Immunocytochemistry of Nuclear Domains and Emery-Dreifuss Muscular Dystrophy Pathophysiology N

Home , Nuclear envelope, Nuclear lamina, Nuclear matrix

review Immunocytochemistry of nuclear domains and Emery-Dreifuss muscular dystrophy pathophysiology N. M. Maraldi,a,b,c G. Lattanzi,b P. Sabatelli,b A. Ognibene,a M. Columbaro,d C. Capanni,a C. Rutigliano,a E. Mattioli,a S. Squarzonib aLaboratory of Cell Biology and Electron Microscopy, I.O.R.; bInstitute of Normal and Pathological Cytomorphology, CNR, Bologna; cDepartment of Human Anatomy, University of Bologna; dLaboratory of Neuromuscular Pathology, IOR; Bologna, Italy

©2003, European Journal of Histochemistry he finding that Emery-Dreifuss muscular dystrophy (EDMD) is genetically related to muta- The present review summarizes recent cytochemical findings Ttions in either emerin or lamin A/C has focused on the functional organization of the nuclear domains, with the attention on the actual involvement of the nuclear a particular emphasis on the relation between nuclear enve- envelope in the pathophysiology of the disease. lope-associated proteins and chromatin. Mutations in two nuclear envelope-associated proteins, emerin and lamin A/C Mutations in lamin A/C gene have been implicated in cause the Emery-Dreifuss muscular dystrophy; the cellular several human diseases, including EDMD, limb-girdle pathology associated with the disease and the functional muscular dystrophy type 1B, hyperthrophic cardiomy- role of emerin and lamin A/C in muscle cells are not well established. On the other hand, a large body of evidence opathy, Dunnigan-type familial partial lipodystrophy, indicates that nuclear envelope-associated proteins are and axonal neuropathy Charcot-Marie-Tooth disorder involved in tissue-specific gene regulation. Moreover, chro- type 2. These disorders have been named nuclear matin remodeling complexes trigger gene expression by utilizing the nuclear matrix-associated actin, which is known to envelopathies, or laminopathies. The common origin interact with both emerin and lamin A/C. It is thus conceiva- of these diseases raises the importance of the func- ble that altered expression of these nuclear envelope-asso- tional role of lamin A/C and nuclear envelope-associ- ciated proteins can account for an impairment of gene ated proteins in the maintenance and/or acquisition of expression mainly during cell differentiation as suggested by recent experimental findings on the involvement of emerin in the cell phenotype. The involvement of different myogenesis. The possibility that Emery-Deifuss muscular nuclear domains in the control of developmental dystrophy pathogenesis could involve alteration of the signa- expression of tissue specific genes has been suggest- ling pathway is considered. ed. However, the functional features of these domains Key words: Emery-Dreifuss muscular dystrophy, emerin, as related to nuclear envelope-associated protein lamin A/C, nuclear envelope, chromatin, nuclear signaling, expression and/or localization are yet not defined. A chromatin remodeling complexes, immunocytochemistry possible functional link among chromatin remodeling complexes, intra-nuclear signaling pathways and Correspondence: Nadir M. Maraldi, Laboratorio di Biologia nuclear envelope-associated proteins will be described Cellulare e Microscopia Elettronica, Istituti Ortopedici in the following sections. “Rizzoli”, via di Barbiano 1/10 40139 Bologna, Italy. Phone: +39-51-6366857. Fax: +39-51-583593. E-mail: [email protected] Immunocytochemistry of the nuclear domains The coordination of all cell functions occurs in the Paper accepted on July 8, 2002 command center located at the nucleus; essential to European Journal of Histochemistry this control is the ability to regulate the transit of 2003; vol. 47 issue 1 [Jan-Mar]:3-16 molecular species into and out of the nucleus. The communication between nucleus and cytoplasm that involves the transport of macromolecular complexes is a key step in gene expression regulation (Dreyfuss and Struhl, 1999). This traffic is regulated by very large and complex structures (30-times the size of a ribosome) called nuclear pore complexes (NPCs) that perforate as gateways the outer and inner membranes of the nuclear envelope (NE). Within the nucleus, the structural organization is less defined

N.M. Maraldi et al. than in the cytoplasm. Cytoplasmic organelles are Organization of nuclear compartmentalization discrete, generally membrane-bound structures, Development in instrumentation and reagents which can be isolated in a considerably pure form to greatly improved the detection of nucleic acids and be analyzed and utilized for in vitro functional assay. proteins by in situ hybridization and immunocyto- The nucleus is far denser than the cytoplasm, due to chemical analyses. The combination of non-isotopic the presence of highly polymeric molecules such as methods with computer-aided imaging provided DNA and RNAs, and contains a variety of subnu- much more complete insights into the actual distri- clear structures which lack membranous bound- bution of genes and regulatory factors that consti- aries. Very complex functions, like transcription, tute the in situ mapping of gene expression. occur at large macro-molecular assemblies, involv- The role of nuclear domains in the control of ing some tenths of components that must interact at developmental expression of cell-growth and tissue- the right time and place (Stein et al., 2000). This specific genes is of fundamental importance also to peculiar arrangement of the nucleus and the com- understand the molecular basis of pathological plexity of the functions occurring in the crowded alterations (Maraldi et al., 1998). However,the con- nuclear space can be studied by a combination of trol of gene expression and the identification of the molecular and morphological approaches. structural organization of the nucleus have been In recent years the cytochemical techniques have longer considered as minimally integrated questions gained formidable insights into the functional organi- (Misteli, 2000).This integration has been recognized zation of the nucleus utilizing complementary in other cases; thus there is longstanding acceptance approaches based on confocal laser scanning micro- that the nucleolus is the domain at which ribosome scopy (CLSM) and electron microscopy (EM). biosynthesis occurs (Olson et al., 2000).The import CLSM provides the three-dimensional information of gene regulatory factors and the export of tran- required to analyze the spatially arranged nuclear scripts have been mapped at the level of nuclear domains (Maraldi et al., 1999a). A drawback of flu- pore complexes (Wente, 2000). DNA replication orescent probe detection within the nucleus is the occurs at distinct foci that associate into larger com- degradation of nuclear organization caused by the plexes and early and late replicating chromatin ter- permeabilization procedure that allows the probe to ritories correspond to R and G bands of metaphase reach the nuclear interior; alterations are also caused chromosomes (Zink et al., 1999). Less defined are by denaturation and deproteination processing the structural bases of the functions that control required to follow DNA replication by BrdU uptake gene expression.This is obviously due to the intrinsic (Visser et al., 2000).The use of green fluorescent pro- complexity of the phenomenon which is not restrict- tein (GFP)-labeled probes expanded the capability of ed in time as DNA replication, or in space, as nuclear CLSM to analyze the nuclear domain activity in vivo. import/export processes. Furthermore gene expres- A new chapter concerns the dynamics of some sion modulation through signaling molecules and nuclear functions by the use of fluorescence recovery transcription factors (TFs) is an impressively com- after photobleaching (FRAP) or fluorescence loss in plex event that involves chromatin remodeling, tar- photobleaching (FLIP) techniques to follow the very geting of TFs to regulatory chromatin sequences, rapid movements of (GFP)-labeled proteins within recruitment of RNA polymerases, polyadenylation of the nuclear domains (Phair and Misteli, 2000). primary transcripts, recruitment of splicing factors, Electron microscopy immunocytochemistry attains a splicing of hnRNAs, maturation and transport of spatial resolution of about 10-20 nm with respect to mRNAs to the nuclear pore complex. All these steps that of 250-700 nm of CLSM methods and a better occur in a limited space corresponding to the inter- preservation of nuclear organization, although three- chromatin area between the chromosome territory dimensional information can be obtained only by containing the transcribed gene and the nearest complex methods of reconstruction of serial sections. nuclear pore (Visser et al., 2000). In this interchro- In several cases, in which the actual function of puta- matin domain several macromolecular aggregates tive nuclear domains is still debated, such as the chro- have been structurally identified and more or less mosome territories (Visser et al., 2000) or the inosi- stringent association with specific functions tol lipid signal transduction systems (Maraldi et al., described. The list of these nuclear components is 1999b), the comparison between CLSM and EM wide, although probably not exaustive. Coiled (Cajal) labeling patterns is mandatory. bodies (CBs), gemini of coiled bodies (Gems), inter-

