Emerin binding to Btf, a death-promoting transcriptional repressor, is disrupted by a missense mutation that causes Emery–Dreifuss muscular dystrophy Tokuko Haraguchi 1 , James M. Holaska 2 , Miho Yamane 1 , Takako Koujin 1 , Noriyo Hashiguchi 1 , Chie Mori 1 , Katherine L. Wilson 2 and Yasushi Hiraoka 1 1 CREST Research Project, Kansai Advanced Research Center, Communications Research Laboratory, Iwaoka-cho, Nishi-ku, Kobe, Japan; 2 Department of Cell Biology, Johns Hopkins University School of Medicine, Baltimore, MD, USA Loss of functional emerin, a nuclear membrane protein, causes X-linked recessive Emery–Dreifuss muscular dystro- phy. In a yeast two-hybrid screen, we found that emerin interacts with Btf, a death-promoting transcriptional repressor, which is expressed at high levels in skeletal muscle. Biochemical analysis showed that emerin binds Btf with an equilibrium affinity (K D )of100 n M . Using a collection of 21 clustered alanine-substitution mutations in emerin, the resi- dues required for binding to Btf mapped to two regions of emerin that flank its lamin-binding domain. Two disease- causing mutations in emerin, S54F and D95–99, disrupted binding to Btf. The D95–99 mutation was relatively unin- formative, as this mutation also disrupts emerin binding to lamin A and a different transcription repressor named germ cell-less (GCL). In striking contrast, emerin mutant S54F, which binds normally to barrier-to-autointegration factor, lamin A and GCL, selectively disrupted emerin binding to Btf. We localized endogenous Btf in HeLa cells by indi- rect immunoflurorescence using affinity-purified antibodies against Btf. In nonapoptotic HeLa cells Btf was found in dot-like structures throughout the nuclear interior. How- ever, within 3 h after treating cells with Fas antibody to induce apoptosis, the distribution of Btf changed, and Btf concentrated in a distinct zone near the nuclear envelope. These results suggest that Btf localization is regulated by apoptotic signals, and that loss of emerin binding to Btf may be relevant to muscle wasting in Emery–Dreifuss muscular dystrophy. Keywords: apoptosis; emerin; Emery–Dreifuss muscular dystrophy; lamin A; MAN1. The loss of emerin function causes X-linked recessive Emery–Dreifuss muscular dystrophy (EDMD) [1], which affects skeletal muscle, heart and major tendons [2,3]. Emerin binds lamins, including lamin A [4,5]. It was therefore intriguing that dominant forms of EDMD arise in people carrying point mutations in LMNA,which encodes A-type lamins [6]. In a fascinating series of discoveries, mutations distributed throughout LMNA were found to cause seven additional diseases: limb-girdle muscular dystrophy type 1B, dilated cardiomyopathy type 1 A, Dunnigan-type familial partial lipodystrophy (FPLD), an axonal neuropathy known as Charcot–Marie–Tooth disorder type 2B1 [7], a bone development disorder named mandibuloacral dysplasia [8–11], and two accelerated ÔagingÕ diseases named Hutchison–Gilford Progeria Syndrome [12,13] and atypical Werner syndrome [14]. With the possible exception of Charcot–Marie–Tooth disorder type 2B1 disorder, the tissues affected in these Ônuclear lamino- pathyÕ disorders may share a common mesenchymal stem cell lineage [15,16]. The mechanisms underlying these diseases are important to understand, due to their clinical significance and because so little is currently known about nuclear envelope function. To explain the tissue-specificity of Emery–Dreifuss muscular dystrophy, emerin and A-type lamins were proposed to influence tissue-specific gene expression [15,17]. Emerin is a 254-residue integral nuclear membrane protein with an apparent molecular mass of 34 kDa (SDS/PAGE). Emerin is expressed in most but not all tissues that have been tested [1,18–20], and is phosphoryl- ated in a cell-cycle dependent manner [21]. EDMD is diagnosed in childhood by ÔcontracturesÕ of tendons in the neck, ankles, and elbow, along with slowly progressive skeletal muscle wasting, and cardiac conduction defects which can cause sudden death [2,22]. Most X-linked EDMD patients, including those with missense mutations, are null for emerin protein due to degradation of the mutant mRNA or protein. However, a few patients express normal amounts of mutant emerin protein, which is correctly localized at the inner nuclear membrane [23,24]. These special mutations include S54F (Ser54fiPhe), P183H and P183T (Pro183fiHis or Thr), and a five-residue deletion (D95–99). These mutations have the potential to reveal Correspondence to T. Haraguchi, Kansai Advanced Research Center, Communications Research Laboratory, 588-2 Iwaoka, Iwaoka-cho, Nishi-ku, Kobe 651-2492, Japan. Fax: + 81 78 969 2249, Tel.: + 81 78 969 224, E-mail: tokuko@crl.go.jp Abbreviations: EDMD, X-linked recessive Emery–Dreifuss muscular dystrophy; GCL, germ cell-less. (Received 9 December 2003, revised 12 January 2004, accepted 20 January 2004) Eur. J. Biochem. 271, 1035–1045 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2004.04007.x disease mechanisms because the mutant proteins are in the right place (inner nuclear membrane) and their expression levels are normal, yet they cause disease. We hypothesized that these proteins must be defective in one or more activities required for emerin function, such as binding to other proteins at the inner nuclear membrane. Two of emerin’s known binding partners are lamin A and barrier-to-autointegration factor (BAF) [4,5]. The 40-residue ÔLEM-domainÕ of emerin binds directly to BAF [5], and is required for emerin to be recruited to BAF on chromatin during nuclear assembly [24]. Emerin and other ÔLEM-domainÕ proteins such as LAP2b and MAN1 [25] constitute a family of BAF-binding proteins [5,26,27]. The expression of an exogenous mutant BAF (G25E) in HeLa cells disrupts the assembly of endogenous BAF, emerin, LAP2b and lamin A/C (but remarkably, not B-type lamins) into reforming nuclear envelopes [24]. Thus, BAF is predicted to recruit or assemble many if not all LEM- domain proteins and A-type lamins during nuclear forma- tion. We know of no disease-causing mutation in emerin that affects its binding to BAF. However the D95–99 mutation disrupts binding to several binding partners including lamin A [5], transcription regulator germ cell-less (GCL [30]); and splicing factor YT521-B [31]. Two ÔspecialÕ disease-causing mutants, S54F and P183H, bind normally to lamin A, GCL, YT521-B and BAF [5,30,31], suggesting that these mutations disrupt emerin’s binding to undiscov- ered binding partners relevant to disease. We used a two-hybrid screen of a HeLa cell cDNA library to search for novel binding partners of emerin, using full length emerin as bait. This screen produced a positive clone encoding a predicted 920-residue protein, previously reported as Btf [28] or KIAA0164 [29]. Btf can act as a transcriptional repressor, and when overexpressed, Btf induces cell death by a mechanism involving the inhibition of antiapoptotic bcl-2 family proteins [28]. Btf has a wide tissue distribution (including heart, brain, placenta, lung, kidney and pancreas) and is highly expressed in skeletal muscle [29]. Our results show that binding to Btf is specifically and selectively disrupted by the disease-associ- ated S54F missense mutation in emerin. The implications of these findings for possible EDMD disease mechanisms are discussed. Materials and methods Cells and reagents HeLa cells were obtained from the Riken Cell Bank (Tsukuba Science City, Tsukuba, Japan). Hoechst 33342, cycloheximide and anti-Fas monoclonal Ig were from Calbiochem (La Jolla, CA, USA), Wako (Osaka, Japan) and MBL (Nagoya, Japan), respectively. Rabbit poly- clonal serum Ôbtf-middleÕ was prepared by immunizing rabbits with a keyhole limpet hemocyanin (KLH)-con- jugated synthetic peptide (CSERITVKKETQSPEQ- CONH 2 ; with amido modification in the C-terminus) corresponding to residues 485–499 of human Btf. Specific antibodies were affinity-purified by chromatography on NHS-activated Sepharose 4B (Amersham Pharmacia Biotech) coupled to the antigenic peptide as described [32]. For indirect immunofluorescence staining, purified antibody was concentrated on Centricon-10 spin columns (Amicon, MA, USA) and eluted with phosphate-buffered saline (Gibco BRL, USA). Yeast two-hybrid screen Emerin interactor(s) were screened by yeast two-hybrid assay using Matchmaker System III (Clontech Inc.) according to manufacturer instructions. Full length emerin cloned in the pGBK-T7 vector was used as bait. The prey HeLa cDNA library in the pGAD-GH vector was provided by Clontech. The bait plasmid was transformed into Saccharomyces cerevisiae strain Y187, and mated with S. cerevisiae AH109 cells pretransformed with the prey library. Positive clones were selected based on growth in the absenceofaminoacidsTrp,Leu,His,andAde(Ôquadruple dropoutÕ), and screened for b-galactosidase production. For one-to-one two-hybrid analysis, cDNAs encoding full length emerin or emerin fragments were first fused to the GAL4 DNA binding domain in the pGBK plasmid, and then transformed into yeast Y187 cells with lithium acetate. These cells were then mated with yeast AH109 cells that expressed either full length Btf (residues 1–920) or Btf fragments 377–920, 377–761, 377–646, 377–574 and 521– 761 fused to the GAL4 activator domain in the pGAD plasmid. After mating, cells were cultured in YPDA medium (1% yeast extract, 2% peptone, 2% dextrose, 0.003% adenine hemisulfate) for 20 h at 30 °C. Diploid cells that grew in the absence of Trp and Leu were selected, and then plated on quadruple-dropout medium to assay two- hybrid-dependent gene expression. Positives were confirmed by b-galactosidase production. Plasmid construction To fuse emerin, Btf or fragments thereof with the two- hybrid vectors, the desired cDNAs were PCR-amplified using the primers and templates in Tables 1 and 2. PCR products were digested with NdeIandBamHI, and inserted into each vector. To construct BD-lamin A plasmids, the coding regions of lamin A were PCR-amplified using the following primers: 5¢-AAGAATTCATGGAGACCCCGT CCCAG-3¢ and 5¢-GCCGTCGACTTACATGATGCTG CAGTTCTGGGG-3¢. PCR products were digested with EcoRI and SalI, and inserted into the p-GBK vector (Clontech Laboratories, Inc., Palo Alto, CA, USA) using the SalIandBamHI sites in the vector. The DNA sequences of all fusion plasmids were confirmed using an ABI377 DNA sequencer (Applied Biosystems, Norwalk, CT, USA). Microtiter well assay for Btf binding to emerin and affinity measurements Btf protein was synthesized and 35 S-labeled in vitro using coupled transcription/translation extracts (Promega Corp.), as described in detail by Holaska et al.[30].Wildtype emerin residues 1–222, or each mutant emerin, were purified as recombinant proteins and adsorbed to micro- titer wells. Typically, 5–50 pmoles of emerin protein were adsorbed per well. [ 35 S]Btf was placed into wells contain- ing each emerin protein, or into BSA-adsorbed wells as negative controls, and incubated 60–90 min at room 1036 T. Haraguchi et al. (Eur. J. Biochem. 271) Ó FEBS 2004 temperature in binding buffer (BB: 20 m M Hepes pH 7.4, 110 m M potassium acetate, 2 m M magnesium acetate, 1m M EGTA). Wells were washed with 200 lL BB, five times, and bound Btf was eluted with 5% SDS, placed in a scintillation vial, and counted. To assay Btf binding by blot overlay, 35 S-labeled Btf was incubated for 16 h at 4 °C with blots of recombinant human emerin proteins, purified by FPLC as described [30]. Western blotting HeLa cells (1 · 10 7 ) were collected by scraping with a rubber policeman after washing twice with DMEM medium, suspended in homogenizing buffer (20 m M Tris/HCl pH 7.5, 2 m M MgCl 2 ,150m M NaCl) supple- mented with protease inhibitor cocktail (Roche) to a final concentration of 1 · 10 7 cells per mL, and homogenized on ice using a Potter homogenizer. Half of each sample was centrifuged at 1000 g for 3 min to separate the nuclear (pellet) and cytoplamic (supernatant) fractions, and the other half was kept as the Ôtotal lysateÕ fraction. Samples corresponding to 1 · 10 5 cells were loaded per lane on SDS 15% polyacrylamide gels. After electrophor- esis, proteins were transferred to poly(vinylidene difluo- ride) membrane at 60 V for 2 h in transfer buffer (50 m M Tris, pH 7.5, 380 m M glycine, 0.1% (v/v) SDS and 20% (v/v) MeOH). After blocking with 5% (v/v) skim milk in NaCl/P i , the membrane was incubated at 4 °Covernight with primary antibody (against Ôbtf-middleÕ antigen; described above) at a dilution of 1 : 1000 in NaCl/P i containing 0.1% (v/v) skim milk, and 0.1% (v/v) Tween- 20. Blots were then incubated at 4 °Cfor2hwithHRP- conjugated anti-rabbit IgG (Cappel) at a dilution of 1 : 1000 and stained by enhanced chemiluminescence (Amersham). Indirect immunofluorescence staining Cells were fixed in 10% (v/v) trichloroacetic