C gene

Available online at www.sciencedirect.com R

Experimental Cell Research 291 (2003) 352–362

www.elsevier.com/locate/yexcr

Nuclear envelope alterations in fibroblasts from LGMD1B patients carrying nonsense Y259X heterozygous or homozygous mutation in lamin A/C gene Antoine Muchir,a Baziel G. van Engelen,b Martin Lammens,b John M. Mislow,c Elizabeth McNally,d Ketty Schwartz,a and Gise`le Bonnea,* a

b

INSERM U582, Institut de Myologie, Groupe Hospitalier Pitie´-Salpe´trie`re, 75013 Paris, France Institute of Neurology and Department of Pathology, Neuromuscular Centre Nijmegen, University Medical Centre Nijmegen, 6500 HB Nijmegen, The Netherlands c Department of Pathology, University of Chicago, Chicago, IL 60637, USA d Department of Medicine, University of Chicago, Chicago, IL 60637, USA Received 20 February 2003, revised version received 19 May 2003

Abstract Mutations in the LMNA gene encoding nuclear lamins A and C are responsible for seven inherited disorders affecting specific tissues. We have analyzed skin fibroblasts from a patient with type 1B limb-girdle muscular dystrophy and from her deceased newborn grandchild carrying, respectively, a heterozygous (⫹/mut) and a homozygous (mut/mut) nonsense Y259X mutation. In fibroblasts⫹/mut, the presence of only 50% lamins A and C promotes no detectable abnormality, whereas in fibroblastsmut/mut the complete absence of lamins A and C leads to abnormally shaped nuclei with lobules in which none of the analyzed nuclear proteins were detected, i.e., B-type lamins, emerin, nesprin-1␣, LAP2␤, and Nup153. These lobules perturb cell division as fibroblastmut/mut cultures with large proportions of cells with dysmorphic nuclei grow more slowly than controls and the cell proliferation normalizes when the number of these abnormally shaped nuclei declines. In all fibroblastsmut/mut, nesprin-1␣-like emerin exhibited aberrant localization in the endoplasmic reticulum. Transfection of wild-type lamin A or C cDNAs restored the correct localization of both emerin and nesprin-1␣. These data demonstrate that lamin C, like lamin A, interacts in vivo directly with nesprin-1␣ and with emerin and that lamin A or C is sufficient for the correct anchorage of emerin and nesprin-1␣ at the nuclear envelope in human cells. © 2003 Elsevier Inc. All rights reserved. Keywords: Lamins A and C; Emerin; Nesprin-1␣; Nuclear envelope; Nuclear lamina; Muscular dystrophy, LGMD1B

Introduction Lamins are members of the intermediate filament protein family that form the nuclear lamina, a fibrous meshwork underlining the inner nuclear membrane (INM) of all eukaryotic cells [1]. Two types of lamins have been identified in somatic cells, A-type and B-type. A-type lamins, A and C, are encoded by alternative splice of the LMNA gene,

* Corresponding author: Inserm U582, Institut de Myologie, Baˆtiment Babinski, GH Pitie´-Salpe´trie`re, 47 boulevard de l’Hoˆpital, 75651 Paris cedex 13, France. Fax: ⫹33-1-42-16-57-00. E-mail address: [email protected] (G. Bonne). 0014-4827/$ – see front matter © 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.yexcr.2003.07.002

whereas B-type lamins, B1 and B2, are encoded by two genes [1]. While lamins B1 and B2, are found in all nucleated somatic cells, the expression of lamins A and C is developmentally regulated and their expression is restricted to differentiated nonproliferating cells [2,3]. Like all intermediate filament proteins, the lamins possess a long central ␣-helical coil-coiled rod domain flanked by a short globular N-terminal head and a long C-terminal tail domain. The proteins form dimers via their rod domains, which then assemble to form the nuclear lamina [4]. The latter is attached to the INM via interactions with nuclear integral proteins. Among them, emerin, nesprin-1␣, and isoforms 1A, 1B, and 1C of the lamina-associated proteins (LAP)

A. Muchir et al. / Experimental Cell Research 291 (2003) 352–362

have specific interactions with lamins A and C [5– 8]. Lamins A and C also have interactions with the nuclear pore complexes (NPC) [9], LAP2␣ [10], chromatin [11,12], and DNA [13]. The nuclear lamina is known to play a role in the maintenance of nuclear architecture and to contribute to the regulation of chromatin organization [14]. Several observations suggest that the lamins colocalize with sites of DNA replication [15,16] and splicing factors [17]. Lamins are also involved in the synthesis of mRNA [18]. Mutations in the LMNA gene have been shown to be responsible for seven diseases. Three of these specifically affect the striated muscle: autosomal dominant and recessive forms of Emery–Dreifuss muscular dystrophy (ADand AR-EDMD) [19,20], autosomal dominant limb-girdle muscular dystrophy with cardiac conduction disturbances (LGMD1B) [21], and an autosomal dominant form of dilated cardiomyopathy with conduction defect (DCM-CD) [22,23]. The fourth specifically involves the adipose tissue: autosomal dominant familial partial lipodystrophy (FPLD) [24]. The fifth, an autosomal recessive form of axonal neuropathy (AR-CMT2), specifically affects the peripheral nervous tissue [25]. Sixth is an autosomal recessive disorder affecting adipose and bone tissues: mandibuloacral dysplasia (MAD) [26]. And finally, the autosomal dominant form of Hutchinson–Gilford progeria syndrome (HGPS) is seventh [27,28]. LGMD1B is an autosomal dominant disorder characterized by symmetrical weakness starting in the proximal lower limb muscles and gradually affecting the upper limb muscles. It is associated with cardiac conduction disturbances that lead to a dilated cardiomyopathy and/or sudden death [29]. To investigate the molecular and cellular alterations involved in LGMD1B, we performed morphological and biochemical analyses on skin fibroblasts cultured from two members of a LGMD1B family, a 66-year-old woman and her deceased newborn grandchild, carrying respectively a heterozygous (⫹/ mut) and homozygous (mut/mut) LMNA nonsense mutation (Y259X) (van Engelen et al., in preparation). In this study we demonstrate that in a homozygous state, this nonsense mutation causes the lack of lamins A and C in fibroblasts from the deceased newborn child. This loss of lamins A and C leads to the destabilization of the nuclear envelope (NE) due to the disorganization of the lamina and alterations in the protein composition of the INM. We show that in the absence of lamins A and C, nesprin-1␣-like emerin is aberrantly localized in the endoplasmic reticulum. These results highlight the important role played by lamins A and C in the organization and structure of the NE in human cells.