review chromatin granule clusters (IGCs) or speckles, and organization imply: i) a system which macromolecu- promyelocytic leukemia (PML) bodies, are quite lar complexes are transiently associated with, at large structures detectable by fluorescence micro- given functional sites, mainly through specific inter- scopy as distinct foci changing in number and size actions with the nuclear matrix components during the cell cycle (Lamond and Earnshaw, 1998). (Kruhlak et al., 2000); ii) a nucleopasmic space in Perichromatin fibrils, perichromatin granules, and which components diffuse freely but concentrate at isolated groups of interchromatin granules are very site of function through reciprocal interactions numerous small structures detectable at the electron (Misteli, 2000). microscope level (Stein et al., 2000). An impressive amount of molecular components have been identi- The nuclear matrix in nuclear domain organization fied by cytochemical methods at these subnuclear in normal and pathological conditions structures; most of them are constituents of the The experimental evidence of the structural organ- transcription machinery or regulatory factors that ization and functional properties of the nuclear modulate transcript synthesis, processing and tar- matrix has been continuously accumulating in the geting (Mintz et al., 1999; Lewis and Tollervey, past twenty years.The nuclear matrix is constituted 2000). by different components, the nuclear lamina, strictly The existence of nuclear domains, though univer- associated to the nuclear envelope, the nucleolar sally accepted, does not imply that the mechanisms remnant, and the inner nuclear matrix, that interacts that coordinate the spatial organization of genes, with the chromosome territories, the interchromatin transcripts and regulatory proteins within the nucle- domains and the nucleolus. Some classes of nuclear us are recognized as well. In fact, the compartmen- matrix proteins have been isolated and the genes talization of regulatory factors that sustain a cog- encoding them cloned (Nickerson, 2001).The major nate nuclear function could be either maintained by proteins of the nuclear lamina are lamins B and A/C an underlying macromolecular framework, or driven (Moir et al., 1995; Stuurman et al., 1998; by diffusion mechanisms coupled with transient Gruenbaum et al., 2000). Lamins B1 and 2 have associations among functionally related components been reported to interact with cytoskeletal interme- (Shopland and Lawrence, 2000). diate filaments (Djabali et al., 1991). Lamins A/C These opposite views of nuclear compartment have been also localized at the nuclear interior, con- organization and origin are based on a large and ceivably interacting with inner nuclear matrix pro- compelling amount of experimental data which are teins (Moir and Spann, 2001). The most character- substantially non contradictory. However, these istic proteins of the inner nuclear matrix are the results are differently interpreted on the basis of the nuclear matrins (Belgrader et al., 1991), the 240- accepted or refused evidence of a non-chromatin, kDa NuMA (He et al., 1995), and the 170-kDa DNA insoluble nuclear matrix within the nucleus (Fey et Topoisomerase II a (Berezney et al., 1995; Zini et al., 1991). DNA replication, ribosome assembly, al., 1994). Another nuclear matrix-associated group transcript synthesis and processing are spatially and of proteins is represented by nuclear actin and by temporally regulated. The main mechanism through several actin-related proteins (Amankwah and Boni, which this occurs is a specific distribution of the fac- 1994; Rando et al., 2000). tors involved in each of these processes at given A class of nuclear matrix proteins specifically bind nuclear sites.The localization of factors into macro- DNA sequence landmarks, defined S/MARs (scaf- molecular assemblies, morphologically detectable by fold/matrix-associated regions), thus defining the immunocytochemical methods, is not determined by base of topologically constrained chromatin loops, membrane-delimited territories but by less defined that either insulate genes from the influence of cis- domains to which specific functions are dynamically acting elements, or play transcription enhancing associated. The question that arises is: which is the activity (Boulikas, 1995). S/MARs as well as some basis of nuclear compartmentalization? An obvious classes of RNPs involved in RNA splicing and trans- reply should be: as in the cytoplasm, a skeletal array port are so specifically bound to nuclear matrix pro- of auto-assembling filaments could both maintain teins that can not be removed by high-salt extraction and dynamically modify the nuclear domains. As but only by apoptotic proteolytic cleavage (Martelli many obvious think, this interpretation is still ques- et al., 1997). A variety of nuclear functions have tioned. Alternative hypotheses on the nucleoplasm been demonstrated to require interactions or local-

N.M. Maraldi et al. ization at the nuclear matrix, including gene expres- depend on mutations of nuclear envelope-associated sion regulation (Nickerson, 2001).Transcription fac- proteins.This fact appears to be particularly signifi- tors are not simply translocated to the nucleoplasm cant in confirming the essential link between nuclear by the presence of nuclear localization signals organization and the pathophysiology of a group of (NLS). Transcription factor transactivation activity hereditary diseases identified as nuclear envelo- depends on their targeting to the nuclear matrix pathies (Hegele, 2000). and, in some instances, to specific nuclear domains, EDMD is a rare form of X-linked muscular dys- through specific localization signals (Stein et al., trophy, displaying variable phenotypical expressivity, 2000). Gene expression and chromatin arrangement characterized by muscle wasting and weakness, with are closely associated events being the first depend- tendon contractures in the lower leg and the upper ent on the second. Chromatin arrangement is active- arm, as well as by progressive atrio-ventricular con- ly modulated in response to physiological factors by duction defects, that was firstly described by Emery macromolecular assemblies of nuclear matrix-asso- and Dreifuss (Emery, 1989).This recessive muscular ciated proteins, the chromatin remodeling complexes dystrophy form (XL-EDMD) is caused by loss of (CRCs), that allow histone-modifying enzymes and expression of emerin (Bione et al., 1994), a nuclear transcription factors to reach their targets membrane protein (Nagano et al., 1996; Manilal et (Hagmann, 1999; Berger, 2000). al., 1996). An autosomal dominant form of EDMD The evidence of a functional relationship between has been then characterized, demonstrating that this nuclear matrix-associated structures and gene form of the disease (AD-EDMD) is caused by muta- expression is consistent with the finding that a mod- tions in the gene that encodes alternatively spliced ified subnuclear organization of genes and regulato- lamins A and C (Bonne et al 1999). Interestingly, the ry factors occurs in pathological conditions like can- two forms of EDMD show very similar clinical fea- cer and neuropathies. In both cases there are modi- tures. Muscle fiber necrosis is rare, while a certain fications in the components of the nuclear architec- amount of fibers show an high size variability and ture involved in the control of gene expression. As internal nuclei. Moreover, some muscles show pecu- examples, in normal cells PML transcription factor liar early contractures, or permanent shortening. At resides in discrete PML bodies associated with the the heart level, cardiomyocyte necrosis and cardiac nuclear matrix, while in promyelocytic leukemic cells conduction system degeneration are not usually the PML protein is genetically rearranged and dis- reported (Emery, 2000). In XL- and AD-EDMD, persed throughout the nucleus (Dyck et al., 1994). ultrastructural alterations have been described The subnuclear distribution of ataxin-1 is altered in affecting both chromatin and the nuclear envelope- spinocerebellar ataxia type 1. Because ataxin-1 is associated structures, such as the nuclear lamina nuclear matrix-associated, the pathogenesis of spin- and the nuclear pores (Ognibene et al., 1999; ocerebellar ataxia involves the progressive disrup- Sullivan et al., 1999). Heterochromatin distribution, tion of the nuclear matrix and a rearrangement of focal absence of heterochromatin, loss of contact the nuclear domains (Skinner et al., 1997; Tait et between heterochromatin and the nuclear lamina, al., 1998). It is therefore evident that the fidelity of nuclear lamina thickening, non uniform nuclear pore the subnuclear organization of factors is essential distribution, are typical ultrastructural changes that for integration of the signals that regulate expression characterize the affected phenotype (Figures 1 and of genes that control cell growth and phenotype 2). In which way emerin and lamin A/C altered (Stein et al., 2000). expression could result in chromatin arrangement modification, and, in turn, in skeletal and cardiac Emery-Dreifuss muscular dystrophy and other muscle functional impairment? nuclear envelopathies Emery-Dreifuss muscular dystrophy (EDMD), The nuclear envelope and the nuclear lamina limb-girdle muscular dystrophy type 1B, hyper- proteins throphic cardiomyopathy (Nagano and Arahata, The finding that a muscular dystrophy, the EDMD, 2000), Dunnigan-type familial partial lipodystrophy arises due to defects in nuclear envelope-associated (Cao and Hegele, 2000), and axonal neuropathy proteins instead of cytoskeletal/plasma membrane- Charcot-Marie-Tooth disorder type 2 (DeSandre- associated proteins, such as in Duchenne muscular Giovannoli et al., 2002), have been demonstrated to dystrophy, was completely unexpected.Therefore, the

review

Figure 1. Skin cultured cells from AD-EDMD patient, fluorescence microscopy. A) Anti-lamin A antibody; in the field two nuclei are vis- ible, the bottom one showing an uniform distribution of the labeling all along the nuclear surface, the top one showing local absence of labeling at irregularly shaped black areas; B) Anti-emerin antibody; the labeling pattern is almost identical to that of lamin A, indicating that the altered expression of lamin A/C prevents a regular arrangement of the nuclear envelope-associated protein emerin. Scale bar = 1 µm.