acid for 10 min at room temperature after a brief wash with DMEM (37 °C), and then permeabilized with 0.1% (v/v) Triton X-100 in NaCl/P i for 5 min, washed three times with NaCl/ P i , and finally incubated 1 h with 1% (v/v) BSA in NaCl/P i , all at room temperature. Antibodies against Btf were then added to cells at 1 : 500 dilution, incubated 18 h at 4 °C, washed four times, and stained with Alexa-conjugated secondary antibody (Molecular Probes Inc.) at a dilution of 1 : 1000 for 3–4 h at room temperature. Finally, cells were washed three times with NaCl/P i , and incubated sequentially with 20, 40, 60, and 80% glycerol containing NaCl/P i , 2.5% 1,4-diazabicyclo-2,2,2-octane (DABCOÒ) and 0.5 lgÆmL )1 4¢,6-diamidino-2-phenylindole (DAPI). Cells were mounted in 90% glycerol containing 2.5% DABCOÒ as an antifading reagent. Table 1. Names of PCR primers used in this work. For cloned regions, the numbers represent the first and last amino acids numbers of cloned regions. Cloned regions Template Forward primer Reverse primer Vector emerin-full GFP-emerin H emerin 1 EGFP-emerin-Nde1–2 pGBKT7 104–254 GFP-emerin H-emerin BamHI310 3¢ H-emerin BamHI pGBKT7 164–254 pGBKT7-emerin 164–5¢ 3¢ H-emerin BamHI pGBKT7 104–228 pGBKT7-emerin H-emerin BamHI310 DTM H-emerin BamHI pGBKT7 btf-full KIAA0164 – – pGADT7 377–920 KIAA0164 377–5¢ End-3¢ pGADT7 377–761 a – – – pGADT7 377–646 KIAA0164 377–5¢ 646–3¢ pGADT7 377–574 KIAA0164 377–5¢ 574–3¢ pGADT7 521–761 KIAA0164 521–5¢ 761–3¢ pGADT7 a This plasmid was selected from the screening of the HeLa cDNA library packaged in Matchmaker Systems III (Clontech). Table 2. Nucleotide sequences of PCR primers. Name of the primer DNA sequence of the primer emerin H-emerin 1 TGC ATA TGG ACA ACT ACG CAG ATC H-emerin BamHI310 CGT GGA TCC TCA TGA CTT ATG GGG AGC CCG A 164–5¢ AAC ATA TGA TCA CGC ACT ACC GCC C EGFP-emerin-Nde1–2 TCC ATA TGC TAG AAG GGG TTG CCT 3 H-emerin BamHI GGC GGA TCC CTA GAA GGG GTT GCC TTC TTC DTM H-emerin BamHI GGG GAT CCC TGG CCC CAG AGC GG btf 377–5¢ AAC ATA TGG ATC AGG AAG CTC TAG ATT AC 521–5¢ AAC ATA TGG CAC GAG AAA AGT CTA CCT TC 574–3¢ TTG GAT CCT TAT GTA CTA GCA AGC AGC C 646–3¢ TTG GAT CCT TAT TGC CGA GTA CTA TGT TC 761–3¢ TTG GAT CCT TAG GGA GAA GAA GGT GAT G end-3¢ AAA GAT CTT TAT TCC TTT TCT TCC TTG CG Ó FEBS 2004 Emerin binds Btf, a transcriptional repressor (Eur. J. Biochem. 271) 1037 Fluorescent images were obtained by a DeltaVision microscope system (Applied Precision Inc. Seattle, WA, USA) based on IX70 (Olympus, Tokyo) using an oil immersion objective lens (PlanApo 60, NA ¼ 1.4) and high-selectivity filters. Serial optical section data (15–30 focal planes at 0.5 lm intervals) were collected on a Peltier- cooled charge-coupled device (Photometrics) and compu- tationally processed by a three-dimensional deconvolution method [33]. Induction of apoptosis HeLa cells were transfected with a cDNA encoding GFP- emerin using Lipofectamine PLUS as recommended by the manufacturer, except that incubation with the DNA solution was reduced to 1.5 h, and cells were cultured for 2 days before use. Anti-Fas Ig and cycloheximide were added to the GFP-emerin-expressing cells on day 2, at a final concentration of 1 lgÆmL )1 and 20 lgÆmL )1 , respect- ively, and then incubated for 3 h at 37 °CinaCO 2 incubator. Results To identify novel binding partners for emerin, we screened a human (HeLa) cDNA library using the yeast two-hybrid assay. Full length emerin, including the transmembrane domain, was fused to the GAL4 DNA-binding domain and used as bait. Positive clones were selected as described in Methods from 1.7 · 10 9 clones screened, and their cDNA inserts were sequenced. Previously known interactors of emerin, such as lamin A, BAF and GCL [5,27,30], were not obtained in our screen. Our positives (total 36 clones) represented a total of three genes, encoding cytochrome c oxidase subunit 3 (six clones), an unknown protein (10 clones – to be reported separately) and a gene previously reported as Btf [28] and KIAA0614 [29] (20 clones). Two splicing isoforms of Btf are known: a long form (Btf L )of918 residues and a short form (Btf S ) which lacks 49 residues (797–846 of Btf L ) near the C-terminus. KIAA0164 encodes two extra Ser residues (inserted between residues 34 and 35 of Btf L ), for a total of 920 residues (predicted mass, 106 120 Da). Our two-hybrid isolate encoded residues Fig. 1. Yeast two-hybrid assay for interaction between emerin and Btf. (A) Yeast two-hybrid assay for emerin truncations in pGBKT7 and Btf truncations in pGADT7. Pairwise interactions between emerin and Btf fragments, as assayed by growth on quadruple-dropout selective medium (minus Trp/Leu/His/Ade; right panel); left panel shows the control plate of cells grown under double-selection (minus Trp/Leu) to maintain both plasmids. (B) Yeast two-hybrid assay for lamin A in pGBKT7 and Btf in pGADT7. Pairwise interaction of pGBKT7 and pGADT7 was assayed by growth on quadruple-dropout medium (right); the left panel is the control, double-selective plate of cells used for the assays. (C) Summary of Btf-interacting domains of emerin in yeast two-hybrid assay. The plus (+) mark at the right represents positive interactions. The C-terminal fragment of emerin (residues 164–254) is sufficient to bind Btf. Interacting domains for emerin are shown at the top of the panel: LEM and transmembrane (TM) domains are indicated. (D) Summary of emerin-interacting domains of Btf in yeast two-hybrid assay. The plus (+) and minus (–) marks at right represent positive and negative interactions, respectively. The central fragment of Btf (residues 377–646) is sufficient to bind emerin. Potential functional domains in Btf are indicated: RS represents the RS domain [45], boxes indicate regions with high (green) and moderate (orange) similarity to transcription complex subunit TRAP150 [42]; the striped bar indicates the putative Bcl-2-binding region of Btf [28]. 1038 T. Haraguchi et al. (Eur. J. Biochem. 