353

from consanguineous parents was born at 30 gestational weeks and died at birth. He carried a homozygous nonsense LMNA mutation Y259X. His maternal grandmother (66 years old) carried a heterozygous nonsense LMNA mutation Y259X. She presented clinically with a LGMD1B phenotype, with limb girdle weakness, which had started at 40 years of age, and bradycardia (40/min) necessitating a pacemaker at 56 years. A full clinical description of this LGMD1B family will be reported elsewhere, the child and his grandmother being family members V-2 and III-13, respectively (van Engelen et al., in preparation). Skin fibroblasts from two control subjects with no known disease (a 12-week-old fetus from a voluntary abortion and a 33-yearold subject who underwent cosmetic surgery) were analyzed as age-matched controls, i.e., fetal age (12 weeks for control vs 30 weeks for patient) and adult age (33 years for control vs 66 years for patient). All fibroblasts were obtained from skin biopsies that were obtained after informed consent of the subjects or the parents for the fetuses and complied with ethical guidelines approved by an institutional review board. Cell culture Skin biopsies were cut into explants approximately 1 mm in diameter, and the explants were incubated in 100-mmdiameter Petri dishes. Fibroblasts were maintained at 37°C in a humidified incubator containing 5% CO2 in DMEM (Gibco/BRL, Paisley, UK) supplemented with 10% fetal calf bovine serum (Gibco-BRL) and 0.1% gentamycin (Gibco-BRL). When fibroblasts had grown out from the explants and the cells were about 80% confluent, the fibroblasts were trypsinated using 0.25% trypsin–EDTA (GibcoBRL) and transferred into new Petri dishes. Each fibroblast trypsination corresponds to a passage. Passage number 1 is the first trypsination after the fibroblasts had grown out from the biopsy. Assays for cell growth Fibroblasts were plated at 2 ⫻ 104 cells per dish in nine uncoated 35 ⫻ 10-mm culture dishes and grown in 2 ml of culture medium as described above. For each fibroblast type (control, heterozygous, homozygous), cell growth was determined daily for 9 days by direct cell counts of one dish. For each dish, fibroblasts were rinsed in PBS and treated with the trypsin solution described above; cells were harvested and centrifuged for 10 min at 1000 rpm. The cell pellet was resuspended in 4 ml of medium and the cell number was determined with a Mallassez hemocytometer.

Materials and methods Indirect immunofluorescence microscopy Patient cells Fibroblasts isolated from skin biopsies of two members of a Dutch LGMD1B family were analyzed. A male child

Fibroblasts grown on coverslips were washed twice with PBS then fixed with 4% formaldehyde in PBS for 5 min and permeabilized with 0.2% Triton X-100 in PBS for 7 min at

354

A. Muchir et al. / Experimental Cell Research 291 (2003) 352–362

room temperature as previously described [30]. After two washes with PBS, fibroblasts were incubated with the primary antibodies diluted in PBS for 1 h at room temperature. The primary antibodies used were anti-lamin A/C monoclonal antibody against the N-terminus domain (Jol5, 1:10; a kind gift from C. Hutchison, University of Durham, UK), anti-lamin B1 polyclonal antibody (1:250), anti-Nup153 monoclonal antibody (used undiluted), anti-LAP2␤ polyclonal antibody (1:8) (these three antibodies were generous gifts from J.C. Courvalin, Institut Jacques Monod, Paris, France), anti-lamin B2 monoclonal antibody (1:50, TEBU, Le Perray en Yvelines, France), anti-emerin polyclonal antibody (1:100), anti-emerin monoclonal antibody MANEM5 at a dilution of 1:50 (these two antibodies were generous gifts from G. Morris, NEWI, UK), anti-LAP2␣ polyclonal antibody (1:500, a gift from R. Foisner, Vienna Biocenter, Austria), anti-BAF polyclonal antibody (1:100, a gift from K. Wilson, Johns Hopkins School of Medicine, Baltimore, MD, USA), anti-nesprin-1␣ polyclonal antibody (1:200) [31], anti-calnexin polyclonal antibody (1:200, StressGen Biotechnologies, San Diego, CA, USA). After washes with PBS, the fibroblasts were incubated with the secondary antibodies (1:500) as described for the primary antibodies. Secondary antibodies used were Alexa fluor 568 goat anti-mouse IgG (Molecular Probes, Leiden, The Netherlands) and fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse IgG (Chemicon, UK). The cells were then washed with PBS and the slides were mounted in Mowiol with 0.1 ␮g/ml DAPI. Immunofluorescence microscopy was performed using an Axiophot microscope (Carl Zeiss). The pictures were processed with Adobe Photoshop 5.5 (Adobe Systems, San Jose, CA, USA). Nuclei with abnormalities were quantified and expressed as a percentage of the total number of nuclei. Immunolabeling experiments were performed in duplicate after each trypsination for each fibroblast culture, and an average of 500 nuclei were counted per experiment. Immunoblot analysis Fibroblasts (3 ⫻ 106) were harvested and homogenized in a 150 mM KCl, 50 mM Tris–HCl, pH 7.5, buffer in the presence of protease inhibitors (1 mM PMSF). Various concentrations of these homogenates were loaded on 10% SDS-polyacrylamide gels. After electrophoresis, proteins were transferred to a 0.2-␮m-pore-size nitrocellulose membrane (Invitrogen, Cergy-Pontoise, France) using the Novex blotting system. The primary antibodies used were antilamin A/C Jol5 monoclonal antibody (1:250) and anti-␤actin monoclonal antibody (1:1000, Sigma, Poole, UK). Secondary antibodies were HRP-conjugated rabbit antimouse and goat anti-rabbit antibodies (1:2000, DAKO, Australia). Immunoblotted bands were visualized by enhanced chemiluminescence (ECL, Amersham International, UK).

Transfection Wild-type prelamin A cDNA and lamin C cDNA fused to the tag at the N-terminal end and cloned into the expres¨ stlund and sion vector pSVK3 were kindly provided by C. O H.J. Worman (Columbia University, New York, NY, USA) [32]. Fibroblasts were transfected in 35-mm Petri dishes using LipofectAMINE (Gibco-BRL), following the manufacturer’s instructions. The cells were overlaid with the lipid–DNA complex for approximately 8 h. The fibroblasts were then allowed to grow in fresh medium for 42 h posttransfection. The transfected fibroblasts were visualized by labeling the tag with monoclonal antibody M5 (1:200, Sigma).

