sign in sign up
<<
Home , Lamin

Research Article 61 Myne-1, a repeat transmembrane of the myocyte inner nuclear membrane, interacts with lamin A/C

John M. K. Mislow1, Marian S. Kim2, Dawn Belt Davis1 and Elizabeth M. McNally2,3,* 1Department of Pathology, The University of Chicago, Chicago, IL 60637, USA 2Department of Medicine, Section of Cardiology, The University of Chicago, Chicago, IL 60637, USA 3Department of Human Genetics, The University of Chicago, Chicago, IL 60637, USA *Author for correspondence (e-mail: [email protected])

Accepted 26 September 2001 Journal of Science 115, 61-70 (2002) © The Company of Biologists Ltd

Summary Mutations in the genes encoding the inner nuclear concomitant with lamin A/C expression. In mature muscle, membrane lamin A/C and emerin produce myne-1 and lamin A/C are perfectly colocalized, although and in humans and colocalization with emerin is only partial. Moreover, we mice. The mechanism by which these broadly expressed show that myne-1 and lamin A/C coimmunoprecipitate gene products result in tissue-specific dysfunction is not from differentiated muscle in vitro. The muscle-specific known. We have identified a protein of the inner nuclear inner expression of myne-1, along with its membrane that is highly expressed in striated and smooth interaction with lamin A/C, indicates that this gene is muscle. This protein, myne-1 (myocyte nuclear envelope), a potential mediator of cardiomyopathy and muscular is predicted to have seven spectrin repeats, an interrupted dystrophy. LEM domain and a single transmembrane domain at its C-terminus. We found that myne-1 is expressed upon Key words: Spectrin repeat, Nuclear membrane, Lamin A/C, early muscle differentiation in multiple intranuclear foci Transmembrane protein