271) Ó FEBS 2004 377–761 of Btf, suggesting that this internal region of Btf is sufficient to interact with emerin. For the studies below, we used the KIAA0164 cDNA as full length Btf. To determine the minimum regions of emerin and Btf required for their interaction, we tested pairwise combina- tions of subfragments of each protein for binding in the two- hybrid assay (Fig. 1C,D). Full length emerin (1–254) and two emerin fragments consisting of residues 104–254 and 164–254 were fused to the GAL4 DNA-binding domain. These emerin fragments were tested for binding to full length Btf (1–920) and Btf fragments 377–920, 377–761, 377–646, 377–574 and 521–761 fused to the GAL4 activator domain (Fig. 1A). Diploid cells expressing both emerin and Btf were tested for interaction by their growth under quadruple selection (media lacking Leu, Trp, His and Ade). Full length emerin (1–254) interacted with full length Btf (1–920) and Btf fragments 377–920, 377–761 or 377–646. Control diploids carrying only the control vectors did not survive this selection, as expected (Fig. 1A). No interaction was detected for full length emerin plus Btf fragment 377– 574 or 521–761, suggesting that Btf residues 377–646 comprised a minimum fragment necessary for binding to emerin. Interestingly, this region of Btf also mediates binding to the antideath protein, Bcl-2 (Fig. 1D, striped bar; [28]). We next tested emerin fragments 104–254 and 164–254 (Fig. 1C) for binding to Btf in two-hybrid assays (Fig. 1A). These fragments represent the C-terminal half of emerin, including its transmembrane span and small lumenal domain. Both fragments interacted with full length Btf (1– 920) and Btf fragments 377–920, 377–761 and 377–646, but not with fragments 377–574 or 521–761. We concluded that the N-terminal half of emerin was not essential for binding to Btf, and that exposed C-terminal residues 164–222 were sufficient to bind Btf. Regions of emerin and Btf important for their interaction in the two-hybrid assay are shown schematically (Fig. 1C,D). As a control, we also used the two-hybrid assay to determine if Btf binds to lamin A (Fig. 1B). No interaction was detected between lamin A (fused to the GAL4 DNA- binding domain) and full length Btf (fused to the GAL4 activator domain; Fig. 1B, right), indicating that Btf does not bind directly to lamin A in this assay. Biochemical analysis of Btf binding to emerin Btf residues 377–646 were sufficient to bind emerin in the yeast two-hybrid assay. To test this result biochemically, we synthesized four 35 S-labeled fragments of Btf in coupled transcription/translation extracts in vitro, and assayed their binding using a microtiter well assay (see Materials). Each well contained a constant amount (5–10 pmole) of immo- bilized (adsorbed), purified recombinant human emerin (nucleoplasmic domain; residues 1–222), and BSA to block nonspecific sites. Wells were not allowed to dry at any time during this assay. 35 S-Labeled Btf fragments were then added, incubated 60–90 min, washed, and bound proteins were eluted using 5% SDS and counted. Consistent with the two-hybrid results, the largest Btf fragment 377–920 was positive for binding to emerin and the smallest, fragment 377–574, did not bind (Fig. 2A). Interestingly, this quanti- tative analysis showed detectable but  50% reduced binding of emerin to Btf fragments 377–761 and 377–646. Similar results were found using emerin-conjugated beads (data not shown). We concluded that Btf residues 761–920 contribute significantly to its affinity for emerin, but are not essential. In contrast, Btf residues 574–646 are essential for binding to emerin. The equilibrium binding affinity (K D )of Btf (fragment 377–920) for emerin was 100 n M (range 60–280 n M ; n ¼ 9; Fig. 2B). The stoichiometry of inter- action was 0.8–1 mole Btf per mole emerin. These results collectively showed that Btf has significant binding affinity for emerin, and revealed regions within each protein that mediate their interaction. Mapping residues in emerin required for binding to Btf A functional map of emerin was defined previously, with respect to binding partners BAF and lamin A [5], and two other binding partners [30,31]. To map the binding site for Btfonemerin,wetested[ 35 S]Btf binding to a collection of 21 purified emerin mutants, each bearing a small cluster of site-directed alanine-substitution mutations. Half of these Fig. 2. Biochemical assay for binding of Btf domains to emerin. (A) Quantitation of Btf fragment binding to emerin in wells. Each [ 35 S]Btf fragment (377–920, 377–761, 377–646 or 377–574) was incu- bated with immobilized emerin (1–222) and its binding to emerin was determined as described in Materials and methods. Bars, S.E.M. (B) Affinity of Btf for emerin was determined by adding increasing [ 35 S]Btf (377–920) to constant amounts of immobilized emerin (residues 1–222) in microtiter wells. Double reciprocal plots were used to accurately determine the affinity constant. Bars, S.E.M. Ó FEBS 2004 Emerin binds Btf, a transcriptional repressor (Eur. J. Biochem. 271) 1039 mutations targeted amino acids conserved between emerin and LAP2b (Table 3; [5]), whereas the remaining mutations affected residues unique to emerin (Table 3 [30]);. Each mutant emerin protein was expressed in bacteria, purified by FPLC, immobilized in microtiter wells, incubated with [ 35 S]Btf (fragment 377–920), and the bound [ 35 S]Btf was counted as described above. The amounts of emerin loaded per well were similar within a factor of  2, as determined by SDS-elution of proteins from parallel wells and immuno- blotting with antibodies against human emerin (data not shown). As emerin was in five-fold excess, slight variations in the amount of emerin per well did not affect the results. We first considered the effects of mutations in ÔconservedÕ residues. Strong binding to Btf was seen for wildtype emerin (residues 1–222) and LEM-domain mutant m24 (Fig. 3A), as expected. The binding of mutants 34, 112, 164 and 179 was slightly reduced, to 70% of wildtype. Binding was significantly reduced, to 20–40% of wildtype, for mutants 70, 76, 196, 207 and 214 (Fig. 3A). For mutations in Ôemerin-specificÕ residues (Fig. 3B), wildtype levels of binding to Btf were detected for mutants 133, 151, 161 and 198, and binding was reduced slightly (to 70% of wildtype) for mutants 122 and 145. However several emerin- specific mutations (45E, 175, 192 and 206) showed signifi- cantly reduced binding to Btf (25–35% of wildtype). Background binding of [ 35 S]Btf to negative control wells containing BSA ranged from 5 to 15% of wildtype (Fig. 3B). Collectively, this analysis implicated two regions of emerin as important for binding to Btf: residues 45–83 and the C-terminal region (residues 175–217) (see Figs 3A,B and 4C). Note that Ômutation cluster 76Õ consists of four alanine substitutions spanning residues 76 through 83 