Results Expression of lamins A and C is reduced in heterozygous fibroblasts and absent in homozygous fibroblasts Our initial aim was to determine whether a heterozygous or homozygous Y259X LMNA nonsense mutation affects the expression of lamins A and C in cultured skin fibroblasts from patients. Cell homogenates were prepared from skin fibroblasts cultured from unaffected fetal and adult tissue, from the adult LGMD1B patient with a heterozygous mutation (⫹/mut) and from her deceased newborn grandchild (30-week-old fetus) with a homozygous mutation (mut/ mut). Western blot analysis was carried out using a monoclonal antibody that recognized the N-terminal domain of lamins A and C (Jol5) to reveal the truncated mutant lamins of 30 kDa potentially produced by the Y259X LMNA mutation. Although similar amounts of total proteins were loaded on the SDS–polyacrylamide gel as revealed by ␤-actin detection, lamins A and C were not detectable in fibroblastsmut/mut and were decreased by 52 ⫾ 9% in fibroblasts⫹/mut compared with age-matched control fibroblasts (Fig. 1). However, the 30-kDa truncated mutant lamin was not detected in patient fibroblasts. These results suggest that the Y259X mutation acts via a haplo-insufficiency mechanism since only lamins A and C produced by the normal allele are expressed and the truncated mutant lamin potentially produced by the Y259X mutation is not detectable. Homozygous Y259X mutation leads to abnormal nuclear morphology and growth We then investigated the impact of the Y259X mutation on nuclear morphology in patient fibroblasts⫹/mut and fibroblastsmut/mut using DAPI staining. Abnormal nuclear morphologies have been reported in fibroblasts carrying LMNA missense mutations [26,33] and in mouse fibroblasts lacking lamin A/C expression [34]. The majority of the control fibroblasts⫹/⫹ have a nucleus with a normal ovoid morphology (Fig. 2A). In contrast, 86.15 ⫾ 4.74% of the nuclei of

A. Muchir et al. / Experimental Cell Research 291 (2003) 352–362

355

Fig. 1. Expression of lamins A and C in fibroblasts⫹/mut and fibroblastsmut/mut. Different concentrations (5, 10 and 20 ␮g protein) of cell homogenates from cultured fibroblasts were loaded on 10% SDS–polyacrylamide gels. The N-terminal lamin A/C antibody (Jo15) was used. Immunoblotting analysis was also performed on control fetal fibroblasts and similar amounts of lamins A and C were detected in both adult and fetal control fibroblasts (data not shown). The ␤-actin antibody was used as an internal control to check that similar amounts of proteins were analyzed. Positions of molecular weight marker proteins are indicated on the left.

fibroblastsmut/mut presented an irregular shape with simple or multiple lobules of their NE (Fig. 2C, arrowheads), at the beginning of culture (passage 3 ⫽ P3) (Fig. 3A). This percentage is considerably higher than that observed in control fetal fibroblast⫹/⫹ (6.90 ⫾ 1.55% at P3, Fig. 3A). During progressive passaging of the cells (P3 to P13), the percentage of abnormal nuclei in the cultured fibroblastsmut/mut decreased to reach 2.95 ⫾ 0.92% at P13 (Fig. 3C). This percentage is identical to that observed in control fibroblasts⫹/⫹. In fibroblasts⫹/mut the majority of the nuclei were normal (Fig. 2B) and the percentage of nuclei with an abnormal morphology was not different from that of adult control fibroblasts⫹/⫹ and fluctuated between 1.25 ⫾ 0.49 and 4.25 ⫾ 1.76% (Fig. 3A). Thus, the Y259X nonsense mutation induces abnormal nuclear morphology in patient cells when present in the homozygous state. In addition, we observed heterogeneous labeling of the nuclear lobules by DAPI in fibroblastsmut/mut (Fig. 2C, arrowheads), suggesting that the nuclear shape abnormalities were accompanied by chromatin reorganization. It could be postulated that fibroblastsmut/mut with dysmorphic nuclei might not be able to divide. To assess whether the abnormalities of the NE have an effect on the fibroblast division, we analyzed the growth rates of cultured fibroblasts at P5 and P9. The proliferation curves are shown in Fig. 3D. At P5, fibroblastsmut/mut exhibiting 80% abnormally shaped nuclei proliferate at a slower rate compared with control fibroblasts (Fig. 3Da). Nine days after seeding, the number of fibroblastsmut/mut was 50% that of control fibroblasts and fibroblasts⫹/mut (95 ⫻ 103 for fibroblastsmut/mut compared with 214 ⫻ 103 for control fibroblasts and 222 ⫻ 103 for fibroblasts⫹/mut), suggesting that fibroblasts with abnormal nuclei have difficul-

ties in dividing. Proliferation curves performed at P9 confirm this observation, as fibroblastsmut/mut that exhibit at this passage only 6% abnormally shaped nuclei grow in a similar manner as control fibroblasts and fibroblasts⫹/mut (Fig. 3Db). Immunofluorescence analysis reveals abnormal localization of nuclear proteins in Y259X fibroblastsmut/mut To further characterize alterations in nuclear protein expression that might be implicated in the abnormal nuclear morphology observed in vitro, we analyzed the subcellular localization of nuclear proteins associated with lamins A and C. The distribution of these different proteins was investigated by immunofluorescence using specific antibodies. In control fibroblasts⫹/⫹, lamins A, C, B1, B2, and emerin were localized at the NE (Fig. 2A). However, in a few cells, lamins B1 and B2 were absent from the pole of the nucleus. In fibroblastsmut/mut, using the N-terminal lamin A/C monoclonal antibody (Jol5), lamins A and C were not detected (Figs. 2Cb, e). This confirms the immunoblotting results where neither lamins A and C nor truncated lamins were detected (Fig. 1). Lamins B1 and B2 were both present at the NE but were absent or greatly decreased in the nuclear lobules as well as at some of the nuclear poles (Figs. 2Cc, h, l). The abnormal pattern of lamin B1 expression was observed in 86.85 ⫾ 4.87% of fibroblastsmut/mut at the beginning of culture (P3), and was considerably higher than that observed in control fetal fibroblasts⫹/⫹ (6.95 ⫾ 2.75% at P3, Fig. 3B). The proportion of fibroblastsmut/mut with abnormal B-type lamin expression decreased throughout the time the cells were allowed to proliferate in culture, to reach

356

A.