attention has been focused on the nuclear envelope, lamin A/C (Fairley et al., 1999). Emerin is a mem- which represents the site at which both emerin, the ber of the lamina-associated nuclear envelope (NE) protein altered in the X-linked recessive form of membrane proteins (Tews, 1999); the other known EDMD, and lamin A/C, the protein linked to the members being: lamin B receptor (LBR), lamina- autosomal dominant form of EDMD, interact. The associated polypeptide 1 (LAP1 A, B, C), and two proteins are not analogous; in fact, emerin is a LAP2β, MAN1, and nurim (Morris and Manilal, transmembrane protein associated to the inner 1999). All these proteins have hydrophobic domains nuclear membrane, whilst lamin A/C is a member of that constitute putative transmembrane anchoring the intermediate filament family located at the sites, and some of them have been reported to pres- nuclear lamina. However, the two proteins interact, ent lamin- and chromatin-binding sequences since emerin is retained at the inner nuclear mem- (Cartegni et al., 1997; Manilal et al., 1999; Östlund brane owing to direct or mediated interactions with et al., 1999). These molecular interactions con-

Figure 2. Skeletal muscle biopsies, electron microscopy. A) XL-EDMD; B) AD-EDMD. Nuclei show typical ultrastructural alterations with local absence of peripheral heterochromatin (black arrows) and thickening of the nuclear lamina (white arrows). Scale bar = 0.5 µm.

N.M. Maraldi et al. tribute to the chromatin arrangement (Figure 3). lize the α-helix interactions at the rod domain, pre- Emerin is a type II integral membrane serine-rich venting dimerization (Hutchison et al., 2001; protein; it is linked to the inner nuclear membrane by Genshel and Schmidt, 2000). Because of differential its hydrophobic C-terminal tail.This typical localiza- expression of lamins and lamin-associated proteins tion depends on interactions with other nuclear in different tissues, the LMNA mutations can differ- membrane proteins as well as on interactions with ently affect the tissue pathophysiology. Moreover, nuclear lamina or nuclear matrix proteins (Wilson, lamin A/C could also interact with TFs such as 2000). Transmembrane sequence is involved in SREBP-1, affecting gene expression (Wilson, emerin localization throughout the ER system, 2000). Finally, abnormal interaction of mutated whilst at least two separate sequences are essential lamins with cytoskeletal or sarcomeric proteins for nuclear envelope targeting; interestingly, emerin, (belonging to the intermediate filament family) when overexpressed, can reach also extra-nuclear might occur in the skeletal and cardiac cells in membrane districts (Östlund et al., 1999).The iden- EDMD and hypertrophic cardiomyopathy 1A tification of emerin also at cytoplasmic sites has (Herrmann and Aebi, 2000). Lamins are type V been reported (Cartegni et al., 1997; Östlund et al., intermediate filament proteins that form orthogonal 1999; Squarzoni et al., 2000), although its localiza- rather than linear arrays. The complexity of the tion at the intercalated discs in the heart (Cartegni nuclear lamina ultrastructure, including both lamins et al., 1997) has been partly disregarded by the evi- and lamin-binding proteins, appears to be strictly dence of nuclear membrane exclusive staining by related with the evolutional complexity in eukaryotes means of monoclonal antibodies (Manilal et al., (Cohen et al., 2001). On the other hand, the evolu- 1999). Emerin can be considered a bridge between tion of nuclear lamina increases the efficiency of the inner nuclear membrane and the nuclear lamina, nuclear envelope disassembly, which is a prerequisite since it remains strongly associated with the nuclear of a more efficient mitotic process and which pro- lamina in purified nuclear matrix preparations vides new mechanisms to modulate chromatin (Squarzoni et al., 1998). Diffusion of emerin is how- expression through exposure to cytoplasmic factors. ever limited by interactions with proteins belonging Furthermore, a more sophisticated nuclear lamina to nuclear structures that face the inner nuclear probably confers selective advantages, possibly due membrane, that is the nuclear lamina and the periph- to improved chromatin organization, nuclear signal- eral heterochromatin. Lamins A and C interact with ing and gene expression (Gotzmann and Foisner, emerin as demonstrated by co-immunoprecipitation 1999). This facts should be taken into account to and blot overlay experiments (Fairley et al., 1999; determine how mutations in nuclear lamina proteins Sakaki et al., 2001), and by knock-out mice model could cause inherited diseases. Nuclear lamina com- (Sullivan et al., 1999). A putative emerin partner is ponents can affect other cell responses, some of the chromatin protein BAF (Barrier to Auto- which could be involved in EDMD pathophysiology, integration Factor), that has been demonstrated to such as apoptosis, chromatin remodeling and tran- bind LAP2 (Furukawa et al., 1998). It has been sug- scriptional regulation. gested that BAF-binding proteins might influence chromatin arrangement and expression (Wilson, Hypotheses on the pathophysyology of EDMD 2000). Some hypotheses have been advanced in order to Lamins are differentially expressed during devel- explain the role of mutant nuclear envelope-associ- opment and differentiation; in man, lamins B1 or B2 ated proteins in the pathophysiology of EDMD are expressed in almost all cell types, while lamins A (Figure 3).The first is that of the mechanical stress, and C are expressed primarily in differentiated non- analogous to that proposed in other muscular dys- proliferating cells, suggesting that they are involved trophies (Manilal et al., 1999). The mechanical in gene expression.The LMNA gene products, lamins stress has been suggested to cause, in muscular dys- A and C, are produced by an alternative splicing at trophies involving cytoskeletal/plasma membrane- exon 10 of LMNA, so that the two proteins share the associated proteins, a progressive damage of the first 566 residues, but have distinctive C termini myofibrils, due to a not coordinated association (Stuurman et al., 1998). Lamin A but not lamin C between the extracellular matrix and the cytoskele- can be modified by farnesylation. LMNA mutations, ton during contraction. A similar situation is less by affecting charge or hydrophobicity, could destabi- plausible in EDMD; however, emerin may be part of

review

Figure 3. Molecular interactions among nuclear envelope-associated proteins, nuclear lamina, nuclear matrix and chromatin. INM (scaffold): only Lamin A and βb-actin are represented; CRC: PI(4,5)P2-dependent Chromatin Remodeling Complex BAF. Different mechanisms possibly involved in EDMD pathophysiology are indicated as affecting specific nuclear targets (arrows). Loss of emerin or altered expression of lamin A/C could result in: damage at the nuclear envelope membranes (mechanical stress); unstable nuclear lamin (apoptosis); altered nuclear matrix/chromatin interactions affecting gene silencing by heterochromatization (chromatin arrangement); altered binding to the nuclear lamina/matrix of trancriptional regulators (gene expression); altered mechanisms of CRC involving PI(4,5)P2-dependent actin polymerization that affect nuclear matrix/chromatin interactions (signal transduction).

a nucleo-cytoskeletal network which protects the only muscle fibers, due to the mechanical stress, nucleus of contractile cells from the mechanical undergo nuclear alterations. In skeletal muscle syn- stress (Maniotis et al., 1997). Accordingly, muta- cytium the damage is limited, because not all nuclei tions of lamin A/C could determine an increased sen- are damaged simultaneously. In contrast the damage sitivity to the stress in cardiac and skeletal muscle of cardiomyocytes will be cumulative and could lead nuclei which lack lamin B1 (Manilal et al., 1999). to conduction blocks (Hutchison et al., 2001). Nuclear fragility as a cause of the defects occurring Another hypothesis has been recently advanced, in EDMD has been suggested in knockout mice lack- that takes into account the fact that muscular dys- ing lamin A/C by the presence of muscle wasting and trophies are degenerative diseases that involve two contractures not at birth, but after some weeks recognized mechanisms for cell death, apoptosis and (Sullivan et al., 1999). These symptoms are accom- necrosis, that might be not mutually exclusive. In panied by an altered nuclear shape, due to the EDMD, that seems not to involve inflammatory absence of lamin A/C and abnormal distribution of responses, the altered expression of nuclear enve- emerin (emerin is mislocalized to the ER, suggesting lope-associated proteins could have no deleterious a lamin A/C-emerin interaction at the NE), that effect on most cells, but it might make cardiac and occurs not only in muscle cells, but also in other tis- skeletal muscle cells more susceptible to pro- sues (Sullivan et al., 1999). Therefore, the nuclei of grammed cell death (Morris, 2000). However, a all tissues, once isolated, are fragile, but conceivably direct demonstration is still lacking, since apoptosis