in emerin [30], extending the ÔimplicatedÕ region to residue 83. Interestingly, disease-causing mutations S54F, D95-99 and P183H also lie within these regions. Disease-causing mutations S54F and D95–99 reduce emerin binding to Btf To determine if Btf binding was sensitive to disease-causing mutations, we first tested [ 35 S]Btf for binding to emerin mutants S54F, D95–99 and P183H on blots (Ôblot overlayÕ assay; Fig. 4A). Western blotting confirmed that similar amounts of emerin protein were present in each lane (Fig. 4A, Emr). We found positive binding of [ 35 S]Btf to wildtype emerin (residues 1–222) and mutant P183H, but no signal for mutants S54F or D95–99 (Fig. 4A). The D95–99 mutation also disrupts binding to lamin A [5]. However, the S54F result was remarkable, because this mutant bound normally to all previously tested binding partners, including BAF and lamin A. Blot overlay assays are insensitive, as the blotted binding partner is often at least partially denatured. To independ- ently verify and quantify this reduced binding of Btf to emerin mutant S54F, we measured the binding of [ 35 S]Btf to disease-associated emerins in the more sensitive and Table 3. Mutations in emerin. Mutations 45A to 206 target Ôemerin- specificÕ residues; mutants 24 to 214 target residues conserved between emerin and LAP2b [5]. Mutated residues are indicated by lines. Name of mutation Wildtype residues Mutant residues 45A 45RRR 45 AAA 45E 45RRR 45 EEE 104 104TYGEPES 104 AYGEAEA 122 122TS 122 AA 145 145EE 145AA 151 151ER 151 AA 161 161YQS 161 AAA 175 175SSL 175 AAA 192 192SSSSS 192 ASAAA 198 198SSWLTR 198 AAAAA 206 206IRPE 206 AAPA 24 24GPVV 24 AAAA 34 34YEKK 34 AAAA S54F 54S 54 F 70 70DADMY 70 AAAMA 76 76LPKKEDAL 76 APAKADAA 112 112GPSRAVRGSVT 112 AASRAVAAAVA 133 133Q 133H 141 141SSSEEECKDR 141 AASAEECKAA 164 164ITHYRPV 164 AAHARPA 179 179LS 179 AA 183 183P 183 H 196 196SS 196 AA 207 207RP 207 AA 214 214GAGL 214 AAGA Fig. 3. Biochemical assay for binding of Btf to emerin mutants. Microtiter well binding assays for [ 35 S]Btf binding to wildtype (WT) emerin residues 1–222, or emerin bearing clusters of site-directed alanine-substitution mutations in either (A) residues conserved between emerin and LAP2b (Table 3,[5]) or (B) residues unique to emerin (not conserved in LAP2b; Table 3). Bars, S.E.M. 1040 T. Haraguchi et al. (Eur. J. Biochem. 271) Ó FEBS 2004 quantitative microtiter well assay (Fig. 4B). Strong binding was seen for wildtype emerin and mutant P183H, whereas binding to mutants S54F and D95-99 was reduced by 50%. Thus, the binding of Btf to mutant S54F was reduced significantly in two independent assays, suggesting that transcription factor Btf is uniquely sensitive to the S54F mutation that causes EDMD. These results for emerin mutants are summarized in Fig. 4C, in relation to regions previously defined as important for binding to BAF and lamin A [5]. Btf relocalizes from intranuclear ‘dot-like’ structures to positions near the nuclear envelope in Fas-antibody-treated (apoptotic) cells To determine if Btf interacts with emerin in vivo,we generated an antibody specific for Btf. This affinity-purified antibody recognized a single band, in the nuclear fraction of HeLa cells, with an apparent mass of 160 KDa in SDS/ PAGE (Fig. 5A, left panel). This band was specific, because it was not recognized by preimmune antiserum (right panel) and was competed for by pretreatment of antibody with the antigen (middle panel). As this putative Btf band migrated more slowly (160 KDa) than predicted from its ORF (106 KDa), we expressed a GFP-Btf fusion protein in HeLa cells and compared this to endogenous Btf by Western blotting (Fig. 5B). The affinity-purified antibody recognized the endogenous 160 KDa band plus a second band with an apparent mass of 180 KDa (Fig. 5B, left lane), which was also recognized by antibodies against GFP (Fig. 5B, right lane). We drew three conclusions: (a) our antibody speci- fically recognized Btf; (b) HeLa cells express Btf, and (c) endogenous Btf migrates in SDS/PAGE with an apparent mass of 160 KDa. We then used this specific antibody to determine if endogenous Btf and emerin could be immunoprecipitated from lysates of HeLa cells. However, the immunopreci- pitations failed due to the insolubility of Btf (data not shown). This same problem was encountered with HA-tagged Btf in transiently transfected HeLa cells (data not shown). We therefore used cytological methods to localize Btf in HeLa cells. Indirect immunofluorescence staining with the affinity-purified antibody showed that Btf was localized inside the nucleus in dot-like structures, distant from emerin (Fig. 5C). Thus, Btf and emerin occupy separate nuclear domains in HeLa cells under normal culture conditions. As muscle wasting in EDMD is thought to involve apoptosis [34], we tested the hypothesis that the subnuclear localization of Btf might change in apoptosis-induced cells. Two days after trans- fection with GFP-emerin encoding plasmids, HeLa cells were induced to enter apoptosis by treatment with Fas antibody plus cycoheximide [35]. In untreated cells, the Btf and GFP-emerin signals are clearly separate (Fig. 5C). However in Fas-antibody-treated cells, within 3 h, the endogenous Btf relocalized to punctate positions near the nuclear envelope, close to GFP-emerin but not spectrally overlapping (Fig. 5D–F). This change in localization occurred relatively early in apoptosis when the nuclei were still relatively spherical and before chromosomes became grossly condensed. These results suggest that the subnuclear localization of Btf is differentially regulated during apoptosis, and that our biochemically–character- ized interaction between Btf and emerin may be physio- logically relevant to regulate Btf at an early stage of cell death. These experiments did not address whether emerin inhibits or promotes Btf’s pro-death activity. However, as muscles might enter apoptosis too readily when emerin is missing, we speculate that Btf is normally inhibited by its association with emerin and potentially other nuclear membrane proteins. Discussion We found that a reportedly pro-apoptotic transcription regulator, Btf, binds emerin with nanomolar affinity in vitro. Importantly, Btf binding to emerin is weakened significantly by the disease-causing S54F mutation, whereas all other previously tested binding partners (BAF, lamin A, GCL and splicing factor YT521-B) bind normally to S54F (reviewed by Bengtsson and Wilson, 2004 [36]). These Fig. 4. Binding of Btf to disease-specific emerin mutants. (A) Blot overlay assays for binding of [ 35 S]Btf to wildtype emerin (WT) and disease-causing emerin mutants S54F, D95–99 (D95) and P183H. (B) Solution binding assays measuring [ 35 S]Btf binding to wildtype (WT) or mutant emerin proteins (numbered as in Fig. 3) immobilized in microtiter wells. Bars, S.E.M. (C) Diagram mapping the proposed Btf-binding domains in emerin, relative to reported binding domains for BAF and lamin A [5]. No binding partner has yet been reported to be disrupted by disease-causing mutant P183H (this report,[5]). Stars indicate disease-causing emerin mutants. Ó FEBS 2004 Emerin binds Btf, a transcriptional repressor (Eur. J. Biochem. 