A. Muchir et al. / Experimental Cell Research 291 (2003) 352–362

a

b

dapi

d

c

lamin A/C

e

dapi

g

f

h

a

lamin B1

lamin A/C

dapi

C.

dapi

d

emerin

i

lamin B2

b

e

g

b

d

e

g

lamin B2

k

emerin

l

emerin

lamin B1

lamin B1

f

lamin A/C

h

dapi

i

c

lamin A/C

dapi

emerin

lamin A/C

dapi

dapi

dapi

lamin B1

f

h

j

B. a

lamin A/C

dapi

emerin

c

emerin

i

lamin B2

emerin

Fig. 2. Localization of lamins and emerin in fibroblasts⫹/⫹, fibroblasts⫹/mut, and fibroblastsmut/mut. Control and LGMD1B fibroblasts were labeled with DAPI; mouse monoclonal antibodies directed against lamins A and C, emerin, and lamin B2; and rabbit polyclonal antibodies directed against lamin B1 and emerin, before analysis by conventional immunofluorescence microscopy. Immunolocalization of these nuclear proteins was carried out on control fibroblasts (A), fibroblasts⫹/mut (B), and fibroblastsmut/mut (C). Bar ⫽ 10 ␮m.

23.10 ⫾ 11.17% at P13. This decrease is parallel but not identical to that observed for the abnormal nuclear morphology (Fig. 3C). In fibroblasts⫹/mut, localization of lamins A, C, B1, and B2 showed no difference from that of control fibroblasts⫹/⫹ (Figs. 2A and B). These data suggest that a 50% reduction in lamins A and C in heterozygous cells has no visible effect on the nuclear lamina, whereas the complete loss in homozygous cells results in disorganization of the nuclear lamina. Next we analyzed the distribution of INM (LAP2␤), nuclear matrix (LAP2␣), chromatin (BAF), and NPC proteins (Nup 153). LAP2␤ and Nup 153 followed the

pattern of expression of lamins B1 and B2 in both heterozygous and homozygous fibroblasts; i.e., they were absent in nuclear lobules and/or poles in fibroblastsmut/mut (Figs. 4Bb, f) and exhibited normal expression in fibroblasts⫹/mut (data not shown). Thus, the loss of lamins A and C alters the organization of the entire NE including the nuclear lamina, the INM, and the NPC. LAP2␣ shows in a few cells a partial relocalization in poles or lobules where neither B-type lamins nor LAP2␤ or Nup153 is localized (Fig. 4Bi). BAF, a DNA binding protein [35], was localized normally in fibroblastsmut/mut, as in control fibroblasts (Figs. 4Ao, 4Bo).

A. Muchir et al. / Experimental Cell Research 291 (2003) 352–362

357

358

A.

A. Muchir et al. / Experimental Cell Research 291 (2003) 352–362

a

b

dapi

d

Nup153

e

dapi

g

j

dapi

m

n

dapi

emerin

BAF

i

dapi

l

α Nesprin-1α

emerin

n

dapi

α Lap2α

emerin

k

m

o

β Lap2β

emerin

dapi

α Nesprin-1α

emerin

lamin B1

f

h

j

l

Nup153

dapi

α Lap2α

c

e

g

i

k

dapi

β Lap2β

emerin

b

d

f

h

a

lamin B1

emerin

dapi

B.

c

o

emerin

BAF

Fig. 4. Immunofluorescence analysis of fibroblasts⫹/⫹ and fibroblastsmut/mut. Control and LGMD1B fibroblasts were labeled with DAPI; mouse monoclonal antibodies directed against lamins A and C, emerin, and Nup153; and rabbit polyclonal antibodies directed against lamin B1, LAP2␤, LAP2␣, nesprin-1␣, and BAF, before analysis by conventional immunofluorescence microscopy. Immunolocalization of these nuclear proteins was carried out on control fibroblasts (A) and fibroblastsmut/mut (B). Bar ⫽ 10 ␮m.

As reported for mouse lmna⫺/⫺ fibroblasts [34], the most dramatic consequence of the absence of lamins A and C in human fibroblastsmut/mut is the aberrant localization of emerin within the cytoplasm (Figs. 2Cf, i, k; 4Be, h, k, n) compared with the normal emerin expression at the NE in fibroblasts⫹/⫹ and fibroblasts⫹/mut (Figs. 2Af, i, n; 2Bf; 4Ae, h, k, n). Only a few fibroblastsmut/mut expressed emerin at the NE, in poles or lobules where B-type lamins were not expressed, similarly to the relocalization of LAP2␣ (Figs. 2Ck, l). Interestingly nesprin-1␣, normally expressed at the NE (Fig. 4Al), exhibits the same aberrant cytoplasmic lo-

calization as emerin in fibroblastsmut/mut (Figs. 4Bk, l). We further determined the exact localization of emerin and nesprin-1␣ within the cytoplasm. The double immunolabeling using specific antibodies against emerin and calnexin, an unglycosylated resident endoplasmic reticulum (ER) transmembrane protein [36], revealed a complete colocalisation of these two proteins in fibroblastsmut/mut (Fig. 5A), demonstrating that in the absence of lamins A and C, emerin is mislocalized to the ER. The further costaining of nesprin-1␣ and emerin revealed that the two proteins also colocalize in fibroblastsmut/mut, demonstrating that nesprin-

Fig. 3. Nuclear abnormalities in LGMD1B fibroblastsmut/mut. (A) Variation in the percentage of morphologically abnormal nuclei observed in fibroblastsmut/mut (solid bars), fibroblasts⫹/mut (dark gray bars), fetal fibroblasts⫹/⫹ (gray bars), and adult fibroblasts⫹/⫹ (open bars) at successive passages in culture. (B) Percentage of loss of lamin B1 from nuclear poles observed in fibroblastsmut/mut (solid bars), fibroblasts⫹/mut (dark gray bars), fetal fibroblasts⫹/⫹ (gray bars), and in adult fibroblasts⫹/⫹ (open bars) with the passages of fibroblasts in culture. (C) Histogram showing the evolution of abnormal morphological nuclei (solid bars) and abnormal lamin B1 expression pattern (gray bars) in fibroblastsmut/mut when the culture is expanded for up to 13 passages. Note the decrease in the percentage of abnormal nuclei with increasing number of cell passages in culture. (D) Proliferation of fibroblasts⫹/⫹, fibroblasts⫹/mut, and fibroblastsmut/mut. The curves show cell growth at passage 5 (a) and passage 9 (b) of fibroblastsmut/mut (black boxes), fibroblasts⫹/mut (gray triangles), and fibroblasts⫹/⫹ (open circles). The number of cells was determined by cell counts using a hematocytometer.