N.M. Maraldi et al. is rather difficult to identify in adult tissues, because signal transduction cascades affecting cell develop- it has to be identified as an ongoing process affect- ment and the acquisition and maintenance of a dif- ing a small proportion of cells, that maintain cycling ferentiate phenotype (Yoshida et al., 2000; Chen et properties, such as satellite cells. al., 2000). NE-associated proteins might affect gene A further hypothesis, which is progressively more expression and cell differentiation being part of accepted, suggests that emerin and lamin A/C may structural platforms for signaling systems (Ellis et regulate gene expression by interacting with specific al., 1998), located at the NE or within the nucleus. transcription factors or DNA sequences (Östlund et The signaling hypothesis partly overlaps the gene al., 1999; Tsuchiya et al., 1999), as well as modu- expression hypothesis, suggesting a way by which lating chromatin arrangement (Wilson, 2000). This NE-associated proteins can interfere not only with hypothesis is sustained by several experimental find- chromatin arrangement but also with the release of ings: a cell cycle-dependent binding of LAP2β to signaling molecules that modulate transcriptional lamin B1 controls the increase of nuclear volume regulators. which occurs in cycling cells and allows the activation of genes involved in the triggering of S phase Signal transduction at the nuclear level and the (Gruenbaum et al., 2000). Emerin too has a role in modulation of chromatin arrangement cell-cycle dependent events, being phosphorylated in The nucleus represents not only the final target of different forms which are expressed in a cell cycle- signaling molecules but also an autonomous cellular dependent way (Manilal et al., 1999). Lamins are compartment where signaling molecules are gener- capable of interacting with Rb, an oncosuppressor ated. In fact, one of the main class of signaling mol- gene product that regulates cell cycle progression by ecules, the phosphoinositides that give rise to the recruiting histone deacetylase complexes that cause second messengers diacylglycerol and IP3, or act chromatin condensation (Mancini et al., 1994). directly as second messengers such as PI(4,5)P2,is Finally, in XL-EDMD patients (Ognibene et al., not only present at the cell membrane (where the 1999) and in lamin-A/C-null mice (Sullivan et al., receptors of several hormones and growth factors 1999), the peripheral heterochromatin layer under- are located), but also at the cell nucleus (Maraldi et lying the NE is altered, suggesting that the chro- al., 1999b).The inositol lipid signaling system at the matin attachment to the NE is an essential mecha- nucleus is characterized by the presence of: i) spe- nism for the regulation of the cell cycle and of gene cific isoforms of the lipid kinases that phosphorylate expression (Wilson, 2000). The modulation of chro- the inositol ring, ii) phospholipases that hydrolyze matin arrangement that might influence gene polyphosphoinositides, iii) protein kinases that are expression should occur both at the nuclear enve- the targets of the lipid-derived second messengers lope-associated chromatin as well as at other (Irvine, 2000; D’Santos et al., 1998). The nuclear intranuclear sites, where lamin A has been localized signaling system, moreover, responds in a specific (Neri et al., 1999; Hutchison et al., 2001). manner to agonists that modulate cell proliferation An alternative hypothesis takes into account the or differentiation (Cocco et al., 2001; Chi and influence that NE-associated proteins can exert on Crabtree, 2000). Interestingly, the responses induced signal transduction. On the other hand, also the by these effectors are not a mere duplication of pathogenesis of muscular dystrophies involving cyto- those occurring elsewhere in the cell (Irvine, 2000). skeletal/cell membrane-associated proteins could be Some elements of the system are located at the explained by altered signal transduction mechanisms nuclear envelope, other at intranuclear domains (Rando, 2001). In fact, an altered localization of fil- involved in RNA splicing, such as interchromatin amin 2 (a sarcoglycan interacting protein), occurs in granules that are linked to the nuclear matrix limb-girdle and Duchenne MD; this protein is (Maraldi et al., 1999b). It is conceivable that alter- involved in cytoskeletal actin reorganization and sig- ation in the expression of NE-associated proteins nal transduction cascades associated with cell dif- can interfere with signal transduction at the nuclear ferentiation in a way that recalls α 7 integrin subunit level. (Thomson et al., 2000), calpains and caveolins One of the most impressive phenomena that fol- (Razani et al., 2000). The dystrophin-glycoprotein lows cell activation (for example, peripheral lympho- complex, originally believed to play a structural role, cyte exposure to antigens) is the massive chromatin is more likely considered to dynamically modulate rearrangement that occurs in few minutes and pre-

review cedes gene activation. The phenomenon consists in proteins, suggesting a direct interface between chro- the decondensation of a large amount of hetero- matin regulation and signal transduction at the chomatin, causing the enlargement of the cell vol- nuclear level (Maraldi et al., 1999b; Zhao et al., ume, and is mediated by multimeric protein com- 1998). plexes, the chromatin remodeling complexes (CRCs) These experimental data support the following (Wang et al., 1996). Chromatin decondensation is conclusions: i) chromatin remodeling is a rapid and therefore a prerequisite that allows transcription efficient mechanism that constitutes a prerequisite factors to activate specific gene promoters. of gene activation (Kadonaga, 1998; Kingston et al., Chromatin organization is thought to be dependent 1996); ii) CRCs are nuclear matrix-associated sys- on the relationships between nucleohistone fibers tems capable of responding to signaling molecules to and the nuclear matrix (Nickerson, 2001); chromo- rapidly modify the nuclear arrangement through some domains that persist into interphase are struc- actin polymerization (Zhao et al., 1998); iii) inosi- turally connected with the inner nuclear matrix and tol lipid-derived second messengers, capable of inter- with the nuclear lamina. It is therefore conceivable fering with actin-binding proteins, constitute a sig- that alteration in the expression of lamins and lamin- naling system present within the nucleus and associated proteins can interfere with chromatin responding to pathophysyological stimuli (Maraldi arrangement (Boyle et al., 2001). Heterochromatin et al., 1999b). dynamics appears to be related to the cell cycle. In It is conceivable that mutations in lamins A/C and late S, in fact, HC is released from the NE and emerin can affect gene expression through a signal- moves towards the replication sites and then returns ing mechanism capable of modulating chromatin to the envelope and could be regulated by lamins and arrangement in response to external stimuli. Emerin- LAP2, which are phosphorylated at specific sites deficient nuclei, or lamin A/C-mutated nuclei might (Dreger et al., 1999). Furthermore each cell type present subtle defects arising from altered chro- has a characteristic nuclear shape and a typical HC matin structure that not necessarily affect gene pattern, which conceivably affects its transcriptional expression in all cell types (Wilson, 2000). Some cell expression. Both these phenotypic characters are types, in fact, like lymphocytes, alternate long quies- acquired by descending cell lineages. cent phases with sudden activation periods; such The molecular basis of chromatin remodeling changes require a profound chromatin remodeling through a nuclear localized signal transduction sys- and dramatic reprogramming of the whole nuclear tem, has been recently identified. Several chromatin size and shape (Zhao et al., 1998). Mesenchymal remodeling complexes (CRCs) have been character- cells arising in the bone marrow differentiate into ized in yeast, Drosophila, and mammals (Wang et the various cells showing alterations in patients al., 1996; 1998). All of the CRCs appear similar in affected by envelopathies. Adipocytes, that are their ability to modify nucleosomal structure and affected in the Dunnigam-type familial partial allow binding of transcription factors. In higher lipodystrophy (Cao and Hegele, 2000), and motor eukaryotes CRCs are multisubunit 2 MDa protein neurons, altered in CMTD Type 2 (DeSandre- complexes, among which the complexes called Giovannoli et al., 2002), do not derive from the mes- Brahma-related gene associated factors (BAF) have enchyma (Procop, 1997) but, as muscle cells, are been extensively studied (Wang et al., 1996). In usually non proliferating. Subtle alteration in chro- human lymphocytes, the BAF complex, constituted matin arrangement affecting gene expression in by BRG1, β-actin and the actin-related protein given situations might negatively affect mainly long- BAF53, is actively induced to associate to the lasting cells. nuclear matrix/chromatin within 10 min after anti- gen receptor stimulation, before detectable increase Nuclear envelope/chromatin interactions in normal in transcription or protein synthesis (Zhao et al., conditions and in EDMD 1998). The activation of the BAF complex, leading Proteins of the inner nuclear membrane, nuclear to an impressive and sudden chromatin decondensa- lamina and chromatin are highly interacting (Figure tion, is triggered by PIP2 levels that control the 3). Emerin, LAP2β and MAN1 are characterized by association of the complex with the nuclear matrix. the LEM box, constituted by a 43-residue motif by In turn, PIP2 modulates actin polymerization in the which these proteins can interact with DNA-binding BAF complex by displacing nuclear actin-binding proteins capable of affecting chromatin arrange-