271) 1041 findings, and our discovery that Btf relocalizes near the nuclear envelope specifically during apoptosis, suggest that Btf is a disease-relevant binding partner for emerin. Btf is highly expressed in skeletal muscle, although it is ubiqui- tously expressed in other tissues tested, including heart and brain [22,29]. Although this expression pattern for Btf does not correlate perfectly with disease-affected tissues, some functions of emerin are known to overlap with both LAP2b [30,37] and MAN1 [38]. It is therefore possible that, in patients who lack emerin, tissues that express ÔbackupÕ LEM-domain proteins might be protected from disease. In this regard, it will be important to determine which (if any) other LEM-domain proteins can interact with Btf. Emerin and its binding partners in EDMD Emerin has many interesting binding partners in the nucleus [36]. Btf joins a small but growing number of emerin- binding proteins that regulate transcription (BAF [5,39], GCL [30]) or splice site selection (YT521-B [31]), or are proposed to regulate transcription (Lmo7; J. M. Holaska and K. L. Wilson, unpublished results). This group of interactors support gene expression models for emerin function and, potentially, the disease mechanism of EDMD and further suggest that emerin and other LEM-domain proteins interact with a variety of overlapping binding partners at the inner nuclear membrane. Our mapping results showed that alanine substitutions in two discrete regions of emerin, residues 45–83 and 175–217, disrupt its binding to Btf. These same regions of emerin, previously designated Ôrepressor binding domains (RBD) 1and2Õ [30] are also critical for binding to GCL [30] and splice site regulator YT521-B [31]. Most (but not all) mutations that disrupted emerin binding to Btf, also disrupt its binding to GCL [30] and YT521-B [31]. Both GCL and Btf are expressed widely in human tissues [29,30]. GCL is Fig. 5. Btf moves near the nuclear envelope in apoptosis-induced HeLa cells. (A) Western blotting to verify specificity of the antibody. Total HeLa cell extracts (total) and corres- ponding cytoplasmic (cytosol) and nuclear (nucleus) fractions were resolved by SDS/ PAGE, and immunoblotted with either affin- ity-purified immune antibodies against Btf (left panel), antigen-pretreated immune anti- bodies (middle panel) or preimmune serum (right panel). The immune antibody specific- ally recognized a nuclear protein that migrated at 160 kDa. (B) Western blots of lysates from HeLa cells that were either nontransfected (–) or transiently transfected to express GFP-Btf (GFP-Btf), probed with the antibodies against Btf (left panel) or GFP (right panel). (C) Subnuclear localization of GFP-emerin and endogenous Btf stained with anti-Btf Ig in nonapoptotic HeLa cells. (D–F) Subnuclear localization of GFP-emerin and endogenous Btf stained with anti-Btf Ig, after 3 h of treatment with Fas-antibody and cyclohexi- mide to induce apoptosis in HeLa cells (Methods). 1042 T. Haraguchi et al. (Eur. J. Biochem. 271) Ó FEBS 2004 most highly expressed in testis and is required for germ cell formation [37,40,41], whereas Btf is highly expressed in skeletal muscle. Our findings for Btf, coupled to previous findings for GCL and YT521-B, strongly support the hypothesis that two regions in the primary amino acid sequence of emerin (RBD1 and RBD2; Fig. 4C and refs [30,31]) form a docking site for gene regulatory proteins. We therefore predict that Btf, GCL and YT521-B are not alone, and that additional proteins involved in gene expression or RNA splicing will also emerge as emerin-binding proteins. Interestingly, the emerin-binding region of Btf is homolog- ous to transcription complex subunit TRAP150 [42], leading us to speculate that TRAP150 might also bind emerin. We suggest that additional binding partners that are both expressed in disease-affected tissues, and potentially disrup- ted by disease-causing mutations in emerin, remain to be discovered. An important next step is to identify genes regulated by emerin-dependent transcription factors, to gain specific insight into the molecular mechanisms of EDMD disease. Our current knowledge suggests that emerin may be anchored and stabilized at the nuclear inner membrane by nesprin-1a and lamin A [43,44]; in turn, emerin may provide regulated binding sites for BAF or other binding partners (Btf, GCL and YT521-B) involved in transcription or splicing. Btf is highly expressed in skeletal muscle, and is therefore presumably important for muscle cell function. A mutation that disrupts Btf binding to emerin would be expected to affect muscles disproportionately, especially if putative ÔbackupÕ; LEM-domain proteins are absent or expressed at levels too low (e.g. LAP2b [30]) to compensate for the absence of emerin. Possible functions for Btf The function of Btf is not fully understood. Btf was discovered as a two-hybrid binding partner for E1B19K, a viral protein similar to Bcl-2 [28]. When fused to the GAL4- DNA binding domain, Btf is sufficient to repress a reporter gene [28], showing that Btf can repress transcription in vivo. Btf also promotes apoptosis when overexpressed in cells [28]. Btf binds to ÔantideathÕ Bcl-2-related proteins such as E1B19K, Bcl-2, and Bcl-XL through its C-terminal region, and this binding is thought to promote apoptosis by blocking their antideath activity [28]. Our two-hybrid mapping results suggest that emerin and Bcl-2 might bind similar regions of Btf (see Fig. 1D). We therefore hypo- thesize that Btf has a choice, and can bind either to Bcl-2 or emerin; in cells that lack functional emerin this balance would be lost, potentially leading to