A binds to and inactivates SKIP (Ski-interacting protein), a co-activator of Notch-dependent target genes . Progeria cells expressing disease-causing lamin A variants that cannot bind SKIP exhibit up-regulated Notch signalling, which in turn causes enhanced osteogenesis and decreased adipogenesis . Another transcription repressor, Oct-1, is sequestered at the nuclear lamina by its interaction with lamin B1, which limits access to its promoters. In lamin B1-deficient cells, Oct-1 is released from the nuclear periphery and down-regulates Oct-1 target genes, many of which are involved in the response to oxidative stress . Thus loss of lamin B function may increase the amount of reactive oxygen species leading to DNA damage and cellular aging.
Lamin complexes mediate chromatin organization and gene positioning
Transcriptional regulation by lamins is also mediated by epigenetic modifier components. For example, direct binding of lamin A/C to an epigenetic regulator ING1 (inhibitor of growth 1) was found to be important for its nuclear localization and function. ING1 protein interacts with core histones, histone deacetylases and histone acetyltransferases and mediates epigenetic regulation . In addition, several lamin-binding proteins of the nuclear lamina are involved in the maintenance of a transcriptionally repressive environment at the nuclear periphery. For instance, LAP2β associates with histone deacetylase 3  and LBR (lamin B receptor) binds HP1 (heterochromatin protein 1), thereby linking the hypoacetylated heterochromatic regions to the NE [29,30]. Furthermore, the inner nuclear membrane proteins LAP2β, emerin, MAN1 and LEM2 share a common motif, the LEM (LAP2–emerin–MAN) domain , which binds a conserved metazoan DNA-binding protein named BAF (barrier to autointegration factor). Thus the DNA-cross-linking activity of BAF in lamin–LEM protein complexes at the nuclear periphery also contributes to the formation of densely packed heterochromatin . Spatial organization of chromatin in the nucleus is not random, and numerous examples suggest that intranuclear positioning of genes provides an additional level of gene expression control. The nuclear periphery is enriched in heterochromatin and has traditionally been seen as a transcriptionally repressive environment. In an attempt to define lamina-associated genome regions in human fibroblasts, the entire genome was mapped for lamin B1-interacting domains using a DamID methyltransferase-tagging approach. More than 1300 sharply defined chromatin regions associated with the lamina were identified and were characterized by low gene expression, enrichment of repeat elements and binding motifs for the insulator protein CTCF (CCCTC-binding factor) at the domain borders . This suggested that anchoring of genes at the nuclear periphery can repress their activities, but the molecular mechanisms of tethering are not fully understood. A recent study showing that ablation of lamin B in Drosophila resulted in loss of peripheral localization of a testis-specific gene cluster in somatic cells and its improper transcriptional up-regulation, provides evidence for a role of lamins in tethering .
Several studies have recently been performed to test whether an artificial tethering of a gene to the nuclear lamina in mammalian cells is sufficient to repress its activity. Although two studies showed that such tethering of an artificial gene mediates at least partial transcriptional repression [35,36], another experimentally similar study failed to demonstrate repression upon tethering to the lamina . Thus the nuclear periphery may have both transcriptionally repressive and permissive domains defined by yet unknown mechanisms. Further experiments are needed to establish how interactions with the nuclear lamina can regulate genes in a physiological context.
Lamins in DNA replication
Early studies have revealed co-localization of B-type lamins with replication factor PCNA (proliferating-cell nuclear antigen) at replication sites in mid-to-late S-phase  and of lamin A with early replication sites . The roles of lamins in DNA replication have been extensively studied in in vitro assembled nuclei from Xenopus egg interphase extracts. Depletion of the major lamin in these extracts  or addition of dominant-negative lamin mutants, which inhibit lamin assembly, inhibited DNA replication [41,42]. Since, under the latter conditions PCNA, but not DNA replication initiation factors, were mislocalized to lamin aggregates, it was concluded that the elongation rather than initiation phase of replication was affected . The recently reported direct interaction of PCNA with the C-terminal Ig-fold of lamins  provides some insight into potential mechanisms, but further studies are needed to understand why and how in metazoans this fundamental process depends on lamins.
Lamins in mitosis
A remarkable feature of the metazoan NE is its complete disintegration at the onset of mitosis, and its rapid reassembly around separated chromatids in telophase (for reviews see [45,46]). The rapid activation of cyclin B1–Cdk1 is the crucial event that commits cells to undergo mitosis and initiates NE disassembly. Whereas in G2-phase, cyclin B shuttles between the nucleus and the cytoplasm, before the onset of NE breakdown, Plk1 (Polo-like kinase 1) phosphorylates cyclin B and causes its accumulation in the nucleus [47,48]. NE rupturing that results from the mechanical force produced by the interaction of microtubules and dynein with the NE facilitates NE breakdown; however, nuclear disassembly occurs also in the absence of microtubules [49,50]. At the onset of mitosis, lamins are phosphorylated by Cdk1 on serine residues flanking the rod domain, leading to lamin filament disassembly [51,52]. Lamin A carrying mutations that prevent phosphorylation at these sites failed to depolymerize . Both A- and B-type lamins are dispersed throughout the cytoplasm during mitosis. Unlike A-type lamins, the B-type lamins remain associated with membranes, presumably due to their constitutive farnesyl modification . Two recent studies provided evidence that depolymerization of lamins may assist NE breakdown during mitosis: lipins, conserved PA (phosphatidic acid) phosphatases catalysing the dephosphorylation of PA to diacylglycerol, an important molecule in lipid metabolism and signalling, were found to be essential for NE breakdown in Caenorhabditis elegans [55,56]. Intriguingly, siRNA (small interfering RNA)-mediated down-regulation of lipin-1 impaired NE breakdown and also inhibited lamina depolymerization, but co-depletion of lipin-1 together with C. elegans lamin rescued NE disassembly. Thus the impaired NE breakdown in lipin-deficient embryos may be caused by a failure of lamin filament disassembly. Although the molecular details of these intriguing lipin-mediated effects on lamina disassembly remain to be resolved, these findings support the idea that depolymerization of lamins at the beginning of mitosis may assist efficient NE breakdown.