N.M. Maraldi et al. ment (Furukawa et al., 1998; Lin et al., 2000). the MARs of the chromatin loops.The inner nuclear Other members of the nuclear envelope/nuclear matrix 10 nm filaments show a non-helical arrange- matrix system interact with DNA-binding proteins, ment and, like intermediate filaments, have an axial such as LBR with the heterochromatin protein HP1 repeat of 23 nm. They should include hnRNP core (Ye and Warman, 1996), and lamins with histones proteins A2/B1, NuMA, lamins, nuclear actin and (Goldberg et al., 1999) and the retinoblastoma (Rb) actin-binding proteins that can bind NuMA protein (Brehm et al., 1998). Particularly interest- (Nickerson, 2001). As suggestively stated in a ing is the case of Rb, which inhibits S-phase related review on the nuclear assembly (Gant and Wilson, genes by recruiting histone deacetylase that repress- 1997), chromatin structure may be the sleeping es transcription affecting chromatin condensation giant of nuclear matrix structure. The decondensa- (Brehm et al., 1998). Emerin, and lamin A/C can tion of mitotically condensed chromatin is thus bind actin (Farley, 1999, Sasseville and Langelier, assumed to be important for, or at least linked to, 1998), a nuclear matrix-associated protein. Other proper nuclear envelope growth (Gant and Wilson, interactions have been recently reported between 1997). It is reasonable that also a proper nuclear HA95, a protein tightly associated with chromatin envelope organization may be essential to the mech- and the nuclear matrix, and a complex constituted anism of controlled chromatin decondensation and by LBR, LAP2β and emerin (Martin et al., 2000). gene expression. Thus, the chromatin provides multiple anchoring site Alterations in either emerin or lamin A/C might for the nuclear envelope, so that the nuclear enve- impair the relationships between the nuclear envelope-associated proteins can modulate the chro- lope and the nuclear matrix-associated actin, thus matin arrangement. affecting the CRC-controlled chromatin deconden- The mechanism of chromatin remodeling by CRC sation (Figure 3). It has been reported in EDMD and involves a PIP2-dependent modulation of the asso- in lamin A-deficient mice that peripheral condensed ciation of the complex to the nuclear matrix through chromatin is altered or absent. Thus it has been the PIP2-responsive actin-regulatory protein hypothesized that a failure to correctly sequester BAF53 (Zhao et al., 1998). Chromatin remodeling transcriptionally inert chromatin at the nuclear is the prerequisite to trigger the release of chromatin periphery might contribute to the pathology of template restriction, allowing specific transcription EDMD by perturbing gene expression.This interpre- factors to bind chromatin itself (Lewin, 1994). A tation is probably misleading, since it implies that specific PI(4,5)P2-binding, actin-regulatory protein, nuclear envelope-associated proteins are involved in the nuclear CapG, is preferentially phosphorylated anchoring gene-poor chromosomes at the nuclear compared with cytosolic CapG (Yu et al., 1990). periphery. This hypothesis, indeed, has been partly Since phosphorylated CapG, in concert with Ca++, contradicted by the finding that in cells lacking can compete with nuclear PI 3-kinase for its emerin, the heterochromatin of chromosome 18 is PI(4,5)P2 substrate, CRC could be modulated in a normally localized (Boyle et al., 2001). A more cor- multiple way by both PLC-β1 and PI 3-kinase rect interpretation takes into account that emerin (Maraldi et al., 2000). The interaction among and lamin A/C are not exclusively localized at the emerin, lamin A/C and nuclear actin can affect the nuclear rim, but, in particular moments of the cell chromatin arrangement that precedes transcription cycle, also within the nucleus (Morris and Manilal, factor-modulated transcriptional activity. The sud- 1999; Hutchison et al., 2001; Maraldi et al., 2002). den transition from an inactive state to an activated At these multiple sites they might interact with actin state, controlled by the nuclear signaling effector at the CRCs which can modulate chromatin arrange- PI(4,5)P2, causes the association of BAF to the ment throughout the nuclear domains. This implies inner nuclear matrix (INM) scaffold. As a conse- that the nuclear matrix supports the structural prop- quence, the solenoid fibers attached to the scaffold erties of the nucleus and accounts for modifications by MAR sequences loose their compact arrangement in gene expression associated with differentiation and become accessible to histone-modifying necessary to sustain phenotypic requirements in spe- enzymes,TFs and RNA polymerase. In fact, many of cialized cells. Therefore, an altered chromatin the enzymes and complexes required for chromatin remodeling could be particularly effective during the remodeling and gene activation are present at first phases of cell differentiation, being the sudden nuclear matrix-associated sites, which also contain chromatin decondensation induced by environmental

review factors a basic mechanism that precedes more sub- also at the cytoplasm, while it is targeted to the NE tle modulations of chromatin arrangements through in developed muscle fibers, suggesting a differential histone acetylation, methylation and phosphoryla- role of the protein in different moments of the mus- tion (Berger,2000) that allow the binding of cell lin- cular phenotype acquisition (Lattanzi et al., 2000). eage-specific transcription factors. Accordingly, in adult muscle, a deficiency in emerin expression in satellite cells could impair regeneration Involvement of nuclear envelope-associated pro- or mass development of the muscle. teins in myogenic differentiation These findings, as well as the restriction of EDMD In different tissues, the presence or absence of dif- defects to defined skeletal districts, suggest that ferent lamins and/or lamin-associated proteins, interaction with other proteins (transcription fac- could result in different phenotypical effects. tors, chromatin-binding proteins, muscle-specific Abnormal interactions with cytoskeletal or sarcom- cytoskeletal proteins) can account for multiple role eric proteins might be responsible of skeletal and of emerin in muscle development (Lin et al., 2000). cardiac myocyte abnormalities (Hegele, 2000). During myogenesis emerin may constitute a network Indeed, mAKAP, an A kinase-associated protein, is between the cytoplasm and the nucleus allowing the located exclusively at the inner nuclear matrix of coordination of signal transduction, as hypothesized skeletal and cardiac muscle cells, so that PKA local- for lamin B (Georgatos and Blobel, 1987). The ization and function should be altered by a damaged organization of the cytoplasmic array of myofila- lamina (Hutchison et al., 2001; Kapiloff et al., ments is under the control of the intermediate fila- 1999). ments desmin, vimentin, nestin (Capetanaky et al., The organization and genesis of skeletal muscle 1997; Li et al., 1997;Vaittinen et al., 1999) and the phenotype appears to be unique for the following correct location of the multiple nuclei may require a characteristics: presence of multiple nuclei whose close connection between nucleoskeletal and influence on the fiber development may be strictly cytoskeletal structures. Once cell differentiation is coordinated by signaling molecules; accumulation of achieved, emerin is no longer expressed at the cyto- metabolic defects in non-proliferating terminal cells; plasmic level but targeted exclusively to the NE, at possibility that the same proteins play different roles least in skeletal myofibers. The absence of emerin in different moments of the complex differentiation gives rise to apparently normal muscle cells, in pattern of the muscle cells (Yun and Wold, 1996). which, during the initial steps of myogenesis, its role The fusion of satellite cells to form myotubes is partly sustained by other differentiating factors. involves re-entry into the cell cycle (Schultz, 1996). However, in the adult, the absence of emerin in the Both emerin (Lattanzi et al., 2000) and lamin A/C satellite cells may result in the impossibility of a (Lourim and Lin, 1989) have been reported to be physiological regeneration by transformation of involved into the myogenic differentiation. Emerin satellite cells into myotubes, thus causing a progres- and lamin A/C absence or altered expression in sive muscle wasting. EDMD might impair the response to cell cycle-medi- In EDMD severe disfunctions occur at the cardiac ated events in satellite cells. This would imply that conduction system that, after the development of cell cycle-dependent functions of NE-associated pro- arrythmias, can lead to sudden death even in the teins are mainly relevant in satellite cells, or that absence of deep alterations of the contractile car- there is a genetic redundancy in other tissues, which diac cells.This suggests that NE-associated proteins is absent from muscle (Ellis et al., 1998). Evidence altered in EDMD do not alter cardiac muscle con- of the involvement of emerin in myogenic differenti- traction but affect structures that mediate the con- ation has been obtained evaluating the expression of duction pathway.The reported presence of emerin at the protein in cultured myoblasts (Lattanzi et al., intercalated discs in normal heart tissue (Cartegni et 2000). In fact, emerin is present in cycling al., 1997) is suggestive of a possible cytoplasmic myoblasts and, in higher amounts in the myotubes, in role of the protein (Squarzoni et al., 2000). The which also the expression of MyoD is high, and less structure of emerin is consistent with a role of represented in resting mononucleated cells, which molecular targeting in different membrane districts, represent quiescent reserve cells where also MyoD is being the interaction through the LEM-box capable down-regulated (Yoshida et al., 1998). Moreover, of recognizing different proteins belonging to struc- during myogenic differentiation, emerin is expressed tural platforms (Lin et al., 2000). Multi-protein