Introduction meshwork that interacts with nuclear membrane receptors and Spectrin repeats are units of approximately 106 residues, with associated proteins (Gruenbaum et al., 2000; Stuurman et al., a characteristic structure that includes three α-helices separated 1998). Nuclear lamins display a typical intermediate filament by two loop regions, forming a triple-helical bundle (Pascual domain profile, including a globular N-terminal head, a central et al., 1996; Pascual et al., 1997; Yan et al., 1993). Spectrin rod domain and a C-terminal globular tail. Lamins are repeat (SR)-containing proteins provide critical structural classified as type A or type B, depending on sequence, roles, and mutations in SR-protein genes have been implicated expression pattern and mitotic behavior. Lamins A and C, in human disease. For example, mutations in the β-spectrin alternatively spliced products of the same precursor gene, are gene are associated with hereditary elliptocytosis (Delaunay, expressed in differentiated cells and tissues, whereas type B 1995). β-spectrin contains 17 SRs within its rod region and is lamins are constitutively expressed within all embryonic and a major contributor to the stability of the erythrocyte somatic tissues (Broers et al., 1997). In addition, during the membrane, where it interacts with an array of proteins mitotic disassembly of the nuclear membrane, A-type lamins including ankryin and (Bennett and Gilligan, 1993). solubilize and are dispersed throughout the cell, whereas B- , the protein product of the Duchenne muscular type lamins remain firmly bound to nuclear membrane vesicles dystrophy locus, contains 24 SRs, and mutations that disrupt (Moir et al., 2000a; Moir et al., 2000b). A growing number of dystrophin lead to muscle membrane instability and muscular inner nuclear membrane proteins have been found to interact dystrophy (Anderson and Kunkel, 1992). The SRs of these and with nuclear lamins in vivo and in vitro and appear to regulate other cytoskeletal proteins serve as sites for protein-protein nuclear membrane function and assembly (Clements et al., interaction, actin and crosslinking, and molecular 2000; Dechat et al., 2000; Gant and Wilson, 1997; Martins et scaffolding and stabilization. Additionally, SRs mediate al., 2000). For example, the lamin B receptor (LBR) contains dimerization in α- and β-spectrin, and α- (Djinovic- a predicted eight-transmembrane segment in its C-terminus Carugo et al., 1999). and a nucleoplasmic N-terminal domain that can be In contrast to the plasma membrane, the stability of the phosphorylated by protein kinase A and cdc2 kinase nuclear membrane derives, in part, from an assembly of (Courvalin et al., 1992; Worman et al., 1990). Additionally, intermediate filaments underlying the inner nuclear membrane. members of the LAP2 (Lamin-associated protein 2) family and Lamins are intermediate filament proteins that provide LBR bind to chromatin (Chu et al., 1998; Foisner and Gerace, significant structure to the inner nuclear membrane, forming a 1993; Furukawa et al., 1998; Ye and Worman, 1994). Such 62 Journal of Cell Science 115 (1) chromatin–nuclear-envelope linker proteins may have essential following primers: 5′TACACGAATGGCCCTCCTCC3′ and 5′ACA- roles in the regulation of gene expression. TGGTGCTTGGGAGGGTC3′. The probe was hybridized to a human Interestingly, a handful of structural proteins at the inner multi-tissue mRNA blot (Clontech) and visualized on a Molecular nuclear membrane have been shown to be important for cardiac Dynamics Phosphorimager (Amersham Pharmacia Biotech) and Kodak MS film. Using an EST-derived sequence, a forward primer and skeletal muscle disease (Cohen et al., 2001; Hegele, 2000; ′ ′ Nagano and Arahata, 2000). Mutations in emerin, an X-linked 5 CTCCTTCTCTCGGCGGACAGTGGCGC3 was designed. A full- length clone of myne-1 was generated by using this forward primer 34 kDa protein, produce muscular dystrophy (Bione et al., and the reverse primer 5′CATGTGATCTGGAGGAGGGCTAAA- 1994; Nagano et al., 1996). In , loss of emerin GCTG3′ to amplify a 3644 bp product from human skeletal muscle can lead to cardiac muscle dysfunction but more commonly cDNA (GenBank accession number AF444779). The product was results in aberrant electrical conduction, or heart block, then cloned into pCRII-TOPO vector (Invitrogen) and fully sequenced suggesting that nuclear membrane proteins are specifically to confirm the sequence of myne-1. Predicted amino-acid sequence important for normal function of the cardiac atrio-ventricular alignments were performed using the MacVector (v. 6.0) program node (Funakoshi et al., 1999). Most recently, mutations lamin (Oxford Molecular Group). Prediction of orientation of A/C have been identified in humans with muscular dystrophy, transmembrane helices was performed on the TMHMM server (v. 2.0) cardiomyopathy and heart block, highlighting the importance at http://www.cbs.dtu.dk/services/TMHMM/. Prediction of amino- of nuclear membrane function in normal heart and skeletal acid domain structure was performed on the Simple Modular Architecture Research Tool (SMART) at http://smart.embl- muscle physiology (Bonne et al., 1999; Fatkin et al., 1999). heidelberg.de/ and the ISREC ProfileScan protein domain analysis Missense and truncating mutations in lamin A/C suggest that server at http://www.isrec.isb-sib.ch/ many of these mutations exert a dominant interfering effect (Bonne et al., 1999; Di Barletta et al., 2000). A specific set of mutations that cluster within a small region of exon 8 of lamin Generation of AM1 A cause an unusual adipocyte wasting disorder, Dunnigan’s The peptide, TSGRSTPNRQKTPRGK, representing residues 970- partial lipodystrophy (DPLD) (Cao and Hegele, 2000; 985 of GenBank accession number AB018339 (KIAA0796) was Shackleton et al., 2000; Speckman et al., 2000). Mice with a synthesized and injected into rabbits to raise a polyclonal antisera homozygous null allele of lamin A/C were shown to develop (Zymed Laboratories, South San Francisco, CA). This sequence early lethality resulting from muscular dystrophy and have an shows no significant to the related sequence KIAA1011/DKFZ 434G173 and no significant homology to any other abnormal fat distribution (Sullivan et al., 1999). Lamin A/C is sequence in the available electronic databases. A glutathione s- broadly expressed, yet the mechanism by which lamin A/C transferase (GST) fusion protein expressing a fragment of myne-1 (aa mutations lead to tissue-specific phenotypes is unknown. 979 to 1105) was expressed in E. coli using pGEX4T-1 (Amersham Disruption of tissue-specific protein interactions may explain Pharmacia Biotech) and was used to affinity purify AM1 as described the phenotypes attributed to lamin A/C mutations. (McNally et al., 1996). Toward this end, we have identified a novel SR protein that is expressed primarily in cardiac, skeletal and smooth muscle and associates with lamin A/C. Recently this protein product Immunocytochemistry and immunoblotting Tissues from a C57/BL6 mouse were harvested and frozen in liquid- was identified as interacting with MuSK, a muscle-specific µ tyrosine kinase of the neuromuscular junction, and it was nitrogen-cooled isopentane. Frozen 7 m sections were fixed in –20°C methanol for two minutes, washed twice in PBS and blocked with 5% termed syne-1 for synaptic nuclear envelope (Apel et al., FBS in PBS. Primary antibodies were used at the following dilutions: 2000). Here, we demonstrate that this protein displays a lamin A/C 1:200; emerin 1:250; anti-smooth muscle actin 1:100; and broader expression pattern than first thought, including high AM1 1:50. Each antibody was diluted in blocking solution and levels of expression at the nuclear membrane of smooth, incubated overnight at 4°C. CY3-conjugated goat anti-rabbit and skeletal and cardiac muscle. Therefore, we propose renaming FITC-conjugated goat anti-mouse secondary antibodies (Jackson this protein myne-1 (myocyte nuclear envelope). Myne-1 is Immunochemicals) were used at 1:2000 in blocking solution. Results predicted to be a type II transmembrane protein with a large were visualized using a Zeiss AxioCam digital camera mounted on a cytoplasmic domain containing seven SR domains, an Zeiss Axiophot 50 microscope (Carl Zeiss Inc.). Images were interrupted LEM domain and a C-terminal membrane- recorded and stored using the Zeiss AxioVision digital imaging spanning domain. We found that myne-1 colocalizes software. Immunoblotting was performed as described in (Davis et al., 2000). AM1 was used at 1:200, and HRP-conjugated goat anti-rabbit completely with lamin A/C in all tissue types tested but antibody (Jackson Immunochemicals) was used at 1:5000. ECL plus shows only partial colocalization with emerin. Coimmuno- (Amersham Pharmacia Biotech) and Kodak MS film were used for precipitation studies show an interaction with lamin A/C. Our detection. Anti-smooth-muscle actin monoclonal antibody, 1A4, was data suggest that the myne-1–lamin A/C interaction may be purchased from Sigma (catalogue number A2547). Anti-lamin A/C one mechanism by which lamin A/C mutations exert a tissue- monoclonal antibody, XB10, was obtained from Covance/BAbCo specific effect. (catalogue number MMS-107P). Anti-emerin monoclonal antibody was from Novocastra (catalogue code NCL-EMERIN). DAPI mounting medium was from Vector Laboratories. Anti-LAP2β monoclonal antibody was from BD Transduction Laboratories Materials and Methods (catalogue number L74520). Identification of myne-1 cDNA, northern blot and sequence analysis A partial cDNA for myne-1 was obtained from the Kazusa DNA Nuclear preparation and fractionation Research Institute, Kisarazu, Japan (http://www.kazusa.or.jp/en/). Nuclear membrane preparations were prepared essentially as Northern blot analysis was carried out by amplifying a 432 bp 32P-α- described in (Davis et al., 2000). Briefly, mouse skeletal muscle tissue dCTP cDNA probe amplified from the myne-1 3′UTR region with the was homogenized in PBS then washed in 1×PBS twice at 4°C and Myne-1 associates with lamin-A/C 63 pelleted. Cells were lysed in hypotonic buffer (10 mM Hepes, pH 7.9, washed four times and proteins eluted in SDS sample buffer and run 1.5 mM MgCl2, 10 mM KCl, 0.5 mM DTT plus protease inhibitor on duplicate 4-12% gradient SDS-PAGE. One gel was stained with cocktail (Boehringer Mannheim GmbH., catalogue number Coomassie Blue stain and destained to visualize proteins and the 1873580)), incubated on ice for 15 minutes, vortexed for 30 seconds, second gel was transferred to a PVDF membrane. Membranes were then nuclei were collected at 14,000 g for 10 seconds at 4°C. The light blocked in 3% BSA/TBS-T for one hour before overnight incubation microsomal fraction was collected from the supernatant by with lamin A/C, 1:500. Secondary antibodies were applied and centrifugation (30,000 g for 30 minutes). The heavy microsomes were visualized as described above. subjected to 105,000 g for 30 minutes. The nuclear membrane and the nucleoplasm were separated by resuspending the initial nuclear pellets in high salt buffer (20 mM Hepes, pH=7.9, 25% glycerol, 0.42 M KCl, Results 1.5 mM MgCl2, 0.2 mM EDTA, 0.5 mM DTT plus protease inhibitor Domain structure, mRNA expression and chromosomal cocktail). Nuclear membranes were pelleted at 14,000 g for five localization of myne-1 minutes at 4°C. Equal volumes of each fraction (15 µl) were loaded. Immunoblotting was performed as described in (Davis et al., 2000). A Basic local alignment search tool (BLAST) search (Altschul AM1 was used at 1:200, and HRP-conjugated goat anti-rabbit et al., 1997) using the dystrophin amino-acid sequence yielded antibody (Jackson Immunochemicals) was used at 1:5000. ECL plus a SR-rich Drosophila sequence, kakapo. Kakapo contains 37 (Amersham Pharmacia Biotech) and Kodak MS film were used for SRs and an α-actinin-type actin-binding domain at its N- detection. terminus and is important for attaching muscle to epidermis (Gregory and Brown, 1998; Prokop et al., 1998; Strumpf and Volk, 1998) and for dendritic sprouting of motor neurons In vitro muscle differentiation and immunofluorescence (Prokop et al., 1998). BLAST searches with kakapo SRs Murine muscle C2C12 cells were obtained from American Type identified a novel SR protein encoded by the EST KIAA0796. Culture Collection and maintained at below 70% confluence to avoid A 5′ EST (GenBank AW952703) was aligned with the existing differentiation. Cells were grown on glass coverslips and ′ differentiated by allowing cells to grow to 70% confluence followed KIAA0796 to encode the 5 end. This consensus sequence by serum starvation. Cells were harvested at sequential stages of contains a Kozak sequence and predicted start methionine, differentiation, fixed, and stained and visualized as described above. completing a 4163 bp cDNA sequence with a 3507 bp open The antibodies used are listed above. reading frame. This sequence was recently reported as encoding syne-1 (synaptic nuclear envelope protein-1) (Apel et al., 2000). We propose the name myne-1 (myocyte nuclear Immunoprecipitation of lamin A/C and myne-1 envelope protein-1) as the protein is preferentially expressed in For immunoprecipitations, differentiated C2C12 cells were harvested, striated and smooth muscle (see below). sonicated on ice in immunoprecipitation buffer (10 mM Hepes, pH Using the Simple modular architecture research tool 7.4, 10 mM KCl, 5 mM EDTA, 1% Triton X-100 and protease (SMART) algorithm (Ponting et al., 1999; Schultz et al., 2000; inhibitor cocktail) and then centrifuged at 16,000 g for 15 minutes at 4°C. Lysate was then precleared with Protein A/G sepharose. Schultz et al., 1998) and TMHMM transmembrane helix server Immunoprecipitations with relevant antibodies (diluted 1:75) were (Krogh et al., 2001), we found that the primary structure of performed at room temperature for three hours, followed by myne-1 predicts a 131 kDa protein containing seven SRs, a incubation with Protein A/G sepharose for two hours and central coiled-coil domain and a type II transmembrane domain centrifugation at 4,000 g for 10 minutes. Immune complexes were followed by a short intraluminal C-terminus (Fig. 1A).