A. Muchir et al. / Experimental Cell Research 291 (2003) 352–362

A.

C.

B. a

359

a

b

c

calnexin

emerin

overlay

α nesprin-1α

emerin

d

e

f

d

e

calnexin

emerin

overlay

α nesprin-1α

emerin

overlay

a

b

c

b

c

dapi

M5

emerin

dapi

M5

α nesprin-1α

d

e

f

d

e

f

dapi

M5 (lamin A)

emerin

dapi

M5 (lamin A)

α nesprin-1α

g

h

i

g

h

i

dapi

M5 (lamin C)

emerin

dapi

M5 (lamin C)

α nesprin-1α

D. a

b

c

overlay

f

Fig. 5. Distribution of emerin and nesprin-1␣ in fibroblastsmut/mut. (A, B) Distribution of emerin (A) and nesprin-1␣ (B) observed in the ER by indirect immunofluorescence. The distribution of DNA was detected using DAPI. The pictures show an overlay of the FITC (green) and Alexa (red) channels where areas of colocalization appear yellow. Bar ⫽ 10 ␮m. (C, D) Transfection of human normal lamin cDNA in fibroblastsmut/mut. In untransfected cells (a– c), emerin and nesprin-1␣ are mislocalized in the ER. Transfected fibroblastsmut/mut expressing normal lamin A (d–f) or normal lamin C (g–i), visualized using M5 antibody directed against the tag, exhibit emerin and nesprin-1␣ localized at the nuclear envelope. Bar ⫽ 10 ␮m.

1␣, like emerin, is localized in the ER in the absence of lamins A and C (Fig. 5B). Lamins A and C are necessary for the correct anchorage of emerin and nesprin-1␣ at the nuclear envelope To confirm that lamins A are C are essential for the correct localization of emerin and nesprin-1␣ in the INM, we transiently transfected fibroblastsmut/mut with human wild-type tagged prelamin A cDNA or tagged lamin C cDNA and analyzed the distribution of emerin and nesprin-1␣ by double immunofluorescence labeling microscopy. The localization of exogenous lamin A or lamin C was monitored using M5 fluorescent antibody against the tag. Although the transfection efficiency was between 0.5 and 1%, the exogenous lamin A and lamin C were correctly localized to the nuclear lamina (Figs. 5Ce, h; 5De, h). Emerin and nesprin-1␣ were localized at the NE in all

transfected fibroblastsmut/mut either with pre-lamin A cDNA or with lamin C cDNA, while they remained distributed in the ER in untransfected fibroblastsmut/mut (Figs. 5C, 5D). These data demonstrate that anchorage of emerin and nesprin-1␣ to the NE depends on the presence of lamins A and C within the nuclear lamina; and that only one of these lamins, A or C, is sufficient to localize emerin and nesprin-1␣ to the NE in human fibroblasts.

Discussion We have reported here the first characterization of human cells carrying heterozygous and homozygous nonsense LMNA mutations. We have analyzed skin fibroblasts from an LGMD1B patient and from her deceased newborn grandchild, carrying respectively a heterozygous and a homozygous Y259X mutation (van Engelen et al., in preparation).

360

A. Muchir et al. / Experimental Cell Research 291 (2003) 352–362

This mutation leads to haplo-insufficiency of lamins A and C and produce, when lamins A and C are completely absent, dramatic modifications in nuclear morphology associated with alterations in the protein composition of the INM and aberrant localization of emerin and nesprin-1␣ in the ER. The Y259X LMNA nonsense mutation most probably acts via a haplo-insufficiency mechanism Western-blot analysis revealed the presence of only proteins produced by the wild-type LMNA allele, i.e., 50% of normal amount of lamins A and C in fibroblasts⫹/mut and no lamins A and C in fibroblastsmut/mut. The 30-kDa truncated lamin potentially resulting from the Y259X mutation was not detected, although an antibody that mapped the Nterminus domain of lamins A and C was used [37]. A similar decrease in lamin A/C expression was observed in the cardiac tissue of AD-EDMD and DCM-CD patients carrying heterozygous Q6X and E111X nonsense LMNA mutations, respectively [23,38]. The truncated mutant proteins resulting from these mutations were also not detected, i.e., a 5-amino acid peptide for Q6X [23] and a 12-kDa protein for E111X [38]. In the mouse model with a lmna exon 8 –11 deletion, the 54-kDa truncated protein resulting from this deletion was not detected, only the wild-type lmna allele produced full-length lamins A and C [34]. All of these data support the idea that nonsense LMNA mutations act via a haplo-insufficiency mechanism. The loss of lamins A and C disturbs nuclear envelope architecture This haplo-insufficiency of lamins A and C, with a 50% reduction in lamins A and C as a consequence of the heterozygous Y259X mutation, produces no visible abnormalities in cultured skin fibroblasts⫹/mut. By contrast, the homozygous Y259X mutation results in a dramatic dysmorphic nuclear morphology with multiple lobules in fibroblastsmut/mut. In addition to the complete loss of lamins A and C, the composition of the NE proteins was greatly altered in these nuclei with a partial or total absence of lamins B1, B2, LAP2␤, and Nup153 in the lobules, similarly to abnormalities reported in mouse lmna⫺/⫺ fibroblasts [34]. Thus, in both human and mouse cells, the absence of lamins A and C induces significant alterations of the NE. Interestingly, the presence of these lobules seems to perturb cell division. Indeed, fibroblastmut/mut cultures containing large proportions of cells with abnormally shaped nuclei grow more slowly than control cells (Fig. 3Da). In addition the rate of cell proliferation normalizes when the number of these abnormally shaped nuclei declines with increased cell passage number (Fig. 3Db). The most likely explanation of this observation is that cells with dysmorphic nuclei are likely not able to divide properly and thus cells with normally shaped nuclei have a growing rate advantage, divide more rapidly, and replace progressively all the cells