N.M. Maraldi et al. platforms are considered as the main functional relationships; iii) within the nucleus, located at both module in signal transduction; protein/protein inter- the nuclear envelope and at the inner nuclear matrix, actions through SH2 domains, lipid/protein interac- an autonomous signaling system, based on poly- tions through PH domains are at the basis of signal- phosphoinositides, has been demonstrated to affect ing systems, whose localization in membrane dis- chromatin arrangement through a direct influence tricts could depend on local domains (lipid micro- on actin-binding proteins of the chromatin remodel- domains, caveolae) or by intramembrane target pro- ing complexes. teins. Emerin and lamins may be part of the signal- Future investigations will demonstrate whether ing system, mainly, but not exclusively, located at the mutations in lamin A/C and emerin can selectively nuclear level. Alterations of these proteins in partic- affect chromatin arrangement and gene expression ular districts could be responsible for the different of the myogenic lineage during differentiation and/or phenotypical alterations that occur in mesenchymal repair. cells in EDMD (skeletal muscle cells, cardiomyocytes), especially during the development. Acknowledgements Interestingly, during muscle cell differentiation from This work was supported by grants: EU project myoblast to myotube, lamin A/C amount increases at Myo-Cluster, contract n° GLG1-CT-1999-00870; the nuclear lamina level (Chaly et al., 1996) in cor- Ministero della Sanità, Italy: PF n° ICS 080.1/ respondence with changes in the peripheral chro- RF97.67 and ICS080.1/RF.99.44; Fondazione matin arrangement. Carisbo PB “Diagnosi, ricerca e trattamento nelle distrofie muscolari” 1999/2000, Bologna, Italy. Conclusions The identification of inherited distinct diseases that are caused by mutations in lamins and lamin- References associated proteins is one of the most intriguing findings not only in the pathophysiology of these diseases Amankwah KS, and Boni U. Ultrastructural localization of filamentous actin within neuronal interphase nuclei in situ. Exp Cell Res 1994; but also in that of basic cell biology. In fact, given 210:315-25. the precedents of seemingly-unrelated diseases, such Belgrader P,Dey R, Berezney R. Molecular cloning of matrin 3. A 125- kilodalton protein of the nuclear matrix contains an extensive acidic as EDMD and Dunnigan lipodystrophy, other nuclear domain. J Biol Chem 1991;266:9893-9. envelopathies are likely to emerge. On the other Berezney R, Mortillaro MJ, Ma H, Wei X, Samarabandu J.The nuclear matrix: A structural milieu for genomic functions. Int Rev Cytol hand, despite the progress in molecular genetic stud- 1995; 162A:1-65. ies on the nuclear envelope proteins, the mechanism Berger SL. Local or global? Nature 2000;408:412-5. by which mutations in emerin or in lamin A/C selec- Bione S, Maestrini E, Rivella S, Mancini M, Regis S, Toniolo D. Identification of a novel X-linked gene responsible for Emery-Dreifuss tively cause damages in skeletal and cardiac muscles muscular dystrophy. Nat Genet 1994;8:323-7. has not been clarified.This indicates that our knowl- Bonne G, Raffaele-DiBarletta MR, Varnous S, Becane HM, Hammouda edge about the role of the nuclear envelope-associ- EH, Merlini L, et al. Mutations in the gene encoding lamin A/C cause autosomal dominant Emery-Dreifuss muscular dystrophy. Nat Genet ated proteins on the functional activities of the 1999;21:285-8. nuclear domains is still unsatisfactory. Provided that Boulikas T. Chromatin domains and predictions of MAR sequences. Int Rev Cytol 1995;162A:279-388. the mechanical stress hypothesis is not completely Boyle S, Gilchrist S, Bridger JM, Mahy NL, Ellis JA, Bickmore WA.The convincing, the possibility that gene expression could spatial organization of human chromosomes within the nuclei of nor- be affected by mutations in lamin A/C and emerin mal and Emerin-mutant cells. Hum Mol Genet 2001;10:211-9. Brehm A, Miska EA, McCance DJ, Reid JL, Bannister AJ, Kouzarides T. appears plausible and testable. Retinoblastoma protein recruits histone deacetylase to repress tran- Future prospects might take into account some scription. Nature 1998;391:597-601. Cao H, Hegele RA. Nuclear lamin A/C R482Q mutation in Canadian kin- experimental findings that are emerging in recent dreds with Dunnigan-type familial partial lipodystrophy. Hum Mol years: i) gene expression regulation occurs in steps, Genet 2000;9:109-12. Capetanaky Y, Milner DJ, Weitzer G. Desmin in muscle formation and being the chromatin decondensation a prerequisite maintenance: knockouts and consequences. Cell Struct Funct 1997; with respect to transcription factor binding to gene 22:103-16. promoters; ii) chromatin arrangement is modulated Cartegni L, Raffaele-DiBarletta MR, Barresi R, Squarzoni S, Sabatelli P, Maraldi NM, et al. Heart-specific localisation of emerin: new by chromatin remodeling complexes, macromolecu- insights into Emery-Dreifuss muscular dystrophy. Hum Mol Genet lar aggregates that respond to signals and utilize 1997;6:2257-64. Chaly N, Munro SB, Swallow MA. Remodelling of the nuclear periphery actin and actin-associated proteins of the nuclear during muscle cell differentiation in vitro. J Cell Biochem 1996;62: matrix to modify the chromatin loop-nuclear matrix 76-89. Chen YW, Zhao P,Boup R, Hoffman EP.Expression profiling in the mus-