Fig. 1. Domain structure of myne-1. (A) The predicted domain structure of myne-1 is shown. It includes seven spectrin repeats (large gray bars) and a transmembrane domain (black bar) at the C-terminus. Between the fifth and sixth spectrin repeat is a region (small gray bars) with homology to the LEM domain found in LAP2, emerin and MAN1. The LEM domain of myne-1 is interrupted in its midportion by an α-helical domain (white bar). Just before the transmembrane domain is a serine-rich domain (hexagon). (B) This region is not present in one splice form of myne. (C) The amino-acid sequence of the interrupted LEM domain of myne-1 is shown. The alignment of the LEM domains of human MAN1, emerin, LAP2, and two C. elegans homologs (F42H11.2/CAEEL; W01G7.5/CAEEL) is shown. Light gray represents homology; dark gray represents identity. 64 Journal of Cell Science 115 (1)

Fig. 3. Subcellular localization of myne-1. Adult mouse skeletal muscle was fractionated and immunoblotted with AM1. The protocol for cell-membrane fractionation is shown in A. The corresponding fractions are shown in lanes 1, 2 and 3. Enrichment of myne-1 is shown in the crude nuclear fraction 3 (lane 3). The protocol for nuclear membrane and nucleoplasm separation is shown in B. The corresponding fractions are shown in lanes 1, 2 and 3. Myne-1 is enriched in the nuclear membrane fraction 3 (lane 3).