with dysmorphic nuclei. The abnormal nuclear shape of the nuclei seems to be linked to the alteration in the protein composition of the NE. The only proteins that show modulation of their alteration in parallel to the disappearance of the abnormal nuclear shape are the B-type lamins, Nup153, and LAP2␤, whereas the absence of lamins A and C, emerin, and nesprin-1␣ from the NE remains constant in all cells independently of their nuclear shape. One could speculate that the absence of lamins A and C promotes a reorganization of the NE protein composition with the regional absence of B-type lamins, Nup153, and LAP2␤ at one pole of the nuclei. This contributes to the formation of lobules and leads eventually to unable cell division or at least to reduce this process, leaving, thus, only cells with normal expression of B-type lamins, Nup153, and LAP2␤, dividing. In this way. RNAi knockdown of B-type lamins or of Nup153 affects cell growth in HeLa cells [39]. However, when most fibroblastsmut/mut exhibit normally shaped nuclei and proliferate as controls cells (Fig. 3Db), around 20% of the cells remain with the absence of lamin B1 at one pole. In our cell growth rate evaluation, we might not detect a difference in proliferation due to only 20% of the cells. Specific evaluation of the growing rate of cells exhibiting abnormal B-type lamin expression at one pole would provide further insight into the implication of B-type lamin and other INM proteins in this process. Lamins A and C are essential for correct anchorage of emerin and nesprin-1␣ at the NE in human cells The most dramatic nuclear alterations observed in fibroblasts carrying a homozygous Y259X LMNA mutation were the mislocalization of emerin and nesprin-1␣ into the ER. These alterations are present in all cells. Delocalization of emerin to the ER was reported in embryonic mouse lmna⫺/⫺ fibroblasts [34] and in human HeLa cells knocked down for lamins A and C by RNAi [39], but this is the first demonstration that nesprin-1␣ is displaced to the ER as emerin is. The direct interaction between lamin A and nesprin-1␣ that has been demonstrated in vitro [5] appears to be necessary for the correct targeting of nesprin-1␣ to the INM. Accordingly, we have demonstrated by transient transfection that anchorage of nesprin-1␣ to the NE, like emerin, depends on the presence of either lamin A or lamin C within the nuclear lamina. Indeed, only one of these lamins, A or C, is sufficient to localize nesprin-1␣ and emerin to the NE, demonstrating for the first time both nesprin-1␣ and emerin have direct links with lamin C. This suggests also that lamin A/C mutations may contribute to striated muscle disease by disrupting the structural network that interlinks emerin, lamin A, lamin C, and nesprin-1␣ at the INM. Interestingly, it was previously reported that transfection with lamin C cDNA was not able to rescue the ER distribution of emerin in either mouse lmna⫺/⫺ fibroblasts [40] or human SW13 cancer cells that do not express lamin A [41].

A. Muchir et al. / Experimental Cell Research 291 (2003) 352–362

In the case of mouse lmna⫺/⫺ fibroblasts, this negative result could be due to the fact that a human cDNA was used to transfect mouse cells, although there is a large homology between mouse and human of lamin C protein. As for SW13 cancer cells, human lamin C cDNA was transfected in human cells. However, the exogenous GFP-lamin C accumulated in intranuclear aggregates and never exhibited homogenous expression at the nuclear rim [41]. Therefore, it would be difficult for this exogenous lamin C to attach emerin at the nuclear rim. In addition, cancer cells present a much more complex phenotype than skin fibroblasts, the latter allowing easier analysis of interaction between nuclear envelope proteins. Finally, we also observed that a few fibroblastsmut/mut had emerin anchored at one pole of their nucleus where no B-type lamins, Nup153, or LAP2␤ were expressed, suggesting that other proteins may interact with emerin to maintain it at the NE. A similar pattern of expression was observed in the hearts of lmna⫺/⫺ mice, where 20% of the cardiac nuclei presented a polar distribution of emerin [34].

Conclusion In human cells, the loss of lamins A and C is associated with nuclear architectural defects including a change in nuclear shape; disorganization of nuclear lamina, INM proteins, and NPC; and mislocalization of nesprin-1␣ and emerin to the ER. Similar tissue or cell fragility has been described previously in syndromes affecting other intermediate filament proteins [42,43]. In laminopathies affecting the striated muscles, the NE is vulnerable to damage in contractile tissues such as skeletal and cardiac muscle. This vulnerability may be increased because, in striated muscle, lamin B1 is expressed at reduced levels [44]. In the deceased newborn child carrying the homozygous nonsense mutation, the absence of lamins A and C may have disturbed the association of the remaining B-type lamins with the NE; the nuclear lamina would lose its structural properties, particularly in striated muscles. Indeed, at autopsy the fetus exhibited multiple joint contractures, severe dystrophy with fibrosis, large variation in fiber diameter, and centralization of nuclei in the deltoid, intercostal, psoas, and quadriceps muscles, and there was almost complete absence of muscle fibers in the diaphragm (van Engelen et al., in preparation). In contrast, we observed that the heterozygous Y259X nonsense mutation does not promote these visible abnormalities in skin fibroblasts, although this heterozygous mutation does lead to a LGMD1B phenotype in the adult. One unique mechanism could not therefore explain the pathophysiology of this disease. In addition to mechanical stress implicated in contractile tissues, other pathophysiological features of the disease may arise from downstream effects on chromatin structure or gene expression that are caused by nuclear lamina disorganization. To obtain a better understanding of the consequences of these lamin A/C mu-

361

tations implicated in muscular disorders, it seems important to analyze other human cells and tissues carrying various LMNA mutations. In summary, the present report demonstrates for the first time the essential role of lamins A and C in human fibroblasts and identifies the molecular and cellular events leading to NE disorganization, which represents the first step toward the disease.

Acknowledgments We are grateful to Dr. Christopher Hutchison, Dr. JeanClaude Courvalin, Dr. Roland Foisner, Dr. Kate Wilson, and Dr. Glenn E. Morris for providing antibodies. We thank Dr. ¨ stlund for providing Howard J. Worman and Dr. Cecilia O wild-type prelamin A and lamin C cDNAs. We thank Emmanuelle Lace`ne, Andre´ e Rouch and Philippe Bozin for technical assistance. We thank Dr. Gillian Butler-Browne for fruitful discussion. This study was supported by grants from the European Union Fifth Framework (MYO-CLUSTER/EUROMEN Contract QLG1-1999-00870), Association Franc¸ aise contre les Myopathies (AFM, Grant 8185), and Human Frontiers Science Program (Grant RGP0057/2001-M101).