review cular dystrophies: identification of novel aspects of molecular patho- sue-specific disease phenotypes? J Cell Sci 2001;114:9-19. physiology. J Cell Biol 2000;151:1321-36. Irvine R. Nuclear lipid signaling. Science 2000;48:1-12. Chi TH, Crabtree GR. Inositol phosphates in the nucleus. Science 2000; Kadonaga JT. Eukaryotic transcription: an interlaced network of tran- 287:1937-8. scription factors and chromatin-modifying machines. Cell 1998;92: Cocco L, Martelli AM, Gilmour RS, Rhee SG, Manzoli FA. Nuclear 307-13. phospolipase C and signaling. Biochem Biopys Acta 2001;1530:1- Kapiloff MS, Schillace RV, Westphal AM, Scott JD. mAKAP: an A- 14. kinase anchoring protein targeted to the nuclear membrane of differ- Cohen M, Lee KK, Wilson KL, Gruenbaum Y.Transcriptional repression, entiated myocytes. J Cell Sci 1999;112:2725-36. apoptosis, human disease and the functional evolution of the nuclear Kingston HE, Bunker CA, Imbalzano AN. Repression and activation by lamina.Trends Biochem Sci 2001;26:41-7. multiprotein complexes that alter chromatin structure. Genes Dev DeSandre-Giovannoli A, Chaouch M, Kozlov S, Vallat J, Tazir M, 1996;10:905-20. Kassouri N, et al. Homozygous defects in LMNA, encoding lamin A/C Kruhlak MJ, Lever MA, Fischle W, Verdin E, Bazerr-Jones DP, Hendzel nuclear-envelope proteins, cause autosomal recessive axonal neu- M. Reduced mobility of the ASF splicing factor through the nucleo- ropathy in human (Charcot-Marie-Tooth Disorder Type 2) and mouse. plasm and steady-state speckle compartments. J Cell Biol 2000; 150: Am J Hum Genet 2002;70:726-33. 41-51. D’Santos CS, Clarke JH, Divecha N. Phospholipid signaling in the nucle- Lamond AI,Earnshaw WC. Structure and function in the nucleus. us. Biochem Biopys Acta 1998;1436:201-32. Science 1998;280:547-53. Djabali K, Portier MM, Gros F, Blobel G, Georgatos SD. Network anti- Lattanzi G, Ognibene A, Sabatelli P, Capanni C, Toniolo D, Columbaro bodies identify nuclear lamin B as a physiological attachment site for M, et al. Emerin expression at the early stages of myogenic differen- peripheral intermediate filaments. Cell 1991;64:109-21. tiation. Different 2000;66:208-17. Dreger M, Otto H, Neubauer G, Mann M, Hucho F. Identification of Lewin B. Chromatin and gene expression: constant questions, but chang- β phosphorylation sites in native lamina-associated polypeptide 2 . ing answers. Cell 1994;70:397-406. Biochem 1999;39:9426-34. Lewis JD, Tollervey D. Like attracts like: Getting RNA processing Dreyfuss G, Struhl K. Nucleus and gene expression: Multiprotein com- together in the nucleus. Science 2000;288:1385-9. plexes, mechanistic connections and nuclear organization. Curr Opin Cell Biol 1999;11:303-6. Li Z, Mericskay M, Agbulut O, Bulter-Browne G, Carlsson L,Thornell L, et al. Desmin is essential for the tensile strength and integrity of Dyck J, Maul GG, Miller WH, Chen JD, Kakizuka A, Evans RM. A novel myofibrils but not for myogenic commitment, differentiation, and macromolecular structure is a target of the promyelocyte-retinoic fusion of skeletal muscle. J Cell Biol 1997;139:129-44. acid receptor oncoprotein. Cell 1994;76:333-43. Lin F,Blake DL, Callebaut I, Skerjanc IS, Holmer L, McBurney MW, et Ellis JA, Craxton M, Yates JRW, Kendrik-Jones J. Aberrant intracellu- al. MAN1, an inner nuclear membrane protein that shares the LEM lar targeting and cell cycle-dependant phosphorylation of emerin con- domain with lamina-associated polypeptide 2 and emerin. J Biol tribute to the Emery-Dreifuss muscular dystrophy phenotype. J Cell Chem 2000;275:4840-7. Sci 1998;111:781-92. Lurim D, Lin JJC. Expression of nuclear lamin A and muscle-specific Emery AEH. Emery-Dreifuss muscular dystrophy-a 40 year retrospec- proteins in differentiating muscle cells in ovo and in vitro. J Cell Biol tive. Neuromusc Disord 2000;10:228-32. 1989; 109:495-504. Emery AEH. Emery-Dreifuss syndrome. J Med Genet 1989;26:637-41. Mancini MA, Shan B, Nickerson JA, Penman S, Lee WH.The retinoblas- Fairley EAL, Kendrick-Jones J, Ellis JA. The Emery-Dreifuss muscular toma gene product is a cell cycle-dependent, nuclear matrix-associat- dystrophy phenotype arises from aberrant targeting and binding of ed protein. Proc Natl Acad Sci USA 1994;91:418-22. emerin to the inner nuclear membrane. J Cell Sci 1999;112:2571- Manilal S, Nguyen TM, Sewry CA, Morris GE.The Emery-Dreifuss mus- 82. cular dystrophy protein emerin, is a nuclear membrane protein. Hum Fey EG, Bangs C, Sparks C, Odgren P. The nuclear matrix: defining Mol Genet 1996;5:801-8. structural and functional roles. Eukar Gene Exp 1991;1:127-43. Manilal S, Sewry CA, Pereboev A, Man N, Gobbi P, Hawkes S, et al. Furukawa K, Fritze CE, Gerace L.The major nuclear envelope targeting Distribution of emerin and lamins in the heart and implications for domain of LAP2 coincides with its lamina binding region but is dis- Emery-Dreifuss muscular dystrophy. Hum Mol Genet 1999;8:353-9. tinct from its chromatin interaction domain. J Biol Chem Maniotis AJ, Chen CS, Ingber DE. Demonstration of mechanical con- 1998;273:4213-9. nections between integrins, cytoskeletal filaments, and nucleoplasm Gant TM, Wilson KL. Nuclear assembly. Annu Rev Cell Dev Biol 1997; that stabilize nuclear structure. Proc Natl Acad Sci USA 1997;94: 13:669-95. 849-54. Genshel J, Schmidt HHJ. Mutations in the LMNA gene encoding lamin Maraldi NM, Capitani S, Cinti C, Neri LM, Santi S, Squarzoni S, et al. A/C. Human Mut 2000;1:6451-9. Chromosome spread for confocal microscopy. Methods Enzymol Georgatos SD, Blobel G. Lamin B constitutes an intermediate filament 1999; 307:190-207. attachment site of the nuclear envelope. J Cell Biol 1987;105:117- Maraldi NM, Zini N, Santi S, Manzoli FA.Topology of inositol lipid sig- 25. nal transduction in the nucleus. J Cell Physiol 1999b;181:203-17. Goldberg M, Harel A, Brandeis M, Rechsteiner T, Richmond TJ, Weiss Maraldi NM, Zini N, Santi S, Riccio M, Falconi M, Capitani S, et al. AM, et al. The tail domain of lamin DmO binds histones H2A and Nuclear domains involved in inositol lipid signal transduction. Advan H2B. Proc Natl Acad Sci USA 1999;96:2852-7. Enzyme Regul 2000;40:219-53. Gotzmann J, Foisner R. Lamins and lamin-binding proteins in function- Maraldi NM, Zini N, Santi S, Ognibene A, Rizzoli R, Mazzotti G, et al. al chromatin organization. Crit Rev Eukar Gene Exp 1999;9:257-65. Cytochemistry of the functional domains of the nucleus in normal and Gruenbaum Y, Wilson KL, Harel A, Goldberg M, Cohen M. Review: in pathologic conditions. Eur J Histochem 1998;42:41-53. Nuclear lamins-structural proteins with fundamental functions. J Maraldi NM, Lattanzi G, Sabatelli P, Ognibene A, Squarzoni S. Struct Biol 2000;129:313-23. Functional domains of the nucleus: implications for Emery-Dreifuss Hagmann M. How chromatin changes its shape. Science 1999;285: muscular dystrophy. Neuromuscular Disorders 2002; 12:815-23. 1200-3. Martelli AM, Bareggi R, Bortul R, Grill V, Narducci P, Zweier M. The He DC, Zeng C, Brinkley BR. Nuclear matrix proteins as structural and nuclear matrix and apoptosis. Histochem Cell Biol 1997;108:1-10. functional components of the mitotic apparatus. Int Rev Cytol 1995; Martins SB, Eide T, Steen RL, Jahnsen T, Skalhegg BS, Collas P.HA95 162B:1-74. is a protein of the chromatin and nuclear matrix regulating nuclear Hegele RA. The nuclear envelope, please: Nuclear lamins and disease. envelope dynamics. J Cell Sci 2000;113:3703-13. Nat Med 2000; 6:136-7. Mintz PJ, Patterson SD, Neuwald AF, Spahr CS, Spector DL. Herrmann H, Aebi U. Intermediate filaments and their associates: Multi- Purification and biochemical characterization of interchromatin gra- talented structural elements specifying cytoarchitecture and cytody- nule clusters. EMBO J 1999;18:4308-20. namics. Curr Opin Cell Biol 2000;12:79-90. Misteli T. Cell biology of transcription and pre-mRNA splicing: nuclear Hutchison CJ, Alvarez-Reyes M, Vaughan OA. Lamins in disease: Why architecture meets nuclear function. J Cell Sci 2000;113:1841-9. do ubiquitously expressed nuclear envelope proteins give rise to tis- Moir RD, Spann TP.The structure and function of nuclear lamins: impli-