Analysis of cDNAs encoding myne-1 reveals that in this region results in the loss of the serine-rich region (Fig. 1B). A BLAST alignment indicates that the first 1020 amino acids of the 1169 amino acid myne-1 share 40% overall homology with the central SR-containing domain of dystrophin. The second and third SRs of myne-1 share 50% overall homology with the C-terminal two SRs found in mAKAP, an A-kinase anchoring protein preferentially expressed in muscle (Kapiloff et al., 1999). Myne-1 residues 595-670 share homology with the LEM domain, a 43 residue Fig. 2. Expression and localization of myne-1. (A) A human multi- region of homology found in LAP2, emerin, and MAN1 (Lin tissue northern blot hybridized to the 3′ UTR of myne-1 is shown. et al., 2000). The LEM domain has been demonstrated to bind Lane 1, heart; lane 2, brain; lane 3, placenta; lane 4, lung; lane 5, to BAF (Shumaker et al., 2001), which in turn binds to liver; lane 6, skeletal muscle; lane 7, kidney; lane 8, pancreas. The chromatin (Zheng et al., 2000). A predicted 20 residue coiled- major transcript is 4.2 kb. (B) An immunoblot from murine tissues using affinity-purified AM1 antibody is shown. G, gastrocnemius; Q, coil domain (Fig. 1C) disrupts the myne-1 LEM domain. Two bipartite nuclear localization signals are found in myne-1 quadriceps; H, heart; B, brain; K, kidney; S, stomach; Bl, bladder. α Loading control panels are shown below. (C) Immunolocalization (amino acids 348-365 and 563-580), located in the last -helix using AM1 in heart and quadriceps muscle shows that myne-1 is at of the third and fifth SRs, respectively. All SRs of myne-1 the nuclear membrane. Double-labeling with DAPI and AM1 is except for the fourth SR are predicted to be acidic (pI 4.6-6.2). shown. The left-hand panels represent nuclei visualized with DAPI The fourth SR is predicted to be basic (pI 9.7). staining. The right-hand panels are stained with AM1. H, heart; S, Using electronic database searches, most of the intron-exon skeletal muscle. Bars represent 20 µm. borders of myne-1 were established (data not shown). The Myne-1 associates with lamin-A/C 65 myne-1 gene maps to human chromosome 6q25.2-6q25.3 and is located close to the marker D6S420. A probe specific to myne-1 was hybridized to multiple mRNAs from human tissues and showed that the prominent myne-1 mRNA was approximately 4.2 kb and was broadly expressed in many tissues, although the highest levels were observed in human heart and skeletal muscle (Fig. 2A). The 4.2 kb transcript represents the major transcript from the gene, but may not be the only transcript since a faint ~9.5 kb mRNA signal was observed throughout all tissues on a human multi-tissue mRNA blot.

Myne-1 is found at the nuclear membrane of skeletal and cardiac muscle To evaluate the protein expression of myne-1, residues 1057-1072 were used as an epitope to generate rabbit antiserum. Affinity-purified AM1 was used to assess protein expression in multiple mouse tissues (Fig. 2B) and showed that myne-1 is a 131 kDa protein that is most abundantly expressed in cardiac, skeletal and smooth muscle. Studies using AM1 on transverse sections of mouse quadriceps and heart, including both atrium and ventricle, showed staining exclusively at the nuclear membrane, identified by the rim surrounding all DAPI-stained nuclei (Fig. 2C). To confirm the nuclear membrane localization of myne-1, we separated the membrane fractions of mouse muscle into heavy microsomes, light microsomes and nuclear membranes. Myne-1 was enriched in the nuclear membrane fraction (Fig. 3A, lane 3). Additionally, muscle nuclei preparations were separated into nuclear membranes and nucleoplasm using high salt extraction. In this preparation, myne-1 was greatly enriched in the nuclear membrane fraction (Fig. 3B, lane 3). In order to examine the tissue localization of myne-1 in more detail, we examined the immunolocalization of myne- 1 in smooth-muscle-containing organs including the stomach (Fig. 4A-D) and bladder (Fig. 4E- H), where staining at the nuclear rim was also seen (red staining). Counterstaining with monoclonal anti-smooth muscle actin (green) Fig. 4. Expression of myne-1 in smooth muscle. A-D represent sections from the identified smooth muscle. Only very faint stomach, whereas E-H are from the bladder. A and E show DAPI-stained nuclei as staining against myne-1 was observed in blue. B and F represent smooth muscle actin (green). C and G show myne-1 staining glandular, epithelial and connective tissues. (red). D represents the merged images from A, B and C. H represents the merged images from E, F and G. Myne-1 is most abundantly expressed in smooth muscle. Bars represent 100 µm. Immunolocalization of myne-1, lamin A/C, and emerin The nuclear distribution of the myne-1 in muscle is similar to transverse sections of mouse quadriceps. We analyzed sections that of emerin and lamin A/C, nuclear envelope proteins that double-labeled with AM1 and antibodies to emerin or lamin organize nuclear architecture and provide structural support to A/C, respectively. In all sections observed, myne-1 and lamin the nuclear envelope (Gruenbaum et al., 2000). To compare A/C displayed an identical coimmunolocalization pattern immunolocalization of myne-1, emerin and lamin A/C, (Fig. 5A-C). Higher magnification showed consistent antibodies to each of these proteins were used to stain coimmunolocalization of myne-1 and lamin A/C (Fig. 5D). 66 Journal of Cell Science 115 (1) Double-labeling with AM1 and emerin antibodies demonstrated that myne-1 and emerin displayed only a partial overlap (Fig. 5E-G). This can be seen under higher magnification (Fig. 5H) that shows exclusion of emerin in the arterial wall nuclei (arrow) where myne-1 is expressed (Fig. 5H). The lower magnification view (Fig. 5G) shows myocyte nuclei that stain for myne-1 but not emerin. Note the scattered red nuclei throughout.

Myne-1 expression and localization during in vitro muscle differentiation The C2C12 cell line, a myoblast cell line capable of in vitro differentiation into myotubes, was used to study the expression of myne-1 during muscle differentiation (Fig. 6A-J). Sparsely plated, undifferentiated cells displayed no detectable levels of myne-1 (Fig. 6B). In contrast, emerin has been shown to be present in undifferentiated C2C12 cells (Fairley et al., 1999). Once cells were placed in differentiation medium, myne-1 expression levels increased (Fig. 6D,F,H,J). Early in differentiation as myoblasts fused to myotubes, myne-1 was located at many distinct multiple foci within nuclei. Later in differentiation, myne-1 became localized to the nuclear membrane of the C2C12 cells (Fig. 6J). We studied colocalization of myne-1 with other known nuclear membrane associated proteins during muscle differentiation. Both myne-1 and lamin A/C display an identical pattern during in vitro differentiation including the localization to intranuclear foci progressing to nuclear membrane localization (Fig. 7A-C). We also studied the localization of myne-1 and LAP2β by double-labeling differentiating C2C12 cells. We similarly found that both myne-1 and LAP2β could be seen in intranuclear foci during in vitro muscle differentiation (Fig. 7D-F).