increased Btf binding to Bcl-2 and entry into apoptosis. This hypothesis is supported by our evidence that Btf relocalizes near emerin in apoptosis-induced cells. Consistent with its ability to repress transcription, Btf localizes in HeLa cell nuclei (this report and [28]), where it is enriched in discrete Ôdot-likeÕ structures adjacent to RNA splicing factor SC35 (T. Haraguchi and Y. Hiraoka, unpublished results). Btf was identified independently in a proteomic analysis of purified interchromatin granule clusters (IGCs), which contain > 200 proteins, including many RNA splicing factors (N. Saitoh and D. Spector, personal communication [Cold Spring Harbor Laborator- ies, New York)]. The N-terminus of Btf includes Arg-Ser repeats (a so-called ÔRS domainÕ), which are characteristic for splicing factors and many other RNA-binding proteins [45]. Thus, the functions of Btf are not yet understood, but might include roles in mRNA metabolism, transcriptional repression [28] or pro-apoptotic responses [28]. We speculate that Btf binding to emerin in vivo is regulated at least during cell death, and potentially also regulated by signals such as hormones, growth/survival factors, exercise or atrophy (in muscle). Testing these models will require further study of Btf function both at the molecular level, and in disease- affected tissues. Acknowledgements We are grateful to Drs Tsuchiya and Arahata for emerin constructs, Dr Nagase (Kazusa DNA) for the KIAA0164 construct, Dr White (Rutgers University) for DNA constructs of Btf L ,Btf S and E1B19K, the Riken Cell Bank for HeLa cells and Drs Saitoh and Spector (Cold Spring Harbor Laboratories, New York) and Dr Morris (Northeast Wales Institute, Wrexham, United Kingdom) for communicating their results prior to publication. We also thank Ms. Kumiko Matsuno for initial cloning of Btf in the yeast two-hybrid assay. This work was supported by grants from the Japan Science and Technology Corporation (CREST; to T. H. and Y. H.), Grant-in-Aid for Scientific Research B (to T. H. and Y. H.), National Institutes of Health Cardiology training grant (T32-HLO-7227-26; to J. M. H.), and grants from the National Institutes of Health (GM48646) and the Scott B. Deutschman memorial Research Award from the American Heart Association (to K. L. W.). References 1. Bione, S., Maestrini, E., Rivella, S., Mancini, M., Regis, S., Romeo, G. & Toniolo, D. (1994) Identification of a novel X-linked gene responsible for Emery–Dreifuss muscular dystrophy. Nat. Genet. 8, 323–327. 2. Emery, A.E.H. (1989) Emery–Dreifuss syndrome. J. Med. Genet. 26, 637–641. 3. Emery, A.E.H. (2000) Emery–Dreifuss muscular dystrophy – a 40 year retrospective. Neuromuscul. Disord. 10, 228–232. 4. Clements, L., Manilal, S., Love, D.R. & Morris, G.E. (2000) Direct interaction between emerin and lamin A. Biochem. Biophys. Res. Commun. 267, 709–714. 5. Lee,K.K.,Haraguchi,T.,Lee,R.S.,Koujin,T.,Hiraoka,Y.& Wilson, K.L. (2001) Distinct functional domains in emerin bind lamin A and DNA-bridging protein BAF. J. Cell Sci. 114, 4567– 4573. 6. Bonne, G., Di Barletta, M.R., Varnous, S., Becane, H M., Hammouda, E.H., Merlini, L., Muntoni, F., Greenberg, C.R., Gary, F., Urtizberea, J A. et al. (1999) Mutations in the gene encoding lamin A/C cause autosomal dominant Emery–Dreifuss muscular dystrophy. Nat. Genet. 21, 285–288. 7. De Sandre-Giovannoli, A., Chaouch, M., Kozlov, S., Vallat, J.M., Tazir, M., Kassouri, N., Szepetowski, P., Hammadouche, T., Vandenberghe, A., Stewart, C.L. et al. (2002) Homozygous defects in LMNA, encoding lamin A/C nuclear-envelope pro- teins, cause autosomal recessive axonal neuropathy in human (Charcot–Marie–Tooth disorder type 2) and mouse. Am. J. Hum. Genet. 70, 726–736. [Erratum appears in Am.J.Hum.Genet. 70, 1075.]. 8. Novelli, G., Muchir, A., Sangiuolo, F., Helbling-Leclerc, A., D’Apice, M.R., Massart, C., Capon, F., Sbraccia, P., Federici, M., Lauro, R. et al. (2002) Mandibuloacral dysplasia is caused by a mutation in LMNA-encoding lamin A/C. Am.J.Hum.Genet.71, 426–431. Ó FEBS 2004 Emerin binds Btf, a transcriptional repressor (Eur. J. Biochem. 271) 1043 9. Wilson, K.L., Zastrow, M. & Lee, K.K. (2001) Nuclear lamins and disease: insights into nuclear infrastructure. Cell 104, 647–650. 10. Burke, B. & Stewart, C.L. (2002) Life at the edge: the nuclear envelope and human disease. Nat.Rev.Mol.CellBiol.3, 575–585. 11. Wehnert, M.S. & Bonne, G. (2002) The nuclear muscular dys- trophies. Semin. Pediatr. Neurol. 9, 100–107. 12. Eriksson, M., Brown, W.T., Gordon, L.B., Glynn, M.W., Singer, J.,Scott,L.,Erdos,M.R.,Robbins,C.M.,Moses,T.Y.,Berglund, P., Dutra, A., Pak, E., Durkin, S., Csoka, A.B., Boehnke, M., Glover, T.W. & Collins, F.S. (2003) Recurrent de novo point mutations in lamin A cause Hutchinson–Gilford progeria syn- drome. Nature 423, 293–298. 13. De Sandre-Giovannoli, A., Bernard, R., Cau, P., Navarro, C., Amiel, J., Boccaccio, I., Lyonnet, S., Stewart, C.L., Munnich, A., Le Merrer, M. & Levy, N. (2003) Lamin A truncation in Hutch- inson–Gilford progeria. Science 300, 2055. 14. Chen,L.,Lee,L.,Kudlow,B.A.,DosSantos,H.G.,Sletvold, O., Shafeghati, Y., Botha, E.G., Garg, A., Hanson, N.B., Martin, G.M., Mian, I.S., Kennedy, B.K. & Oshima, J. (2003) LMNA mutations in atypical Werner’s syndrome. Lancet 362, 440–445. 15. Wilson, K.L. (2000) The nuclear envelope, muscular dystrophy and gene expression. Trends Cell Biol. 10, 125–129. 16. Nagano, A. & Arahata, K. (2000) Nuclear envelope proteins and associated diseases. Curr. Opinion Neurol. 13, 533–539. 17. Cohen, M., Lee, K.K., Wilson, K.L. & Gruenbaum, Y. (2001) Transcriptional repression, apoptosis, human disease and the functional evolution of the nuclear lamina. Trends Bioch. Sci. 26, 41–47. 18. Manilal, S., Nguyen, T.M., Sewry, C.A. & Morris, G.E. (1996) The Emery–Dreifuss muscular dystrophy protein, emerin, is a nuclear membrane protein. Hum. Mol. Genet. 5, 801–808. 19. Nagano, A., Koga, R., Ogawa, M., Kurano, Y., Kawada, J., Okada, R., Hayashi, Y.K., Tsukahara, T. & Arahata, K. (1996) Emerin deficiency at the nuclear membrane in patients with Emery–Dreifuss muscular dystrophy. Nat. Genet. 12, 254–259. 20. Yorifuji, H., Tadano, Y., Tsuchiya, Y., Ogawa, M., Goto, K., Umetani, A., Asaka, Y. & Arahata, K. (1997) Emerin, deficiency of which causes Emery–Dreifuss muscular dystrophy, is localized at the inner nuclear membrane. Neurogenetics 1, 135–140. 