References [1] H.J. Worman, J.C. Courvalin, The inner nuclear membrane, J. Membr. Biol. 177 (2000) 1–11. [2] C. Stewart, B. Burke, Teratocarcinoma stem cells and early mouse embryos contain only a single major lamin polypeptide closely resembling lamin B, Cell 51 (1987) 383–392. [3] M.N. Guilly, J.P. Kolb, F. Gosti, F. Godeau, J.C. Courvalin, Lamins A and C are not expressed at early stages of human lymphocyte differentiation, Exp. Cell Res. 189 (1990) 145–147. [4] U. Aebi, J. Cohn, L. Buhle, L. Gerace, The nuclear lamina is a meshwork of intermediate-type filaments, Nature 323 (1986) 560–564. [5] J.M. Mislow, J.M. Holaska, M.S. Kim, K.K. Lee, M. Segura-Totten, K.L. Wilson, E.M. McNally, Nesprin-1alpha self-associates and binds directly to emerin and lamin A in vitro, FEBS Lett. 525 (2002) 135–140. [6] L. Clements, S. Manilal, D.R. Love, G.E. Morris, Direct interaction between emerin and lamin A, Biochem. Biophys. Res. Commun. 267 (2000) 709 –714. [7] M. Sakaki, H. Koike, N. Takahashi, N. Sasagawa, S. Tomioka, K. Arahata, S. Ishiura, Interaction between emerin and nuclear lamins, J. Biochem. 129 (2001) 321–327. [8] R. Foisner, L. Gerace, Integral membrane proteins of the nuclear envelope interact with lamins and chromosomes, and binding is modulated by mitotic phosphorylation, Cell 73 (1993) 1267–1279. [9] R.P. Aaronson, G. Blobel, Isolation of nuclear pore complexes in association with a lamina, Proc. Natl. Acad. Sci. USA 72 (1975) 1007–1011. [10] T. Dechat, B. Korbei, O.A. Vaughan, S. Vlcek, C.J. Hutchison, R. Foisner, Lamina-associated polypeptide 2alpha binds intranuclear A-type lamins, J. Cell Sci. 113 (Pt. 19) (2000) 3473–3484. [11] B. Burke, On the cell-free association of lamins A and C with metaphase chromosomes, Exp. Cell Res. 186 (1990) 169 –176. [12] C.A. Glass, J.R. Glass, H. Taniura, K.W. Hasel, J.M. Blevitt, L. Gerace, The alpha-helical rod domain of human lamins A and C contains a chromatin binding site, EMBO J. 12 (1993) 4413– 4424.

362

A. Muchir et al. / Experimental Cell Research 291 (2003) 352–362

[13] V.V. Stierle, J. Couprie, C. Ostlund, I. Krimm, S. Zinn-Justin, P. Hossenlopp, H.J. Worman, J.C. Courvalin, I. Duband-Goulet, The carboxyl-terminal region common to lamins A and C contains a DNA binding domain, Biochemistry 42 (2003) 4819 – 4828. [14] R.D. Goldman, Y. Gruenbaum, R.D. Moir, D.K. Shumaker, T.P. Spann, Nuclear lamins: building blocks of nuclear architecture, Genes Dev. 16 (2002) 533–547. [15] T.P. Spann, R.D. Moir, A.E. Goldman, R. Stick, R.D. Goldman, Disruption of nuclear lamin organization alters the distribution of replication factors and inhibits DNA synthesis, J. Cell Biol. 136 (1997) 1201–1212. [16] R.D. Moir, T.P. Spann, H. Herrmann, R.D. Goldman, Disruption of nuclear lamin organization blocks the elongation phase of DNA replication, J. Cell Biol. 149 (2000) 1179 –1192. [17] G. Jagatheesan, S. Thanumalayan, B. Muralikrishna, N. Rangaraj, A. Karande, V. K. Parnaik, Colocalization of intranuclear lamin foci with RNA splicing factors, J. Cell Sci. 112 (1999) 4651– 4661. [18] T.P. Spann, A.E. Goldman, C. Wang, S. Huang, R.D. Goldman, Alteration of nuclear lamin organization inhibits RNA polymerase II-dependent transcription, J. Cell Biol. 156 (2002) 603– 608. [19] G. Bonne, M.R. Di Barletta, S. Varnous, H. Becane, E.H. Hammouda, L. Merlini, F. Muntoni, C.R. Greenberg, F. Gary, J.A. Urtizberea, D. Duboc, M. Fardeau, D. Toniolo, K. Schwartz, Mutations in the gene encoding lamin A/C cause autosomal dominant Emery– Dreifuss muscular dystrophy, Nat. Genet. 21 (1999) 285–288. [20] M.R. di Barletta, E. Ricci, G. Galluzzi, P. Tonali, M. Mora, L. Morandi, A. Romorini, T. Voit, K.H. Orstavik, L. Merlini, C. Trevisan, V. Biancalana, I. Housmanowa-Petrusewicz, S. Bione, R. Ricotti, K. Schwartz, G. Bonne, D. Toniolo, Different mutations in the LMNA gene cause autosomal dominant and autosomal recessive Emery–Dreifuss muscular dystrophy, Am. J. Hum. Genet. 66 (2000) 1407–1412. [21] A. Muchir, G. Bonne, A.J. van der Kooi, M. van Meegen, F. Baas, P.A. Bolhuis, M. de Visser, K. Schwartz, Identification of mutations in the gene encoding lamins A/C in autosomal dominant limb girdle muscular dystrophy with atrioventricular conduction disturbances (LGMD1B), Hum. Mol. Genet. 9 (2000) 1453–1459. [22] D. Fatkin, C. MacRae, T. Sasaki, M.R. Wolff, M. Porcu, M. Frenneaux, J. Atherton, H.J. Vidaillet, S. Spudich, U. De Girolami, J.G. Seidman, C.E. Seidman, F. Muntoni, G. Muehle, W. Johnson, B. McDonough, Missense mutations in the rod domain of the lamin A/C gene as causes of dilated cardiomyopathy and conduction-system disease, N. Engl. J. Med. 341 (1999) 1715–1724. [23] H.-M. Be´ cane, G. Bonne, S. Varnous, A. Muchir, V. Ortega, E.H. Hammouda, J.-A. Urtizberea, T. Lavergne, M. Fardeau, B. Eymard, S. Weber, K. Schwartz, D. Duboc, High incidence of sudden death of conduction system and myocardial disease due to lamins A/C gene mutation, Pacing Clin. Electrophysiol. 23 (2000) 1661–1666. [24] S. Shackleton, D.J. Lloyd, S.N. Jackson, R. Evans, M.F. Niermeijer, B.M. Singh, H. Schmidt, G. Brabant, S. Kumar, P.N. Durrington, S. Gregory, S. O’Rahilly, R.C. Trembath, LMNA, encoding lamin A/C, is mutated in partial lipodystrophy, Nat. Genet. 24 (2000) 153–156. [25] A. De Sandre-Giovannoli, M. Chaouch, S. Kozlov, J.M. Vallat, M. Tazir, N. Kassouri, P. Szepetowski, T. Hammadouche, A. Vandenberghe, C.L. Stewart, D. Grid, N. Levy, Homozygous defects in LMNA. encoding lamin A/C nuclear-envelope proteins, cause autosomal recessive axonal neuropathy in human (Charcot–Marie–Tooth disorder type 2) and mouse, Am. J. Hum. Genet. 70 (2002) 726 –736. [26] G. Novelli, A. Muchir, F. Sangiuolo, A. Helbling-Leclerc, M. Rosaria d’Apice, C. Massart, F. Capon, P. Sbraccia, M. Federici, R. Lauro, C. Tudisco, R. Pallota, G. Scarano, B. Dallapiccola, L. Merlini, G. Bonne, Mandibuloacral dysplasia is caused by a mutation in LMNA encoding lamins A/C, Am. J. Hum. Genet. 71 (2002) 426 – 431. [27] De Sandre-GiovannoliA.BernardR.Cau, P. Navarro, C. Amiel, J. Boccacio, I. Lyonnet, S. Stewart, C. L. Munnich, A. Le Merrer, M. Levy. N. Lamin A truncation in Hutchinson–Gilford progeria, Science 2003.