Myne-1 interaction with A-type lamin Immunolocalization studies demonstrated that myne- 1 displays a nuclear membrane localization identical to lamin A/C. To test the potential interaction Fig. 5. Colocalization of myne-1, lamin A/C and emerin in skeletal muscle. between myne-1 and lamin A/C, myne-1 was Transverse sections of quadriceps muscle were stained with AM1 (A,C,D,E,G immunoprecipitated using the AM1 antibody from and H). Staining with lamin A antibody is shown in B, and double staining fully differentiated C2C12 cells. The precipitates were with AM1 and lamin is shown in C and D. Staining with emerin is shown in F, and double staining with AM1 and emerin is shown in G and H. D shows a separated by electrophoresis and immunoblotted with higher magnification view of the boxed region in C. Similarly, H shows the a monoclonal antibody against lamin A/C (XB10). boxed region in G at higher magnification. Note colocalization of myne-1 and Lamin A/C was found associated with the lamin A appears as yellow staining in C and D. In contrast, in G and H, some immunoprecipitate using AM1 (Fig. 8, right panel, nuclei that express myne-1, but not emerin, are shown as red. The arrow in H lane 4), but was not seen with protein A/G beads alone indicates a cell within the artery wall that expressed myne-1 but not emerin. or in the absence of muscle extracts (Fig. 8, right Low magnification bars represent 100 µm. High magnification bars represent panel, lanes 2 and 3). 25 µm.

nuclear membrane. Protein levels of myne-1 increased during Discussion myocyte differentiation, and the intracellular localization of Myne-1 is a type II transmembrane protein with seven SRs and myne-1 progressed from discrete intranuclear foci to nuclear an interrupted LEM domain found at the nuclear membrane of rim staining during myocyte differentiation. Interestingly, cardiac, skeletal and smooth muscle. Myne-1 colocalized with we found this same developmental pattern for lamin A/C, a lamin A/C, an intermediate filament protein found at the inner nuclear-membrane-associated protein, and for LAP2β, a Myne-1 associates with lamin-A/C 67 involve regulation of the gene expression that accompanies differentiation. Previously, this protein was identified as syne-1 (synaptic nuclear expressed protein-1) (Apel et al., 2000). Because of our findings, we propose renaming this as myne-1 (myocyte nuclear envelope protein-1) to account for its expression pattern and potential interactions outside of postsynaptic nuclei. Apel et al. identified the transmembrane domain as a klarisht-like domain. klarisht is a protein critical for migration of nuclei to the cell periphery in Drosophila. This is consistent with the peripheral localization of nuclei in skeletal myocytes but is inconsistent with the centrally located nuclei of cardiac and smooth muscle. Using the SMART and TMHMM server algorithms, we found that this region is highly likely to encode a type II transmembrane domain. This prediction was tested experimentally by a nuclear membrane extraction performed on mouse muscle where myne-1 was greatly enriched in the nuclear membrane fractions. In addition to SRs and the transmembrane domain, we found that myne-1 contains an interrupted or ‘broken’ LEM domain. The LEM domain is a region of 43 amino acids and is so named for its presence in LAP2, emerin and MAN-1 protein of the inner nuclear membrane (Lin et al., 2000). The function of the LEM domain is not fully understood, but recent data (Shumaker et al., 2001) suggest that this residue is critical for binding BAF, a small molecular weight protein that binds to double stranded DNA (Zheng et al., 2000). Therefore, the LEM domain may be important for crosslinking chromatin to the inner nuclear membrane. The disrupted LEM sequences in myne-1 may function similarly or may have additional functions given to the tissue-specific expression of myne-1. We identified myne-1 by its homology to the SRs of kakapo. SRs are normally associated with a number of cytoskeletal proteins, such as spectrin, dystrophin, , α-actinin, and ACF7, and participate in protein-protein interactions with other cytoskeletal proteins such as actin and zyxin (Amann et al., 1998; Crawford et al., 1992; Rybakova et al., 1996; Rybakova and Ervasti, 1997). Because SR proteins are generally associated with the and the plasma membrane, the nuclear membrane localization of myne-1 is unusual. mAKAP, a protein kinase A anchoring protein Fig. 6. Immunolocalization of myne-1 in differentiating C2C12 cells. targeted to the nuclear membrane of differentiated myocytes, A,C,E,G and I represent nuclei stained with DAPI. AM1 labeling is possesses three SRs, two of which are critical for targeting shown in B,D,F,H and J. UN indicates undifferentiated C2C12 cells. mAKAP to the nuclear membrane (Kapiloff et al., 1999). D1, D3, D5 andD9 indicate 1 day, 3 days, 5 days and 9 days of Downstream regulation of cAMP-dependent proteins such as differentiation, respectively. During differentiation, myne-1 is protein kinase A (PKA) is mediated by anchoring proteins localized to discrete intranuclear foci, whereas late in differentiation, (AKAPs) that sequester PKA to discrete subcellular locations. it becomes localized to the nuclear membrane. The bar represents This compartmentalization is critical for cellular function, as 30 µm. specificity of cAMP-mediated signaling and function is based in a large part on distinct spatial positioning. In the case of nuclear transmembrane protein. In both cases, the intranuclear mAKAP, it is the SRs of the protein that are critical for the foci colocalized with myne-1. Intranuclear foci during myocyte compartmentalization of the protein, targeting mAKAP to the development have previously been described for lamin A/C nuclear membrane (Kapiloff et al., 1999). Like mAKAP, myne- and the nuclear-membrane-associated protein mAKAP 1 may serve as a scaffolding protein for kinases. Using the (Kapiloff et al., 1999; Pugh et al., 1997). Intranuclear foci yeast two-hybrid system, Apel et al., demonstrated that syne- within developing myocytes cells may represent either 1 binds to MuSK, a tyrosine kinase expressed in postsynaptic punctate invaginations or membranous structures within the myocytes (Apel et al., 2000). The in vitro interaction between nucleus. The role of these intranuclear foci during myocyte syne-1 and MuSK occurs in the cytoplasmic domain of MuSK. development is not known, but given the potential role of This cytoplasmic domain contains the tyrosine kinase domain nuclear membrane proteins in chromatin interaction, it may of MuSK (Valenzuela et al., 1995) and demonstrates high 68 Journal of Cell Science 115 (1)

Fig. 7. Immunolocalization of myne-1, lamin A/C and LAP2β in C2C12 cells. (A) Fully differentiated C2C12 stained with lamin A/C is shown. (B) The same cells stained with AM1 demonstrating the coexpression and colocalization of myne-1 and lamin A is shown. (C) A partially differentiated C2C12 culture stained with AM1 is shown; (D) shows the same cells stained with LAP2β demonstrating the coexpression and colocalization of myne-1 and LAP2β. Bars represent 20 µm.