21. Ellis, J.A., Craxton, M., Yates, J.R. & Kendrick-Jones, J. (1998) Aberrant intracellular targeting and cell cycle-dependent phos- phorylation of emerin contribute to the Emery–Dreifuss muscular dystrophy phenotype. J. Cell Sci. 111, 781–792. 22. Morris, G.E. & Manilal, S. (1999) Heart to heart: from nuclear proteins to Emery–Dreifuss muscular dystrophy. Hum. Mol. Genet. 8, 1847–1851. 23. Fairley, E.A.L., Kendrick-Jones, J. & Ellis, J.A. (1999) The Emery–Dreifuss muscular dystrophy phenotype arises from aberrant targeting and binding of emerin at the inner nuclear membrane. J. Cell Sci. 112, 2571–2582. 24. Haraguchi, T., Koujin, T., Segura-Totten, M., Lee, K.K., Mats- uoka, Y., Yoneda, Y., Wilson, K.L. & Hiraoka, Y. (2001) BAF is required for emerin assembly into the reforming nuclear envelope. J. Cell Sci. 114, 4575–4585. 25. Lin, F., Blake, D.L., Callebaut, I., Skerjanc, I.S., Holmer, L., McBurney, M.W., Paulin-Levasseur, M. & Worman, H.J. (2000) MAN1, an inner nuclear membrane protein that shares the LEM domain with lamina-associated polypeptide 2 and emerin. J. Biol. Chem. 275, 4840–4847. 26.Shumaker,D.K.,Lee,K.K.,Tanhehco,Y.C.,Craigie,R.& Wilson, K.L. (2001) LAP2 binds to BAF-DNA complexes: requirement for the LEM-domain and modulation by variable regions. EMBO J. 20, 1754–1764. 27. Segura-Totten, M., Kowalski, A.K., Craigie, R. & Wilson, K.L. (2002) Barrier-to-autointegration factor: major roles in chromatin decondensation and nuclear assembly. J. Cell Biol. 158, 475– 485. 28. Kasof, G.M., Goyal, L. & White, E. (1999) Btf, a novel death- promoting transcriptional repressor that interacts with Bcl-2 related proteins. Mol. Cell Biol. 19, 4390–4404. 29. Nagase, T., Seki, N., Ishikawa, K., Tanaka, A. & Nomura, N. (1996) Prediction of the coding sequences of unidentified human genes. V. The coding sequences of 40 new genes (KIAA0161- KIAA0200) deduced by analysis of cDNA clones from human cell line KG-1. DNA Res. 3, 17–24. 30. Holaska, J.M., Lee, K.K., Kowalski, A.K. & Wilson, K.L. (2003) Transcriptional repressor germ cell-less (GCL) and BAF compete for binding to emerin. J. Biol. Chem. 278, 6969–6975. 31. Wilkinson, F.L., Holaska, J.M., Zhang, Z., Sharma, A., Manilal, S.,Holt,I.,Stamm,S.,Wilson,K.L.&Morris,G.E.(2003) Emerin interacts in vitro with the splicing-associated factor, YT521-B. Eur. J. Biochem. 270, 2459–2466. 32. Matsuoka, Y., Nishizawa, K., Yano, T., Shibata, M., Ando, S., Takahashi, T. & Inagaki, M. (1992) Two different protein kinases act on a different time schedule as glial filament kinases during mitosis. EMBO J. 11, 2895–2902. 33. Agard, D., Hiraoka, Y., Show, P. & Sedat, J.W. (1989) Fluores- cence microscopy in three dimensions. Meth.CellBiol.30, 353–377. 34. Bonne, G., Yaou, R.B., Beroud, C., Boriani, G., Brown, S., de Visser, M., Duboc, D., Ellis, J., Hausmanowa-Petrusewicz, I., Lattanzi, G., Merlini, L., Morris, G., Muntoni, F., Opolski, G., Pinto, Y.M., Sangiuolo, F., Toniolo, D., Trembath, R., van Berlo, J.H., van der Kooi, A.J. & Wehnert, M. (2003) 108th ENMC International Workshop, 3rd Workshop of the MYO- CLUSTER project: EUROMEN, 7th International Emery– Dreifuss Muscular Dystrophy (EDMD) Workshop, 13–15 Sep- tember 2002, Naarden, The Netherlands. Neuromuscul. Disord. 13, 508–515. 35. Yonehara, S., Ishii, A. & Yonehara, M. (1989) A cell-killing monoclonal antibody (anti-Fas) to a cell surface antigen co- downregulated with the receptor of tumor necrosis factor. J. Exp. Med. 169, 1747–1756. 36. Bengtsson, L. & Wilson, K.L. (2004) Multiple and surprising new functions for emerin, a nuclear membrane protein. Curr. Opin. Cell Biol. in press. 37. Nili, E., Cojocaru, G.S., Kalma, Y., Ginsberg, D., Copeland, N.G., Gilbert, D.J., Jenkins, N.A., Berger, R., Shaklai, S., Amariglio, N. et al. (2001) Nuclear membrane protein LAP2b mediates transcriptional repression alone and together with its binding partner GCL (germ-cell-less). J. Cell Sci. 114, 3297–3307. 38. Liu, J., Lee, K.K., Segura-Totten, M., Neufeld, E., Wilson, K.L. & Gruenbaum, Y. (2003) MAN1 and emerin have overlapping function(s) essential for chromosome segregation and cell division in Caenorhabditis elegans. Proc. Natl Acad. Sci. USA 100, 4598– 4603. 39. Wang, X., Xu, S., Rivolta, C., Li, L.Y., Peng, G.H., Swain, P.K., Sung, C.H., Swaroop, A., Berson, E.L., Dryja, T.P. et al. (2002) Barrier to autointegration factor interacts with the cone-rod homeobox and represses its transactivation function. J. Biol. Chem. 277, 43288–43300. 40. Kimura,T.,Yomogida,K.,Iwai,N.,Kato,Y.&Nakano,T. (1999) Molecular cloning and genomic organization of mouse homologue of Drosophila germ cell-less and its expression in germ lineage cells. Biochem. Biophys. Res. Commun. 262, 223–230. 41. Robertson, S.E., Dockendorff, T.C., Leatherman, J.L., Faulkner, D.L. & Jongens, T.A. (1999) Germ cell-less is required only during the establishment of the germ cell lineage of Drosophila and has activities which are dependent and independent of its localization to the nuclear envelope. Dev. Biol. 215, 288–297. 1044 T. Haraguchi et al. (Eur. J. Biochem. 271) Ó FEBS 2004 [...]... 2004 Emerin binds Btf, a transcriptional repressor (Eur J Biochem 271) 1045 42 Ito, M., Yuan, C.-X., Malik, S., Gu, W., Fondell, J.D., Yamamura, S., Fu, Z.-Y., Zhang, X., Qin, J & Roeder, R.G (1999) Identity between TRAP and SMCC complexes indicates novel pathways for the function of nuclear receptors and diverse mammalian activators Mol Cell 3, 361–370 43 Mislow, J.M., Kim, M.S., Davis, D.B & McNally,... (200 2a) Myne-1, a spectrin repeat transmembrane protein of the myocyte inner nuclear membrane, interacts with lamin A/ C J Cell Sci 115, 61–70 44 Mislow, J.M., Holaska, J.M., Kim, M.S., Lee, K.K., SeguraTotten, M., Wilson, K.L & McNally, E.M (2002b) Nesprin- 1a self-associates and binds directly to emerin and lamin A in vitro FEBS Lett 525, 135–140 45 Valcarcel, J & Green, M.R (1996) The SR protein family:... E.M (2002b) Nesprin- 1a self-associates and binds directly to emerin and lamin A in vitro FEBS Lett 525, 135–140 45 Valcarcel, J & Green, M.R (1996) The SR protein family: pleiotropic functions in pre-mRNA splicing Trends Biochem Sci 21, 296–301 . Emerin binding to Btf, a death-promoting transcriptional repressor, is disrupted by a missense mutation that causes Emery–Dreifuss muscular dystrophy Tokuko. ATC AGG AAG CTC TAG ATT AC 521–5¢ AAC ATA TGG CAC GAG AAA AGT CTA CCT TC 574–3¢ TTG GAT CCT TAT GTA CTA GCA AGC AGC C 646–3¢ TTG GAT CCT TAT TGC CGA GTA