[28] M. Eriksson, W.T. Brown, L.B. Gordon, M.W. Glynn, J. Singer, L. Scott, M.R. Erdos, C.M. Robbins, T.Y. Moses, P. Berglund, A. Dutra, E. Pak, S. Durkin, A.B. Csoka, M. Boehnke, T.W. Glover, F.S. Collins, Recurrent de novo point mutations in lamin A cause Hutchinson–Gilford progeria syndrome, Nature 25 (2003) 25. [29] A.J. van der Kooi, T.M. Ledderhof, W.G. de Voogt, C.J. Res, G. Bouwsma, D. Troost, H.F. Busch, A.E. Becker, M. de Visser, A newly recognized autosomal dominant limb girdle muscular dystrophy with cardiac involvement, Ann. Neurol. 39 (1996) 636 – 642. [30] L. Yang, T. Guan, L. Gerace, Integral membrane proteins of the nuclear envelope are dispersed throughout the endoplasmic reticulum during mitosis, J. Cell Biol. 137 (1997) 1199 –1210. [31] J.M. Mislow, M.S. Kim, D.B. Davis, E.M. McNally, Myne-1, a spectrin repeat transmembrane protein of the myocyte inner nuclear membrane, interacts with lamin A/C, J. Cell Sci. 115 (2002) 61–70. [32] C. Ostlund, G. Bonne, K. Schwartz, H.J. Worman, Properties of lamin A mutants found in Emery–Dreifuss muscular dystrophy, cardiomyopathy and Dunnigan-type partial lipodystrophy, J. Cell Sci. 114 (2001) 4435– 4445. [33] C. Vigouroux, M. Auclair, E. Dubosclard, M. Pouchelet, J. Capeau, J.C. Courvalin, B. Buendia, Nuclear envelope disorganization in fibroblasts from lipodystrophic patients with heterozygous R482Q/W mutations in lamin A/C gene, J. Cell Sci. 114 (2001) 4459 – 4468. [34] T. Sullivan, D. Escalante-Alcalde, H. Bhatt, M. Anver, N. Bhat, K. Nagashima, C.L. Stewart, B. Burke, Loss of A-type lamin expression compromises nuclear envelope integrity leading to muscular dystrophy, J. Cell Biol. 147 (1999) 913–920. [35] M.S. Lee, R. Craigie, Protection of retroviral DNA from autointegration: involvement of a cellular factor, Proc. Natl. Acad. Sci. USA 91 (1994) 9823–9827. [36] L.W. Tjoelker, C.E. Seyfried, R.L. Eddy Jr., M.G. Byers, T.B. Shows, J. Calderon, R.B. Schreiber, P.W. Gray, Human, mouse, and rat calnexin cDNA cloning: identification of potential calcium binding motifs and gene localization to human chromosome 5, Biochemistry 33 (1994) 3229 –3236. [37] J.A. Dyer, I.R. Kill, G. Pugh, R.A. Quinlan, E.B. Lane, C.J. Hutchison, Cell cycle changes in A-type lamin associations detected in human dermal fibroblasts using monoclonal antibodies, Chromosome Res. 5 (1997) 383–394. [38] E. Arbustini, A. Pilotto, A. Repetto, M. Grasso, A. Negri, M. Diegoli, C. Campana, L. Scelsi, E. Baldini, A. Gavazzi, L. Tavazzi, Autosomal dominant dilated cardiomyopathy with atrioventricular block: a lamin A/C defect-related disease, J. Am. Coll. Cardiol. 39 (2002) 981–990. [39] J. Harborth, S.M. Elbashir, K. Bechert, T. Tuschl, K. Weber, Identification of essential genes in cultured mammalian cells using small interfering RNAs, J. Cell Sci. 114 (2001) 4557– 4565. [40] W.H. Raharjo, P. Enarson, T. Sullivan, C.L. Stewart, B. Burke, Nuclear envelope defects associated with LMNA mutations cause dilated cardiomyopathy and Emery–Dreifuss muscular dystrophy, J. Cell Sci. 114 (2001) 4447– 4457. [41] O.A. Vaughan, M. Alvarez-Reyes, J.M. Bridger, J.L.V. Broers, F.C.S. Ramaekers, M. Wehnert, G.E. Morris, W.G.F. Whitfield, C.J. Hutchison, Both emerin and lamin C depend on lamin A for localization at the nuclear envelope, J. Cell Sci. 114 (2001) 2577–2590. [42] Y.M. Chan, Q.C. Yu, J.D. Fine, E. Fuchs, The genetic basis of Weber-Cockayne epidermolysis bullosa simplex, Proc. Natl. Acad. Sci. USA 90 (1993) 7414 –7418. [43] E.L. Rugg, S.M. Morley, F.J. Smith, M. Boxer, M.J. Tidman, H. Navsaria, I.M. Leigh, E.B. Lane, Missing links: Weber–Cockayne keratin mutations implicate the L12 linker domain in effective cytoskeleton function, Nat. Genet. 5 (1993) 294 –300. [44] S. Manilal, C.A. Sewry, A.P.N. Man, P. Gobbi, S. Hawkes, D.R. Love, G.E. Morris, Distribution of emerin and lamins in the heart and implications for Emery–Dreifuss muscular dystrophy, Hum. Mol. Genet. 8 (1999) 353–359.