Fig. 8. Co-immunoprecipitation of myne-1 and lamin A/C. Fully differentiated C2C12 cells were lysed and the soluble fraction was processed for immunoprecipitations using myne-1-specific antibody AM1 coupled to Protein A/G sepharose beads. The precipitate was separated by electrophoresis on parallel SDS-PAGE gels. One was stained with Coomassie blue (A), and the second was immunoblotted with a lamin-A/C-specific antibody XB10 (B). Lane 1, total cell lysate; lane 2, lysate precipitated with Protein A/G sepharose beads lacking AM1; lane 3, AM1 immunoprecipitated with Protein A/G sepharose in the absence of lysate; lane 4, lysate immunoprecipitated with AM1 on Protein A/G sepharose beads. homology to many tyrosine kinases when subjected to a are broadly expressed, and the mode by which tissue-specific BLAST search. Thus, syne-1/myne-1 may act as a scaffold for effects develop from mutations in these genes is not known. any number of tyrosine kinases, or serine/threonine kinases, as We hypothesize that mutations in these genes may disrupt its serine-rich C-terminal domain implies. tissue-specific protein interactions. For example, mutations We observed complete colocalization between myne-1 and within exon 8 of lamin A/C are associated with a unique lamin A/C, but only partial overlap between myne-1 and adipocyte-wasting disorder, DLPD (Speckman et al., 2000). It emerin. Although this observation does not exclude the is possible that particular lamin A/C mutations disrupt the possibility of a myne-1–emerin interaction, it indicates that an localization or function of myne-1. Further dissection of the interaction is not necessary for myne-1 and emerin to localize myne-1–lamin-A/C interaction will determine whether there is to the nuclear membrane of cells. Myne-1 and lamin A/C direct binding between these proteins and what other proteins coimmunoprecipitate from muscle extracts, indicating an may participate in the subcortical network of the inner nuclear interaction between the two proteins. At this point, it is not membrane. A better understanding of the cell biology of these known whether the lamin-A/C–myne-1 interaction is a direct interactions may shed light on the tissue-specific mechanisms or indirect interaction. It is possible that other nuclear of pathogenesis in these disorders. membrane or nuclear-membrane-associated proteins participate in or in some way mediate the lamin-A/C-myne-1 interaction. References Mutations in the genes encoding emerin and lamin A/C Altschul, S. F., Madden, T. L., Schaffer, A. A., Zhang, J., Zhang, Z., Miller, specifically alter the phenotype of skeletal muscle, cardiac W. and Lipman, D. J. (1997). Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25, muscle and, in particular, the atrio-ventricular node of the 3389-3402. cardiac conduction system (Becane et al., 2000; Bonne et al., Amann, K. J., Renley, B. A. and Ervasti, J. M. (1998). A cluster of basic 2000; Fatkin et al., 1999; Funakoshi et al., 1999). The repeats in the dystrophin rod domain binds F-actin through an electrostatic phenotypic spectrum of lamin A/C and emerin mutations interaction. J. Biol. Chem. 273, 28419-28423. Anderson, M. S. and Kunkel, L. M. (1992). The molecular and overlaps strikingly and is distinct from a number of other biochemical basis of Duchenne muscular dystrophy. Trends Biochem. Sci. muscular dystrophies that alter genes encoding plasma 17, 289-292. membrane proteins (Hack et al., 2000). Lamin A/C and emerin Apel, E. D., Lewis, R. M., Grady, R. M. and Sanes, J. R. (2000). Syne-1, a Myne-1 associates with lamin-A/C 69