Early studies have already described a small amount of lamin B associated with the mitotic spindle during mitosis [49,57], but only recent studies provided evidence that the spindle-bound lamin B is the ‘long-sought-after’ structural component of the mitotic spindle matrix . Depletion of lamin B in HeLa cells and Xenopus mitotic egg extracts resulted in defective spindles, suggesting that lamin B was important for the formation of functional spindles. Interestingly, lamin B and several known spindle assembly factors remained as part of a matrix structure following nocodazole-mediated disruption of microtubules. The formation of the lamin B-containing spindle matrix required the presence of RanGTP, as well as two spindle assembly factors Nudel and dynein . Altogether, the data demonstrated that the mitotic lamin B may form the spindle matrix that tethers spindle assembly factors in order to support spindle assembly.
Lamins in nuclear reformation
Following sister chromatid separation, a new NE is formed around the chromosomes. Mitotic endoplasmic reticulum membranes are targeted to chromatin, most likely through the interaction of inner nuclear membrane proteins with chromatin, thus forming a new nuclear membrane around chromosomes . The assembly of A- and B-type lamins into the lamina differs both temporally and spatially. Whereas lamin B is detectable at condensed chromosomes already during late anaphase/telophase [5,61], A-type lamins enter the newly formed nucleus only after B-type lamins and other NE components, including nuclear pore complexes, have been assembled. FRAP (fluorescence recovery after photobleaching) experiments revealed that, upon formation of a new nucleus, the majority of GFP (green fluorescent protein)-tagged lamin A is mobile in the nuclear interior, and during early G1-phase gradually translocates to the nuclear periphery, forming a less mobile polymerized structure. The kinetics of lamin assembly during post-mitotic nuclear reformation indicates that A- and B-type lamins assemble into a filamentous lamina by independent pathways. This is consistent with the finding that inhibition of B-type lamin polymerization at the end of mitosis does not interfere with the assembly of A-type lamins .
Whereas lamina disassembly requires lamin phosphorylation, lamin assembly is mediated by lamin dephosphorylation . PP1, which is targeted to the NE at the end of mitosis through its interaction with AKAP149 (A-kinase-anchoring protein 149), is crucial for lamin B assembly in a cell-free system and is therefore a likely candidate for the lamin dephosphorylating enzyme at the end of mitosis [62,64].
The importance of lamins for nuclear reassembly has been controversial, since first experiments examining NE assembly in cell-free mitotic extracts from which lamins have been immunodepleted yielded contradictory results (reviewed in ). The apparent discrepancies are thought to have arisen as a consequence of inefficient immunodepletion of lamins from the extracts in some experimental set-ups. A small amount of lamin might remain associated with the chromatin upon immunodepletion and, although insufficient to form a detectable lamina, it could be sufficient for its function in NE formation. Taken together, the available data indicate that a fully assembled peripheral lamina is not essential for NE formation, although a small amount of lamins may be required for efficient NE reassembly.
Lamins have long been seen as pure structural elements of the nucleus, but recent exciting developments in the field have provided evidence for regulatory functions of lamins in key cellular processes such as cell-cycle control and differentiation. To this end, numerous transcription factors and signalling pathways have been reported to be affected by lamins, but it is still a long way to understand the molecular details of lamin-mediated regulatory pathways in the context of tissues and organisms. Nevertheless, from the currently available data, several non-mutually exclusive mechanisms can be envisaged. First, lamins can provide a platform for the recruitment of signalling and regulatory factors favouring their interactions and ensuring efficient transmission of signals. Secondly, the nuclear lamina can scavenge transcription factors, thereby removing them from their target promoters and decreasing the active pool (see also ). Thirdly, the lamina can regulate gene expression by tethering genes to the nuclear periphery. Understanding the details of these lamin-mediated regulatory pathways will also be essential to shed light on the underlying molecular mechanisms of lamin-linked diseases and for the development of effective therapeutic strategies.
Nuclear Envelope Disease and Chromatin Organization 2009: Independent Meeting held at College of St Hild and St Bede, University of Durham, Durham, U.K., 22–23 April 2009. Organized and Edited by Chris Hutchison (Durham, U.K.).
activator protein 1
barrier to autointegration factor
extracellular-signal-regulated kinase 1/2
inhibitor of growth 1
proliferating-cell nuclear antigen
sterol-regulatory-element-binding protein 1
We gratefully acknowledge grant support from the Austrian Science Research Fund [grant number FWF P17871] and the EURO-Laminopathies research project of the European Commission [contract number LSHM-CT-2005-018690] to R.F. M.B. is a recipient of an European Molecular Biology Organization postdoctoral fellowship, J.B. is supported by the University of Vienna [grant number I031-B].
These authors contributed equally to this paper.