N.M. Maraldi et al. cation for disease. Cell Mol Life Sci 2001;58:1748-57. Stuurman N, Heins S, Aebi U. Nuclear lamins: their structure, assembly Moir RD, Spann TP, Goldman RD. The dynamic and possible functions and interactions. J Struct Biol 1998;122:42-66. of nuclear lamins. Int Rev Cytol 1995;162B:141-82. Sullivan T, Escalante-Alcalde D, Bhatt H, Anver M, Bhat N, Nagashima Morris G, Manilal S. Heart to heart: from nuclear proteins to Emery- K, et al. Loss of A-type lamin expression compromises nuclear enve- Dreifuss muscular dystrophy. Hum Mol Gen 1999;10:1847-51. lope integrity leading to muscular dystrophy. J Cell Biol Morris GE. Nuclear proteins and cell death in inherited neuromuscular 1999;147:913-20. disease. Neuromusc Dis 2000;10:217-27. Tait D, Riccio M, Sittler A, Scherzinger E, Santi S, Ognibene A, et al. Nagano A, Arahata K. Nuclear envelope proteins and associated dis- Ataxin-3 is transported into the nucleus and associates with the eases. Curr Opin Neurol 2000;13:533-9. nuclear matrix. Hum Mol Genet 1998;7:991-7. Nagano A, Koga R, Ogawa M, Kurano Y, Kawada J, Okada R, et al. Tews DS. Emerin. Int J Bioch Cell Biol 1999;31:891-4. Emerin deficiency at the nuclear membrane in patients with Emery- Thomson TG, Chan YM, Hack AA, Brosius M, Rajala M, Lidov HGW, et Dreifuss muscular dystrophy. Nat Genet 1996;12:254-9. al. Filamin 2 (FLN2): A muscle-specific sarcoglycan interacting pro- Neri LM, Raymond Y, Giordano A, Capitani S, Martelli AM. Lamin A is tein. J Cell Biol 2000;148:115-26. part of the internal nucleoskeleton of human erythroleukemia cells. J Tsuchiya Y, Hase A, Ogawa M, Yorifuji A, Arahata K. Distinct regions Cell Physiol 1999;178:284-95. specify the nuclear membrane targeting of emerin, the responsible Nickerson JA. Experimental observations of a nuclear matrix. J Cell Sci protein for Emery-Dreifuss muscular dystrophy. Eur J Biochem 2001;114:463-74. 1999;259: 859-65. Novelli G, Muchir A, Sangiuolo F, Helbling-Leclerc A, D’Apice MR, Vaittinen S, Lukka R, Sahlgren C, Rantanen J, Hurme T, Lendhal U, ert Massart C, et al. Mandibuloacral dysplasia is caused by a mutation in al. Specific and innervation-regulated expression of intermediate fila- LMNA encoding lamin A/C. Am J hum genet 2002; 71:426-31. ment protein nestin at neuromuscular and myotendinous junctions in Ognibene A, Sabatelli P,Petrini S, Squarzoni S, Riccio M, Santi S, et al. skeletal muscle. Am J Pathol 1999;154:591-600. Nuclear changes in a case of X-linked Emery-Dreifuss muscular dys- Visser AE, Jaunin F,Fakan S, Aten JA. High resolution analysis of inter- trophy. Muscle Nerve 1999;22:864-9. phase chromosome domains. J Cell Sci 2000;113:2585-93. Olson MDJ, Dundr M, Szebeni A. The nucleolus: an old factory with Wang W, Chi T, Xue Y, Zhou S, Kuo A, Crabtree GR. Architectural DNA unexpected capabilities.Trends Cell Biol 2000;10:189-96. binding by a high-mobility-group/kinesin-like subunit in mammalian Östlund C, Ellenberg J, Hallberg E, Lippincott-Schwartz J,Worman HJ. SWI/SNF-related complexes. Proc Natl Acad Sci USA Intracellular trafficking of emerin, the Emery-Dreifuss muscular dys- 1998;95:492-8. trophy protein. J. Cell Sci 1999;112:1709-19. Wang W, Xue Y, Zhou S, Kuo A, Cairns B, Crabtree GR. Diversity and Phair RD, Misteli T. High mobility of proteins in the nucleus of mam- specialization of mammalian SWI/SNF complexes. Genes Dev 1996; malian cells. Nature 2000;404:604-9. 10:2117-30. Procop DJ. Marrow stromal cells for nonhematopoietic tissues. Science Wente SR. Gatekeepers of the nucleus. Science 2000;288:1374-9. 1997;276:71-4. Wilson KL. The nuclear envelope, muscular dystrophy and gene expres- Rando TA. The dystrophin-glycoprotein complex, cellular signaling, and sion.Trends Cell Biol 2000;10:125-9. the regulation of cell survival in the muscular dystrophies. Muscle Nerve 2002;24:1575-94. Ye Q,Worman HJ. Interaction between an integral protein of the nuclear Rando OJ, Zhao K, Crabtree GR. Searching for a function for nuclear envelope inner membrane and human chromodomain proteins homol- actin.Trends Cell Biol 2000;10:92-7. ogous to Drosophila HP1. J Biol Chem 1996;271:14653-6. Razani B, Schlegel A, Lisanti MP. Caveolin proteins in signaling, onco- Yoshida M, Hama H, Ishikawa-Sakurai M, Iamamura M, Mizuno Y, genic transformation and muscular dystrophy. J Cell Sci Araishi K, et al. Biochemical evidence for association of dystrobrevin 2000;13:2103-9. with the sarcoglycan-sarcospan complex as a basis for understanding sarcoglycanopathy. Hum Mol Genet 2000;9:1033-40. Sakaki M, Koike H, Takahashi N, Sasagawa N, Tomioka S, Arahata K, et al. Interaction between emerin and nuclear lamins. J Biochem Yoshida N, Yoshida S, Koishi K, Masuda K, Nabeshima Y. Cell hetero- 2001;129:321-7. geneity upon myogenic differentiation: down-regulation of MyoD and Sasseville AMJ, Langener Y. In vitro interaction of the carboxy-terminal Myf-5 generates ‘reserve cells’. J Cell Sci 1998;111:769-79. domain of lamin A with actin. FEBS Lett 1998;425:485-9. Yu FX, Johnston PA, Sudhof TC, Yin HL. gCap39, a calcium ion and Shopland LS, Lawrence JB. Seeking common ground in nuclear com- polyphosphoinositide-regulated actin capping protein. Science 1990; plexity. J Cell Biol 2000;150:F1-F4. 250:1413-5. Shultz E. Satellite cell proliferative compartments in growing skeletal Yun K, Wold B. Skeletal muscle determination and differentiation. story muscles. Dev Biol 1996;17:84-94. of a core regulatory network and its context. Curr Opin Cell Biol 1996;8:877-89. Skinner PJ, Koshy BT, Cummings CJ, Klement IA, Helin K, Servadio A, et al. Ataxin-1 with an expanded glutamine tract alters nuclear Zhao K, Wang W, Rando OJ, Xue Y, Swiderek K, Kuo A, Crabtree GR. matrix-associated structures. Nature 1997;389:971-4. Rapid and phosphoinositol-dependent binding of the SWI/SNF-like Squarzoni S, Sabatelli P, Capanni C, Petrini S, Ognibene A, Toniolo D, BAF complex to chromatin after T lymphocyte receptor signaling. Cobianchi F, et al. Emerin presence in platelets. Acta Neuropathol Cell 1998; 95:625-36. 2000; 100:291-8. Zini N, Santi S, Ognibene A, Bavelloni A, Neri LM, Valmori A, et al. Squarzoni S, Sabatelli P, Ognibene A, Toniolo D, Cartegni L, Cobianchi Discrete localization of different DNA topoisomerases in HeLa and F, et al. Immunocytochemical detection of emerin within the nuclear K562 cell nuclei and subnuclear fractions. Exp Cell Res matrix. Neuromusc Dis 1998;8:338-44. 1994;210:336-48. Stein GS, van Wijnen AJ, Stein JL, Lian JB, Montecino M, Choi J, et al. Zink D, Bornfleth H, Visser AC, Cremer C, Cremer T. Organization of Intranuclear traffiking of transcription factors: Implications for bio- early and late replicating DNA in human chromosome territories. Exp logical control. J Cell Sci 2000;113:2527-33. Cell Res 1999;247:176-88.