dystrophin- and Klarsicht-related protein associated with synaptic nuclei at cardiomyopathy in Emery-Dreifuss muscular dystrophy. Neuromuscul. the neuromuscular junction. J. Biol. Chem. 275, 31986-31995. Disord. 9, 108-114. Becane, H. M., Bonne, G., Varnous, S., Muchir, A., Ortega, V., Furukawa, K., Fritze, C. E. and Gerace, L. (1998). The major nuclear Hammouda, E. H., Urtizberea, J. A., Lavergne, T., Fardeau, M., envelope targeting domain of LAP2 coincides with its lamin binding region Eymard, B. et al. (2000). High incidence of sudden death with conduction but is distinct from its chromatin interaction domain. J. Biol. Chem. 273, system and myocardial disease due to lamins A and C gene mutation. Pacing 4213-4219. Clin. Electrophysio.l 23, 1661-1666. Gant, T. M. and Wilson, K. L. (1997). Nuclear assembly. Annu. Rev. Cell Bennett, V. and Gilligan, D. M. (1993). The spectrin-based membrane Dev. Biol. 13, 669-695. skeleton and micron-scale organization of the plasma membrane. Annu. Rev. Gregory, S. L. and Brown, N. H. (1998). kakapo, a gene required for adhesion Cell. Biol. 9, 27-66. between and within cell layers in Drosophila, encodes a large cytoskeletal Bione, S., Maestrini, E., Rivella, S., Mancini, M., Regis, S., Romeo, G. and linker protein related to plectin and dystrophin. J. Cell Biol. 143, 1271-1282. Toniolo, D. (1994). Identification of a novel X-linked gene responsible for Gruenbaum, Y., Wilson, K. L., Harel, A., Goldberg, M. and Cohen, M. Emery-Dreifuss muscular dystrophy. Nat. Genet. 8, 323-327. (2000). Nuclear lamins–structural proteins with fundamental functions. J. Bonne, G., Di Barletta, M. R., Varnous, S., Becane, H. M., Hammouda, Struct. Biol. 129, 313-323. E. H., Merlini, L., Muntoni, F., Greenberg, C. R., Gary, F., Urtizberea, Hack, A. A., Groh, M. E. and McNally, E. M. (2000). Sarcoglycans in J. A. et al. (1999). Mutations in the gene encoding lamin A/C cause muscular dystrophy. Microsc. Res. Tech. 48, 167-180. autosomal dominant Emery-Dreifuss muscular dystrophy. Nat. Genet. 21, Hegele, R. A. (2000). The envelope, please: nuclear lamins and disease. Nat. 285-288. Med. 6, 136-137. Bonne, G., Mercuri, E., Muchir, A., Urtizberea, A., Becane, H. M., Recan, Kapiloff, M. S., Schillace, R. V., Westphal, A. M. and Scott, J. D. (1999). D., Merlini, L., Wehnert, M., Boor, R., Reuner, U. et al. (2000). Clinical mAKAP: an A-kinase anchoring protein targeted to the nuclear membrane and molecular genetic spectrum of autosomal dominant Emery-Dreifuss of differentiated myocytes. J. Cell Sci. 112, 2725-2736. muscular dystrophy due to mutations of the lamin A/C gene. Ann. Neurol. Krogh, A., Larsson, B., von Heijne, G. and Sonnhammer, E. L. (2001). 48, 170-180. Predicting transmembrane protein topology with a hidden Markov model: Broers, J. L., Machiels, B. M., Kuijpers, H. J., Smedts, F., van den application to complete genomes. J. Mol. Biol. 305, 567-580. Kieboom, R., Raymond, Y. and Ramaekers, F. C. (1997). A- and B-type Lin, F., Blake, D. L., Callebaut, I., Skerjanc, I. S., Holmer, L., McBurney, lamins are differentially expressed in normal human tissues. Histochem. Cell M. W., Paulin-Levasseur, M. and Worman, H. J. (2000). MAN1, an inner Biol. 107, 505-517. nuclear that shares the LEM domain with lamina- Cao, H. and Hegele, R. A. (2000). Nuclear lamin A/C R482Q mutation in associated polypeptide 2 and emerin. J. Biol. Chem. 275, 4840-4847. canadian kindreds with Dunnigan-type familial partial lipodystrophy. Hum. Martins, S. B., Eide, T., Steen, R. L., Jahnsen, T., Skalhegg, B. S. and Mol. Genet. 9, 109-112. Collas, P. (2000). HA95 is a protein of the chromatin and nuclear matrix Chu, A., Rassadi, R. and Stochaj, U. (1998). Velcro in the nuclear envelope: regulating nuclear envelope dynamics. J. Cell Sci. 113, 3703-3713. LBR and LAPs. FEBS Lett. 441, 165-169. McNally, E. M., Duggan, D., Gorospe, J. R., Bonnemann, C. G., Fanin, Clements, L., Manilal, S., Love, D. R. and Morris, G. E. (2000). Direct M., Pegoraro, E., Lidov, H. G., Noguchi, S., Ozawa, E., Finkel, R. S. et interaction between emerin and lamin A. Biochem. Biophys. Res. Commun. al. (1996). Mutations that disrupt the carboxyl-terminus of gamma- 267, 709-714. sarcoglycan cause muscular dystrophy. Hum. Mol. Genet. 5, 1841-1847. Cohen, M., Lee, K. K., Wilson, K. L. and Gruenbaum, Y. (2001). Moir, R. D., Spann, T. P., Lopez-Soler, R. I., Yoon, M., Goldman, A. E., Transcriptional repression, , human disease and the functional Khuon, S. and Goldman, R. D. (2000a). The dynamics of the nuclear evolution of the . Trends Biochem. Sci. 26, 41-47. lamins during the cell cycle-relationship between structure and function. J. Courvalin, J. C., Segil, N., Blobel, G. and Worman, H. J. (1992). The lamin Struct. Biol. 129, 324-334. B receptor of the inner nuclear membrane undergoes -specific Moir, R. D., Yoon, M., Khuon, S. and Goldman, R. D. (2000b). Nuclear phosphorylation and is a substrate for p34cdc2-type protein kinase. EMBO lamins A and B1: different pathways of assembly during nuclear envelope J. 11, 4027-4036. formation in living cells. J. Cell Biol. 151, 1155-1168. Crawford, A. W., Michelsen, J. W. and Beckerle, M. C. (1992). Nagano, A. and Arahata, K. (2000). Nuclear envelope proteins and An interaction between zyxin and alpha-actinin. J. Cell Biol. 116, 1381- associated diseases. Curr. Opin. Neurol. 13, 533-539. 1393. Nagano, A., Koga, R., Ogawa, M., Kurano, Y., Kawada, J., Okada, R., Davis, D. B., Delmonte, A. J., Ly, C. T. and McNally, E. M. (2000). Hayashi, Y. K., Tsukahara, T. and Arahata, K. (1996). Emerin deficiency Myoferlin, a candidate gene and potential modifier of muscular dystrophy. at the nuclear membrane in patients with Emery-Dreifuss muscular Hum. Mol. Genet. 9, 217-226. dystrophy. Nat. Genet. 12, 254-259. Dechat, T., Korbei, B., Vaughan, O. A., Vlcek, S., Hutchison, C. J. and Pascual, J., Pfuhl, M., Rivas, G., Pastore, A. and Saraste, M. (1996). The Foisner, R. (2000). Lamina-associated polypeptide 2alpha binds spectrin repeat folds into a three-helix bundle in solution. FEBS Lett. 383, intranuclear A-type lamins. J. Cell Sci. 113, 3473-3484. 201-207. Delaunay, J. (1995). Genetic disorders of the red cell membranes. FEBS Lett. Pascual, J., Pfuhl, M., Walther, D., Saraste, M. and Nilges, M. (1997). 369, 34-37. Solution structure of the spectrin repeat: a left-handed antiparallel triple- Di Barletta, M., Ricci, E., Galluzzi, G., Tonali, P., Mora, M., Morandi, L., helical coiled-coil. J. Mol. Biol. 273, 740-751. Romorini, A., Voit, T., Orstavik, K. H., Merlini, L. et al. (2000). Different Ponting, C. P., Schultz, J., Milpetz, F. and Bork, P. (1999). SMART: mutations in the LMNA gene cause autosomal dominant and autosomal identification and annotation of domains from signalling and extracellular recessive Emery-Dreifuss muscular dystrophy. Am. J. Hum. Genet. 66, protein sequences. Nucleic Acids Res. 27, 229-232. 1407-1412. Prokop, A., Uhler, J., Roote, J. and Bate, M. (1998). The kakapo mutation Djinovic-Carugo, K., Young, P., Gautel, M. and Saraste, M. (1999). affects terminal arborization and central dendritic sprouting of Drosophila Structure of the alpha-actinin rod: molecular basis for cross-linking of actin motorneurons. J. Cell Biol. 143, 1283-1294. filaments. Cell 98, 537-546. Pugh, G. E., Coates, P. J., Lane, E. B., Raymond, Y. and Quinlan, R. A. Fairley, E. A., Kendrick-Jones, J. and Ellis, J. A. (1999). The Emery- (1997). Distinct nuclear assembly pathways for lamins A and C lead to their Dreifuss muscular dystrophy phenotype arises from aberrant targeting and increase during quiescence in Swiss 3T3 cells. J. Cell Sci. 110, 2483-2493. binding of emerin at the inner nuclear membrane. J. Cell Sci. 112, 2571- Rybakova, I. N. and Ervasti, J. M. (1997). Dystrophin-glycoprotein complex 2582. is monomeric and stabilizes actin filaments in vitro through a lateral Fatkin, D., MacRae, C., Sasaki, T., Wolff, M. R., Porcu, M., Frenneaux, association. J. Biol. Chem. 272, 28771-28778. M., Atherton, J., Vidaillet, H. J., Spudich, S., De Girolami, U. et al. Rybakova, I. N., Amann, K. J. and Ervasti, J. M. (1996). A new model for (1999). Missense mutations in the rod domain of the lamin A/C gene as the interaction of dystrophin with F-actin. J. Cell Biol. 135, 661-672. causes of and conduction-system disease. N. Engl. Schultz, J., Milpetz, F., Bork, P. and Ponting, C. P. (1998). SMART, a simple J. Med. 341, 1715-1724. modular architecture research tool: identification of signaling domains. Foisner, R. and Gerace, L. (1993). Integral membrane proteins of the nuclear Proc. Natl. Acad. Sci. USA 95, 5857-5864. envelope interact with lamins and chromosomes, and binding is modulated Schultz, J., Copley, R. R., Doerks, T., Ponting, C. P. and Bork, P. (2000). by mitotic phosphorylation. Cell 73, 1267-1279. SMART: a web-based tool for the study of genetically mobile domains. Funakoshi, M., Tsuchiya, Y. and Arahata, K. (1999). Emerin and Nucleic Acids Res. 28, 231-234. 70 Journal of Cell Science 115 (1)

Shackleton, S., Lloyd, D. J., Jackson, S. N., Evans, R., Niermeijer, M. F., lamin expression compromises nuclear envelope integrity leading to Singh, B. M., Schmidt, H., Brabant, G., Kumar, S., Durrington, P. N. et muscular dystrophy. J. Cell Biol. 147, 913-920. al. (2000). LMNA, encoding lamin A/C, is mutated in partial lipodystrophy. Valenzuela, D. M., Stitt, T. N., DiStefano, P. S., Rojas, E., Mattsson, K., Nat. Genet. 24, 153-156. Compton, D. L., Nunez, L., Park, J. S., Stark, J. L., Gies, D. R. et al. Shumaker, D. K., Lee, K. K., Tanhehco, Y. C., Craigie, R. and Wilson, K. (1995). Receptor tyrosine kinase specific for the skeletal muscle lineage: L. (2001). LAP2 binds to BAF small middle dotDNA complexes: expression in embryonic muscle, at the neuromuscular junction, and after requirement for the LEM domain and modulation by variable regions. injury. Neuron 15, 573-584. EMBO J. 20, 1754-1764. Worman, H. J., Evans, C. D. and Blobel, G. (1990). The lamin B receptor Speckman, R. A., Garg, A., Du, F., Bennett, L., Veile, R., Arioglu, E., of the nuclear envelope inner membrane: a polytopic protein with eight Taylor, S. I., Lovett, M. and Bowcock, A. M. (2000). Mutational and potential transmembrane domains. J. Cell Biol. 111, 1535-1542. haplotype analyses of families with familial partial lipodystrophy (Dunnigan Yan, Y., Winograd, E., Viel, A., Cronin, T., Harrison, S. C. and Branton, variety) reveal recurrent missense mutations in the globular C-terminal D. (1993). Crystal structure of the repetitive segments of spectrin. Science domain of lamin A/C. Am. J. Hum. Genet. 66, 1192-1198. 262, 2027-2030. Strumpf, D. and Volk, T. (1998). Kakapo, a novel cytoskeletal-associated Ye, Q. and Worman, H. J. (1994). Primary structure analysis and lamin B protein is essential for the restricted localization of the neuregulin-like and DNA binding of human LBR, an integral protein of the nuclear envelope factor, vein, at the muscle-tendon junction site. J. Cell Biol. 143, 1259-1270. inner membrane. J. Biol. Chem. 269, 11306-11311. Stuurman, N., Heins, S. and Aebi, U. (1998). Nuclear lamins: their structure, Zheng, R., Ghirlando, R., Lee, M. S., Mizuuchi, K., Krause, M. and assembly, and interactions. J. Struct. Biol. 122, 42-66. Craigie, R. (2000). Barrier-to-autointegration factor (BAF) bridges DNA in Sullivan, T., Escalante-Alcalde, D., Bhatt, H., Anver, M., Bhat, N., a discrete, higher-order nucleoprotein complex. Proc. Natl. Acad. Sci. USA Nagashima, K., Stewart, C. L. and Burke, B. (1999). Loss of A-type 97, 8997-9002.